--- /dev/null
+Now that you have the lsof distribution, I suggest:
+
+* If you're unfamiliar with lsof, read 00README for information on
+ Configuring and building lsof, 00QUICKSTART for tips on using lsof.
+
+ If you're too impatient for that, do this:
+
+ $ ./Configure <put your UNIX dialect's abbreviation here>
+ (Do the inventory step, as you prefer.)
+ (Do the customization step, as you prefer.)
+ $ make
+ $ ./lsof -h
+
+ To get a list of UNIX dialect abbreviations:
+
+ $ Configure -h
+
+ Please don't be impatient -- read the documentation first.
+
+* Read the current distribution's details in 00DIST.
+
+* If you want technical details, read 00DCACHE and 00PORTING.
+
+* If you want to cross-configure, read 00XCONFIG.
+
+* Use the test suite, described in 00TEST, by:
+
+ $ cd tests
+ $ make
+
+ and possibly:
+
+ $ make opt
+
+* If you're having trouble, read 00FAQ. (Please read 00FAQ before
+ you send a bug report.)
+
+* Lsof contributors may find their names in 00CREDITS. (Thanks, again.)
+
+* Read the lsof.man page file. Its nroff source is in lsof.8.
+
+* Consider subscribing to the lsof-l mailing list -- read 00LSOF-L
+ for details.
+
+
+Vic Abell <abe@purdue.edu>
+April 19, 2002
--- /dev/null
+A tour of the lsof_4.82 distribution:
+
+ 00.README.FIRST_4.82 is this file.
+
+ README.lsof_4.82 contains distribution and security information.
+
+ RELEASE.SUMMARY_4.82 contains a summary of the lsof 4.82
+ distribution.
+
+ lsof_4.82_src.tar is the lsof 4.82 source tar archive.
+
+ lsof_4.82_src.tar.sig is a detached GPG certificate for
+ lsof_4.82_src.tar.
+
+I suggest you follow these steps:
+
+1. Read 00.README.FIRST_4.82.
+
+2. Read README.lsof_4.82 and follow its instructions to verify
+ the authenticity of lsof_4.82_src.tar.
+
+3. Unpack lsof_4.82_src.tar -- use `tar xf lsof_4.82_src.tar`.
+ That will produce an lsof_4.82_src sub-directory.
+
+4. Change to the lsof_4.82_src sub-directory and read its
+ 00.README.FIRST file.
+
+Vic Abell <abe@purdue.edu>
+Wed Mar 25 15:26:14 EDT 2009
--- /dev/null
+
+ Lsof Credits
+
+I owe an enormous debt to the users of lsof who have contributed
+to its steady growth. The size of the list of people who have
+helped me, while it has grown too large to include in the lsof man
+page any more, is a testimonial to their generosity.
+
+First I acknowledge a debt to the work of Dan Bernstein, Michael
+``Ford'' Ditto, Tom Dunigan, Alexander Dupuy, Vik Lall, Ray Moody,
+C. Spencer, Michael Spitzer and those who wrote Berkeley's fstat
+program, all contributors to lsof's predecessors.
+
+I thank Doug McKenzie for his HP-UX proctor program and Rich Kulawiec
+for pointing it out.
+
+Finally I thank all the following people who have used lsof, pointed
+out its flaws, described its shortcomings, offered suggestions for
+improving it, supplied code for it, gave me technical advice, and
+provided test systems where I was able to do development work.
+
+ Szilveszter Adam
+ David Addison
+ Elias Halldor Agustsson
+ Per Allansson
+ Jim Ankenbrandt
+ Richard Allen
+ Thomas Anders
+ Ric Anderson
+ Stuart Anderson
+ Michael Antlitz
+ Marc Auslander
+ Tigran Aivazian
+ Jos Backus
+ David Bacon
+ Alexis Ballier
+ Scott Ballew
+ Ade Barkah
+ Alon Bar-Lev
+ Anthony Baxter
+ John Beacom
+ Bruce Beare
+ M. Jay Beck
+ Bill Behr
+ Michael Beirne
+ Marc Bejarano
+ Andrew Bell
+ Steve Bellenot
+ Robert Benites
+ Ulrich Bernhard
+ Peter J. Bertoncini
+ Dave Bianchi
+ Mark Bixby
+ Allan Black
+ Jan Blunck
+ Achim Bohnet
+ Steve Bonds
+ Mark Bonsack
+ Volker Borchert
+ Bill Bormann
+ Ermin Borovac
+ Heddy Boubaker
+ Pieter Bowman
+ Michael Bracewell
+ H. Merijn Brand
+ Danny Braniss
+ Thomas Braunbeck
+ Kieran Broadfoot
+ Dean Brock
+ Hal Brooks
+ Andrew Brown
+ Jim Brown
+ Michael Bryan
+ Matthew Burt
+ Pierfrancesco Caci
+ Bill Campbell
+ David Capshaw
+ John Caruso
+ Jon Champlin
+ Kris Chandrasekhar
+ Albert Chin-A-Young
+ Bernt Christandl
+ Marc Christensen
+ Hans Petter Christiansen
+ Tom Christiansen
+ Yves Christophe
+ Richard Chycoski
+ A. Channing Clark
+ Axel Clauberg
+ John Clear
+ David Clissold
+ Richard Coley
+ John Colgrave
+ David Comay
+ Lionel Cons
+ Bob Cook
+ Patrick Connor
+ Carl Cook
+ Jim Cooper
+ Roger Cornelius
+ Doug Crabill
+ Eric Cronin
+ Kim Culhan
+ Dave Curry
+ Robert Dahlem
+ Guy Dallaire
+ D. Chris Daniels
+ Renata Maria Dart
+ Ian Darwin
+ Carl E. Davidson
+ Will Day
+ Frederic Delanoy
+ Mike Depot
+ Steve Dibbell
+ Hugh Dickins
+ David DiGiacomo
+ Casper Dik
+ John DiMarco
+ Don Draper
+ Michel Dubois
+ Eric Dumazet
+ Dick Dunbar
+ Marc Duponcheel
+ Jan Dvorak
+ Calle Dybedahl
+ John Dzubera
+ Jeff Earickson
+ Greg Earle
+ Bernd Eckenfels
+ Niklas Edmundsson
+ Philip Edwards
+ Robert Ehrlich
+ Mark W. Eichin
+ Doug Eldred
+ Scott Ellentuch
+ Tom Endo
+ Craig Everhart
+ Chris Evert
+ Bob Farmer
+ Sami Farin
+ Mike Feldman
+ Quentin Fennessy
+ Ian Fitchet
+ Toralf Foerster
+ Bob Foertsch
+ Pierre-Yves Fontaniere
+ Ralph Forsythe
+ Jason Fortezzo
+ Mike Fraser
+ Curt Freeland
+ Terry Friedrichsen
+ Harvey Garner
+ Carson Gaspar
+ Stuart D. Gathman
+ Brian L. Gentry
+ Dave Gilbert
+ Steve Ginsberg
+ Edwin Groothuis
+ Jin Guojun
+ Kurt Gollhardt
+ Roman Gollent
+ Steve Gonczi
+ Julian Gordon
+ Marcin Gozdalik
+ Henry Grebler
+ Richard Green
+ Chaskiel Grundman
+ Armin Gruner
+ David Gutierrez
+ Robert Hall
+ Garner Halloran
+ Adam Hammer
+ Charles Hannum
+ Vlad Harchev
+ Craig Harmer
+ Michael Haro
+ Peter Harvey
+ Steinar Haug
+ Sheldon Hearn
+ John Heasley
+ Wolfgang Hecht
+ Janet Hempstead
+ Michael Hennecke
+ Randolph J. Herber
+ Andrew Hill
+ Kurt Hillig
+ Steven Hinkle
+ Paul Hite
+ Billy Ho
+ Brett Hogden
+ Gaylord Holder
+ Kjetil Torgrim Homme
+ Pekka Honkanen
+ Jeffrey C. Honig
+ Heidi Hornstein
+ Michael A. Hovan III
+ Barbara Howe
+ J. Nelson Howell
+ Jeff Howie
+ Louis Huemiller
+ John Hughes
+ Gerrit Huizenga
+ Peter Ilieve
+ Mayer Ilovitz
+ Gregory A. Ivanov
+ John Jackson
+ Kurt Jaeger
+ Edward Jajko
+ Marian Jancar
+ Paul Jarc
+ Jakub Jelinek
+ Robert Jelinek
+ Bruce Jerrick
+ Carl Johnson
+ Dion Johnson
+ Jeff Johnson
+ Douglas B. Jones
+ LaMont Jones
+ Peter Jordan
+ Arne H. Juul
+ Pasi Kaara
+ Frank Kaefer
+ Keith Kalet
+ Claus Kalle
+ Henri Karrenbeld
+ Amir Katz
+ Henry Katz
+ Kawaljeet Kaur
+ Kris Kennaway
+ Terry Kennedy
+ Shane Kenney
+ Andrew Kephart
+ Robert Kiessling
+ Joshua Kinard
+ Don Kirouac
+ Steve Kirsch
+ Philip Kizer
+ Thomas Klausner
+ Roger Klorese
+ Peter Klosky
+ Przemek Klosowski
+ Angelos D. Keromytis
+ Radko Keves
+ Valdis Kletnieks
+ Chris Kordish
+ Alek O. Komarnitsky
+ Joseph Kowalski
+ Christian Krackowizer
+ Paul Kranenburg
+ Troyan Krastev
+ Brad Krebs
+ Alex Kreis
+ Johannes Kroeger
+ Vincent Kujala
+ Ken Laing
+ Shirley Lam
+ Erwin Lansing
+ Victoria H. Lau
+ Markus Lautenbacher
+ Steve Lacey
+ Marc Aurele La France
+ Chad R. Larson
+ Steve Laubscher
+ Andrei V. Lavreniyuk
+ Loc Le
+ Tin Le
+ Diane Lebel
+ Francis Le Bourse
+ Kyungjoon Lee
+ Marty Leisner
+ Maciej Lesniewski
+ Stuart Levy
+ Ben Lewis
+ Michael Lewis
+ Angel Li
+ Ambrose Li
+ Wendy Lin
+ Carl E. Lindberg
+ Onno van der Linden
+ Johan Lindquist
+ James Lingard
+ Jason Lingohr
+ Robert Lipe
+ Gabor Liptak
+ Friedel Loinger
+ Michael Long
+ Pete Lord
+ Steve Logue
+ Bela Lubkin
+ Pav Lucistnik
+ Horst Luehrsen
+ Andreas Luik
+ Timothy J. Luoma
+ Michael Mackenzie
+ Lawrence MacIntyre
+ Benson Margulies
+ Claude Marinier
+ Chris Markle
+ Roy Marples
+ Eberhard Mater
+ James Mathiesen
+ Tom Matthews
+ Fletcher Mattox
+ David Mazieres
+ Brian McAllister
+ Scott McClung
+ Dale McCluskey
+ Terry McCoy
+ Sean McDermott
+ Duncan McEwan
+ Dwight McKay
+ William McVey
+ Eric McWhorter
+ Marjo F. Mercado
+ Dan Mercer
+ Bill Melvin
+ Andrew Merril
+ Richard van Meurs
+ Jim Mewes
+ Gary Millen
+ Timothy Miller
+ Davin Milun
+ Yuliy Minchev
+ Jim Mintha
+ Mike Miscevic
+ Arkadiusz Miskiewicz
+ Janardhan Molumuri
+ Nasser Momtaheni
+ Laurent Montaron
+ Phillip Moore
+ Dmitry Morozovsky
+ John Paul Morrison
+ John Gardiner Myers
+ Jeffrey Mogul
+ Dave Morrison
+ Pat Myrto
+ Toshiya Nakamura
+ Filippo Natali
+ Allan Nathanson
+ Chance Neale
+ Dan Nelson
+ Vladislav Nespor
+ Bjorn S. Nilsson
+ Anders Nordby
+ Joseph J. Nuspl Jr.
+ David O'Brien
+ Alexandre Oliva
+ Craig B. Olofson
+ Dave Olson
+ Rainer Orth
+ Sergey A. Osokin
+ Keith Parks
+ Will Partain
+ Vasco Pedro
+ Mark Peek
+ Ezra Peisach
+ Bill Pemberton
+ Lee Penn
+ Gildas Perrot
+ Jesse Perry
+ Nathan Peterson
+ Dominique Petitpierre
+ Hung Pham
+ Ray Phillips
+ Francois Pinard
+ Alex Podlecki
+ Lutz Poetschulat,
+ John Polstra
+ Scott Presnell
+ Mark Price
+ Philippe-Andre Prindeville
+ David Putz
+ Tom Qin
+ Kurtis Rader
+ Peter Radig
+ Jean-Pierre Radley
+ Tim Ramsey
+ Dewan Rashid
+ Richard J. Rauenzahn
+ Louis Rayman
+ Brian Redman
+ Eric S. Raymond
+ Erwin Reyns
+ Aaron Rhodes
+ Jim Reid
+ Jean-Luc Richier
+ Ingimar Robertson
+ Sylvain Robitaille
+ Larry Rogers
+ Malgorzata Roos
+ Larry Rosenman
+ Stephan Rossi
+ Kevin Ruderman
+ Wolfgang Rupprecht
+ Pavol Rusnak
+ Conrad J. Sabatier
+ Klaus Saggerer
+ Chris Schanzle
+ Igor Schein
+ Horst Scheuermann
+ Michael Schmitz
+ Larry Schwimmer
+ Hendrik G. Seliger
+ Igor V. Semenyuk
+ Jonathan Sergent
+ Frank Sanders
+ Berkley Shands
+ Gregory Neil Shapiro
+ Eyal Shaynis
+ Michael Shields
+ Wesley Shields
+ Philip Shin
+ Anthony Shortland
+ Dave Sill
+ John Silva
+ Chuck Silvers
+ Gerry Singleton
+ Leonard Sitongia
+ Kevin Smallwood
+ Curt Smith
+ Ben Smithurst
+ Douglas R. Smith
+ Kevin Smith
+ Chang Song
+ Josh Soref
+ John Speno
+ Kenneth Stailey
+ Piet Starreveld
+ David Steiner
+ Charles Stephens
+ Marc Stephenson
+ Chip Stettler
+ Dave Stevens
+ Jeff Stewart
+ Diana Stockdale
+ Andreas Stolcke
+ Jeff Stoner
+ Sushila Subramanian
+ Jan Ole Suhr
+ Mike Sullivan
+ Patrick D. Sullivan
+ Peter Svensson
+ Chris Sylvain
+ Paul Szabo
+ Dale Talcott
+ Jon A. Tankersley
+ Jan Tax
+ Samuel Thibault
+ Andy Thomas
+ Matthew Thurmaier
+ Chris Timmons
+ Andrzej Tobola
+ R. Lindsay Todd
+ Zdenko Tomasic
+ Michael Townsend
+ Linus Torvalds
+ Mike Tracy
+ Dan Trinkle
+ Lars Tunkrans
+ Lenny Turetsky
+ Kevin Vajk
+ Peter Valchev
+ John R. Vanderpool
+ Peter Van Epp
+ Peter C. Vernam
+ Peter Vines
+ Bob Ward
+ Jules van Weerden
+ Tom Weaver
+ Fernando A.B. Whitaker
+ Tom Whitty
+ Carson Wilson
+ David J. Wilson
+ Frank Winkler
+ Marc Winkler
+ Mark Vasoll
+ Holger VanKoll
+ Robert Vernon
+ Joep Vesseur
+ Larry Virden
+ Jos Vos
+ Jun Biao Wang
+ Christopher J Warweg
+ Bill Watson
+ Florian M. Weps
+ Joel White
+ Paul Wickman
+ Eric Williams
+ Steve Williams
+ Steve Wilson
+ Erich Wimmer
+ Wally Winzer, Jr.
+ Patrick Wolfe
+ Stephen Woods
+ James Woodward
+ Scott Worley
+ Joshua Wright
+ Sailu Yallapragada
+ Donna Yobs
+ Ron Young
+ Blair Zajac
+ Karel Zak
+ Donald Zoch
+ Malcom Zung
+ and Waldemar Zurowski
+
+If I have omitted a contributor's name, the fault is wholly mine,
+and I apologize for the error.
+
+
+Vic Abell <abe@purdue.edu>
+March 25, 2009
--- /dev/null
+
+ Configuring The Device Cache File Path
+
+ Contents
+
+ A. Introduction and History
+ B. Device Cache File Format
+ 1. Integrity Checks
+ 2. The Setgid and Setuid-root States
+ C. Device Cache File Path Options
+ 1. Path Named by ``-D''
+ 2. Path Named in Environment Variable
+ 3. Default System-wide Path
+ a. Build Procedure
+ 4. Default Personal Path
+ 5. Modified Default Personal Path
+ D. Displaying the Default Path
+ Appendix A, Unix Dialects Without a Device Cache
+ Appendix B, Lsof Dialects and Their Permissions
+ 1. Setuid-root Lsof Dialects
+ 2. Setgid Lsof Dialects That Surrender Setgid
+ Permission
+
+
+A. Introduction and History
+===========================
+
+Lsof writes a file of information about the contents of the nodes
+in /dev (or /devices) to reduce its startup overhead on later calls.
+It does this for all Unix dialects, except those noted in Appendix A.
+
+This file, called the device cache file, enables lsof to avoid
+calling the kernel stat(2) function on every node in /dev (or
+/devices) from which it builds a table of correspondence between
+major/minor device numbers and device names.
+
+A full scan of /dev (or /devices) on some systems may involve
+calling the sometimes-slow stat(2) function 10,000 times or more.
+Furthermore, each stat(2) call consumes space in the kernel's name
+cache, forcing from it path name components that would be more
+useful when lsof tries to associate them with open files.
+
+While it's hard to question the usefulness of the device cache,
+it's also hard to decide where it should be written. When the
+feature was first added, the device cache file was written to /tmp,
+and its ownership was set to that of the real user ID (UID) under
+which the creating lsof process was run. However, to enable any
+process to update it when /dev (or /devices) changed, lsof set its
+modes to 0666, thus allowing anyone to read or write it.
+
+The writing of a world-readable and world-writable device cache
+file to any place has security weaknesses. A clever intruder who
+carefully preserves the integrity of the file might be able to
+remove devices that would prevent lsof from observing the intruder's
+files. A clever intruder might also be able to put a symbolic link
+in place and trick lsof into writing to the link's destination with
+its effective permissions, thus bypassing the real user's (possibly
+weaker) permissions.
+
+Later the location of the device cache file was changed. It was
+converted to a personal file, located in the home directory of each
+real UID that executed lsof, and owned by that UID. Thus it was
+no longer possible for one user to affect lsof's access to the
+device cache file, nor was it possible for a user to mount a symbolic
+link attack on a restricted file, but the result was that each lsof
+user had a private copy of the device cache file.
+
+The device cache file feature has undergone some further refinements
+in path name formation to reach its present state. This documentation
+describes the path name formation options open to the lsof builder
+and user after those refinements, and how lsof attempts to insure
+that none of the options presents a security risk.
+
+
+B. Device Cache File Format
+===========================
+
+The device cache file is a flat file of ASCII text. It has an
+initial statement of how many sections the file might contain --
+the possible sections are character devices, block devices, clone
+devices, pseudo devices, and checksum. The character devices and
+checksum sections are always present.
+
+Each section has a header that numbers the entries in the section.
+
+The last section is a checksum section that contains a 16 bit cyclic
+redundancy (CRC) checksum of everything in the file but the checksum
+section itself.
+
+Lsof always sets the permission modes of the device cache file to
+0600, and the owner to the real UID of the process that executes
+lsof; the group, the real group ID (GID) of the lsof process.
+
+Setting the permission modes to 0600 means that a system-wide device
+cache file won't be usable unless the procedure that builds it
+changes the modes after lsof has written it. A suitable procedure
+for building a system-wide device cache that shows how to adjust
+these inadequate permission modes is given in the Default System-wide
+Path section.
+
+
+B.1. Integrity Checks
+=====================
+
+When lsof opens the device cache file it makes these integrity
+checks:
+
+ 1. Lsof must gain permission from access(2) to be able to
+ open the file for reading. If lsof is writing the file,
+ it usually cedes permission control to the applicable
+ directory and file modes and ownerships. (Some additional
+ checks apply and they're described in the sections on path
+ options.)
+
+ By explicit design lsof never writes to the system-wide
+ device cache file, even when the real UID of its process
+ is root. The system-wide device cache file must be written
+ with a root-owned procedure via the ``-D[b|u<path>'' options
+ -- i.e., under the system administrator's control. (See
+ the Build Procedure sub-section of the Default System-wide
+ Path section.)
+
+ 2. The device cache file's modes must be 0600 (0644 if lsof
+ is reading a system-wide device cache file) and its size
+ must be non-zero.
+
+ 3. There must be a correctly formatted section count line
+ at the beginning of the file.
+
+ 4. Each section must have a header line with a count that
+ properly numbers the lines in the section. The first words
+ of legal section titles are "device", "block", "clone",
+ "pseudo", and "CRC".
+
+ 5. The lines of a section must have the proper format.
+
+ 6. All lines are included in a 16 bit CRC, and it is recorded
+ in a non-checksummed section line at the end of the file.
+
+ 7. The checksum computed when the file is read must match the
+ checksum recorded when the file was written.
+
+ 8. The checksum section line must be followed by end-of-
+ information.
+
+ 9. Lsof must be able to get matching results from stat(2)
+ on a randomly chosen entry of the device section.
+
+
+B.2. The Setgid and Setuid-root States
+======================================
+
+There are two fundamental ways in which lsof is granted access to
+restricted system resources. Both access methods are related to the
+effective permissions given the lsof binary or executable.
+
+The first and preferable way to grant lsof access to system resources
+through the permissions endowed on its executable is the giving of
+set group ID (setgid) permission. The group is the one that has
+permission to read the kernel memory and swap devices -- e.g., /dev/kmem,
+/dev/mem, /dev/swap, etc.
+
+This method of granting access is called setgid mode because it
+enables lsof to run with an effective group ID set to the one
+granted by the permissions of its executable file and by the group
+that owns the executable file. See the getegid(2) man page for a
+further discussion of effective group ID.
+
+Usually lsof only needs setgid permission to open access to the
+kernel memory files. After they're open, lsof drops its setgid
+permission.
+
+The second and least preferable way to grant lsof access to system
+resources through the permissions endowed on its executable is the
+giving of set user ID to root (setuid-root) permission. This is
+much too strong a permission, but necessary: to use the -X option
+fully for the version of lsof for AIX 5 and above; to use the
+version of lsof for HP-UX 11.11 and above; and to use the version
+of lsof for Linux 2.1.72 and above. These lsof implementations
+require setuid-root permission to be able to access restricted
+resources -- e.g., the individual files of the /proc file system.
+(But note that the setuid-root Linux lsof doesn't need and has no
+device cache support.)
+
+Lsof never drops setuid-root permission, because it needs that
+power throughout its execution. However, when the lsof process is
+setuid-root, lsof disallows these device cache file path options:
+
+ 1. It ignores the ``-D[b|r|u]<path>'' options. It accepts
+ only the ``-Di'' and ``-Dr'' options.
+
+ 2. It refuses to recognize a path supplied via an environment
+ variable.
+
+ 3. It refuses to accept an additional path component from an
+ environment variable to be inserted in the middle of a
+ personal device cache file path.
+
+Each restriction is imposed because setuid-root power might allow
+a malicious user to form a device cache file path that would give
+read access to a normally inaccessible place (That's bad enough.),
+or write access to a critical system file (That's the worst case.)
+
+There is one further state that lsof can enter that is slightly
+different from the setuid-root and setgid states. That state occurs
+when lsof is being run from a root shell -- i.e., the lsof real
+user ID is root. To avoid accidental complications, when lsof is
+in this state, it ignores all environment variable options.
+
+In the rest of this document you will find more detailed discussion
+of the special restrictions caused by the type of permission that
+has been given the lsof executable.
+
+
+C. Device Cache File Path Options
+=================================
+
+Lsof offers five options for constructing the path to the device
+cache file. Each has special conditions and safeguards that
+surround its use. The options are:
+
+ 1. A device cache file that is named in the <path> component
+ of the parameters of lsof's ``-D'' option.
+
+ =========================================================
+ * This is a default option of the lsof distribution. *
+ * *
+ * Paths specified with this option are read-only unless *
+ * the real UID of the lsof process is root (0), or the *
+ * lsof process is able to surrender setgid permission *
+ * (See Appendix B) and it is not setuid-root. *
+ =========================================================
+
+ 2. A device cache file whose name is specified by an environment
+ variable.
+
+ =========================================================
+ * This is a default option of the lsof distribution. *
+ * *
+ * This option is enabled when the lsof dialect is able *
+ * to surrender setgid permission (See Appendix B.), and *
+ * the lsof process is not setuid-root. *
+ * *
+ * The environment variable path is read-only if the *
+ * lsof process does not surrender setgid permission *
+ * (See Appendix B.) *
+ =========================================================
+
+ 3. A system-wide default device cache file, located at a path
+ determined by the builder of lsof. The lsof builder is also
+ responsible for building the device cache file, using a
+ different lsof path formation option at a suitable time --
+ e.g., when the system is booted.
+
+ =========================================================
+ * This is option is disabled by default in the lsof *
+ * distribution. *
+ * *
+ * The path specified with this option is read-only. *
+ =========================================================
+
+ 4. A default personal device cache file, located in the UID's
+ home directory.
+
+ =========================================================
+ * This is a default option of the lsof distribution. *
+ =========================================================
+
+ 5. A personal device cache file whose name is modified by an
+ environment variable.
+
+ =========================================================
+ * This is a default option of the lsof distribution. *
+ * *
+ * The modified personal path is read-only if the lsof *
+ * process does not surrender setgid permission. *
+ * *
+ * This option is disabled when the lsof process is *
+ * setuid-root or its real UID is root (0). *
+ =========================================================
+
+When there are multiple choices for the device cache file path,
+lsof chooses from the above list in the order the list is given,
+subject to restrictions based on the effective group and user IDs
+that are in effect.
+
+Each possible path name is discussed in a later section that
+describes the restrictions that apply to it and the method for
+building lsof to use it.
+
+In one special case lsof will use two paths in order. When a
+system-wide device cache file is enabled, and lsof finds that it
+doesn't exist, lsof will attempt to use a personal device cache
+file.
+
+
+C.1. Path Named by ``-D''
+=========================
+
+The ``-D[b|r|u]<path>'' option can name a path for the device cache
+file where it is unconditionally built (`b'); read, but never
+rebuilt (`r'); and read and rebuilt, if necessary (`u').
+
+If the lsof process is setuid-root, no path may be specified with
+the ``-D'' option -- i.e., only the `i' function is accepted. The
+`r' option may be used if it doesn't have a path argument.
+
+If the lsof process is not setuid-root, nor is the real UID of the
+lsof process root, a path may accompany the `b', `r', and `u'
+functions if the lsof process surrenders setgid permission. (See
+Appendix B.) If the process doesn't surrender setgid permission,
+then a path may accompany only `r'.
+
+Lsof's permission to access a device cache file at a path specified
+with ``-D[b|r|u]<path>'' depends completely on the permission modes
+and ownerships of the file and its directory components.
+
+When the real UID of the lsof process is root (0), paths may be
+specified with ``-D[b|r|u]''.
+
+====================================================================
+* *
+* The ``-D[b|r|u]<path>'' option is enabled by default in the lsof *
+* distribution by the following definition in the dialect's *
+* machine.h header file: *
+* *
+* #define HASDCACHE 1 *
+* *
+* To disable all device cache file options, including all ``-D'' *
+* forms, change the above line in the dialect's machine.h file to: *
+* *
+* /* #define HASDCACHE 1 */ *
+* *
+* or remove it. *
+* *
+* The ``-D[b|r|u]<path>'' options are disabled when the lsof *
+* process is setuid-root. If the lsof process isn't setuid-root, *
+* nor is its real UID root (0), and if the lsof process surrenders *
+* setgid permission, ``-D[b|r|u]'' may be accompanied by a path. *
+* *
+* A path may accompany ``-D[b|u]'' when the real UID of the lsof *
+* process is root. *
+* *
+* ``-Dr'' without a path name argument is always acceptable. *
+* *
+====================================================================
+
+
+C.2. Path Named in Environment Variable
+=======================================
+
+A device cache file path may be declared in an environment variable.
+This option is defined in the dialect's machine.h header file with
+the HASENVDC definition. The value of the HASENVDC definition is
+the environment variable's name.
+
+Lsof will use the value of the environment variable named by HASENVDC
+for the device cache file path unless either of the following
+conditions apply:
+
+ 1. The lsof process is in the setuid-root state.
+or
+ 2. The effective and real UIDs of the lsof process are root
+ (0).
+
+Lsof uses the value of the HASENVDC environment variable as the
+device cache file path after it senses there is no path declared by
+a ``-D'' option.
+
+A path from an environment variable is read-only unless the lsof
+process surrenders setgid permission. (See Appendix B.)
+
+====================================================================
+* *
+* The path name environment variable option is enabled by default, *
+* and the environment variable is named LSOFDEVCACHE in the lsof *
+* distribution by the following definition in the dialect's *
+* machine.h header file: *
+* *
+* #define HASENVDC "LSOFDEVCACHE" *
+* *
+* To disable the path name environment variable option, change *
+* the above line in the dialect's machine.h header file to: *
+* *
+* /* #define HASENVDC "LSOFDEVCACHE" */ *
+* *
+* or remove it. To change the name of the environment variable, *
+* change the quoted value of the HASENVDC definition -- e.g., this *
+* form changes the environment variable name to "FOOBAR": *
+* *
+* #define HASENVDC "FOOBAR" *
+* *
+* You can disable the path name environment option by disabling *
+* all device cache file processing when you remove or by disabling *
+* the HASDCACHE definition in the dialect's machine.h header file. *
+* *
+* The path name environment option is disabled when the lsof *
+* process is setuid-root or when the real UID of the lsof process *
+* is root (0). *
+* *
+* The path named in an environment variable is read-only unless *
+* the lsof process surrenders setgid permission. (See Appendix *
+* B.) *
+* *
+====================================================================
+
+
+C.3. Default System-wide Path
+=============================
+
+When a default system-wide device cache file path is defined (It's
+not enabled by default in the lsof distribution.), lsof will use
+it after it discovers no path has been specified by a ``-D'' option
+and no path has been specified in the environment variable named
+in the string #define HASENVDC of the dialect's machine.h header
+file.
+
+Lsof must be able to open the system-wide device cache file --
+i.e., it must have read access to the file and search access to
+the directories that lead it. As part of its integrity checks,
+lsof requires that the system-wide device cache file's permission
+modes be 0644.
+
+When lsof discovers that the named system-wide device cache file
+doesn't exist, it will attempt to open a personal device cache file
+should that path formation option be enabled. This is the *only*
+case where lsof will attempt to use two device cache file paths.
+
+The system-wide device cache file is read-only; lsof will never
+attempt to write to it. However, when the real UID of the lsof
+process is root, that process may name the system-wide device
+cache file with ``-D[b|u]<path>''.
+
+====================================================================
+* *
+* The system-wide file path option is disabled by default in the *
+* lsof distribution. This place-marking definition in a dialect's *
+* machine.h header file may be altered to enable a system-wide *
+* device cache file path: *
+* *
+* /* #define HASSYSDC "/your/choice/of/path" */ *
+* *
+* To enable the system-wide name option, declaring that its path *
+* is ``/foo/bar/lsof.dc'', change the above line in the dialect's *
+* machine.h header file to: *
+* *
+* #define HASSYSDC "/foo/bar/lsof.dc" *
+* *
+* or change the quoted string of the definition to the path of *
+* your choice. *
+* *
+* You can disable the path name environment option by disabling *
+* all device cache file processing when you remove or disable the *
+* HASDCACHE definition in the dialect's machine.h header file. *
+* *
+* The system-wide device cache file is read-only. *
+* *
+====================================================================
+
+
+C.3.a. Build Procedure
+======================
+
+The system administrator must build the system-wide device cache
+file at an appropriate time -- e.g., each time the system is booted,
+and each time a node is added, deleted or modified in /dev (or
+/devices). The procedure that builds the system-wide device cache
+file must use lsof's ``-D[b|u]<path>'' options to build the file,
+and must change the file's permission modes to 0644 after it has
+been built.
+
+Here's a simple shell script procedure to build a system-wide device
+cache file. It assumes:
+
+ 1. The Unix dialect's kernel supports the interpreter script
+ execution option -- i.e., a script whose first line has
+ the form ``#!<path_to_interpreter>''.
+
+ 2. The chmod, echo, rm, sh, and test programs are located
+ in ``/bin''.
+
+ 3. The string value of the HASSYSDC definition in the dialect's
+ machine.h header file is the path ``/your/choice/of/path''.
+
+ 4. The lsof executable is located in ``/usr/local/etc''.
+
+ #!/bin/sh
+ #
+ # Simple script to build a system-wide device cache file
+ # for lsof.
+
+ HASSYSDC=/your/choice/of/path
+ /bin/rm -f $HASSYSDC
+ /usr/local/etc/lsof -Du$HASSYSDC > /dev/null 2>&1
+ if /bin/test $? -ne 0
+ then
+ /bin/echo "WARNING: failed to create $HASSYSDC"
+ exit 1
+ fi
+ /bin/chmod 0644 $HASSYSDC
+ exit 0
+
+The invocation of lsof uses the ``-Du$HASSYSDC'' option to read
+the device cache file and recreate it if necessary. The invocation
+can be made more efficient if a known process PID -- e.g., ``-p1''
+-- can be specified. However, if that PID is not always active
+when lsof is called, lsof might set its exit code non-zero, causing
+the subsequent test to believe that the lsof call failed. When in
+doubt, omit the PID specification and accept the extra lsof processing
+time for reporting and discarding all open file information.
+
+
+C.4. Default Personal Path
+==========================
+
+The default personal path option is defined by default in the lsof
+distribution. The path is formed of the home directory of the real
+UID of the lsof process, followed optionally by the contents of
+the HASPERSDCPATH environment variable, followed by ``.lsof_'',
+followed by the first component (characters up to the first period)
+of the name returned by gethostname(2).
+
+If gethostname(2) returns nothing, then nothing will follow the
+``.lsof_'' string. If the first character of what gethostname(2)
+returns is a `.', then all the gethostname(2) value will follow
+the ``/lsof_'' string. (See the ``%l'' conversion for a way to
+make lsof include the entire host name in the path.)
+
+====================================================================
+* *
+* The personal path option is enabled by default in the lsof *
+* distribution. The HASPERSDC #define in a dialect's machine.h *
+* header is a format specification that tells lsof how to form the *
+* personal device cache file path. The conversions in the format *
+* specification begin with `%' , ala the printf(3) function of the *
+* standard I/O library. These conversions are supported: *
+* *
+* ``%%'' causes a single `%' to appear in the path. *
+* *
+* ``%0'' is a separator that marks the beginning of a path *
+* for a setuid-root lsof process or one whose real *
+* UID is 0. When lsof reaches this conversion and *
+* the process is setuid-root or has a real UID of *
+* root, it erases any previously formed path and *
+* restarts with the next HASPERSDC format character. *
+* If lsof reaches this conversion and the process is *
+* not setuid-root and its real UID is not root, path *
+* formation is ended. *
+* *
+* ``%h'' causes the home directory of the real UID of the *
+* lsof process to appear in the path. *
+* *
+* ``%l'' causes the full name returned by gethostname(2) to *
+* appear in the path. *
+* *
+* ``%L'' causes the first component of the name returned by *
+* gethostname(2) to appear in the path. The first *
+* component is defined to be what appears to the *
+* left of the first `.'. If nothing appears to the *
+* left then everything will appear in the path. *
+* *
+* ``%p'' causes the value of (HASPERSDCPATH) from the *
+* process environment to appear in the path. If the *
+* (HASPERSDCPATH) value doesn't end in a '/', one *
+* will be added. *
+* *
+* ``%u'' causes the login name associated with the real UID *
+* of the lsof process to appear in the path. *
+* *
+* ``%U'' causes the real UID of the lsof process, converted *
+* to a decimal string, to appear in the path. *
+* *
+* All other characters are copied from the format to the *
+* path. CAUTION: THINK VERY CAREFULLY ABOUT THE EFFECT OF *
+* USING CHARACTERS THAT FORM AN ABSOLUTE COMPONENT LIKE *
+* ``/tmp'' IN THE FORMAT. Consider what power your dialect *
+* might have (e.g., if it is setuid-root) when lsof must *
+* create a device cache file at the path. Consider using a *
+* ``%0'' conversion to declare an alternate path for lsof *
+* processes that are setuid-root or whose real uid is root. *
+* See the "How do I put the personal device cache file in *
+* /tmp?" question and answer in 00FAQ for an explanation of *
+* this example: *
+* *
+* #define HASPERSDC "/tmp/.lsof_%u_%l_pers%0%h/.lsof_%L" *
+* *
+* This is the format specification that appears in the machine.h *
+* header files of the lsof distribution: *
+* *
+* #define HASPERSDC "%h/%p.lsof_%L" *
+* *
+* It causes the path to be formed from the home directory of the *
+* real UID of the lsof process (``%h''), followed by `/', followed *
+* by the contents of the environment variable named by *
+* HASPERSDCPATH and a trailing `/', as needed (``%p''), followed *
+* by the string ``.lsof_'', and terminated with the first *
+* component of the host's name (``%L''). *
+* *
+* To change the personal path option, change the HASPERSDC string *
+* and recompile lsof. To disable the personal path option, remove *
+* or disable HASPERSDC. The personal path option is disabled when *
+* HASDCACHE is not defined. *
+* *
+====================================================================
+
+
+C.5. Modified Default Personal Path
+===================================
+
+The modified default personal path form is a special case of the
+default personal path. In this form the value of the environment
+variable named by the HASPERSDCPATH #define is inserted in the
+personal path when the ``%p'' conversion appears in the HASPERSDC
+format specification.
+
+This allows, for example, the lsof user to move personal device
+cache files to another branch of the home directory, perhaps to a
+sub-directory where multiple device cache files may appear from
+different machines that use the same NFS- mounted home directory.
+
+The HASPERSDCPATH definition of the dialect's machine.h header file
+names the environment variable. By default in the lsof distribution
+it is LSOFPERSDCPATH.
+
+The modified personal path component is ignored when lsof process
+is setuid-root is root, lest it be maliciously or accidentally used in
+some convoluted form to access paths the real UID cannot. The
+modified personal path component is also ignored when the real UID
+of the lsof process is root (0), so that lsof will not accidentally
+use a personal environment value.
+
+If the lsof process surrenders setgid permission (See Appendix B.),
+lsof can read from and write to the modified personal path. If,
+however, the lsof process doesn't surrender setgid permission, the
+modified personal path is read-only.
+
+If your dialect runs setuid-root or doesn't surrender its setgid
+permission, and you want to use the LSOFPERSDCPATH environment
+variable to address a collection of device cache files in a
+subdirectory, you will have to gather the collection in the
+subdirectory yourself with shell copy or move commands.
+
+====================================================================
+* *
+* The modified personal path option is enabled by default in the *
+* lsof distribution with these definitions in the dialect's *
+* machine.h header file: *
+* *
+* #define HASPERSDCPATH "LSOFPERSDCPATH" *
+* and *
+* #define HASPERSDC "%h/%p.lsof_%L" *
+* *
+* The value of the definition is the name of the environment *
+* variable that contains the modified personal path name *
+* component that is inserted in the personal path when ``%p'' *
+* appears in HASPERSDC. See the Default Personal Path section *
+* for a complete description of the ``%p'' conversion. *
+* *
+* To disable the modified personal path name component, disable *
+* the HASPERSDCPATH definition in the dialect's machine.h header *
+* file -- e.g., change it to: *
+* *
+* /* #define HASPERSDCPATH "LSOFPERSDCPATH" */ *
+* *
+* or remove the definition altogether. If you do this, don't *
+* forget to remove any ``%p'' conversion from HASPERSDC. *
+* *
+* The modified personal path option is disabled when HASDCACHE is *
+* not defined. *
+* *
+* The modified personal path environment variable value is ignored *
+* when the lsof process is setuid-root or when the real UID of *
+* the lsof process is root (0). *
+* *
+* The modified personal path is read-only when the lsof process *
+* doesn't surrender its setgid permission. *
+* *
+====================================================================
+
+
+D. Displaying the Default Path
+==============================
+
+Whatever device cache file path formation options you decide to
+use, remember that the lsof help output, displayed in response to
+its ``-h'' or ``-?'' help options, will display the read-mode
+default (the highest numbered) path that lsof has been enabled to
+form from which it will read.
+
+Since some paths are read-only, the path displayed in help option
+output may not be the one to which lsof will write, should that
+become necessary. To see the read-only and write device cache file
+paths, environment variable names, and the personal device cache
+file format specification (HASPERSDC), use the -D? option.
+
+
+Appendix A, Unix Dialects Without a Device Cache
+================================================
+
+Linux lsof implementations that obtain their information from files
+in the /proc file system do not have device cache support. Generally
+lsof for Linux versions 2.1.72 and greater are /proc based.
+
+
+Appendix B, Lsof Dialects and Their Permissions
+===============================================
+
+These are the permissions recommended in the lsof distribution.
+
+
+Appendix B.1 Setuid-root Lsof Dialects
+======================================
+
+These dialect versions of lsof need root permission. For general
+use they may have to be installed setuid-root.
+
+ Apple Darwin 9 (Mac OS X 10.5)
+ HP-UX 11.11 and 11.23
+ Linux (no device cache support needed)
+
+
+Appendix B.2 Setgid Lsof Dialects That Surrender Setgid Permission
+==================================================================
+
+Lsof versions for these dialects have WILLDROPGID defined in their
+machine.h header files.
+
+ AIX 5.[12] and 5.3-ML1
+ FreeBSD 4.x, 4.1x, 5.x and [67].x for x86-based systems
+ FreeBSD 5.x and [67].x for Alpha, AMD64 and Sparc64-based
+ systems
+ HP-UX 11.00
+ NetBSD 1.[456], 2.x and 3.x for Alpha, x86, and SPARC-based
+ systems
+ NEXTSTEP 3.[13]
+ OpenBSD 2.[89] and 3.[0-9] for x86-based systems
+ OPENSTEP 4.x
+ SCO OpenServer Release 5.0.4 for x86-based systems
+ SCO|Caldera UnixWare 7.1.4 for x86-based systems
+ Solaris 2.6, 8, 9 and 10
+ Tru64 UNIX 5.1
+
+
+Vic Abell <abe@purdue.edu>
+March 25, 2009
--- /dev/null
+ AIX 5.3
+ Apple Darwin 9 (Mac OS X 10.5)
+ FreeBSD 4.9 for x86-based systems
+ FreeBSD 7.[012] and 8.0 for AMD64-based systems
+ Linux 2.1.72 and above for x86-based systems
+ Solaris 9 and 10
--- /dev/null
+
+ Notes for the distribution of lsof version 4
+
+********************************************************************
+| The latest release of lsof is always available via anonymous ftp |
+| from lsof.itap.purdue.edu. Look in pub/tools/unix/lsof. |
+********************************************************************
+
+ Contents
+
+ Dialects Supported
+ How Lsof Works
+ Lsof Output
+ Getting Started Quickly
+ Limiting, Filtering, and Selecting Lsof Output
+ Parsing Lsof Output with Another Program
+ Repeat Mode
+ Distribution Restrictions
+ Cautions
+ Distribution Contents
+ Warranty
+ Bug Reports
+ The lsof-l Mailing List
+ Version 3 Release Notes
+ 3.0, May 24, 1994
+ ...
+ 3.88, February 17, 1997
+ What's New in Version 4
+ Version 4 Release Notes
+ 4.0, February 24, 1997
+ 4.01, March 3, 1997
+ 4.02, March 21, 1997
+ 4.03, April 7, 1997
+ 4.04, April 17, 1997
+ 4.04 supplement, April 18, 1997
+ 4.05, April 24, 1997
+ 4.06, April 30, 1997
+ 4.07, May 12, 1997
+ 4.08, May 23, 1997
+ 4.09, June 1, 1997
+ 4.10, June 8, 1997
+ 4.11, June 12, 1997
+ 4.12, June 24, 1997
+ 4.13, July 9, 1997
+ 4.14, July 22, 1997
+ 4.15, August 15, 1997
+ 4.16, September 25, 1997
+ 4.17, October 14, 1997
+ 4.18, October 25, 1997
+ 4.19, October 30, 1997
+ 4.20, November 11, 1997
+ 4.21, December 1, 1997
+ 4.22, December 15, 1997
+ 4.23, January 16, 1998
+ 4.24, January 28, 1998
+ 4.25, February 7, 1998
+ 4.26, February 17, 1998
+ 4.27, March 6, 1998
+ 4.28, March 10, 1998
+ 4.29, March 26, 1998
+ 4.30, April 9, 1998
+ 4.31, April 21, 1998
+ 4.32, May 13, 1998
+ 4.33, May 22, 1998
+ 4.34, June 26, 1998
+ 4.35, July 17, 1998
+ 4.36, August 4, 1998
+ 4.37, September 15, 1998
+ 4.38, November 25, 1998
+ 4.39, December 29, 1998
+ 4.40, January 25, 1999
+ 4.41, February 27, 1999
+ 4.42, March 30, 1999
+ 4.43, May 11, 1999
+ 4.44, June 24, 1999
+ 4.45, July 30, 1999
+ 4.46, October 23, 1999
+ 4.47, November 29, 1999
+ 4.48, January 14, 2000
+ 4.49, April 3, 2000
+ 4.50, June 29, 2000
+ 4.51, August 21, 2000
+ 4.52, November 8, 2000
+ 4.53, December 6, 2000
+ 4.54, January 19, 2001
+ 4.55, February 15, 2001
+ 4.56, May 3, 2001
+ 4.57, July 19, 2001
+ 4.58, September 13, 2001
+ 4.59, October 20, 2001
+ 4.60, November 9, 2001
+ 4.61, January 22, 2002
+ 4.62, March 7, 2002
+ 4.63, April 23, 2002
+ 4.64, June 26, 2002
+ 4.65, October 10, 2002
+ 4.66, December 22, 2002
+ 4.67, March 27, 2003
+ 4.68, June 18, 2003
+ 4.69, October 16, 2003
+ 4.70, January 16, 2004
+ 4.71, March 11, 2004
+ 4.72, July 13, 2004
+ 4.73, October 21, 2004
+ 4.74, January 17, 2005
+ 4.75, May 16, 2005
+ 4.76, August 30, 2005
+ 4.77, April 10, 2006
+ 4.78, April 24, 2007
+ 4.79, April 15, 2008
+ 4.80, May 12, 2008
+ 4.81, October 21, 2008
+ 4.82, March 25, 2009
+
+
+Dialects Supported
+==================
+
+Lsof (for LiSt Open Files) lists files opened by processes on
+selected Unix systems. Version 4 is a source reorganization of
+version 3, itself a major revision of version 2. Version 4 has
+been tested on:
+
+ AIX 5.3
+ Apple Darwin 9 (Mac OS X 10.5)
+ FreeBSD 4.9 for x86-based systems
+ FreeBSD 7.[012] and 8.0 for AMD64-based systems
+ Linux 2.1.72 and above for x86-based systems
+ Solaris 9 and 10
+
+(The pub/tools/unix/lsof/contrib directory on lsof.itap.purdue.edu
+contains information on other ports.)
+
+If your favorite Unix dialect is not in the list, or if your version
+of it is more recent than the ones listed, please contact me at
+<abe@purdue.edu>.
+
+Version 3 of lsof was tested on:
+
+ AIX 3.2.5, 4.1[.[1234]], and 4.2
+ BSDI BSD/OS 2.0, 2.0.1, and 2.1 for x86-based systems
+ DC/OSx 1.1 for Pyramid systems
+ Digital UNIX (DEC OSF/1) 2.0, 3.0, 3.2, and 4.0
+ EP/IX 2.1.1 for the CDC 4680
+ FreeBSD 1.1.5.1, 2.0, 2.0.5, 2.1, 2.1.5 for x86-based
+ systems
+ HP-UX 8.x, 9.x, 10.01, 10.10, and 10.20
+ IRIX 5.2, 5.3, 6.0, 6.0.1, and 6.[124]
+ Linux through 2.0.27 for x86-based systems
+ NetBSD 1.0, 1.1, and 1.2 for x86 and SPARC-based
+ systems
+ NEXTSTEP 2.1 and 3.[0123]
+ OpenBSD 1.2 and 2.0 for x86-based systems
+ Reliant UNIX 5.43 for Pyramid systems
+ RISC/os 4.52 for MIPS R2000-based systems
+ SCO OpenServer Release 1.1, 3.0, and 5.0.x for x86-based
+ systems
+ SCO UnixWare 2.1 and 2.1.1 for x86-based systems
+ Sequent PTX 2.1.[1569], 4.0.[23], 4.1.[024], 4.2[.1],
+ and 4.3
+ Solaris 2.[12345], 2.5.1, and 2.6-Beta
+ SunOS 4.1.x
+ Ultrix 4.2, 4.3, 4.4, and 4.5
+
+Version 3 and its predecessor, version 2, may be found at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/OLD
+
+
+How Lsof Works
+==============
+
+Using available kernel data access methods -- getproc(), getuser(),
+kvm_*(), nlist(), pstat(), read(), readx(), /proc -- lsof reads
+process table entries, task table entries, user areas and file
+pointers to reach the underlying structures that describe files
+opened by processes.
+
+Lsof interprets most file node structures -- advfsnodes, autonodes,
+cnodes, cdrnodes, devnodes, fifonodes, gnodes, hsnodes, inodes,
+mfsnodes, pcnodes, procnodes, rnodes, snodes, specnodes, s5inodes,
+tmpnodes. It understands NFS connections. It recognizes FIFOs,
+multiplexed files, Unix and Internet sockets. It knows about
+streams. It understands /proc file systems for some dialects. On
+many dialects it recognizes execution text and library references.
+It knows about AFS on some Unix dialects.
+
+
+Lsof Output
+===========
+
+The lsof output describes:
+
+ * the identification number of the process (PID) that has opened
+ the file;
+
+ * the process group identification number (PGID) of the process
+ (optional);
+
+ * the process identification number of the parent process (PPID)
+ (optional);
+
+ * the command the process is executing;
+
+ * the owner of the process;
+
+ * for all files in use by the process, including the executing
+ text file and the shared libraries it is using:
+
+ * the file descriptor number of the file, if applicable;
+
+ * the file's access mode;
+
+ * the file's lock status;
+
+ * the file's device numbers;
+
+ * the file's inode number;
+
+ * the file's size or offset;
+
+ * the name of the file system containing the file;
+
+ * any available components of the file's path name;
+
+ * the names of the file's stream components;
+
+ * the file's local and remote network addresses;
+
+ * the TLI network (typically UDP) state of the file;
+
+ * the TCP state, read queue length, and write queue length
+ of the file;
+
+ * the file's TCP window read and write lengths (Solaris
+ only);
+
+ * other file or dialect-specific values.
+
+
+Getting Started Quickly
+=======================
+
+If you want to get started using lsof quickly, or see some examples
+of how lsof can be used, consult the 00QUICKSTART file of the lsof
+distribution.
+
+The 00QUICKSTART file won't help you build or install lsof, but it
+will cut through the density of the lsof man page, giving you more
+readily an idea of what you can do with lsof.
+
+For information on building and installing lsof, consult the 00README
+file of the lsof distribution.
+
+
+Limiting, Filtering, and Selecting Lsof Output
+==============================================
+
+Lsof accepts options to limit, filter, and select its output.
+These are the possible criteria:
+
+ * Process ID (PID) number -- to list the open files for a given
+ process;
+
+ * Process Group ID (PGID) -- to list the open files for all
+ the processes of a given process group;
+
+ * User ID number or login name -- to list the open files for
+ all the processes of a given user;
+
+ * Internet address -- to list the open files using a given
+ Internet address (host name), protocol, or port (number or
+ name); or to list all open Internet files;
+
+ * command name;
+
+ * file descriptor name or number;
+
+ * list all open NFS files;
+
+ * list all open Unix domain socket files;
+
+ * list all uses of a specific file;
+
+ * list all open files on a file system.
+
+Selection options are normally ORed -- i.e., an open file meeting
+any of the criteria is listed. The selection options may be ANDed
+so that an open file will be listed only if it meets all the
+criteria.
+
+In the absence of any selection criteria, lsof lists files open to
+all processes.
+
+
+Parsing Lsof Output with Another Program
+========================================
+
+The lsof -F option directs it to produce "field" output that can
+easily be parsed by another program. The lsof distribution contains
+sample awk, perl 4, and perl 5 scripts in its scripts subdirectory
+that show how to post-process field output.
+
+
+Repeat Mode
+===========
+
+Lsof can be directed to produce output, delay for a specified time,
+then repeat the output, cycling until stopped by an interrupt or
+quit signal. This mode is useful for monitoring the status of some
+file operation -- e.g., an ftp transfer or a tape backup operation.
+
+Repeat mode is more efficient when combined with lsof's selection
+options, since they limit lsof overhead.
+
+It's possible to use lsof's field output options to supply repeat
+mode output to another process for its manipulation. The scripts
+subdirectory of the lsof distribution has sample Perl scripts
+showing how to consume lsof repeat mode output from a pipe.
+
+
+Distribution Restrictions
+=========================
+
+Lsof may be used and distributed freely, subject to these limitations:
+
+1. Neither the author nor Purdue University is responsible for
+ any consequences of the use of this software.
+
+2. The origin of this software must not be misrepresented, either
+ by explicit claim or by omission. Credit to the author and
+ Purdue University must appear in documentation and sources.
+
+3. Altered versions must be plainly marked as such, and must not
+ be misrepresented as being the original software.
+
+4. This notice may not be removed from or altered in the lsof source
+ files.
+
+
+Cautions
+========
+
+Lsof is a tool that is closely tied to the Unix operating system
+version. It uses header files that describe kernel structures and
+reads kernel structures that typically change from OS version to
+OS version.
+
+DON'T TRY TO USE AN LSOF BINARY, COMPILED FOR ONE UNIX OS VERSION,
+ON ANOTHER.
+
+On some Unix dialects, notably SunOS and Solaris, lsof versions
+may be even more restricted by architecture type. An lsof binary,
+compiled for SunOS 4.1.3 on a sun4c machine, for example, won't
+work on a sun4m machine.
+
+AN LSOF BINARY, COMPILED FOR ONE SOLARIS 1.X ARCHITECTURE, ISN'T
+GUARANTEED TO WORK ON A DIFFERENT SOLARIS 1.X ARCHITECTURE.
+
+
+Distribution Contents
+=====================
+
+The lsof distribution is checked for completeness when it is
+constructed and by the Inventory script when you run the Configure
+script. (See The Inventory Script section of the 00README file of
+this distribution.)
+
+Lsof is organized in these parts:
+
+ * The main lsof directory, containing common sources,
+ configuration and setup scripts and three subdirectories:
+ dialects/, lib/, and scripts/.
+
+ Lsof is compiled in the main lsof directory after configuration.
+ The selected dialect sources are copied or linked from the
+ specified subdirectory. (Symbolic linking is the standard
+ method.)
+
+ Common lsof definitions may be found in lsof.h; common
+ function prototypes, proto.h; and common storage, store.c.
+
+ * The dialects/ subdirectory contains subdirectories with
+ sources specific to UNIX dialect implementations -- e.g.,
+ the dialects/sun/ subdirectory contains sources for the
+ SunOS (Solaris 1.x) and Solaris (2.x) implementations of
+ lsof. The dialects subdirectories also contain Makefiles
+ and scripts for assisting dialect source configuration.
+
+ Dialect configuration definitions may be found in dlsof.h;
+ other dialect definitions, dlsof.h; dialect prototypes,
+ dproto.h; and dialect storage, dstore.c.
+
+ * The lib/ subdirectory contains sources for common lsof
+ functions. Not all dialects use the functions -- some have
+ their own versions of them. The lib/ functions are enabled
+ and customized with #define's in the dialect machine.h header
+ files.
+
+ * The scripts/ subdirectory contains sample scripts for
+ processing lsof field (-F) output. The scripts are written
+ in AWK, Perl 4, and Perl 5.
+
+The 00PORTING file of the lsof distribution has more information
+on lsof components, configuration, and construction.
+
+
+Warranty
+========
+
+Lsof is provided as-is without any warranty of any kind, either
+expressed or implied, including, but not limited to, the implied
+warranties of merchantability and fitness for a particular purpose.
+The entire risk as to the quality and performance of lsof is with
+you. Should lsof prove defective, you assume the cost of all
+necessary servicing, repair, or correction.
+
+
+Bug Reports
+===========
+
+Now that the obligatory disclaimer is out of the way, let me hasten
+to add that I accept lsof bug reports and try hard to respond to
+them. I will also consider and discuss requests for new features,
+ports to new dialects, or ports to new OS versions.
+
+PLEASE DON'T SEND A BUG REPORT ABOUT LSOF TO THE UNIX DIALECT
+VENDOR.
+
+At worst such a bug report will confuse the vendor; at best, the
+vendor will forward the bug report to me.
+
+Please send all bug reports, requests, etc. to me via email at
+<abe@purdue.edu>.
+
+
+The lsof-l Mailing List
+=======================
+
+Information about lsof, including notices about the availability
+of new revisions, may be found in mailings of the lsof-l listserv.
+For more information about it, including instructions on how to
+subscribe, read the 00LSOF-L file of the lsof distribution.
+
+
+Version 3 Release Notes
+=======================
+
+See 00DIST in the last lsof 3 revision 3.88, for its complete
+set of release notes. Lsof revision 3.88 may be found at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/OLD
+
+3.0 May 24, 1994
+ This is the first official release of lsof 3.
+
+...
+
+3.88 February 17, 1997
+
+ +======================================+
+ | This is the last version 3 revision. |
+ +======================================+
+
+ Added documentation files -- 00.README.FIRST[_<version>]
+ and 00RELEASE.SUMMARY_<version> -- to the distribution.
+
+
+What's new in Version 4
+=======================
+
+The main goal of version 4 was to eliminate the confusing common/
+fragment source file technique. Changing the version number also
+provided an opportunity to restart the numbering, which at 3.88
+had risen to a large value.
+
+The sources that appeared in the dialects/common subdirectory of
+version 3 in fragment files have been incorporated into the version
+4 liblsof.a library as *.c files. This results in significant
+changes to many source files, scripts, and Makefiles of all dialect
+versions. It allows elimination of some source files -- ddev.c,
+dfile.c, dmnt.c -- for dialects now obtaining functions from
+liblsof.a that formerly came from making dialect source files by
+combining fragment files.
+
+The version 4 liblsof.a sources are stored in the lib/ subdirectory
+of the main lsof directory. The liblsof.a functions are activated
+and conditioned in their source files by values #define'd in the
+dialect dlsof.h and machine.h header files.
+
+Dialects that provide a private version of a library function refrain
+from #define'ing the symbol that would activate the library function
+code.
+
+
+Version 4 Release Notes
+=======================
+
+4.0 February 24, 1997
+
+ +====================================+
+ | This is the first lsof 4 revision. |
+ +====================================+
+
+ Reorganized sources: eliminated code fragment files
+ and created a library in their place. Modified or
+ deleted many dialect source and header files.
+ Changed documentation accordingly.
+
+ Added a warning to sgi/Makefile and 00FAQ that advises
+ against using the IRIX C compiler -n32 option when
+ compiling lsof. Thanks go to Peter Ilieve
+ <peter@memex.co.uk> for bringing this to my attention.
+
+ Dropped IRIX 5.2 in mid-stream, because my 5.2 test
+ system was upgraded to 5.3.
+
+4.01 March 3, 1997
+ Added TFS support for Pyramid dialects.
+
+ Added test to Configure and to the IRIX dnode.c
+ for the different cnode struct that appears in
+ <cachefs/cachefs_fs.h> on the 6.2 IMPACT distribution.
+ Heddy Boubaker <boubaker@amfou.cenatls.cena.dgac.fr>
+ alerted me to the cnode change and helped test this
+ lsof adjustment.
+
+ Shut down the lsof child process before doing a -r
+ sleep(). A comment from Dan Mercer <dam@mmm.com>
+ prompted this.
+
+4.02 March 21, 1997
+
+ Based on a report from Pasi Kaara <Pasi.Kaara@atk.tpo.fi>,
+ disabled HP-UX CCIT support in lsof for HP-UX
+ versions 10 and above. Pasi's report also led to
+ changes in the HP-UX machine.h to support use of
+ gcc to compile lsof for HP-UX 10.20 and warnings
+ against using `cc -Aa` or `gcc -ansi` to compile
+ lsof under HP-UX 10.x.
+
+ With help from Richard Allen <ra@hp.is> taught
+ HP-UX 10.x lsof to name file systems better by
+ using the virtual file system device number. Elias
+ Halldor Agustsson <elias@rhi.hi.is> provided a test
+ system.
+
+ Changed NEXTSTEP and UNIXWARE Makefiles to use
+ safer quoting when generating version.h. The change
+ was suggested by Bob Farmer <ucs_brf@unx1.shsu.edu>.
+
+ Added SHELL=/bin/sh string to all Makefiles.
+
+ Added support for Linux 2.1.28 on a test system,
+ kindly provided by Jonathan Sergent <sergent@purdue.edu>.
+ Configure tests the Linux 2.1.x's C library lseek()
+ function for proper handling of kernel offsets.
+ If lseek() appears suspect, Configure activates
+ the use of a private lseek() function. Changed
+ the private nlist() function to nlist_private()
+ and taught it to use the query_module() syscall in
+ place of the deprecated get_kernel_syms() one.
+ Added rudimentary AX.25 support for Pierfrancesco
+ Caci <ik5pvx@infogroup.it> who helped test it.
+ Updated the old get_kernel_syms() code to recognize
+ and skip module name entries.
+
+ Prompted by Marty Leisner <leisner@sdsp.mc.xerox.com>,
+ eased the requirement that service name lookup for
+ the -i option be accompanied by a protocol name. The
+ name is not needed if both TCP and UDP names yield the
+ same port number.
+
+ Added xusers.awk script from Dan Mercer <damercer@mmm.com>
+ to the distribution scripts/ subdirectory.
+
+ Changed Configure script to use LSOF_VERS for all
+ UNIX dialect version numbers and to pass LSOF_VERS
+ to the dialect Mksrc functions. Also added the
+ ability for a dialect stanza to declare a different
+ dialect Makefile source. Modified dialect Mksrc
+ files -- e.g., linux and sun -- accordingly.
+
+ Added support for BSD/OS 3.0 with help from Jim
+ Reid <jim@mpn.cp.philips.com>. Terry Kennedy
+ <TERRY@spcvxa.spc.edu> kindly provided a test
+ system. During the port corrected a bug that
+ prevented proper handling of revoked files.
+
+4.03 April 7, 1997
+ At the suggestion of Dan Mercer <damercer@mmm.com>,
+ made HP-UX building of lsof aware of differences
+ between the HP-UX bundled and unbundled C compilers.
+
+ Added the ability for the lsof builder to define the
+ default warning message issuance state. By default the
+ issuance of warning messages is disabled; defining
+ WARNINGSTATE in machine.h disables it. The Customize
+ script was updated to handle WARNINGSTATE. Dan Mercer
+ suggested this.
+
+ Eliminated compiler complaint about improperly cast
+ get_Nl_value() argument in ncache_load() in lib/rnch.c.
+
+ Corrected zeromem() argument error in SCO dproc.c.
+ Sped up parent directory cache lookup slightly.
+
+ Updated for PTX 4.4, including additional VxFS (EFS)
+ file system support.
+
+4.04 April 17, 1997
+ At the suggestion of Bela Lubkin <belal@sco.COM>
+ changed device cache handling to be more tolerant
+ of a device cache file whose [cm]times are older
+ than the ones on /dev or /devices. The change
+ required adding information to Solaris device cache
+ file clone lines, so the first time lsof 4.04 is
+ run under Solaris it will complain about a bad
+ cached clone device in a previous device cache
+ file, then regenerate it.
+
+ Added boot file path detection for SCO OSR 5 and
+ above, based on information supplied by Bela.
+
+ Fixed two bugs in DEC OSF/1 lsof -- an error in
+ reporting locks and a missing continue statement
+ in readdev() after a failure to open a directory.
+ Jan Ole Suhr <josuhr@informatik.tu-clausthal.de>
+ reported the second bug and supplied a fix.
+
+ Fixed XFS problems with IRIX 6.2 by abandoning the
+ idea that SGI will distribute XFS header files and
+ defining an lsof-private xfs_inode structure. John
+ Paul Morrison <John.Paul.Morrison@MultiActive.com>
+ helped develop and test the 5.3 definition. John
+ R. Vanderpool <fish@daacdev1.gsfc.nasa.gov> helped
+ develop and test the 6.2 definition.
+
+ Remove obsolete comments about common/*.frag files.
+
+ Updated Linux lsof for Linux version 2.1.35.
+
+4.04 April 18, 1997
+Supplement Regenerated the 4.04 distribution to correct a non-
+ device-cache #define misplacement in the Solaris and
+ SunOS dlsof.h. Alexandre Oliva <oliva@dcc.unicamp.br>
+ reported the problem.
+
+4.05 April 24, 1997
+ Corrected an error in 00DCACHE.
+
+ Made sure SCO /etc/ps/booted.systems is closed.
+
+ Based on an observation by Bela Lubkin <belal@sco.COM>
+ that the lsof child had needless file descriptors
+ open, closed all but the open pipes between the
+ lsof parent and child.
+
+ Decommissioned CDC EP/IX support; I no longer have a
+ test system.
+
+ Based on a suggestion from Patrick Connor
+ <connor@phreak.csd.sgi.com>, added -xansi to CFLAGS
+ for IRIX 5.3 and 6.[234].
+
+ Also at Patrick's suggestion changed Configure to
+ propagate exact SunOS 4.1.x version to the main
+ and library Makefiles. This allowed the sunos413
+ and sunos413cc Configure abbreviations to be
+ shortened to sunos and sunoscc.
+
+ Updated obsolete argument uses (-H changed to -n)
+ in count_pf.perl* and watch_a_file.perl scripts.
+
+ Adjusted Solaris 2.6 lsof for Beta_Update with tips
+ from Casper Dik <casper@holland.Sun.COM>.
+
+ Fixed a Solaris 2.4 TCP address reporting bug.
+
+4.06 April 30, 1997
+ Added a step to the Makefile clean rules that does
+ a make clean in the lib subdirectory; suggested by
+ Casper Dik <casper@holland.Sun.COM>. (Configure's
+ -clean argument already did this.)
+
+ Fixed an incorrect awk argument in the sunos*)
+ Configure stanza, reported by Alexandre Oliva
+ <oliva@dcc.unicamp.br>.
+
+ Added CD9660 (aka ISO) file system support to
+ FreeBSD, NetBSD, and OpenBSD with mods and help
+ from Kenneth Stailey <kstailey@disclosure.com>.
+ (BSDI already had CD9660 support.) While at it,
+ added file descriptor system support to BSDI and
+ FreeBSD.
+
+ Added /kern file system support to OpenBSD. The
+ support wasn't extended to BSDI, FreeBSD, or NetBSD,
+ because it requires Kenneth Stailey's changes to
+ /sys/miscfs/kernfs/kernfs.h.
+
+ Updated IRIX 6.3 support after getting access to
+ a test system, provided by John Paul Morrison
+ <John.Paul.Morrison@MultiActive.com>. Improved
+ the handling of IRIX 5.1 and greater FIFOs.
+
+4.07 May 12, 1997
+ Based on AIX problem reports from David Capshaw
+ <David.Capshaw@SEMATECH.Org>, changed the aix*
+ Configure script stanza to avoid -bnolibpath for
+ gcc (which the GNU loader doesn't grok) and AIX
+ below 4.1.4 (where -bnolibpath hasn't been tested
+ or is known to be unimplemented), and to refuse to
+ use gcc for compiling lsof in AIX versions below
+ 4.1 (because of possible structure alignment
+ problems). Updated 00FAQ appropriately.
+
+ Added OpenBSD support for EXT2FS. This support
+ has yet to be tested.
+
+ Tested lsof under OpenBSD 2.1.
+
+ Activated /kern file system support for NetBSD when
+ Configure senses that /sys/miscfs/kernfs/kernfs.h
+ defines the kern_target structure. This support
+ has not been tested under NetBSD, although it has
+ been tested under OpenBSD.
+
+ Made some simple changes to the BSDI machine.h,
+ suggested by Jeffrey C. Honig <jch@bsdi.com>.
+
+ Improved handling of alternate dialect Configure
+ abbreviations -- aix and aixgcc, hpux and hpuxgcc,
+ solaris and solariscc, and sunos and sunoscc.
+
+4.08 May 23, 1997
+ Cleaned up dialect Makefile's, staring with a suggestion
+ from Christopher Schanzle <chris@cam.nist.gov>.
+
+ Improved Configure's -clean processing.
+
+ Corrected bugs in Solaris lock reporting.
+
+ Changed NetBSD Configure stanza to put -I/usr/include
+ before -I/sys.
+
+4.09 June 1, 1997
+ Adjusted for latest FreeBSD 3.0 release. This
+ required adding a new kernel name cache module for
+ reading BSD-form hashed kernel name cache entries,
+ rnmh.c, to the lsof library, and adding a #define
+ to each machine.h to select it.
+
+ Activated rnmh.c for BSDI 2.1, BSDI 3.0, NetBSD
+ 1.2, and OpenBSD 2.1.
+
+4.10 June 8, 1997
+ Adjusted for Linux 2.1.x (x > 35) kernels with
+ hashed task structure pointers. Marty Leisner
+ <leisner@sdsp.mc.xerox.com> and Jonathan Sergent
+ <sergent@io.com> tested the adjustment.
+
+ Replaced readdev() stat() calls with lstat() to
+ reduce device table and cache entries with the same
+ device number and inode values. Added code to
+ remove all remaining duplicates. This fixes a
+ Linux problem reported by Jonathan Sergent and
+ makes device node name output predictable.
+
+ Corrected a bug in UnixWare stream file handling
+ that prevented searching for the stream file by
+ its associated character device name.
+
+ Added Pyramid code to determine Reliant UNIX clone
+ major device number differently from that of DC/OSx.
+
+4.11 June 12, 1997
+ Changed Configure to sense that the PTX inp_[fl]addr
+ members of the inpcb structure of <netinet/in_pcb.h>
+ have a struct type and set HASINADDRSTR for use in
+ PTX dnode.c and dsock.c tests.
+
+ Changed PTX version 4.1.4 tests to use 4.1.3 instead.
+ Carson Wilson <carson@mcs.com> reported the need
+ to do this and tested the change.
+
+ Fixed a block device table indexing bug in lib/rdev.c,
+ reported by Carson Wilson. The same bug was squashed
+ in pyramid/ddev.c.
+
+ Added code to the Pyramid Reliant UNIX kread()
+ function to compensate for an address boundary
+ error in the kernel's /dev/kmem driver.
+
+ Verified that lsof compiles and works under AIX
+ 4.2.1. Added an AIX test for the presence of NFS
+ header files, defined HAS_NFS and adjusted AIX
+ dialect sources accordingly.
+
+ Based on a suggestion from Gaylord Holder
+ <holder@phy.ucsf.EDU>, added DEC OSF/1 code to
+ auto-detect the booted file, whence kernel symbol
+ addresses are obtained.
+
+4.12 June 24, 1997
+ Corrected a device number sign extension problem
+ in the reading and writing of device cache file.
+ The problem was reported by Bela Lubkin <belal@sco.com>
+ and he suggested a fix.
+
+ Fixed an SCO stream device lookup problem. The
+ report and solution came from Bela Lubkin
+
+ Enhanced the Configure script to enable cross-
+ configuration of lsof, based on suggestions from
+ Marty Leisner <leisner@sdsp.mc.xerox.com>. A new
+ documentation file, 00XCONFIG, describes the process.
+
+ Made Pyramid OBJFS support conditional on the
+ presence of supporting header files. Corrected
+ the Pyramid MkKernOpts script so it generates the
+ necessary -D's for the Nile/Jolt architecture.
+ Richard Coley <rcoley@pyra.co.uk> helped.
+
+ Added another IRIX xfs_inode variant for 6.2, 32
+ bits, no XFS rollup patch.
+
+ Tested under UnixWare 2.1.2.
+
+4.13 July 9, 1997
+ Taught Pyramid lsof to grok ttyfs vnodes with help
+ from Richard Coley <rcoley@pyra.co.uk>. Fixed some
+ minor bugs in Pyramid FIFO reporting. Eliminated
+ use of the Pyramid UCB compatibility library at
+ Richard's suggestion.
+
+ Eliminated reporting of "strange" inode numbers
+ for SCO OSR 3.2v5.0.x HPPS files with help from
+ Bela Lubkin <belal@sco.com>
+
+ Modified port to service name lookup to use a small
+ number of getservbyport() calls before reading the
+ entire map with getservent(). Changed port reporting
+ to represent a zero as `*' to be consistent with
+ other prt number reporting tools like netstat.
+ Casper Dik <casper@holland.Sun.COM> suggested these
+ changes -- the getserv*() one to improve performance
+ for large NIS service name maps.
+
+ Changed all readdev() functions to make the absence
+ of block devices a warning instead of a fatal error
+ after Brian Redman <ber@ms.com> reported his IRIX
+ 6.4 system had no block devices. (It really did
+ have block devices, but readdev()'s lstat() use
+ caused it to miss them in a directory symbolically
+ linked from /dev/dsk->/hw/disk.) Fixed Brian's
+ real problem by changing the IRIX readdev() to use
+ stat() on /dev nodes if a Configure test shows /hw
+ is readable. Extended the potential to do the same
+ to all readdev() functions.
+
+ For consistency and convenience changed some
+ Configure abbreviations and dialect subdirectory
+ names: "decosf" abbreviation and "osf" dialect
+ subdirectory name to "du"; "netbsd" dialect
+ subdirectory name to "n+obsd"; "next3" abbreviation
+ and "next" dialect subdirectory name to "ns"; "sco"
+ abbreviation and dialect subdirectory name to "osr";
+ "sgi" dialect subdirectory name to "irix"; and
+ "unixware" abbreviation and dialect subdirectory
+ name to "uw".
+
+ Added #if/#endif clauses to the AIX rmdupdev()
+ function to avoid clone processing for AIX versions
+ less than 4.1.4. The problem was reported by Toralf
+ Foerster <toralf.foerster@io-warnemuende.de>, who
+ supplied corrective code.
+
+ Added support for new style NetBSD inode with i_ffs
+ and i_e2fs union members.
+
+ Improved Configure and 00FAQ information on Digital
+ UNIX configuration subdirectory with suggestions
+ from Brad Krebs <brad@EECS.Berkeley.EDU>.
+
+4.14 July 22, 1997
+ Reorganized the Solaris handling of the inode
+ structure header file, ufs_inode.h, to eliminate
+ VxFS structure definition conflicts for Solaris
+ 2.4, based on information from Greg Earle
+ <earle@netbsd4me.jpl.nasa.gov>.
+
+ Cleaned up some typos and confusion in Configure's
+ help output, based on comments from Bela Lubkin
+ <belal@sco.com>
+
+ Added a 00DIALECTS file, containing UNIX dialect
+ version numbers, that can be used by Configure and
+ the man page.
+
+4.15 August 15, 1997
+ Aligned `Configure -help` output better. Removed
+ Configure's 2.6 Beta test adjustments.
+
+ Added improved Solaris VxFS configuration and
+ handling, based on information from Greg Earle
+ <earle@netbsd4me.jpl.nasa.gov>.
+
+ Added socket state -- TCO or TPI -- for socket
+ files at the suggestion of Ian Fitchet
+ <I.D.Fitchet@ftel.co.uk>.
+
+4.16 September 25, 1997
+ Added reporting of TCP/TPI queue lengths and window
+ sizes ala netstat to NAME column. Added -T option
+ to select or de-select TCP/TPI info reporting.
+ (Window sizes are only reported for Solaris.)
+ Fixed anomalies along the way in SIZE/OFF processing
+ for some dialects.
+
+ Fixed service name argument processor to allow
+ minus signs as part of the name. Consequently this
+ disallows names with embedded minus signs from
+ being specified as the start of a range.
+
+ Added 00FAQ entries explaining why lsof won't find
+ a file being edited with vi, why window sizes aren't
+ reported for all dialects, and what the "no more
+ information" message means.
+
+ Forced Pyramid CC to be /usr/ccs/bin/cc to avoid
+ accidental use of the BSD variant in /usr/ucb/cc.
+
+ Added support for Linux glibc2, including a Configure
+ test; cross-Configure support (00XCONFIG); and much
+ unfortunate and risky sleight-of-hand in lsof Linux
+ dialect header and source files, forced upon lsof
+ by incompatibilities between Linux kernel and glibc2
+ header files.
+
+ Included in scripts/identd.perl5 a Perl 5 implementation
+ of an identd server, using lsof, provided by Kapil
+ Chowksey <kchowksey@hss.hns.com>.
+
+ Updated IRIX 6.4 xfs_inode guess.
+
+4.17 October 14, 1997
+ Added -V option for verbose search result reporting.
+ Verbose reports are prepared for failure to locate
+ file names, command names, Internet addresses or
+ files, login names, NFS files, PIDs, PGIDs, and UIDs.
+
+ Augmented Linux NFS file test to cope with kernels
+ whose NFS code is in a loadable module. Need for
+ the test was pointed out by Jonathan Sergent
+ <sergent@csociety.ecn.purdue.edu>. The change
+ required that Linux have private dmnt.c source,
+
+ Completed a Linux 2.1.57 port on a system provided
+ by Jonathan Sergent.
+
+4.18 October 25, 1997
+ Eliminated memory leaks in alloc_lfile(), lkup_port(),
+ and NEXTSTEP's process_text() function.
+
+ Added recognition of OpenBSD 2.2 in Configure,
+ supplied by Kenneth Stailey <kstailey@disclosure.com>.
+
+ Consolidated print_file() functions to use the one
+ in lib/prtf.c. Made it configurable and changed
+ it to size print columns dynamically.
+
+ !!! WARNING !!!
+
+ WITH DYNAMICALLY SIZED PRINT COLUMNS LSOF 4.18
+ PRODUCES OUTPUT SIGNIFICANTLY DIFFERENT FROM THAT
+ OF PREVIOUS REVISIONS. LINES ARE GENERALLY SHORTER
+ AND THERE IS GENERALLY LESS BLANK SPACE BETWEEN
+ COLUMNS AND THE ITEMS IN THEM. THERE ARE NO LONGER
+ ANY SPACES BETWEEN DEVICE NUMBER ELEMENTS, ONLY
+ COMMAS.
+
+ !!! WARNING !!!
+
+ Added special types and print specification modifiers
+ for file size and offset to handle UNIX dialects
+ with 64 bit sizes and offsets. Paul Eggert
+ <eggert@twinsun.com> reported the need for this
+ addition.
+
+ With Paul Eggert's help picked lint from the lsof
+ library, the main level lsof sources, and the Sun
+ dialect sources.
+
+ Added documentation, including the file 00LSOF-L,
+ about the lsof-l LISTSERV.
+
+ Added support for Reliant UNIX on the RM600. Bob
+ Passarella <rmpassar@pyramid.com> supplied the
+ changes. Kevin Smith <kevin@pyramid.com> helped
+ arrange test systems. While incorporating Bob's
+ changes, modified lib/rnch.c to handle kernel ncache
+ structs whose name is accessed via a char *, rather
+ than in a char array.
+
+ Changed #include order of <sys/socketvar.h> for
+ Solaris 2.x. W. Richard Stevens <rstevens@kohala.com>
+ pointed out the need to do this.
+
+4.19 October 30, 1997
+ Changed Pyramid Reliant RM600 proc scan to skip
+ SSYS (p_flag) processes, since they don't seem to
+ have a readable u_cdir vnode.
+
+ Enabled Pyramid Reliant UNIX kread() work-around
+ for DC/OSx, too, since its read(/dev/kmem) kernel
+ driver seems to share the page boundary bug this
+ work-around circumvents.
+
+ Changed SzOffFtm_d and SzOffFtm_dv (new formats at
+ 4.18 to print size and offset) from signed to
+ unsigned. Setting them signed at 4.18 was an
+ oversight.
+
+ Plugged a memory leak that caused the loss of 130
+ bytes per repeat-mode pass. Fixed it with a simple
+ work-around in main(). Lionel Cons <Lionel.Cons@cern.ch>
+ reported the leak.
+
+4.20 November 11, 1997
+ Tested under BSDI 3.1.
+
+ Added support for Reliant UNIX Mesh IPC files with
+ help from Billy Ho <bho@pyramid.com>.
+
+ Added support to Digital UNIX lsof that uses the
+ libmsfs tag_to_path() function (when it exists) to
+ look up AdvFS path names. The idea and sample code
+ came from Dean Brock <brock@cs.unca.edu>. Converted
+ Dean's code into more general purpose support for
+ private name cache lookups via the HASPRIVNMCACHE
+ #define in the dialect machine.h file and code
+ conditional on it in the printname() function.
+
+ Taught Digital UNIX lsof to recognize NFS3 file
+ systems. Corrected Digital UNIX lsof DEVICE column
+ alignment.
+
+4.21 December 1, 1997
+ Squashed bug, introduced at revision 4.18, that
+ resulted in double reporting of each selected PID
+ when terse mode (-t) was specified.
+
+ Corrected minor bug, also introduced at 4.18, that
+ might cause an extra print_proc() pass when one
+ PID has been specified.
+
+ Added -R to lsof options in scripts/idrlogin.perl*.
+ The option should have been there -- it was supposed
+ to be mandatory for PGID reporting -- but a bug,
+ corrected in revision 4.18, previously made -R
+ unnecessary.
+
+ Enabled configuring for BSDI BSD/OS 4.0 per a
+ suggestion from Jeff Honig <jch@bsdi.com>.
+
+ Enabled replacement of scoff_t with off64_t (scoff_t
+ is used to type r_size and r_localsize in the rnode
+ struct) for IRIX 5.3 systems that have the NFS
+ kernel rollup patch (1477). This compensates for
+ SGI's failure to distribute an updated <sys/fs/rnode.h>
+ with their patch.
+
+ Validated under Linux 2.0.3[12], Linux 2.1.64, and
+ NetBSD 1.3.
+
+ Added FreeBSD root directory reporting, courtesy
+ of Dan Nelson <dnelson@emsphone.com>.
+
+4.22 December 15, 1997
+ Made adjustments for Linux 2.1.7[02].
+
+ Improved NAME information for Linux UNIX domain
+ sockets.
+
+ Added option +|-M to control the reporting of
+ portmapper registration information in square
+ brackets after the TCP or UDP port or service name.
+ Kenneth Stailey <kstailey@disclosure.com> suggested
+ the feature and provided sample code from OpenBSD.
+ Reporting is disabled by default in the distribution
+ and may be enabled with +M; if lsof is compiled
+ with HASPMAPENABLED (e.g., from machine.h), reporting
+ will be enabled by default and can be disabled with
+ -M.
+
+ Changed the -w option to +|-w to match the syntax
+ of the +|-M option and to eliminate any options
+ that flip meaning when a symbol is defined at
+ compile time. For both +|-M and +|-w, specifying
+ `-' when the default state is disabled or specifying
+ `+' when the default state is enabled causes no
+ problems.
+
+ !!!WARNING The -w option has changed in lsof 4.22. WARNING!!!
+
+ Made the +|- prefix legal for most options, but
+ didn't document it in the man page or help panel.
+ Most options that disable something -- e.g., -b,
+ -C, -n, -P -- now disable when the prefix is `-'
+ and enable when it is `+'. Since the states these
+ options disable are enabled by default, I chose to
+ avoid documentation complexity and confusion by
+ not mentioning that they can be used with the `+'
+ prefix.
+
+ Condensed the help panel.
+
+ Made sure Digital UNIX Configure stanza puts normal
+ include path (e.g., /usr/include) before system
+ include paths.
+
+ Added IPX socket information reporting to Linux
+ with help from Jonathan Sergent <sergent@purdue.edu>.
+
+4.23 January 16, 1998
+ Fixed conflict arising from the quondam replacement
+ of the Sun Solaris <netdb.h> with a BIND/BSD version.
+
+ With help from Jonathan Sergent <sergent@purdue.edu>
+ developed a /proc file system based Linux lsof.
+ It needs some Linux 2.1.x release to work -- I'm
+ not sure which, but I tested under 2.1.72, 2.1.76,
+ and 2.1.79. The Configure script selects special
+ sources for this lsof, so the full lsof distribution
+ now contains both /dev/kmem and /proc based sources
+ for Linux lsof. An optional kernel mod, written
+ by Jonathan, enhances the /proc-based lsof ability
+ to recognize IPX socket files. Reorganized and
+ augmented the Linux sections in 00FAQ to explain
+ the two types of Linux lsof.
+
+ Defined DOSTAT_FUNCTION for dostat() in misc.c to
+ select the function, stat() or lstat(), it will use.
+ DOSTAT_FUNCTION is normally undefined, defaults to
+ lstat(), and is only defined for the /proc-based
+ Linux lsof in its dlsof.h.
+
+ Made conditional on the presence of IRIX 6.4 XFS
+ rollup patch #6 an XFS node change introduced in
+ revision 4.16. Identified the patch with help
+ from John R. Vanderpool <fish@daacdev1.gsfc.nasa.gov>.
+
+ Added NFS node compensation for NetBSD 1.3. The
+ code and suggestion for it was supplied by Jean-Luc
+ Richier <richier@imag.fr>.
+
+ Added diagnostic messages to the /dev/kmem-based
+ Linux Mksrc script to report errors during the
+ construction of the kernel name cache header file,
+ kncache.h. Added 00FAQ information on kncache.h.
+
+ Added a new Linux test host, running 2.0.33 and
+ GlibC, provided by Steve Logue <stevel@mail.cdsnet.net>.
+
+ Ported to PTX 4.1.3 and 4.4.2. Adjusted lib/rnch.c
+ for 4.4.2 to allow customization f additional ncache
+ struct element names.
+
+4.24 January 28, 1998
+ Changed /proc-based Linux lsof offset test to use "/"
+ instead of "/etc/passwd".
+
+ To assist Jim Mintha <jim@geog.ubc.ca> with the
+ packaging of lsof for Debian Linux, added a
+ DEBIAN_LINUX_LSOF #define to trigger the activation
+ of special system map file location code in the
+ /dev/kmem-based dproc.c.
+
+ Applied modification to dialects/bsdi/dlsof.h from
+ Ingimar Robertson <iar@skyrr.is>, enabling lsof to
+ compile for BSDI BSD/OS 2.0.
+
+ Corrected a documentation error in 00DCACHE, pointed
+ out by Thomas Anders <anders@hmi.de>. The error was
+ created when the -V option was added at lsof 4.17.
+
+ Made IRIX 5.3 through 6.3 lsof aware of IRIX SCSI
+ tape devices (e.g., /dev/tape). Dave Olson of SGI
+ and Randolph J. Herber of FNAL provided valuable
+ advice, and Igor Schein <ischein@air-boston.com>
+ helped test.
+
+ Added a machine.h symbol (NEVER_HASDCACHE) that
+ prevents Customize from offering to change HASDCACHE.
+ The symbol may appear anywhere in machine.h --
+ e.g., in a comment. Included the symbol in a
+ comment of the HASDCACHE section of the /proc-based
+ Linux lsof machine.h, and accompanied it with
+ warnings against #define'ing HASDCACHE. Did the
+ same thing for WARNDEVACCESS (NEVER_WARNDEVACCESS
+ is the suppressant.)
+
+4.25 February 7, 1998
+ Corrected an IRIX mis-cast of file offset (position).
+ Igor Schein <ischein@air-boston.com> reported the
+ problem. This was offered as a patch to 4.24.
+ Picked some lint Igor pointed out.
+
+ At Igor's suggestion added an optional decimal
+ digit size argument to the -o option. This argument
+ specifies how many file offset decimal digits can
+ follow "0t" before lsof switches to a "0x..." form.
+ The argument size specification doesn't count the
+ two characters of the "0t". A size of 0 means
+ unlimited. The default is OFFDECDIG (8), preserving
+ compatibility with existing lsof output; it can be
+ changed by the lsof builder. When size is specified
+ with -o it does not force offset display; -o without
+ a size still must be used to do that.
+
+ Added an IRIX 6.2, 32 bit system, XFS node patch,
+ courtesy of Ulrich Bernhard <rzubu@rzu.unizh.ch>.
+
+ For my own convenience enabled Configure to use
+ /usr/local/bin/gcc for NEXTSTEP. This allows
+ circumvention of a gcc 2.8.0 ranlib problem on
+ my test 3.1 `040 cube.
+
+ Added flags recommended by the RISC/os and Ultrix
+ compilers for the updated (and longer) main.c.
+
+ Updated FreeBSD cd9660_node.h Configure test.
+
+4.26 February 17, 1998
+ Added shared process group processing for IRIX 5.3,
+ and IRIX 6.1 and above, based on investigation of
+ a bug report from Igor Schein <ischein@air-boston.com>.
+ Igor helped test this addition.
+
+ Improved handling of file system name arguments.
+ It's now done in a manner similar to fuser. The
+ -f argument forces path names to be considered as
+ simple files, rather than as file system names.
+ The +f flag forces them to be considered as file
+ system names. Normally path arguments are considered
+ file system names when they match a mounted-on
+ directory in the system's mount table, or when they
+ match a mounted file system's block device. Igor
+ Schein helped test this change.
+
+ Igor also suggests that the proper compilation of
+ the IRIX 6.4 proc structure after patch 2536 has
+ been installed may need -DPIOMEMOPS. So lsof's
+ MkKernOpts script was updated to propagate that
+ option from CCOPTS in /var/sysgen/system/irix.sm,
+ even though patch 2536 doesn't add -DPIOMEMOPS to
+ it. Added a 00FAQ item on this patch.
+
+ Added a fatal warning message about names forced
+ to be file system names (with +f) that have no
+ match in the mount table.
+
+ Improved the -V message for files and file systems
+ for which no open files were found. Added reporting
+ of /proc file and file system search failures.
+
+ Did some code reorganization to combine the multiple
+ ck_file_arg() functions into one. Moved the new
+ function from the library to the top level and put
+ it in arg.c; moved the usage function from arg.c
+ to a new top-level source file, usage.c, to balance
+ top-level source file size. The new usage.c depends
+ on version.h; arg.c no longer does.
+
+ Added flag recommended by the DU compiler for the
+ updated (and longer) main.c.
+
+4.27 March 6, 1998
+ At the request of Igor Schein <ischein@air-boston.com>
+ added a conditional repeat mode option, using the
+ `+' prefix to the `r' option. +r operates as does
+ -r with the exception that it exits the first time
+ no open files have been listed during a cycle.
+ The exit code will be zero when any open files have
+ been listed; one, if none were ever listed.
+
+ Ported lsof to HP-UX 11.0 with the help of Richard
+ Allen. This port hasn't been tested on a 64 bit
+ kernel; I'm sure it won't work there without more
+ mods. It may not work on PA 2 architectures; I've
+ only tested it under PA 1 and a separate, busy
+ tester reported PA 2 problems that I've been unable
+ to investigate.
+
+ In anticipation of getting access to a 64 bit HP-UX
+ kernel and the pending start of the Solaris 2.7
+ Beta test (It will have 64 bit kernel addressing.),
+ started adding support for 64 bit kernel pointers.
+ This includes: ubiquitous use of the KA_T cast
+ for kernel pointers; a format to print them,
+ KA_T_FMT_X; a function to print them, print_kptr();
+ and modifications to most kernel-related functions
+ -- e.g., process_file(), process_node(),
+ process_socket(), readvfs() -- to process kernel
+ addresses as KA_T types.
+
+ Fixed minor bug in handling path name arguments
+ that end with a `/'.
+
+ Removed support for RISC/os; its test system is no
+ longer available.
+
+ Made modifications to insure that lsof output
+ doesn't contain non-printable characters. All such
+ characters are now printed in the printf form
+ "\x%02x". Several new common functions were
+ installed in misc.c to support "safe" printing.
+ This second major modification in 4.27 to common
+ and dialect code could have introduced bugs not
+ yet detected.
+
+4.28 March 10, 1998
+ Refined unprintable format to use \b, \f, \r, \n,
+ \t, and ^* (for CTRL) forms. Corrected omission
+ of safestrprt() use for field output command name.
+ These changes were offered as patches to 4.27.
+
+ Made space an unprintable character (\x20) in the
+ COMMAND column; printable elsewhere, including the
+ NAME column, field output, and error messages.
+
+ Made sure FD column is parseable as a single entity
+ -- i.e., has no embedded space. Thus, if the access
+ mode is unknown but there is a known lock mode, (a
+ very rare case) the access mode will be printed as
+ `-'.
+
+ Picked lint with gcc 2.8.0 under Solaris 2.6.
+
+ With the help of Dave Olson of SGI identified a
+ proc struct element that should have been added to
+ <sys/proc.h> by IRIX 6.4 patch 2536. Added a
+ work-around for it to the lsof Configure script.
+ Igor Schein <ischein@air-boston.com> identified
+ that the patch caused a proc structure length
+ complaint from lsof. Removed an obsolete 00FAQ
+ item on the patch, installed at lsof 4.26, explaining
+ that no solution was yet available.
+
+ Added a 00FAQ item on how BIND installs its own
+ header files, including <netdb.h>, which may cause
+ the rpcent struct definition to vanish. Solaris
+ has an automatic lsof work-around, but that hasn't
+ been (and probably can't be) propagated to all
+ dialects supported by lsof. The 00FAQ item recommends
+ re-installation of the vendor header files that
+ BIND has replaced. (Others include <rpcent.h>,
+ <sys/bitypes.h>, and <sys/ctypes.h>.)
+
+ Made AIX AFS fixes.
+
+4.29 March 26, 1998
+ Corrected bug in Internet address matching. The
+ matching formerly stopped if the foreign address
+ matched, thus failing to check the local address
+ for a match. That led to a possible false "Internet
+ address not located" warning (i.e., in response to
+ -V) about the local address, when both foreign and
+ local addresses were specified with -i. This
+ correction was offered as a patch to 4.28.
+
+ Changed readmnt() usage in an attempt to defer
+ mount readlink() and stat() delays until they are
+ necessary.
+
+ Corrected two bugs in the Digital UNIX readdev()
+ function. Made the correction available as a patch
+ to 4.28 and regenerated the 4.28 DU binaries.
+
+ Added a missing argument to a print-kptr() call in
+ the HP-UX dsock.c. The missing argument causes a
+ fatal gcc error. The problem was reported by Eyal
+ Shaynis <eyal.shaynis@telrad.co.il>. The fix was
+ offered as a 4.28 patch.
+
+ Adjusted for Digital UNIX 4.0D; the spec_node
+ structure is now defined in <sys/specdev.h>. Kris
+ Chandrasekhar <Kris.Chandrasekhar@digital.com>
+ identified the need for the adjustment.
+
+ Incorporated a bug fix from Brian McAllister
+ <mcallister@mit.edu> to the DU readmnt() function.
+ This fix was offered as a patch to 4.28.
+
+ Added "safe" printing to a SunOS clone device error
+ message.
+
+ Corrected bug in tabling of Linux /proc-based lock
+ info.
+
+ Corrected bug in handling of SunOS TLI streams.
+ Dan Farmer <zen@trouble.org> reported the problem.
+
+ Added a Solaris 2.6 work-around to keep the BIND
+ <sys/bitypes.h> from colliding with the Solaris
+ <sys/int_types.h>.
+
+ Strengthened the Configure test for /proc-based
+ Linux lsof, based on a report from Marty Leisner
+ <leisner@sdsp.mc.xerox.com>.
+
+ Tested on OpenBSD 2.3.
+
+ Made AIX changes that allow use with 3.2.5. The
+ changes were suggested and tested by Brett Hogden
+ <hogden@rge.com>.
+
+ Added Solaris 2.6 AFS support. Disabled reporting
+ of some node numbers for Solaris 2.5 and above open
+ AFS files. The node number computation algorithms
+ used for SunOS 4.1.x and Solaris less than 2.5 no
+ longer always work under Solaris 2.5 and above.
+
+4.30 April 9, 1998
+ Corrected a pid structure member naming error for
+ UnixWare < 2.1.2. The problem was reported by
+ Richard van Meurs <vanmeurs.anva@atriserv.nl>. He
+ supplied the correction. This was offered as a
+ patch to 4.29.
+
+ Had a report from Igor Schein <ischein@air-boston.com>
+ that IRIX 6.4 patch 2839 is another SGI kernel
+ patch, along with 2536, that changes the size of
+ the proc structure in the kernel without changing
+ the proc structure in <sys/proc.h>. Upon further
+ investigation found that the effect of these patches
+ on the proc structure is not consistent. Therefore,
+ dropped the Configure patch test for IRIX 6.4 and
+ made the code in irix/dproc.c slightly more tolerant
+ of proc structure size differences for IRIX 6.4.
+ Igor help test the change.
+
+ Corrected Solaris >= 2.5 AFS inode number generation.
+ Craig Everhart <Craig_Everhart@transarc.com> helped
+ find the cause of the problem. This was offered as
+ a patch to 4.29.
+
+ Refined the Linux /dev/kmem-based glibc evasion
+ for the timeval structure to make it work with
+ glibc version 2.0.7. This required defining a new
+ global symbol, TIMEVAL_LSOF, default timeval, that
+ the /dev/kmem-based Linux lsof can set to its
+ private glibc timeval name, distinct from the kernel
+ timeval name.
+
+ Added support for Alpha to the /dev/kmem-based
+ Linux lsof. Alexandre Oliva <oliva@dcc.unicamp.br>
+ provided a test system. Added an item to 00FAQ
+ about lsof, the Alpha processor, and Linux.
+
+ Added a 00FAQ item about lsof year 2000 compliance.
+ Basically it says lsof is probably compliant,
+ because its only date or time computations are done
+ with time_t values, but I haven't done any specific
+ Y2K validation. I don't have plans to do any.
+
+ Added support for UnixWare 7. Chris Daniels
+ <chrisd@dlpco.com> provided a test system and Don
+ Draper <dond@sco.COM> provided technical information.
+ Added BFS and SFS file system support to lsof for
+ UW 2.1.[12] and 7.
+
+ Updated Solaris VxFS support for VxFS 3.2.1. Greg
+ Earle <earle@netbsd4me.jpl.nasa.gov> reported the
+ need for the update. Greg and Roger Klorese
+ <rogerk@veritas.com> provided technical information.
+ Scott McClung <mcclung@primenet.com> tested.
+
+ Changed IRIX XFS patch detection in anticipation of
+ learning there are multiple XFS patches for IRIX 6.4
+ that require different versions of the lsof-invented
+ xfs_inode structure.
+
+4.31 April 21, 1998
+ Added a VxFS #if/#endif wrap to a section of the
+ HP-UX dnode.c that wasn't properly protected. The
+ problem was reported by Peter Klosky <PKlosky@bdm.com>.
+ This was offered as a patch to 4.30.
+
+ Added support for Solaris 2.7 (first Beta release).
+ Mike Sullivan <Mike.Sullivan@Eng.Sun.COM> provided
+ technical advice and helped test. Charles Stephens
+ <cfs@jurassic.eng.Sun.COM> also helped test.
+
+ Fixed bug in /proc-based Linux that caused it to
+ access /proc/mounts excessively. Marty Leisner
+ <leisner@sdsp.mc.xerox.com> provided a syscall
+ trace that identified the bug. The fix was offered
+ as a patch to 4.30.
+
+ Adjusted the IRIX 6.4 private structure definition
+ for the XFS node to accommodate patch 2970. Igor
+ Schein <ischein@air-boston.com> identified the
+ patch and the required adjustment.
+
+4.32 May 11, 1998
+ Corrected Solaris 2.7 code for reporting PCFS
+ (floppy disk) node numbers. Casper Dik
+ <casper@holland.sun.com> supplied the fix. The
+ fix was offered as a patch to 4.31.
+
+ Corrected a bug in conditional repeat mode handling
+ pointed out by Igor Schein <ischein@air-boston.com>.
+ This was offered as a patch to 4.31.
+
+ Improved reporting of AIX open(/dev/memory device)
+ errors.
+
+ Corrected a Solaris < 2.5 KA_T declaration error,
+ pointed out by Robert Kiessling <robert@easynet.de>.
+ Changed KA_T from a #define to a typedef for all
+ dialects to prevent future problems of this kind.
+
+ Changed the sample Perl 5 script big_brother.perl5
+ to report a four digit year from localtime().
+
+ Added support for AIX 4.3[.1]. Bill Pemberton
+ <wfp5p@tigger.itc.virginia.edu> provided a test
+ system. Andrew Kephart <akephart@austin.ibm.com>
+ and Tom Weaver <tvweaver@austin.ibm.com> provided
+ technical assistance. Niklas Edmundsson
+ <nikke@ing.umu.se> did 4.3.1 testing.
+
+ Added -qmaxmem option to CFLAGs for an AIX compilation
+ with an xlc version 4.x compiler.
+
+ Adjusted Linux socket handling for changes in the
+ AX25 members of the sock struct. Richard Green
+ <rtg@tir.com> pointed out the problem. Tested
+ /dev/kmem-based lsof under Linux 2.0.34.
+
+4.33 May 22, 1998
+ Added generic IPv6 support to common lsof sources
+ and specific IPv6 support to AIX sources. Andrew
+ Kephart <akephart@austin.ibm.com> supplied the
+ additions and helped with testing. Bill Pemberton
+ <wfp5p@tigger.itc.virginia.edu> provided a test
+ system. The modification affected sources for
+ every dialect, whether it supports IPv6 or not, by
+ changing the interfaces to the common Internet
+ address function ent_inaddr().
+
+ Added support for the NetBSD UVM virtual memory
+ system. Paul Kranenburg <pk@cs.few.eur.nl> supplied
+ technical details.
+
+ Bracketed HP-UX 11 use of <sys/spinlock.h> with
+ #if/#endif _KERNEL.
+
+ Corrected printing of PCB address in DEVICE column
+ for IRIX.
+
+4.34 June 26, 1998
+ Updated 00FAQ to discuss TCP and UDP ports private
+ to the AIX kernel and 00README to describe how ACLs
+ can be used to give lsof permission to read the
+ kernel memory devices. Add information to 00FAQ
+ and 00README about other OpenBSD architectures
+ where lsof is reported to compile and run. Added
+ section to 00FAQ discussing how an incorrect loader
+ path environment variable value can prevent lsof
+ from loading correctly.
+
+ Improved Solaris namefs and doorfs support so that
+ it is now possible to search for an open VDOOR file
+ by the path name of its fattached file system
+ object. Igor Schein <igor@txc.com> requested the
+ ability to do such a search. Even with the change,
+ lsof can't always identify path names for open
+ VDOOR files.
+
+ Also at Igor's request, improved reporting of
+ information on open Solaris VCHR files that share
+ a common vnode, and Solaris UNIX domain socket
+ files.
+
+ Corrected print_kptr() argument error in PTX dnode.c,
+ reported by Mark Price <mprice@sequent.com>.
+ Compensated for ncache element naming differences,
+ introduced at PTX 4.4.2; Kurtis D. Rader
+ <krader@sequent.com> reported the problem.
+
+ Changed output column title from INODE to NODE to
+ better reflect the column's contents of node IDs
+ for more than just inodes.
+
+ Improved Configuration and processing for Solaris
+ AFS. Corrected AIX AFS 3.4 afs_rwlock_t simulation.
+
+ Corrected a cast problem with two AIX knlist()
+ calls, thus quieting an AIX 4.2.1 compiler argument
+ type warning. Jon Champlin <champlin@us.ibm.com>
+ reported the problem.
+
+ Added support to most dialect versions (exception:
+ /proc-based Linux) to warn when the identity of
+ the kernel where lsof was compiled doesn't match
+ the running identity. The warning can be suppressed
+ with -w. Note: determining AIX state requires
+ calling oslevel, a potentially slow operation.
+ Jon Champlin <champlin@us.ibm.com> suggested this
+ addition.
+
+ !!!! WARNING !!!! !!!! WARNING !!!! !!!! WARNING !!!!
+
+ Those using the lsof cross-configuration capability
+ (see 00XCONFIG), should be aware that the kernel
+ identity test feature introduces two new basic
+ cross configuration environment variables, LSOF_ARCH
+ and LSOF_VSTR.
+
+ !!!! WARNING !!!! !!!! WARNING !!!! !!!! WARNING !!!!
+
+ Identified a situation where a Solaris UNIX domain
+ socket name is known and can be searched for by
+ name; added the necessary code.
+
+4.35 July 17, 1998
+ Made the kernel identity check an option with the
+ HASKERNIDCK #define in machine.h. Enabled altering
+ of HASKERNIDCK with the Customize script. Added
+ a clause to the help output that indicates the
+ build-time HASKERNIDCK status.
+
+ Added more information to the NAME column for
+ Solaris UNIX domain sockets. Made them searchable
+ by their clone device path name. Igor Schein
+ <igor@txc.com> requested this.
+
+ Completed the HP-UX 11 port with support for its
+ optional 64 bit kernel. Rich Rauenzahn
+ <rrauenza@cup.hp.com> provided a test system.
+ Corrected errors with HP-UX 11 lock reporting and
+ private kernel structure and type definitions.
+ Added support for HP-UX NFS3 files.
+
+ Limited mount table warnings -- e.g., when -b is
+ used -- to one set per mount point.
+
+ Fixed some mount table scanning and usage bugs,
+ including one in Solaris, reported by Kjetil Torgrim
+ Homme <kjetilho@ifi.uio.no>.
+
+4.36 August 4, 1998
+ Made corrections and additions to IPv6 support and
+ to AF_ROUTE socket handling, supplied by Jean-Luc
+ Richier <Jean-Luc.Richier@imag.fr>. Jean-Luc's
+ additions provide IPv6 support for the Inria IPv6
+ implementations on FreeBSD and NetBSD.
+
+ Fixed two Solaris 2.5, 2.5.1, 2.6 and 2.7 TCP and
+ UDP host name or IP address reporting bugs, reported
+ by James Mathiesen <James-Mathiesen@deshaw.com>.
+ This fix was offered as a patch to 4.35.
+
+ Updated the Customize script to cause ENTER to use
+ all defaults. Amir J. Katz <amir@ndsoft.com>
+ suggested this and helped test the changes.
+
+ Updated Solaris ICMP and IP stream handling, based
+ on a report from Igor Schein <igor@txc.com>.
+
+ Fixed a bug in the Digital UNIX mount table handling,
+ reported by Bob Ward <bward@thehartford.com>.
+ While working on the bug, found and updated some
+ obsolete AdvFS code. This fix was offered as a
+ patch to 4.35.
+
+4.37 September 15, 1998
+ Deactivated SGI IRIX support and archived revision
+ 4.36 sources and binaries in pub/tools/unix/lsof/OLD.
+
+ Improved performance of FD searching. This was
+ offered as a patch to 4.36.
+
+ Amir J. Katz <amir@ndsoft.com> pointed out that
+ ranlib isn't needed for AIX or Solaris. Made
+ appropriate Configure script changes.
+
+ Fixed a file offset reporting bug for HP-UX VCHR
+ and VBLK device nodes located on a VxFS root. Doug
+ Siebert <douglas-siebert@iowa.edu> reported the
+ bug. The fix was offered as a patch to 4.36.
+
+ Resolved an HP-UX root device name reporting bug,
+ partly caused by an out-dated local copy of the
+ <sys/mount.h> mount structure, by generating a
+ local header file with the structure that can be
+ compiled without needing _KERNEL defined. Doug
+ Siebert also reported this bug.
+
+ Changed some dialect source code -- Digital UNIX,
+ Solaris, SunOS, and UnixWare -- to make more
+ consistent with ps the user ID lsof reports in the
+ USER column. Added a 00FAQ entry about it. Igor
+ Schein <igor@txc.com> reported the Solaris and
+ SunOS lsof inconsistencies with what ps(1) reports.
+
+ Ported lsof to Pyramid ReliantUNIX 5.44.
+
+ Added brackets as comments to case, do, done, else,
+ endif, esac, if, and while statements in Configure
+ to assist in navigating its clauses.
+
+ Added more Linux 2.0.x glibc work-arounds.
+
+ Added support for UnixWare 7.0.1.
+
+ Ralph Forsythe <ralph@contact-paging.com> provided
+ a new FreeBSD test system.
+
+4.38 November 25, 1998
+ Added support for recent FreeBSD 3.0 distributions.
+ A 3.0 test system was provided by David O'Brien
+ <obrien@NUXI.com>. This was offered as a patch
+ to 4.37.
+
+ Updated the scripts/idrlogin.perl* files to look
+ for sshd processes in addition to rlogind and
+ telnetd ones.
+
+ Added support for DU 5.0 Beta. Berkley Shands
+ <berkley@cs.wustl.edu> provided a test system.
+
+ Added support for OpenBSD 2.4 with changes supplied
+ by Kenneth Stailey <kstailey@disclosure.com>.
+
+ Changed the Solaris 2.7 tests and documentation to
+ Solaris 7.
+
+ Made some changes to the header files for NEXTSTEP
+ 3.3 and added support for OPENSTEP 4.x with help
+ from Michael A. Hovan III <mhovan@BLaCKSMITH.com>
+ and Carl Lindberg <Carl_Lindberg@BLaCKSMITH.com>.
+ The combined dialect subdirectory is named n+os.
+ One of Carl's changes propagates RC_CFLAGS to the
+ library Makefile. Timothy J. Luoma <luomat@peak.org>
+ helped test under NEXTSTEP 3.3 and OPENSTEP 4.2.
+
+ Made UW 7.x version sensitive to the presence of
+ ptf7038. Added peer PCB address to Unix domain
+ socket Name column, even when a path name has been
+ located. Information for these changes was supplied
+ by Francis Le Bourse <flebourse@intelcom.fr>. Lee
+ Penn <lee@dlpco.com> provided a test system.
+
+ Tested lsof under OSR 5.0.5 on a test system also
+ provided by Lee Penn.
+
+ Made path name argument processing more tolerant
+ of errors per a suggestion from Julian Gordon
+ <julian@cadence.com>.
+
+ Acquired a new UnixWare 2.x test system, generously
+ provided by Computer Classroom, Inc. -- Matthew
+ Thurmaier <matt@compclass.com>, Ken Laing
+ <ken@compclass.com>, and Andrew Merril
+ <andrew@compclass.com>. Updated Configure to accept
+ a UnixWare version of 2.1.3.
+
+ Updated kmem-based lsof for Linux 2.0.36.
+
+ Updated NetBSD sources for a change in a UVM virtual
+ mapping header file.
+
+ Corrected a cache allocation bug in Sun format
+ kernel name cache handling. The bug only shows up
+ when the kernel name cache is inaccessible.
+
+4.39 December 29, 1998
+ Corrected problems with large device number handling
+ for 64 bit Solaris 7. The problems were reported
+ by Steve Bellenot <bellenot@math.fsu.edu>. Steve
+ helped test the fixes. The fixes were offered as
+ two patches to lsof 4.38.
+
+ Improved FreeBSD Configure operations for header
+ files that must be obtained from the kernel source
+ tree, based on a suggestion from David O'Brien
+ <obrien@NUXI.com>.
+
+ For Bela Lubkin <filbo@deepthought.armory.com> made
+ optional with +f[cfn] the display of file structure
+ address, shared use count, and node structure
+ address. /proc-based Linux doesn't implement this
+ feature, because it doesn't read kernel structures
+ from kernel memory. Modified the PTX -X option to
+ take advantage of the new file structure display
+ option. Added shared.perl5 to the scripts/
+ subdirectory to provide an example of how +f[fn]
+ might be used to track shared file descriptors and
+ files.
+
+ Added more /dev/kmem-based Linux glibc evasions,
+ provided by Jeff Johnson <jbj@redhat.com> and Maciej
+ Lesniewski <nimir@kis.p.lodz.pl>. Jeff helped test
+ them on various Linux architectures.
+
+ Tested on AIX 4.3.2; no changes were required.
+ Doug Crabill <dgc@purdue.edu> provided a test
+ system.
+
+ Fixed -c option to detect missing command name when
+ following option begins with `+'.
+
+4.40 January 25, 1999
+ Added support for using the CDS compiler for Reliant
+ Unix 5.44 and above. Made Reliant Unix MIPC support
+ optional, dependent on the presence of <sys/mipc.h>.
+
+ Based on a report from Michael Schmitz <MSchmitz@lbl.gov>
+ that /dev/kmem-based lsof misbehaves on a Linux
+ 2.0.x m68k kernel without module support, made the
+ absence of query_module() or get_kernel_syms()
+ Linux kernel support a fatal error. Updated relevant
+ sections of 00FAQ to reflect the change.
+
+ Added the ability to force the Linux Configure
+ stanza to use the /proc or /dev/kmem source base
+ via a LINUX_BASE environment variable specification.
+ This is a cross-configuration assist.
+
+ Added "+D <dir>" and "+d <dir>" options for directory
+ searching. +D searches the entire tree, starting
+ at <dir>, including <dir>, its contents, and its
+ subdirectory branches; +d searches only <dir> and
+ its contents, but not its subdirectory branches.
+ Improved lsof's searching of the specified name
+ list to compensate for anticipated long lists from
+ +d and +D.
+
+ Made an egrep in the Solaris Configure stanza usable
+ by the standard and XPG4 egrep's. Kenneth Stailey
+ <kstailey@disclosure.com> pointed out the improvement.
+
+ Fixed bugs in /dev/kmem-based Linux and UnixWare
+ Unix domain socket name searching.
+
+ Changed a Linux Alpha #include to be conditional
+ on the presence of its named header file, so that
+ lsof will compile on Red Hat 5.1 and 5.2 (Linux
+ kernel 2.0.35) where the header file is absent.
+ The problem was reported by Alexandre Oliva
+ <oliva@dcc.unicamp.br>.
+
+ Fixed an AIX 4.3+ bug in procinfo struct space
+ allocation, reported by Jeff Stewart <jws@purdue.edu>.
+ This was offered as a patch to 4.39.
+
+ Added an lstatsafely() function to offer the same
+ isolation for lstat() calls that statsafely() offers
+ for stat() calls. This made DOSTAT_FUNCTION no
+ longer necessary, so deleted it.
+
+ With help from Laurent P. Montaron <lpm@sequent.com>
+ ported lsof to PTX 4.4.4. Laurent did a monumental
+ job of identifying TCP/IP changes by their TCP
+ version, rather than by their PTX (With mix 'n
+ match PTX and TCP/IP versions, the PTX version
+ often has no bearing on the TCP/IP version.), and
+ changed the Configure script and pre-processor
+ #if/#else/#endif blocks to match. He also updated
+ Unix domain socket handling for PTX TCP/IP versions
+ 4.5 and above.
+
+ Updated CLIENT handle acquisition of fill_portmap()
+ in print.c to use the more modern RPC function
+ clnt_create() in place of clnttcp_create() where
+ possible. PTX 4.4.4 requires clnt_create().
+
+4.41 February 27, 1999
+ Added FreeBSD 3.1 and and 4.0 support with help
+ from Sheldon Hearn <axl@iafrica.com>, David O'Brien
+ <obrien@NUXI.com>, and John Polstra <jdp@polstra.com>.
+
+ Corrected bungled AIX 4.3+ patch that went into
+ lsof 4.40.
+
+ Reorganized the Configure script to improve Makefile
+ construction. A specific impetus for this was to
+ allow FreeBSD system-wide make flags to be propagated
+ to the lsof Makefiles, but other goals were to make
+ sure that the DEBUG= make entry can over-ride
+ standard CFLAGS values, and to better manage the
+ identification of compilers and their versions.
+ Two compiler-related values may now be supplied in
+ environment variables: 1) the compiler path in
+ LSOF_CC; and 2) the compiler version in LSOF_CCV.
+ 00XCONFIG documents them.
+
+ Added support for Pyramid Reliant Unix bsdsfs,
+ msockfs, and sockfs file systems.
+
+ Added an optional LSOF_CINFO string to Configure,
+ producing a CINFO string in selected Makefiles,
+ producing a #define LSOF_CINFO in selected version.h
+ header files. The purpose of this is to allow
+ Configure the option to propagate information to
+ the lsof -v output. It is now used for Linux to
+ identify the code base, and for HP-UX 10.30 and
+ 11.0 and Solaris 7 to identify the kernel bit size.
+
+ Added system information to NEXTSTEP and OPENSTEP
+ -v output, from the second line of hostinfo's
+ output.
+
+ Fixed a login name buffer overflow problem in the
+ processing of -u option values. This was offered
+ as a patch to 4.40. !!!THIS IS A SERIOUS STACK
+ OVERFLOW BUG; A LINUX EXPLOIT EXISTS FOR IT THAT
+ OPENS A BASH SHELL WITH LSOF'S AUTHORITY -- E.G,
+ SETGID(KMEM) POWER!!!
+
+ Improved the Solaris mount table filter so the
+ volume manager's fake mount point, "/vol", is
+ ignored and doesn't supplant "/" in NAME column
+ path assemblies. Igor Schein <igor@txc.com> reported
+ this bug and provided important help in finding
+ it. This was offered as a patch to 4.40.
+
+ Changed the Linux /dev/kmem-based lock ownership
+ test to answer a problem reported by Tom Christiansen
+ <tchrist@jhereg.perl.com>. This was offered as a
+ patch to 4.40.
+
+ Installed an HP-UX 11 patch, suggested by Kevin
+ Vajk <kvajk@cup.hp.com>, that adjusts a private
+ lsof kernel header file, derived via Q4, to correspond
+ to an HP-UX patch bundle.
+
+ Made NetBSD 1.3I sockproto structure adjustment.
+
+4.42 March 30, 1999
+ Fixed a typo in the HP-UX dfile.c that caused +fF
+ and +fN output controls to swap effect.
+
+ Enabled for OpenBSD 2.5 per notice from Kenneth
+ Stailey <kstailey@kstailey.tzo.com>
+
+ Made more VM accommodations for FreeBSD 4.0.
+
+ Improved file system search reporting to include
+ path name components when they're available, instead
+ of mindlessly reporting the file system name in
+ the NAME column. Guy Dallaire <gdallair@geocities.com>
+ brought the need for this change to my attention.
+
+ Updated Solaris 2.6 VxFS for Veritas Oracle Database
+ Edition 2.0, VxFS version 3.3, and VxVm version
+ 2.5.4, based on a report from Chris Kordish
+ <chris.kordish@East.Sun.COM>. Chris kindly provided
+ a test system.
+
+ Improved HP-UX ipc_s patch detection in Configure,
+ response in .../dialects/hpux/hpux11/ipc_s.h, and
+ documentation in 00FAQ, Kevin Vajk <kvajk@cup.hp.com>
+ helped test.
+
+ Added to Customize the option to suppress HASKERNIDCK
+ selection for specified dialects. Suppressed it
+ for /proc-based Linux lsof, and removed its test
+ and code from there. Tin Le <tin@netimages.com>
+ alerted me to the need for this update.
+
+ Ported to official Digital UNIX 5.0 release.
+
+ Changed DU lsof to use the knlist(3) function when
+ no kernel file has been specified with -k. This
+ change was suggested by Erich Wimmer
+ <Erich.Wimmer@digital.com>.
+
+ Updated Configure for latest NetBSD (1.3I?) with
+ UVM support the default.
+
+4.43 May 11, 1999
+ Corrected a typo in the Solaris gcc discussion in
+ 00FAQ. Made changes to the Solaris 2.5[.1] private
+ tcp_s structure. Both changes were done in response
+ to reports from Igor Schein <igor@txc.com>, who
+ tested the Solaris 2.5 change.
+
+ Made more IPv6 adjustments to lsof for Tru64 UNIX
+ (Digital UNIX) 5.0, based on information obtained
+ from Compaq by Berkley Shands <berkley@cs.wustl.edu>.
+
+ Corrected HP-UX error message about HP-UX 11 q4 usage.
+ Amir Katz <amir@ndsoft.com> reported the correction.
+
+ Fixed a GlibC 2.1 conflict in /proc-based Linux lsof.
+
+ Fixed a man page typo reported by Vlad Harchev
+ <hvv@hippo.ru>.
+
+ Changed some Solaris 2.7 references to Solaris 7
+ in Configure and 00XPORTING.
+
+ Added a Solaris example to the echo statements that
+ are the install rule in the SunOS/Solaris Makefile.
+
+ Added a field to the file structure output --
+ FILE-FLAG (file structure open flags, f_flag[s],
+ and process file flags, typically u_pofile)) --
+ enabled with +f[gG]. Its field output character
+ is 'G'.
+
+ Figured out another piece of the HP-UX 11 patched
+ ipc_s structure puzzle with the help of Keith Kalet
+ <KEITH_KALET@HP-USA-om41.om.hp.com>.
+
+ Fixed a PTX real vnode to real inode interpretation
+ bug.
+
+ Added link count to lsof output. Eric Dumazet
+ <dumazet@risgw.ris.fr> requested and helped test
+ it. The new +L option enables and filters it.
+ Its field output character is `k'.
+
+ Updated Configure script to recognize NetBSD 1.4.
+
+ Updated AFSConfig to handle default answers to
+ questions.
+
+ Incorporated patch from Jonathan Sergent <sergent@io.com>
+ that enables /proc-based Linux lsof to run on both
+ 32 and 64 bit kernels.
+
+ Updated Configure script with a patch from David
+ O'Brien <obrien@NUXI.com> that recognizes FreeBSD 3.2.
+
+4.44 June 24, 1999
+ Corrected use of nlink member of hsnode for SunOS
+ 4.1.x High Sierra File System files. John Dzubera
+ <zube@tlaloc.stat.colostate.edu> reported the
+ problem and helped test the fix. Also fixed a
+ SunOS segmentation fault bug. These fixes were
+ offered as a patch to 4.43.
+
+ Improved handling of /proc-based Linux UNIX PCB
+ address.
+
+ Fixed a NEXTSTEP and OPENSTEP bug that made repeat
+ option (-r) processing malfunction. This fix was
+ offered as a patch to 4.43.
+
+ Fixed Configure so it doesn't use -O in the Cflags
+ for the bundled HP-UX C compiler. Jim Ankenbrandt
+ <jankenbrandt@penton.com> reported the problem.
+
+ Corrected output ordering of parent PID and process
+ group ID when both -R and -g are specified.
+
+ Enhanced the pdev.c and pdvn.c library modules for
+ wider use. These dialect versions use the new
+ library modules: DEC OSF/1, Digital UNIX, and Tru64
+ UNIX; Pyramid DC/OSx and Reliant UNIX; SCO OSR and
+ UnixWare; and Sequent PTX.
+
+ Added basic clone device support to /dev/kmem-based
+ HP-UX lsof for HP-UX 10.30 and higher.
+
+ Added raw socket support to /proc-based Linux lsof.
+
+ Changed NODE-ADDR column title to NODE-ID in
+ anticipation of using more general identification
+ information in the column.
+
+ Ported to UnixWare 7.1, using a test system kindly
+ provided by Matt Thurmaier <matt@compclass.com>
+ and Don Draper <dond@sco.com>.
+
+ Updated for NetBSD 1.4C VM changes, and a new
+ current and root working directory structure.
+
+ Made minor adjustment for latest Tru64 UNIX 5.0
+ Beta release.
+
+4.45 July 30, 1999
+ Fixed quoting problem in DEC OSF/1, Digital Unix,
+ and Tru64 UNIX Makefile's install rule. The problem
+ was reported by Berkley Shands <berkley@cs.wustl.edu>.
+ Fixed bug in Tru64 UNIX 4 lsof that caused FDs to
+ be skipped. These fixes were offered in a patch
+ to 4.44.
+
+ Fixed a repeat-mode /proc-based Linux lsof bug,
+ reported by Sami Farin <sfarin@ratol.fi>. This
+ was offered as a patch to 4.44.
+
+ Picked lint, some reported by Sami Farin.
+
+ Corrected a 00DCACHE documentation error in a sample
+ shell script. The problem was reported by Chad R.
+ Larson <chad@larsons.org>. Changed commented-out
+ entries in machine.h files so they require more
+ thought and work when the comments are removed,
+ based on a remark by Chad.
+
+ Compensated for the practice of Solaris 7 and above
+ to record the dev= value in /etc/mnttab in 32 bit
+ mode, even on 64 bit systems. This was offered as
+ a patch to 4.44.
+
+ Added a C library test for /proc-based Linux lsof,
+ so that the #include files can be adjusted for a
+ non-GlibC environment. The need for this was
+ reported by Andrew Hill <andrewh@tirin.openworld.co.uk>.
+ This was offered as a patch to 4.44.
+
+ Added support for Auspex LFS 1.8.1 and 1.9.2 to
+ SunOS 4.1.4 lsof. The support was requested by
+ Quentin Fennessy <quentin@dvorak.amd.com>, who
+ provided information and did testing.
+
+ Enabled IPv6 support code for NetBSD and OpenBSD,
+ conditional on Configure script tests. Wolfgang
+ Rupprecht <wolfgang@wsrcc.com> supplied the NetBSD
+ code and tested it. The OpenBSD code I constructed
+ has been compiled but not tested.
+
+ Updated the identd Perl 5 script, based on a report
+ from Wendy Lin <af5@taiyang.cc.purdue.edu> that
+ the space in its response line in front of the user
+ name violates RFC 1413.
+
+ Added IPv6 support to /proc-based Linux lsof.
+ Jonathan Sergent <sergent@ETLA.NET> and Andrew
+ Thomas Sydelko <sydelko@ecn.purdue.edu> kindly
+ provided a test system.
+
+ Updated man page description of AIX multiplexed
+ files to indicate that they might be /dev/ptc or
+ /dev/pts, depending on the AIX version. The
+ correction was suggested by Onno van der Linden
+ <onno@simplex.nl>.
+
+ Sylvain Robitaille <syl@alcor.concordia.ca> reports
+ lsof passes his Y2K tests.
+
+4.46 October 23, 1999
+ Corrected /proc-based Linux lsof to detect that an
+ IPv6 address is a mapped IPv4 address. The problem
+ was reported and analyzed by Arkadiusz Miskiewicz
+ <misiek@misiek.eu.org>, who also tested the fix.
+
+ Added a libc5 library /dev/kmem-based Linux lsof
+ circumvention, supplied by Jason Lingohr
+ <lingman@lucid.net.au>.
+
+ Corrected a bug in -t (terse) AIX output, reported
+ by Wendy Lin <af5@taiyang.cc.purdue.edu>. I
+ introduced the bug at revision 4.43 when adding
+ FILE_FLAG reporting. This was offered as a patch
+ to 4.45.
+
+ Added a work-around for a problem in the OpenBSD
+ 2.3 <sys/pipe.h> header file. Volker Borchert
+ <bt@teknon.de> provided and tested it.
+
+ Improved description of cross-building lsof for a
+ 64 bit Solaris 7 system on a 32 bit system with
+ suggestions from Phillip Edwards
+ <Philip.Edwards@sn.wpafb.af.mil>.
+
+ Fixed a gawk POSIX-mode pattern error in the Linux
+ /dev/kmem-based Mksrc script, based on a tip from
+ Ambrose C. Li <acli@mingpaoxpress.com>.
+
+ Fixed a bug in the Tru64 UNIX IPv6 handling, courtesy
+ of a report from Casper Dik <casper@holland.sun.com>.
+
+ Enabled support for OpenBSD 2.6.
+
+ Enabled support for BSDI BSD/OS 4.1, based on a
+ report from Jeffrey C Honig <jch@bsdi.com> that
+ only a Configure script change is necessary.
+
+ Enabled Configure script to use gcc for building
+ lsof for a 64 bit Solaris 7 and 8 kernels, if the
+ gcc version is 2.95 or above.
+
+ Improved -i option handling for systems with IPv6
+ support so that it will search for a host name in
+ both IPv4 and IPv6 families, when that is possible.
+ As a companion modification, changed -V processing
+ to report a single error when a multiple host name
+ match is requested. Casper Dik <casper@holland.Sun.COM>
+ helped test.
+
+ Fixed a DEC OSF/1, Digital UNIX, Tru64 UNIX repeat
+ mode bug, reported by Mayer Ilovitz <mayer@cooper.edu>.
+ Mayer helped test the fix. The fix was offered as a
+ patch to 4.45.
+
+ Changed Solaris socket file recognition scheme, so it
+ is (nearly) the same through Solaris 8, where the
+ previous clone device scheme no longer works.
+
+ With significant assistance from Casper Dik, added
+ support for Solaris 8 Beta and Beta refresh. The
+ IPv6 support in Solaris 8 is still in some flux,
+ so there are temporary compensations for the
+ differences between Beta IPv6 support and Beta
+ refresh IPv6 support. Casper and I hope those
+ differences disappear by FCS.
+
+ Improved the delivery of information on Solaris
+ 2.5.1, 2.6, 7, and 8 door files.
+
+ Fixed a repeat mode bug that surfaces when /etc/passwd
+ changes between cycles. The bug report and diagnostic
+ help were supplied by Igor Schein <igor@txc.com>.
+ The fix was offered as a patch to 4.45.
+
+ Added support for INRIA IPv6 to NetBSD. Jean-Luc
+ Richier <Jean-Luc.Richier@imag.fr> provided patches
+ and a test system on which to verify them.
+
+ Added support for AIX 4.3.3. Jeff W. Stewart
+ <jws@anaconda.cc.purdue.edu> provided a test system.
+
+ Made adjustments for FreeBSD 4.0-current.
+
+ Improved reporting of information for AIX sockets that
+ lack protocol control blocks.
+
+4.47 November 29, 1999
+ Based on a query from Jean-Pierre Radley <jpr@jpr.com>,
+ changed the lsof top-level Makefile to propagate
+ CFGF to the library Makefile. (DEBUG was already
+ being propagated.) Added osrgcc and scogcc Configure
+ abbreviations (to use gcc) for Jean-Pierre.
+
+ In response to a query from Igor Schein <igor@txc.com>,
+ improved the Configure script test for Solaris 7
+ and 8 that decides if the compiler can produce 64
+ bit executables.
+
+ Made an ugly hack, based on making a private rnode
+ structure definition from q4 output, to compensate
+ for HP-UX 10.20 and lower recent NFS3 patches. HP
+ didn't supply an updated <nfs/rnode.h> with the
+ patches. The problem was reported by Will Partain
+ <partain@mekb2.sps.mot.com>. Elias Halldor Agustsson
+ <elias@hi.is> helped identify the patches as
+ PHNE_18173, PHNE_19426, PHNE_19937, and PHNE_20091,
+ and provided a test system.
+
+ Switched BSDI test system from 2.1 and 3.1 to 4.0.1,
+ courtesy of Terry Kennedy <terry@tmk.com>.
+
+ Added some more dev_t hacks for Alpha FreeBSD 4.0.
+
+ Added support for IPv6 on BSD 4.x. The support hasn't
+ yet been tested, just compiled.
+
+ Added support for the mnt file system (mntfs or
+ /etc/mnttab) on Solaris 8. Tested on Solaris 8
+ BETA-Refresh.
+
+ Made selection of optional fields (e.g., PPID with
+ -FR) in a field output specification select the
+ optional field, too, so that the option selector
+ for the field (e.g., -R) isn't also required. This
+ change was made in response to an inquiry from John
+ DuBois <spcecdt@armory.com>. This may require some
+ revision to scripts that parse all field output;
+ two scripts in the lsof distribution's scripts/
+ subdirectory had to be updated.
+
+ Corrected handling of Linux IPv4 addresses mapped
+ in IPv6 addresses.
+
+ Tested under OpenBSD 2.6.
+
+4.48 January 14, 2000
+ Modified -i argument processing of colon-separated
+ IPv6 addresses to recognize an IPv4 address mapped
+ in an IPv6 address and handle it as an IPv4 address.
+ This was offered as a patch to 4.47.
+
+ Added a defined symbol (NOWARNBLKDEV) to control
+ (inhibit) the issuance of a warning when no block
+ devices are found. This was done anticipating its
+ need in FreeBSD 4.x, but that dialect version no
+ longer has any block devices, so HASBLKDEV was
+ disabled for it instead. NOWARNBLKDEV was left in
+ place for possible use in the future.
+
+ Enabled KAME IPv6 Configure support for FreeBSD
+ when <netinet6/in6.h> is found.
+
+ Disabled use of gcc to compile lsof for 64 bit
+ HP-UX 11.
+
+ Updated Configure to recognized FreeBSD 3.4.
+
+ Based on suggestions from Bernt Christandl
+ <beb@MPA-Garching.MPG.DE> improved AFS configuration
+ for AIX and Solaris, and updated AIX AFS 3.5 support.
+ Johannes Tax <tax@bluedog.oit.unc.edu>, Hung T.
+ Pham <hung_pham@unc.edu>, and Curt Freeland
+ <curt@grumpy.cse.nd.edu> provided test systems.
+
+ Updated lsof's private rnode definition for AIX
+ 4.3.3, since IBM still doesn't ship the
+ <oncplus/nfs/rnode.h> header file and the rnode
+ structure definition in <nfs/rnode.h> doesn't match
+ what the kernel uses. This was offered as a patch
+ to 4.47.
+
+ Weakened the test in the Linux /proc-based lsof of
+ the field count of data lines in /proc/net/{tcp,udp}.
+ It appears that recent 2.3.x Linux kernels have
+ added untitled fields to these files. The bug
+ report came from Gabor Liptak <gaborliptak@usa.net>.
+
+ Adjusted for a FreeBSD 4.0 change in the definition
+ of [_]KERNEL. David O'Brien <obrien@NUXI.com> reported
+ the problem and provided a test system.
+
+ Removed the HASPPID bracket from Fppid (the -R
+ option state variable) so that the field select
+ table will compile even when HASPPID is not defined.
+ This problem was introduced at revision 4.47 with
+ code that causes some field output characters to
+ set option states. The problem was reported by
+ David Bacon <bacon@birch.eecs.lehigh.edu>.
+
+4.49 April 3, 2000
+ Made clearer in man page that "Lxx" FDs are AIX
+ loader table references. Also updated the 00FAQ
+ discussion of the Stale Segment ID bug to include
+ AIX 4.3.x.
+
+ Modified support for NetBSD 1.4Q to include the
+ <sys/buf.h> header file to cope with an MFS change.
+
+ Added support for OpenBSD UVM virtual memory.
+
+ Added support for AIX systems with > 2GB of memory.
+ Chris Sylvain <csylvain@itg.ummc.umaryland.edu>
+ reported the problem and provided the solution.
+ Chris also supplied some minor code cleanup. This
+ was offered as a patch to 4.48.
+
+ Based on new information from Igor Schein <igor@txc.com>
+ made additional compensation in Configure script
+ for 64 bit Solaris 7 and 8 gcc.
+
+ Added some 00FAQ info on the effect ordering of
+ the +fg and -FG options has on output format.
+
+ Improved NetBSD IPv6 configuration, based on a
+ suggestion from Thomas Klausner
+ <wiz@danbala.ifoer.tuwien.ac.at>. Added code to
+ convert IPv4-mapped-in-IPv6 addresses to IPv4
+ addresses.
+
+ Updated the information in 00FAQ and the HP-UX 11
+ binary directory README files on the HP-UX 11 ipis_s
+ patch with new information supplied by Eric McWhorter
+ <emcwhorter@xsis.xerox.com>.
+
+ Added documentation on changes to HASFSTYPE and
+ HASNCACHE, and the new HASPRIVPRIPP.
+
+ Adjusted Configure for FreeBSD 5.0. Made additional,
+ necessary changes to Configure and the BSDI sources
+ to eliminate load errors.
+
+ Added KAME IPv6 support to FreeBSD at the request
+ of Ollivier Robert <roberto@eurocontrol.fr>, who
+ provided a test system.
+
+ Corrected the script that generates the CHECKSUMS
+ files for binaries to correctly name the detached
+ PGP certificate. The documentation bug was reported
+ by Michael Hennecke <hennecke@rz.uni-karlsruhe.de>.
+
+4.50 June 29, 2000
+ Added a NetBSD alpha test host, courtesy of Ray
+ Phillips <r.phillips@mailbox.uq.edu.au>. An lsof
+ 4.49 binary, built on Ray's 1.4.1 system was made
+ available prior to the 3.50 release.
+
+ Upgraded the system map file tests in /dev/kmem-based
+ Linux lsof, making the use of DEBIAN_LINUX_LSOF
+ unnecessary. Tested the changes on a system made
+ available by Vincent Kujala <kujala@geog.ubc.ca>
+ and Jim Mintha <jim@ic.uva.nl>.
+
+ Forced AIX to use the large-file-enabled versions
+ of lstat (lstat64) and stat (stat64) if <sys/stat.h>
+ contains stat64. This should allow lsof to stat()
+ AIX files > 2GB even when the builder has not
+ defined the "large file enabled programming
+ environment." Configure tests <sys/stat.h> and
+ puts -DHASSTAT64 in the Makefile's CFLAGS to make
+ this happen. Fernando A.B. Whitaker
+ <whitaker@cenapad.unicamp.br> reported the problem.
+ This was offered as a patch to 4.48.
+
+ Enabled Configure script to handle OpenBSD 2.7.
+ Angelos D. Keromytis <angelos@dsl.cis.upenn.edu>
+ reported the availability of OpenBSD 2.7 and supplied
+ the Configure script patch.
+
+ Improved handling of DOOR and fattach()'d files in
+ Solaris.
+
+ Changed message about missing kernel symbol file
+ from "not yet determined" to "none found".
+
+ Updated FreeBSD, NetBSD, NEXTSTEP, OpenBSD, and
+ OPENSTEP support to report "no PCB" and the values
+ of the SO_CANTSENDMORE and SO_CANTRCVMORE state
+ flags when a socket structure has no inpcb pointer.
+ This modification was made to AIX lsof at revision
+ 4.46. Added an entry to 00FAQ about sockets that
+ have no inpcb pointer.
+
+ Upgraded support for FreeBSD 5.0-CURRENT. Ben
+ Smithurst <ben@scientia.demon.co.uk> supplied
+ patches and did testing. David O'Brien <obrien@NUXI.com>
+ supplied a test system. The update included dropping
+ the Fctty part of file descriptor file system
+ support, conditional on a Configure script test.
+ I propagated those changes to BSDI, NetBSD, and
+ OpenBSD in anticipation of their having the
+ modification in the future. David also arranged
+ with Michael Haro <mharo@area51.fremont.ca.us> for
+ a FreeBSD 3.4 test system.
+
+ In response to an lsof 3.72 bug report from Jim Mewes
+ <jim@corp.phone.com>, added more kernel address
+ filtering to the lsof function, kread(), that reads
+ Solaris kernel data.
+
+ In response to a report from Marc Duponcheel
+ <marc@offline.be>, added tests to the /proc-based
+ Linux lsof to ignore file systems of types "autofs"
+ and "pipfs".
+
+ Based on a report and information supplied by Casper
+ Dik <casper@holland.Sun.COM>, updated the ncache_load()
+ function in lib/rnch.c with new code that deals
+ with a post Solaris 8 change in kernel name cache
+ (DNLC) handling. Casper tested the update, which
+ should be invisible to Solaris versions without
+ the new DNLC code.
+
+ Added support for Solaris VxFS QIO files, based on
+ a report from Kieran Broadfoot <kieran.broadfoot@gs.com>.
+ Kieran help test the support.
+
+ Added support for PTX 4.4.6 and 4.5[.1] with help
+ from the usual cast of good people at Sequent.
+
+ Added support for 64 bit file sizes and offsets on
+ BSDI, FreeBSD, NetBSD, and OpenBSD, based on a
+ report from Dan Nelson <dnelson@emsphone.com>.
+ Dan supplied a patch and did FreeBSD testing.
+
+ Added Configure script recognition of NetBSD 1.5,
+ based on a report from Andrew Brown <atatat@atatdot.net>.
+ Thomas Klausner <wiz@danbala.ifoer.tuwien.ac.at> updated
+ the NetBSD port package to use a pre-release of this
+ addition.
+
+ At the last minute saw a notice via deja.com's
+ UseNet search service that FreeBSD 3.5 had been
+ released and lsof didn't grok it. Added recognition
+ of 3.5 to lsof's Configure script, but didn't have
+ the opportunity to test lsof on 3.5.
+
+4.51 August 21, 2000
+ Added Configure script support for the upcoming
+ Solaris 9 release based on suggestions from Casper
+ Dik <Casper.Dik@holland.sun.com>.
+
+ Changed sample Perl scripts to assume that
+ /usr/local/bin/perl is Perl 5 and Perl 4 may be
+ found in /usr/local/bin/perl4.
+
+ Updated Configure to recognize FreeBSD 4.1 and made
+ a FreeBSD pre-release distribution available.
+
+ Bela Lubkin <belal@sco.COM> tested lsof on the
+ upcoming SCO OSR 5.0.6 release and reports that
+ lsof appears to work properly.
+
+ Updated the AIX compiler test in Configure to
+ recognize its version 5.
+
+ Updated AIX 4.3.3 support with automatic recognition
+ of the proper rnode structure, based on machine
+ bit width. Also added code to detect when processing
+ the -X option that lsof has been compiled with the
+ "other" AIX 4.3.3 user structure and to apply
+ compensations. When a compensation method works,
+ it's applied during subsequent -X processing; when
+ none works, further -X processing is disabled.
+
+ Added Tru64 UNIX 5.1 support. Updated Tru64 UNIX
+ library text file support to recognize new kernel
+ support for AdvFS library files. Berkley Shands
+ <berkley@cs.wustl.edu> and Klaus Saggerer USG
+ [saggerer@zk3.dec.com> helped put me in contact
+ with Chang Song <song@zk3.dec.com>, the developer
+ of 5.1's new kernel name cache and he helped me
+ develop new code in lsof to access it.
+
+ Corrected reporting of PTX fattach()'d address.
+
+ Changed Configure and dlsof.h for NetBSD and OpenBSD
+ to use /usr/include/uvm header files when available.
+ Andrew Brown <atatat@atatdot.net>, Thomas Klausner
+ <wiz@danbala.ifoer.tuwien.ac.at>, and Wolfgang
+ Rupprecht <wolfgang@wsrcc.com> pointed out the need
+ to do this for NetBSD. Andrew provided access to
+ a NetBSD 1.5 system for verifying the changes.
+
+ Installed snprintf() support, including a private
+ version in the lsof library for those UNIX dialects
+ without the function. Changed all sources to use
+ it instead of sprintf() and strcpy().
+
+ Fixed a memory leak in the readvfs() functions of
+ BSDI, DEC/OSF1, Digital UNIX, FreeBSD, NetBSD,
+ OpenBSD, and Tru64 UNIX.
+
+ Tested on Linux 2.4.
+
+ Modified the Pyramid MkKernOpts script to compensate
+ for `uname -s` configuration alternatives. Robert
+ Dahlem <Robert.Dahlem@ffm2.siemens.de> supplied
+ the modification.
+
+ Obtained access to an FCS Solaris 8 64 bit system
+ and built lsof on it, using Sun Workshop C 5.0 and
+ gcc 2.96 20000814 (experimental). Both compilers
+ produce a working lsof.
+
+
+4.52 November 8, 2000
+ Completed work on an HP-UX 11.11 port that uses a
+ pstat(2) interface provided by HP. To distinguish
+ it from its predecessors for HP-UX, this lsof
+ version is called PSTAT-based and the predecessor
+ versions are now called /dev/kmem-based. I am
+ indebted to the far-sightedness and support of
+ these good people at HP for making PSTAT-based lsof
+ possible: Carl Davidson, Louis Huemiller, Rich
+ Rauenzahn, and Sailu Yallapragada. The PSTAT-based
+ sources are in lsof_4.52/dialects/hpux/pstat, the
+ /dev/kmem-based ones in lsof_4.52/dialects/hpux/kmem.
+
+ Ported to IBM Monterey for Merced|Itanium, aka AIX
+ 5L. It configures via the Configure script's "aix"
+ abbreviation and has been tested on AIX 5L Beta 3.
+ Jay Beck, Steve Dibbell, Loc Le, Nasser Momtaheni,
+ and Malcom Zung of IBM provided generous support.
+ Since AIX 5L is still in Beta testing, this port
+ can't be considered complete.
+
+ Added Configure support for OpenBSD 2.8. David
+ Mazieres <dm@cs.nyu.edu> provided a test system.
+
+ Based on a report from Marc Christensen
+ <marc@mecworks.com> added sockfs to the mount scan
+ exemption list for /proc-based Linux lsof.
+
+ Added large file, CDFS, and DOSFS for UnixWare 7.x.
+ Added UnixWare device memory mapping support. All
+ UnixWare changes were supplied by Eric Dumazet
+ <edumazet@cosmosbay.com> Eric also supplied some
+ miscellaneous bug fixes.
+
+ Deferred name cache loading until printname() needs
+ to use the name cache.
+
+ Terminated Pyramid, SunOS 4.1.x, and Ultrix support,
+ because test systems are no longer available.
+ Final Pyramid and Ultrix source code distributions
+ for lsof revision 4.51 may be found on lsof.itap.purdue.edu
+ in pub/tools/unix/lsof/OLD/src. The no longer
+ supported SunOS 4.1.x source code is still distributed
+ with the Solaris source code.
+
+ Added code to set Solaris node address to real vnode
+ address, when applicable.
+
+ John Speno <speno@lopan.isc-net.upenn.edu> provided
+ information that enabled me to update the Tru64
+ AdvFS (MSFS) node definition for AdvFS version 5.
+
+ Added Tru64 5.x CFS support with help from Kris
+ Chandrasekhar <Kris.Chandrasekhar@compaq.com>,
+ Diane Lebel <lebel@zk3.dec.com>, and John Speno.
+ The support only provides information about cached
+ file attributes.
+
+ Installed a Configure patch for HP-UX 11 supplied by
+ Kenneth Stailey <kstailey@disclosure.com> that adds
+ another command to q4 input.
+
+ Tested on FreeBSD 4.2.
+
+ Will Day <willday@rom.oit.gatech.edu> and Frank
+ Winkler <frank.winkler@germany.sun.com> graciously
+ supplied Solaris 8 binaries.
+
+ Added Solaris 9 text file support, supplied by
+ Casper Dik <Casper.Dik@holland.sun.com>.
+
+4.53 December 6, 2000
+ Added the AIX 5L j2_lock.h to the distribution with
+ a Configure script step to use it when it's missing
+ from /usr/include/j2.
+
+ Removed SunOS 4.1.x support.
+
+ Removed Linux 2.0.x /dev/kmem support.
+
+ Fixed VBLK and VCHR special device file reporting
+ to handle /dev information more accurately.
+
+ Added a Apple Darwin / Mac OS X 1.2 port, provided
+ by Allan Nathanson <ajn@apple.com>. Allan also
+ arranged for a test system so I can maintain this
+ port. An additional test system was provided by
+ Dale Talcott.
+
+ Dropped claims of support for all UnixWare versions
+ except 7.1.0, since that is the only version on
+ which I can test lsof. Even though lsof 4.53 is
+ deprecated for UnixWare 2.1.3, installed a patch
+ for it with testing done by A. Channing Clark
+ <clark.channing@heb.com>.
+
+ Dropped claims of support for all SCO OpenServer
+ versions except 5.0.5, since that is the only
+ version on which I can test lsof.
+
+4.54 January 19, 2001
+ Added compensation for a change that made the
+ FreeBSD mount structure invisible. I can only test
+ back to 3.2 and the compensation works there, so
+ it's been #ifdef'd for 3.2 and above. David O'Brien
+ <obrien@FreeBSD.org> provided the necessary clue.
+
+ Based on a report from Valdis Kletnieks
+ <Valdis.Kletnieks@vt.edu>, changed all IPv6 support
+ to report a TYPE of IPv6 for sockets with IPv4
+ addresses mapped in IPv6 addresses. The previous
+ lsof behavior was to report their TYPE as IPv4.
+
+ Restored the Linux GlibC test to Configure, removed
+ at revision 4.53, based on a report from John Dzubera
+ <zube@cs.colostate.edu>, that RedHat Linux 6.0 still
+ needs the test.
+
+ Made setting of link count for Solaris more selective.
+
+ Limited Readlink() recursion to MAXSYMLINKS. The bug
+ was reported by Jan Dvorak <johnydog@go.cz>.
+
+ Dropped the *claim* that lsof runs on Solaris 2.5.1.
+ It may well do so, but I no longer have access to a
+ test system.
+
+ Fixed an #endif comment typo, reported by Igor Schein.
+
+ Fixed a typo in a cast for a Tru64 UNIX 5.1 function
+ and updated Configure for Tru64 UNIX 5.0 and 5.1 with
+ information from Jesse Perry <jesse.perry@compaq.com>.
+
+ Corrected non-fatal typos in the AdvFS support in
+ dnode.c for Tru64 UNIX.
+
+ Added msdos file system support for NetBSD and OpenBSD.
+ Andrew Brown <atatat@atatdot.net> requested and helped
+ test it.
+
+4.55 February 15, 2001
+ Based on a report from Bernd Eckenfels <ecki@lina.inka.de>
+ added support in lsof for files in /proc/<PID>/maps
+ that have been deleted.
+
+ Changed PGRP output title to PGID, conforming to
+ the most common current abbreviation for Process
+ Group ID (PGID). While some systems continue to
+ use *pgrp for internal kernel variable names, most
+ systems that support the display of PGID via ps(1)
+ now title it PGID. The lsof -g and -Fg options
+ operations are unchanged in function; only titles
+ and descriptions have changed. Also changed internal
+ variable names from *PGRP and *pgrp to *PGID and
+ *pgid where possible.
+
+ Dropped the *claim* that lsof runs on HP-UX 9.x.
+ It may well do so, but I no longer have access to
+ a test system.
+
+ In response to a suggestion from Jeff Howie
+ <jeff.howie@federated.ca> added support for command
+ name selection by regular expression. A new form
+ of the -c option value is use to identify and
+ specify a regular expression.
+
+ Restore the *claim* that lsof works on UnixWare
+ 7.0, since I re-acquired a test system.
+
+4.56 May 3, 2001
+ Corrected some problems Amir Katz <Amir_Katz@bmc.com>
+ found with Insure++, one in lib/dvch.c, the rest
+ in Solaris sources. Amir's report also helped me
+ find an error in an snpf() call that caused (the
+ unsupported) Solaris 2.5.1 lsof to crash. Wally
+ Winzer, Jr. <wally.winzer@ChampUSA.COM> helped test.
+
+ Added support for UnixWare 7.1.1 and above in-kernel
+ UNIX sockets. John Hughes <john@Calva.COM> kindly
+ provided code and access to a test system. John
+ also provided a test system and advice for adding
+ UnixWare 7.1.1 NonStop Cluster and CFS support.
+ More help with that effort came from Kurt Gollhardt
+ (SCO), Barbara Howe (SCO), Bela Lubkin (SCO), and
+ Dewan Rashid <Dewan.Rashid@ir.com>.
+
+ Archived a set of compilation hints (patches) from
+ Bill Melvin <Bill.Melvin@esc.edu> that make it
+ possible to compile the old, unsupported lsof 3.08
+ sources on UnixWare 1.x without NFS or CDFS support.
+
+ Installed support supplied by Allan Nathanson
+ <ajn@apple.com> for the Darwin "Gold Master" release,
+ Mac OS X 10.0 (aka Darwin 1.3 in its public source
+ version). Added Allan's CVS repository suggestions
+ to the script that gets additional header files
+ from an open source repository.
+
+ Tested an HP-UX 11.11 kernel patch from Sailu
+ Yallapragada that enables reporting of TCP/IP
+ information for telnetd processes that use the
+ telnet multiplexor. I don't yet know the kernel
+ patch ID.
+
+ Made the Solaris inclusion of <inet/mi.h> conditional
+ on the Solaris version. (It's apparently not needed
+ at 2.6 and above.) Bill Watson <bill.watson@uk.sun.com>
+ brought this to my attention.
+
+ Added alternate Linux 2.4.x lock extent test, supplied
+ by Jim Mintha <jim@ic.uva.nl>.
+
+ Rearranged the lines and pre-processor tests in
+ regex.h, lib/regex.c, and lib/snpf.c so that unifdef
+ can be used to eliminate copyright and GPL statements
+ when the files aren't being used for a particular
+ dialect. (USE_LIB_* definitions in a dialect's
+ machine.h header file determine if one or more of
+ those three files are to be used.)
+
+ Added preliminary support for Solaris 8 with VxFS
+ 3.4. This support will be refined as I get
+ information from Veritas about how they will
+ distribute the kernel header files lsof needs.
+ Those header files were omitted from the standard
+ VxFS 3.4 distribution. Technical assistance and
+ testing were provided by Calle Dybedahl <cdy@algonet.se>,
+ Gary Millen <gary.millen@veritas.com>, Rainer Orth
+ <ro@TechFak.Uni-Bielefeld.DE>, Peter C. Vernam
+ <pvernam@draper.com>, and Donna Yobs
+ <Donna.Yobs@veritas.com>
+
+ Tested on FreeBSD 4.3-STABLE.
+
+ Dropped the *claim* that lsof works on UNIX dialects
+ where I no longer have test systems: BSDI 2.1,
+ 3.[01] and 4.0; DEC OSF/1, Digital UNIX and True
+ 64 UNIX 2.0 and 3.2; FreeBSD 2.1.[67], 2.2[.x],
+ 3.[012345] and 4.[01]; HP-UX 10.20; NetBSD 1.[234];
+ SCO OpenServer 5.0.5; and SCO UnixWare 7.0
+
+ Tested on Solaris 9 BETA, s81_36.
+
+4.57 July 19, 2001
+ Help (-h) and version (-v) output now have URLs
+ for the newly created and timeliest lsof FAQ
+ (00FAQ in the lsof distribution) at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/FAQ
+
+ and the man page for the current lsof distribution
+ at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/lsof_man
+
+ Based on a report from Steve Laubscher
+ <slaubs@woodward.com>, modified dlsof.h for PTX
+ 4.6[.1] to avoid a temporary dnlc_t definition
+ needed at PTX 4.5.1.
+
+ Corrected test for old Linux kernels in Configure.
+ Henri Karrenbeld <ishtar@cal044202.student.utwente.nl>
+ brought the error to my attention. Limited Linux
+ claims to 2.1.72 and above in the documentation.
+
+ Improved HP-UX 11 Configure stanza and stream socket
+ handling.
+
+ Constructed a work-around for the HP-UX 11 optional
+ OnlineJFS package. The work-around sadly requires
+ lsof to have a private version of the vx_inode
+ structure, since the OnlineJFS package doesn't
+ update <sys/fs/vx_inode.h>. Troyan Krastev
+ <Troyan.Krastev@ricoh-usa.com> brought the bug to
+ my attention and Michael Bracewell
+ <michael@ra.TSS.PeachNet.EDU> provided a test system
+ where I developed the work-around.
+
+ Added locale support to lsof's isprint() test,
+ based on a suggestion from Dan Mercer <damercer@mmm.com>.
+ Lsof will use setlocale(), when that function and
+ its supporting <locale.h> header file are available.
+
+ Added OpenBSD 2.9 support.
+
+ Based on a report from Aaron Rhodes <arhodes@psionic.com>
+ and with testing help from Aaron, made the lsof
+ 4.56 revision compile and work on OpenBSD 2.6.
+ While that OpenBSD version is no longer supported,
+ Aaron's report exposed a Configure script bug
+ affecting OpenBSD versions lsof does support.
+
+ Updated for FreeBSD 5.0-CURRENT. Szilveszter Adam
+ <sziszi@petra.hos.u-szeged.hu> help test. The lsof
+ FreeBSD ports packager, David O'Brien <obrien@FreeBSD.org>,
+ assisted.
+
+ Tested on AIX 5.1. Loc Le and Nasser Momtaheni of
+ IBM provided test systems.
+
+4.58 September 13, 2001
+ Added options to safestrprt() and safestrprtn() to
+ surround the string with '"' and to suppress the
+ printing of an ending '\n'. Use of these functions
+ in device cache file error message reporting answers
+ a suggestion for better error reporting from John
+ Jackson <jrj@purdue.edu>.
+
+ Fixed a Solaris 2.6 and above problem related to
+ searching for "large" (O_LARGEFILE) files by name;
+ lsof was using the wrong version of [l]stat(2).
+ The bug was reported by Daniel Trinkle
+ <trinkle@cs.purdue.edu>.
+
+ Added AIX 4.1.4 and above XTI socket support.
+
+ Added OSR Xenix Shared Data and Semaphore file type
+ support with modifications supplied by Bela Lubkin.
+
+ Updated OPENSTEP support with modifications from Carl
+ E. Lindberg <lindberg@clindberg.org>. The changes
+ enable the correct reporting of executable and
+ library open files ("txt" type).
+
+ Limited claims of OpenServer support to the versions
+ where I currently test, 5.0.4 and 5.0.6. (Lsof
+ probably works on 5.0.5.)
+
+ Enabled processing of -C option for PSTAT-based HP-UX
+ lsof.
+
+ Enabled and tested on FreeBSD 4.4.
+
+ Corrected a file system test example in 00QUICKSTART,
+ based on a report from Jun Biao WANG <wangjunb@cn.ibm.com>.
+
+ Made available for re-distribution a user-contributed
+ port of lsof 4.51 to Reliant UNIX 5.45. Thomas
+ Mauterer <Thomas.Mauterer@philosys.de> contributed
+ the port.
+
+4.59 October 20, 2001
+ With the closing of the Sequent Synergy Links Lab
+ by IBM, terminated lsof support for PTX. The last
+ tested PTX lsof revision, 4.58, is available on
+ lsof.itap.purdue.edu in .../lsof/OLD/src.
+
+ Adjusted for FreeBSD 5.0-CURRENT NFS header file
+ changes, based on a report from Jos Backus
+ <josb@cncdsl.com>.
+
+ Corrected a bug in the way Linux lsof identifies
+ the owner of a process. Lionel Cons <lionel.cons@cern.ch>
+ reported the problem and tested the fix. Added
+ code to avoid stat(2) calls on regular Linux files
+ whenever possible. Lionel reported the need to do
+ this (AFS files) and tested the new code.
+
+ Added new output field for raw device number in
+ hex. The field is identified with 'r'. This field
+ is NOT selected when -F or -F0 is specified so that
+ its appearance won't disturb existing scripts that
+ process field output.
+
+ Added support for OpenUNIX 8. A test system was
+ provided by Larry Rosenman <ler@lerctr.org>.
+ Matthew Thurmaier <matt@compclass.com> and many
+ people from Caldera provided technical assistance.
+
+ Added an additional UVM test to the NetBSD Configure
+ stanza. Andrew Brown <atatat@atatdot.net> supplied
+ the test; it recognizes NetBSD 1.5Y UVM changes to
+ the vnode structure recently committed by Chuck
+ Silvers.
+
+ Applied Configure and get-xnu-headers.sh script
+ changes suppled by Allan Nathanson <ajn@apple.com>
+ for Darwin 1.4.
+
+ Added for Bela Lubkin <belal@mammoth.ca.caldera.com>
+ OSR-specific environment variables to supply values
+ to the Configure script. The variables are described
+ in 00XCONFIG.
+
+ Added an IP version selector to the -i option
+ parameters.
+
+4.60 November 9, 2001
+ Added special handling to and corrected bugs in
+ the matching of IPv4 in IPv6 addresses to -i6:<...>
+ selectors.
+
+ Made 00FAQ corrections and updates, based on
+ discussions with Igor Schein <igor@txc.com>.
+
+ Modified Configure script to detect a 64 bit capable
+ gcc compiler and permit it to be used to build 64
+ bit (PA-RISC 2) lsof for HP-UX 11.00. Tested with
+ HP's gcc package, which Rich Rauenzahn of HP kindly
+ installed on a test system at HP. Stefan Marquardt
+ <stefan.marquardt@hagebau.de> helped test.
+
+ Made lsof's method of killing its child process
+ more robust, based on a suggestion from Bela Lubkin
+ <belal@caldera.com>.
+
+ Modified all dialect Makefile segments to accept
+ select -v #define's from the environment -- a
+ builder's comment, host, logname, system information
+ and user name. This was done for Bela Lubkin, so
+ he can "tune" the -v output when he packages lsof
+ in the upcoming Caldera OSR 5.0.7 release.
+
+ Changed Perl scripts in scripts/ to put the lsof
+ path consistently in $LSOF. Also added a fix from
+ Bela Lubkin to scripts/big_brother.perl5 that allows
+ it to tolerate SCO OSR "ago" clauses in open UDP
+ file information. Strengthened emphasis in
+ scripts/00README that the scripts are examples that
+ shouldn't be expected to run on all UNIX dialects
+ without modification.
+
+ At Bela Lubkin's suggestion changed the device
+ cache file format examples in 00DCACHE and 00FAQ
+ to avoid "%U%". That's an SCCS escape sequence.
+
+ Added support for OpenBSD 3.0.
+
+ Added +DAportable to CFLAGS for 32 bit HP-UX 11.
+ Amir Katz <Amir_Katz@bmc.com> suggested the addition.
+
+4.61 January 22, 2002
+ Updated field output example Perl scripts in the
+ scripts/ subdirectory to discover the lsof path,
+ starting at .. and proceeding through the PATH
+ environment variable's directories.
+
+ Added minor OSR Configure script fixes, provided
+ by Bela Lubkin <belal@caldera.com>.
+
+ In response to a report from Joshua Wright
+ <Joshua.Wright@jwu.edu> modified NetBSD and OpenBSD
+ Configure stanzas and sources so that lsof can be
+ built when there is no system source tree (e.g.,
+ /usr/src/sys).
+
+ In response to a report from Peter Valchev
+ <pvalchev@openbsd.org> improved the UVM test in
+ the OpenBSD Configure stanza.
+
+ Updated Configure script to recognize FreeBSD 4.5.
+ Updated for FreeBSD 5.0 procfs and pseudofs changes.
+
+ Updated HP-UX stanza to see if the compiler named
+ in the LSOF_CC environment variable is the bundled
+ compiler. If it is, "-O" is omitted from the
+ compiler flags.
+
+ Updated Digital UNIX 4.x and Tru64 UNIX error message
+ related to kernel name list failures. Added an FAQ
+ section about how a kloadsrv daemon failure can cause
+ knlist(3) to fail. The condition was reported by
+ Douglas B. Jones <douglas@gpc.peachnet.edu>
+
+ Based on a report from Mark W. Eichin <eichin@thok.org>
+ made Linux lsof capable of handling and reporting
+ file sizes greater than 32 bits.
+
+ Tested on Solaris 9 BETA-Refresh.
+
+ Corrected a bug in the matching of IPv4 addresses,
+ mapped in IPv6 addresses, to an IPv4 parameter to
+ an -i option.
+
+ Ported to 64 bit Power AIX 5.1 kernel with advice
+ from David Clissold <cliss@austin.ibm.com> and Marc
+ Stephenson <marc@austin.ibm.com>, and on a test
+ system provided by Loc Le <lple@us.ibm.com>.
+
+4.62 March 7, 2002
+ Updated 00README to reflect the usefulness of gcc
+ for building AIX lsof. Documented a report from
+ Brian L. Gentry <BGentry@nationsrent.com> of success
+ on AIX 4.3.3. I documented my success on 32 bit
+ Power AIX 5.1 and my lack of success on ia64 AIX
+ 5.1 and 64 bit Power AIX 5.1.
+
+ Improved UnixWare >=7.1.1 reporting of UNIX socket
+ NAME field information for NonStop Cluster systems
+ with a patch provided by John Hughes <john@Calva.COM>.
+ Offered John's improvement as a patch to lsof 4.61.
+
+ Corrected bugs in handling of open files on block
+ devices by OSR lsof. The bugs were reported by
+ Bela Lubkin <filbo@deepthought.armory.com>.
+
+ Fixed bug in writing >32 bit device numbers for
+ block devices to the device cache file.
+
+ Added support for reporting block special nodes
+ not in /dev (or /devices). That required "like
+ device special" be changed to "like block special"
+ and "like character special". (00FAQ was updated.)
+
+ Based on a report from Peter Valchev <pvalchev@openbsd.org>
+ improved the definition of the source for NetBSD
+ and OpenBSD kernel symbols (the nlist() source
+ file). NetBSD now defaults to getbootfile(3) if
+ it is available, /netbsd otherwise. OpenBSD now
+ defaults to /dev/ksyms if it is available, /bsd
+ otherwise.
+
+ Made possible compilation under BSD/OS (BSDI) 5.0
+ with changes to Configure, dialects/bsdi/dlsof,
+ dialects/bsdi/dproc.c and lib/rnmh.c. The changes
+ were suggested by Steven Hinkle <hinkle@bsdi.com>.
+ Note that these changes do not substantiate a claim
+ that lsof works on BSDI 5.0, because I haven't
+ tested it there.
+
+ Updated OpenUNIX private <sys/fs/memfs_mnode.h>,
+ based on a report from Larry Rosenman <ler@lerctr.org>
+ that it had been updated by Caldera patch OU800PK3.
+ Unfortunately the patch only corrects some of the
+ problems with the header file, so it is still
+ necessary to distribute a private patched version
+ of it with the lsof sources.
+
+ Applied a man page correction reported by Frederic
+ Delanoy <max_ok@yahoo.com>.
+
+ Corrected cast bugs related to using the HP-UX
+ bundled C compiler on HP-UX 11.11.
+
+4.63 April 23, 2002
+ Added HPUX_BOOTFILE environment variable for use
+ by the Configure script in determining HP-UX kernel
+ configuration information -- e.g., the state of
+ the ipis_s structure in the HP-UX 11 kernel. The
+ change was suggested by Marc Bejarano <beej@alum.mit.edu>.
+ Marc also suggested some changes to the HP-UX
+ section in 00FAQ that discusses Configure's use of
+ q4 for HP-UX 11.
+
+ Fixed a bug in the Solaris lsof file system matching
+ code. It was not reporting that VCHR files in
+ /devices were in / when /devices was in /, too.
+
+ Corrected bugs in device number, file size, file
+ offset, and raw device number field output generation.
+
+ Added recognition of OpenBSD 3.1 to the Configure
+ script with a suggestion from Peter Valchev
+ <pvalchev@sightly.net>. Note that this change does
+ not constitute a claim that lsof works on OpenBSD
+ 3.1, because I haven't tested it there.
+
+ Built an automated test suite. (See 00TEST and
+ the tests/ sub-directory of the lsof main directory).
+ Bela Lubkin requested it. Dale Talcott, John
+ Hughes, and Larry Rosenman helped me validate it
+ on their systems.
+
+ During the development of the test suite I discovered
+ the following lsof bugs or missing features, and
+ corrected or supplied them.
+
+ * Corrected the reporting of locks for:
+ o Digital UNIX 4.0d and Tru64 Unix 5.[01];
+ o HP-UX 10.30 and 11.00;
+ o OpenUNIX 8;
+ o UnixWare 7.1.1.
+
+ * Enabled HP-UX 10.30 and 11.00 to report open NFS
+ file link counts.
+
+ * Corrected the reporting of UNIX domain socket
+ names for Apple Darwin, FreeBSD 4.5 and above,
+ NetBSD 1.4.1 and above, and for OpenBSD 3.0 and
+ above.
+
+ * Enabled HP-UX 11.11 to stat(2) large files.
+
+ * Fixed handling of combination 32 and 64 bit
+ device numbers in AIX 64 bit architectures.
+
+ Updated the AIX 4.3.3 NFS rnode recognition code,
+ first installed at revision 4.51. It looks like
+ some IBM update has restored a single rnode structure
+ independent of the machine bit width.
+
+ Updated the NetBSD and OpenBSD sources so NetBSD
+ can process DTYPE_PIPE files, as OpenBSD was already
+ able to do.
+
+ Updated Darwin get-xnu-headers.sh script to reflect
+ information about a recent reorganization of the
+ Darwin CVS hierarchy, supplied by Allan Nathanson
+ <ajn@apple.com>.
+
+ Added defense against the standard I/O descriptor
+ attack.
+
+4.64 June 26, 2002
+ Corrected some FreeBSD pre-processor directives.
+ David O'Brien <obrien@NUXI.com> pointed them out.
+
+ Updated lsof's main() function to: 1) close all
+ open file descriptors above 2 before starting; and
+ 2) to set a non-interfering umask. Moved GET_MAX_FD
+ test from misc.c to proto.h, so that main() could
+ use it. Added multiple-include protection to
+ proto.h.
+
+ Moved FAQ's test suite Q's & A's to a more appropriate
+ section. Added a Q&A on HASSECURITY option and
+ its affect on searching for open files. (That was
+ already in the man page.)
+
+ Updated hpux/kmem/dnode.c for HP-UX < 11 compilation
+ with information from John Dzubera <Zube@CS.ColoState.EDU>.
+ While lsof doesn't support HP-UX < 11 any more, I
+ try to avoid disabling it there when possible, and
+ a locking fix for HP-UX >= 11 in lsof 4.63
+ inadvertently disabled compilation of lsof for
+ HP-UX < 11. Fixed long-standing bug in HP-UX 10.20
+ lock reporting.
+
+ Removed language from the test suite programs that
+ requires an ANSI-C compiler. This allowed the test
+ suite to be validated with cc and gcc on the un-
+ supported HP-UX 10.20.
+
+ At the suggestion of Manuel Bouyer <bouyer@antioche.eu.org>
+ switched NetBSD and OpenBSD lsof from using nlist()
+ to using kvm_nlist(). Made the same change for
+ BSDI, Darwin, and FreeBSD.
+
+ Validated test suite on OPENSTEP 4.2.
+
+ In response to a suggestion from Jeff Stoner
+ <jstoner@blackboard.com> enhanced support for the
+ FD list of the -d option to allow it to be either
+ an exclusion or inclusion list, using the '^' prefix
+ to denote exclusions.
+
+ Made adjustments for FreeBSD 4.6 and 5.0-CURRENT.
+ Fixed a FreeBSD /etc/make.conf CFLAGS extraction
+ bug, reported by Kris Kennaway <kris@obsecurity.org>,
+ and new a bug in the fix, reported by Eric Cronin
+ <ecronin@eecs.umich.edu>
+
+ Added nullfs support for FreeBSD, NetBSD, and OpenBSD
+ at the request of Andrew Brown <atatat@atatdot.net>.
+
+ Modified all readmnt() functions to ignore mounted-on
+ directory names that don't begin with '/'.
+
+ Tested on NetBSD 1.6A and OpenBSD 3.1.
+
+ Upgraded to Solaris 9 FCS with two changes to the
+ BETA-Refresh support: 1) an adjustment to dnode.c
+ for a change in the so_so (sonode) structure; and
+ 2) addition of Solaris 9 FCS specific DNLC code.
+ David Comay <David.Comay@Eng.Sun.COM> sent me the
+ dnode.c change and Casper Dik <Casper.Dik@sun.com>
+ helped with the new DNLC support code.
+
+ Applied OpenUNIX changes that permit lsof to compile
+ and run on the upcoming 8.0.1 release. The changes
+ were supplied by Robert Lipe <robertl@caldera.com>.
+ Larry Rosenman <ler@lerctr.org> provided a test
+ system.
+
+ Added Solaris fd file system support.
+
+4.65 October 10, 2002
+ Adjusted for change in FreeBSD 5.0-CURRENT inode
+ structure, reported by David O'Brien <obrien@NUXI.com>.
+ Adjusted for changes in FreeBSD 5.0-CURRENT <sys/vnode.h>.
+ One change was reported by Anders Nordby
+ <anders@FreeBSD.org>. Adjusted for FreeBSD 5.0-CURRENT
+ on sparc64 architecture.
+
+ Enhanced the error reporting of Solaris lsof when
+ it detects a kvm_open() failure, and added a 00FAQ
+ entry on the cause, based on a report from Peter
+ J. Bertoncini <pjb@anl.gov>.
+
+ Enabled compiling of lsof for NetBSD 1.5 with the
+ NULL file system, using a patch from Andrew Brown
+ <atatat@atatdot.net>.
+
+ Removed a hack in the LTbigf test program that was
+ once needed when it was compiled on Solaris 9 BETA-
+ Refresh with gcc. The hack isn't needed on Solaris
+ 9 FCS. Janet Hempstead <jan@library.carleton.ca>
+ brought the need for this change to my attention.
+
+ Applied a patch, supplied by Andrew Brown
+ <atatat@atatdot.net>, that updates lsof for NetBSD
+ version 1.6F. Corrected handling of the NetBSD
+ nullfs.
+
+ Updated to BSDI BSD/OS 4.3 on a test system kindly
+ provided by Terry Kennedy <terry@tmk.com>.
+
+ Updated to FreeBSD 4.7.
+
+ Updated to Apple Darwin 1.5, 5.x and 6.x with
+ patches supplied by Allan Nathanson <ajn@apple.com>.
+ The patches include IPv6 support.
+
+ Updated Configure to use the -bnolibpath loader
+ option when building lsof on a PowerPC, running
+ AIX 5 or greater. Valdis Kletnieks
+ <Valdis.Kletnieks@vt.edu> informed me this was
+ needed. Lsof for AIX 5.x was initially developed
+ on the IA64, where -bnolibpath can't be used and
+ I didn't think to restore it to PowerPC loads when
+ AIX 5.x became available for that architecture.
+
+ Updated to UnixWare 7.1.3 on a test system provided
+ by Larry Rosenman <ler@lerctr.org>. Removed claims
+ that lsof works on OpenUNIX 8.0.1, because UnixWare
+ 7.1.3 is the release name of OpenUNIX 8.0.1.
+
+ Based on a comment that his e-mail address was
+ wrong in the lsof distribution from Kenneth Stailey
+ <kstailey@disclosure.com>, removed all e-mail
+ addresses from lsof documentation files except this
+ one, 00DIST. The addresses in 00DIST are used to
+ send revision release notices to those who contributed
+ to a revision, but the addresses in this file for
+ previous revisions and in other documentation files
+ sometimes grow stale and are never validated.
+
+4.66 December 22, 2002
+ Acquired Solaris 7 and 8 test systems, courtesy of
+ John Dzubera <Zube@CS.ColoState.EDU>. Updated
+ 00TEST and tests/TestDB accordingly.
+
+ Clarified FreeBSD 5.0 architecture claims at the
+ suggestion of David O'Brien <obrien@NUXI.com>.
+ Also implemented David's suggestion to change
+ Intel to x86.
+
+ Installed changes to DNLC handling in OSR lsof in
+ preparation for handling changes in the OSR 5.0.7
+ DNLC cache. Information about the changes and
+ patches to handle them were supplied by Bela Lubkin
+ <filbo@deepthought.armory.com>.
+
+ Upgraded True 64 UNIX support to the 5.1B release
+ on a test system provided by Berkley Shands
+ <berkley@cse.wustl.edu> Had to used relaxed ANSI
+ compilation because of an error in a system header
+ file and other lsof source usages.
+
+ Implemented the HASNOSOCKSECURITY compile-time
+ option. When it and HASSECURITY are defined, lsof
+ will be built to list only the user's open files,
+ but will also list anyone else's open socket files,
+ provided the "-i" option selects their listing.
+ Updated the Customize script to ask about setting
+ HASNOSOCKSECURITY. Left it undefined in all dialect
+ machine.h header files. This change was requested
+ by Kenneth Stailey <kstailey@speakeasy.net> for
+ use with ntop.
+
+ Added support for OpenBSD 3.2 and its kernel trace
+ file.
+
+ Improved lsof help (-h) and version (-v) information
+ reporting.
+
+ Fixed a FreeBSD 4.7 and above off-by-two UNIX domain
+ socket path termination bug, reported by Ken Stailey
+ <kstailey@speakeasy.net>
+
+4.67 March 27, 2003
+ Began the transition of the lsof ftp server host
+ name from vic.cc.purdue.edu to lsof.itap.purdue.edu.
+ That reflects Purdue organizational changes. This
+ first step makes the new name an alias to the old
+ one. The old name, vic.cc.purdue.edu, will remain
+ usable for an extended period.
+
+ Corrected a revision number reference in section
+ 17.17 of 00FAQ on the appearance of Solaris negative
+ DNLC caching handing.
+
+ Updated 00FAQ discussion of compilers for 64 bit
+ Solaris.
+
+ Validated test suite for 64 bit Solaris 8 and gcc.
+
+ At the request of Alek O. Komarnitsky <alek@komar.org>
+ added the "+c <width>" option to enable optional
+ changing of the COMMAND column output maximum width
+ from the default to <width>. The default maximum
+ width remains CMDL, as defined in lsof.h.
+
+ Fixed three AIX kernel bit size detection bugs,
+ one in the AIX Configure script stanza, the second
+ and third in the AIX dproc.c get_kernel_access()
+ function. The bugs were reported by Pierre-Yves
+ Fontaniere <pyf@cc.in2p3.fr>, who tested the fixes.
+
+ Added kernel event queue file support for FreeBSD,
+ NetBSD and OpenBSD. Andrew Brown <atatat@atatdot.net>
+ supplied the code.
+
+ Updated to AIX 5.2 on a test system provided by
+ Dale Talcott <dtalcott@purdue.edu>. Had to build
+ work-arounds for two missing AIX 5.2 header files,
+ <j2/j2_snapshot> and <proc/proc_public.h>. Corrected
+ an off-by-one UNIX socket addressing bug. Taught
+ AIX lsof to handle both jfs and jfs2 files at the
+ same time. Adjusted for an IBM mistake in the
+ sizing of the fdsinfo structure in <procinfo.h>
+ Toshiya Nakamura <TOSHIYAN@jp.ibm.com> helped test,
+
+ Updated to FreeBSD 4.8. Corrected another bug in
+ FreeBSD UNIX domain socket name handling.
+
+ Corrected gcc build problems on HP-UX 11i, reported
+ by Yuliy Minchev <yuliy@mobiltel.bg>.
+
+ Updated BSDI BSD/OS support to 4.3.1.
+
+ Augmented a lock ID test on NetBSD to check if the
+ ID is an LWP pointer.
+
+4.68 June 18, 2003
+ Enhanced Configure script's cleanup operations.
+
+ Added support for OpenBSD 3.3, based on a report
+ from Peter Valchev <pvalchev@sightly.net>.
+
+ Improved the description of the detached PGP
+ signature certificate file in the main lsof README
+ file, based on a suggestion from Diana Stockdale
+ <diana@mpl.ucsd.edu>.
+
+ Installed a work-around for FreeBSD 5.0-CURRENT on
+ Alpha to avoid a compiler register use complaint.
+
+ Corrected a 'c' option error message. Gnele
+ <blaadeleng@yahoo.com> reported the problem.
+
+ Upgraded EXT2FS and UFS support for NetBSD and
+ OpenBSD to handle new inode information, and the
+ fast UFS1 and UFS2 file systems.
+
+ With the help of Andrew Brown <atatat@atatdot.net>
+ determined the NetBSD snapshot (1.6F) at which
+ <sys/mount.h> could be included under _KERNEL, thus
+ eliminating the lsof netexport.h hack. The same
+ change applies to OpenBSD versions 3.3 and above.
+
+ Applied a patch from Armin Gruner <ag@muc.de> that
+ corrects the use of the HASPROCFS definition in the
+ FreeBSD dialect sources.
+
+ Corrected spelling errors in 00FAQ and in the
+ generated 00.README.FIRST_<version> file of the
+ distribution archive. John Jackson <jrj@purdue.edu>
+ and Ray Phillips <r.phillips@jkmrc.uq.edu.au>
+ spotted and reported the errors.
+
+ Corrected a spelling error in a comment and incorrect
+ use of an alarm function in the LTsock test program.
+
+ At the suggestion of Stuart Anderson <sba@srl.caltech.edu>
+ added preliminary (and incomplete) SAM-FS file system
+ support to Solaris lsof. Completion awaits availability
+ of SAM-FS internals.
+
+ Fixed a Solaris device name printing bug, reported by
+ Ric Anderson <ric@tick.Telcom.Arizona.EDU>, only
+ visible when HASDCACHE is not defined. Ric helped
+ test the fix.
+
+ Fixed an AIX kernel bit size handling bug related
+ to the NFS node (rnode) structure.
+
+ Corrected a print_kptr() function call error in the AIX
+ AFS code, reported by David Steiner
+ <david.r.steiner@Dartmouth.EDU>. Upon further reflection
+ and because I no longer have appropriate AIX AFS test
+ systems, disabled AIX AFS support in the Configure script
+ for AIX versions above 4.3.3.0 or AIX AFS versions above 3.5.
+
+ Added support for FreeBSD 5.1.
+
+ With advice from Allan Nathanson <ajn@apple.com> adjusted
+ the Darwin get-xnu-headers.sh script to access the kernel
+ header files needed by lsof from a new form of the Apple
+ open source repository.
+
+ Installed Linux and lsof library bug fixes and
+ improvements, supplied by Marian Jancar <mjancar@suse.cz>.
+ One Linux improvement handles mount strings that
+ have octal escapes in them, eg., \040 for embedded
+ blanks. Marian tested the changes.
+
+4.69 October 16, 2003
+ Received and applied an OpenBSD patch from Peter Valchev
+ <pvalchev@sightly.net> that replaces a ctob() call with
+ a sysconf() call. Peter claims sysconf() is needed for
+ OpenBSD on SPARC. (It is not needed for NetBSD on SPARC.)
+
+ With the upgrade of my only Solaris 7 test system
+ to, Solaris 8, dropped the *claim* that lsof works
+ on Solaris 7. That doesn't mean it won't work
+ there, so those who want lsof for Solaris 7 probably
+ should be able to build it there and it probably
+ will work there.
+
+ Revised lsof's DNLC handling for BSD derivatives,
+ including: BSDI; Darwin, DEC OSF/1, Digital UNIX
+ and Tru64 UNIX; FreeBSD; NetBSD; and OpenBSD. The
+ latest NetBSD distribution's dropping of the vnode
+ capability ID (v_id) required the revision.
+
+ Adjusted to the latest FreeBSD 5.1-CURRENT.
+
+ Added NetBSD support for using kvm_getproc2().
+
+ Added a patch from Andrew Brown <atatat@atatdot.net>
+ to handle NetBSD enum conflicts and changes in the
+ <miscfs/kernfs/kernfs.h> and <miscfs/procfs/procfs.h>
+ header files.
+
+ Added a "#define _KERNEL" to the AIX dnode2.c source
+ file for compatibility with a new <j2/j2_inode.h>
+ AIX 5.2 header file version. The addition was
+ supplied by Dick Dunbar <Dick.Dunbar@Siebel.com>
+ and was offered as a patch to lsof 4.68/
+
+ Added support for a second type of Solaris SAMFS.
+ Stuart Anderson <sba@srl.caltech.edu> provided the
+ support. SAMFS support in lsof SOLARIS remains
+ scanty, because Sun won't release any details on
+ its kernel structures.
+
+ Dropped the *claim* that lsof works on AIX 4.3.3,
+ because I was unable to test it there. That doesn't
+ mean it won't work there, so those who want lsof
+ for AIX 4.3.3 probably should be able to build it
+ there and it probably will work there.
+
+ Updated for Solaris 10 on test systems provided by
+ Mike Miscevic <miscevic@hotpop.com>. Casper Dik
+ <casper@holland.sun.com> provided significant help.
+ During the Solaris 10 port found and fixed an lofs
+ handling bug that prevented reporting of open lofs
+ file lock status.
+
+ Updated the DNLC test, LTdnlc, to provide a possible
+ explanation about file systems on which the test
+ might fail.
+
+ Modified the procedure for obtaining missing Darwin
+ XNU kernel header files. The new one requires more
+ manual intervention, but is the best that can be
+ done with the way Apple open sources are now
+ organized. 00FAQ explains the new procedures for
+ those not used to downloading Apple open source
+ files.
+
+ Added support for Apple Darwin 7.0 (Mac OS X 10.3)
+ with patches supplied by Allan Nathanson <ajn@apple.com>.
+ Dropped the *claim* that lsof builds and works on
+ Apple Darwin below 6.0.
+
+ Validated lsof on FreeBSD 4.9, using a test system
+ provided by Ben Lewis <bl@purdue.edu>.
+
+ Validated lsof on FreeBSD 5.1-CURRENT for Amd64.
+ David O'Brien <obrien@FreeBSD.org> provided a test
+ system.
+
+ Changed the NetBSD Configure stanza to do header
+ file searches in /usr/include by default. The
+ LSOF_INCLUDE and NETBSD_SYS environment variables
+ may still be used to specify other search paths.
+ Discussions with Andrew Brown and Wolfgang S.
+ Rupprecht <wolfgang@wsrcc.com> led to the change.
+
+4.70 January 16, 2004
+ Improved shell-portability of the linux stanza of
+ the Configure script with a patch from Paul Jarc
+ <prj@po.cwru.edu>.
+
+ Added a "silent" rule to tests/Makefile for Paul.
+ Updated, extended and clarified the test suite
+ documentation in 00FAQ and 00TEST.
+
+ Fixed Solaris 10 dlsof.h typo, reported by Mike
+ Miscevic <miscevic@hotpop.com>. The typo prevents lsof
+ from loading cleanly in Solaris 10 builds past 40.
+
+ Fixed a Solaris HSFS node number reporting bug and
+ added a structure definition work-around for Solaris
+ 10.
+
+ Converted PGP signing to GPG. My previous PGP key can
+ be used, but the gpg "--allow-non-selfsigned-uid"
+ option may have to be used when it is imported into a
+ GPG key ring.
+
+ Added bz2 compression.
+
+ Updated for OpenBSD 3.4.
+
+ Added a work-around for a missing header file in the
+ s10_44 Solaris 10 build.
+
+ Added support for FreeBSD 5.2-BETA and 5.2-CURRENT.
+
+ Updated Linux AX25 support with modifications supplied
+ by Lutz Poetschulat <dl9cu@db0zwi.de>.
+
+ Added raw IPv6 support to Linux lsof.
+
+ Improved handling of parameters after "-i@".
+
+ Improved file name test in LTdnlc.c.
+
+ Added loop count controls to the reading of Solaris
+ lock chains. The change was implemented as a result of
+ a report from Steve Gonczi <steve@relicore.com>.
+
+ Based on a report from John Jackson <jrj@purdue.edu>,
+ enabled a Solaris 10 <sys/lgrp.h> work-around for
+ Solaris 9, too. (Patch 112233 installs an lgrp.h on
+ Solaris 9 that needs the work-around.)
+
+ With help from Andrew Brown <atatat@atatdot.net> and
+ John Heasley <heas@netbsd.org> added log-structured
+ file system (LFS) support for NetBSD and OpenBSD.
+
+ Added AMD64 to the list of FreeBSD 5.x-CURRENT
+ supported architectures. FreeBSD.org provides a test
+ system, courtesy of (I believe) David O'Brien
+ <obrien@FreeBSD.org>.
+
+ Added a cast to lseek() in the HP-UX /dev/kmem-based
+ kread() function to make it work properly with the
+ bundled HP C compiler.
+
+4.71 March 11, 2004
+ Added text file support to Apple Darwin lsof and
+ enabled the lsof executable portion of the LTbasic
+ test. Added support for Darwin kernel queue, POSIX
+ semaphore and POSIX shared memory files. Tested on
+ Darwin 7.2 (aka Mac OS 10.3.2).
+
+ Added process_kqueue() function prototypes for FreeBSD,
+ NetBSD and OpenBSD.
+
+ Picked some lint in AIX sources, lib/rnmh.c and
+ tests/LTsock.c.
+
+ Added "-x [fl]" cross-over option, which enables +d and
+ +D processing to cross over symbolic links and|or file
+ system mount points. Discussion with Johan Lindquist
+ <johan@smilfinken.net> and Eric Williams (aka The Ghost
+ In The Machine) <ewill3@earthlink.net> on Linux news
+ groups revealed the need for the option.
+
+ Updated support for UnixWare 7.1.4.
+
+ Added support for the optional reporting of socket
+ options, socket states and TCP flags for most currently
+ supported dialects. John Smith <lbalbalba@hotmail.com>
+ and Tristan Nefzger <tn@bhtrader.com> requested the
+ information. The dialects and their versions for which
+ this feature has become available include:
+
+ AIX 4.3.2 and 5.[12]
+ Apple Darwin 7.2
+ BSDI BSD/OS 4.3.1
+ Digital UNIX and Tru64 UNIX 4.0
+ FreeBSD 4.9 and 5.2
+ HP-UX 11 and 11.11 (aka 11i)
+ NetBSD 1.6ZH
+ OpenBSD 3.4
+ OPENSTEP 4.2
+ OpenUNIX 8
+ SCO OpenServer Release 5.0.6
+ Solaris 2.6, 8, 9 and 10
+ UnixWare 7.1.[134]
+
+ Modified the Configure stanza for HP-UX 11 with better
+ q4 detection. Steve Bonds <3vhmxxm02@sneakemail.com>
+ supplied the modification.
+
+ Applied a patch from Mike Miscevic <miscevic@hotpop.com>
+ to enable lsof to compile with the zone support in the
+ Solaris 10 s10_b51 release. Added information on lsof
+ zone behavior to 00FAQ.
+
+ Added a "-z [z]" option to Solaris 10 lsof. It enables
+ the listing of zone name and can also be used to select
+ the listing of processes and their files from specified
+ zones.
+
+4.72 July 13, 2004
+ Corrected Solaris 10 ZONE column title display bug with
+ a patch from Joep Vesseur <Joep.Vesseur@Sun.COM>. Joep's
+ fix was offered as a patch to 4.71.
+
+ Based on a report from Jean-Pierre Radley <jpr@jpr.com>
+ about an unexpected GNU uname Configure interaction on
+ OSR, and working from information received from Bela
+ Lubkin, changed the OSR Configure stanza to use
+ /bin/uname instead of uname. Added an FAQ entry about
+ Configure version detection problems.
+
+ Added the +m and "+m m" options in response to a dialog
+ with Robert T. Brown <rbrown@netmentor.com>. The
+ options allow the creation of a mount table supplement
+ file which can be used on selected dialects to get
+ device numbers when stat(2) and lstat(2) can't deliver
+ them. (That's generally the result of an inaccessible
+ NFS server.) Currently the new options are supported
+ only on Linux.
+
+ Made cpumask_t typedef _KERNEL compensation for FreeBSD
+ 5.2-CURRENT. Refined it for 5.2.1-RELEASE with testing
+ help from Scott Ellentuch <tuc@ttsg.com>.
+
+ Added support for FreeBSD 4.10. Larry Rosenmann
+ <ler@lerctr.org> kindly provided a test system.
+
+ Added support for NetBSD 2.0 with patches supplied by
+ Andrew Brown <atatat@atatdot.net>. Andrew also
+ provided two test systems.
+
+ Made handling of Linux maps file more robust, based on
+ a report from Jan Blunck <J.Blunck@tu-harburg.de>. As
+ a side benefit, made handling of generated stat(2)
+ information more flexible.
+
+ As a result of a discussion with Jason Fortezzo
+ <fortezza@mechanicalism.net>, adjusted lsof for Solaris
+ to obtain the maximum user name length from ut_name of
+ the utmpx structure, if <utmpx.h> exists.
+
+ Tested under OpenBSD 3.5.
+
+ Updated 00README information about using gcc (via the
+ Configure aixgcc abbrevisiation) to compile lsof on
+ AIX. Ann Janssen <ajanssen@nebook.com> made me aware
+ the information was out of date.
+
+ Added an AIX SIGDANGER handler and some 00FAQ sections
+ on lsof memory usage after a discussion with Tom Qin
+ <tom.qin@citigroup.com> about lsof memory usage.
+
+ Added scripts/sort_res.perl5, contributed by Fabian
+ Frederick <fabian.frederick@gmx.fr>. The script
+ displays lsof output sorted by size and path name.
+
+ Improved handling of files on Linux NFS mount points
+ that use the root_squash option, based on discussions
+ with Paul Szabo <psz@maths.usyd.edu.au>.
+
+ Updated FreeBSD 5.2-CURRENT support, based on a problem
+ report from Filippo Natali <filippo@widestore.net>.
+
+ Corrected improper FreeeBSD 5.x-CURRENT #if condition,
+ reported by Kim Culhan <kimc@kim.net>.
+
+ Added a Configure script work-around for AIX 5.2 lsof
+ with JFS2, compiled by gcc >= 3.3. The work-around
+ was supplied by Florian M. Weps <fmw@hactrn.ch>.
+
+4.73 October 21, 2004
+ Added an __XPG4_CHAR_CLASS__ #define before
+ #include'ing <ctype.h> on Solaris to restore lsof's
+ ability to display special characters such as acute-e.
+
+ Added wide-character (e.g., UTF-8) support where
+ possible, prompted by a request from Kyungjoon Lee
+ <kjoonlee@gmail.com>. Some older dialects -- e.g.,
+ NetBSD 1.4.1 -- don't support wide characters, so the
+ wide character support is enabled by definitions in
+ each dialect's machine.h. Dialects with wide-
+ character support are listed in 00FAQ.
+
+ Make a FreeBSD 5.2-CURRENT adjustment for <sys/pipe.h>,
+ supplied by Sergey A. Osokin <osa@FreeBSD.ORG>.
+
+ Implemented a Linux feature request made by Jakub
+ Jelinek <jakub@redhat.com> that enhances lsof's ability
+ to locate UNIX domain sockets whose paths are named as
+ arguments. Jakub supplied suggested code.
+
+ Dropped *claims* that lsof works on AIX below 5.1, SCO
+ Dropped *claims* that lsof works on AIX below 5.1, SCO
+ Openserver 5.0.4, Tru64 UNIX 5.0, and UnixWare below
+ 7.1.4. Lsof will probably build and work on those UNIX
+ dialect versions, but I no longer have any way to test
+ lsof on them.
+
+ Added support for FreeBSD 5.3 and 6.0. The FreeBSD
+ 5.3 support hasn't been tested.
+
+ Added FD test code that will allow dialect versions to
+ test FD option selections. Used the new code in the
+ PSTAT-based HP-UX lsof to enable it to avoid scanning
+ the mount table when its information is not needed.
+ The addition was made in response to a query from
+ Harvey Garner <Harvey.Garner@championusa.com> about
+ lsof performance in a busy NFS environment.
+
+ Upgraded lsof's AIX support level to AIX 5.3, based on
+ a report from Dick Dunbar <Dick.Dunbar@Siebel.com>. (I
+ have not tested lsof under AIX 5.3.) Based on Dick's
+ recommendation and local testing changed the C for AIX
+ version 6 and higher -qmaxmem option value to -1.
+
+ Made LSOF_AR environment variable more useful and
+ documented it in 00XCONFIG.
+
+ Corrected the use of sum(1) to generate signatures for
+ the lsof distribution and binaries to match the
+ documentation that claims it is sum -r output. Jin
+ Guojun <jin@george.lbl.gov> noticed and reported the
+ problem.
+
+ Tested under OpenBSD 3.6.
+
+ Added checksum and GPG certificate files for the bz2,
+ gz and Z lsof distribution archives. The new files
+ reside with the distribution archives and supplement
+ the signature information already inside the archives.
+
+ Validated on Solaris 10, i8xpc, build s10_63.
+
+4.74 January 17, 2005
+ Fixed a Solaris segment fault bug on systems that lack
+ a /dev/allkmem device. Offered the fix as a patch to
+ lsof 4.73. The bug was reported by Donald Zoch
+ <donald.zoch@amd.com>.
+
+ Updated lsof for FreeBSD 6.0 and higher for a change in
+ <sys/vnode.h>, based on a report from Sergey A. Osokin
+ <osa@FreeBSD.ORG>. Made the update available in a 4.74
+ 'A' edition pre-release.
+
+ Filed an HP bug report about missing pstat(2) CWD info
+ for LOFS on HP-UX 11.11 and higher. The missing CWD
+ info was noticed by Ermin Borovac <e.borovac@bom.gov.au>.
+ Added info to 00FAQ about the problem, which can cause
+ the lsof test suite's LTbasic test to fail.
+
+ Updated the q4-generated tcp_s.h in the lsof
+ distribution and added socket option support for HP-UX
+ 11.00. Erwin Reyns <ereyns@europarl.eu.int> helped
+ test.
+
+ Updated for Solaris 10, build s10_69, with a patch
+ supplied by Mike Miscevic <miscevic@hotpop.com>.
+
+ Added v_path support to Solaris 10 lsof. That relieves
+ it of having to read and decode the kernel DNLC, and
+ delivers full paths more reliably.
+
+ Added specialized NFS4 support to Solaris 10 lsof.
+
+ Applied Solaris 10 patches to lsof supplied by Casper
+ Dik <casper@holland.sun.com>.
+
+ Updated lsof for NetBSD 2.99.10 and tested it on a
+ system provided by Andrew Brown <atatat@atatdot.net>.
+
+ Added support for the FreeBSD 6.0-CURRENT f_vnode
+ pointer in the file structure.
+
+ Added BSDI, FreeBSD, NetBSD and OpenBSD support for the
+ *effnlink member of the inode structure. This makes
+ the lsof LTnlink test run faster on all modified
+ dialects and correctly on OpenBSD.
+
+ Added ptyfs support for NetBSD, using modifications
+ provided by Andrew Brown.
+
+ Changed the netbsd Configure stanza to look by default
+ for system header files in both /usr/include and
+ /usr/src. (The NETBSD_SYS environment variable can
+ still be used to select an alternate for /usr/src.)
+
+ Corrects two FreeBSD 4.10 RPC/XDR type definitions.
+
+ Added an FAQ Q&A about setuid and setgid restrictions
+ in HP-UX 11.11. The information in the answer was
+ supplied by Frank Sanders <frank.sanders@siemens.com>.
+
+ Added abbreviations for AXI FCIO and FSNAPSHOT file
+ flags. Holger VanKoll <Holger.VanKoll@swisscom.com>
+ reported the missing FCIO.
+
+ Adjusted lsof's private AIX 64 bit rnode structure for
+ 64 bit AIX 5.2 systems. (IBM doesn't distribute a
+ correct <nfs/rnode.h> for it.)
+
+ Corrected a Linux socket inode printing bug reported by
+ Igor Schein <igor@txc.com>.
+
+ Updated for FreeBSD 4.11. The support compiles but
+ hasn't been tested.
+
+ Back-ported a FreeBSD 6.0-CURRENT fix to FreeBSD
+ 5.3-RELEASE-p1. That was done to solve a compilation
+ problem reported by Radko Keves <rado@daemon.sk>.
+
+4.75 May 16, 2005
+ Dropped the *claim* that lsof works on DEC OSF/1 and
+ Digital UNIX, since my last 4.0 test system has been
+ removed. The last tested distribution of lsof on DEC OSF/1
+ and Digital UNIX was revision 4.74. It has been archived
+ on lsof.itap.purdue.edu in pub/tools/unix/lsof/OLD/src.
+
+ Added negation forms to the values in the -g (PGID) and
+ -p (PID) lists. Negated PGID and PID values, like
+ negated UID or login name values, are applied without
+ ORing or ANDing and take effect before any other
+ selection criteria are applied.
+
+ At the request of Marcin Gozdalik <gozdal@gmail.com>
+ added a -X option for Linux. The option inhibits the
+ reading of the /proc/net/tcp* and /proc/net/udp*
+ files.
+
+ Based on a report from David Gutierrez
+ <davegu1@hotmail.com> changed DEC OSF/1 process table
+ allocation to request memory in smaller increments.
+
+ Based on a report from jayjwa <jayjwa@atr2.ath.cx>
+ updated the Customize script to use "tail -n 1" where
+ possible.
+
+ Enabled support for FreeBSD 5.4.
+
+ Improved the BSDI, FreeBSD, NetBSD, OpenBSD and Solaris
+ kvm_open() and kvm_openfiles() error messages.
+
+ Enabled support for NetBSD 2.99.12.
+
+ Improved HP-UX Configure stanza with help from Piet
+ Starreveld <pstarrev@csc.com>. Picked some lint Piet
+ found.
+
+ Enabled IPv6 support for HP-UX > 11. Piet Starreveld
+ helped test it on 11.23, among others.
+
+ Updated for HP-UX 11.23 on the ia64 architecture.
+
+ Updated to latest FreeBSD 6.0-CURRENT, using a test
+ system provided by Andrzej Tobola <ato@iem.pw.edu.pl>.
+
+ Added support for SCO OSR 6.0.0 and UnixWare 7.1.4 with
+ help from Richard at SCO.
+
+ Corrected a Linux bug in NFS handling, reported by Karel Zak
+ <kzak@redhat.com>. Karel supplied a patch.
+
+ Improved the code for accessing an AIX 3.2 and higher
+ sockaddr_un structure, thus eliminating a segmentation
+ fault possibility.
+
+ Updated for AIX 5.3.
+
+ Added preliminary (DEBUG) support for the AIX SANFS
+ file system.
+
+ Fixed a bug in the Solaris 10 processing of the vnode's
+ v_path pointer with code supplied by Edward Jajko
+ <ejajko@portal.com>. The fix was offered as a patch to
+ 4.74.
+
+ Dropped support for OpenUNIX 8, since a test system is
+ no longer available. Archived an OpenUNIX-only
+ distribution of the last revision (4.74) tested on
+ OpenUNIX in pub/tools/unix/lsof/OLD/src.
+
+ Tested under Openbsd 3.7.
+
+ Tested under Darwin 7.7.0.
+
+ Enabled building on amd64 Solaris 10 with hints from
+ Marc Aurele La France <tsi@ualberta.ca>. Marc provided
+ a test system.
+
+ Supplied a missing quote in the FreeBSD Configure
+ stanza. Carl Cook <Info@quantum-sci.com> reported the
+ problem.
+
+ Removed "-O" option from tests/Makefile so that the
+ HP-UX bundled compiler won't complain.
+
+4.76 August 30, 2005
+ Corrected an example and spelling errors in man page.
+
+ Updated for Apple Darwin 8.x with changes supplied by
+ Allan Nathanson <ajn@apple.com>. Allan also provided a
+ test system.
+
+ Completed documentation of CLRLFILEADD in all machine.h
+ files.
+
+ At the request of Chris Markle <cmarkle@sendmail.com>
+ added partial listen queue length to socket options
+ displayed when -Tf is specified. Partial queue length
+ is not reported for all dialects. (00FAQ lists the
+ ones where it is reported.)
+
+ Updated for FreeBSD 7.0 with information supplied by
+ Andrzej Tobola <ato@iem.pw.edu.pl>.
+
+ Updated Solaris VxFS support for VxFS versions 4 and
+ above with technical advice from Craig Harmer
+ <craig_harmer@symantec.com>, Gary Millen
+ <gary_millen@symantec.com> and Chuck Silvers
+ <charles_silvers@symantec.com>. Testing help was
+ provided by Michael Antlitz <mantlitz@prophasys.com>,
+ Steve Ginsberg <steve@dhapdigital.com> and Kenneth
+ Stailey <kstailey@yahoo.com>.
+
+ Fixed a Solaris address space map processing bug.
+ Janardhan Molumuri <mjanardhan@gmail.com> reported the
+ bug and help me identify it. Made the fix available as
+ a patch to 4.75.
+
+ Added support for Solaris 10 port and CTFS files. The
+ CTFS support is imcomplete, because I don't know how
+ to get inode number, size and link count. (There's
+ a new 00FAQ entry about that.)
+
+ Investigated a report from Christopher J Warweg
+ <warwegc@GAO.GOV> that the CHECKSUMS for the lsof 4.75
+ binary for 64 bit Solaris 8 was incorrect. It was my
+ packaging error. I rebuilt and repackaged the binary.
+
+ Enabled support for Linux map file names with embedded
+ spaces.
+
+4.77 April 10, 2006
+ Added -X option support for Solaris 10 and above. When
+ -X is specified lsof will report cached v_node path
+ names for unlinked files, followed by "(deleted)".
+ Improved cached vnode path name handling by adding
+ "(?)" to the end of path names of questionable accuracy.
+ Updated 00FAQ to reflect these changes.
+
+ Updated for FreeBSD 7.0-CURRENT.
+
+ Fixed name addition spacing bug, reported by Stuart
+ Anderson <anderson@ligo.caltech.edu>. Also updated
+ Solaris 10 SAMFS support at Stuart's request.
+
+ Added missing "break;" and another HASSTATVFS test to
+ the NetBSD and OpenBSD dnode.c. Bill Behr
+ <bbehr@networkstoragecorp.com> reported those needs.
+
+ Fixed an HP-UX 11 file descriptor "chunk" size problem,
+ reported by Per Allansson <per@appgate.com>. Per helped
+ devise the fix and tested it. This fix was offered as
+ a patch to lsof 4.76.
+
+ Updated for FreeBSD 6.0-STABLE and FreeBSD
+ 6.1-PRERELEASE.
+
+ Updated scripts/sort_res.perl5 with changes supplied by
+ Frederick Fabian <fabian.frederick@skynet.be>, the
+ author of the script.
+
+ Corrected +|-M man page documentation error, reported
+ by Roger Cornelius <rac@tenzing.org>.
+
+ Improved FreeBSD user device random seed generation in
+ response to a problem report from Danny Braniss
+ <danny@cs.huji.ac.il>.
+
+ Eliminated three syntax error bugs and other compiler
+ complaints from the PSTAT-based lsof. H. Merijn Brand
+ <h.m.brand@xs4all.nl> reported the problems and tested
+ the fixes.
+
+ Eliminated compiler complaints in the test suite.
+
+ Investigated problems with the building of lsof on
+ PA-RISC HP-UX 11.23, based on a report from John
+ Orndorff <John.Orndorff@sungard.com>. Found that
+ neither the HP bundled C compiler nor gcc would build
+ lsof, but the the HP unbundled ANSI C compiler would.
+ Concluded that HP bundled C compiler can't handle
+ <gssapi/gssapi.h>. Devised a work-around to gcc's
+ omission of the rpcent structure definition of
+ <netdb.h> that allows it to compile lsof's print.c, but
+ the resulting binary doesn't run reliably. Documented
+ the situation in 00FAQ.
+
+ Changed reporting of unknown file types. The number of
+ an unknown type is now reported as four octets. The
+ change was made in response to a Linux lsof bug report
+ from Karel Zak <kzak@redhat.com>.
+
+ Dropped the *claim* that lsof works on BSDI BSD/OS
+ since my last test system has been removed. The last
+ tested distribution of lsof for BSDI BSD/OS was
+ revision 4.76. It has been archived on
+ lsof.itap.purdue.edu in pub/tools/unix/lsof/OLD/src.
+
+ As a result of discussing the lsof source tar's MD5
+ checksum with Andrew Bell <andrew.bell.ia@gmail.com>,
+ changed the description of a suitable MD5 tool in the
+ lsof distribution's documentation to name the openssl
+ "dgst" command.
+
+ Enabled compilation on Solaris 10 1/06 with a fix sent
+ by Jason Fortezzo <fortezza@mechanicalism.net>. Made
+ the fix available as a patch to 4.76.
+
+ Adjusted to FreeBSD 5.5-PRERELEASE.
+
+ Corrected a bug in the lsof library's process_file()
+ function to enable the locating of AIX XTI sockets by
+ their TCP/IP address values. The bug was reported by
+ Michel Dubois <Dubois@sears.ca>.
+
+ Based on a bug report from Karel Zak <kzak@redhat.com>
+ added command name length checking to as many dialects
+ as possible (Linux for Karel) for the "-c c" option.
+
+ Updated for OpenBSD 3.[89]. Tested the 3.9 update on a
+ system provided by David Mazieres. I have not tested
+ on OpenBSD 3.8, but David reports lsof 4.76 worked
+ there.
+
+ Ended regression testing of lsof on 32 bit Solaris 8
+ with the ending of access to a test system. Lsof
+ continues to be tested on 64 bit Solaris 8.
+
+4.78 April 24, 2007
+ Added more information to the lsof FAQ about missing
+ link counts and sizes on Linux files.
+
+ Simplified Linux stat() and lstat() usage.
+
+ Relocated #define's that prevent OpenBSD compilation on
+ systems without a /proc file system. Pieter Bowman
+ <bowman@math.utah.edu> reported the problem.
+
+ Added code to avoid processing Linux /proc/<PID>/maps
+ file entries with zero device and node numbers. Some
+ such entries now have names associated with them that
+ are not path names -- e.g., "[heap]", "[stack]" or
+ "[vdso]". Scott Worley <sworley@chkno.net> reported
+ lsof's mishandling of such entries.
+
+ Added SELinux security context support, provided by
+ James Antill <james.antill@redhat.com>. I have not
+ tested this, but James and Karel Zak <kzak@redhat.com>
+ have.
+
+ Added the #include of <sys/types.h> to Solaris lsof to
+ enabled compilation on Solaris 10 6/06. Peter Harvey
+ Peter.Harvey@Sun.COM diagnosed the problem and supplied
+ a patch.
+
+ Added better support for JFS2 on AIX 5.2 and 5.3, based
+ on bug reports and help from Thomas Braunbeck
+ <BRAUNBEC@de.ibm.com> and Tom Whitty <TWHITTY@cerner.com>.
+
+ Documented that lsof supports AIX 5.3 only up through
+ maintenance level 1 (ML1).
+
+ Enabled Solaris lsof to locate the AFS vnode operation
+ address for OpenAFS 1.4.1. The fix was supplied by
+ Robert Jelinek <Robert.Jelinek@MorganStanley.com>.
+
+ Enabled support for Solaris 10 ZFS. If the necessary
+ ZFS header files aren't found, lsof offers the option
+ to drop ZFS support, to use internal, possibly
+ inaccurate structure definitions, or to supply a path
+ to the missing header files. Horst Scheuermann
+ <Horst.Scheuermann@uni-trier.de> provided a development
+ system and helped test the support.
+
+ Corrected a typo in the man page, reported by Eric S.
+ Raymond <esr@thyrsus.com>.
+
+ Changed the spelling of macroes to macros in lsof
+ source and documentations files, based on a suggestion
+ from Josh Soref <timeless@gmail.com> and verification
+ with the OED.
+
+ The following dialects are no longer supported: 32 bit
+ AIX 5.2, HP-UX 11, OpenStep 4.2, Solaris 2.6, Solaris
+ 8, True Unix 64 and UnixWare 7.1.4. Lsof may work on
+ them, but I no longer have test systems for them.
+ Support for OpenBSD ends at its version 3.9 for lack of
+ interest in the port.
+
+4.79 April 15, 2008
+
+ **************** IMPORTANT NOTE ******************
+ * *
+ * Lsof support has been reduced to the following *
+ * dialects: AIX, FreeBSD, Linux and Solaris, and *
+ * only in selected versions of those dialects. *
+ * The selected versions are listed in this file *
+ * and in other lsof documentation. *
+ * *
+ * I have made this move because of retirement *
+ * and because I no longer have many test systems *
+ * available to me. *
+ * *
+ * Vic Abell *
+ * *
+ **************************************************
+
+ Fixed a Solaris VXFS permission problem when accessing
+ the VXFS inode offsets. The bug was reported by
+ Gregory A. Ivanov <ivga@mts.ru>. Gregory tested the
+ fix.
+
+ Moved an #include <string.h> later in FreeBSD dlsof.h
+ to enable compilation on recent FreeBSD releases. The
+ change was supplied by Roy Marples <uberlord@gentoo.org>.
+
+ Improved Linux /proc file stream reading speed by applying
+ an expanded version of a patch from Eric Dumazet
+ <dada1@cosmosbay.com> that allocates a page size buffer
+ to each stream. Improved TCP, TCP6, UDP and UDP6 hashing
+ by determining the hash bucket count from the /proc/net
+ sockstat and sockstat6 files. The improvement was
+ suggested by Eric and he provided sample code. Eric also
+ tested both improvements.
+
+ Modified Configure script to build lsof on FreeBSD
+ 6.2. Tested it on a system provided by Larry Rosenman
+ <ler@lerctr.org>.
+
+ Fixed a Linux maps file processing bug that prevented path
+ names from having an embedded colon. James Lingard
+ <jchl@arastra.com> reported the bug and helped with its
+ fix.
+
+ Based on reports from Eric Dumazet and Samuel Thibault
+ <samuel.thibault@ens-lyon.org> added support for the
+ Linux 2.6.22 kernel's /proc/<PID>/fdinfo files -- i.e.,
+ file offset and flags. Samuel Thibault provided a test
+ system.
+
+ Fixed a Linux UNIX socket memory leak, reported by
+ Philip Shin <pshin@xceedium.com>. Phillip supplied the
+ fix.
+
+ With generous assistance from HP added support for an HP-UX
+ 11.23 patch that makes TLI/XTI socket address information
+ available.
+
+ Fixed a header file problem for FreeBSD 6.2 on the Alpha
+ architecture. The problem was reported by Pekka Honkanen
+ <phonkane@cc.hut.fi>. Pekka tested the fix.
+
+ Based on a report and using suggested fixes from Karel Zak
+ <kzak@redhat.com>, made these changes to Linux lsof: corrected
+ a getpidcon() error message; insured that inode numbers are
+ handled correctly for their unsigned long long type; and
+ improved SELinux handling. At the request of Alon Bar-Lev
+ <alonbl@gentoo.org> added the LINUX_HASSELINUX environment
+ variable to enable or inhibit SElinux support unconditionally.
+
+ Updated Configure for FreeBSD 8.0-CURRENT and tested lsof on
+ AMD64 there.
+
+ Added a patch provided by Oles Hnatkevych
+ <don_oles@able.com.ua> for FreeBSD systems where the root
+ file system is on a CD9660 device.
+
+ Added compensation for the disappearance of FMARK and FDEFER
+ from the FreeBSD 8.0-CURRENT <sys/file.h>.
+
+ Updated FreeBSD lsof with ZFS support. Larry Rosenman
+ <ler@lerctr.org>, Erwin Lansing <erwin@FreeBSD.org>, Wesley
+ Shields <wxs@atarininja.org> and Dmitry Morozovsky
+ <marck@rinet.ru> provided test systems.
+
+ Fixed a socket file identification problem reported by
+ Pavol Rusnak <stick@gk2.sk>. Pavol also reported the
+ cause of the problem.
+
+ Added the ability to format the repeat mode marker line
+ with strftime(3), where the dialect supports the
+ localtime(3) and strftime(3) C library functions. The
+ addition was suggested by Mike Depot <mdepot@comcast.net>,
+ who also tested it. The addition required creating a new
+ main lsof source module, util.c, that contains functions
+ whose compilation conflicts with the general header file
+ tree defined by lsof.h and dlsof.h.
+
+ Based on reports from Andrei V. Lavreniyuk
+ <andy.lavr@reactor-xg.kiev.ua> and Pav Lucistnik
+ <pav@FreeBSD.org> updated the FreeBSD 7.0 and above
+ file lock handling to use new locking structures. The
+ update requires a terrible hack to get a definition for
+ the lock owner structure from a kernel source module
+ into a local lsof header file.
+
+4.80 May 12, 2008
+ Updated for a FreeBSD 7.0 and above byte level locking
+ change. The problem was reported by Conrad J. Sabatier
+ <conrads@cox.net>, who helped test the update. Wesley
+ Shields <wxs@FreeBSD.org> provided an 8.0-CURRENT test
+ system.
+
+ Propagated the FreeBSD 7.0 and above locking changes to
+ FreeBSD 6.x, based on a report from Edwin Groothuis
+ <edwin@FreeBSD.org>.
+
+ Added warnings for unsupported dialects or versions.
+
+ Added Linux support for the UDPLITE protocol. Eric
+ Dumazet <dada1@cosmosbay.com> supplied a patch.
+
+ Added a missing quote to the Configure script's
+ FreeBSD stanza.
+
+ Added a usage.o rule to the HP-UX PSTAT-based
+ Makefile. I mistakenly deleted the rule at revision
+ 4.79. The missing rule was reported by Kawaljeet Kaur
+ <kawaljeet.malviya@gmail.com> who tested the corrected
+ Makefile.
+
+4.81 October 21, 2008
+ Updated the Darwin libproc sources with changes from
+ Allan Nathanson <ajn@apple.com>. Tested them on a iMac
+ mini, provided by Apple Inc.
+
+ Changed dummy declarations in library source files to
+ eliminate complaints about unused variables and empty
+ object files. This change may not work on dialects I
+ can no longer test; it has been tested on some versions
+ of AIX, Darwin, FreeBSD, Linux and Solaris.
+
+ At the request of Hal Brooks <hal@uga.edu> added support
+ for Linux /proc/net/packet files. Hal tested it.
+
+ Added socket file only performance enhancements to Linux
+ and PSTAT-based HPUX lsof.
+
+ Added htonl call around improper usage of INADDR_LOOPBACK;
+ report from an Apple engineer forwarded by Allan Nathanson
+ <ajn@apple.com>.
+
+ Adjusted for FreeBSD-8.0 change in device number handling.
+ The adjustment should work for FreeBSD 5 and above, should
+ the 8.0 change be propagated downward. The problem was
+ reported by Pav Lucistnik <pav@FreeBSD.org>. An updated
+ test system was provided by Erwin Lansing <erwin@FreeBSD.org>.
+
+ Reduced AIX support to version 5.3, since test systems with
+ older versions are no longer available to me.
+
+ At the request of Marjo F. Mercado <mmercado@xceedium.com>
+ and Phil Shin <pshin@xceedium.com> applied some speed
+ improvements to lsof, particularly when the files of
+ interest are /Internet files -- i.e., selected with lsof's
+ -i" option. Added a two new options to assist the
+ improvements: 1) "-c^<command>" to tell lsof to exclude the
+ named command(s); and 2) "-stcp|ud>:[^]state' to tell lsof
+ to include in its reporting or exclude ('^') from its
+ reporting Internet files in the named states (e.g., LISTEN,
+ ^CLOSE_WAIT, IDLE, etc.) For the most part these changes
+ apply only to AIX, Darwin, FreeBSD, PSTAT-based HP-UX, Linux
+ and Solaris, since those are the only places I could test
+ them. They are controlled by the HASTCPUDPSTATE definition
+ in each dialect's machine.h header file. Marjo and Phil
+ provided HP-UX 11.23 and 11.31 test systems.
+
+ Fixed a stat(2) problem on HP-UX 11.31 while testing the
+ speed improvements.
+
+ Adjusted for kernel header file changes in FreeBSD
+ 8.0-CURRENT. Larry Rosenman <ler@lerctr.org> provided
+ a test system.
+
+ Added a warning for Solaris systems where VxFS node info
+ can't be obtained from the VxFS utility library. The
+ warning was requested by Tom Matthews <Tom.MATTHEWS@rbs.com>.
+
+ Corrected mishandling of file system path name arguments
+ that have trailing slashes, except, of course, the root
+ file system, "/". Allan Nathanson <ajn@apple.com> reported
+ the bug.
+
+4.82 March 25, 2009
+ Corrected an over-zealous exclusion test that caused
+ lsof to report nothing when it was given no arguments
+ and built with HASSECURITY and HASNOSOCKSECURITY enabled.
+ Joshua Kinard <kumba@gentoo.org> reported the bug and
+ supplied information for reproducing it.
+
+ Based on a report from Dan Trinkle <trinkle@cs.purdue.edu>
+ corrected use of <sys/utsname.h> for 32 bit Solaris 10
+ and above compilations. Simultaneously eliminated a
+ casting complaint in arg.c and updated Configure to use
+ the appropriate 64 bit compilation option (-xarch=v9 or
+ -m64) with the Solaris Sun C compiler.
+
+ Updated for FreeBSD 7.1-PRERELEASE with information
+ supplied by Larry Rosenman <ler@lerctr.org>.
+
+ Updated the Darwin libproc sources with changes from
+ Allan Nathanson <ajn@apple.com>. Tested them on a iMac
+ mini, provided by Apple Inc. Allan also provided man
+ page corrections.
+
+ Updated the FreeBSD Makefile to use the ${MAKE} variable
+ for ZFS dnode2.c module compilation, based on a suggestion
+ from Alexis Ballier <aballier@gentoo.org>.
+
+ Improved the Solaris VxFS library location test, based on a
+ suggestion from Jason Fortezzo <fortezza@mechanicalism.net>.
+ Jason tested the change.
+
+ Updated Solaris 10 ZFS support for ZFS version 4 and ZFS
+ pool version 10, using a test system kindly provided by
+ Vladislav Nespor <vladislav.nespor@id.ethz.ch>. Renata
+ Maria Dart <renata@slac.stanford.edu> tested on ZFS
+ version 4, verifying that the update works there, too.
+ (ZFS pool version 10 is apparently the ZFS version shipped
+ with the 10/08 update to Solaris. The original ZFS
+ support targeted ZFS version 3.)
+
+ I still consider ZFS support in Solaris lsof a hack,
+ because it depends on a znode structure definition that
+ I developed using dbx. Sun is remiss in not distributing
+ the ZFS header files used to build the distributed kernel.
+
+ Because of the znode structure definition hack, I can't
+ guarantee that lsof ZFS support will work for any other
+ versions of ZFS.
+
+ Solaris 10: adjusted to a change in the way devices are
+ stored in the kernel; fixed a problem in zone handling;
+ and added rudimentary sharedfs support. Carson Gaspar
+ <carson@taltos.org> reported the device number problem,
+ provided a test system, and tested the changes. Peter
+ Vines <psv2b@eservices.virginia.edu> reported the zone
+ handling problem and tested the fix.
+
+ Adapted to FreeBSD 8.0-CURRENT changes in device number
+ computation. Problem was reported by Erwin Lansing
+ <erwin@FreeBSD.org>. Larry Rosenman <ler@lerctr.org>
+ provided a test system.
+
+ Corrected Solaris Configure test for appropriate VxFS
+ library when using gcc to compile lsof.
+
+ Updated for loss of KAME IPv6 FreeBSD accommodations.
+
+ Adapted to FreeBSD 7.2. Made Configure script recognized
+ FreeBSD 6.3.
+
+Vic Abell <abe@purdue.edu>
+March 25, 2009
--- /dev/null
+
+ Frequently Asked Questions about lsof
+
+**********************************************************************
+| The latest release of lsof is always available via anonymous ftp |
+| from lsof.itap.purdue.edu. Look in pub/lsof.README for its |
+| location. |
+**********************************************************************
+
+______________________________________________________________________
+
+This file contains frequently asked questions about lsof and answers
+to them.
+
+Vic Abell <abe@purdue.edu>
+March 25, 2009
+______________________________________________________________________
+
+Table of Contents:
+
+1.0 General Concepts
+1.1 Lsof -- what is it?
+1.2 Where do I get lsof?
+1.2.1 Are there mirror sites?
+1.2.2 Are lsof executables available?
+1.2.3 How do I check the validity of an lsof distribution?
+1.2.4 Why can't I get the sum(1) result reported in
+ README.lsof_<revision>?
+1.2.5 Why won't gpg accept the lsof-signing PGP public key?
+1.3 Where can I get more lsof documentation?
+1.4 How do I report an lsof bug?
+1.5 Where can I get the lsof FAQ?
+1.5.1 How timely is the on-line FAQ?
+1.6 Is there a test suite?
+1.7 Is lsof vulnerable to the standard I/O descriptor attack?
+1.8 Can I alter lsof's make(1) behavior?
+1.9 Is there an lsof license?
+1.10 Language locale support
+1.10.1 Does lsof support language locales? How do I use the support?
+1.10.2 Does lsof support wide characters in language locales?
+1.11 Are any files in the lsof distribution copyrighted?
+1.12 Are there other lsof-related resources?
+
+2.0 Lsof Ports
+2.1 What ports exist?
+2.2 What about a new port?
+2.2.1 User-contributed Ports
+2.3 Why isn't there an AT&T SVR4 port?
+2.4 Why isn't there an SGI IRIX port?
+2.5 Why does lsof's Configure script report "WARNING: unsupported
+ dialect or version"?
+
+3.0 Lsof Problems
+3.1 Configuration Problems
+3.1.1 Why can't Configure determine the UNIX dialect version?
+3.2 Compilation Problems
+3.2.1 Why does the compiler complain about missing header files?
+3.2.2 Why does gcc complain about the contents of header files
+ distributed by the system's vendor?
+3.2.3 Other header file problems
+3.3 Why doesn't lsof report full path names?
+3.3.1 Why do lsof -r reports show different path names?
+3.3.2 Why does lsof report the wrong path names?
+3.3.3 Why doesn't lsof report path names for unlinked (rm'd) files?
+3.3.4 Why doesn't lsof report the "correct" hard linked file path
+ name?
+3.3.5 When will lsof report path names for deleted files?
+3.4 Why is lsof so slow?
+3.5 Why doesn't lsof's setgid or setuid permission work?
+3.6 Does lsof have security problems?
+3.7 Will lsof show remote hosts using files via NFS?
+3.8 Why doesn't lsof report locks held on NFS files?
+3.8.1 Why does lsof report a one byte lock on byte zero as a full
+ file lock?
+3.9 Why does lsof report different values for open files on the
+ same file system (the automounter phenomenon)?
+3.10 Why don't lsof and netstat output match?
+3.10.1 Why can't lsof find accesses to some TCP and UDP ports?
+3.11 Why does lsof update the device cache file?
+3.12 Why doesn't lsof report state for UDP socket files?
+3.13 I am editing a file with vi; why doesn't lsof find the file?
+3.14 Why doesn't lsof report TCP/TPI window and queue sizes for my
+ dialect?
+3.14.1 Why doesn't lsof report socket options, socket states, and TCP
+ flags and values for my dialect?
+3.14.2 Why doesn't lsof report the partial listen queue connection
+ count for my dialect?
+3.15 What does "no more information" in the NAME column mean?
+3.16 Why doesn't lsof find a process that ps finds?
+3.17 Why doesn't -V report a search failure?
+3.18 Portmap problems
+3.18.1 Why isn't a name displayed for the portmap registration?
+3.18.2 How can I display only portmap registrations?
+3.18.3 Why doesn't lsof report portmap registrations for some ports?
+3.19 Why is `lsof | wc` bigger than my system's open file limit?
+3.20 Why doesn't lsof report file offset (position)?
+3.20.1 What does lsof report for size when the file doesn't really have
+ one?
+3.21 Problems with path name arguments
+3.21.1 How do I ask lsof to search a file system?
+3.21.2 Why doesn't lsof find all the open files in a file system?
+3.21.3 Why does the lsof exit code report it didn't find open files
+ when some files were listed?
+3.21.4 Why won't lsof find all the open files in a directory?
+3.21.5 Why are the +D and +d options so slow?
+3.21.6 Why do the +D and +d options produce warning messages?
+3.22 Why can't my C compiler find the rpcent structure definition?
+3.23 Why doesn't lsof report fully on file "foo" on UNIX dialect
+ "bar?"
+3.24 Why do I get a complaint when I execute lsof that some library
+ file can't be found?
+3.25 Why does lsof complain it can't open files?
+3.26 Why does lsof warn "compiled for x ... y; this is z."?
+3.27 How can I disable the kernel identity check?
+3.28 Why don't ps(1) and lsof agree on the owner of a process?
+3.29 Why doesn't lsof find an open socket file whose connection
+ state is past CLOSE_WAIT?
+3.30 Why don't machine.h definitions work when the surrounding
+ comments are removed?
+3.31 What do "can't read inpcb at 0x...", "no protocol control
+ block", "no PCB, CANTSENDMORE, CANTRCVMORE", etc. mean?
+3.32 What do the "unknown file system type" warnings mean?
+3.33 Installation
+3.33.1 How do I install lsof?
+3.33.2 How do I install a common lsof when I have machines that
+ need differently constructed lsof binaries?
+3.34 Why do lsof 4.53 and above reject device cache files built
+ by earlier lsof revisions?
+3.35 What do "like block special" and "like character special" mean
+ in the NAME column?
+3.36 Why does an lsof make fail because of undefined symbols?
+3.37 Command Regular Expressions (REs)
+3.37.1 What are basic and extended regular expressions?
+3.37.2 Why can't I put a slash in a command regular expression?
+3.37.3 Why does lsof say my command regular expression wasn't found?
+3.38 Why doesn't lsof report on shared memory segments?
+3.39 Why does lsof report two instances of itself?
+3.40 Why does lsof report '\n' in device cache file error messages?
+3.41 Kernel Symbol and Address Problems
+3.41.1 What does "lsof: WARNING: name cache hash size length error: 0"
+ mean?
+3.41.2 Why does lsof produce "garbage" output?
+3.42 Why does lsof report open files when run as super user that
+ it doesn't report when run with lesser privileges?
+3.43 Test Suite Problems
+3.43.1 Errors all tests can report:
+3.43.1.1 Why do tests complain "ERROR!!! can't execute ../lsof"?
+3.43.1.2 Why do tests complain "ERROR!!! can't find ..." a file?
+3.43.1.3 Why do some tests fail to compile?
+3.43.1.4 Why do some tests always fail?
+3.43.1.5 Why does the test suite say it hasn't been validated on
+ my dialect?
+3.43.1.6 Why do the tests complain they can't stat() or open()
+ /dev/mem or /dev/kmem?
+3.43.2 LTbigf test issues
+3.43.2.1 Why does the LTbigf test say that the dialect doesn't
+ support large files?
+3.43.2.2 Why does LTbigf complain about operations on its config.LTbigf*
+ file?
+3.43.2.3 Why does LTbigf warn that lsof doesn't return file offsets?
+3.43.3 Why does the LTbasic test complain "ERROR!!! lsof this ..."
+ and "ERROR!!! lsof that ..."?
+3.43.4 LTnfs test issues
+3.43.4.1 Why does the LTnfs test complain "couldn't find NFS file ..."?
+3.43.5 LTnlink test issues
+3.43.5.1 Why does the LTnlink test complain that its test file is on
+ an NFS file system?
+3.43.5.2 Why does LTnlink delay and report "waiting for link count
+ update: ..."?
+3.43.6 LTdnlc test issues
+3.43.6.1 Why won't the LTdnlc test run?
+3.43.6.2 What does the LTdnlc test mean by "... <path> found: 100.00%"?
+3.43.6.3 Why does the DNLC test fail?
+3.43.7 Why hasn't the test suite been qualified for 64 bit HP-UX
+ 11 when lsof is compiled with gcc?
+3.43.8 LTszoff test issues
+3.43.8.1 Why does LTszoff warn that lsof doesn't return file offsets?
+3.43.9 LTlock test issues
+3.44 File descriptor list (the ``-d'' option) problems
+3.44.1 Why does lsof reject a ``-d'' FD list?
+3.44.2 Why are file descriptors other than those in my FD list
+ reported?
+3.45 How can I supply device numbers for inaccessible NFS file
+ systems?
+3.46 Why won't lsof find open files on over-mounted file systems?
+3.47 What can be done when lsof reports no more space?
+3.48 What if the lsof build encounters ar and ld problems?
+
+4.0 AIX Problems
+4.1 What is the Stale Segment ID bug and why is -X needed?
+4.1.1 Stale Segment ID APAR
+4.2 Gcc Work-around for AIX 4.1x
+4.3 Gcc and AIX 4.2
+4.4 Why won't lsof's Configure allow the use of gcc for AIX
+ below 4.1?
+4.5 What is an AIX SMT file type?
+4.6 Why does AIX lsof start so slowly?
+4.7 Why does exec complain it can't find libc.a[shr.o]?
+4.8 What does lsof mean when it says, "TCP no PCB, CANTSENDMORE,
+ CANTRCVMORE" in a socket file's NAME column?
+4.9 When the -X option is used on AIX 4.3.3, why does lsof disable
+ it, saying "WARNING: user struct mismatch; -X option disabled?"
+4.10 Why doesn't the -X option work on my AIX 5L or 5.[123] system?
+4.11 Why doesn't /usr/bin/oslevel report the correct AIX version?
+4.11.1 Why doesn't /usr/bin/oslevel report the correct AIX version
+ on AIX 5.1?
+4.12 Why does lsof for AIX 5.1 or above Power architecture
+ complain about kernel bit size?
+4.13 What can't gcc be used to compile lsof on the ia64 architecture
+ for AIX 5 and above?
+4.14 Why does lsof get a segmentation fault when compiled with gcc
+ for a 64 bit Power architecture AIX 5.1 kernel?
+4.15 Why does lsof ignore AFS on my AIX system?
+4.16 Why does lsof report "system paging space is low" and exit?
+4.17 Why does lsof have compilation and execution problems on AIX
+ 5.3 above maintenance level 1?
+
+5.0 Apple Darwin Problems
+5.1 What do /dev/kmem-based and libproc-based mean?
+5.2 /dev/kmem-based Apple Darwin Questions
+5.2.1 Why does Configure ask for a path to the Darwin XNU kernel
+ header files?
+5.2.1.1 Why does Configure complain that Darwin XNU kernel header
+ files are missing?
+5.2.2 Why doesn't Apple Darwin lsof report text file information?
+5.2.3 Why doesn't Apple Darwin lsof support IPv6?
+5.2.4 Why does lsof complain about a mismatch between the release
+ for which lsof was compiled and the booted Mac OS X release?
+5.2.5 Why does lsof for Apple Darwin 8 and higher report
+ "stat(...): ..." in the NAME column?
+5.2.6 What are the limitations of Apple Darwin lsof link count
+ reporting?
+5.3 Libproc-based Apple Darwin Questions
+
+6.0 BSD/OS BSDI Problems
+6.0.5 Statement of deprecation
+
+7.0 DEC OSF/1, Digital UNIX, and Tru64 UNIX Problems
+7.1 Why does lsof complain about non-existent /dev/fd entries?
+7.2 Why does the Digital UNIX V3.2 ld complain about Ots* symbols?
+7.3 Why can't lsof locate named pipes (FIFOs) under V3.2?
+7.4 Why does lsof use the wrong configuration header files?
+ For example, why can't the lsof compilation find cpus.h?
+7.5 Why does lsof indicate incomplete paths with " -- " for Tru64
+ UNIX 5.1 files?
+7.6 Why doesn't lsof report link count, node number, and size
+ for some Tru64 5.x CFS files?
+7.7 Why does lsof say it can't read the kernel name list or
+ proc table on Digital UNIX 4.x or Tru64 UNIX?
+
+8.0 FreeBSD Problems
+8.1 Why doesn't lsof report on open kernfs files?
+8.2 Why doesn't lsof work on my FreeBSD system?
+8.3 Why doesn't lsof work on the RELEASE version of CURRENT?
+8.4 Why can't kvm_open() can't find some file?
+8.5 FreeBSD ZFS Problems
+8.5.1 Why does FreeBSD lsof report "WARNING: no ZFS support has been
+8.6 Why can't Configure create lsof_owner.h for FreeBSD 6 and above?
+8.6.1 Why are there lockf structure compiler errors for FreeBSD 6.0
+ and higher lsof?
+8.6.2 Why don't /usr/src/sys/sys/lockf.h and /usr/include/sys/lockf.h
+ match?
+
+9.0 HP-UX Problems
+9.1 What do /dev/kmem-based and PSTAT-based mean?
+9.2 /dev/kmem-based HP-UX lsof Questions
+9.2.1 Why doesn't a /dev/kmem-based HP-UX lsof compilation use -O?
+9.2.2 Why doesn't the /dev/kmem-based CCITT support work under 10.x?
+9.2.3 Why can't /dev/kmem-based lsof be compiled with `cc -Aa` or
+ `gcc -ansi` under HP-UX 10.x?
+9.2.4 Why does /dev/kmem-based lsof complain about no C compiler?
+9.2.5 Why does Configure complain about q4 for /dev/kmem-based lsof
+ for HP-UX 11?
+9.2.6 When compiling /dev/kmem-based lsof for HP-UX 11 what do the
+ "aCC runtime: ERROR..." messages mean?
+9.2.7 Why doesn't /dev/kmem-based lsof for HP-UX 11 report VxFS file
+ link counts, node numbers, and sizes correctly?
+9.2.8 Why can't /dev/kmem-based lsof be built with gcc for 64 bit
+ HP-UX 11?
+9.2.8.1 How can I acquire a gcc for building lsof for 64 bit HP-UX 11?
+9.2.9 Why does /dev/kmem-based lsof for HP-UX 11 report "unknown file
+ system type" for VxFS files?
+9.2.10 Why does the ANSI-C compiler complain about comments in HP-UX
+ 11 header files?
+9.2.11 Why does dnode1.c cause the HP-UX 11 compiler to complain that
+ <sys/fs/vx_inode.h> is missing or incorrect?
+9.3 PSTAT-based HP-UX lsof Questions
+9.3.1 Why does PSTAT-based lsof complain about pst_static and
+ other PSTAT structures?
+9.3.2 Why does PSTAT-based lsof complain it can't read pst_*
+ structures?
+9.3.3 Why does PSTAT-based lsof rebuild the device cache file
+ after each reboot?
+9.3.4 Why doesn't PSTAT-based lsof report TCP addresses for
+ telnetd's open socket files?
+9.3.5 Why does PSTAT-based lsof cause an HP-UX 11.11 kernel panic?
+9.3.6 Why doesn't PSTAT-based lsof report a CWD that is on a loopback
+ (LOFS) file system?
+9.3.7 Why do some swinstall packages for PSTAT-based HP-UX 11.11
+ packages complain about setgid and setuid bits?
+9.3.8 Why won't the bundled C compiler build PSTAT-based lsof for
+ PA-RISC HP-UX 11.23?
+9.3.9 Why won't gcc build PSTAT-based lsof for PA-RISC HP-UX 11.23?
+9.3.10 Why does PSTAT-based lsof complain, "FATAL: pst_stream_size
+ should be: 672; is 72" on HP-UX 11.11 and above?
+9.4 Why won't the HP-UX depot install?
+
+10.0 Linux Problems
+10.1 What do /dev/kmem-based and /proc-based lsof mean?
+10.2 /proc-based Linux lsof Questions
+10.2.1 Why doesn't /proc-based lsof report file offsets (positions)?
+10.2.2 Why does /proc-based lsof report "can't identify protocol" for
+ some socket files?
+10.2.3 Why does /proc-based lsof warn about unsupported formats?
+10.2.4 Why does /proc-based lsof report "(deleted)" after a path name?
+10.2.5 Why doesn't /proc-based lsof report full open file information
+ for all processes?
+10.2.6 Why won't Customize offer to change HASDCACHE or WARNDEVACCESS
+ for /proc-based lsof?
+10.2.7 /proc-based lsof Linux NFS questions
+10.2.7.1 Why can't lsof find files on an accessible NFS file system?
+10.2.7.2 Why can't lsof find files on an inaccessible NFS file system?
+10.2.8 Why doesn't /proc-based Linux lsof report socket options and
+ values, socket state flags, and TCP options and values?
+10.2.9 Does /proc-based Linux lsof use a device cache?
+10.2.10 Why doesn't /proc-based Linux lsof report any or all file structure
+ values for its +fcfgGn option?
+10.3 Special Linux file types
+10.3.1 Why is ``DEL'' reported as a Linux file type?
+10.3.2 Why is ``unknown'' reported as a Linux file type?
+10.4 Linux ``mem'' Entry Problems
+10.4.1 What do ``path dev=xxx'' and ``path inode=yyy'' mean in the
+ NAME column of Linux ``mem'' file types?
+10.4.2 Why is neither link count nor size reported for some Linux
+ ``DEL'' and ``mem'' file types?
+10.5 Special Linux NAME column messages
+10.5.1 What does ``(stat: xxx)'' mean in the NAME column of Linux
+ files?
+10.5.2 What does ``(readlink: xxx)'' mean in the NAME column of
+ Linux files?
+10.6 Why is ``NOFD'' reported as a Linux file type?
+10.7 Why does Linux lsof report a NAME column value that begins with
+ ``/proc''?
+10.8 Linux /proc/net/tcp* and /proc/net/udp* issues
+10.8.1 Why use the Linux -X option?
+10.8.2 Why does lsof say ``-i is useless when -X is specified''?
+10.8.3 Why does lsof say ``can't identify protocol (-X specified)''?
+
+11.0 NetBSD Problems
+11.1 Why doesn't lsof report on open kernfs files?
+11.2 Why doesn't lsof report on open files on: file descriptor
+ file systems; /proc file systems; 9660 (CD-ROM) file systems;
+ MS-DOS (floppy disk) file systems; or kernel file systems?
+11.3 Why does lsof produce confusing results for nullfs file
+ systems?
+11.4 NetBSD header file problems
+11.4.1 Why can't the compiler find some NetBSD header files?
+11.4.2 Why does NetBSD lsof produce incorrect output?
+11.5 Why isn't lsof feature xxx enabled for NetBSD?
+
+12.0 NEXTSTEP and OPENSTEP Problems
+12.1 Why can't lsof report on 3.1 lockf() or fcntl(F_SETLK)
+ locks?
+12.2 Why doesn't lsof compile for NEXTSTEP with AFS?
+
+13.0 OpenBSD Problems
+13.1 Why doesn't lsof support kernfs on my OpenBSD system?
+13.2 Will lsof work on OpenBSD on non-x86-based architectures?
+13.3 <sys/pipe.h> problems
+13.3.1 Why does the compiler claim nbpg isn't defined?
+13.3.2 What value should I assign to nbpg?
+13.4 Why doesn't lsof report on open MS-DOS file system (floppy
+ disk) files?
+13.5 Why isn't lsof feature xxx enabled for OpenBSD?
+
+14.0 Output problems
+14.1 Why do the lsof column sizes change?
+14.2 Why does the offset have ``0t' and ``0x'' prefixes?
+14.3 What are the values printed in the FILE_FLAG column
+ and why is 0x<value> sometimes included?
+14.3.1 Why doesn't lsof display FILE_FLAG values for my dialect?
+14.4 Network Addresses
+14.4.1 Why does lsof's -n option cause IPv4 addresses, mapped to
+ IPv6, to be displayed in IPv6 notation?
+14.5 Why does lsof output \x, ^x, or \xnn for characters
+ sometimes?
+14.5.1 Why is space considered a non-printable character in command
+ names?
+14.6 Why doesn't lsof print all the characters of a command name?
+14.7 Why does lsof reject some -c command names, saying their lengths
+ are "> what system provides (nn)"?
+14.8 Why does lsof sometimes print TYPE numbers instead of names?
+14.9 Marker line format problems
+14.9.1 Why won't lsof accept a marker line format?
+14.9.2 Why does lsof reject the NL (%n) marker line format?
+14.10 How are protocol state name exclusion and inclusion used?
+14.10.1 Why doesn't my dialect support state name exclusion and inclusion?
+
+15.0 Pyramid Version Problems
+15.0.5 Statement of deprecation
+
+16.0 SCO Problems
+16.1 SCO OpenServer Problems
+16.1.1 How can I avoid segmentation faults when compiling lsof?
+16.1.2 Where is libsocket.a?
+16.1.3 Why do I get "warning C4200" messages when I compile lsof?
+16.2 SCO|Caldera UnixWare Problems
+16.2.1 Why doesn't lsof compile on my UnixWare 7.1.1 or above
+ system?
+16.2.2 Why does lsof complain about node_self() on my UnixWare
+ 7.1.1 or above system?
+16.2.3 Why does UnixWare 7.1.1 or above complain about -lcluster,
+ node_self(), or libcluster.so?
+16.2.4 Why does UnixWare 7.1.1 or above lsof complain it can't
+ read the kernel name list?
+16.2.5 Why doesn't lsof report link count, node number, and size
+ for some UnixWare 7.1.1 or above CFS files?
+16.2.6 Why doesn't lsof report open files on all UnixWare 7.1.1
+ NonStop Cluster (NSC) nodes?
+16.2.7 Why doesn't lsof report the UnixWare 7.1.1 NonStop Cluster
+ (NSC) node a process is using?
+16.2.8 Why does the compiler complain about missing UnixWare 2.1[.x]
+ header files?
+
+17.0 Sun Problems
+17.0.5 Statement of deprecation
+17.1 My Sun gcc-compiled lsof doesn't work -- why?
+17.2 How can I make lsof compile with gcc under Solaris 2.[456],
+ 2.5.1, 7, 8 or 9?
+17.3 Why does Solaris Sun C complain about system header files?
+17.4 Why doesn't lsof work under my Solaris 2.4 system?
+17.5 Where are the Solaris header files?
+17.6 Where is the Solaris /usr/src/uts/<architecture>/sys/machparam.h?
+17.7 Why does Solaris lsof say ``can't read proc table''?
+17.8 Why does Solaris lsof complain about a bad cached clone device?
+17.9 Why doesn't Solaris make generate .o files?
+17.10 Why does lsof report some Solaris 2.3 and 2.4 lock types as `N'?
+17.11 Why does lsof Configure say "WARNING: no cc in ..."?
+17.12 Solaris 7, 8 and 9 Problems
+17.12.1 Why does lsof say the compiler isn't adequate for Solaris
+ 7, 8 or 9?
+17.12.2 Why does Solaris 7, 8 or 9 lsof say "FATAL: lsof was compiled
+ for..."?
+17.12.3 How do I build lsof for a 64 bit Solaris kernel under a 32
+ bit Solaris kernel?
+17.12.4 How do I install lsof for Solaris 7, 8 or 9?
+17.12.5 Why does my Solaris 7, 8 or 9 system say it cannot execute
+ lsof?
+17.12.6 What gcc will produce 64 bit Solaris 7, 8 and 9 executables?
+17.12.7 Why does lsof on my Solaris 7, 8 or 9 system say, "can't
+ read namelist from /dev/ksyms?"
+17.13 Solaris and COMMON
+17.13.1 What does COMMON mean in the NAME column for a Solaris VCHR
+ file?
+17.13.2 Why does a COMMON Solaris VCHR file sometimes seem to have an
+ incorrect minor device number?
+17.14 Why don't lsof and Solaris pfiles reports always match?
+17.15 Why does lsof say, "kvm_open(namelist=default, core=default):
+ Permission denied?"
+17.16 Why is lsof slow on my busy Solaris UFS file system?
+17.17 Why is lsof so slow on my Solaris 8 or 9 system?
+17.18 Solaris and VxFS
+17.18.1 Why doesn't lsof support VxFS 3.4 on Solaris 2.6, and above?
+17.18.2 Why does lsof report "vx_inode: vxfsu_get_ioffsets error"
+ for open Solaris 2.6 and above VxFS 3.4 and above files?
+17.18.3 Why does Solaris Configure claim there is no VxFS library?
+17.18.4 Why doesn't Solaris lsof report VxFS path name components?
+17.18.5 Why does Solaris 10 lsof report scrambled VxFS paths?
+17.19 Large file problems
+17.19.1 Why does lsof complain it can't stat(2) a Solaris 2.5.1
+ large file?
+17.20 Why does lsof get a segmentation fault on 64 bit Solaris
+ 8 using NIS+?
+17.21 Will lsof crash the Solaris kernel?
+17.22 Why does lsof on Solaris 7, 8, or 9 report a kvm_open()
+ failure?
+17.23 Solaris and SAM-FS
+17.23.1 Why does Solaris lsof report "(limited SAM-FS info)"?
+17.23.2 Why can't lsof locate named SAM-FS files?
+17.24 Lsof and Solaris 10 zones
+17.24.1 How can I make lsof list the Solaris zone?
+17.24.2 Why doesn't lsof work in a Solaris 10 zone?
+17.24.3 Why does lsof complain it can't stat() Solaris 10 zone file
+ systems?
+17.25 Solaris 10 problems
+17.25.1 Why does Solaris 10 lsof sometimes report the wrong path name?
+17.25.2 Why does Solaris 10 lsof sometimes report only the mounted-on
+ directory and device?
+17.25.3 What does "(deleted)" mean in the NAME column of a Solaris 10
+ open file?
+17.25.4 What does "(?)" mean in the NAME column of a Solaris 10 open
+ file?
+17.26 Solaris contract file problems
+17.26.1 Why doesn't lsof report size, link count and node number for
+ Solaris 10 contract files?
+17.26.2 Why can't lsof locate a Solaris 10 contract file by path name?
+17.27 Solaris 10 ZFS probblems
+17.27.1 Why does Configure ask for the location of ZFS header files?
+17.27.2 Why do -h and -v output warn about possibly inaccurate ZFS
+ structure definitions?
+17.27.3 Why don't the Open Solaris ZFS header files provide correct
+ ZFS kernel structure definitions?
+
+18.0 Lsof Features
+18.1 Why doesn't lsof doesn't report on /proc entries on my
+ system?
+18.2 How do I disable the device cache file feature or alter
+ it's behavior?
+18.2.1 What's the risk with a perverted device cache file?
+18.2.2 How do I put the full host name in a personal device cache file
+ path?
+18.2.3 How do I put the personal device cache file in /tmp?
+18.3 Why doesn't lsof know about AFS files on my favorite dialect?
+18.3.1 Why doesn't lsof report node numbers for all AFS volume files,
+ or how do I reveal dynamic module addresses to lsof?
+______________________________________________________________________
+
+
+1.0 General Concepts
+
+1.1 Lsof -- what is it?
+
+ Lsof is a UNIX-specific tool. Its name stands for LiSt
+ Open Files, and it does just that. It lists information
+ about files that are open by the processes running on a
+ UNIX system.
+
+ See the lsof man page, the 00DIST file, the 00QUICKSTART
+ file, and the 00README file of the lsof distribution for
+ more information.
+
+1.2 Where do I get lsof?
+
+ Lsof is available via anonymous ftp from lsof.itap.purdue.edu.
+ Look in the pub/tools/unix/lsof sub-directory.
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof
+
+ Bzip2'd, compressed and gzip'd tar files with GPG certificates
+ are available.
+
+1.2.1 Are there mirror sites?
+
+ On September 3, 2003 these sites appeared to mirror lsof:
+
+ ftp://ftp.cerias.purdue.edu/pub/tools/unix/sysutils/lsof
+ ftp://ftp.tau.ac.il/pub/unix/admin
+ ftp://ftp.cert.dfn.de/pub/tools/admin/lsof
+ ftp://ftp.fu-berlin.de/pub/unix/tools/lsof
+ ftp://gd.tuwien.ac.at/utils/admin-tools/lsof
+ ftp://sunsite.ualberta.ca/pub/Mirror/lsof
+
+1.2.2 Are lsof executables available?
+
+ Some lsof executables are available in the subdirectory
+ tree pub/tools/unix/lsof/binaries These are neither guaranteed
+ to be current nor cover every dialect and machine architecture.
+
+ I don't recommend you use pre-compiled lsof binaries; I
+ recommend you obtain the sources and build your own binary.
+ Even if you're a Sun user without a Sun C compiler, you
+ can use gcc to compile lsof.
+
+ If you must use a binary file, please be conscious of the
+ security and configuration implications in using an executable
+ of unknown or different origin. The lsof binaries are
+ accompanied by GPG certificates. Please use them!
+
+ Three additional cautions apply to executables:
+
+ 1. Don't try to use an lsof executable, compiled for one
+ version of a UNIX dialect, on another. Patches can
+ make the dialect version different.
+
+ 2. If you want to use an lsof binary on multiple systems,
+ they must be running the same dialect OS version and
+ have the same patches and feature support.
+
+1.2.3 How do I check the validity of an lsof distribution?
+
+ There are two ways to check the validity of an lsof
+ distribution:
+
+ 1. Follow the instructions in the CHECKSUMS_<revision>
+ file found with the lsof distribution.
+
+ Checking with GPG is the best method.
+
+ 2. Follow the instructions in the "Security" section of the
+ README.lsof_<revision> file found inside the lsof
+ distribution.
+
+ Again, checking with GPG is the best method.
+
+1.2.4 Why can't I get the sum(1) result reported in
+ README.lsof_<revision>?
+
+ The "Security" section of the README.lsof_<revision> file found
+ inside the lsof distribution gives md5, sum, and GPG certificate
+ information.
+
+ The simplest, the sum(1) signature, seems to be the trickiest.
+ That's because there are different sum(1) methods, BSD systems
+ usually have cksum(1) instead of sum(1), and different systems
+ compute the block size value differently.
+
+ First, the lsof sum results are computed with the old,
+ "alternate" algorithm. On newer systems, you can use sum's
+ "-r" option to get that computation result.
+
+ Second, on BSD systems you usually must use cksum(1) instead
+ of sum(1), because they have no sum(1). To tell cksum(1)
+ to use the old, "alternate" algorithm, use its "-o1" option.
+
+ Third, the second value that sum reports, the block count, may
+ be computed differently on different systems -- usually block
+ size is considered to be 512 or 1,024. The lsof block counts
+ were computed on a system with a sum(1) option that considers
+ block size to be 512. The BSD system cksum(1) -o1 option
+ considers block size to be 1,024. If your sum(1) or cksum(1)
+ doesn't report a block count that matches the sum(1) signature
+ given in README.lsof_<revision>, check its man page to see what
+ block size it uses, then adjust its reported block count
+ appropriately.
+
+1.2.5 Why won't gpg accept the lsof-signing PGP public key?
+
+ An older PGP key that once signed lsof distributions is
+ included in lsof revisions prior to 4.70. The PGP key is
+ indeed my key, but is incompatible with GPG. It was created
+ about ten years ago and is still acceptable to PGP versions
+ 2.6.2 through 6.5.2.
+
+ Lsof revisions 4.70 and above are signed with a copy of my PGP
+ key that has been made acceptable for use with GPG by importing
+ it under GPG's "--allow-non-selfsigned-uid" option.
+
+ You can find my GPG compatible key in lsof revisions 4.70 and
+ above and at:
+
+ ftp://lsof.itap.purdue.edu/pub/Victor_A_Abell.gpg
+
+ If you have an older lsof revision with my PGP key, there are
+ two possible ways to use it:
+
+ * Use it with a PGP version from 2.6.2 through 6.5.2.
+
+ * Use GPG's "--allow-non-selfsigned-uid" option when you
+ import my PGP key into your GPG key ring.
+
+ $ gpg --allow-non-selfsigned-uid --import Victor_A_Abell.pgp
+
+1.3 Where can I get more lsof documentation?
+
+ A significant set of documentation may be found in the lsof
+ distribution (See "Where can I get lsof?). There is a
+ manual page, copious documentation in files whose names
+ begin with 00, and a copy of this FAQ in the file 00FAQ
+ (perhaps slightly less recent than this file if you're
+ reading it via a web browser.)
+
+ Two URLs provide some documentation that appears in the
+ lsof distribution:
+
+ FAQ: ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/FAQ
+
+ man page: ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/lsof_man
+
+1.4 How do I report an lsof bug?
+
+ If you believe you have discovered a bug in lsof, you can
+ report it via e-mail to <abe@purdue.edu>. Do NOT report lsof
+ bugs to the UNIX dialect vendor. Make sure "lsof" appears in
+ the "Subject:" line so my e-mail filter won't classify your
+ letter as Spam.
+
+ Before you send me a bug report, please read the "Bug Reports"
+ section of the 00README file of the lsof distribution. It
+ lists the steps you should take before and when reporting a
+ suspected bug.
+
+1.5 Where can I get the lsof FAQ?
+
+ This lsof FAQ is available in the file 00FAQ in the lsof
+ distribution and at the URL:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/FAQ
+
+1.5.1 How timely is the on-line FAQ?
+
+ The on-line FAQ is sometimes too timely. :-)
+
+ I update it as soon as new information is available. That may
+ include information about support that won't appear in the lsof
+ source distribution until the next revision. If you encounter
+ something like that, please send me e-mail at <abe@purdue.edu>. I
+ may be able to point you at a pre-release distribution that contains
+ the support of interest. Make sure "lsof" appears in the "Subject:"
+ line so my e-mail filter won't classify your letter as Spam.
+
+1.6 Is there a test suite?
+
+ Yes, as of lsof revision 4.63 there's an automated lsof
+ test suite in the tests/ sub-directory of the lsof top-level
+ directory.
+
+ More information on using the test suite, what it does,
+ how to use it and how to configure it may be found in the
+ 00TEST file of the lsof distribution. That file also
+ explains where the test suite has been tested.
+
+ Frequently asked questions about the test suite will be
+ asked and answered here in the FAQ. (See "Test Suite
+ Problems.")
+
+ After lsof has been configured with the Configure script,
+ lsof can be made and tested with:
+
+ $ make
+ $ cd tests
+ $ make
+
+ Under normal conditions -- i.e., unless the lsof tree has
+ been cleaned or purged severely -- all tests or individual
+ tests may be run by:
+
+ $ cd test
+ $ make
+ or
+ $ <run a single test> (See 00TEST.)
+
+1.7 Is lsof vulnerable to the standard I/O descriptor attack?
+
+ Lsof revisions 4.63 and above are not vulnerable.
+
+ Lsof revisions 4.62 and below are vulnerable, but no damage
+ scenarios have so far been demonstrated.
+
+ The standard I/O descriptor attack is a local programmed
+ assault on setuid and setgid programs that tricks them into
+ opening a sensitive file with write access on a standard
+ descriptor, usually stderr (2), and writing error messages
+ to stderr. If the attacker can control the content of the
+ error message, the attacker may gain elevated privileges.
+
+ The attack was first described in Pine Internet Advisory
+ PINE-CERT-20020401, available at:
+
+ http://www.pine.nl/advisories/pine-cert-20020401.txt
+
+ If you are using an lsof revision below 4.63, you should
+ remove any setuid or setgid permissions you might have
+ given its executable. Then you should upgrade to lsof
+ revision 4.63.
+
+1.8 Can I alter lsof's make(1) behavior?
+
+ Yes. There are at least two ways to do that.
+
+ You can put replacements for lsof Makefile strings in your
+ environment. If you specify the -e make option, make will
+ give environment variable values precedence over strings
+ from the Makefile. For example, to change the compiler
+ string CC from the environment, you might do this with the
+ Bourne shell:
+
+ $ CC=foobar; export CC
+ $ make -e
+
+ You can also replace lsof Makefile strings in the make
+ command invocation. Here's the previous example done that
+ way:
+
+ $ make CC=foobar
+
+ Changing the CFGF, CFGL, and DEBUG strings used in lsof
+ Makefiles, either from the environment or from the make
+ invocation, can significantly alter lsof make(1) behavior.
+ I commonly use DEBUG to change the -O option to -g so I
+ can build an lsof executable for debugging -- e.g.,
+
+ $ make DEBUG=-g
+
+ (Look for DEBUG in this FAQ for other examples of its use.)
+
+ Consult the Makefiles to see what CFGL, CFGL, and other
+ lsof Makefile strings contain, and to see what influence
+ their alteration might have on lsof make(1) behavior.
+
+1.9 Is there an lsof license?
+
+ No.
+
+ The only restriction on the use or redistribution of lsof
+ is contained in this copyright statement, found in every
+ lsof source file. (The copyright year in or format of the
+ notice may vary slightly.)
+
+ /*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette,
+ * Indiana 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American
+ * Telephone and Telegraph Company or the Regents of the
+ * University of California.
+ *
+ * Permission is granted to anyone to use this software for
+ * any purpose on any computer system, and to alter it and
+ * redistribute it freely, subject to the following
+ * restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible
+ * for any consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented,
+ * either by explicit claim or by omission. Credit to the
+ * authors and Purdue University must appear in documentation
+ * and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must
+ * not be misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+1.10 Language locale support
+
+1.10.1 Does lsof support language locales? How do I use the support?
+
+ Most UNIX dialect versions of lsof support 8 bit language
+ locale characters -- e.g., the ability to print 8 bit
+ characters that have accents and other marks over them.
+
+ See the answer to the "Does lsof support wide characters in
+ language locales?" question for information on when lsof's
+ language locale support covers characters wider than 8 bits.
+
+ To see if lsof supports language locales for your dialect, look
+ in the dialect's machine.h header file for the HASSETLOCALE
+ definition. If it is present and not disabled, then lsof has
+ language locale support for the dialect.
+
+ To enable lsof's language locale support, you must specify in a
+ locale environment variable (e.g., LANG) a language locale
+ known to your system that supports the printing of marked
+ characters -- e.g, en_US. (On some dialects locale(1) may be
+ used to list the known language locales.)
+
+ Note that LANG=C and LANG=POSIX are NOT language locales that
+ support the printing of marked characters.
+
+ If the language locale doesn't support the printing of marked
+ characters, lsof's OUTPUT of them follows the rules for
+ non-printable characters described in the OUTPUT section of
+ lsof(8).
+
+ Consult your dialect's setlocale(3) man page for the names of
+ environment variables other than LANG -- e.g., LC_ALL,
+ LC_TYPE, etc. -- which may be used to define language locales.
+
+1.10.2 Does lsof support wide characters in language locales?
+
+ When lsof's language locale support is enabled with the
+ HASSETLOCALE definition, for selected dialects lsof will also
+ print wide characters (e.g., from UTF-8) when iswprint(3)
+ reports them to be printable.
+
+ Wide character support is available when HASWIDECHAR is defined
+ in a dialect's machine.h header file. As of this writing on
+ July 22, 2004, the following dialect versions have wide character
+ support:
+
+ AIX >= 4.3.2
+ Apple Darwin >= 7.3.0
+ FreeBSD >= 5.2
+ HP-UX >= 11.00
+ /proc-based Linux
+ NetBSD >= 1.6
+ SCO OpenServer >= 5.0.6
+ Solaris >= 2.6
+ Tru64 UNIX 5.1
+
+1.11 Are any files in the lsof distribution copyrighted?
+
+ Yes. Most files carry the copyright of the Purdue Research
+ Foundation and may be redistributed under the terms that
+ accompany the copyright notice. Those terms may also be found
+ in the answer to the question, "Is there an lsof license?")
+
+ A few files carry other copyright notices. Some are BSD
+ notices and they explain the terms under which they are
+ included in the lsof distribution.
+
+ Those that carry vendor copyright notices have been reproduced
+ in their original or modified forms with permission from the
+ copyright owners. That permission is indicated in the README
+ files that accompany the files.
+
+1.12 Are there other lsof-related resources?
+
+ There are other resources available, connected to lsof. Among
+ them are FreeBSD and Linux packages whose products use lsof and
+ two particularly interesting resources.
+
+ The two interesting resources are a Gnome Tool Kit (GTK) GUI
+ for lsof and a Perl wrapper module.
+
+ The GTK GUI is called Glsof and was developed by Gnele. It can
+ be found at:
+
+ http://www.sourceforge.net
+
+ The Perl wrapper module by Marc Beyer can be found at:
+
+ http://search.cpan.org/dist/Unix-Lsof/
+
+
+2.0 Lsof Ports
+
+2.1 What ports exist?
+
+ The pub/lsof.README file carries the latest port information:
+
+ AIX 5.[23] and 5.3
+ FreeBSD 4.9 for x86-based systems
+ FreeBSD 7.[012] and 8.0 for AMD64-based systems
+ Linux 2.1.72 and above for x86-based systems
+ Solaris 9 and 10
+
+ In the above list the only UNIX dialects present are ones for
+ which I test the current lsof revision. Lsof may still support
+ unlisted dialect versions -- e.g., HP-UX 10.20, Solaris 7, etc.
+ -- but I don't have access to systems where I could test lsof
+ on them, so I can't claim lsof works on them. If your dialect
+ isn't in the list, you should try building lsof on it anyway.
+
+ Lsof version 4 predecessors, versions 2 and 3, may support older
+ version of some dialects. Contact me via e-mail at <abe@purdue.edu>
+ if you're interested in their distributions. Make sure "lsof"
+ appears in the "Subject:" line so my e-mail filter won't classify
+ your letter as Spam.
+
+2.2 What about a new port?
+
+ The 00PORTING file in the distribution gives hints on doing
+ a port. I will consider doing a port in exchange for
+ permanent access to a test host. I require permanent access
+ so I can test new lsof revisions, because I will not offer
+ distributions of dialect ports I cannot upgrade and test.
+
+2.2.1 User-contributed Ports
+
+ Sometimes I receive contributions of ports of lsof to
+ systems where I can't test future revisions of lsof. Hence,
+ I don't incorporate these contributions into my lsof
+ distribution.
+
+ However, I do make descriptions of these contributions
+ available. You can find them in the 00INDEX and README
+ files at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/contrib
+
+ Consult the 00INDEX file in the contrib/ directory for a
+ list of the available contributions and consult README
+ there for information on how to obtain them.
+
+2.3 Why isn't there an AT&T SVR4 port?
+
+ I haven't produced an AT&T SVR4 port because I haven't seen
+ a UNIX dialect that is strictly limited to the AT&T System
+ V, Release 4 source code. Every one I have seen is a
+ derivative with vendor additions.
+
+ The vendor additions are significant to lsof because they
+ affect the internal kernel structures with which lsof does
+ business. While some vendor derivatives of SVR4 are similar,
+ each one I have encounted so far has been different enough
+ from its siblings to require special source code.
+
+ If you're interested in an SVR4 version of lsof, here are
+ some existing ports you might consider:
+
+ DC/OSx (This obsolete port is only available upon
+ special request.)
+ Reliant UNIX (This obsolete port is only available
+ upon special request.)
+ SCO|Caldera UnixWare (This is the most likely choice.)
+ Solaris
+
+2.4 Why isn't there an SGI IRIX port?
+
+ Lsof support for IRIX was terminated at lsof revision 4.36,
+ because it had become increasingly difficult for me to
+ obtain information on the IRIX kernel structures lsof needs
+ to access.
+
+ At IRIX 6.5 I decided the obstacles were too large for me
+ to overcome, and I stopped supporting lsof on IRIX. I have
+ sources to the last revision of lsof (4.36) for IRIX, but
+ that version of lsof does not work on IRIX 6.5 and is
+ vulnerable to the standard I/O descriptor attack. (See
+ the "Is lsof vulnerable to the standard I/O descriptor
+ attack?" Q&A for more information.) Contact me to discuss
+ obtaining those sources.
+
+ If you wish to pursue the issue, don't contact me, contact
+ SGI. This case was opened with SGI on the subject:
+
+ Case ID: 0982584
+ Category: Unix
+ Priority: 30-Moderate Impact
+
+ Problem Summary:
+ kernel structure header files needed for continued lsof
+ support
+
+ Problem Description:
+ Email In 07/17/98 19:09:23
+
+2.5 Why does lsof's Configure script report "WARNING: unsupported
+ dialect or version"?
+
+ Lsof's Configure script issues this message when it encounters
+ a dialect or its version that lsof once supported, but no
+ longer does. Usually I drop support for a dialect or version
+ when I can no longer test lsof on it.
+
+ However, it's worth trying to compile and use lsof. Be sure to
+ run the test suite. (See the answer to the "Is there a test
+ suite? question for information on the test suite.)
+
+ If you have problems with an unsupported dialect or version,
+ contact me via e-mail at <abe@purdue.edu> and I may be able to help.
+ Make sure "lsof" appears in the "Subject:" line so my e-mail filter
+ won't classify your letter as Spam.
+
+
+3.0 Lsof Problems
+
+3.1 Configuration Problems
+
+3.1.1 Why can't Configure determine the UNIX dialect version?
+
+ The lsof Configure script uses UNIX shell commands, often in a
+ command pipeline, to determine the UNIX dialect version.
+ (Consult the dialect stanza in Configure to determine which
+ commands are used.) If Configure can't determine the dialect
+ version, probably one of the commands is not behaving as
+ Configure expects.
+
+ Symptoms of the failure include Configure warning messages and
+ incorrect version definitions in the Makefile CFLAGS.
+
+ If you suspect that the lsof Configure script is failing to
+ determine the dialect version correctly, try running the
+ commands from Configure stanza one at a time. That will
+ usually reveal the source of the problem. Be particularly
+ mindful that the PATH environment variable can cause commands
+ to be executed from non-standard directories.
+
+ If you can't determine the source of the problem, there is a
+ work-around. You can supply the UNIX dialect version in the
+ LSOF_VSTR environment variable. Use Configure as a guide to
+ forming what it expects in LSOF_VSTR. There is also some
+ information on LSOF_VSTR in the 00XCONFIG documentation file
+ of the lsof distribution.
+
+3.2 Compilation Problems
+
+3.2.1 Why does the compiler complain about missing header files?
+
+ When you use make to build lsof, the compiler may complain
+ that it can't find header files -- e.g.,
+
+ $ make
+ (cd lib; make DEBUG="-O" CFGF="-DAIXA=0 -DAIXV=4330 \
+ -DLSOF_VSTR=\"4.3.3.0\"")
+ gcc -DAIXA=0 -DAIXV=4330 -DLSOF_VSTR="4.3.3.0" -O \
+ -c ckkv.c
+ In file included from ckkv.c:33: ../machine.h:70: \
+ sys/types.h: A file or directory in the path name \
+ does not exist. \
+
+ That type of complaint doesn't represent an lsof problem.
+ It represents a problem with a missing system header file
+ that probably should be found in /usr/include or in the
+ system source tree.
+
+ As a first step try using find(1) to locate the problem
+ header file. If it's a system header file and can't be
+ found, here are some possible causes:
+
+ 1. The file set, RPM or package containing the header files
+ has not been installed. instructions for doing that
+ are specific to the UNIX dialect and beyond the scope
+ of this document.
+
+ 2. If the compiler is gcc, the private gcc header files:
+
+ * May not have been installed;
+
+ * May have been installed incorrectly;
+
+ * May not have been updated properly after the last
+ compiler or system update;
+
+ * Ones from a previous installation may not have been
+ removed.
+
+ A path leading to the gcc private header files can be
+ found with `gcc -v`. Consult the gcc documentation for
+ instructions on proper installation of the private gcc
+ header files.
+
+ 3. On some dialects -- e.g., FreeBSD, NetBSD, OpenBSD --
+ lsof may need to use header files that are located in
+ the system source tree -- /sys or /usr/src/sys, for
+ example. Make sure the system source tree has been
+ installed.
+
+3.2.2 Why does gcc complain about the contents of header files
+ distributed by the system's vendor?
+
+ When you use make to build lsof and gcc to compile it, gcc
+ may complain that it finds errors in system header files
+ -- e.g.,
+
+ $ make
+ (cd lib; make DEBUG="-O" CFGF="-Dsolaris=80000 \
+ -DHASPR_GWINDOWS -m64 -DHASIPv6 -DHAS_VSOCK \
+ -DLSOF_VSTR=\"5.8\"")
+ gcc -Dsolaris=80000 -DHASPR_GWINDOWS -m64 -DHASIPv6 \
+ -DHAS_VSOCK -DLSOF_VSTR="5.8" -O -c dvch.c
+ In file included from /usr/include/sys/proc.h:31, \
+ from /homes/abe/gnu/gcc-3.2.1/lib/gcc-lib/sparcv9-sun-solaris2/ \
+ 3.2.1/include/sys/user.h:267, from /usr/include/kvm.h:13, \
+ from ../dlsof.h:53, from ../lsof.h:172, from dvch.c:43: \
+ /homes/abe/gnu/gcc-3.2.1/lib/gcc-lib/sparcv9-sun-solaris2/\
+ 3.2.1/include/sys/task.h:59: parse error before "uint_t"
+
+ Errors like the above are most likely not problems in the
+ system's header files, but in the private copies of them
+ that were created when gcc was made or installed. Note
+ the presense of
+ ".../gcc-3.2.1/lib/gcc-lib/sparcv9-sun-solaris2/3.2.1/include/..."
+ in the paths for user.h and task.h. It indicates both
+ header files are gcc-specific.
+
+ To solve errors like this requires comparing the header
+ files in the vendor's /usr/include tree to the gcc-specific
+ ones in gcc's private gcc-lib/.../include tree. It may be
+ necessary to regenerate gcc-specific header files, correct
+ them or remove them. See the gcc distribution for the
+ appropriate tools.
+
+ A possible temporary work-around is to direct gcc to use
+ the vendor's header files instead of its temporary ones by
+ declaring -I/usr/include in the compilation flags.
+
+3.2.3 Other header file problems
+
+ Don't overlook any vendor tools that might validate the
+ vendor header files installed on the system -- e.g., the
+ Solaris pkgchk tool can be used to check the header files
+ that were installed from the SUNWhea package.
+
+ For other header file problems contact me at <abe@purdue.edu>.
+ Please follow the reporting guidelines in the "How do I
+ report an lsof bug?" section of this FAQ.
+
+3.3 Why doesn't lsof report full path names?
+
+ Lsof reports the full path name when it is specified as a
+ search argument for open files that match the argument.
+ However, if the argument is a file system mounted-on
+ directory, and lsof finds additional path name components
+ from the kernel name cache, it will report them.
+
+ Lsof reports path name for file system types that have path
+ name lookup features -- e.g., some versions of AdvFS for
+ Digital and Tru64 UNIX. The Linux /proc-based lsof reports
+ full path names, because the Linux /proc file system provides
+ them. Lsof on recent builds of Solaris 10 also report full
+ path names, because those Solaris kernels record the full path
+ name in the vnode structure.
+
+ Otherwise, lsof uses the kernel name cache, where it exists
+ and can be accessed, and reports some or all path name
+ components (e.g., the sys and proc.h components of
+ /usr/include/sys/proc.h) for these dialects:
+
+ Apple Darwin
+ DC/OSx
+ FreeBSD
+ HP-UX, /dev/kmem and PSTAT based
+ Linux, /dev/kmem-based
+ NetBSD
+ NEXTSTEP
+ OpenBSD
+ OPENSTEP
+ Reliant UNIX
+ SCO OpenServer
+ SCO|Caldera UnixWare
+ Solaris 2.x, 7, 8 and 9 (except for some VxFS versions;
+ see the "Why doesn't Solaris
+ lsof report VxFS path name
+ components?" section for more
+ information)
+ Solaris 10 (early builds) Tru64 UNIX
+
+ As far as I can determine, AFS path lookups don't share in
+ kernel name cache operations, so lsof can't identify open AFS
+ path name components. Apparently Solaris VxFS versions 4 and
+ above don't share in kernel name cache operations, either, so
+ lsof can't display path name components for those open files.
+
+ Since the size of the kernel name cache is limited and the
+ cache is in constant flux, it does not always contain the names
+ of all components in an open file's path; sometimes it contains
+ none of them.
+
+ Lsof reports the file system directory name and whatever
+ components of the file's path it finds in the cache, starting
+ with the last component and working backwards through the
+ directories that contain it. If lsof finds no path
+ components, lsof reports the file system device name instead.
+
+ When lsof does report some path components in the NAME
+ column, it prefixes them with the file system directory
+ name, followed by " -- ", followed by the components --
+ e.g., /usr -- sys/path.h for /usr/include/sys/path.h. The
+ " -- " is omitted when lsof finds all the path name components
+ of a file's name.
+
+ The PSTAT-based HP-UX lsof relies on kernel name cache
+ contents, too, even though its information comes to lsof
+ via pstat() function calls. Consequently, PSTAT-based
+ HP-UX lsof won't always report full paths, but may use the
+ " -- " partial path name notation, or may occasionally
+ report no path name at all but just the file system mounted-on
+ directory and device names.
+
+ Lsof can't obtain path name components from the kernel name
+ caches of the following dialects:
+
+ AIX
+
+ Only the Linux kernel records full path names in the
+ structures it maintains about open files; instead, most
+ kernels convert path names to device and node number doublets
+ and use them for subsequent file references once files have
+ been opened.
+
+ To convert the device and node number doublet into a
+ complete path name, lsof would have to start at the root
+ node (root directory) of the file system on which the node
+ resides, and search every branch for the node, building
+ possible path names along the way. That would be a time
+ consuming operation and require access to the raw disk
+ device (usually implying setuid-root permission).
+
+ If the prospect of all that local disk activity doesn't
+ concern you, think about the cost when the device is
+ NFS-mounted.
+
+ Try using the file system mount point and node number lsof
+ reports as parameters to find -- e.g.,
+
+ $ find <mount_point> -inum <node_number> -print
+
+ and you may get an appreciation of what a file system
+ directory tree search would cost.
+
+3.3.1 Why do lsof -r reports show different path names?
+
+ When you run lsof with its repeat (``-r'') option, you may
+ notice that the extent to which it reports path names for
+ the same files may vary from cycle to cycle. That happens
+ because other processes are making kernel calls affecting
+ the cache and causing entries to be removed from and added
+ to it.
+
+3.3.2 Why does lsof report the wrong path names?
+
+ Under some circumstances lsof may report an incorrect path
+ name component, especially for files in a rapidly changing
+ directory like /tmp.
+
+ In a rapidly changing directory, like /tmp, if the kernel
+ doesn't clear the cache entry when it removes a file, a
+ new file may be given the same keys and lead lsof to believe
+ that the old cache entry with the same keys belongs to the
+ new file.
+
+ Lsof tries to avoid this error by purging duplicate entries
+ from its copy of the kernel name cache when they have the
+ same device and inode number, but different names.
+
+ This error is less likely to occur in UNIX dialects where the
+ keys to the name cache are node address and possibly a
+ capability ID. The Apple Darwin, Digital UNIX, FreeBSD, HP-UX,
+ NEXTSTEP, OPENSTEP, Solaris, Tru64 UNIX, and UnixWare dialects
+ use node address. Apple Darwin, FreeBSD, NetBSD, OpenBSD,
+ Tru64 UNIX, and also use a capability ID to further identify
+ name cache entries.
+
+3.3.3 Why doesn't lsof report path names for unlinked (rm'd) files?
+
+ When lsof gets path name components from the kernel's name
+ cache, it does not report the path names of a file that has
+ been unlinked from its parent directory -- e.g., deleted via
+ rm, or the unlink() system call -- even when some process may
+ still hold the file open; lsof reports only the file system's
+ mounted-on directory and device. That's because path name
+ components are removed from the kernel name cache when the file
+ is unlinked.
+
+ Unlinked open files are sometimes used by applications for
+ temporary, but invisible storage (i.e., ls won't show them,
+ and no other process can open them.) However, they may
+ occasionally consume disk space to excess and cause concern
+ for a system administrator, who will be unable to locate
+ them with find, ls, du, or other tools that rely on finding
+ files by examining the directory tree.
+
+ By using lsof's +L option you can see the link count of
+ open files -- in the NLINK column. An unlinked file will
+ have an NLINK value of zero. By using the option +L1 you
+ can tell lsof to display only files whose link count is
+ less than one (i.e., zero).
+
+ There are some UNIX dialect-specific exceptions to lsof's
+ inability to report unlinked path names. They are described in
+ the answer to the "When will lsof report path names for deleted
+ files?" question.
+
+3.3.4 Why doesn't lsof report the "correct" hard linked file path
+ name?
+
+ When lsof reports a rightmost path name component for a
+ file with hard links, the component may come from the
+ kernel's name cache. Since the key which connects an open
+ file to the kernel name cache may be the same for each
+ differently named hard link, lsof may report only one name
+ for all open hard-linked files. Sometimes that will be
+ "correct" in the eye of the beholder; sometimes it will
+ not. Remember, the file identification keys significant
+ to the kernel are the device and node numbers, and they're
+ the same for all the hard linked names.
+
+3.3.5 When will lsof report path names for deleted files?
+
+ Lsof will report path names for deleted files for two
+ dialects: Linux and later builds of Solaris 10.
+
+ Deleted Linux path names are reported by default and have
+ "(deleted)" at their ends.
+
+ The display of Solaris 10 deleted path names may be selected
+ with the -X option. When selected they are also reported with
+ "(deleted)" at their ends.
+
+3.4 Why is lsof so slow?
+
+ Lsof may appear to be slow if network address to host name
+ resolution is slow. This can happen, for example, when the
+ name server is unreachable, or when a Solaris PPP cache daemon
+ is malfunctioning.
+
+ To see if name lookup is causing lsof to be slow, turn it off
+ with the ``-n'' option.
+
+ Port service name lookup or portmap registration lookup may
+ also be causes of slow-down. To suppress port service name
+ lookup, specify the ``-P'' option.
+
+ Lsof doesn't usually make direct portmap calls -- only when +M
+ is specified, or when HASPMAPENABLED is defined during lsof
+ construction. (The lsof help panel, produced with `lsof -h`
+ will display the default portmap registration reporting
+ state.) The quickest first step in checking if lsof is slow
+ because of the portmapper is to use lsof's ``-M'' option.
+
+ Lsof may be slow if UID to login name lookups are slow.
+ Suppress them with ``-l''.
+
+ On dialects where lsof uses the kernel name cache, try
+ disabling its use with ``-C''. (You can tell if lsof uses the
+ kernel name cache by looking for ``-C'' in lsof's ``-h''
+ output.) Of course, disabling kernel name cache use will mean
+ that lsof won't report full or partial path names, just file
+ system and character device names.
+
+ If you're just interested in the open files of one process, try
+ using the ``-p <Process-ID>'' option to limit lsof to that
+ process. (The ``-p'' option may also be followed with a list
+ of Process-IDs.)
+
+ If you're interested in including or excluding certain
+ commands, try lsof's "-c[^]cmd" option.
+
+ If you're interested in certain Internet TCP and UDP states
+ (e.g, ESTABLISHED) or in excluding some (e.g., CLOSE_WAIT), try
+ lsof's "-s p:s" option. More information on it may be found in
+ the answer to the "How are protocol state name exclusion and
+ inclusion used?" question.
+
+ Your UNIX dialect may not support "-s p:s" and its associated
+ performance improvments to Internet-only file processing. You
+ can find more information on those topics in the answer to the
+ "Why doesn't my dialect support state name exclusion and
+ inclusion?" question.
+
+ Older AIX lsof may be slow to start because of its oslevel
+ identity comparison. (Newer AIX lsof uses uname(2).) See the
+ "Why does AIX lsof start so slowly?" and "Why does lsof warn
+ "compiled for x ... y; this is z.?" sections for more
+ information.
+
+3.5 Why doesn't lsof's setgid or setuid permission work?
+
+ If you install lsof on an NFS file system that has been
+ mounted with the nosuid option, lsof may not be able to
+ use the setgid or setuid permission you give it, complaining
+ it can't open the kernel memory device -- e.g., /dev/kmem.
+
+ The only solution is to install lsof on a file system that
+ doesn't inhibit setgid or setuid permission.
+
+3.6 Does lsof have security problems?
+
+ I don't think so. However, lsof does usually start with
+ setgid permission, and sometimes with setuid-root permission.
+ Any program that has setgid or setuid-root permission,
+ should always be regarded with suspicion.
+
+ Lsof drops setgid power, holding it only while it opens
+ access to kernel memory devices (e.g., /dev/kmem, /dev/mem,
+ /dev/swap). That allows lsof to bypass the weaker security
+ of access(2) in favor of the stronger checks the kernel
+ makes when it examines the right of the lsof process to
+ open files declared with -k and -m. Lsof also restricts
+ some device cache file naming options when it senses the
+ process has setuid-root power.
+
+ On a few dialects lsof requires setuid-root permission
+ during its full execution in order to access files in the
+ /proc file system. These dialects include:
+
+ DC/OSx 1.1 for Pyramid systems
+ Reliant UNIX 5.4[34] for Pyramid systems
+
+ When lsof runs with setuid-root permission it severely
+ restricts all file accesses it might be asked to make with
+ its options.
+
+ The device cache file (typically .lsof_hostname in the home
+ directory of the real user ID that executes lsof) has 0600
+ modes. (The suffix, hostname, is the first component of
+ the host's name returned by gethostname(2).) However, even
+ when lsof runs setuid-root, it makes sure the file's
+ ownerships are changed to that of the real user and group.
+ In addition, lsof checks the file carefully before using
+ it (See the question "How do I disable the device cache
+ file feature or alter it's behavior?" for a description of
+ the checks.); discards the file if it fails the scrutiny;
+ complains about the condition of the file; then rebuilds
+ the file.
+
+ See the 00DCACHE file of the lsof distribution for more
+ information about device cache file handling and the risks
+ associated with the file.
+
+3.7 Will lsof show remote hosts using files via NFS?
+
+ No. Remember, lsof displays open files for the processes
+ of the host on which it runs. If the host on which lsof
+ is running is an NFS server, the remote NFS client processes
+ that are accessing files on the server leave no process
+ records on the server for lsof to examine.
+
+3.8 Why doesn't lsof report locks held on NFS files?
+
+ Generally lock information held by local processes on remote
+ NFS files is not recorded by the UNIX dialect kernel. Hence,
+ lsof can't report it.
+
+ One exception is some patch levels of Solaris 2.3, and all
+ versions of Solaris 2.4 and above. Lsof for those dialects
+ does report on locks held by local processes on remotely
+ mounted NFS files.
+
+3.8.1 Why does lsof report a one byte lock on byte zero as a full
+ file lock?
+
+ When a process has a lock of length one, starting at byte
+ zero, lsof can't distinguish it from a full file lock.
+ That's because most UNIX dialects represent both locks the
+ same way in their file lock (flock or eflock) structures.
+
+3.9 Why does lsof report different values for open files on the
+ same file system (the automounter phenomenon)?
+
+ On UNIX dialects where file systems may be mounted by an
+ automounter with the ``direct'' type, lsof may sometimes
+ report difference DEVICE, SIZE/OFF, INODE and NAME values
+ when asked to report files open on the file system.
+
+ This happens because some files open on the file system --
+ e.g., the current directory of a shell that changed its
+ directory to the file system as the file system's first
+ reference -- may be characterized in the kernel with
+ temporary automounter node information. The cd doesn't
+ cause the file system to be mounted.
+
+ A subsequent reference to the file system -- e.g., an ls
+ of any place in it -- will cause the file system to be
+ mounted. Processes with files open to the mounted file
+ system are characterized in the kernel with data that
+ reflects the mounted file system's parameters.
+
+ Unfortunately some kernels (e.g., some versions of Solaris
+ 2.x) don't revisit the process that did only a change-directory
+ for the purpose of updating the data associated with the
+ open directory file. The file continues to be characterized
+ with temporary automounter information until it does another
+ directory change, even a trivial ``cd .''.
+
+ Lsof will report on both reference types, when supplied
+ the file system name as an argument, but the data lsof
+ reports will reflect what it finds in the kernel. For the
+ different types lsof will display different data, including
+ different major and minor device numbers in the DEVICE
+ column, different lengths in the SIZE/OFF column, different
+ node numbers in the INODE column, and slightly different
+ file system names in the NAME column.
+
+ In contrast, fuser, where available, can only report on
+ one reference type when supplied the file system name as
+ an argument. Usually it will report on the one that is
+ associated with the mounted file system information. If
+ the only reference type is the temporary automounter one,
+ fuser will often be silent about it.
+
+3.10 Why don't lsof and netstat output match?
+
+ Lsof and netstat output don't match because lsof reports
+ the network information it finds in open file system objects
+ -- e.g., socket files -- while netstat often gets its
+ information from separate kernel tables.
+
+ The information available to netstat may describe network
+ activities never or no longer associated with open files,
+ but necessary for proper network state machine operation.
+
+ For example, a TCP connection in the FIN_WAIT_[12] state
+ may no longer have an associated open file, because the
+ connection has been closed at the application layer and is
+ now being closed at the TCP/IP protocol layer.
+
+3.10.1 Why can't lsof find accesses to some TCP and UDP ports?
+
+ Lsof stands for LiSt Open Files. If there is no open file
+ connected to a TCP or UDP port, lsof won't find it. That's
+ the most common reason why lsof doesn't find a port netstat
+ might report open.
+
+ One reason I've found on some UNIX dialects is that their
+ kernels set aside TCP and UDP ports for communicating with
+ support activities, running in application layer servers
+ -- the automounter daemons, and the NFS biod and nfsd
+ daemons are examples. Netstat may report the ports are in
+ use, but lsof doesn't.
+
+ Another reason is that netstat may also be able to report
+ a port is open on a particular dialect, because it uses a
+ source of data different from what lsof uses -- e.g.,
+ netstat might examine kernel tables or use streams messages
+ to MIB2, while lsof relies on the information it finds in
+ open file structures and their descendants.
+
+ Sometimes it's possible to search the data netstat and lsof
+ use. For example, on Linux /proc/tcp and /proc/udp can be
+ examined. There might an entry there for a particular
+ protocol and port, but if the line on which the port appears
+ doesn't have an inode number that matches an inode number
+ of an open file, lsof won't be able to identify the process
+ using the port.
+
+ This is a tough question to which there is no easy answer.
+
+3.11 Why does lsof update the device cache file?
+
+ At the end of the lsof output you may see the message:
+
+ lsof: WARNING: /Homes/abe/.lsof_vic was updated.
+
+ In this message /Homes/abe/.lsof_vic is the path to the
+ private device cache file for login abe. (See 00DCACHE.)
+
+ Lsof issues this message when it finds it necessary to
+ recheck the system device directory (e.g., /dev or /devices)
+ and rebuild the device cache file during the open file
+ scan. Lsof may need to do these things it finds that a
+ device directory node has changed, or if it cannot find a
+ device in the cache.
+
+3.12 Why doesn't lsof report state for UDP socket files?
+
+ Lsof reports UDP TPI connection state -- TS_IDLE (Idle),
+ TS_BOUND (Bound), etc. -- for some, but not all dialects.
+ TPI state is stream-based TCP/IP information that isn't
+ available in many dialects.
+
+ A fairly weak general rule is if netstat(1) reports UDP
+ TPI state, lsof may be able to report it, too. But don't
+ be surprised if lsof fails to report UDP TPI state for your
+ dialect. Other factors influence lsof's ability to report
+ UDP TPI state, including the availability of state number
+ data in kernel structures, and state number to state name
+ conversion data.
+
+3.13 I am editing a file with vi; why doesn't lsof find the file?
+
+ Classic implementations of vi usually don't keep open the file
+ being edited. (Newer ones may do so in order to maintain an
+ advisory lock.) Instead classic vi opens the file, makes a
+ temporary copy (usually in /tmp or /usr/tmp), and does its work
+ in that file. When you save the file being edited from a
+ classic vi implementation, it reopens and rewrites the file.
+
+ During a classic vi session, except for the brief periods when
+ vi is reading or rewriting the file, lsof won't find an open
+ reference to the file from the vi process, because there is
+ none.
+
+3.14 Why doesn't lsof report TCP/TPI window and queue sizes for my
+ dialect?
+
+ Lsof only reports TCP/TPI window sizes for Solaris, because
+ only its netstat reports them. The intent of providing
+ TCP/TPI information in lsof NAME column output is to make
+ it easier to match netstat output to lsof output.
+
+ In general lsof only reports queue sizes for both TCP and
+ UDP (TPI) connections on BSD-derived UNIX dialects, where
+ both sets of values appear in kernel socket queue structures.
+ SYSV-derived UNIX dialects whose TCP/IP implementations
+ are based on streams generally provide only TCP queue sizes,
+ not UDP (TPI) ones.
+
+ While you may find that netstat on some SYSV-derived UNIX
+ dialects with streams TCP/IP may report UDP (TPI) queue
+ sizes, you will probably also find that the sizes are always
+ zero -- netstat supplies a constant zero for UDP (TPI)
+ queue sizes to make its headers align the same for TCP and
+ UDP (TPI) connections. Solaris seems to get it right --
+ i.e., its netstat does not report UDP (TPI) queue sizes.
+
+ When in doubt, I chose to avoid reporting UDP (TPI) queue
+ sizes for UNIX dialects whose netstat-reported values I
+ knew to be a constant zero or whose origin I couldn't
+ determine. OSR is a dialect in this category.
+
+3.14.1 Why doesn't lsof report socket options, socket states, and TCP
+ flags and values for my dialect?
+
+ The lsof -T argument, 'f', that selects the reporting of socket
+ options, socket states and TCP flags was implemented at lsof
+ revision 4.71 for the following UNIX dialects, providing the
+ indicated information:
+
+ AIX 4.3.2 and 5.1 and above
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ Apple Darwin 7.2 and above
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ Digital UNIX and Tru64 UNIX 4.0
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ FreeBSD 4.9 and above
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ HP-UX 11.00 (/dev/kmem-based lsof)
+ All socket options and values are reported. No socket
+ states are reported. Only the TF_NODELAY TCP flag and
+ the TF_MSS value are reported.
+ HP-UX 11.11 and iiiv2 (PSTAT-based lsof)
+ All socket options and values, and socket states are
+ reported. No TCP flags or values are reported.
+ Linux
+ No socket options and values, socket states, or TCP
+ flags and values are reported. The support for "-Tf"
+ could not be added to Linux, because socket options,
+ socket states, and TCP flags and values are not
+ available via the /proc file system.
+ NetBSD 1.6G and above
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ OpenBSD 3.4 and above
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ OPENSTEP 4.2
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ OpenUNIX 8
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ SCO OpenServer Release 5.0.6
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+ Solaris 2.6, 8 and above
+ The socket option display is limited to BROADCAST,
+ DEBUG, DGRAM_ERRIND, DONTROUTE and OOBINLINE. Socket
+ values are limited to KEEPALIVE and LINGER. No socket
+ states are reported. The TCP DELACK, NODELAY and
+ SENTFIN flags are reported. The TCP MSS value is
+ reported.
+ UnixWare 7.1.[134]
+ All socket options and values, socket states, and TCP
+ flags and values described in lsof(8) are reported.
+
+3.14.2 Why doesn't lsof report the partial listen queue connection
+ count for my dialect?
+
+ The reporting of partial listen queue connections was added to
+ -Tf processing at lsof revision 4.76. Currently it is reported
+ for these dialects:
+
+ AIX 4.3.2
+ This dialect is no longer supported, so no attempt
+ was made to add partial listen queue length support
+ for it.
+ AIX 5.1 and above
+ Partial listen queue information is available.
+ Apple Darwin 7.2 and above
+ Partial listen queue information is available.
+ Digital UNIX 4.0
+ This dialect is no longer supported, so no attempt
+ was made to add partial listen queue length support
+ for it.
+ FreeBSD 4.9 and above
+ Partial listen queue information is available.
+ HP-UX 11.00 (/dev/kmem-based lsof)
+ No partial listen queue information is available.
+ HP-UX 11.11 and iiiv2 (PSTAT-based lsof)
+ No partial listen queue information is available.
+ Linux
+ No partial listen queue information is available.
+ NetBSD 1.6G and above
+ Partial listen queue information is available.
+ OpenBSD 3.4 and above
+ Partial listen queue information is available.
+ OPENSTEP 4.2
+ Partial listen queue information is available.
+ OpenUNIX 8
+ This dialect is no longer supported, so no attempt
+ was made to add partial listen queue length support
+ for it.
+ SCO OpenServer Release 5.0.6
+ No partial listen queue information is available.
+ Solaris 2.6, 8 and above
+ Partial listen queue information is available.
+ Tru64 UNIX 5.0
+ This dialect is no longer supported, so no attempt
+ was made to add partial listen queue length support
+ for it.
+ Tru64 UNIX 5.1
+ Partial listen queue information is available.
+ UnixWare 7.1.[134]
+ Partial listen queue information is available.
+
+
+3.15 What does "no more information" in the NAME column mean?
+
+ When lsof can find no successor structures -- a gnode,
+ inode, socket, or vnode -- connected to the file structure
+ of an open descriptor of a process, it reports "no more
+ information" in the NAME column. The TYPE, DEVICE, SIZE/OFF,
+ and INODE columns will be blank.
+
+ Because the file structure is supposed to contain a pointer
+ to the next structure of a file's processing support, if
+ the pointer is NUL, lsof can go no further.
+
+ Some UNIX dialects have file structures for system processes
+ -- e.g., the sched process -- that have no successor
+ structure pointers. The "no more information" NAME will
+ commonly appear for these processes in lsof output.
+
+ It may also be the case that lsof has read the file structure
+ while it is being assembled and before a successor structure
+ pointer value has been set. The "no more information" NAME
+ will again result.
+
+ Unless lsof output is filled with "no more information"
+ NAME column messages, the appearance of a few should be no
+ cause for alarm.
+
+3.16 Why doesn't lsof find a process that ps finds?
+
+ If lsof fails to display open files for a process that ps
+ indicates exists, there may be several reasons for the
+ difference.
+
+ The process may be a "zombie" for which ps displays the
+ "(defunct)" state. In that case, the process has exited
+ and has no open file information lsof can display. It does
+ still have a process structure, sufficient for the needs
+ of ps.
+
+ Another possible explanation is that kernel tables and
+ structures may have been changing when lsof looked for the
+ process, making lsof unable to find all relevant process
+ structures. Try repeating the lsof request.
+
+3.17 Why doesn't -V report a search failure?
+
+ The usual reason that -V won't report a search failure is
+ that lsof located the search item, but was prevented from
+ listing it by an option that doesn't participate in search
+ failure reporting.
+
+ For example, this lsof invocation:
+
+ $ lsof -V -i TCP@foobar -a -d 999
+
+ won't report it can't find the Internet address TCP@foobar,
+ even if there is an open file connected to that address,
+ unless the open file also has a file descriptor number of
+ 999 (the ``-a -d 999'' options).
+
+ Compile-time options can also affect -V results in much the
+ same way. For example, if HASSECURITY and HASNOSOCKSECURITY
+ are defined at compile time, this lsof invocation, run by a
+ non-root user:
+
+ $ lsof -V -c inetd
+
+ won't report that it can't find the inetd command, even if
+ there is a process running the inetd command, because the
+ HASSECURITY and HASNOSOCKSECURITY options prevent the
+ listing of all but the socket files of another user, and
+ no socket file selector (e.g., "-i") was specified.
+
+
+3.18 Portmap problems
+
+3.18.1 Why isn't a name displayed for the portmap registration?
+
+ When portmap registration reporting is enabled, any time
+ there is a registration for a local TCP or UDP port, lsof
+ displays it in square brackets, following the port number
+ or service name -- e.g., ``:1234[name]'' or ``:name[100083]''.
+
+ The TCP or UDP port number or service number (what follows
+ the `:') is displayed under the control of the lsof -P
+ option. The registration identity is held by the portmapper
+ and may be a name or a number, depending on how the
+ registration's owner declared it. Lsof reports what the
+ port map holds and cannot derive a registration name from
+ a registration number.
+
+ Lsof can be compiled with registration reporting enabled
+ or disabled by default, under the control of the HASPMAPENABLED
+ #define (usually in machine.h). The lsof help panel (`lsof
+ -h`) will show the default. Lsof is distributed with
+ reporting disabled by default.
+
+3.18.2 How can I display only portmap registrations?
+
+ Lsof doesn't have an option that will display only TCP or
+ UDP ports with portmap registrations. The +M option only
+ enables the reporting of registration information when
+ Internet socket files are displayed; +M doesn't select
+ the displaying of Internet socket files -- the -i option
+ does that.
+
+ This simple lsof pipe to grep will do the job:
+
+ $ lsof -i +M | grep "\["
+
+ This works because -i selects Internet socket files, +M
+ enables portmap registration reporting, and only output
+ lines with opening square brackets will have registrations.
+
+ When portmap registration reporting is enabled by default,
+ because the lsof builder constructed it that way, +M is
+ not necessary. (The lsof help panel, produced with `lsof
+ -h` will display the default portmapper registration
+ reporting state.) However, specifying +M when reporting
+ is already enabled is acceptable, as is specifying -M when
+ reporting is already disabled.
+
+ Digression: lsof will accept `+' or `-' as a prefix to most
+ options. (That isn't documented in the man page or help
+ panel to reduce confusion and complexity.) The -i option
+ is as acceptable as +i, so the above example could be
+ written a little more tersely as:
+
+ $ lsof +Mi | grep "\["
+
+ But be careful to use the ``Mi'' ordering, since ``iM''
+ implies M is an address argument to `i'.
+
+3.18.3 Why doesn't lsof report portmap registrations for some ports?
+
+ Lsof reports portmap registrations for local TCP and UDP
+ ports only. It identifies local ports this way:
+
+ * The port appears in the local address section of the
+ kernel structure that contains it.
+
+ * The port appears in the foreign address section of a
+ kernel structure whose local and foreign Internet
+ addresses are the same.
+
+ * The port appears in the foreign address section of a
+ kernel address structure whose Internet address is
+ INADDR_LOOPBACK (127.0.0.1).
+
+ Following these rules, lsof ignores foreign portmapped
+ ports. That's done for reasons of efficiency and possible
+ security prohibitions. Contacting all remote portmappers
+ could take a long time and be blocked by network difficulties
+ (i.e., be inefficient). Many firewalls block portmapper
+ access for security reasons.
+
+ Lsof may occasionally ignore portmap registration information
+ for a legitimate local port by virtue of its local port
+ rules. This can happen when a port appears in the foreign
+ part of its kernel structure and the local and foreign
+ Internet addresses don't match (perhaps because they're on
+ different interfaces), and the foreign Internet address
+ isn't INADDR_LOOPBACK (127.0.0.1).
+
+3.19 Why is `lsof | wc` bigger than my system's open file limit?
+
+ There is a strong temptation to count open files by piping
+ lsof output to wc. If your purpose is to compare the number
+ you get to some Unix system parameter that defines the
+ number of open files your system can have, resist the
+ temptation.
+
+ One reason is that lsof reports a number of "files" that
+ don't occupy Unix file table space -- current working
+ directories, root directories, jail directories, text files,
+ library files, memory mapped files are some. Another reason
+ is that lsof can report a file shared by more than one
+ process that itself occupies only one file table slot.
+
+ If you want to know the number of open files that occupy
+ file table slots, use the +ff option and process the lsof
+ output's FILE_ADDR column information with standard Unix
+ tools like cut, grep, sed, and sort.
+
+ You might also consider using use lsof's field output with
+ +ff, selecting the file struct address with -FF, and
+ processing the output with an AWK or Perl script. See the
+ list_fields.awk, list_fields.perl, and shared.perl5 scripts
+ in the scripts/ subdirectory of the lsof distribution for
+ hints on file struct post-processing filters.
+
+3.20 Why doesn't lsof report file offset (position)?
+
+ Lsof won't report a file offset (position) value if the -s
+ option has been specified, or if the dialect doesn't support
+ the displaying of file offset (position).
+
+ That lsof is reporting only file size is indicated by the
+ fact that the appropriate column header says SIZE instead
+ of SIZE/OFF.
+
+ If lsof doesn't support the displaying of file offset
+ (position) -- e.g., for Linux /proc-based lsof -- the -h
+ or -? output panel won't list the -o option.
+
+ Sometimes the availability of file offset information
+ depends on the dialect's kernel. This is particularly true
+ for socket file offsets.
+
+ Maintenance of offsets for pseudo-terminal devices varies
+ by UNIX dialect and is related to how the dialect kernel
+ implements pseudo-terminal support. Kernels like AIX, for
+ example, that short-circuit the transfer of data between
+ socket and pseudo devices to reduce TCP/IP daemon interrupt
+ rates won't advance offsets in the TCP/IP daemon socket
+ files. Instead they will advance offsets in the open
+ standard I/O files of the shell child precess where the
+ pseudo-terminal devices are used.
+
+ When in doubt about the behavior of lsof in reporting file
+ offset information, do some carefully measured experiments,
+ consult the lsof sources, or contact me at <abe@purdue.edu>
+ to discuss the matter. Please follow the reporting guidelines
+ in the "How do I report an lsof bug?" section of this FAQ.
+
+3.20.1 What does lsof report for size when the file doesn't really have
+ one?
+
+ When a file has no true size -- e.g., it's a socket, a
+ FIFO, or a pipe -- lsof tries to report the information it
+ finds in the kernel that describes the contents of associated
+ kernel buffers.
+
+ Thus, for example, size for most TCP/IP files is socket
+ buffer size. The size of the socket read buffer is reported
+ for read-only files; the size of the write buffer for
+ write-only files; and the sum of the buffers sizes for
+ read-write files.
+
+3.21 Problems with path name arguments
+
+3.21.1 How do I ask lsof to search a file system?
+
+ You can ask lsof to search for all open files on a file
+ system by specifying its mounted path name as an lsof
+ argument -- e.g.,
+
+ $ lsof /
+
+ Output of the mount command will show file system mounted
+ path names. It will also show the mounted-on device path
+ for the file system.
+
+ If the mounted-on device is a block device (the permission
+ field in output of `ls -l <device>` starts with a `b/),
+ you can specify it's name, too -- e.g.,
+
+ $ lsof /dev/sd0a
+
+ If the mounted-on device isn't a block device -- for example,
+ some UNIX dialects call a CD-ROM device a character device
+ (ls output starts with a `c') -- you can force lsof to
+ assume that the specified device names a file system with
+ the +f option -- e.g.,
+
+ $ lsof +f -- /dev/sd0a
+
+ (Note: you must use ``--'' after +f or -f if a file name
+ follows immediately, because +f and -f can be followed by
+ characters that specify flag output selections.)
+
+ When you use +f and lsof can't match the device to a file
+ system, lsof will issue a complaint.
+
+ The +f option may be used in some dialects to ask lsof to
+ search for an NFS file system by its server name and server
+ mount point. If the mount application reports an NFS file
+ system mounted-on value that way, then this sample lsof
+ request should work.
+
+ $ lsof +f -- fleet:/home/fleet/u5
+
+ Finally, you can use -f if you don't want a mounted file
+ system path name to be considered a request to report all
+ open files on the file system. This is useful when you
+ want to know if anyone is using the file system's mounted
+ path name. This example directs lsof to report on open
+ access to the `/' directory, including when it's being used
+ as a current working or root directory.
+
+ $ lsof -f -- /
+
+ The lsof -f option performs the same function as -f does
+ in some fuser implementations. However, since the lsof -c
+ option was chosen for another purpose before the `f' option
+ was added to lsof, +f was selected as the analogue to the
+ fuser -c option. (Sorry for the potential confusion.)
+
+3.21.2 Why doesn't lsof find all the open files in a file system?
+
+ Lsof may not find all the open files in a file system for
+ several reasons.
+
+ First, some processes with files open on the file system
+ may have been changing status when lsof examined the process
+ table, and lsof "missed" them. Remember, the kernel changes
+ much faster than lsof can respond to the changes.
+
+ Second, be sure you have specified the file system correctly.
+ Perhaps you specified a file instead. You can use lsof's
+ -V option to have lsof report in detail on what it couldn't
+ find. Make sure the report for the file system you specified
+ says "file system." Here's some -V output:
+
+ $ /lsof -V /tmp ./lsof.h ./lsof
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ lsof 2688 abe txt VREG 18,1,7 1428583 226641 ./lsof
+ lsof 2689 abe txt VREG 18,1,7 1428583 226641 ./lsof
+ lsof: no file use located: ./lsof.h
+
+ You can also use lsof's +f option to force it to consider
+ a path name as a file system. If lsof can't find a file
+ system by the specified name, it will issue a complaint --
+ e.g.,
+
+ $ lsof +f -- /usr
+ lsof: not a file system: /usr
+
+ (/usr is a directory in the / file system.)
+
+3.21.3 Why does the lsof exit code report it didn't find open files
+ when some files were listed?
+
+ Sometimes lsof will list some open files, yet return a
+ non-zero exit code, suggesting it hasn't found all the
+ specified files.
+
+ The first thing you should when you suspect lsof is incorrect
+ is to repeat the request, adding the -V option. In the
+ resulting report you may find that your file system
+ specification really wasn't a file system specification,
+ just a file specification.
+
+ Finally, if you specify two files or two file systems twice,
+ lsof will credit all matches to the first of the two and
+ believe that there were no matches for the second. It's
+ possible to specify a single file system twice with different
+ path names by using both its mounted directory path name
+ and mounted-one device name.
+
+ $ lsof +f -V spcuna:/sysprog /sysprog
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ksh 11092 abe cwd VDIR 39,0,1 1536 226562 /sysprog
+ (spcuna:/sysprog)
+ ...
+ lsof: no file system use located: spcuna:/sysprog
+
+ All matches were credited to /sysprog; none to spcuna:/sysprog.
+
+3.21.4 Why won't lsof find all the open files in a directory?
+
+ When you give lsof a simple directory path name argument
+ (not a file system mounted-on name), you are asking it to
+ search for processes that have the directory open as a
+ file, or as a process-specific directory -- e.g., root or
+ current working directory.
+
+ If you want to list instances of open files inside the
+ directory, you need to specify the individual path names
+ of those files, or use the lsof +D and +d options.
+
+ See the answer to the question "Why are the +D and +d
+ options so slow?" before you use +D or +d casually.
+
+ See the answer to the question "Why do the +D and +d options
+ produce warning messages?" for an explanation of some
+ process authority limitations of +D and +d.
+
+3.21.5 Why are the +D and +d options so slow?
+
+ The +D and +d options cause lsof to build a path name search
+ list for a specified directory. +D causes lsof to descend
+ the directory to its furthest subdirectory, while +d
+ restricts it to the top level. In both cases, the specified
+ directory itself is included in the search list. In both
+ symbolic links are ignored.
+
+ Building such a search list can take considerable time,
+ especially when the specified directory contains many files
+ and subdirectories -- lsof must call the system readlink()
+ and stat() functions for each file and directory. Storing
+ the search list can cause lsof to use more than its normal
+ amount of dynamic memory -- each file recorded in the search
+ list consumes dynamic memory for its path name, characteristics,
+ and search linkages. Using the list means lsof must search
+ it for every open file in the system.
+
+ Building the search list for a directory specified on some
+ file systems can be slow -- e.g., for an NFS directory with
+ many files. Some file systems have special logging features
+ that can introduce additional delays to the building of
+ the search list -- e.g., NFS logging, or logging on a
+ Solaris UFS file system. The bottom line is that slow
+ search list construction may not be so much an lsof problem
+ as a file system problem. (Hint: if you're using Solaris
+ UFS logging, consider specifying the "logging,noatime"
+ option pair to reduce the number of atime writes to the
+ UFS logging queue and disk.)
+
+ A somewhat risky way to speed up lsof's building of the
+ search list is to use lsof's ``-O'' option. It forces lsof
+ to do all system calls needed to build the search list
+ directly, rather than in a child process. While direct
+ system calls are much faster, they can block in the kernel
+ -- e.g., when an NFS server stops responding -- stopping
+ lsof until the kernel operation unblocks.
+
+ As an example of the load +D can impose, consider that an
+ `lsof +D /` on a lightly loaded NeXT '040 cube with a 1GB
+ root file system disk took 4+ minutes of real time. It
+ also generated several hundred error messages about files
+ and directories the lsof process didn't have permission to
+ access with stat(2).
+
+ The bottom line is that +D and +d should be used cautiously.
+ +D is more costly than +d for deeply nested directory trees,
+ because of the full directory descent it causes. So use
+ +d where possible. And you might need to consider the
+ performance of the file system that holds the directory
+ you name with +d or +D.
+
+ In view of these warnings, when is it appropriate to use
+ +D or +d? Probably the most appropriate time is when you
+ would specify the directory's contents to lsof with a shell
+ globbing construct -- e.g., `lsof *`. If that's what you
+ need to do, `lsof +d .` is probably more efficient than
+ having the shell produce a directory list, form it into an
+ argument vector, and pass the vector to lsof for it to
+ unravel.
+
+ See the answer to the question "Why do the +D and +d options
+ produce warning messages?" for an explanation of some
+ process authority limitations of +D and +d.
+
+3.21.6 Why do the +D and +d options produce warning messages?
+
+ +D and +d option processing is limited by the authority of
+ the lsof process -- i.e., lsof can only examine (with
+ lstat(2) and stat(2)) files the owner of the process can
+ access.
+
+ If the ownership, group membership, or permissions of the
+ specified directory, file within it, or directory within
+ it prevents the owner of the lsof process from using lstat(2)
+ or stat(2) on it, lsof will issue a warning message, naming
+ the path and giving the system's (lstat(2's or stat(2)'s)
+ reason (errno explanation text) for refusing access.
+
+ As an example, assume user abc has a subdirectory in /tmp,
+ owned by abc and readable, writable and searchable by only
+ its owner. If user def asks lsof to search for all /tmp
+ references with +D or +d, lsof will be unable to lstat(2)
+ or stat(2) anything in abc's private subdirectory, and will
+ issue an appropriate warning.
+
+ Lsof warnings can usually be suppressed with the -w option.
+ However, using -w with +D or +d means that there will be
+ no indication why lsof couldn't find an open reference to
+ a restricted directory or something contained in it.
+
+ Hint: if you need to use +D or +d and avoid authority
+ warnings, and if you have super-user power, su and use lsof
+ with +D or +d as root.
+
+3.22 Why can't my C compiler find the rpcent structure definition?
+
+ When you try to compile lsof your compiler may complain
+ that the rpcent structure is undefined. The complaints
+ may look like this:
+
+ >print.c: In function `fill_portmap':
+ >print.c:213: dereferencing pointer to incomplete type
+ >...
+
+ The most likely cause is that someone has allowed a BIND
+ installation to update /usr/include/netdb.h (or perhaps
+ /usr/include/rpc/netdb.h), removing the rpcent structure
+ definition that lsof expects to find there.
+
+ Only Solaris has an automatic work-around. (See dlsof.h
+ in dialects/sun.). The Solaris work-around succeeds because
+ there is another header file, <rpc/rpcent.h>, with the rpcent
+ structure definition, and there is a Solaris C pre-processor
+ test that can tell when the BIND <netdb.h> is in place and
+ hence <rpc/rpcent.h> must be included.
+
+ Doubtlessly there are similar work-arounds possible in
+ other UNIX dialects whose header files have been "touched"
+ by BIND, but in general I recommend restoration of the
+ vendor's <netdb.h> and any other header files BIND might
+ have replaced. (I think BIND replaces <resolv.h>,
+ <sys/bitypes.h>, <sys/cdefs.h> -- and maybe others.)
+
+3.23 Why doesn't lsof report fully on file "foo" on UNIX dialect
+ "bar?"
+
+ Lsof sometimes won't report much information on a given
+ file, or may even report an error message in its NAME
+ column. That's usually because the file is of a special
+ type -- e.g., in a file system specific to the UNIX dialect
+ -- and I haven't used a system where the file appeared
+ during my testing.
+
+ If you encounter such a situation, send me e-mail at
+ <abe@purdue.edu> and we may be able to devise an addition to
+ lsof that will report on the file in question. Please follow
+ the reporting guidelines in the "How do I report an lsof bug?"
+ section of this FAQ. Make sure "lsof" appears in the
+ "Subject:" line so my e-mail filter won't classify your letter
+ as Spam.
+
+3.24 Why do I get a complaint when I execute lsof that some library
+ file can't be found?
+
+ On systems where the LIBPATH (or the equivalent) environment
+ variable is used to record the library search path in
+ executable files when they are built, an incorrect value
+ may make it impossible for the system to find the shared
+ libraries needed to load lsof for execution.
+
+ This may be particularly true on systems like AIX >= 4.1.4,
+ where the lsof Makefile takes the precautionary step of
+ using the -bnolibpath loader flag to insure that the path
+ to the private static lsof library is not recorded in the
+ lsof binary. Should LIBPATH be invalid when lsof is built,
+ it will be recorded in the lsof binary as the default
+ library path search order and lead to an inability to find
+ libraries when lsof is executed.
+
+ So, if you get missing library complaints when you try to
+ execute lsof, check LIBPATH, or whatever environment variable
+ is used on your system to define library search order in
+ executable files. Use the tools at your disposal to look
+ at the library paths recorded in the lsof binary -- e.g.,
+ chatr on HP-UX, dump on AIX, ldd on Solaris.
+
+ Make sure, too, that when the correct library search path
+ has been recorded in the executable file, the required
+ library files exist at one or more of the search paths.
+
+
+3.25 Why does lsof complain it can't open files?
+
+ When lsof begins execution, unless it has been asked to
+ report only help or version information, typically it will
+ attempt to access kernel memory and symbol files -- e.g.,
+ /unix, /dev/kmem. Even though lsof needs only permission
+ to open these files for reading, read access to them might
+ be restricted by ownerships and permission modes.
+
+ So the first step to diagnosing lsof problems with opening
+ files is to use ls(1) to examine the ownerships and permission
+ modes of the files that lsof wants to open. You may find
+ that lsof needs to be installed with some type of special
+ ownership or permission modes to enable it to open the
+ necessary files for reading. See the "Installing Lsof"
+ section of 00README for more information.
+
+3.26 Why does lsof warn "compiled for x ... y; this is z."?
+
+ Unless warnings are suppressed (with -w) or the kernel
+ identity check symbol (HASKERNIDCK) definition has been
+ deleted, all but one lsof dialect version (exception:
+ /proc-based Linux lsof) compare the identity of the running
+ kernel to that of the one for which lsof was constructed.
+ If the identities don't match, lsof issues a warning like
+ this:
+
+ lsof: WARNING: compiled for Solaris release 5.7; this is 5.6.
+
+ Two kernel identity differences can generate this warning
+ -- the version number and the release number.
+
+ Build and running identity differences are usually significant,
+ because they usually indicate kernels whose structures are
+ different -- kernel structures commonly change at dialect
+ version releases. Since lsof reads data from the kernel
+ in the form of structures, it is sensitive to changes in
+ them. The general rule is that an lsof compiled for one
+ UNIX dialect version will not work correctly when run on
+ a different version.
+
+ There are three work-arounds: 1) use -w to suppress the
+ warning -- and risk missing other warnings; 2) permanently
+ disable the identity check by deleting the definition of
+ HASKERNIDCK in the dialect's machine.h header file -- with
+ the same risk; or 3) rebuild lsof on the system where it
+ is to be run. (Deleting HASKERNIDCK can be done with the
+ Customize script or by editing machine.h.)
+
+ Generally checking kernel identity is a quick operation
+ for lsof. However, it is potentially slow under AIX, where
+ lsof must run /usr/bin/oslevel. To speed up lsof, use -w
+ to suppress the /usr/bin/oslevel test. See "Why does AIX
+ lsof start so slowly?" for more information.
+
+3.27 How can I disable the kernel identity check?
+
+ The kernel identity check is controlled by the HASKERNIDCK
+ definition. When it is defined, most dialects (exclusion:
+ /proc-based Linux lsof) will compare the build-time kernel
+ identity with the run-time one.
+
+ To disable the kernel identity check, disable the HASKERNIDCK
+ definition in the dialect's machine.h header file. The
+ Customize script can be used to do that in its section
+ about the kernel identity check.
+
+ Caution: while disabling the kernel identity check may
+ result in smaller lsof startup overhead, it comes with the
+ risk of executing an lsof that may produce warning messages,
+ error messages, incorrect output, or no output at all.
+
+3.28 Why don't ps(1) and lsof agree on the owner of a process?
+
+ Generally the user ID lsof reports in its USER column is
+ the process effective user ID, as found in the process
+ structure. Sometimes that may not agree with what ps(1)
+ reports for the same process.
+
+ There are sundry reasons for the difference. Sometimes
+ ps(1) uses a different source for process information,
+ e.g., the /proc file system or the psinfo structure.
+ Sometimes the kernel is lax or confused (e.g., Solaris
+ 2.5.1) about what ID to report as the effective user ID.
+ Sometimes the system carries only one user ID in its process
+ structure (some BSD derivatives), leaving lsof no choice.
+
+ The differences between lsof and ps(1) user identifications
+ should be small and normally it will be apparent that the
+ confusion is over a process whose application has changed
+ to an effective user ID different from the real one.
+
+3.29 Why doesn't lsof find an open socket file whose connection
+ state is past CLOSE_WAIT?
+
+ TCP/IP connections in states past CLOSE_WAIT -- e.g.,
+ FIN_WAIT_1, CLOSING, LAST_ACK, FIN_WAIT_2, and TIME_WAIT
+ -- don't always have open files associated with them. When
+ they don't, lsof can't identify them. When the connection
+ state advances from CLOSE_WAIT, sometimes the open file
+ associated with the connection is deleted.
+
+3.30 Why don't machine.h definitions work when the surrounding
+ comments are removed?
+
+ The machine.h header files in dialect subdirectories have
+ some commented-out definitions like:
+
+ /* #define HASSYSDC "/your/choice/of/path */
+
+ You can't simply remove the comments and expect the definition
+ to work. That's intended to make you think about what
+ value you are assigning to the symbol. The assigned value
+ might have a system-specific convention. HASSYSDC, for
+ example, might be /var/db/lsof.dc for FreeBSD, but it might
+ be /var/adm/lsof.dc for Solaris.
+
+ Symbols defined in the lsof documentation are described in
+ 00PORTING, other machine.h comments, and other lsof
+ documentation files. HASSYSDC, for example, is discussed
+ in 00DCACHE. When comments and documentation don't suffice,
+ consult the source code for hints on how the symbol is
+ used.
+
+3.31 What do "can't read inpcb at 0x...", "no protocol control
+ block", "no PCB, CANTSENDMORE, CANTRCVMORE", etc. mean?
+
+ Sometimes lsof will report "can't read inpcb at 0x00000000",
+ "no protocol control block", "no PCB, CANTSENDMORE,
+ CANTRCVMORE" or a similar message in the NAME column for
+ open TCP socket files. These messages mean the file's socket
+ structure lacks a pointer to the INternet Protocol Control
+ Block (inpcb) where lsof expects to find connection addresses
+ -- local and foreign ports, local and foreign IP addresses.
+ The socket file has probably been submitted to the shutdown(2)
+ function for processing.
+
+ In some implementations lsof issues the "no PCB, CANTSENDMORE,
+ CANTRCVMORE" message, which tries to explain the absence
+ of a protocol control block by showing the socket state
+ settings that have been made by the shutdown(2) function.
+
+ If a non-zero address follows the "0x" in the "can't read
+ inpcb" message, it means lsof couldn't read inpcb contents
+ from the indicated address in kernel memory.
+
+3.32 What do the "unknown file system type" warnings mean?
+
+ Lsof may report a message similar to"
+
+ unknown file system type, v_op: 0x10472f10
+
+ in the NAME column for some files.
+
+ This means that lsof has encountered a vnode for the file
+ whose operation switch address (from v_op) references a
+ file system type for which there is no support in lsof.
+ After lsof identifies the file system type, it uses
+ pre-compiled code to locate the file system specific node
+ for the file where lsof finds information like file size,
+ device number, node number, etc.
+
+ To get some idea of what the file system type might be,
+ use nm on your kernel symbol file to locate the symbol name
+ that corresponds to the v_op address -- e.g., on Solaris
+ do:
+
+ $ nm -x /dev/ksyms | grep 0x10472f10
+ 0x10472f10 ... |file_system_name_vnodeops
+
+ Where "file_system_name" is the clue to the unsupported
+ file system.
+
+ Lsof doesn't use the v_op address to identify file system
+ types on all dialects. Sometimes it uses an index number
+ it finds in the vnode. It will translate that symbol to
+ a short name in the warning message -- e.g., "nfs3" -- if
+ possible.
+
+3.33 Installation
+
+3.33.1 How do I install lsof?
+
+ There is no "standard" way to install lsof. Too much
+ depends on local conditions for me to be able to provide
+ working install rules in the lsof make files. (The skeleton
+ install rules you will find just give "hints.") See the
+ "Installing Lsof" section of 00README for a fuller explanation.
+
+ To install lsof you will need to consider these questions:
+
+ * Who should be able to use lsof? (See HASSECURITY and
+ HASNOSOCKSECURITY in the "Security" section of 00README.)
+
+ * Where should lsof be installed? This is a decision
+ mostly dictated by local conditions. Somewhere in
+ /usr/local -- etc/ or sbin/ -- is a common choice.
+
+ * What permissions should I give the lsof executable?
+ The answer to this varies by dialect. The make files
+ have install rules that give hints. The "Installing
+ Lsof" section of 00README gives information, too.
+
+ * What if I want to install lsof in a shared file system
+ for machines that require different lsof configurations?
+ See the next question and answer, "How do I install a
+ common lsof when I have machines that need differently
+ constructed lsof binaries?"
+
+3.33.2 How do I install a common lsof when I have machines that
+ need differently constructed lsof binaries?
+
+ A dilemma that faces some system administrators when they
+ install lsof in a shared file system -- e.g., NFS -- is
+ that they must have different lsof executables for different
+ systems.
+
+ The answer is to build an lsof wrapper script that is
+ executed in place of lsof. The script can use system
+ commands to determine which lsof binary should be executed.
+
+ Consider this example. You have HP-UX machines with 32
+ and 64 bit kernels that share the /usr/local/sbin directory
+ where you want to install lsof. Consequently, on each
+ system you must use a different lsof executable, built for
+ the system's bit size. (That's because lsof reads kernel
+ structures, sized by the kernel's bit size.)
+
+ One answer is to install three things in /usr/local/sbin:
+ 1) a 32 bit lsof as lsof32; 2) a 64 bit lsof as lsof64;
+ and 3) an lsof script. The script might look like this
+ one, based on work by Amir J. Katz:
+
+ #!/bin/sh
+ x=`/usr/bin/getconf KERNEL_BITS` # returns 32 or 64
+ if /usr/bin/test "X$x" = "X32"
+ then
+ lsof32 $*
+ else
+ if /usr/bin/test "X$x" = "X64"
+ then
+ lsof64 $*
+ else
+ echo "Can't determine which lsof executable to use;"
+ echo "getconf KERNEL_BITS says: $x"
+ exit 1
+ fi
+ fi
+
+ Solaris users should consult "How do I install lsof for
+ Solaris 7, 8 or 9?" for information on a similar trick
+ using the Solaris isaexec command.
+
+ Users of other dialects might be able to use a command like
+ uname(1) that can identify a distinguishing feature of the
+ system to be incorporated in pre-installed lsof executable
+ names. For example, use `uname -r` and install binaries
+ with suffixes that match `uname -r` output.
+
+3.34 Why do lsof 4.53 and above reject device cache files built
+ by earlier lsof revisions?
+
+ When lsof revisions 4.53 run and encounter a device cache
+ file built by an earlier revision, it will reject the file
+ and build a new one. The rejection will be advertised with
+ these messages:
+
+ lsof: WARNING: no /dev device in <name>: 2 sections
+ ...
+ lsof: WARNING: created device cache file: <name>
+
+ This happens because the header line of the device cache
+ file was changed at revision 4.53 to contain the number of
+ the device on which the device directory resides. The old
+ device cache file header line -- the "2 sections" line in
+ the above warning message, node reads "2 sections, dev=600".
+
+ This is not a serious problem, since lsof automatically
+ rebuilds the device cache file with the correct header
+ line.
+
+3.35 What do "like block special" and "like character special" mean
+ in the NAME column?
+
+ When lsof comes across an open block or character file
+ whose device, raw device and inode place it somewhere other
+ than /dev (or /devices), lsof doesn't report the /dev (or
+ /devices) name in the NAME column. Instead lsof reports
+ the file system name and device or path name in the NAME
+ column and parenthetically adds "like block special <path>"
+ or "like character special <path>".
+
+ The value for <path> will point to a block or character
+ device in /dev (or /devices) whose raw device number matches
+ that of the open file being reported, but whose device
+ number or node number (or both) don't match.
+
+ Such an open file is connected to a device node that has
+ been created in a directory other than /dev (or /devices.)
+ See mknod(8) for information on how such nodes are created.
+ (Generally one needs root power to create device nodes with
+ mknod.)
+
+3.36 Why does an lsof make fail because of undefined symbols?
+
+ When lsof is compiled via the `make` step and the final
+ load step fails because of missing symbols, the problem
+ may not be lsof. The problem may be that ld, called by
+ the compiler as part of the `make` step, can't find some
+ library that lsof needs.
+
+ First check the last compiler line of the make operation
+ -- e.g., the last line with cc or gcc in it before the
+ undefined symbol report -- for loader arguments, i.e.,
+ ones beginning with "-l". Except for "-llsof" the rest
+ name system libraries. ("-L./lib" precedes "-llsof" to
+ tell the loader its location.)
+
+ Check that all the named system libraries exist. Look in
+ /lib and /usr/lib as a start, but that may not be the only
+ place system libraries live. Consult your dialect's
+ documentation, e.g., the compiler and loader man pages,
+ for other possible locations.
+
+ If some system library doesn't exist, that may mean it was
+ never installed or was removed. You'll have to re-install
+ the missing library.
+
+ You may find that all the system libraries lsof uses exist.
+ Your next step might be to use nm and grep to see if any
+ of them contain the undefined symbols.
+
+ $ nm library | grep symbol
+
+ If the undefined symbol exists in some library named by
+ the lsof make step, then you might have a problem with some
+ environment variable that controls the load step. The most
+ common is LD_LIBRARY_PATH. It may have a setting that
+ causes ld to ignore a directory containing a library lsof
+ names. If this is the case, try unsetting LD_LIBRARY_PATH
+ in the environment of the ld process -- e.g., do:
+
+ $ unset LD_LIBRARY_PATH
+ or
+ % unsetenv LD_LIBRARY_PATH
+
+ Consult your ld man page for other environment variables
+ that might affect library searching -- e.g., LIBPATH, LPATH,
+ SHLIB_PATH, etc.
+
+ If the undefined function doesn't exist in any libraries
+ lsof names, check other libraries. See if the function
+ has a man page that names its library. If the latter is
+ true, please let me know, because that is an lsof problem
+ I need to fix.
+
+ If none of these solutions work for you, send me some
+ documentation via e-mail at <abe@purdue.edu>. Include `uname
+ -a` output, the output of the lsof `Configure ...` and `make`
+ steps, and the contents of the environment in force when the
+ `make` step was executed -- e.g., `env` or `printenv` output.
+ If you've located the libraries lsof names, send me that
+ information, too. Make sure "lsof" appears in the "Subject:"
+ line so my e-mail filter won't classify your letter as Spam.
+
+3.37 Command Regular Expressions (REs)
+
+3.37.1 What are basic and extended regular expressions?
+
+ Lsof's ``-c'' option allows the specification of regular
+ expressions (REs), enclosed in two slash ('/') characters and
+ followed by these modifiers:
+
+ b the RE is a basic RE.
+ i ignore case.
+ x the RE is an extended RE (the default).
+
+ Note: the characters of the regular expression may need to
+ be quoted to prevent their expansion by the shell.
+
+ Example: this RE is an extended RE that matches exactly
+ four characters, whose third may be an upper ('O') or lower
+ case ('o') oh:
+
+ -c /^..o.$/i
+
+ For simplicity's sake, an RE that is acceptable to egrep(1)
+ is usually called an extended RE.
+
+ REs suitable for the old line editor, ed(1), are often
+ called basic REs (and sometimes also called obsolete).
+
+ These are some ways basic REs usually differ from extended
+ REs. (There are other differences.)
+
+ * `|', `+', `?', '{', and '}' are ordinary characters.
+
+ * `^' is an ordinary character except at the beginning of
+ the RE.
+
+ * `$' is an ordinary character except at the end of the
+ RE.
+
+ * `*' is an ordinary character if it appears at the
+ beginning of the RE.
+
+ For more information on REs and the distinction between
+ basic and extended REs, consult your dialect's man pages
+ for ed(1), egrep(1), sed(1), and possibly regex(5) or
+ regex(7).
+
+3.37.2 Why can't I put a slash in a command regular expression?
+
+ Since a UNIX command name is the last part of a path to
+ the command's executable, the lsof command regular expression
+ (RE) syntax uses slash ('/') to mark the beginning and end
+ of an RE. Slash may not appear in the RE and the `\'
+ back-slash escape is ineffective for "hiding" it.
+
+ More likely than not, if you try to put a slash in an lsof
+ command RE, you'll get this response:
+
+ $ lsof -s/.\// ...
+ lsof: invalid regexp modifier: /
+
+ Lsof is complaining the the first character it found after
+ the second slash isn't an lsof command RE modifier -- 'b',
+ 'i', or 'x'.
+
+3.37.3 Why does lsof say my command regular expression wasn't found?
+
+ When you use both forms of lsof's -c option --
+ ``-c <command>'' and ``-c /RE/[m]'' -- and ask that lsof
+ do a verbose search (``-V''), you may be surprised that
+ lsof will say that the regular expression wasn't found.
+
+ This can happen if the ``-c <command>'' form matches first,
+ because then the ``-c/RE/[m]'' test will never have been
+ applied. For example:
+
+ $ ./lsof -clsof -c/^..o.$/ -V -adcwd
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
+ lsof 7850 abe cwd VDIR 6,0 2048 96442 / (/dev/sd0a)
+ lsof: no command found for regex: ^..o.$
+
+ The ``-clsof'' option matched first, so the ``-c/^..o.$/
+ option wasn't tested.
+
+3.38 Why doesn't lsof report on shared memory segments?
+
+ Lsof reports on shared memory segments only if they're
+ associated with an open file. That's consistent with lsof's
+ mission -- to LiSt Open Files. Shared memory segments with
+ no file associations aren't open files.
+
+ That's not to say that a report on shared memory segments
+ and their associated processes wouldn't be useful. But it
+ calls for a new tool, not more baggage for lsof.
+
+3.39 Why does lsof report two instances of itself?
+
+ When you ask lsof to report all open files and it has
+ permission to do so, you may see two lsof processes in the
+ output. The processes are connected via pipes -- e.g.,
+ here's an HP-UX 11 example.
+
+ COMMAND PID USER FD TYPE DEVICE ...
+ ...
+ lsof 29450 abe 7w PIPE 0x48732408 ...
+ lsof 29450 abe 8r PIPE 0x48970808 ...
+ ...
+ lsof 29451 abe 6r PIPE 0x48732408 ...
+ lsof 29451 abe 9w PIPE 0x48970808 ...
+
+ The first process will usually be the lsof you initiated;
+ the second, an lsof child process that is used to isolate
+ its parent process from kernel functions that can block --
+ e.g., readlink() or stat().
+
+ Information to and from the kernel functions is exchanged
+ via the two pipes. When the parent process detects that
+ the child process has become blocked, it attempts to kill
+ the child. Depending on the UNIX dialect that may succeed
+ or fail, but the parent won't be blocked in any event.
+
+ See the "BLOCKS AND TIMEOUTS" and "AVOIDING KERNEL BLOCKS"
+ sections of the lsof man page for more information on why
+ the child process is used and how you can specify lsof
+ options to avoid it. (Caution: that may be risky.)
+
+3.40 Why does lsof report '\n' in device cache file error messages?
+
+ Lsof revisions prior to 4.58 may report '\n' in error
+ messages it delivers about problems in the device cache
+ file -- e.g.,
+
+ lsof: WARNING: no ...: 4 sections\n
+
+ That's deliberately done to show the exact contents of the
+ device cache file line about which lsof is complaining,
+ including its terminating NL (New Line) '\n' character.
+ In the above example the line in the device cache file
+ causing the lsof complaint contains "4 sections" and ends
+ with a '\n'.
+
+ At revision 4.58 and above, device cache error messages
+ like the one in the above example have been changed to
+ read:
+
+ lsof: WARNING: no ...: line "4 sections"
+
+ The terminal '\n' is no longer reported, the line contents
+ are enclosed in double quote marks ('"'), and the word
+ "line" has been added as a prefix to denote that what
+ follows is a line from the device cache file.
+
+3.41 Kernel Symbol and Address Problems
+
+3.41.1 What does "lsof: WARNING: name cache hash size length error: 0"
+ mean?
+
+ When run on some systems, lsof may issue this warning:
+
+ lsof: WARNING: name cache hash size length error: 0
+
+ That is an example from a FreeBSD system where lsof reads
+ the kernel's _nchash variable and finds its value is zero.
+
+ Similar warnings include:
+
+ WARNING: kernel name cache size:
+ WARNING: can't read kernel's name cache:
+ WARNING: no name cache address
+ WARNING: name cache hash size length error:
+ WARNING: unusable name cache size:
+
+ These warnings are issued when lsof is attempting to read
+ the kernel's name cache information. They are usually the
+ result of a mis-match between the addresses for kernel
+ symbols lsof gets via nlist(2) and the addresses in use by
+ the kernel.
+
+ Lsof usually gets kernel symbol addresses from what it
+ believes to be the kernel boot file. In FreeBSD, for
+ example, that's the path returned by getbootfile(3), usually
+ /kernel. The boot file can have other names in other UNIX
+ dialects -- /unix, /vmunix, /bsd, /netbsd, /mach, /stand/vmunix,
+ etc.
+
+ Lsof will get incorrect (mismatched) addresses from the
+ boot file if it has been replaced by a newer one which
+ hasn't yet been booted -- e.g., if this is done in FreeBSD:
+
+ # mv /kernel /kernel.OLD
+ # mv /kernel.NEW /kernel
+
+ Until the FreeBSD system is rebooted, the booted kernel is
+ /kernel.OLD, but getbootfile() says it is /kernel. If
+ symbol addresses important to lsof in /kernel.OLD and
+ /kernel don't match, the lsof WARNING messages result.
+
+3.41.2 Why does lsof produce "garbage" output?
+
+ Kernel name cache warnings may not be the only sign that
+ lsof is using incorrect symbol addresses to read kernel
+ values. If there's no reasonable test lsof can make on
+ what it reads from the kernel, it may issue other warnings
+ or even report nonsensical results.
+
+ The warnings may appear on STDERR, such as:
+
+ lsof: can't read proc table info
+
+ Or the warnings may appear in the NAME column as messages
+ saying lsof can't read or interpret some kernel structure --
+ e.g.,
+
+ ... NAME
+ ... can't read file struct from 0x12345
+
+ One possible work-around is to point lsof's kernel symbol
+ address gathering at the proper boot file. That can be
+ done with lsof's -k option -- e.g.,
+
+ $ lsof -k /kernel.OLD
+
+ The best work-around is to make sure the standard boot file
+ is properly sited -- e.g., if you've moved a new /kernel
+ in place, boot it.
+
+3.42 Why does lsof report open files when run as super user that
+ it doesn't report when run with lesser privileges?
+
+ The most likely cause is that the HASSECURITY option was
+ selected when the lsof executable was built.
+
+ If HASSECURITY is defined when lsof is built, and lsof is
+ run with the privileges of a non-ROOT user, it will only
+ list open files belonging to the user. The same lsof
+ executable, when run with root user privileges, will list
+ all open files.
+
+ However, if HASSECURITY and HASNOSOCKSECURITY are both
+ defined when lsof is built, lsof will list open files
+ belonging to the user and will also list anyone else's open
+ socket files, provided their listing is selected with the
+ "-i" option.
+
+ So first ask yourself if the process whose open files lsof
+ won't list belong to a user other than the one under which
+ you're running lsof, and are not open socket files. If
+ either is true, use lsof's help (-h or -?) option and look
+ for a line near the bottom of the help panel that says:
+
+ "... can list all files..."
+
+ If the leading "..." says "Only root" then HASSECURITY was
+ defined when lsof was built. If the trailing "..." says
+ ", but anyone can list socket files" then HASNOSOCKSECURITY
+ was also defined.
+
+ Should you want an lsof not built with HASSECURITY defined,
+ rerun the lsof Configure script. If you let Configure do
+ customization, make sure you answer 'n' when it asks if
+ you want to enable HASSECURITY and HASNOSOCKSECURITY. If
+ you don't need to do customization, you can rebuild lsof
+ with the "-n" option to Configure. Here's an example of
+ such a rebuild sequence:
+
+ $ Configure -clean
+ $ Configure -n <dialect-abbreviation>
+ $ make
+
+ More information on the HASSECURITY and HASNOSOCKSECURITY
+ options may be found in the "Security" section of the
+ 00README file of the lsof distribution.
+
+3.43 Test Suite Problems
+
+3.43.1 Errors all tests can report:
+
+3.43.1.1 Why do tests complain "ERROR!!! can't execute ../lsof"?
+
+ All tests in the test suite expect an executable lsof file
+ to exist in the tests parent directory, ../lsof.
+
+ If there's none there, the tests/Makefile has a rule to
+ make it, but there are probably circumstances where that
+ rule may fail.
+
+ The work-around is to re-Configure and re-make lsof, then
+ run the test suite.
+
+3.43.1.2 Why do tests complain "ERROR!!! can't find ..." a file?
+
+ Many tests create (or use from a supplied environment
+ variable path) a test file and use lsof to find it. When
+ lsof can't file the file, the tests report the error with
+ messages of the form:
+
+ ERROR!!! can't find ... : <some file path>
+ or
+ ERROR!!! lsof couldn't find ...
+
+ These type of error messages mean that the lsof field output
+ delivered to the test didn't contain a file that the test
+ could identify as the one it intended lsof to find. It
+ might also mean that the process information -- command
+ name, PID or parent PID -- didn't match what the test
+ expected.
+
+ This could imply a bug in the test or a bug in lsof. Try
+ using lsof to find a known file that is open. For example,
+ while in the tests sub-directory, do this:
+
+ $ sleep 30 < Makefile
+ $ ../lsof Makefile
+
+ If lsof doesn't report that Makefile is open, then the
+ fault may be with lsof. If lsof reports the file is open,
+ search further in the test code for the failure cause.
+
+3.43.1.3 Why do some tests fail to compile?
+
+ If a test suite program fails to compile, it may be because
+ I've never had an opportunity to compile the test on the
+ particular UNIX version you are using.
+
+ See Appendix B in 00TEST for a list of the UNIX dialects
+ where the test suite has been validate.
+
+3.43.1.4 Why do some tests always fail?
+
+ There are several tests in the optional group that have
+ conflicting or special requirements:
+
+ LTbigf needs a dialect and file system that support
+ large files.
+
+ LTlock won't work if the tests/ sub-directory is
+ on an NFS file system.
+
+ LTnfs won't work if the tests/ sub-directory is
+ not on an NFS file system.
+
+ So for two tests in particular, LTlock and LTnfs, one will
+ generally fail.
+
+ Some failing tests can be run successfully by supplying to
+ them a path to the appropriate type of file system with
+ the -p option.
+
+3.43.1.5 Why does the test suite say it hasn't been validated on
+ my dialect?
+
+ When you use the default rule of the test suite's Makefile,
+ it may issue this complaint:
+
+ $ cd tests
+ $ make
+ !!!WARNING!!!
+
+ This dialect or its particular version may not have
+ been validated with the lsof test suite. Consequently
+ some tests may fail or may not even compile.
+
+ !!!WARNING!!!
+
+ You are then given the opportunity to answer 'y' to have
+ the test suite operation continue.
+
+ This message means that the tests/TestDB file in the tests
+ sub-directory doesn't show that the test suite has been
+ run with the combination of compiler flags found in
+ tests/config.cflags. The tests might nor run; they may
+ encounter compiler failures.
+
+ See 00TEST for more information on the UNIX dialects where
+ the test suite has been validated and on the workings of
+ TestDB and its supporting scripts.
+
+ When the tests/Makefile "auto" rule is used, the message
+ is more terse and the condition is fatal.
+
+ This suite has not been validated on:
+
+ <dialect_description>
+
+ No opportunity to continue is offered.
+
+ The tests/Makefile "silent" rule will skip checking for
+ the validation footprint.
+
+3.43.1.6 Why do the tests complain they can't stat() or open()
+ /dev/mem or /dev/kmem?
+
+ When the tests detect that lsof for the dialect reads its
+ information from kernel memory (i.e., the LT_KMEM definition
+ is present in tests/config.cflags), and when the lsof
+ executable path is ../lsof, the tests make sure they can
+ stat() and open() for read access the relevant kernel memory
+ devices, /dev/kmem and possibly /dev/mem.
+
+ If those stat() or open() operations fail, the tests issue
+ an error message and quit. The message explains why the
+ system rejected the operation in terms of system "errno"
+ symbols and messages. More often than not the explanation
+ will be that the process lacks permission to access the
+ indicated device node.
+
+ One work-around is to give the lsof executable being tested
+ the necessary permission -- e.g., via chgrp, chmod, etc.
+ -- and set its path in the LT_LSOF_PATH environment variable.
+ (See 00TEST.)
+
+ Another work-around is to make sure the process that runs
+ the tests has the necessary permissions -- e.g., run it as
+ root, or enable the process login to access the resources.
+ For example, I can run the tests on my personal work-station
+ because /dev/kmem and /dev/mem are readable by the "kmem"
+ group and my login is in that group.
+
+
+3.43.2 LTbigf test issues
+
+3.43.2.1 Why does the LTbigf test say that the dialect doesn't
+ support large files?
+
+ Large file support is defined dialect by dialect in the
+ lsof source files and Configure script. If large file
+ support isn't defined there, it isn't defined in the LTbigf
+ test.
+
+ If you think that's wrong for a particular dialect, contact me
+ via e-mail at <abe@purdue.edu>. Make sure "lsof" appears in the
+ "Subject:" line so my e-mail filter won't classify your letter
+ as Spam.
+
+3.43.2.2 Why does LTbigf complain about operations on its config.LTbigf*
+ file?
+
+ The LTbigf must be able to write a large file test (size
+ > 32 bits) and seek within it and the process file ulimit
+ size must permit the operation. If the default location
+ for the test file, tests/, isn't on a file system enabled
+ for large file operations or if the process ulimit file
+ block size is too small, lsof will get file operation
+ errors, particularly when seeking
+
+ There may be a work-around. Specify the path to a file
+ LTbigf can write in a file system enabled for large file
+ operations a the -poption. Make sure that the ulimit file
+ block size permits writing a large file. For example,
+ presuming /scratch23 is large-file-enabled, and presuming
+ you have permission to raise the ulimit file block size,
+ this shell commands will allow the LTbigf test to run on
+ AIX:
+
+ $ ./LTbigf -p /scratch23/abe/bigfile
+
+ (Note: syntax for the ulimit command varies by dialect and
+ by shell. Discovering the proper variant is left to the
+ reader.)
+
+ More information on this subject can be found in the LTbigf
+ description in the 00TEST file. If course, the LTbigf.c
+ source file in tests/ is the ultimate source of information,
+
+3.43.2.3 Why does LTbigf warn that lsof doesn't return file offsets?
+
+ On some dialects (e.g., Linux) lsof can't report file
+ offsets, because the data access method underlying lsof
+ doesn't provide them. If LTbigf knows that lsof can't
+ report file offsets for the dialect, it issues this warning:
+
+ LTbigf ... WARNING!!! lsof can't return file offsets
+ for this dialect, so offset tests have
+ been disabled.
+
+ LTbigf then performs the size test and skips the offset
+ tests.
+
+ For more information see 00TEST and the "Why doesn't
+ /proc-based lsof report file offsets (positions)?" Q&A of
+ this file.
+
+3.43.3 Why does the LTbasic test complain "ERROR!!! lsof this ..."
+ and "ERROR!!! lsof that ..."?
+
+ The LTbasic test program uses lsof to examine a running
+ lsof process. It looks for the lsof current working
+ directory, executable (if possible), and kernel memory file
+ (if applicable).
+
+ Failures to find those things result in the LTbasic error
+ messages. More information on how LTbasic produces the error
+ messages may be found in the LTbasic.c source file.
+
+ On HP-UX 11.11 and higher, for example, if the test's current
+ working directory is on a loopback (LOFS) file system, LTbasic
+ won't be able to find the current working directory of the lsof
+ process because of a bug in the HP-UX kernel.
+
+ The solution for that HP-UX problem is to install an HP-UX
+ patch. See the answer to the "Why doesn't PSTAT-based lsof
+ report a CWD that is on a loopback (LOFS) file system?"
+ question for more information on the patch.
+
+3.43.4 NFS test issues
+
+3.43.4.1 Why does the LTnfs test complain "couldn't find NFS file ..."?
+
+ The LTnfs test must work with an NFS test file. After it
+ opens the file it asks lsof to find it on an NFS file system.
+ If the file isn't on an NFS file system, lsof won't find it,
+ and the NFS test script complains and fails.
+
+ The work-around is to use -p option to supply a path to a
+ regular NFS file (not a directory) that is on an NFS file
+ system that LTnfs can read. Presuming /share/bin/file is
+ such a file and can be opened for reading by the LTnfs
+ test, this sample shell command could be used to run the
+ LTnfs test successfully:
+
+ $ ./LTnfs -p /share/bin/file
+
+ (If the NFS file system is enabled for large files, the
+ NFS test will produce the error message described in the
+ following Q&A.)
+
+3.43.5 LTnlink test issues
+
+3.43.5.1 Why does the LTnlink test complain that its test file is on
+ an NFS file system?
+
+ The LTnlink test may complain:
+
+ LTnlink ... WARNING!!! test file <path> is NFS mounted.
+
+ and then issue an explanation and a hint about using the
+ -p option.
+
+ The LTnlist test does this because of the way NFS file
+ links are managed when an NFS file is unlinked and the
+ unlinking process still has the file open. Unlike with
+ files on a local file system, when an NFS file that is
+ still open is unlinked, its link count is not reduced.
+
+ The file name is changed to a name of the form .nfsxxxx
+ and the link count is left unchanged until the process
+ holding the file open closes it. That's done by NFS so it
+ can keep proper track of the file on NFS clients and servers.
+
+ Since the link count isn't reduced when the LTnlink test
+ program closes the NFS test file it still has open, lsof
+ won't find it for LTnlink with a link count of zero.
+ Consequently, LTnlink disables that test section and issues
+ its warning.
+
+ The warning suggests that the unlink test section can be
+ run by giving LTnlink a path to a test file with the -p
+ option. That path must name a file LTnlink can write and
+ unlink. Presuming /scratch23/abe/nlinkfile is on a local
+ file system and the LTnlink test can write to it and unlink
+ it, this sample shell command can be used to run the complete
+ LTnlink test successfully:
+
+ $ LTnlink -p /scratch23/abe/nlinkfile
+
+3.43.5.2 Why does LTnlink delay and report "waiting for link count
+ update: ..."?
+
+ On some UNIX dialects and file system combinations the
+ updating of link count after a file has been unlinked can
+ be delayed. Consequently, lsof won't be able to report
+ the updated link count to LTnlink for a while.
+
+ When lsof doesn't report the proper link count to LTnlink,
+ it sleeps and repeats the lsof call, using the "waiting
+ for link count update: ..." message as a signal that it is
+ waiting for the expected lsof response. The wait cycle
+ duration is limited to approximately one minute.
+
+3.43.6 LTdnlc test issues
+
+3.43.6.1 Why won't the LTdnlc test run?
+
+ Lsof is unable to access the DNLC cache on AIX, because the
+ kernel symbols for the DNLC aren't exported. Contact IBM
+ to learn why that decision was made.
+
+ The LTdnlc test won't work on Apple Darwin because lsof
+ can't obtain reliable DNLC information.
+
+ The LTdnlc test may fail on other dialects. Failure causes
+ include: a busy system with a DNLC that is changing rapidly;
+ path name components too large for the DNLC; a file system
+ -- e.g., NFS, /tmp, loopback -- which doesn't fully
+ participate in the DNLC; or DNLC limitations (Many DNLC
+ implementations will only store path name components if
+ they are 31 characters or less.)
+
+ If you suspect the file system doesn't fully participate
+ in kernel DNLC processing, as a work-around rebuild and
+ test lsof on one that does.
+
+3.43.6.2 What does the LTdnlc test mean by "... <path> found: 100.00%"?
+
+ Even when it succeeds the LTdnlc test will report:
+
+ LTdnlc ... /export/home/abe/src/lsof4/tests found: 100.00%
+
+ This message means that the LTdnlc test asked lsof to find
+ the file at the indicated path five times and lsof found
+ the full path name in the indicated percentage of calls.
+ The LTdnlc test considers it a failure if the percentage
+ falls below 50.0%
+
+3.43.6.3 Why does the DNLC test fail?
+
+ The DNLC test may fail when some component of the lsof
+ tests/ sub-directory can't be cached by the kernel DNLC.
+ Some kernels have a limit on the length of individual
+ components (typically) 32.
+
+3.43.7 Why hasn't the test suite been qualified for 64 bit HP-UX
+ 11 when lsof is compiled with gcc?
+
+ When I attempted to qualify lsof for HP-UX 11, compiled
+ with gcc 3.0, the LTsock test failed. I traced the failure
+ to a gcc compilation error. Because LTsock is an important
+ test, I didn't feel that the test suite was qualified if
+ it failed.
+
+ LTsock compiles and runs correctly on 64 bit HP-UX 11 when
+ compiled with HP's ANSI-C.
+
+3.43.8 LTszoff test issues
+
+3.43.8.1 Why does LTszoff warn that lsof doesn't return file offsets?
+
+ On some dialects (e.g., Linux) lsof can't report file
+ offsets, because the data access method underlying lsof
+ doesn't provide them. If LTszoff knows that lsof can't
+ report file offsets for the dialect, it issues this warning:
+
+ LTszoff ... WARNING!!! lsof can't return file offsets
+ for this dialect, so offset tests have
+ been disabled.
+
+ LTszoff then performs the size test and skips the offset
+ tests.
+
+ For more information see 00TEST and the "Why doesn't
+ /proc-based lsof report file offsets (positions)?" Q&A of
+ this file.
+
+3.43.9 LTlock test issues
+
+3.44 File descriptor list (the ``-d'' option) problems
+
+3.44.1 Why does lsof reject a ``-d'' FD list?
+
+ Lsof rejects ``-d'' FD lists that contain both exclusions
+ and inclusions with messages like:
+
+ lsof: exclude in an include list: ^1
+ lsof: include in an exclude list: 2
+
+ That's because ``-d'' FD lists are processed as ORed lists,
+ so it makes no sense for them to contain both exclusions
+ and inclusions.
+
+ I.e.,, if a ``-d'' FD list were to contain ``^cwd,1'', the
+ ``^cwd'' member is useless, because the ``1'' member
+ dominates by saying "include only FD 1". That effectively
+ excludes ``cwd'' FD.
+
+ Note that lists may have multiple members of the same type,
+ exclude or include. They are processed as an ORed set.
+ If an FD isn't excluded by any member of an exclude list,
+ it is selected. If an FD is included by any member of an
+ include list, it is selected.
+
+3.44.2 Why are file descriptors other than those in my FD list
+ reported?
+
+ The FD list that follows ``-d'' excludes or includes file
+ descriptors, but unless the ``-a'' (AND) option is specified,
+ the FD list selections are ORed to the other selections.
+
+ For example, the following lsof command will cause all file
+ descriptors to be listed for the lsof command, and all but
+ the cwd descriptor for all other commands, probably not
+ what was intended.
+
+ $ lsof -clsof -d^cwd
+
+ Hint: use ``-a'' -- e.g.,
+
+ $ lsof -clsof -a -d^cwd
+
+3.45 How can I supply device numbers for inaccessible NFS file
+ systems?
+
+ When lsof can't get device numbers for inaccessible NFS file
+ systems via stat(2) or lstat(2), it attempts to get them from
+ the mount table's dev=xxx options. Successes are reported with
+ a warning message that indicates the source of the device
+ number and that output might be incomplete as a consequence of
+ the warnings.
+
+ Some system mount tables -- e.g., Linux /proc/mounts -- don't
+ have a dev=xxx option. In that case, and provided lsof for the
+ dialect supports them, you can use the +m option to create a
+ mount table supplement file and the "+m m" option to use it.
+
+ First check the lsof -h (help) output to see if the +m and
+ "+m m" options are supported. If they are, use +m to create a
+ mount table supplement file when all mounted file systems are
+ accessible. Use "+m m" later to make the supplement available
+ when some mounted file systems might not be available.
+
+ Here's an example that creates a mount supplement file in
+ $HOME/mnt-sup and later makes it available to lsof.
+
+ $ rm -f $HOME/mnt-sup
+ $ lsof +m > $HOME/mnt-sup
+ ...
+ $ lsof +m $HOME/mnt-sup <other lsof options>
+
+ If lsof has to get the device number from the supplement, it
+ will issue an informative warning message. The warning can be
+ suppressed with lsof's -w option.
+
+ Caution! Since the mount table supplement file is static, it
+ is its supplier's responsibility to update it as file system
+ mounts change.
+
+ For more information, consult the lsof man page. The
+ "ALTERNATE DEVICE NUMBERS" section has useful information on
+ how lsof acquires device numbers when stat(2) or lstat(2)
+ fail.
+
+3.46 Why won't lsof find open files on over-mounted file systems?
+
+ When a file system, /xyz for example, is mounted on the same
+ mount point as another file system, /abc for example, running
+ lsof with an argument of the path of the first file system's
+ mount point -- the over-mounted one, /abc -- probably will not
+ reveal any files open on /abc.
+
+ That's because lsof looks for open files on a file system by
+ looking for files with the file system's device number. The
+ two file systems usually have different device numbers and lsof
+ determines the device number search key from the supplied name
+ of the second file system.
+
+ A general work-around exists only for Linux. On that UNIX
+ dialect, when you know the over-mounted file system's mount
+ point path, you can ask lsof to report on all open files and
+ grep that output for the path of the over-mounted file system
+ mount point.
+
+3.47 What can be done when lsof reports no more space?
+
+ Many lsof methods cache information in memory, using the
+ dialects malloc() library function. When malloc() can't
+ allocate the requested amount of memory, lsof exits with
+ warning messages similar to this AIX message:
+
+ lsof: no more dev-ch space at pid 2257750: 0x82a8e600
+
+ Lsof then exits immediately and produces no more output.
+
+ A possible work-around is to increase the memory foot print
+ of the shell that runs lsof. That is often done with the
+ ulimit(1) shell command.
+
+3.48 What if the lsof build encounters ar and ld problems?
+
+ The lsof main and library Makefiles use the library archiver,
+ ar, and the system loader, ld, applications. Improperly
+ located, installed or configured versions of them may cause the
+ lsof build to encounter errors with them.
+
+ The application producing the error should identify itself in
+ its error messages.
+
+ The first thing to check the path of the application that is
+ being used. Try `which ar` or `which ld` to see if perhaps the
+ PATH used during the build might be causing the wrong archiver
+ or loader to be used.
+
+ If the problem is with the use of the wrong archiver, and it's
+ not possible to correct the PATH to it, try using the LSOF_AR
+ environment variable to specify the path to and arguments for
+ the correct archiver. See 00XCONFIG for more information and
+ note that LSOF_AR must specify the path to the archive
+ application and the arguments for it, less the terminating
+ library and module name arguments.
+
+ If the problem is with the loader, there is no lsof work-
+ around. That's because lsof calls the loader via the C
+ compiler, so the problem must be fixed at the compiler (system)
+ level.
+
+
+4.0 AIX Problems
+
+4.1 What is the Stale Segment ID bug and why is -X needed?
+
+ Kevin Ruderman reports that he has been informed by IBM
+ that processes using the AIX 3.2.x, 4.1[.12345]], 4.2[.1],
+ and 4.3.x kernel's readx() function can cause other AIX
+ processes to hang because of what appears to be file system
+ corruption.
+
+ This failure, known as the Stale Segment ID bug, is caused
+ by an error in the AIX kernel's journaled segment memory
+ handler that causes the kernel's dir_search() function
+ erroneously to believe directory entries contain zeroes.
+ The process using the readx() call need not be doing anything
+ wrong. Usually the system must be under such heavy load
+ that the segment ID being used in the readx() call has been
+ freed and then reallocated to another process since it was
+ obtained from kernel memory.
+
+ Lsof uses the readx() function to access library entry
+ structures, based on the segment ID it finds in the proc
+ structure of a process. Since IBM probably will never fix
+ the kernel bug, I've added an AIX-specific option to lsof
+ that controls its use of the readx() function.
+
+ By default lsof readx() use is disabled; specifying the
+ ``-X'' option enables readx() use.
+
+ If you want to change the default readx() behavior of AIX
+ lsof, change the HASXOPT, HASXOPT_ROOT, and HASXOPT_VALUE
+ definitions in dialects/aix/machine.h. You can also use
+ these definitions to enable or disable readx() -- consult
+ the comments in machine.h. You may want to disable readx()
+ use permanently if you plan to make lsof publicly executable.
+
+ When HASXOPT_ROOT is defined, lsof will restrict use of
+ the -X option to processes whose real UID is root; if
+ HASXOPT_ROOT isn't defined, any user may specify the -X
+ option. The Customize script offers the option to change
+ HASXOPT_ROOT when HASXOPT is defined and HASXOPT_ROOT is
+ named in any dialect's machine.h header file.
+
+ I have never seen lsof cause a problem with its use of
+ readx(), but I believe there is some chance it could, given
+ the right circumstances.
+
+4.1.1 Stale Segment ID APAR
+
+ Here are the details of the Stale Segment ID bug and IBM's
+ response, provided by Kevin Ruderman.
+
+ AIX V3
+ APAR=ix49183
+ user process hangs forever in kernel due to file
+ system corruption
+ STAT=closed prs TID=tx2527 ISEV=2 SEV=2
+ (A "closed prs" is one closed with a Permanent
+ ReStriction.)
+ RCOMP=575603001 aix v3 for rs/6 RREL=r320
+
+ AIX V4 (internal defect, no apar #)
+ prefix p
+ name 175671
+ abstract KERMP: loop for ever in dir_search()
+
+ Problem description:
+
+ 1. Some user application -- e.g., lsof -- gets the segment
+ ID (SID) for the process private segment of a target
+ process from the process table.
+
+ 2. The target process exits, deleting the process private
+ segment.
+
+ 3. The SID is reallocated for use as a persistent segment.
+
+ 4. The user application runs again and tries to read the
+ user area structure from /dev/mem, using the SID it read
+ from the process table.
+
+ 5. The loads done by the driver for /dev/mem cause faults
+ in the directory; new blocks are allocated; the size
+ changed; and zero pages created.
+
+ 6. The next application that looks for a file in the affected
+ directory hangs in the kernel's dir_search() function
+ because of the zero pages. This occurs because the
+ kernel's dir_search() function loops through the variable
+ length entries one at a time, moving from one to the
+ next by adding the length of the current entry to its
+ address to get the address of the next entry. This
+ process should end when the current pointer passes the
+ end of the known directory length.
+
+ However, while the directory length has increased, the
+ entry length data has not, so when dir_search() reaches
+ the zero pages, it loops forever, adding a length of
+ zero to the current pointer, never passing the end of
+ the directory length. The application process is hung;
+ it can't be killed or stopped.
+
+ IBM closed the problem with a PRS code (Permanent ReStriction)
+ under AIX Version 3 and had targeted a fix for AIX 4.2. They
+ have recently (I became aware of it September 10, 1996)
+ cancelled the defect report altogether and have indicated they
+ are not going to fix the defect.
+
+4.2 Gcc Work-around for AIX 4.1x
+
+ When gcc is used to compile lsof for AIX 4.1x, it doesn't
+ align one element of the user structure correctly. Xlc
+ sees the U_irss element as a type "long long" and aligns
+ it on an 8 byte boundary. That's because the default mode
+ of xlc is -qlonglong; when -qlonglong is enabled, the
+ _LONG_LONG symbol is also defined.
+
+ Gcc sees U_irss as a two element array of type long, because
+ _LONG_LONG isn't defined. Hence gcc aligns the U_irss
+ element array on a 4 byte boundary, rather than an 8 byte
+ one, making the gcc incantation of the user structure 4
+ bytes shorter than xlc's.
+
+ When the length of gcc's user structure is supplied as
+ argument 4 to the undocumented getuser() function of the
+ AIX kernel, getuser() rejects it as an incorrect size and
+ returns EINVAL.
+
+ Lsof has a work-around for this problem. It involves a
+ special test in the Configure script when the "aixgcc"
+ Configure abbreviation is used -- e.g.,
+
+ $ Configure -n aixgcc
+
+ The test is to compile a small program with gcc and check
+ the alignment of U_irss. If it's not aligned on an 8 byte
+ boundary, the Configure script makes a special copy of
+ <sys/user.h> in ./dialects/aix/aix<AIX_version> whose
+ U_irss will align properly, and generates compile time
+ options to use it.
+
+ While I have tested this work-around only with 4.1.4, it
+ should work with earlier versions of AIX 4.1. It does not
+ work for AIX 4.2; a different work-around is employed there.
+ (See the next section.)
+
+ If you want to use this technique to compile other AIX
+ 4.1x programs with gcc for using getuser(), check the
+ Configure script.
+
+ Stuart D. Gathman identified this gcc AIX alignment problem.
+
+4.3 Gcc and AIX 4.2[.1]
+
+ Alignment problems with gcc and AIX 4.2[.1] inside the user
+ structure are more severe, because there are some new 64
+ bit types in AIX that gcc doesn't yet (as of 2.7.x) support.
+ The <sys/user.h> U_irss element problem, discussed in 4.3
+ above, doesn't exist in 4.2[.1].
+
+ The AIX lsof machine.h header file has a work-around,
+ provided by Henry Grebler, that bypasses gcc alignment
+ problems. Later versions of gcc (e.g., 2.8.x) will probably
+ bypass the problems as well.
+
+4.4 Why won't lsof's Configure allow the use of gcc for AIX
+ below 4.1?
+
+ Gcc can't reliably be used to compile lsof for AIX versions
+ below AIX 4.1 because of possible kernel structure element
+ alignment differences between it and xlc.
+
+4.5 What is an AIX SMT file type?
+
+ When you run AIX X clients with the DISPLAY environment
+ variable set to ``:0.0'' they communicate with the AIX X
+ server via files whose kernel file structure has an undefined
+ type (f_type == 0xf) -- at least there's no definition for
+ it in <sys/file.h>.
+
+ These are Shared Memory Transport (SMT) sockets, an artifact
+ of AIXWindows, designed for more efficient data transfers
+ between the X server and its clients.
+
+ Henry Grebler and David J. Wilson alerted me to the existence
+ of these files. Mike Feldman and others helped me identify
+ them as SMT sockets.
+
+ The curious reader can find more about SMT sockets in
+ /usr/lpp/X11/README.SMT.
+
+4.6 Why does AIX lsof start so slowly?
+
+ When AIX lsof starts it compares the running kernel's
+ identity to the one for which it was built, using
+ /usr/bin/oslevel. That comparison can sometimes take a
+ long time to complete, depending on the system's maintenance
+ level and how recently it was examined with oslevel.
+
+ AIX revisions 4.67 and above for AIX 5 and above don't use
+ oslevel to determine the kernel identity. They use uname(2)
+ instead, and it is much faster.
+
+ You can skip the oslevel test by suppressing warning messages
+ with lsof's -w option. Doing that carries with it the risk
+ of missing other warning messages, however.
+
+ You can also disable the kernel identity check by disabling
+ the definition of the HASKERNIDCK symbol by editing AIX
+ machine.h header file or by using the Customize script to
+ disable it.
+
+ See the "Why does lsof warn "compiled for x ... y; this is
+ z.?" section for more information.
+
+4.7 Why does exec complain it can't find libc.a[shr.o]?
+
+ When you try to execute lsof you may get this complaint:
+
+ exec(): 0509-036 Cannot load program ./lsof because of
+ the following errors:
+ 0509-022 Cannot load library libc.a[shr.o].
+ 0509-026 System error: A file or directory in
+ the path name does not exist.
+
+ This is probably the result of making lsof when the LIBPATH
+ environment variable contained a directory path that doesn't
+ contain libc.a. You can see what LIBPATH contained when
+ lsof was made by using the dump application on lsof. For
+ example, if LIBPATH contained /foo/bar when lsof was made,
+ you will see this (partial) dump output:
+
+ $ dump -H lsof
+ ...
+ ***Import File Strings***
+ INDEX PATH BASE ...
+ 0 /foo/bar
+
+ To correct the problem, revisit the lsof source directory
+ and remake lsof this way:
+
+ $ unset LIBPATH; make (sh or ksh)
+ or
+ % unsetenv LIBPATH; make (csh or tcsh)
+
+4.8 What does lsof mean when it says, "no PCB, CANTSENDMORE,
+ CANTRCVMORE" in a socket file's NAME column?
+
+ When an AIX application calls shutdown(2) on an open socket
+ file, but hasn't called close(2) on the file, the file will
+ remain visible to lsof as an open socket file without any
+ extended protocol information.
+
+ Lsof reports that state in the NAME column by saying that
+ there is "no PCB" (Protocol Control Block) for the protocol
+ (e.g., TCP in the NODE column). If the open socket file
+ has the state variables SO_CANTSENDMORE and SO_CANTRCVMORE
+ set -- i.e., from the shutdown(2) call -- lsof reports them
+ with the CANTSENDMORE and CANTRCVMORE notes in the NAME
+ column.
+
+4.9 When the -X option is used on AIX 4.3.3, why does lsof disable
+ it, saying "WARNING: user struct mismatch; -X option disabled?"
+
+ The -X option causes lsof to read the loader information
+ of the user structure from virtual memory via the readx()
+ system call. It does that with the user structure definition
+ from <sys/user.h> that was compiled into the lsof executable.
+
+ On AIX 4.3.3 there are two different user structure
+ definitions in two separate <sys/user.h> header files,
+ distributed at different times by IBM. If lsof was compiled
+ with one and the kernel on which lsof is being run was
+ compiled with the other, lsof normally won't get correct
+ loader information when it calls readx().
+
+ In an attempt to compensate for that difference, lsof makes
+ an independent check of the loader information by getting
+ the user structure's open file count via readx() and
+ comparing it to the open file count obtained independently
+ via getprocs(). When the two counts don't match, lsof
+ tries to read the count (and re-read the loader information)
+ with two offsets, based on observed differences between
+ the two user structures.
+
+ When one of the three attempts produces a correct open file
+ count, lsof uses its corresponding offset on subsequent
+ readings of the loader information.
+
+ When none of the three attempts produces a correct open
+ file count, lsof issues the WARNING message and disables
+ -X processing.
+
+ To eliminate this problem, obtain an lsof binary that
+ matches the kernel of the AIX 4.3.3 system where you want
+ to run lsof. Compiling lsof on the target system is the
+ preferred way to get a matching binary.
+
+4.10 Why doesn't the -X option work on my AIX 5L or 5.[123] system?
+
+ If your AIX 5L or 5.[123] system uses the ia64 architecture,
+ lsof needs setuid-root permission to be able to do the
+ processing that -X requires.
+
+ Check the output of `uname -a` to determine the architecture
+ type.
+
+ The work-around is to give lsof setuid-root permission.
+
+4.11 Why doesn't /usr/bin/oslevel report the correct AIX version?
+
+ The oslevel man page says, "The oslevel command reports
+ the level of the operating system using a subset of all
+ filesets installed on your system."
+
+ You can see which fileset is below the expected level with
+ oslevel's -l option. For example, if you believe your
+ system is at AIX level 4.3.3, but oslevel reports 4.3.2,
+ use this oslevel command to find the filesets below 4.3.3:
+
+ $ /usr/bin/oslevel -l 4.3.3.0
+
+ If you don't know what level argument to supply to oslevel's
+ -l option, use oslevel's -q option first.
+
+4.11.1 Why doesn't /usr/bin/oslevel report the correct AIX version
+ on AIX 5.1?
+
+ The subset list for oslevel on AIX 5.1 seems to include at
+ least two filesets, xlsmp.msg.en_US.rte and xlsmp.rte, that
+ do not install from AIX 5.1 media with a 5.1.0.0 level.
+ Hence, oslevel reports 5.0.0.0 instead of the expected
+ 5.1.0.0.
+
+ If either xlsmp.msg.en_US.rte or xlsmp.rte is installed,
+ lsof's Configure script and run-time tests will identify
+ the AIX version incorrectly. The run-time test will
+ issue a complaint message of this form:
+
+ lsof: WARNING: compiled for AIX version xxx; this is yyy.
+
+ You can correct the Configure test by pre-defining the
+ oslevel value, setting the correct value in the LSOF_VSTR
+ environment variable before running the Configure script
+ -- e.g., to pre-define AIX 5.1 when using ksh, do this:
+
+ $ LSOF_VSTR=5.1.0.0 Configure -n aix
+
+ You can't affect oslevel output without uninstalling
+ xlsmp.msg.en_US.rte and xlsmp.rte. If you can't do that,
+ you'll have to put up with the run-time complaint.
+
+4.12 Why does lsof for AIX 5.1 or above Power architecture
+ complain about kernel bit size?
+
+ When you run an lsof binary on an AIX 5.1 or above Power
+ system, it might complain:
+
+ lsof: FATAL: compiled for a 32 bit kernel.
+ The bit size of this kernel is 64.
+ or
+ exec: 0509-036 Cannot load program ./lsof because of
+ the following errors:
+ 0509-032 Cannot run a 64-bit program on a 32-bit
+ machine.
+
+ Starting at lsof revision 4.61, lsof binaries for Power
+ architecture systems running AIX 5.1 or above are closely
+ tied to the kernel bit size. Lsof must do that so it can
+ read and understand kernel structures.
+
+ Lsof's Configure script tunes the lsof configuration so
+ that the binary built in the make(1) step is adjusted to
+ the kernel bit size.
+
+ An lsof binary knows the bit size for which it was constructed,
+ tests the bit size of the kernel under which it is running,
+ and objects if the two sizes don't match. To see the bit
+ size for which lsof was constructed, run it with its -v
+ option and look for these lines in the output:
+
+ configuration info: 32 bit kernel
+ or
+ configuration info: 64 bit kernel
+
+ (Note: these lines will appear only in -v output for AIX
+ 5.1 and above lsof binaries, built for Power architecture.)
+
+ You can see the kernel bit size test method in the aix
+ stanza of the lsof Configure script and in the get_kernel_access()
+ function of the lsof .../dialects/aix/dproc.c source file.
+
+ There is more information on pre-defining the kernel bit
+ size when building lsof in Configure, 00PORTING, and
+ 00XCONFIG.
+
+ The only work-around is to use an lsof binary built to
+ match the running kernel bit size.
+
+4.13 What can't gcc be used to compile lsof on the ia64 architecture
+ for AIX 5 and above?
+
+ Gcc can't be used to compile lsof on the ia64 architecture
+ for AIX 5 and above because I haven't had access to a system
+ that has a working gcc compiler. The gcc compiler on my
+ one and only ia64 AIX 5.1 test system, provided by IBM,
+ didn't work at all.
+
+4.14 Why does lsof get a segmentation fault when compiled with gcc
+ for a 64 bit Power architecture AIX 5.1 kernel?
+
+ When lsof is configured with the lsof "aixgcc" Configure
+ abbreviation, the resulting lsof executable may cause a
+ segmentation violation when it is run. I've observed this
+ with gcc version 2.9-aix43-010414-7.
+
+ As far as I have been able to tell, the segmentation fault
+ is the result of a gcc compilation, loading, or library
+ error. Watching lsof run with gcc's companion debugger,
+ gdb, shows no error in the lsof source code that might
+ explain the fault.
+
+ The only work-around I know is to use the IBM C compiler
+ in place of gcc -- i.e., use the "aix" lsof Configure
+ abbreviation.
+
+4.15 Why does lsof ignore AFS on my AIX system?
+
+ The lsof Configure script quits on AIX when AFS is present,
+ the AIX version is greater than 4.3.3.0 or the AFS version
+ is greater than 3.5. That's because I have no test systems
+ available for those AIX and AFS version combinations.
+
+ When the lsof Configure script detects an AIX and AFS
+ version combination that is unsupported, it will report:
+
+ !!!FATAL: Lsof does not support AFS on this combination of
+ AIX and AFS versions. To disable AFS, set the
+ value of the AIX_HAS_AFS environment variable to
+ "no".
+
+ The only work-around is to set the AIX_HAS_AFS environment
+ variable as explained in the error message:
+
+ $ AIX_HAS_NSF=no; export AIX_HAS_NFS
+ $ ./Configure -n aix
+
+4.16 Why does lsof report "system paging space is low" and exit?
+
+ When AIX paging space runs low, the AIX kernel sends a SIGDANGER
+ signal to processes, warning them that they should reduce their
+ memory usage.
+
+ When lsof receives that signal, it issues the following fatal
+ error message and exits:
+
+ lsof: FATAL: system paging space is low.
+
+ A possible work-around is to limit the amount of information
+ lsof must cache in its process memory with the "-c", "-g", "-l"
+ and "-p" options.
+
+ Also see the answer to the "What can be done when lsof reports
+ no more space?" question.
+
+4.17 Why does lsof have a compilation problem on AIX 5.3 above
+ maintenance level 1?
+
+ On some AIX 5.3 systems with maintenance levels 2 and higher
+ installed, lsof 4.77 and below may not compile properly. The
+ compiler complains the snapshotObject structure definition,
+ needed by <j2/j2_inode.h>, is missing.
+
+ That problem is fixed in the 4.78 revision.
+
+
+5.0 Apple Darwin Problems
+
+5.1 What do /dev/kmem-based and libproc-based mean?
+
+ Lsof for Apple Darwin currently uses /dev/kmem to read kernel
+ data structures from which it gathers and reports open file
+ information. That version of lsof is called /dev/kmem-based
+ lsof.
+
+ At an upcoming release lsof will use a library called libproc
+ to obtain information about open files. That version of lsof
+ wil be called libproc-based lsof.
+
+ The /dev/kmem-based lsof sources may be found in the kmem
+ subdirectory of the dialects/darwin branch of the lsof source
+ tree. When the supporting version of Apple Darwin is released,
+ the libproc-based lsof sources will be found in
+ .../dialects/darwin/libproc.
+
+5.2 /dev/kmem-based Apple Darwin Questions
+
+5.2.1 Why does Configure ask for a path to the Darwin XNU kernel
+ header files?
+
+ When lsof was ported to Apple Darwin by Allan Nathanson at
+ revision 4.53, some kernel header files needed by lsof
+ weren't being exported by the developers. (That's still
+ true at lsof revision 4.76.)
+
+ At first a shell script that Allan provided would get the
+ missing header files by checking them out from the CVS
+ root. Although the script was updated from time to time,
+ eventually the re-organization of Darwin sources has made
+ it impossible to update the script to do an automatic
+ download of the missing header files.
+
+ At lsof revision 4.69 and above it is necessary for the Darwin
+ lsof builder to download the Darwin XNU kernel headers before
+ attempting to build lsof. The download my be done via a web
+ browser, starting at this URL:
+
+ http://www.opensource.apple.com/darwinsource/index.html
+
+ Once there, select the link to the Mac OS X version that
+ matches the one on the system where lsof is to be built.
+
+ Follow that link's "[ Source ]" link. Once there, select the
+ tar.gz link of the xnu* entry near the bottom of the page.
+ That entry should have a name that matches the xnu* name shown
+ by `uname -a` -- e.g., if uname reports:
+
+ $ uname -a
+ ... root:xnu/xnu-517.7.21 ...
+
+ Then the appropriate xnu* entry is xnu-517.7.21. Clicking
+ its link should lead to an "Apple Open Source" page requesting
+ an Apple ID and password.
+
+ Enter them if they're available. If an Apple ID and password
+ are not available, get them by following the instructions on
+ the page -- i.e., follow the signin.apple.com link.
+
+ Once a valid Apple ID and its password have been entered,
+ the download will begin. Select the saving of the downloaded
+ xnu*.tar.gz file in an appropriate place on the Mac OS X
+ system.
+
+ Once the download completes, install it. Use gunzip to
+ decompress the download and tar to extract the archive -- e.g.,
+
+ $ gunzip -c xnu-517.7.21.tar.gz | tar xf -
+
+ Remember the absolute path to the extracted archive. That is
+ its installed place. E.g., if the xnu-517.7.21.tar archive was
+ extracted to the lsof builder's home directory, its full
+ installation path will be something like:
+
+ ~/xnu-517.7.21
+
+ Now run the lsof Configure script. When it asks for the path
+ to the installed Darwin XNU kernel header files, supply the
+ path to the gunzip'd and extracted xnu* archive -- e.g.,
+ ~/xnu-517.7.21.
+
+ The path to the Darwin XNU kernel headers may also be
+ supplied to the Configure script in the DARWIN_XNUDIR
+ environment variable, eliminating the need to enter it
+ interactively -- e.g.,
+
+ $ DARWIN_XNUDIR=~/xnu-344.49 ./Configure -n darwin
+
+5.2.1.1 Why does Configure complain that Darwin XNU kernel header
+ files are missing?
+
+ These are some reasons why the lsof Configure script might
+ claim that Darwin XNU header files are missing:
+
+ * The wrong path to them was specified.
+
+ * The files and directories in the path are not readable
+ and searchable -- i.e., check the modes and ownerships.
+
+ * The downloaded archive doesn't match the Mac OS X
+ version of the system.
+
+ If in doubt, revisit the Darwin XNU kernel header file
+ download instructions in the answer to the question "Why
+ does Configure ask for a path to the Darwin XNU kernel
+ header files?"
+
+ If Configure still can't find Darwin XNU kernel header
+ files, contact me via e-mail at <abe@purdue.edu> for help.
+ Make sure "lsof" appears in the "Subject:" line so my e-mail
+ filter won't classify your letter as Spam.
+
+5.2.2 Why doesn't Apple Darwin lsof report text file information?
+
+ At the first port of lsof to Apple Darwin, revision 4.53,
+ insufficient information was available -- logic and header
+ files -- to permit the installation of VM space scanning
+ for text files. As of lsof 4.70 it is sill not available.
+
+ Text file support will be added to Apple Darwin lsof after
+ the necessary information becomes available.
+
+5.2.3 Why doesn't Apple Darwin lsof support IPv6?
+
+ At the first port of lsof to Apple Darwin, revision 4.53,
+ Apple Darwin lacked IPv6 support. IPv6 became available
+ in Apple Darwin version 1.5 and support for it was added
+ to lsof then.
+
+5.2.4 Why does lsof complain about a mismatch between the release
+ for which lsof was compiled and the booted Mac OS X release?
+
+ When lsof is started on the "Gold Master" Darwin release
+ (aka Mac OS X), it complains:
+
+ lsof: compiled for 1.0 release; this is 1.3.2.
+
+ This happens because the lsof binary released with Mac OS
+ X was built on a system whose release number (1.0) doesn't
+ match that of the released system -- usually 1.3.x Lsof
+ makes this check because UNIX dialect OS changes are often
+ accompanied by header file changes that affect lsof.
+
+ In this specific case, this error can be ignored. If you
+ don't want to do that, get the lsof distribution and build
+ lsof so its built-on and running-on Mac OS X release numbers
+ match.
+
+5.2.5 Why does lsof for Apple Darwin 8 and higher report
+ "stat(...): ..." in the NAME column?
+
+ Lsof for Apple Darwin 8 may report messages like these in the
+ NAME column:
+
+ stat(/private/var/run/asl_prune): No such file or directory
+ or
+ stat(/private/var/db/netinfo/local.nidb/Config): Permission denied
+
+ Those messages indicate that lsof was unable to collect open
+ file information for the paths enclosed in "stat(...)" with the
+ stat(2) function, because the function encountered the reported
+ error.
+
+ A work-around for the "Permission denied" error is to run lsof
+ with elevated privileges -- e.g., when logged on as the super
+ user.
+
+ If the stat(2) error message is "No such file or directory",
+ the file probably has been unlinked (removed) and there is no
+ lsof work-around.
+
+5.2.6 What are the limitations of Apple Darwin lsof link count
+ reporting?
+
+ Lsof for Apple Darwin cannot report link count information
+ reliably.
+
+ For Apple Darwin below 8 link count information is not always
+ available in the kernel node structures available to lsof.
+ When link count information is available, however, it includes
+ link counts of zero. Thus, using lsof's +L1 option may result
+ in the finding of some files whose link counts are zero.
+
+ Lsof can report only some link count information for Apple
+ Darwin 8 and above. Link count information is only available
+ for files where lsof can assemble the full file path and has
+ permission to apply stat(2) to it. (See the answer to the "Why
+ does lsof for Apple Darwin 8 and higher report "stat(...): ..."
+ in the NAME column?" question for more information on stat(2)
+ failures.)
+
+ Apple Darwin 8 and above files that have been unlinked and thus
+ have a link count of zero cannot be found by stat(2) -- i.e.,
+ stat(2) returns a "No such file or directory" error. As a
+ result lsof never displays link counts of zero and the use of
+ lsof's +L1 option to find them always fails.
+
+5.3 Libproc-based Apple Darwin Questions
+
+
+6.0 BSD/OS BSDI Problems
+
+6.0.5 Statement of deprecation
+
+ As of lsof revision 4.76 support for BSDI BSD/OS has been
+ dropped. The 4.76 distribution of lsof for BSDI BSD/OS may be
+ found on lsof.itap.purdue.edu in pub/tools/unix/lsof/OLD/src.
+
+
+7.0 DEC OSF/1, Digital UNIX, and Tru64 UNIX Problems
+
+7.1 Why does lsof complain about non-existent /dev/fd entries?
+
+ When you run lsof for Digital UNIX 3.2, lsof may complain:
+
+ lsof: can't lstat /dev/fd/xxx: No such file or directory
+ lsof: can't lstat /dev/fd/yyy: No such file or directory
+
+ (Or it may warn about other missing /dev/fd paths.) When
+ you do an ``ls /dev/fd'' none of the missing paths are listed.
+
+ This is caused by a bug in the DEC library function
+ getdirentries(). For some reason, when /dev/fd is a file
+ system mount point, getdirentries() returns an incorrect
+ size for it to readdir(). (Lsof calls readdir() in its
+ ddev.c readdev() function.) Because of the incorrect size,
+ readdir() goes past the end of the /dev/fd directory buffer,
+ encounters random paths and returns them to lsof. Lsof
+ then attempts to lstat(2) the random paths, gets error
+ replies from lstat(2), and complains about the paths.
+
+ Duncan McEwan discovered this error and has reported it to
+ DEC. Duncan also supplied an alternate readdir() function
+ as a work-around. I've incorporated his readdir() in
+ dialects/osf/ddev.c (as the static ReadDir() function) with
+ some slight modifications, and enabled its use when the
+ USELOCALREADDIR symbol is defined.
+
+ The Configure script defines USELOCALREADDIR for Digital
+ UNIX version and 3.2. If you don't want to use Duncan's
+ local readdir() function, edit the Makefile and remove
+ -DUSELOCALREADDIR from the CFGF string. When DEC releases
+ a corrected getdirentries() function, I'll modify the
+ Configure script to stop defining USELOCALREADDIR.
+
+7.2 Why does the Digital UNIX V3.2 ld complain about Ots* symbols?
+
+ When you compile lsof on your Digital UNIX V3.2 system, ld
+ may complain:
+
+ ld:
+ Unresolved:
+ knlist
+ _OtsRemainder32Unsigned
+ _OtsDivide64Unsigned
+ _OtsRemainder64Unsigned
+ _OtsDivide32Unsigned
+ _OtsMove
+ _OtsDivide32
+ _OtsRemainder32
+ *** Exit 1
+
+ Chris Eleveld reports this happens on Digital UNIX V3.2
+ systems after the Fortran compiler has been installed.
+
+ The best work-around seems to be to remove -lmld from the
+ CFGL string in the Makefile produced by Configure -- i.e.,
+ change:
+
+ CFGL= -lmld
+ to
+ CFGL=
+
+ According to the V3.2 man page for nlist(3), this shouldn't
+ work, but my testing shows that it does. Although I haven't
+ been able to test this second work-around, you might try
+ adding -lots to CFGL, rather than removing -lmld -- i.e.,
+ change:
+
+ CFGL= -lmld
+ to
+ CFGL= -lmld -lots
+
+ WARNING: my testing also shows that the V2.0 nlist(3) man
+ page means what it says when it calls for -lmld -- lsof
+ loaded without -mld under V2.0 can't locate the proc
+ (process) table address.
+
+ DON'T REMOVE -lmld FROM THE DIGITAL UNIX V2.0 MAKEFILE.
+
+ If you run into this problem, please let me know what
+ problem you encountered and how you solved it.
+
+7.3 Why can't lsof locate named pipes (FIFOs) under V3.2?
+
+ While lsof for V3.2 can report on named pipes (FIFOs), it
+ can't find them by name. That appears to happen because
+ of the way the V3.2 kernel lstat(2) function reports named
+ pipe device numbers.
+
+ The V3.2 kernel reports the device number as 0xfffffff,
+ while the kernel structures for named pipes that lsof
+ examines contain the device number of the file system on
+ which the named pipe resides.
+
+ Consequently, lsof can't match the device and inode number
+ pair it receives from applying lstat(2) to the named pipe
+ with any device and inode number pair it finds when scanning
+ kernel structures.
+
+ I don't have a work-around. You can, of course, ask for
+ full lsof output and use a post-processing filer (e.g.,
+ grep) to locate the named pipe of interest.
+
+ This problem doesn't exist under V2.0.
+
+7.4 Why does lsof use the wrong configuration header files?
+ For example, why can't the lsof compilation find cpus.h?
+
+ DEC OSF/1, Digital UNIX, and Tru64 UNIX configuration header
+ files describe the hardware and software environment for
+ which your kernel boot file was constructed. For example,
+ /sys/<name>/cpus.h defines the number of CPUs in its NCPUS
+ #define.
+
+ Lsof searches for the configuration header file subdirectory
+ in /sys (/usr/sys for Digital UNIX version 4.0 and Tru64
+ UNIX) by converting the first host name component to capital
+ letters -- e.g., TOMIS is derived from tomis.bio.purdue.edu.
+ If that subdirectory exists, lsof uses header files from
+ it. (Configure reports what subdirectory is being used.)
+
+ If Configure doesn't find a host-name derived subdirectory,
+ it prompts you for the entry of a subdirectory name. If
+ you can't find one, quit Configure and run the kernel
+ generation process to create a proper configuration sub-
+ directory. If you don't identify a proper configuration
+ subdirectory and you try to compile lsof, the compiler will
+ complain about missing header files -- e.g., a missing
+ cpus.h.
+
+ Once you have located or generated a proper configuration
+ subdirectory, rerun Configure. If you have generated a
+ configuration subdirectory whose name is derived from the
+ host name, Configure will find and use it. If not, you
+ will have to specify its name to Configure.
+
+7.5 Why does lsof indicate incomplete paths with " -- " for Tru64
+ UNIX 5.1 files?
+
+ When lsof can't find a component of a path in the kernel's
+ name cache (aka DNLC), or can't determine that the left-most
+ component has as its parent the file system root, it uses
+ an "incomplete path" notation. That notation begins with
+ the file system root name, followed by " -- ", followed by
+ the consecutive path name components lsof was able to find
+ in the DNLC -- e.g., "/ -- init".
+
+ Because the DNLC was significantly redesigned in Tru64 UNIX
+ 5.1, lsof's handling of the cache had to be completely
+ redone. As part of the DNLC redesign a name cache entry
+ parameter lsof formerly used to locate the file system root
+ of a path was removed. With help from Chang Song I've been
+ able to implement an alternate method for detecting the
+ root of these file system types: AdvFS (MSFS), CDFS, DVDFS,
+ FDFS, NFS, NFS3, and UFS.
+
+ When lsof doesn't know how to identify the root for a file
+ system type, it will resort to the " -- " incomplete path
+ notation.
+
+7.6 Why doesn't lsof report link count, node number, and size
+ for some Tru64 5.x CFS files?
+
+ Lsof reports link count, node number, and size for open
+ CFS files as recorded in their kernel node structure's
+ cached attributes. Sometimes not all attributes are cached
+ on the system where lsof runs, so lsof cannot report them.
+
+7.7 Why does lsof say it can't read the kernel name list or
+ proc table on Digital UNIX 4.x or Tru64 UNIX?
+
+ By default on Digital UNIX 4 and Tru64 UNIX lsof reads the
+ addresses for kernel symbols with the knlist(3) function.
+ That function can fail, for example, when the kloadsrv
+ daemon isn't running or is malfunctioning. When that
+ happens, lsof may abort with one of these error messages:
+
+ lsof: can't read kernel name list from knlist(3): ...
+ or
+ lsof: can't read proc table info
+
+ The first message suggests a complete knlist(3) or kloadsrv
+ failure; the second, a partial one.
+
+ If you know the name of the file from which the running
+ system was booted, e.g., /vmunix, you can use lsof's -k
+ option to direct it to read kernel symbol addresses from
+ the name list of that file --
+
+ $ lsof -k /vmunix ...
+
+ If that works, then knlist(3) is malfunctioning and you
+ need to fix it.
+
+
+8.0 FreeBSD Problems
+
+8.1 Why doesn't lsof report on open kernfs files?
+
+ Lsof doesn't report on open FreeBSD kernfs files because
+ the structures lsof needs aren't defined in the kernfs.h
+ header file in /sys/misc/kernfs.
+
+8.2 Why doesn't lsof work on my FreeBSD system?
+
+ If lsof doesn't work on your FreeBSD system, first make
+ sure you have the latest lsof revision. See the answer to
+ the "Where do I get lsof?" question for information on how
+ to get the latest lsof revision.
+
+ Once you have gotten the latest lsof revision, Configure
+ and make it. If Configure fails -- e.g., it complains
+ about an unknown FreeBSD version -- then lsof probably
+ hasn't been ported to your FreeBSD version yet, and there's
+ no need to go any further. Follow the answer to the "How
+ do I report an lsof bug" to report the Configure complaint
+ to me.
+
+ If you are able to Configure and make lsof, run its test
+ suite. (See the answer to the "Is there a test suite?"
+ question for more information on how to use lsof's test
+ suite.)
+
+ If lsof still fails, make sure your kernel sources, kernel
+ header files, kernel boot file, standard header files and
+ libraries are synchronized. They should all be built from
+ the same CVS refresh. If they aren't, then the KVM library
+ or lsof may be using kernel structure definitions that
+ don't match the booted kernel.
+
+ If you have synchronized your kernel, header files and
+ libraries, and still can't get lsof to work, follow the
+ steps in the answer to the "How do I report an lsof bug"
+ question to report the problem to me.
+
+8.3 Why doesn't lsof work on the RELEASE version of CURRENT?
+
+ Lsof tracks the CURRENT release of the current leading edge
+ FreeBSD version, because my access to leading edge FreeBSD is
+ limited to FreeBDSD.org reference systems, all running the
+ CURRENT release.
+
+ Sometimes that tracking leads to changes in lsof that won't
+ work on an earlier RELEASE version of the current leading edge
+ version.
+
+ When that happens, please send e-mail to me <abe@purdue.edu>.
+ Make sure "lsof" appears in the "Subject:" line so my e-mail
+ filter won't classify your letter as Spam.
+
+8.4 Why does kvm_open() complain it can't find some file?
+
+ If lsof issues this complaint:
+
+ lsof: kvm_open(execfile=/boot/kernel/kernel,
+ corefile=/dev/mem: No such file or directory
+
+ Your FreeBSD system might not have a /dev/mem device. If
+ not, create one -- e.g., as root do:
+
+ # mknod /dev/mem c <major> 0
+ # chmod 440 /dev/mem
+ # chgrp kmem /dev/mem
+
+ For <major> use /dev/kmem's major device number.
+
+ You may have to run kldload, too -- again as root do:
+
+ # kldload mem
+
+8.5 FreeBSD ZFS Problems
+
+8.5.1 Why does FreeBSD lsof report "WARNING: no ZFS support has been
+ defined."?
+
+ Lsof issues that message when it detects a file on a ZFS file
+ system, but has not been built with support for ZFS. Lsof's
+ Configure script detects support can be added for ZFS when it
+ finds this file:
+
+ /usr/src/sys/contrib/opensolaris/uts/common/fs/zfs/sys/zfs_znode.h
+
+ That header file and others in the OpenSolaris files in
+ /usr/src enable lsof to extract information about ZFS files
+ from the kernel structures associated with them.
+
+8.6 Why can't Configure create lsof_owner.h for FreeBSD 6 and above?
+
+ Lsof may report:
+
+ Creating ./lockf_owner.h from /usr/src/sys/kern/kern_lockf.c
+ FATAL ERROR: can't read /usr/src/sys/kern/kern_lockf.c
+ FATAL ERROR: ./lockf_owner.h creation failed (see 00FAQ)
+ or
+ Creating ./lockf_owner.h from /usr/src/sys/kern/kern_lockf.c
+ FATAL ERROR: ./lockf_owner.h creation failed (see 00FAQ)
+
+ Those messages mean that lsof's Configure script failed to
+ create a local header file, ./lockf_owner.h, needed to use the
+ new kernel file locking code of some versions of FreeBSD 6 and
+ above.
+
+ The changes that implement that new locking code alter the
+ lockf structure in <sys/lockf.h> and introduce a new structure,
+ lockf_entry, to that header file. When Configure detects the
+ presence of the lockf_entry definition in <sys/lockf.h>, it
+ tries to construct the local header file, ./lockf_owner.h.
+
+ Configure has to do that because an unfortunate side effect of
+ the new kernel file locking code is that <sys/lockf.h> doesn't
+ contain the lockf_owner structure definition referenced in its
+ own lockf structure. Lsof needs to access elements of that
+ lockf_owner structure to determine if a lock belongs to the
+ process that has a file open.
+
+ The missing lockf_owner structure definition is in the kernel
+ source file, typically /usr/src/sys/kern/kern_lockf.c.
+ Configure tries to extract the lockf_owner structure definition
+ from kern_lockf.c into lsof's local header file, ./lockf_owner.h.
+ If Configure can't do that, it reports:
+
+ FATAL ERROR: ./lockf_owner.h creation failed
+
+ If Configure can't even read kern_lockf.c, it first reports:
+
+ FATAL ERROR: can't read /usr/src/sys/kern/kern_lockf.c
+
+ The work-around for this problem is to update the FreeBSD
+ kernel /usr/src tree (e.g., do a CVSup or csup) on the system
+ where lsof is to be built and then do a "make buildworld"
+ followed by a "make installworld".
+
+8.6.1 Why are there lockf structure compiler errors for FreeBSD 6.0
+ and higher lsof?
+
+ If, when compiling lsof, the compiler complains with error
+ messages like:
+
+ dnode.c: In function 'get_lock_state':
+ dnode.c:113: error: 'struct lockf' has no member named 'lf_flags'
+ dnode.c:115: error: 'struct lockf' has no member named 'lf_id'
+ ...
+
+ Then lsof is being built on a system that has new kernel file
+ locking code and lsof's Configure script failed to build a
+ local lockf_owner.h header file with a structure definition
+ lsof needs.
+
+ See the "Why can't Configure create lsof_owner.h for FreeBSD 6
+ and above?" section for more information and a work-around.
+
+8.6.2 Why don't /usr/src/sys/sys/lockf.h and /usr/include/sys/lockf.h
+ match?
+
+ This mismatch can cause the errors explained in the answer to
+ the "Why are there lockf structure compiler errors for FreeBSD
+ 6.0 and higher lsof?" question.
+
+ If /usr/src/sys/sys/lockf.h has been updated with a CVSup or
+ csup, the new lockf.h won't be propagated to /usr/include/sys
+ until the "make buildworld" and "make installworld" steps have
+ been completed.
+
+
+9.0 HP-UX Problems
+
+9.1 What do /dev/kmem-based and PSTAT-based mean?
+
+ Lsof for HP-UX 11.0 and below uses /dev/kmem to read kernel
+ data structures from which it gathers and reports open file
+ information. That version of lsof is called /dev/kmem-based
+ lsof.
+
+ Starting with HP-UX 10.10, finding definitions for the
+ necessary kernel structures became more difficult as HP no
+ longer distributed header files in /usr/include that defined
+ all kernel structures. So I started "inventing" structure
+ definitions by using Q4 to display them.
+
+ By HP-UX 11, the process of invention became extremely
+ intensive to support. Following a patch to the ipc_s
+ structure in early 1999, my invented definition of that
+ structure became incorrect. Although I was able to devise
+ a work-around test for the patch with Q4, it was clear that
+ my inventions were bound to cause more problems.
+
+ Discussion with HP about the patch led to my proposing that
+ an lsof API in the HP-UX kernel was the proper solution.
+ Much to my surprise, HP agreed. I believe Carl Davidson
+ was the prime mover behind that decision, but I know others
+ participated, among them Louis Huemiller, Rich Rauenzahn,
+ and Sailu Yallapragada. I am indebted to these folks and
+ HP for their willingness to do this work.
+
+ The API was added to the PSTAT interface in a project named
+ PEGL, Pstat Enhancements for Glance and Lsof. Louis and
+ Sailu did the bulk of the design and implementation work
+ and testing began in March, 2000
+
+ HP-UX 11.11 is the first version that provides PSTAT support
+ for lsof. HP-UX versions in between 11.0 and 11.11 -- all
+ Beta versions as far as I can determine -- have no lsof
+ support.
+
+ See the "PSTAT-based HP-UX lsof Questions" section for
+ questions and answers specific to PSTAT-based HP-UX lsof.
+ The next section, "Why doesn't a /dev/kmem-based HP-UX lsof
+ compilation use -O?" covers /dev/kmem-based HP-UX lsof.
+
+ The /dev/kmem-based lsof sources may be found in the kmem
+ subdirectory of the dialects/hpux branch of the lsof source
+ tree. The PSTAT-based lsof sources may be found in
+ .../dialects/hpux/pstat.
+
+9.2 /dev/kmem-based HP-UX lsof Questions
+
+ The sources for /dev/kmem-based lsof for HP-UX may be found
+ in lsof_<revision>/dialects/hpux/kmem.
+
+ Lsof's Configure shell script decides to use these sources
+ when it finds that the /usr/include/sys/pstat subdirectory
+ doesn't exist.
+
+ Lsof can be forced to use the /dev/kmem sources by setting
+ "/dev/kmem" in the HPUX_BASE environment variable. Consult
+ the Configure shell script and 00XPORTING for more information.
+
+9.2.1 Why doesn't a /dev/kmem-based HP-UX lsof compilation use -O?
+
+ If you only have the standard (bundled) HP-UX C compiler
+ and haven't purchased and installed the optional one, then
+ you can't use cc's -O option. The HP-UX cc(1) man page
+ says this:
+
+ "Options
+ Note that in the following list, the cc and c89 options
+ -A , -G , -g , -O , -p , -v , -y , +z , and +Z are
+ not supported by the C compiler provided as part of
+ the standard HP-UX operating system. They are supported
+ by the C compiler sold as an optional separate product."
+
+ Lsof's Configure script tries to detect what C compiler
+ product you have installed by examining your compiler. If
+ that examination reveals a standard (bundled) compiler,
+ lsof avoids using -O.
+
+ If the Configure compiler test fails, the C compiler will
+ complain that it doesn't support -O. You can suppress that
+ complaint with this make invocation:
+
+ $ make DEBUG=""
+
+9.2.2 Why doesn't the /dev/kmem-based CCITT support work under 10.x?
+
+ Pasi Kaara, who originally provided the HP-UX CCITT support,
+ reports that it no longer works under HP-UX 10.x.
+ Consequently, at lsof revision 4.02 it has been disabled.
+
+9.2.3 Why can't /dev/kmem-based lsof be compiled with `cc -Aa` or
+ `gcc -ansi` under HP-UX 10.x?
+
+ Some HP-UX 10.x header files, needed by lsof, can't be
+ compiled properly in ANSI_C mode; structure element definition
+ and alignment problems result. The f_offset member of the
+ file structure, for example, is incorrect.
+
+ This ANSI-C obstacle extends to using the -Aa option of
+ the HP C compiler and the -ansi option of gcc.
+
+9.2.4 Why does /dev/kmem-based lsof complain about no C compiler?
+
+ Lsof's Configure script looks in /bin and /usr/ccs/bin for
+ an HP C compiler, because it needs to know if the compiler
+ is the standard (bundled) one or the optional separate
+ product. If it finds no compiler in either place, Configure
+ quits after complaining:
+
+ No executable cc in /bin or /usr/ccs/bin
+
+ If you don't have a C compiler in either of these standard
+ places, you should consider installing it. If you have
+ gcc installed, you can use it by declaring the ``hpuxgcc''
+ abbreviation to lsof's Configure script.
+
+ If you have a C compiler in a non-standard location, you
+ can use the HPUX_CCDIR[12] environment variables to name
+ the path to it. Consult the 00XCONFIG file of the lsof
+ distribution for more information.
+
+9.2.5 Why does Configure complain about q4 for /dev/kmem-based lsof
+ for HP-UX 11?
+
+ When you run Configure on an HP-UX 11 system, it may complain:
+
+ !!!ERROR!!! !!!ERROR!!! !!!ERROR!!! !!!ERROR!!!
+ Configure can't use /usr/contrib/bin/q4 to examine the ipis_s
+ structure. You must do that yourself, report the result in
+ the HPUX_IPC_S_PATCH environment variable, then repeat the
+ Configure step. Consult the Configure script's use of
+ /usr/contrib/bin/q4 and the 00XCONFIG file for information
+ on ipis_s testing and the setting of HPUX_IPC_S_PATCH.
+ !!!ERROR!!! !!!ERROR!!! !!!ERROR!!! !!!ERROR!!!
+
+ This message states that Configure cannot use q4 from
+ /usr/contrib/bin to examine the kernel's boot image for
+ the ipis_s structure. Maybe q4 hasn't been installed, or
+ perhaps Configure can't execute it.
+
+ Lsof needs to gather information about ipis_s to determine
+ if the ipis_s structure is defined in the kernel boot image,
+ if the ipis_s structure of the kernel boot image has an
+ ipis_msgsqueued member, and if the ipc_s structure of the
+ kernel boot image uses has an ipc_ipis member.
+
+ The ipis_s structure isn't described in any header file
+ HP-UX releases with HP-UX 11. It appears in the private
+ lsof header file .../dialects/hpux/kmem/hpux11/ipc_s.h.
+ Lsof gets local and remote connection addresses (IP and
+ port numbers) from ipc_s, so an incorrect ipc_s definition
+ may cause incorrect reporting of TCP/IP connection addresses.
+ It definitely will cause incorrect reporting on 32 bit
+ kernels. In any case lsof should be compiled with a correct
+ ipc_s definition no matter the kernel bit size, so the
+ Configure script always tests for it when the HP-UX version
+ is 11.
+
+ For lsof's Configure script to gather the necessary ipis_s
+ information q4 needs to be installed in /usr/contrib/bin
+ and the kernel boot image, /stand/vmunix, needs to have
+ been processed with pxdb. If either is untrue, lsof issues
+ the above error message, perhaps preceded by q4 messages.
+ (Note: lsof's use of q4 may also fail if q4 can't execute
+ nm -- e.g., it can't find /usr/bin/nm, or there is a
+ conflicting, private version of nm earlier in the path.)
+
+ If /stand/vmunix hasn't been processed by pxdb, the q4
+ messages will include:
+
+ q4: (error) vmunix not pxdb'd
+ or
+ q4: (warning) /stand/vmunix has not been processed by pxdb.
+
+ It's possible to make a suitable private copy of /stand/vmunix
+ for configuring lsof. That requires /opt/langtools/bin/pxdb
+ or the q4 version of pxdb from /usr/contrib/bin/q4pxdb.
+ The path to the result is supplied to the lsof Configure
+ script in the HPUX_BOOTFILE environment variable. Configure
+ still requires /usr/contrib/bin/q4.
+
+ The following sample Bourne shell commands make a private
+ copy of /stand/vmunix in /tmp, process it with pxdb or
+ q4pxdb, and supply its path to lsof's Configure script in
+ HPUX_BOOTFILE.
+
+ $ cp /stand/vmunix /tmp/vmunix.lsof
+
+ $ /opt/langtools/bin/pxdb /tmp/vmunix.lsof
+ or
+ $ /usr/contrib/bin/q4pxdb /tmp/vmunix.lsof
+
+ ... pxdb messages ...
+ $ HPUX_BOOTFILE=/tmp/vmunix.lsof Configure -n hpux
+
+ It may also be necessary to use q4 outside the lsof Configure
+ script. In that case q4 can be to determine the state of
+ ipis_s and ipc_s with these q4 commands:
+
+ $ /usr/contrib/bin/q4 /stand/vmunix
+ ...
+ q4> fields -c struct ipc_s
+ ...
+ q4> fields -c struct ipis_s
+
+ Look in the q4 output for the ipc_ipis member of the ipc_s
+ structure, and look in the q4 output for the ipis_s structure
+ for the ipis_msgsqueued member. If ipc_s has ipc_ipis but
+ ipis_s lacks ipis_msgsqueued, set HPUX_IPC_S_PATCH environment
+ variable to "1". If ipc_s has ipc_ipis and ipis_s has
+ ipis_msgsqueued, set HPUX_IPC_S_PATCH to "2" -- e.g.,
+
+ $ HPUX_IPC_S_PATCH=1 Configure -n hpux
+ or
+ $ HPUX_IPC_S_PATCH=2 Configure -n hpux
+
+ If ipc_s has no ipc_ipis member, set HPUX_IPC_S_PATCH to
+ "N" -- e.g., use this Configure step:
+
+ $ HPUX_IPC_S_PATCH=N Configure -n hpux
+
+9.2.6 When compiling /dev/kmem-based lsof for HP-UX 11 what do the
+ "aCC runtime: ERROR..." messages mean?
+
+ When the lsof Makefile asks the HP-UX unbundled compiler
+ to load lsof, it may complain:
+
+ /bin/cc -o lsof -DHPUXV=1100 -DHASVXFS -DHPUXKERNBITS=64 \
+ -I/home/abe/src/lsof4/dialects/hpux/kmem/hpux11 +DD64 \
+ -DHAS_IPC_S_PATCH=2 -I/home/abe/src/lsof4/dialects/hpux/kmem \
+ -DLSOF_VSTR=\"B.11.00\" -g dfile.o dmnt.o dnode.o dnode1.o \
+ dnode2.o dproc.o dsock.o dstore.o arg.o main.o misc.o \
+ node.o print.o proc.o store.o usage.o -L./lib -llsof -lelf \
+ -lnsl
+ aCC runtime: ERROR: Unexpected use of shared libraries
+ aCC runtime: ERROR: Read aCC manpage, +A option
+ /usr/lib/nls/loc/locales.1//is_IS.iso88591
+
+ This is a bug in the HP-UX national language support.
+ (Notice the last message with "locales" in it?) Complain
+ to HP -- then use this work-around before executing make:
+
+ $ unset LANG
+ $ make
+
+9.2.7 Why doesn't /dev/kmem-based lsof for HP-UX 11 report VxFS file
+ link counts, node numbers, and sizes correctly?
+
+ This is usually the result of running an lsof binary whose
+ revision number is less than 4.57 on a system that has
+ OnlineJFS support installed. It can also happen with lsof
+ 4.57 binaries when the OnlineJFS support with which they
+ were built doesn't match the OnlineJFS status of the system
+ on which they are run.
+
+ The OnlineJFS status of lsof 4.57 and higher binaries can
+ be determined by running:
+
+ $ lsof -v 2>&1 | grep HASONLINEJFS
+
+ If that shell pipe produces output, lsof was compiled with
+ OnlineJFS support enabled; no output, disabled.
+
+ If OnlineJFS is installed on an HP-UX 11 system the
+ /sbin/fs/vxfs/subtype executable exists and outputs "vxfs3.3"
+ when run.
+
+ The problem occurs because the optional OnlineJFS support
+ installation doesn't update <sys/fs/vx_inode.h>. Consequently
+ lsof can be compiled with an incorrect definition of the
+ vx_inode structure and look for for link counts, node
+ numbers, and sizes in the wrong places in the structure.
+
+ The current response I have gotten from HP is that no
+ <sys/fs/vx_inode.h> update will be provided for OnlineJFS.
+
+ I've addressed this problem temporarily with a work-around
+ (hack) in lsof revision 4.57.
+
+9.2.8 Why can't /dev/kmem-based lsof be built with gcc for 64 bit
+ HP-UX 11?
+
+ When Configure is given the "hpuxgcc" abbreviation, the
+ HP-UX version is 11, and the kernel bit size is 64, the
+ lsof Configure script may abort with the messages:
+
+ !!!!!!!!!!!!!!!!! FATAL ERROR !!!!!!!!!!!!!!!!!!
+
+ APPARENTLY GCC CANNOT BUILD 64 BIT EXECUTABLES.
+ A COMPILER MUST BE USED THAT CAN. SEE 00FAQ
+ FOR MORE INFORMATION.
+
+ (This is the "more information" in 00FAQ.)
+
+ This means the Configure script compiled a test program
+ with gcc the result wasn't an ELF-64 binary. Lsof tries
+ two gcc modes, one with no options and another with the
+ -mlp64 option, before it concludes gcc can't be used.
+
+ See the "How can I acquire a gcc for building lsof for 64
+ bit HP-UX 11?" answer for information on where you might
+ be able to get a gcc for HP-UX 11 that can produce ELF-64
+ executables.
+
+9.2.8.1 How can I acquire a gcc for building lsof for 64 bit HP-UX 11?
+
+ Check this HP URL:
+
+ http://h21007.www2.hp.com/dspp/tech/tech_TechSoftwareDetailPage_IDX/1,1703,547,00.html
+
+ (That's one very long link; be careful you cut 'n paste it
+ all.)
+
+ In November 2001 that URL led to a web page whose title
+ was "gcc for hp-ux 11." The page offered a link for
+ downloading a 64 bit gcc 3.0 compiler for HP-UX 11.0 and
+ 11i. Rich Rauenzahn of HP installed that compiler on an
+ HP test system he allows me to use and I successfully built
+ a 64 bit lsof with it.
+
+ The HP package may install the 64 bit capable gcc in
+ /usr/local/pa20_64/bin/gcc, so you may have to adjust your
+ path or set the LSOF_CC environment variable to compensate.
+
+9.2.9 Why does /dev/kmem-based lsof for HP-UX 11 report "unknown file
+ system type" for some open files?
+
+ The lsof binary being used probably doesn't have support for
+ the VxFS file system.
+
+ To confirm that, check `lsof -v` output for "-DHASVXFS". If
+ it's not present, lsof doesn't have VxFS support.
+
+ You also need to establish that lsof really is complaining
+ about VxFS files by checking the kernel boot file for the
+ symbol associated with the hexadecimal address reported in the
+ "unknown file system type" message -- e.g., "v_op: 0x8711c8."
+ Use nm(1) to do that:
+
+ $ nm -x /stand/vmunix | grep 8711c8
+
+ If nm reports the symbol associated with the address is
+ vx_vnodeops, then lsof is complaining about an open VxFS file.
+
+ The solution in that case is to build lsof yourself (The
+ bundled C compiler will do it.), making sure that lsof's
+ Configure script detects the presence of VxFS. Configure does
+ that by finding these two header files:
+
+ /usr/include/sys/fs/vx_hpux.h
+ /usr/include/sys/fs/vx_inode.h
+
+ If the system where you are building lsof doesn't have those
+ header files, but does have VxFS, you might be able to install
+ the header files by installing the HP JournalFS package from
+ the CoreOS CD -- in particular the file set JournalFS.VXFS-PRG
+ and its associated patch, PHKL_18543. (My thanks to Steve
+ Bonds for that information.)
+
+ Finally, if you find that lsof isn't complaining about VxFS
+ when it complains about an unknown file system type, send
+ e-mail to me <abe@purdue.edu> for further assistance. Make
+ sure "lsof" appears in the "Subject:" line so my e-mail filter
+ won't classify your letter as Spam.
+
+9.2.10 Why does the ANSI-C compiler complain about comments in HP-UX
+ 11 header files?
+
+ When compiling lsof on HP-UX 11, the HP ANSI-C compiler's
+ pre-processor, cpp, may complain about comments in HP-UX header
+ files -- e.g.,
+
+ cpp: "/usr/include/sys/cdfs.h", line 232: warning 2028:
+ Found comment inside comment started on line 232.
+ cpp: "/usr/include/sys/cdnode.h", line 196: warning 2028:
+ Found comment inside comment started on line 196.
+ cpp: "/usr/include/nfs/snode.h", line 30: warning 2028:
+ Found comment inside comment started on line 30
+
+ This is not a problem with lsof. It is a problem with the
+ HP-UX header files; they have non-compliant ANSI-C comment
+ sequences in them -- e.g.,
+
+ <sys/cdfs.h>: 232
+ /* struct cdfs *cdfs_link; /* linked list of file systems */
+
+ The initial "/*" is not terminated by an ending "*/" before the
+ appearance of a second "/*".
+
+9.2.11 Why does dnode1.c cause the HP-UX 11 compiler to complain that
+ <sys/fs/vx_inode.h> is missing or incorrect?
+
+ If CFLAGS in the lsof Makefile for an HP-UX 11 compilation
+ includes HASONLINEJFS, indicating the system has OnlineJFS
+ support, lsof needs the <sys/fs/vx_inode.h> header file.
+ Sometimes it is missing from /usr/include/sys/fs.
+
+ <sys/fs/vx_inode.h> is a header file that must be obtained from
+ Veritas. If that proves impossible, please contact me via
+ e-mail at <abe@purdue.edu>. Make sure "lsof" appears in the
+ "Subject:" line so my e-mail filter won't classify your letter
+ as Spam.
+
+
+9.3 PSTAT-based HP-UX lsof Questions
+
+ The sources for PSTAT-based lsof for HP-UX may be found in
+ lsof_<revision>/dialects/hpux/pstat.
+
+ Lsof's Configure shell script decides to use these sources
+ when it finds that the /usr/include/sys/pstat subdirectory
+ exists.
+
+ Lsof can be forced to use the PSTAT-based sources by setting
+ "pstat" in the HPUX_BASE environment variable. Consult
+ the Configure shell script and 00XPORTING for more information.
+
+9.3.1 Why does PSTAT-based lsof complain about pst_static and
+ other PSTAT structures?
+
+ When lsof starts it may issue one of these fatal error
+ messages:
+
+ lsof: FATAL: can't determine PSTAT static size
+ lsof: FATAL: can't read <n> bytes of pst_static
+ lsof: FATAL: pst_static doesn't contain <name>_size
+ lsof: FATAL: <name>_size should be <n>
+
+ These messages indicate that lsof's tests for the proper
+ level of PSTAT support have failed. The structure names,
+ given in <name>, and sizes, given in <n>, identify the
+ support deficiency more precisely.
+
+ You may need to upgrade the PSTAT support in your kernel
+ to be able to use PSTAT-based lsof.
+
+9.3.2 Why does PSTAT-based lsof complain it can't read pst_*
+ structures?
+
+ Lsof may put messages like the following in the NAME
+ column of its output.
+
+ can't read cwd pst_filedetails: Permission denied
+ can't read mem pst_filedetails: Permission denied
+ can't read rtd pst_filedetails: Permission denied
+ can't read txt pst_filedetails: Permission denied
+ can't read pst_filedetails: Permission denied
+ can't read 3 stream structures: Permission denied
+ can't read pst_socket: Permission denied
+
+ These messages indicate that the lsof binary lacks the
+ authority to read the name structures for processes other
+ than ones belonging to the UID under which lsof is running.
+ Authority to read the structures of other processes is
+ limited to root processes -- i.e., lsof must have setuid-root
+ permission if it is to list open files for arbitrary
+ processes.
+
+ If you want to eliminate these errors, you must run lsof
+ as root or install it with setuid-root permission.
+
+9.3.3 Why does PSTAT-based lsof rebuild the device cache file
+ after each reboot?
+
+ After each HP-UX rebuild, the first time a user runs lsof it
+ will report:
+
+ lsof: WARNING: device cache mismatch: /dev/tun...
+ lsof: WARNING: created device cache file: /<user_path>
+
+ This happens because the device numbers on /dev/tun* device
+ nodes are recalculated at each reboot. When lsof detects
+ a change in the device number of a /dev/tun* file, it rebuilds
+ its local device cache file.
+
+9.3.4 Why doesn't PSTAT-based lsof report TCP addresses for
+ telnetd's open socket files?
+
+ When lsof can't report TCP addresses for telnetd's open
+ socket files it is because an unpatched PSTAT kernel
+ interface doesn't report the addresses to lsof.
+
+ This has been addressed in PSTAT kernel patch PHKL_24047.
+ It is available from the HP IT Resource Center at:
+
+ http://itrc.hp.com
+
+ In the page's "maintenance / support" box select the
+ "individual patches" link. Once at its page, select the
+ "hp-ux" link. On that page select the "Series 800" or
+ "Series 700" radio button and select "11.11" from the
+ pull-down list to the right of the button. Under "search
+ or browse the path list" select "Search by Patch IDs" from
+ the pull down list, enter PHKL_24047 in the following text
+ box, and select search. That should lead to information
+ about PHKL_24047 and a link for downloading it. (You may
+ have to log in first and you may have to create a login
+ identity by registering before you can log in.)
+
+ Some time in March 2006 the PHKL_24047 patch was "lost"
+ by the HP-UX networking lab. It has been "found" again
+ in August 2006 and will be re-released as a GRO patch
+ "some time." I don't yet know when that will be. You
+ must contact HP to learn about the availability of the
+ GRO patch.
+
+9.3.5 Why does PSTAT-based lsof cause an HP-UX 11.11 kernel panic?
+
+ When PSTAT-based lsof runs on some HP-UX 11.11 kernels,
+ the kernel may panic. Symptoms include:
+
+ Console message:
+ 0xFBE000301100EF00 00000000 0000EF00 -
+ type 31 = legacy PA HEX chassis-code
+
+ /var/adm/syslog:
+ ... vmunix: Trap Type 15 (Data page fault)
+ ... vmunix: Instruction Address (pcsq.pcoq) = 0x...
+
+ The panic is caused by a bug in the way PSTAT's pstat_getstream()
+ function obtains module names from streams managed by the
+ otsam stream driver (part of OSI Transport Services). Lsof
+ calls pstat_getstream() when it encounters an open otsam
+ stream file. An HP-UX 11.11 system uses otsam if otsam
+ appears in /stand/system.
+
+ HP-UX 11.11 patch PHKL_24507 (available some time after
+ July 15, 2001) fixes the pstat_getstream() bug. See the
+ information in the answer to the "Why doesn't PSTAT-based
+ lsof report TCP addresses for telnetd's open socket files?"
+ question for information on how to obtain the patch.
+
+9.3.6 Why doesn't PSTAT-based lsof report a CWD that is on a loopback
+ (LOFS) file system?
+
+ When PSTAT-based lsof reports on processes whose current
+ working directory (CWD) is on a loopback file system, lsof
+ can't report the open CWD file. The reason is that the HP-UX
+ 11.11 and above kernel's loopback file system code is not
+ passing the CWD file ID to the kernel's pstat(2) code. Hence
+ lsof is given no information on the lofs CWD.
+
+ The problem was first reported to me by Ermin Borovac and an
+ internal bug report was filed with the HP-UX file system group
+ on October 26, 2004. That report has now been answered by the
+ patch PHKL_33200 -- s700_800 11.11 lofs cumulative patch. The
+ HP IT Resource Center (http://itrc.hp.com) is a source for the
+ patch.
+
+9.3.7 Why do some swinstall packages for PSTAT-based HP-UX 11.11
+ packages complain about setgid and setuid bits?
+
+ First, let me explain that I do not provide lsof swinstall
+ packages for lsof. Others provide them and they should be
+ contacted about problems with their packages.
+
+ However, I have become aware of a problem with one package
+ about which I have some information I can share. The problem
+ shows up in these swinstall messages:
+
+ ERROR: Unknown owner and/or group for file
+ "/usr/local/bin/lsof". SUID and/or SGID bit was
+ not set.
+ ERROR: Failed installing fileset "lsof.lsof-RUN,r=4.73".
+ Check the above output for details.
+
+ The swpackage SUID/SGID functionality was restricted by changes
+ for POSIX compliance, breaking backward compatibility. The
+ patch PHCO_27671 allows SUID/SGID for uid/gid of 0 only, as a
+ compromise between backward compatibility and POSIX conformance.
+
+ If the setuid bit is to be set on the executable, the UID and
+ GID of the executable must be 0 (zero).
+
+9.3.8 Why won't the bundled C compiler build PSTAT-based lsof for
+ PA-RISC HP-UX 11.23?
+
+ A PA-RISC HP-UX 11.23 bundled C compiler dated May 2005 or
+ later will not build PSTAT-based lsof. It will deliver error
+ messages related to the system's <gssapi/gssapi.h> header
+ file.
+
+ There is nothing wrong with that header file or lsof. The
+ problem is that the bundled C compiler can't cope with the
+ gssapi.h header file.
+
+ The work-around is to use the HP ANSI C compiler. Using gcc
+ is not a satisfactory work-around. See the answer to the "Why
+ won't gcc build PSTAT-based lsof for PA-RISC HP-UX 11.23?"
+ question for more information.
+
+9.3.9 Why won't gcc build PSTAT-based lsof for PA-RISC HP-UX 11.23?
+
+ Gcc will not even compile PSTAT-based lsof revisions below 4.77
+ for PA-RISC HP-UX 11.23 dated May 2005 or later. It reports
+ errors in lsof's print.c fill_portmap() function about missing
+ members of the rpcent structure. That happens because gcc
+ defines _XOPEN_SOURCE_EXTENDED which disables the definition of
+ the rpcent structure in <netdb.h>.
+
+ Using the HP bundled C compiler is not a viable work-around.
+ That is explained in the answer to the "Why won't the bundled C
+ compiler build PSTAT-based lsof for PA-RISC HP-UX 11.23?"
+
+ While an lsof revision 4.77 or higher can be compiled with gcc,
+ the results are unreliable. Lsof will compile, but it
+ occasionally produces segment faults when it runs. I have not
+ been able to reproduce the failure reliably or locate a
+ debugger that will work with the gcc-compiled lsof.
+
+ The only reliable work-around is to use the HP ANSI C
+ compiler.
+
+9.3.10 Why does PSTAT-based lsof complain, "FATAL: pst_stream_size
+ should be: 672; is 72" on HP-UX 11.11 and above?
+
+ This message indicates a mismatch between the PSTAT header
+ files used to build lsof (<sys/pstat.h> and those in the
+ /usr/include/sys/pstat subdirectory), and those that built the
+ running kernel.
+
+ Unfortunately the June 2008 patch set for HP-UX 11.23 creates
+ this inconsistency, because it does not contain all the patches
+ needed to match the kernel with the PSTAT header files. Even
+ more serious is that the missing patches update the kernel's
+ PSTAT support to provide TCP/UDP endpoint information to lsof
+ from TCP/TLI streams.
+
+ The patch inconsistency comes about because, while the following
+ patch is installed,
+
+ PHKL_36577 1.0 PM-PSTAT section 2 manpage changes
+
+ other kernel patches are not.
+
+ The PHKL_36577 patch updates the PSTAT header files and manual
+ pages to match kernel changes that other patches with the
+ following numbers (or patches that contain or supersede them)
+ contain:
+
+ PHNE_36575 1.0 Cumulative STREAMS Patch
+ PHNE_37670 1.0 cumulative ARPA Transport patch
+ PHNE_37851 1.0 NFS cumulative patch
+
+ Those patches implement the kernel changes that support the
+ delivery of information promised in patch PHKL_36577.
+
+ The work-around is to install the missing patches.
+
+9.4 Why won't the HP-UX depot install?
+
+ I don't distribute lsof depts, so I can't support them.
+
+ From time to time depots prepared by various sites -- e.g.,
+ usually HP-UX software collection sites -- will contain errors
+ that cause installation of the depot to fail.
+
+ Do not contact me when this happens. Instead, contact the
+ administrator of the site that prepared the depot.
+
+ As should be clear from the bulk of the lsof documentation, I
+ do not recommend you use pre-built lsof binaries in any form.
+ Instead, I recommend you obtain the lsof source distribution
+ and build lsof yourself.
+
+
+10.0 Linux
+
+10.1 What do /dev/kmem-based and /proc-based lsof mean?
+
+ At approximately Linux 2.1.72 and exactly at lsof revision
+ 4.23 support for Linux forks. The first fork, containing
+ the oldest lsof form is based on access to kernel memory
+ structures, and is called /dev/kmem-based lsof. A
+ /dev/kmem-based lsof is heavily intertwined with the Linux
+ kernel version, its header files, and its system map file.
+ Typically a /dev/kmem-based lsof needs only setgid permission
+ to local all open file information.
+
+ After approximately Linux 2.1.72 and at revision 4.23 lsof
+ obtains all its information from the /proc file system.
+ That lsof is called the /proc-based lsof. A /proc-based
+ lsof does not read kernel memory, needs neither kernel
+ header files nor the system map file, and is less likely
+ to be affected by Linux kernel changes. However, it does
+ require setuid-root permission to list all open files, and
+ it can't report file offsets (positions).
+
+ After revision 4.52 the /dev/kmem-based Linux sources for
+ lsof are no longer distributed. Information about them
+ may be found in the 00INDEX and README files at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/OLD/src
+
+10.2 /proc-based Linux lsof Questions
+
+10.2.1 Why doesn't /proc-based lsof report file offsets (positions)?
+
+ /proc-based lsof revisions 4.79 and above can only report file
+ offsets (positions) for the files of Linux kernels 2.6.22 and
+ above.
+
+ During its initialization /proc-based lsof tests to see if
+ offset information can be obtained. If it cannot, lsof
+ disables offset reporting. If the -o option was selected, lsof
+ also issues this warning:
+
+ lsof: WARNING: can't report offset; disregarding -o.
+
+
+10.2.2 Why does /proc-based lsof report "can't identify protocol" for
+ some socket files?
+
+ /proc-based lsof may report:
+
+ COMMAND PID ... TYPE ... NODE NAME
+ pump 226 ... sock ... 309 can't identify protocol
+
+ This means that it can't identify the protocol (i.e., the
+ AF_* designation) being used by the open socket file. Lsof
+ identifies protocols by matching the node number associated
+ with the /proc/<PID>/fd entry to the node numbers found in
+ selected files of the /proc/net sub-directory. Currently
+ /proc-based lsof examines these protocol files:
+
+ /proc/net/ax25 (untested)
+ /proc/net/ipx (needs kernel patch)
+ /proc/net/raw
+ /proc/net/raw6
+ /proc/net/tcp
+ /proc/net/tcp6
+ /proc/net/udp
+ /proc/net/udp6
+ /proc/net/unix
+
+ If /proc-based lsof says it can't identify the protocol
+ for an open socket file, you may be able to identify the
+ protocol yourself by using grep to look for the specific
+ node number in the files of /proc/net -- e.g.,
+
+ $ grep <node_number> /proc/net/*
+
+ You may not be able to find the desired node number, because
+ not all kernel protocol modules fully support /proc/net
+ information.
+
+ If you find a matching node number in a /proc/net file that is
+ not currently being processed by lsof, contact me via e-mail at
+ <abe@purdue.edu>. I'll discuss adding support to /proc-based
+ lsof for the protocol of the /proc/net file with you. Make
+ sure "lsof" appears in the "Subject:" line so my e-mail filter
+ won't classify your letter as Spam.
+
+ The code that matches node numbers of open IPX protocol
+ socket files to those in /proc/net/ipx requires Jonathan
+ Sergent's Linux 2.1.79 patch to /usr/src/linux/net/ipx/af_ipx.c.
+ The patch, suitable for input to Larry Wall's patch program,
+ may be found in the lsof distribution file:
+
+ .../dialects/linux/proc/patches/net_ipx_af_ipx.c.patch
+
+10.2.3 Why does /proc-based lsof warn about unsupported formats?
+
+ Lsof may issue the following warning:
+
+ lsof: WARNING: unsupported format: /proc/net/<file>
+
+ if the header line of the indicated <file> in /proc/net --
+ ax25, ipx, raw, tcp, udp, or unix -- doesn't match what
+ lsof expects to find.
+
+ When the header line of a /proc/net file isn't what lsof
+ expects, lsof probably can't parse the rest of the file
+ correctly and doesn't try. As a result, lsof can't report
+ any NAME column information (e.g., local and remote addresses)
+ for socket files bound to the indicated network protocol.
+
+ If you get this warning, please send me e-mail at <abe@purdue.edu>.
+ Include the contents of the file lsof claims has an unsupported
+ format. Make sure "lsof" appears in the "Subject:" line so my
+ e-mail filter won't classify your letter as Spam.
+
+10.2.4 Why does /proc-based lsof report "(deleted)" after a path name?
+
+ The "(deleted)" notation following a path name in /proc-based
+ lsof's NAME column comes from the /proc/<PID>/fd/<FD> entry
+ for the open file. It's the Linux kernel's way of indicating
+ the file is open but has been unlinked (rm'd).
+
+10.2.5 Why doesn't /proc-based lsof report full open file information
+ for all processes?
+
+ /proc-based lsof can only report on processes whose /proc
+ files it has permission to read. /proc normally grants
+ permission to read all its files only to root or to the
+ owning user ID.
+
+ Without permission to read most /proc files, lsof can only
+ report full information for processes belonging to the user
+ who is running lsof. /proc-based lsof may be able to report
+ some information for all processes, depending on the
+ permissions of their associated /proc files, but usually
+ /proc-based lsof won't be able to access the files in
+ /proc/<PID>/fd/ that describe regular open files.
+
+ If you want /proc-based lsof to report on all processes, you
+ must install it with setuid-root permission.
+
+10.2.6 Why won't Customize offer to change HASDCACHE or WARNDEVACCESS
+ for /proc-based lsof?
+
+ /proc-based lsof doesn't read device information from /dev
+ or the device cache file, so it makes no sense to change
+ the state of device cache processing or /dev node accessibility
+ warnings.
+
+10.2.7 /proc-based lsof Linux NFS questions
+
+10.2.7.1 Why can't lsof find files on an accessible NFS file system?
+
+ On occasion lsof may be unable to identify that an open
+ file is on an NFS file system. This is most likely the
+ result of a bug in the way the Linux kernel supplies
+ information to the reader of /proc/mounts (lsof) -- sometimes
+ that pseudo-file is truncated by the kernel.
+
+ One way to see if this is the case is to search for the
+ NFS file system in /proc/mounts -- e.g.,
+
+ $ grep <NFS_file_system_mount_point> /proc/mounts
+
+ If you get no output or the third word of the output isn't
+ "nfs", then lsof won't consider the file system an NFS file
+ system.
+
+ A second test is to look at the end of /proc/mounts --
+ e.g.,
+
+ $ tail /proc/mounts
+
+ If tail reports "# truncated" then /proc/mounts is incomplete
+ because of a Linux kernel bug. The bug is documented at:
+
+ http://www.xss.co.at/sysinfo/mounts.html
+
+ The bug is fixed in Linux kernel 2.4.18, and possibly in
+ some earlier Linux kernel versions.
+
+10.2.7.2 Why can't lsof find files on an inaccessible NFS file system?
+
+ If lsof issues this message about a Linux file system,
+ mounted from an NFS server:
+
+ lsof: WARNING: can't stat() nfs file system /xxx/yyy
+
+ Then lsof won't be able to find any open files on the file
+ system.
+
+ That's because of an inadequacy in the Linux /proc file
+ system. Its /proc/mounts file doesn't give the device
+ doublet (major and minor numbers) of the file system as do
+ many UNIX systems (e.g., Solaris). The only way lsof can
+ get the device doublet for a Linux file system is to call
+ stat(2) on the file system path, which fails if the NFS
+ server isn't accessible.
+
+ When lsof doesn't know the device doublet of a file system,
+ it can't find open files on the inaccessible file system,
+ because it can't match the doublets of open files to the
+ doublet of the inaccessible file system.
+
+ This topic is covered extensively in lsof(8) it its ALTERNATE
+ DEVICE NUMBERS and BLOCKS AND TIMEOUTS sections.
+
+10.2.8 Why doesn't /proc-based Linux lsof report socket options and
+ values, socket state flags, and TCP options and values?
+
+ The Linux /proc file system doesn't report socket options
+ and values, socket states, and TCP options and values to
+ lsof.
+
+10.2.9 Does /proc-based Linux lsof use a device cache?
+
+ No. The Linux /proc/<PID>/fd/* entries provide device names to
+ lsof via readlink(2). It is not necessary to enable device
+ cache processing for /proc-based Linux lsof via the Customize
+ script or modifications to the Linux machine.h header file.
+
+10.2.10 Why doesn't /proc-based Linux lsof report any or all file structure
+ values for its +fcfgGn option?
+
+ /proc-based lsof revisions 4.79 and above can only report some
+ file structure values for Linux kernels below 2.6.22.
+
+ When running on Linux kernels at 2.6.22 and above lsof 4.79 can
+ report some file flag values -- i.e., in response to the +fg or
+ +fG options. The flag values are obtained from the
+ /proc/<PID>/fdinfo/ files introduced at Linux kernel 2.6.22.
+
+ /proc-based Linux lsof tests its availability to obtain file
+ flag values at initialization. If values are not available,
+ lsof disables file flag reporting. If the flags were requested
+ with +fg or +fG, lsof displays this warning:
+
+ lsof: WARNING: can't report file flags; disregarding +f.
+
+ As a special note, when Linux lsof can report flag bits, it
+ will not report 'R' for a read-only file. There is no
+ read-only flag bit O_* symbol in <fcntl.h> (or <bits/fcntl.h>)
+ and lsof reports only bits that are set. The absence of O_RDWR
+ and O_WRONLY flag bits implies the file is read-only.
+
+10.3 Special Linux file types
+
+10.3.1 Why is ``DEL'' reported as a Linux file type?
+
+ Lsof usually reports entries from the Linux /proc/<PID>/maps
+ file with ``mem'' in the TYPE column. However, when lsof can't
+ stat(2) a path in the process' ``maps'' file and the ``maps''
+ file entry contains ``(deleted)'', indicating the file was
+ deleted after it had been opened, lsof reports the file type as
+ ``DEL''.
+
+10.3.2 Why is ``unknown'' reported as a Linux file type?
+
+ Lsof may report a Linux file's type as ``unknown'' in the TYPE
+ column when lsof can't obtain complete stat(2) results for the
+ file.
+
+ Usually the NAME column will contain a ``(stat: xxx)'' error
+ message, but that could have been suppressed with the lsof
+ ``-w'' option.
+
+10.4 Linux ``mem'' Entry Problems
+
+10.4.1 What do ``path dev=xxx'' and ``path inode=yyy'' mean in the
+ NAME column of Linux ``mem'' file types?
+
+ When the device or inode number in the process' ``maps'' file
+ entry doesn't match the stat(2) results from the file path,
+ lsof reports the inconsistent information from the stat(2) of
+ the path parenthetically after the path in the NAME column
+ in one of these forms:
+
+ (path dev=xxx) only the device number,
+ ``xxx'', from a stat(2) of the
+ ``maps'' file entry path
+ differs from the ``maps'' file
+ entry value reported in the
+ DEVICE column.
+
+ (path inode=yyy) only the inode number,
+ ``yyy'', from a stat(2) of the
+ ``maps'' file entry path
+ differs from the ``maps'' file
+ entry value reported in the
+ NODE column.
+
+ (path dev=xxx inode=yyy) Both device and inode numbers
+ differ.
+
+ Lsof reports the ``maps'' file device number in the DEVICE
+ column and the inode number in the NODE column.
+
+ When device and inode mismatches occur, lsof suppresses the
+ reporting of link count and size. See the answer to the "Why
+ is neither link count nor size reported for some Linux ``DEL''
+ and ``mem'' file types?" question for more information.
+
+ Device and inode inconsistencies can occur when a file at a
+ ``maps'' path is replaced after the process has started, or
+ when a different file system with similar path names is mounted
+ on top of the original file system.
+
+ The device inconsistency parenthetical messages can be
+ suppressed with lsof's ``-w'' option.
+
+10.4.2 Why is neither link count nor size reported for some Linux
+ ``DEL'' and ``mem'' file types?
+
+ Link count and size are not reported for some entries from the
+ process' ``maps'' file because a stat(2) of the entry file path
+ failed or stat(2) delivered device or inode numbers that don't
+ match the ones in the ``maps'' entry.
+
+ When the stat(2) device or inode numbers don't match those in
+ the ``maps'' file entry, it is likely that the stat(2) results
+ don't apply to the file that was originally mapped by the
+ process and whose path appears in the ``maps'' file entry, so
+ lsof tries to avoid reporting possibly incorrect information.
+
+ See the answer to the "What do ``path dev=xxx'' and ``path
+ inode=yyy'' mean in the NAME column of Linux ``mem'' file
+ types?" for more information on how mismatched stat(2) device
+ and inode numbers are reported.
+
+10.5 Special Linux NAME column messages
+
+10.5.1 What does ``(stat: xxx)'' mean in the NAME column of Linux
+ files?
+
+ When lsof tried to stat(2) the path in the NAME column, the
+ stat(2) system call failed and produced an error message of
+ ``xxx''.
+
+ This situation usually occurs if the lsof process lacks
+ permission to stat(2) the path -- e.g., the lsof executable
+ lacks root permission, or lsof is attempting to stat(2) a path
+ on an NFS device mounted with the root_squash option.
+
+ The message can be suppressed with lsof's ``-w'' option.
+
+10.5.2 What does ``(readlink: xxx)'' mean in the NAME column of
+ Linux files?
+
+ When lsof tried to convert the /proc/<PID>/fd path, reported in
+ the NAME column, to its full and more meaningful path, the
+ readlink(2) system call used to do the conversion failed. The
+ readlink(2) failure message is ``xxx''.
+
+ This situation usually occurs if the lsof process lacks
+ permission to readlink(2) some part of the path -- e.g., the
+ lsof executable lacks root permission, or lsof is attempting to
+ stat(2) a path on an NFS device mounted with the root_squash
+ option.
+
+ The message can be suppressed with lsof's ``-w'' option.
+
+10.6 Why is ``NOFD'' reported as a Linux file type?
+
+ When lsof lacks permission to use opendir() on the fd/
+ subdirectory of a process' /proc/<PID> directory, it reports a
+ single file of the type ``NOFD'' (for no file descriptors).
+
+ Lsof reports the the /proc/<PID>/path in the NAME column,
+ followed by "(opendir: xxx)", where ``xxx'' is the error
+ message returned by opendir().
+
+ The ``NOFD'' entry can be suppressed with lsof's ``-w'' option.
+
+10.7 Why does Linux lsof report a NAME column value that begins with
+ ``/proc''?
+
+ When lsof has problems processing a ``/proc/<PID>'' entry --
+ e.g., it can't convert the entry to a full and more meaningful
+ path name, or it can't access the /proc/<PID>/fd subdirectory
+ with opendir() -- it will report the /proc/<PID> path in the
+ NAME column.
+
+10.8 Linux /proc/net/tcp* and /proc/net/udp* issues
+
+10.8.1 Why use the Linux -X option?
+
+ If you're not interested in TCP/IP socket information for a
+ particular use of lsof, adding the -X option will make lsof run
+ more quickly, because -X inhibits the reading of the
+ /proc/net/tcp* and /proc/net/udp* files. For example, you may
+ only be interested in knowing what process has a particular
+ file open.
+
+ When the Linux system has a large number of open TCP/IP socket
+ files, the time savings provided by -X can be significant.
+
+10.8.2 Why does lsof say ``-i is useless when -X is specified''?
+
+ If -X is specified, lsof can't report much information on open
+ TCP/IP socket files. However, lsof's -i option requests that
+ information. Hence, the two options conflict and can't be used
+ together.
+
+10.8.3 Why does lsof say ``can't identify protocol (-X specified)''?
+
+ If the Linux lsof -X option is specified and an open socket
+ file can't be identified without accessing the /proc/net/tcp*
+ and /proc/net/udp* files, lsof will report that it can't
+ identify the socket's protocol and that the failure may be
+ caused by the -X specification
+
+
+11.0 NetBSD Problems
+
+11.1 Why doesn't lsof report on open kernfs files?
+
+ Lsof doesn't report on open NetBSD kernfs files because the
+ structures lsof needs aren't defined in the kernfs.h header
+ file in /sys/misc/kernfs.
+
+11.2 Why doesn't lsof report on open files on: file descriptor
+ file systems; /proc file systems; 9660 (CD-ROM) file systems;
+ MS-DOS (floppy disk) file systems; or kernel file systems?
+
+ Lsof is not able to report on open files on certain file
+ system if /usr/src/sys/msdosfs didn't exist when the lsof
+ Configure script ran and lsof was made. /usr/src/sys/msdosfs
+ contains header files lsof needs for collecting data on
+ certain file system files.
+
+ You can tell if an lsof executable above) lacks support
+ for a file system if the following test of `lsof -v` produces
+ nothing:
+
+ $ lsof -v 2>&1 | grep <support_enabled_definition>
+
+ The <support-enabled_definition> will be:
+
+ File System Type Definition Note
+ ---------------- ---------- ----
+ File descriptor HASFDESCFS
+ /proc HASPROCFS
+ 9660 HAS9660FS
+ MS-DOS HASMSDOSFS (lsof 4.61 and above)
+ Kernel HASKERNFS
+
+ The work-around is to install /usr/src/sys, rerun the lsof
+ Configure script, and remake lsof.
+
+11.3 Why does lsof produce confusing results for nullfs file
+ systems?
+
+ Consider this report from /sbin/mount:
+
+ /usr/home on /home type null (local)
+
+ (According to /sbin/mount /usr/home is the mounted-on device
+ and /home is the mounted-on directory.)
+
+ When lsof is asked to report on open files on /home, it
+ will report them as files on /usr/home instead. That's an
+ artifact of the NetBSD kernel's dynamic name lookup cache
+ (DNLC) and the way the kernel handles nullfs mounted-on
+ directories.
+
+ While lsof will report all open files on /home when given
+ /home as a file system directory argument, even though
+ reporting them as located on /usr/home, lsof will not find
+ the same files when asked to report on all open files on
+ /usr/home when given /usr/home as a file system device
+ argument. That's because from the mount perspective
+ /usr/home is equivalent to a device, but from the device
+ perspective it is still a directory.
+
+ So, what this lsof command reports:
+
+ $ lsof /home
+ ... NAME
+ ... /usr/home/...
+
+ Won't be duplicated by this lsof command:
+
+ $ lsof /usr/home
+
+ Another way to look at this confusing /home and /usr/home
+ example is to consider what stat(2) reports. For /home
+ stat(2) reports a device doublet that matches what lsof
+ finds in open file node structures, while the device doublet
+ stat(2) reports for /usr/home won't match what lsof finds.
+ Nor does the mode reported by stat(2) indicate a block
+ devices, as is the expected case.
+
+ There is no simple answer to this confusion, nor is there
+ even a simple explanation. Simply be aware that when
+ supplying file system arguments to lsof on NetBSD, use the
+ mounted-on directory name for a nullfs as the lsof argument,
+ and don't be surprised when the NAME column reports the
+ mounted-on device name.
+
+11.4 NetBSD header file problems
+
+11.4.1 Why can't the compiler find some NetBSD header files?
+
+ If the compiler's pre-processor complains it can't find some
+ header files when it compiles lsof source files, /usr/include
+ and /usr/src may not have all the header files lsof needs.
+
+ As a work-around use the NETBSD_SYS environment variable
+ to specify to lsof the location of the additional header
+ files -- e.g.,
+
+ % setenv NETBSD_SYS /my_source
+ % ./Configure -n netbsd
+
+ or
+ $ NETBSD_SYS=/mys_source ./Configure -n netbsd
+
+ Caution: using this work-around may cause the lsof Configure
+ script to activate or omit different features, depending
+ on where it finds the header files that determine the state
+ of the features.
+
+11.4.2 Why does NetBSD lsof produce incorrect output?
+
+ If the NetBSD system's kernel was built from header files that
+ don't match those in /usr/include -- e.g., //usr/src has the
+ ones from which the kernel was built -- lsof may build, but
+ won't produce correct output.
+
+ As a possible work-around, try directing the C compiler to
+ select header files from /usr/src before it selects them from
+ /usr/include. That can be done with the DEBUG make string --
+ e.g.,
+
+ $ make DEBUG="-I/usr/src -I/usr/include"
+
+ If that work-around fails, try using the LSOF_INCLUDE and
+ NETBSD_SYS environment variables to swap /usr/include and
+ /usr/src when running the Configure script, then use the make
+ DEBUG string when running make -- e.g.,
+
+ $ LSOF_INCLUDE=/usr/src; export LSOF_INCLUDE
+ $ NETBSD_SYS=/usr/include; export NETBSD_SYS
+ $ ./Configure -n netbsd
+ $ make DEBUG="-I/usr/src -I/usr/include"
+
+11.5 Why isn't lsof feature xxx enabled for NetBSD?
+
+ Lsof's Configure script enables NetBSD features by locating
+ and examining header files associated with the features,
+ and based on what it finds, setting compile-time definitions
+ in Makefiles. (See 00PORTING for a list of the definitions.)
+
+ When Configure doesn't find header files or doesn't find
+ appropriate values in header files, that may mean the header
+ file tree lsof is searching is incomplete or out of date.
+
+ Lsof normally looks for NetBSD header files in /usr/include.
+ It can also be directed to look in other directories --
+ e.g., /sys -- if told to do so with the contents of the
+ LSOF_INCLUDE and NETBSD_SYS environment variables.
+
+ To determine what header file enables a missing feature,
+ check the NetBSD stanza in the Configure script. Then
+ check the locations it checks for the indicated header
+ files and contents.
+
+ See 00XCONFIG for more information on LSOF_INCLUDE and
+ and NETBSD_SYS.
+
+
+12.0 NEXTSTEP and OPENSTEP Problems
+
+12.1 Why can't lsof report on 3.1 lockf() or fcntl(F_SETLK)
+ locks?
+
+ Lsof has code to test for locks defined with lockf() or
+ fcntl(F_SETLK) under NEXTSTEP 3.1, but that code has never
+ been tested. I couldn't test it, because my NEXTSTEP 3.1
+ lockf() and fcntl(F_SETLK) functions return "Invalid
+ argument" every way I have tried to invoke them.
+
+ If your NEXTSTEP 3.1 system does allow you to use lockf()
+ and fcntl(F_SETLK) and lsof doesn't report locks set with
+ them, then the code in .../dialects/next/dnode.c probably
+ isn't correct. Please contact me via e-mail at <abe@purdue.edu>
+ and tell me how you got your lockf() and fcntl(F_SETLK) system
+ calls to work. Make sure "lsof" appears in the "Subject:" line
+ so my e-mail filter won't classify your letter as Spam.
+
+12.2 Why doesn't lsof compile for NEXTSTEP with AFS?
+
+ I no longer have a NEXTSTEP test system that has AFS.
+ Changes to lsof since I once had a test system have caused
+ me to change the AFS code in NEXTSTEP without being able
+ to test the changes.
+
+ If you need AFS support for NEXTSTEP and can't get it to
+ compile, please contact me. Perhaps we can jointly fix
+ the problems.
+
+
+13.0 OpenBSD Problems
+
+13.1 Why doesn't lsof support kernfs on my OpenBSD system?
+
+ Lsof supports the kernel file system on OpenBSD versions
+ whose /sys/miscfs/kernfs/kernfs.h (or <miscfs/kernfs/kernfs.h>
+ header file correctly defines the kern_target structure.
+ The lsof Configure script's openbsd stanza checks for the
+ presence of the structure's kt_name element and activates
+ kernfs support for the CFLAGS -DHASKERNFS definition only
+ when it finds kt_name.
+
+ The kernfs.h header file is scheduled to be updated in the
+ OpenBSD 2.1 release, according to Kenneth Stailey, who
+ authored its changes.
+
+13.2 Will lsof work on OpenBSD on non-x86-based architectures?
+
+ I've not tested lsof on an OpenBSD system that uses a
+ non-x86-based architecture, but I've had one report that
+ lsof 4.33 compiles and works on OpenBSD for the pmax
+ architecture (decstation 3100).
+
+13.3 <sys/pipe.h> problems
+
+13.3.1 Why does the compiler claim nbpg isn't defined?
+
+ When compiling lsof on some (older) OpenBSD SPARC versions,
+ the compiler may complain:
+
+ In file included from ../dlsof.h:191,
+ from ../lsof.h:166,
+ from fino.c:52:
+ /usr/include/sys/pipe.h:83: `nbpg' undeclared here
+ (not in a function)
+ /usr/include/sys/pipe.h:83: size of array `ms' has
+ non-integer type
+
+ This happens because <sys/pipe.h> uses NBPG from
+ <machine/param.h> to size the `ms' array, and some OpenBSD
+ systems define NBPG in terms of a kernel integer variable,
+ nbpg.
+
+ Lsof revisions 4.46 and above have a hack to dlsof.h,
+ developed by Volker Borchert that avoids the compiler
+ problem for SPARC OpenBSD 2.3. The hack might work for
+ other OpenBSD SPARC versions, but hasn't been tested there.
+
+ If you want to enable the hack for your OpenBSD SPARC
+ version, modify this code in .../dialects/n+obsd/dlsof.h:
+
+ # if defined(OPENBSDV)
+ # if OPENBSDV==2030 && defined(__sparc__)
+ # if defined(nbpg)
+ #undef nbpg
+ # endif /* defined(nbpg) */
+ #define nbpg 4096 /* WARNING!!! ... */
+ # endif /* OPENBSDV==2030 && defined(__sparc__) */
+ #include <sys/pipe.h>
+ #endif /* defined(OPENBSDV) */
+
+ You will probably want to change the second #if test to
+ match your OpenBSD version. You may also want to change
+ what value is assigned to nbpg. See the next section,
+ "What value should I assign to nbpg?"
+
+13.3.2 What value should I assign to nbpg?
+
+ If you need to enable the nbpg hack, described in "Why does
+ the compiler claim nbpg isn't defined?", you may also need
+ to assign a value other than 4096 to nbpg. 4096 works for
+ the sun4c processor and should work for sun4m, but 8192
+ may be needed for sun4.
+
+ Check <machine/param.h> and other OpenBSD documentation to
+ determine the correct nbpg assignment.
+
+13.4 Why doesn't lsof report on open MS-DOS file system (floppy
+ disk) files?
+
+ Lsof is not able to report on open MS-DOS file system files
+ if /usr/src/sys/msdosfs didn't exist when the lsof Configure
+ script ran and lsof was made. /usr/src/sys/msdosfs contains
+ header files lsof needs for collecting data on MS-DOS file
+ system files.
+
+ You can tell if an lsof executable (revisions 4.61 and
+ above) lacks MS-DOS file system support if the following
+ command reports nothing:
+
+ $ lsof -v 2>&1 | grep HASMSDOSFS
+
+ The work-around is to install /usr/src/sys, rerun the lsof
+ Configure script, and remake lsof.
+
+13.5 Why isn't lsof feature xxx enabled for OpenBSD?
+
+ Lsof's Configure script enables OpenBSD features by locating
+ and examining header files associated with the features,
+ and based on what if finds, setting compile-time definitions
+ in Makefiles. (See 00PORTING for a list of the definitions.)
+
+ When Configure doesn't find header files or doesn't find
+ appropriate values in header files, that may mean the header
+ file tree lsof is searching is incomplete or out of date.
+
+ Lsof normally looks for OpenBSD header files in /usr/include
+ and /sys. It can also be directed to look in other
+ directories if told to do so with the contents of the
+ LSOF_INCLUDE and NETBSD_SYS environment variables.
+
+ To determine what header file enables a missing feature,
+ check the OpenBSD stanza in the Configure script. Then
+ check the locations it checks for the indicated header
+ files and contents.
+
+ See 00XCONFIG for more information on LSOF_INCLUDE and
+ and NETBSD_SYS.
+
+
+14.0 Output Problems
+
+14.1 Why do the lsof column sizes change?
+
+ Lsof dynamically sizes its output columns each time it runs
+ to make sure that each column takes the minimum space.
+ Column parsing -- e.g., with awk -- is possible, because
+ each column is guaranteed to be separated from the preceding
+ one by at lease one space, and no column except the last
+ (NAME) contains embedded spaces.
+
+14.2 Why does the offset have ``0t' and ``0x'' prefixes?
+
+ The offset value that appears in the SIZE/OFF column has
+ ``0t' and ``0x'' prefixes to distinguish it from size values
+ that may appear in the same column.
+
+ Normally if the offset value is less than 100,000,000 (8
+ digits), it appears in decimal with a ``0t' prefix; over
+ 99,999,999, in hexadecimal with a ``0x'' prefix.
+
+ A decimal offset is handy, for example, when tracking the
+ progress of an outbound ftp transfer. When lsof reports
+ on the ftp process, it will report the size of the file
+ being sent with its open descriptor; it will report the
+ progress of the transfer via the offset of the outbound
+ open ftp data socket descriptor.
+
+ The ``-o [n]'' option may be used to specify the maximum
+ number of decimal digits to be printed after ``0t'' before
+ lsof switches to the hexadecimal digits after `0x''. As
+ already noted, the default decimal digit count is 8.
+
+14.3 What are the values printed in the FILE_FLAG column
+ and why is 0x<value> sometimes included?
+
+ The two comma separated lists, separated by a semicolon,
+ printed in the FILE-FLAG column (when the "+fg" option is
+ specified), are short-hand names or hexadecimal values for
+ the bits lsof finds in the f_flag or f_flags member of file
+ structures for files (the first list, the one before the
+ semicolon), and process open files flags found in various
+ kernel structures, often named "pofile" (the second list,
+ the one after the semicolon).
+
+ Lsof determines the short-hand names from symbols in the
+ <fcntl.h>, <linux/fs.h>, <sys/fcntl.h>, <sys/fcntlcom.h>,
+ o<sys/file.h>, and <sys/user.h> header files.
+
+ See the discussion of FILE-FLAG in the OUTPUT section of
+ the lsof man page, and the FF_* and POF_* symbols in lsof.h
+ for a list of the names.
+
+ Bits with no names defined for them are represented by an
+ 0x<value> member of the comma-separated list -- a hexadecimal
+ integer. When "+fG" is specified (instead of "+fg"), lsof
+ will list all flag values as two hexadecimal integers,
+ separated by a semicolon.
+
+ When "-FG" is specified to get the flags in an output field,
+ the format defaults to hexadecimal. You can get names
+ instead by following "-FG" with "+fg" -- e.g.,
+
+ $ lsof -FG +fg ...
+
+ However, when you precede "-FG" with "+fg" -- e.g.,
+
+ $ lsof +fg -FG
+
+ the format will be hexadecimal; order is important.
+
+14.3.1 Why doesn't lsof display FILE_FLAG values for my dialect?
+
+ All versions of lsof except the /proc-based Linux lsof
+ report FILE-FLAG values. Lsof can't obtain FILE-FLAG
+ information from the Linux /proc interface.
+
+14.4 Network Addresses
+
+14.4.1 Why does lsof's -n option cause IPv4 addresses, mapped to
+ IPv6, to be displayed in IPv6 notation?
+
+ When you use the -n option to tell lsof to display numeric
+ network addresses, and an IPv4 address has been mapped to
+ IPv6, lsof displays the address in IPv6 format and puts
+ "ipv4" in the TYPE column. That combination indicates the
+ IPv4 address has been mapped to IPv6.
+
+ For example, the IPv4 address 1.2.3.4, when mapped to an
+ IPv6 address, will be displayed by lsof as:
+
+ [::ffff:1.2.3.4]
+
+ The enclosing brackets are lsof's signal that this is an
+ IPv6 address. Inside the brackets is a standard IPv6
+ address, reported by inet_ntop(). The first two colons,
+ signifying zeroes in the first 64 bits of the IPv6 address,
+ and the hexadecimal ffff in the next 32 bits, indicate that
+ the last 32 bits contains a mapped IPv4 address, which is
+ then displayed in IPv4 dot notation.
+
+14.5 Why does lsof output \x, ^x, or \xnn for characters
+ sometimes?
+
+ Lsof displays only printable ASCII characters. Lsof
+ considers a character printable if isprint(3) says it
+ is. If isprint(3) says a character isn't printable,
+ the lsof may page explains:
+
+ "... Non-printable characters are printed in one of
+ three forms: the C ``\[bfrnt]'' form; the control
+ character `^' form (e.g., ``^@''); or hexadecimal
+ leading ``\x'' form (e.g., ``\xab''). Space is
+ non-printable in the COMMAND column (``\x20'') and
+ printable elsewhere."
+
+14.5.1 Why is space considered a non-printable character in command
+ names?
+
+ Space is considered an unprintable character in command
+ names because it is sometimes possible to hide the full
+ command name from scripts that parse ps(1) output by
+ embedding a space in the name.
+
+14.6 Why doesn't lsof print all the characters of a command name?
+
+ By default lsof prints the first nine characters of the
+ names of commands associated with processes. If more
+ characters are required, the "w" value of the "+c w" option
+ may be used to specify a larger width.
+
+ If "w" is zero ('0') lsof will print all characters of all
+ command names up to the limit of the number of characters
+ supplied by the particular UNIX dialect. When reporting
+ command names, lsof replaces non-printable characters as
+ discussed in the answer to " Why does lsof output \x, ^x, or
+ \xnn for characters sometimes?"
+
+ See the answer to the "Why is space considered a non-printable
+ character in command names?" question for an explanation of why
+ spaces are replaced by the ``\x20'' representation in command
+ names.
+
+ The number of command name characters supplied to lsof by UNIX
+ dialects in files and structures varies by dialect. For
+ example, Linux 2.4.27 supplies lsof the first 15 characters of
+ command names and Solaris 9 supplies 16. Thus, even if "w" is
+ zero ('0'), lsof can't report more characters for command names
+ on those two UNIX dialects than they provide lsof.
+
+14.7 Why does lsof reject some -c command names, saying their lengths
+ are "> what system provides (nn)"?
+
+ The command name length that a specific system provides varies
+ from dialect to dialect. As noted in the answer to the "Why
+ doesn't lsof print all the characters of a command name?"
+ question, Linux and Solaris provide a limited number of command
+ name characters.
+
+ When more characters are specified in the parameter to the -c
+ option, lsof considers it an error and issues a fatal error
+ message -- e.g.,
+
+ lsof: "-c xxxxyyyy" length (8) > what system provides (7)
+
+ The only work-around is to specify no more characters to -c
+ that the system provides to lsof.
+
+14.8 Why does lsof sometimes print TYPE numbers instead of names?
+
+ When lsof can't convert a type number to a name for printing in
+ the TYPE column, it will report the number as four octets.
+
+14.9 Marker line format problems
+
+14.9.1 Why won't lsof accept a marker line format?
+
+ Lsof's Configure script must find the localtime(3) and
+ strftime(3) functions in the dialect's C library in order to
+ enable support for marker line formats.
+
+ Check the output of lsof's -v option for the presence of
+ -DHAS_STRFTIME in the compiler flags. If it isn't there,
+ Configure didn't find the necessary two C library functions.
+
+ If you think lsof should have found the functions, make a copy
+ of the C test program in the Configure script that it uses to
+ find the functions. Then use the copy, or a more informative
+ modification of it, to learn why Configure can't find the
+ functions. You can find that program by searching for
+ strftime.
+
+14.9.2 Why does lsof reject the NL (%n) marker line format?
+
+ When repeat mode and field output (with -F) have both been
+ specified, lsof won't allow new line (NL) formats to be
+ specified with ``%n''. That's because the marker line is
+ always guaranteed to be a single line.
+
+ There is no work-around to this restriction.
+
+14.10 How are protocol state name exclusion and inclusion used?
+
+ Protocol state name inclusion and exclusion with the ``-s p:s''
+ option and its arguments have some issues to consider.
+
+ First, there is the problem of determining what state names, if
+ any, the dialect produces. Try running this lsof command to
+ find them:
+
+ $ lsof -i
+
+ Knowing the state names of interest, the next problem is to
+ decide on the lsof options and their parameters that will
+ produce the desired output. Here some examples are probably
+ the most useful.
+
+ To list only TCP socket files in LISTEN and CLOSE_WAIT states,
+ use:
+
+ $ lsof -itcp -stcp:listen,close_wait
+ or
+ $ lsof -iTCP -sTCP:LISTEN,CLOSE_WAIT
+
+ Case isn't important to lsof in protocol and state names.
+
+ To exclude TCP socket files in CLOSE_WAIT state, use:
+
+ $ lsof -itcp -stcp:^close_wait
+
+ Note the `^' preceding close_wait; it selects exclusion. You
+ can mix included and excluded names in a comma separated list,
+ but you may not include and exclude the same name for the same
+ protocol.
+
+ To list TCP files in LISTEN state and UDP files in Idle state,
+ use:
+
+ $ lsof -i -stcp:listen -sudp:idle
+
+ Note: if you don't accompany the ``-s p:s'' list option and
+ argguments with the -i option, lsof will list all other regular
+ files, while applying the specified inclusion and exclusion
+ specifications to network files. Generally, then, you want to
+ use -i with -s.
+
+14.10.1 Why doesn't my dialect support state name exclusion and inclusion?
+
+ When state name inclusion and exclusion was added, I had access
+ to test systems for AIX, Darwin, FreeBSD, Linux, PSTAT-based
+ HP-UX and Solaris.
+
+ Therefore, I was unable to add and test the support to any other
+ UNIX dialects.
+
+ If a dialect has the support, then the HASTCPUDPSTATE definition
+ in its machine.h header file will be active; if not, it will be
+ absent or commented out.
+
+ If your dialect doesn't have the support and you want it added,
+ you will have to provide me Internet access to a test host, where
+ I can compile lsof and have the credentials to test the changes
+ the support requires. If that's possible for you, please contact
+ me via e-mail at <abe@purdue.edu>. Make sure "lsof" appears in
+ the "Subject:" line so my e-mail filter won't classify your letter
+ as Spam.
+
+
+15.0 Pyramid Version Problems
+
+15.0.5 Statement of deprecation
+
+ As of lsof revision 4.52 support for all Pyramid versions has
+ been dropped. Contact me via e-mail at <abe@purdue.edu> if you
+ are interested in obtaining the last lsof Pyramid distribution.
+ Make sure "lsof" appears in the "Subject:" line so my e-mail
+ filter won't classify your letter as Spam.
+
+
+16.0 SCO Problems
+
+16.1 SCO OpenServer Problems
+
+16.1.1 How can I avoid segmentation faults when compiling lsof?
+
+ If you have an older SCO OpenServer compiler, it may get
+ a segmentation fault when compiling some lsof modules.
+ That appears to happen because of the -Ox optimization
+ action requested in the lsof Makefile.
+
+ Try changing -Ox to -O with this make invocation:
+
+ $ make DEBUG=-O
+
+ Bela Lubkin supplied this tip and Steve Williams verified
+ it.
+
+16.1.2 Where is libsocket.a?
+
+ If you compile lsof and the loader says it can't find the
+ socket library, libsocket.a, called by the -lsocket option
+ in the lsof compile flags, you probably are running an SCO
+ OpenServer release earlier than 5.0 and don't have the
+ TCP/IP Development System package installed.
+
+ You may have the necessary header files, because you have
+ the TCP/IP run-time package installed, but if you don't
+ have the TCP/IP Development System package installed, you
+ won't have libsocket.a.
+
+ Your choices are to install the TCP/IP Development System
+ package or upgrade to OpenServer Release 5.0. You will
+ find libsocket.a in 5.0 -- you'll find all the libraries
+ and header files there, in fact -- and you can use gcc to
+ compile lsof if you don't want to install the 5.0 Development
+ System package.
+
+16.1.3 Why do I get "warning C4200" messages when I compile lsof?
+
+ When you compile lsof under OSR 3.2v4.2 (and perhaps under
+ earlier versions as well), you may get many compiler warning
+ messages of the form:
+
+ node.c(183) : warning C4200: previous declarator is not
+ compatible with default argument promotion
+
+ In my opinion this is a bug in the OSR compiler. Because
+ the compiler cannot handle full ANSI-C prototypes, it
+ assumes default types for function parameters as it encounters
+ untyped in a function prototype -- e.g., in this function
+ declaration from node.c,
+
+ readrnode(ra, r)
+ KA_T ra;
+ struct rnode *r;
+ {
+ ...
+
+ the compiler assigns default int types to the ra and r
+ arguments.
+
+ Then, when the compiler encounters the fully typed parameters
+ after the function skeleton and sees parameters with types
+ that don't match the assumptions it previously made, it
+ whines about its own assumptions.
+
+ You can ignore these messages.
+
+16.2 SCO|Caldera UnixWare Problems
+
+16.2.1 Why doesn't lsof compile on my UnixWare 7.1.1 or above
+ system?
+
+ When you Configure lsof with the "uw" abbreviation and try
+ to compile it for UnixWare 7.1.1, you may get compiler
+ error messages like this:
+
+ UX:acomp: ERROR: "dproc.c", line 98:
+ undefined struct/union member: p_pgidp
+
+ This suggest that you probably have a non-stop cluster
+ UnixWare 7.1.1 system. Its <sys/proc.h> header file differs
+ from the one on the system where I did the lsof port to
+ UnixWare 7.1.1. I currently don't have access to a non-stop
+ cluster system to be able to develop changes to lsof that
+ would make it compile and work there.
+
+ If you have a non-stop cluster UnixWare 7.1.1 system, want lsof
+ for it, and can offer me a test account on the system, please
+ contact me via e-mail at <abe@purdue.edu>. Make sure "lsof"
+ appears in the "Subject:" line so my e-mail filter won't
+ classify your letter as Spam.
+
+ If you have a system with nsc_cfs and can offer me a test
+ account on it, please contact me via e-mail at <abe@purdue.edu>.
+ Make sure "lsof" appears in the "Subject:" line so my e-mail
+ filter won't classify your letter as Spam.
+
+16.2.2 Why does lsof complain about node_self() on my UnixWare
+ 7.1.1 or above system?
+
+ If lsof exits immediately after issuing this message:
+
+ can't identify process NSC node; node_self(): <message>
+
+ It means that lsof has been built to run on a NonStop
+ Cluster (NSC) UnixWare 7.1.1 or higher system and can't
+ get the number of the node on which it is running. Lsof
+ uses the node number to determine the path to the kernel
+ boot file.
+
+ You can tell if lsof has been built for NSC by looking for
+ "-DHAS_UW_NSC" in lsof's "-v" option output.
+
+ If the system on which you're trying to run lsof isn't
+ running an NSC kernel, you will need to build a non-NSC
+ lsof.
+
+16.2.3 Why does UnixWare 7.1.1 or above complain about -lcluster,
+ node_self(), or libcluster.so?
+
+ When you build, compile, and load lsof for UnixWare 7.1.1
+ and above, ld may complain that it can't find the -lcluster
+ library or that the node_self symbol is undefined. When
+ you try to run an existing lsof binary it may complain that
+ libcluster.so can't be found.
+
+ These messages mean the tests made by Configure on your
+ system led it to believe your system is running a NonStop
+ Cluster (NSC) kernel, or the lsof binary you're trying to
+ use was built on a NonStop Cluster system. If an lsof
+ binary was built for NSC, this shell command produces
+ output:
+
+ $ strings <lsof_binary> | grep HAS_UW_NSC
+
+ If that's not the case, and you can rebuild lsof, set the
+ UW_HAS_NSC environment variable to "N" and do this:
+
+ $ Configure -n clean
+ $ UW_HAS_NSC=N
+ $ export UW_HAS_NSC
+ $ Configure -n uw
+ $ make
+
+ You can also edit Makefile and lib/Makefile. Remove
+ -DHAS_UW_NSC from the CFGF strings. Remove -lcluster from
+ the CFGL strings. Then run make again.
+
+ If you have an existing NSC lsof binary and you want one
+ for a non-NSC system, you will have to build lsof yourself
+ on the system where you want to use it. (That's always a
+ good idea anyway.)
+
+
+16.2.4 Why does UnixWare 7.1.1 or above lsof complain it can't
+ read the kernel name list?
+
+ If lsof complains:
+
+ can't read kernel name list from <path>
+
+ It means that lsof can't find the booted kernel image file
+ at <path>. On NonStop Cluster (NSC) UnixWare 7.1.1 or
+ higher systems lsof determines the booted file path by
+ examining this file:
+
+ /stand/`node_self`/boot
+
+ If examining that file doesn't lead to an NSC path, lsof
+ uses:
+
+ /stand/1/unix
+
+ On non-NSC systems lsof expects the booted kernel image to
+ be in /stand/unix.
+
+ If your booted kernel image is in a different place, use
+ lsof's "-k <path>" option to specify its path.
+
+16.2.5 Why doesn't lsof report link count, node number, and size
+ for some UnixWare 7.1.1 or above CFS files?
+
+ Lsof reports link count, node number, and size for open
+ CFS files as recorded in their kernel node structure's
+ cached attributes. Sometimes not all attributes are cached
+ on the node where lsof runs, so lsof cannot report them.
+
+16.2.6 Why doesn't lsof report open files on all UnixWare 7.1.1
+ NonStop Cluster (NSC) nodes?
+
+ Lsof can only report on files open on the node on which it
+ runs, because the information lsof reports comes from the
+ private kernel memory of the node. This may mean that
+ asking lsof to find a specific open file, or use of a
+ specific Internet address or port, may not report all open
+ instances on nodes other than the one used to run lsof.
+
+ You can use the NSC onnode(1) command to run lsof on specific
+ nodes, or the onall(1) command to run lsof on all nodes --
+ e.g.,
+
+ $ onall lsof [options] 2>&1 | less
+ or
+ $ onnode node-number lsof [options] 2>&1 | less
+
+ Note that, when lsof is run all nodes, the path name
+ component assembly results it reports in its NAME column
+ may vary, because the dynamic name cache from which lsof
+ gets the components is private to the kernel of each node.
+
+ Also note the use of shell redirection in the examples to
+ merge the standard error file information from onnode and
+ onall with lsof's standard output file output. That will
+ put the onnode and onall node announcements in proper
+ sequence with lsof's output.
+
+16.2.7 Why doesn't lsof report the UnixWare 7.1.1 NonStop Cluster
+ (NSC) node a process is using?
+
+ To induce lsof to report the node on which a process runs
+ would be a significant, non-standard modification to lsof.
+ It has much wider implications than merely the printing of
+ a number in an output column. I'm not currently (April
+ 2001) prepared to undertake such a modification.
+
+ If you want node-specific NSC information about open files,
+ run lsof under the control of onall(1) or onnode(1).
+
+ $ onall lsof [options] 2>&1 | less
+ or
+ $ onnode node-number lsof [options] 2>&1 | less
+
+16.2.8 Why does the compiler complain about missing UnixWare 2.1[.x]
+ header files?
+
+ SCO|Caldera didn't ship the following header files with
+ UnixWare 2.1 through 2.1.3:
+
+ <fs/proc/prdata.h>
+ <fs/procfs/prdata.h>
+ <sys/fs/fifonode.h>
+ <sys/fs/namenode.h>
+
+ Lsof needs those header files for its compilation. Contact
+ SCO|Caldera to get copies of those header files.
+
+ If you can't get the header files from SCO|Caldera, please
+ contact me via e-mail at <abe@purdue.edu>. Make sure "lsof"
+ appears in the "Subject:" line so my e-mail filter won't
+ classify your letter as Spam.
+
+
+17.0 Sun Problems
+
+17.0.5 Statement of deprecation
+
+ Lsof support for SunOS 4.1.x was last tested at revision 4.51.
+ Contact me via e-mail at <abe@purdue.edu> if you're interested in
+ obtaining it. Make sure "lsof" appears in the "Subject:" line so
+ my e-mail filter won't classify your letter as Spam.
+
+17.1 My Sun gcc-compiled lsof doesn't work -- why?
+
+ Gcc can be used to build lsof successfully. However, an
+ improperly installed Sun gcc compiler will usually not
+ produce a working lsof.
+
+ If your Sun gcc-compiled lsof doesn't report anything, or
+ reports ``can't read proc table,'' or gcc refuses to compile
+ lsof without error, check that the gcc step that "fixes"
+ Sun header files was run on the system where you're using
+ gcc to compile lsof. As an alternative, if you have the
+ SunPro C 5.0 compiler or later available, use it to compile
+ lsof -- e.g., use the solariscc Configure abbreviations.
+
+17.2 How can I make lsof compile with gcc under Solaris 2.[456],
+ 2.5.1, 7, 8 or 9?
+
+ Presuming your gcc-specific header files are wrong for
+ Solaris, edit the lsof Configure-generated Makefile and
+ lib/Makefile and make this change:
+
+ CFGF= -Dsolaris=20400 ...
+ to
+ CFGF= -Dsolaris=20400 -D__STDC__=0 -I/usr/include ...
+
+ or change:
+
+ CFGF= -Dsolaris=20500 ...
+ to
+ CFGF= -Dsolaris=20500 -D__STDC__=0 -I/usr/include ...
+
+ or change:
+
+ CFGF= -Dsolaris=20501 ...
+ to
+ CFGF= -Dsolaris=20501 -D__STDC__=0 -I/usr/include ...
+
+ This is only a temporary work-around. You really should
+ instruct gcc to to update your gcc-specific header files
+ or install a recent gcc (e.g., 3.2), which has no need for
+ private copies of Solaris include files.
+
+17.3 Why does Solaris Sun C complain about system header files?
+
+ You're probably trying to use /usr/ucb/cc if you get compiler
+ complaints like:
+
+ cc -O -Dsun -Dsolaris=20300 ...
+ "/usr/include/sys/machsig.h", line 81: macro BUS_OBJERR
+ redefines previous macro at "/usr/ucbinclude/sys/signal.h",
+ line 444
+
+ Note the reference to "/usr/ucbinclude/sys/signal.h". It
+ reveals that the BSD Compatibility Package C compiler is
+ in use. Lsof requires the ANSI C version of the Solaris
+ C compiler, usually found in /usr/opt/bin/cc or
+ /opt/SUNWspro/bin/cc.
+
+ Try adding a CC string to the lsof Makefile that points to
+ the Sun ANSI C version of the Sun C compiler -- e.g.,
+
+ CC= /usr/opt/bin/cc
+ or
+ CC= /opt/SUNWspro/bin/cc.
+
+17.4 Why doesn't lsof work under my Solaris 2.4 system?
+
+ If lsof doesn't work under your Solaris 2.4 system -- e.g.,
+ it produces no output, little output, or the output is
+ missing command names or file descriptors -- you may have
+ a pair of conflicting Sun patches installed.
+
+ Solaris patch 101945-32 installs a kernel that was built
+ with a <sys/auxv.h> header file whose NUM_*_VECTORS
+ definitions don't match the ones in the <sys/auxv.h> updated
+ by Solaris patch 102303-02.
+
+ NUM_*_VECTORS in the kernel of patch 101945-32 are smaller
+ than the ones in the <sys/auxv.h> of patch 102303-02. The
+ consequence is that when lsof is compiled with the <sys/auxv.h>
+ whose NUM_*_VECTORS definitions are larger than the ones
+ used to compile the patched kernel, lsof's user structure
+ does not align with the one that the kernel employs.
+
+ If you have these two patches installed, contact Sun and
+ complain about the mis-match.
+
+ You may be able to work around the problem by editing
+ /usr/include/sys/auxv.h to have the following NUM_*_VECTORS
+ definitions:
+
+ #define NUM_GEN_VECTORS 4
+ #define NUM_SUN_VECTORS 8
+
+ The Configure script issues a prominent WARNING that you should
+ try the work-around.
+
+ I thank Leif Hedstrom for identifying the offending patches.
+
+17.5 Where are the Solaris header files?
+
+ If you try to compile lsof under Solaris and get a compiler
+ complaint that it can't find system header files, perhaps
+ you forgot to add the header file package, SUNWhea.
+
+17.6 Where is the Solaris /usr/src/uts/<architecture>/sys/machparam.h?
+
+ When you try to Configure lsof for Solaris 2.[23456], 2.5.1,
+ and 7 -- e.g., on a `uname -m` == sun4m system -- Configure
+ complains:
+
+ grep: /usr/src/uts/sun4m/sys/machparam.h:
+ No such file or directory
+ grep: /usr/src/uts/sun4m/sys/machparam.h:
+ No such file or directory
+
+ And when you try to compile the configured lsof, cc or gcc
+ complains:
+
+ dproc.c:530: `KERNELBASE' undeclared (first use this function)
+
+ The explanation is that somehow your Solaris system doesn't
+ have the header files in /usr/src/uts it should have. Perhaps
+ someone removed the directory to save space. Perhaps you're
+ using a gcc installation, copied from another system. In any
+ event, you will have to load the header files from the SUNWhea
+ package of your Solaris distribution.
+
+ KERNELBASE is an important symbol to lsof -- it keeps lsof
+ from sending an illegal kernel value to kvm_read() where
+ a segmentation violation might result (a bug in the kvm
+ library). Lsof can get illegal kernel values because it
+ reads kernel values slowly with kvm_read() calls that the
+ kernel is changing rapidly.
+
+ Lsof doesn't need KERNELBASE at Solaris 2.5 and above,
+ because it has a KERNELBASE value whose address lsof can
+ find with /dev/ksyms and whose value it can read with
+ kvm_read(). Under Solaris 2.5 /usr/src/uts has moved to
+ /usr/platform.
+
+17.7 Why does Solaris lsof say ``can't read proc table''?
+
+ When lsof collects data on processes, using the kvm_*()
+ functions to scan the kernel's proc structure table, it
+ checks to make sure it has identified a reasonable number
+ of them -- a minimum of three. When lsof can't identify
+ three processes during a scan, it repeats the scan.
+
+ When five scans fail to yield three processes, lsof issues
+ the fatal message:
+
+ lsof: can't read proc table
+
+ and exits.
+
+ Usually lsof fails to identify three processes during a
+ scan because its idea of the form of the proc structure
+ differs from that being used by the kernel. Since the proc
+ structure is defined in <sys/proc.h> and other /usr/include
+ header files, the root cause of a proc structure discrepancy
+ usually can be found in the composition of /usr/include.
+
+ One common way that /usr/include header files can be
+ incorrect is that gcc was used to compile lsof, gcc used
+ its special (i.e., "fixed") header files instead of the
+ ones in /usr/include, and the special gcc header files
+ weren't updated when Solaris was. Answers to these questions:
+
+ My Sun gcc-compiled lsof doesn't work -- why?
+
+ How can I make lsof compile with gcc under Solaris 2.[456],
+ 2.5.1, 7, 8 or 9?
+
+ Why does Solaris Sun C complain about system header files?
+
+ discuss the gcc header file problem and offer suggestions
+ on how to fix it or work around it.
+
+ It may also be that you are trying to run a version of lsof
+ that was compiled on an older version of Solaris. For
+ example, an lsof executable, compiled for Solaris 2.4, will
+ produce the ``can't read proc table'' message if you try
+ to run it under Solaris 2.5. If you have compiled lsof
+ under Solaris 2.5 and it still won't work, see if the header
+ files in /usr/include have been updated to 2.5, or still
+ represent a previous version of Solaris.
+
+ Another source of header file discrepancies to consider is
+ the Solaris patch level and whether a binary kernel patch
+ was not matched with a corresponding header file update.
+ See the "Why doesn't lsof work under my Solaris 2.4 system?"
+ question for an example of one in Solaris 2.4 -- there may
+ be other such patch conflicts I don't know about.
+
+17.8 Why does Solaris lsof complain about a bad cached clone device?
+
+ When lsof revisions below 4.04 have been run on a Solaris
+ system and have been allowed to create a device cache file,
+ the running of revisions 4.04 and above on the same systems
+ may produce this complaint:
+
+ lsof: bad cached clone device: ...
+ lsof: WARNING: created device cache file: ...
+
+ This is the result of a change in the device cache file
+ that took place at lsof revision 4.04. The change introduced
+ a node number into the clone device lines of the device
+ cache file and was done in such a way that lsof could detect
+ device cache files whose clone lines don't have node numbers
+ (lines created by previous lsof revisions) and recognize
+ the need to regenerate the device cache file.
+
+17.9 Why doesn't Solaris make generate .o files?
+
+ Solaris /usr/ccs/bin/make won't generate .o files from .c
+ files if /usr/share/lib/make/make.rules is missing. It
+ may be found in and installed from the SUNWsport package.
+
+17.10 Why does lsof report some Solaris 2.3 and 2.4 lock types as `N'?
+
+ For Solaris 2.3 with patch P101318 installed at level 45
+ or above, and for all versions of Solaris 2.4, NFS locks
+ are represented by a NFS-specific kernel lock structure
+ that sometimes lacks a read or write lock type indicator.
+ When lsof encounters such a lock structure, it reports the
+ lock type as `N'.
+
+17.11 Why does lsof Configure say "WARNING: no cc in ..."?
+
+ When lsof's Configure script is executed with the solariscc
+ abbreviation it tries to make sure it's using the Sun C
+ compiler and not the UCB substitute from /usr/ucb/cc.
+ Thus, it looks for cc in the "standard" Sun compiler
+ location, /opt/SUNWspro/bin.
+
+ If Configure can't find cc there, it issues the warning:
+
+ lsof: WARNING: no cc in /opt/SUNWspro/bin;
+ using cc without path.
+
+ and uses cc for the compiler name, letting the shell find
+ cc with its PATH environment variable.
+
+ You can tell Configure where to find your cc with the
+ SOLARIS_CCDIR cross-configuration environment variable.
+ (See 00XCONFIG for more information on SOLARIS_CCDIR).
+ For example, use this Configure shell command:
+
+ SOLARIS_CCDIR=/usr/special/bin Configure -n solariscc
+
+ (SOLARIS_CCDIR should be the full path to the directory
+ containing your cc.)
+
+17.12 Solaris 7, 8 and 9 Problems
+
+17.12.1 Why does lsof say the compiler isn't adequate for Solaris
+ 7, 8 or 9?
+
+ Solaris 7, 8 and 9 kernels come in two flavors, 32 and 64
+ bit. 64 bit kernels run on machines that support the SPARC
+ v9 instruction set architecture. Separate executables for
+ some programs, -- e.g., ones using libkvm like lsof -- must
+ be built for 32 and 64 bit kernels.
+
+ Previous Sun (e.g., SC4.0) and earlier gcc compilers will
+ build lsof for 32 bit kernels, but they won't build it for
+ 64 bit kernels. Compilers that will build lsof for 64 bit
+ Solaris 7, 8 and 9 kernels are the Sun WorkShop Compilers
+ C 5.0 and above, and recent gcc versions, e.g., 3.2.
+
+ When given the ``-xarch=v9'' flag, the C 5.0 compiler and
+ above, and associated loader and 64 bit libraries will
+ build a 64 bit lsof executable; when given the "-m64" or
+ "-mcpu=v9" (deprecated) flags, an appropriate gcc compiler
+ will build a 64 bit lsof executable.
+
+ When the lsof Configure script detects a 64 bit kernel is
+ in use (e.g., by executing `/bin/isainfo -kv`), and when
+ it finds that the specified compiler is inappropriate,
+ it complains with these messages:
+
+ For gcc:
+
+ "!!!WARNING!!!=========!!!WARNING!!!=========!!!WARNING!!!"
+ "! !"
+ "! LSOF NEEDS TO BE CONFIGURED FOR A 64 BIT KERNEL, BUT !"
+ "! THIS GCC DOESN'T SUPPORT THE BUILDING OF 64 BIT !"
+ "! SOLARIS EXECUTABLES. LSOF WILL BE CONFIGURED FOR A !"
+ "! 32 BIT echo KERNEL. !"
+ "! !"
+ "!!!WARNING!!!=========!!!WARNING!!!=========!!!WARNING!!!"
+
+ For Sun C:
+
+ !!!WARNING!!!==========!!!WARNING!!!==========!!!WARNING!!!
+ ! !
+ ! LSOF NEEDS TO BE CONFIGURED FOR A 64 BIT KERNEL, BUT |
+ ! THE VERSION OF SUN C AVAILABLE DOESN'T SUPPORT THE !
+ ! -xarch=v9 FLAG. LSOF WILL BE CONFIGURED FOR A 32 BIT !
+ ! KERNEL. !
+ ! !
+ !!!WARNING!!!==========!!!WARNING!!!==========!!!WARNING!!!
+
+17.12.2 Why does Solaris 7, 8 or 9 lsof say "FATAL: lsof was compiled
+ for..."?
+
+ Solaris 7, 8 or 9 lsof may say:
+
+ lsof: FATAL: lsof was compiled for a xx bit kernel,
+ but this machine has booted a yy bit kernel.
+
+ Where: xx = 32 or 64
+ yy = 64 or 32
+
+ (xx and yy won't match.)
+
+ This message indicates that lsof was compiled for one size
+ kernel and is being asked to execute on a different size
+ one. That's not possible for programs like lsof that use
+ libkvm.
+
+ Depending on the instruction sets for which you need Solaris
+ 7, 8 or 9 lsof, you may need two or more versions of lsof,
+ compiled for each kernel size, installed for use with
+ /usr/lib/isaexec. See the "How do I install lsof for
+ Solaris 7, 8 or 9?" section of this document for more
+ information on that.
+
+17.12.3 How do I build lsof for a 64 bit Solaris kernel under a 32
+ bit Solaris kernel?
+
+ If your Solaris system has an appropriate compiler (e.g.,
+ WorkShop Compilers C 5.0 and above, or a recent gcc like
+ 3.2) and the 64 bit libraries have been installed, you can
+ force lsof's Configure script to build a 64 bit version of
+ lsof with:
+
+ $ SOLARIS_KERNBITS=64 Configure -n solariscc
+
+ The SOLARIS_KERNBITS environment variable is part of the
+ lsof cross-configuration support, described in the 00XCONFIG
+ file of the lsof distribution.
+
+17.12.4 How do I install lsof for Solaris 7, 8 or 9?
+
+ If you are installing lsof where it will be used only under
+ the bit size kernel for which it was built, no special
+ installation is required.
+
+ If, however, you are installing different versions of lsof
+ for different bit sizes -- e.g., for use on a 64 bit NFS
+ server and from its 32 bit clients -- you should read the
+ man page for isaexec(3C) and install lsof according to its
+ instructions.
+
+ The executable at the directory where lsof is to be found
+ should be a hard link to /usr/lib/isaexec or a copy of it.
+ In the directory there must be instruction architecture
+ subdirectories -- e.g., .../sparc/ and .../sparcv9/. The
+ lsof for 64 bit size kernels is installed in the .../sparcv9/
+ subdirectory; the one for 32 bit size kernels, in .../sparc/.
+
+ For example, if you're installing 32 and 64 bit lsof
+ executables in /usr/local/etc, you would:
+
+ # cd /usr/local/etc
+ # ln /usr/lib/isaexec lsof
+ # mkdir sparc sparcv9
+ # install the 32 bit lsof as sparc/lsof
+ # install the 64 bit lsof as sparcv9/lsof
+ # chmod, chown, and chgrp sparc/lsof and
+ sparcv9/lsof appropriately
+
+ Lsof permissions and ownerships are the same whether one
+ or more lsof executables are being installed, with or
+ without the /usr/lib/isaexec hard link.
+
+17.12.5 Why does my Solaris 7, 8 or 9 system say it cannot execute
+ lsof?
+
+ When you attempt to execute lsof, your Solaris 7, 8 or 9
+ shell may complain:
+
+ ksh: ./lsof: cannot execute
+
+ If the lsof executable exists and has the proper execution
+ permissions, this error may be the result of trying to
+ execute an lsof, built for a 64 bit kernel, on a 32 bit
+ kernel.
+
+ This will tell you about the lsof executable:
+
+ $ file lsof
+ lsof: ELF 64-bit MSB executable SPARCV9 Version 1,
+ dynamically linked, not stripped
+
+ The "64-bit" notation indicates the binary was built for
+ a 64 bit kernel. To see the running kernel bit size, use
+ this command:
+
+ $ isainfo -kv
+ 32-bit sparc kernel modules
+
+ The "32-bit" notation indicates a 32 bit kernel has been
+ booted.
+
+ The only work-around is to obtain, or Configure and make,
+ an lsof for the appropriate kernel bit size. If you
+ Configure and make lsof on the kernel where you wish to
+ run it the proper compiler, the lsof Configure step will
+ generate Makefiles that can be used with make to build an
+ appropriate lsof executable.
+
+ To compile a 64 bit lsof, you must have an appropriate
+ compiler -- i.e., Sun WorkShop Compilers C 5.0 or higher
+ or a recent gcc like 3.2.
+
+17.12.6 What gcc will produce 64 bit Solaris 7, 8 and 9 executables?
+ 8 and 9 executables?
+
+ Properly built and installed recent gcc versions -- e.g.,
+ 3.2 -- will build lsof for 64 bit Solaris kernels.
+
+ If you update your gcc version to 3.2 or later, make sure
+ the private gcc header files become current -- i.e., clear
+ out any private header files from a previous gcc or Solaris
+ installation before installing the new ones, or build to
+ a new --prefix root and replace the old root with it after
+ the build and installation are complete.
+
+17.12.7 Why does lsof on my Solaris 7, 8 or 9 system say, "can't
+ read namelist from /dev/ksyms?"
+
+ You're probably trying to use an lsof executable built for
+ an earlier Solaris release on a 64 bit Solaris 7, 8 or 9
+ kernel. The output from `lsof -v` will tell you the build
+ environment of your lsof executable. You should also have
+ gotten a warning message that lsof is compiled for a
+ different Solaris version than the one under which it is
+ running -- something like this:
+
+ lsof: WARNING: compiled for Solaris release X; this is Y
+
+ You need to build lsof on the system where you want to use
+ it. For 64 bit Solaris 7, 8 and 9 you need a compiler that
+ can generate 64 bit Solaris executables -- e.g., the Sun
+ Workshop 5 C compiler or later, or a recent gcc version
+ like 3.2. See the "Why does lsof say the compiler isn't
+ adequate for Solaris 7, 8 or 9?" section and the ones
+ following it for a discussion of building lsof for 64 bit
+ Solaris 7, 8 or 9.
+
+17.13 Solaris and COMMON
+
+17.13.1 What does COMMON mean in the NAME column for a Solaris VCHR
+ file?
+
+ When lsof puts COMMON or (COMMON) in the NAME column of a
+ Solaris VCHR file, it means that the file is handled by
+ the special file system functions of the kernel through a
+ common vnode.
+
+17.13.2 Why does a COMMON Solaris VCHR file sometimes seem to have an
+ incorrect minor device number?
+
+ When lsof reports on an open file in a Solaris special file
+ system that uses a COMMON vnode, and the file is a VCHR
+ file, lsof tries to locate the associated device node by
+ looking for matches on the major and minor device numbers
+ first.
+
+ If no major and minor match results, lsof then looks for
+ a match on pseudo and clone device files. (See /devices/pseudo.)
+ Those device nodes are matched specially by either their
+ major or minor device numbers, but not both. Hence, when
+ lsof finds a match under those special conditions, it may
+ report a value in its output DEVICE column that differs
+ from one of the major and minor numbers of the device node.
+
+ Here's an example from a sun4m Solaris 7 system:
+
+ $ ls -li /devices/pseudo/pm@0:pm
+ 151261 crw-rw-rw- 1 root sys 117, 0 ...
+ $ lsof /devices/pseudo/pm@0:pm
+ COMMAND ... DEVICE ... NODE NAME
+ powerd 117,1 ... 151261 /devices/pseudo/pm@0:pm (COMMON)
+ Xsun ... 117,0 ... 151261 /devices/pseudo/pm@0:pm
+
+ Note that the DEVICE value for the file with (COMMON) in
+ its name field has a different minor device number (1) from
+ what ls reports (0), while the DEVICE value for the file
+ without (COMMON) matches the ls output exactly. Both match
+ on the major device number, 117. The minor device number
+ mis-match is a result of the way the Solaris kernel handles
+ special file system common vnodes, and it's the reason lsof
+ puts (COMMON) after the name to signal that a mis-match is
+ possible.
+
+17.14 Why don't lsof and Solaris pfiles reports always match?
+
+ /usr/proc/bin/pfiles for Solaris 2.6, 7, 8, and 9 also
+ reports information on open files for processes. Sometimes
+ the information it reports differs from what lsof reports.
+
+ There are several reasons why this might be true. First,
+ because pfiles is a Sun product, based on Sun kernel
+ features, its developers have a better chance of knowing
+ exactly how open file information is organized. I sometimes
+ have to guess at how kernel file structure linkages are
+ constructed by gleaning hints from header files.
+
+ Second, lsof is aimed at providing information, specifically
+ device and node numbers, that can be used to identify named
+ file system objects -- i.e., path names. Thus, lsof tries
+ to make sure its device and node numbers match those reported
+ by stat(2). Pfiles doesn't always report numbers that
+ match stat(2) -- e.g., for files using clone and pseudo
+ devices via common vnodes like the nlist() /dev/ksyms usage.
+
+ Here's the Solaris 7 COMMON VCHR example again with additional
+ pfiles output:
+
+ $ ls -li /devices/pseudo/pm@0:pm
+ 151261 crw-rw-rw- 1 root sys 117, 0 ...
+ $ lsof /devices/pseudo/pm@0:pm
+ vic1: 10 = lsof /dev/pm
+ COMMAND ... DEVICE ... NODE NAME
+ powerd ... 117,1 ... 151261 /devices/pseudo/pm@0:pm (COMMON)
+ Xsun ... 117,0 ... 151261 /devices/pseudo/pm@0:pm
+ $ pfiles ...
+ 0: S_IFCHR ... dev:32,24 ino:61945 ... rdev:117,1
+ ...
+ 14: S_IFCHR ... dev:32,24 ino:151261 ... rdev:117,0
+
+ Note that the NODE number, reported by lsof, matches what
+ ls(1) and stat(2) report, while the ino value pfiles reports
+ doesn't. Lsof also indicates with the (COMMON) notation
+ that the DEVICE number is a pseudo one, derived from the
+ character device's value. The lsof DEVICE value matches
+ the pfiles rdev value, correct behavior for a character
+ device, but pfiles gives no sign that it's not possible to
+ find that character device number in /devices with ls(1)
+ or stat(2).
+
+17.15 Why does lsof say, "kvm_open(namelist=default, core=default):
+ Permission denied?"
+
+ Lsof needs permission to read from the /dev/kmem and /dev/mem
+ memory devices. Access to them is opened via a call to
+ the kvm_open() library function and it reports the indicated
+ message.
+
+ You must give lsof permission to read the memory devices.
+ The super user can almost always do that, but other lsof
+ users can do it if some group -- e.g., sys -- has permission
+ to read the memory devices, and the lsof binary is installed
+ with the group's ownership and with the setgid permission
+ bit enabled.
+
+17.16 Why is lsof slow on my busy Solaris UFS file system?
+
+ Lsof may be slow on a busy Solaris UFS file system when
+ UFS logging has been enabled with the "logging" mount
+ option. That option can significantly increase disk
+ operations under certain conditions -- e.g., when a lot of
+ files are accessed quickly.
+
+ When only the "logging" option is specified to mount, all
+ file accesses (atime updates) are logged to the UFS logging
+ queue. Each atime update requires two writes to the disk
+ to complete it.
+
+ If you want to do UFS logging -- and there are reliability
+ advantages to it -- consider using the "logging,noatime"
+ mount options instead. That will shift atime updates from
+ the logging queue to fewer and independent asynchronous
+ operations, consequently making the UFS logging queue a
+ smaller bottleneck.
+
+ Consult mount_ufs(1M) for more information on the logging
+ and noatime options.
+
+ (My thanks to Casper Dik for this tip on improving the
+ performance of UFS logging.)
+
+17.17 Why is lsof so slow on my Solaris 8 or 9 system?
+
+ Solaris 8 has a post-release feature upgrade modifying
+ kernel name cache (DNLC) handling that can slow lsof
+ throughput dramatically. The feature, sometimes called
+ negative DNLC caching, is standard in Solaris 9.
+
+ As best I can tell, when you install the Solaris 8 MU1
+ package, you get negative DNLC caching. If this pipe
+ produces any output, your system has negative DNLC caching.
+
+ $ nm /dev/ksyms | grep negative_cache_vnode
+
+ The reason negative DNLC caching perturbs lsof is that a
+ single vnode address (found in the negative_cache_vnode
+ kernel variable) is used to mark entries in the DNLC that
+ are not (the negative part) found on disk.
+
+ Since a single vnode address (the DNLC key lsof uses) can
+ represent many (I've seen upwards of 30,000.) DNLC entries,
+ their presence overloads lsof's internal DNLC hashing
+ function. An overloaded hash function is a slow hash
+ function, and lsof's slows to a crawl when it encounters
+ thousands of keys that produce the same value when the lsof
+ DNLC hash function is applied to them.
+
+ The solution is simple -- ignore negative DNLC cache keys.
+ They don't represent path name components lsof can use.
+ Lsof revisions 4.50 and above have an addition that ignores
+ them and the performance of those lsof revisions improves
+ significantly when presented with negative DNLC cache keys.
+
+ If you don't have an lsof revision at 4.51 or later, there's
+ a work-around. Use lsof's ``-C'' option. It disables
+ lsof's DNLC caching. Of course, that also inhibits the
+ reporting of any path name components from the kernel DNLC.
+ When ``-c'' is used, lsof will continue to report file
+ system and character device paths.
+
+17.18 Solaris and VxFS
+
+17.18.1 Why doesn't lsof support VxFS 3.4 on Solaris 2.6, and above?
+
+ Lsof will not support VxFS version 3.4 on Solaris 2.6 and above
+ unless some files from VxFS Update 2 have been installed. VxFS
+ 3.4 FCS and VxFS 3.4 update 1 lack the header files lsof
+ normally uses to obtain information from the VxFS 3.4 kernel
+ node structure, vx_inode. VxFS 3.4 Update 2 provides a method
+ whereby lsof can obtain the necessary vx_inode information from
+ the vxfsu_get_ioffsets() function in Veritas utility
+ libraries.
+
+ The utility libraries (32 bit and 64 bit versions) may be
+ found in /opt/VRTSvxfs/lib. An ancillary header file may
+ be found in /opt/VRTSvxfs/include/sys/fs/vx_libutil.h.
+ Documentation of the vxfsu_get_ioffsets(3) function may be
+ found in /opt/VRTS/man/man3/vxfsu_get_ioffsets.3.
+
+ Those files of VxFS 3.4 Update 2 may be downloaded from:
+
+ ftp://ftp.veritas.com/pub/support/vxfs_34.i64243.tar
+
+ The vxfs_34.i64243.tar archive will unpack into an i64243
+ directory containing these files:
+
+ $ ls i64243
+ README
+ libvxfsutil.sol26.sums
+ libvxfsutil.sol26.tar.Z
+ libvxfsutil.sol27.sums
+ libvxfsutil.sol27.tar.Z
+ libvxfsutil.sol28.sums
+ libvxfsutil.sol28.tar.Z
+
+ Read README. Select the *.tar.Z file appropriate for your
+ Solaris version. Its contents will unpack into /opt/VRTS
+ and /opt/VRTSvxfs, so you will need sufficient permission
+ -- e.g., do it as root -- to unpack the uncompressed archive.
+ Once you've done that, it's a good idea to compare the
+ checksums of the archive you unpacked with the ones recorded
+ in the appropriate *.sums file. Use `sum -r` to verify
+ the checksums.
+
+ For example, if you want the Solaris 8 version, uncompress
+ and unpack libvxfsutil.sol28.tar.Z -- e.g.,
+
+ $ su
+ ...
+ # cd i6423
+ # zcat libvxfsutil.sol28.tar.Z | tar xf -
+
+ That should create these new files and subdirectories with
+ the indicated checksums:
+
+ File or subdirectory sum -r
+
+ /opt/VRTSvxfs/include/vxfsutil.h 03938
+ /opt/VRTSvxfs/lib/libvxfsutil.a 51794
+ /opt/VRTSvxfs/lib/sparcv9/
+ /opt/VRTSvxfs/lib/sparcv9/libvxfsutil.a 07420
+ /opt/VRTS/man/man3/
+ /opt/VRTS/man/man3/vxfsu_get_ioffsets.3 62480
+
+ Once these files are in place, run lsof's Configure script
+ for the solaris or solariscc abbreviation. Configure will
+ locate the appropriate VxFS 3.4 Update 2 files and set up
+ for the making of an lsof that will properly display open
+ VxFS 3.4 file information.
+
+17.18.2 Why does lsof report "vx_inode: vxfsu_get_ioffsets error"
+ for open Solaris 2.6 and above VxFS 3.4 and above files?
+
+ Even when lsof supports VxFS 3.4 and above on Solaris 2.6 and
+ above, it may report "vx_inode: vxfsu_get_ioffsets error" in
+ the NAME column for all VxFS files.
+
+ The usual cause is that lsof doesn't have permission to
+ read the file at the end of the /dev/vxportal symbolic
+ link. If, for example, lsof has been installed setgid(sys),
+ then the /dev/vxportal symbolic link destination should be
+ owned by the sys group and readable by it.
+
+ Update 2 for VxFS 3.4 sets the modes of the /dev/vxportal
+ symbolic link destination to 0640 and the group ownership
+ to sys. But I have had a report that the modes are wrong
+ in a VxFS 4.0 installation.
+
+ Another cause may be that the system has more than one version
+ of VxFS installed (Only one can be active.), and lsof's
+ Configure script did not choose the header files and libraries
+ for the active VxFS version. Configure opts for VxFS 4.0 and
+ above header files and libraries (in /opt/VRTS) in preference
+ to those for VxFS below 4.0 (in /opt/VRTSvxfs).
+
+ Look for the directories /opt/VRTS and /opt/VRTSvxfs. If you
+ have /opt/VRTS, make sure its header and library symbolic links
+ point to those of the active VxFS version.
+
+ If you have both directories, look at the CFLAGS that Configure
+ constructed for making lsof and see which directory path
+ follows a -I option. If that doesn't match the directory path
+ of the active VxFS version, try pointing Configure at the
+ correct directory with the SOLARIS_VXFSINCL environment
+ variable -- e.g.,
+
+ $ SOLARIS_VXFSINCL=/opt/.../include ./Configure -n solaris
+
+17.18.3 Why does Solaris Configure claim there is no VxFS library?
+
+ The lsof Configure script, when configuring for Solaris, may
+ report:
+
+ FATAL: no VxFS .../libvxfsutil.a
+
+ That fatal error message indicates lsof has found the VxFS
+ utility library's header files, but can't find the library
+ itself in the expected location adjacent to the header files.
+
+ One possible cause is an incorrect symbolic link from
+ /opt/VRTS/lib/sparcv9/libvxfsutil.a to the library's real
+ location. (Some VxFS distributions declared the link
+ incorrectly.) Use `ls -lL` on that path to see if it exists.
+ If it doesn't exist, the link may be missing an additional
+ leading "../" component.
+
+ If the problem is a missing "../" from the library's link, you
+ can correct the link or check with Veritas/Symantec for the
+ patch that corrects it.
+
+ If the problem is not a missing "../", and you know the
+ libvxfsutil.a location, you can define its path in the
+ SOLARIS_VXFSLIB environment variable before running the lsof
+ Configure script. (See 00XCONFIG for information about using
+ the SOLARIS_VXFSLIB environment variable.)
+
+ If you have no libvxfsutil.a, you must obtain it from
+ Veritas/Symantec or find it in your VxFS installation package.
+
+17.18.4 Why doesn't Solaris lsof report VxFS path name components?
+
+ Solaris lsof will report path name components for VxFS versions
+ that use the common Solaris Dynamic Name Lookup Cache (DNLC) or
+ on some file systems of VxFS versions that support the VxFS
+ Reverse Name Lookup (RNL) facility.
+
+ VxFS versions 3.3 (approximately) and below use the common
+ Solaris DNLC. (I haven't been able to determine exactly when
+ VxFS stopped using the DNLC.) For versions above that boundary,
+ but below 4.0, lsof can't report path name components.
+
+ At VxFS 4.0 and above, lsof can be compiled to use the VxFS RNL
+ facility for reporting path names. If "-DHASVXFSRNL" appears
+ in the compiler flags section of lsof "-v" option output, then
+ the lsof Configure script detected the VxFS RNL facility and
+ lsof has been compiled to use it.
+
+ Lsof's use of the RNL facility can fail when the VxFS file
+ system disk layout version is below 6. In that case, lsof can
+ report no path name components. For more information, see the
+ vxfs_inotopath(3) manual page. any of the following commands
+ will show the disk layout version for a VxFS file system, when
+ supplied the block device or mount point on which the file
+ system is mounted.
+
+ fstyp -v <block_device>
+ or
+ mkfs -m <block_device>
+ or
+ vxupgrade <mount_point>
+
+ You must have permission to read the block device -- e.g., be
+ the root user.
+
+ You may also be able to upgrade an older disk layout to one
+ that will work with the RNL. See the vxupgrade(1M) man page
+ for more information on that.
+
+ When lsof can't report VxFS path name components, it reports
+ the file system mount point and the path name of device on
+ which it is mounted. The device path name is enclosed in
+ parentheses.
+
+17.18.5 Why does Solaris 10 lsof report scrambled VxFS paths?
+
+ Solaris 10 lsof may report a bogus, scrambled path for an open
+ VxFS file, when lsof obtains the path from a vnode's cached
+ path. Veritas/Symantec reports that their Solaris 10
+ implementation has bugs in the way it handles the Solaris 10
+ vnode cached path and those bugs will be fixed in an upcoming
+ patch some time after August 15, 2005.
+
+ When Solaris 10 lsof reports a path for an open VxFs file
+ obtained via the VxFS Reverse Name Lookup facility, the path
+ will be correct.
+
+ Also see the answers to the questions "Why does Solaris 10 lsof
+ sometimes report the wrong path name?" and "Why doesn't Solaris
+ lsof report VxFS path name components?"
+
+17.19 Large file problems
+
+17.19.1 Why does lsof complain it can't stat(2) a Solaris 2.5.1
+ large file?
+
+ When given an argument that is the path to a Solaris 2.5.1
+ file, enable for large file operations with the O_LARGEFILE
+ open(2) option, lsof complains that it can't stat(2) the
+ file. That's because lsof isn't using a stat(2) call and
+ associated structure enabled for large files.
+
+ This error has been fixed, starting at lsof revision 4.58
+ for Solaris 2.6 and above. That fix won't work on Solaris
+ 2.5.1 and I no longer have access to a Solaris 2.5.1 test
+ system to develop a separate fix.
+
+ The work-around is to avoid specifying a O_LARGEFILE path
+ as an argument to lsof on Solaris 2.5.1. Instead use a
+ combination of lsof and grep to achieve the same results,
+ albeit more clumsily.
+
+17.20 Why does lsof get a segmentation fault on 64 bit Solaris
+ 8 using NIS+?
+
+ I have received a report from Gary Craig that lsof produces
+ a segmentation fault on his 64 bit Solaris 8 system using
+ NIS+. Via an independent test program we have exonerated
+ lsof and tracked the fault to the NIS+ __nis_server_name()
+ function in the C name server library, -lnsl.
+
+ Lsof causes the __nis_server_name() NIS+ function to be
+ called by calling getservent() to read entries of the port
+ number to service name map.
+
+ The only Sun bug ID that appears to describe the problem
+ is 4304244, although its text is unclear enough to leave
+ room for doubt.
+
+ Until Sun eliminates the __nis_server_name() segmentation
+ fault cause, a work-around for lsof is to use its "-P"
+ option, causing lsof to avoid port to service name lookups.
+
+17.21 Will lsof crash the Solaris kernel?
+
+ I've received and investigated one report that it has when
+ the Sun hardware (a QME interface) was faulty. Today (May
+ 23, 2002) I've learned that Sun has reports of kernel
+ crashes caused by adb, lsof, and mdb.
+
+ The Sun investigation pinpointed a problem in the /dev/kmem
+ kernel driver and there is a Sun bug report, 4344513, about
+ the problem. There is a fix in Solaris 9, and patches for
+ Solaris 7 and 8 (SPARC and x86).
+
+ To see if your Solaris system is fixed, look for a
+ /devices/pseudo/*allkmem node.
+
+ Extensive address filtering was added to lsof revision 4.50
+ to forestall what I then (July 2001) believed to be only
+ the possibility that lsof might crash Solaris. However,
+ the filtering isn't perfect, since a filtered address might
+ become invalid after lsof has filtered it but before lsof
+ has delivered it to /dev/kmem. That filtering work is
+ described in .../dialects/sun/solaris_kaddr_filters, also
+ available at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/solaris_kaddr_filters
+
+ The best and safest work-around is to upgrade to Solaris
+ 9 or install an appropriate patch or its equivalent from
+ this list:
+
+ Solaris SPARC x86
+ Version Patch Patch
+ ======= ===== =====
+ 7 106541-20 106542-20
+ 8 108528-14 108529-14
+
+17.22 Why does lsof on Solaris 7, 8, or 9 report a kvm_open()
+ failure?
+
+ When lsof is started on some Solaris 7, 8, and 9 systems
+ it may report:
+
+ lsof: kvm_open(namelist=default, corefile=default): \
+ No such file or directory
+
+ Lsof revisions 4.65 and later will first report:
+
+ lsof: cannot stat /dev/allkmem
+
+ The second message, not delivered in lsof revisions below
+ 4.65, explains the cause of the kvm_open() failure; it
+ can't find /dev/allkmem.
+
+ /dev/allkmem is a device added to Solaris 7 and 8 in patches
+ and in the Solaris 9 FCS. See the preceding "Will lsof
+ crash the Solaris kernel?" section for more information on
+ /dev/allkmem and the patches.
+
+ The kvm_open(3KVM) function in the KVM library of patched
+ Solaris 7 and 8 systems and in Solaris 9 expects to find
+ /dev/allkmem and exits on error when it does not.
+
+ If you have installed the patch that updated your KVM
+ library to a version that expects /dev/allkmem to be present
+ and it is not, you may need to reconfigure your system's
+ devices with devfsadm(1M) or enter "boot -r" to the OpenBoot
+ monitor's prompt (usually "ok").
+
+17.23 Solaris and SAM-FS
+
+17.23.1 Why does Solaris lsof report "(limited SAM-FS info)"?
+
+ Lsof 4.68 and above report "(limited SAM-FS info)" on
+ Solaris in the NAME column after the path or file system
+ name for all files it finds on SAM-FS file systems.
+
+ That's because no more information is known about the
+ composition of the nodes that follow SAM-FS vnodes. If
+ you can provide that information, please contact me via
+ e-mail at <abe@purdue.edu>. Make sure "lsof" appears in the
+ "Subject:" line so my e-mail filter won't classify your letter
+ as Spam.
+
+17.23.2 Why can't lsof locate named SAM-GS files?
+
+ Solaris lsof 4.68 and above can't locate files on SAM-FS
+ file systems when the files are named as lsof arguments
+ because lsof doesn't know how to locate open SAM-FS file
+ device and node number information. (See also 'Why does
+ Solaris lsof report "(limited SAM-FS info)?')
+
+17.24 Lsof and Solaris 10 zones
+
+17.24.1 How can I make lsof list the Solaris zone?
+
+ Use the lsof "-z [z]" option.
+
+17.24.2 Why doesn't lsof work in a Solaris 10 zone?
+
+ When run from within a Solaris 10 zone, lsof will usually
+ report:
+
+ lsof: can't stat(/devices): No such file or directory
+
+ That's because a Solaris zone usually has no /devices
+ subdirectory, a restriction of the zone implementation intended
+ to limit the ability of zone processes to control global system
+ resources, including physical devices.
+
+ While a zone may have a /dev subdirectory, that subdirectory
+ usually lacks the /dev/allkmem, /dev/mem and /dev/kmem devices
+ lsof and the KVM library it uses require.
+
+ The work-around is to run lsof in the global zone. When it is
+ run in a global zone lsof will be able to report on processes
+ running in any zone, including the global zone.
+
+17.24.3 Why does lsof complain it can't stat() Solaris 10 zone file
+ systems?
+
+ When run from the global zone on Solaris 10 lsof may complain:
+
+ lsof: WARNING: can't stat() 15 zone file systems;
+ using dev= options
+
+ The warning message means lsof found the reported number of
+ file system entries in the mount table for which it didn't have
+ permission to get stat(2) results, but which had "zone=" and
+ "dev=" mount table options.
+
+ That is a normal restriction of Solaris 10 zones. Since the
+ lsof warning message indicates it was able to find "dev="
+ options for the file systems, lsof will probably work
+ correctly.
+
+ One work-around is to relax the restrictions on zone mount
+ points, so that lsof can stat() them. While that may be
+ possible by changing directory modes or group ownerships, it is
+ probably not a good idea, because it weakens the restrictions
+ zones are intended to provide.
+
+ Another work-around is to suppress the warning message with
+ lsof's "-w" option. The down side of that is that it causes
+ the suppression of all warning messages, leading to the
+ possibility that some non-stat() warning messages will be
+ suppressed.
+
+17.25 Solaris 10 problems
+
+17.25.1 Why does Solaris 10 lsof sometimes report the wrong path name?
+
+ When a path name component is renamed -- e.g., with mv(1) --
+ Solaris 10 lsof may report the old component for an open file
+ that used the component in its path before the rename. That's
+ because Solaris 10 lsof reports the path name cached in the
+ open file's vnode and the Solaris 10 kernel doesn't update the
+ open vnode's cached path name when a component of it is changed.
+
+ When an open file is deleted -- e.g., with rm(1) -- the path
+ name by which it was opened remains cached in the vnode. Lsof
+ can be instructed to display that path name with the -X option.
+ The path name might be incorrect because of the rename problem
+ described above. See the answer to the 'What does "(deleted)"
+ mean in the NAME column of a Solaris 10 open file?' question
+ for more information.
+
+ Lsof is sometimes able to detect that cached path name is
+ incorrect. In that case lsof may report only the mounted-on
+ directory and device of the file system or it may report that
+ the path name is of questionable accuracy by appending a
+ trailing "(?)" to it in the NAME column.
+
+ See the answer to the "Why does Solaris 10 lsof sometimes
+ report only the mounted-on directory and device?" and 'What
+ does "(?)" mean in the NAME column of a Solaris 10 open file?'
+ questions for more information.
+
+17.25.2 Why does Solaris 10 lsof sometimes report only the mounted-on
+ directory and device?
+
+ For some regular open files lsof may report only the mounted-on
+ directory and device of the file system on which the file
+ resides. That's because lsof was able to determine that the
+ path name cached in the open file's vnode is incorrect.
+
+ Lsof detects the cached path name is incorrect by applying
+ stat(2) to it, provided that no error was detected when stat(2)
+ was applied to the file system mounted-on directory during lsof
+ setup. If a mounted-on directory stat(2) error was detected
+ during setup, lsof does no cached path name analysis and simply
+ reports it.
+
+ When the application of stat(2) to the cached path name returns
+ a no-entry reply (the ENOENT error number), lsof concludes the
+ path no longer exists (i.e., has been unlinked) and reports the
+ mounted-on directory and device of the file system. That
+ behavior can be modified with the -X option in lsof revisions
+ 4.77 and above. See the answer to the 'What does "(deleted)"
+ mean in the NAME column of a Solaris 10 open file?' for more
+ information.
+
+ When the application of stat(2) to the cached path name returns
+ a permission error reply (the EACCES or EPERM error numbers),
+ lsof reports the cached path name and adds a trailing "(?)" to
+ indicate the reported path name is of questionable accuracy.
+ See the answer to the question 'What does "(?)" mean in the
+ NAME column of a Solaris 10 open file?' for more information.
+
+ If the application of stat(2) to the cached path name yields
+ any other error reply, lsof reports the mounted-on directory
+ and device of the file system.
+
+ When the application of stat(2) to the cached path name
+ succeeds, lsof compares the reported device and node numbers to
+ what it has obtained for the open file from kernel structures.
+ If they match, lsof reports the cached path name. If they
+ don't match, lsof instead reports the mounted-on directory and
+ device of the file system.
+
+ A work-around that allows lsof to apply stat(2) successfully to
+ cached path names is to give lsof sufficient permission to do
+ it -- i.e., run lsof as the root user.
+
+17.25.3 What does "(deleted)" mean in the NAME column of a Solaris 10
+ open file?
+
+ When the -X option is specified to Solaris 10 lsof, it will
+ report in its NAME column the path name cached for a deleted
+ file in its vnode. The path name will be followed by
+ "(deleted)".
+
+ Note that the path name cached in a file's vnode is the path
+ name by which the file was opened. It is not updated by the
+ Solaris kernel when any path name component is changed. Hence,
+ it may not represent the final path name the open file had.
+
+ See the answer to the "Why does Solaris 10 lsof sometimes
+ report the wrong path name?" question for more information on
+ how changing a path name component affects the correctness of a
+ what lsof reports.
+
+17.25.4 What does "(?)" mean in the NAME column of a Solaris 10 open
+ file?
+
+ When lsof encounters a path name cached in the open file's
+ vnode that stat(2) reports lsof lacks permission to access,
+ lsof adds "(?)" to the path name reported in the NAME column to
+ indicate the path name is of questionable accuracy.
+
+ See the answers to the "Why does Solaris 10 lsof sometimes
+ report the wrong path name?" and "Why does Solaris 10 lsof
+ sometimes report only the mounted-on directory and device?"
+ questions for more information on why lsof may report a path
+ name of questionable accuracy.
+
+ A work-around that allows lsof to apply stat(2) successfully to
+ cached path names is to give lsof sufficient permission to do
+ it -- i.e., run lsof as the root user.
+
+17.26 Solaris contract file problems
+
+17.26.1 Why doesn't lsof report size, link count and node number for
+ Solaris 10 contract files?
+
+ Lsof doesn't report size, link count or node number for Solaris
+ 10 contract files because I don't know how to obtain them from
+ contract file kernel structures.
+
+17.26.2 Why can't lsof locate a Solaris 10 contract file by path name?
+
+ Because lsof can't find the node number of Solaris contract
+ files, it can't match the device and node numbers it gets from
+ applying stat(2) to the contract file path name with what it
+ finds in kernel data.
+
+17.27 Solaris 10 ZFS probblems
+
+17.27.1 Why does Configure ask for the location of ZFS header files?
+
+ To provide ZFS support when lsof is compiled it needs to have
+ access to the definitions of ZFS structures used by the kernel.
+ Those definitions are contained in header files that Sun does
+ not distribute with Solaris 10.
+
+ When the Configure script detects the system might support ZFS
+ by finding the <sys/fs/zfs.h> header file, it checks for the
+ required ZFS kernel structure definition header files. When
+ they are not found, it asks if ZFS support should be included
+ and if so, with what definitions.
+
+ Two sources can be declared for definitions: 1) lsof's internal
+ ones; or 2) the necessary header files at some location other
+ than /usr/include/sys.
+
+ The necessary header files are:
+
+ dmu.h zfs_acl.h zfs_debug.h zfs_rlock.h zil.h
+ spa.h zfs_context.h zfs_dir.h zfs_vfsops.h zio.h
+ txg.h zfs_ctldir.h zfs_ioctl.h zfs_znode.h zio_impl.h
+
+ The necessary header files might have been obtained, for
+ example, from the http://src.opensolaris.org Open Solaris
+ source site. However, the Open Solaris header files might not
+ be adequate -- see the answer to the "Why don't the Open
+ Solaris ZFS header files provide correct ZFS kernel structure
+ definitions?' question for information on why the Open Solaris
+ ZFS header files might not provide correct ZFS kernel structure
+ definitions.
+
+ The supplied header file path must point to a directory that
+ contains a sys/ subdirectory where the matching header files
+ are located -- e.g., if the supplied path is /tmp/zfs, then
+ /tmp/zfs/sys must contain the necessary header files.
+
+ See the answer to the "Why do -h and -v output warn about
+ possibly inaccurate ZFS kernel structure definitions?" question
+ for information on the risks associated with using lsof's
+ internal definitions.
+
+17.27.2 Why do -h and -v output warn about possibly inaccurate ZFS
+ kernel structure definitions?
+
+ When lsof is configured to use its internal ZFS kernel
+ structure definitions, there is the chance that the definitions
+ may be out of date. Hence, an lsof built with them may report
+ incorrect device numbers, file sizes, node numbers and link
+ counts.
+
+ Please contact me via e-mail at <abe@purdue.edu> for work-around
+ suggestions. Make sure "lsof" appears in the "Subject:" line so
+ my e-mail filter won't classify your letter as Spam.
+
+17.27.3 Why don't the Open Solaris ZFS header files provide correct
+ ZFS kernel structure definitions?
+
+ The Open Solaris ZFS header files may not provide correct ZFS
+ kernel definitions. An lsof built with them may report
+ incorrect device numbers, file sizes, node numbers and link
+ counts.
+
+ The definitions may be incorrect because the Open Solaris
+ header files are not synchronized with any particular Solaris
+ 10 release. Consequently they might contain modifications to
+ the structures not in the structure definitions used to build
+ the running kernel. It may be that only Sun has header files
+ that match the kernel structures of the running Solaris 10
+ kernel.
+
+ Please contact me via e-mail at <abe@purdue.edu> for work-around
+ suggestions. Make sure "lsof" appears in the "Subject:" line so
+ my e-mail filter won't classify your letter as Spam.
+
+
+18.0 Lsof Features
+
+18.1 Why doesn't lsof doesn't report on /proc entries on my
+ system?
+
+ /proc file system support is generally available only for
+ BSD, SYSV R4 dialects, and Tru64 UNIX (Digital UNIX, DEC
+ OSF/1). It's also available for Linux, and Pyramid DC/OSx
+ and Reliant UNIX.
+
+ Even on some SYSV R4 dialects I encountered many problems
+ while trying to incorporate /proc file system support.
+ The chief problem is that some vendors don't distribute
+ the header file that describes the /proc file system node
+ -- usually called prdata.h.
+
+18.2 How do I disable the device cache file feature or alter
+ it's behavior?
+
+ To disable the device cache file feature for a dialect,
+ remove the HASDCACHE definition from the machine.h file of
+ the dialect's machine.h header file. You can also use
+ HASDCACHE to change the default prefix (``.lsof'') of the
+ device cache file.
+
+ Be sure you consider disabling the device cache file feature
+ carefully. Having a device cache file significantly reduces
+ lsof startup overhead by eliminating a full scan of /dev
+ (or /devices) once the device cache file has been created.
+ That full scan also overloads the kernel's name cache with
+ the names of the /dev (or /devices) nodes, reducing the
+ opportunity for lsof to find path name components of open
+ files.
+
+ If you're worried about the presence of mode 0600 device
+ cache files in the home directories of the real user IDs
+ that execute lsof, consider these checks that lsof makes
+ on the file before using it:
+
+ 1. To read the device cache file, lsof must gain
+ permission from access(2).
+
+ 2. The device cache file's modes must be 0600 (0644
+ if lsof is reading a system-wide device cache file)
+ and its size non-zero.
+
+ 3. There must be a correctly formatted section count
+ line at the beginning of the file.
+
+ 4. Each section must have a header line with a count
+ that properly numbers the lines in the section.
+ Legal sections are device, clone, pseudo-device,
+ and CRC.
+
+ 5. The lines of a section must have the proper format.
+
+ 6. All lines are included in a 16 bit CRC, and it is
+ recorded in a non-checksummed section line at the
+ end of the file.
+
+ 7. The checksum computed when the file is read must
+ match the checksum recorded when the file was
+ written.
+
+ 8. The checksum section line must be followed by
+ end-of-information.
+
+ 9. Lsof must be able to get matching results from
+ stat(2) on a randomly chosen entry of the device
+ section.
+
+ For more information on the device cache file, read the
+ 00DCACHE file of the lsof distribution.
+
+18.2.1 What's the risk with a perverted device cache file?
+
+ Even with the checks that lsof makes on the device cache
+ file, it's conceivable that an intruder could modify it so
+ it would pass lsof's tests.
+
+ The only serious consequence I know of this change is the
+ removal of a file whose major device number identifies a
+ socket from some user ID's device cache file. When such
+ a device has been removed from the device cache file, and
+ when lsof doesn't detect the removal, lsof may not be able
+ to identify socket files when executed by the affected user
+ ID. Only certain dialects are at risk to this attack --
+ e.g., SCO OpenServer and Solaris 2.x, 7, 8, and 9.
+
+ If you're tracking a network intruder with lsof, that could
+ be important to you. If you suspect that someone has
+ corrupted the device cache file you're using, I recommend
+ you use lsof's -Di option to tell it to ignore it and use
+ the contents of /dev (or /devices) instead; or remove the
+ device cache file (usually .lsof_hostname, where hostname
+ is the first component of the host's name returned by
+ gethostname(2)) from the user ID's home directory and let
+ lsof create a new one for you.
+
+18.2.2 How do I put the full host name in a personal device cache file
+ path?
+
+ Lsof constructs the personal device cache file path name
+ from a format specified in the HASPERSDC #define in the
+ dialect's machine.h header file. As distributed HASPERSDC
+ declares the path to be ``.lsof_'' plus the first component
+ of the host name with the format ``.lsof_%L''.
+
+ If you want to change the way lsof constructs the personal
+ device cache file path name, you can change the HASPERSDC
+ #define and recompile lsof. If, for example, you #define
+ HASPERSDC to be ``.lsof_%l'' (note the lower case `l'),
+ Configure and remake lsof, then the personal device cache
+ file path will be ``.lsof_'' plus the host name returned
+ by gethostname(2).
+
+ See the 00DCACHE file of the lsof distribution for more
+ information on the formation of the personal device cache
+ file path and the use of the HASPERSDC #define.
+
+18.2.3 How do I put the personal device cache file in /tmp?
+
+ Change the HASPERSDC definition in your dialect's machine.h
+ header file.
+
+ When you redefine HASPERSDC, make sure you put at least
+ one user identification conversion in it to keep separate
+ the device cache files for each user of lsof. Also give
+ some thought to including the ``%0'' conversion to define
+ an alternate path for setuid-root and root processes.
+
+ Here's a definition that puts a personal device cache file
+ in /tmp with the name ``.lsof_login_hostname_pers''.
+
+ #define HASPERSDC "/tmp/.lsof_%u_%l_pers"
+
+ Thus the /tmp personal device cache file path for login
+ "abe" on host "lsof.itap.purdue.edu" would be:
+
+ /tmp/.lsof_abe_lsof.itap.purdue.edu_pers
+
+ You can add the User ID (UID) with the "%U" conversion and
+ the first host name component with the ``%L'' conversion.
+
+ CAUTION: be careful using absolute paths like /tmp lest
+ lsof processes that are setuid-root or whose real UID is
+ root be used to exploit some security weakness via /tmp.
+ Elect instead to add an alternate path for those processes
+ with the ``%0'' conversion. Here's an extension of the
+ previous HASPERSDC format for /tmp that declares an alternate
+ path:
+
+ #define HASPERSDC "/tmp/.lsof_%u_%l_pers%0%h/.lsof_%L"
+
+ When the lsof process is setuid-root or its real UID is
+ root, presuming root's home directory is `/' and the host's
+ name is ``lsof.itap.purdue.edu'', the extended format yields:
+
+ /.lsof_vic
+
+18.3 Why doesn't lsof know about AFS files on my favorite dialect?
+
+ Lsof currently supports AFS for these dialects:
+
+ AIX 4.1.4 (AFS 3.4a)
+ Linux 1.2.13 (AFS 3.3)
+ NEXTSTEP 3.2 (AFS 3.3)
+ Solaris 2.[56] (AFS 3.4a)
+
+ It may recognize AFS files on other versions of these
+ dialects, but I have no way to test that. Lsof may report
+ correct information for AFS files on other dialects, but
+ I can't test that either.
+
+ AFS support must be custom crafted for each UNIX dialect
+ and then tested. If lsof supports your favorite dialect,
+ but doesn't recognize its AFS files, probably I don't have
+ access to a test system. If you want AFS support badly
+ for your dialect, consider helping me do the development
+ and testing.
+
+18.3.1 Why doesn't lsof report node numbers for all AFS volume files,
+ or how do I reveal dynamic module addresses to lsof?
+
+ When AFS is implemented via dynamic kernel modules -- e.g.,
+ in NEXTSTEP -- lsof can't obtain the addresses of AFS
+ variables in the kernel that it uses to identify AFS vnodes.
+ It can guess that a vnode is assigned to an AFS file and
+ it can obtain other information about AFS files, but it
+ has trouble computing AFS volume node numbers.
+
+ To determine node numbers for AFS volumes other than the
+ root volume, /afs, lsof needs access to a hashed volume
+ structure pointer table. When it can't find the address
+ of that table, because AFS support is implemented via
+ dynamic kernel modules, lsof will return blanks in the
+ INODE column for AFS volume files. Lsof can identify the
+ root volume's node number (0), and can compute the node
+ numbers for all other AFS files.
+
+ If you have a name list file that contains the addresses
+ of the AFS dynamic modules -- e.g., you saved module symbols
+ when you created a loadable module kernel with modload(8)
+ by specifying -sym -- lsof may be able to find the kernel
+ addresses it needs in that file.
+
+ Lsof looks up AFS dynamic kernel addresses for these dialects
+ at these default paths:
+
+ NEXTSTEP 3.2 /usr/vice/etc/afs_loadable
+
+ A different path to a name list file with AFS dynamic kernel
+ addresses may be specified with the -A option, when the -A
+ option description appears in lsof's -h or -? (help) output.
+
+ If any addresses appear in the -A name list file that also
+ appear in the regular kernel name list file -- e.g., /vmunix
+ -- they must match, or lsof will silently ignore the -A
+ addresses on the presumption that they are out of date.
--- /dev/null
+
+ The Lsof Mailing List, lsof-l
+
+Information on lsof is available via a GNU Mailman mailing list, named
+lsof-l. The server is located on the host rcac.purdue.edu.
+
+
+Subscribing
+===========
+
+You may subscribe to the lsof-l mailing list by sending e-mail to:
+
+ lsof-l-subscribe@rcac.purdue.edu
+
+The body of your e-mail may be empty. You will receive a confirmation
+reply, explaining one further step you must take to complete your
+subscription.
+
+The list manager uses the e-mail address and real name in the "From:"
+line of your request to set those values in your subscription. If you
+want different values in your subscription, consult the Mailman help
+information to learn how to specify them on your subscription request.
+(See the next "Get Help" section on how to obtain Mailman help
+information.)
+
+
+Get Help
+========
+
+More information about the rcac.purdue.edu GNU Mailman server is
+available by sending e-mail to lsof-l-request@rcac.purdue.edu with
+"help" in the subject line. The body of your e-mail may be empty.
+
+The other information will be delivered by return e-mail.
+
+You can also obtain information on the Mailman e-mail commands in
+section 3.2 of the GNU Mailman documentation at:
+
+ http://www.gnu.org/software/mailman/mailman-member/mailman-member.html
+
+
+The Web Interface
+=================
+
+There is a web interface at:
+
+ https://lists.rcac.purdue.edu/listinfo/lsof-l
+
+You can use it to manage your lsof-l list entry.
+
+
+Posting and Moderation
+======================
+
+Once you have subscribed to lsof-l (and have an e-mail confirmation
+that your subscription was accepted), you may post messages to the list
+by sending e-mail directly to:
+
+ lsof-l@rcac.purdue.edu
+
+I moderate the lsof-l mailing list and try to keep its traffic low,
+mainly limiting it to announcements of new revisions, patches and
+security issues. Postings don't appear until I've approved them.
+
+
+Send Bug Reports to Me Via E-Mail
+=================================
+
+DON'T SEND BUG REPORTS TO lsof-l. Send them directly to me via e-mail
+at <abe@purdue.edu>. Make sure lsof appears in the "Subject:" line and
+make sure you first read the "Bug Reports" section of the 00README file
+of the lsof distribution.
+
+
+Unsubscribing
+=============
+
+You can unsubscribe from lsof-l by sending e-mail to:
+
+ lsof-l-unsubscribe@rcac.purdue.edu
+
+The body of your e-mail may be empty. You will receive a confirmation
+reply, explaining one further step you must take to complete the
+removal of your subscription.
+
+
+Archive
+=======
+
+There is an archive; use the link:
+
+ https://lists.rcac.purdue.edu/listinfo/lsof-l
+
+The archive link is the first one on the web page. You will need the
+password you received or set when you subscribed, or later set via
+lsof-l-request or the web interface.
+
+
+Vic Abell <abe@purdue.edu>
+May 8, 2008
--- /dev/null
+.:
+00.README.FIRST
+00CREDITS
+00DCACHE
+00DIALECTS
+00DIST
+00FAQ
+00LSOF-L
+00MANIFEST
+00PORTING
+00QUICKSTART
+00README
+00TEST
+00XCONFIG
+AFSConfig*
+Configure*
+Customize*
+Inventory*
+arg.c
+dialects/
+lib/
+lsof.8
+lsof.h
+lsof.man
+lsof_fields.h
+main.c
+misc.c
+node.c
+print.c
+proc.c
+proto.h
+regex.h
+scripts/
+store.c
+tests/
+usage.c
+util.c
+version
+
+./dialects:
+aix/
+darwin/
+du/
+freebsd/
+hpux/
+linux/
+n+obsd/
+n+os/
+osr/
+sun/
+uw/
+
+./dialects/aix:
+Makefile
+Mksrc*
+aix5/
+ddev.c
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dnode1.c
+dnode2.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/aix/aix5:
+README
+j2/
+
+./dialects/aix/aix5/j2:
+j2_lock.h
+private_j2_snapshot.h
+
+./dialects/darwin:
+get-hdr-loc.sh*
+kmem/
+libproc/
+
+./dialects/darwin/kmem:
+Makefile
+Mksrc*
+ddev.c
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dnode1.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/darwin/libproc:
+Makefile
+Mksrc*
+ddev.c
+dfile.c
+dlsof.h
+dmnt.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/du:
+Makefile
+Mksrc*
+ddev.c
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/freebsd:
+Makefile
+Makefile.zfs
+Mksrc*
+dlsof.h
+dmnt.c
+dnode.c
+dnode1.c
+dnode2.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+dzfs.h
+include/
+machine.h
+
+./dialects/freebsd/include:
+procfs/
+
+./dialects/freebsd/include/procfs:
+pfsnode.h
+
+./dialects/hpux:
+kmem/
+pstat/
+
+./dialects/hpux/kmem:
+Makefile
+Mksrc*
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dnode1.c
+dnode2.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+hpux11/
+machine.h
+
+./dialects/hpux/kmem/hpux11:
+ipc_s.h
+kernbits.h
+lla.h
+nfs_clnt.h
+proc.h
+rnode.h
+sth.h
+tcp_s.h
+udp_s.h
+vnode.h
+
+./dialects/hpux/pstat:
+Makefile
+Mksrc*
+dfile.c
+dlsof.h
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/linux:
+Makefile
+Mksrc*
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/n+obsd:
+Makefile
+Mksrc*
+dlsof.h
+dmnt.c
+dnode.c
+dnode1.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/n+os:
+Makefile
+Mksrc*
+dlsof.h
+dnode.c
+dnode1.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+
+./dialects/osr:
+Makefile
+Mksrc*
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+include/
+machine.h
+
+./dialects/osr/include:
+netdb.h
+sys/
+
+./dialects/osr/include/sys:
+cdefs.h
+
+./dialects/sun:
+Makefile
+Mksrc*
+ddev.c
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dnode1.c
+dnode2.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+get-hdr-loc.sh*
+machine.h
+solaris_kaddr_filters
+zfs/
+
+./dialects/sun/zfs:
+sys/
+zfs_vnops.c
+
+./dialects/sun/zfs/sys:
+dmu.h
+spa.h
+txg.h
+zfs_acl.h
+zfs_context.h
+zfs_ctldir.h
+zfs_debug.h
+zfs_dir.h
+zfs_ioctl.h
+zfs_rlock.h
+zfs_vfsops.h
+zfs_znode.h
+zil.h
+zio.h
+zio_impl.h
+
+./dialects/uw:
+Makefile
+Mksrc*
+dfile.c
+dlsof.h
+dmnt.c
+dnode.c
+dnode1.c
+dnode2.c
+dnode3.c
+dproc.c
+dproto.h
+dsock.c
+dstore.c
+machine.h
+uw7/
+
+./dialects/uw/uw7:
+README
+fs/
+sys/
+vm/
+
+./dialects/uw/uw7/fs:
+nsc_cfs/
+procfs/
+
+./dialects/uw/uw7/fs/nsc_cfs:
+cnode.h
+
+./dialects/uw/uw7/fs/procfs:
+README
+prdata.h
+
+./dialects/uw/uw7/sys:
+fs/
+
+./dialects/uw/uw7/sys/fs:
+README
+fifonode.h
+namenode.h
+
+./dialects/uw/uw7/vm:
+
+./lib:
+Makefile.skel
+ckkv.c
+cvfs.c
+dvch.c
+fino.c
+isfn.c
+lkud.c
+pdvn.c
+prfp.c
+ptti.c
+rdev.c
+regex.c
+rmnt.c
+rnam.c
+rnch.c
+rnmh.c
+snpf.c
+
+./scripts:
+00MANIFEST
+00README
+big_brother.perl5*
+count_pf.perl*
+count_pf.perl5*
+identd.perl5*
+idrlogin.perl*
+idrlogin.perl5*
+list_NULf.perl5*
+list_fields.awk
+list_fields.perl*
+shared.perl5*
+sort_res.perl5*
+watch_a_file.perl*
+xusers.awk*
+
+./tests:
+00README
+Add2TestDB*
+CkTestDB*
+LTbasic.c
+LTbigf.c
+LTdnlc.c
+LTlib.c
+LTlock.c
+LTnfs.c
+LTnlink.c
+LTsock.c
+LTszoff.c
+LTunix.c
+LsofTest.h
+Makefile
+TestDB
--- /dev/null
+
+ Guide to Porting lsof 4 to Unix OS Dialects
+
+**********************************************************************
+| The latest release of lsof is always available via anonymous ftp |
+| from lsof.itap.purdue.edu. Look in pub/lsof.README for its |
+| location. |
+**********************************************************************
+
+ Contents
+
+ How Lsof Works
+ /proc-based Linux Lsof -- a Different Approach
+ General Guidelines
+ Organization
+ Source File Naming Conventions
+ Coding Philosophies
+ Data Requirements
+ Dlsof.h and #include's
+ Definitions That Affect Compilation
+ Options: Common and Special
+ Defining Dialect-Specific Symbols and Global Storage
+ Coding Dialect-specific Functions
+ Function Prototype Definitions and the _PROTOTYPE Macro
+ The Makefile
+ The Mksrc Shell Script
+ The MkKernOpts Shell Script
+ Testing and the lsof Test Suite
+ Where Next?
+
+
+How Lsof Works
+--------------
+
+Before getting on with porting guidelines, just a word or two about
+how lsof works.
+
+Lsof obtains data about open UNIX dialect files by reading the
+kernel's proc structure information, following it to the related
+user structure, then reading the open file structures stored
+(usually) in the user structure. Typically lsof uses the kernel
+memory devices, /dev/kmem, /dev/mem, etc. to read kernel data.
+
+Lsof stores information from the proc and user structures in an
+internal, local proc structure table. It then processes the open
+file structures by reading the file system nodes that lie behind
+them, extracting and storing relevant data in internal local file
+structures that are linked to the internal local process structure.
+
+Once all data has been gathered, lsof reports it from its internal,
+local tables.
+
+There are a few variants on this subject. Some systems don't have
+just proc structures, but have task structures, too, (e.g., NeXTSTEP
+and OSF/1 derivatives). For some dialects lsof gets proc structures
+or process information (See "/proc-based Linux Lsof -- a Different
+Approach) from files of the /proc file system. It's not necessary
+for lsof to read user structures on some systems (recent versions
+of HP-UX), because the data lsof needs can be found in the task or
+proc structures. In the end lsof gathers the same data, just from
+slightly different sources.
+
+
+/proc-based Linux Lsof -- a Different Approach
+==============================================
+
+For a completely different approach to lsof construction, take a
+look at the /proc-based Linux sources in .../dialects/linux/proc.
+(The sources in .../dialects/linux/kmem are for a traditional lsof
+that uses /dev/kmem to read information from kernel structures.)
+
+The /proc-based lsof obtains all its information from the Linux
+/proc file system. Consequently, it is relatively immune to changes
+in Linux kernel structures and doesn't need to be re-compiled each
+time the Linux kernel version changes.
+
+There are some down-sides to the Linux /proc-based lsof:
+
+ * It must run setuid-root in order to be able to read the
+ /proc file system branches for all processes. In contrast,
+ the /dev/kmem-based Linux lsof usually needs only setgid
+ permission.
+
+ * It depends on the exact character format of /proc files, so
+ it is sensitive to changes in /proc file composition.
+
+ * It is limited to the information a /proc file system
+ implementor decides to provide. For example, if a
+ /proc/net/<protocol> file lacks an inode number, the
+ /proc-based lsof can't connect open socket files to that
+ protocol. Another deficiency is that the /proc-based may
+ not be able to report file offset (position) information,
+ when it isn't available in the /proc/<PID>/fd/ entry for a
+ file.
+
+ In contrast the /dev/kmem-based lsof has full access to
+ kernel structures and "sees" new data as soon as it appears.
+ Of course, that new data requires that lsof be recompiled
+ and usually also requires changes to lsof.
+
+Overall the switch from a /dev/kmem base to a /proc one is an
+advantage to Linux lsof. The switch was made at lsof revision 4.23
+for Linux kernel versions 2.1.72 (approximately) and higher. The
+reason I'm not certain at which Linux kernel version a /proc-based
+lsof becomes possible is that the /proc additions needed to implement
+it have been added gradually to Linux 2.1.x in ways that I cannot
+measure.
+
+/proc-based lsof functions in many ways the same as /dev/kmem-based
+lsof. It scans the /proc directory, looking for <PID>/ subdirectories.
+Inside each one it collects process-related data from the cwd, exe,
+maps, root, and stat information files.
+
+It collects open file information from the fd/ subdirectory of each
+<PID>/ subdirectory. The lstat(2), readlink(2), and stat(2) system
+calls gather information about the files from the kernel.
+
+Lock information comes from /proc/locks. It is matched to open
+files by inode number. Mount information comes from /proc/mounts.
+Per domain protocol information comes from the files of /proc/net;
+it's matched to open socket files by inode number.
+
+The Linux /proc file system implementors have done an amazing job
+of providing the information lsof needs. The /proc-based lsof
+project has so far generated only two kernel modification:
+
+ * A modification to /usr/src/linux/net/ipx/af_ipx.c adds the
+ inode number to the entries of /proc/net/ipx.
+
+ Jonathan Sergent did this kernel modification.
+
+ It may be found in the .../dialects/linux/proc/patches
+ subdirectory of the lsof distribution.
+
+ * An experimental modification to /usr/src/linux/fs/stat.c
+ allows lstat(2) to return file position information for
+ /proc/<PID>/fd/<FD> files.
+
+ Contact me for this modification.
+
+
+One final note about the /proc-based Linux lsof: it doesn't need
+any functions from the lsof library in the lib/ subdirectory.
+
+
+General Guidelines
+------------------
+
+These are the general guidelines for porting lsof 4 to a new Unix
+dialect:
+
+ * Understand the organization of the lsof sources and the
+ philosophies that guide their coding.
+
+ * Understand the data requirements and determine the methods
+ of locating the necessary data in the new dialect's kernel.
+
+ * Pick a name for the subdirectory in lsof4/dialects for your
+ dialect. Generally I use a vendor operating system name
+ abbreviation.
+
+ * Locate the necessary header files and #include them in the
+ dialect's dlsof.h file. (You may not be able to complete
+ this step until you have coded all dialect-specific functions.)
+
+ * Determine the optional library functions of lsof to be used
+ and set their definitions in the dialect's machine.h file.
+
+ * Define the dialect's specific symbols and global storage
+ in the dialect's dlsof.h and dstore.c files.
+
+ * Code the dialect-specific functions in the appropriate
+ source files of the dialect's subdirectory.
+
+ Include the necessary prototype definitions of the dialect-
+ specific functions in the dproto.h file in the dialect's
+ subdirectory.
+
+ * Define the dialect's Makefile and source construction shell
+ script, Mksrc.
+
+ * If there are #define's that affect how kernel structures
+ are organized, and those #define's are needed when compiling
+ lsof, build a MkKernOpts shell script to locate the #define's
+ and supply them to the Configure shell script.
+
+
+Organization
+------------
+
+The code in a dialect-specific version of lsof comes from three
+sources:
+
+ 1) functions common to all versions, located in the top level
+ directory, lsof4;
+
+ 2) functions specific to the dialect, located in the dialect's
+ subdirectory -- e.g., lsof4/dialects/sun;
+
+ 3) functions that are common to several dialects, although
+ not to all, organized in a library, liblsof.a. The functions
+ in the library source can be selected and customized with
+ definitions in the dialect machine.h header files.
+
+The tree looks like this:
+
+ lsof4 ----------------------+ 3) library --
+ | \ lsof4/lib
+ 1) fully common functions + \
+ e.g., lsof4/main.c + lsof4/dialects/
+ / / / / \
+ + + + + +
+ 2) dialect-specific subdirectories -- e.g., lsof4/dialects/sun
+
+The code for a dialect-specific version is constructed from these
+three sources by the Configure shell script in the top level lsof4
+directory and definitions in the dialect machine.h header files.
+Configure uses the Mksrc shell script in each dialect's subdirectory,
+and may use an optional MkKernOpts shell script in selected dialect
+subdirectories.
+
+Configure calls the Mksrc shell script in each dialect's subdirectory
+to assemble the dialect-specific sources in the main lsof directory.
+Configure may call MkKernOpts to determine kernel compile-time
+options that are needed for compiling kernel structures correctly
+for use by lsof. Configure puts the options in a dialect-specific
+Makefile it build, using a template in the dialect subdirectory.
+
+The assembly of dialect-specific sources in the main lsof directory
+is usually done by creating symbolic links from the top level to
+the dialect's subdirectory. The LSOF_MKC environment variable may
+be defined prior to using Configure to change the technique used
+to assemble the sources -- most commonly to use cp instead of ln -s.
+
+The Configure script completes the dialect's Makefile by adding
+string definitions, including the necessary kernel compile-time
+options, to a dialect skeleton Makefile while copying it from the
+dialect subdirectory to the top level lsof4 directory. Optionally
+Makefile may call the dialect's MkKernOpts script to add string
+definitions.
+
+When the lsof library, lsof4/lib/liblsof.a, is compiled its
+functions are selected and customized by #define's in the dialect
+machine.h header file.
+
+
+Source File Naming Conventions
+------------------------------
+
+With one exception, dialect-specific source files begin with a
+lower case `d' character -- ddev.c, dfile.c, dlsof.h. The one
+exception is the header file that contains dialect-specific
+definitions for the optional features of the common functions.
+It's called machine.h for historical reasons.
+
+Currently all dialects use almost the same source file names. One
+exception to the rule happens in dialects where there must be
+different source files -- e.g., dnode[123].c -- to eliminate node
+header file structure element name conflicts. The source modules
+in a few subdirectories are organized that way.
+
+Unusual situations occur for NetBSD and OpenBSD, and for NEXTSTEP
+and OPENSTEP. Each pair of dialects is so close in design that
+the same dialect sources from the n+obsd subdirectory serves NetBSD
+and OpenBSD; from n+os, NEXTSTEP and OPENSTEP.
+
+These are common files in lsof4/:
+
+ Configure the configuration script
+
+ Customize does some customization of the selected lsof
+ dialect
+
+ Inventory takes an inventory of the files in an lsof
+ distribution
+
+ version the version number
+
+ dialects/ the dialects subdirectory
+
+These are the common function source files in lsof4/:
+
+ arg.c common argument processing functions
+
+ lsof.h common header file that #include's the dialect-specific
+ header files
+
+ main.c common main function for lsof 4
+
+ misc.c common miscellaneous functions -- e.g., special versions
+ of stat() and readlink()
+
+ node.c common node reading functions -- readinode(), readvnode()
+
+ print.c common print support functions
+
+ proc.c common process and file structure functions
+
+ proto.h common prototype definitions, including the definition of
+ the _PROTOTYPE() macro
+
+ store.c common global storage version.h the current lsof version
+ number, derived from the file version by the Makefile
+
+ usage.c functions to display lsof usage panel
+
+These are the dialect-specific files:
+
+ Makefile the Makefile skeleton
+
+ Mksrc a shell script that assists the Configure script
+ in configuring dialect sources
+
+ MkKernOpts an optional shell script that identifies kernel
+ compile-time options for selected dialects -- e.g.,
+ Pyramid DC/OSx and Reliant UNIX
+
+ ddev.c device support functions -- readdev() -- may be
+ eliminated by functions from lsof4/lib/
+
+ dfile.c file processing functions -- may be eliminated by
+ functions from lsof4/lib/
+
+ dlsof.h dialect-specific header file -- contains #include's
+ for system header files and dialect-specific global
+ storage declarations
+
+ dmnt.c mount support functions -- may be eliminated by
+ functions from lsof4/lib/
+
+ dnode.c node processing functions -- e.g., for gnode or vnode
+
+ dnode?.c additional node processing functions, used when node
+ header files have duplicate and conflicting element
+ names.
+
+ dproc.c functions to access, read, examine and cache data about
+ dialect-specific process structures -- this file contains
+ the dialect-specific "main" function, gather_proc_info()
+
+ dproto.h dialect-specific prototype declarations
+
+ dsock.c dialect-specific socket processing functions
+
+ dstore.c dialect-specific global storage -- e.g., the nlist()
+ structure
+
+ machine.h dialect specific definitions of common function options --
+ e.g., a HASINODE definition to activate the readinode()
+ function in lsof4/node.c
+
+ The machine.h header file also selects and customizes
+ the functions of lsof4/lib/.
+
+These are the lib/ files. Definitions in the dialect machine.h
+header files select and customize the contained functions that are
+to be compiled and archived to liblsof.a.
+
+ Makefile.skel is a skeleton Makefile, used by Configure
+ to construct the Makefile for the lsof
+ library.
+
+ cvfs.c completevfs() function
+
+ USE_LIB_COMPLETEVFS selects it.
+
+ CVFS_DEVSAVE, CVFS_NLKSAVE, CVFS_SZSAVE,
+ and HASFSINO customize it.
+
+ dvch.c device cache functions
+
+ HASDCACHE selects them.
+
+ DCACHE_CLONE, DCACHE_CLR, DCACHE_PSEUDO,
+ DVCH_CHOWN, DVCH_DEVPATH, DVCH_EXPDEV,
+ HASBLKDEV, HASENVDC, HASSYSDC, HASPERSDC,
+ HASPERSDCPATH, and NOWARNBLKDEV customize
+ them.
+
+ fino.c find block and character device inode functions
+
+ HASBLKDEV and USE_LIB_FIND_CH_INO select them.
+
+ isfn.c hashSfile() and is_file_named() functions
+
+ USE_LIB_IS_FILE_NAMED selects it.
+
+ lkud.c device lookup functions
+
+ HASBLKDEV and USE_LIB_LKUPDEV select them.
+
+ pdvn.c print device name functions
+
+ HASBLKDEV and USE_LIB_PRINTDEVNAME select them.
+
+ prfp.c process_file() function
+
+ USE_LIB_PROCESS_FILE selects it.
+
+ FILEPTR, DTYPE_PIPE, HASPIPEFN, DTYPE_GNODE,
+ DTYPE_INODE, DTYPE_PORT, DTYPE_VNODE,
+ HASF_VNODE, HASKQUEUE, HASPRIVFILETYPE,
+ HASPSXSHM and HASPSXSEM customize it.
+
+ ptti.c print_tcptpi() function
+
+ USE_LIB_PRINT_TCPTPI selects it.
+
+ HASSOOPT, HASSBSTATE, HASSOSTATE, AHSTCPOPT,
+ HASTCPTPIQ and HASTCPTPIW customize it.
+
+ rdev.c readdev() function
+
+ USE_LIB_READDEV selects it.
+
+ DIRTYPE, HASBLKDEV, HASDCACHE, HASDNAMLEN,
+ RDEV_EXPDEV, RDEV_STATFN, USE_STAT, and
+ WARNDEVACCESS customize it.
+
+ rmnt.c readmnt() function
+
+ USE_LIB_READMNT selects it.
+
+ HASFSTYPE, MNTSKIP, RMNT_EXPDEV, RMNT_FSTYPE,
+ and MOUNTS_FSTYPE customize it.
+
+ rnam.c BSD format name cache functions
+
+ HASNCACHE and USE_LIB_RNAM select them.
+
+ HASFSINO, NCACHE, NCACHE_NC_CAST, NCACHE_NM,
+ NCACHE_NMLEN, NCACHE_NODEADDR, NCACHE_NODEID,
+ NCACHE_NO_ROOT, NCACHE_NXT, NCACHE_PARADDR,
+ NCACHE_PARID, NCACHE_SZ_CAST, NCHNAMLEN,
+ X_NCACHE, and X_NCSIZE, customize them.
+
+ rnch.c Sun format name cache functions
+
+ HASNCACHE and USE_LIB_RNCH select them.
+
+ ADDR_NCACHE, HASDNLCPTR, HASFSINO, NCACHE_DP,
+ NCACHE_NAME, NCACHE_NAMLEN, NCACHE_NEGVN,
+ NCACHE_NODEID, NCACHE_NXT, NCACHE_PARID,
+ NCACHE_VP, X_NCACHE, and X_NCSIZE, customize
+ them.
+
+ snpf.c Source for the snprintf() family of functions
+
+ USE_LIB_SNPF selects it.
+
+
+The comments and the source code in these library files give more
+information on customization.
+
+
+Coding Philosophies
+-------------------
+
+A few basic philosophies govern the coding of lsof 4 functions:
+
+ * Use as few #if/#else/#endif constructs as possible, even at
+ the cost of nearly-duplicate code.
+
+ When #if/#else/#endif constructs are necessary:
+
+ o Use the form
+
+ #if defined(s<symbol>)
+
+ in preference to
+
+ #ifdef <symbol>
+
+ to allow easier addition of tests to the #if.
+
+ o Indent them to signify their level -- e.g.,
+
+ #if /* level one */
+ # if /* level two */
+ # endif /* level two */
+ #else /* level one */
+ #endif /* level one */
+
+ o Use ANSI standard comments on #else and #endif statements.
+
+ * Document copiously.
+
+ * Aim for ANSI-C compatibility:
+
+ o Use function prototypes for all functions, hiding them
+ from compilers that cannot handle them with the _PROTOTYPE()
+ macro.
+
+ o Use the compiler's ANSI conformance checking wherever
+ possible -- e.g., gcc's -ansi option.
+
+
+Data Requirements
+-----------------
+
+Lsof's strategy in obtaining open file information is to access
+the process table via its proc structures, then obtain the associated
+user area and open file structures. The open file structures then
+lead lsof to file type specific structures -- cdrnodes, fifonodes,
+inodes, gnodes, hsfsnodes, pipenodes, pcnodes, rnodes, snodes,
+sockets, tmpnodes, and vnodes.
+
+The specific node structures must yield data about the open files. The
+most important items and device number (raw and cooked) and node
+number. (Lsof uses them to identify files and file systems named as
+arguments.) Link counts and file sizes are important, too, as are the
+special characteristics of sockets, pipes, FIFOs, etc.
+
+This means that to begin an lsof port to a new Unix dialect you
+must understand how to obtain these structures from the dialect's
+kernel. Look for kernel access functions -- e.g., the AIX readx()
+function, Sun and Sun-like kvm_*() functions, or SGI's syssgi()
+function. Look for clues in header files -- e.g. external declarations
+and macros.
+
+If you have access to them, look at sources to programs like ps(1),
+or the freely available monitor and top programs. They may give
+you important clues on reading proc and user area structures. An
+appeal to readers of dialect-specific news groups may uncover
+correspondents who can help.
+
+Careful reading of system header files -- e.g., <sys/proc.h> --
+may give hints about how kernel storage is organized. Look for
+global variables declared under a KERNEL or _KERNEL #if. Run nm(1)
+across the kernel image (/vmunix, /unix, etc.) and look for references
+to structures of interest.
+
+Even if there are support functions for reading structures, like the
+kvm_*() functions, you must still understand how to read data from
+kernel memory. Typically this requires an understanding of the
+nlist() function, and how to use /dev/kmem, /dev/mem, and /dev/swap.
+
+Don't overlook the possibility that you may have to use the process
+file system -- e.g., /proc. I try to avoid using /proc when I can,
+since it usually requires that lsof have setuid(root) permission
+to read the individual /proc "files".
+
+Once you can access kernel structures, you must understand how
+they're connected. You must answer questions like:
+
+ * How big are kernel addresses? How are they type cast?
+
+ * How are kernel variable names converted to addresses?
+ Nlist()?
+
+ * How are the proc structures organized? Is it a static
+ table? Are the proc structures linked? Is there a
+ kernel pointer to the first proc structure? Is there a
+ proc structure count?
+
+ * How does one obtain copies of the proc structures? Via
+ /dev/kmem? Via a vendor API?
+
+ * If this is a Mach derivative, is it necessary to obtain the
+ task and thread structures? How?
+
+ * How does one obtain the user area (or the utask area in Mach
+ systems) that corresponds to a process?
+
+ * Where are the file structures located for open file
+ descriptors and how are they located? Are all file
+ structures in the user area? Is the file structure space
+ extensible?
+
+ * Where do the private data pointers in file structures lead?
+ To gnodes? To inodes? To sockets? To vnodes? Hint: look
+ in <sys/file.h> for DTYPE_* instances and further pointers.
+
+ * How are the nodes organized? To what other nodes do they
+ lead and how? Where are the common bits of information in
+ nodes -- device, node number, size -- stored? Hint: look
+ in the header files for nodes for macros that may be used
+ to obtain the address of one node from another -- e.g., the
+ VTOI() macro that leads from a vnode to an inode.
+
+ * Are text reference nodes identified and how? Is it
+ necessary to examine the virtual memory map of a process or
+ a task to locate text references? Some kernels have text
+ node pointers in the proc structures; some, in the user
+ area; Mach kernels may have text information in the task
+ structure, reached in various ways from the proc, user area,
+ or user task structure.
+
+ * How is the device table -- e.g., /dev or /devices --
+ organized? How is it read? Using direct or dirent structures?
+
+ How are major/minor device numbers represented? How are
+ device numbers assembled and disassembled?
+
+ Are there clone devices? How are they identified?
+
+ * How is mount information obtained? Getmntinfo()? Getmntent()?
+ Some special kernel call?
+
+ * How are sockets identified and organized? BSD-style? As
+ streams? Are there streams?
+
+ * Are there special nodes -- CD-ROM nodes, FIFO nodes, etc.?
+
+ * How is the kernel's name cache organized? Can lsof access
+ it to get partial name components?
+
+
+Dlsof.h and #include's
+----------------------
+
+Once you have identified the kernel's data organization and know
+what structures it provides, you must add #include's to dlsof.h to
+access their definitions. Sometimes it is difficult to locate the
+header files -- you may need to introduce -I specifications in the
+Makefile via the DINC shell variable in the Configure script.
+
+Sometimes it is necessary to define special symbols -- e.g., KERNEL,
+_KERNEL, _KMEMUSER -- to induce system header files to yield kernel
+structure definitions. Sometimes making those symbol definitions
+cause other header file and definition conflicts. There's no good
+general rule on how to proceed when conflicts occur.
+
+Rarely it may be necessary to extract structure definitions from
+system header files and move them to dlsof.h, create special versions
+of system header files, or obtain special copies of system header
+files from "friendly" (e.g., vendor) sources. The dlsof.h header
+file in lsof4/dialects/sun shows examples of the first case; the
+second, no examples; the third, the irix5hdr subdirectory in
+lsof4/dialects/irix (a mixture of the first and third).
+
+Building up the necessary #includes in dlsof.h is an iterative
+process that requires attention as you build the dialect-specific
+functions that references kernel structures. Be prepared to revisit
+dlsof.h frequently.
+
+
+Definitions That Affect Compilation
+-----------------------------------
+
+The source files at the top level and in the lib/ subdirectory
+contain optional functions that may be activated with definitions
+in a dialect's machine.h header file. Some are functions for
+reading node structures that may not apply to all dialects -- e.g.
+CD-ROM nodes (cdrnode), or `G' nodes (gnode) -- and others are
+common functions that may occasionally be replaced by dialect-specific
+ones. Once you understand your kernel's data organization, you'll
+be able to decide the optional common node functions to activate.
+
+Definitions in machine.h and dlsof.h also enable or disable other
+optional common features. The following is an attempt to list all
+the definitions that affect lsof code, but CAUTION, it is only
+attempt and may be incomplete. Always check lsof4 source code in
+lib/ and dialects/, and dialect machine.h header files for other
+possibilities
+
+ AFS_VICE See 00XCONFIG.
+
+ AIX_KERNBITS specifies the kernel bit size, 32 or 64, of the Power
+ architecture AIX 5.x kernel for which lsof was built.
+
+ CAN_USE_CLNT_CREATE is defined for dialects where the more modern
+ RPC function clnt_create() can be used in
+ place of the deprecated clnttcp_create().
+
+ CLONEMAJ defines the name of the variable that
+ contains the clone major device number.
+ (Also see HAS_STD_CLONE and HAVECLONEMAJ.)
+
+ DEVDEV_PATH defines the path to the directory where device
+ nodes are stored, usually /dev. Solaris 10
+ uses /devices.
+
+ DIALECT_WARNING may be defined by a dialect to provide a
+ warning message that will be displayed with
+ help (-h) and version (-v) output.
+
+ FSV_DEFAULT defines the default file structure values to
+ list. It may be composed of or'd FSV_*
+ (See lsof.h) values. The default is none (0).
+
+ GET_MAJ_DEV is a macro to get major portion from device
+ number instead of via the standard major()
+ macro.
+
+ GET_MIN_DEV is a macro to get minor portion from device
+ number instead of via the standard minor()
+ macro.
+
+ GET_MAX_FD the name of the function that returns an
+ int for the maximum open file descriptor
+ plus one. If not defined, defaults to
+ getdtablesize.
+
+ HAS9660FS enables CD9660 file system support in a
+ BSD dialect.
+
+ HAS_ADVLOCK_ARGS is defined for NetBSD and OpenBSD dialects
+ whose <sys/lockf.h> references vop_advlock_args.
+
+ HAS_AFS enables AFS support code for the dialect.
+
+ HAS_ATOMIC_T indicates the Linux version has an
+ <asm/atomic.h> header file and it contains
+ "typedef struct .* atomic_t;"
+
+ HASAOPT indicates the dialect supports the AFS -A
+ option when HAS_AFS is also defined.
+
+ HAS_ASM_TERMIOBITS indicates for Linux Alpha that the
+ <asm/termiobits.h> header file exists.
+
+ HASAX25CBPTR indicates that the Linux sock struct has an
+ ax25_db pointer.
+
+ HASBLKDEV indicates the dialect has block device support.
+
+ HASBUFQ_H indicates the *NSD dialect has the <sys/bufq.h>
+ header file.
+
+ HASCACHEFS enables cache file system support for the
+ dialect.
+
+ HAS_CDFS enables CDFS file system support for the
+ dialect.
+
+ HASCDRNODE enables/disables readcdrnode() in node.c.
+
+ HAS_CONST indicates that the compiler supports the
+ const keyword.
+
+ HASCPUMASK_T indicates the FreeBSD 5.2 or higher dialect
+ has cpumask_t typedef's.
+
+ HAS_CRED_IMPL_H indicates the Solaris 10 dialect has the
+ <sys/cred_impl.h> header file available.
+
+ HASCWDINFO indicates the cwdinfo structure is defined
+ in the NetBSD <sys/filedesc.h>.
+
+ HASDCACHE enables device file cache file support.
+ The device cache file contains information
+ about the names, device numbers and inode
+ numbers of entries in the /dev (or /device)
+ node subtree that lsof saves from call to
+ call. See the 00DCACHE file of the lsof
+ distribution for more information on this
+ feature.
+
+ HASDENTRY indicates the Linux version has a dentry
+ struct defined in <linux/dcache.h>.
+
+ HASDEVKNC indicates the Linux version has a kernel
+ name cached keyed on device number.
+
+ HAS_DINODE_U indicates the OpenBSD version has a dinode_u
+ union in its inode structure.
+
+ HASDNLCPTR is defined when the name cache entry of
+ <sys/dnlc.h> has a name character pointer
+ rather than a name character array.
+
+ HASEFFNLINK indicates the *BSD system has the i_effnlink
+ member in the inode structure.
+
+ HASENVDC enables the use of an environment-defined
+ device cache file path and defines the name
+ of the environment variable from which lsof
+ may take it. (See the 00DCACHE file of
+ the lsof distribution for information on
+ when HASENVDC is used or ignored.)
+
+ HASEXT2FS is defined for BSD dialects for which ext2fs
+ file system support can be provided. A value
+ of 1 indicates that the i_e2din member does not
+ exist; 2, it exists.
+
+ HASF_VNODE indicates the dialect's file structure has an
+ f_vnode member in it.
+
+ HASFDESCFS enables file descriptor file system support
+ for the dialect. A value of 1 indicates
+ <miscfs/fdesc.h> has a Fctty definition; 2,
+ it does not.
+
+ HASFDLINK indicates the file descriptor file system
+ node has the fd_link member.
+
+ HASFIFONODE enables/disables readfifonode() in node.c.
+
+ HAS_FL_FD indicates the Linux version has an fl_fd
+ element in the lock structure of <linux/fs.h>.
+
+ HAS_FL_FILE indicates the Linux version has an fl_file
+ element in the lock structure of <linux/fs.h>.
+
+ HAS_FL_WHENCE indicates the Linux version has an fl_whence
+ element in the lock structure of <linux/fs.h>.
+
+ HAS_F_OPEN indicates the UnixWare 7.x dialect has the
+ f_open member in its file struct.
+
+ HASFSINO enables the inclusion of the fs_ino element
+ in the lfile structure definition in lsof.h.
+ This contains the file system's inode number
+ and may be needed when searching the kernel
+ name cache. See dialects/osr/dproc.c for
+ an example.
+
+ HAS_JFS2 The AIX >= 5.0 dialect has jfs2 support.
+
+ HASFSTRUCT indicates the dialect has a file structure
+ the listing of whose element values can be
+ enabled with +f[cfn]. FSV_DEFAULT defines
+ the default listing values.
+
+ HASFSTYPE enables/disables the use of the file system's
+ stat(2) st_fstype member.
+
+ If the HASFSTYPE value is 1, st_fstype is
+ treated as a character array; 2, it is
+ treated as an integer.
+
+ See also the RMNT_EXPDEV and RMNT_FSTYPE
+ documentation in lib/rmnt.c
+
+ HASGETBOOTFILE indicates the NetBSD or OpenBSD dialect has
+ a getbootfile() function.
+
+ HASGNODE enables/disables readgnode() in node.c.
+
+ HASHASHPID is defined when the Linux version (probably
+ above 2.1.35) has a pidhash_next member in
+ its task structure.
+
+ HASHSNODE enables/disables readhsnode() in node.c.
+
+ HASI_E2FS_PTR indicates the BSD dialect has a pointer in
+ its inode to the EXTFS dinode.
+
+ HASI_FFS indicates the BSD dialect has i_ffs_size
+ in <ufs/ufs/inode.h>.
+
+ HASI_FFS1 indicates the BSD dialect supports the fast
+ UFS1 and UFS2 file systems.
+
+ HAS_INKERNEL indicates the SCO OSR 6.0.0 or higher, or
+ UnixWare 7.1.4 or higher system uses the
+ INKERNEL symbol in <netinet/in_pcb.h> or
+ <netinet/tcp_var.h>.
+
+ HASINODE enables/disables readinode() in node.c.
+
+ HASINOKNC indicates the Linux version has a kernel
+ name cache keyed on inode address.
+
+ HASINADDRSTR is defined when the inp_[fl]addr members
+ of the inpcb structure are structures.
+
+ HASINRIAIPv6 is defined if the dialect has the INRIA IPv6
+ support. (HASIPv6 will also be defined.)
+
+ HASINT16TYPE is defined when the dialect has a typedef
+ for int16 that may conflict with some other
+ header file's redefinition (e.g., <afs/std.h>).
+
+ HASINT32TYPE is defined when the dialect has a typedef
+ for int32 that may conflict with some other
+ header file's redefinition (e.g., <afs/std.h>).
+
+ HASINTSIGNAL is defined when signal() returns an int.
+
+ HAS_IPCLASSIFIER_H is defined for Solaris dialects that have the
+ <inet/ipclassifier.h> header file.
+
+ HAS_IPC_S_PATCH is defined when the HP-UX 11 dialect has the
+ ipc_s patch installed. It has a value of
+ 1 if the ipc_s structure has an ipc_ipis
+ member, but the ipis_s structure lacks the
+ ipis_msgsqueued member; 2, if ipc_s has
+ ipc_ipis, but ipis_s lacks ipis_msgsqueued.
+
+ HASIPv6 indicates the dialect supports the IPv6
+ Internet address family.
+
+ HASKERNELKEYT indicates the Linux version has a
+ __kernel_key_t typedef in <linux/types.h>.
+
+ HASKERNFS is defined for BSD dialects for which
+ /kern file system support can be provided.
+
+ HASKERNFS_KFS_KT indicates *kfs_kt is in the BSD dialect's
+ <miscfs/kernfs/kernfs.h>.
+
+ HASKOPT enables/disables the ability to read the
+ kernel's name list from a file -- e.g., from
+ a crash dump file.
+
+ HASKQUEUE indicates the dialect supports the kqueue
+ file type.
+
+ HASKVMGETPROC2 The *BSD dialect has the kvm_gettproc2()
+ function.
+
+ HAS_KVM_VNODE indicates the FreeBSD 5.3 or higher dialect has
+ "defined(_KVM_VNODE)" in <sys/vnode.h>.
+
+ HASLFILEADD defines additional, dialect-specific elements
+ SETLFILEADD in the lfile structure (defined in lsof.h).
+ HASLFILEADD is a macro. The accompanying SETFILEADD
+ macro is used in the alloc_lfile() function of
+ proc.c to preset the additional elements.
+
+ HAS_LF_LWP is defined for BSD dialects where the lockf
+ structure has an lf_lwp member.
+
+ HASLFS indicates the *BSD dialect has log-structured
+ file system support.
+
+ HAS_LOCKF_ENTRY indicates the FreeBSD version has a lockf_entry
+ structure in its <sys/lockf.h> header file.
+
+ HAS_LWP_H is defined for BSD dialects that have the
+ <sys/lwp.h> header file.
+
+ HASMOPT enables/disables the ability to read kernel
+ memory from a file -- e.g., from a crash
+ dump file.
+
+ HASMSDOSFS enables MS-DOS file system support in a
+ BSD dialect.
+
+ HASMNTSTAT indicates the dialect has a stat(2) status
+ element in its mounts structure.
+
+ HASMNTSUP indicates the dialect supports the mount supplement
+ option.
+
+ HASNAMECACHE indicates the FreeBSD dialect has a namecache
+ structure definition in <sys/namei.h>.
+
+ HASNCACHE enables the probing of the kernel's name cache
+ to obtain path name components. A value
+ of 1 directs printname() to prefix the
+ cache value with the file system directory
+ name; 2, avoid the prefix.
+
+ HASNCVPID The *BSD dialect namecache struct has an
+ nc_vpid member.
+
+ HASNETDEVICE_H indicates the Linux version has a netdevice.h
+ header file.
+
+ HAS_NFS enables NFS support for the dialect.
+
+ HASNFSKNC indicates the LINUX version has a separate
+ NFS name cache.
+
+ HASNFSPROTO indicates the NetBSD or OpenBSD version
+ has the nfsproto.h header file.
+
+ HASNFSVATTRP indicates the n_vattr member of the nfsnode of
+ the *BSD dialect is a pointer.
+
+ HASNLIST enables/disables nlist() function support.
+ (See NLIST_TYPE.)
+
+ HASNOFSADDR is defined if the dialect has no file structure
+ addresses. (HASFSTRUCT must be defined.)
+
+ HASNOFSCOUNT is defined if the dialect has no file structure counts.
+ (HASFSTRUCT must be defined.)
+
+ HASNOFSFLAGS is defined if the dialect has no file structure flags.
+ (HASFSTRUCT must be defined.)
+
+ HASNOFSNADDR is defined if the dialect has no file structure node
+ addresses. (HASFSTRUCT must be defined.)
+
+ HAS_NO_6PORT is defined if the FreeBSD in_pcb.h has no in6p_.port
+ definitions.
+
+ HAS_NO_6PPCB is defined if the FreeBSD in_pcb.h has no in6p_ppcb
+ definition.
+
+ HAS_NO_ISO_DEV indicates the FreeBSD 6 and higher system has
+ no i_dev member in its iso_node structure.
+
+ HAS_NO_LONG_LONG indicates the dialect has no support for the C
+ long long type. This definition is used by
+ the built-in snprintf() support of lib/snpf.c.
+
+ HAS_NO_SI_UDEV indicates the FreeBSD 6 and higher system has
+ no si_udev member in its cdev structure.
+
+ HASNOSOCKSECURITY enables the listing of open socket files,
+ even when HASSECURITY restricts listing of
+ open files to the UID of the user who is
+ running lsof, provided socket file listing
+ is selected with the "-i" option. This
+ definition is only effective when HASSECURITY
+ is also defined.
+
+ HASNULLFS indicates the dialect (usually *BSD) has a
+ null file system.
+
+ HASOBJFS indicates the Pyramid version has OBJFS
+ support.
+
+ HASONLINEJFS indicates the HP-UX 11 dialect has the optional
+ OnlineJFS package installed.
+
+ HASPERSDC enables the use of a personal device cache
+ file path and specifies a format by which
+ it is constructed. See the 00DCACHE file
+ of the lsof distribution for more information
+ on the format.
+
+ HASPERSDCPATH enables the use of a modified personal
+ device cache file path and specifies the
+ name of the environment variable from which
+ its component may be taken. See the 00DCACHE
+ file of the lsof distribution for more
+ information on the modified personal device
+ cache file path.
+
+ HASPINODEN declares that the inode number of a /proc file
+ should be stored in its procfsid structure.
+
+ HASPIPEFN defines the function that processes DTYPE_PIPE
+ file structures. It's used in the prfp.c
+ library source file. See the FreeBSD
+ dialect source for an example.
+
+ HASPIPENODE enables/disables readpipenode() in node.c.
+
+ HASPMAPENABLED enables the automatic reporting of portmapper
+ registration information for TCP and UDP
+ ports that have been registered.
+
+ HASPPID indicates the dialect has parent PID support.
+
+ HASPR_LDT indicates the Solaris dialect has a pr_ldt
+ member in the pronodetype enum.
+
+ HASPR_GWINDOWS indicates the Solaris dialect has a pr_windows
+ member in the pronodetype enum.
+
+ HASPRINTDEV this value defines a private function for
+ printing the dialect's device number. Used
+ by print.c/print_file(). Takes one argument:
+
+ char *HASPRINTDEV(struct lfile *)
+
+ HASPRINTINO this value names a private function for
+ printing the dialect's inode number. Used
+ by print.c/print_file(). Takes one argument:
+
+ char *HASPRINTINO(struct lfile *)
+
+ HASPRINTNM this value names a private function for
+ printing the dialect's file name. Used by
+ print.c/print_file(). Takes one argument:
+
+ void HASPRINTNM(struct lfile *)
+
+ HASPRINTOFF this value names a private function for
+ printing the dialect's file offset. Used
+ by print.c/print_file(). Takes two arguments:
+
+ char *HASPRINTOFF(struct lfile *, int ty)
+
+ Where ty == 0 if the offset is to be printed
+ in 0t<decimal> format; 1, 0x<hexadecimal>.
+
+ HASPRINTSZ this value names a private function for
+ printing the dialect's file size. Used
+ by print.c/print_file(). Takes one argument:
+
+ char *HASPRINTSZ(struct lfile *)
+
+ void HASPRINTNM(struct lfile *)
+
+ HASPRIVFILETYPE enables processing of the private file
+ type, whose number (from f_type of the file
+ struct) is defined by PRIVFILETYPE.
+ HASPRIVFILETYPE defines the function that
+ processes the file struct's f_data member.
+ Processing is initiated from the process_file()
+ function of the prfp.c library source file
+ or from the dialect's own process_file()
+ function.
+
+ HASPRIVNMCACHE enables printing of a file path from a
+ private name cache. HASPRIVNMCACHE defines
+ the name of the printing function. The
+ function takes one argument, a struct lfile
+ pointer to the file, and returns non-zero
+ if it prints a cached name to stdout.
+
+ HASPRIVPRIPP is defined for dialects that have a private
+ function for printing the IP protocol name.
+ When this is not defined, the function to
+ do that defaults to printiproto().
+
+ HASPROCFS defines the name (if any) of the process file
+ system -- e.g., /proc.
+
+ HASPROCFS_PFSROOT indicates PFSroot is in the BSD dialect's
+ <miscfs/procfs/procfs.h>.
+
+ HASPSEUDOFS indicates the FreeBSD dialect has pseudofs
+ file system support.
+
+ HASPSXSEM indicates the dialect has support for the POSIX
+ semaphore file type.
+
+ HASPSXSHM indicates the dialect has support for the POSIX
+ shared memory file type.
+
+ HASPTYFS indicates the *BSD dialect has a ptyfs file system.
+
+ HASRNODE enables/disables readrnode() in node.c.
+
+ HASRNODE3 indicates the HPUX 10.20 or lower dialect has NFS3
+ support with a modified rnode structure.
+
+ HASRPCV2H The FreeBSD dialect has <nfs/rpcv2.h>.
+
+ HAS_SANFS indicates the AIX system has SANFS file system
+ support.
+
+ HASSBSTATE indicates the dialect has socket buffer state
+ information (e.g., SBS_* symbols) available.
+
+ HASSECURITY enables/disables restricting open file
+ information access. (Also see HASNOSOCKSECURITY.)
+
+ HASSELINUX indicates the Linux dialect has SELinux security
+ context support available.
+
+ HASSETLOCALE is defined if the dialect has <locale.h> and
+ setlocale().
+
+ HAS_SI_PRIV indicates the FreeBSD 6.0 and higher cdev
+ structure has an si_priv member.
+
+ HASSOUXSOUA indicates that the Solaris <sys/socketvar.h> has
+ soua_* members in its so_ux_addr structure.
+
+ HASSPECDEVD indicates the dialect has a special device
+ directory and defines the name of a function
+ that processes the results of a successful
+ stat(2) of a file in that directory.
+
+ HASSPECNODE indicates the DEC OSF/1, or Digital UNIX,
+ or Tru64 UNIX <sys/specdev.h> has a spec_node
+ structure definition.
+
+ HASSNODE indicates the dialect has snode support.
+
+ HAS_SOCKET_SK indicates that the Linux socket structure
+ has the ``struct sock *sk'' member.
+
+ HASSOOPT indicates the dialect has socket option
+ information (e.g., SO_* symbols) available.
+
+ HASSOSTATE indicates the dialect has socket state
+ information (e.g., SS_* symbols) available.
+
+ HASSTATVFS indicates the NetBSD dialect has a statvfs
+ struct definition.
+
+ HASSTAT64 indicates the dialect's <sys/stat.h> contains
+ stat64.
+
+ HAS_STD_CLONE indicates the dialect uses a standard clone
+ device structure that can be used in common
+ library function clone processing. If the
+ value is 1, the clone table will be built
+ by readdev() and cached when HASDCACHE is
+ defined; if the value is 2, it is assumed
+ the clone table is built independently.
+ (Also see CLONEMAJ and HAVECLONEMAJ.)
+
+ HASSTREAMS enables/disables streams. CAUTION, requires
+ specific support code in the dialect sources.
+
+ HAS_STRFTIME indicates the dialect has the gmtime() and
+ strftime() C library functions that support
+ the -r marker format option. Configure tests
+ for the functions and defines this symbol.
+
+ HASSYSDC enables the use of a system-wide device
+ cache file and defines its path. See the
+ 00DCACHE file of the lsof distribution for
+ more information on the system-wide device
+ cache file path option.
+
+ HAS_SYS_PIPEH indicates the dialect has a <sys/pipe.h>
+ header file.
+
+ HAS_SYS_SX_H indicates the FreeBSD 7.0 and higher system has
+ a <sys/sx.h> header file.
+
+ HASTAGTOPATH indicates the DEC OSF/1, Digital UNIX, or
+ Tru64 UNIX dialect has a libmsfs.so,
+ containing tag_to_path().
+
+ HASTMPNODE enables/disables readtnode() in node.c.
+
+ HASTCPOPT indicates the dialect has TCP option
+ information (i.e., from TF_* symbols)
+ available.
+
+ HASTCPTPIQ is defined when the dialect can duplicate
+ the receive and send queue sizes reported
+ by netstat.
+
+ HASTCPTPIW is defined when the dialect can duplicate
+ the receive and send window sizes reported
+ by netstat.
+
+ HASTCPUDPSTATE is defined when the dialect has support for
+ TCP and UDP state, including the "-s p:s"
+ option and associated speed ehancements.
+
+ HASTFS indicates that the Pyramid dialect has TFS
+ file system support.
+
+ HAS_UFS1_2 indicates the FreeBSD 6 and higher system has
+ UFS1 and UFS2 members in its inode structure.
+
+ HAS_UM_UFS indicates the OpenBSD version has UM_UFS[12]
+ definitions.
+
+ HASUNMINSOCK indicates the Linux version has a user name
+ element in the socket structure; a value of
+ 0 says there is no unix_address member; 1,
+ there is.
+
+ HASUINT16TYPE is defined when the dialect has a typedef
+ for u_int16 that may conflict with some other
+ header file's redefinition (e.g., <afs/std.h>).
+
+ HASUTMPX indicates the dialect has a <utmpx.h> header
+ file.
+
+ HAS_UVM_INCL indicates the NetBSD or OpenBSD dialect has
+ a <uvm> include directory.
+
+ HAS_UW_CFS indicates the UnixWare 7.1.1 or above dialect
+ has CFS file system support.
+
+ HAS_UW_NSC indicates the UnixWare 7.1.1 or above dialect
+ has a NonStop Cluster (NSC) kernel.
+
+ HAS_V_LOCKF indicates the FreeBSD version has a v_lockf
+ member in the vode structure, defined in
+ <sys/vnode.h>.
+
+ HASVMLOCKH indicates the FreeBSD dialect has <vm/lock.h>.
+
+ HASVNODE enables/disables readvnode() function in node.c.
+
+ HAS_V_PATH indicates the dialect's vnode structure has a
+ v_path member.
+
+ HAS_VSOCK indicates that the Solaris version has a VSOCK
+ member in the vtype enum
+
+ HASVXFS enables Veritas VxFS file system support for
+ the dialect. CAUTION, the dialect sources
+ must have the necessary support code.
+
+ HASVXFSDNLC indicates the VxFS file system has its own
+ name cache.
+
+ HASVXFS_FS_H indicates <sys/fs/vx_fs.h> exists.
+
+ HASVXFS_MACHDEP_H indicates <sys/fs/vx_machdep.h> exists.
+
+ HASVXFS_OFF64_T indicates <sys/fs/vx_solaris.h> exists and
+ has an off64_t typedef.
+
+ HASXVFSRNL indicates the dialect has VxFS Reverse Name
+ Lookup (RNL) support.
+
+ HASVXFS_SOL_H indicates <sys/fs/vx_sol.h> exists.
+
+ HASVXFS_SOLARIS_H indicates <sys/fs/vx_solaris.h> exists.
+
+ HASVXFS_U64_T if HASVXFS_SOLARIS_H is defined, this
+ variable indicates that <sys/fs/vx_solaris.h>
+ has a vx_u64_t typedef.
+
+ HASVXFSUTIL indicates the Solaris dialect has VxFS 3.4
+ or higher and has the utility libraries,
+ libvxfsutil.a (32 bit) and libvxfsutil64.a
+ (64 bit).
+
+ HASVXFS_VX_INODE indicates that <sys/fs/vx_inode.h> contains
+ a vx_inode structure.
+
+ HASWIDECHAR indicates the dialect has the wide-character
+ support functions iswprint(), mblen() and mbtowc().
+
+ HASXNAMNODE indicates the OSR dialect has <sys/fs/xnamnode.h>.
+
+ HASXOPT defines help text for dialect-specific X option
+ and enables X option processing in usage.c and
+ main.c.
+
+ HASXOPT_ROOT when defined, restricts the dialect-specific
+ X option to processes whose real user ID
+ is root.
+
+ HAS_ZFS indicates the dialect has support for the ZFS file
+ system.
+
+ HASXOPT_VALUE defines the default binary value for the X option
+ in store.c.
+
+ HASZONES the Solaris dialect has zones.
+
+ HAVECLONEMAJ defines the name of the status variable
+ that indicates a clone major device number
+ is available in CLONEMAJ. (Also see CLONEMAJ
+ and HAS_STD_CLONE.)
+
+ HPUX_KERNBITS defines the number of bits in the HP-UX 10.30
+ and above kernel "basic" word: 32 or 64.
+
+ KA_T defines the type cast required to assign
+ space to kernel pointers. When not defined
+ by a dialect header file, KA_T defaults to
+ unsigned long.
+
+ KA_T_FMT_X defines the printf format for printing a
+ KA_T -- the default is "%#lx" for the
+ default unsigned long KA_T cast.
+
+ LSOF_ARCH See 00XCONFIG.
+
+ LSOF_BLDCMT See 00XCONFIG.
+
+ LSOF_CC See 00XCONFIG.
+
+ LSOF_CCV See 00XCONFIG.
+
+ LSOF_HOST See 00XCONFIG.
+
+ LSOF_INCLUDE See 00XCONFIG.
+
+ LSOF_LOGNAME See 00XCONFIG.
+
+ LSOF_MKC See the "The Mksrc Shell Script" section of
+ this file.
+
+ LSOF_SYSINFO See 00XCONFIG.
+
+ LSOF_USER See 00XCONFIG.
+
+ LSOF_VERS See 00XCONFIG.
+
+ LSOF_VSTR See 00XCONFIG.
+
+ MACH defines a MACH system.
+
+ N_UNIXV defines an alternate value for the N_UNIV symbol.
+
+ NCACHELDPFX defines C code to be executed before calling
+ ncache_load().
+
+ NCACHELDSFX defines C code to be executed after calling
+ ncache_load().
+
+ NEVER_HASDCACHE keeps the Customize script from offering to
+ change HASDCACHE by its presence anywhere
+ in a dialect's machine.h header file --
+ e.g., in a comment. See the Customize
+ script or machine.h in dialects/linux/proc.
+
+ NEVER_WARNDEVACCESS keeps the Customize script from offering to
+ change WARNDEVACCESS by its presence anywhere
+ in a dialect's machine.h header file --
+ including in a comment. See the Customize
+ script or machine.h in dialects/linux/proc.
+
+ NLIST_TYPE is the type of the nlist table, Nl[], if it is
+ not nlist. HASNLIST must be set for this
+ definition to be effective.
+
+ NOWARNBLKDEV specifies that no warning is to be issued
+ when no block devices are found. This
+ definiton is used only when HASBLKDEV is
+ also defined.
+
+ OFFDECDIG specifies how many decimal digits will be
+ printed for the file offset in a 0t form
+ before switching to a 0x form. The count
+ includes the "0t". A count of zero means
+ the size is unlimited.
+
+ PRIVFILETYPE is the number of a private file type, found
+ in the f_type member of the file struct, to
+ be processed by the HASPRIVFILETYPE function.
+ See the AIX dialect sources for an example.
+
+ _PSTAT_STREAM_GET_XPORT
+ indicates the HP-UX PSTAT header files require
+ this symbol to be defined for proper handling of
+ stream export data.
+
+ TIMEVAL_LSOF defines the name of the timeval structure.
+ The default is timeval. /dev/kmem-based
+ Linux lsof redefines timeval with this
+ symbol to avoid conflicts between glibc
+ and kernel definitions.
+
+ TYPELOGSECSHIFT defines the type of the cdfs_LogSecShift
+ member of the cdfs structure for UnixWare
+ 7 and higher.
+
+ UID_ARG_T defines the cast on a User ID when passed
+ as a function argument.
+
+ USE_LIB_COMPLETEVFS
+ selects the use of the completevfs() function
+ in lsof4/lib/cvfs.c.
+
+ USE_LIB_FIND_CH_INO
+ selects the use of the find_ch_ino() inode
+ function in lsof4/lib/fino.c.
+
+ Note: HASBLKDEV selects the has_bl_ino()
+ function.
+
+ USE_LIB_IS_FILE_NAMED
+ selects the use of the is_file_named() function
+ in lsof4/lib/isfn.c.
+
+ USE_LIB_LKUPDEV selects the use of the lkupdev() function
+ in lsof4/lib/lkud.c.
+
+ Note: HASBLKDEV selects the lkupbdev() function.
+
+ USE_LIB_PRINTDEVNAME
+ selects the use of the printdevname() function
+ in lsof4/lib/pdvn.c.
+
+ Note: HASBLKDEV selects the printbdevname()
+ function.
+
+ USE_LIB_PRINT_TCPTPI
+ selects the use of the print_tcptpi() function
+ in lsof4/lib/ptti.c.
+
+ USE_LIB_PROCESS_FILE
+ selects the use of the process_file() function
+ in lsof4/lib/prfp.c.
+
+ USE_LIB_READDEV selects the use of the readdev() and stkdir()
+ functions in lsof4/lib/rdev.c.
+
+ USE_LIB_READMNT selects the use of the readmnt() function
+ in lsof4/lib/rmnt.c.
+
+ USE_LIB_RNAM selects the use of the device cache functions
+ in lsof4/lib/rnam.c.
+
+ Note: HASNCACHE must also be defined.
+
+ USE_LIB_RNCH selects the use of the device cache functions
+ in lsof4/lib/rnch.c.
+
+ Note: HASNCACHE must also be defined.
+
+ USE_STAT is defined for those dialects that must
+ use the stat(2) function instead of lstat(2)
+ to scan /dev -- i.e., in the readdev()
+ function.
+
+ VNODE_VFLAG is an alternate name for the vnode structure's
+ v_flag member.
+
+ WARNDEVACCESS enables the issuing of a warning message when
+ lsof is unable to access /dev (or /device)
+ or one of its subdirectories, or stat(2)
+ a file in them. Some dialects (e.g., HP-UX)
+ have many inaccessible subdirectories and
+ it is appropriate to inhibit the warning
+ for them with WARNDEVACCESS. The -w option
+ will also inhibit these warnings.
+
+ WARNINGSTATE when defined, disables the default issuing
+ of warning messages. WARNINGSTATE is
+ undefined by default for all dialects in
+ the lsof distribution.
+
+ WIDECHARINCL defines the header file to be included (if any)
+ when wide-character support is enabled with
+ HASWIDECHAR.
+
+ zeromem() defines a macro to zero memory -- e.g., using
+ bzero() or memset().
+
+Any dialect's machine.h file and Configure stanza can serve as a
+template for building your own. All machine.h files usually have
+all definitions, disabling some (with comment prefix and suffix)
+and enabling others.
+
+
+Options: Common and Special
+---------------------------
+
+All but one lsof option is common; the specific option is ``-X''.
+If a dialect does not support a common option, the related #define
+in machine.h -- e.g., HASCOPT -- should be deselected.
+
+The specific option, ``-X'', may be used by any dialect for its
+own purpose. Right now (May 30, 1995) the ``-X'' option is binary
+(i.e., it's not allowed arguments of its own, and its value must
+be 0 or 1) but that could be changed should the need arise. The
+option is enabled with the HASXOPT definition in machine.h; its
+default value is defined by HASXOPT_VALUE.
+
+The value of HASXOPT should be the text displayed for ``-X'' by
+the usage() function in usage.c. HASXOPT_VALUE should be the
+default value, 0 or 1.
+
+AIX for the IBM RICS System/6000 defines the ``-X'' option to
+control readx() usage, since there is a bug in AIX kernels that
+readx() can expose for other processes.
+
+
+Defining Dialect-Specific Symbols and Global Storage
+----------------------------------------------------
+
+A dialect's dlsof.h and dstore.c files contain dialect-specific
+symbol and global storage definitions. There are symbol definitions,
+for example, for function and data casts, and for file paths.
+Dslof.h defines lookup names the nlist() table -- X_* symbols --
+when nlist() is being used.
+
+Global storage definitions include such things as structures for
+local Virtual File System (vfs) information; mount information;
+search file information; and kernel memory file descriptors --
+e.g., Kmem for /dev/kmem, Mem for /dev/mem, Swap for /dev/drum.
+
+
+Coding Dialect-specific Functions
+---------------------------------
+
+Each supported dialect must have some basic functions that the
+common functions of the top level may call. Some of them may be
+obtained from the library in lsof4/lib, selected and customized by
+#define's in the dialect machine.h header file. Others may have
+to be coded specifically for the dialect.
+
+Each supported dialect usually has private functions, too. Those
+are wholly determined by the needs of the dialect's data organization
+and access.
+
+These are some of the basic functions that each dialect must supply
+-- they're all defined in proto.h:
+
+ initialize() function to initialize the dialect
+
+ is_file_named() function to check if a file was named
+ by an optional file name argument
+ (lsof4/lib/isfn.c)
+
+ gather_proc_info() function to gather process table
+ and related information and cache it
+
+ printchdevname() function to locate and optionally
+ print the name of a character device
+ (lsof4/lib/pdvn.c)
+
+ print_tcptpistate() function to print the TCP or TPI
+ state for a TCP or UDP socket file,
+ if the one in lib/ptti.c isn't
+ suitable (define USE_LIB_PRINT_TCPTPI
+ to activate lib/ptti.c)
+
+ process_file() function to process an open file
+ structure (lsof4/lib/prfp.c)
+
+ process_node() function to process a primary node
+
+ process_socket() function to process a socket
+
+ readdev() and stkdir() functions to read and cache device
+ information (lsof4/lib/rdev.c)
+
+ readmnt() function to read mount table information
+ (lsof4/lib/rmnt.c)
+
+Other common functions may be needed, and might be obtained from
+lsof4/lib, depending on the needs of the dialect's node and socket
+file processing functions.
+
+Check the functions in lsof4/lib and specific lsof4/dialects/*
+files for examples.
+
+As you build these functions you will probably have to add #include's
+to dlsof.h.
+
+
+Function Prototype Definitions and the _PROTOTYPE Macro
+-------------------------------------------------------
+
+Once you've defined your dialect-specific definitions, you should
+define their prototypes in dproto.h or locally in the file where
+they occur and are used. Do this even if your compiler is not ANSI
+compliant -- the _PROTOTYPE macro knows how to cope with that and
+will avoid creating prototypes that will confuse your compiler.
+
+
+The Makefile
+------------
+
+Here are some general rules for constructing the dialect Makefile.
+
+ * Use an existing dialect's Makefile as a template.
+
+ * Make sure the echo actions of the install rule are appropriate.
+
+ * Use the DEBUG string to set debugging options, like ``-g''.
+ You may also need to use the -O option when forking and
+ SIGCHLD signals defeat your debugger.
+
+ * Don't put ``\"'' in a compiler flags -D<symbol>=<string>
+ clause in your Makefile. Leave off the ``\"'' even though
+ you want <string> to be a string literal and instead adapt
+ the N_UNIX* macros you'll find in Makefiles for FreeBSD
+ and Linux. That will allow the Makefile's version.h rule
+ to put CFLAGS into version.h without having to worry about
+ the ``\"'' sequences.
+
+ * Finally, remember that strings can be passed from the top
+ level's Configure shell script. That's an appropriate way
+ to handle options, especially if there are multiple versions
+ of the Unix dialect to which you are porting lsof 4.
+
+
+The Mksrc Shell Script
+----------------------
+
+Pattern your Mksrc shell script after an existing one from another
+dialect. Change the D shell variable to the name of your dialect's
+subdirectory in lsof4/dialects. Adjust any other shell variable
+to your local conditions. (Probably that won't be necessary.)
+
+Note that, if using symbolic links from the top level to your
+dialect subdirectory is impossible or impractical, you can set the
+LSOF_MKC shell variable in Configure to something other than
+"ln -s" -- e.g., "cp," and Configure will pass it to the Mksrc
+shell script in the M environment variable.
+
+
+The MkKernOpts Shell Script
+---------------------------
+
+The MkKernOptrs shell script is used by some dialects -- e.g.,
+Pyramid DC/OSx and Reliant UNIX -- to determine the compile-time
+options used to build the current kernel that affect kernel structure
+definitions, so those same options can be used to build lsof.
+Configure calls MkKernOpts for the selected dialects.
+
+If your kernel is built with options that affect structure definitions.
+-- most commonly affected are the proc structure from <sys/proc.h>
+and the user structure from <sys/user.h> -- check the MkKernOpts
+in lsof4/dialects/irix for a comprehensive example.
+
+
+Testing and the Lsof Test Suite
+-------------------------------
+
+Once you have managed to create a port, here are some tips for
+testing it.
+
+* First look at the test suite in the tests/ sub-directory of the
+ lsof distribution. While it will need to be customized to be
+ usable with a new port, it should provide ideas on things to
+ test. Look for more information about the test suite in the
+ 00TEST file.
+
+* Pick a simple process whose open files you are likely to
+ know and see if the lsof output agrees with what you know.
+ (Hint: select the process with `lsof -p <process_PID>`.)
+
+ Are the device numbers and device names correct?
+
+ Are the file system names and mount points correct?
+
+ Are inode numbers and sizes correct?
+
+ Are command names, file descriptor numbers, UIDs, PIDs, PGIDs,
+ and PPIDs correct?
+
+ A simple tool that does a stat(2) of the files being examined
+ and reports the stat struct contents can provide a reference for
+ some values; so can `ls -l /dev/<device>`.
+
+* Let lsof list information about all open files and ask the
+ same questions. Look also for error messages about not being
+ able to read a node or structure.
+
+* Pick a file that you know is open -- open it and hold it
+ that way with a C program (not vi), if you must. Ask lsof to
+ find the file's open instance by specifying its path to lsof.
+
+* Create a C program that opens a large number of files and holds
+ them open. Background the test process and ask lsof to list
+ its files.
+
+* Generate some locks -- you may need to write a C program to
+ do this, hold the locked file open, and see if lsof can identify
+ the lock properly. You may need to write several C programs
+ if your dialect supports different lock functions -- fnctl(),
+ flock(), lockf(), locking().
+
+* Identify a process with known Internet file usage -- inetd
+ is a good one -- and ask lsof to list its open files. See if
+ protocols and service names are listed properly.
+
+ See if your lsof identifies Internet socket files properly for
+ rlogind or telnetd processes.
+
+* Create a UNIX domain socket file, if your dialect allows it,
+ hold it open by backgrounding the process, and see if lsof can
+ identify the open UNIX domain socket file properly.
+
+* Create a FIFO file and see what lsof says about it.
+
+* Watch an open pipe -- `lsof -u <your_login> | less` is a
+ good way to do this.
+
+* See if lsof can identify NFS files and their devices properly.
+ Open and hold open an NFS file and see if lsof can find the open
+ instance by path.
+
+* If your test system has CD-ROM and floppy disk devices, open
+ files on them and see if lsof reports their information correctly.
+ Such devices often have special kernel structures associated
+ with them and need special attention from lsof for their
+ identification. Pay particular attention to the inode numbers
+ lsof reports for CD-ROM and floppy disk files -- often they are
+ calculated dynamically, rather than stored in a kernel node
+ structure.
+
+* If your implementation can probe the kernel name cache, look
+ at some processes with open files whose paths you know to see
+ if lsof identifies any name components. If it doesn't, make
+ sure the name components are in the name cache by accessing
+ the files yourself with ls or a similar tool.
+
+* If your dialect supports the /proc file system, use a C program
+ to open files there, background a test process, and ask lsof to
+ report its open files.
+
+* If your dialect supports fattach(), create a small test program
+ to use it, background a test process, and ask lsof to report
+ its open files.
+
+I can supply some quick-and-dirty tools for reporting stat buffer
+contents, holding files open, creating UNIX domain files, creating
+FIFOs, etc., if you need them.
+
+
+Where Next?
+-----------
+
+Is this document complete? Certainly not! One might wish that it
+were accompanied by man pages for all lsof functions, by free beer
+or chocolates, by ... (You get the idea.)
+
+But those things are not likely to happen as long as lsof is a
+privately supported, one man operation.
+
+So, if you need more information on how lsof is constructed or
+works in order to do a port of your own, you'll have to read the
+lsof source code. You can also ask me questions via email, but
+keep in mind the private, one-man nature of current lsof support.
+
+
+Vic Abell <abe@purdue.edu>
+March 25, 2009
--- /dev/null
+
+ A Quick Start for Lsof
+
+1. Introduction
+================
+
+ Agreed, the lsof man page is dense and lsof has a plethora of
+ options. There are examples, but the manual page format buries
+ them at the end. How does one get started with lsof?
+
+ This file is an attempt to answer that question. It plunges
+ immediately into examples of lsof use to solve problems that
+ involve looking at the open files of Unix processes.
+
+
+ Contents
+
+ 1. Introduction
+ 2. Finding Uses of a Specific Open File
+ 3. Finding Open Files Filling a File System
+ a. Finding an Unlinked Open File
+ 4. Finding Processes Blocking Umount
+ 5. Finding Listening Sockets
+ 6. Finding a Particular Network Connection
+ 7. Identifying a Netstat Connection
+ 8. Finding Files Open to a Named Command
+ 9. Deciphering the Remote Login Trail
+ a. The Fundamentals
+ b. The idrlogin.perl[5] Scripts
+ 10. Watching an Ftp or Rcp Transfer
+ 11. Listing Open NFS Files
+ 12. Listing Files Open by a Specific Login
+ a. Ignoring a Specific Login
+ 13. Listing Files Open to a Specific Process Group
+ 14. When Lsof Seems to Hang
+ a. Kernel lstat(), readlink(), and stat() Blockages
+ b. Problems with /dev or /devices
+ c. Host and Service Name Lookup Hangs
+ d. UID to Login Name Conversion Delays
+ 15. Output for Other Programs
+ 16. The Lsof Exit Code and Shell Scripts
+ 17. Strange messages in the NAME column
+
+ Options
+
+ A. Selection Options
+ B. Output Options
+ C. Precautionary Options
+ D. Miscellaneous Lsof Options
+
+
+2. Finding Uses of a Specific Open File
+========================================
+
+ Often you're interested in knowing who is using a specific file.
+ You know the path to it and you want lsof to tell you the processes
+ that have open references to it.
+
+ Simple -- execute lsof and give it the path name of the file of
+ interest -- e.g.,
+
+ $ lsof /etc/passwd
+
+ Caveat: this only works if lsof has permission to get the status
+ (via stat(2)) of the file at the named path. Unless the lsof
+ process has enough authority -- e.g., it is being run with a
+ real User ID (UID) of root -- this AIX example won't work:
+
+ Further caveat: this use of lsof will fail if the stat(2) kernel
+ syscall returns different file parameters -- particularly device
+ and inode numbers -- than lsof finds in kernel node structures.
+ This condition is rare and is usually documented in the 00FAQ
+ file of the lsof distribution.
+
+ $ lsof /etc/security/passwd
+ lsof: status error on /etc/security/passwd: Permission denied
+
+
+3. Finding Open Files Filling a File System
+============================================
+
+ Oh! Oh! /tmp is filling and ls doesn't show that any large files
+ are being created. Can lsof help?
+
+ Maybe. If there's a process that is writing to a file that has
+ been unlinked, lsof may be able to discover the process for you.
+ You ask it to list all open files on the file system where /tmp
+ is located.
+
+ Sometimes /tmp is a file system by itself. In that case,
+
+ $ lsof /tmp
+
+ is the appropriate command. If, however, /tmp is part of another
+ file system, typically /, then you may have to ask lsof to list
+ all files open on the containing file system and locate the
+ offending file and its process by inspection -- e.g.,
+
+ $ lsof / | more
+ or
+ $ lsof / | grep ...
+
+ Caveat: there must be a file open to a for the lsof search to
+ succeed. Sometimes the kernel may cause a file reference to
+ persist, even where there's no file open to a process. (Can you
+ say kernel bug? Maybe.) In any event, lsof won't be able to
+ help in this case.
+
+ a. Finding an Unlinked Open File
+ =================================
+
+ A pesky variant of a file that is filling a file system is an
+ unlinked file to which some process is still writing. When a
+ process opens a file and then unlinks it, the file's resources
+ remain in use by the process, but the file's directory entries
+ are removed. Hence, even when you know the directory where the
+ file once resided, you can't detect it with ls.
+
+ This can be an administrative problem when the unlinked file is
+ large, and the process that holds it open continues to write to
+ it. Only when the process closes the file will its resources,
+ particularly disk space, be released.
+
+ Lsof can help you find unlinked files on local disks. It has an
+ option, +L, that will list the link counts of open files. That
+ helps because an unlinked file on a local disk has a zero link
+ count. Note: this is NOT true for NFS files, accessed from a
+ remote server.
+
+ You could use the option to list all files and look for a zero
+ link count in the NLINK column -- e.g.,
+
+ $lsof +L
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF NLINK NODE NAME
+ ...
+ less 25366 abe txt VREG 6,0 40960 1 76319 /usr/...
+ ...
+ > less 25366 abe 3r VREG 6,0 17360 0 98768 / (/dev/sd0a)
+
+ Better yet, you can specify an upper bound to the +L option, and
+ lsof will select only files that have a link count less than the
+ upper bound. For example:
+
+ $ lsof +L1
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF NLINK NODE NAME
+ less 25366 abe 3r VREG 6,0 17360 0 98768 / (/dev/sd0a)
+
+ You can use lsof's -a (AND) option to narrow the link count search
+ to a particular file system. For example, to look for zero link
+ counts on the /home file system, use:
+
+ $ lsof -a +L1 /home
+
+ CAUTION: lsof can't always report link counts for all file types
+ -- e.g., it may not report them for FIFOs, pipes, or sockets.
+ Remember also that link counts for NFS files on an NFS client
+ host don't behave as do link counts for files on local disks.
+
+
+4. Finding Processes Blocking Umount
+=====================================
+
+ When you need to unmount a file system with the umount command,
+ you may find the operation blocked by a process that has a file
+ open on the file systems. Lsof may be able to help you find the
+ process. In response to:
+
+ $ lsof <file_system_name>
+
+ Lsof will display all open files on the named file system. It
+ will also set its exit code zero when it finds some open files
+ and non-zero when it doesn't, making this type of lsof call
+ useful in shell scripts. (See section 16.)
+
+ Consult the output of the df command for file system names.
+
+ See the caveat in the preceding section about file references
+ that persist in the kernel without open file traces. That
+ situation may hamper lsof's ability to help with umount, too.
+
+
+5. Finding Listening Sockets
+=============================
+
+ Sooner or later you may wonder if someone has installed a network
+ server that you don't know about. Lsof can list for you all the
+ network socket files open on your machine with:
+
+ $ lsof -i
+
+ The -i option without further qualification lists all open Internet
+ socket files. You can add network names or addresses, protocol
+ names, and service names or port numbers to the -i option to
+ refine the search. (See the next section.)
+
+
+6. Finding a Particular Network Connection
+===========================================
+
+ When you know the source or destination of a network connection
+ whose open files and process you'd like to identify, the -i option
+ may help.
+
+ If, for example, you want to know what process has a connection
+ open to or from the Internet host named aaa.bbb.ccc, you can ask
+ lsof to search for it with:
+
+ $ lsof -i@aaa.bbb.ccc
+
+ If you're interested in a particular protocol -- TCP or UDP --
+ and a specific port number or service name, you can add those
+ discriminators to the -i information:
+
+ $ lsof -iTCP@aaa.bbb.ccc:ftp-data
+
+ If you're interested in a particular IP version -- IPv4 or IPv6
+ -- and your UNIX dialect supports both (It does if "IPv[46]"
+ appears in the lsof -h output.), you can add the '4' or '6'
+ selector immediately after -i:
+
+ $ lsof -i4
+ $ lsof -i6
+
+
+7. Identifying a Netstat Connection
+====================================
+
+ How do I identify the process that has a network connection
+ described in netstat output? For example, if netstat says:
+
+ Proto Recv-Q Send-Q Local Address Foreign Address (state)
+ tcp 0 0 vic.1023 ipscgate.login ESTABLISHED
+
+ What process is connected to service name ``login'' on ipscgate?
+
+ Use lsof's -i option:
+
+ $lsof -iTCP@ipscgate:login
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ rlogin 25023 abe 3u inet 0x10144168 0t184 TCP lsof.itap.purdue.edu:1023->ipscgate.cc.purdue.edu:login
+ ...
+
+ There's another way. Notice the 0x10144168 in the DEVICE column
+ of the lsof output? That's the protocol control block (PCB)
+ address. Many netstat applications will display it when given
+ the -A option:
+
+ $ netstat -A
+ PCB Proto Recv-Q Send-Q Local Address Foreign Address (state)
+ 10144168 tcp 0 0 vic.1023 ipscgate.login ESTABLISHED
+ ...
+
+ Using the PCB address, lsof, and grep, you can find the process this
+ way, too:
+
+ $ lsof -i | grep 10144168
+ rlogin 25023 abe 3u inet 0x10144168 0t184 TCP lsof.itap.purdue.edu:1023->ipscgate.cc.purdue.edu:login
+ ...
+
+
+8. Finding Files Open to a Named Command
+=========================================
+
+ When you want to look at the files open to a particular command,
+ you can look up the PID of the process running the command and
+ use lsof's -p option to specify it.
+
+ $ lsof -p <PID>
+
+ However, there's a quicker way, using lsof's -c option, provided
+ you don't mind seeing output for every process running the named
+ command.
+
+ $ lsof -c <first_characters_of_command_name_that_interest_you>
+
+ The lsof -c option is useful when you want to see how many instances
+ of a given command are executing and what their open files are.
+ One useful example is for the sendmail command.
+
+ $ lsof -c sendmail
+
+
+9. Deciphering the Remote Login Trail
+======================================
+
+ If the network connection you're interested in tracing has been
+ initiated externally and is connected to an rlogind, sshd, or
+ telnetd process, asking lsof to identify that process might not
+ give a wholly satisfying answer. The report may be that the
+ connection exists, but to a process owned by root.
+
+ a. The Fundamentals
+ ====================
+
+ How do you get from there to the login name really using the
+ connection? You have to know a little about how real and pseudo
+ ttys are paired in your system, and then use several lsof probes
+ to identify the login.
+
+ This example comes from a Solaris 2.4 system, named klaatu.cc.
+ I've logged on to it via rlogin from lsof.itap. The first lsof
+ probe,
+
+ $ lsof -i@lsof.itap
+
+ yields (among other things):
+
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ in.rlogin 7362 root 0u inet 0xfc0193b0 0t242 TCP klaatu.cc.purdue.edu:login->lsof.itap.purdue.edu:1023
+ ...
+
+ This confirms that a connection exists. A second lsof probe
+ shows:
+
+ $ lsof -p7362
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ...
+ in.rlogin 7362 root 0u inet 0xfc0193b0 0t242 TCP klaatu.cc.purdue.edu:login->lsof.itap.purdue.edu:1023
+ ...
+ in.rlogin 7362 root 3u VCHR 23, 0 0t66 52928 /devices/pseudo/clone@0:ptmx->pckt->ptm
+
+ 7362 is the Process ID (PID) of the in.rlogin process, discovered
+ in the first lsof probe. (I've abbreviated the output to simplify
+ the example.) Now comes a need to understand Solaris pseudo-ttys.
+ The key indicator is in the DEVICE column for FD 3, the major/minor
+ device number of 23,0. This translates to /dev/pts/0, so a third
+ lsof probe,
+
+ $ lsof /dev/pts/0
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ksh 7364 abe 0u VCHR 24, 0 0t2410 53410 /dev/pts/../../devices/pseudo/pts@0:0
+
+ shows in part that login abe has a ksh process on /dev/pts/0.
+ (The NAME that lsof shows is not /dev/pts/0 but the full expansion
+ of the symbolic link that lsof finds at /dev/pts/0.)
+
+ Here's a second example, done on an HP-UX 9.01 host named ghg.ecn.
+ Again, I've logged on to it from lsof.itap, so I start with:
+
+ $ lsof -i@lsof.itap
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ rlogind 10214 root 0u inet 0x041d5f00 0t1536 TCP ghg.ecn.purdue.edu:login->lsof.itap.purdue.edu:1023
+ ...
+
+ Then,
+
+ $ lsof -p10214
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ...
+ rlogind 10214 root 0u inet 0x041d5f00 0t2005 TCP ghg.ecn.purdue.edu:login->lsof.itap.purdue.edu:1023
+ ...
+ rlogind 10214 root 3u VCHR 16,0x000030 0t2037 24642 /dev/ptym/ptys0
+
+ Here the key is the NAME /dev/ptym/ptys0. In HP-UX 9.01 tty and
+ pseudo tty devices are paired with the names like /dev/ptym/ptys0
+ and /dev/pty/ttys0, so the following lsof probe is the final step.
+
+ $ lsof /dev/pty/ttys0
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ksh 10215 abe 0u VCHR 17,0x000030 0t3399 22607 /dev/pty/ttys0
+ ...
+
+ Here's a third example for an AIX 4.1.4 system. I've used telnet
+ to connect to it from lsof.itap.purdue.edu. I start with:
+
+ $ lsof -i@lsof.itap.purdue.edu
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ...
+ telnetd 15616 root 0u inet 0x05a93400 0t5156 TCP cloud.cc.purdue.edu:telnet->lsof.itap.purdue.edu:3369
+
+ Then I look at the telnetd process:
+
+ $ lsof -p15616
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ...
+ telnetd 15616 root 0u inet 0x05a93400 0t5641 TCP cloud.cc.purdue.edu:telnet->lsof.itap.purdue.edu:3369
+ ...
+ telnetd 15616 root 3u VCHR 25, 0 0t5493 103 /dev/ptc/0
+
+ Here the key is /dev/ptc/0. In AIX it's paired with /dev/pts/0.
+ The last probe for that shows:
+
+ $ lsof /dev/pts/0
+ COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME
+ ...
+ ksh 16642 abe 0u VCHR 26, 0 0t6461 360 /dev/pts/0
+
+ b. The idrlogin.perl[5] Scripts
+ ================================
+
+ There's another, perhaps easier way, to go about the job of
+ tracing a network connection. The lsof distribution contains
+ two Perl scripts, idrlogin.perl (Perl 4) and idrlogin.perl5
+ (Perl 5), that use lsof field output to display values for
+ shells that are parented by rlogind, sshd, or telnetd, or
+ connected directly to TCP sockets. The lsof test suite contains
+ a C library that can be adapted for use with C programs that
+ need to call lsof and process its field output.
+
+ The two Perl scripts use the lsof -R option; it causes the
+ paRent process ID (PPID) to be listed in the lsof output. The
+ scripts identify all shell processes -- e.g., ones whose command
+ names end in ``sh'' -- and determine if: 1) the ultimate ancestor
+ process before a PID greater than 2 (e.g., init's PID is 1) is
+ rlogind, sshd, or telnetd; or 2) the shell process has open
+ TCP socket files.
+
+ Here's an example of output from idlogin.perl on a Solaris 2.4
+ system:
+
+ centurion: 1 = cd src/lsof4/scripts
+ centurion: 2 = ./idrlogin.perl
+ Login Shell PID Via PID TTY From
+ oboyle ksh 12640 in.telnetd 12638 pts/5 opal.cc.purdue.edu
+ icdtest ksh 15158 in.rlogind 15155 pts/6 localhost
+ sh csh 18207 in.rlogind 18205 pts/1 babylon5.cc.purdue.edu
+ root csh 18242 in.rlogind 18205 pts/1 babylon5.cc.purdue.edu
+ trouble ksh 19208 in.rlogind 18205 pts/1 babylon5.cc.purdue.edu
+ abe ksh 21334 in.rlogind 21332 pts/2 lsof.itap.purdue.edu
+
+ The scripts assume that its parent directory contains an
+ executable lsof. If you decide to use one of the scripts, you
+ may want to customize it for your local lsof and perl paths.
+
+ Note that processes executing as remote shells are also
+ identified.
+
+ Here's another example from a UnixWare 7.1.0 system.
+
+ tweeker: 1 = cd src/lsof4/scripts
+ tweeker: 9 = ./idrlogin.perl
+ Login Shell PID Via PID TTY From
+ abe ksh 9438 in.telnetd 9436 pts/3 lsof.itap.purdue.edu
+
+
+10. Watching an Ftp or Rcp Transfer
+===================================
+
+ The nature of the Internet being one of unpredictable performance
+ at times, occasionally you want to know if a file transfer, being
+ done by ftp or rcp, is making any progress.
+
+ To use lsof for watching a file transfer, you need to know the
+ PID of the file transfer process. You can use ps to find that.
+ Then use lsof,
+
+ $ lsof -p<PID>
+
+ to examine the files open to the transfer process. Usually the
+ ftp files or interest are at file descriptors 9 and 10 or 10 and
+ 11; for rcp, 3 and 4. They describe the network socket file and
+ the local data file.
+
+ If you want to watch only those file descriptors as the file
+ transfer progresses, try these lsof forms (for ftp in the example):
+
+ $ lsof -p<PID> -ad9,10 -r
+ or
+ $ lsof -p<PID> -ad10,11 -r
+
+ Some options need explaining:
+
+ -p<PID> specifies that lsof is to restrict its attention
+ to the process whose ID is <PID>. You can specify
+ a set of PIDs by separating them with commas.
+
+ $ lsof -p 1234,5678,9012
+
+ -a specifies that lsof is to AND its tests together.
+ The two tests that are specified are tests on the
+ PID and tests on file descriptions (``d9,10'').
+
+ d9,10 specifies that lsof is to test only file descriptors
+ 9 and 10. Note that the `-' is absent, since ``-a''
+ is a unary option and can be followed immediately
+ by another lsof option.
+
+ -r tells lsof to list the requested open file information,
+ sleep for a default 15 seconds, then list the open
+ file information again. You can specify a different
+ time (in seconds) after -r and override the default.
+ Lsof issues a short line of equal signs between
+ each set of output to distinguish it.
+
+ For an rcp transfer, the above example becomes:
+
+ $ lsof -p<PID> -ad3,4 -r
+
+
+11. Listing Open NFS Files
+==========================
+
+ Lsof will list all files open on remote file systems, supported
+ by an NFS server. Just use:
+
+ $ lsof -N
+
+ Note, however, that when run on an NFS server, lsof will not list
+ files open to the server from one of its clients. That's because
+ lsof can only examine the processes running on the machine where
+ it is called -- i.e., on the NFS server.
+
+ If you run lsof on the NFS client, using the -N option, it will
+ list files open by processes on the client that are on remote
+ NFS file systems.
+
+
+12. Listing Files Open by a Specific Login
+==========================================
+
+ If you're interested in knowing what files the processes owned
+ by a particular login name have open, lsof can help.
+
+ $ lsof -u<login>
+ or
+ $ lsof -u<User ID number>
+
+ You can specify either the login name or the UID associated with
+ it. You can specify multiple login names and UID numbers, mixed
+ together, by separating them with commas.
+
+ $ lsof -u548,abe
+
+ On the subject of login names and UIDs, it's worth noting that
+ lsof can be told to report either. By default it reports login
+ names; the -l option switches reporting to UIDs. You might want
+ to use -l if login name lookup is slow for some reason.
+
+ a. Ignoring a Specific Login
+ =============================
+
+ The -u option can also be used to direct lsof to ignore a
+ specific login name or UID, or a list of them. Simply prefix
+ the login names or UIDs with a `^' character, as you might do
+ in a regular expression. The `^' prefix is useful, for example,
+ when you want to have lsof ignore the files open to system
+ processes, owned by the root (UID 0) login. Try:
+
+ $ lsof -u ^root
+ or
+ $ lsof -u ^0
+
+
+13. Listing Files Open to a Specific Process Group
+==================================================
+
+ There's a Unix collection of processes called a process group.
+ The name indicates that the processes of the group have a common
+ association and are grouped so that a signal sent to one (e.g.,
+ a keyboard kill stroke) is delivered to all.
+
+ This causes Unix to create a two element process group:
+
+ $ lsof | less
+
+ You can use lsof to look at the open files of all members of a
+ process group, if you know the process group ID number. Assuming
+ that it is 12717 for the above example, this lsof command:
+
+ $ lsof -g12717 -adcwd
+
+ would produce on a Solaris 8 system:
+
+ $ lsof -g12717 -adcwd
+ COMMAND PID PGID USER FD TYPE DEVICE SIZE/OFF NODE NAME
+ sshd 11369 12717 root cwd VDIR 0,2 189 1449175 /tmp (swap)
+ sshd 12717 12717 root cwd VDIR 136,0 1024 2 /
+
+ The ``-g12717'' option specifies the process group ID of interest;
+ the ``-adcwd'' option specifies that options are to be ANDed and
+ that lsof should limit file output to information about current
+ working directory (``cwd'') files.
+
+
+14. When Lsof Seems to Hang
+===========================
+
+ On occasion when you run lsof it seems to hang and produce no
+ output. This may result from system conditions beyond the control
+ of lsof. Lsof has a number of options that may allow you to
+ bypass the blockage.
+
+ a. Kernel lstat(), readlink(), and stat() Blockages
+ ====================================================
+
+ Lsof uses the kernel (system) calls lstat(), readlink(), and
+ stat() to locate mounted file system information. When a file
+ system has been mounted from an NFS server and that server is
+ temporarily unavailable, the calls lsof uses may block in the
+ kernel.
+
+ Lsof will announce that it is being blocked with warning messages
+ (unless they have been suppressed by the lsof builder), but
+ only after a default waiting period of fifteen seconds has
+ expired for each file system whose server is unavailable. If
+ you have a number of such file systems, the total wait may be
+ unacceptably long.
+
+ You can do two things to shorten your suffering: 1) reduce the
+ wait time with the -S option; or 2) tell lsof to avoid the
+ kernel calls that might block by specifying the -b option.
+
+ $ lsof -S 5
+ or
+ $ lsof -b
+
+ Avoiding the kernel calls that might block may result in the
+ lack of some information that lsof needs to know about mounted
+ file systems. Thus, when you use -b, lsof warns that it might
+ lack important information.
+
+ The warnings that result from using -b (unless suppressed by
+ the lsof builder) can themselves be annoying. You can suppress
+ them by adding the -w option. (Of course, if you do, you won't
+ know what warning messages lsof might have issued.)
+
+ $ lsof -bw
+
+ Note: if the lsof builder suppressed warning message issuance,
+ you don't need to use -w to suppress them. You can tell what
+ the default state of message warning issuance is by looking at
+ the -h (help) output. If it says ``-w enable warnings'' then
+ warnings are disabled by default; ``-w disable warnings'', they
+ are enabled by default.
+
+ b. Problems with /dev or /devices
+ ==================================
+
+ Lsof scans the /dev or /devices branch of your file system to
+ obtain information about your system's devices. (The scan isn't
+ necessary when a device cache file exists.)
+
+ Sometimes that scan can take a very long time, especially if
+ you have a large number of devices, and if your kernel is
+ relatively slow to process the stat() system call on device
+ nodes. You can't do anything about the stat() system call
+ speed.
+
+ However, you can make sure that lsof is allowed to use its
+ device cache file feature. When lsof can use a device cache
+ file, it retains information it gleans via the stat() calls
+ on /dev or /devices in a separate file for later, faster
+ access.
+
+ The device cache file feature is described in the lsof man
+ page. See the DEVICE CACHE FILE, LSOF PERMISSIONS THAT AFFECT
+ DEVICE CACHE FILE ACCESS, DEVICE CACHE FILE PATH FROM THE -D
+ OPTION, DEVICE CACHE PATH FROM AN ENVIRONMENT VARIABLE,
+ SYSTEM-WIDE DEVICE CACHE PATH, PERSONAL DEVICE CACHE PATH
+ (DEFAULT), and MODIFIED PERSONAL DEVICE CACHE PATH sections.
+
+ There is also a separate file in the lsof distribution, named
+ 00DCACHE, that describes the device cache file in detail,
+ including information about possible security problems.
+
+ One final observation: don't overlook the possibility that your
+ /dev or /devices tree might be damaged. See if
+
+ $ ls -R /dev
+ or
+ $ ls -R /devices
+
+ completes or hangs. If it hangs, then lsof will probably hang,
+ too, and you should try to discover why ls hangs.
+
+ c. Host and Service Name Lookup Hangs
+ ======================================
+
+ Lsof can hang up when it tries to convert an Internet dot-form
+ address to a host name, or a port number to a service name. Both
+ hangs are caused by the lookup functions of your system.
+
+ An independent check for both types of hangs can be made with
+ the netstat program. Run it without arguments. If it hangs,
+ then it is probably having lookup difficulties. When you run
+ it with -n it shouldn't hang and should report network and port
+ numbers instead of names.
+
+ Lsof has two options that serve the same purpose as netstat's
+ -n option. The lsof -n option tells it to avoid host name
+ lookups; and -P, service name lookups. Try those options when
+ you suspect lsof may be hanging because of lookup problems.
+
+ $ lsof -n
+ or
+ $ lsof -P
+ or
+ $ lsof -nP
+
+ d. UID to Login Name Conversion Delays
+ =======================================
+
+ By default lsof converts User IDentification (UID) numbers to
+ login names when it produces output. That conversion process
+ may sometimes hang because of system problems or interlocks.
+
+ You can tell lsof to skip the lookup with the -l option; it
+ will then report UIDs in the USER column.
+
+ $ lsof -l
+
+
+15. Output for Other Programs
+=============================
+
+ The -F option allows you to specify that lsof should describe
+ open files with a special form of output, called field output,
+ that can be parsed easily by a subsequent program. The lsof
+ distribution comes with sample AWK, Perl 4, and Perl 5 scripts
+ that post-process field output. The lsof test suite has a C
+ library that could be adapted for use by C programs that want to
+ process lsof field output from an in-bound pipe.
+
+ The lsof manual page describes field output in detail in its
+ OUTPUT FOR OTHER PROGRAMS section. A quick look at a sample
+ script in the scripts/ subdirectory of the lsof distribution will
+ also give you an idea how field output works.
+
+ The most important thing about field output is that it is relatively
+ homogeneous across Unix dialects. Thus, if you write a script
+ to post-process field output for AIX, it probably will work for
+ HP-UX, Solaris, and Ultrix as well.
+
+
+16. The Lsof Exit Code and Shell Scripts
+========================================
+
+ When lsof exits successfully it returns an exit code based on
+ the result of its search for specified files. (If no files were
+ specified, then the successful exit code is 0 (zero).)
+
+ If lsof was asked to search for specific files, including any
+ files on specified file systems, it returns an exit code of 0
+ (zero) if it found all the specified files and at least one file
+ on each specified file system. Otherwise it returns a 1 (one).
+
+ If lsof detects an error and makes an unsuccessful exit, it
+ returns an exit code of 1 (one).
+
+ You can use the exit code in a shell script to search for files
+ on a file system and take action based on the result -- e.g.,
+
+ #!/bin/sh
+ lsof <file_system_name> > /dev/null 2>&1
+ if test $? -eq 0
+ then
+ echo "<file_system_name> has some users."
+ else
+ echo "<file_system_name> may have no users."
+ fi
+
+
+17. Strange messages in the NAME column
+=======================================
+
+ When lsof encounters problems analyzing a particular file, it may
+ put a message in the file's NAME column. Many of those messages
+ are explained in the 00FAQ file of the lsof distribution.
+
+ So consult 00FAQ first if you encounter a NAME column message you
+ don't understand. (00FAQ is a possible source of information
+ about other unfamiliar things in lsof output, too.)
+
+ If you can't find help in 00FAQ, you can use grep to look in the
+ lsof source files for the message -- e.g.,
+
+ $ cd .../lsof_4.76_src
+ $ grep "can't identify protocol" *.[ch]
+
+ The code associated with the message will usually make clear the
+ reason for the message.
+
+ If you have an lsof source tree that has been processed by the
+ lsof Configure script, you need grep only there. If, however,
+ your source tree hasn't been processed by Configure, you may
+ have to look in the top-level lsof source directory and in the
+ dialects sub-directory for the UNIX dialect you are using - e.g.,
+
+ $ cd .../lsof_4.76_src
+ $ grep "can't identify protocol" *.[ch]
+ $ cd dialects/Linux
+ $ grep "can't identify protocol" *.[ch]
+
+ In rare cases you may have to look in the lsof library, too --
+ e.g.,
+
+ $ cd .../lsof_4.76_src
+ $ grep "can't identify protocol" *.[ch]
+ $ cd dialects/Linux
+ $ grep "can't identify protocol" *.[ch]
+ $ cd ../../lib
+ $ grep "can't identify protocol" *.[ch]
+
+
+Options
+=======
+
+ The following appendices describe the lsof options in detail.
+
+
+A. Selection Options
+====================
+
+ Lsof has a rich set of options for selecting the files to be
+ displayed. These include:
+
+ -a tells lsof to AND the set of selection options that
+ are specified. Normally lsof ORs them.
+
+ For example, if you specify the -p<PID> and -u<UID>
+ options, lsof will display all files for the
+ specified PID or for the specified UID.
+
+ By adding -a, you specify that the listed files
+ should be limited to PIDs owned by the specified
+ UIDs -- i.e., they match the PIDs *and* the UIDs.
+
+ $ lsof -p1234 -au 5678
+
+ -c specifies that lsof should list files belonging
+ to processes having the associated command name.
+
+ Hint: if you want to select files based on more than
+ one command name, use multiple -c<name> specifications.
+
+ $ lsof -clsof -cksh
+
+ -d tells lsof to select by the associated file descriptor
+ (FD) set. An FD set is a comma-separated list of
+ numbers and the names lsof normally displays in
+ its FD column: cwd, Lnn, ltx, <number>, etc. See
+ the OUTPUT section of the lsof man page for the
+ complete list of possible file descriptors. Example:
+
+ $ lsof -dcwd,0,1,2
+
+ -g tells lsof to select by the associated process
+ group ID (PGID) set. The PGID set is a comma-separated
+ list of PGID numbers. When -g is specified, it also
+ enables the display of PGID numbers.
+
+ Note: when -g isn't followed by a PGID set, it
+ simply selects the listing of PGID for all processes.
+ Examples:
+
+ $ lsof -g
+ $ lsof -g1234,5678
+
+ -i tells lsof to display Internet socket files. If no
+ protocol/address/port specification follows -i,
+ lsof lists all Internet socket files.
+
+ If a specification follows -i, lsof lists only the
+ socket files whose Internet addresses match the
+ specification.
+
+ Hint: multiple addresses may be specified with
+ multiple -i options. Examples:
+
+ $ lsof -iTCP
+ $ lsof -i@lsof.itap.purdue.edu:sendmail
+
+ -N selects the listing of files mounted on NFS devices.
+
+ -U selects the listing of socket files in the Unix
+ domain.
+
+
+B. Output Options
+==================
+
+ Lsof has these options to control its output format:
+
+ -F produce output that can be parsed by a subsequent
+ program.
+
+ -g print process group (PGID) IDs.
+
+ -l list UID numbers instead of login names.
+
+ -n list network numbers instead of host names.
+
+ -o always list file offset.
+
+ -P list port numbers instead of port service names.
+
+ -s always list file size.
+
+
+C. Precautionary Options
+=========================
+
+ Lsof uses system functions that can block or take a long time,
+ depending on the health of the Unix dialect supporting it. These
+ include:
+
+ -b directs lsof to avoid system functions -- e.g.,
+ lstat(2), readlink(2), stat(2) -- that might block
+ in the kernel. See the BLOCKS AND TIMEOUTS
+ section of the lsof man page.
+
+ You might want to use this option when you have
+ a mount from an NFS server that is not responding.
+
+ -C tells lsof to ignore the kernel's name cache. As
+ a precaution this option will have little effect on
+ lsof performance, but might be useful if the kernel's
+ name cache is scrambled. (I've never seen that
+ happen.)
+
+ -D might be used to direct lsof to ignore an existing
+ device cache file and generate a new one from /dev
+ (and /devices). This might be useful if you have
+ doubts about the integrity of an existing device
+ cache file.
+
+ -l tells lsof to list UID numbers instead of login
+ names -- this is useful when UID to login name
+ conversion is slow or inoperative.
+
+ -n tells lsof to avoid converting Internet addresses
+ to host numbers. This might be useful when your
+ host name lookup (e.g., DNS) is inoperative.
+
+ -O tells lsof to avoid its strategy of forking to
+ perform potentially blocking kernel operations.
+ While the forking allows lsof to detect that a
+ block has occurred (and possibly break it), the
+ fork operation is a costly one. Use the -O option
+ with care, lest your lsof be blocked.
+
+ -P directs lsof to list port numbers instead of trying
+ to convert them to port service names. This might
+ be useful if port to service name lookups (e.g.,
+ via NIS) are slow or failing.
+
+ -S can be used to change the lstat/readlink/stat
+ timeout interval that governs how long lsof waits
+ for response from the kernel. This might be useful
+ when an NFS server is slow or unresponsive. When
+ lsof times out of a kernel function, it may have
+ less information to display. Example:
+
+ $ lsof -S2
+
+ -w tells lsof to avoid issuing warning messages, if
+ they are enabled by default, or enable them if they
+ are disabled by default. Check the -h (help) output
+ to determine their status. If it says ``-w enable
+ warnings'', then warning messages are disabled by
+ default; ``-w disable warnings'', they are enabled
+ by default.
+
+ This may be a useful option, for example, when you
+ specify -b, if warning messages are enabled, because
+ it will suppress the warning messages lsof issues
+ about avoiding functions that might block in the
+ kernel.
+
+
+D. Miscellaneous Lsof Options
+==============================
+
+ There are some lsof options that are hard to classify, including:
+
+ -? these options select help output.
+ -h
+
+ -F selects field output. Field output is a mode where
+ lsof produces output that can be parsed easily by
+ subsequent programs -- e.g., AWK or Perl scripts.
+ See ``15. Output for Other Programs'' for more
+ information.
+
+ -k specifies an alternate kernel symbol file -- i.e.,
+ where nlist() will get its information. Example:
+
+ $ lsof -k/usr/crash/vmunix.1
+
+ -m specifies an alternate kernel memory file from
+ which lsof will read kernel structures in place
+ of /dev/kmem or kvm_read(). Example:
+
+ $ lsof -m/usr/crash/vmcore.n
+
+ -r tells lsof to repeat its scan every 15 seconds (the
+ default when no associated value is specified). A
+ repeat time, different from the default, can follow
+ -r. Example:
+
+ $ lsof -r30
+
+ -v displays information about the building of the
+ lsof executable.
+
+ -- The double minus sign option may be used to
+ signal the end of options. It's particularly useful
+ when arguments to the last option are optional and
+ you want to supply a file path that could be confused
+ for arguments to the last option. Example:
+
+ $ lsof -g -- 1
+
+ Where `1' is a file path, not PGID ID 1.
+
+
+Vic Abell <abe@purdue.edu>
+March 27, 2006
--- /dev/null
+
+ Making and Installing lsof 4
+
+********************************************************************
+| The latest release of lsof is always available via anonymous ftp |
+| from lsof.itap.purdue.edu. Look in pub/tools/unix/lsof. |
+********************************************************************
+
+ Contents
+
+ Pre-built Lsof Binaries
+ Making Lsof
+ Other Configure Script Options
+ Environment Variables
+ Security
+ Run-time Warnings
+ Device Access Warnings
+ NFS Blocks
+ Caches -- Name and Device
+ Raw Sockets
+ Other Compile-time Definitions
+ The AFSConfig Script
+ The Inventory Script
+ The Customize Script
+ Cautions
+ Warranty
+ License
+ Bug Reports
+ The 00FAQ File
+ The lsof-l Mailing List
+ Field Output Example Scripts
+ Field Output C Library
+ Testing Lsof
+ Dialect Notes
+ AFS
+ AIX
+ Apple Darwin
+ Auspex LFS (no longer maintained)
+ BSDI BSD/OS
+ DEC OSF/1, Digital UNIX, Tru64 UNIX
+ FreeBSD
+ HP-UX
+ IPv6
+ Linux
+ NetBSD
+ NEXTSTEP and OPENSTEP
+ OpenBSD
+ Pyramid DC/OSx and Reliant UNIX (no longer available)
+ Caldera OpenUNIX
+ SCO OpenServer
+ SCO|Caldera UnixWare
+ Solaris 2.x, 7, 8, 9 and 10
+ Ultrix (no longer available)
+ Veritas VxFS and VxVM
+ User-contributed Dialect Support
+ Dialects No Longer Supported
+ Installing Lsof
+ Setuid-root Lsof Dialects
+ Setgid Lsof Dialects
+ Porting lsof 4 to a New UNIX Dialect
+ Quick Start to Using lsof
+ Cross-configuring Lsof
+ Environment Variables Affecting the Configure Script
+
+
+=======================
+Pre-built Lsof Binaries
+=======================
+
+Avoid using pre-built lsof binaries if you can; build your own
+instead.
+
+I do not support lsof binaries built and packaged by third parties nor
+lsof binaries built from anything but the latest lsof revision. (See
+the Bug Reports section for more information on the details of lsof
+support.)
+
+One important reasone for those support restrictions is that when lsof
+is built its Configure script tunes lsof to the features available on
+the building system, often embodied in supporting header files and
+libraries. If the building system doesn't have support for a
+particular feature, lsof won't be built to support the feature on any
+system.
+
+The Veritas VxFS file system is a good example of a feature that
+requires build-time support.
+
+UNIX dialect version differences -- Solaris 8 versus 9, AIX 4.3.3
+vesus 5.2, etc. -- can also render a pre-built lsof binary useless
+on a different version. So can kernel bit size.
+
+There are so many potential pitfalls to using an lsof binary
+improperly that I strongly recommend lsof be used only where it is
+built.
+
+
+===========
+Making Lsof
+===========
+
+ $ cd <lsof source directory>
+ $ ./Configure <your dialect's abbreviation>
+ $ make
+
+(Consult the 00FAQ and 00XCONFIG files of the lsof distribution
+for information about using make command invocations and environment
+variables to override lsof default Makefile strings.)
+
+This lsof distribution can be used with many UNIX dialects. However,
+it must be configured specifically for each dialect. Configuration
+is done in three ways: 1) by changing definitions in the machine.h
+header file of the UNIX dialect of interest; 2) by defining
+environment variable values prior to calling Configure (see the
+00XCONFIG file, the Environment Variabls and Environment Variables
+Affecting the Configure Script sections of this file); and 3) by
+running the Configure shell script found in the top level of the
+distribution directory.
+
+You may not need to change any machine.h definitions, but you might
+want to look at them anyway. Pay particular attention to the
+definitions that are discussed in the Security section of this
+file. Please read that section.
+
+The Configure script calls three other scripts in the lsof
+distribution: AFSConfig; Inventory; and Customize. The AFSConfig
+script is called for selected dialects (AIX, HP-UX, NEXTSTEP, and
+Solaris) to locate AFS header files and determine the AFS version.
+See The AFSConfig Script section of this file for more information.
+
+The Inventory script checks the completeness of the lsof distribution.
+Configure calls Inventory after it has accepted the dialect
+abbreviation, but before it configures the top-level directory for
+the dialect. See The Inventory Script section of this file for
+more information.
+
+Configure calls the Customize script after it has configured the
+top-level lsof directory for the declared dialect. Customize helps
+you modify some of the important compile-time definitions of
+machine.h. See the The Customize Script section.
+
+You should also think about where you will install lsof and its
+man page, and whom you will let execute lsof. Please read the
+Installing Lsof section of this file for information on installation
+considerations.
+
+Once you have inspected the machine.h file for the dialect for
+which you want to build lsof, and made any changes you need, run
+the Configure script, supplying it with the abbreviation for the
+dialect. (See the following table.) Configure selects the
+appropriate options for the dialect and runs the Mksrc shell script
+in the dialect sub-directory to construct the appropriate source
+files in the top-level distribution directory.
+
+Configure may also run the MkKernOpts script in the dialect
+sub-directory to propagate kernel build options to the dialect
+Makefile. This is done for only a few dialects -- e.g., DC/OSx,
+and Reliant UNIX.
+
+Configure creates a dialect-specific Makefile. You may want to
+inspect or edit this Makefile to make it conform to local conventions.
+If you want the Makefile to install lsof and its man page, you will
+have to create an appropriate install rule.
+
+Lsof may be configured using UNIX dialect abbreviations from the
+following table. Alternative abbreviations are indicated by a
+separating `|'. For example, for SCO OpenServer you can use either
+the ``osr'' or the ``sco'' abbreviation:
+
+ $ Configure osr
+ or
+ $ Configure sco
+
+ Abbreviations UNIX Dialect
+ ------------- ------------
+
+ aix IBM AIX 5.[23] and 5.3-ML1 using IBM's C Compiler
+ aixgcc IBM AIX 5.[12] and 5.3-ML1 using gcc
+ darwin Apple Darwin 7.x and 8.x for Power Macintosh systems
+ decosf DEC OSF/1, Digital UNIX, Tru64 UNIX 4.0 and 5.1
+ digital_unix Digital UNIX, DEC OSF/1, Tru64 UNIX 4.0 and 5.1
+ du Digital UNIX, DEC OSF/1, Tru64 UNIX 4.0 and 5.1
+ freebsd FreeBSD 4.x, 4.1x, 5.x and [67].x
+ hpux HP-UX 11.00, 11.11 and 11.23, using HP's C
+ Compiler, both /dev/kmem-based and PSTAT-based
+ hpuxgcc HP-UX 11.00, 11.11 and 11.23, using gcc, both
+ /dev/kmem-based and PSTAT-based
+ linux Linux 2.1.72 and above for x86-based systems
+ netbsd NetBSD 1.[456], 2.x and 3.x
+ next NEXTSTEP 3.[13]
+ nextstep NEXTSTEP 3.[13]
+ ns NEXTSTEP 3.[13]
+ nxt NEXTSTEP 3.[13]
+ openbsd OpenBSD 2.[89] and 3.[0-9]
+ openstep OPENSTEP 4.x
+ os OPENSTEP 4.x
+ osr SCO OpenServer Release 5.0.6, using the C compiler
+ from the SCO developer's kit
+ osrgcc SCO OpenServer Release 5.0.6, using gcc
+ osr6 SCO Openserver 6.0.0, using the SCO C compiler
+ sco SCO OpenServer Release 5.0.6, using the C compiler
+ from the SCO developer's kit
+ scogcc SCO OpenServer Release 5.0.6, using gcc
+ solaris Solaris 2.x, 7, 8, 9 and 10 using gcc
+ solariscc Solaris 2.x, 7, 8, 9 and 10 using Sun's cc
+ tru64 Tru64 UNIX, DEC OSF/1, Digital UNIX 4.0 and 5.1
+ unixware SCO|Caldera UnixWare 7.1.4
+ uw SCO|Caldera UnixWare 7.1.4
+
+If you have an earlier version of a dialect not named in the above
+list, lsof may still work on your system. I have no way of testing
+that myself. Try configuring for the named dialect -- e.g., if
+you're using Solaris 2.1, try configuring for Solaris 2.5.1.
+
+After you have configured lsof for your UNIX dialect and have
+selected options via the Customize script (See the The Customize
+Script section.) , use the make command to build lsof -- e.g.,
+
+ $ make
+
+
+Other Configure Script Options
+==============================
+
+There are three other useful options to the Configure script besides
+the dialect abbreviation:
+
+ -clean may be specified to remove all traces of
+ a dialect configuration, including the
+ Makefile, symbolic links, and library files.
+
+ -h may be specified to obtain a list of
+ -help Configure options, including dialect
+ abbreviations.
+
+ -n may be specified to stop the Configure
+ script from calling the Customize and
+ Inventory scripts.
+
+ Caution: -n also suppresses the AFSConfig
+ step.
+
+
+
+Environment Variables
+=====================
+
+Lsof configuration, building, and execution may be affected by
+environment variable settings. See the Definitions That Affect
+Compilation section in the 00PORTING file, the General Environment
+Variables section in the 00XCONFIG file, the Dialect-Specific
+Environment Variables section in the 00XCONFIG file, and the
+Environment Variables Affecting the Configure Script section of
+this file for more information.
+
+Note in the General Environment Variables section of the 00XCONFIG
+file that there are five environment variables that can be used to
+pre-define values in lsof's -v output: LSOF_BLDCMT, LSOF_HOST,
+LSOF_LOGNAME, LSOF_SYSINFO, and LSOF_USER.
+
+
+Security
+========
+
+If the symbol HASSECURITY is defined, a security mode is enabled,
+and lsof will allow only the root user to list all open files.
+Non-root users may list only open files whose processes have the
+same user ID as the real user ID of the lsof process (the one that
+its user logged on with).
+
+However, if HASNOSOCKSECURITY is also defined, anyone may list
+anyone else's open socket files, provided their listing is enabled
+with the "-i" option.
+
+Lsof is distributed with the security mode disabled -- HASSECURITY
+is not defined. (When HASSECURITY is not defined, the definition
+of HASNOSOCKSECURITY has no meaning.) You can enable the security
+mode by defining HASSECURITY in the Makefile or in the machine.h
+header file for the specific dialect you're using -- e.g.
+dialects/aix/machine.h.
+
+The Customize script, run by Configure when it has finished its
+work, gives you the opportunity to define HASSECURITY and
+HASNOSOCKSECURITY. (See the The Customize Script section.)
+
+The lsof -h output indicates the state HASSECURITY and HASNOSOCKSECURITY
+had when lsof was built, reporting:
+
+ "Only root can list all files;"
+ if HASSECURITY was defined and HASNOSOCKSECURITY wasn't
+ defined;
+
+ "Only root can list all files, but anyone can list socket files."
+ if HASSECURITY and HASNOSOCKSECURITY were both defined;
+
+ "Anyone can list all files;"
+ if HASSECURITY wasn't defined. (The definition of
+ HASNOSOCKSECURITY doesn't matter when HASSECURITY isn't
+ defined.)
+
+You should carefully consider the implications of using the default
+security mode. When lsof is compiled in the absence of the
+HASSECURITY definition, anyone who can execute lsof may be able to
+see the presence of all open files. This may allow the lsof user
+to observe open files -- e.g., log files used to track intrusions
+-- whose presence you would rather not disclose.
+
+All pre-compiled binaries on lsof.itap.purdue.edu and mirrored from
+it were constructed without the HASSECURITY definition.
+
+As distributed, lsof writes a user-readable and user-writable device
+cache file in the home directory of the real user ID executing
+lsof. There are other options for constructing the device cache file
+path, and they each have security implications.
+
+The 00DCACHE file in the lsof distribution discusses device cache
+file path construction in great detail. It tells how to disable
+the various device cache file path options, or how to disable the
+entire device cache file feature by removing the HASDCACHE definition
+from the dialect's machine.h file. There is also information on
+the device cache file feature in the 00FAQ file. (The 00DCACHE
+and 00FAQ files are part of the lsof distribution package.)
+
+The Customize script, run by Configure after it has finished its
+work, gives you the opportunity to change the compile-time options
+related to the device cache file. (See The Customize Script
+section.)
+
+Since lsof may need setgid or setuid-root permission (See the Setgid
+Lsof Dialects and Setuid-root Lsof Dialects sections.), its security
+should always be viewed with skepticism. Lest the setgid and
+setuid-root permissions allow lsof to read kernel name list or
+memory files, declared with the -k and -m options, that the lsof
+user can't normally access, lsof uses access(2) to establish its
+real user's authority to read such files when it can't surrender
+its power before opening them. This change was added at the
+suggestion of Tim Ramsey.
+
+Lsof surrenders setgid permission on most dialects when it has
+gained access to the kernel's memory devices. There are exceptions
+to this rule, and some lsof implementations need to run setuid-root.
+(The Setgid Lsof Dialects and Setuid-root Lsof Dialects sections
+contains a list of lsof implementations and the permissions
+recommended in the distribution's Makefiles.)
+
+The surrendering of setgid permission is controlled by the WILLDROPGID
+definition in the dialect machine.h header files.
+
+In the end you must judge for yourself and your installation the
+risks that lsof presents and restrict access to it according to
+your circumstances and judgement.
+
+
+Run-time Warnings
+=================
+
+Lsof can issue warning messages when it runs -- e.g., about the
+state of the device cache file, about an inability to access an
+NFS file system, etc. Issuance of warnings are enabled by default
+in the lsof distribution.
+
+Issuance or warnings may be disabled by default by defining
+WARNINGSTATE in the dialect's machine.h. The Customize script may
+also be used to change the default warning message issuance state.
+(See The Customize Script section.)
+
+The ``-w'' option description of the ``-h'' option (help) output
+will indicate the default warning issuance state. Whatever the
+state may be, it can be reversed with ``-w''.
+
+
+Device Access Warnings
+======================
+
+When lsof encounters a /dev (or /devices) directory, one of its
+sub-directories, or one of their files that it cannot access with
+opendir(3) or stat(2), it issues a warning message and continues.
+Lsof will be more likely to issue such a warning when it has been
+installed with setgid(<some group name>) permission; it won't have
+trouble if it has been installed with setuid(root) permission or
+is being run under the root login.
+
+The lsof caller can inhibit or enable the warning with the -w
+option, depending on the issuance state of run-time warnings. (See
+the Run-time Warnings section.)
+
+The warning messages do not appear when lsof obtains device
+information from a device cache file that it has built and believes
+to be current or when warning message issuance is disabled by
+default. (See the "Caches -- Name and Device" section for more
+information on the device cache file.)
+
+The lsof builder can inhibit the warning by disabling the definition
+of WARNDEVACCESS in the dialect's machine.h or disable all warnings
+by defining WARNINGSTATE. WARNDEVACCESS is defined by default for
+most dialects. However, some dialects have some device directory
+elements that are private -- e.g., HP-UX -- and it is more convenient
+for the lsof user if warning messages about them are inhibited.
+
+Output from lsof's -h option indicates the status of WARNDEVACCESS.
+If it was defined when lsof was compiled, this message will appear:
+
+ /dev warnings = enabled
+
+If WARNDEVACCESS was not defined when lsof was compiled, this
+message will appear instead:
+
+ /dev warnings = disabled
+
+The Customize script, run by Configure after it has finished its
+work, gives you the opportunity to change the WARNDEVACCESS
+definition. (See The Customize Script section.)
+
+
+NFS Blocks
+==========
+
+Lsof is susceptible to NFS blocks when it tries to lstat() mounted
+file systems and when it does further processing -- lstat() and
+readlink() -- on its optional file and file system arguments.
+
+Lsof tries to avoid being stopped completely by NFS blocks by doing
+the lstat() and readlink() functions in a child process, which
+returns the function response via a pipe. The lsof parent limits
+the wait for data to arrive in the pipe with a SIGALRM, and, if
+the alarm trips, terminates the child process with a SIGINT and a
+SIGKILL.
+
+This is as reliable and portable a method for breaking NFS deadlocks
+as I have found, although it still fails under some combinations
+of NFS version, UNIX dialect, and NFS file system mount options.
+It generally succeeds when the "intr" or "soft" mount options are
+used; it generally fails when the "hard" mount option is used.
+
+When lsof cannot kill the child process, a second timeout causes
+it to stop waiting for the killed child to complete. While the
+second timeout allows lsof to complete, it may leave behind a hung
+child process. Unless warnings are inhibited by default or with
+the -w option, lsof reports the possible hung child.
+
+NFS block handling was updated with suggestions made by Andreas
+Stolcke. Andreas suggested using the alternate device numbers that
+appear in the mount tables of some dialects when it is not possible
+to stat(2) the mount points.
+
+The -b option was added to direct lsof to avoid the stat(2) and
+readlink(2) calls that might block on NFS mount points and always
+use the alternate device numbers. If warning message issuance is
+enabled and you don't want warning messages about what lsof is
+doing, use the -w option, too.
+
+The -O option directs lsof to avoid doing the potentially blocking
+operations in child processes. Instead, when -O is specified, lsof
+does them directly. While this consumes far less system overhead,
+it can cause lsof to hang, so I advise you to use -O sparingly.
+
+
+Caches -- Name and Device
+==========================
+
+Robert Ehrlich suggested that lsof obtain path name components for
+open files from the kernel's name cache. Where possible, lsof
+dialect implementations do that. The -C option inhibits kernel
+name cache examination.
+
+Since AFS apparently does not use the kernel's name cache, where
+lsof supports AFS it is unable to identify AFS files with path name
+components.
+
+Robert also suggested that lsof cache the information it obtains
+via stat(2) for nodes in /dev (or /devices) to reduce subsequent
+running time. Lsof does that, too.
+
+In the default distribution the device cache file is stored in
+.lsof_hostname, mode 0600, in the home directory of the login of
+the user ID that executes lsof. The suffix, hostname, is the first
+component of the host's name returned by gethostname(2). If lsof
+is executed by a user ID whose home directory is NFS-mounted from
+several hosts, the user ID's home directory may collect several
+device cache files, one for each host from which it was executed.
+
+Lsof senses accidental or malicious damage to the device cache file
+with extensive integrity checks, including the use of a 16 bit CRC.
+It also tries to sense changes in /dev (or /devices) that indicate
+the device cache file is out of date.
+
+There are other options for forming the device cache file path.
+Methods the lsof builder can use to control and employ them are
+documented in the separate 00DCACHE file of the lsof distribution.
+
+
+Raw Sockets
+===========
+
+On many UNIX systems raw sockets use a separate network control
+block structure. Display of files for applications using raw
+sockets -- ping, using ICMP, for example -- need special support
+for displaying their information. This support is so dialect-specific
+and information to provide it so difficult to find that not all
+dialect revisions of lsof handle raw sockets completely.
+
+
+Other Compile-time Definitions
+==============================
+
+The machine.h and dlsof.h header files for each dialect contains
+definitions that affect the compilation of lsof. Check the
+Definitions That Affect Compilation section of the 00PORTING file
+of the lsof distribution for their descriptions. (Also see The
+Customize Script section.)
+
+
+The AFSConfig Script
+====================
+
+Lsof supports AFS on some combinations of UNIX dialect and AFS
+version. See the AFS section of this document for a list of
+supported combinations.
+
+When configuring for dialects where AFS is supported, the Configure
+script calls the AFSConfig script to determine the location of AFS
+header files and the AFS version. Configure will not call AFSConfig,
+even for the selected dialects, unless the file /usr/vice/etc/ThisCell
+exists.
+
+The AFS header file location is recorded in the AFSHeaders file;
+version, AFSVersion. Once these values have been recorded, Configure
+can be told to skip the calling of AFSConfig by specifying its
+(Configure's) -n option.
+
+
+The Inventory Script
+====================
+
+The lsof distribution contains a script, called Inventory, that
+checks the distribution for completeness. It uses the file 00MANIFEST
+in the distribution as a reference point.
+
+After the Configure script has accepted the dialect abbreviation,
+it normally calls the Inventory script to make sure the distribution
+is complete.
+
+After Inventory has run, it creates the file ".ck00MAN" in the
+top-level directory to record for itself the fact that the inventory
+has been check. Should Inventory be called again, it senses this
+file and asks the caller if another check is in order, or if the
+check should be skipped.
+
+The -n option may be supplied to Configure to make it bypass the
+calling of the Inventory script. (The option also causes Configure
+to avoid calling the Customize script.)
+
+The lsof power user may want to define (touch) the file ".neverInv".
+Configure avoids calling the Inventory script when ".neverInv"
+exists.
+
+
+The Customize Script
+====================
+
+Normally when the Configure script has finished its work, it calls
+another shell script in the lsof distribution called Customize.
+(You can tell Configure to bypass Customize with its -n option.)
+
+Customize leads you through the specification of these important
+compile-time definitions for the dialect's machine.h header file:
+
+ HASDCACHE device cache file control
+ HASENVDC device cache file environment
+ variable name
+ HASPERSDC personal device cache file path
+ format
+ HASPERSDCPATH name of environment variable that
+ provides an additional component
+ of the personal device cache file
+ path
+ HASSYSDC system-wide device cache file path
+ HASKERNIDCK the build-time to run-time kernel
+ identity check
+ HASSECURITY the security option
+ HASNOSOCKSECURITY the open socket listing option whe
+ HASSECURITY is defined
+ WARNDEVACCESS /dev (or /devices) warning message
+ control
+ WARNINGSTATE warning message issuance state
+
+The Customize script accompanies its prompting for entry of new
+values for these definitions with brief descriptions of each of
+them. More information on these definitions may be found in this
+file or in the 00DCACHE and 00FAQ files of the lsof distribution.
+
+You don't need to run Customize after Configure. You can run it
+later or you can edit machine.h directly.
+
+The -n option may be supplied to Configure to make it bypass the
+calling of the Customize script. (The option also causes Configure
+to avoid calling the Inventory script.)
+
+The lsof power user may want to define (touch) the file ".neverCust".
+Configure avoids calling the Customize script when ".neverCust"
+exists.
+
+Customize CAUTION: the Customize script works best when it is
+applied to a newly configured lsof source base -- i.e., the machine.h
+header file has not been previously modified by the Customize
+script. If you have previously configured lsof, and want to rerun
+the Customize script, I recommend you clean out the previous
+configuration and create a new one:
+
+ $ Configure -clean
+ $ Configure <dialect_abbreviation>
+ ...
+ Customize in response to the Customize script prompts.
+
+
+Cautions
+========
+
+Lsof is a tool that is closely tied to the UNIX operating system
+version. It uses header files that describe kernel structures and
+reads kernel structures that typically change from OS version to
+OS version, and even within a version as vendor patches are applied.
+
+DON'T TRY TO USE AN LSOF BINARY, COMPILED FOR ONE UNIX OS VERSION,
+ON ANOTHER. VENDOR PATCHES INFLUENCE THE VERSION IDENTITY.
+
+On some UNIX dialects lsof versions may be even more restricted by
+architecture type.
+
+The bottom line is use lsof where you built it. If you intend to
+use a common lsof binary on multiple systems, make sure all systems
+run exactly the same OS version and have exactly the same patches.
+
+
+Warranty
+========
+
+Lsof is provided as-is without any warranty of any kind, either
+expressed or implied, including, but not limited to, the implied
+warranties of merchantability and fitness for a particular purpose.
+The entire risk as to the quality and performance of lsof is with
+you. Should lsof prove defective, you assume the cost of all
+necessary servicing, repair, or correction.
+
+
+License
+=======
+
+Lsof has no license. Its use and distribution are subject to these
+terms and conditions, found in each lsof source file. (The copyright
+year in or format of the notice may vary slightly.)
+
+ /*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette,
+ * Indiana 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American
+ * Telephone and Telegraph Company or the Regents of the
+ * University of California.
+ *
+ * Permission is granted to anyone to use this software for
+ * any purpose on any computer system, and to alter it and
+ * redistribute it freely, subject to the following
+ * restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible
+ * for any consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented,
+ * either by explicit claim or by omission. Credit to the
+ * authors and Purdue University must appear in documentation
+ * and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must
+ * not be misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+Bug Reports
+===========
+
+Now that the obligatory disclaimer is out of the way, let me hasten to
+add that I accept lsof bug reports and try hard to respond to them. I
+will also consider and discuss requests for new features, ports to new
+dialects, or ports to new OS versions.
+
+PLEASE DON'T SEND BUG REPORTS ABOUT LSOF TO THE UNIX DIALECT OR DIALECT
+OPTION VENDOR.
+
+At worst such bug reports will confuse the vendor; at best, the vendor
+will forward the bug report to me.
+
+PLEASE DON'T SEND BUG REPORTS ABOUT LSOF BINARIES BUILT OR DISTRIBUTED
+BY SOMEONE ELSE, BECAUSE I CAN'T SUPPORT THEM.
+
+I do support binaries I built, obtained ONLY from lsof.itap.purdue.edu.
+Before reporting a problem with a lsof.itap.purdue.edu binary, please
+verify the correctness of the signatures found in its associated
+CHECKSUMS file.
+
+Before you send me a bug report, please do these things:
+
+ * Make sure you try the latest lsof revision.
+
+ + Download the latest revision from:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof
+
+ + Verify the signatures of what you have downloaded;
+
+ + While connected to lsof.itap.purdue.edu, check for patches:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/patches
+
+ + If patches exist, install them in the latest revision
+ you just downloaded. Then build the latest revision and
+ see if it fixes your bug.
+
+ * If you're having trouble compiling lsof with gcc, try the
+ UNIX dialect vendor's compiler. I don't have access to gcc on
+ all test systems, so my support for it is hit-and-miss, and so
+ is my ability to respond to gcc compilation problem reports.
+
+ * Check the lsof frequently asked questions file, 00FAQ,
+ to see if there's a question and answer relevant to your
+ problem.
+
+ * Make sure you're running the lsof you think you are by
+ checking the path to it with which(1). When in doubt, use an
+ absolute path to lsof. Make sure that lsof binary has
+ sufficient permissions to do what you ask, including internal
+ permissions given it (e.g., restrictions on what files lsof may
+ report for whom) during its build.
+
+When you send a bug report, make sure you include output from your
+running of lsof's Configure script. If you were able to compile a
+running lsof, please also include:
+
+ * Output from which(1) that shows the absolute path to the
+ lsof binary in question;
+
+ * Output from running lsof with its -h and -v options at
+ lsof's absolute path;
+
+ * Output from "ls -l" directed to lsof's absolute path.
+
+If you weren't able to compile a running lsof, please send me: the
+compiler error output; identification of the lsof revision you're using
+(contents of the lsof version.c file); identification of your system
+(full uname output or output from whatever other tool identifies the
+system); and compiler identification (e.g., gcc -v output).
+
+Either set of output will help me understand how lsof was configured
+and what UNIX dialect and lsof revision is involved.
+
+Please send all bug reports, requests, etc. to me via e-mail at
+<abe@purdue.edu>. Make sure "lsof" appears in the "Subject:" line so
+my e-mail filter won't classify your letter as Spam.
+
+
+The 00FAQ File
+==============
+
+The lsof distribution contains an extensive frequently asked
+questions file on lsof features and problems. I recommend you
+consult it before sending me e-mail. Use your favorite editor or
+pager to search 00FAQ -- e.g., supplying as a search argument some
+fixed text from an lsof error message.
+
+
+The lsof-l Mailing List
+=======================
+
+Information about lsof, including notices about the availability
+of new revisions, may be found in mailings of the lsof-l listserv.
+For more information about it, including instructions on how to
+subscribe, read the 00LSOF-L file of the lsof distribution.
+
+
+Field Output Example Scripts
+============================
+
+Example AWK and Perl 4 or 5 scripts for post-processing lsof field
+output are locate in the scripts sub-directory of the lsof distribution.
+The scripts sub-directory contains a 00README file with information
+about the scripts.
+
+
+Field Output C Library
+======================
+
+The lsof test suite (See "Testing Lsof."), checks basic lsof
+operations using field output. The test suite has its own library
+of C functions for common test program operations, including
+processing of field output. The library or selections of its
+functions could be adapted for use by C programs that want to
+process lsof field output. See the library in the file LTlib.c
+in the tests/ sub-directory
+
+
+Testing Lsof
+============
+
+Lsof has an automated test suite in the tests/ sub-directory that
+can be used to test some basic lsof features -- once lsof has been
+configured and made. Tests are arranged in three groups: basic
+tests that should run on all dialects; standard tests that should
+run on all dialects; and optional tests that may not run on all
+dialects or may need special resources to run. See 00TEST for more
+information.)
+
+CAUTION!!! Before you attempt to use the test suite make sure that
+the lsof you want to test can access the necessary kernel resources
+-- e.g., /dev/mem, /dev/kmem, /proc, etc. Usually you want to test
+the lsof you just built, so this is an important check. (See
+00TEST.)
+
+To run the basic and standard tests, using the lsof in the parent
+directory of tests/, do this:
+
+ $ cd tests
+ $ make test
+ or $ make std
+ or $ make standard
+
+The basic and standard tests may be run as silently as possible,
+using the lsof in the parent directory of tests/, with:
+
+ $ cd tests
+ $ make auto
+
+This is the "automatic" test mode, designed for use by scripts that
+build lsof. The caller is expected to test the make exit code to
+determine if the tests succeeded. The caller should divert standard
+output and standard error to /dev/null to suppress make's error
+exit message.
+
+The optional tests may be run, using the lsof in the parent directory
+of tests/, with:
+
+ $ cd tests
+ $ make opt
+ or $ make optional
+
+It's possible to excute individual tests, too. See the 00TEST file
+of this distribution for more informaiton on the tests, what they
+do, and how to run and possibly customize each test.
+
+It's possible to run the tests, using an lsof other than the one
+in the parent directory of /tests, too. See 00TEST for information
+about using the LT_LSOF_PATH environment variable to do that.
+
+
+=============
+Dialect Notes
+=============
+
+
+AFS
+===
+
+Lsof recognizes AFS files on the following combinations of UNIX
+dialect and AFS versions:
+
+ AIX 4.1.4 (AFS 3.4a)
+ Linux 1.2.13 (AFS 3.3)
+ NEXTSTEP 3.2 (AFS 3.3) (untested on recent lsof revisions)
+ Solaris 2.6 (AFS 3.4a)
+ Ultrix 4.2 RISC (AFS 3.2b) (no longer available)
+
+Lsof has not been tested under other combinations -- e.g. HP-UX
+10.10 and AFS 3.4a -- and probably won't even compile there. Often
+when a UNIX dialect version or AFS version changes, the new header
+files come into conflict, causing compiler objections.
+
+
+AIX
+===
+
+Specify the aix Configure abbreviation for AIX 4.1.[45], 4.2[.1],
+4.3[.123], 5L, and 5.[123].
+
+Specify aixgcc on AIX above 4.1 to use the gcc compiler. (Gcc can't be
+used to compile lsof on AIX 4.1 and below because of kernel structure
+alignment differences between it and xlc.) Gcc results sometimes
+depend on the version of the gcc compiler that is used.
+
+Compilation of lsof with gcc on AIX 4.3[.123], 5L, and 5.[123] has been
+sparsely tested with varying degrees of success: it has been reported
+to succeed on AIX 4.3.3 and 32 bit Power AIX 5.1; to fail on ia64 AIX
+5.1 and 64 bit Power AIX 5.1; and to succeed on 32 and 64 bit Power AIX
+5.2. Lsof compilation with gcc hasn't been tested on AIX 5.3.
+
+At revision 4.61 and above lsof is configured and built to match the
+bit size of the kernel of Power architecture AIX 5.1 systems. Lsof
+binaries built for 32 and 64 bit kernels are not interchangeable. See
+00FAQ for more information.
+
+The Configure script uses /usr/bin/oslevel to determine the AIX version
+for AIX less than 5 and ``uname -rv'' for AIX 5 and higher. If
+/usr/bin/oslevel isn't executable on AIX less than 5, the Configure
+script issues a warning message and uses ``uname -rv'' to determine the
+AIX version.
+
+When Configure must use ``uname -rv'' on AIX less than 5 to determine
+the AIX version, the result will lack a correct third component --
+e.g., the `4' of ``4.1.4''. If your AIX less than 5 system lacks lacks
+an executable oslevel, I suggest you edit the Configure-produced
+Makefile and complete the _AIXV definition in the CFGF string.
+
+By default lsof avoids using the kernel's readx() function, causing
+it to be unable to report information on some text and library file
+references. The ``-X'' option allows the lsof user to ask for the
+information readx() supplies.
+
+Lsof avoids readx() to avoid the possibility of triggering a kernel
+problem, known as the Stale Segment ID kernel bug. Kevin Ruderman
+reported this bug to me. The bug shows up when the kernel's
+dir_search() function hangs, hanging the application process that
+called it so completely that the application process can neither
+be killed nor stopped. The hang is the consequence of another
+process (perhaps lsof) making legitimate use of the kernel's readx()
+function to access the kernel memory that dir_search() is examining.
+IBM has indicated they have no plans to fix the bug.
+
+A fuller discussion of this bug may be found in the 00FAQ file of
+the lsof distribution. There you will find a description of the
+Stale Segment ID bug, the APAR on it, and a discussion of the
+sequence of events that exposes it.
+
+I added the ``-X'' function so you can tell lsof to use readx(),
+but if you use ``-X'', you should be alert to its possibly serious
+side effects. Although readx() is normally disabled, its state is
+controlled with the HASXOPT, HASXOPT_ROOT, and HASXOPT_VALUE
+definitions in dialects/aix/machine.h, and you can change its
+default state by changing those definitions. You can also change
+HASXOPT_ROOT via the Customize script.
+
+You can also compile lsof with readx() use permanently enabled or
+disabled -- see the comments about the definitions in the
+dialects/aix/machine.h header file. You may want to permanently
+disable lsof's use of readx() if you plan to make lsof publicly
+executable. You can also restrict -X to processes whose real UID
+is root by defining HASXOPT_ROOT.
+
+I have never seen lsof cause the Stale Segment ID bug to occur and
+haven't had a report that it has, but I believe there is a possibility
+it could.
+
+AFS support for AIX was added with help help from Bob Cook and Jan
+Tax who provided test systems.
+
+Henry Grebler and David J. Wilson helped with lsof for AIX 4.2.
+
+Bill Pemberton provided an AIX 4.3 test system. Andrew Kephart
+and Tom Weaver provided AIX 4.3 technical assistance. Niklas
+Edmundsson did 4.3.1 testing. Doug Crabill provided an AIX 4.3.2
+test system. Jeff W. Stewart provided an AIX 4.3.3 test system.
+
+The SMT file type for AIX 4.1.[45], 4.2[.1], and 4.3[.12] is my
+fabrication. See the 00FAQ file more information on it.
+
+Loc Le and Nasser Momtaheni of IBM provided test systems for AIX 5L and
+5.1. Lsof for AIX 5L and 5.1 needs setuid-root permission to process
+the -X option on systems whose architecture type is ia64.
+
+Dale Talcott of Purdue provided AIX 5.1 and 5.2 test systems. Dale and
+John Jackson of Purdue provided an AIX 5.3 test system.
+
+
+Apple Darwin
+============
+
+The Apple Darwin port was provided by Allan Nathanson for version
+1.2. Allan also arranged for access to a test system for maintenance
+and regression testing. Dale Talcott provided a test system, too.
+
+Allan supplied patches for updates to 1.4, 5.x, 6.x, 7.x and 8.x.
+
+
+BSDI BSD/OS
+===========
+
+As of lsof revision 4.77 support for BSDI BSD/OS has been
+discontinued. Lsof revision 4.76 with BSDI BSD/OS support may be found
+on lsof.itap.purdue.edu in pub/tools/unix/lsof/src.
+
+
+DEC OSF/1, Digital UNIX, Tru64 UNIX
+===================================
+
+Robert Benites, Dean Brock, Angel Li, Dwight McKay, Berkley Shands,
+Ron Young and Steve Wilson have kindly provided test systems.
+Jeffrey Mogul has provided technical assistance. Dave Morrison
+and Lawrence MacIntyre did Digital UNIX V3.2 testing.
+
+Lsof supports the ADVFS/MSFS layered file system product. Lsof
+can locate all the open files of an ADVFS/MSFS file system when
+its path is specified, provided the file system is listed in
+/etc/fstab with an ``advfs'' type. (This /etc/fstab caveat applies
+only to Digital UNIX 2.0.) At Digital UNIX 4.0 and Tru64 UNIX,
+using code provided by David Brock, lsof 4.20 and above can locate
+ADVFS file paths.
+
+Testing of lsof on DEC OSF/1 and Digital UNIX 4.0 ended with lsof
+revision 4.74. Hence, the lsof documentation has dropped the claim
+that it works there. For a distribution of lsof 4.74 that was tested
+on DEC OSF/1 and Digital UNIX 4.0, check pub/tools/unix/lsof/OLD/src
+on the lsof ftp home, lsof.itap.purdue.edu.
+
+Lsof revisions past 4.74 have only been tested on Tru64 UNIX 5.1.
+
+
+FreeBSD
+=======
+
+Bill Bormann of Purdue University provided access to several FreeBSD
+test systems. Ade Barkah, John Clear, Ralph Forsythe, Michael
+Haro, Kurt Jaeger, and William McVey have also provided FreeBSD
+test systems.
+
+The FreeBSD distribution header files are augmented by header files
+in the dialects/freebsd/include directory.
+
+David O'Brien maintains the lsof FreeBSD port package.
+
+
+HP-UX
+=====
+
+Lsof has two HP-UX bases: /dev/kmem for HP-UX 11.0 and earlier;
+and PSTAT for HP-UX 11.11 and later. The lsof Configure script
+will pick the appropriate base.
+
+To use the CCITT x.25 socket support for HP-UX, you must have the
+x.25 header files in /etc/conf/x25
+
+Pasi Kaara helped with the HP-UX port, especially with its CCITT
+x.25 socket support.
+
+Richard Allen provided HP-UX 10.x and 11.x test systems, as did
+Mark Bixby, and Elias Halldor Agustsson. Marc Winkler helped test
+the 10.20 port. Richard J. Rauenzahn provided a 64 bit HP-UX 11
+test system and an HP-UX 11.11 development system.
+
+AFS support for HP-UX was added thanks to help from Chaskiel Moses
+Grundman, who provided a test system.
+
+The /dev/kmem-based HP-UX 11.00 support is extremely fragile. It
+depends on privately developed kernel structure definitions. (See
+.../dialects/hpux/hpux11 for the header files making the definitions.)
+Those header files and their definitions will not be updated by
+HP-UX 11.00 patches, making it likely that any patch changing a
+kernel structure critical to lsof will break lsof in some way.
+
+It's possible to build a 64 bit lsof for 64 bit HP-UX 11.00 with
+gcc, but you must have a gcc compiler capable of producing 64 bit
+executables. See the 00FAQ file for more information.
+
+The PSTAT-based lsof for HP-UX 11.11 and later is much more solid.
+I am indebted to the vision of HP for providing an lsof kernel API
+through the PSTAT implementation. Specifically I appreciate the
+help of HP staff members Carl Davidson, Louis Huemiller, Rich
+Rauenzahn, and Sailu Yallapragada that made PSTAT-based HP-UX lsof
+possible.
+
+
+IPv6
+====
+
+Lsof has IPv6 support that has been tested for these UNIX dialects:
+AIX 4.3.x; Apple Darwin 5.[12] and 6.0; the INRIA and KAME FreeBSD IPv6
+implementations; PSTAT-based HP-UX; /proc-based Linux; the INRIA and
+KAME NetBSD implementations; and Solaris 8 and 9. Lsof has IPv6
+support that hasn't been tested for: OpenBSD (KAME); OpenUNIX 8; Tru64
+Unix 5.[01]; and UnixWare 7.1.[34].
+
+Please let me know if your UNIX dialect has IPv6 support and I'll
+see if it can be supported by lsof.
+
+
+Linux
+=====
+
+Tim Korb, Steve Logue, Joseph J. Nuspl Jr., and Jonathan Sergent
+have provided Linux test systems.
+
+Michael Shields helped add and test automatic handling of ELF/COFF
+form names in /System.map, Marty Leisner and Keith Parks have helped
+test many lsof revisions. Marty has provided valuable suggestions,
+Linux hints, and code, too.
+
+The 00FAQ file gives some Linux tips, including information on
+coping with system map file problems.
+
+To determine the state of the Linux 2.1.x C library lseek() function,
+the lsof Configure script runs a test program that must have
+permission to read /dev/kmem. The test determines if the lseek()
+function properly handles kernel offsets, which appear to be negative
+because their high order bit is set. If the lseek() test reveals
+a faulty lseek(), Configure activates the use of a private lseek()
+function for kernel offset positioning. See the Linux problems
+section of the 00FAQ file of the lsof distribution for more
+information.
+
+
+NetBSD
+======
+
+Greg Earle and Paul Kranenburg have assisted with the NetBSD ports.
+Paul has provided test systems. Ray Phillips provided a NetBSA
+Alpha test system. Andrew Brown also provided a test system.
+
+The NetBSD dialect version of lsof is compiled using the dialect
+sources it shares with OpenBSD in the n+obsd dialect sub-directory.
+
+
+NEXTSTEP and OPENSTEP
+=====================
+
+Virtual memory header files that allow lsof to display text references
+were derived from the contents of /usr/include/vm of NEXTSTEP 2.0.
+NeXT did not ship the virtual memory header files with other NEXTSTEP
+or OPENSTEP versions.
+
+You may use the RC_FLAGS environment variable to declare compiler
+options outside the Makefile. A common use of this variable is to
+define the architecture types to be included in a "fat" executable.
+See the comments in dialects/next/Makefile for an example.
+
+
+OpenBSD
+=======
+
+David Mazieres has provided OpenBSD test systems. The OpenBSD
+dialect version of lsof is compiled using the dialect sources it
+shares with NetBSD in the n+obsd dialect sub-directory.
+
+Kenneth Stailey has provided OpenBSD testing and advice.
+
+John Dzubera (Zube) reports, "lsof 4.33 compiles and runs on OpenBSD
+2.3 for the pmax architecture (decstation 3100)."
+
+I have not tested lsof on OpenBSD 3.8, but David Mazieres reports
+revision 4.76 worked on OpenBSD 3.8.
+
+
+Pyramid DC/OSx and Reliant UNIX
+===============================
+
+As of lsof revision 4.52 support for all Pyramid dialects has been
+discontinued. Lsof revision 4.51 with Pyramid support may be
+obtained upon request. Send the request to abe@purdue.edu.
+
+These two UNIX dialects are very similar and share dialect-specific
+source files from the pyramid sub-directory.
+
+The Reliant Unix Pyramid C compiler issues warning messages that
+I haven't found a convenient way to suppress. You can ignore
+warning messages about casts and conversions that lose bits. The
+message "warning: undefining __STDC__" is intentionally caused by
+the lsof MkKernOpts configuration script to suppress warning messages
+about cast and conversion problems in standard system header files,
+such as <stdio.h> and <string.h>.
+
+Bruce Beare and Kevin Smith provided test systems.
+
+
+Caldera OpenUNIX
+================
+
+Larry Rosenman provided an OpenUNIX 8 test system. Matthew Thurmaier
+provided technical assistance, along with these people from Caldera:
+Jack Craig, Robert Lipe, and Bela Lubkin.
+
+Robert Lipe supplied changes to lsof for OpenUNIX 8.0.1. Those
+changes were also incorporated in UnixWare 7.1.3 when it became
+the release name for OpenUNIX 8.0.1.
+
+Support for lsof on OpenUNIX ended at lsof revision 4.74. The last
+lsof revision, 4.74, tested on OpenUNIX, may be found at the lsof
+"home" ftp site, lsof.itap.purdue.edu, in pub/tools/unix/lsof/OLD/src.
+
+
+SCO OpenServer
+==============
+
+Dion Johnson, Bela Lubkin, and Nathan Peterson of SCO gave me copies
+of SCO OpenServer and the SCO OpenServer Development System 3.0
+and provided technical advice for the lsof port.
+
+Hugh Dickins, Bela Lubkin, Craig B. Olofson, and Nathan Peterson
+provided version 5.0 and gave technical advice for porting lsof to
+it. Bela provided the 5.0.4 changes. D. Chris Daniels provided
+a 5.0.4 test system, Lee Penn provided one for 5.0.5, and John
+Dubois for 5.0.6.
+
+The <netdb.h> header file was accidentally omitted from some SCO
+OpenServer Development System releases. The Configure script will
+sense its absence and substitute an equivalent from the BSD
+distribution. The BSD <netdb.h> and the <sys/cdefs.h> header file
+it includes are located in the dialects/os/include sub-directory
+tree.
+
+To compile lsof from its distribution sources you must have the
+TCP/IP and NSF headers in /usr/include. While those are optional
+OpenServer packages, I have access to no system that doesn't have
+them, so I'm unable to build lsof for such a configuration. However,
+it should be possible to modify the lsof Configure script and
+sources so lsof would compile and work without those optional
+packages.
+
+If you have an OpenServer system configured without the TCP/IP and
+NFS packages, and want to tackle the job of building lsof for it,
+contact me via e-mail at <abe@purdue.edu>. I'll identify the
+Configure script, header file, and source file changes you will
+need to make. (Caution: this is not a simple task, or I would have
+already done it.)
+
+The optional osrgcc and scogcc Configure abbreviations construct
+Makefiles for compiling lsof with gcc.
+
+The UnixWare 7.1.4 sources are used for OpenServer Release 6.0.0.
+Hence there is a separate Configure abbreviation for it, "osr6".
+Richard of SCO provided a test system and technical assistance.
+
+
+SCO|Caldera UnixWare
+============
+
+D. Chris Daniels, John Hughes, Ken Laing, Andrew Merril, Lee Penn, and
+Matthew Thurmaier provided test systems. Bela Lubkin provided
+technical assistance. Larry Rosenman provided 7.1.[34] test systems.
+
+
+Solaris 2.x, 7, 8, 9 and 10
+===========================
+
+SEE THE CAUTIONS SECTION OF THIS DOCUMENT.
+
+The latest Solaris revision of lsof 4 might work under Solaris
+2.[1-4] and 2.5[.1] and 7 but hasn't been tested there. I have no
+test systems for those Solaris versions.
+
+Lsof will compile with gcc and the Sun C compiler under Solaris.
+If you want to use the Sun compiler, use the solariscc Configure
+abbreviation. If you use a gcc version less than 2.8 on Solaris,
+make sure the gcc-specific includes have been updated for your
+version of Solaris -- i.e., run the gcc fixincludes script.
+
+Solaris 7, 8, 9 and 10 support for 64 bit kernels depends on a Sun
+WorkShop or Forte C compiler version that supports the "-xarch=v9"
+flag -- usually 5.0 or greater. Gcc versions 2.95 and above *may*
+be configured and built for 64 bit support, but it takes some extra
+work, the resulting compiler may be fragile, and the gcc developers
+discourage it. I've built 64 bit capable gcc compilers for Solaris
+7, 8 and 9 from gcc versions 2.95 through 3.0.1 and produced working
+lsof executables with them. More information on 64 bit gcc for
+Solaris may be found in the 00FAQ file.
+
+Solaris 10 ZFS support is questionable, because Sun does not distribute
+the ZFS kernel structure definition header files. The lsof Configure
+script and source code use some risky work-arounds. ZFS file system
+support was made possible with help from Horst Scheuermann.
+
+Dave Curry and Steve Kirsch provided resources for Solaris 2.x
+ports. Casper Dik and Gerry Singleton consulted and provided
+valuable assistance.
+
+Henry Katz, Joseph Kowalski, Charles Stephens, Mike Sullivan, and
+Mike Tracy provided technical assistance.
+
+AFS support was added to Solaris lsof with help from Curt Freeland,
+Heidi Hornstein, Michael L. Lewis, Terry McCoy, Phillip Moore, and
+Sushila R. Subramanian.
+
+Casper Dik provided valuable assistance for the Solaris 8 support.
+
+Sun has graciously provided me access to BETA versions of Solaris
+2.5, 2.6, 7, 8, and 9.
+
+John Dzubera provided Solaris 7 and 8 test systems.
+
+Mike Miscevic provided Solaris 10 test systems.
+
+
+Ultrix
+======
+
+As of lsof revision 4.52 support for Ultrix is no longer available,
+because I no longer have an Ultrix test system.
+
+Terry Friedrichsen, Dwight McKay, and Jeffrey Mogul helped me with
+this port.
+
+DECnet support was added to Ultrix lsof with the help of John
+Beacom, who kindly provided a test system. The Configure script
+decides that DECnet support is available if /usr/lib/libdnet.a and
+/usr/include/netdnet/dn.h exist and are readable.
+
+
+Veritas VxFS and VxVM
+=====================
+
+Lsof supports some versions of Veritas VxFS and VxVM on some UNIX
+dialects. Consult the lsof Configure script for the specific
+dialect, and consult the lsof dialect-specific source files for
+the UNIX dialect of interest. Veritas support will usually be
+found in a source file named dnode[1-9].c.
+
+Since Veritas rarely has a version number that can be extracted
+with shell commands, lsof doesn't use it. Instead, when lsof
+supports Veritas, the Configure script will form compile-time
+definitions starting with HASVXFS. Check the lsof 00PORTING
+documentation file for more information.
+
+Lsof Veritas support requires that the supporting Veritas header
+files be installed -- e.g., in /usr/include/sys/fs. (The location
+will depend in the dialect's header file conventions.)
+
+Some information on lsof support for Veritas extensions may be
+found in the lsof 00DIST file.
+
+Chris Kordish and Andy Thomas have provided Solaris VxFS test
+systems.
+
+
+================================
+User-contributed Dialect Support
+================================
+
+There are some user-contributed dialect versions of lsof; more
+information on them can be found at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/contrib
+
+Check the 00INDEX file there for details.
+
+
+============================
+Dialects No Longer Supported
+============================
+
+Because I don't have access to test systems, these UNIX dialects
+are no longer supported by lsof:
+
+ CDC EP/IX
+ /dev/kmem-based Linux
+ MIPS RISC/os
+ Motorola V/88
+ Pyramid DC/OSx
+ Pyramid Reliant UNIX
+ Sequent DYNIX
+ SGI IRIX
+ SunOS 4.x
+ Ultrix
+ UnixWare below 7.0
+
+Remnants of the support lsof once provided for these dialects may
+be found in:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/OLD/binaries
+and
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/OLD/dialects
+
+
+===============
+Installing Lsof
+===============
+
+The distributed Makefiles do not have actions that will install
+lsof. I've come to the conclusion there is no standard for installing
+lsof or its man page, so I no longer distribute make rules for
+installing them. You should adjust the Makefile for your local
+preferences.
+
+The Makefile does have an install rule that will cause lsof to
+compile by virtue of its dependency clause. Some Makefiles also
+have a dependency that causes the production of a man page that is
+ready to install. However, the actions of the install rule will
+not cause the lsof executable or its man page to be installed in
+any UNIX system-wide directory.
+
+Instead, after the compilation and optional man page production
+are completed, the install rule will produce a brief description
+of what actions you might add to the install rule. The description
+will suggest the possible modes, ownerships, permissions, and
+destinations your install rule might employ to install the lsof
+executable and man page.
+
+As you form your install rule, keep in mind that lsof usually needs
+some type of special permission to do its job. That may be permission
+to read memory devices such as /dev/kmem, /dev/mem, or /dev/swap,
+or it may be authorization to read entries in the /proc file system.
+
+Memory device access can usually be provided by setting the modes
+of the lsof executable so that it's effective group identifier when
+it runs is the same as the group that has permission to read the
+memory devices -- i.e., it is setgid-group. The privileged group
+is usually kmem, sys, or system.
+
+Don't overlook using ACLs -- e.g., on AIX or Solaris 8 -- to give
+lsof permission to access memory devices. ACLs, coupled to a
+separate group like kmem, can be safer than giving lsof setgid
+authorization to a commonly used system group.
+
+When lsof needs to read /proc file system entries, it must be
+installed with modes that make its effective user identifier root
+when it runs -- i.e., it must be setuid-root. If lsof must be
+installed setuid-root (only the AIX 5L, PSTAT-based HPUX, and
+/proc-based Linux, ports need such power.), then access to memory
+devices is automatic (or not needed in the case of /proc-based
+Linux).
+
+Your choice of permissions for lsof may also be affected by your
+desire to allow anyone to use it or your need to restrict its usage
+to specific individuals. You will have to be guided by local policy
+and convention in this case.
+
+The next two sections, Setgid Lsof Dialect Versions and Setuid-root
+Lsof Dialect Versions, list recommended install permissions.
+
+The system directory where you install the lsof executable is also
+open to choice. A traditional place for a tool like lsof is
+/usr/local/etc, but recent changes in directory structure organization
+suggest that somewhere in /opt may be more suitable.
+
+Bear one other factor in mind when choosing a location for the lsof
+executable -- it usually is a shared executable, requiring access
+to shared libraries. Thus, locations like /sbin or /usr/sbin are
+probably unsuitable.
+
+Once you've chosen a location for the executable you may find that
+the location for the man page follows -- e.g., if the executable
+goes in /usr/local/etc, then the man page goes in /usr/local/man.
+If the executable location doesn't imply a location for the man
+page, you'll have to let local custom guide you.
+
+
+Setuid-root Lsof Dialect Versions
+=================================
+
+These dialect versions should be installed with setuid-root
+permission -- i.e., the lsof binary should be owned by root and
+its setuid execution bit (04000) should be set.
+
+ AIX 5L and above for full use of the -X option
+ Apple Darwin 8.x for Power Macintosh systems
+ PSTAT-based HP-UX 11.11 and 11.23
+ /proc-based Linux (generally 2.1.72 and above)
+
+
+Setgid Lsof Dialect Versions
+============================
+
+These dialect versions should be installed with setgid permission,
+owned by the group that can read kernel memory devices such as
+/dev/drum, /dev/kmem, /dev/ksyms, /dev/mem, /dev/swap. ACLs may
+be another mechanism (e.g., under AIX or Solaris 8) you can use to
+grant read permission to the kernel memory devices.
+
+ AIX 4.1.[45], 4.2[.1], and 4.3[.123]
+ Apple Darwin 7.x for Power Macintosh systems
+ DEC OSF/1, Digital UNIX, Tru64 UNIX 2.0, 3.2, 4.0, and 5.[01]
+ FreeBSD 2.1.6, 2.2[.x], 3.x, 4.x, 5.x and [67].x
+ /dev/kmem-based 11.00
+ NetBSD 1.[456], 2.x and 3.x
+ NEXTSTEP 3.[13]
+ OpenBSD 2.[89] and 3.[0-9]
+ OPENSTEP 4.x
+ Caldera OpenUNIX 8
+ SCO OpenServer 5.0.[46]
+ SCO UnixWare 7.0 and 7.1.[0134]
+ Solaris 2.6, 8, 9 and 10
+ Ultrix 4.2 (no longer available)
+
+====================================
+Porting lsof 4 to a New UNIX Dialect
+====================================
+
+If you're brave enough to consider this, look at the 00PORTING
+file. Please contact me before you start. I might be able to help
+you or even do the port myself.
+
+Don't overlook the contrib/ directory in pub/tools/unix/lsof on my
+ftp server, lsof.itap.purdue.edu. It contains user-contributed ports
+of lsof to dialects I don't distribute, because I can't test new
+revisions of lsof on them.
+
+
+=========================
+Quick Start to Using lsof
+=========================
+
+For information on how to get started quickly using lsof, consult
+the 00QUICKSTART file of the lsof distribution. It cuts past the
+formal density of the lsof man page to provide quick examples of
+using lsof to solve common open file display problems.
+
+
+======================
+Cross-configuring Lsof
+======================
+
+Using environment variables it is possible to Configure (and possibly
+build) lsof for one UNIX dialect on a different one -- e.g., you
+are running Configure on a Linux 2.3 system and you want to Configure
+and build lsof for Linux 2.4.
+
+See the 00XCONFIG file of the lsof distribution for a discussion
+of how to do this.
+
+
+====================================================
+Environment Variables Affecting the Configure Script
+====================================================
+
+Configure script actions can be modified by introducing values to
+the script via environment variables. In many cases the environment
+variable values take the place of test operations the Configure
+script makes.
+
+For more information on environment variables that can affect
+Configure, consult the 00XCONFIG file of the lsof distribution.
+See the General Environment Variables sections for descriptions of
+ones that affect all dialects. Consult the Dialect-Specific
+Environment Variables section for ones that might affect the dialect
+you are trying to configure.
+
+
+Vic Abell <abe@purdue.edu>
+April 24, 2007
--- /dev/null
+
+ The Lsof Test Suite
+
+ Contents
+
+ A. Introduction
+ 1. Test Suite Goals
+ 2. Not a FAQ
+ 3. Where have the tests been tested?
+ B. Test Methodology
+ 1. Test Limitations
+ 2. Test Data Base and Scripts
+ 3. The Makefile
+ 3.1 The CkTestDB Script
+ 4. The Lsof Executable and LT_LSOF_PATH
+ 5. Automated Testing
+ C. Configure Script Participation
+ 1. config.cc
+ 2. config.cflags
+ 2.1 config.cflags Contents
+ 3. config.ldflags
+ 4. config.xobj
+ D. Cleaning -- Quick or Spotless
+ E. Test Libraries
+ 1. LTlib.c
+ F. The Individual Tests
+ 1. LTbasic, a Basic Lsof Test
+ 2. LTbigf, Test Sizes and Offsets for Large
+ (> 32 bit) Files
+ 3. LTdnlc, Test the Kernel's Dynamic Name Lookup
+ Cache
+ 4. LTlock, Lock Tests
+ 5. LTnfs, NFS Test
+ 6. LTnlink, Link Count Test
+ 7. LTsock, Test IPv4 Sockets
+ 8. LTszoff, Test Sizes and Offsets for Small
+ (< 32 bit) Files
+ 9. LTunix, Test UNIX Domain Sockets
+ Appendix A, Test Files
+ Appendix B, Test Validations
+ Appendix C, Test Failures
+
+
+A. Introduction
+===============
+
+Lsof has an automated test suite whose components are located in
+the tests/ sub-directory of the lsof top-level directory. Configuring,
+building and testing lsof can be done with these shell commands:
+
+ $ Configure -n <dialect-abbreviation>
+ $ make
+ $ cd tests
+ $ make
+
+That's all there is to it!
+
+But read on for more dirty details.
+
+A.1. Test Suite Goals
+=====================
+
+The lsof test suite attempts to test basic lsof features. It does
+not promise to test every lsof feature for every supported dialect.
+(That's a nearly impossible goal.)
+
+As a result, the test suite cannot promise that every lsof feature
+works as documented. At best the test suite gives some assurance
+that basic, standard and some optional lsof features work.
+
+A.2. Not a FAQ
+==============
+
+One caution: this is not a frequently asked questions (FAQ) file
+for the lsof test suite. FAQs on the lsof test suite will be found
+in the one and only lsof FAQ in file 00FAQ of the lsof distribution,
+or on-line at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/FAQ
+
+A.3. Where have the tests been tested?
+======================================
+
+OK, I just said this isn't an FAQ and here comes a question and
+answer that looks like an FAQ. Consider it a very frequently asked
+question and indulge me -- let me answer it here.
+
+The lsof test suite hasn't been tested everywhere it might be
+possible to build lsof successfully. That "everywhere" includes
+dialect versions -- e.g., Solaris < 2.6 -- to which I no longer
+have access. On some dialect versions to which I have access, some
+tests won't run because the test system lacks support.
+
+In "Appendix B, Test Validations" I've tried to list where I compiled
+and tested the test suite and information on any tests I was unable
+to run. In "Appendix C, Test Failures" I list where the test suite
+fails and why it failed.
+
+A.4 Where are the tests?
+========================
+
+This is another FAQ whose answer is that the tests are in the tests/
+sub-directory of the main lsof source directory.
+
+
+B. Test Methodology
+===================
+
+The test suite is made up of individual C programs. Test setup is
+performed by the lsof Configure script itself, which writes a set
+of dialect configuration files in the tests/ subdirectory. (See
+"C. Configure Script Participation.")
+
+Each program or script performs a specialized tests. Those tests
+are described below in "F. The Individual Tests".
+
+Each test measures lsof functionality by putting a specific lsof
+command execution at the end of an in-bound (to the test) pipe.
+The caller asks lsof to write its results to the pipe in field
+output form. (See the -F option in the lsof man page.)
+
+Using an in-bound lsof pipe allows the tests to measure a great
+deal of lsof functionality, including as an interesting side effect,
+the performance of field output. Consequently, there's no special
+field output test.
+
+B.1. Test Limitations
+=====================
+
+One limitation of the tests is that they don't measure lsof formatted
+output -- i.e., the output normally see when lsof is run. There
+are just too many variants of lsof output produced across the range
+of dialects where lsof runs, so field output is a more consistent
+way to process lsof output.
+
+But using field output does mean that the test suite doesn't check
+for lsof formatting problems, except in the field output itself.
+
+B.2. Test Data Base and Scripts
+===============================
+
+The TestDB file contains a simple data base that describes where
+the tests have been validated. Entries are formed from a combination
+of the lines in the config.cflags file produced by the lsof Configure
+script. The entries can be considered "lsof dialect footprints,"
+hereafter simply called "dialect footprints" or just "footprints."
+
+Two shell scripts support TestDB. The first, Add2TestDB, will add
+a footprint to TestDB. I don't recommend you use this script.
+Mostly it's for my use when I find that the test suite has been
+validated on a new dialect.
+
+It's also possible to add a footprint to TestDB by simply editing
+TestDB and pasting into it a copy of the footprint reported by a
+failed Makefile rule. I don't generally recommend this be done,
+either.
+
+There are Makefile rules that use (and avoid) the CkTestDB script.
+(See "B.3 The Makefile".)
+
+The default (i.e., "all") Makefile rule uses the CkTestDB script
+to look for the footprint in TestDB. If no footprint is found, the
+script issues a warning, displays the unfound footprint, and asks
+if running the test suite should continue.
+
+The "auto" Makefile rule also uses CkTestDB, but with a special
+call that causes CkTestDB to look in TestDB for the footprint,
+report it when it can't be found, and then fail. That CkTestDB
+failure causes the "auto" rule to fail, too.
+
+The "silent" Makefile rule doesn't use CkTestDB to look in TestDB
+for the footprint. It runs the standard and basic tests as silently
+as possible, then returns a failure or success exit code that
+signals the result of the running of the tests. (Use the "silent"
+rule carefully, because it will skip proving the tests have previously
+run on the dialect.)
+
+B.3. The Makefile
+=======================
+
+The Makefile runs the tests in the test suite. It has these rules.
+
+ all runs the basic test and the standard tests,
+ interacting with the caller should the footprint
+ not be found in TestDB.
+
+ (This is the default rule.)
+
+ auto runs the basic test and the standard tests on a
+ previously validated system as silently as possible.
+
+ The footprint must be found in TestDB for this rule
+ to succeed. (See the "silent" rule for one that
+ avoids checking TestDB.)
+
+ This rule is designed for lsof build scripts that
+ want a quick noiseless test to make sure what they
+ built works as it previously did.
+
+ This rule calls CkTestDB in a way that inhibits
+ its normal go-ahead request. (See "B.2.1 The CkTestDB
+ Script".) If CkTestDB finds the footprint and all
+ tests succeed, this rule returns a zero exit code
+ (success). If the footprint isn't found or if any
+ test fails, a non-zero exit code (failure) is
+ returned.
+
+ ckDB calls the CkTestDB script to look for a footprint.
+ If none is found, the way CkTestDB was called (See
+ "B.3.1 The CkTestDB Script".) causes it to return
+ a non-zero exit code (failure) to this rule, and
+ the rule then returns a non-zero exit code (failure)
+ itself.
+
+ This rule is used by the "auto" rule. If this rule
+ succeeds (zero exit code), the "auto" rule then
+ uses the "silent" rule.
+
+ clean removes test and compiler output. (See the "D.
+ Cleaning -- Quick or Spotless" section.)
+
+ opt runs the optional tests.
+ optional
+
+ silent runs the lsof basic and standard tests as silently
+ as possible (as the "auto" rule does), but without
+ using CkTestDB to look for a footprint. If all
+ tests succeed, the rule returns a zero exit code
+ (success). If any test fails, the rule returns a
+ non-zero exit code (failure).
+
+ Use the "silent" rule carefully, because it will
+ skip proving the tests have previously run on the
+ dialect.
+
+ spotless does what the clean rule does and also removes the
+ config.* files created by ../Configure. (See the
+ "D. Cleaning -- Quick or Spotless" section.)
+
+ std runs the basic test and the standard tests.
+ standard
+
+The Makefile cleaning rules are further described in "D. Cleaning
+-- Quick or Spotless."
+
+B.3.1 The CkTestDB Script
+=========================
+
+Some Makefile rules (e.g., "all" and "auto") use the CkTestDB script
+to make sure the tests have been run previously on the dialect.
+CkTestDB does that by looking for the dialect's footprint in TestDB.
+
+If no footprint is found, and if standard input is a TTY, CkTestDB
+asks for a go-ahead signal. If standard input isn't a TTY, CkTestDB
+aborts the test run. (See "B.2. Test Data Base and Scripts".)
+
+The Makefile "silent" rule does not call CkTestDB. use the "silent"
+rule carefully, because it will skip proving the tests have previously
+run on the dialect.
+
+B.4. The Lsof Executable and LT_LSOF_PATH
+=========================================
+
+Normally the programs in the test suite use the lsof executable in
+their parent directory, ../lsof. Usually that lsof has just been
+built and testing it is the next logical step.
+
+Be careful that ../lsof has sufficient permission to access the
+necessary kernel resources -- e.g., /dev/kmem, /dev/mem, /proc,
+etc. If it doesn't the tests will fail. (The tests do check to
+see if they can open /dev/mem and /dev/kmem for read access if
+LT_KMEM is defined in config.cflags and if the path to the lsof
+executable is ../lsof.)
+
+Here are two possible ways you can make sure the lsof being tested
+has sufficient permission: 1) use chmod and chgrp to enable its
+running and put its path in LT_LSOF_PATH, thus disabling the check
+in the tests for kernel memory access; or 2) run the tests (and
+hence the lsof being tested) under a login that has sufficient
+permission -- e.g., is in a group that can read /dev/kmem.
+
+You can direct the tests to use a different lsof executable by
+specifying its path in the LT_LSOF_PATH environment variable. To
+test an lsof executable already installed in /usr/local/etc --
+provided that lsof is at revision 4.63 or higher -- do this:
+
+ $ LT_LSOF_PATH=/usr/local/etc/lsof
+ $ export LT_LSOF_PATH
+ $ cd .../lsof_<version>/tests
+ $ make
+
+When you specify an alternate executable path via LT_LSOF_PATH,
+that also prevents the tests from checking to see if they have
+kernel memory access.
+
+B.5 Automated Testing
+=====================
+
+Normally the lsof test suite is wordy and may require interaction.
+When you want to avoid those interferences, use the Makefile "auto"
+or "silent" rules. (See the description of the "auto" and "silent"
+rules in "B.3 The Makefile".)
+
+The footprint must be present in TestDB in order to use the "auto"
+rule. If it is not, the "auto" rule will fail and report the
+missing footprint. Footprints in TestDB proclaim that the tests
+have previously succeeded on the dialect.
+
+The footprint need not be present in TestDB in order to use the
+"silent" rule. Use the "silent" rule carefully, because it will
+skip proving the tests have previously run on the dialect.
+
+
+C. Configure Script Participation
+=================================
+
+An important step in setting up the test suite is performed by the
+Configure script in the lsof home directory (the parent to tests/.)
+
+Configure writes four files in tests/ that describe how the tests
+are to be compiled, configured and loaded. The files also describe
+options that Configure selected that are important to the test
+suite.
+
+C.1. config.cc
+==============
+
+This file, config.cc, contains the name of or the path to the C
+compiler used to compile lsof. The Makefile uses this file in
+place of the standard make(1) CC string with a shell in-place
+execution statement -- i.e., `cat config.cc`.
+
+If the LSOF_CC environment variable is supplied to the lsof Configure
+script, its value will appear in the config.cc file.
+
+C.2. config.cflags
+==================
+
+This file, config.cflags, contains C compiler flags that Makefile
+uses to compile the C programs in the test suite. As with the
+compiler file, config.cc, the make rules incorporate the contents
+of this file into C compiler options with `cat config.cflags`.
+
+This file is also used by the Add2TestDB and CkTestDB shell scripts
+to build and match footprints. (See "B.2. Test Data Base and
+Scripts.")
+
+C.2.1 config.cflags Contents
+============================
+
+The config.cflags file may contain the following C compiler flags.
+
+
+ -DLT_AIXA is present if lsof was built for AIX.
+ It contains the AIX architecture flag.
+ (See the lsof Configure script or
+ dialects/aix/dlsof.h for more information
+ on the AIX architecture flag.)
+
+ -DLT_BIGF is present if lsof was built for a dialect
+ that has large file (sizes and offsets >
+ 32 bits).
+
+ -DLT_CC is present if the lsof compiler is cc.
+
+ -DLT_DIAL_<abbr> always ends in (the <abbr> part) the
+ "canonical" -- i.e., usually the most
+ common abbreviation by which the dialect
+ is known.
+
+ Example: -DLT_DIAL_solaris
+
+ -DLT_GCC is present if the lsof compiler is gcc.
+
+ -DLT_K64 is present if lsof has been built for a
+ 64 bit kernel
+
+ -DLT_KMEM is present if lsof has been built to
+ use /dev/kmem to obtain kernel values.
+
+ -DLT_VERS=<vn> contains the version number for the
+ dialect, as used in lsof pre-processor
+ tests.
+
+ Example for Solaris 10: -DLT_VERS=100000
+
+ -DLT_VPATH is present if the dialect has the v_path
+ member in the vnode structure (e.g., some
+ versions of Solaris 10).
+
+The config.cflags file may also contain dialect-specific compiler
+flags needed to activate a specific feature on the dialect. For
+example, for HP-UX config.cflags might contain:
+
+ -D_LARGEFILE64_SOURCE This compiler flag enables the use of
+ large-file system library functions
+ --e.g., open64().
+
+ The lsof Configure script stanzas for
+ the dialects select these options.
+
+
+C.3. config.ldflags
+===================
+
+This file, config.ldflags, contains the dialect loader flags --
+i.e., the equivalent to make(1) LFLAGS -- for loading the test
+programs.
+
+Example for Solaris: -lsocket this adds the socket library
+ to the loading of the lsof
+ test programs.
+
+Example for UnixWare: -lsocket -lnsl this adds the socket and
+ name server libraries to
+ the loading of the lsof
+ test programs.
+
+
+C.4. config.xobj
+================
+
+This file, config.xobj, contains the paths to any extra object
+files (.o's) that must be loaded when the test suite C programs
+are loaded. Like config.cc and config.cflags, it's incorporated
+into the loader statements of the Makefile's rules with `cat
+config.xobj`.
+
+Examples for DEC OSF/1 and NEXTSTEP:
+
+ ../lib/snpf.o this loads the private lsof object file
+ that contains an snprintf() function. (The
+ DEC OSF/1 4.0 and NEXTSTEP 3.1 C libraries
+ don't have snprintf().)
+
+
+D. Cleaning -- Quick or Spotless
+================================
+
+There are two Makefile rules that clean the tests/ subdirectory --
+"clean" and "spotless". They cause different degrees of cleaning.
+
+ clean a "quick" clean that removes compiled object files,
+ executables and test files. It does NOT remove
+ the configuration files that ../Configure and the
+ config.perl rule wrote.
+
+ spotless cleans out everything clean does -- plus the
+ configuration files that ../Configure and the
+ config.perl rule wrote.
+
+ This is the rule used when `./Configure -clean` is
+ specified. If this rule is used, `../Configure -n
+ <abbr>` and `../make`) must be run again before
+ the test suite can be used.
+
+
+E. Test Library
+===============
+
+The lsof test suite provides a C library.
+
+E.1. LTlib.c
+============
+
+This is a C library of common functions used by tests. Configured
+at compile time by the contents of config.cflags, it uses the single
+header file LsofTest.h. LsofTest.h tailors its definitions to the
+dialect at compile time, using the LT_DIAL_* definitions in
+config.cflags.
+
+Two particularly useful functions in the library are: ExecLsof(),
+which will execute an lsof child process; and RdFromLsof(), which
+will read from the in-bound lsof pipe, and decode the fields into
+structures that are easy for C programs to process.
+
+This library is a good model for processing field output in a C
+program from an in-bound lsof pipe.
+
+The source for the library, LTlib.c, contains more documentation.
+
+
+F. The Individual Tests
+=======================
+
+The individual tests are listed in this section. The listings
+explain what the tests do, a few errors they might report, and how
+to use options and environment variables to customize the tests.
+
+The test descriptions are listed in this section in alphabetical
+order, not in the order they are run by Makefile.
+
+The Makefile runs the tests in three groups, basic tests, standard
+tests, and optional tests. The default make "all" rule runs the
+basic and standard tests. (The "standard", "std", and "test"
+Makefile rules are synonyms to "all".) If the standard tests succeed,
+Makefile suggests running the optional tests with the "opt" (or
+"optional") rule.
+
+The Makefile "auto" and "silent" rules run only the basic and
+standard tests. They do not run or suggest you run the optional
+tests.
+
+ The basic test:
+ LTbasic
+
+ Standard tests:
+ LTnlink
+ LTsock
+ LTszoff
+ LTunix
+
+ Optional tests:
+ LTbigf
+ LTdnlc
+ LTlock
+ LTnfs
+
+The basic and standard tests should all succeed on all dialects,
+although LTnlink may warn that it can't perform its unlink test on
+an NFS file system.
+
+The optional tests may run, they may be disabled for specific
+dialects, or they may fail because of special resource needs --
+e.g., LTbigf will run only on UNIX dialects where it knows how to
+handle files whose lengths exceed 32 bits, and LTnfs needs access
+to an NFS file system mounted from a remote NFS server.
+
+Tests that need special resources usually provide a hint about the
+resources when they fail. Information about special resource needs
+may also be found in the following sections about the individual
+tests.
+
+G.1. LTbasic, a Basic Lsof Test
+===============================
+
+This is the basic lsof test. If it doesn't run, it's not likely
+any other tests will run, either. Hence, if it fails, no Makefile
+rule runs any other tests.
+
+This test uses lsof to locate files in the lsof process, including
+current working directory, the lsof executable, and the /dev/kmem
+open file.
+
+Finding the lsof executable may not be possible on AIX if lsof was
+compiled without support for its -X option.
+
+Finding /dev/kmem use by lsof is only possible on dialects where
+lsof uses /dev/kmem. The -DLT_KMEM define indicates that.
+
+Run this test:
+
+ $ ./LTbasic
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.2. LTbigf, Test Sizes and Offsets for Large (> 32 bit) Files
+==============================================================
+
+This is a test in the optional test group.
+
+This test is effective only when ../Configure has put -DLT_BIGF in
+config.cflags. Without that definition this test simply reports
+that the dialect doesn't support large files. That report is
+accompanied by a successful test exit code, so that the runner of
+the test (e.g., the Makefile) won't believe the test failed.
+
+When a dialect does support large files, the test attempts to create
+a file that looks very large -- e.g., has a length as reported by
+ls(1) of 0x140000000 bytes. However, the file really has only a
+small amount of data in it, the rest of the file consists of a
+large standard UNIX file system "hole."
+
+By default the test file is named config.LTbigf<PID>, where PID is
+the Process ID of the LTbigf process.
+
+When that file is not on a file system enabled for large files, or
+when the process that runs LTbigf can't create a big file, LTbigf
+will report an error. The error will be accompanied by hints that
+the -p option may need to be used to define a path where the test
+can write a large file, or the process ulimit file block size may
+need to be raised -- e.g., to "unlimited."
+
+LTbigf can't test file offset reporting on Linux kernels below
+2.6.22, because the /proc file systems of those kernels don't make
+file offsets available to lsof.
+
+Run this test:
+
+ $ ./LTbigf [-p <path>]
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.3. LTdnlc, Test the Kernel's Dynamic Name Lookup Cache
+========================================================
+
+This is a test in the optional test group.
+
+This test asks lsof to locate the current working directory of its
+own process and report the path it has assembled from the components
+it found in the kernel's Dynamic Name Lookup Cache (DNLC) or via
+other dialect-specific methods. (E.g., Linux, HP-UX 11.11, and
+some Tru64 UNIX versions have private name lookup methods.)
+
+The test checks what lsof reports as the current working directory
+path for any missing components and counts the number of full paths
+returned. (Symbolic link complications prevent testing for exact
+path matches.) The test is repeated. If full paths are returned
+at least half the time, the test considers itself successful.
+
+This test can't be run on AIX, because lsof can't access the DNLC
+there. It can't be run on Apple Darwin versions below 8.0, either,
+because insufficiently reliable DNLC information is available there.
+This test may fail on other dialects when the file system -- e.g., NFS.
+/tmp, loopback -- doesn't fully participate in the dialect's DNLC.
+
+Run this test:
+
+ $ ./LTdnlc
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.4. LTlock, Lock Tests
+=======================
+
+This is a test in the optional test group.
+
+This test uses flock() and fcntl() to set and clear file locks,
+and measures lsof's ability to report them. The choice of system
+lock call is based on the dialect. (There are LT_DIAL_* pre-processor
+tests in LTlock.c.)
+
+This test can't be run on an NFS client file system, because NFS
+lock information is kept on the server. Lsof on the client can't
+see that server kernel data.
+
+By default the test attempts to create a file named config.LTlock<PID>,
+where PID is the Process ID of the locking test process. It uses
+lsof to determine if the file is on a client NFS file system. If
+it is, the test aborts, hinting that the -p option can be used to
+specify a non-client-NFS test file path.
+
+This test can't be run on Darwin, because insufficient file system
+lock information is available to lsof there.
+
+Run this test:
+
+ $ ./LTlock [-p <path>]
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.6. LTnfs, NFS Test
+====================
+
+This is a test in the optional test group.
+
+This test verifies that lsof can locate files mounted on a client
+NFS system from an NFS server.
+
+By default it creates a test file, config.LTnfs<PID>, where PID is
+the Process ID of the test process. The test then uses lsof to
+find the file on an NFS file system.
+
+If lsof can't find the file the test warns that the test file might
+not be on an NFS file system and hints that the -p option may be
+used to specify the path of an NFS file, provided the test can have
+read access to it there. The test warning also states that the
+file at the path specified with -p must be a regular file, not a
+directory.
+
+This test can't be run on Darwin versions below 8.0, because
+insufficient NFS file information is available to lsof there.
+
+Run this test:
+
+ $ ./LTnfs [-p <path>]
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.7. LTnlink, Link Count Test
+=============================
+
+This is a test in the standard test group.
+
+The test checks lsof's reporting of link count (nlink in UNIX
+argot.)
+
+It creates a test file in the current working directory named
+config.LTnlink<PID>, where PID is the Process ID of the test
+process. It then uses stat(2) and lsof to measure the link count
+of the file.
+
+If LTnlink creates the test file in the current working directory
+and it is on an NFS file system, LTnlink won't be able to perform
+one section of its test. In that section the test file is unlinked
+so its link count will be zero and lsof is asked to find it among
+the set of files whose link counts are zero.
+
+When an NFS file is unlinked its link count isn't reduced until
+the last open instance is closed on either the NFS clients or the
+NFS. That's a consequence of NFS statelessness and leads to the
+occasional presence of files with names of the form .nfsxxxx.
+
+Should LTnlink find its test file on an NFS file system, it disables
+the unlink section of its tests and issues a warning. It also
+issues a hint that the test file path can be named via the -p option
+to give a test file location that isn't on an NFS file system.
+
+This test can't be run on Darwin, because insufficient file system link
+count information is available to lsof there.
+
+Because some UNIX dialects delay the reporting of a link count
+update after a file has been unlinked, LTnlink may not get its
+expected response from lsof for a while after the test file has
+been unlinked. In that cause LTnlink may delay for up to a minute,
+calling lsof once every two seconds and displaying a "waiting for
+link count update: ..." message.
+
+Run this test:
+
+ $ ./LTnlink [-p <path>]
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.7. LTsock, Test IPv4 Sockets
+==============================
+
+This is a test in the standard test group.
+
+This test uses lsof to locate open IPv4 socket files that the test
+has created itself. The test opens a server socket, then forks a
+child process to connect to that socket. After both are running,
+the test uses lsof to find the open socket files at their known
+host and port addresses.
+
+Run this test:
+
+ $ ./LTsock
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.8. LTszoff, Test Sizes and Offsets for Small (< 32 bit) Files
+===============================================================
+
+This is a test in the standard test group.
+
+This test checks lsof's reporting of file size and offset for small
+(< 32 bits) files.
+
+It creates a test file in the current working directory named
+config.LTszoff<PID>. PID is the Process ID of the test process.
+
+LTszoff can't test file offset reporting on Linux kernels below
+2.6.22, because the /proc file systems of those kernels don't make
+file offsets available to lsof.
+
+Run this test:
+
+ $ ./LTszoff [-p <path>]
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+G.9. LTunix, Test UNIX Domain Sockets
+======================================
+
+This is a test in the standard test group.
+
+This test checks lsof's reporting of UNIX domain sockets.
+
+The test creates a pair of UNIX domain sockets and uses bind(2) to
+associate the file system names config.LT0U<PID> (client) and
+config.LT1U<PID> (server) with them. (PID is the test process ID.)
+The test then uses lsof to find the two open UNIX domain socket
+files.
+
+Run this test:
+
+ $ ./LTunix
+
+Environment variables: LT_LSOF_PATH defines the path to the lsof
+ executable. (The default is ../lsof.)
+
+
+Appendix A, Test Files
+======================
+
+These files may be created by suite tests.
+
+ Created
+ ./tests Name by Test Use
+ ============ ======= ===
+
+ config.LTbifg** LTbigf to test lsof's large file size
+ and offset reporting
+
+ config.LTlock* LTlock for locking tests
+
+ config.LTnfs* LTnfs for NFS tests
+
+ config.LTnlink* LTnlink for link count tests
+
+ config.LTszoff* LTszoff for small file size and and
+ offset reporting
+
+ config.LT[01]U* LTunix two UNIX domain sockets, used
+ to determine if lsof can report
+ their open instances properly
+
+
+Appendix B, Test Validations
+============================
+
+This appendix lists the UNIX dialects and their versions where I
+have validated the test suite. The list indicates the particular
+tests I was unable to run, mostly LTnfs because the test systems
+I used had no NFS file systems mounted.
+
+The information in the following table is encoded in a test data
+base file, TestDB, as footprints, using the tests compiler options
+written to config.cflags by the lsof Configure script. See "B.2.
+Test Data Base and Scripts" for more information on the test data
+base, footprints, and the scripts that support them.
+
+ UNIX
+ Dialect Dialect Description Untested Tests
+ ======= =================== ==============
+ AIX 4.3.3, Power, cc
+ 5.1, Power-32, cc
+ 5.1, Power-32, gcc
+ 5.1, Power-64, cc
+ 5.2, Power-32, cc
+ 5.2, Power-32, gcc
+ 5.2, Power-64, cc
+ 5.2, Power-64, gcc
+ 5.3, Power-64, cc
+ Darwin 1.4, 5.5, 6.x, 7.x gcc Darwin lsof doesn't
+ have adequate support
+ to allow the LTbigf,
+ Ltdnlc, LTlock, LTnfs,
+ and LTnlink tests to
+ run.
+ 8.0, gcc Darwin lsof doesn't
+ have adequate support
+ to allow the LTbigf,
+ LTlock and LTnlink
+ tests to run.
+ 9.0, gcc Darwin lsof doesn't
+ have adequate support
+ to allow the LTlock
+ test to run.
+ FreeBSD 4.5, i386, gcc
+ 4.6, i386, gcc
+ 4.7, i386, gcc
+ 4.8, i386, gcc
+ 4.9, i386, gcc
+ 4.10, i386 gcc
+ 5.0, Alpha, gcc
+ 5.0, Sparc, gcc
+ 5.0, i386, gcc
+ 5.1, Alpha, gcc
+ 5.1, Amd64, gcc
+ 5.1, Sparc, gcc
+ 5.1, i386, gcc
+ 5.2, i386, gcc
+ 5.2, Alpha, gcc
+ 5.2, Amd64, gcc
+ 5.2, Sparc, gcc
+ 5.3, Alpha, gcc
+ 5.4, Alpha, gcc
+ 5.5, Alpha, gcc
+ 6.0, Alpha, gcc
+ 6.0, Amd64, gcc
+ 6.0, Sparc, gcc
+ 6.1, i386, gcc
+ 7.0 Alpha, gcc
+ 7.0 Amd64, gcc
+ 7.1 Amd64, gcc
+ 8.0 Amd64, gcc
+ DEC OSF/1 4.0, cc
+ HP-UX 10.20, cc LTbigf
+ 10.20, gcc (1) LTbigf
+ 11.00-32, ANSI-C LTbigf, LTnfs
+ 11.00-64, ANSI-C
+ 11.11, ANSI-C
+ 11.23, ANSI-C
+ Linux 2.4.12-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.18-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.21-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.23-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.24-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.25-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.26-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.27-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.28-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.29-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.4.30-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.6.1-rc2 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.6.18-686 LTbigf, no offset tests
+ LTszoff, no offset tests
+ 2.6.22-686 (Note: this Linux kernel
+ supplies file offsets to
+ lsof.)
+ NEXTSTEP 3.1, gcc LTnfs
+ NetBSD 1.4.1, Alpha, gcc LTnfs
+ 1.5x, x86, gcc LTnfs
+ 1.6x, Alpha, gcc LTnfs
+ 1.6x, x86, gcc LTnfs
+ 2.0x, alpha, gcc LTnfs
+ 2.0x, sparc64, gcc LTnfs
+ 2.0x, x86, gcc LTnfs
+ 2.99.9, x86, gcc LTnfs
+ 2.99.10, x86, gcc LTnfs
+ 2.99.11, x86, gcc LTnfs
+ 2.99.12, x86, gcc LTnfs
+ 3.99., x86, gcc LTnfs
+ OpenBSD 3.0, gcc
+ 3.1, gcc
+ 3.2, gcc
+ 3.3, gcc
+ 3.4, gcc
+ 3.5, gcc
+ 3.6, gcc
+ 3.7, gcc
+ 3.9, gcc
+ OPENSTEP 4.2, gcc LTnfs
+ OSR 5.04, cc LTnfs
+ 5.06, cc LTnfs
+ Solaris 2.6, cc LTnfs
+ 2.6, gcc LTnfs
+ 7-32, cc
+ 7-32, gcc LTnfs
+ 8-32, cc
+ 8-32, gcc
+ 8-64, cc
+ 8-64, gcc
+ 9-64, Beta-Refresh, cc
+ 9-64, Beta-Refresh, gcc
+ 9-64, FCS, cc
+ 9-64, FCS, gcc
+ 10-32, i86pc, gcc
+ 10-32, i86pc, cc
+ 10-64, Sparc, cc
+ 10-64, Sparc, gcc
+ Tru64 UNIX 5.0, cc
+ Tru64 UNIX 5.0, cc
+ 5.1, cc
+ UnixWare 7.1.1, NSC, cc LTnfs
+ 7.1.3, cc
+ 7.1.4, cc
+
+If you are able to run the test suite on dialect versions other
+than the ones listed above, please send e-mail to <abe@purdue.edu>,
+indicating the dialect version where you were able to run the test
+suite. Please send me the footprint formed by CkTestDB, or run
+the Add2TestDB script and send me the footprint it reports.
+
+If you encounter problems compiling the tests or running them on
+a dialect version listed above, please send e-mail to <abe@purdue.edu>,
+naming the dialect version and providing the output from the lsof
+Configure script and make operation.
+
+1) John Dzubera did the HP-UX 10.20 gcc testing and provided its
+ footprint.
+
+
+Appendix C, Test Failures
+=========================
+
+I was unable to make the test suite run on the following dialects.
+
+ UNIX Dialect
+ and Description Failure
+ =============== =======
+ HP-UX 11-64, gcc 64 bit gcc 3.0 didn't compile the LTsock
+ test correctly on my 64 bit HP-UX 11 test
+ system.
+
+
+Vic Abell <abe@purdue.edu>
+March 25, 2009
--- /dev/null
+
+ Cross-configuring Lsof
+
+Introduction
+============
+
+Lsof cross-configuration is useful when the target dialect or target
+dialect version for which lsof is to be configured and built differs
+from the one on which the Configure operation is done.
+
+Marty Leisner <leisner@sdsp.mc.xerox.com> suggested the method
+described here for lsof cross-configuration, and he supplied
+modifications to the Configure script for cross-configuring Linux
+lsof.
+
+Marty says:
+
+ "I used this to successfully compile (lsof) on the same machine
+ for (Linux) 2.0.30 and 2.1.42. (I normally don't bring up a
+ 2.1.42 machine all the time). Also it (the 2.0.30 system)
+ doesn't have much storage and compiles on it are slow.
+
+ Set LSOF_VERS if it's not the (version of the) current system.
+ (Actually, you should get the version out of include/linux/version.h.)
+
+ Define LINUX_KERNEL to (the path) where the kernel sources
+ are (located). (No longer necessary as of lsof revision 4.53.)
+
+ This should work on most systems; they put a kernel in
+ /usr/src/linux, which is the default.
+
+ Now I can just do:
+
+ LINUX_KERNEL=/some/other/kernel LSOF_VERS=2142 ./Configure linux
+
+ Comments? Its very convenient when running multiple kernels.
+ (It would be (have been) very handy when the structures changed
+ between 2.0.2* and 2.0.30 , or whatever.)
+
+ I run multiple OSes at a time (not to mention multiple
+ architectures. It's very pleasant to cross-build either
+ operating systems or versions."
+
+So, the situation is that you have lsof sources on a UNIX dialect
+version, and you want to configure them to build lsof for some
+other version of the same dialect, or perhaps for some other UNIX
+dialect altogether.
+
+
+The Cross-Configure Method
+==========================
+
+The lsof cross-configure method uses environment variables to tell
+the lsof Configure script about the target dialect. The environment
+variables may specify alternate locations for Configure to examine
+when it determines characteristics of the target, or they may
+specify the values Configure would discover when it examined the
+target's characteristics.
+
+Consult each environment variable description for the UNIX dialect
+in which you're interested to see how it affects the operation of
+the Configure script.
+
+The number and values of the variables differ by dialect. Each
+variable begins with an upper case version of the dialect's Configure
+abbreviation -- e.g., AIX for aix or aixgcc, LINUX for linux,
+UW for uw (UnixWare), etc.
+
+Of course, the UNIX dialect's version is probably different from
+that of the system on which you're doing the cross-configuration,
+so you will need to specify the new version, too. For example, to
+configure for FreeBSD 3.0 on a 2.1.7 system, where the standard
+3.0 header files are in /3.0/usr/include and the 3.0 system sources
+are in /3.0/sys, do this:
+
+ LSOF_VERS=300 LSOF_INCLUDE=/3.0/usr/include \
+ FREEBSD_SYS=/3.0/sys Configure -n freebsd
+
+
+General Environment Variables
+=============================
+
+There are some environment variables whose names don't begin with
+an upper case rendering of a dialect abbreviation. Generally they
+apply to all dialects.
+
+AFS_VICE is for AFS configuration. It need be set only if
+ lsof supports AFS on your dialect and you want to
+ specify an alternate path to the VICE files.
+
+ default: /usr/vice
+
+LSOF_AR is the path to and arguments for the library archive
+ application that is used to build the lsof library,
+ liblsof.a. When this value is placed in the library
+ Makefile as the contents of the AR make string, it is
+ followed by the path to the library and the relative
+ paths of the library module
+
+ default: ar cr
+
+LSOF_ARCH is the architecture type string for the system.
+ Usually this is the output of `uname -m`. Consult
+ the Configure script for details. The LSOF_ARCH
+ value may have to be quoted if it contains spaces.
+
+ default: auto-detection (e.g., from `uname -m`)
+
+LSOF_BLDCMT may be used to introduce a builder's comment into
+ lsof's -v output. It defaults to the null string,
+ causing no builder's comment to appear in -v output.
+
+ default: none
+
+LSOF_CC is the path to the C compiler. You may need to
+ specify it if your C compiler is in a non-standard
+ place, not found by your path. If you specify a
+ compiler different from the expected default, you
+ may have to change the compile time flags by
+ specifying new CFGF, CFGL, and DEBUG strings on
+ the make command line.
+
+ default: normally cc, but some dialects have other
+ defaults and some have auto-detection.
+
+ Check the dialect stanza in the lsof Configure
+ script to see how LSOF_CC is set by default.
+
+LSOF_CCV is the C compiler version. You should specify it
+ if you have specified a compiler path in LSOF_CC.
+
+ default: the lsof Configure script knows how to find
+ the version number of gcc and some other
+ dialect-specific compilers.
+
+ Check the dialect stanza in the lsof Configure
+ script to see how lsof_CCV is set by default.
+
+LSOF_HOST may be used to specify a value in lsof's -v output
+ other than the name of the host where lsof was
+ built. A value of "none" inhibits host name display
+ in -v output.
+
+ default: the dialect's host name application -- e.g.,
+ hostname or uname -n
+
+LSOF_INCLUDE is the path to the standard header files. You may
+ need to specify it if you want Configure to test
+ header files in a tree different from /usr/include,
+ and you want to compile lsof from the header files
+ in that different tree.
+
+LSOF_LOGNAME may be used to specify a value in lsof's -v output
+ other than the one in the LOGNAME environment
+ variable for the login name of the person who built
+ lsof. A value of "none" inhibits login name display
+ in -v output.
+
+ default: the LOGNAME environment variable
+
+LSOF_MKC may be used to specify an alternate method of
+ connecting dialect sources to the top-level lsof
+ directory. See 00PORTING for more information.
+
+ default: ln -s
+
+LSOF_SYSINFO may be used to specify a value in lsof's -v output
+ other than the standard system identification --
+ e.g., output from uname. A value of "none" inhibits
+ system information display in -v output.
+
+ default: the dialect's standard system identification
+ application output -- e.g., uname, sysinfo
+
+LSOF_USER may be used to specify a value in lsof's -v output
+ other than the one in the USER environment variable
+ for the login name of the person who built lsof.
+ A value of "none" inhibits login name display in
+ -v output.
+
+ default: the USER environment variable
+
+LSOF_VERS is the target dialect version number. It must be
+ stated in the dialect's form -- e.g., FreeBSD 2.0.5
+ is given as 205, Solaris 7 as 70000, etc. The
+ table, "Abbreviations, Variable Prefixes, and
+ Version Numbers," in this file gives the form for
+ LSOF_VERS for each dialect lsof supports.
+
+ default: auto-detection (e.g., from `uname -r`)
+
+LSOF_VSTR is the version string from which LSOF_VERS is
+ derived. Usually this is the output of `uname -r`
+ or `uname -v`. Consult the Configure script for
+ details. The LSOF_VSTR value may have to be quoted
+ if it contains spaces.
+
+ default: auto-detection (e.g., output from
+ `hostname`, `uname -r`, or `uname -v)
+
+
+Make Strings
+============
+
+The CFGF, CFGL, and DEBUG strings can be specified on the make
+command line to change default values placed in the top-level and
+library Makefiles by Configure. For example, Configure usually
+defines the compiler optimization level to be -O, but you can change
+that with "DEBUG=-g" on the make command -- e.g.,
+
+ $ make DEBUG=-g lsof
+
+Similarly, the CFGF string contains miscellaneous compile-time
+options, and CFGL contains loader options. Consult the Makefiles
+generated by Configure for the values it defines by default for
+CFGF and CFGL.
+
+As an example, Configure might define CFGL to be "-L./lib -llsof -w"
+for NextStep 3.1; to remove "-w", use this make invocation:
+
+ $ make CFGL="-L./lib -llsof"
+
+
+Abbreviations, Variable Prefixes, and Version Numbers
+=====================================================
+
+The following table describes the relationship between Configure
+abbreviations, environment variable prefixes, and lsof UNIX dialect
+version numbers. The lsof UNIX dialect version number must be
+declared exactly in the listed form when supplied via the LSOF_VERS
+environment variable.
+
+ Dialect Lsof Version
+ Configure Variable Version Number for
+Abbreviation* Prefix Number LSOF_VERS
+
+ aix AIX 3.2.5 3250
+ aixgcc 4.1.0 4100
+ 4.1.4 4140
+ 4.1.4 4150
+ 4.2.0 4200
+ 4.2.1 4210
+ 4.3 4300
+ 4.3.1 4310
+ 4.3.2 4320
+ 4.3.3 4330
+ 5.0.x 5000
+ 5.1.x 5100
+ 5.2.x 5200
+ 5.3.x 5300
+ darwin DARWIN 1.2* 120
+ 1.3* 130
+ 1.4* 140
+ 5.[012] 500
+ 5.[3-9] 530
+ 6.x 600
+ 7.x 700
+ 8.x 800
+ du DU 2.0 20000
+ 3.0 30000
+ 3.2 30200
+ 4.0 40000
+ 5.0 50000
+ 5.1 50100
+ freebsd FREEBSD 1.x 1000
+ 2.x 2000
+ 2.0.5 2005
+ 2.1.x 2010
+ 2.2.x 2020
+ 3.x 30x0
+ 4.x 40x0
+ 4.1x 41x0
+ 5.x 50x0
+ 6.x 60x0
+ hpux HPUX 9.1 901
+ hpuxgcc HPUX 9.5 905
+ 10.0 1000
+ 10.10 1010
+ 10.20 1020
+ 11.00 1100
+ 11.11 1111
+ linux LINUX 2.1.x 21xxx
+ 2.2.x 22xxx
+ 2.3.x 23xxx
+ 2.4.x 24xxx
+ 2.6.x 26xxx
+ netbsd NETBSD 1.2 1002000
+ 1.3 1003000
+ 1.4 1004000
+ 1.5 1005000
+ 1.6 1006000
+ 2.0 2000000
+ 2.99.9 2099009
+ 2.99.10 2099010
+ ns NEXTSTEP 3.1 31
+ openbsd OPENBSD 1.2 1020
+ 2.0 2000
+ 2.1 2010
+ 2.2 2020
+ 2.3 2030
+ 2.4 2040
+ 2.5 2050
+ 2.6 2060
+ 2.7 2070
+ 2.8 2080
+ 2.9 2090
+ 3.0 3000
+ 3.1 3010
+ 3.2 3020
+ 3.3 3030
+ 3.4 3040
+ 3.5 3050
+ 3.6 3060
+ os OPENSTEP 4.x 4x
+ osr OSR 3.2v2.0 20
+ 3.2v2.1 21
+ 3.2v4.0 40
+ 3.2v4.1 41
+ 3.2v4.2 42
+ 3.2v5.0.0 500
+ 3.2v5.0.2 502
+ 3.2v5.0.4 504
+ 3.2v5.0.6 506
+ ou OU 8.0.0 80000
+ solaris SOLARIS 2.3 20300
+ solariscc SOLARIS 2.4 20400
+ 2.5 20500
+ 2.5.1 20501
+ 2.6 20600
+ 7 70000
+ 8 80000
+ 9 90000
+ 10 100000
+ uw UW 7.0 70000
+ 7.1.0 70100
+ 7.1.1 70101
+ 7.1.3 70103
+
+* -- The optional Configure abbreviations -- e.g., the ``decosf''
+ and ``digital_unix'' alternatives to ``du'' -- aren't listed
+ here.
+
+
+Dialect-Specific Environment Variables
+======================================
+
+Here are the dialect-specific environment variables, listed
+alphabetically. The first part of any environment variable will
+be the dialect abbreviation, as specified to Configure, converted
+to upper case characters. See the `Configure -help` output for a
+listing of the abbreviations.
+
+AIX_ARCH specifies the AIX architecture when the AIX version is
+ 5.0 or higher. A value of "" signifies POWER; "ia64",
+ 64 bit x86 (Itanium).
+
+ default: none (tested via `uname -a`)
+
+AIX_HAS_AFS specifies the state of AIX ADS support when the AIX
+ version is 4.3.3 or lower. (Lsof doesn't support AFS
+ above AIX 4.3.3.) A value of "" allows the Configure
+ script to determine the AFS support state; "no",
+ disables AFS support; and "yes", forces the enabling of
+ AFS support.
+
+ default: none (tested via presence of AFS files and the
+ lsof AFSConfig shell script)
+
+AIX_KERNBITS specifies the kernel bit size, 32 or 64, of the Power
+ architecture AIX 5.x kernel for which lsof was built.
+
+ default: determined by the Configure script with a test
+ program that uses <sys/systemcfg.h> macros.
+
+AIX_USHACK If this environment variable has a value of "Y" or "y",
+ and if the aixgcc Configure abbreviation is selected,
+ the AIX 4.1 and greater gcc user structure hack is
+ activated; any other non-NULL value, it's not set; a
+ NULL value, it's tested by compilation.
+
+ default: none (tested by compilation)
+
+DARWIN_XNUDIR If this environment variable has a value, the value is
+ used as the path to the Darwin XNU kernel source code.
+
+ default: none (entry requested)
+
+DARWIN_XNU_HEADERS If this environment variable has a value, the value is
+ used as the path to the Darwin XNU kernel header files.
+ This path would match the DSTROOT environment variable
+ used when a "make installhdrs" was executed from the
+ Darwin XNU kernel source directory.
+
+ default: none
+
+DU_ADVFSV specifies the DEC OSF/1, Digital UNIX, or Tru64 UNIX
+ ADVFS file system version -- e.g., 200 for 2.0, 400
+ for 4.0, etc.
+
+ default: determined via /usr/sbin/setld
+
+DU_CDIR specifies the name of the DEC OSF/1, Digital UNIX, or
+ Tru64 UNIX system configuration directory.
+
+ default: first host name component, converted to upper
+ case
+
+DU_SHLIB specifies the DEC OSF/1, Digital UNIX, or Tru64 UNIX
+ shared library directory path.
+
+ default: /usr/shlib
+
+DU_SYSDIR DEC OSF/1, Digital UNIX, or Tru64 UNIX system
+ directory path.
+
+ 2.x and 3.x default: /sys
+ 4.x default: /usr/sys
+
+FREEBSD_KERNEL specifies the path to the FreeBSD kernel for FreeBSD
+ version less than 2.0.
+
+ default: /386bsd
+
+FREEBSD_SYS specifies the path to the FreeBSD system source
+ directory.
+
+ default: /sys
+
+HPUX_BASE specifies the HP-UX lsof source code base, kmem or
+ pstat, to be used.
+
+ default: determined by testing for the
+ /usr/include/sys/pstat subdirectory
+
+HPUX_BOOTFILE specifies the file in which lsof's Configure script can
+ find kernel information. This specification may be
+ useful for defining the path to a copy of /stand/vmunix
+ that has been processed by pxdb or q4pxdb.
+
+ default: /stand/vmunix
+
+HPUX_CCDIR1 specifies the first directory where Configure might
+ find an HP-UX C compiler. This is ignored when
+ LSOF_CC has been specified.
+
+ default: /bin
+
+HPUX_CCDIR2 specifies the second directory where Configure might
+ find an HP-UX C compiler. This is ignored when
+ LSOF_CC has been specified.
+
+ default: /usr/ccs/bin
+
+HPUX_HASONLINEJFS If this environment variable has a value of "Y" or "y",
+ the HASONLINEJFS definition will be enabled in the
+ Makefile CFLAGS. That will cause dnode1.c to use an
+ alternate vx_inode.h header file in the hpux11 sub-
+ directory of dialects/hpux/kmem.
+
+ default: determined using nm and grep
+
+HPUX_IPC_S_PATCH If this environment variable has a value of "1", the
+ ipc_s structure of the HP-UX 11 kernel is assumed to
+ have an ipc_ipis member, but it is assumed the ipis_s
+ structure lacks the ipis_msgsqueued member; "2", ipc_s
+ has ipc_ipis, but ipis_s has ipis_msgsqueued; "n" or
+ "N", ipc_s lacks ipc_ipis; any other non-NULL value is
+ considered an error; a NULL value, HPUX_IPC_S_PATCH is
+ determined by testing.
+
+ default: determined with q4 and grep
+
+HPUX_KERNBITS specifies the number of bits (32 or 64) in the HP-UX
+ 11 "basic kernel word.
+
+ default: `getconf _SC_KERNEL_BITS`
+
+HPUX_LIBC1 specifies the first directory that might contain the
+ HP-UX C library, libc.sl.
+
+ default: /usr/lib
+
+HPUX_LIBC2 specifies the second directory that might contain the
+ HP-UX C library, libc.sl.
+
+ default: /lib
+
+HPUX_RNODE3 If this environment variable has a value of "1", the
+ Configure script will define HASRNODE3 in the Makefile
+ CFGF flags. If it is defined, but not "1", Configure
+ will not define HASRNODE2.
+
+ default: determined using `nm -x /stand/vmunix` and
+ `grep r_fh3 /usr/include/nfs/rnode.h`
+
+HPUX_X25DIR specifies path to the HP-UX X25 directory that contains
+ configuration header files.
+
+ default: /etc/conf
+
+LINUX_CLIB specifies the definition of the Linux C library:
+
+ default: "" (standard C library)
+ others: -DGLIBCV=2 (glibc2)
+
+LINUX_HASSELINUX If this environment variable has a value of "Y" or "y",
+ Configure unconditionally activates SELinux support.
+ If it has any other value, Configure unconditionally
+ inhibits SELinux suport.
+
+ Default: assumed to be "Y" if <selinux/selinux.h>
+ exists
+
+LINUX_LSEEK If this environment variable has a value of "Y" or "y",
+ Configure uses Makefile.lseek in place of Makefile in
+ order to enable use of the private lseek() function for
+ 2.1.x kernels; any other non-NULL value, Makefile.lseek
+ will isn't used; a NULL value, the alternate lseek()
+ need is determined by compilation.
+
+ default: determined by test program
+
+NETBSD_SYS specifies the path to the NetBSD system source
+ directory.
+
+ default: /usr/include
+
+NETBSD_UVM If this environment variable has a value of "Y" or "y",
+ the NetBSD system uses the UVM virtual memory system;
+ any other non-NULL value, it does not; a NULL value,
+ it will be determined by the contents of /etc/mk.conf.
+
+ default: tested by grep'ping /etc/mk.conf
+
+OPENBSD_SYS specifies the path to the OpenBSD system source
+ directory.
+
+ default: /sys
+
+OPENBSD_UVM If this environment variable has a value of "Y" or "y",
+ the OpenBSD system uses the UVM virtual memory system;
+ any other non-NULL value, it does not; a NULL value,
+ it will be determined by examining /bsd.
+
+ default: tested by grep'ping `nm /bsd` output
+
+
+OSR_CFGF The value of this environment variable is made the
+ initial value for the compiler flags the lsof Configure
+ script constructs for the Makefile CFGF macro.
+
+ default: ""
+
+OSR_CFGL The value of this environment variable is made the
+ initial value for the loader flags the lsof Configure
+ script constructs for the Makefile CFGL macro.
+
+ default: ""
+
+OSR_STATLSTAT If this environment variable has a value of "Y" or "y",
+ HAS_STATLSTAT is defined in the Makefile's CFGL string;
+ any other non-NULL value, it's not defined; a NULL
+ value, it is determined with nm and grep.
+
+ default: determined with nm and grep
+
+
+SOLARIS_23P101318 If this environment variable has a non-NULL value, the
+ value is interpreted as the patch level of the Solaris
+ 2.3 P101318 patch.
+
+ default: pkginfo tested with grep
+
+SOLARIS_24P101945 If this environment variable has a non-NULL value, the
+ value is interpreted as the patch level of the Solaris
+ 2.4 P101945 patch.
+
+ default: pkginfo tested with grep
+
+SOLARIS_24P102303 If this environment variable has a non-NULL value, the
+ value is interpreted as the patch level of the Solaris
+ 2.4 P102303 patch.
+
+ default: pkginfo tested with grep
+
+SOLARIS_26PR_GWINDOWS If this environment variable has a value of "Y" or "y",
+ the HASPR_GWINDOWS definition is set in the Solaris 2.6
+ and 7 Makefile's CFG string; any other non-NULL value,
+ it's not set; a NULL value, it's tested by compilation.
+
+ default: tested by compilation
+
+SOLARIS_26PR_LDT If this environment variable has a value of "Y" or "y",
+ the HASPR_LDT definition is set in the Solaris 2.6
+ Makefile's CFGL string; any other non-NULL value, it's
+ not set; a NULL value, it's tested by compilation.
+
+ default: tested by compilation
+
+SOLARIS_CCDIR specifies the path to the Sun C compiler -- i.e., when
+ `Configure solariscc` is used. This is ignored when
+ LSOF_CC has been specified.
+
+ default: /opt/SUNWspro/bin
+
+SOLARIS_INSTR specifies the Sun C compiler target instruction set
+ when building lsof for a 64 bit kernel -- i.e., when
+ the Configure abbreviation is "solariscc". Possible
+ values include amd64 and sparcv9. This is ignored when
+ the Configure abbreviation is "solaris" -- i.e., the
+ compiler is gcc.
+
+ default: tested with /bin/isainfo -k
+
+SOLARIS_KERNBITS specifies the number of bits in the Solaris 7, 8, 9 or
+ 10 kernel: 32 or 64.
+
+ default: tested with /bin/isainfo -kv
+
+SOLARIS_VSOCK If this environment variable has a value of "Y" or "y",
+ the HAS_VSOCK definition is in the Solaris Makefile's
+ CFGL string; any other non-NULL value, it's not set; a
+ NULL value, it's tested by compilation.
+
+ default: tested by compilation
+
+SOLARIS_VXFSINCL This environment variable defines the path to the
+ header files of the VxFS 3.4 or greater version. If
+ SOLARIS_VXFSINCL is not set, the default is used.
+
+ default: VxFS < 4.0:
+ /opt/VRTSvxfs/include
+ VxFS 4.0 and above:
+ /opt/VRTSfssdk/<version>/include
+
+SOLARIS_VXFSLIB This environment variable defines the path to the
+ VxFS 3.4 or greater utility libraries, libvxfsutil.a
+ (32 bit) and libvxfsutil64.a (64 bit). If
+ SOLARIS_VXFSLIB is not set, the default is used.
+
+ Note: end SOLARIS_VXFSLIB at the "/lib" component; do
+ NOT put "/sparcv9" at its end. The lsof
+ Configure script will add "/sparcv9" if it is
+ required; hence, if Configure finds that
+ "/sparcv9" is needed, your SOLARIS_VXFSLIB
+ directory tree must have a sparcv9 subdirectory.
+
+ default: `dirname $SOLARIS_VXFSINCL`/lib
+
+SUN_AFSAPATHDEF specifies the path to the AFS library modload file
+ for either Solaris or SunOS.
+
+ default: /usr/vice/etc/modload/libafs
+ Verified with ls.
+
+ Note: the SunOS support is no longer maintained.
+
+UW_HAS_NSC If this environment variable has a value of "Y" or "y",
+ lsof will be configured for a UnixWare 7.1.1 or above
+ NonStop Cluster (NSC) system.
+
+ default: tested via /bin/node_self
+
+Vic Abell <abe@purdue.edu>
+April 15, 2008
--- /dev/null
+#!/bin/sh
+#
+# $Id: AFSConfig,v 1.2 99/05/09 14:49:54 abe Exp $
+#
+# AFSConfig: configure for AFS
+
+AFSD=/usr/vice/etc/afsd
+AH=AFSHeaders
+AV=AFSVersion
+STD=/usr/afsws/include
+
+# Establish trap and stty handling.
+
+ISIG=":"
+trap 'rm -f $AH $AV; $ISIG; exit 1' 1 2 3 15
+stty -a 2>&1 | grep isig > /dev/null
+if test $? -eq 0
+then
+ stty -a 2>&1 | egrep -e -isig > /dev/null
+ if test $? -eq 0
+ then
+ ISIG="stty -isig"
+ stty isig
+ fi
+fi
+
+# Decide how to use echo.
+
+ECHO=`echo -n ""`
+if test "X$ECHO" = "X-n "
+then
+ EC="\c"
+ EO=""
+else
+ EC=""
+ EO="-n"
+fi
+
+# Decide (perhaps for a second time) that AFS is installed.
+
+CELL=""
+if test -r /usr/vice/etc/ThisCell
+then
+ cell=`awk '{print $1}' /usr/vice/etc/ThisCell`
+ if test -d /afs/$cell
+ then
+ CELL=$cell
+ else
+ CELL=`echo $cell | sed 's/\([^.]*\)\..*/\1/'`
+ if test "X$CELL" != "X"
+ then
+ if test ! -d /afs/$CELL
+ then
+ CELL=""
+ fi
+ fi
+ fi
+fi
+if test "X$CELL" = "X"
+then
+ echo ""
+ echo "This system does not appear to have AFS installed."
+ exit 1
+fi
+
+# See if AFS configuration is wanted.
+
+cat << .CAT_MARK
+
+AFS appears to be installed on this system; cell name "$CELL".
+
+Lsof needs to be configured for AFS by identifying: 1) the directory
+that includes the AFS header files needed to compile AFS support into
+lsof; and 2) the version of AFS that is installed.
+.CAT_MARK
+
+END=0
+while test $END = 0
+do
+ echo ""
+ echo $EO "Do you want to configure lsof for AFS (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ exit 1
+ fi
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ END=1
+ else
+ echo ""
+ echo "Please answer y or n."
+ fi
+done
+
+# See if $AH exists and points to a likely place.
+
+AHOK=""
+echo ""
+echo "====================================================================="
+echo ""
+if test -r $AH
+then
+ AHP=`cat $AH`
+ if test -r $AHP
+ then
+ if test -r $AHP/afs/afs.h
+ then
+ cat << .CAT_MARK
+The location of the AFS header files required by lsof has been
+previously identified as "$AHP".
+
+Since <afs/afs.h> can be found there, that's probably correct.
+.CAT_MARK
+
+ END=0
+ while test $END = 0
+ do
+ echo ""
+ echo $EO "Do you want to use $AHP again (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ AHOK="ok"
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ rm -f $AH
+ AHP=""
+ END=1
+ else
+ echo ""
+ echo "Please answer y or n."
+ fi
+ fi
+ done
+ else
+ echo "\"$AHP\" has been previously specified as the location of the"
+ echo "AFS header files, but it lacks an afs/afs.h header file."
+ rm -f $AH
+ AHP=""
+ fi
+ else
+ echo "The file ./$AH exists, but has no AFS header file path in it."
+ rm -f $AH
+ AHP=""
+ fi
+else
+ echo "No previous header location has been specified."
+ rm -f $AH
+ AHP=""
+fi
+
+# See if the header files are in the "standard" place.
+
+if test "X$AHOK" != "Xok"
+then
+ if test -r $STD
+ then
+ echo ""
+ echo "====================================================================="
+ echo ""
+ echo "The AFS header files appear to be in the \"standard\" location --"
+ echo "i.e.: \"$STD\"."
+ END=0
+ while test $END = 0
+ do
+ echo ""
+ echo $EO "Do you want to let lsof use them (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ echo $STD > $AH
+ AHOK="ok"
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ END=1
+ else
+ echo ""
+ echo "Please answer y or n."
+ fi
+ fi
+ done
+ fi
+fi
+
+# Ask for the AFS header file location.
+
+if test "X$AHOK" != "Xok"
+then
+ echo ""
+ echo "====================================================================="
+ echo ""
+ echo "Please specify the full path where lsof can find the AFS header"
+ echo "files. A possible location is: \"/afs/$CELL/<sysname>/include\"."
+ cat << .CAT_MARK
+The <sysname> component of the path is the AFS system name that
+was used to configure and build AFS on this system. It is usually
+constructed from a manufacturer or Unix operating system designation,
+followed by a version number -- e.g., hp800_ux90, sun4m_54, vax_ul43,
+etc. You may have to consult your AFS documentation to determine
+what <sysname> applies to your configuration.
+.CAT_MARK
+
+ END=0
+ while test $END = 0
+ do
+ echo ""
+ echo $EO "Do you want to see the contents of /afs/$CELL (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ END=1
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ echo ""
+ ls -C /afs/$CELL
+ END=1
+ else
+ echo ""
+ echo "Please answer y or n."
+ fi
+ fi
+ done
+ END=0
+ while test $END = 0
+ do
+ echo ""
+ echo $EO "AFS header file path: $EC"
+ read ANS EXCESS
+ fc=`expr "${ANS}X" : '\(.\).*'`
+ if test "X$fc" = "X/"
+ then
+ if test -r $ANS/afs/afs.h
+ then
+ echo $ANS > $AH
+ AHOK="ok"
+ END=1
+ else
+ echo ""
+ echo "$ANS/afs/afs.h doesn't exist."
+ echo "Please enter a path whose afs subdirectory contains afs.h"
+ fi
+ else
+ echo ""
+ echo "Please enter an absolute path name."
+ fi
+ done
+fi
+if test "X$AHOK" != "Xok"
+then
+ echo "AFSConfig: unknown error"
+ exit 1
+fi
+
+# Determine AFS version.
+
+if test -r $AV
+then
+ echo ""
+ echo "====================================================================="
+ echo ""
+ AVN=`cat $AV`
+ cat << .CAT_MARK
+The AFS version was previously specified as: $AVN
+.CAT_MARK
+
+ END=0
+ while test $END = 0
+ do
+ echo ""
+ echo $EO "Is this the correct version number (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ exit 0
+ fi
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ rm -f $AV
+ END=1
+ else
+ echo "Please answer y or n."
+ fi
+ done
+fi
+
+# See if the version number can be determined.
+
+if test -r $AFSD
+then
+ ANS=`strings $AFSD | grep "Base configuration afs" | sed 's/^.*ion afs\([^ ]*\) .*/\1/'`
+ TV=`echo $ANS | sed 's/^\([0-9]*\)\.\([0-9]*\)\(.*\)/\1 \2 \3/' | awk '{printf "%d.%d%s\n",$1,$2,$3}'`
+ if test "X$ANS" = "X$TV"
+ then
+ echo ""
+ echo "====================================================================="
+ echo ""
+ cat << .CAT_MARK
+Examining $AFSD the AFS version number appears to be: $TV
+.CAT_MARK
+
+ END=0
+ while test $END = 0
+ do
+ echo ""
+ echo $EO "Do you want to use this version number (y/n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ END=1
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ echo $TV > $AV
+ exit 0
+ else
+ echo ""
+ echo "Please answer y or n."
+ fi
+ fi
+ done
+ fi
+fi
+
+# Ask for the version number.
+
+echo ""
+echo "====================================================================="
+END=0
+while test $END = 0
+do
+ echo ""
+ echo $EO "Please enter the AFS version number: $EC"
+ read ANS EXCESS
+ TV=`echo $ANS | sed 's/^\([0-9]*\)\.\([0-9]*\)\(.*\)/\1 \2 \3/' | awk '{printf "%d.%d%s\n",$1,$2,$3}'`
+ if test "X$ANS" = "X$TV"
+ then
+ echo $TV > $AV
+ exit 0
+ fi
+done
--- /dev/null
+#!/bin/sh
+#
+# Configure -- configure lsof
+#
+# See the LSOF_HLP here document for usage.
+#
+# See the lsof distribution file 00XCONFIG for information on setting
+# environment variables for cross-configuring lsof -- e.g., for configuring
+# for Linux 2.3 on a machine running 2.4. Marty Leisner suggested this
+# support and provided the Linux Configure stanza modifications.
+#
+# When configuring for a particular dialect, <target-dialect>, this script
+# requires that the subdirectory ./dialects/<target-dialect> contain a
+# shell script, named $LSOF_MK, that places its source modules in this
+# directory.
+#
+# $Id: Configure,v 1.157 2009/03/25 19:23:55 abe Exp $
+
+#LSOF_CFLAGS_OVERRIDE=1 may be introduced through the environment to cause
+# the library Makefile's CFLAGS definition to override any in the
+# environment.
+
+#LSOF_DISTRIBKVM may be introduced through the environment to specify the
+# Sun4 kernel virtual memory type of distrib.cf
+
+LSOF_F="ddev.c dfile.c dlsof.h dmnt.c dnode*.c dproc.c dproto.h dsock.c dstore.c dzfs.h kernelbase.h machine.h machine.h.old new_machine.h __lseek.s"
+LSOF_HLP_BASE=./cfghlp.
+LSOF_HLP=${LSOF_HLP_BASE}$$
+
+#LSOF_LOCALSUFFIX may be introduced through the environment to select a local
+# version of a Makefile. It is used as a suffix to $LSOF_MKF.
+
+LSOF_MK=Mksrc
+
+#LSOF_MKC is the dialect's Mksrc create command -- default "ln -s".
+
+# LSOF_MKFC may be introduced though the environment to change the name
+# used for the created make file.
+
+if test "X$LSOF_MKFC" = "X" # {
+then
+ LSOF_MKFC=Makefile
+fi # }
+
+LSOF_LIB=lib
+LSOF_MKF=Makefile
+LSOF_LIBMKF=Makefile
+LSOF_LIBMKFSKEL=Makefile.skel
+
+LSOF_VF=version
+
+# Make sure no other variable important to Makefile construction is
+# already set in the environment.
+#
+# $AFS_VICE locatiion of AFS VICE directory
+# (default = /usr/vice)
+# $LSOF_AFS AFS temporary
+# $LSOF_AFS_NQ AFS-not-qualified flag
+# $LSOF_AFSV AFS version
+# $LSOF_AR archive command and its arguments for making the
+# lsof library
+# $LSOF_ARCH Unix dialect architecture as a string -- may be
+# supplied externally
+# $LSOF_CC C compiler name (may be supplied externally)
+# $LSOF_CCV C compiler version (may be supplied externally)
+# $LSOF_CDIR configuration directory
+# $LSOF_CFGD depend options
+# $LSOF_CFGDN depend file name
+# $LSOF_CFGF C flags -- e.g., -D's
+# $LSOF_CFGL last lsof library loader flags -- e.g., -l's
+# $LSOF_CINFO Configure information for LSOF_CINFO in version.h
+# $LSOF_DEBUG Makefile's DEBUG string
+# $LSOF_DINC include flags -- -I's
+# $LSOF_DINC_ADD include flags status
+# $LSOF_DOC special document (man page) directory path
+# $LSOF_ERR internal error flag
+# $LSOF_FCFGL first lsof library loader flags -- e.g., -l's
+# that must precede $LSOF_LIB
+# $LSOF_FBSD_ZFS FreeBSD $LSOF_FBSD_ZFS_MKF status
+# $LSOF_FBSD_ZFS_CFGF FreeBSD ZFS configure flags
+# $LSOF_FBSD_ZFS_MKF FreeBSD ZFS Makefile name
+# $LSOF_FBSD_ZFS_SYS FreeBSD ZFS system sources location
+# $LSOF_HOST host name (e.g., from uname -n)
+# $LSOF_INCLUDE directory where header files are found
+# (default = /usr/include)
+# $LSOF_LD loader name if not $LSOF_CC
+# $LSOF_LIB_NO if "N" don't configure the lsof library
+# $LSOF_LINUX_INCL temporary variable for holding path to Linux
+# include files
+# $LSOF_LOCALSUFFIX local suffix for Makefile
+# $LSOF_NBSD_BUFQH NetBSD <sys/bufq.h> copy status
+# $LSOF_NBSD_PTYFS NetBSD ${NETBSD_SYS}/sys/fs/ptyfs/ copy status
+# $LSOF_N_UNIXV *BSD system's kernel file
+# $LSOF_PL patch level
+# $LSOF_RANLIB randomizing command for the lsof library
+# $LSOF_SCRIPT_CALL Customize and Inventory scripts call status
+# $LSOF_SPMKF Special Makefile name
+# $LSOF_TGT canonical target abbreviation (shortest)
+# $LSOF_TMP1 internal temporary
+# $LSOF_TMP2 internal temporary
+# $LSOF_TMP3 internal temporary
+# $LSOF_TMP4 internal temporary
+# $LSOF_TMP5 internal temporary
+# $LSOF_TMP6 internal temporary
+# $LSOF_TMPC_BASE base name for $LSOF_TMPC
+# $LSOF_TMPC temporary C source file base name
+# $LSOF_TSTBIGF big file capability (for $LSOF_TSTCFLG)
+# $LSOF_TSTCC tests CC file
+# $LSOF_TSTCFLG tests CFLAGS file
+# $LSOF_TSTDFLG dialect-specific values for $LSOF_TSTCFLG
+# $LSOF_TSTK64 status of 64 bit kernel (for $LSOF_TSTCFLG)
+# $LSOF_TSTKMEM /dev/kmem usage status (for $LSOF_TSTCFLG)
+# $LSOF_TSTLFF tests LDFLAGS file
+# $LSOF_TSTLFLG tests LDFLAGS values
+# $LSOF_TSTSUBD test subdirectory
+# $LSOF_TSTVPATH test v_path state (for $LSOF_TSTCFLG)
+# $LSOF_TSTXO test extra objects (for $LSOF_TSTXOC)
+# $LSOF_TSTXOC test extra objects file
+# $LSOF_UNSUP Lsof is unsupported on this dialect
+# $LSOF_VERS Unix dialect version as a decimal number -- may
+# be supplied externally
+# $LSOF_VSTR Unix dialect version as a string -- may be supplied
+# externally
+
+if test "X$AFS_VICE" = "X" # {
+then
+ AFS_VICE="/usr/vice"
+fi # }
+LSOF_AFS=""
+LSOF_AFS_NQ=""
+LSOF_AFSV=""
+if test "X$LSOF_ARCH" = "X" # {
+then
+ LSOF_ARCH=""
+fi # }
+LSOF_CDIR=""
+LSOF_CFGD=""
+LSOF_CFGDN=""
+LSOF_CFGF=""
+LSOF_CFGL=""
+LSOF_CINFO=""
+LSOF_DEBUG=""
+LSOF_DOC=""
+LSOF_ERR=""
+LSOF_FCFGL=""
+LSOF_FBSD_ZFS=0
+LSOF_FBSD_ZFS_CFGF=""
+LSOF_FBSD_ZFS_MKF="Makefile.zfs"
+LSOF_FBSD_ZFS_SYS=""
+LSOF_HOST=""
+if test "X$LSOF_INCLUDE" = "X" # {
+then
+ LSOF_DINC=""
+ LSOF_INCLUDE="/usr/include"
+else
+ LSOF_DINC="-I$LSOF_INCLUDE"
+fi # }
+LSOF_LD=""
+LSOF_LIB_NO=""
+LSOF_LINUX_INCL=""
+LSOF_PL=""
+LSOF_RANLIB="ranlib \${LIB}"
+LSOF_SCRIPT_CALL="yes"
+LSOF_SPMKF=""
+LSOF_TMP1=""
+LSOF_TMP2=""
+LSOF_TMPC_BASE=./lsof_Configure_tmp_
+LSOF_TMPC=${LSOF_TMPC_BASE}$$
+LSOF_TSTBIGF=""
+LSOF_TSTSUBD="./tests"
+LSOF_TSTCC="${LSOF_TSTSUBD}/config.cc"
+LSOF_TSTCFLG="${LSOF_TSTSUBD}/config.cflags"
+LSOF_TSTDFLG=""
+LSOF_TSTK64=0
+LSOF_TSTKMEM=1
+LSOF_TSTLFF="${LSOF_TSTSUBD}/config.ldflags"
+LSOF_TSTLFLG=""
+LSOF_TSTVPATH=0
+LSOF_TSTXO=""
+LSOF_TSTXOC="${LSOF_TSTSUBD}/config.xobj"
+LSOF_UNSUP="WARNING: unsupported dialect or version"
+if test "X$LSOF_VERS" = "X" # {
+then
+ LSOF_VERS=""
+fi # }
+if test "X$LSOF_VSTR" = "X" # {
+then
+ LSOF_VSTR=""
+fi # }
+
+# Establish echo type -- Berkeley or SYSV.
+
+j=`echo -n ""`
+if test "X$j" = "X-n "
+then
+ EC="\c"
+ EO=""
+else
+ EC=""
+ EO="-n"
+fi
+
+# Make sure temporary files are removed before an abnormal exit.
+
+trap 'rm -f ${LSOF_HLP_BASE}* ${LSOF_TMPC_BASE}*; exit 1' 1 2 3 15
+
+rm -f $LSOF_HLP
+cat > $LSOF_HLP << LSOF_HLP
+Usage: Configure <options> <target-dialect>
+ <options>: -clean : clean up previous configuration
+ -d|-dialects : display a list of supported dialect versions
+ -h|-help : display help information
+ -n : avoid AFS, customization, and inventory checks
+ <target-dialect> (****USE -d TO GET TESTED DIALECT VERSION NUMBERS****):
+ aix|aixgcc : IBM AIX xlc (aix) or gcc (aixgcc)
+ darwin : Apple Darwin
+ decosf : DEC OSF/1
+ digital_unix|du : Digital UNIX
+ freebsd : FreeBSD
+ hpux|hpuxgcc : HP-UX cc (hpux) or gcc (hpuxgcc)
+ linux : Linux
+ netbsd : NetBSD
+ nextstep|next|ns|nxt : NEXTSTEP
+ openbsd : OpenBSD
+ openstep|os : OPENSTEP
+ osr|sco : SCO OpenServer < 6.0.0, SCO devloper's compiler
+ osrgcc|scogcc : SCO OpenServer < 6.0.0, gcc compiler
+ osr6 : SCO OpenServer 6.0.0, SCO compiler
+ solaris|solariscc : Solaris gcc (solaris) or cc (solariscc)
+ tru64 : Tru64 UNIX
+ unixware|uw : SCO|Caldera UnixWare
+LSOF_HLP
+
+LSOF_TGT="no-target"
+
+args=$#
+while test $args -gt 0 # {
+do
+ case $1 in # {
+ -clean)
+ if test -r $LSOF_MKFC # {
+ then
+ echo "make -f $LSOF_MKFC clean"
+ make -f $LSOF_MKFC clean
+ else
+ if test -r ${LSOF_LIB}/${LSOF_LIBMKF} # {
+ then
+ echo "(cd ${LSOF_LIB}; make -f ${LSOF_LIBMKF} clean)"
+ (cd ${LSOF_LIB}; make -f ${LSOF_LIBMKF} clean)
+ else
+ if test -r ${LSOF_LIB}/${LSOF_LIBMKF}.skel # {
+ then
+ echo "(cd ${LSOF_LIB}; make -f ${LSOF_LIBMKF}.skel clean)"
+ (cd ${LSOF_LIB}; make -f ${LSOF_LIBMKF}.skel clean)
+ fi # }
+ fi # }
+ fi # }
+ if test -r ${LSOF_TSTSUBD}/Makefile # {
+ then
+ echo "(cd ${LSOF_TSTSUBD}; make spotless)"
+ (cd ${LSOF_TSTSUBD}; make spotless)
+ else
+ echo '(cd ${LSOF_TSTSUBD}; rm *.o config.*)'
+ (cd ${LSOF_TSTSUBD}; rm *.o config.*)
+ fi # }
+ rm -f $LSOF_F $LSOF_MKFC $LSOF_FBSD_ZFS_MKF ${LSOF_TMPC_BASE}*
+ echo rm -f $LSOF_F $LSOF_MKFC $LSOF_FBSD_ZFS_MKF ${LSOF_TMPC_BASE}*
+ rm -rf AFSHeaders AFSVersion version.h vnode_if.h
+ echo "rm -rf AFSHeaders AFSVersion version.h vnode_if.h"
+ rm -f ${LSOF_HLP_BASE}* cd9660_node.h lockf_owner.h fbsd_minor.h
+ echo "rm -f ${LSOF_HLP_BASE}* cd9660_node.h lockf_owner.h fbsd_minor.h"
+ rm -f dialects/aix/aix5/j2/j2_snapshot.h
+ echo "rm -f dialects/aix/aix5/j2/j2_snapshot.h"
+ rm -f dialects/sun/solaris10 # DEBUG -- for s10_44
+ echo "rm -f dialects/sun/solaris10" # DEBUG -- for s10_44
+ rm -f dialects/du/du5_sys_malloc.h
+ echo "rm -f dialects/du/du5_sys_malloc.h"
+ rm -f dialects/hpux/kmem/hpux_mount.h
+ echo "rm -f dialects/hpux/kmem/hpux_mount.h"
+ rm -rf dialects/n+obsd/include
+ echo "rm -rf dialects/n+obsd/include"
+ rm -f dialects/uw/uw7/vm/swap.h
+ echo "rm -f dialects/uw/uw7/vm/swap.h"
+ rm -f ${LSOF_LIB}/${LSOF_LIBMKF}
+ echo "rm -f ${LSOF_LIB}/${LSOF_LIBMKF}"
+ exit 0
+ ;;
+
+ -d|-dialects)
+ if test -r ./00DIALECTS -a -r ./version # {
+ then
+ V=`sed '/VN/s/.ds VN \(.*\)/\1/' version`
+ echo "lsof $V has been *tested* on these UNIX dialect versions:"
+ cat 00DIALECTS
+ echo Although "$V hasn't been tested on other versions of these dialects,"
+ echo "it may work. Try \`Configure <dialect>\` and \`make\` to see."
+ rm -f $LSOF_HLP
+ exit 0
+ else
+ echo "Can't display UNIX dialect version information:"
+ if test ! -r ./00DIALECTS # {
+ then
+ echo " ./00DIALECTS is inaccessible."
+ fi # }
+ if test ! -r ./version # {
+ then
+ echo " ./version is inaccessible."
+ fi # }
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ ;;
+
+ -h|-help) cat $LSOF_HLP
+ rm -f $LSOF_HLP
+ exit 0
+ ;;
+
+ -n*)
+ LSOF_SCRIPT_CALL="no"
+ ;;
+
+ *)
+ if test "X$LSOF_TGT" != "Xno-target" # {
+ then
+ echo "Only one dialect may be configured at a time."
+ echo 'Both "$LSOF_TGT" and "$1" were specified.'
+ cat $LSOF_HLP
+ rm -f $LSOF_HLP
+ exit 1
+ else
+ LSOF_TGT=$1
+ fi # }
+ ;;
+ esac # }
+ shift
+ args=`expr $args - 1`
+done # }
+
+case $LSOF_TGT in # {
+ no-target)
+ echo "No target dialect was specified."
+ cat $LSOF_HLP
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+
+# Configure for AIX xlc and AIX gcc.
+
+ aix|aixgcc)
+
+ # AIXA stands for AIX architecture. It is assigned these values in this
+ # stanza:
+ #
+ # 0 The AIX version is < 5.0, or the AIX 5.0 architecture is
+ # Power and the kernel bit size is 32.
+ #
+ # 1 The AIX version is >= 5.0, the AIX architecture is Power,
+ # and the kernel bit size is 64.
+ #
+ # 2 The AIX version is >= 5.0 and the architecture is IA64.
+
+ LSOF_RANLIB="@echo \\\\\\\\c" # AIX make doesn't like a null ${RANLIB}.
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+
+ # If the AIX version isn't pre-defined, determine it.
+
+ LSOF_TMP1=`uname -v`
+ if test "X$LSOF_TMP1" = "X5" # {
+ then
+
+ # If the AIX version is 5, build the version string with `uname -rv`
+ # output.
+
+ LSOF_VSTR=`uname -r | awk '{printf "5.%d.0.0\n",\$1}'`
+ echo "Uname reports the version is $LSOF_VSTR."
+ else
+
+ # See if oslevel can determine the version.
+
+ LSOF_TMP1=/usr/bin/oslevel
+ if test -x $LSOF_TMP1 # {
+ then
+ echo "Determining AIX version with $LSOF_TMP1."
+ echo "This may take a while, depending on your maintenance level."
+ LSOF_VSTR=`$LSOF_TMP1 | sed 's/[^0-9]*\([0-9\.]*\).*/\1/'`
+ echo "$LSOF_TMP1 reports the version is $LSOF_VSTR."
+ else
+
+ # If oslevel can't be used, build the version string with
+ # `uname -rv` and issue a warning.
+
+ LSOF_VSTR=`uname -rv | awk '{printf "%d.%d.0.0\n",\$2,\$1}'`
+ echo "WARNING: can't execute $LSOF_TMP1; uname -rv reports"
+ echo " the version is $LSOF_VSTR; edit CFGF in Makefile and"
+ echo " lib/Makefile to refine AIXV and LSOF_VSTR."
+ fi # }
+ fi # }
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+ LSOF_VERS=`echo $LSOF_VSTR | sed 's/\.//g'`
+ fi # }
+ if test $LSOF_VERS -ge 4320 # {
+ then
+ LSOF_TSTBIGF=" "
+ fi # }
+ if test "X$LSOF_CC" = "X" # {
+ then
+ if test "X$LSOF_TGT" = "Xaixgcc" # {
+ then
+ LSOF_CC=gcc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ else
+ LSOF_CC=cc
+ fi # }
+ fi # }
+ LSOF_TGT="aix"
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+
+ # Prevent use of gcc for AIX below 4.1.
+
+ if test $LSOF_VERS -lt 4100 # {
+ then
+ echo "********************************************************"
+ echo "* Sorry, but gcc can't be used to compile lsof for AIX *"
+ echo "* versions less than 4.1, because of possible kernel *"
+ echo "* structure alignment differences between it and xlc. *"
+ echo "********************************************************"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+
+ # Test for AFS.
+
+ if test "X$AIX_HAS_AFS" != "X" # {
+ then
+ LSOF_AFS=$AIX_HAS_AFS
+ fi # }
+ if test "X$LSOF_AFS" != "Xno" # {
+ then
+ if test "X$LSOF_AFS" = "Xyes" -o -r ${AFS_VICE}/etc/ThisCell # {
+ then
+ if test "X$LSOF_AFS" != "Xyes" # {
+ then
+ if test "X$LSOF_SCRIPT_CALL" = "Xno" # {
+ then
+ if test -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ else
+ if test ! -x ./AFSConfig # {
+ then
+ echo "Can't find or execute the AFSConfig script"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ ./AFSConfig
+ if test $? -eq 0 -a -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ fi # }
+ fi # }
+ if test "X$LSOF_AFS" = "Xyes" # {
+ then
+ if test "X$LSOF_AFSV" = "X" # {
+ then
+ if test -r ./AFSVersion # {
+ then
+ LSOF_AFSV=`cat ./AFSVersion | sed 's/^\([0-9]*\)\.\([0-9]*\).*/\1 \2/' | awk '{printf "%d%02d\n",\$1,\$2}'`
+ else
+ echo "!!!FATAL: no ./AFSVersion file. It should have been"
+ echo " created by a previous AFS configuration run."
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+ if test $LSOF_VERS -gt 4330 -o LSOF_AFSV -gt 305 # {
+ then
+ echo "!!!FATAL: Lsof does not support AFS on this combination of"
+ echo " AIX ($LSOF_VERS) and AFS ($LSOF_AFSV) versions."
+ echo " To disable AFS, set the value of the AIX_HAS_AFS"
+ echo " environment variable to \"no\"."
+ rm -f $LSOF_HLP
+ exit 1
+ else
+ LSOF_CFGF="$LSOF_CFGF -DHAS_AFS=$LSOF_AFSV"
+ LSOF_DINC="$LSOF_DINC -I`cat ./AFSHeaders`"
+ if test -r ${LSOF_INCLUDE}/sys/inttypes.h # {
+ then
+ grep "^typedef.*int16;" ${LSOF_INCLUDE}/sys/inttypes.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASINT16TYPE"
+ fi # }
+ grep "^typedef.*u_int32;" ${LSOF_INCLUDE}/sys/inttypes.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASUINT16TYPE"
+ fi # }
+ grep "^typedef.*int32;" ${LSOF_INCLUDE}/sys/inttypes.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASINT32TYPE"
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+
+ # Miscellaneous AIX tests
+
+ if test -d ${LSOF_INCLUDE}/nfs # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_NFS"
+ fi # }
+ echo $LSOF_CC | grep cc | grep -v gcc > /dev/null
+ if test $? -eq 0 -a $LSOF_VERS -ge 4140 -a $LSOF_VERS -lt 5000 # {
+ then
+ LSOF_CFGL="$LSOF_CFGL -bnolibpath"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/socket.h # {
+ then
+ grep AF_INET6 ${LSOF_INCLUDE}/sys/socket.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/stat.h # {
+ then
+ grep stat64 ${LSOF_INCLUDE}/sys/stat.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASSTAT64"
+ fi # }
+ fi # }
+#DEBUG SANFS if test -r ${LSOF_INCLUDE}/sys/sanfs/sanfsnode.h??? # {
+#DEBUG SANFS then
+#DEBUG SANFS LSOF_CFGF="$LSOF_CFGF -DHAS_SANFS"
+#DEBUG SANFS fi # }
+ if test $LSOF_VERS -ge 5000 # {
+ then
+
+ # This is AIX 5 or greater.
+
+ if test -d ${LSOF_INCLUDE}/j2 # {
+ then
+
+ # The AIX > 5.0 system has jfs2 support. Make the necesssary definitions
+ # and adjustments.
+
+ rm -f dialects/aix/aix5/j2/j2_snapshot.h
+ (cd dialects/aix/aix5/j2; ln -s private_j2_snapshot.h j2_snapshot.h)
+ LSOF_CFGF="$LSOF_CFGF -DHAS_JFS2"
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/aix/aix5"
+ if test $LSOF_VERS -ge 5200 # {
+ then
+ if test -r ${LSOF_INCLUDE}/j2/j2_snapshot.h # {
+ then
+
+ # The system has its own j2_snapshot.h, so make sure the
+ # private lsof copy is discarded.
+
+ rm -f dialects/aix/aix5/j2/j2_snapshot.h
+ fi # }
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+
+ # Test gcc version for AIX 5.2.
+
+ LSOF_TMP1=`echo $LSOF_CCV | awk -F . '{printf "%d%02d",$1,$2}'`
+ if test $LSOF_TMP1 -ge 303 # {
+ then
+
+ # Add gcc >= 3.3 option to handle use of i_dev from the wInode
+ # anonymous structure reference in the JFS2 inode structure of
+ # <j2/j2_inode.h>.
+
+ LSOF_CFGF="$LSOF_CFGF -fms-extensions"
+ fi # }
+ fi #}
+ fi # }
+ fi # }
+
+ # Determine the AIX architecture type and set AIXA accordingly.
+
+ if test "X$AIX_ARCH" = "X" # {
+ then
+ uname -a | grep -i ia64 > /dev/null
+ if test $? -eq 0 # {
+ then
+ AIX_ARCH="ia64"
+ else
+ AIX_ARCH=""
+ fi # }
+ fi # }
+ if test "X$AIX_ARCH" = "Xia64" # {
+ then
+
+ # This is AIX >= 5 on ia64.
+
+ LSOF_TSTK64=1
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+
+ # Quit if gcc was specified as the compiler, since the gcc options to
+ # do an ia64 lsof compilation are unknown.
+
+ echo "*************************************************************"
+ echo "* *"
+ echo "* !!!!!!!!!!!!!!!!!!!!! FATAL ERROR !!!!!!!!!!!!!!!!!!!!!!! *"
+ echo "* *"
+ echo "* Gcc can't be used to compile lsof for AIX 5 and above on *"
+ echo "* the ia64 architecture. Consult lsof's FAQ (in the file *"
+ echo "* 00FAQ) for more information. *"
+ echo "* *"
+ echo "*************************************************************"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ LSOF_TMP1=2
+ if test "X$LSOF_AR" = "X" # {
+ then
+ LSOF_AR="/usr/bin/ar cr"
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -q64"
+ LSOF_CFGL="$LSOF_CFGL -lelf"
+ else
+
+ # This is AIX >= 5 on Power architecture.
+
+ echo $LSOF_CC | grep cc | grep -v gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGL="$LSOF_CFGL -bnolibpath"
+ fi # }
+ if test "X$AIX_KERNBITS" = "X" # {
+ then
+
+ # The kernel bit size wasn't predefined. Determine it by compiling
+ # and executing a test program.
+
+ rm -f ${LSOF_TMPC}.*
+ echo "#include <sys/systemcfg.h>" > ${LSOF_TMPC}.c
+ echo 'main(){ if (__KERNEL_32()) printf("32\\n");' >> ${LSOF_TMPC}.c
+ echo 'else if (__KERNEL_64()) printf("64\\n");' >> ${LSOF_TMPC}.c
+ echo 'else printf("0\\n");' >> ${LSOF_TMPC}.c
+ echo "return(0); }" >> ${LSOF_TMPC}.c
+ echo "Testing kernel bit size with $LSOF_CC"
+ $LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x
+ if test ! -x ${LSOF_TMPC}.x # {
+ then
+ echo "!!!FATAL: can't compile test program, ${LSOF_TMPC}.c."
+ rm -f $LSOF_HLP rm -f ${LSOF_TMPC}.*
+ exit 1
+ fi # }
+ AIX_KERNBITS=`./${LSOF_TMPC}.x`
+ rm -f ${LSOF_TMPC}.*
+ fi # }
+
+ # Use the kernel bit size specification to select archiver and compiler
+ # options, and to update AIXA.
+
+ case $AIX_KERNBITS in # {
+ 32)
+ if test "X$LSOF_AR" = "X" # {
+ then
+ LSOF_AR="/usr/bin/ar cr"
+ fi # }
+ LSOF_TMP1=0
+ ;;
+ 64)
+ if test "X$LSOF_AR" = "X" # {
+ then
+ LSOF_AR="/usr/bin/ar -X 64 -v -q"
+ fi # }
+ LSOF_TSTK64=1
+ LSOF_TMP1=1
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -maix64"
+ else
+ LSOF_CFGF="$LSOF_CFGF -q64"
+ fi # }
+ ;;
+ *)
+ echo "!!!FATAL: unrecognized kernel bit size: $AIX_KERNBITS"
+ rm -f $LSOF_HLP
+ exit 1
+ esac # }
+
+ # Put kernel bit size information in $LSOF_CINFO and $LSOF_CFGF.
+
+ echo "Kernel bit size: $AIX_KERNBITS"
+ LSOF_TMP2="${AIX_KERNBITS} bit kernel"
+ if test "X$LSOF_CINFO" != "X" # {
+ then
+ LSOF_CINFO="${LSOF_CINFO} ${LSOF_TMP2}"
+ else
+ LSOF_CINFO="${LSOF_TMP2}"
+ fi # }
+ LSOF_CFGF="${LSOF_CFGF} -DAIX_KERNBITS=${AIX_KERNBITS}"
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -DAIXA=$LSOF_TMP1"
+ if test "X$LSOF_TSTDFLG" = "X" # {
+ then
+ LSOF_TSTDFLG="-DLT_AIXA=$LSOF_TMP1"
+ else
+ LSOF_TSTDFLG="$LSOF_TSTDFLG -DLT_AIXA=$LSOF_TMP1"
+ fi # }
+ else
+
+ # AIX is < 5, so set AIXA accordingly.
+
+ LSOF_CFGF="$LSOF_CFGF -DAIXA=0"
+ if test "X$LSOF_TSTDFLG" = "X" # {
+ then
+ LSOF_TSTDFLG="-DLT_AIXA=0"
+ else
+ LSOF_TSTDFLG="$LSOF_TSTDFLG -DLT_AIXA=0"
+ fi # }
+ fi #}
+ LSOF_CFGF="$LSOF_CFGF -DAIXV=$LSOF_VERS"
+ LSOF_DIALECT_DIR=aix
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+
+ # Do gcc tests.
+
+ if test $LSOF_VERS -ge 4100 -a $LSOF_VERS -lt 4200 # {
+ then
+ if test "X$AIX_USHACK" = "X" # {
+ then
+
+ # Compile and run a gcc test program to evaluate the user structure.
+
+ rm -f ${LSOF_TMPC}.*
+ echo "#include <stddef.h>" > ${LSOF_TMPC}.c
+ echo "#include <sys/user.h>" >> ${LSOF_TMPC}.c
+ echo "main(){exit((offsetof(struct user, U_irss) & 0x7) ? 1 : 0);}" >>${LSOF_TMPC}.c
+ echo "Testing user.h with $LSOF_CC"
+ $LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x
+ if ! ${LSOF_TMPC}.x # {
+ then
+ LSOF_TMP1=1
+ else
+ LSOF_TMP1=0
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ else
+ if test "$AIX_USHACK" = "Y" -o "$AIX_USHACK" = "y" # {
+ then
+ LSOF_TMP1=1
+ else
+ LSOF_TMP1=0
+ fi # }
+ fi # }
+ if test ${LSOF_TMP1} -eq 1 # {
+ then
+ echo "Applying gcc AIX 4.1+ user struct alignment hack"
+ rm -rf ./dialects/aix/aix$LSOF_VERS
+ mkdir ./dialects/aix/aix$LSOF_VERS
+ mkdir ./dialects/aix/aix${LSOF_VERS}/sys
+ sed 's/U_irss\[/dummy_for_alignment, U_irss\[/' < ${LSOF_INCLUDE}/sys/user.h > ./dialects/aix/aix${LSOF_VERS}/sys/user.h
+ LSOF_CFGF="$LSOF_CFGF -U_LONG_LONG -I`pwd`/dialects/aix/aix$LSOF_VERS"
+ fi # }
+ fi # }
+ else
+
+ # Get xlc version number
+
+ rm -f ${LSOF_TMPC}.*
+ echo "main(){}" > ${LSOF_TMPC}.c
+ echo "Getting version number of ${LSOF_CC}."
+ $LSOF_CC -c ${LSOF_TMPC}.c -I${LSOF_INCLUDE} -o ${LSOF_TMPC}.o -qlist > /dev/null 2>&1
+ LSOF_CCV=`head -1 ${LSOF_TMPC}.lst | sed 's/\(.*\) ---.*/\1/'`
+ rm ${LSOF_TMPC}.*
+ echo "The version is \"${LSOF_CCV}\"."
+ echo $LSOF_CCV | grep "Version [0-9]" > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP=`echo $LSOF_CCV | sed 's/.*Version \([0-9]*\).*/\1/'`
+ if test "X$LSOF_TMP" != "X" -a $LSOF_TMP -ge 4 # {
+ then
+ if test $LSOF_TMP -ge 6 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -qmaxmem=-1"
+ else
+ LSOF_CFGF="$LSOF_CFGF -qmaxmem=16384"
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ if test $LSOF_VERS -ge 5300 # {
+ then
+ LSOF_UNSUP=""
+ fi # }
+ ;;
+
+# Configure for Apple Darwin.
+
+ darwin)
+ if test "X$LSOF_CC" = "X" # {
+ then
+ LSOF_CC=cc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ fi # }
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the Darwin / Mac OS X version isn't pre-defined, determine it.
+
+ case $LSOF_VSTR in # {
+ 1.2*)
+ LSOF_VERS=120
+ ;;
+ 1.3*)
+ LSOF_VERS=130
+ ;;
+ 1.4*)
+ LSOF_VERS=140
+ ;;
+ 5.[012]*)
+ LSOF_VERS=500
+ ;;
+ 5.[3-9]*)
+ LSOF_VERS=530
+ ;;
+ 6.*)
+ LSOF_VERS=600
+ ;;
+ 7.*) # Mac OS X 10.3 (Panther)
+ LSOF_VERS=700
+ ;;
+ 8.*) # Mac OS X 10.4 (Tiger)
+ LSOF_VERS=800
+ ;;
+ 9.*) # Mac OS X 10.5 (Leopard)
+ LSOF_VERS=900
+ ;;
+ 10.*) # a coming version of Mac OS?
+ LSOF_VERS=1000
+ ;;
+ *)
+ echo Unknown Darwin release: `uname -r`
+ echo Assuming Darwin 9.0
+ LSOF_VERS=900
+ ;;
+ esac # }
+ fi # }
+
+ # Do Darwin version-specific stuff.
+
+ case $LSOF_VERS in # {
+ 120|130)
+ LSOF_TMP1="hfs/hfs.h hfs/hfs_macos_defs.h miscfs/devfs/devfsdefs.h miscfs/devfs/devfs_proto.h miscfs/fdesc/fdesc.h"
+ ;;
+ 140|500)
+ LSOF_TMP1="hfs/hfs.h hfs/hfs_macos_defs.h hfs/rangelist.h miscfs/devfs/devfsdefs.h miscfs/devfs/devfs_proto.h miscfs/fdesc/fdesc.h"
+ ;;
+ 530)
+ LSOF_TMP1="hfs/hfs.h hfs/hfs_macos_defs.h hfs/rangelist.h miscfs/devfs/devfsdefs.h miscfs/devfs/devfs_proto.h miscfs/fdesc/fdesc.h net/ndrv.h net/ndrv_var.h"
+ ;;
+ 600)
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ LSOF_TMP1="hfs/hfs.h hfs/hfs_catalog.h hfs/hfs_cnode.h hfs/hfs_macos_defs.h hfs/rangelist.h miscfs/devfs/devfsdefs.h miscfs/devfs/devfs_proto.h miscfs/fdesc/fdesc.h net/ndrv_var.h net/raw_cb.h netinet/ip_var.h netinet/tcp_var.h"
+ ;;
+ 700)
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ LSOF_TMP1="hfs/hfs.h hfs/hfs_catalog.h hfs/hfs_cnode.h hfs/hfs_macos_defs.h hfs/rangelist.h miscfs/devfs/devfsdefs.h miscfs/devfs/devfs_proto.h miscfs/fdesc/fdesc.h net/ndrv_var.h net/raw_cb.h netinet/ip_var.h netinet/tcp_var.h sys/eventvar.h"
+ ;;
+ 800)
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ LSOF_TMP1="net/ndrv_var.h net/raw_cb.h netinet/ip_var.h netinet/tcp_var.h sys/eventvar.h sys/file_internal.h sys/mount_internal.h sys/proc_internal.h sys/vnode_internal.h"
+ ;;
+ 900)
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ LSOF_TMP1=""
+ LSOF_UNSUP=""
+ LSOF_TSTBIGF=" " # enable LTbigf test
+ ;;
+ 1000)
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ LSOF_TMP1=""
+ ;;
+ *)
+ echo "Unsupported Darwin version: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ LSOF_TMP2=""
+ LSOF_TMP3=""
+ LSOF_TMP4=""
+ LSOF_CFGF="$LSOF_CFGF -mdynamic-no-pic"
+ LSOF_CFGL="-lcurses"
+
+ if test "X$DARWIN_XNUDIR" != "X" # {
+ then
+ LSOF_TMP2="${DARWIN_XNUDIR}/bsd"
+ LSOF_TMP3="${DARWIN_XNUDIR}/osfmk"
+ LSOF_TMP4=""
+ else
+ LSOF_TMP2="${DARWIN_XNU_HEADERS}/System/Library/Frameworks/Kernel.framework/Versions/A/PrivateHeaders"
+ LSOF_TMP3="${DARWIN_XNU_HEADERS}/System/Library/Frameworks/System.framework/Versions/B/PrivateHeaders"
+ LSOF_TMP4=""
+ if test "X$DARWIN_XNU_HEADERS" != "X" # {
+ then
+ LSOF_TMP4="${DARWIN_XNU_HEADERS}/usr/include"
+ fi # }
+ fi # }
+
+ # Test Darwin base.
+
+ if test "X$DARWIN_BASE" = "X" -o "X$DARWIN_BASE" = "Xlibproc" # {
+ then
+ LSOF_TMP5=""
+ if test $LSOF_VERS -ge 800 -o "X$DARWIN_BASE" = "Xlibproc" # {
+ then
+ if test -r ${LSOF_INCLUDE}/libproc.h # {
+ then
+ DARWIN_BASE="libproc"
+ else
+ if test -r ${LSOF_INCLUDE}/../local/include/libproc.h # {
+ then
+ DARWIN_BASE="libproc"
+ LSOF_TMP5="-I${LSOF_INCLUDE}/../local/include"
+ else
+ echo "FATAL: can't find libproc.h"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+ else
+
+ # The default Darwin base is /dev/kmem.
+
+ DARWIN_BASE="/dev/kmem"
+ fi # }
+ fi # }
+ if test "X$DARWIN_BASE" = "Xlibproc" # {
+ then
+
+ # Configure for libproc-based Darwin lsof.
+
+ echo "Configuring libproc-based Darwin lsof"
+ LSOF_CINFO="libproc-based"
+ LSOF_DIALECT_DIR=darwin/libproc
+ if test $LSOF_VERS -lt 1000 # {
+ then
+ LSOF_CFGL="$LSOF_CFGL -lproc"
+ fi # }
+ LSOF_TSTKMEM=0
+ LSOF_DINC="$LSOF_DINC $LSOF_TMP5"
+ if test ! -r ${LSOF_INCLUDE}/sys/proc_info.h # {
+ then
+ if test "X$LSOF_TMP5" = "X" -o ! -r ${LSOF_TMP5}/sys/proc_info.h # {
+ then
+ echo "FATAL: can't find sys/proc_info.h"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+
+ # Add header file paths for libproc-based Darwin lsof.
+
+ for i in $LSOF_TMP3 $LSOF_TMP4 $LSOF_INCLUDE # {
+ do
+ if test -d $i -a "X$i" != "X/usr/include" # {
+ then
+ LSOF_DINC="$LSOF_DINC -I${i}"
+ fi # }
+ done # }
+
+ # Do other libproc-based Darwin lsof setups.
+
+ if test -r ${LSOF_INCLUDE}/utmpx.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASUTMPX"
+ fi # }
+ else
+ if test "X$DARWIN_BASE" != "X/dev/kmem" # {
+ then
+ echo "Darwin base unrecognized: $DARWIN_BASE"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+
+ # Configure for /dev/kmem-based Darwin lsof.
+
+ echo "Configuring /dev/kmem-based Darwin lsof"
+ LSOF_CINFO="/dev/kmem-based"
+ LSOF_DIALECT_DIR=darwin/kmem
+
+ # Make sure needed /dev/kmem-base XNU Darwin kernel header files are
+ # present.
+
+ LSOF_TMP5=""
+ for i in $LSOF_TMP1 # {
+ do
+ LSOF_TMP6=0
+ for j in $LSOF_TMP2 $LSOF_TMP3 $LSOF_TMP4 $LSOF_INCLUDE # {
+ do
+ if test "X${j}" != "X" -a -r ${j}/${i} # {
+ then
+ LSOF_TMP6=1
+ break
+ fi # }
+ done # }
+ if test $LSOF_TMP6 -ne 1 # {
+ then
+ if test "X$LSOF_TMP5" = "X" # {
+ then
+ LSOF_TMP5=$i
+ else
+ LSOF_TMP5="$LSOF_TMP5 $i"
+ fi # }
+ fi # }
+ done # }
+ if test "X$LSOF_TMP5" != "X" # {
+ then
+
+ # If any Darwin XNU kernel header files are missing, call the
+ # get-hdr-loc.sh script to find the path.
+
+ LSOF_TMP6=`pwd`/dialects/darwin/get-hdr-loc.sh
+ if test ! -x $LSOF_TMP6 # {
+ then
+ echo "FATAL: can't execute: $LSOF_TMP6"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ DARWIN_XNUDIR=`$LSOF_TMP6 $LSOF_TMP5`
+ if test $? -ne 0 # {
+ then
+ echo "FATAL: $LSOF_TMP6 returns: $DARWIN_XNUDIR"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ LSOF_TMP2="${DARWIN_XNUDIR}/bsd"
+ LSOF_TMP3="${DARWIN_XNUDIR}/osfmk"
+ LSOF_TMP4=""
+ fi # }
+
+ # Add header file paths for /dev/kmem-based Darwin lsof.
+
+ for i in $LSOF_TMP2 $LSOF_TMP3 $LSOF_TMP4 $LSOF_INCLUDE # {
+ do
+ if test -d $i -a "X$i" != "X/usr/include" # {
+ then
+ LSOF_DINC="$LSOF_DINC -I${i}"
+ fi # }
+ done # }
+
+ # Make conditional feature definitions for /dev/kmem-based Darwin lsof.
+
+ for i in $LSOF_TMP2 $LSOF_TMP3 $LSOF_TMP4 $LSOF_INCLUDE # {
+ do
+ if test "X${i}" != "X" -a -r ${i}/sys/namei.h # {
+ then
+ grep -q nc_vpid ${i}/sys/namei.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNCVPID"
+ fi # }
+ break
+ fi # }
+ done # }
+ for i in $LSOF_TMP2 $LSOF_TMP3 $LSOF_TMP4 $LSOF_INCLUDE # {
+ do
+ if test "X${i}" != "X" # {
+ then
+ if test $LSOF_VERS -ge 800 # {
+ then
+ if test -r ${i}/sys/file_internal.h # {
+ then
+ grep -q DTYPE_KQUEUE ${i}/sys/file_internal.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASKQUEUE"
+ fi # }
+ break
+ fi # }
+ else
+ if test $LSOF_VERS -ge 700 # {
+ then
+ if test -r ${i}/sys/file.h # {
+ then
+ grep -q DTYPE_KQUEUE ${i}/sys/file.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASKQUEUE"
+ fi # }
+ fi # }
+ break
+ fi # }
+ fi # }
+ fi # }
+ done # }
+ LSOF_CFGF="$LSOF_CFGF -DHAS9660FS"
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -DDARWINV=$LSOF_VERS"
+ LSOF_CFLAGS_OVERRIDE=1
+ ;;
+
+# Configure for DEC OSF/1, Digital UNIX, or Tru64 UNIX.
+
+ digital_unix|du|decosf|tru64)
+ LSOF_TGT="du"
+ LSOF_TSTBIGF=" "
+ LSOF_TSTK64=1
+ if test "X$LSOF_DINC" = "X" # {
+ then
+ LSOF_DINC="-I/usr/include"
+ fi # }
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the DEC OSF/1, Digital UNIX, or Tru64 UNIX version isn't
+ # predefined, determine it.
+
+ case $LSOF_VSTR in # {
+ V2.0)
+ LSOF_VERS=20000
+ ;;
+ V3.0)
+ LSOF_VERS=30000
+ ;;
+ V3.2)
+ LSOF_VERS=30200
+ ;;
+ ?4.0)
+ LSOF_TSTXO="../lib/snpf.o"
+ LSOF_VERS=40000
+ ;;
+ ?5.0)
+ LSOF_VERS=50000
+ ;;
+ ?5.1)
+ LSOF_VERS=50100
+ ;;
+ *)
+ echo "WARNING: unknown version; assuming version is 2.0"
+ LSOF_VERS=20000
+ ;;
+ esac # }
+ fi # }
+
+ # Do DEC OSF/1, Digital UNIX, or Tru64 UNIX version specific stuff.
+
+ case $LSOF_VERS in # {
+ 20000)
+ LSOF_CFGF="-Olimit 1024"
+ LSOF_TMP1="/sys"
+ ;;
+ 30000)
+ LSOF_CFGF="-Olimit 1024"
+ LSOF_TMP1="/sys"
+ LSOF_TMP2=-DUSELOCALREADDIR
+ ;;
+ 30200)
+ LSOF_CFGF="-Olimit 1024"
+ LSOF_TMP1="/sys"
+ LSOF_TMP2=-DUSELOCALREADDIR
+ ;;
+ 40000)
+ LSOF_TMP1="/usr/sys"
+ ;;
+ 50000|50100)
+ LSOF_CFGF="-DUSE_STAT"
+ LSOF_TMP1="/usr/sys"
+ ;;
+ *)
+ echo "WARNING: unknown version: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ if test "X$DU_SYSDIR" = "X" # {
+ then
+ DU_SYSDIR=$LSOF_TMP1
+ fi # }
+ LSOF_HOST=`uname -n`
+ if test "X$DU_CDIR" = "X" # {
+ then
+ LSOF_CDIR=`expr $LSOF_HOST : '\([^\.]*\)\..*$'`
+ if test "X$LSOF_CDIR" = "X" # {
+ then
+ LSOF_CDIR=$LSOF_HOST
+ fi # }
+ LSOF_CDIR=`echo $LSOF_CDIR | tr a-z A-Z`
+ else
+ LSOF_CDIR=$DU_CDIR
+ fi # }
+ LSOF_LOOP=1
+ while test $LSOF_LOOP = 1 # {
+ do
+ if test -d ${DU_SYSDIR}/$LSOF_CDIR # {
+ then
+ echo "Using header files in ${DU_SYSDIR}/$LSOF_CDIR"
+ LSOF_LOOP=0
+ else
+ cat << .CAT_MARK
+
+Please enter the name of the subdirectory in $DU_SYSDIR that contains the
+configuration files for this host. Usually its name would be $LSOF_CDIR, but
+that subdirectory doesn't seem to exist. The lsof compilation needs header
+files specific to this machine's configuration found in that directory.
+
+If you can't specify the appropriate configuration subdirectory, quit this
+Configure step now and generate a proper configuration subdirectory with the
+kernel generation process.
+
+.CAT_MARK
+
+ echo "$DU_SYSDIR contains:"
+ echo ""
+ ls -CF $DU_SYSDIR
+ echo ""
+ echo -n "Configuration subdirectory name? "
+ read LSOF_CDIR LSOF_EXCESS
+ if test "X$LSOF_CDIR" = "X" -o ! -d ${DU_SYSDIR}/$LSOF_CDIR # {
+ then
+ echo ""
+ echo Cannot access directory ${DU_SYSDIR}/$LSOF_CDIR.
+ fi # }
+ fi # }
+ done # }
+
+ # Determine the ADVFS file system version.
+
+ if test "X$DU_ADVFSV" = "X" # {
+ then
+ echo "Determining the ADVFS version -- this will take a while."
+ LSOF_ADVFSV=`/usr/sbin/setld -i | grep "^OSFADVFSBIN[0-9]" | sed 's/\([^ ]*\).*/\1/' | sort -u | tail -1 | sed 's/OSFADVFSBIN//'`
+ else
+ LSOF_ADVFSV=$DU_ADVFSV
+ fi # }
+ case $LSOF_ADVFSV in # {
+ 1*)
+ LSOF_ADVFSV=100
+ echo "The ADVFS version is 1."
+ ;;
+ 2*)
+ LSOF_ADVFSV=200
+ echo "The ADVFS version is 2."
+ ;;
+ 3*)
+ LSOF_ADVFSV=300
+ echo "The ADVFS version is 3."
+ ;;
+ 4*)
+ LSOF_ADVFSV=400
+ echo "The ADVFS version is 4."
+ ;;
+ 5*)
+ LSOF_ADVFSV=500
+ echo "The ADVFS version is 5."
+ ;;
+ *)
+ echo "The ADVFS version is unknown; it will be assumed to be 1."
+ LSOF_ADVFSV=100
+ ;;
+ esac # }
+ LSOF_CFGF="$LSOF_CFGF -DDUV=$LSOF_VERS -DADVFSV=$LSOF_ADVFSV $LSOF_TMP2"
+ if test "X$DU_SYSINC" = "X" # {
+ then
+ DU_SYSINC="/usr/sys/include"
+ fi # }
+ LSOF_DINC="$LSOF_DINC -I${DU_SYSDIR}/$LSOF_CDIR -I$DU_SYSINC"
+ LSOF_CFGL="-lmld"
+ if test "X${DU_SHLIB}" = "X" # {
+ then
+ DU_SHLIB=/usr/shlib
+ fi # }
+ if test -r ${DU_SHLIB}/libmsfs.so # {
+ then
+ nm ${DU_SHLIB}/libmsfs.so | grep tag_to_path > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASTAGTOPATH"
+ LSOF_CFGL="$LSOF_CFGL -lmsfs"
+ fi # }
+ fi # }
+ grep "^struct spec_node {" ${DU_SYSDIR}/include/sys/specdev.h > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASSPECNODE"
+ fi # }
+ if test $LSOF_VERS -ge 50000 # {
+ then
+
+ # Make du5_sys_malloc.h for DU 5.0 and above. Enable strict ANSI checking
+ # on 5.0 and 5.1A, but not 5.1B. Enable IPv6 handling.
+
+ LSOF_TMP1="-std1"
+ if test $LSOF_VERS -ge 50100 # {
+ then
+ LSOF_TMP1="-std"
+ if test -x /usr/sbin/sizer # {
+ then
+ /usr/sbin/sizer -v | grep -q 5.1A
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1="-std1"
+ fi # }
+ fi # }
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF $LSOF_TMP1"
+ LSOF_TMP1=${LSOF_INCLUDE}/sys/malloc.h
+ if test -r $LSOF_TMP1 # {
+ then
+ LSOF_TMP2=dialects/du/du5_sys_malloc.h
+ rm -f $LSOF_TMP2
+ echo "#if !defined(MANUFACTURED_DU5_SYS_MALLOC_H)" > $LSOF_TMP2
+ echo "/* By lsof Configure:" `date` " */" >> $LSOF_TMP2
+ echo "#define MANUFACTURED_DU5_SYS_MALLOC_H" >> $LSOF_TMP2
+ grep "^#define[ ]MALLOC_NUM_BUCKETS" $LSOF_TMP1 >> $LSOF_TMP2
+ echo "struct percpukmembuckets {" >> $LSOF_TMP2
+ sed '1,/^struct percpukmembuckets/d' $LSOF_TMP1 | sed -n '1,/^};/p' >> $LSOF_TMP2
+ echo "#endif" >> $LSOF_TMP2
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/du"
+ fi # }
+
+ # Enable IPv6 for Tru64 UNIX 5.0 and above.
+
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/namei.h
+ then
+ grep -q nc_vpid ${LSOF_INCLUDE}/sys/namei.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNCVPID"
+ fi # }
+ fi # }
+ LSOF_DIALECT_DIR=du
+ ;;
+
+# Configure for FreeBSD.
+
+ freebsd)
+ LSOF_FBSD_ZFS=0
+ if test "X$LSOF_CC" = "X" # {
+ then
+ LSOF_CC=cc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ fi # }
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the FreeBSD version isn't pre-defined, determine it.
+
+ case $LSOF_VSTR in # {
+ 1.*)
+ LSOF_VERS=1000
+ ;;
+ 2.0-*)
+ LSOF_VERS=2000
+ ;;
+ 2.0.5-*)
+ LSOF_VERS=2005
+ ;;
+ 2.1*)
+ LSOF_VERS=2010
+ ;;
+ 2.2*)
+ LSOF_VERS=2020
+ ;;
+ 3.0*)
+ LSOF_VERS=3000
+ ;;
+ 3.1*)
+ LSOF_VERS=3010
+ ;;
+ 3.2*)
+ LSOF_VERS=3020
+ ;;
+ 3.3*)
+ LSOF_VERS=3030
+ ;;
+ 3.4*)
+ LSOF_VERS=3040
+ ;;
+ 3.5*)
+ LSOF_VERS=3050
+ ;;
+ 3*)
+ LSOF_VERS=3050
+ echo "!!!WARNING!!! Unsupported FreeBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for FreeBSD 3.5"
+ ;;
+ 4.0*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4000
+ ;;
+ 4.1-*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4010
+ ;;
+ 4.2*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4020
+ ;;
+ 4.3*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4030
+ ;;
+ 4.4*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4040
+ ;;
+ 4.5*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4050
+ ;;
+ 4.6*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4060
+ ;;
+ 4.7*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4070
+ ;;
+ 4.8*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4080
+ ;;
+ 4.9*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4090
+ ;;
+ 4.10*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4100
+ ;;
+ 4.11*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=4110
+ ;;
+ 4*)
+ LSOF_VERS=4100
+ echo "!!!WARNING!!! Unsupported FreeBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for FreeBSD 4.10"
+ ;;
+ 5.0*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=5000
+ ;;
+ 5.1*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=5010
+ ;;
+ 5.2*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=5020
+ ;;
+ 5.3*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=5030
+ ;;
+ 5.4*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=5040
+ ;;
+ 5.5*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=5050
+ ;;
+ 5*)
+ LSOF_VERS=5050
+ echo "!!!WARNING!!! Unsupported FreeBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for FreeBSD 5.5"
+ ;;
+ 6.0*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=6000
+ ;;
+ 6.1*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=6010
+ ;;
+ 6.2*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=6020
+ ;;
+ 6.3*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=6030
+ ;;
+ 6.4*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=6040
+ ;;
+ 6*)
+ LSOF_VERS=6000
+ echo "!!!WARNING!!! Unsupported FreeBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for FreeBSD 6.0"
+ ;;
+ 7.0*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=7000
+ ;;
+ 7.1*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=7010
+ ;;
+ 7.2*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=7020
+ ;;
+ 7*)
+ LSOF_VERS=7000
+ echo "!!!WARNING!!! Unsupported FreeBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for FreeBSD 7.0"
+ ;;
+ 8*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=8000
+ ;;
+ *)
+ echo Unknown FreeBSD release: `uname -r`
+ echo Assuming FreeBSD 2.x
+ LSOF_CFGL="-lkvm"
+ LSOF_VERS=2000
+ LSOF_N_UNIXV=`/usr/sbin/sysctl -n kern.bootfile`
+ if test "X$LSOF_N_UNIXV" = "X" # {
+ then
+ LSOF_N_UNIXV="/kernel"
+ fi # }
+ ;;
+ esac # }
+ fi # }
+
+ # Clear LSOF_UNSUP message for supported versions of FreeBSD.
+
+ case $LSOF_VERS in # {
+ 4090|7000|7010|7020|8000)
+ LSOF_UNSUP=""
+ ;;
+ esac # }
+
+ # Get system CFLAGS from FREEBSD_MAKE_CONF (default=/etc/make.conf).
+
+ if test "X$FREEBSD_MAKE_CONF" = "X" # {
+ then
+ FREEBSD_MAKE_CONF="/etc/make.conf"
+ fi # }
+ if test -r $FREEBSD_MAKE_CONF # {
+ then
+ LSOF_CFGF=`echo "all:\n.include <bsd.prog.mk>" | make -f- -VCFLAGS`
+ LSOF_TMP=1
+ while test $LSOF_TMP -eq 1 # {
+ do
+ echo $LSOF_CFGF | grep -q -e '-O'
+ if test $? -eq 0 # {
+ then
+ if test "X$LSOF_DEBUG" = "X"
+ then # {
+ LSOF_DEBUG=`echo $LSOF_CFGF | sed 's/.*\(-O[^ $]*\).*/\1/'`
+ fi # }
+ LSOF_CFGF=`echo $LSOF_CFGF | sed 's/\(.*\)-O[^ $]*\(.*\)/\1 \2/' | sed 's/^ *//g' | sed 's/ */ /g' | sed 's/ *$//'`
+ else
+ LSOF_TMP=0
+ fi # }
+ LSOF_FBSD_ZFS_CFGF="$LSOF_CFGF"
+ done # }
+ fi # }
+
+ # Determine path to FreeBSD sources.
+
+ LSOF_DINC_ADD=0
+ if test "X$FREEBSD_SYS" = "X" # {
+ then
+ if test -d /usr/src/sys # {
+ then
+ FREEBSD_SYS=/usr/src/sys
+ else
+ if test -d /sys # {
+ then
+ FREEBSD_SYS="/sys"
+ else
+ echo "!!!WARNING!!! No kernel sources in /usr/src/sys or /sys"
+ fi # }
+ fi # }
+ fi # }
+
+ # Do FreeBSD version-specific stuff.
+
+ case $LSOF_VERS in # {
+ 1000)
+ LSOF_CFGF="$LSOF_CFGF -DHASPROCFS"
+ LSOF_CFGL="-lutil"
+ LSOF_DINC="$LSOF_DINC -I`pwd`/dialects/freebsd/include"
+ if test "X$FREEBSD_KERNEL" = "X" # {
+ then
+ LSOF_N_UNIXV="/386bsd"
+ else
+ LSOF_N_UNIXV=$FREEBSD_KERNEL
+ fi # }
+ ;;
+ 2000|2005|2010)
+ LSOF_CFGL="-lkvm"
+ ;;
+ 2020)
+ LSOF_CFGL="-lkvm"
+ if test -r ${LSOF_INCLUDE}/vm/lock.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVMLOCKH"
+ fi # }
+ ;;
+ 3000|3010|3020|3030|3040|3050)
+ LSOF_CFGL="-lkvm"
+ if test -r ${LSOF_INCLUDE}/nfs/rpcv2.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASRPCV2H"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/vm/lock.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVMLOCKH"
+ fi # }
+ ;;
+ 4000|4010|4020|4030|4040|4050|4060|4070|4080|4090|4100|4110|5000|5010|5020|5030|5040|5050|6000|6010|6020|6030|6040|7000|7010|7020|8000)
+ if test -r ${LSOF_INCLUDE}/sys/namei.h # {
+ then
+ grep -q "^struct[ ]*namecache[ ]*{" ${LSOF_INCLUDE}/sys/namei.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNAMECACHE"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/ufs/ufs/inode.h # {
+ then
+ grep -q i_effnlink ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASEFFNLINK=i_effnlink"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/file.h # {
+ then
+ grep -q f_vnode ${LSOF_INCLUDE}/sys/file.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASF_VNODE"
+ fi # }
+ fi # }
+ LSOF_CFGL="-lkvm"
+ if test $LSOF_VERS -ge 5000 # {
+ then
+
+ # Do FreeBSD 5 and higher version-specific stuff.
+
+ if test -r ${LSOF_INCLUDE}/sys/vnode.h # {
+ then
+ grep VT_FDESC ${LSOF_INCLUDE}/sys/vnode.h > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ if test ! -r ${LSOF_INCLUDE}/fs/devfs/devfs.h # {
+ then
+ if test -r ${FREEBSD_SYS}/fs/devfs/devfs.h # {
+ then
+ LSOF_DINC_ADD=1
+ else
+ echo "!!!FATAL: lsof cannot locate the devfs.h header file"
+ echo " in ${LSOF_INCLUDE}/fs/devfs/devfs.h or"
+ echo " ${FREEBSD_SYS}/fs/devfs/devfs.h. Consult"
+ echo " 00FAQ for an explanation."
+ exit 1
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+
+ # Do FreeBSD 5.2 and higher version-specific stuff.
+
+ if test $LSOF_VERS -ge 5020 # {
+ then
+
+ # Determine the status of the cpumask_t typedef.
+
+ rm -f ${LSOF_TMPC}.*
+ cat > ${LSOF_TMPC}.c << .LSOF_END_HERE_DOC3
+#undef _KERNEL
+#include <sys/types.h>
+main() {
+cpumask_t c;
+}
+.LSOF_END_HERE_DOC3
+ $LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ LSOF_TMP1=$?
+ rm -f ${LSOF_TMPC}.*
+ if test $LSOF_TMP1 -ne 0 # {
+ then
+
+ # The cpumask_t typedef is unknown when _KERNEL is not defined.
+
+ if test -r ${LSOF_INCLUDE}/sys/types.h \
+ -a -r ${LSOF_INCLUDE}/machine/_types.h # {
+ then
+ grep -q cpumask_t ${LSOF_INCLUDE}/sys/types.h
+ if test $? -eq 0 # {
+ then
+ grep -q __cpumask_t ${LSOF_INCLUDE}/machine/_types.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASCPUMASK_T"
+ else
+ $LSOF_CC -E ${LSOF_INCLUDE}/machine/_types.h 2>/dev/null | grep -q __cpumask_t
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASCPUMASK_T"
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/socketvar.h # {
+ then
+ grep -q SBS_CANT ${LSOF_INCLUDE}/sys/socketvar.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASSBSTATE"
+ fi # }
+ fi # }
+ fi # }
+ if test $LSOF_VERS -ge 5030 # {
+ then
+
+ # Do FreeBSD 5.3 and higher version-specific stuff.
+
+ if test -r ${LSOF_INCLUDE}/sys/vnode.h # {
+ then
+ grep -q "defined(_KVM_VNODE)" ${LSOF_INCLUDE}/sys/vnode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_KVM_VNODE"
+ fi #}
+ fi # }
+ fi # }
+ if test $LSOF_VERS -ge 6000 # {
+ then
+
+ # Do FreeBSD 6.0 and higher version-specific stuff.
+
+ if test -r ${LSOF_INCLUDE}/ufs/ufs/inode.h # {
+ then
+ grep -q i_din2 ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_UFS1_2"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/conf.h # {
+ then
+ grep -q "^#define minor(" ${LSOF_INCLUDE}/sys/conf.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_CONF_MINOR"
+ rm -f fbsd_minor.h
+ if test -r ${LSOF_INCLUDE}/sys/types.h # {
+ then
+ LSOF_TMP1=`grep "^#define[ ]minor(" ${LSOF_INCLUDE}/sys/types.h`
+ if test "X$LSOF_TMP1" != "X" # {
+ then
+ echo "Creating fbsd_minor.h"
+ cat > fbsd_minor.h << FBSD_MINOR1
+/*
+ * fbsd_minor.h -- created by lsof Configure script on
+FBSD_MINOR1
+ echo $EO " * $EC" >> ./fbsd_minor.h
+ date >> ./fbsd_minor.h
+ cat >> ./fbsd_minor.h << FBSD_MINOR2
+ */
+
+#if !defined(FBSD_MINOR_H)
+#define FBSD_MINOR_H
+
+FBSD_MINOR2
+ echo $EO "${LSOF_TMP1}${EC}" >> fbsd_minor.h
+ cat >> ./fbsd_minor.h << FBSD_MINOR3
+
+#endif /* defined(FBSD_MINOR_H) */
+FBSD_MINOR3
+ fi # }
+ fi # }
+ else
+ if test -r ${FREEBSD_SYS}/fs/devfs/devfs_int.h # {
+ then
+ grep -q cdev2priv ${FREEBSD_SYS}/fs/devfs/devfs_int.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_CDEV2PRIV"
+ fi # }
+ fi # }
+ fi # }
+ grep -q "si_udev;" ${LSOF_INCLUDE}/sys/conf.h
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_NO_SI_UDEV"
+ fi # }
+ grep -q si_priv ${LSOF_INCLUDE}/sys/conf.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_SI_PRIV"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/sx.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_SYS_SX_H"
+ fi # }
+
+ # Do ZFS test. Try for the newer OpenSolaris files first -- i.e.,
+ # the ones in ${FREEBSD_SYS}/cddl/contrib/opensolaris. If that fails,
+ # try for the older ones in ${FREEBSD}/contrib/opensolaris.
+
+ LSOF_FBSD_ZFS_SYS=${FREEBSD_SYS}/cddl
+ if test ! -r ${LSOF_FBSD_ZFS_SYS}/contrib/opensolaris/uts/common/fs/zfs/sys/zfs_znode.h # {
+ then
+ LSOF_FBSD_ZFS_SYS=${FREEBSD_SYS}
+ if test ! -r ${LSOF_FBSD_ZFS_SYS}/contrib/opensolaris/uts/common/fs/zfs/sys/zfs_znode.h # {
+ then
+ LSOF_FBSD_ZFS_SYS=""
+ fi # }
+ fi # }
+ if test "X$LSOF_FBSD_ZFS_SYS" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_ZFS"
+ LSOF_FBSD_ZFS=1
+ LSOF_FBSD_ZFS_CFGF="$LSOF_FBSD_ZFS_CFGF -DFREEBSDV=$LSOF_VERS"
+ LSOF_FBSD_ZFS_CFGF="$LSOF_FBSD_ZFS_CFGF -DHAS_ZFS"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/vnode.h # {
+ then
+
+ # See if the vnode contains the byte level lock pointer.
+
+ grep -q v_lockf ${LSOF_INCLUDE}/sys/vnode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_V_LOCKF"
+ if test $LSOF_FBSD_ZFS -eq 1 # {
+ then
+ LSOF_FBSD_ZFS_CFGF="$LSOF_FBSD_ZFS_CFGF -DHAS_V_LOCKF"
+ fi # }
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/lockf.h # {
+ then
+
+ # Determine the type of locking structure to which the inode or
+ # vnode points.
+
+ grep -q "^struct lockf_entry" ${LSOF_INCLUDE}/sys/lockf.h
+ if test $? -eq 0 # {
+ then
+
+ # Build the ./lockf_owner.h header file.
+
+ LSOF_TMP1=""
+ LSOF_TMP2=0
+ echo "Creating ./lockf_owner.h from ${FREEBSD_SYS}/kern/kern_lockf.c"
+ rm -f ./lockf_owner.h
+ if test -r ${FREEBSD_SYS}/kern/kern_lockf.c # {
+ then
+ LSOF_TMP1=`grep -n "^struct lock_owner" ${FREEBSD_SYS}/kern/kern_lockf.c | sed 's/\([0-9]*\):.*$/\1/'`
+ if test "X$LSOF_TMP1" != "X" # {
+ then
+ LSOF_TMP2=0
+ for i in `grep -n "};" ${FREEBSD_SYS}/kern/kern_lockf.c | sed 's/\([0-9]*\):.*$/\1/'` # {
+ do
+ if test $LSOF_TMP2 -eq 0 -a $i -gt $LSOF_TMP1 # {
+ then
+ LSOF_TMP2=$i
+ fi # }
+ done # }
+ if test $LSOF_TMP2 -eq 0 # {
+ then
+ LSOF_TMP1=""
+ else
+ cat > ./lockf_owner.h << LOCKF_OWNER1
+/*
+ * lockf_owner.h -- created by lsof Configure script on
+LOCKF_OWNER1
+ echo $EO " * $EC" >> ./lockf_owner.h
+ date >> ./lockf_owner.h
+ cat >> ./lockf_owner.h << LOCKF_OWNER2
+ */
+
+#if !defined(LOCKF_OWNER_H)
+#define LOCKF_OWNER_H
+
+LOCKF_OWNER2
+ ed -s ${FREEBSD_SYS}/kern/kern_lockf.c >> ./lockf_owner.h << LOCKF_OWNER3
+${LSOF_TMP1},${LSOF_TMP2}p
+LOCKF_OWNER3
+ if test $? -ne 0 # {
+ then
+ LSOF_TMP1=""
+ else
+ cat >> ./lockf_owner.h << LOCKF_OWNER4
+
+#endif /* defined(LOCKF_OWNER_H) */
+LOCKF_OWNER4
+ fi # }
+ fi # }
+ fi # }
+ else
+ echo "FATAL ERROR: can't read ${FREEBSD_SYS}/kern/kern_lockf.c"
+ fi # }
+ if test "X$LSOF_TMP1" != "X" -a "X$LSOF_TMP2" != "X0" # {
+ then
+ echo "./lockf_owner.h creation succeeded."
+ LSOF_CFGF="$LSOF_CFGF -DHAS_LOCKF_ENTRY"
+ else
+ echo "FATAL ERROR: ./lockf_owner.h creation failed (see 00FAQ)"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+
+ # Test for in6p_.port in inpcb structure.
+
+ if test -r ${LSOF_INCLUDE}/netinet/in_pcb.h # {
+ then
+ grep -q 'in6p_.port' ${LSOF_INCLUDE}/netinet/in_pcb.h
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_NO_6PORT"
+ fi # }
+ fi # }
+
+ # Test for in6p_ppcb in inpcb structure.
+
+ if test -r ${LSOF_INCLUDE}/netinet/in_pcb.h # {
+ then
+ grep -q 'in6p_ppcb' ${LSOF_INCLUDE}/netinet/in_pcb.h
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_NO_6PPCB"
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ ;;
+ *)
+ echo "Unknown FreeBSD release: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ LSOF_CFGF="$LSOF_CFGF -DFREEBSDV=$LSOF_VERS"
+ if test $LSOF_VERS -lt 2000 -a "X$FREEBSD_KERNEL" = "X" # {
+ then
+ if test ! -x $LSOF_N_UNIXV # {
+ then
+ echo "Hmmm -- $LSOF_N_UNIXV doesn't appear to be your kernel file."
+ echo "Please enter the name of the file in / that contains"
+ echo "the kernel for this host. It must be a regular file,"
+ echo "not a directory, and must be executable."
+ LSOF_LOOP=1
+ while test $LSOF_LOOP = 1 # {
+ do
+ echo ""
+ echo "/ contains:"
+ echo ""
+ ls -CF /
+ echo ""
+ echo -n "Kernel file name? "
+ read LSOF_N_UNIXV LSOF_EXCESS
+ LSOF_N_UNIXV="/$LSOF_N_UNIXV"
+ if test ! -d $LSOF_N_UNIXV -a -x $LSOF_N_UNIXV # {
+ then
+ LSOF_LOOP=0
+ else
+ echo ""
+ echo $LSOF_N_UNIXV is not a regular executable file.
+ fi # }
+ done # }
+ fi # }
+ LSOF_N_UNIXV=`echo $LSOF_N_UNIXV | sed 's#^/*#/#'`
+ LSOF_CFGF="$LSOF_CFGF -DN_UNIXV=$LSOF_N_UNIXV"
+ fi # }
+ if test -r ${FREEBSD_SYS}/miscfs/fdesc/fdesc.h # {
+ then
+ LSOF_TMP1=${FREEBSD_SYS}/miscfs/fdesc/fdesc.h
+ else
+ if test $LSOF_VERS -ge 5000 -a -r ${LSOF_INCLUDE}/fs/fdescfs/fdesc.h # {
+ then
+ LSOF_TMP1=${LSOF_INCLUDE}/fs/fdescfs/fdesc.h
+ else
+ LSOF_TMP1=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP1" != "X" # {
+ then
+ grep -q Fctty $LSOF_TMP1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASFDESCFS=1"
+ else
+ LSOF_CFGF="$LSOF_CFGF -DHASFDESCFS=2"
+ fi # }
+ grep -q fd_link $LSOF_TMP1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASFDLINK"
+ fi # }
+ LSOF_DINC_ADD=1
+ fi # }
+ if test $LSOF_VERS -ge 5000 # {
+ then
+ LSOF_TMP1="fs"
+ else
+ LSOF_TMP1="miscfs"
+ fi # }
+ if test $LSOF_VERS -lt 5000 # {
+ then
+ if test -d ${FREEBSD_SYS}/${LSOF_TMP1}/procfs # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPROCFS"
+ LSOF_DINC_ADD=1
+ fi # }
+ else
+ if test -d ${FREEBSD_SYS}/${LSOF_TMP1}/pseudofs # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPSEUDOFS"
+ LSOF_DINC_ADD=1
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/${LSOF_TMP1}/nullfs/null.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNULLFS"
+ else
+ if test -r ${FREEBSD_SYS}/${LSOF_TMP1}/nullfs/null.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNULLFS"
+ LSOF_DINC_ADD=1
+ fi # }
+ fi # }
+ if test -r ${FREEBSD_SYS}/isofs/cd9660/cd9660_node.h # {
+ then
+ rm -f cd9660_node.h
+ grep -q "^#ifdef [_]*KERNEL" ${FREEBSD_SYS}/isofs/cd9660/cd9660_node.h
+ if test $? -eq 0 # {
+ then
+ ln -s ${FREEBSD_SYS}/isofs/cd9660/cd9660_node.h cd9660_node.h
+ else
+ sed -e '/^ \* Prototypes for ISOFS vnode operations/,$c\
+ \ The ISOFS prototypes were removed by Configure. */' \
+ < ${FREEBSD_SYS}/isofs/cd9660/cd9660_node.h > cd9660_node.h
+ echo "" >> cd9660_node.h
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -DHAS9660FS"
+ if test $LSOF_VERS -ge 6000 # {
+ then
+ grep -q "i_dev;" cd9660_node.h
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_NO_ISO_DEV"
+ fi # }
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/namei.h
+ then
+ grep -q nc_vpid ${LSOF_INCLUDE}/sys/namei.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNCVPID"
+ fi # }
+ fi # }
+ if test $LSOF_DINC_ADD -eq 1 # {
+ then
+ if test "X$LSOF_DINC" = "X" # {
+ then
+ LSOF_DINC="-I${FREEBSD_SYS}"
+ else
+ LSOF_DINC="$LSOF_DINC -I${LSOF_INCLUDE} -I${FREEBSD_SYS}"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/netinet/in.h # {
+ then
+ grep IPV6_INRIA_VERSION ${LSOF_INCLUDE}/netinet/in.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6 -DHASINRIAIPv6"
+ fi # }
+ fi # }
+ echo $CFGF | grep HASIPv6 > /dev/null
+ if test $? -ne 0 -a -r ${LSOF_INCLUDE}/netinet6/in6.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+ LSOF_DIALECT_DIR=freebsd
+ ;;
+
+# Configure for HP-UX and HP-UX gcc.
+
+ hpux|hpuxgcc)
+ LSOF_CFGL=""
+ LSOF_RANLIB=""
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the HP-UX version isn't pre-defined, determine it.
+
+ LSOF_VERS=`echo $LSOF_VSTR | awk -F. '{printf "%d%02d",\$2,\$3}'`
+ fi # }
+ if test $LSOF_VERS -ge 1020 # {
+ then
+ LSOF_TSTBIGF="-D_LARGEFILE64_SOURCE"
+ fi # }
+
+ # Determine compiler.
+
+ if test "X$LSOF_CC" = "X" # {
+ then
+ if test "X$LSOF_TGT" = "Xhpuxgcc" # {
+ then
+ LSOF_CC=gcc
+ else
+ if test "X$HPUX_CCDIR1" = "X" # {
+ then
+ HPUX_CCDIR1="/bin"
+ fi # }
+ if test "X$HPUX_CCDIR2" = "X" # {
+ then
+ HPUX_CCDIR2="/usr/ccs/bin"
+ fi # }
+ if test -x ${HPUX_CCDIR1}/cc # {
+ then
+ LSOF_CC=${HPUX_CCDIR1}/cc
+ else
+ if test -x ${HPUX_CCDIR2}/cc # {
+ then
+ LSOF_CC=${HPUX_CCDIR2}/cc
+ else
+ echo "No executable cc in $HPUX_CCDIR1 or $HPUX_CCDIR2"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+ $LSOF_CC -O < /dev/null 2>&1 | grep -q Bundled
+ if test $? -eq 0 # {
+ then
+ LSOF_DEBUG="No-O" # to disable -O
+ if test "X$HPUX_LIBC1" = "X" # {
+ then
+ HPUX_LIBC1="/usr/lib"
+ fi # }
+ if test -r ${HPUX_LIBC1}/libc.sl # {
+ then
+ LSOF_FCFGL="-L$HPUX_LIBC -lc"
+ else
+ if test "X$HPUX_LIBC2" = "X" # {
+ then
+ HPUX_LIBC2="/usr/lib"
+ fi # }
+ if test -r ${HPUX_LIBC2}/libc.sl # {
+ then
+ LSOF_FCFGL="-L$HPUX_LIBC2 -lc"
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ else
+ $LSOF_CC -O < /dev/null 2>&1 | grep -q Bundled
+ if test $? -eq 0 # {
+ then
+ LSOF_DEBUG="No-O" # to disable -O
+ fi # }
+ fi # }
+ LSOF_TGT=hpux
+
+ # Test for "const void" support.
+
+ rm -f ${LSOF_TMPC}.*
+ echo "main() { const void *x; return(0); }" >> $LSOF_TMPC.c
+ $LSOF_CC $LSOF_TMPC.c -o $LSOF_TMPC.x > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="-DHAS_CONST"
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+
+ # Test HP-UX base.
+
+ if test "X$HPUX_BASE" = "X" # {
+ then
+ if test -d $LSOF_INCLUDE/sys/pstat -a $LSOF_VERS -ge 1111 # {
+ then
+ HPUX_BASE="pstat"
+ else
+ HPUX_BASE="/dev/kmem"
+ fi # }
+ fi # }
+ if test "X$HPUX_BASE" = "Xpstat" # {
+ then
+
+ # Configure for pstat-based HP-UX lsof.
+
+ LSOF_CINFO="PSTAT-based"
+ echo "Configuring PSTAT-based HP-UX lsof"
+ LSOF_DIALECT_DIR=hpux/pstat
+ LSOF_CFGF="$LSOF_CFGF -DHPUXV=$LSOF_VERS -D_PSTAT64"
+ LSOF_CFGL="$LSOF_CFGL -lnsl"
+ LSOF_TSTKMEM=0
+ LSOF_TSTK64=1
+ ls -l $LSOF_CC | grep -q ansic
+ LSOF_TMP1=$?
+ ls -l $LSOF_CC | grep -q aCC
+ if test $? -eq 0 -o $LSOF_TMP1 -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -Ae +DD32"
+ else
+ echo $LSOF_CC | grep -q gcc
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF +DD32"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/netinet/in6.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/pstat/stream_pstat_body.h # {
+ then
+ grep -q PS_STR_XPORT_DATA ${LSOF_INCLUDE}/sys/pstat/stream_pstat_body.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -D_PSTAT_STREAM_GET_XPORT"
+ fi # }
+ fi # }
+ if test $LSOF_VERS -ge 1123 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -D_LARGEFILE64_SOURCE"
+ fi # }
+ else
+ if test "X$HPUX_BASE" = "X/dev/kmem" # {
+ then
+
+ # Configure for /dev/kmem-based HP-UX lsof.
+
+ if test "X$HPUX_BOOTFILE" = "X" # {
+ then
+ HPUX_BOOTFILE="/stand/vmunix"
+ fi # }
+ if test $LSOF_VERS -gt 1100 # {
+ then
+ echo ""
+ echo "************************************************"
+ echo "* *"
+ echo "* !!!!!!!!!!!!!!! FATAL ERROR !!!!!!!!!!!!!!!! *"
+ echo "* *"
+ echo "* LSOF DOES NOT SUPPORT THIS VERSION OF HP-UX. *"
+ echo "* *"
+ echo "************************************************"
+ echo ""
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -DHPUXV=$LSOF_VERS"
+ LSOF_CINFO="/dev/kmem-based"
+ LSOF_DIALECT_DIR=hpux/kmem
+ echo "Configuring /dev/kmem-based HP-UX lsof"
+ if test $LSOF_VERS -lt 1000 # {
+ then
+ if test "X$HPUX_X25DIR" = "X" # {
+ then
+ HPUX_X25DIR="/etc/conf"
+ else
+ HPUX_X25DIR=$HPUX_X25DIR
+ fi # }
+ if test -r ${HPUX_X25DIR}/x25/x25addrstr.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHPUX_CCITT"
+ LSOF_DINC="$LSOF_DINC -I$HPUX_X25DIR"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_inode.h -a -r ${LSOF_INCLUDE}/sys/fs/vx_hpux.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS"
+ fi # }
+ if test $LSOF_VERS -ge 1030 # {
+ then
+ if test "X$HPUX_KERNBITS" = "X" # {
+ then
+ HPUX_KERNBITS=`getconf _SC_KERNEL_BITS`
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -DHPUXKERNBITS=${HPUX_KERNBITS} -I`pwd`/dialects/hpux/kmem/hpux11"
+ if test $HPUX_KERNBITS -eq 64 # {
+ then
+ LSOF_TSTK64=1
+ echo ""
+ echo "*****************************************"
+ echo "* *"
+ echo "* NOTICE! Configuring for 64 bit HP-UX *"
+ echo "* *"
+ echo "*****************************************"
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+
+ # Test gcc for 64 bit support, trying gcc with no options, then
+ # with -mlp64, testing the result with file.
+
+ echo ""
+ echo "Testing $LSOF_CC for 64 bit support"
+ rm -f ${LSOF_TMPC}.*
+ echo "main(){}" > ${LSOF_TMPC}.c
+ LSOF_TMP1=""
+ $LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ /bin/file ${LSOF_TMPC}.x | /bin/grep 64 > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1=" "
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP1" = "X" # {
+ then
+ rm -f ${LSOF_TMPC}.x
+ $LSOF_CC ${LSOF_TMPC}.c -mlp64 -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ /bin/file ${LSOF_TMPC}.x | /bin/grep 64 > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1="-mlp64"
+ fi # }
+ fi # }
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ if test "X$LSOF_TMP1" = "X" # {
+ then
+ echo ""
+ echo "***************************************************"
+ echo "* *"
+ echo "* !!!!!!!!!!!!!!!!! FATAL ERROR !!!!!!!!!!!!!!!!! *"
+ echo "* *"
+ echo "* APPARENTLY GCC CANNOT BUILD 64 BIT EXECUTABLES. *"
+ echo "* A COMPILER MUST BE USED THAT CAN. SEE 00FAQ *"
+ echo "* FOR MORE INFORMATION. *"
+ echo "* *"
+ echo "***************************************************"
+ echo ""
+ rm -f $LSOF_HLP
+ exit 1
+ else
+ if test "X$LSOF_TMP1" != "X " # {
+ then
+ LSOF_CFGF="$LSOF_CFGF $LSOF_TMP1"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lelf"
+ LSOF_CINFO="${LSOF_CINFO}, 64 bit HP-UX"
+ fi # }
+ else
+
+ # Set options for the HP-UX compiler.
+
+ LSOF_CFGF="$LSOF_CFGF +DD64"
+ LSOF_CFGL="$LSOF_CFGL -lelf"
+ LSOF_CINFO="${LSOF_CINFO}, 64 bit HP-UX"
+ fi # }
+ else
+ LSOF_CFGF="$LSOF_CFGF -D_FILE_OFFSET_BITS=64"
+ LSOF_CINFO="${LSOF_CINFO}, 32 bit HP-UX"
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF +DAportable"
+ fi # }
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lnsl"
+ else
+
+ # When HP-UX is less than 10.30, but greater than or equal to 10,
+ # check NFS3 rnode status.
+
+ if test $LSOF_VERS -ge 1000 # {
+ then
+ LSOF_TMP1=0
+ if test "X$HPUX_RNODE3" = "X" # {
+ then
+ nm -x $HPUX_BOOTFILE | grep -q nfs_vnodeops3
+ if test $? -eq 0 # {
+ then
+ if test -r ${LSOF_INCLUDE}/nfs/rnode.h # {
+ then
+ grep -q r_fh3 ${LSOF_INCLUDE}/nfs/rnode.h
+ if test $? -ne 0 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ else
+ if test "X$HPUX_RNODE3" = "X1" # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASRNODE3"
+ fi # }
+ fi # }
+ fi # }
+ if test $LSOF_VERS -eq 1100 # {
+ then
+
+ # Test for the ipis_s structure. If it is present, set HAS_IPC_S_PATCH.
+
+ if test "X$HPUX_IPC_S_PATCH" = "X" # {
+ then
+ if test -x /usr/contrib/Q4/bin/q4exe # {
+ then
+ LSOF_TMP=/usr/contrib/Q4/bin/q4exe
+ else
+ LSOF_TMP=/usr/contrib/bin/q4
+ fi # }
+ if test -x $LSOF_TMP # {
+ then
+ rm -f ${LSOF_TMPC}.out
+ echo ""
+ echo $EO "Looking in $HPUX_BOOTFILE for ipis_s with $LSOF_TMP ... $EC"
+ echo "yes\\nfields -c struct ipis_s" | $LSOF_TMP $HPUX_BOOTFILE > ${LSOF_TMPC}.out 2>&1
+ if test $? -ne 0 # {
+ then
+ echo ""
+ echo ""
+ echo "!!!ERROR!!! $LSOF_TMP failed and produced the following output."
+ echo ""
+ cat ${LSOF_TMPC}.out
+ HPUX_IPC_S_PATCH=fail
+ else
+ grep ipis_s ${LSOF_TMPC}.out > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ echo "ipis_s exists."
+
+ # See if ipis_msgsqueued is present.
+
+ grep ipis_msgsqueued ${LSOF_TMPC}.out > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ HPUX_IPC_S_PATCH=2
+ else
+ HPUX_IPC_S_PATCH=1
+ fi # }
+ else
+ echo "ipis_s doesn't exist."
+ HPUX_IPC_S_PATCH=N
+ fi # }
+ fi # }
+ rm -f ${LSOF_TMPC}.out
+ else
+ echo "Can't locate or execute $LSOF_TMP"
+ echo $EO "ls says: $EC"
+ ls -ld $LSOF_TMP
+ HPUX_IPC_S_PATCH=fail
+ fi # }
+ fi # }
+ if test "X$HPUX_IPC_S_PATCH" = "Xfail" # {
+ then
+ echo ""
+ echo "!!!ERROR!!! !!!ERROR!!! !!!ERROR!!! !!!ERROR!!!"
+ echo "Configure can't use $LSOF_TMP to examine the ipis_s"
+ echo "structure. You must do that yourself, report the result in"
+ echo "the HPUX_IPC_S_PATCH environment variable, then repeat the"
+ echo "Configure step. Consult the Configure script's use of"
+ echo "$LSOF_TMP and the 00XCONFIG file for information"
+ echo "on ipis_s testing and the setting of HPUX_IPC_S_PATCH."
+ echo "!!!ERROR!!! !!!ERROR!!! !!!ERROR!!! !!!ERROR!!!"
+ echo ""
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ if test "X$HPUX_IPC_S_PATCH" = "X1" -o "X$HPUX_IPC_S_PATCH" = "X2" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_IPC_S_PATCH=$HPUX_IPC_S_PATCH"
+ else
+ if test "X$HPUX_IPC_S_PATCH" != "Xn" -a "X$HPUX_IPC_S_PATCH" != "XN" # {
+ then
+ echo "Illegal value for HPUX_IPC_S_PATCH: $HPUX_IPC_S_PATCH"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+ fi #}
+
+ # Manufacture an hpux_mount.h header file with a mount struct in it, as
+ # required.
+
+ if test -r ${LSOF_INCLUDE}/sys/mount.h # {
+ then
+ LSOF_TMP1="dialects/${LSOF_DIALECT_DIR}/hpux_mount.h"
+ rm -f $LSOF_TMP1
+ echo "#if !defined(MANUFACTURED_HPUX_SYS_MOUNT_H)" > $LSOF_TMP1
+ echo "#define MANUFACTURED_HPUX_SYS_MOUNT_H" >> $LSOF_TMP1
+ echo "/* By lsof Configure:" `date` " */" >> $LSOF_TMP1
+ echo "struct mount" >> $LSOF_TMP1
+ sed '1,/struct mount/d' ${LSOF_INCLUDE}/sys/mount.h | sed -n '1,/m_dev/p' >> $LSOF_TMP1
+ echo "};" >> $LSOF_TMP1
+ echo "#endif" >> $LSOF_TMP1
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/${LSOF_DIALECT_DIR}"
+ fi # }
+
+ # Test for OnlineJFS.
+
+ if test $LSOF_VERS -ge 1100 # {
+ then
+ if test "X$HPUX_HASONLINEJFS" = "X" -a -x /sbin/fs/vxfs/subtype # {
+ then
+ LSOF_TMP1=`/sbin/fs/vxfs/subtype`
+ if test "X$LSOF_TMP1" = "Xvxfs3.3"
+ then
+ HPUX_HASONLINEJFS="Y"
+ fi # }
+ fi # }
+ if test "X$HPUX_HASONLINEJFS" = "XY" -o "X$HPUX_HASONLINEJFS" = "Xy"
+ # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASONLINEJFS"
+ fi # }
+ fi # }
+
+ # Test for AFS.
+
+ if test -r ${AFS_VICE}/etc/ThisCell # {
+ then
+ if test "X$LSOF_SCRIPT_CALL" = "Xno" # {
+ then
+ if test -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ else
+ if test ! -x ./AFSConfig # {
+ then
+ echo "Can't find or execute the AFSConfig script"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ ./AFSConfig
+ if test $? -eq 0 -a -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ fi # }
+ if test "X$LSOF_AFS" = "Xyes" # {
+ then
+ LSOF_AFSV=`cat ./AFSVersion | sed 's/^\([0-9]*\)\.\([0-9]*\).*/\1 \2/' | awk '{printf "%d%02d\n",\$1,\$2}'`
+ LSOF_CFGF="$LSOF_CFGF -DHAS_AFS=$LSOF_AFSV"
+ LSOF_DINC="$LSOF_DINC -I`cat ./AFSHeaders`"
+ fi # }
+ fi # }
+ else
+ echo "HP-UX base unrecognized: $HPUX_BASE"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+ ;;
+
+# Configure for Linux.
+
+ linux)
+ LSOF_TSTBIGF="-D_FILE_OFFSET_BITS=64"
+ LSOF_TSTKMEM=0
+ if test "X$LSOF_CC" = "X" # {
+ then
+ LSOF_CC=cc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ fi # }
+ LSOF_DIALECT_DIR=""
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the Linux version isn't predefined, determine it.
+
+ LSOF_VERS=`echo $LSOF_VSTR | sed 's/\./ /g' | awk '{printf "%d%d%03d",\$1,\$2,\$3}'`
+ fi # }
+ LSOF_CFGF="-DLINUXV=$LSOF_VERS"
+ if test $LSOF_VERS -lt 21072 # {
+ then
+ echo ""
+ echo "!!!WARNING!!!==!!!WARNING!!!==!!!WARNING!!!==!!!WARNING!!!"
+ echo "! !"
+ echo "! THE /PROC-BASED LSOF SOURCES HAVE NOT BEEN TESTED ON !"
+ echo "! LINUX KERNELS BELOW 2.1.72, AND MAY NOT WORK ON THIS !"
+ echo "! KERNEL. IT SHOULD USE A /DEV/KMEM-BASED LSOF. !"
+ echo "! !"
+ echo "!!!WARNING!!!==!!!WARNING!!!==!!!WARNING!!!==!!!WARNING!!!"
+ echo ""
+ else
+ LSOF_UNSUP=""
+ fi # }
+
+ # If the Linux C library type isn't predefined, determine it.
+
+ if test "X$LINUX_CLIB" = "X" # {
+ then
+ echo -n "Testing C library type with $LSOF_CC ... "
+ rm -f ${LSOF_TMPC}.*
+ cat > $LSOF_TMPC.c << .LSOF_END_HERE_DOC1
+#include <features.h>
+main() {
+#if defined(__GLIBC__) && defined(__GLIBC_MINOR__)
+printf("-DGLIBCV=%d\n",__GLIBC__*100+__GLIBC_MINOR__);
+#elif defined(__GLIBC__)
+printf("-DGLIBCV=%d00\n",__GLIBC__);
+#else
+printf("\n");
+#endif
+return(0); }
+.LSOF_END_HERE_DOC1
+ $LSOF_CC ${LSOF_TMPC}.c -I$LSOF_INCLUDE -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ if test -x ${LSOF_TMPC}.x # {
+ then
+ LINUX_CLIB=`${LSOF_TMPC}.x`
+ LSOF_TMP=$?
+ else
+ LINUX_CLIB=""
+ LSOF_TMP=1
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ echo "done"
+ if test $LSOF_TMP -ne 0 # {
+ then
+ echo "Cannot determine C library type; assuming it is not glibc."
+ LINUX_CLIB=""
+ else
+ if test "X$LINUX_CLIB" = "X" # {
+ then
+ echo "The C library type is not glibc."
+ else
+ echo "The C library type is glibc, version \"$LINUX_CLIB\"."
+ fi # }
+ fi # }
+ fi # }
+ if test "X$LINUX_CLIB" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF $LINUX_CLIB"
+ fi # }
+
+ # Test for IPv6 support.
+
+ if test -r ${LSOF_INCLUDE}/netinet/ip6.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+
+ # Test for SELinux support.
+
+ LSOF_TMP1=0
+ if test "X$LINUX_HASSELINUX" = "X" # {
+ then
+ if test -r ${LSOF_INCLUDE}/selinux/selinux.h # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ else
+ if test "X$LINUX_HASSELINUX" = "XY" -o "X$LINUX_HASSELINUX" = "xY" # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASSELINUX"
+ LSOF_CFGL="$LSOF_CFGL -lselinux"
+ fi # }
+ LSOF_DIALECT_DIR="linux"
+ LSOF_CFGF="$LSOF_CFGF -D_FILE_OFFSET_BITS=64 -D_LARGEFILE64_SOURCE"
+ ;;
+
+# Configure for NetBSD.
+
+ netbsd)
+ if test "X$LSOF_CC" = "X" # {
+ then
+ LSOF_CC=cc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ fi # }
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # Validate the NetBSD version.
+
+ case $LSOF_VSTR in # {
+ 1.2*)
+ LSOF_VERS="1002000"
+ ;;
+ 1.3*)
+ LSOF_VERS="1003000"
+ ;;
+ 1.4*)
+ LSOF_VERS="1004000"
+ ;;
+ 1.5*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="1005000"
+ ;;
+ 1.6*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="1006000"
+ ;;
+ 1*)
+ LSOF_VERS="1006000"
+ echo "!!!WARNING!!! Unsupported NetBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for NetBSD 1.6"
+ ;;
+ 2.0*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="2000000"
+ ;;
+ 2.99.9)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="2099009"
+ ;;
+ 2.99.10)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="2099010"
+ ;;
+ 2.99.*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="2099010"
+ ;;
+ 2*)
+ LSOF_VERS="2000000"
+ echo "!!!WARNING!!! Unsupported NetBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for NetBSD 2.0"
+ ;;
+ 3.0*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="3000000"
+ ;;
+ 3.99.*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS="3099000"
+ ;;
+ 3*)
+ LSOF_VERS="3000000"
+ echo "!!!WARNING!!! Unsupported NetBSD version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for NetBSD 3.0"
+ ;;
+ *)
+ echo "Unknown NetBSD release: $LSOF_VSTR"
+ echo Assuming NetBSD 1.6
+ LSOF_VERS="1006000"
+ ;;
+ esac # }
+ fi # }
+
+ # Test for legal NetBSD version.
+
+ case $LSOF_VERS in # {
+ 1002000|1003000|1004000|1005000|1006000)
+ ;;
+ 2000000|2099009|2099010)
+ ;;
+ 3000000|3099000)
+ ;;
+ *)
+ echo "Unknown NetBSD version: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ LSOF_CFGF="-DNETBSDV=$LSOF_VERS"
+ LSOF_TMP1="-DN_UNIXV=/netbsd"
+ if test -r ${LSOF_INCLUDE}/util.h # {
+ then
+ grep -q getbootfile ${LSOF_INCLUDE}/util.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGL="-lutil"
+ LSOF_TMP1="-DHASGETBOOTFILE"
+ fi # }
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF $LSOF_TMP1"
+ if test -r ${LSOF_INCLUDE}/kvm.h # {
+ then
+ grep -q kvm_getproc2 ${LSOF_INCLUDE}/kvm.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASKVMGETPROC2"
+ fi # }
+ fi # }
+
+ # Here begin the dual tests on header files that may be in $LSOF_INCLUDE
+ # or $NETBSD_SYS.
+ #
+ # Note that $LSOF_TMP1 holds an indicator of the need for -I$NETBSD_SYS.
+ # LSOF_TMP4 contains a temporary indicator of the use of $NETBSD_SYS.
+
+ LSOF_TMP1=0
+ if test "X$NETBSD_SYS" = "X" # {
+ then
+ if test -d /usr/src # {
+ then
+ NETBSD_SYS="/usr/src/sys"
+ else
+ NETBSD_SYS=$LSOF_INCLUDE
+ fi # }
+ fi # }
+ LSOF_TMP2="nfs/nfsproto.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNFSPROTO"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ LSOF_TMP2="netinet/ip6.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ else
+ LSOF_TMP2="netinet/in.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q IPV6_INRIA_VERSION $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6 -DHASINRIAIPv6"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="miscfs/fdesc/fdesc.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q Fctty $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASFDESCFS=1"
+ else
+ LSOF_CFGF="$LSOF_CFGF -DHASFDESCFS=2"
+ fi # }
+ grep -q fd_link $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASFDLINK"
+ fi # }
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ LSOF_TMP2="miscfs/nullfs/null.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNULLFS"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ LSOF_TMP2="miscfs/procfs"
+ if test -d ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -d ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPROCFS"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ if test -r ${LSOF_TMP3}/procfs.h # {
+ then
+ grep -q PFSroot ${LSOF_TMP3}/procfs.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPROCFS_PFSROOT"
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/bufq.h"
+ LSOF_NBSD_BUFQH=0
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASBUFQ_H"
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ if test $NETBSD_SYS != $LSOF_INCLUDE # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASBUFQ_H"
+ LSOF_NBSD_BUFQH=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="isofs/cd9660"
+ if test -d ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -d ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS9660FS=1"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ else
+ LSOF_TMP2="fs/cd9660"
+ if test -d ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -d ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS9660FS=1"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="msdosfs"
+ if test -d ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -d ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASMSDOSFS=1"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ else
+ LSOF_TMP2="fs/msdosfs"
+ if test -d ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -d ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASMSDOSFS=2"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="miscfs/kernfs/kernfs.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q "kt_name;" $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASKERNFS"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ grep -q "*kfs_kt;" $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASKERNFS_KFS_KT"
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/namei.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q nc_vpid $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNCVPID"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="ufs/ufs/inode.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q i_ffs_size $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASI_FFS"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ else
+ grep -q i_ffs1_size $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASI_FFS1"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ grep -q i_ffs_effnlink $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASEFFNLINK=i_ffs_effnlink"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/vnode.h"
+ LSOF_NBSD_PTYFS=0
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "XLSOF_TMP3" != "X" # {
+ then
+ grep -q VT_EXT2FS $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASEXT2FS"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ LSOF_TMP2="ufs/ufs/inode.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP5="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP6=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP5="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP6=1
+ else
+ LSOF_TMP5=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP5" != "X" # {
+ then
+ grep -q "*e2fs_din" $LSOF_TMP5
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASI_E2FS_PTR"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=$LSOF_TMP6
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ grep -q VT_LFS $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASLFS"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ grep -q VT_PTYFS $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP2="fs/ptyfs/ptyfs.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPTYFS"
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ if test $NETBSD_SYS != $LSOF_INCLUDE # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPTYFS"
+ LSOF_NBSD_PTYFS=1
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ if test "X$NETBSD_UVM" = "X" # {
+ then
+ grep -q UVM $LSOF_TMP3
+ if test $? -ne 0 # {
+ then
+ egrep -q "v_uvm;|v_uobj;" $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ NETBSD_UVM="Y"
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="nfs/nfsnode.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q "*n_vattr" $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNFSVATTRP"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/lockf.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q vop_advlock_args $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_ADVLOCK_ARGS"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ grep -q lf_lwp $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_LF_LWP"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/lwp.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_LWP_H"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/filedesc.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ grep -q "^struct cwdinfo {" $LSOF_TMP3
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASCWDINFO"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/pipe.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ LSOF_TMP4=0
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ LSOF_TMP4=1
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_SYS_PIPEH"
+ if test $LSOF_TMP1 -eq 0 -a $LSOF_TMP4 -eq 1 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/statvfs.h # {
+ then
+ grep -q '^struct statvfs {' ${LSOF_INCLUDE}/sys/statvfs.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASSTATVFS"
+ fi # }
+ fi # }
+
+ # Here end the dual NetBSD tests for header files in $LSOF_INCLUDE or
+ # NETBSD_SYS.
+ #
+ # After this LSOF_TMP1 may be reused.
+
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_DINC="-I$LSOF_INCLUDE -I$NETBSD_SYS"
+ fi # }
+
+ # Build special header files, as required.
+
+ rm -rf dialects/n+obsd/include
+ if test "X$NETBSD_UVM" = "XY" -o "X$NETBSD_UVM" = "Xy" # {
+ then
+ mkdir dialects/n+obsd/include
+ touch dialects/n+obsd/include/opt_uvmhist.h
+ touch dialects/n+obsd/include/opt_lockdebug.h
+ LSOF_CFGF="$LSOF_CFGF -DUVM -I`pwd`/dialects/n+obsd/include"
+ if test -d ${LSOF_INCLUDE}/uvm # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_UVM_INCL"
+ fi # }
+ fi # }
+ LSOF_TMP2="sys/mount.h"
+ if test -r ${LSOF_INCLUDE}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${LSOF_INCLUDE}/$LSOF_TMP2"
+ else
+ if test -r ${NETBSD_SYS}/$LSOF_TMP2 # {
+ then
+ LSOF_TMP3="${NETBSD_SYS}/$LSOF_TMP2"
+ else
+ LSOF_TMP3=""
+ fi # }
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+
+ # Build a local NetBSD netexport.h header file for possible use by
+ # <msdosfs/msdosfsmount.h>. Make sure CFGL contains a -I for it.
+
+ LSOF_TMP1=${LSOF_TMPC}.edscr
+ LSOF_TMP2=${LSOF_TMPC}.netcred
+ LSOF_TMP3=${LSOF_TMPC}.netexport
+ LSOF_TMP4=dialects/n+obsd/include/netexport.h
+ if test ! -d dialects/n+obsd/include # {
+ then
+ mkdir dialects/n+obsd/include
+ fi # }
+ rm -f $LSOF_TMP1 $LSOF_TMP2 $LSOF_TMP3 $LSOF_TMP4
+ echo "/^struct netcred" > $LSOF_TMP1
+ echo "1,.-1d" >> $LSOF_TMP1
+ echo "/^};" >> $LSOF_TMP1
+ echo "1,.w $LSOF_TMP2" >> $LSOF_TMP1
+ ed ${LSOF_INCLUDE}/sys/mount.h < $LSOF_TMP1 > /dev/null 2>&1
+ rm -f $LSOF_TMP1
+ echo "/^struct netexport" > $LSOF_TMP1
+ echo "1,.-1d" >> $LSOF_TMP1
+ echo "/^};" >> $LSOF_TMP1
+ echo "1,.w $LSOF_TMP3" >> $LSOF_TMP1
+ ed ${LSOF_INCLUDE}/sys/mount.h < $LSOF_TMP1 > /dev/null 2>&1
+ echo "/*" > $LSOF_TMP4
+ echo " * netexport.h" >> $LSOF_TMP4
+ echo -n " * Created by Configure: " >> $LSOF_TMP4
+ echo `date` >> $LSOF_TMP4
+ echo " */" >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ echo "#if !defined(NETEXPORT_H)" >> $LSOF_TMP4
+ echo "#define NETEXPORT_H" >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ echo "#include <net/radix.h>" >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ if test -r $LSOF_TMP2 # {
+ then
+ cat $LSOF_TMP2 >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ fi # }
+ if test -r $LSOF_TMP3 # {
+ then
+ cat $LSOF_TMP3 >> $LSOF_TMP4
+ fi # }
+ echo "#endif /* !defined(NETEXPORT_H) */" >> $LSOF_TMP4
+ rm -f $LSOF_TMP1 $LSOF_TMP2 $LSOF_TMP3
+ echo $LSOF_CFGF | grep /dialects/n+obsd/include > /dev/null 2>&1
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/n+obsd/include"
+ fi # }
+ fi # }
+ if test $LSOF_NBSD_BUFQH -eq 1 # {
+ then
+
+ # Make a local copy of $NETBSD_SYS/sys/bufq.h.
+
+ if test ! -d dialects/n+obsd/include # {
+ then
+ mkdir dialects/n+obsd/include
+ fi # }
+ if test ! -d dialects/n+obsd/include/sys # {
+ then
+ mkdir dialects/n+obsd/include/sys
+ fi # }
+ cp $NETBSD_SYS/sys/bufq.h dialects/n+obsd/include/sys
+ echo $LSOF_CFGF | grep /dialects/n+obsd/include > /dev/null 2>&1
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/n+obsd/include"
+ fi # }
+ fi # }
+ if test $LSOF_NBSD_PTYFS -eq 1 # {
+ then
+
+ # Make a local copy of $NETBSD_SYS/sys/fs/ptyfs/.
+
+ if test ! -d dialects/n+obsd/include # {
+ then
+ mkdir dialects/n+obsd/include
+ fi # }
+ if test ! -d dialects/n+obsd/include/fs # {
+ then
+ mkdir dialects/n+obsd/include/fs
+ fi # }
+ rm -rf dialects/n+obsd/include/fs/ptyfs
+ mkdir dialects/n+obsd/include/fs/ptyfs
+ cp $NETBSD_SYS/fs/ptyfs/*.h dialects/n+obsd/include/fs/ptyfs
+ echo $LSOF_CFGF | grep /dialects/n+obsd/include > /dev/null 2>&1
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/n+obsd/include"
+ fi # }
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lkvm"
+ LSOF_DIALECT_DIR=n+obsd
+ ;;
+
+# Configure for NeXTSTEP or OPENSTEP.
+
+ nextstep|next|ns|nxt|openstep|os)
+ LSOF_TGT="ns"
+ LSOF_TSTXO="../lib/snpf.o"
+ if test "X$LSOF_AR" = "X" # {
+ then
+ LSOF_AR="rm -f \${LIB}; ar cr"
+ fi # }
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`hostinfo | sed -n 's/.*NeXT Mach \([0-9\.]*\).*/\1/p'`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the NeXSTEP version isn't predefined, determine it.
+
+ LSOF_VERS=`echo $LSOF_VSTR | sed -n 's/\([0-9]*\)\.\([0-9]*\)/\1\2/p'`
+ fi # }
+ if test "X$LSOF_CC" = "X" # {
+ then
+ if test -x /usr/local/bin/gcc # {
+ then
+ LSOF_CC=/usr/local/bin/gcc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ else
+ LSOF_CC=cc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ fi # }
+ fi # }
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGL="-w"
+ LSOF_DEBUG="-pedantic -O"
+ else
+ LSOF_CFGL=""
+ fi # }
+ LSOF_CFGF="-DSTEPV=$LSOF_VERS"
+ LSOF_DIALECT_DIR=n+os
+
+ # Test for AFS.
+
+ if test -r ${AFS_VICE}/etc/ThisCell # {
+ then
+ if test "X$LSOF_SCRIPT_CALL" = "Xno" # {
+ then
+ if test -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ else
+ if test ! -x ./AFSConfig # {
+ then
+ echo "Can't find or execute the AFSConfig script"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ ./AFSConfig
+ if test $? -eq 0 -a -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ fi # }
+ if test "X$LSOF_AFS" = "Xyes" # {
+ then
+ LSOF_AFSV=`cat ./AFSVersion | sed 's/^\([0-9]*\)\.\([0-9]*\).*/\1 \2/' | awk '{printf "%d%02d\n",\$1,\$2}'`
+ LSOF_CFGF="$LSOF_CFGF -DHAS_AFS=$LSOF_AFSV"
+ LSOF_DINC="$LSOF_DINC -I`cat ./AFSHeaders`"
+ fi # }
+ fi # }
+ ;;
+
+# Configure for OpenBSD. (OpenBSD uses NetBSD dialect sources and version
+# numbering.
+
+ openbsd)
+ if test "X$LSOF_CC" = "X" # {
+ then
+ LSOF_CC=cc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ fi # }
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the OpenBSD version isn't pre-defined, determine it.
+
+ case $LSOF_VSTR in # {
+ 1*)
+ LSOF_VERS=1020
+ echo "!!!WARNING!!! Unsupported OpenBSD 1.x version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for OpenBSD 1.2"
+ ;;
+ 2.5*)
+ LSOF_VERS=2050
+ ;;
+ 2.6*)
+ LSOF_VERS=2060
+ ;;
+ 2.7*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=2070
+ ;;
+ 2.8*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=2080
+ ;;
+ 2.9*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=2090
+ ;;
+ 2*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=2090
+ echo "!!!WARNING!!! Unsupported OpenBSD 2.x version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for OpenBSD 2.9"
+ ;;
+ 3.0*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3000
+ ;;
+ 3.1*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3010
+ ;;
+ 3.2*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3020
+ ;;
+ 3.3*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3030
+ ;;
+ 3.4*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3040
+ ;;
+ 3.5*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3050
+ ;;
+ 3.6*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3060
+ ;;
+ 3.7*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3070
+ ;;
+ 3.8*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3080
+ ;;
+ 3.9*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3090
+ ;;
+ 3*)
+ LSOF_TSTBIGF=" "
+ LSOF_VERS=3090
+ echo "!!!WARNING!!! Unsupported OpenBSD 3.x version: $LSOF_VSTR"
+ echo "!!!WARNING!!! Configuring for OpenBSD 3.9"
+ ;;
+ *)
+ echo "Unknown OpenBSD release: $LSOF_VSTR"
+ echo Assuming OpenBSD 3.9
+ LSOF_VERS=3090
+ ;;
+ esac # }
+ fi # }
+
+ # Test for legal OpenBSD version.
+
+ case $LSOF_VERS in # {
+ 1020|2050|2060|2070|2080|2090|3000|3010|3020|3030|3040|3050|3060|3070|3080|3090)
+ ;;
+ *)
+ echo "Unknown OpenBSD version: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ LSOF_CFGF="-DOPENBSDV=$LSOF_VERS"
+ if test -r /dev/ksyms # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DN_UNIXV=/dev/ksyms"
+ else
+ LSOF_CFGF="$LSOF_CFGF -DN_UNIXV=/bsd"
+ fi
+ if test -r ${LSOF_INCLUDE}/nfs/nfsproto.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNFSPROTO"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/netinet6/in6.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+ LSOF_TMP1=0
+ if test "X$OPENBSD_SYS" = "X" # {
+ then
+ OPENBSD_SYS="/sys"
+ fi # }
+ if test -r ${OPENBSD_SYS}/miscfs/fdesc/fdesc.h # {
+ then
+ grep -q Fctty ${OPENBSD_SYS}/miscfs/fdesc/fdesc.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASFDESCFS=1"
+ else
+ LSOF_CFGF="$LSOF_CFGF -DHASFDESCFS=2"
+ fi # }
+ grep -q fd_link ${OPENBSD_SYS}/miscfs/fdesc/fdesc.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASFDLINK"
+ fi # }
+ LSOF_TMP1=1
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/vnode.h # {
+ then
+ grep -q VT_LFS ${LSOF_INCLUDE}/sys/vnode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASLFS"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/miscfs/nullfs/null.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNULLFS"
+ else
+ if test -r ${OPENBSD_SYS}/miscfs/nullfs/null.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNULLFS"
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test -d ${OPENBSD_SYS}/miscfs/procfs # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPROCFS"
+ LSOF_TMP1=1
+ fi # }
+ if test -d ${OPENBSD_SYS}/isofs/cd9660 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS9660FS=1"
+ LSOF_TMP1=1
+ else
+ if test -d ${OPENBSD_SYS}/fs/cd9660 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS9660FS=2"
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test -d ${OPENBSD_SYS}/msdosfs # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASMSDOSFS=1"
+ LSOF_TMP1=1
+ else
+ if test -d ${OPENBSD_SYS}/fs/msdosfs # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASMSDOSFS=2"
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test -r ${OPENBSD_SYS}/miscfs/kernfs/kernfs.h # {
+ then
+ grep -q "kt_name;" ${OPENBSD_SYS}/miscfs/kernfs/kernfs.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASKERNFS"
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test $LSOF_TMP1 -eq 1 -a "X$LSOF_INCLUDE" != "X$OPENBSD_SYS" # {
+ then
+ LSOF_DINC="-I$LSOF_INCLUDE -I$OPENBSD_SYS"
+ fi # }
+ grep -q VT_EXT2FS ${LSOF_INCLUDE}/sys/vnode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1=1
+ if test -r ${LSOF_INCLUDE}/ufs/ufs/inode.h # {
+ then
+ grep -q "*e2fs_din" ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASI_E2FS_PTR"
+ fi # }
+ grep -q "^#define[ ]i_e2din" ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1=2
+ fi # }
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -DHASEXT2FS=$LSOF_TMP1"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/ufs/ufs/inode.h # {
+ then
+ grep -q i_effnlink ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASEFFNLINK=i_effnlink"
+ fi # }
+ grep -q dinode_u ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_DINODE_U"
+ fi # }
+ grep -q i_ffs1_size ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASI_FFS1"
+ fi # }
+ grep -q UM_UFS ${LSOF_INCLUDE}/ufs/ufs/inode.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_UM_UFS"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/namei.h
+ then
+ grep -q nc_vpid ${LSOF_INCLUDE}/sys/namei.h
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASNCVPID"
+ fi # }
+ fi # }
+ if test "X$OPENBSD_UVM" = "X" # {
+ then
+ if test -r ${LSOF_INCLUDE}/sys/vnode.h # {
+ then
+ grep -q UVM ${LSOF_INCLUDE}/sys/vnode.h
+ if test $? -ne 0 # {
+ then
+ egrep -q "v_uvm;|v_uobj;" ${LSOF_INCLUDE}/sys/vnode.h
+ if test $? -eq 0 # {
+ then
+ OPENBSD_UVM="Y"
+ fi # }
+ fi # }
+ fi # }
+ fi # }
+ if test "X$OPENBSD_UVM" = "XY" -o "X$OPENBSD_UVM" = "Xy" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DUVM"
+ if test -d ${LSOF_INCLUDE}/uvm # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_UVM_INCL"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/mount.h -a $LSOF_VERS -lt 3030 # {
+ then
+
+ # Build a local OpenBSD netexport.h header file for possible use by
+ # <msdosfs/msdosfsmount.h>. Make sure CFGL contains a -I for it.
+
+ LSOF_TMP1=${LSOF_TMPC}.edscr
+ LSOF_TMP2=${LSOF_TMPC}.netcred
+ LSOF_TMP3=${LSOF_TMPC}.netexport
+ LSOF_TMP4=dialects/n+obsd/include/netexport.h
+ if test ! -d dialects/n+obsd/include # {
+ then
+ mkdir dialects/n+obsd/include
+ fi # }
+ rm -f $LSOF_TMP1 $LSOF_TMP2 $LSOF_TMP3 $LSOF_TMP4
+ echo "/^struct netcred" > $LSOF_TMP1
+ echo "1,.-1d" >> $LSOF_TMP1
+ echo "/^};" >> $LSOF_TMP1
+ echo "1,.w $LSOF_TMP2" >> $LSOF_TMP1
+ ed ${LSOF_INCLUDE}/sys/mount.h < $LSOF_TMP1 > /dev/null 2>&1
+ rm -f $LSOF_TMP1
+ echo "/^struct netexport" > $LSOF_TMP1
+ echo "1,.-1d" >> $LSOF_TMP1
+ echo "/^};" >> $LSOF_TMP1
+ echo "1,.w $LSOF_TMP3" >> $LSOF_TMP1
+ ed ${LSOF_INCLUDE}/sys/mount.h < $LSOF_TMP1 > /dev/null 2>&1
+ echo "/*" > $LSOF_TMP4
+ echo " * netexport.h" >> $LSOF_TMP4
+ echo -n " * Created by Configure: " >> $LSOF_TMP4
+ echo `date` >> $LSOF_TMP4
+ echo " */" >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ echo "#if !defined(NETEXPORT_H)" >> $LSOF_TMP4
+ echo "#define NETEXPORT_H" >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ echo "#include <net/radix.h>" >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ if test -r $LSOF_TMP2 # {
+ then
+ cat $LSOF_TMP2 >> $LSOF_TMP4
+ echo "" >> $LSOF_TMP4
+ fi # }
+ if test -r $LSOF_TMP3 # {
+ then
+ cat $LSOF_TMP3 >> $LSOF_TMP4
+ fi # }
+ echo "#endif /* !defined(NETEXPORT_H) */" >> $LSOF_TMP4
+ rm -f $LSOF_TMP1 $LSOF_TMP2 $LSOF_TMP3
+ echo $LSOF_CFGF | grep /dialects/n+obsd/include > /dev/null 2>&1
+ if test $? -ne 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/n+obsd/include"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/lockf.h # {
+ then
+ grep vop_advlock_args ${LSOF_INCLUDE}/sys/lockf.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_ADVLOCK_ARGS"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/pipe.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_SYS_PIPEH"
+ fi # }
+ LSOF_CFGL="-lkvm"
+ LSOF_DIALECT_DIR=n+obsd
+ ;;
+
+# Configure for SCO OpenServer.
+
+ osr|osrgcc|sco|scogcc)
+ LSOF_RANLIB=""
+
+ if test "X$OSR_CFGF" != "X" # {
+ then
+
+ # Adopt LSOF_CFGF from OSR_CFGF in environment.
+
+ LSOF_CFGF=$OSR_CFGF
+ fi # }
+ if test "X$OSR_CFGL" != "X" # {
+ then
+
+ # Adopt LSOF_CFGL from OSR_CFGL in environment.
+
+ LSOF_CFGL=$OSR_CFGL
+ fi # }
+
+ # Evaluate compiler specification.
+
+ if test "X$LSOF_CC" = "X" # {
+ then
+ if test "X$LSOF_TGT" = "Xosr" -o "X$LSOF_TGT" = "Xsco" # {
+ then
+ LSOF_CC=cc
+ LSOF_TMP1=1
+ else
+ LSOF_CC=gcc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ LSOF_TMP1=2
+ fi # }
+ else
+ LSOF_TMP1=0
+ fi # }
+ LSOF_TGT="osr"
+
+ # Determine version.
+
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR="`LANG=C_C.C /bin/uname -X 2>/dev/null | grep Release | sed 's/Release = \(.*\)/\1/'`"
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the SCO OpenServer release version isn't predefined, determine it.
+
+ case $LSOF_VSTR in # {
+ 3.2v2.0)
+ LSOF_VERS="20"
+ ;;
+ 3.2v2.1)
+ LSOF_VERS="21"
+ ;;
+ 3.2v4.0)
+ LSOF_VERS="40"
+ ;;
+ 3.2v4.1)
+ LSOF_VERS="41"
+ ;;
+ 3.2v4.2)
+ LSOF_VERS="42"
+ ;;
+ 3.2v5.*)
+ LSOF_TSTLFLG="-lsocket"
+ LSOF_VERS="`echo $LSOF_VSTR | sed 's/3\.2v//; s/\.//g'`"
+ ;;
+ *)
+ echo Unknown SCO OpenServer release: $LSOF_VSTR
+ echo Assuming 3.2.0 or 3.2.1
+ LSOF_VERS="0"
+ ;;
+ esac # }
+ fi # }
+
+ # Do SCO OpenServer specific stuff.
+
+ case $LSOF_VERS in # {
+ 0)
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -nointl"
+ LSOF_DEBUG="-Ox"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lrpc -lsocket -lc_s"
+ LSOF_MKC="cp"
+ ;;
+ 20)
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_DEBUG="-Ox"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lrpc -lsocket -lc_s"
+ LSOF_MKC="cp"
+ ;;
+ 21)
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -nointl"
+ LSOF_DEBUG="-Ox"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lrpc -lsocket -lc_s"
+ LSOF_MKC="cp"
+ ;;
+ 40)
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -nointl"
+ LSOF_DEBUG="-Ox"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lrpc -lsocket -lc_s"
+ ;;
+ 41)
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -nointl"
+ LSOF_DEBUG="-Ox"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lrpc -lsocket -lc_s"
+ ;;
+ 42)
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -nointl"
+ LSOF_DEBUG="-Ox"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lrpc -lsocket -lc_s"
+ ;;
+ 5*)
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -belf"
+ LSOF_DEBUG="-O3 -Kspace"
+ fi # }
+ LSOF_CFGL="$LSOF_CFGL -lsocket"
+ ;;
+ *)
+ echo "Unsupported SCO OpenServer release: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ LSOF_CFGF="$LSOF_CFGF -DOSRV=$LSOF_VERS"
+ if test "X$OSR_STATLSTAT" = "X" # {
+ then
+ echo "Testing libc.a for statlstat"
+ /bin/nm /lib/libc.a | grep statlstat > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_STATLSTAT"
+ fi # }
+ else
+ if test "X$OSR_STATLSTAT" = "XY" -o "X$OSR_STATLSTAT" = "Xy" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_STATLSTAT"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/nfs/rnode.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_NFS"
+ fi # }
+ if test ! -r ${LSOF_INCLUDE}/netdb.h # {
+ then
+ LSOF_DINC="$LSOF_DINC -I`pwd`/dialects/osr/include"
+ fi # }
+ LSOF_DIALECT_DIR=osr
+ ;;
+
+# Configure for Sun Solaris, SunPro C and gcc.
+
+ solaris|solariscc)
+ LSOF_RANLIB=""
+ if test "X$LSOF_CC" = "X" # {
+ then
+ if test "X$LSOF_TGT" = "Xsolariscc" # {
+ then
+ if test "X$SOLARIS_CCDIR" = "X" # {
+ then
+ SOLARIS_CCDIR="/opt/SUNWspro/bin"
+ fi # }
+ if test -x ${SOLARIS_CCDIR}/cc # {
+ then
+ LSOF_CC=${SOLARIS_CCDIR}/cc
+ else
+ echo "WARNING: no cc in ${SOLARIS_CCDIR}; using cc without path."
+ LSOF_CC=cc
+ fi # }
+ LSOF_CCV=`$LSOF_CC -V 2>&1 | sed -n 's/^cc: \(.*\)/\1/p'`
+ else
+ LSOF_CC=gcc
+ LSOF_CCV=`$LSOF_CC -v 2>&1 | sed -n 's/.*version \(.*\)/\1/p'`
+ fi # }
+ fi # }
+ LSOF_TGT="solaris"
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -r`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the Solaris version isn't predefined, determine it.
+
+ case $LSOF_VSTR in # {
+ 5.[0-2])
+ LSOF_VERS="20300"
+ ;;
+ 5.3)
+ LSOF_VERS="20300"
+ ;;
+ 5.4)
+ LSOF_VERS="20400"
+ ;;
+ 5.5)
+ LSOF_VERS="20500"
+ ;;
+ 5.5.1)
+ LSOF_VERS="20501"
+ ;;
+ 5.6*)
+ LSOF_TSTLFLG="-lsocket -lnsl"
+ LSOF_VERS="20600"
+ ;;
+ 5.7*)
+ LSOF_TSTBIGF=" "
+ LSOF_TSTLFLG="-lsocket -lnsl"
+ LSOF_VERS="70000"
+ ;;
+ 5.8*)
+ LSOF_TSTBIGF=" "
+ LSOF_TSTLFLG="-lsocket -lnsl"
+ LSOF_VERS="80000"
+ ;;
+ 5.9*)
+ LSOF_TSTBIGF=" "
+ LSOF_TSTLFLG="-lsocket -lnsl"
+ LSOF_VERS="90000"
+ ;;
+ 5.10*)
+ LSOF_TSTBIGF=" "
+ LSOF_TSTLFLG="-lsocket -lnsl"
+ LSOF_VERS="100000"
+ ;;
+ *)
+ echo Unknown Solaris version: $LSOF_VSTR
+ rm -f $LSOF_HLP
+ exit 1
+ esac # }
+ fi # }
+
+ # Clear LSOF_UNSUP message for selected Solaris versions.
+
+ case $LSOF_VERS in # {
+ 90000|100000)
+ LSOF_UNSUP=""
+ ;;
+ esac # }
+
+ # Do Solaris version-specific stuff.
+
+ case $LSOF_VERS in # {
+ 20300)
+
+ # Solaris patch 101318-32 creates a longer kernel tcp_s structure,
+ # and 101318-45 changes the way the vnode's v_filocks member is
+ # handled. The following code creates a symbol definition for
+ # patch 101318 whose value is the patch level. No symbol is defined
+ # if the patch level is not greater than zero.
+
+ if test "X$SOLARIS_23P101318" = "X" # {
+ then
+ LSOF_PL=`grep -h SUNW_PATCHID=101318 /var/sadm/pkg/SUNWcar*/pkginfo | sed 's/.*-//' | sort -u | tail -1`
+ if test "X$LSOF_PL" = "X" # {
+ then
+ LSOF_PL=0
+ fi # }
+ else
+ LSOF_PL=$SOLARIS_23P101318
+ fi # }
+ if test $LSOF_PL -gt 0 # {
+ then
+ LSOF_CFGF="-DP101318=$LSOF_PL"
+ fi # }
+ ;;
+ 20400)
+ if test "X$SOLARIS_24P101945" = "X" # {
+ then
+ LSOF_PL=`grep -h SUNW_PATCHID=101945 /var/sadm/pkg/SUNWcar*/pkginfo | sed 's/.*-//' | sort -u | tail -1`
+ if test "X$LSOF_PL" = "X" # {
+ then
+ LSOF_PL=0
+ fi # }
+ else
+ LSOF_PL=$SOLARIS_24P101945
+ fi # }
+ if test $LSOF_PL -ge 32 # {
+ then
+ if test "X$SOLARIS_24P102303" = "X" # {
+ then
+ LSOF_PL=`grep -h SUNW_PATCHID=102303 /var/sadm/pkg/SUNWhea*/pkginfo | sed 's/.*-//' | sort -u | tail -1`
+ if test "X$LSOF_PL" = "X" # {
+ then
+ LSOF_PL=0
+ fi # }
+ else
+ LSOF_PL=$SOLARIS_24P102303
+ fi # }
+ if test $LSOF_PL -ge 2 # {
+ then
+ echo "WARNING: your Solaris 2.4 system appears to have patches 101945-32 and 102303-2"
+ echo " installed. This probably means the NUM_*_VECTORS definitions in"
+ echo " <sys/auxv.h> don't match the ones used to build your kernel. Consult"
+ echo " the Sun Problems section of the 00FAQ file of the lsof distribution"
+ echo " for more information on how to work around the problem."
+ fi # }
+ fi # }
+ ;;
+ 20500|20501)
+ ;;
+ 20600|70000|80000|90000|100000)
+ if test "X$SOLARIS_26PR_GWINDOWS" = "X" # {
+ then
+ rm -f ${LSOF_TMPC}.*
+ echo "#define _KMEMUSER" > ${LSOF_TMPC}.c
+ echo "#include <sys/proc/prdata.h>" >> ${LSOF_TMPC}.c
+ echo "main(){" >> ${LSOF_TMPC}.c
+ echo "enum prnodetype p=PR_GWINDOWS;}" >> ${LSOF_TMPC}.c
+ echo "Testing prdata.h for PR_GWINDOWS, using $LSOF_CC"
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+ $LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ else
+ $LSOF_CC ${LSOF_TMPC}.c -I$LSOF_INCLUDE -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ fi # }
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPR_GWINDOWS"
+ fi # }
+ else
+ if test "X$SOLARIS_26PR_GWINDOWS" = "XY" -o "X$SOLARIS_26PR_GWINDOWS" = "Xy" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPR_GWINDOWS"
+ fi # }
+ fi # }
+ if test "X$SOLARIS_26PR_LDT" = "X" # {
+ then
+ rm -f ${LSOF_TMPC}.*
+ echo "#define _KMEMUSER" > ${LSOF_TMPC}.c
+ echo "#include <sys/proc/prdata.h>" >> ${LSOF_TMPC}.c
+ echo "main(){" >> ${LSOF_TMPC}.c
+ echo "enum prnodetype p=PR_LDT;}" >> ${LSOF_TMPC}.c
+ echo "Testing prdata.h for PR_LDT, using $LSOF_CC"
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+ $LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ else
+ $LSOF_CC ${LSOF_TMPC}.c -I$LSOF_INCLUDE -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ fi # }
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPR_LDT"
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ else
+ if test "X$SOLARIS_26PR_LDT" = "XY" -o "X$SOLARIS_26PR_LDT" = "Xy" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASPR_LDT"
+ fi # }
+ fi # }
+ if test $LSOF_VERS -ge 70000 # {
+ then
+
+ # Do tests for Solaris 7 and above.
+
+ if test "X$SOLARIS_KERNBITS" = "X" # {
+ then
+ SOLARIS_KERNBITS=`/bin/isainfo -kv`
+ fi # }
+ if test "X$SOLARIS_INSTR" = "X" # {
+ then
+ SOLARIS_INSTR=`/bin/isainfo -k`
+ fi #}
+ echo $SOLARIS_KERNBITS | grep 64 > /dev/null
+ if test $? -eq 0 # {
+ then
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+
+ # Test gcc for 64 bit support.
+
+ echo "Testing $LSOF_CC for 64 bit support"
+ rm -f ${LSOF_TMPC}.*
+ echo "main(){}" > ${LSOF_TMPC}.c
+ LSOF_TMP1=""
+
+ # First try gcc's -m64 option -- it's the most current possibility.
+
+ $LSOF_CC ${LSOF_TMPC}.c -m64 -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ /bin/file ${LSOF_TMPC}.x | /bin/grep 64 > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1="-m64"
+ fi # }
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ if test "X$LSOF_TMP1" = "X" # {
+ then
+
+ # Try using the older -mcpu=v9 option with gcc instead of -m64.
+
+ echo "main(){}" > ${LSOF_TMPC}.c
+ $LSOF_CC ${LSOF_TMPC}.c -mcpu=v9 -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ /bin/file ${LSOF_TMPC}.x | /bin/grep 64 > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1="-mcpu=v9"
+ fi # }
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ fi # }
+ if test "X$LSOF_TMP1" = "X" # {
+ then
+ echo ""
+ echo "!!!WARNING!!!=========!!!WARNING!!!=========!!!WARNING!!!"
+ echo "! !"
+ echo "! LSOF NEEDS TO BE CONFIGURED FOR A 64 BIT KERNEL, BUT !"
+ echo "! THIS GCC DOESN'T SUPPORT THE BUILDING OF 64 BIT !"
+ echo "! SOLARIS EXECUTABLES. LSOF WILL BE CONFIGURED FOR A !"
+ echo "! 32 BIT KERNEL. !"
+ echo "! !"
+ echo "!!!WARNING!!!=========!!!WARNING!!!=========!!!WARNING!!!"
+ echo ""
+ else
+ echo ""
+ echo "*********************************"
+ echo "* Configuring for 64 bit kernel *"
+ echo "*********************************"
+ echo ""
+ LSOF_CFGF="$LSOF_CFGF $LSOF_TMP1"
+ LSOF_CINFO="64 bit kernel"
+ LSOF_TSTK64=1
+ fi # }
+ else
+
+ # Test Sun compiler for 64 bit support.
+
+ case $SOLARIS_INSTR in # {
+ amd64*)
+ LSOF_TMP1="amd64"
+ LSOF_TMP2="amd64"
+ ;;
+ sparc*)
+ LSOF_TMP1="v9"
+ LSOF_TMP2="sparcv9"
+ ;;
+ *)
+ LSOF_TMP1=""
+ ;;
+ esac # }
+ if test "X$LSOF_TMP1" != "X" # {
+ then
+ echo "Testing $LSOF_CC for 64 bit $LSOF_TMP2 support"
+ rm -f ${LSOF_TMPC}.*
+ LSOF_TMP3="-xarch=$LSOF_TMP1"
+ echo "main(){}" > ${LSOF_TMPC}.c
+ LSOF_TMP4=`$LSOF_CC ${LSOF_TMPC}.c $LSOF_TMP3 -o ${LSOF_TMPC}.x 2>&1`
+ if test $? -eq 0 # {
+ then
+ /bin/file ${LSOF_TMPC}.x | /bin/grep 64 > /dev/null
+ if test $? -ne 0 # {
+ then
+ LSOF_TMP3=""
+ else
+ echo "X$LSOF_TMP4" | grep "use -m64" > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP3=-m64
+ fi # }
+ fi # }
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ else
+ LSOF_TMP3=""
+ fi # }
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ echo ""
+ echo "*********************************"
+ echo "* Configuring for 64 bit kernel *"
+ echo "*********************************"
+ echo ""
+ LSOF_CFGF="$LSOF_CFGF $LSOF_TMP3"
+ LSOF_CINFO="64 bit kernel"
+ LSOF_TSTK64=1
+ else
+ echo ""
+ echo "!!!WARNING!!!==========!!!WARNING!!!==========!!!WARNING!!!"
+ echo "!"
+ echo "! LSOF NEEDS TO BE CONFIGURED FOR A 64 BIT KERNEL, BUT"
+ echo "! THE VERSION OF SUN C AVAILABLE DOESN'T SUPPORT THE"
+ echo "! \"$LSOF_TMP2\" INSTRUCTION SET."
+ echo "!"
+ echo "! LSOF WILL BE CONFIGURED FOR A 32 BIT KERNEL."
+ echo "!"
+ echo "!!!WARNING!!!==========!!!WARNING!!!==========!!!WARNING!!!"
+ echo ""
+ fi # }
+ fi # }
+ else
+ echo ""
+ echo "*********************************"
+ echo "* Configuring for 32 bit kernel *"
+ echo "*********************************"
+ echo ""
+ LSOF_CINFO="32 bit kernel"
+ fi # }
+ fi # }
+
+ # Do tests specific to Solaris 8 and above.
+
+ if test $LSOF_VERS -ge 80000 # {
+ then
+ if test -r ${LSOF_INCLUDE}/netinet/ip6.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+ fi # }
+
+ # Do tests specific to Solaris 9 and above.
+
+ if test $LSOF_VERS -ge 90000 # {
+ then
+ if test -r ${LSOF_INCLUDE}/sys/socketvar.h # {
+ then
+ grep soua_vp ${LSOF_INCLUDE}/sys/socketvar.h > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASSOUXSOUA"
+ fi # }
+ fi # }
+ fi # }
+
+ # Do tests specific to Solaris 10 and above.
+
+ if test $LSOF_VERS -ge 100000 # {
+ then
+ if test -r ${LSOF_INCLUDE}/inet/ipclassifier.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_IPCLASSIFIER_H"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/cred_impl.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_CRED_IMPL_H"
+
+ # DEBUG -- Begin temporary hack for Solaris 10, build s10_44.
+
+ grep "c2/audit.h" ${LSOF_INCLUDE}/sys/cred_impl.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ rm -rf `pwd`/dialects/sun/solaris10
+ mkdir `pwd`/dialects/sun/solaris10
+ mkdir `pwd`/dialects/sun/solaris10/c2
+ touch `pwd`/dialects/sun/solaris10/c2/audit.h
+ LSOF_CFGF="$LSOF_CFGF -I`pwd`/dialects/sun/solaris10"
+ fi # }
+
+ # DEBUG -- End temporary hack for Solaris 10, build s10_44.
+
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/vnode.h # {
+ then
+ grep v_path ${LSOF_INCLUDE}/sys/vnode.h > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_V_PATH"
+ LSOF_TSTVPATH=1
+ fi # }
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/zone.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASZONES"
+ fi # }
+
+ # Check for Solaris 10 or higher ZFS.
+
+ if test -r ${LSOF_INCLUDE}/sys/fs/zfs.h # {
+ then
+
+ # Check for required ZFS kernel header files.
+
+ LSOF_TMP1="dmu.h zfs_acl.h zfs_debug.h zfs_rlock.h zil.h spa.h zfs_context.h zfs_dir.h zfs_vfsops.h zio.h txg.h zfs_ctldir.h zfs_ioctl.h zfs_znode.h zio_impl.h"
+ LSOF_TMP2=""
+ for i in $LSOF_TMP1 # {
+ do
+ if test ! -r ${LSOF_INCLUDE}/sys/$i # {
+ then
+ if test "X$LSOF_TMP2" = "X" # {
+ then
+ LSOF_TMP2=$i
+ else
+ LSOF_TMP2="$LSOF_TMP2 $i"
+ fi # }
+ fi # }
+ done # }
+ if test "X$LSOF_TMP2" = "X" # {
+ then
+ LSOF_TMP4="${LSOF_INCLUDE}/sys"
+ LSOF_TMP5=0
+ else
+ LSOF_TMP3=`pwd`/dialects/sun/get-hdr-loc.sh
+ if test ! -x $LSOF_TMP3 # {
+ then
+ echo "FATAL: can't execute: $LSOF_TMP3"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ LSOF_TMP4=`$LSOF_TMP3 $LSOF_TMP2`
+ LSOF_TMP5=$?
+ fi #}
+ if test $LSOF_TMP5 -eq 0 # {
+ then
+
+ # ZFS support has been requested.
+
+ if test "X$LSOF_TMP4" = "X" # {
+ then
+
+ # Use of lsof interal ZFS structure definitions has been
+ # requested.
+
+ LSOF_CFGF="$LSOF_CFGF -DHAS_ZFS=1"
+ else
+
+ # Use of ZFS header files at the returned location has
+ # been requested.
+
+ LSOF_CFGF="$LSOF_CFGF -DHAS_ZFS=2"
+ if test "X$LSOF_TMP4" != "X${LSOF_INCLUDE}/sys" # {
+ then
+
+ # Add the returned location unless it is ${LSOF_INCLUDE}/sys.
+
+ LSOF_CFGF="$LSOF_CFGF -I$LSOF_TMP4"
+ fi # }
+ fi # }
+
+ # Identify the ZFS version.
+
+ LSOF_TMP1=0
+ if test -x /usr/sbin/zpool # {
+ then
+ LSOF_TMP2=`/usr/sbin/zpool upgrade -v|grep -i running|grep -i zfs`
+ if test "X$LSOF_TMP2" != "X" # {
+ then
+ LSOF_TMP3=`echo $LSOF_TMP2 | sed 's/^.* \([0-9][0-9]*\)\.$/\1/'`
+ if test "X$LSOF_TMP3" != "X" # {
+ then
+ LSOF_TMP1=$LSOF_TMP3
+ fi # }
+ fi # }
+ fi # }
+ LSOF_CFGF="$LSOF_CFGF -DZFS_VERS=$LSOF_TMP1"
+ else
+
+ # The get-hdr-loc.sh script returned a non-zero exit value.
+ # If there was a message on STDOUT, an error was detected.
+ # if there was no message, ZFS support is not required.
+
+ if test "X$LSOF_TMP4" != "X" # {
+ then
+ echo "$LSOF_TMP3 detected the error: $LSOF_TMP4"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ fi # }
+ fi # }
+ ;;
+ *)
+ echo "Unsupported Solaris version: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ LSOF_CFGF="-Dsolaris=$LSOF_VERS $LSOF_CFGF"
+
+ # Test for <utmpx.h>
+ if test -r ${LSOF_INCLUDE}/utmpx.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASUTMPX"
+ fi # }
+
+ # Test for VSOCK.
+
+ if test "X$SOLARIS_VSOCK" = "X" # {
+ then
+ rm -f ${LSOF_TMPC}.*
+ echo "#include <sys/vnode.h>" > ${LSOF_TMPC}.c
+ echo "main(){" >> ${LSOF_TMPC}.c
+ echo "enum vtype p=VSOCK;}" >> ${LSOF_TMPC}.c
+ echo "Testing vnode.h for VSOCK, using $LSOF_CC"
+ echo $LSOF_CC | grep gcc > /dev/null
+ if test $? -eq 0 # {
+ then
+ $LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ else
+ $LSOF_CC ${LSOF_TMPC}.c -I$LSOF_INCLUDE -o ${LSOF_TMPC}.x > /dev/null 2>&1
+ fi # }
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_VSOCK"
+ fi # }
+ rm -f ${LSOF_TMPC}.*
+ else
+ if test "X$SOLARIS_VSOCK" = "XY" -o "X$SOLARIS_VSOCK" = "Xy" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_VSOCK"
+ fi # }
+ fi # }
+
+ # Test for AFS.
+
+ if test -r ${AFS_VICE}/etc/ThisCell # {
+ then
+ if test "X$LSOF_SCRIPT_CALL" = "Xno" # {
+ then
+ if test -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ else
+ if test ! -x ./AFSConfig # {
+ then
+ echo "Can't find or execute the AFSConfig script"
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ ./AFSConfig
+ if test $? -eq 0 -a -r ./AFSHeaders -a -r ./AFSVersion # {
+ then
+ LSOF_AFS="yes"
+ fi # }
+ fi # }
+ if test "X$LSOF_AFS" = "Xyes" # {
+ then
+ if test "X$SUN_AFSAPATHDEF" = "X" # {
+ then
+ ls /usr/vice/etc/modload/libafs > /dev/null 2>&1
+ if test $? -ne 0 # {
+ then
+ LSOF_TMP1=`ls /usr/vice/etc/modload/libafs* 2>/dev/null | wc -l`
+ if test $LSOF_TMP1 -ne 0 # {
+ then
+ SUN_AFSAPATHDEF=`ls -t /usr/vice/etc/modload/libafs* | head -1`
+ fi # }
+ fi # }
+ fi # }
+ if test "X$SUN_AFSAPATHDEF" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DAFSAPATHDEF=\\\"$SUN_AFSAPATHDEF\\\""
+ fi # }
+ LSOF_AFSV=`cat ./AFSVersion | sed 's/^\([0-9]*\)\.\([0-9]*\).*/\1 \2/' | awk '{printf "%d%02d\n",\$1,\$2}'`
+ LSOF_CFGF="$LSOF_CFGF -DHAS_AFS=$LSOF_AFSV"
+ LSOF_DINC="$LSOF_DINC -I`cat ./AFSHeaders`"
+ fi # }
+ fi # }
+
+ # Test for VxFS.
+ #
+ # If the location of the VxFS header files hasn't been defined in the
+ # environment, establish their likely locations.
+
+ LSOF_TMP2=$SOLARIS_VXFSINCL
+ if test -d /opt/VRTS/include # {
+ then
+ LSOF_TMP2="$LSOF_TMP2 /opt/VRTS/include"
+ fi # }
+ if test -d /opt/VRTSvxfs/include # {
+ then
+ LSOF_TMP2="$LSOF_TMP2 /opt/VRTSvxfs/include"
+ fi # }
+ LSOF_TMP1=0
+ for i in $LSOF_TMP2 # {
+ do
+ if test -r ${i}/vxfsutil.h # {
+ then
+ LSOF_TMP1=1
+ SOLARIS_VXFSINCL=$i
+ break
+ fi # }
+ done # }
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+
+ # The VxFS header files are for VxFS version 3.4 or above. Enable VxFS
+ # for those versions.
+
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS -DHASVXFSUTIL -I$SOLARIS_VXFSINCL"
+
+ # Determine which libvxfsutil.a is required -- 32 or 64 bit.
+
+ LSOF_TMP2="" # assume 32 bit
+ echo "X$LSOF_CINFO" | grep "^X64" > /dev/null 2>&1
+ if test $? -eq 0-a "X$SOLARIS_INSTR" != "X" # {
+ then
+ case $SOLARIS_INSTR in # {
+ amd64*)
+ LSOF_TMP2="/amd64"
+ ;;
+ sparcv9*)
+ LSOF_TMP2="/sparcv9"
+ ;;
+ esac # }
+ fi # }
+
+ # See if the correct library has been specified and exists.
+
+ if test "X$SOLARIS_VXFSLIB" = "X" # {
+ then
+ SOLARIS_VXFSLIB=`dirname $SOLARIS_VXFSINCL`/lib
+ fi # }
+ LSOF_TMP3="${SOLARIS_VXFSLIB}${LSOF_TMP2}/libvxfsutil.a"
+ if test ! -r $LSOF_TMP3 # {
+ then
+ echo "!!!FATAL: no VxFS $LSOF_TMP3"
+ exit 1
+ fi # }
+ LSOF_CFGL="-L$SOLARIS_VXFSLIB${LSOF_TMP2} -lvxfsutil -ldl"
+
+ # See if the library has the Reverse Name Lookup (RNL) function.
+
+ nm $LSOF_TMP3 | grep vxfs_inotopath > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFSRNL -DHASVXFSDNLC"
+ fi # }
+ else
+
+ # See if there are VxFS header files for VxFS versions below 3.4.
+
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_inode.h # {
+ then
+
+ # Define VxFS for VxFS versions below 3.4. Make additional header
+ # file tests.
+
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS"
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_fs.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_FS_H"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_sol.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_SOL_H"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_machdep.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_MACHDEP_H"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_solaris.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_SOLARIS_H"
+ grep "off32_t;" ${LSOF_INCLUDE}/sys/fs/vx_machdep.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_OFF32_T"
+ fi # }
+ grep "off64_t;" ${LSOF_INCLUDE}/sys/fs/vx_solaris.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_OFF64_T"
+ fi # }
+ grep "vx_u64_t;" ${LSOF_INCLUDE}/sys/fs/vx_solaris.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_U64_T"
+ fi # }
+ fi # }
+ egrep "struct[ ]vx_inode[ ]\{" ${LSOF_INCLUDE}/sys/fs/vx_inode.h > /dev/null
+ # } (dummy '}' to match '{' in above egrep)
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS_VX_INODE"
+ fi # }
+ fi # }
+ fi # }
+
+ # Set libraries and dialect subdirectory.
+
+ LSOF_CFGL="$LSOF_CFGL -lkvm -lelf -lsocket -lnsl"
+ LSOF_DIALECT_DIR=sun
+
+ # Set local-specific stuff.
+
+ if test "X$LSOF_LOCALSUFFIX" = "XLOCAL" # {
+ then
+ LSOF_DOC="\${DESTDIR}/usr/local/man"
+ fi # }
+ ;;
+
+# Configure for SCO|Caldera OpenServer Release 6.0.0 and UnixWare.
+
+ osr6|unixware|uw)
+ LSOF_TMP1=$LSOF_TGT
+ LSOF_TGT="uw"
+ LSOF_RANLIB=""
+ if test "X$LSOF_VSTR" = "X" # {
+ then
+ LSOF_VSTR=`uname -v`
+ fi # }
+ if test "X$LSOF_VERS" = "X" # {
+ then
+
+ # If the Openserver Release 6.0.0 or UnixWare version isn't pre-defined,
+ # determine it.
+
+ LSOF_VERS=`echo $LSOF_VSTR | sed 's/\([0-9\.]*\).*/\1/; s/\./ /g' | awk '{printf "%d%02d%02d\n", $1, $2, $3;}'`
+ fi # }
+ if test $LSOF_TMP1 = "osr6" # {
+ then
+ LSOF_CINFO="OSR6 support via UnixWare sources"
+
+ # Convert the OpenServer Release 6.0.0 version number to a UnixWare one.
+
+ case $LSOF_VERS in # {
+ 60000)
+ LSOF_VERS=70104
+ ;;
+ *)
+ echo "Unknown OpenServer Release version: $LSOF_VERS"
+ rm -f $LSOF_HLP
+ exit 1
+ esac # }
+ fi # }
+ LSOF_CFGF="-DUNIXWAREV=$LSOF_VERS"
+
+ # Do OpenServer Release 6.0.0 and UnixWare version-specific stuff.
+
+ case $LSOF_VERS in # {
+ 20100|20101|20102|20103)
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_inode.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS"
+ fi # }
+ LSOF_CFGL="-lsocket -lnsl -lelf -lgen"
+ ;;
+ 70000|70001|70100|70101|70103|70104)
+ LSOF_TSTBIGF=" "
+ LSOF_TSTLFLG="-lsocket -lnsl"
+ if test $LSOF_VERS -lt 70103 # {
+ then
+ LSOF_DINC="$LSOF_DINC -I`pwd`/dialects/uw/uw7"
+ else # $LSOF_VERS -ge 70103
+
+ # Process 7.1.3 and above.
+
+ if test -r ${LSOF_INCLUDE}/netinet/in6.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASIPv6"
+ fi # }
+ if test $LSOF_VERS -ge 70104 # {
+ then
+
+ # Process 7.1.4 and above.
+
+ LSOF_TMP1=0
+ if test -r ${LSOF_INCLUDE}/netinet/in_pcb.h # {
+ then
+ grep INKERNEL ${LSOF_INCLUDE}/netinet/in_pcb.h > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test $LSOF_TMP1 -eq 0 -a -r ${LSOF_INCLUDE}/netinet/tcp_var.h # {
+ then
+ grep INKERNEL ${LSOF_INCLUDE}/netinet/tcp_var.h > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP1=1
+ fi # }
+ fi # }
+ if test $LSOF_TMP1 -eq 1 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_INKERNEL"
+ fi # }
+ fi # }
+ fi # }
+ if test ! -r ${LSOF_INCLUDE}/vm/swap.h -a -r ${LSOF_INCLUDE}/sys/swap.h # {
+ then
+ (cd ./dialects/uw/uw7/vm; rm -f swap.h; ln -s ${LSOF_INCLUDE}/sys/swap.h swap.h)
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/vx_gemini.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASVXFS"
+ fi # }
+ LSOF_CFGL="-lsocket -lnsl -lelf -lgen"
+ /bin/pkginfo 2> /dev/null | grep -i patch | grep -i ptf7038 > /dev/null
+ if test -r ${LSOF_INCLUDE}/sys/file.h # {
+ then
+ grep f_open ${LSOF_INCLUDE}/sys/file.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_F_OPEN"
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/cdfs_fs.h # {
+ then
+ grep "cdfs_LogSecShift;" ${LSOF_INCLUDE}/sys/fs/cdfs_fs.h > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ LSOF_TMP=`grep "cdfs_LogSecShift;" ${LSOF_INCLUDE}/sys/fs/cdfs_fs.h | sed 's/^[ ]*\([^ ]*\).*/\1/'`
+ if test "X$LSOF_TMP" != "X" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DTYPELOGSECSHIFT=$LSOF_TMP"
+ fi # }
+ fi # }
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/proc.h # {
+ then
+ grep p_pgid ${LSOF_INCLUDE}/sys/proc.h > /dev/null
+ if test $? -eq 0 # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_P_PGID"
+ fi # }
+ fi # }
+ if test $LSOF_VERS -ge 70101 # {
+ then
+
+ # Do OpenServer Release 6.0.0 and UnixWare 7.1.1 and above tests, as
+ # required.
+
+ if test "X$UW_HAS_NSC" = "X" # {
+ then
+ UW_HAS_NSC=N
+ if test -x /bin/node_self # {
+ then
+ /bin/node_self > /dev/null 2>&1
+ if test $? -eq 0 # {
+ then
+ UW_HAS_NSC=Y
+ fi # }
+ fi # }
+ fi # }
+ if test "X$UW_HAS_NSC" = "XY" -o "X$UW_HAS_NSC" = "Xy" # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_UW_NSC"
+ LSOF_CFGL="$LSOF_CFGL -lcluster"
+ fi # }
+ if test -r ${LSOF_INCLUDE}/sys/nsc_synch.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_UW_CFS"
+ fi # }
+ fi # }
+ ;;
+ *)
+ echo Unsupported UnixWare version: `uname -v`
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+ esac # }
+ if test -r ${LSOF_INCLUDE}/sys/fs/xnamnode.h # {
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHASXNAMNODE"
+ fi # }
+ LSOF_DIALECT_DIR=uw
+ ;;
+
+# Handle unknown abbreviation.
+
+ *)
+ echo "Can't configure for $LSOF_TGT."
+ cat $LSOF_HLP
+ rm -f $LSOF_HLP
+ exit 1
+ ;;
+
+# End of LSOF_TGT cases
+
+esac # }
+
+# Do an inventory of the distribution, as required.
+
+if test "X$LSOF_SCRIPT_CALL" = "Xyes" -a ! -r ./.neverInv # {
+then
+ if test ! -f ./Inventory # Want -x, but Ultrix doesn't grok it. # {
+ then
+ echo "Can't find Inventory script."
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+ ./Inventory
+fi # }
+
+# Make sure target directory exists.
+
+if test ! -d ./dialects/$LSOF_DIALECT_DIR # {
+then
+ echo "Can't configure for $LSOF_TGT -- ./dialects/$LSOF_DIALECT_DIR doesn't exist."
+ rm -f $LSOF_HLP
+ exit 1
+fi # }
+
+# Make sure $LSOF_MK exists in the target directory.
+
+if test ! -r ./dialects/$LSOF_DIALECT_DIR/$LSOF_MK # {
+then
+ echo "Can't configure for $LSOF_TGT -- ./dialects/$LSOF_DIALECT_DIR/$LSOF_MK doesn't exist."
+ rm -f $LSOF_HLP
+ exit 1
+fi # }
+
+# Make sure $LSOF_MKF, $LSOF_SPMKF, or $LSOF_MKF.$LSOF_LOCALSUFFIX) exists
+# in the target directory.
+
+if test "X$LSOF_SPMKF" != "X" # {
+then
+ LSOF_TMP1=$LSOF_SPMKF
+else
+ LSOF_TMP1=$LSOF_MKF
+fi # }
+if test "X$LSOF_LOCALSUFFIX" != "X" # {
+then
+ LSOF_REST=$LSOF_TMP1.$LSOF_LOCALSUFFIX
+else
+ LSOF_REST=$LSOF_TMP1
+fi # }
+if test ! -r ./dialects/$LSOF_DIALECT_DIR/$LSOF_REST # {
+then
+ echo "Can't configure for $LSOF_TGT -- ./dialects/$LSOF_DIALECT_DIR/$LSOF_REST doesn't exist."
+ rm -f $LSOF_HLP
+ exit 1
+fi # }
+
+# If this is FreeBSD, make sure $LSOF_FBSD_ZFS_MKF exists.
+
+if test $LSOF_FBSD_ZFS -eq 1 # {
+then
+ if test ! ./dialects/$LSOF_DIALECT_DIR/$LSOF_FBSD_ZFS_MKF # {
+ then
+ echo "Can't configure for $LSOF_TGT -- ./dialects/$LSOF_DIALECT_DIR/$LSOF_FBSD_ZFS_MKF doesn't exist."
+ rm -f $LSOF_HLP
+ exit 1
+ fi # }
+fi # }}
+
+# Make sure $LSOF_VF exists. Extract the version number from it.
+
+if test ! -r $LSOF_VF # {
+then
+ echo "Version number file, ./$LSOF_VF, doesn't exist."
+ rm -f $LSOF_HLP
+ exit 1
+else
+ LSOF_VN=`sed "s/.ds VN \(.*\)/\1/" < $LSOF_VF`
+fi # }
+
+# Clean up in advance.
+
+rm -f $LSOF_F $LSOF_MKFC $LSOF_FBSD_ZFS_MKF $LSOF_TSTCFLG $LSOF_TSTCC
+rm -f $LSOF_TSTXOC $LSOF_TSTLFF
+echo "rm -f $LSOF_F $LSOF_MKFC $LSOF_FBSD_ZFS_MKF $LSOF_TSTCFLG"
+echo "rm -f $LSOF_TSTCC $LSOF_TSTXOC $LSOF_TSTLFF"
+
+# Make sure there's a C compiler name.
+
+if test "X$LSOF_CC" = "X" # {
+then
+ LSOF_CC=cc
+fi # }
+
+# Do common feature analyses.
+
+# Check for localtime(3) and strftime(3).
+
+rm -f ${LSOF_TMPC}.*
+cat > $LSOF_TMPC.c << .LSOF_END_HERE_DOC2
+#include <time.h>
+main(){
+ time_t cl;
+ struct tm *ts;
+ char bf[32];
+ if ((cl = time(NULL)) == (time_t)-1)
+ return(1);
+ ts = localtime(&cl);
+ if (strftime(bf, sizeof(bf), "%D", ts) != 8)
+ return(1);
+ if ((bf[2] != '/') || (bf[5] != '/'))
+ return (1);
+ return(0);
+}
+.LSOF_END_HERE_DOC2
+echo $EO "Testing C library for localtime() and strftime(), using $LSOF_CC ... $EC"
+$LSOF_CC ${LSOF_TMPC}.c -o ${LSOF_TMPC}.x > /dev/null 2>&1
+if test -x ${LSOF_TMPC}.x # {
+then
+ ./${LSOF_TMPC}.x
+ if test $? -eq 0 # }
+ then
+ LSOF_CFGF="$LSOF_CFGF -DHAS_STRFTIME"
+ echo "present"
+ else
+ echo "unusable"
+ fi # }
+else
+ echo "missing"
+fi # }
+rm -f ${LSOF_TMPC}.[cox]
+
+# Make the dialect sources.
+
+if test "X$LSOF_MKC" = "X" # {
+then
+ LSOF_MKC="ln -s"
+fi # }
+LSOF_MKC=$LSOF_MKC ./dialects/$LSOF_DIALECT_DIR/$LSOF_MK $LSOF_TGT $LSOF_VERS
+
+# Make $LSOF_MKFC and ${LSOF_LIB}/$LSOF_LIBMKF.
+
+echo "# $LSOF_TGT Makefile for lsof revision $LSOF_VN" > $LSOF_MKFC
+echo "" >> $LSOF_MKFC
+echo "CC= $LSOF_CC" >> $LSOF_MKFC
+if test "X$LSOF_CCV" != "X" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "CCV= $LSOF_CCV" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_LIB_NO" = "X" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "LIB= ${LSOF_LIB}/liblsof.a" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_LD" != "X" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "LD= $LSOF_LD" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_CINFO" != "X" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "CINFO= $LSOF_CINFO" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_CFGD" != "X" # {
+then
+ echo "CFGD= $LSOF_CFGD" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_CFGDN" != "X" # {
+then
+ echo "CFGDN= $LSOF_CFGDN" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_ARCH" != "X" # {
+then
+ LSOF_CFGF="$LSOF_CFGF -DLSOF_ARCH=\\\"$LSOF_ARCH\\\""
+fi # }
+if test "X$LSOF_VSTR" != "X" # {
+then
+ LSOF_TMP=`echo $LSOF_VSTR | sed 's/(/\\\\(/g' | sed 's/)/\\\\)/g'`
+ LSOF_CFGF="$LSOF_CFGF -DLSOF_VSTR=\\\"$LSOF_TMP\\\""
+fi # }
+echo "" >> $LSOF_MKFC
+echo "CFGF= $LSOF_CFGF" >> $LSOF_MKFC
+if test "X$LSOF_LIB_NO" = "X" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "CFGL= $LSOF_FCFGL -L./$LSOF_LIB -llsof $LSOF_CFGL" >> $LSOF_MKFC
+fi # }
+echo "" >> $LSOF_MKFC
+if test "X$LSOF_DEBUG" = "X" # {
+then
+ LSOF_DEBUG="-O"
+else
+ if test "X$LSOF_DEBUG" = "XNo-O" # {
+ then
+ LSOF_DEBUG=""
+ fi # }
+fi # }
+echo "DEBUG= $LSOF_DEBUG" >> $LSOF_MKFC
+if test "X$LSOF_DINC" != "X" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "DINC= $LSOF_DINC" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_DOC" != "X" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "DOC=$LSOF_DOC" >> $LSOF_MKFC
+fi # }
+if test "X$LSOF_DISTRIBKVM" != "X" -a "X$LSOF_DISTRIBKVM" != "XKVM" # {
+then
+ echo "" >> $LSOF_MKFC
+ echo "KVM= $LSOF_DISTRIBKVM" >> $LSOF_MKFC
+fi # }
+rm -f ${LSOF_LIB}/$LSOF_LIBMKF
+if test "X$LSOF_LIB_NO" = "X" # {
+then
+ cp $LSOF_MKFC ${LSOF_LIB}/$LSOF_LIBMKF
+fi # }
+cat ./dialects/$LSOF_DIALECT_DIR/$LSOF_REST >> $LSOF_MKFC
+if test "X$LSOF_LIB_NO" = "X" # {
+then
+
+ # Put archiving and optional randomizing strings in ${LSOF_LIB}/$LSOF_LIBMKF.
+ #
+ # Process optional CFLAGS override.
+ #
+ # Add the library Makefile skeleton section.
+
+ echo "" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ if test "X$LSOF_AR" = "X" # {
+ then
+ echo "AR= ar cr \${LIB} \${OBJ}" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ else
+ echo "AR= $LSOF_AR \${LIB} \${OBJ}" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ fi # }
+ if test "X$LSOF_RANLIB" != "X" # {
+ then
+ echo "" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ echo "RANLIB= $LSOF_RANLIB" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ fi # }
+ echo "" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ if test "X$LSOF_CFLAGS_OVERRIDE" = "X" # {
+ then
+ echo "CFLAGS= \${CDEFS} \${INCL} \${DEBUG}" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ else
+ echo "override CFLAGS=\${CDEFS} \${INCL} \${DEBUG}" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ fi # }
+ echo "" >> ${LSOF_LIB}/$LSOF_LIBMKF
+ cat ${LSOF_LIB}/$LSOF_LIBMKFSKEL >> ${LSOF_LIB}/$LSOF_LIBMKF
+ echo $LSOF_MKFC and ${LSOF_LIB}/$LSOF_LIBMKF created.
+else
+ echo $LSOF_MKFC created.
+fi # }
+
+# If this is FreeBSD, create $LSOF_FBSD_ZFS_MKF.
+
+if test $LSOF_FBSD_ZFS -eq 1 # {
+then
+ rm -f $LSOF_FBSD_ZFS_MKF
+ echo "# $LSOF_TGT ZFS Makefile for lsof revision $LSOF_VN" > $LSOF_FBSD_ZFS_MKF
+ echo "" >> $LSOF_FBSD_ZFS_MKF
+ echo "CC= $LSOF_CC" >> $LSOF_FBSD_ZFS_MKF
+ echo "" >> $LSOF_FBSD_ZFS_MKF
+ echo "CFLAGS= $LSOF_FBSD_ZFS_CFGF" >> $LSOF_FBSD_ZFS_MKF
+ echo "" >> $LSOF_FBSD_ZFS_MKF
+ if test "X$LSOF_DEBUG" = "X" # {
+ then
+ LSOF_DEBUG="-O"
+ else
+ if test "X$LSOF_DEBUG" = "XNo-O" # {
+ then
+ LSOF_DEBUG=""
+ fi # }
+ fi # }
+ echo "DEBUG= $LSOF_DEBUG" >> $LSOF_FBSD_ZFS_MKF
+ echo "" >> $LSOF_FBSD_ZFS_MKF
+ echo "OPENSOLARIS= $LSOF_FBSD_ZFS_SYS" >> $LSOF_FBSD_ZFS_MKF
+ echo "" >> $LSOF_FBSD_ZFS_MKF
+ cat ./dialects/$LSOF_DIALECT_DIR/$LSOF_FBSD_ZFS_MKF >> $LSOF_FBSD_ZFS_MKF
+ echo $LSOF_FBSD_ZFS_MKF created.
+fi # }
+
+# Create test cc file.
+
+echo "$LSOF_CC" > $LSOF_TSTCC
+echo "$LSOF_TSTCC created"
+
+# Create test cflags file.
+
+echo "-DLT_DIAL_$LSOF_TGT" > $LSOF_TSTCFLG
+if test "X$LSOF_TSTBIGF" != "X" # {
+then
+ echo "-DLT_BIGF" >> $LSOF_TSTCFLG
+ if test "X$LSOF_TSTBIGF" != "X " # {
+ then
+ for i in $LSOF_TSTBIGF # {
+ do
+ echo "$i" >> $LSOF_TSTCFLG
+ done # }
+ fi # }
+fi # }
+if test "X$LSOF_TSTDFLG" != "X" # {
+then
+ for i in $LSOF_TSTDFLG # {
+ do
+ echo "$i" >> $LSOF_TSTCFLG
+ done # }
+fi # }
+echo $LSOF_CC | grep gcc > /dev/null 2>&1
+if test $? -eq 0 # {
+then
+ echo "-DLT_GCC" >> $LSOF_TSTCFLG
+else
+ echo "-DLT_CC" >> $LSOF_TSTCFLG
+fi # r}
+if test $LSOF_TSTKMEM -eq 1 # {
+then
+ echo "-DLT_KMEM" >> $LSOF_TSTCFLG
+fi # }
+if test $LSOF_TSTK64 -eq 1 # {
+then
+ echo "-DLT_K64" >> $LSOF_TSTCFLG
+fi # }
+echo "-DLT_VERS=$LSOF_VERS" >> $LSOF_TSTCFLG
+if test $LSOF_TSTVPATH -eq 1 # {
+then
+ echo "-DLT_VPATH" >> $LSOF_TSTCFLG
+fi # }
+echo "$LSOF_TSTCFLG created"
+
+# Create tests loader flags file.
+
+echo $LSOF_TSTLFLG > $LSOF_TSTLFF
+echo "$LSOF_TSTLFF created"
+
+# Create test extra objects file.
+
+echo "$LSOF_TSTXO" > $LSOF_TSTXOC
+echo "$LSOF_TSTXOC created"
+
+rm -f $LSOF_HLP
+
+# Call Customize, as required.
+
+if test "X$LSOF_SCRIPT_CALL" = "Xyes" -a ! -r ./.neverCust # {
+then
+ if test ! -f ./Customize # { Want -x, but Ultrix doesn't grok it.
+ then
+ echo "Can't find Customize script."
+ exit 1
+ fi # }
+ ./Customize $LSOF_DIALECT_DIR
+fi # }
+
+# Issue unsupported warning, as appropriate.
+
+if test "X$LSOF_UNSUP" != "X" # {
+then
+ echo "$LSOF_UNSUP"
+fi #}
+exit 0
--- /dev/null
+#!/bin/sh
+#
+# $Id: Customize,v 1.9 2005/05/11 13:02:18 abe Exp $
+#
+# Customize: customize dialect's machine.h header file.
+#
+# Allows easy modification of some important compile-time definitions for
+# lsof, made in the dialect's machine.h header file, including:
+#
+# HASSECURITY the security option
+# HASNOSOCKSECURITY
+# the socket oberalization of HASSECURITY
+# HASDCACHE enabling/disabling the device cache file
+# (Note: changing the device cache file option isn't
+# offered when machine.h contains NEVER_HASDCACHE
+# anywhere, including in a comment.)
+# HASENVDC enabling/disabling device cache path from environment
+# HASKERNIDCK enabling/disabling the kernel identity check
+# (not done for some dialects)
+# HASPERSDC enabling/disabling personal device cache path
+# construction
+# HASPERSDCPATH enabling/disabling additional personal device cache
+# path component
+# HASSYSDC enabling/disabling system-wide device cache file path
+# HASXOPT_ROOT enabling/disabling root use of the -X option
+# WARNDEVACCESS enabling inaccessible /dev node warnings
+# (Note: changing the inaccessible /dev/node warning
+# option isn't offered when machine.h contains
+# NEVER_WARNDEVACCESS anywhere, including in a
+# comment.)
+# WARNINGSTATE enable/disabling default warning message state
+#
+# Usage: Customize [dialect_directory]
+#
+# where: dialect_directory (optional) is the directory in which the dialect's
+# dialect's sources, Makefile and scripts are found
+
+OLD=machine.h
+NEW=new_machine.h
+
+# Save optional dialect directory.
+
+if test $# -eq 1
+then
+ DialDir=$1
+else
+ DialDir=""
+fi
+
+# Establish trap and stty handling.
+
+ISIG=":"
+trap 'rm -f $NEW; $ISIG; exit 1' 1 2 3 15
+stty -a 2>&1 | grep isig > /dev/null
+if test $? -eq 0
+then
+ stty -a 2>&1 | egrep -e -isig > /dev/null
+ if test $? -eq 0
+ then
+ ISIG="stty -isig"
+ stty isig
+ fi
+fi
+
+# Decide how to use echo.
+
+ECHO=`echo -n ""`
+if test "X$ECHO" = "X-n "
+then
+ EC="\c"
+ EO=""
+else
+ EC=""
+ EO="-n"
+fi
+
+# Decide how to use tail(1).
+
+TMP1=`tail -n 1 $0 2> /dev/null`
+if test $? -eq 0 -a "X$TMP1" = "X#LAST_LINE"
+then
+ TA="-n 1"
+else
+ TA="-1"
+fi
+
+# Display the introduction and basic explanation.
+
+cat << .CAT_MARK
+
+You may now customize the machine.h header file for this UNIX
+dialect. The customizations will take effect when you compile
+lsof. You may also choose to skip customization and proceed to
+the compilation of lsof.
+
+If you don't know if you need to customize or want to know more
+about what you can customize, consult the 00DCACHE, 00FAQ, 00PORTING,
+and 00README files of the lsof distribution. You might also find
+it helpful to examine the machine.h header file for the dialect
+you're customizing.
+
+You don't need to use this procedure to customize lsof; you can
+edit the machine.h header file directly. If you later decide you
+want to use this procedure to customize machine.h, execute the
+./Customize script.
+.CAT_MARK
+
+END=0
+while test $END -eq 0
+do
+ echo ""
+ echo $EO "Do you want to customize (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ exit 0
+ fi
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "Customizing ..."
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [y]."
+ fi
+done
+
+# See if $OLD exists.
+
+if test ! -r $OLD
+then
+ echo ""
+ echo "FATAL: The file \"$OLD\" doesn't exist. Customization can't"
+ echo "continue without it."
+ echo ""
+ echo "Did you run the Configure script?"
+ echo ""
+ echo "Customize quits."
+ echo ""
+ exit 1
+fi
+
+# See if $NEW exists.
+
+if test -r $NEW
+then
+ echo ""
+ echo "====================================================================="
+ echo ""
+ echo "WARNING: \"$NEW\" exists. Customization will replace it."
+ END=0
+ while test $END -eq 0
+ do
+ echo ""
+ echo $EO "Do you want to remove $NEW (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "Removing $NEW"
+ echo ""
+ rm -f $NEW
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ echo ""
+ echo "FATAL: Customize quits; it must be able to create \"$NEW\"."
+ echo ""
+ exit 1
+ else
+ echo ""
+ echo "Please answer y|n [y]."
+ fi
+ fi
+ done
+fi
+
+# Process HASSECURITY.
+
+cat << .CAT_MARK
+
+=====================================================================
+
+When HASSECURITY is enabled, only the root user may use lsof to
+examine all open files; other users may examine only the files
+belonging to the real user ID of their lsof process. If
+HASNOSOCKSECURITY is also defined, anyone may list anyone else's
+open socket files, provided their listing is selected with the "-i"
+option.
+
+When HASSECURITY is disabled, anyone may use lsof to examine all
+open files.
+
+.CAT_MARK
+
+grep HASSECURITY $OLD | tail $TA | egrep "^#define" > /dev/null
+if test $? -eq 0
+then
+ echo "HASSECURITY is enabled."
+ NSEC=1
+else
+ echo "HASSECURITY is disabled."
+ NSEC=0
+fi
+END=0
+while test $END -eq 0
+do
+ echo ""
+ if test $NSEC -eq 1
+ then
+ echo $EO "Disable HASSECURITY (y|n) [n]? $EC"
+ else
+ echo $EO "Enable HASSECURITY (y|n) [n]? $EC"
+ fi
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ echo ""
+ if test $NSEC -eq 1
+ then
+ NSEC=0
+ echo "HASSECURITY will be disabled."
+ else
+ NSEC=1
+ echo "HASSECURITY will be enabled."
+ fi
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASSECURITY will not be changed."
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+done
+
+# If HASSECURITY is enabled, see if HASNOSOCKSECURITY should also be defined.
+
+if test $NSEC -eq 1
+then
+ cat << .CAT_MARK
+
+====================================================================
+
+When HASSECURITY is enabled, you may also define HASNOSOCKSECURITY.
+
+When both are defined, no one but root may list all of anyone else's
+open files -- only their own open files -- but anyone may list
+anyone else's open socket files.
+
+This option is useful with ntop (http://www.ntop.org).
+
+.CAT_MARK
+
+ grep HASNOSOCKSECURITY $OLD | tail $TA | egrep "^#define" > /dev/null
+ if test $? -eq 0
+ then
+ echo "HASNOSOCKSECURITY is enabled."
+ SOCKSEC=1
+ else
+ echo "HASNOSOCKSECURITY is disabled."
+ SOCKSEC=0
+ fi
+ END=0
+ while test $END -eq 0
+ do
+ echo ""
+ if test $SOCKSEC -eq 1
+ then
+ echo $EO "Disable HASNOSOCKSECURITY (y|n) [n]? $EC"
+ else
+ echo $EO "Enable HASNOSOCKSECURITY (y|n) [n]? $EC"
+ fi
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ echo ""
+ if test $SOCKSEC -eq 1
+ then
+ SOCKSEC=0
+ echo "HASNOSOCKSECURITY will be disabled."
+ else
+ SOCKSEC=1
+ echo "HASNOSOCKSECURITY will be enabled."
+ fi
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASNOSOCKSECURITY will not be changed."
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ done
+else
+ SOCKSEC=0
+fi
+
+# Process WARNINGSTATE.
+
+cat << .CAT_MARK
+
+=====================================================================
+
+When WARNINGSTATE is enabled, lsof will will issue whatever warning
+messages it finds necessary. When WARNINGSTATE is disabled, lsof
+will issue no warning messages. For individual uses of lsof, -w
+disables warning state and +w enables it.
+
+.CAT_MARK
+
+grep WARNINGSTATE $OLD | tail $TA | egrep "^#define" > /dev/null
+if test $? -eq 0
+then
+ echo "WARNINGSTATE is disabled."
+ WST=0
+else
+ echo "WARNINGSTATE is enabled."
+ WST=1
+fi
+END=0
+NWST=$WST
+while test $END -eq 0
+do
+ echo ""
+ if test $NWST -eq 0
+ then
+ echo $EO "Enable WARNINGSTATE? (y|n) [n]? $EC"
+ else
+ echo $EO "Disable WARNINGSTATE? (y|n) [n]? $EC"
+ fi
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ echo ""
+ if test $NWST -eq 0
+ then
+ echo "WARNINGSTATE will be enabled."
+ NWST=1
+ else
+ echo "WARNINGSTATE will be disabled."
+ NWST=0
+ fi
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "WARNINGSTATE will not be changed."
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+done
+
+# Process WARNDEVACCESS, unless the dialect's machine.h header file contains
+# NEVER_WARNDEVACCESS.
+
+grep NEVER_WARNDEVACCESS $OLD > /dev/null
+if test $? -eq 0
+then
+ NEVERWDA=1
+ NWDA=0
+else
+ NEVERWDA=0
+ cat << .CAT_MARK
+
+=====================================================================
+
+When WARNDEVACCESS is enabled, lsof will issue warning messages
+when it can't access nodes in /dev (or /devices), subject to the
+default or explicit (-w) WARNINGSTATE.
+
+When WARNDEVACCESS is disabled, lsof will silently skip nodes in
+/dev (or /devices) that it can't access.
+
+.CAT_MARK
+
+ grep WARNDEVACCESS $OLD | tail $TA | egrep "^#define" > /dev/null
+ if test $? -eq 0
+ then
+ echo "WARNDEVACCESS is enabled."
+ WDA=1
+ else
+ echo "WARNDEVACCESS is disabled."
+ WDA=0
+ fi
+ END=0
+ NWDA=$WDA
+ while test $END -eq 0
+ do
+ echo ""
+ if test $NWDA -eq 1
+ then
+ echo $EO "Disable WARNDEVACCESS (y|n) [n]? $EC"
+ else
+ echo $EO "Enable WARNDEVACCESS (y|n) [n]? $EC"
+ fi
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ echo ""
+ if test $NWDA -eq 1
+ then
+ echo "WARNDEVACCESS will be disabled."
+ NWDA=0
+ else
+ echo "WARNDEVACCESS will be enabled."
+ NWDA=1
+ fi
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "WARNDEVACCESS will not be changed."
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ done
+fi
+
+# Process HASDCACHE, unless the dialect's machine.h header file contains
+# NEVER_HASDCACHE.
+
+ENVV=""
+ENVN=0
+PDCV=""
+PDCN=0
+PDCPV=""
+PDCPN=0
+SDCV=""
+SDCN=0
+grep NEVER_HASDCACHE $OLD > /dev/null
+if test $? -eq 0
+then
+ NEVERDC=1
+ CDC=0
+ DC=0
+ NDC=0
+else
+ NEVERDC=0
+ cat << .CAT_MARK
+
+=====================================================================
+
+When HASDCACHE is enabled, lsof will write a device cache file that
+contains information about the nodes in /dev (or /devices). The
+options HASENVDC, HASPERSDC, HASPERSDCPATH, and HASSYSDC define
+the device cache file path.
+
+When HASDCACHE is disabled, lsof won't write a device cache file.
+
+Consult the 00DCACHE and 00FAQ files of the lsof distribution for
+more information.
+
+.CAT_MARK
+
+ grep HASDCACHE $OLD | tail $TA | egrep "^#define" > /dev/null
+ if test $? -eq 0
+ then
+ echo "HASDCACHE is enabled."
+ DC=1
+ else
+ echo "HASDCACHE is disabled."
+ DC=0
+ fi
+ END=0
+ NDC=$DC
+ while test $END -eq 0
+ do
+ echo ""
+ if test $NDC -eq 1
+ then
+ echo $EO "Disable HASDCACHE (y|n) [n]? $EC"
+ else
+ echo $EO "Enable HASDCACHE (y|n) [n]? $EC"
+ fi
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ echo ""
+ if test $NDC -eq 1
+ then
+ echo "HASDCACHE will be disabled."
+ NDC=0
+ else
+ echo "HASDCACHE will be enabled."
+ NDC=1
+ fi
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASDCACHE will not be changed."
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ done
+
+ # See if other device cache options need to be declared.
+
+ if test $DC -eq 1 -a $NDC -eq 1
+ then
+ cat << .CAT_MARK
+
+=====================================================================
+
+You have decided that HASDCACHE should be defined. There are other
+definitions associated with HASDCACHE that specify options for the
+formation of the device cache file path. You may change them.
+
+Consult the 00DCACHE and 00FAQ files of the lsof distribution for
+more information.
+
+The current path options are:
+
+.CAT_MARK
+
+ grep HASENVDC $OLD | tail $TA | egrep "^#define"
+ egrep "HASPERSDC$|HASPERSDC[ ]" $OLD | tail $TA | egrep "^#define"
+ grep HASPERSDCPATH $OLD | tail $TA | egrep "^#define"
+ grep HASSYSDC $OLD | tail $TA | egrep "^#define"
+ END=0
+ while test $END -eq 0
+ do
+ echo ""
+ echo $EO "Do you want to change path options (y|n) [n]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ CDC=1
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ CDC=0
+ END=1
+ else
+ if test "X$ANS" = "X"
+ then
+ echo ""
+ echo "The path options will not be changed."
+ CDC=0
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ fi
+ done
+ else
+ CDC=0
+ fi
+ if test \( $NDC -eq 1 -a $DC -eq 0 \) -o \( $DC -eq 1 -a $CDC -eq 1 \)
+ then
+ cat << .CAT_MARK
+
+=====================================================================
+
+You may specify for HASENVDC the name of the environment variable
+from which lsof should take the device cache file path for non-root
+users. Press ENTER to use the current value of HASENVDC:
+
+.CAT_MARK
+
+ echo $EO " $EC"
+ TMP1=`grep HASENVDC $OLD | tail $TA | egrep "^#define"`
+ if test "X$TMP1" != "X"
+ then
+ TMP1=`echo "$TMP1" | sed 's/^#define[ ]HASENVDC[ ]"\([^"]*\)".*$/\1/'`
+ echo "$TMP1"
+ else
+ echo "no current HASENVDC value"
+ fi
+ END=0
+ GV=0
+ while test $END -eq 0
+ do
+ echo ""
+ echo $EO "Do you want to define a name for HASENVDC (y|n) [n]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ ENVV=""
+ END=1
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ GV=1
+ END=1
+ else
+ if test "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASENVDC will not be changed."
+ ENVV=$TMP1
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ fi
+ done
+ if test $GV -eq 1
+ then
+ echo ""
+ echo $EO "Please enter the HASENVDC name (no quotes): $EC"
+ read TMP1 EXCESS
+ ENVV=`echo $TMP1 | sed 's/^\"//' | sed 's/\"$//'`
+ if test "X$ENVV" = "X"
+ then
+ ENVN=1
+ fi
+ fi
+ cat << .CAT_MARK
+
+=====================================================================
+
+HASPERSDC is a format that specifies how the personal device cache
+path is constructed. Consult the 00DCACHE and 00FAQ files of the
+lsof distribution for information on the conversions supported in
+HASPERSDC. Press ENTER to use the curent HASPERSDC format:
+
+.CAT_MARK
+
+ echo $EO " $EC"
+ TMP1=`egrep "HASPERSDC$|HASPERSDC[ ]" $OLD | tail $TA | egrep "^#define"`
+ if test "X$TMP1" != "X"
+ then
+ TMP1=`echo "$TMP1" | sed 's/^#define[ ]HASPERSDC[ ]"\([^"]*\)".*$/\1/'`
+ echo "$TMP1"
+ else
+ echo "no current HASPERSDC format"
+ fi
+ END=0
+ GV=0
+ while test $END -eq 0
+ do
+ echo ""
+ echo $EO "Do you want to define a format for HASPERSDC (y|n) [n]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ END=1
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ GV=1
+ END=1
+ else
+ if test "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASPERSDC will not be changed."
+ PDCV=$TMP1
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ fi
+ done
+ if test $GV -eq 1
+ then
+ echo ""
+ echo $EO "Please enter the HASPERSDC format (no quotes): $EC"
+ read TMP1 EXCESS
+ PDCV=`echo $TMP1 | sed 's/^\"//' | sed 's/\"$//'`
+ if test "X$PDCV" = "X"
+ then
+ PDCN=1
+ fi
+ fi
+ cat << .CAT_MARK
+
+=====================================================================
+
+Specify for HASPERSDCPATH the name of the environment variable from
+which lsof should take a path name component to insert at the %p
+conversion in the HASPERSDC format.
+
+Consult the 00FAQ and 00DCACHE files of the lsof distribution for
+more information on HASPERSDCPATH usage.
+
+Press ENTER to use the current value for HASPERSDCPATH:
+
+.CAT_MARK
+
+ echo $EO " $EC"
+ TMP1=`grep HASPERSDCPATH $OLD | tail $TA | egrep "^#define"`
+ if test "X$TMP1" != "X"
+ then
+ TMP1=`echo "$TMP1" | sed 's/^#define[ ]HASPERSDCPATH[ ]"\([^"]*\)".*$/\1/'`
+ echo "$TMP1"
+ else
+ echo "no current HASPERSDCPATH value"
+ fi
+ END=0
+ GV=0
+ while test $END -eq 0
+ do
+ echo ""
+ echo $EO "Do you want to change HASPERSDCPATH (y|n) [n]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASPERSDCPATH will not be changed."
+ PDCPV=$TMP1
+ END=1
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ GV=1
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ done
+ if test $GV -eq 1
+ then
+ echo ""
+ echo $EO "Please enter the HASPERSDCPATH name (no quotes): $EC"
+ read TMP1 EXCESS
+ PDCPV=`echo $TMP1 | sed 's/^\"//' | sed 's/\"$//'`
+ if test "X$PDCPV" = "X"
+ then
+ PDCPN=1
+ fi
+ fi
+ cat << .CAT_MARK
+
+=====================================================================
+
+Specify for HASSYSDC the system-wide device cache file path. Press
+ENTER to use the current HASSYSDC value:
+
+.CAT_MARK
+
+ echo $EO " $EC"
+ TMP1=`grep HASSYSDC $OLD | tail $TA | egrep "^#define"`
+ if test "X$TMP1" != "X"
+ then
+ TMP1=`echo "$TMP1" | sed 's/^#define[ ]HASSYSDC[ ]"\([^"]*\)".*$/\1/'`
+ echo "$TMP1"
+ else
+ echo "no current HASSYSDC value"
+ fi
+ END=0
+ GV=0
+ while test $END -eq 0
+ do
+ echo ""
+ echo $EO "Do you want to define a system-device path (y|n) [n]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ END=1
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ GV=1
+ END=1
+ else
+ if test "X$ANS" = "X"
+ then
+ echo ""
+ echo "No HASSYSDC change will be made."
+ SDCV=$TMP1
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ fi
+ done
+ if test $GV -eq 1
+ then
+ echo ""
+ echo $EO "Please enter the system-wide path (no quotes): $EC"
+ read TMP1 EXCESS
+ SDCV=`echo $TMP1 | sed 's/^\"//' | sed 's/\"$//'`
+ if test "X$SDCV" = "X"
+ then
+ SDCN=1
+ fi
+ fi
+ fi
+fi
+
+# If HASXOPT is defined, and HASXOPT_ROOT is mentioned,
+# ask about changing HASXOPT_ROOT.
+
+HXRC=0
+grep HASXOPT $OLD | tail $TA | egrep "^#define" > /dev/null
+if test $? -eq 0
+then
+ grep HASXOPT_ROOT $OLD > /dev/null
+ if test $? -eq 0
+ then
+ cat << .CAT_MARK
+
+=====================================================================
+
+HASXOPT is defined. If the dialect for which you are customizing
+appears in the following list, you may want to change the definition
+of HASXOPT_ROOT to restrict the use of the X option to lsof processes
+whose real user ID is root, or enable use of it by all user IDs.
+
+ AIX the -X option enables the risky operation of letting
+ lsof read library entry structures with readx().
+ If HASXOPT_ROOT is defined, only processes whose
+ real user ID is root will be allowed to use -X.
+ If HASXOPT_ROOT is undefined, any process will be
+ allowed to use -X. Consult the 00FAQ file of the
+ lsof distribution for more information on why
+ readx() may be risky.
+
+.CAT_MARK
+
+ grep HASXOPT_ROOT $OLD | tail $TA | egrep "^#define" > /dev/null
+ if test $? -eq 0
+ then
+ echo "HASXOPT_ROOT is defined."
+ HXR="undefine"
+ HXRS=1
+ else
+ echo "HASXOPT_ROOT is not defined."
+ HXR="define"
+ HXRS=0
+ fi
+ END=0
+ while test $END -eq 0
+ do
+ echo ""
+ echo $EO "Do you want to $HXR HASXOPT_ROOT (y|n) [n]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ HXRA=1
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASXOPT_ROOT will not be changed."
+ HXRA=0
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ done
+ if test $HXRA -eq 1
+ then
+ HXRC=1
+ fi
+ fi
+fi
+
+# Process HASKERNIDCK. Skip processing for selected dialect directories.
+
+case $DialDir in
+ linux/proc)
+ NIDCK=0
+ ;;
+ *)
+ cat << .CAT_MARK
+
+=====================================================================
+
+When HASKERNIDCK is enabled, lsof compares the identity of the
+kernel where it was built to the identity of the kernel where it
+is running. This check can detect an lsof executable inappropriate
+for the system on which it is being run.
+
+The kernel identity check can take considerable time on some UNIX
+dialects -- e.g., AIX -- so there may be occasions when it is
+desirable to disable it, in spite of the increased risk of using
+an inappropriate lsof executable.
+
+.CAT_MARK
+
+ grep HASKERNIDCK $OLD | tail $TA | grep "^#define" > /dev/null
+ if test $? -eq 0
+ then
+ echo "HASKERNIDCK is enabled."
+ IDCK=1
+ else
+ echo "HASKERNIDCK is disabled."
+ IDCK=0
+ fi
+ END=0
+ NIDCK=$IDCK
+ while test $END -eq 0
+ do
+ echo ""
+ if test $NIDCK -eq 1
+ then
+ echo $EO "Disable HASKERNIDCK (y|n) [n]? $EC"
+ else
+ echo $EO "Enable HASKERNIDCK (y|n) [n]? $EC"
+ fi
+ read ANS EXCESS
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ echo ""
+ if test $NIDCK -eq 1
+ then
+ NIDCK=0
+ echo "HASKERNIDCK will be disabled."
+ else
+ NIDCK=1
+ echo "HASKERNIDCK will be enabled."
+ fi
+ END=1
+ else
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ echo ""
+ echo "HASKERNIDCK will not be changed."
+ END=1
+ NIDCK=$IDCK
+ else
+ echo ""
+ echo "Please answer y|n [n]."
+ fi
+ fi
+ done
+ ;;
+esac
+
+# Initialize new machine.h.
+
+rm -f $NEW
+cp $OLD $NEW
+chmod 0644 $NEW
+echo "" >> $NEW
+echo "/*" >> $NEW
+echo $EO " * Added by Customize on $EC" >> $NEW
+date >> $NEW
+echo " */" >> $NEW
+echo "" >> $NEW
+
+# Change HASSECURITY and HASNOSOCKSECURITY, as required.
+
+echo "#undef HASSECURITY" >> $NEW
+echo "#undef HASNOSOCKSECURITY" >> $NEW
+if test $NSEC -eq 1
+then
+ echo "#define HASSECURITY 1" >> $NEW
+ if test $SOCKSEC -eq 1
+ then
+ echo "#define HASNOSOCKSECURITY 1" >> $NEW
+ fi
+fi
+
+# Change WARNDEVACCESS, as required.
+
+if test $NEVERWDA -eq 0
+then
+ echo "#undef WARNDEVACCESS" >> $NEW
+ if test $NWDA -eq 1
+ then
+ echo "#define WARNDEVACCESS 1" >> $NEW
+ fi
+fi
+
+# Change WARNINGSTATE, as required.
+
+echo "#undef WARNINGSTATE" >> $NEW
+if test $NWST -eq 0
+then
+ echo "#define WARNINGSTATE 1" >> $NEW
+fi
+
+# Change device cache definitions, as required.
+
+if test \( $NDC -eq 1 -a $DC -eq 0 \) -o \( $DC -eq 1 -a $CDC -eq 1 \)
+then
+ if test "X$ENVV" = "X" -a "X$PDCV" = "X" -a "X$SDCV" = "X"
+ then
+ cat << .CAT_MARK
+
+FATAL: HASDCACHE is defined, but there is no definition for
+ any of HASENVDC, HASPERSDC, or HASSYSDC
+
+ No new machine.h has been created.
+
+ Customize quits.
+
+ Restart Customize and define at least one of HASENVDC,
+ HASPERSDC, or HASSYSDC.
+
+.CAT_MARK
+
+ rm -f $NEW
+ exit 1
+ fi
+fi
+if test "X$PDCV" != "X"
+then
+ echo "$PDCV" | grep "%p" > /dev/null
+ if test $? -eq 0 -a $PDCPN -eq 1
+ then
+ cat << .CAT_MARK
+
+FATAL: HASDCACHE is defined and HASPERSDC has a %p conversion,
+ but HASPERSDCPATH is NULL.
+
+ No new machine.h has been created.
+
+ Customize quits.
+
+ Restart Customize and define HASPERSDCPATH.
+
+.CAT_MARK
+
+ rm -f $NEW
+ exit 1
+ fi
+fi
+echo "#undef HASDCACHE" >> $NEW
+if test $NEVERDC -eq 1
+then
+ echo "#undef HASENVDC" >> $NEW
+ echo "#undef HASPERSDC" >> $NEW
+ echo "#undef HASPERSDCPATH" >> $NEW
+ echo "#undef HASSYSDC" >> $NEW
+else
+ if test $NDC -eq 1
+ then
+ echo "#define HASDCACHE 1" >> $NEW
+ if test "X$ENVV" != "X" -o $ENVN -eq 1
+ then
+ echo "#undef HASENVDC" >> $NEW
+ if test $ENVN -eq 0
+ then
+ echo "#define HASENVDC \"$ENVV\"" >> $NEW
+ fi
+ fi
+ if test "X$PDCV" != "X" -o $PDCN -eq 1
+ then
+ echo "#undef HASPERSDC" >> $NEW
+ if test $PDCN -eq 0
+ then
+ echo "#define HASPERSDC \"$PDCV\"" >> $NEW
+ fi
+ fi
+ if test "X$PDCPV" != "X" -o $PDCPN -eq 1
+ then
+ echo "#undef HASPERSDCPATH" >> $NEW
+ if test $PDCPN -eq 0
+ then
+ echo "#define HASPERSDCPATH \"$PDCPV\"" >> $NEW
+ fi
+ fi
+ if test "X$SDCV" != "X" -o $SDCN -eq 1
+ then
+ echo "#undef HASSYSDC" >> $NEW
+ if test $SDCN -eq 0
+ then
+ echo "#define HASSYSDC \"$SDCV\"" >> $NEW
+ fi
+ fi
+ fi
+fi
+
+# Change HASXOPT_ROOT, as required.
+
+if test $HXRC -eq 1
+then
+ if test $HXRS -eq 1
+ then
+ echo "#undef HASXOPT_ROOT" >> $NEW
+ else
+ echo "#define HASXOPT_ROOT 1" >> $NEW
+ fi
+fi
+
+# Change HASKERNIDCK, as required.
+
+echo "#undef HASKERNIDCK" >> $NEW
+if test $NIDCK -eq 1
+then
+ echo "#define HASKERNIDCK 1" >> $NEW
+fi
+
+# Replace the current machine.h with the new one, as requested.
+
+echo ""
+echo "====================================================================="
+echo ""
+echo "A new $OLD file has been created in \"$NEW\"."
+END=0
+while test $END -eq 0
+do
+ echo ""
+ echo "Do you want to rename $OLD to ${OLD}.old and replace it with"
+ echo $EO "$NEW (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ END=1
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ rm -f ${OLD}.old
+ mv $OLD ${OLD}.old
+ mv $NEW $OLD
+ END=1
+ else
+ echo ""
+ echo "Please answer y|n [y]."
+ fi
+ fi
+done
+echo ""
+echo "You may now run the make command -- e.g.,"
+echo ""
+echo " $ make"
+echo ""
+exit 0
+#LAST_LINE
--- /dev/null
+#!/bin/sh
+#
+# Inventory -- take an inventory of the lsof distribution's 00MANIFEST
+
+# Establish trap and stty handling.
+
+ISIG=":"
+trap '$ISIG; exit 1' 1 2 3 15
+stty -a 2>&1 | grep isig > /dev/null
+if test $? -eq 0
+then
+ stty -a 2>&1 | egrep -e -isig > /dev/null
+ if test $? -eq 0
+ then
+ ISIG="stty -isig"
+ stty isig
+ fi
+fi
+
+# Establish echo type -- Berkeley or SYSV.
+
+j=`echo -n ""`
+if test "X$j" = "X-n "
+then
+ EC="\c"
+ EO=""
+else
+ EC=""
+ EO="-n"
+fi
+
+# Display the introduction and basic explanation.
+
+cat << .CAT_MARK
+
+This configuration step (the Inventory script) takes inventory of
+the lsof distribution. The script runs for a minute or two while
+it checks that all the subdirectories, information files, scripts,
+header files and source files that should be present really are.
+
+It's not absolutely necessary that you take inventory, but it's a
+good idea to do it right after the lsof distribution has been
+unpacked. Once the inventory has been taken, this script creates
+the file ./.ck00MAN as a signal that the inventory step has been
+done.
+
+You can call the Inventory script directly at any time to take
+inventory. You can inhibit the inventory step permanently by
+creating the file ./.neverInv, and you can tell the Configure script
+to skip the inventory and customization steps with the -n option.
+.CAT_MARK
+
+END=0
+while test $END = 0
+do
+ echo ""
+ echo $EO "Do you want to take inventory (y|n) [y]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ exit 0
+ fi
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY" -o "X$ANS" = "X"
+ then
+ END=1
+ else
+ echo ""
+ echo "Please answer y or n."
+ fi
+done
+
+# The current directory is assumed to be the lsof distribution home.
+
+D=`pwd`
+
+# If .ck00MAN exists, the manifest has already been checked.
+# See if the caller wants to check it again.
+
+CK=$D/.ck00MAN
+if test -r $CK
+then
+ cat << .CAT_MARK
+
+======================================================================
+
+The lsof distribution inventory in 00MANIFEST has already been checked.
+.CAT_MARK
+
+ END=0
+ while test $END = 0
+ do
+ echo ""
+ echo $EO "Do you want to check the inventory again (y|n) [n]? $EC"
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN" -o "X$ANS" = "X"
+ then
+ exit 0
+ else
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ END=1
+ else
+ echo ""
+ echo "Please answer y or n."
+ fi
+ fi
+ done
+fi
+echo ""
+
+# See if manifest exists. Exit if it does not.
+
+if test ! -r 00MANIFEST
+then
+ echo "FATAL: 00MANIFEST file not found or not readable; Inventory exits."
+ echo ""
+ exit 1
+fi
+
+# Start the inventory.
+
+S=""
+echo "Conducting an inventory of the lsof distribution; this will take a while."
+echo ""
+echo $EO "Examining ${D}:$EC"
+ERR=0
+OK=1
+for i in `cat 00MANIFEST | sed 's/\*$//'`
+do
+ if test "X$i" != "X"
+ then
+ j=`expr $i : '\(.*\)/$'`
+ if test "X$j" != "X" -a "X$j" != "X0"
+ then
+
+ # Check a subdirectory reference.
+
+ if test ! -d ${D}/${S}/$j
+ then
+ if test $OK = 1
+ then
+ echo ""
+ fi
+ echo " Subdirectory ${S}/$j is missing. ++++"
+ ERR=1
+ OK=0
+ fi
+ else
+ s=`expr $i : '\(.*\):$'`
+ if test "X$s" != "X" -a "X$s" != "X0"
+ then
+
+ # Process a subdirectory change.
+
+ if test $OK -eq 1
+ then
+ echo " OK"
+ fi
+ OK=1
+ S=$s
+ echo $EO "Examining $S:$EC"
+ if test ! -d ${D}/$S
+ then
+ echo " ERROR"
+ echo " Subdirectory $S is missing. ++++"
+ ERR=1
+ OK=0
+ fi
+ else
+
+ # Process a file reference.
+
+ if test ! -r ${D}/${S}/$i
+ then
+ if test $OK -eq 1
+ then
+ echo " ERROR"
+ fi
+ echo " File ${S}/$i is missing. ++++"
+ ERR=1
+ OK=0
+ fi
+ fi
+ fi
+ fi
+done
+if test $OK -eq 1
+then
+ echo " OK"
+fi
+echo ""
+if test $ERR -ne 0
+then
+ echo "+++++++++++++++++++++++++++++++++++++++++++++++"
+ echo "+ +"
+ echo "+ SOME FILES OR DIRECTORIES MAY BE MISSING! +"
+ echo "+ +"
+ echo "+++++++++++++++++++++++++++++++++++++++++++++++"
+else
+ echo "This lsof distribution seems to be complete."
+fi
+echo ""
+echo "" >> $CK
+exit $ERR
--- /dev/null
+
+ Information About This Lsof Distribution
+
+
+What You Have
+=============
+
+If you got this far without being confused, then you are probably
+familiar with the construction of the lsof distribution or you have
+read RELEASE.SUMMARY_4.82. If either is the case, please skip to
+the Inventory section. If you haven't read RELEASE.SUMMARY_4.82,
+I suggest you do it now, because it explains how the lsof distribution
+is constructed and other useful things about lsof, including a
+summary of changes for the past few lsof revisions.
+
+Even though you may have thought you were getting lsof.tar.bz2,
+lsof.tar.gz or lsof.tar.Z with ftp, you really got lsof_4.82.tar.bz2,
+lsof_4.82.tar.gz or lsof_4.82.tar.Z. That's because the triplet of
+lsof.tar.* files are symbolic links to their longer-named counterparts.
+
+The bzip2'd, gzip'd or compressed tar files with lsof_, followed by a
+number, are wrapper archives, designed to package the lsof source
+archive, this file, other documentation files, and a GPG authentication
+certificate together.
+
+The number, 4.82, is the lsof revision number. When you bunzip2'd,
+gunzip'd or uncompressed lsof_4.82.tar.* and used tar to unpack
+lsof_4.82.tar, you got: 00.README.FIRST_4.82, describing the contents
+of lsof_4.82; README.lsof_4.82; lsof_4.82_src.tar; and
+lsof_4.82_src.tar.sig. All are identified with the revision number.
+You're reading README.lsof_4.82. lsof_4.82_src.tar.sig is a GPG
+certificate that authenticates the lsof source archive,
+lsof_4.82_src.tar.
+
+After you read the Inventory and Security sections, and hopefully
+after you check the GPG certificate, unpack the lsof_4.82_src.tar
+source archive and you will get a sub-directory, named lsof_4.82_src,
+that contains the lsof 4.82 source distribution.
+
+
+Inventory
+=========
+
+Once you have unpacked lsof_4.82_src.tar.tar, you can check
+lsof_4.82_src for completeness by changing to that sub-directory
+and running the Inventory script. The lsof_4.82_src/Configure
+script runs the Inventory script, too. The Configure script also
+calls a customization script, called Customize. You can direct
+Configure to avoid calling Inventory and Customize with the -n
+option.
+
+See the Distribution Contents section of the 00DIST file and The
+Inventory Script section of the 00README file for more information
+on the contents of the lsof distribution, and the Configure,
+Customize and Inventory scripts. The 00DIST and 00README files
+will be found in the lsof_4.82_src sub-directory you just created.
+
+
+Security
+========
+
+The md5 checksum for lsof_4.82_src.tar is:
+
+ MD5 (lsof_4.82_src.tar) = 91efe55e2d5d06292556fb454d42c800
+
+A good source for an MD5 checksum computation tool is the OpenSSL
+project whose work may be found at:
+
+ www.openssl.org
+
+You can use the openssl "dgst" operator to compute an MD5 checksum --
+e.g.,
+
+ $ openssl dgst -md5 lsof_4.82_src
+
+The old-style sum(1) checksum for lsof_4.82_src.tar (Please read
+the next paragraph if you don't get this value.) is:
+
+ 11381 8606 lsof_4.82/lsof_4.82_src.tar
+
+If your dialect's sum(1) program defaults to the new style algorithm
+(e.g., Solaris), you may have to use its -r option (or use the
+Solaris /usr/ucb/sum). If your Unix dialect doesn't have a sum(1)
+program (e.g., FreeBSD, or NetBSD), use its cksum(1) program with
+the -o1 option to get an old-style checksum. You may also need to
+ignore the block count, depending on the block size used on your
+your system (i.e., 512 or 1,024). The sum(1) that produced the
+above checksum considers block size to be 512; in contrast the BSD
+cksum(1) programs' -o1 option considers block size to be 1,024.
+
+lsof_4.82_src.tar.sig is a GPG certificate file, using my public
+key. My key may be available on some public key servers under the
+names:
+
+ Victor A. Abell <abe@cc.purdue.edu>
+ or
+ Victor A. Abell <abe@purdue.edu>
+
+You will also find it at:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/Victor_A_Abell.gpg
+
+Get my key and install it in your public key ring.
+
+Once my key is installed, use this command to check the certificate
+of lsof_4.82_src.tar:
+
+ gpg --verify lsof_4.82_src.tar.sig lsof_4.82_src.tar
+
+If the certificate check isn't good, lsof_4.82_src.tar is suspect.
+Report the problem to me via e-mail at <abe@purdue.edu>.
+
+If you don't have GPG, you can compare the md5 checksum of
+lsof_4.82_src.tar to the value listed in this file. However, that
+is a less reliable authentication method, since it can't detect
+changes to both lsof_4.82_src.tar and the md5 checksum value listed
+in this tile.
+
+Other Security
+==============
+
+Signature information for the distribution file that contains
+this file may be found in the CHECKSUMS file that is located
+where the distribution file was found.
+
+
+Victor A. Abell <abe@purdue.edu>
+Wed Mar 25 15:26:12 EDT 2009
--- /dev/null
+/*
+ * arg.c - common argument processing support functions for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1994 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: arg.c,v 1.49 2009/03/25 19:20:30 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Local definitions
+ */
+
+#define CMDRXINCR 32 /* CmdRx[] allocation increment */
+
+
+/*
+ * Local static variables
+ */
+
+static int NCmdRxA = 0; /* space allocated to CmdRx[] */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static int ckfd_range,(char *first, char *dash, char *last, int *lo, int *hi));
+_PROTOTYPE(static int enter_fd_lst,(char *nm, int lo, int hi, int excl));
+_PROTOTYPE(static int enter_nwad,(struct nwad *n, int sp, int ep, char *s, struct hostent *he));
+_PROTOTYPE(static struct hostent *lkup_hostnm,(char *hn, struct nwad *n));
+_PROTOTYPE(static char *isIPv4addr,(char *hn, unsigned char *a, int al));
+
+
+/*
+ * ckfd_range() - check fd range
+ */
+
+static int
+ckfd_range(first, dash, last, lo, hi)
+ char *first; /* starting character */
+ char *dash; /* '-' location */
+ char *last; /* '\0' location */
+ int *lo; /* returned low value */
+ int *hi; /* returned high value */
+{
+ char *cp;
+/*
+ * See if the range character pointers make sense.
+ */
+ if (first >= dash || dash >= last) {
+ (void) fprintf(stderr, "%s: illegal FD range for -d: ", Pn);
+ safestrprt(first, stderr, 1);
+ return(1);
+ }
+/*
+ * Assemble and check the high and low values.
+ */
+ for (cp = first, *lo = 0; *cp && cp < dash; cp++) {
+ if (!isdigit((unsigned char)*cp)) {
+
+FD_range_nondigit:
+
+ (void) fprintf(stderr, "%s: non-digit in -d FD range: ", Pn);
+ safestrprt(first, stderr, 1);
+ return(1);
+ }
+ *lo = (*lo * 10) + (int)(*cp - '0');
+ }
+ for (cp = dash+1, *hi = 0; *cp && cp < last; cp++) {
+ if (!isdigit((unsigned char)*cp))
+ goto FD_range_nondigit;
+ *hi = (*hi * 10) + (int)(*cp - '0');
+ }
+ if (*lo >= *hi) {
+ (void) fprintf(stderr, "%s: -d FD range's low >= its high: ", Pn);
+ safestrprt(first, stderr, 1);
+ return(1);
+ }
+ return(0);
+}
+
+
+/*
+ * ck_file_arg() - check file arguments
+ */
+
+int
+ck_file_arg(i, ac, av, fv, rs, sbp)
+ int i; /* first file argument index */
+ int ac; /* argument count */
+ char *av[]; /* argument vector */
+ int fv; /* Ffilesys value (real or temporary) */
+ int rs; /* Readlink() status if argument count == 1:
+ * 0 = undone; 1 = done */
+ struct stat *sbp; /* if non-NULL, pointer to stat(2) buffer
+ * when argument count == 1 */
+{
+ char *ap, *fnm, *fsnm, *path;
+ short err = 0;
+ int fsm, ftype, j, k;
+ MALLOC_S l;
+ struct mounts *mp;
+ static struct mounts **mmp = (struct mounts **)NULL;
+ int mx, nm;
+ static int nma = 0;
+ struct stat sb;
+ struct sfile *sfp;
+ short ss = 0;
+
+#if defined(CKFA_EXPDEV)
+ dev_t dev, rdev;
+#endif /* defined(CKFA_EXPDEV) */
+
+#if defined(HASPROCFS)
+ unsigned char ad, an;
+ int pfsnl = -1;
+ pid_t pid;
+ struct procfsid *pfi;
+#endif /* defined(HASPROCFS) */
+
+/*
+ * Loop through arguments.
+ */
+ for (; i < ac; i++) {
+ if (rs && (ac == 1) && (i == 0))
+ path = av[i];
+ else {
+ if (!(path = Readlink(av[i]))) {
+ ErrStat = 1;
+ continue;
+ }
+ }
+ /*
+ * Remove terminating `/' characters from paths longer than one.
+ */
+ j = k = strlen(path);
+ while ((k > 1) && (path[k-1] == '/')) {
+ k--;
+ }
+ if (k < j) {
+ if (path != av[i])
+ path[k] = '\0';
+ else {
+ if (!(ap = (char *)malloc((MALLOC_S)(k + 1)))) {
+ (void) fprintf(stderr, "%s: no space for copy of %s\n",
+ Pn, path);
+ Exit(1);
+ }
+ (void) strncpy(ap, path, k);
+ ap[k] = '\0';
+ path = ap;
+ }
+ }
+ /*
+ * Check for file system argument.
+ */
+ for (ftype = 1, mp = readmnt(), nm = 0;
+ (fv != 1) && mp;
+ mp = mp->next)
+ {
+ fsm = 0;
+ if (strcmp(mp->dir, path) == 0)
+ fsm++;
+ else if (fv == 2 || (mp->fs_mode & S_IFMT) == S_IFBLK) {
+ if (mp->fsnmres && strcmp(mp->fsnmres, path) == 0)
+ fsm++;
+ }
+ if (!fsm)
+ continue;
+ ftype = 0;
+ /*
+ * Skip duplicates.
+ */
+ for (mx = 0; mx < nm; mx++) {
+ if (strcmp(mp->dir, mmp[mx]->dir) == 0
+ && mp->dev == mmp[mx]->dev
+ && mp->rdev == mmp[mx]->rdev
+ && mp->inode == mmp[mx]->inode)
+ break;
+ }
+ if (mx < nm)
+ continue;
+ /*
+ * Allocate space for and save another mount point match and
+ * the type of match -- directory name (mounted) or file system
+ * name (mounted-on).
+ */
+ if (nm >= nma) {
+ nma += 5;
+ l = (MALLOC_S)(nma * sizeof(struct mounts *));
+ if (mmp)
+ mmp = (struct mounts **)realloc((MALLOC_P *)mmp, l);
+ else
+ mmp = (struct mounts **)malloc(l);
+ if (!mmp) {
+ (void) fprintf(stderr,
+ "%s: no space for mount pointers\n", Pn);
+ Exit(1);
+ }
+ }
+ mmp[nm++] = mp;
+ }
+ if (fv == 2 && nm == 0) {
+ (void) fprintf(stderr, "%s: not a file system: ", Pn);
+ safestrprt(av[i], stderr, 1);
+ ErrStat = 1;
+ continue;
+ }
+ /*
+ * Loop through the file system matches. If there were none, make one
+ * pass through the loop, using simply the path name.
+ */
+ mx = 0;
+ do {
+
+ /*
+ * Allocate an sfile structure and fill in the type and link.
+ */
+ if (!(sfp = (struct sfile *)malloc(sizeof(struct sfile)))) {
+ (void) fprintf(stderr, "%s: no space for files\n", Pn);
+ Exit(1);
+ }
+ sfp->next = Sfile;
+ Sfile = sfp;
+ sfp->f = 0;
+ if ((sfp->type = ftype)) {
+
+ /*
+ * For a non-file system path, use the path as the file name
+ * and set a NULL file system name.
+ */
+ fnm = path;
+ fsnm = (char *)NULL;
+ /*
+ * Stat the path to obtain its characteristics.
+ */
+ if (sbp && (ac == 1))
+ sb = *sbp;
+ else {
+ if (statsafely(fnm, &sb) != 0) {
+ int en = errno;
+
+ (void) fprintf(stderr, "%s: status error on ", Pn);
+ safestrprt(fnm, stderr, 0);
+ (void) fprintf(stderr, ": %s\n", strerror(en));
+ Sfile = sfp->next;
+ (void) free((FREE_P *)sfp);
+ ErrStat = 1;
+ continue;
+ }
+
+#if defined(HASSPECDEVD)
+ (void) HASSPECDEVD(fnm, &sb);
+#endif /* defined(HASSPECDEVD) */
+
+ }
+ sfp->i = (INODETYPE)sb.st_ino;
+ sfp->mode = sb.st_mode & S_IFMT;
+
+#if defined(CKFA_EXPDEV)
+ /*
+ * Expand device numbers before saving, so that they match the
+ * already-expanded local mount info table device numbers.
+ * (This is an EP/IX 2.1.1 and above artifact.)
+ */
+ sfp->dev = expdev(sb.st_dev);
+ sfp->rdev = expdev(sb.st_rdev);
+#else /* !defined(CKFA_EXPDEV) */
+ sfp->dev = sb.st_dev;
+ sfp->rdev = sb.st_rdev;
+#endif /* defined(CKFA_EXPDEV) */
+
+#if defined(CKFA_MPXCHAN)
+ /*
+ * Save a (possible) multiplexed channel number. (This is an
+ * AIX artifact.)
+ */
+ sfp->ch = getchan(path);
+#endif /* defined(CKFA_MPXCHAN) */
+
+ } else {
+ mp = mmp[mx++];
+ ss++;
+
+#if defined(HASPROCFS)
+ /*
+ * If this is a /proc file system, set the search flag and
+ * abandon the sfile entry.
+ */
+ if (mp == Mtprocfs) {
+ Sfile = sfp->next;
+ (void) free((FREE_P *)sfp);
+ Procsrch = 1;
+ continue;
+ }
+#endif /* defined(HASPROCFS) */
+
+ /*
+ * Derive file name and file system name for a mount point.
+ *
+ * Save the device numbers, inode number, and modes.
+ */
+ fnm = mp->dir;
+ fsnm = mp->fsname;
+ sfp->dev = mp->dev;
+ sfp->rdev = mp->rdev;
+ sfp->i = mp->inode;
+ sfp->mode = mp->mode & S_IFMT;
+ }
+ ss = 1; /* indicate a "safe" stat() */
+ /*
+ * Store the file name and file system name pointers in the sfile
+ * structure, allocating space as necessary.
+ */
+ if (!fnm || fnm == path) {
+ sfp->name = fnm;
+
+#if defined(HASPROCFS)
+ an = 0;
+#endif /* defined(HASPROCFS) */
+
+ } else {
+ if (!(sfp->name = mkstrcpy(fnm, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for file name: ", Pn);
+ safestrprt(fnm, stderr, 1);
+ Exit(1);
+ }
+
+#if defined(HASPROCFS)
+ an = 1;
+#endif /* defined(HASPROCFS) */
+
+ }
+ if (!fsnm || fsnm == path) {
+ sfp->devnm = fsnm;
+
+#if defined(HASPROCFS)
+ ad = 0;
+#endif /* defined(HASPROCFS) */
+
+ } else {
+ if (!(sfp->devnm = mkstrcpy(fsnm, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for file system name: ", Pn);
+ safestrprt(fsnm, stderr, 1);
+ Exit(1);
+ }
+
+#if defined(HASPROCFS)
+ ad = 1;
+#endif /* defined(HASPROCFS) */
+
+ }
+ if (!(sfp->aname = mkstrcpy(av[i], (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for argument file name: ", Pn);
+ safestrprt(av[i], stderr, 1);
+ Exit(1);
+ }
+
+#if defined(HASPROCFS)
+ /*
+ * See if this is an individual member of a proc file system.
+ */
+ if (!Mtprocfs || Procsrch)
+ continue;
+
+# if defined(HASFSTYPE) && HASFSTYPE==1
+ if (strcmp(sb.st_fstype, HASPROCFS) != 0)
+ continue;
+# endif /* defined(HASFSTYPE) && HASFSTYPE==1 */
+
+ if (pfsnl == -1)
+ pfsnl = strlen(Mtprocfs->dir);
+ if (!pfsnl)
+ continue;
+ if (strncmp(Mtprocfs->dir, path, pfsnl) != 0)
+ continue;
+ if (path[pfsnl] != '/')
+
+# if defined(HASPINODEN)
+ pid = 0;
+# else /* !defined(HASPINODEN) */
+ continue;
+# endif /* defined(HASPINODEN) */
+
+ else {
+ for (j = pfsnl+1; path[j]; j++) {
+ if (!isdigit((unsigned char)path[j]))
+ break;
+ }
+ if (path[j] || (j - pfsnl - 1) < 1
+ || (sfp->mode & S_IFMT) != S_IFREG)
+
+# if defined(HASPINODEN)
+ pid = 0;
+# else /* !defined(HASPINODEN) */
+ continue;
+# endif /* defined(HASPINODEN) */
+
+ else
+ pid = atoi(&path[pfsnl+1]);
+ }
+ if (!(pfi = (struct procfsid *)malloc((MALLOC_S)
+ sizeof(struct procfsid))))
+ {
+ (void) fprintf(stderr, "%s: no space for %s ID: ",
+ Pn, Mtprocfs->dir);
+ safestrprt(path, stderr, 1);
+ Exit(1);
+ }
+ pfi->pid = pid;
+ pfi->f = 0;
+ pfi->nm = sfp->aname;
+ pfi->next = Procfsid;
+ Procfsid = pfi;
+
+# if defined(HASPINODEN)
+ pfi->inode = (INODETYPE)sfp->i;
+# endif /* defined(HASPINODEN) */
+
+ /*
+ * Abandon the Sfile entry, lest it be used in is_file_named().
+ */
+ Sfile = sfp->next;
+ if (ad)
+ (void) free((FREE_P *)sfp->devnm);
+ if (an)
+ (void) free((FREE_P *)sfp->name);
+ (void) free((FREE_P *)sfp);
+#endif /* defined(HASPROCFS) */
+
+ } while (mx < nm);
+ }
+ if (!ss)
+ err = 1;
+ return((int)err);
+}
+
+
+#if defined(HASDCACHE)
+/*
+ * ctrl_dcache() - enter device cache control
+ */
+
+int
+ctrl_dcache(c)
+ char *c; /* control string */
+{
+ int rc = 0;
+
+ if (!c) {
+ (void) fprintf(stderr,
+ "%s: no device cache option control string\n", Pn);
+ return(1);
+ }
+/*
+ * Decode argument function character.
+ */
+ switch (*c) {
+ case '?':
+ if (*(c+1) != '\0') {
+ (void) fprintf(stderr, "%s: nothing should follow -D?\n", Pn);
+ return(1);
+ }
+ DChelp = 1;
+ return(0);
+ case 'b':
+ case 'B':
+ if (Setuidroot
+
+#if !defined(WILLDROPGID)
+ || Myuid
+#endif /* !defined(WILLDROPGID) */
+
+ )
+ rc = 1;
+ else
+ DCstate = 1;
+ break;
+ case 'r':
+ case 'R':
+ if (Setuidroot && *(c+1))
+ rc = 1;
+ else
+ DCstate = 2;
+ break;
+ case 'u':
+ case 'U':
+ if (Setuidroot
+
+#if !defined(WILLDROPGID)
+ || Myuid
+#endif /* !defined(WILLDROPGID) */
+
+ )
+ rc = 1;
+ else
+ DCstate = 3;
+ break;
+ case 'i':
+ case 'I':
+ if (*(c+1) == '\0') {
+ DCstate = 0;
+ return(0);
+ }
+ /* fall through */
+ default:
+ (void) fprintf(stderr, "%s: unknown -D option: ", Pn);
+ safestrprt(c, stderr, 1);
+ return(1);
+ }
+ if (rc) {
+ (void) fprintf(stderr, "%s: -D option restricted to root: ", Pn);
+ safestrprt(c, stderr, 1);
+ return(1);
+ }
+/*
+ * Skip to optional path name and save it.
+ */
+ for (c++; *c && (*c == ' ' || *c == '\t'); c++)
+ ;
+ if (strlen(c)) {
+ if (!(DCpathArg = mkstrcpy(c, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: no space for -D path: ", Pn);
+ safestrprt(c, stderr, 1);
+ Exit(1);
+ }
+ }
+ return(0);
+}
+#endif /* defined(HASDCACHE) */
+
+
+/*
+ * enter_cmd_rx() - enter command regular expression
+ */
+
+int
+enter_cmd_rx(x)
+ char *x; /* regular expression */
+{
+ int bmod = 0;
+ int bxmod = 0;
+ int i, re;
+ int imod = 0;
+ int xmod = 0;
+ int co = REG_NOSUB|REG_EXTENDED;
+ char reb[256], *xb, *xe, *xm;
+ MALLOC_S xl;
+ char *xp = (char *)NULL;
+/*
+ * Make sure the supplied string starts a regular expression.
+ */
+ if (!*x || (*x != '/')) {
+ (void) fprintf(stderr, "%s: regexp doesn't begin with '/': ", Pn);
+ if (x)
+ safestrprt(x, stderr, 1);
+ return(1);
+ }
+/*
+ * Skip to the end ('/') of the regular expression.
+ */
+ xb = x + 1;
+ for (xe = xb; *xe; xe++) {
+ if (*xe == '/')
+ break;
+ }
+ if (*xe != '/') {
+ (void) fprintf(stderr, "%s: regexp doesn't end with '/': ", Pn);
+ safestrprt(x, stderr, 1);
+ return(1);
+ }
+/*
+ * Decode any regular expression modifiers.
+ */
+ for (i = 0, xm = xe + 1; *xm; xm++) {
+ switch(*xm) {
+ case 'b': /* This is a basic expression. */
+ if (++bmod > 1) {
+ if (bmod == 2) {
+ (void) fprintf(stderr,
+ "%s: b regexp modifier already used: ", Pn);
+ safestrprt(x, stderr, 1);
+ }
+ i = 1;
+ } else if (xmod) {
+ if (++bxmod == 1) {
+ (void) fprintf(stderr,
+ "%s: b and x regexp modifiers conflict: ", Pn);
+ safestrprt(x, stderr, 1);
+ }
+ i = 1;
+ } else
+ co &= ~REG_EXTENDED;
+ break;
+ case 'i': /* Ignore case. */
+ if (++imod > 1) {
+ if (imod == 2) {
+ (void) fprintf(stderr,
+ "%s: i regexp modifier already used: ", Pn);
+ safestrprt(x, stderr, 1);
+ }
+ i = 1;
+ } else
+ co |= REG_ICASE;
+ break;
+ case 'x': /* This is an extended expression. */
+ if (++xmod > 1) {
+ if (xmod == 2) {
+ (void) fprintf(stderr,
+ "%s: x regexp modifier already used: ", Pn);
+ safestrprt(x, stderr, 1);
+ }
+ i = 1;
+ } else if (bmod) {
+ if (++bxmod == 1) {
+ (void) fprintf(stderr,
+ "%s: b and x regexp modifiers conflict: ", Pn);
+ safestrprt(x, stderr, 1);
+ }
+ i = 1;
+ } else
+ co |= REG_EXTENDED;
+ break;
+ default:
+ (void) fprintf(stderr, "%s: invalid regexp modifier: %c\n",
+ Pn, (int)*xm);
+ i = 1;
+ }
+ }
+ if (i)
+ return(1);
+/*
+ * Allocate space to hold expression and copy it there.
+ */
+ xl = (MALLOC_S)(xe - xb);
+ if (!(xp = (char *)malloc(xl + 1))) {
+ (void) fprintf(stderr, "%s: no regexp space for: ", Pn);
+ safestrprt(x, stderr, 1);
+ Exit(1);
+ }
+ (void) strncpy(xp, xb, xl);
+ xp[(int)xl] = '\0';
+/*
+ * Assign a new CmdRx[] slot for this expression.
+ */
+ if (NCmdRxA >= NCmdRxU) {
+
+ /*
+ * More CmdRx[] space must be assigned.
+ */
+ NCmdRxA += CMDRXINCR;
+ xl = (MALLOC_S)(NCmdRxA * sizeof(lsof_rx_t));
+ if (CmdRx)
+ CmdRx = (lsof_rx_t *)realloc((MALLOC_P *)CmdRx, xl);
+ else
+ CmdRx = (lsof_rx_t *)malloc(xl);
+ if (!CmdRx) {
+ (void) fprintf(stderr, "%s: no space for regexp: ", Pn);
+ safestrprt(x, stderr, 1);
+ Exit(1);
+ }
+ }
+ i = NCmdRxU;
+ CmdRx[i].exp = xp;
+/*
+ * Compile the expression.
+ */
+ if ((re = regcomp(&CmdRx[i].cx, xp, co))) {
+ (void) fprintf(stderr, "%s: regexp error: ", Pn);
+ safestrprt(x, stderr, 0);
+ (void) regerror(re, &CmdRx[i].cx, &reb[0], sizeof(reb));
+ (void) fprintf(stderr, ": %s\n", reb);
+ if (xp) {
+ (void) free((FREE_P *)xp);
+ xp = (char *)NULL;
+ }
+ return(1);
+ }
+/*
+ * Complete the CmdRx[] table entry.
+ */
+ CmdRx[i].mc = 0;
+ CmdRx[i].exp = xp;
+ NCmdRxU++;
+ return(0);
+}
+
+
+/*
+ * enter_fd() - enter file descriptor list for searching
+ */
+
+int
+enter_fd(f)
+ char *f; /* file descriptor list pointer */
+{
+ char c, *cp1, *cp2, *dash;
+ int err, excl, hi, lo;
+ char *fc;
+/*
+ * Check for non-empty list and make a copy.
+ */
+ if (!f || (strlen(f) + 1) < 2) {
+ (void) fprintf(stderr, "%s: no file descriptor specified\n", Pn);
+ return(1);
+ }
+ if (!(fc = mkstrcpy(f, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: no space for fd string: ", Pn);
+ safestrprt(f, stderr, 1);
+ Exit(1);
+ }
+/*
+ * Isolate each file descriptor in the comma-separated list, then enter it
+ * in the file descriptor string list. If a descriptor has the form:
+ *
+ * [0-9]+-[0-9]+
+ *
+ * treat it as an ascending range of file descriptor numbers.
+ *
+ * Accept a leading '^' as an excusion on match.
+ */
+ for (cp1 = fc, err = 0; *cp1;) {
+ if (*cp1 == '^') {
+ excl = 1;
+ cp1++;
+ } else
+ excl = 0;
+ for (cp2 = cp1, dash = (char *)NULL; *cp2 && *cp2 != ','; cp2++) {
+ if (*cp2 == '-')
+ dash = cp2;
+ }
+ if ((c = *cp2) != '\0')
+ *cp2 = '\0';
+ if (cp2 > cp1) {
+ if (dash) {
+ if (ckfd_range(cp1, dash, cp2, &lo, &hi))
+ err = 1;
+ else {
+ if (enter_fd_lst((char *)NULL, lo, hi, excl))
+ err = 1;
+ }
+ } else {
+ if (enter_fd_lst(cp1, 0, 0, excl))
+ err = 1;
+ }
+ }
+ if (c == '\0')
+ break;
+ cp1 = cp2 + 1;
+ }
+ (void) free((FREE_P *)fc);
+ return(err);
+}
+
+
+/*
+ * enter_fd_lst() - make an entry in the FD list, Fdl
+ */
+
+static int
+enter_fd_lst(nm, lo, hi, excl)
+ char *nm; /* FD name (none if NULL) */
+ int lo; /* FD low boundary (if nm NULL) */
+ int hi; /* FD high boundary (if nm NULL) */
+ int excl; /* exclusion on match */
+{
+ char buf[256], *cp;
+ int n;
+ struct fd_lst *f, *ft;
+/*
+ * Don't allow a mixture of exclusions and inclusions.
+ */
+ if (FdlTy >= 0) {
+ if (FdlTy != excl) {
+ if (!Fwarn) {
+
+ /*
+ * If warnings are enabled, report a mixture.
+ */
+ if (nm) {
+ (void) snpf(buf, sizeof(buf) - 1, "%s%s",
+ excl ? "^" : "", nm);
+ } else {
+ if (lo != hi) {
+ (void) snpf(buf, sizeof(buf) - 1, "%s%d-%d",
+ excl ? "^" : "", lo, hi);
+ } else {
+ (void) snpf(buf, sizeof(buf) - 1, "%s%d",
+ excl ? "^" : "", lo);
+ }
+ }
+ buf[sizeof(buf) - 1] = '\0';
+ (void) fprintf(stderr,
+ "%s: %s in an %s -d list: %s\n", Pn,
+ excl ? "exclude" : "include",
+ FdlTy ? "exclude" : "include",
+ buf);
+ }
+ return(1);
+ }
+ }
+/*
+ * Allocate an fd_lst entry.
+ */
+ if (!(f = (struct fd_lst *)malloc((MALLOC_S)sizeof(struct fd_lst)))) {
+ (void) fprintf(stderr, "%s: no space for FD list entry\n", Pn);
+ Exit(1);
+ }
+ if (nm) {
+
+ /*
+ * Process an FD name. First see if it contains only digits; if it
+ * does, convert them to an integer and set the low and high
+ * boundaries to the result.
+ *
+ * If the name has a non-digit, store it as a string, and set the
+ * boundaries to impossible values (i.e., low > high).
+ */
+ for (cp = nm, n = 0; *cp; cp++) {
+ if (!isdigit((unsigned char)*cp))
+ break;
+ n = (n * 10) + (int)(*cp - '0');
+ }
+ if (*cp) {
+ if (!(f->nm = mkstrcpy(nm, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for copy of: %s\n", Pn, nm);
+ Exit(1);
+ }
+ lo = 1;
+ hi = 0;
+ } else {
+ f->nm = (char *)NULL;
+ lo = hi = n;
+ }
+ } else
+ f->nm = (char *)NULL;
+/*
+ * Skip duplicates.
+ */
+ for (ft = Fdl; ft; ft = ft->next) {
+ if (f->nm) {
+ if (!ft->nm || strcmp(f->nm, ft->nm))
+ continue;
+ } else if ((lo != ft->lo) || (hi != ft->hi))
+ continue;
+ (void) free((FREE_P *)f);
+ return(0);
+ }
+/*
+ * Complete the fd_lst entry and link it to the head of the chain.
+ */
+ f->hi = hi;
+ f->lo = lo;
+ f->next = Fdl;
+ Fdl = f;
+ FdlTy = excl;
+ return(0);
+}
+
+
+/*
+ * enter_dir() - enter the files of a directory for searching
+ */
+
+#define EDDEFFNL 128 /* default file name length */
+
+int
+enter_dir(d, descend)
+ char *d; /* directory path name pointer */
+ int descend; /* subdirectory descend flag:
+ * 0 = don't descend
+ * 1 = descend */
+{
+ char *av[2];
+ dev_t ddev;
+ DIR *dfp;
+ char *dn = (char *)NULL;
+ MALLOC_S dnl, dnamlen;
+ struct DIRTYPE *dp;
+ int en, sl;
+ int fct = 0;
+ char *fp = (char *)NULL;
+ MALLOC_S fpl = (MALLOC_S)0;
+ MALLOC_S fpli = (MALLOC_S)0;
+ struct stat sb;
+/*
+ * Check the directory path; reduce symbolic links; stat(2) it; make sure it's
+ * really a directory.
+ */
+ if (!d || !*d || *d == '+' || *d == '-') {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: +d not followed by a directory path\n", Pn);
+ return(1);
+ }
+ if (!(dn = Readlink(d)))
+ return(1);
+ if (statsafely(dn, &sb)) {
+ if (!Fwarn) {
+ en = errno;
+ (void) fprintf(stderr, "%s: WARNING: can't stat(", Pn);
+ safestrprt(dn, stderr, 0);
+ (void) fprintf(stderr, "): %s\n", strerror(en));
+ }
+ if (dn && dn != d) {
+ (void) free((FREE_P *)dn);
+ dn = (char *)NULL;
+ }
+ return(1);
+ }
+ if ((sb.st_mode & S_IFMT) != S_IFDIR) {
+ if (!Fwarn) {
+ (void) fprintf(stderr, "%s: WARNING: not a directory: ", Pn);
+ safestrprt(dn, stderr, 1);
+ }
+ if (dn && dn != d) {
+ (void) free((FREE_P *)dn);
+ dn = (char *)NULL;
+ }
+ return(1);
+ }
+
+#if defined(HASSPECDEVD)
+ (void) HASSPECDEVD(dn, &sb);
+#endif /* defined(HASSPECDEVD) */
+
+ ddev = sb.st_dev;
+/*
+ * Stack the directory and record it in Sfile for searching.
+ */
+ Dstkn = Dstkx = 0;
+ Dstk = (char **)NULL;
+ (void) stkdir(dn);
+ av[0] = (dn == d) ? mkstrcpy(dn, (MALLOC_S *)NULL) : dn;
+ av[1] = (char *)NULL;
+ dn = (char *)NULL;
+ if (!ck_file_arg(0, 1, av, 1, 1, &sb)) {
+ av[0] = (char *)NULL;
+ fct++;
+ }
+/*
+ * Unstack the next directory and examine it.
+ */
+ while (--Dstkx >= 0) {
+ if (!(dn = Dstk[Dstkx]))
+ continue;
+ Dstk[Dstkx] = (char *)NULL;
+ /*
+ * Open the directory path and prepare its name for use with the
+ * files in the directory.
+ */
+ if (!(dfp = OpenDir(dn))) {
+ if (!Fwarn) {
+ if ((en = errno) != ENOENT) {
+ (void) fprintf(stderr,
+ "%s: WARNING: can't opendir(", Pn);
+ safestrprt(dn, stderr, 0);
+ (void) fprintf(stderr, "): %s\n", strerror(en));
+ }
+ }
+ (void) free((FREE_P *)dn);
+ dn = (char *)NULL;
+ continue;
+ }
+ dnl = strlen(dn);
+ sl = ((dnl > 0) && (*(dn + dnl - 1) == '/')) ? 0 : 1;
+ /*
+ * Define space for possible addition to the directory path.
+ */
+ fpli = (MALLOC_S)(dnl + sl + EDDEFFNL + 1);
+ if ((int)fpli > (int)fpl) {
+ fpl = fpli;
+ if (!fp)
+ fp = (char *)malloc(fpl);
+ else
+ fp = (char *)realloc(fp, fpl);
+ if (!fp) {
+ (void) fprintf(stderr,
+ "%s: no space for path to entries in directory: %s\n",
+ Pn, dn);
+ Exit(1);
+ }
+ }
+ (void) snpf(fp, (size_t)fpl, "%s%s", dn, sl ? "/" : "");
+ (void) free((FREE_P *)dn);
+ dn = (char *)NULL;
+ /*
+ * Read the contents of the directory.
+ */
+ for (dp = ReadDir(dfp); dp; dp = ReadDir(dfp)) {
+
+ /*
+ * Skip: entries with no inode number;
+ * entries with a zero length name;
+ * ".";
+ * and "..".
+ */
+ if (!dp->d_ino)
+ continue;
+
+#if defined(HASDNAMLEN)
+ dnamlen = (MALLOC_S)dp->d_namlen;
+#else /* !defined(HASDNAMLEN) */
+ dnamlen = (MALLOC_S)strlen(dp->d_name);
+#endif /* defined(HASDNAMLEN) */
+
+ if (!dnamlen)
+ continue;
+ if (dnamlen <= 2 && dp->d_name[0] == '.') {
+ if (dnamlen == 1)
+ continue;
+ if (dp->d_name[1] == '.')
+ continue;
+ }
+ /*
+ * Form the entry's path name.
+ */
+ fpli = (MALLOC_S)(dnamlen - (fpl - dnl - sl - 1));
+ if ((int)fpli > 0) {
+ fpl += fpli;
+ if (!(fp = (char *)realloc(fp, fpl))) {
+ (void) fprintf(stderr, "%s: no space for: ", Pn);
+ safestrprt(dn, stderr, 0);
+ putc('/', stderr);
+ safestrprtn(dp->d_name, dnamlen, stderr, 1);
+ Exit(1);
+ }
+ }
+ (void) strncpy(fp + dnl + sl, dp->d_name, dnamlen);
+ fp[dnl + sl + dnamlen] = '\0';
+ /*
+ * Lstatsafely() the entry; complain if that fails.
+ *
+ * Stack entries that represent subdirectories.
+ */
+ if (lstatsafely(fp, &sb)) {
+ if ((en = errno) != ENOENT) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: can't lstat(", Pn);
+ safestrprt(fp, stderr, 0);
+ (void) fprintf(stderr, "): %s\n", strerror(en));
+ }
+ }
+ continue;
+ }
+
+#if defined(HASSPECDEVD)
+ (void) HASSPECDEVD(fp, &sb);
+#endif /* defined(HASSPECDEVD) */
+
+ if (!(Fxover & XO_FILESYS)) {
+
+ /*
+ * Unless "-x" or "-x f" was specified, don't cross over file
+ * system mount points.
+ */
+ if (sb.st_dev != ddev)
+ continue;
+ }
+ if ((sb.st_mode & S_IFMT) == S_IFLNK) {
+
+ /*
+ * If this is a symbolic link and "-x_ or "-x l" was specified,
+ * Statsafely() the entry and process it.
+ *
+ * Otherwise skip symbolic links.
+ */
+ if (Fxover & XO_SYMLINK) {
+ if (statsafely(fp, &sb)) {
+ if ((en = errno) != ENOENT) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: can't stat(", Pn);
+ safestrprt(fp, stderr, 0);
+ (void) fprintf(stderr,
+ ") symbolc link: %s\n", strerror(en));
+ }
+ }
+ continue;
+ }
+ } else
+ continue;
+ }
+ if (av[0]) {
+ (void) free((FREE_P *)av[0]);
+ av[0] = (char *)NULL;
+ }
+ av[0] = mkstrcpy(fp, (MALLOC_S *)NULL);
+ if ((sb.st_mode & S_IFMT) == S_IFDIR && descend)
+
+ /*
+ * Stack a subdirectory according to the descend argument.
+ */
+ stkdir(av[0]);
+ /*
+ * Use ck_file_arg() to record the entry for searching. Force it
+ * to consider the entry a file, not a file system.
+ */
+ if (!ck_file_arg(0, 1, av, 1, 1, &sb)) {
+ av[0] = (char *)NULL;
+ fct++;
+ }
+ }
+ (void) CloseDir(dfp);
+ if (dn && dn != d) {
+ (void) free((FREE_P *)dn);
+ dn = (char *)NULL;
+ }
+ }
+/*
+ * Free malloc()'d space.
+ */
+ if (dn && dn != d) {
+ (void) free((FREE_P *)dn);
+ dn = (char *)NULL;
+ }
+ if (av[0] && av[0] != fp) {
+ (void) free((FREE_P *)av[0]);
+ av[0] = (char *)NULL;
+ }
+ if (fp) {
+ (void) free((FREE_P *)fp);
+ fp = (char *)NULL;
+ }
+ if (Dstk) {
+ (void) free((FREE_P *)Dstk);
+ Dstk = (char **)NULL;
+ }
+ if (!fct) {
+
+ /*
+ * Warn if no files were recorded for searching.
+ */
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: no files found in directory: ", Pn);
+ safestrprt(d, stderr, 1);
+ }
+ return(1);
+ }
+ return(0);
+}
+
+
+/*
+ * enter_id() - enter PGID or PID for searching
+ */
+
+int
+enter_id(ty, p)
+ enum IDType ty; /* type: PGID or PID */
+ char *p; /* process group ID string pointer */
+{
+ char *cp;
+ int err, i, id, j, mx, n, ni, nx, x;
+ struct int_lst *s;
+
+ if (!p) {
+ (void) fprintf(stderr, "%s: no process%s ID specified\n",
+ Pn, (ty == PGID) ? " group" : "");
+ return(1);
+ }
+/*
+ * Set up variables for the type of ID.
+ */
+ switch (ty) {
+ case PGID:
+ mx = Mxpgid;
+ n = Npgid;
+ ni = Npgidi;
+ nx = Npgidx;
+ s = Spgid;
+ break;
+ case PID:
+ mx = Mxpid;
+ n = Npid;
+ ni = Npidi;
+ nx = Npidx;
+ s = Spid;
+ break;
+ default:
+ (void) fprintf(stderr, "%s: enter_id \"", Pn);
+ safestrprt(p, stderr, 0);
+ (void) fprintf(stderr, "\", invalid type: %d\n", ty);
+ Exit(1);
+ }
+/*
+ * Convert and store the ID.
+ */
+ for (cp = p, err = 0; *cp;) {
+
+ /*
+ * Assemble ID.
+ */
+ for (i = id = x = 0; *cp && *cp != ','; cp++) {
+ if (!i) {
+ i = 1;
+ if (*cp == '^') {
+ x = 1;
+ continue;
+ }
+ }
+
+#if defined(__STDC__)
+ if (!isdigit((unsigned char)*cp))
+#else /* !defined(__STDC__) */
+ if (!isascii(*cp) || ! isdigit((unsigned char)*cp))
+#endif /* __STDC__ */
+
+ {
+ (void) fprintf(stderr, "%s: illegal process%s ID: ",
+ Pn, (ty == PGID) ? " group" : "");
+ safestrprt(p, stderr, 1);
+ return(1);
+ }
+ id = (id * 10) + *cp - '0';
+ }
+ if (*cp)
+ cp++;
+ /*
+ * Avoid entering duplicates and conflicts.
+ */
+ for (i = j = 0; i < n; i++) {
+ if (id == s[i].i) {
+ if (x == s[i].x) {
+ j = 1;
+ continue;
+ }
+ (void) fprintf(stderr,
+ "%s: P%sID %d has been included and excluded.\n",
+ Pn,
+ (ty == PGID) ? "G" : "",
+ id);
+ err = j = 1;
+ break;
+ }
+ }
+ if (j)
+ continue;
+ /*
+ * Allocate table table space.
+ */
+ if (n >= mx) {
+ mx += IDINCR;
+ if (!s)
+ s = (struct int_lst *)malloc(
+ (MALLOC_S)(sizeof(struct int_lst) * mx));
+ else
+ s = (struct int_lst *)realloc((MALLOC_P *)s,
+ (MALLOC_S)(sizeof(struct int_lst) * mx));
+ if (!s) {
+ (void) fprintf(stderr, "%s: no space for %d process%s IDs",
+ Pn, mx, (ty == PGID) ? " group" : "");
+ Exit(1);
+ }
+ }
+ s[n].f = 0;
+ s[n].i = id;
+ s[n++].x = x;
+ if (x)
+ nx++;
+ else
+ ni++;
+ }
+/*
+ * Save variables for the type of ID.
+ */
+ if (ty == PGID) {
+ Mxpgid = mx;
+ Npgid = n;
+ Npgidi = ni;
+ Npgidx = nx;
+ Spgid = s;
+ } else {
+ Mxpid = mx;
+ Npid = Npuns = n;
+ Npidi = ni;
+ Npidx = nx;
+ Spid = s;
+ }
+ return(err);
+}
+
+
+/*
+ * enter_network_address() - enter Internet address for searching
+ */
+
+int
+enter_network_address(na)
+ char *na; /* Internet address string pointer */
+{
+ int ae, i, pr;
+ int ep = -1;
+ int ft = 0;
+ struct hostent *he = (struct hostent *)NULL;
+ char *hn = (char *)NULL;
+ MALLOC_S l;
+ struct nwad n;
+ char *p, *wa;
+ int pt = 0;
+ int pu = 0;
+ struct servent *se, *se1;
+ char *sn = (char *)NULL;
+ int sp = -1;
+ MALLOC_S snl = 0;
+
+#if defined(HASIPv6)
+ char *cp;
+#endif /* defined(HASIPv6) */
+
+ if (!na) {
+ (void) fprintf(stderr, "%s: no network address specified\n", Pn);
+ return(1);
+ }
+ zeromem((char *)&n, sizeof(n));
+ wa = na;
+/*
+ * Process an IP version type specification, IPv4 or IPv6, optionally followed
+ * by a '@' and a host name or Internet address, or a ':' and a service name or
+ * port number.
+ */
+ if ((*wa == '4') || (*wa == '6')) {
+ if (*wa == '4')
+ ft = 4;
+ else if (*wa == '6') {
+
+#if defined(HASIPv6)
+ ft = 6;
+#else /* !defined(HASIPv6) */
+ (void) fprintf(stderr, "%s: IPv6 not supported: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+#endif /* defined(HASIPv6) */
+
+ }
+ wa++;
+ if (!*wa) {
+
+ /*
+ * If nothing follows 4 or 6, then all network files of the
+ * specified IP version are selected. Sequential -i, -i4, and
+ * -i6 specifications interact logically -- e.g., -i[46] followed
+ * by -i[64] is the same as -i.
+ */
+ if (!Fnet) {
+ Fnet = 1;
+ FnetTy = ft;
+ } else {
+ if (FnetTy) {
+ if (FnetTy != ft)
+ FnetTy = 0;
+ } else
+ FnetTy = ft;
+ }
+ return(0);
+ }
+ } else if (Fnet)
+ ft = FnetTy;
+/*
+ * If an IP version has been specified, use it to set the address family.
+ */
+ switch (ft) {
+ case 4:
+ n.af = AF_INET;
+ break;
+
+#if defined(HASIPv6)
+ case 6:
+ n.af = AF_INET6;
+ break;
+#endif /* defined(HASIPv6) */
+
+ }
+/*
+ * Process protocol name, optionally followed by a '@' and a host name or
+ * Internet address, or a ':' and a service name or port number.
+ */
+ if (*wa && *wa != '@' && *wa != ':') {
+ for (p = wa; *wa && *wa != '@' && *wa != ':'; wa++)
+ ;
+ if ((l = wa - p)) {
+ if (!(n.proto = mkstrcat(p, l, (char *)NULL, -1, (char *)NULL,
+ -1, (MALLOC_S *)NULL)))
+ {
+ (void) fprintf(stderr,
+ "%s: no space for protocol name from: -i ", Pn);
+ safestrprt(na, stderr, 1);
+nwad_exit:
+ if (n.proto)
+ (void) free((FREE_P *)n.proto);
+ if (hn)
+ (void) free((FREE_P *)hn);
+ if (sn)
+ (void) free((FREE_P *)sn);
+ return(1);
+ }
+ /*
+ * The protocol name should be "tcp", "udp" or "udplite".
+ */
+ if ((strcasecmp(n.proto, "tcp") != 0)
+ && (strcasecmp(n.proto, "udp") != 0)
+ && (strcasecmp(n.proto, "udplite") != 0))
+ {
+ (void) fprintf(stderr,
+ "%s: unknown protocol name (%s) in: -i ", Pn, n.proto);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ /*
+ * Convert protocol name to lower case.
+ */
+ for (p = n.proto; *p; p++) {
+ if (*p >= 'A' && *p <= 'Z')
+ *p = *p - 'A' + 'a';
+ }
+ }
+ }
+/*
+ * Process an IPv4 address (1.2.3.4), IPv6 address ([1:2:3:4:5:6:7:8]),
+ * or host name, preceded by a '@' and optionally followed by a colon
+ * and a service name or port number.
+ */
+ if (*wa == '@') {
+ wa++;
+ if (!*wa || *wa == ':') {
+
+#if defined(HASIPv6)
+unacc_address:
+#endif /* defined(HASIPv6) */
+
+ (void) fprintf(stderr,
+ "%s: unacceptable Internet address in: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+
+ if ((p = isIPv4addr(wa, n.a, sizeof(n.a)))) {
+
+ /*
+ * Process IPv4 address.
+ */
+ if (ft == 6) {
+ (void) fprintf(stderr,
+ "%s: IPv4 addresses are prohibited: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ wa = p;
+ n.af = AF_INET;
+ } else if (*wa == '[') {
+
+#if defined(HASIPv6)
+ /*
+ * Make sure IPv6 addresses are permitted. If they are, assemble
+ * one.
+ */
+ if (ft == 4) {
+ (void) fprintf(stderr,
+ "%s: IPv6 addresses are prohibited: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ if (!(cp = strrchr(++wa, ']')))
+ goto unacc_address;
+ *cp = '\0';
+ i = inet_pton(AF_INET6, wa, (void *)&n.a);
+ *cp = ']';
+ if (i != 1)
+ goto unacc_address;
+ for (ae = i = 0; i < MAX_AF_ADDR; i++) {
+ if ((ae |= n.a[i]))
+ break;
+ }
+ if (!ae)
+ goto unacc_address;
+ if (IN6_IS_ADDR_V4MAPPED((struct in6_addr *)&n.a[0])) {
+ if (ft == 6) {
+ (void) fprintf(stderr,
+ "%s: IPv4 addresses are prohibited: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ for (i = 0; i < 4; i++) {
+ n.a[i] = n.a[i+12];
+ }
+ n.af = AF_INET;
+ } else
+ n.af = AF_INET6;
+ wa = cp + 1;
+#else /* !defined(HASIPv6) */
+ (void) fprintf(stderr,
+ "%s: unsupported IPv6 address in: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+#endif /* defined(HASIPv6) */
+
+ } else {
+
+ /*
+ * Assemble host name.
+ */
+ for (p = wa; *p && *p != ':'; p++)
+ ;
+ if ((l = p - wa)) {
+ if (!(hn = mkstrcat(wa, l, (char *)NULL, -1, (char *)NULL,
+ -1, (MALLOC_S *)NULL)))
+ {
+ (void) fprintf(stderr,
+ "%s: no space for host name: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+
+#if defined(HASIPv6)
+
+ /*
+ * If no IP version has been specified, look up an IPv6 host
+ * name first. If that fails, look up an IPv4 host name.
+ *
+ * If the IPv6 version has been specified, look up the host
+ * name only under its IP version specification.
+ */
+ if (!ft)
+ n.af = AF_INET6;
+ if (!(he = lkup_hostnm(hn, &n)) && !ft) {
+ n.af = AF_INET;
+ he = lkup_hostnm(hn, &n);
+ }
+#else /* !defined(HASIPv6) */
+ if (!ft)
+ n.af = AF_INET;
+ he = lkup_hostnm(hn, &n);
+#endif /* defined(HASIPv6) */
+
+ if (!he) {
+ fprintf(stderr, "%s: unknown host name (%s) in: -i ",
+ Pn, hn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ }
+ wa = p;
+ }
+ }
+/*
+ * If there is no port number, enter the address.
+ */
+ if (!*wa)
+ goto nwad_enter;
+/*
+ * Process a service name or port number list, preceded by a colon.
+ *
+ * Entries of the list are separated with commas; elements of a numeric range
+ * are specified with a separating minus sign (`-'); all service names must
+ * belong to the same protocol; embedded spaces are not allowed. An embedded
+ * minus sign in a name is taken to be part of the name, the starting entry
+ * of a range can't be a service name.
+ */
+ if (*wa != ':' || *(wa + 1) == '\0') {
+
+unacc_port:
+ (void) fprintf(stderr,
+ "%s: unacceptable port specification in: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ for (++wa; wa && *wa; wa++) {
+ for (ep = pr = sp = 0; *wa; wa++) {
+ if (*wa < '0' || *wa > '9') {
+
+ /*
+ * Convert service name to port number, using already-specified
+ * protocol name. A '-' is taken to be part of the name; hence
+ * the starting entry of a range can't be a service name.
+ */
+ for (p = wa; *wa && *wa != ','; wa++)
+ ;
+ if (!(l = wa - p)) {
+ (void) fprintf(stderr,
+ "%s: invalid service name: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ if (sn) {
+ if (l > snl) {
+ sn = (char *)realloc((MALLOC_P *)sn, l + 1);
+ snl = l;
+ }
+ } else {
+ sn = (char *)malloc(l + 1);
+ snl = l;
+ }
+ if (!sn) {
+ (void) fprintf(stderr,
+ "%s: no space for service name: -i ", Pn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ (void) strncpy(sn, p, l);
+ *(sn + l) = '\0';
+ if (n.proto) {
+
+ /*
+ * If the protocol has been specified, look up the port
+ * number for the service name for the specified protocol.
+ */
+ if (!(se = getservbyname(sn, n.proto))) {
+ (void) fprintf(stderr,
+ "%s: unknown service %s for %s in: -i ",
+ Pn, sn, n.proto);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ pt = (int)ntohs(se->s_port);
+ } else {
+
+ /*
+ * If no protocol has been specified, look up the port
+ * numbers for the service name for both TCP and UDP.
+ */
+ if((se = getservbyname(sn, "tcp")))
+ pt = (int)ntohs(se->s_port);
+ if ((se1 = getservbyname(sn, "udp")))
+ pu = (int)ntohs(se1->s_port);
+ if (!se && !se1) {
+ (void) fprintf(stderr,
+ "%s: unknown service %s in: -i ", Pn, sn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ if (se && se1 && pt != pu) {
+ (void) fprintf(stderr,
+ "%s: TCP=%d and UDP=%d %s ports conflict;\n",
+ Pn, pt, pu, sn);
+ (void) fprintf(stderr,
+ " specify \"tcp:%s\" or \"udp:%s\": -i ",
+ sn, sn);
+ safestrprt(na, stderr, 1);
+ goto nwad_exit;
+ }
+ if (!se && se1)
+ pt = pu;
+ }
+ if (pr)
+ ep = pt;
+ else {
+ sp = pt;
+ if (*wa == '-')
+ pr++;
+ }
+ } else {
+
+ /*
+ * Assemble port number.
+ */
+ for (; *wa && *wa != ','; wa++) {
+ if (*wa == '-') {
+ if (pr)
+ goto unacc_port;
+ pr++;
+ break;
+ }
+ if (*wa < '0' || *wa > '9')
+ goto unacc_port;
+ if (pr)
+ ep = (ep * 10) + *wa - '0';
+ else
+ sp = (sp * 10) + *wa - '0';
+ }
+ }
+ if (!*wa || *wa == ',')
+ break;
+ if (pr)
+ continue;
+ goto unacc_port;
+ }
+ if (!pr)
+ ep = sp;
+ if (ep < sp)
+ goto unacc_port;
+ /*
+ * Enter completed port or port range specification.
+ */
+
+nwad_enter:
+
+ for (i = 1; i;) {
+ if (enter_nwad(&n, sp, ep, na, he))
+ goto nwad_exit;
+
+#if defined(HASIPv6)
+ /*
+ * If IPv6 is enabled, a host name was specified, and the
+ * associated * address is for the AF_INET6 address family,
+ * try to get and address for the AF_INET family, too, unless
+ * IPv4 is prohibited.
+ */
+ if (hn && (n.af == AF_INET6) && (ft != 6)) {
+ n.af = AF_INET;
+ if ((he = lkup_hostnm(hn, &n)))
+ continue;
+ }
+#endif /* defined(HASIPv6) */
+
+ i = 0;
+ }
+ if (!*wa)
+ break;
+ }
+ if (sn)
+ (void) free((FREE_P *)sn);
+ return(0);
+}
+
+/*
+ * enter_nwad() - enter nwad structure
+ */
+
+static int
+enter_nwad(n, sp, ep, s, he)
+ struct nwad *n; /* pointer to partially completed
+ * nwad (less port) */
+ int sp; /* starting port number */
+ int ep; /* ending port number */
+ char *s; /* string that states the address */
+ struct hostent *he; /* pointer to hostent struct from which
+ * network address came */
+{
+ int ac;
+ unsigned char *ap;
+ static int na = 0;
+ struct nwad nc;
+ struct nwad *np;
+/*
+ * Allocate space for the argument specification.
+ */
+ if (strlen(s)) {
+ if (!(n->arg = mkstrcpy(s, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for Internet argument: -i ", Pn);
+ safestrprt(s, stderr, 1);
+ Exit(1);
+ }
+ } else
+ n->arg = (char *)NULL;
+/*
+ * Loop through all hostent addresses.
+ */
+ for (ac = 1, nc = *n;;) {
+
+ /*
+ * Test address specification -- it must contain at least one of:
+ * protocol, Internet address or port. If correct, link into search
+ * list.
+ */
+ if (!nc.proto
+ && !nc.a[0] && !nc.a[1] && !nc.a[2] && !nc.a[3]
+
+#if defined(HASIPv6)
+ && (nc.af != AF_INET6
+ || (!nc.a[4] && !nc.a[5] && !nc.a[6] && !nc.a[7]
+ && !nc.a[8] && !nc.a[9] && !nc.a[10] && !nc.a[11]
+ && !nc.a[12] && !nc.a[13] && !nc.a[14] && !nc.a[15]))
+#endif /* defined(HASIPv6) */
+
+ && sp == -1) {
+ (void) fprintf(stderr,
+ "%s: incomplete Internet address specification: -i ", Pn);
+ safestrprt(s, stderr, 1);
+ return(1);
+ }
+ /*
+ * Limit the network address chain length to MAXNWAD for reasons of
+ * search efficiency.
+ */
+ if (na >= MAXNWAD) {
+ (void) fprintf(stderr,
+ "%s: network address limit (%d) exceeded: -i ",
+ Pn, MAXNWAD);
+ safestrprt(s, stderr, 1);
+ return(1);
+ }
+ /*
+ * Allocate space for the address specification.
+ */
+ if ((np = (struct nwad *)malloc(sizeof(struct nwad))) == NULL) {
+ (void) fprintf(stderr,
+ "%s: no space for network address from: -i ", Pn);
+ safestrprt(s, stderr, 1);
+ return(1);
+ }
+ /*
+ * Construct and link the address specification.
+ */
+ *np = nc;
+ np->sport = sp;
+ np->eport = ep;
+ np->f = 0;
+ np->next = Nwad;
+ Nwad = np;
+ na++;
+ /*
+ * If the network address came from gethostbyname(), advance to
+ * the next address; otherwise quit.
+ */
+ if (!he)
+ break;
+ if (!(ap = (unsigned char *)he->h_addr_list[ac++]))
+ break;
+
+#if defined(HASIPv6)
+ {
+ int i;
+
+ for (i = 0;
+ (i < (he->h_length - 1)) && (i < (MAX_AF_ADDR - 1));
+ i++)
+ {
+ nc.a[i] = *ap++;
+ }
+ nc.a[i] = *ap;
+ }
+#else /* !defined(HASIPv6) */
+ nc.a[0] = *ap++;
+ nc.a[1] = *ap++;
+ nc.a[2] = *ap++;
+ nc.a[3] = *ap;
+#endif /* defined(HASIPv6) */
+
+ }
+ return(0);
+}
+
+
+#if defined(HASTCPUDPSTATE)
+/*
+ * enter_state_spec() -- enter TCP and UDP state specifications
+ */
+
+int
+enter_state_spec(ss)
+ char *ss; /* state specification string */
+{
+ char *cp, *ne, *ns, *pr;
+ int err, d, f, i, tx, x;
+ size_t len;
+ static char *ssc = (char *)NULL;
+ char *ty;
+/*
+ * Check the protocol specification.
+ */
+ if (!strncasecmp(ss, "tcp:", 4)) {
+ pr = "TCP";
+ tx = 0;
+ }
+
+#if !defined(USE_LIB_PRINT_TCPTPI)
+ else if (!strncasecmp(ss, "UDP:", 4)) {
+ pr = "UDP";
+ tx = 1;
+ }
+
+#endif /* !defined(USE_LIB_PRINT_TCPTPI) */
+
+ else {
+ (void) fprintf(stderr, "%s: unknown -s protocol: \"%s\"\n",
+ Pn, ss);
+ return(1);
+ }
+ cp = ss + 4;
+ if (!*cp) {
+ (void) fprintf(stderr, "%s: no %s state names in: %s\n",
+ Pn, pr, ss);
+ return(1);
+ }
+ (void) build_IPstates();
+ if (!(tx ? UdpSt : TcpSt)) {
+ (void) fprintf(stderr, "%s: no %s state names available: %s\n",
+ Pn, pr, ss);
+ return(1);
+ }
+/*
+ * Allocate the inclusion and exclusion tables for the protocol.
+ */
+ if (tx) {
+ if (UdpNstates) {
+ if (!UdpStI) {
+ if (!(UdpStI = (unsigned char *)calloc((MALLOC_S)UdpNstates,
+ sizeof(unsigned char))))
+ {
+ ty = "UDP state inclusion";
+
+no_IorX_space:
+
+ (void) fprintf(stderr, "%s: no %s table space\n",
+ Pn, ty);
+ Exit(1);
+ }
+ }
+ if (!UdpStX) {
+ if (!(UdpStX = (unsigned char *)calloc((MALLOC_S)UdpNstates,
+ sizeof(unsigned char))))
+ {
+ ty = "UDP state exclusion";
+ goto no_IorX_space;
+ }
+ }
+ }
+ } else {
+ if (TcpNstates) {
+ if (!TcpStI) {
+ if (!(TcpStI = (unsigned char *)calloc((MALLOC_S)TcpNstates,
+ sizeof(unsigned char))))
+ {
+ ty = "TCP state inclusion";
+ goto no_IorX_space;
+ }
+ }
+ if (!TcpStX) {
+ if (!(TcpStX = (unsigned char *)calloc((MALLOC_S)TcpNstates,
+ sizeof(unsigned char))))
+ {
+ ty = "TCP state exclusion";
+ goto no_IorX_space;
+ }
+ }
+ }
+ }
+/*
+ * Convert the state names in the rest of the string to state indexes and
+ * record them in the appropriate inclusion or exclusion table.
+ */
+ if (ssc)
+ (void) free((MALLOC_P *)ssc);
+ if (!(ssc = mkstrcpy(cp, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no temporary state argument space for: %s\n", Pn, ss);
+ Exit(1);
+ }
+ cp = ssc;
+ err = 0;
+ while (*cp) {
+
+ /*
+ * Determine inclusion or exclusion for this state name.
+ */
+ if (*cp == '^') {
+ x = 1;
+ cp++;
+ } else
+ x = 0;
+ /*
+ * Find the end of the state name. Make sure it is non-null in length
+ * and terminated with '\0'.
+ */
+ ns = cp;
+ while (*cp && (*cp != ',')) {
+ cp++;
+ }
+ ne = cp;
+ if (*cp) {
+ *cp = '\0';
+ cp++;
+ }
+ if (!(len = (size_t)(ne - ns))) {
+ (void) fprintf(stderr, "%s: NULL %s state name in: %s\n",
+ Pn, pr, ss);
+ err = 1;
+ continue;
+ }
+ /*
+ * Find the state name in the appropriate table.
+ */
+ f = 0;
+ if (tx) {
+ if (UdpSt) {
+ for (i = 0; i < UdpNstates; i++) {
+ if (!strcasecmp(ns, UdpSt[i])) {
+ f = 1;
+ break;
+ }
+ }
+ }
+ } else {
+ if (TcpSt) {
+ for (i = 0; i < TcpNstates; i++) {
+ if (!strcasecmp(ns, TcpSt[i])) {
+ f = 1;
+ break;
+ }
+ }
+ }
+ }
+ if (!f) {
+ (void) fprintf(stderr, "%s: unknown %s state name: %s\n",
+ Pn, pr, ns);
+ err = 1;
+ continue;
+ }
+ /*
+ * Set the inclusion or exclusion status in the appropriate table.
+ */
+ d = 0;
+ if (x) {
+ if (tx) {
+ if (!UdpStX[i]) {
+ UdpStX[i] = 1;
+ UdpStXn++;
+ } else
+ d = 1;
+ } else {
+ if (!TcpStX[i]) {
+ TcpStX[i] = 1;
+ TcpStXn++;
+ } else
+ d = 1;
+ }
+ } else {
+ if (tx) {
+ if (!UdpStI[i]) {
+ UdpStI[i] = 1;
+ UdpStIn++;
+ } else
+ d = 1;
+ } else {
+ if (!TcpStI[i]) {
+ TcpStI[i] = 1;
+ TcpStIn++;
+ } else
+ d = 1;
+ }
+ }
+ if (d) {
+
+ /*
+ * Report a duplicate.
+ */
+ (void) fprintf(stderr, "%s: duplicate %s %sclusion: %s\n",
+ Pn, pr,
+ x ? "ex" : "in",
+ ns);
+ err = 1;
+ }
+ }
+/*
+ * Release any temporary space and return.
+ */
+ if (ssc) {
+ (void) free((MALLOC_P *)ssc);
+ ssc = (char *)NULL;
+ }
+ return(err);
+}
+#endif /* defined(HASTCPUDPSTATE) */
+
+
+/*
+ * enter_str_lst() - enter a string on a list
+ */
+
+int
+enter_str_lst(opt, s, lp, incl, excl)
+ char *opt; /* option name */
+ char *s; /* string to enter */
+ struct str_lst **lp; /* string's list */
+ int *incl; /* included count */
+ int *excl; /* excluded count */
+{
+ char *cp;
+ short i, x;
+ MALLOC_S len;
+ struct str_lst *lpt;
+
+ if (!s || *s == '-' || *s == '+') {
+ (void) fprintf(stderr, "%s: missing %s option value\n",
+ Pn, opt);
+ return(1);
+ }
+ if (*s == '^') {
+ i = 0;
+ x = 1;
+ s++;
+ } else {
+ i = 1;
+ x = 0;
+ }
+ if (!(cp = mkstrcpy(s, &len))) {
+ (void) fprintf(stderr, "%s: no string copy space: ", Pn);
+ safestrprt(s, stderr, 1);
+ return(1);
+ }
+ if ((lpt = (struct str_lst *)malloc(sizeof(struct str_lst))) == NULL) {
+ (void) fprintf(stderr, "%s: no list space: ", Pn);
+ safestrprt(s, stderr, 1);
+ (void) free((FREE_P *)cp);
+ return(1);
+ }
+ lpt->f = 0;
+ lpt->str = cp;
+ lpt->len = (int)len;
+ lpt->x = x;
+ if (i)
+ *incl += 1;
+ if (x)
+ *excl += 1;
+ lpt->next = *lp;
+ *lp = lpt;
+ return(0);
+}
+
+
+/*
+ * enter_uid() - enter User Identifier for searching
+ */
+
+int
+enter_uid(us)
+ char *us; /* User IDentifier string pointer */
+{
+ int err, i, j, lnml, nn;
+ unsigned char excl;
+ MALLOC_S len;
+ char lnm[LOGINML+1], *lp;
+ struct passwd *pw;
+ char *s, *st;
+ uid_t uid;
+
+ if (!us) {
+ (void) fprintf(stderr, "%s: no UIDs specified\n", Pn);
+ return(1);
+ }
+ for (err = 0, s = us; *s;) {
+
+ /*
+ * Assemble next User IDentifier.
+ */
+ for (excl = i = j = lnml = nn = uid = 0, st = s;
+ *s && *s != ',';
+ i++, s++)
+ {
+ if (lnml >= LOGINML) {
+ while (*s && *s != ',') {
+ s++;
+ lnml++;
+ }
+ (void) fprintf(stderr,
+ "%s: -u login name > %d characters: ", Pn,
+ (int)LOGINML);
+ safestrprtn(st, lnml, stderr, 1);
+ err = j = 1;
+ break;
+ }
+ if (i == 0 && *s == '^') {
+ excl = 1;
+ continue;
+ }
+ lnm[lnml++] = *s;
+ if (nn)
+ continue;
+
+#if defined(__STDC__)
+ if (isdigit((unsigned char)*s))
+#else /* !defined(__STDC__) */
+ if (isascii(*s) && isdigit((unsigned char)*s))
+#endif /* defined(__STDC__) */
+
+ uid = (uid * 10) + *s - '0';
+ else
+ nn++;
+ }
+ if (*s)
+ s++;
+ if (j)
+ continue;
+ if (nn) {
+ lnm[lnml++] = '\0';
+ if ((pw = getpwnam(lnm)) == NULL) {
+ (void) fprintf(stderr, "%s: can't get UID for ", Pn);
+ safestrprt(lnm, stderr, 1);
+ err = 1;
+ continue;
+ } else
+ uid = pw->pw_uid;
+ }
+
+#if defined(HASSECURITY) && !defined(HASNOSOCKSECURITY)
+ /*
+ * If the security mode is enabled, only the root user may list files
+ * belonging to user IDs other than the real user ID of this lsof
+ * process. If HASNOSOCKSECURITY is also defined, then anyone may
+ * list anyone else's socket files.
+ */
+ if (Myuid && uid != Myuid) {
+ (void) fprintf(stderr,
+ "%s: ID %d request rejected because of security mode.\n",
+ Pn, uid);
+ err = 1;
+ continue;
+ }
+#endif /* defined(HASSECURITY) && !defined(HASNOSOCKSECURITY) */
+
+ /*
+ * Avoid entering duplicates.
+ */
+ for (i = j = 0; i < Nuid; i++) {
+ if (uid != Suid[i].uid)
+ continue;
+ if (Suid[i].excl == excl) {
+ j = 1;
+ continue;
+ }
+ (void) fprintf(stderr,
+ "%s: UID %d has been included and excluded.\n",
+ Pn, (int)uid);
+ err = j = 1;
+ break;
+ }
+ if (j)
+ continue;
+ /*
+ * Allocate space for User IDentifier.
+ */
+ if (Nuid >= Mxuid) {
+ Mxuid += UIDINCR;
+ len = (MALLOC_S)(Mxuid * sizeof(struct seluid));
+ if (!Suid)
+ Suid = (struct seluid *)malloc(len);
+ else
+ Suid = (struct seluid *)realloc((MALLOC_P *)Suid, len);
+ if (!Suid) {
+ (void) fprintf(stderr, "%s: no space for UIDs", Pn);
+ Exit(1);
+ }
+ }
+ if (nn) {
+ if (!(lp = mkstrcpy(lnm, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: no space for login: ", Pn);
+ safestrprt(lnm, stderr, 1);
+ Exit(1);
+ }
+ Suid[Nuid].lnm = lp;
+ } else
+ Suid[Nuid].lnm = (char *)NULL;
+ Suid[Nuid].uid = uid;
+ Suid[Nuid++].excl = excl;
+ if (excl)
+ Nuidexcl++;
+ else
+ Nuidincl++;
+ }
+ return(err);
+}
+
+
+/*
+ * isIPv4addr() - is host name an IPv4 address
+ */
+
+static char *
+isIPv4addr(hn, a, al)
+ char *hn; /* host name */
+ unsigned char *a; /* address receptor */
+ int al; /* address receptor length */
+{
+ int dc = 0; /* dot count */
+ int i; /* temorary index */
+ int ov[MIN_AF_ADDR]; /* octet values */
+ int ovx = 0; /* ov[] index */
+/*
+ * The host name must begin with a number and the return octet value
+ * arguments must be acceptable.
+ */
+ if ((*hn < '0') || (*hn > '9'))
+ return((char *)NULL);
+ if (!a || (al < MIN_AF_ADDR))
+ return((char *)NULL);
+/*
+ * Start the first octet assembly, then parse tge remainder of the host
+ * name for four octets, separated by dots.
+ */
+ ov[0] = (int)(*hn++ - '0');
+ while (*hn && (*hn != ':')) {
+ if (*hn == '.') {
+
+ /*
+ * Count a dot. Make sure a preceding octet value has been
+ * assembled. Don't assemble more than MIN_AF_ADDR octets.
+ */
+ dc++;
+ if ((ov[ovx] < 0) || (ov[ovx] > 255))
+ return((char *)NULL);
+ if (++ovx > (MIN_AF_ADDR - 1))
+ return((char *)NULL);
+ ov[ovx] = -1;
+ } else if ((*hn >= '0') && (*hn <= '9')) {
+
+ /*
+ * Assemble an octet.
+ */
+ if (ov[ovx] < 0)
+ ov[ovx] = (int)(*hn - '0');
+ else
+ ov[ovx] = (ov[ovx] * 10) + (int)(*hn - '0');
+ } else {
+
+ /*
+ * A non-address character has been detected.
+ */
+ return((char *)NULL);
+ }
+ hn++;
+ }
+/*
+ * Make sure there were three dots and four non-null octets.
+ */
+ if ((dc != 3)
+ || (ovx != (MIN_AF_ADDR - 1))
+ || (ov[ovx] < 0) || (ov[ovx] > 255))
+ return((char *)NULL);
+/*
+ * Copy the octets as unsigned characters and return the ending host name
+ * character position.
+ */
+ for (i = 0; i < MIN_AF_ADDR; i++) {
+ a[i] = (unsigned char)ov[i];
+ }
+ return(hn);
+}
+
+
+/*
+ * lkup_hostnm() - look up host name
+ */
+
+static struct hostent *
+lkup_hostnm(hn, n)
+ char *hn; /* host name */
+ struct nwad *n; /* network address destination */
+{
+ unsigned char *ap;
+ struct hostent *he;
+ int ln;
+/*
+ * Get hostname structure pointer. Return NULL if there is none.
+ */
+
+#if defined(HASIPv6)
+ he = gethostbyname2(hn, n->af);
+#else /* !defined(HASIPv6) */
+ he = gethostbyname(hn);
+#endif /* defined(HASIPv6) */
+
+ if (!he)
+ return(he);
+/*
+ * Copy first hostname structure address to destination structure.
+ */
+
+#if defined(HASIPv6)
+ if (n->af != he->h_addrtype)
+ return((struct hostent *)NULL);
+ if (n->af == AF_INET6) {
+
+ /*
+ * Copy an AF_INET6 address.
+ */
+ if (he->h_length > MAX_AF_ADDR)
+ return((struct hostent *)NULL);
+ (void) memcpy((void *)&n->a[0], (void *)he->h_addr, he->h_length);
+ if ((ln = MAX_AF_ADDR - he->h_length) > 0)
+ zeromem((char *)&n->a[he->h_length], ln);
+ return(he);
+ }
+#endif /* defined(HASIPv6) */
+
+/*
+ * Copy an AF_INET address.
+ */
+ if (he->h_length != 4)
+ return((struct hostent *)NULL);
+ ap = (unsigned char *)he->h_addr;
+ n->a[0] = *ap++;
+ n->a[1] = *ap++;
+ n->a[2] = *ap++;
+ n->a[3] = *ap;
+ if ((ln = MAX_AF_ADDR - 4) > 0)
+ zeromem((char *)&n->a[4], ln);
+ return(he);
+}
--- /dev/null
+
+# Linux /proc-based Makefile
+#
+# $Id: Makefile,v 1.11 2008/04/15 13:30:01 abe Exp $
+
+PROG= lsof
+
+BIN= ${DESTDIR}
+
+DOC= ${DESTDIR}
+
+I=/usr/include
+S=/usr/include/sys
+L=/usr/include/local
+P=
+
+CDEF=
+CDEFS= ${CDEF} ${CFGF}
+DEP= ${CFGD} ${CFGDN}
+INCL= ${DINC}
+CFLAGS= ${CDEFS} ${INCL} ${DEP} ${DEBUG}
+
+GRP=
+
+HDR= lsof.h lsof_fields.h dlsof.h machine.h proto.h dproto.h
+
+SRC= dfile.c dmnt.c dnode.c dproc.c dsock.c dstore.c \
+ arg.c main.c misc.c node.c print.c proc.c store.c usage.c \
+ util.c
+
+OBJ= dfile.o dmnt.o dnode.o dproc.o dsock.o dstore.o \
+ arg.o main.o misc.o node.o print.o proc.o store.o usage.o \
+ util.o
+
+MAN= lsof.8
+
+OTHER=
+
+SHELL= /bin/sh
+
+SOURCE= Makefile ${OTHER} ${MAN} ${HDR} ${SRC}
+
+all: ${PROG}
+
+${PROG}: ${P} ${LIB} ${OBJ}
+ ${CC} -o $@ ${OBJ} ${CFGL}
+
+clean: FRC
+ rm -f Makefile.bak ${PROG} a.out core errs lint.out tags *.o version.h ${CFGDN}
+ rm -f machine.h.old new_machine.h
+ (cd lib; ${MAKE} -f Makefile.skel clean)
+
+install: all FRC
+ @echo ''
+ @echo 'Please write your own install rule. Lsof should be installed'
+ @echo 'setuid to root if you wish any lsof user to be able to examine'
+ @echo 'all open files. Your install rule actions might look something'
+ @echo 'like this:'
+ @echo ''
+ @echo ' install -m 4xxx -o root -g $${GRP} $${PROG} $${BIN}'
+ @echo ' install -m 444 $${MAN} $${DOC}'
+ @echo ''
+ @echo 'You will have to complete the 4xxx modes, the GRP value, and'
+ @echo 'the skeletons for the BIN and DOC strings, given at the'
+ @echo 'beginning of this Makefile, e.g.,'
+ @echo ''
+ @echo ' BIN= $${DESTDIR}/usr/local/etc'
+ @echo ' DOC= $${DESTDIR}/usr/man/man8'
+ @echo ' GRP= sys'
+ @echo ''
+
+${LIB}: FRC
+ (cd lib; ${MAKE} DEBUG="${DEBUG}" CFGF="${CFGF}")
+
+version.h: FRC
+ @echo Constructing version.h
+ @rm -f version.h
+ @echo '#define LSOF_BLDCMT "${LSOF_BLDCMT}"' > version.h;
+ @echo '#define LSOF_CC "${CC}"' >> version.h
+ @echo '#define LSOF_CCV "${CCV}"' >> version.h
+ @echo '#define LSOF_CCDATE "'`date`'"' >> version.h
+ @echo '#define LSOF_CCFLAGS "'`echo ${CFLAGS} | sed 's/\\\\(/\\(/g' | sed 's/\\\\)/\\)/g' | sed 's/"/\\\\"/g'`'"' >> version.h
+ @echo '#define LSOF_CINFO "${CINFO}"' >> version.h
+ @if [ "X${LSOF_HOST}" = "X" ]; then \
+ echo '#define LSOF_HOST "'`uname -n`'"' >> version.h; \
+ else \
+ if [ "${LSOF_HOST}" = "none" ]; then \
+ echo '#define LSOF_HOST ""' >> version.h; \
+ else \
+ echo '#define LSOF_HOST "${LSOF_HOST}"' >> version.h; \
+ fi \
+ fi
+ @echo '#define LSOF_LDFLAGS "${CFGL}"' >> version.h
+ @if [ "X${LSOF_LOGNAME}" = "X" ]; then \
+ echo '#define LSOF_LOGNAME "${LOGNAME}"' >> version.h; \
+ else \
+ if [ "${LSOF_LOGNAME}" = "none" ]; then \
+ echo '#define LSOF_LOGNAME ""' >> version.h; \
+ else \
+ echo '#define LSOF_LOGNAME "${LSOF_LOGNAME}"' >> version.h; \
+ fi; \
+ fi
+ @if [ "X${LSOF_SYSINFO}" = "X" ]; then \
+ echo '#define LSOF_SYSINFO "'`uname -a`'"' >> version.h; \
+ else \
+ if [ "${LSOF_SYSINFO}" = "none" ]; then \
+ echo '#define LSOF_SYSINFO ""' >> version.h; \
+ else \
+ echo '#define LSOF_SYSINFO "${LSOF_SYSINFO}"' >> version.h; \
+ fi \
+ fi
+ @if [ "X${LSOF_USER}" = "X" ]; then \
+ echo '#define LSOF_USER "${USER}"' >> version.h; \
+ else \
+ if [ "${LSOF_USER}" = "none" ]; then \
+ echo '#define LSOF_USER ""' >> version.h; \
+ else \
+ echo '#define LSOF_USER "${LSOF_USER}"' >> version.h; \
+ fi \
+ fi
+ @sed '/VN/s/.ds VN \(.*\)/#define LSOF_VERSION "\1"/' < version >> version.h
+
+FRC:
+
+# DO NOT DELETE THIS LINE - make depend DEPENDS ON IT
+
+dfile.o: ${HDR} dfile.c
+
+dmnt.o: ${HDR} dmnt.c
+
+dnode.o: ${HDR} dnode.c
+
+dproc.o: ${HDR} dproc.c
+
+dsock.o: ${HDR} dsock.c
+
+dstore.o: ${HDR} dstore.c
+
+arg.o: ${HDR} arg.c
+
+main.o: ${HDR} main.c
+
+misc.o: ${HDR} misc.c
+
+node.o: ${HDR} node.c
+
+print.o: ${HDR} print.c
+
+proc.o: ${HDR} proc.c
+
+store.o: ${HDR} store.c
+
+usage.o: ${HDR} version.h usage.c
+
+util.o: ${HDR} util.c
+
+# *** Do not add anything here - It will go away. ***
--- /dev/null
+#!/bin/sh
+#
+# Mksrc - make Linux source files for /proc-based lsof
+#
+# WARNING: This script assumes it is running from the main directory
+# of the lsof, version 4 distribution.
+#
+# One environment variable applies:
+#
+# LSOF_MKC is the method for creating the source files.
+# It defaults to "ln -s". A common alternative is "cp".
+#
+# $Id: Mksrc,v 1.2 2000/12/04 14:31:02 abe Exp $
+
+
+D=dialects/linux
+L="dfile.c dlsof.h dmnt.c dnode.c dproc.c dproto.h dsock.c dstore.c machine.h"
+
+for i in $L
+do
+ rm -f $i
+ $LSOF_MKC $D/$i $i
+ echo "$LSOF_MKC $D/$i $i"
+done
+
--- /dev/null
+/*
+ * dfile.c - Linux file processing functions for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: dfile.c,v 1.7 2002/02/26 15:20:15 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * printdevname() - print character device name
+ *
+ * Note: this function should not be needed in /proc-based lsof, but
+ * since it is called by printname() in print.c, an ersatz one
+ * is provided here.
+ */
+
+int
+printdevname(dev, rdev, f, nty)
+ dev_t *dev; /* device */
+ dev_t *rdev; /* raw device */
+ int f; /* 1 = follow with '\n' */
+ int nty; /* node type: N_BLK or N_chr */
+{
+ char buf[128];
+
+ (void) snpf(buf, sizeof(buf), "%s device: %d,%d",
+ (nty == N_BLK) ? "BLK" : "CHR",
+ (int)GET_MAJ_DEV(*rdev), (int)GET_MIN_DEV(*rdev));
+ safestrprt(buf, stdout, f);
+ return(1);
+}
--- /dev/null
+/*
+ * dlsof.h - Linux header file for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * $Id: dlsof.h,v 1.18 2008/10/21 16:17:21 abe Exp $
+ */
+
+
+#if !defined(LINUX_LSOF_H)
+#define LINUX_LSOF_H 1
+
+#include <dirent.h>
+#define DIRTYPE dirent /* for arg.c's enter_dir() */
+#define __USE_GNU /* to get all O_* symbols in fcntl.h */
+#include <fcntl.h>
+#include <malloc.h>
+#include <signal.h>
+#include <stdlib.h>
+#include <string.h>
+#include <setjmp.h>
+#include <unistd.h>
+#include <netinet/in.h>
+
+# if defined(GLIBCV)
+#include <netinet/tcp.h>
+# else /* !defined(GLIBCV) */
+#include <linux/tcp.h>
+# endif /* defined(GLIBCV) */
+
+#include <rpc/rpc.h>
+#include <rpc/pmap_prot.h>
+
+#if defined(HASSELINUX)
+#include <selinux/selinux.h>
+#endif /* defined(HASSELINUX) */
+
+#include <sys/sysmacros.h>
+#include <sys/socket.h>
+#include <arpa/inet.h>
+#include <linux/if_ether.h>
+
+
+/*
+ * This definition is needed for the common function prototype definitions
+ * in "proto.h", but isn't used in /proc-based lsof.
+ */
+
+typedef unsigned long KA_T;
+
+
+/*
+ * Local definitions
+ */
+
+#define COMP_P const void
+#define DEVINCR 1024 /* device table malloc() increment */
+#define FSNAMEL 4
+#define MALLOC_P void
+#define FREE_P MALLOC_P
+#define MALLOC_S size_t
+#define MAXSYSCMDL 15 /* max system command name length
+ * This value should be obtained from a
+ * header file #define, but no consistent one
+ * exists. Some versions of the Linux kernel
+ * have a hard-coded "char comm[16]" command
+ * name member of the task structured
+ * definition in <linux/sched.h>, while others
+ * have a "char comm[TASK_COMM_LEN]" member
+ * with TASK_COMM_LEN #define'd to be 16.
+ * Hence, a universal, local definition of
+ * 16 is #define'd here. */
+#define PROCFS "/proc"
+#define QSORT_P void
+#define READLEN_T size_t
+
+/*
+ * Definitions that indicate what values are present in a stat(2) or lstat(2)
+ * buffer.
+ */
+
+#define SB_DEV 0x01 /* st_dev */
+#define SB_INO 0x02 /* st_ino */
+#define SB_MODE 0x04 /* st_mode */
+#define SB_NLINK 0x08 /* st_nlink */
+#define SB_RDEV 0x10 /* st_rdev */
+#define SB_SIZE 0x20 /* st_size */
+#define SB_ALL (SB_DEV | SB_INO | SB_MODE | SB_NLINK | SB_RDEV | \
+ SB_SIZE) /* all values */
+
+#define STRNCPY_L size_t
+#define STRNML 32
+
+# if defined(_FILE_OFFSET_BITS) && _FILE_OFFSET_BITS==64
+#define SZOFFTYPE unsigned long long
+ /* size and offset internal storage
+ * type */
+#define SZOFFPSPEC "ll" /* SZOFFTYPE print specification
+ * modifier */
+# endif /* defined(_FILE_OFFSET_BITS) && _FILE_OFFSET_BITS==64 */
+
+#define XDR_PMAPLIST (xdrproc_t)xdr_pmaplist
+#define XDR_VOID (xdrproc_t)xdr_void
+
+
+/*
+ * Global storage definitions (including their structure definitions)
+ */
+
+struct mounts {
+ char *dir; /* directory (mounted on) */
+ char *fsname; /* file system
+ * (symbolic links unresolved) */
+ char *fsnmres; /* file system
+ * (symbolic links resolved) */
+ dev_t dev; /* directory st_dev */
+ dev_t rdev; /* directory st_rdev */
+ INODETYPE inode; /* directory st_ino */
+ mode_t mode; /* directory st_mode */
+ int ds; /* directory status -- i.e., SB_*
+ * values */
+ mode_t fs_mode; /* file system st_mode */
+ int ty; /* node type -- e.g., N_REGLR, N_NFS */
+ struct mounts *next; /* forward link */
+};
+
+struct sfile {
+ char *aname; /* argument file name */
+ char *name; /* file name (after readlink()) */
+ char *devnm; /* device name (optional) */
+ dev_t dev; /* device */
+ dev_t rdev; /* raw device */
+ mode_t mode; /* S_IFMT mode bits from stat() */
+ int type; /* file type: 0 = file system
+ * 1 = regular file */
+ INODETYPE i; /* inode number */
+ int f; /* file found flag */
+ struct sfile *next; /* forward link */
+};
+
+extern int HasNFS;
+extern int OffType;
+
+#endif /* LINUX_LSOF_H */
--- /dev/null
+/*
+ * dmnt.c -- Linux mount support functions for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: dmnt.c,v 1.17 2008/04/15 13:32:26 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Local definitions
+ */
+
+#if defined(HASMNTSUP)
+#define HASHMNT 128 /* mount supplement hash bucket count
+ * !!!MUST BE A POWER OF 2!!! */
+#endif /* defined(HASMNTSUP) */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static char *cvtoe,(char *os));
+
+#if defined(HASMNTSUP)
+_PROTOTYPE(static int getmntdev,(char *dn, struct stat *s, int *ss));
+_PROTOTYPE(static int hash_mnt,(char *dn));
+#endif /* defined(HASMNTSUP) */
+
+
+/*
+ * Local structure definitions.
+ */
+
+#if defined(HASMNTSUP)
+typedef struct mntsup {
+ char *dn; /* directory name */
+ dev_t dev; /* device number */
+ int ln; /* line on which defined */
+ struct mntsup *next; /* next entry */
+} mntsup_t;
+#endif /* defined(HASMNTSUP) */
+
+
+/*
+ * Local static definitions
+ */
+
+static struct mounts *Lmi = (struct mounts *)NULL; /* local mount info */
+static int Lmist = 0; /* Lmi status */
+static mntsup_t **MSHash = (mntsup_t **)NULL; /* mount supplement
+ * hash buckets */
+
+
+/*
+ * cvtoe() -- convert octal-escaped characters in string
+ */
+
+static char *
+cvtoe(os)
+ char *os; /* original string */
+{
+ int c, cl, cx, ol, ox, tx;
+ char *cs;
+ int tc;
+/*
+ * Allocate space for a copy of the string in which octal-escaped characters
+ * can be replaced by the octal value -- e.g., \040 with ' '. Leave room for
+ * a '\0' terminator.
+ */
+ if (!(ol = (int)strlen(os)))
+ return((char *)NULL);
+ if (!(cs = (char *)malloc(ol + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for octal-escaping.\n",
+ Pn, ol + 1);
+ Exit(1);
+ }
+/*
+ * Copy the string, replacing octal-escaped characters as they are found.
+ */
+ for (cx = ox = 0, cl = ol; ox < ol; ox++) {
+ if (((c = (int)os[ox]) == (int)'\\') && ((ox + 3) < ol)) {
+
+ /*
+ * The beginning of an octal-escaped character has been found.
+ *
+ * Convert the octal value to a character value.
+ */
+ for (tc = 0, tx = 1; os[ox + tx] && (tx < 4); tx++) {
+ if (((int)os[ox + tx] < (int)'0')
+ || ((int)os[ox + tx] > (int)'7'))
+ {
+
+ /*
+ * The escape isn't followed by octets, so ignore the
+ * escape and just copy it.
+ */
+ break;
+ }
+ tc <<= 3;
+ tc += (int)(os[ox + tx] - '0');
+ }
+ if (tx == 4) {
+
+ /*
+ * If three octets (plus the escape) were assembled, use their
+ * character-forming result.
+ *
+ * Otherwise copy the escape and what follows it until another
+ * escape is found.
+ */
+ ox += 3;
+ c = (tc & 0xff);
+ }
+ }
+ if (cx >= cl) {
+
+ /*
+ * Expand the copy string, as required. Leave room for a '\0'
+ * terminator.
+ */
+ cl += 64; /* (Make an arbitrary increase.) */
+ if (!(cs = (char *)realloc(cs, cl + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't realloc %d bytes for octal-escaping.\n",
+ Pn, cl + 1);
+ Exit(1);
+ }
+ }
+ /*
+ * Copy the character.
+ */
+ cs[cx++] = (char)c;
+ }
+/*
+ * Terminate the copy and return its pointer.
+ */
+ cs[cx] = '\0';
+ return(cs);
+}
+
+
+#if defined(HASMNTSUP)
+/*
+ * getmntdev() - get mount device from mount supplement
+ */
+
+static int
+getmntdev(dn, s, ss)
+ char *dn; /* mount point directory name */
+ struct stat *s; /* stat(2) buffer receptor */
+ int *ss; /* stat(2) status result -- i.e., SB_*
+ * values */
+{
+ static int err = 0;
+ int h;
+ mntsup_t *mp, *mpn;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+
+ if (err)
+ return(0);
+ if (!MSHash) {
+
+ /*
+ * No mount supplement hash buckets have been allocated, so read the
+ * mount supplement file and create hash buckets for its entries.
+ */
+ char buf[(MAXPATHLEN*2) + 1], *dp, path[(MAXPATHLEN*2) + 1];
+ dev_t dev;
+ FILE *fs;
+ int ln = 0;
+ size_t sz;
+
+ if ((MntSup != 2) || !MntSupP)
+ return(0);
+ if (!is_readable(MntSupP, 1)) {
+
+ /*
+ * The mount supplement file isn't readable.
+ */
+ err = 1;
+ return(0);
+ }
+ if (!(fs = open_proc_stream(MntSupP, "r", &vbuf, &vsz, 0))) {
+
+ /*
+ * The mount supplement file can't be opened for reading.
+ */
+ if (!Fwarn)
+ (void) fprintf(stderr, "%s: can't open(%s): %s\n",
+ Pn, MntSupP, strerror(errno));
+ err = 1;
+ return(0);
+ }
+ buf[sizeof(buf) - 1] = '\0';
+ /*
+ * Read the mount supplement file.
+ */
+ while (fgets(buf, sizeof(buf) - 1, fs)) {
+ ln++;
+ if ((dp = strchr(buf, '\n')))
+ *dp = '\0';
+ if (buf[0] != '/') {
+
+ /*
+ * The mount supplement line doesn't begin with the absolute
+ * path character '/'.
+ */
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: %s line %d: no path: \"%s\"\n",
+ Pn, MntSupP, ln, buf);
+ err = 1;
+ continue;
+ }
+ if (!(dp = strchr(buf, ' ')) || strncmp(dp + 1, "0x", 2)) {
+
+ /*
+ * The path on the mount supplement line isn't followed by
+ * " 0x".
+ */
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: %s line %d: no device: \"%s\"\n",
+ Pn, MntSupP, ln, buf);
+ err = 1;
+ continue;
+ }
+ sz = (size_t)(dp - buf);
+ (void) strncpy(path, buf, sz);
+ path[sz] = '\0';
+ /*
+ * Assemble the hexadecimal device number of the mount supplement
+ * line.
+ */
+ for (dev = 0, dp += 3; *dp; dp++) {
+ if (!isxdigit((int)*dp))
+ break;
+ if (isdigit((int)*dp))
+ dev = (dev << 4) + (int)*dp - (int)'0';
+ else
+ dev = (dev << 4) + (int)tolower(*dp) - (int)'a' + 10;
+ }
+ if (*dp) {
+
+ /*
+ * The device number couldn't be assembled.
+ */
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: %s line %d: illegal device: \"%s\"\n",
+ Pn, MntSupP, ln, buf);
+ err = 1;
+ continue;
+ }
+ /*
+ * Search the mount supplement hash buckets. (Allocate them as
+ * required.)
+ */
+ if (!MSHash) {
+ if (!(MSHash = (mntsup_t **)calloc(HASHMNT,
+ sizeof(mntsup_t *)))
+ ) {
+ (void) fprintf(stderr,
+ "%s: no space for mount supplement hash buckets\n",
+ Pn);
+ Exit(1);
+ }
+ }
+ h = hash_mnt(path);
+ for (mp = MSHash[h]; mp; mp = mp->next) {
+ if (!strcmp(mp->dn, path))
+ break;
+ }
+ if (mp) {
+
+ /*
+ * A path match was located. If the device number is the
+ * same, skip this mount supplement line. Otherwise, issue
+ * a warning.
+ */
+ if (mp->dev != dev) {
+ (void) fprintf(stderr,
+ "%s: %s line %d path duplicate of %d: \"%s\"\n",
+ Pn, MntSupP, ln, mp->ln, buf);
+ err = 1;
+ }
+ continue;
+ }
+ /*
+ * Allocate and fill a new mount supplement hash entry.
+ */
+ if (!(mpn = (mntsup_t *)malloc(sizeof(mntsup_t)))) {
+ (void) fprintf(stderr,
+ "%s: no space for mount supplement entry: %d \"%s\"\n",
+ Pn, ln, buf);
+ Exit(1);
+ }
+ if (!(mpn->dn = (char *)malloc(sz + 1))) {
+ (void) fprintf(stderr,
+ "%s: no space for mount supplement path: %d \"%s\"\n",
+ Pn, ln, buf);
+ Exit(1);
+ }
+ (void) strcpy(mpn->dn, path);
+ mpn->dev = dev;
+ mpn->ln = ln;
+ mpn->next = MSHash[h];
+ MSHash[h] = mpn;
+ }
+ if (ferror(fs)) {
+ if (!Fwarn)
+ (void) fprintf(stderr, "%s: error reading %s\n",
+ Pn, MntSupP);
+ err = 1;
+ }
+ (void) fclose(fs);
+ if (err) {
+ if (MSHash) {
+ for (h = 0; h < HASHMNT; h++) {
+ for (mp = MSHash[h]; mp; mp = mpn) {
+ mpn = mp->next;
+ if (mp->dn)
+ (void) free((MALLOC_P *)mp->dn);
+ (void) free((MALLOC_P *)mp);
+ }
+ }
+ (void) free((MALLOC_P *)MSHash);
+ MSHash = (mntsup_t **)NULL;
+ }
+ return(0);
+ }
+ }
+/*
+ * If no errors have been detected reading the mount supplement file, search
+ * its hash biuckets for the supplied directory path.
+ */
+ if (err)
+ return(0);
+ h = hash_mnt(dn);
+ for (mp = MSHash[h]; mp; mp = mp->next) {
+ if (!strcmp(dn, mp->dn)) {
+ memset((void *)s, 0, sizeof(struct stat));
+ s->st_dev = mp->dev;
+ *ss |= SB_DEV;
+ return(1);
+ }
+ }
+ return(0);
+}
+
+
+/*
+ * hash_mnt() - hash mount point
+ */
+
+static int
+hash_mnt(dn)
+ char *dn; /* mount point directory name */
+{
+ register int i, h;
+ size_t l;
+
+ if (!(l = strlen(dn)))
+ return(0);
+ if (l == 1)
+ return((int)*dn & (HASHMNT - 1));
+ for (i = h = 0; i < (int)(l - 1); i++) {
+ h ^= ((int)dn[i] * (int)dn[i+1]) << ((i*3)%13);
+ }
+ return(h & (HASHMNT - 1));
+}
+#endif /* defined(HASMNTSUP) */
+
+
+/*
+ * readmnt() - read mount table
+ */
+
+struct mounts *
+readmnt()
+{
+ char buf[MAXPATHLEN], *cp, **fp;
+ char *dn = (char *)NULL;
+ int ds;
+ char *fp0 = (char *)NULL;
+ char *fp1 = (char *)NULL;
+ char *ln;
+ struct mounts *mp;
+ FILE *ms;
+ int nfs;
+ struct stat sb;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+
+ if (Lmi || Lmist)
+ return(Lmi);
+/*
+ * Open access to /proc/mounts, assigning a page size buffer to its stream.
+ */
+ (void) snpf(buf, sizeof(buf), "%s/mounts", PROCFS);
+ ms = open_proc_stream(buf, "r", &vbuf, &vsz, 1);
+/*
+ * Read mount table entries.
+ */
+ while (fgets(buf, sizeof(buf), ms)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) < 3
+ || !fp[0] || !fp[1] || !fp[2])
+ continue;
+ /*
+ * Convert octal-escaped characters in the device name and mounted-on
+ * path name.
+ */
+ if (fp0) {
+ (void) free((FREE_P *)fp0);
+ fp0 = (char *)NULL;
+ }
+ if (fp1) {
+ (void) free((FREE_P *)fp1);
+ fp1 = (char *)NULL;
+ }
+ if (!(fp0 = cvtoe(fp[0])) || !(fp1 = cvtoe(fp[1])))
+ continue;
+ /*
+ * Ignore an entry with a colon in the device name, followed by
+ * "(pid*" -- it's probably an automounter entry.
+ *
+ * Ignore autofs, pipefs, and sockfs entries.
+ */
+ if ((cp = strchr(fp0, ':')) && !strncasecmp(++cp, "(pid", 4))
+ continue;
+ if (!strcasecmp(fp[2], "autofs") || !strcasecmp(fp[2], "pipefs")
+ || !strcasecmp(fp[2], "sockfs"))
+ continue;
+ /*
+ * Interpolate a possible symbolic directory link.
+ */
+ if (dn)
+ (void) free((FREE_P *)dn);
+ dn = fp1;
+ fp1 = (char *)NULL;
+ if (!(ln = Readlink(dn))) {
+ if (!Fwarn){
+ (void) fprintf(stderr,
+ " Output information may be incomplete.\n");
+ }
+ continue;
+ }
+ if (ln != dn) {
+ (void) free((FREE_P *)dn);
+ dn = ln;
+ }
+ if (*dn != '/')
+ continue;
+ /*
+ * Detect an NFS mount point.
+ */
+ if (!(nfs = strcasecmp(fp[2], "nfs")) && !HasNFS)
+ HasNFS = 1;
+ /*
+ * Stat() the directory.
+ */
+ if (statsafely(dn, &sb)) {
+ if (!Fwarn) {
+ (void) fprintf(stderr, "%s: WARNING: can't stat() ", Pn);
+ safestrprt(fp[2], stderr, 0);
+ (void) fprintf(stderr, " file system ");
+ safestrprt(dn, stderr, 1);
+ (void) fprintf(stderr,
+ " Output information may be incomplete.\n");
+ }
+
+#if defined(HASMNTSUP)
+ if ((MntSup == 2) && MntSupP) {
+ ds = 0;
+ if (!getmntdev(dn, &sb, &ds) || !(ds & SB_DEV))
+ continue;
+ (void) fprintf(stderr,
+ " assuming dev=%#lx from %s\n",
+ (long)sb.st_dev, MntSupP);
+ } else
+ continue;
+#else /* !defined(HASMNTSUP) */
+ continue;
+#endif /* defined(HASMNTSUP) */
+
+ } else
+ ds = SB_ALL;
+ /*
+ * Allocate and fill a local mount structure.
+ */
+ if (!(mp = (struct mounts *)malloc(sizeof(struct mounts)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate mounts struct for: ", Pn);
+ safestrprt(dn, stderr, 1);
+ Exit(1);
+ }
+ mp->dir = dn;
+ dn = (char *)NULL;
+ mp->next = Lmi;
+ mp->dev = ((mp->ds = ds) & SB_DEV) ? sb.st_dev : 0;
+ mp->rdev = (ds & SB_RDEV) ? sb.st_rdev : 0;
+ mp->inode = (INODETYPE)((ds & SB_INO) ? sb.st_ino : 0);
+ mp->mode = (ds & SB_MODE) ? sb.st_mode : 0;
+ if (!nfs) {
+ mp->ty = N_NFS;
+ if (HasNFS < 2)
+ HasNFS = 2;
+ } else
+ mp->ty = N_REGLR;
+
+#if defined(HASMNTSUP)
+ /*
+ * If support for the mount supplement file is defined and if the
+ * +m option was supplied, print mount supplement information.
+ */
+ if (MntSup == 1)
+ (void) printf("%s %#lx\n", mp->dir, (long)mp->dev);
+#endif /* defined(HASMNTSUP) */
+
+ /*
+ * Interpolate a possible file system (mounted-on) device name link.
+ */
+ dn = fp0;
+ fp0 = (char *)NULL;
+ mp->fsname = dn;
+ ln = Readlink(dn);
+ dn = (char *)NULL;
+ /*
+ * Stat() the file system (mounted-on) name and add file system
+ * information to the local mount table entry.
+ */
+ if (!ln || statsafely(ln, &sb))
+ sb.st_mode = 0;
+ mp->fsnmres = ln;
+ mp->fs_mode = sb.st_mode;
+ Lmi = mp;
+ }
+/*
+ * Clean up and return the local mount info table address.
+ */
+ (void) fclose(ms);
+ if (dn)
+ (void) free((FREE_P *)dn);
+ if (fp0)
+ (void) free((FREE_P *)fp0);
+ if (fp1)
+ (void) free((FREE_P *)fp1);
+ Lmist = 1;
+ return(Lmi);
+}
--- /dev/null
+/*
+ * dnode.c - Linux node functions for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: dnode.c,v 1.20 2009/03/25 19:22:39 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Local definitions
+ */
+
+#define OFFSET_MAX ((off_t)0x7fffffff) /* this is defined in
+ * .../src/fs/locks.c and not
+ * in a header file */
+#define PIDBUCKS 64 /* PID hash buckets */
+#define HASHPID(pid) (((int)((pid * 31415) >> 3)) & (PIDBUCKS - 1))
+
+
+/*
+ * Local structure definitions
+ */
+
+struct llock {
+ int pid;
+ dev_t dev;
+ INODETYPE inode;
+ char type;
+ struct llock *next;
+};
+
+
+/*
+ * Local definitions
+ */
+
+struct llock **LckH = (struct llock **)NULL; /* PID-hashed locks */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void check_lock,(void));
+
+
+/*
+ * check_lock() - check lock for file *Lf, process *Lp
+ */
+
+static void
+check_lock()
+{
+ int h;
+ struct llock *lp;
+
+ h = HASHPID(Lp->pid);
+ for (lp = LckH[h]; lp; lp = lp->next) {
+ if (Lp->pid == lp->pid
+ && Lf->dev == lp->dev
+ && Lf->inode == lp->inode)
+ {
+ Lf->lock = lp->type;
+ return;
+ }
+ }
+}
+
+
+/*
+ * get_fields() - separate a line into fields
+ */
+
+int
+get_fields(ln, sep, fr, eb, en)
+ char *ln; /* input line */
+ char *sep; /* separator list */
+ char ***fr; /* field pointer return address */
+ int *eb; /* indexes of fields where blank or an
+ * entry from the separator list may be
+ * embedded and are not separators
+ * (may be NULL) */
+ int en; /* number of entries in eb[] (may be
+ * zero) */
+{
+ char *bp, *cp, *sp;
+ int i, j, n;
+ MALLOC_S len;
+ static char **fp = (char **)NULL;
+ static int nfpa = 0;
+
+ for (cp = ln, n = 0; cp && *cp;) {
+ for (bp = cp; *bp && (*bp == ' ' || *bp == '\t'); bp++);
+ ;
+ if (!*bp || *bp == '\n')
+ break;
+ for (cp = bp; *cp; cp++) {
+ if (*cp == '\n') {
+ *cp = '\0';
+ break;
+ }
+ if (*cp == '\t') /* TAB is always a separator */
+ break;
+ if (*cp == ' ') {
+
+ /*
+ * See if this field may have an embedded space.
+ */
+ if (!eb || !en)
+ break;
+ else {
+ for (i = j = 0; i < en; i++) {
+ if (eb[i] == n) {
+ j = 1;
+ break;
+ }
+ }
+ if (!j)
+ break;
+ }
+ }
+ if (sep) {
+
+ /*
+ * See if the character is in the separator list.
+ */
+ for (sp = sep; *sp; sp++) {
+ if (*sp == *cp)
+ break;
+ }
+ if (*sp) {
+
+ /*
+ * See if this field may have an embedded separator.
+ */
+ if (!eb || !en)
+ break;
+ else {
+ for (i = j = 0; i < en; i++) {
+ if (eb[i] == n) {
+ j = 1;
+ break;
+ }
+ }
+ if (!j)
+ break;
+ }
+ }
+ }
+ }
+ if (*cp)
+ *cp++ = '\0';
+ if (n >= nfpa) {
+ nfpa += 32;
+ len = (MALLOC_S)(nfpa * sizeof(char *));
+ if (fp)
+ fp = (char **)realloc((MALLOC_P *)fp, len);
+ else
+ fp = (char **)malloc(len);
+ if (!fp) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for field pointers.\n",
+ Pn, len);
+ Exit(1);
+ }
+ }
+ fp[n++] = bp;
+ }
+ *fr = fp;
+ return(n);
+}
+
+
+/*
+ * get_locks() - get lock information from /proc/locks
+ */
+
+void
+get_locks(p)
+ char *p; /* /proc lock path */
+{
+ unsigned long bp, ep;
+ char buf[MAXPATHLEN], *ec, **fp;
+ dev_t dev;
+ int ex, i, h, mode, pid;
+ INODETYPE inode;
+ struct llock *lp, *np;
+ FILE *ls;
+ long maj, min;
+ char type;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+/*
+ * Destroy previous lock information.
+ */
+ if (LckH) {
+ for (i = 0; i < PIDBUCKS; i++) {
+ for (lp = LckH[i]; lp; lp = np) {
+ np = lp->next;
+ (void) free((FREE_P *)lp);
+ }
+ LckH[i] = (struct llock *)NULL;
+ }
+ } else {
+
+ /*
+ * If first time, allocate the lock PID hash buckets.
+ */
+ LckH = (struct llock **)calloc((MALLOC_S)PIDBUCKS,
+ sizeof(struct llock *));
+ if (!LckH) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d lock hash bytes\n",
+ Pn, sizeof(struct llock *) * PIDBUCKS);
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc lock file, assign a page size buffer to its stream,
+ * and read it.
+ */
+ if (!(ls = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ while (fgets(buf, sizeof(buf), ls)) {
+ if (get_fields(buf, ":", &fp, (int *)NULL, 0) < 10)
+ continue;
+ if (!fp[1] || strcmp(fp[1], "->") == 0)
+ continue;
+ /*
+ * Get lock type.
+ */
+ if (!fp[3])
+ continue;
+ if (*fp[3] == 'R')
+ mode = 0;
+ else if (*fp[3] == 'W')
+ mode = 1;
+ else
+ continue;
+ /*
+ * Get PID.
+ */
+ if (!fp[4] || !*fp[4])
+ continue;
+ pid = atoi(fp[4]);
+ /*
+ * Get device number.
+ */
+ ec = (char *)NULL;
+ if (!fp[5] || !*fp[5]
+ || (maj = strtol(fp[5], &ec, 16)) == LONG_MIN || maj == LONG_MAX
+ || !ec || *ec)
+ continue;
+ ec = (char *)NULL;
+ if (!fp[6] || !*fp[6]
+ || (min = strtol(fp[6], &ec, 16)) == LONG_MIN || min == LONG_MAX
+ || !ec || *ec)
+ continue;
+ dev = (dev_t)makedev((int)maj, (int)min);
+ /*
+ * Get inode number.
+ */
+ ec = (char *)NULL;
+ if (!fp[7] || !*fp[7]
+ || (inode = strtoull(fp[7], &ec, 0)) == ULONG_MAX
+ || !ec || *ec)
+ continue;
+ /*
+ * Get lock extent. Convert it and the lock type to a lock character.
+ */
+ if (!fp[8] || !*fp[8] || !fp[9] || !*fp[9])
+ continue;
+ ec = (char *)NULL;
+ if ((bp = strtoul(fp[8], &ec, 0)) == ULONG_MAX || !ec || *ec)
+ continue;
+ if (!strcmp(fp[9], "EOF")) /* for Linux 2.4.x */
+ ep = OFFSET_MAX;
+ else {
+ ec = (char *)NULL;
+ if ((ep = strtoul(fp[9], &ec, 0)) == ULONG_MAX || !ec || *ec)
+ continue;
+ }
+ ex = ((off_t)bp == (off_t)0 && (off_t)ep == OFFSET_MAX) ? 1 : 0;
+ if (mode)
+ type = ex ? 'W' : 'w';
+ else
+ type = ex ? 'R' : 'r';
+ /*
+ * Look for this lock via the hash buckets.
+ */
+ h = HASHPID(pid);
+ for (lp = LckH[h]; lp; lp = lp->next) {
+ if (lp->pid == pid
+ && lp->dev == dev
+ && lp->inode == inode
+ && lp->type == type)
+ break;
+ }
+ if (lp)
+ continue;
+ /*
+ * Allocate a new llock structure and link it to the PID hash bucket.
+ */
+ if (!(lp = (struct llock *)malloc(sizeof(struct llock)))) {
+ (void) snpf(buf, sizeof(buf), InodeFmt_d, inode);
+ (void) fprintf(stderr,
+ "%s: can't allocate llock: PID %d; dev %x; inode %s\n",
+ Pn, pid, (int)dev, buf);
+ Exit(1);
+ }
+ lp->pid = pid;
+ lp->dev = dev;
+ lp->inode = inode;
+ lp->type = type;
+ lp->next = LckH[h];
+ LckH[h] = lp;
+ }
+ (void) fclose(ls);
+}
+
+
+/*
+ * process_proc_node() - process file node
+ */
+
+void
+process_proc_node(p, s, ss, l, ls)
+ char *p; /* node's readlink() path */
+ struct stat *s; /* stat() result for path */
+ int ss; /* *s status -- i.e., SB_* values */
+ struct stat *l; /* lstat() result for FD (NULL for
+ * others) */
+ int ls; /* *l status -- i.e., SB_* values */
+{
+ mode_t access;
+ mode_t type = 0;
+ char *cp;
+ struct mounts *mp = (struct mounts *)NULL;
+ size_t sz;
+ char *tn;
+/*
+ * Set the access mode, if possible.
+ */
+ if (l && (ls & SB_MODE) && ((l->st_mode & S_IFMT) == S_IFLNK)) {
+ if ((access = l->st_mode & (S_IRUSR | S_IWUSR)) == S_IRUSR)
+ Lf->access = 'r';
+ else if (access == S_IWUSR)
+ Lf->access = 'w';
+ else
+ Lf->access = 'u';
+ }
+/*
+ * Determine node type.
+ */
+ if (ss & SB_MODE) {
+ type = s->st_mode & S_IFMT;
+ switch (type) {
+ case S_IFBLK:
+ Ntype = N_BLK;
+ break;
+ case S_IFCHR:
+ Ntype = N_CHR;
+ break;
+ case S_IFIFO:
+ Ntype = N_FIFO;
+ break;
+ case S_IFSOCK:
+ process_proc_sock(p, s, ss, l, ls);
+ return;
+ }
+ }
+ if (Selinet)
+ return;
+/*
+ * Save the device. If it is an NFS device, change the node type to N_NFS.
+ */
+ if (ss & SB_DEV) {
+ Lf->dev = s->st_dev;
+ Lf->dev_def = 1;
+ }
+ if ((Ntype == N_CHR || Ntype == N_BLK)) {
+ if (ss & SB_RDEV) {
+ Lf->rdev = s->st_rdev;
+ Lf->rdev_def = 1;
+ }
+ }
+ if (Ntype == N_REGLR && (HasNFS == 2)) {
+ for (mp = readmnt(); mp; mp = mp->next) {
+ if ((mp->ty == N_NFS)
+ && (mp->ds & SB_DEV) && (Lf->dev == mp->dev)
+ ) {
+ Ntype = N_NFS;
+ break;
+ }
+ }
+ }
+/*
+ * Save the inode number.
+ */
+ if (ss & SB_INO) {
+ Lf->inode = (INODETYPE)s->st_ino;
+ Lf->inp_ty = 1;
+ }
+/*
+ * Check for a lock.
+ */
+ if (Lf->dev_def && (Lf->inp_ty == 1))
+ (void) check_lock();
+/*
+ * Save the file size.
+ */
+ switch (Ntype) {
+ case N_BLK:
+ case N_CHR:
+ case N_FIFO:
+ if (!Fsize && l && (ls & SB_SIZE) && OffType) {
+ Lf->off = (SZOFFTYPE)l->st_size;
+ Lf->off_def = 1;
+ }
+ break;
+ default:
+ if (Foffset) {
+ if (l && (ls & SB_SIZE) && OffType) {
+ Lf->off = (SZOFFTYPE)l->st_size;
+ Lf->off_def = 1;
+ }
+ } else if (!Foffset || Fsize) {
+ if (ss & SB_SIZE) {
+ Lf->sz = (SZOFFTYPE)s->st_size;
+ Lf->sz_def = 1;
+ }
+ }
+ }
+/*
+ * Record the link count.
+ */
+ if (Fnlink && (ss & SB_NLINK)) {
+ Lf->nlink = (long)s->st_nlink;
+ Lf->nlink_def = 1;
+ if (Nlink && (Lf->nlink < Nlink))
+ Lf->sf |= SELNLINK;
+ }
+/*
+ * Format the type name.
+ */
+ if (ss & SB_MODE) {
+ switch (type) {
+ case S_IFBLK:
+ tn = "BLK";
+ break;
+ case S_IFCHR:
+ tn = "CHR";
+ break;
+ case S_IFDIR:
+ tn = "DIR";
+ break;
+ case S_IFIFO:
+ tn = "FIFO";
+ break;
+ case S_IFREG:
+ tn = "REG";
+ break;
+ case S_IFLNK:
+ tn = "LINK";
+ break;
+ case S_ISVTX:
+ tn = "VTXT";
+ break;
+ default:
+ (void) snpf(Lf->type, sizeof(Lf->type), "%04o",
+ ((type >> 12) & 0xf));
+ tn = (char *)NULL;
+ }
+ } else
+ tn = "unknown";
+ if (tn)
+ (void) snpf(Lf->type, sizeof(Lf->type), "%s", tn);
+ Lf->ntype = Ntype;
+/*
+ * Record an NFS file selection.
+ */
+ if (Ntype == N_NFS && Fnfs)
+ Lf->sf |= SELNFS;
+/*
+ * Test for specified file.
+ */
+ if (Sfile
+ && is_file_named((char *)NULL,
+ ((type == S_IFCHR) || (type == S_IFBLK)) ? 1 : 0))
+ Lf->sf |= SELNM;
+/*
+ * If no NAME information has been stored, store the path.
+ *
+ * Store the remote host and mount point for an NFS file.
+ */
+ if (!Namech[0]) {
+ (void) snpf(Namech, Namechl, "%s", p);
+ if ((Ntype == N_NFS) && mp && mp->fsname) {
+ cp = endnm(&sz);
+ (void) snpf(cp, sz, " (%s)", mp->fsname);
+ }
+ }
+ if (Namech[0])
+ enter_nm(Namech);
+}
--- /dev/null
+/*
+ * dproc.c - Linux process access functions for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: dproc.c,v 1.22 2008/10/21 16:17:21 abe Exp $";
+#endif
+
+#include "lsof.h"
+
+
+/*
+ * Local definitions
+ */
+
+#define FDINFO_FLAGS 1 /* fdinfo flags available */
+#define FDINFO_POS 2 /* fdinfo position available */
+#define FDINFO_ALL (FDINFO_FLAGS | FDINFO_POS)
+#define LSTAT_TEST_FILE "/"
+#define LSTAT_TEST_SEEK 1
+
+#if !defined(ULLONG_MAX)
+#define ULLONG_MAX 18446744073709551615ULL
+#endif /* !defined(ULLONG_MAX) */
+
+
+/*
+ * Local structures
+ */
+
+struct l_fdinfo {
+ int flags; /* flags: line value */
+ off_t pos; /* pos: line value */
+};
+
+
+/*
+ * Local variables
+ */
+
+static short Cckreg; /* conditional status of regular file
+ * checking:
+ * 0 = unconditionally check
+ * 1 = conditionally check */
+static short Ckscko; /* socket file only checking status:
+ * 0 = none
+ * 1 = check only socket files */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static int get_fdinfo,(char *p, struct l_fdinfo *fi));
+_PROTOTYPE(static int getlinksrc,(char *ln, char *src, int srcl));
+_PROTOTYPE(static int nm2id,(char *nm, int *id, int *idl));
+_PROTOTYPE(static int read_id_stat,(int ty, char *p, int id, char **cmd,
+ int *ppid, int *pgid));
+_PROTOTYPE(static void process_proc_map,(char *p, struct stat *s, int ss));
+_PROTOTYPE(static int process_id,(char *idp, int idpl, char *cmd, UID_ARG uid,
+ int pid, int ppid, int pgid));
+_PROTOTYPE(static int statEx,(char *p, struct stat *s, int *ss));
+
+
+#if defined(HASSELINUX)
+_PROTOTYPE(static int cmp_cntx_eq,(char *pcntx, char *ucntx));
+
+
+#include <fnmatch.h>
+
+
+/*
+ * cmp_cntx_eq -- compare program and user security contexts
+ */
+
+static int
+cmp_cntx_eq(pcntx, ucntx)
+ char *pcntx; /* program context */
+ char *ucntx; /* user supplied context */
+{
+ return !fnmatch(ucntx, pcntx, 0);
+}
+
+
+/*
+ * enter_cntx_arg() - enter name ecurity context argument
+ */
+
+int
+enter_cntx_arg(cntx)
+ char *cntx; /* context */
+{
+ cntxlist_t *cntxp;
+/*
+ * Search the argument list for a duplicate.
+ */
+ for (cntxp = CntxArg; cntxp; cntxp = cntxp->next) {
+ if (!strcmp(cntxp->cntx, cntx)) {
+ if (!Fwarn) {
+ (void) fprintf(stderr, "%s: duplicate context: %s\n",
+ Pn, cntx);
+ }
+ return(1);
+ }
+ }
+/*
+ * Create and link a new context argument list entry.
+ */
+ if (!(cntxp = (cntxlist_t *)malloc((MALLOC_S)sizeof(cntxlist_t)))) {
+ (void) fprintf(stderr, "%s: no space for context: %s\n", Pn, cntx);
+ Exit(1);
+ }
+ cntxp->f = 0;
+ cntxp->cntx = cntx;
+ cntxp->next = CntxArg;
+ CntxArg = cntxp;
+ return(0);
+}
+#endif /* defined(HASSELINUX) */
+
+
+/*
+ * gather_proc_info() -- gather process information
+ */
+
+void
+gather_proc_info()
+{
+ char *cmd;
+ struct dirent *dp;
+ struct stat sb;
+ int lwp, n, nl, pgid, pid, ppid, rv, tx;
+ static char *lwppath = (char *)NULL;
+ static int lwppathl = 0;
+ static char *path = (char *)NULL;
+ static int pathl = 0;
+ static char *pidpath = (char *)NULL;
+ static MALLOC_S pidpathl = 0;
+ static MALLOC_S pidx = 0;
+ static DIR *ps = (DIR *)NULL;
+ static char *taskpath = (char *)NULL;
+ static int taskpathl = 0;
+ DIR *ts;
+ UID_ARG uid;
+
+/*
+ * Do one-time setup.
+ */
+ if (!pidpath) {
+ pidx = strlen(PROCFS) + 1;
+ pidpathl = pidx + 64 + 1; /* 64 is growth room */
+ if (!(pidpath = (char *)malloc(pidpathl))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for \"%s/\"<pid>\n",
+ Pn, pidpathl, PROCFS);
+ Exit(1);
+ }
+ (void) snpf(pidpath, pidpathl, "%s/", PROCFS);
+ }
+/*
+ * Get lock and net information.
+ */
+ (void) make_proc_path(pidpath, pidx, &path, &pathl, "locks");
+ (void) get_locks(path);
+ (void) make_proc_path(pidpath, pidx, &path, &pathl, "net/");
+ (void) set_net_paths(path, strlen(path));
+/*
+ * If only socket files have been selected, or socket files have been selected
+ * ANDed with other selection options, enable the skipping of regular files.
+ *
+ * If socket files and some process options have been selected, enable
+ * conditional skipping of regular file; i.e., regular files will be skipped
+ * unless they belong to a process selected by one of the specified options.
+ */
+ if (Selflags & SELNW) {
+
+ /*
+ * Some network files selection options have been specified.
+ */
+ if (Fand || !(Selflags & ~SELNW)) {
+
+ /*
+ * Selection ANDing or only network file options have been
+ * specified, so set unconditional skipping of regular files
+ * and socket file only checking.
+ */
+ Cckreg = 0;
+ Ckscko = 1;
+ } else {
+
+ /*
+ * If ORed file selection options have been specified, or no ORed
+ * process selection options have been specified, enable
+ * unconditional file checking and clear socket file only checking.
+ *
+ * If only ORed process selection options have been specified,
+ * enable conditional file skipping and socket file only checking.
+ */
+ if ((Selflags & SELFILE) || !(Selflags & SELPROC))
+ Cckreg = Ckscko = 0;
+ else
+ Cckreg = Ckscko = 1;
+ }
+ } else {
+
+ /*
+ * No network file selection options were specified. Enable
+ * unconditional file checking and clear socket file only checking.
+ */
+ Cckreg = Ckscko = 0;
+ }
+/*
+ * Read /proc, looking for PID directories. Open each one and
+ * gather its process and file information.
+ */
+ if (!ps) {
+ if (!(ps = opendir(PROCFS))) {
+ (void) fprintf(stderr, "%s: can't open %s\n", Pn, PROCFS);
+ Exit(1);
+ }
+ } else
+ (void) rewinddir(ps);
+ while ((dp = readdir(ps))) {
+ if (nm2id(dp->d_name, &pid, &n))
+ continue;
+ /*
+ * Build path to PID's directory.
+ */
+ if ((pidx + n + 1 + 1) > pidpathl) {
+ pidpathl = pidx + n + 1 + 1 + 64;
+ if (!(pidpath = (char *)realloc((MALLOC_P *)pidpath, pidpathl)))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for \"%s/%s/\"\n",
+ Pn, pidpathl, PROCFS, dp->d_name);
+ Exit(1);
+ }
+ }
+ (void) snpf(pidpath + pidx, pidpathl - pidx, "%s/", dp->d_name);
+ n += (pidx + 1);
+ /*
+ * Process the PID's stat info.
+ */
+ if (stat(pidpath, &sb))
+ continue;
+ uid = (UID_ARG)sb.st_uid;
+ /*
+ * Process the PID's process information.
+ */
+ (void) make_proc_path(pidpath, n, &path, &pathl, "stat");
+ rv = read_id_stat(0, path, pid, &cmd, &ppid, &pgid);
+ if (rv == 1)
+ continue;
+ else if (rv == 0) {
+ (void) process_id(pidpath, n, cmd, uid, pid, ppid, pgid);
+ continue;
+ }
+ /*
+ * The process is a zombie. Check for a non-zombie task.
+ */
+ (void) make_proc_path(pidpath, n, &taskpath, &taskpathl, "task");
+ tx = n + 4;
+ if ((ts = opendir(taskpath))) {
+
+ /*
+ * Process the PID's tasks (lightweight processes.) Record the
+ * open files of the first one whose LWP ID does not match the PID
+ * and which is not a itself a zombie.
+ */
+ while ((dp = readdir(ts))) {
+
+ /*
+ * Get the LWP ID. Skip the task if its LWP ID matches the
+ * process PID.
+ */
+ if (nm2id(dp->d_name, &lwp, &nl))
+ continue;
+ if (lwp == pid)
+ continue;
+ /*
+ * Check the LWP state.
+ */
+ if (read_id_stat(1, path, lwp, &cmd, &ppid, &pgid))
+ continue;
+ /*
+ * Attempt to record the LWP.
+ */
+ if ((tx + 1 + nl + 1) > lwppathl) {
+ lwppathl = tx + 1 + n + 1 + 64;
+ if (lwppath)
+ lwppath = (char *)realloc((MALLOC_P *)lwppath,
+ lwppathl);
+ else
+ lwppath = (char *)malloc((MALLOC_S)lwppathl);
+ if (!lwppath) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d task bytes", Pn,
+ lwppathl);
+ (void) fprintf(stderr,
+ " for \"%s/%s/\"\n", taskpath, dp->d_name);
+ Exit(1);
+ }
+ }
+ (void) snpf(lwppath, lwppathl, "%s/%s/", taskpath,
+ dp->d_name);
+ if (!process_id(lwppath, (tx + 1 + nl+ 1), cmd, uid, pid,
+ ppid, pgid))
+ {
+ break;
+ }
+ }
+ (void) closedir(ts);
+ }
+ }
+}
+
+
+/*
+ * get_fdinfo() - get values from /proc/<PID>fdinfo/FD
+ */
+
+static int
+get_fdinfo(p, fi)
+ char *p; /* path to fdinfo file */
+ struct l_fdinfo *fi; /* pointer to local fdinfo values
+ * return structure */
+{
+ char buf[MAXPATHLEN + 1], *ep, **fp;
+ FILE *fs;
+ int rv = 0;
+ unsigned long ul;
+ unsigned long long ull;
+/*
+ * Signal no values returned (0) if no fdinfo pointer was provided or if the
+ * fdinfo path can't be opened.
+ */
+ if (!fi)
+ return(0);
+ if (!p || !*p || !(fs = fopen(p, "r")))
+ return(0);
+/*
+ * Read the fdinfo file.
+ */
+ while (fgets(buf, sizeof(buf), fs)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) < 2)
+ continue;
+ if (!fp[0] || !*fp[0] || !fp[1] || !*fp[1])
+ continue;
+ if (!strcmp(fp[0], "flags:")) {
+
+ /*
+ * Process a "flags:" line.
+ */
+ ep = (char *)NULL;
+ if ((ul = strtoul(fp[1], &ep, 0)) == ULONG_MAX
+ || !ep || *ep)
+ continue;
+ fi->flags = (unsigned int)ul;
+ if ((rv |= FDINFO_FLAGS) == FDINFO_ALL)
+ break;
+ } else if (!strcmp(fp[0], "pos:")) {
+
+ /*
+ * Process a "pos:" line.
+ */
+ ep = (char *)NULL;
+ if ((ull = strtoull(fp[1], &ep, 0)) == ULLONG_MAX
+ || !ep || *ep)
+ continue;
+ fi->pos = (off_t)ull;
+ if ((rv |= FDINFO_POS) == FDINFO_ALL)
+ break;
+ }
+ }
+ fclose(fs);
+/*
+ * Signal via the return value what information was obtained. (0 == none)
+ */
+ return(rv);
+}
+
+
+/*
+ * getlinksrc() - get the source path name for the /proc/<PID>/fd/<FD> link
+ */
+
+
+static int
+getlinksrc(ln, src, srcl)
+ char *ln; /* link path */
+ char *src; /* link source path return address */
+ int srcl; /* length of src[] */
+{
+ char *cp;
+ int ll;
+
+ if ((ll = readlink(ln, src, srcl - 1)) < 1
+ || ll >= srcl)
+ return(-1);
+ src[ll] = '\0';
+ if (*src == '/')
+ return(ll);
+ if ((cp = strchr(src, ':'))) {
+ *cp = '\0';
+ ll = strlen(src);
+ }
+ return(ll);
+}
+
+
+/*
+ * initialize() - perform all initialization
+ */
+
+void
+initialize()
+{
+ int fd;
+ struct l_fdinfo fi;
+ char path[MAXPATHLEN];
+ struct stat sb;
+/*
+ * Test for -i and -X option conflict.
+ */
+ if (Fxopt && (Fnet || Nwad)) {
+ (void) fprintf(stderr, "%s: -i is useless when -X is specified.\n",
+ Pn);
+ usage(1, 0, 0);
+ }
+/*
+ * Open LSTAT_TEST_FILE and seek to byte LSTAT_TEST_SEEK, then lstat the
+ * /proc/<PID>/fd/<FD> for LSTAT_TEST_FILE to see what position is reported.
+ * If the result is LSTAT_TEST_SEEK, enable offset reporting.
+ *
+ * If the result isn't LSTAT_TEST_SEEK, next check the fdinfo file for the
+ * open LSTAT_TEST_FILE file descriptor. If it exists and contains a "pos:"
+ * value, and if the value is LSTAT_TEST_SEEK, enable offset reporting.
+ */
+ if ((fd = open(LSTAT_TEST_FILE, O_RDONLY)) >= 0) {
+ if (lseek(fd, (off_t)LSTAT_TEST_SEEK, SEEK_SET)
+ == (off_t)LSTAT_TEST_SEEK) {
+ (void) snpf(path, sizeof(path), "%s/%d/fd/%d", PROCFS, Mypid,
+ fd);
+ if (!lstat(path, &sb)) {
+ if (sb.st_size == (off_t)LSTAT_TEST_SEEK)
+ OffType = 1;
+ }
+ }
+ if (!OffType) {
+ (void) snpf(path, sizeof(path), "%s/%d/fdinfo/%d", PROCFS,
+ Mypid, fd);
+ if (get_fdinfo(path, &fi) & FDINFO_POS) {
+ if (fi.pos == (off_t)LSTAT_TEST_SEEK)
+ OffType = 2;
+ }
+ }
+ (void) close(fd);
+ }
+ if (!OffType) {
+ if (Foffset && !Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't report offset; disregarding -o.\n",
+ Pn);
+ Foffset = 0;
+ Fsize = 1;
+ }
+ if (Fsv && (OffType != 2)) {
+ if (!Fwarn && FsvByf)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't report file flags; disregarding +f.\n",
+ Pn);
+ Fsv = 0;
+ }
+/*
+ * Make sure the local mount info table is loaded if doing anything other
+ * than just Internet lookups. (HasNFS is defined during the loading of the
+ * local mount table.)
+ */
+ if (Selinet == 0)
+ (void) readmnt();
+}
+
+
+/*
+ * make_proc_path() - make a path in a /proc directory
+ *
+ * entry:
+ * pp = pointer to /proc prefix
+ * lp = length of prefix
+ * np = pointer to malloc'd buffer to receive new file's path
+ * nl = length of new file path buffer
+ * sf = new path's suffix
+ *
+ * return: length of new path
+ * np = updated with new path
+ * nl = updated with new path length
+ */
+
+int
+make_proc_path(pp, pl, np, nl, sf)
+ char *pp; /* path prefix -- e.g., /proc/<pid>/ */
+ int pl; /* strlen(pp) */
+ char **np; /* malloc'd receiving buffer */
+ int *nl; /* strlen(*np) */
+ char *sf; /* suffix of new path */
+{
+ char *cp;
+ MALLOC_S rl, sl;
+
+ sl = strlen(sf);
+ if ((rl = pl + sl + 1) > *nl) {
+ if ((cp = *np))
+ cp = (char *)realloc((MALLOC_P *)cp, rl);
+ else
+ cp = (char *)malloc(rl);
+ if (!cp) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for %s%s\n",
+ Pn, rl, pp, sf);
+ Exit(1);
+ }
+ *nl = rl;
+ *np = cp;
+ }
+ (void) snpf(*np, *nl, "%s", pp);
+ (void) snpf(*np + pl, *nl - pl, "%s", sf);
+ return(rl - 1);
+}
+
+
+/*
+ * nm2id() - convert a name to an integer ID
+ */
+
+static int
+nm2id(nm, id, idl)
+ char *nm; /* pointer to name */
+ int *id; /* pointer to ID receiver */
+ int *idl; /* pointer to ID length receiver */
+{
+ register int tid, tidl;
+
+ for (*id = *idl = tid = tidl = 0; *nm; nm++) {
+
+#if defined(__STDC__) /* { */
+ if (!isdigit((unsigned char)*nm))
+#else /* !defined(__STDC__) } { */
+ if (!isascii(*nm) || !isdigit((unsigned char)*cp))
+#endif /* defined(__STDC__) } */
+
+ {
+ return(1);
+ }
+ tid = tid * 10 + (int)(*nm - '0');
+ tidl++;
+ }
+ *id = tid;
+ *idl = tidl;
+ return(0);
+}
+
+
+/*
+ * open_proc_stream() -- open a /proc stream
+ */
+
+FILE *
+open_proc_stream(p, m, buf, sz, act)
+ char *p; /* pointer to path to open */
+ char *m; /* pointer to mode -- e.g., "r" */
+ char **buf; /* pointer tp setvbuf() address
+ * (NULL if none) */
+ size_t *sz; /* setvbuf() size (0 if none or if
+ * getpagesize() desired */
+ int act; /* fopen() failure action:
+ * 0 : return (FILE *)NULL
+ * <>0 : fprintf() an error message
+ * and Exit(1)
+ */
+{
+ FILE *fs; /* opened stream */
+ static size_t psz = (size_t)0; /* page size */
+ size_t tsz; /* temporary size */
+/*
+ * Open the stream.
+ */
+ if (!(fs = fopen(p, m))) {
+ if (!act)
+ return((FILE *)NULL);
+ (void) fprintf(stderr, "%s: can't fopen(%s, \"%s\"): %s\n",
+ Pn, p, m, strerror(errno));
+ Exit(1);
+ }
+/*
+ * Return the stream if no buffer change is required.
+ */
+ if (!buf)
+ return(fs);
+/*
+ * Determine the buffer size required.
+ */
+ if (!(tsz = *sz)) {
+ if (!psz)
+ psz = getpagesize();
+ tsz = psz;
+ }
+/*
+ * Allocate a buffer for the stream, as required.
+ */
+ if (!*buf) {
+ if (!(*buf = (char *)malloc((MALLOC_S)tsz))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for %s stream buffer\n",
+ Pn, (int)tsz, p);
+ Exit(1);
+ }
+ *sz = tsz;
+ }
+/*
+ * Assign the buffer to the stream.
+ */
+ if (setvbuf(fs, *buf, _IOFBF, tsz)) {
+ (void) fprintf(stderr, "%s: setvbuf(%s)=%d failure: %s\n",
+ Pn, p, (int)tsz, strerror(errno));
+ Exit(1);
+ }
+ return(fs);
+}
+
+
+/*
+ * process_id - process ID: PID or LWP
+ *
+ * return: 0 == ID processed
+ * 1 == ID not processed
+ */
+
+static int
+process_id(idp, idpl, cmd, uid, pid, ppid, pgid)
+ char *idp; /* pointer to ID's path */
+ int idpl; /* pointer to ID's path length */
+ char *cmd; /* pointer to ID's command */
+ UID_ARG uid; /* ID's UID */
+ int pid; /* ID's PID */
+ int ppid; /* parent PID */
+ int pgid; /* parent GID */
+{
+ int av;
+ static char *dpath = (char *)NULL;
+ static int dpathl = 0;
+ short enls, enss, lnk, oty, pn, pss, sf;
+ int fd, i, ls, n, ss, sv;
+ struct l_fdinfo fi;
+ DIR *fdp;
+ struct dirent *fp;
+ static char *ipath = (char *)NULL;
+ static int ipathl = 0;
+ int j = 0;
+ struct stat lsb, sb;
+ char nmabuf[MAXPATHLEN + 1], pbuf[MAXPATHLEN + 1];
+ static char *path = (char *)NULL;
+ static int pathl = 0;
+ static char *pathi = (char *)NULL;
+ static int pathil = 0;
+ int txts = 0;
+
+#if defined(HASSELINUX)
+ cntxlist_t *cntxp;
+#endif /* defined(HASSELINUX) */
+
+/*
+ * See if process is excluded.
+ */
+ if (is_proc_excl(pid, pgid, uid, &pss, &sf)
+ || is_cmd_excl(cmd, &pss, &sf))
+ return(1);
+ if (Cckreg) {
+
+ /*
+ * If conditional checking of regular files is enabled, enable
+ * socket file only checking, based on the process' selection
+ * status.
+ */
+ Ckscko = (sf & SELPROC) ? 0 : 1;
+ }
+ alloc_lproc(pid, pgid, ppid, uid, cmd, (int)pss, (int)sf);
+ Plf = (struct lfile *)NULL;
+/*
+ * Process the ID's current working directory info.
+ */
+ if (!Ckscko) {
+ (void) make_proc_path(idp, idpl, &path, &pathl, "cwd");
+ alloc_lfile(CWD, -1);
+ if (getlinksrc(path, pbuf, sizeof(pbuf)) < 1) {
+ if (!Fwarn) {
+ (void) memset((void *)&sb, 0, sizeof(sb));
+ lnk = ss = 0;
+ (void) snpf(nmabuf, sizeof(nmabuf), "(readlink: %s)",
+ strerror(errno));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ pn = 1;
+ } else
+ pn = 0;
+ } else {
+ lnk = pn = 1;
+ ss = SB_ALL;
+ if (HasNFS) {
+ if ((sv = statsafely(path, &sb)))
+ sv = statEx(pbuf, &sb, &ss);
+ } else
+ sv = stat(path, &sb);
+ if (sv) {
+ ss = 0;
+ if (!Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "(stat: %s)",
+ strerror(errno));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ }
+ }
+ if (pn) {
+ (void) process_proc_node(lnk ? pbuf : path,
+ &sb, ss,
+ (struct stat *)NULL, 0);
+ if (Lf->sf)
+ link_lfile();
+ }
+ }
+/*
+ * Process the ID's root directory info.
+ */
+ if (!Ckscko) {
+ (void) make_proc_path(idp, idpl, &path, &pathl, "root");
+ alloc_lfile(RTD, -1);
+ if (getlinksrc(path, pbuf, sizeof(pbuf)) < 1) {
+ if (!Fwarn) {
+ (void) memset((void *)&sb, 0, sizeof(sb));
+ lnk = ss = 0;
+ (void) snpf(nmabuf, sizeof(nmabuf), "(readlink: %s)",
+ strerror(errno));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ pn = 1;
+ } else
+ pn = 0;
+ } else {
+ lnk = pn = 1;
+ ss = SB_ALL;
+ if (HasNFS) {
+ if ((sv = statsafely(path, &sb)))
+ sv = statEx(pbuf, &sb, &ss);
+ } else
+ sv = stat(path, &sb);
+ if (sv) {
+ ss = 0;
+ if (!Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "(stat: %s)",
+ strerror(errno));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ }
+ }
+ if (pn) {
+ (void) process_proc_node(lnk ? pbuf : path,
+ &sb, ss,
+ (struct stat *)NULL, 0);
+ if (Lf->sf)
+ link_lfile();
+ }
+ }
+/*
+ * Process the ID's execution info.
+ */
+ if (!Ckscko) {
+ (void) make_proc_path(idp, idpl, &path, &pathl, "exe");
+ alloc_lfile("txt", -1);
+ if (getlinksrc(path, pbuf, sizeof(pbuf)) < 1) {
+ (void) memset((void *)&sb, 0, sizeof(sb));
+ lnk = ss = 0;
+ if (!Fwarn) {
+ if ((errno != ENOENT) || uid) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "(readlink: %s)",
+ strerror(errno));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ pn = 1;
+ } else
+ pn = 0;
+ } else {
+ lnk = pn = 1;
+ ss = SB_ALL;
+ if (HasNFS) {
+ if ((sv = statsafely(path, &sb))) {
+ sv = statEx(pbuf, &sb, &ss);
+ if (!sv && (ss & SB_DEV) && (ss & SB_INO))
+ txts = 1;
+ }
+ } else
+ sv = stat(path, &sb);
+ if (sv) {
+ ss = 0;
+ if (!Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "(stat: %s)",
+ strerror(errno));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ } else
+ txts = 1;
+ }
+ if (pn) {
+ (void) process_proc_node(lnk ? pbuf : path,
+ &sb, ss,
+ (struct stat *)NULL, 0);
+ if (Lf->sf)
+ link_lfile();
+ }
+ }
+/*
+ * Process the ID's memory map info.
+ */
+ if (!Ckscko) {
+ (void) make_proc_path(idp, idpl, &path, &pathl, "maps");
+ (void) process_proc_map(path, txts ? &sb : (struct stat *)NULL,
+ txts ? ss : 0);
+ }
+
+#if defined(HASSELINUX)
+/*
+ * Process the PID's SELinux context.
+ */
+ if (Fcntx) {
+
+ /*
+ * If the -Z (cntx) option was specified, match the valid contexts.
+ */
+ errno = 0;
+ if (getpidcon(pid, &Lp->cntx) == -1) {
+ Lp->cntx = (char *)NULL;
+ if (!Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf),
+ "(getpidcon: %s)", strerror(errno));
+ if (!(Lp->cntx = strdup(nmabuf))) {
+ (void) fprintf(stderr,
+ "%s: no context error space: PID %ld",
+ Pn, (long)Lp->pid);
+ Exit(1);
+ }
+ }
+ } else if (CntxArg) {
+
+ /*
+ * See if context includes the process.
+ */
+ for (cntxp = CntxArg; cntxp; cntxp = cntxp->next) {
+ if (cmp_cntx_eq(Lp->cntx, cntxp->cntx)) {
+ cntxp->f = 1;
+ Lp->pss |= PS_PRI;
+ Lp->sf |= SELCNTX;
+ break;
+ }
+ }
+ }
+ }
+#endif /* defined(HASSELINUX) */
+
+/*
+ * Process the ID's file descriptor directory.
+ */
+ if ((i = make_proc_path(idp, idpl, &dpath, &dpathl, "fd/")) < 3)
+ return(0);
+ dpath[i - 1] = '\0';
+ if ((OffType == 2)
+ && ((j = make_proc_path(idp, idpl, &ipath, &ipathl, "fdinfo/")) >= 7))
+ oty = 1;
+ else
+ oty = 0;
+ if (!(fdp = opendir(dpath))) {
+ if (!Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "%s (opendir: %s)",
+ dpath, strerror(errno));
+ alloc_lfile("NOFD", -1);
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ link_lfile();
+ }
+ return(0);
+ }
+ dpath[i - 1] = '/';
+ while ((fp = readdir(fdp))) {
+ if (nm2id(fp->d_name, &fd, &n))
+ continue;
+ (void) make_proc_path(dpath, i, &path, &pathl, fp->d_name);
+ (void) alloc_lfile((char *)NULL, fd);
+ if (getlinksrc(path, pbuf, sizeof(pbuf)) < 1) {
+ (void) memset((void *)&sb, 0, sizeof(sb));
+ lnk = ss = 0;
+ if (!Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "(readlink: %s)",
+ strerror(errno));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ pn = 1;
+ } else
+ pn = 0;
+ } else {
+ lnk = 1;
+ if (HasNFS) {
+ if (lstatsafely(path, &lsb)) {
+ (void) statEx(pbuf, &lsb, &ls);
+ enls = errno;
+ } else {
+ enls = 0;
+ ls = SB_ALL;
+ }
+ if (statsafely(path, &sb)) {
+ (void) statEx(pbuf, &sb, &ss);
+ enss = errno;
+ } else {
+ enss = 0;
+ ss = SB_ALL;
+ }
+ } else {
+ ls = lstat(path, &lsb) ? 0 : SB_ALL;
+ enls = errno;
+ ss = stat(path, &sb) ? 0 : SB_ALL;
+ enss = errno;
+ }
+ if (!ls && !Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "lstat: %s)",
+ strerror(enls));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ if (!ss && !Fwarn) {
+ (void) snpf(nmabuf, sizeof(nmabuf), "(stat: %s)",
+ strerror(enss));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ if (Ckscko) {
+ if ((ss & SB_MODE) && ((sb.st_mode & S_IFMT) == S_IFSOCK))
+ pn = 1;
+ else
+ pn = 0;
+ } else
+ pn = 1;
+ }
+ if (pn) {
+ if (oty) {
+ (void) make_proc_path(ipath, j, &pathi, &pathil,
+ fp->d_name);
+ if ((av = get_fdinfo(pathi, &fi)) & FDINFO_POS) {
+ ls |= SB_SIZE;
+ lsb.st_size = fi.pos;
+ } else
+ ls &= ~SB_SIZE;
+
+#if !defined(HASNOFSFLAGS)
+ if ((av & FDINFO_FLAGS) && (Fsv & FSV_FG)) {
+ Lf->ffg = (long)fi.flags;
+ Lf->fsv |= FSV_FG;
+ }
+# endif /* !defined(HASNOFSFLAGS) */
+
+ }
+ process_proc_node(lnk ? pbuf : path,
+ &sb, ss, &lsb, ls);
+ if (Lf->sf)
+ link_lfile();
+ }
+ }
+ (void) closedir(fdp);
+ return(0);
+}
+
+
+/*
+ * process_proc_map() - process the memory map of a process
+ */
+
+static void
+process_proc_map(p, s, ss)
+ char *p; /* path to process maps file */
+ struct stat *s; /* executing text file state buffer */
+ int ss; /* *s status -- i.e., SB_* values */
+{
+ char buf[MAXPATHLEN + 1], *ep, fmtbuf[32], **fp, nmabuf[MAXPATHLEN + 1];
+ dev_t dev;
+ int ds, en, i, mss, nf, sv;
+ int eb = 6;
+ INODETYPE inode;
+ MALLOC_S len;
+ long maj, min;
+ FILE *ms;
+ int ns = 0;
+ struct stat sb;
+ struct saved_map {
+ dev_t dev;
+ INODETYPE inode;
+ };
+ static struct saved_map *sm = (struct saved_map *)NULL;
+ static int sma = 0;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+/*
+ * Open the /proc/<pid>/maps file, assign a page size buffer to its stream,
+ * and read it/
+ */
+ if (!(ms = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ while (fgets(buf, sizeof(buf), ms)) {
+ if ((nf = get_fields(buf, ":", &fp, &eb, 1)) < 7)
+ continue; /* not enough fields */
+ if (!fp[6] || !*fp[6])
+ continue; /* no path name */
+ /*
+ * See if the path ends in " (deleted)". If it does, strip the
+ * " (deleted)" characters and remember that they were there.
+ */
+ if (((ds = (int)strlen(fp[6])) > 10)
+ && !strcmp(fp[6] + ds - 10, " (deleted)"))
+ {
+ *(fp[6] + ds - 10) = '\0';
+ } else
+ ds = 0;
+ /*
+ * Assemble the major and minor device numbers.
+ */
+ ep = (char *)NULL;
+ if (!fp[3] || !*fp[3]
+ || (maj = strtol(fp[3], &ep, 16)) == LONG_MIN || maj == LONG_MAX
+ || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[4] || !*fp[4]
+ || (min = strtol(fp[4], &ep, 16)) == LONG_MIN || min == LONG_MAX
+ || !ep || *ep)
+ continue;
+ /*
+ * Assemble the device and inode numbers. If they are both zero, skip
+ * the entry.
+ */
+ dev = (dev_t)makedev((int)maj, (int)min);
+ if (!fp[5] || !*fp[5])
+ continue;
+ ep = (char *)NULL;
+ if ((inode = strtoull(fp[5], &ep, 0)) == ULLONG_MAX
+ || !ep || *ep)
+ continue;
+ if (!dev && !inode)
+ continue;
+ /*
+ * See if the device + inode pair match that of the executable.
+ * If they do, skip this map entry.
+ */
+ if (s && (ss & SB_DEV) && (ss & SB_INO)
+ && (dev == s->st_dev) && (inode == (INODETYPE)s->st_ino))
+ continue;
+ /*
+ * See if this device + inode pair has already been processed as
+ * a map entry.
+ */
+ for (i = 0; i < ns; i++) {
+ if (dev == sm[i].dev && inode == sm[i].inode)
+ break;
+ }
+ if (i < ns)
+ continue;
+ /*
+ * Record the processing of this map entry's device and inode pair.
+ */
+ if (ns >= sma) {
+ sma += 10;
+ len = (MALLOC_S)(sma * sizeof(struct saved_map));
+ if (sm)
+ sm = (struct saved_map *)realloc(sm, len);
+ else
+ sm = (struct saved_map *)malloc(len);
+ if (!sm) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for saved maps, PID %d\n",
+ Pn, len, Lp->pid);
+ Exit(1);
+ }
+ }
+ sm[ns].dev = dev;
+ sm[ns++].inode = inode;
+ /*
+ * Allocate space for the mapped file, then get stat(2) information
+ * for it.
+ */
+ alloc_lfile("mem", -1);
+ if (HasNFS) {
+ sv = statsafely(fp[6], &sb);
+ } else
+ sv = stat(fp[6], &sb);
+ if (sv) {
+ en = errno;
+ /*
+ * Applying stat(2) to the file failed, so manufacture a partial
+ * stat(2) reply from the process' maps file entry.
+ *
+ * If the file has been deleted, reset its type to "DEL"; otherwise
+ * generate a stat() error name addition.
+ */
+ (void) memset((void *)&sb, 0, sizeof(sb));
+ sb.st_dev = dev;
+ sb.st_ino = (ino_t)inode;
+ sb.st_mode = S_IFREG;
+ mss = SB_DEV | SB_INO | SB_MODE;
+ if (ds)
+ alloc_lfile("DEL", -1);
+ else {
+ (void) snpf(nmabuf, sizeof(nmabuf), "(stat: %s)",
+ strerror(en));
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ } else if ((sb.st_dev != dev) || ((INODETYPE)sb.st_ino != inode)) {
+
+ /*
+ * The stat(2) device and inode numbers don't match those obtained
+ * from the process' maps file.
+ *
+ * If the file has been deleted, reset its type to "DEL"; otherwise
+ * generate inconsistency name additions.
+ *
+ * Manufacture a partial stat(2) reply from the maps file
+ * information.
+ */
+ if (ds)
+ alloc_lfile("DEL", -1);
+ else if (!Fwarn) {
+ char *sep;
+
+ if (sb.st_dev != dev) {
+ (void) snpf(nmabuf, sizeof(nmabuf),
+ "(path dev=%d,%d%s",
+ GET_MAJ_DEV(sb.st_dev), GET_MIN_DEV(sb.st_dev),
+ ((INODETYPE)sb.st_ino == inode) ? ")" : ",");
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ sep = "";
+ } else
+ sep = "(path ";
+ if ((INODETYPE)sb.st_ino != inode) {
+ (void) snpf(fmtbuf, sizeof(fmtbuf), "%%sinode=%s)",
+ InodeFmt_d);
+ (void) snpf(nmabuf, sizeof(nmabuf), fmtbuf,
+ sep, (INODETYPE)sb.st_ino);
+ nmabuf[sizeof(nmabuf) - 1] = '\0';
+ (void) add_nma(nmabuf, strlen(nmabuf));
+ }
+ }
+ (void) memset((void *)&sb, 0, sizeof(sb));
+ sb.st_dev = dev;
+ sb.st_ino = (ino_t)inode;
+ sb.st_mode = S_IFREG;
+ mss = SB_DEV | SB_INO | SB_MODE;
+ } else
+ mss = SB_ALL;
+ /*
+ * Record the file's information.
+ */
+ process_proc_node(fp[6], &sb, mss, (struct stat *)NULL, 0);
+ if (Lf->sf)
+ link_lfile();
+ }
+ (void) fclose(ms);
+}
+
+
+/*
+ * read_id_stat() - read ID (PID or LWP ID) status
+ *
+ * return: -1 == ID is a zombie
+ * 0 == ID OK
+ * 1 == ID unavailable
+ */
+
+static int
+read_id_stat(ty, p, id, cmd, ppid, pgid)
+ int ty; /* type: 0 == PID, 1 == LWP */
+ char *p; /* path to status file */
+ int id; /* ID: PID or LWP */
+ char **cmd; /* malloc'd command name */
+ int *ppid; /* returned parent PID for PID type */
+ int *pgid; /* returned process group ID for PID
+ * type */
+{
+ char buf[MAXPATHLEN], *cp, *cp1, **fp;
+ static char *cbf = (char *)NULL;
+ static MALLOC_S cbfa = 0;
+ FILE *fs;
+ MALLOC_S len;
+ int nf;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+/*
+ * Open the stat file path, assign a page size buffer to its stream,
+ * and read the file's first line.
+ */
+ if (!(fs = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return(1);
+ cp = fgets(buf, sizeof(buf), fs);
+ (void) fclose(fs);
+ if (!cp)
+ return(1);
+/*
+ * Separate the line into fields on white space separators. Expect five fields
+ * for a PID type and three for an LWP type.
+ */
+ if ((nf = get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0))
+ < (ty ? 5 : 3))
+ {
+ return(1);
+ }
+/*
+ * Convert the first field to an integer; its conversion must match the
+ * ID argument.
+ */
+ if (!fp[0] || (atoi(fp[0]) != id))
+ return(1);
+/*
+ * Get the command name from the second field. Strip a starting '(' and
+ * an ending ')'. Allocate space to hold the result and return the space
+ * pointer.
+ */
+ if (!(cp = fp[1]))
+ return(1);
+ if (cp && *cp == '(')
+ cp++;
+ if ((cp1 = strrchr(cp, ')')))
+ *cp1 = '\0';
+ if ((len = strlen(cp) + 1) > cbfa) {
+ cbfa = len;
+ if (cbf)
+ cbf = (char *)realloc((MALLOC_P *)cbf, cbfa);
+ else
+ cbf = (char *)malloc(cbfa);
+ if (!cbf) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for command \"%s\"\n",
+ Pn, cbfa, cp);
+ Exit(1);
+ }
+ }
+ (void) snpf(cbf, len, "%s", cp);
+ *cmd = cbf;
+/*
+ * If the type is PID, convert and return parent process (fourth field)
+ * and process group (fifth field) IDs.
+ */
+ if (!ty) {
+ if (fp[3] && *fp[3])
+ *ppid = atoi(fp[3]);
+ else
+ return(1);
+ if (fp[4] && *fp[4])
+ *pgid = atoi(fp[4]);
+ else
+ return(1);
+ }
+/*
+ * Check the state in the third field. If it is 'Z', return that indication.
+ */
+ return((fp[2] && !strcmp(fp[2], "Z")) ? -1 : 0);
+}
+
+
+/*
+ * statEx() - extended stat() to get device numbers when a "safe" stat has
+ * failed and the system has an NFS mount
+ *
+ * Note: this function was suggested by Paul Szabo as a way to get device
+ * numbers for NFS files when an NFS mount point has the root_squash
+ * option set. In that case, even if lsof is setuid(root), the identity
+ * of its requests to stat() NFS files lose root permission and may fail.
+ *
+ * This function should be used only when links have been successfully
+ * resolved in the /proc path by getlinksrc().
+ */
+
+static int
+statEx(p, s, ss)
+ char *p; /* file path */
+ struct stat *s; /* stat() result -- NULL if none
+ * wanted */
+ int *ss; /* stat() status -- SB_* values */
+{
+ static size_t ca = 0;
+ static char *cb = NULL;
+ char *cp;
+ int ensv = ENOENT;
+ struct stat sb;
+ int st = 0;
+ size_t sz;
+/*
+ * Make a copy of the path.
+ */
+ sz = strlen(p);
+ if ((sz + 1) > ca) {
+ if (cb)
+ cb = (char *)realloc((MALLOC_P *)cb, sz + 1);
+ else
+ cb = (char *)malloc(sz + 1);
+ if (!cb) {
+ (void) fprintf(stderr,
+ "%s: PID %ld: no statEx path space: %s\n",
+ Pn, (long)Lp->pid, p);
+ Exit(1);
+ }
+ ca = sz + 1;
+ }
+ (void) strcpy(cb, p);
+/*
+ * Trim trailing leaves from the end of the path one at a time and do s safe
+ * stat() on each trimmed result. Stop when a safe stat() succeeds or doesn't
+ * fail because of EACCES or EPERM.
+ */
+ for (cp = strrchr(cb, '/'); cp && (cp != cb);) {
+ *cp = '\0';
+ if (!statsafely(cb, &sb)) {
+ st = 1;
+ break;
+ }
+ ensv = errno;
+ if ((ensv != EACCES) && (ensv != EPERM))
+ break;
+ cp = strrchr(cb, '/');
+ }
+/*
+ * If a stat() on a trimmed result succeeded, form partial results containing
+ * only the device and raw device numbers.
+ */
+ memset((void *)s, 0, sizeof(struct stat));
+ if (st) {
+ errno = 0;
+ s->st_dev = sb.st_dev;
+ s->st_rdev = sb.st_rdev;
+ *ss = SB_DEV | SB_RDEV;
+ return(0);
+ }
+ errno = ensv;
+ *ss = 0;
+ return(1);
+}
--- /dev/null
+/*
+ * dproto.h - Linux function prototypes for /proc-based lsof
+ *
+ * The _PROTOTYPE macro is defined in the common proto.h.
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * $Id: dproto.h,v 1.7 2008/04/15 13:32:26 abe Exp $
+ */
+
+
+#if defined(HASSELINUX)
+_PROTOTYPE(extern int enter_cntx_arg,(char *cnxt));
+#endif /* defined(HASSELINUX) */
+
+_PROTOTYPE(extern int get_fields,(char *ln, char *sep, char ***fr, int *eb, int en));
+_PROTOTYPE(extern void get_locks,(char *p));
+_PROTOTYPE(extern int is_file_named,(char *p, int cd));
+_PROTOTYPE(extern int make_proc_path,(char *pp, int lp, char **np, int *npl, char *sf));
+_PROTOTYPE(extern FILE *open_proc_stream,(char *p, char *mode, char **buf, size_t *sz, int act));
+_PROTOTYPE(extern void process_proc_node,(char *p, struct stat *s, int ss, struct stat *l, int ls));
+_PROTOTYPE(extern void process_proc_sock,(char *p, struct stat *s, int ss, struct stat *l, int ls));
+_PROTOTYPE(extern void set_net_paths,(char *p, int pl));
--- /dev/null
+/*
+ * dsock.c - Linux socket processing functions for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: dsock.c,v 1.34 2009/03/25 19:22:39 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Local definitions
+ */
+
+#define INOBUCKS 128 /* inode hash bucket count -- must be
+ * a power of two */
+#define INOHASH(ino) ((int)((ino * 31415) >> 3) & (INOBUCKS - 1))
+#define TCPUDPHASH(ino) ((int)((ino * 31415) >> 3) & (TcpUdp_bucks - 1))
+#define TCPUDP6HASH(ino) ((int)((ino * 31415) >> 3) & (TcpUdp6_bucks - 1))
+
+
+/*
+ * Local structures
+ */
+
+struct ax25sin { /* AX25 socket information */
+ char *da; /* destination address */
+ char *dev_ch; /* device characters */
+ char *sa; /* source address */
+ INODETYPE inode;
+ unsigned long sq, rq; /* send and receive queue values */
+ unsigned char sqs, rqs; /* send and receive queue states */
+ int state;
+ struct ax25sin *next;
+};
+
+struct ipxsin { /* IPX socket information */
+ INODETYPE inode;
+ char *la; /* local address */
+ char *ra; /* remote address */
+ int state;
+ unsigned long txq, rxq; /* transmit and receive queue values */
+ struct ipxsin *next;
+};
+
+struct packin { /* packet information */
+ INODETYPE inode;
+ int ty; /* socket type */
+ int pr; /* protocol */
+ struct packin *next;
+};
+
+struct rawsin { /* raw socket information */
+ INODETYPE inode;
+ char *la; /* local address */
+ char *ra; /* remote address */
+ char *sp; /* state characters */
+ MALLOC_S lal; /* strlen(la) */
+ MALLOC_S ral; /* strlen(ra) */
+ MALLOC_S spl; /* strlen(sp) */
+ struct rawsin *next;
+};
+
+struct tcp_udp { /* IPv4 TCP and UDP socket
+ * information */
+ INODETYPE inode;
+ unsigned long faddr, laddr; /* foreign & local IPv6 addresses */
+ int fport, lport; /* foreign & local ports */
+ unsigned long txq, rxq; /* transmit & receive queue values */
+ int proto; /* 0 = TCP, 1 = UDP, 2 = UDPLITE */
+ int state; /* protocol state */
+ struct tcp_udp *next;
+};
+
+#if defined(HASIPv6)
+struct tcp_udp6 { /* IPv6 TCP and UDP socket
+ * information */
+ INODETYPE inode;
+ struct in6_addr faddr, laddr; /* foreign and local IPv6 addresses */
+ int fport, lport; /* foreign & local ports */
+ unsigned long txq, rxq; /* transmit & receive queue values */
+ int proto; /* 0 = TCP, 1 = UDP, 2 = UDPLITE */
+ int state; /* protocol state */
+ struct tcp_udp6 *next;
+};
+#endif /* defined(HASIPv6) */
+
+struct uxsin { /* UNIX socket information */
+ INODETYPE inode;
+ char *pcb;
+ char *path;
+ unsigned char sb_def;
+ dev_t sb_dev;
+ INODETYPE sb_ino;
+ dev_t sb_rdev;
+ struct uxsin *next;
+};
+
+
+/*
+ * Local static values
+ */
+
+static char *AX25path = (char *)NULL; /* path to AX25 /proc information */
+static struct ax25sin **AX25sin = (struct ax25sin **)NULL;
+ /* AX25 socket info, hashed by inode */
+static char *ax25st[] = {
+ "LISTENING", /* 0 */
+ "SABM SENT", /* 1 */
+ "DISC SENT", /* 2 */
+ "ESTABLISHED", /* 3 */
+ "RECOVERY" /* 4 */
+};
+#define NAX25ST (sizeof(ax25st) / sizeof(char *))
+static char *Ipxpath = (char *)NULL; /* path to IPX /proc information */
+static struct ipxsin **Ipxsin = (struct ipxsin **)NULL;
+ /* IPX socket info, hashed by inode */
+static struct packin **Packin = (struct packin **)NULL;
+ /* packet info, hashed by inode */
+static char *Packpath = (char *)NULL; /* path to packer /proc information */
+struct packpr { /* packet protocol conversions */
+ int pr; /* protocol number */
+ char *nm; /* protocol name */
+} static Packpr[] = {
+
+#if defined(ETH_P_LOOP)
+ { ETH_P_LOOP, "LOOP" },
+#endif /* defined(ETH_P_LOOP) */
+
+#if defined(ETH_P_PUP)
+ { ETH_P_PUP, "PUP" },
+#endif /* defined(ETH_P_PUP) */
+
+#if defined(ETH_P_PUPAT)
+ { ETH_P_PUPAT, "PUPAT" },
+#endif /* defined(ETH_P_PUPAT) */
+
+#if defined(ETH_P_IP)
+ { ETH_P_IP, "IP" },
+#endif /* defined(ETH_P_IP) */
+
+#if defined(ETH_P_X25)
+ { ETH_P_X25, "X25" },
+#endif /* defined(ETH_P_X25) */
+
+#if defined(ETH_P_ARP)
+ { ETH_P_ARP, "ARP" },
+#endif /* defined(ETH_P_ARP) */
+
+#if defined(ETH_P_BPQ)
+ { ETH_P_BPQ, "BPQ" },
+#endif /* defined(ETH_P_BPQ) */
+
+#if defined(ETH_P_IEEEPUP)
+ { ETH_P_IEEEPUP, "I3EPUP" },
+#endif /* defined(ETH_P_IEEEPUP) */
+
+#if defined(ETH_P_IEEEPUPAT)
+ { ETH_P_IEEEPUPAT, "I3EPUPA" },
+#endif /* defined(ETH_P_IEEEPUPAT) */
+
+#if defined(ETH_P_DEC)
+ { ETH_P_DEC, "DEC" },
+#endif /* defined(ETH_P_DEC) */
+
+#if defined(ETH_P_DNA_DL)
+ { ETH_P_DNA_DL, "DNA_DL" },
+#endif /* defined(ETH_P_DNA_DL) */
+
+#if defined(ETH_P_DNA_RC)
+ { ETH_P_DNA_RC, "DNA_RC" },
+#endif /* defined(ETH_P_DNA_RC) */
+
+#if defined(ETH_P_DNA_RT)
+ { ETH_P_DNA_RT, "DNA_RT" },
+#endif /* defined(ETH_P_DNA_RT) */
+
+#if defined(ETH_P_LAT)
+ { ETH_P_LAT, "LAT" },
+#endif /* defined(ETH_P_LAT) */
+
+#if defined(ETH_P_DIAG)
+ { ETH_P_DIAG, "DIAG" },
+#endif /* defined(ETH_P_DIAG) */
+
+#if defined(ETH_P_CUST)
+ { ETH_P_CUST, "CUST" },
+#endif /* defined(ETH_P_CUST) */
+
+#if defined(ETH_P_SCA)
+ { ETH_P_SCA, "SCA" },
+#endif /* defined(ETH_P_SCA) */
+
+#if defined(ETH_P_RARP)
+ { ETH_P_RARP, "RARP" },
+#endif /* defined(ETH_P_RARP) */
+
+#if defined(ETH_P_ATALK)
+ { ETH_P_ATALK, "ATALK" },
+#endif /* defined(ETH_P_ATALK) */
+
+#if defined(ETH_P_AARP)
+ { ETH_P_AARP, "AARP" },
+#endif /* defined(ETH_P_AARP) */
+
+#if defined(ETH_P_8021Q)
+ { ETH_P_8021Q, "8021Q" },
+#endif /* defined(ETH_P_8021Q) */
+
+#if defined(ETH_P_IPX)
+ { ETH_P_IPX, "IPX" },
+#endif /* defined(ETH_P_IPX) */
+
+#if defined(ETH_P_IPV6)
+ { ETH_P_IPV6, "IPV6" },
+#endif /* defined(ETH_P_IPV6) */
+
+#if defined(ETH_P_SLOW)
+ { ETH_P_SLOW, "SLOW" },
+#endif /* defined(ETH_P_SLOW) */
+
+#if defined(ETH_P_WCCP)
+ { ETH_P_WCCP, "WCCP" },
+#endif /* defined(ETH_P_WCCP) */
+
+#if defined(ETH_P_PPP_DISC)
+ { ETH_P_PPP_DISC, "PPP_DIS" },
+#endif /* defined(ETH_P_PPP_DISC) */
+
+#if defined(ETH_P_PPP_SES)
+ { ETH_P_PPP_SES, "PPP_SES" },
+#endif /* defined(ETH_P_PPP_SES) */
+
+#if defined(ETH_P_MPLS_UC)
+ { ETH_P_MPLS_UC, "MPLS_UC" },
+#endif /* defined(ETH_P_MPLS_UC) */
+
+#if defined(ETH_P_MPLS_MC)
+ { ETH_P_MPLS_MC, "MPLS_MC" },
+#endif /* defined(ETH_P_MPLS_MC) */
+
+#if defined(ETH_P_ATMMPOA)
+ { ETH_P_ATMMPOA, "ATMMPOA" },
+#endif /* defined(ETH_P_ATMMPOA) */
+
+#if defined(ETH_P_MPLS_MC)
+ { ETH_P_MPLS_MC, "MPLS_MC" },
+#endif /* defined(ETH_P_MPLS_MC) */
+
+#if defined(ETH_P_ATMFATE)
+ { ETH_P_ATMFATE, "ATMFATE" },
+#endif /* defined(ETH_P_ATMFATE) */
+
+#if defined(ETH_P_AOE)
+ { ETH_P_AOE, "AOE" },
+#endif /* defined(ETH_P_AOE) */
+
+#if defined(ETH_P_TIPC)
+ { ETH_P_TIPC, "TIPC" },
+#endif /* defined(ETH_P_TIPC) */
+
+#if defined(ETH_P_802_3)
+ { ETH_P_802_3, "802.3" },
+#endif /* defined(ETH_P_802_3) */
+
+#if defined(ETH_P_AX25)
+ { ETH_P_AX25, "AX25" },
+#endif /* defined(ETH_P_AX25) */
+
+#if defined(ETH_P_ALL)
+ { ETH_P_ALL, "ALL" },
+#endif /* defined(ETH_P_ALL) */
+
+#if defined(ETH_P_802_2)
+ { ETH_P_802_2, "802.2" },
+#endif /* defined(ETH_P_802_2) */
+
+#if defined(ETH_P_SNAP)
+ { ETH_P_SNAP, "SNAP" },
+#endif /* defined(ETH_P_SNAP) */
+
+#if defined(ETH_P_DDCMP)
+ { ETH_P_DDCMP, "DDCMP" },
+#endif /* defined(ETH_P_DDCMP) */
+
+#if defined(ETH_P_WAN_PPP)
+ { ETH_P_WAN_PPP, "WAN_PPP" },
+#endif /* defined(ETH_P_WAN_PPP) */
+
+#if defined(ETH_P_PPP_MP)
+ { ETH_P_PPP_MP, "PPP MP" },
+#endif /* defined(ETH_P_PPP_MP) */
+
+#if defined(ETH_P_LOCALTALK)
+ { ETH_P_LOCALTALK, "LCLTALK" },
+#endif /* defined(ETH_P_LOCALTALK) */
+
+#if defined(ETH_P_PPPTALK)
+ { ETH_P_PPPTALK, "PPPTALK" },
+#endif /* defined(ETH_P_PPPTALK) */
+
+#if defined(ETH_P_TR_802_2)
+ { ETH_P_TR_802_2, "802.2" },
+#endif /* defined(ETH_P_TR_802_2) */
+
+#if defined(ETH_P_MOBITEX)
+ { ETH_P_MOBITEX, "MOBITEX" },
+#endif /* defined(ETH_P_MOBITEX) */
+
+#if defined(ETH_P_CONTROL)
+ { ETH_P_CONTROL, "CONTROL" },
+#endif /* defined(ETH_P_CONTROL) */
+
+#if defined(ETH_P_IRDA)
+ { ETH_P_IRDA, "IRDA" },
+#endif /* defined(ETH_P_IRDA) */
+
+#if defined(ETH_P_ECONET)
+ { ETH_P_ECONET, "ECONET" },
+#endif /* defined(ETH_P_ECONET) */
+
+#if defined(ETH_P_HDLC)
+ { ETH_P_HDLC, "HDLC" },
+#endif /* defined(ETH_P_HDLC) */
+
+#if defined(ETH_P_ARCNET)
+ { ETH_P_ARCNET, "ARCNET" },
+#endif /* defined(ETH_P_ARCNET) */
+
+};
+#define NPACKPR (sizeof(Packpr) / sizeof(struct packpr))
+struct packty { /* packet socket type conversions */
+ int ty; /* type number */
+ char *nm; /* type name */
+} static Packty[] = {
+
+#if defined(SOCK_STREAM)
+ { SOCK_STREAM, "STREAM" },
+#endif /* defined(SOCK_STREAM) */
+
+#if defined(SOCK_DGRAM)
+ { SOCK_DGRAM, "DGRAM" },
+#endif /* defined(SOCK_DGRAM) */
+
+#if defined(SOCK_RAW)
+ { SOCK_RAW, "RAW" },
+#endif /* defined(SOCK_RAW) */
+
+#if defined(SOCK_RDM)
+ { SOCK_RDM, "RDM" },
+#endif /* defined(SOCK_RDM) */
+
+#if defined(SOCK_SEQPACKET)
+ { SOCK_SEQPACKET, "SEQPACKET" },
+#endif /* defined(SOCK_SEQPACKET) */
+
+#if defined(SOCK_PACKET)
+ { SOCK_PACKET, "PACKET" },
+#endif /* defined(SOCK_PACKET) */
+};
+#define NPACKTY (sizeof(Packty) / sizeof(struct packty))
+static char *Rawpath = (char *)NULL; /* path to raw socket /proc
+ * information */
+static struct rawsin **Rawsin = (struct rawsin **)NULL;
+ /* raw socket info, hashed by inode */
+static char *SockStatPath = (char *)NULL;
+ /* path to /proc/net socket status */
+static char *TCPpath = (char *)NULL; /* path to TCP /proc information */
+static struct tcp_udp **TcpUdp = (struct tcp_udp **)NULL;
+ /* IPv4 TCP & UDP info, hashed by
+ * inode */
+static int TcpUdp_bucks = 0; /* dynamically sized hash bucket
+ * count for TCP and UDP -- will
+ * be a power of two */
+
+#if defined(HASIPv6)
+static char *Raw6path = (char *)NULL; /* path to raw IPv6 /proc information */
+static struct rawsin **Rawsin6 = (struct rawsin **)NULL;
+ /* IPv6 raw socket info, hashed by
+ * inode */
+static char *SockStatPath6 = (char *)NULL;
+ /* path to /proc/net IPv6 socket
+ * status */
+static char *TCP6path = (char *)NULL; /* path to IPv6 TCP /proc information */
+static struct tcp_udp6 **TcpUdp6 = (struct tcp_udp6 **)NULL;
+ /* IPv6 TCP & UDP info, hashed by
+ * inode */
+static int TcpUdp6_bucks = 0; /* dynamically sized hash bucket
+ * count for IPv6 TCP and UDP -- will
+ * be a power of two */
+static char *UDP6path = (char *)NULL; /* path to IPv6 UDP /proc information */
+static char *UDP6LITEpath = (char *)NULL;
+ /* path to IPv6 UDPLITE /proc
+ * information */
+#endif /* defined(HASIPv6) */
+
+static char *UDPpath = (char *)NULL; /* path to UDP /proc information */
+static char *UDPLITEpath = (char *)NULL;
+ /* path to UDPLITE /proc information */
+static char *UNIXpath = (char *)NULL; /* path to UNIX /proc information */
+static struct uxsin **Uxsin = (struct uxsin **)NULL;
+ /* UNIX socket info, hashed by inode */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static struct ax25sin *check_ax25,(INODETYPE i));
+_PROTOTYPE(static struct ipxsin *check_ipx,(INODETYPE i));
+_PROTOTYPE(static struct packin *check_pack,(INODETYPE i));
+_PROTOTYPE(static struct rawsin *check_raw,(INODETYPE i));
+_PROTOTYPE(static struct tcp_udp *check_tcpudp,(INODETYPE i, char **p));
+_PROTOTYPE(static struct uxsin *check_unix,(INODETYPE i));
+_PROTOTYPE(static void get_ax25,(char *p));
+_PROTOTYPE(static void get_ipx,(char *p));
+_PROTOTYPE(static void get_pack,(char *p));
+_PROTOTYPE(static void get_raw,(char *p));
+_PROTOTYPE(static void get_tcpudp,(char *p, int pr, int clr));
+_PROTOTYPE(static void get_unix,(char *p));
+_PROTOTYPE(static void print_ax25info,(struct ax25sin *ap));
+_PROTOTYPE(static void print_ipxinfo,(struct ipxsin *ip));
+
+#if defined(HASIPv6)
+_PROTOTYPE(static struct rawsin *check_raw6,(INODETYPE i));
+_PROTOTYPE(static struct tcp_udp6 *check_tcpudp6,(INODETYPE i, char **p));
+_PROTOTYPE(static void get_raw6,(char *p));
+_PROTOTYPE(static void get_tcpudp6,(char *p, int pr, int clr));
+_PROTOTYPE(static int net6a2in6,(char *as, struct in6_addr *ad));
+#endif /* defined(HASIPv6) */
+
+
+/*
+ * build_IPstates() -- build the TCP and UDP state tables
+ */
+
+void
+build_IPstates()
+{
+ if (!TcpSt) {
+ (void) enter_IPstate("TCP", "ESTABLISHED", TCP_ESTABLISHED);
+ (void) enter_IPstate("TCP", "SYN_SENT", TCP_SYN_SENT);
+ (void) enter_IPstate("TCP", "SYN_RECV", TCP_SYN_RECV);
+ (void) enter_IPstate("TCP", "FIN_WAIT1", TCP_FIN_WAIT1);
+ (void) enter_IPstate("TCP", "FIN_WAIT2", TCP_FIN_WAIT2);
+ (void) enter_IPstate("TCP", "TIME_WAIT", TCP_TIME_WAIT);
+ (void) enter_IPstate("TCP", "CLOSE", TCP_CLOSE);
+ (void) enter_IPstate("TCP", "CLOSE_WAIT", TCP_CLOSE_WAIT);
+ (void) enter_IPstate("TCP", "LAST_ACK", TCP_LAST_ACK);
+ (void) enter_IPstate("TCP", "LISTEN", TCP_LISTEN);
+ (void) enter_IPstate("TCP", "CLOSING", TCP_CLOSING);
+ (void) enter_IPstate("TCP", "CLOSED", 0);
+ (void) enter_IPstate("TCP", (char *)NULL, 0);
+ }
+}
+
+
+/*
+ * check_ax25() - check for AX25 socket file
+ */
+
+static struct ax25sin *
+check_ax25(i)
+ INODETYPE i; /* socket file's inode number */
+{
+ struct ax25sin *ap;
+ int h;
+
+ h = INOHASH(i);
+ for (ap = AX25sin[h]; ap; ap = ap->next) {
+ if (i == ap->inode)
+ return(ap);
+ }
+ return((struct ax25sin *)NULL);
+}
+
+
+/*
+ * check_ipx() - check for IPX socket file
+ */
+
+static struct ipxsin *
+check_ipx(i)
+ INODETYPE i; /* socket file's inode number */
+{
+ int h;
+ struct ipxsin *ip;
+
+ h = INOHASH(i);
+ for (ip = Ipxsin[h]; ip; ip = ip->next) {
+ if (i == ip->inode)
+ return(ip);
+ }
+ return((struct ipxsin *)NULL);
+}
+
+
+/*
+ * check_pack() - check for packet file
+ */
+
+static struct packin *
+check_pack(i)
+ INODETYPE i; /* packet file's inode number */
+{
+ int h;
+ struct packin *pp;
+
+ h = INOHASH(i);
+ for (pp = Packin[h]; pp; pp = pp->next) {
+ if (i == pp->inode)
+ return(pp);
+ }
+ return((struct packin *)NULL);
+}
+
+
+
+/*
+ * check_raw() - check for raw socket file
+ */
+
+static struct rawsin *
+check_raw(i)
+ INODETYPE i; /* socket file's inode number */
+{
+ int h;
+ struct rawsin *rp;
+
+ h = INOHASH(i);
+ for (rp = Rawsin[h]; rp; rp = rp->next) {
+ if (i == rp->inode)
+ return(rp);
+ }
+ return((struct rawsin *)NULL);
+}
+
+
+/*
+ * check_tcpudp() - check for IPv4 TCP or UDP socket file
+ */
+
+static struct tcp_udp *
+check_tcpudp(i, p)
+ INODETYPE i; /* socket file's inode number */
+ char **p; /* protocol return */
+{
+ int h;
+ struct tcp_udp *tp;
+
+ h = TCPUDPHASH(i);
+ for (tp = TcpUdp[h]; tp; tp = tp->next) {
+ if (i == tp->inode) {
+ switch (tp->proto) {
+ case 0:
+ *p = "TCP";
+ break;
+ case 1:
+ *p = "UDP";
+ break;
+ case 2:
+ *p = "UDPLITE";
+ break;
+ default:
+ *p = "unknown";
+ }
+ return(tp);
+ }
+ }
+ return((struct tcp_udp *)NULL);
+}
+
+
+#if defined(HASIPv6)
+/*
+ * check_raw6() - check for raw IPv6 socket file
+ */
+
+static struct rawsin *
+check_raw6(i)
+ INODETYPE i; /* socket file's inode number */
+{
+ int h;
+ struct rawsin *rp;
+
+ h = INOHASH(i);
+ for (rp = Rawsin6[h]; rp; rp = rp->next) {
+ if (i == rp->inode)
+ return(rp);
+ }
+ return((struct rawsin *)NULL);
+}
+
+
+/*
+ * check_tcpudp6() - check for IPv6 TCP or UDP socket file
+ */
+
+static struct tcp_udp6 *
+check_tcpudp6(i, p)
+ INODETYPE i; /* socket file's inode number */
+ char **p; /* protocol return */
+{
+ int h;
+ struct tcp_udp6 *tp6;
+
+ h = TCPUDP6HASH(i);
+ for (tp6 = TcpUdp6[h]; tp6; tp6 = tp6->next) {
+ if (i == tp6->inode) {
+ switch (tp6->proto) {
+ case 0:
+ *p = "TCP";
+ break;
+ case 1:
+ *p = "UDP";
+ break;
+ case 2:
+ *p = "UDPLITE";
+ break;
+ default:
+ *p = "unknown";
+ }
+ return(tp6);
+ }
+ }
+ return((struct tcp_udp6 *)NULL);
+}
+#endif /* defined(HASIPv6) */
+
+
+/*
+ * check_unix() - check for UNIX domain socket
+ */
+
+static struct uxsin *
+check_unix(i)
+ INODETYPE i; /* socket file's inode number */
+{
+ int h;
+ struct uxsin *up;
+
+ h = INOHASH(i);
+ for (up = Uxsin[h]; up; up = up->next) {
+ if (i == up->inode)
+ return(up);
+ }
+ return((struct uxsin *)NULL);
+}
+
+
+/*
+ * get_ax25() - get /proc/net/ax25 info
+ */
+
+static void
+get_ax25(p)
+ char *p; /* /proc/net/ipx path */
+{
+ struct ax25sin *ap, *np;
+ FILE *as;
+ char buf[MAXPATHLEN], *da, *dev_ch, *ep, **fp, *sa;
+ int h, nf;
+ INODETYPE inode;
+ unsigned long rq, sq, state;
+ MALLOC_S len;
+ unsigned char rqs, sqs;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+/*
+ * Do second time cleanup or first time setup.
+ */
+ if (AX25sin) {
+ for (h = 0; h < INOBUCKS; h++) {
+ for (ap = AX25sin[h]; ap; ap = np) {
+ np = ap->next;
+ if (ap->da)
+ (void) free((FREE_P *)ap->da);
+ if (ap->dev_ch)
+ (void) free((FREE_P *)ap->dev_ch);
+ if (ap->sa)
+ (void) free((FREE_P *)ap->sa);
+ (void) free((FREE_P *)ap);
+ }
+ AX25sin[h] = (struct ax25sin *)NULL;
+ }
+ } else {
+ AX25sin = (struct ax25sin **)calloc(INOBUCKS,
+ sizeof(struct ax25sin *));
+ if (!AX25sin) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d AX25 hash pointer bytes\n",
+ Pn, INOBUCKS * sizeof(struct ax25sin *));
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc/net/ax25 file, assign a page size buffer to the stream,
+ * and read it. Store AX25 socket info in the AX25sin[] hash buckets.
+ */
+ if (!(as = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ while (fgets(buf, sizeof(buf) - 1, as)) {
+ if ((nf = get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0)) < 24)
+ continue;
+ /*
+ * /proc/next/ac25 has no title line, a very poor deficiency in its
+ * implementation.
+ *
+ * The ax25_get_info() function in kern mnodule .../net/ax25/af_ax25.c
+ * says the format of the lines in the file is:
+ *
+ * magic dev src_addr dest_addr,digi1,digi2,.. st vs vr va t1 t1 \
+ * t2 t2 t3 t3 idle idle n2 n2 rtt window paclen Snd-Q Rcv-Q \
+ * inode
+ *
+ * The code in this function is forced to assume that format is in
+ * effect..
+ */
+
+ /*
+ * Assemble the inode number and see if it has already been recorded.
+ * If it has, skip this line.
+ */
+ ep = (char *)NULL;
+ if (!fp[23] || !*fp[23]
+ || (inode = strtoull(fp[23], &ep, 0)) == ULONG_MAX
+ || !ep || *ep)
+ continue;
+ h = INOHASH((INODETYPE)inode);
+ for (ap = AX25sin[h]; ap; ap = ap->next) {
+ if (inode == ap->inode)
+ break;
+ }
+ if (ap)
+ continue;
+ /*
+ * Assemble the send and receive queue values and the state.
+ */
+ rq = sq = (unsigned long)0;
+ rqs = sqs = (unsigned char)0;
+ ep = (char *)NULL;
+ if (!fp[21] || !*fp[21]
+ || (sq = strtoul(fp[21], &ep, 0)) == ULONG_MAX || !ep || *ep)
+ continue;
+ sqs = (unsigned char)1;
+ ep = (char *)NULL;
+ if (!fp[22] || !*fp[22]
+ || (rq = strtoul(fp[22], &ep, 0)) == ULONG_MAX || !ep || *ep)
+ continue;
+ rqs = (unsigned char)1;
+ ep = (char *)NULL;
+ if (!fp[4] || !*fp[4]
+ || (state = strtoul(fp[4], &ep, 0)) == ULONG_MAX || !ep || *ep)
+ continue;
+ /*
+ * Allocate space for the destination address.
+ */
+ if (!fp[3] || !*fp[3])
+ da = (char *)NULL;
+ else if ((len = strlen(fp[3]))) {
+ if (!(da = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d destination AX25 addr bytes: %s\n",
+ Pn, len + 1, fp[3]);
+ Exit(1);
+ }
+ (void) snpf(da, len + 1, "%s", fp[3]);
+ } else
+ da = (char *)NULL;
+ /*
+ * Allocate space for the source address.
+ */
+ if (!fp[2] || !*fp[2])
+ sa = (char *)NULL;
+ else if ((len = strlen(fp[2]))) {
+ if (!(sa = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d source AX25 address bytes: %s\n",
+ Pn, len + 1, fp[2]);
+ Exit(1);
+ }
+ (void) snpf(sa, len + 1, "%s", fp[2]);
+ } else
+ sa = (char *)NULL;
+ /*
+ * Allocate space for the device characters.
+ */
+ if (!fp[1] || !*fp[1])
+ dev_ch = (char *)NULL;
+ else if ((len = strlen(fp[1]))) {
+ if (!(dev_ch = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d destination AX25 dev bytes: %s\n",
+ Pn, len + 1, fp[1]);
+ Exit(1);
+ }
+ (void) snpf(dev_ch, len + 1, "%s", fp[1]);
+ } else
+ dev_ch = (char *)NULL;
+ /*
+ * Allocate space for an ax25sin entry, fill it, and link it to its
+ * hash bucket.
+ */
+ if (!(ap = (struct ax25sin *)malloc(sizeof(struct ax25sin)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d byte ax25sin structure\n",
+ Pn, sizeof(struct ax25sin));
+ Exit(1);
+ }
+ ap->da = da;
+ ap->dev_ch = dev_ch;
+ ap->inode = inode;
+ ap->rq = rq;
+ ap->rqs = rqs;
+ ap->sa = sa;
+ ap->sq = sq;
+ ap->sqs = sqs;
+ ap->state = (int)state;
+ ap->next = AX25sin[h];
+ AX25sin[h] = ap;
+ }
+ (void) fclose(as);
+}
+
+
+/*
+ * get_ipx() - get /proc/net/ipx info
+ */
+
+static void
+get_ipx(p)
+ char *p; /* /proc/net/ipx path */
+{
+ char buf[MAXPATHLEN], *ep, **fp, *la, *ra;
+ int fl = 1;
+ int h;
+ INODETYPE inode;
+ unsigned long rxq, state, txq;
+ struct ipxsin *ip, *np;
+ MALLOC_S len;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+ FILE *xs;
+/*
+ * Do second time cleanup or first time setup.
+ */
+ if (Ipxsin) {
+ for (h = 0; h < INOBUCKS; h++) {
+ for (ip = Ipxsin[h]; ip; ip = np) {
+ np = ip->next;
+ if (ip->la)
+ (void) free((FREE_P *)ip->la);
+ if (ip->ra)
+ (void) free((FREE_P *)ip->ra);
+ (void) free((FREE_P *)ip);
+ }
+ Ipxsin[h] = (struct ipxsin *)NULL;
+ }
+ } else {
+ Ipxsin = (struct ipxsin **)calloc(INOBUCKS,
+ sizeof(struct ipxsin *));
+ if (!Ipxsin) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d IPX hash pointer bytes\n",
+ Pn, INOBUCKS * sizeof(struct ipxsin *));
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc/net/ipx file, assign a page size buffer to the stream,
+ * and read it. Store IPX socket info in the Ipxsin[] hash buckets.
+ */
+ if (!(xs = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ while (fgets(buf, sizeof(buf) - 1, xs)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) < 7)
+ continue;
+ if (fl) {
+
+ /*
+ * Check the column labels in the first line.
+ */
+ if (!fp[0] || strcmp(fp[0], "Local_Address")
+ || !fp[1] || strcmp(fp[1], "Remote_Address")
+ || !fp[2] || strcmp(fp[2], "Tx_Queue")
+ || !fp[3] || strcmp(fp[3], "Rx_Queue")
+ || !fp[4] || strcmp(fp[4], "State")
+ || !fp[5] || strcmp(fp[5], "Uid")
+ || !fp[6] || strcmp(fp[6], "Inode"))
+ {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: unsupported format: %s\n",
+ Pn, p);
+ }
+ break;
+ }
+ fl = 0;
+ continue;
+ }
+ /*
+ * Assemble the inode number and see if the inode is already
+ * recorded.
+ */
+ ep = (char *)NULL;
+ if (!fp[6] || !*fp[6]
+ || (inode = strtoull(fp[6], &ep, 0)) == ULONG_MAX
+ || !ep || *ep)
+ continue;
+ h = INOHASH(inode);
+ for (ip = Ipxsin[h]; ip; ip = ip->next) {
+ if (inode == ip->inode)
+ break;
+ }
+ if (ip)
+ continue;
+ /*
+ * Assemble the transmit and receive queue values and the state.
+ */
+ ep = (char *)NULL;
+ if (!fp[2] || !*fp[2]
+ || (txq = strtoul(fp[2], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[3] || !*fp[3]
+ || (rxq = strtoul(fp[3], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[4] || !*fp[4]
+ || (state = strtoul(fp[4], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ /*
+ * Allocate space for the local address, unless it is "Not_Connected".
+ */
+ if (!fp[0] || !*fp[0] || strcmp(fp[0], "Not_Connected") == 0)
+ la = (char *)NULL;
+ else if ((len = strlen(fp[0]))) {
+ if (!(la = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d local IPX address bytes: %s\n",
+ Pn, len + 1, fp[0]);
+ Exit(1);
+ }
+ (void) snpf(la, len + 1, "%s", fp[0]);
+ } else
+ la = (char *)NULL;
+ /*
+ * Allocate space for the remote address, unless it is "Not_Connected".
+ */
+ if (!fp[1] || !*fp[1] || strcmp(fp[1], "Not_Connected") == 0)
+ ra = (char *)NULL;
+ else if ((len = strlen(fp[1]))) {
+ if (!(ra = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d remote IPX address bytes: %s\n",
+ Pn, len + 1, fp[1]);
+ Exit(1);
+ }
+ (void) snpf(ra, len + 1, "%s", fp[1]);
+ } else
+ ra = (char *)NULL;
+ /*
+ * Allocate space for an ipxsin entry, fill it, and link it to its
+ * hash bucket.
+ */
+ if (!(ip = (struct ipxsin *)malloc(sizeof(struct ipxsin)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d byte ipxsin structure\n",
+ Pn, sizeof(struct ipxsin));
+ Exit(1);
+ }
+ ip->inode = inode;
+ ip->la = la;
+ ip->ra = ra;
+ ip->txq = txq;
+ ip->rxq = rxq;
+ ip->state = (int)state;
+ ip->next = Ipxsin[h];
+ Ipxsin[h] = ip;
+ }
+ (void) fclose(xs);
+}
+
+
+/*
+ * get_pack() - get /proc/net/packet info
+ */
+
+static void
+get_pack(p)
+ char *p; /* /proc/net/raw path */
+{
+ char buf[MAXPATHLEN], *ep, **fp;
+ int h, lc, ty;
+ INODETYPE inode;
+ struct packin *np, *pp;
+ unsigned long pr;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+ FILE *xs;
+/*
+ * Do second time cleanup or first time setup.
+ */
+ if (Packin) {
+ for (h = 0; h < INOBUCKS; h++) {
+ for (pp = Packin[h]; pp; pp = np) {
+ np = pp->next;
+ (void) free((FREE_P *)pp);
+ }
+ Packin[h] = (struct packin *)NULL;
+ }
+ } else {
+ Packin = (struct packin **)calloc(INOBUCKS,
+ sizeof(struct packin *));
+ if (!Packin) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d packet hash pointer bytes\n",
+ Pn, INOBUCKS * sizeof(struct packin *));
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc/net/packet file, assign a page size buffer to its stream,
+ * and read the file. Store packet info in the Packin[] hash buckets.
+ */
+ if (!(xs = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ lc = 0;
+ while (fgets(buf, sizeof(buf) - 1, xs)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) < 9)
+ continue;
+ lc++;
+ if (lc == 1) {
+
+ /*
+ * Check the column labels in the first line.
+ */
+ if (!fp[2] || strcmp(fp[2], "Type")
+ || !fp[3] || strcmp(fp[3], "Proto")
+ || !fp[8] || strcmp(fp[8], "Inode"))
+ {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: unsupported format: %s\n",
+ Pn, p);
+ }
+ break;
+ }
+ continue;
+ }
+ /*
+ * Assemble the inode number and see if the inode is already
+ * recorded.
+ */
+ ep = (char *)NULL;
+ if (!fp[8] || !*fp[8]
+ || (inode = strtoull(fp[8], &ep, 0)) == ULONG_MAX
+ || !ep || *ep)
+ continue;
+ h = INOHASH(inode);
+ for (pp = Packin[h]; pp; pp = pp->next) {
+ if (inode == pp->inode)
+ break;
+ }
+ if (pp)
+ continue;
+ /*
+ * Save the socket type and protocol.
+ */
+ if (!fp[2] || !*fp[2] || (strlen(fp[2])) < 1)
+ continue;
+ ty = atoi(fp[2]);
+ ep = (char *)NULL;
+ if (!fp[3] || !*fp[3] || (strlen(fp[3]) < 1)
+ || ((pr = strtoul(fp[3], &ep, 16)) == ULONG_MAX) || !ep || *ep)
+ continue;
+ /*
+ * Allocate space for a packin entry, fill it, and link it to its
+ * hash bucket.
+ */
+ if (!(pp = (struct packin *)malloc(sizeof(struct packin)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d byte packet structure\n",
+ Pn, sizeof(struct packin));
+ Exit(1);
+ }
+ pp->inode = inode;
+ pp->pr = (int)pr;
+ pp->ty = ty;
+ pp->next = Packin[h];
+ Packin[h] = pp;
+ }
+ (void) fclose(xs);
+}
+
+
+/*
+ * get_raw() - get /proc/net/raw info
+ */
+
+static void
+get_raw(p)
+ char *p; /* /proc/net/raw path */
+{
+ char buf[MAXPATHLEN], *ep, **fp, *la, *ra, *sp;
+ int h;
+ INODETYPE inode;
+ int nf = 12;
+ struct rawsin *np, *rp;
+ MALLOC_S lal, ral, spl;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+ FILE *xs;
+/*
+ * Do second time cleanup or first time setup.
+ */
+ if (Rawsin) {
+ for (h = 0; h < INOBUCKS; h++) {
+ for (rp = Rawsin[h]; rp; rp = np) {
+ np = rp->next;
+ if (rp->la)
+ (void) free((FREE_P *)rp->la);
+ if (rp->ra)
+ (void) free((FREE_P *)rp->ra);
+ (void) free((FREE_P *)rp);
+ }
+ Rawsin[h] = (struct rawsin *)NULL;
+ }
+ } else {
+ Rawsin = (struct rawsin **)calloc(INOBUCKS,
+ sizeof(struct rawsin *));
+ if (!Rawsin) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d raw hash pointer bytes\n",
+ Pn, INOBUCKS * sizeof(struct rawsin *));
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc/net/raw file, assign a page size buffer to its stream,
+ * and read the file. Store raw socket info in the Rawsin[] hash buckets.
+ */
+ if (!(xs = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ while (fgets(buf, sizeof(buf) - 1, xs)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) < nf)
+ continue;
+ if (nf == 12) {
+
+ /*
+ * Check the column labels in the first line.
+ */
+ if (!fp[1] || strcmp(fp[1], "local_address")
+ || !fp[2] || strcmp(fp[2], "rem_address")
+ || !fp[3] || strcmp(fp[3], "st")
+ || !fp[11] || strcmp(fp[11], "inode"))
+ {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: unsupported format: %s\n",
+ Pn, p);
+ }
+ break;
+ }
+ nf = 10;
+ continue;
+ }
+ /*
+ * Assemble the inode number and see if the inode is already
+ * recorded.
+ */
+ ep = (char *)NULL;
+ if (!fp[9] || !*fp[9]
+ || (inode = strtoull(fp[9], &ep, 0)) == ULONG_MAX
+ || !ep || *ep)
+ continue;
+ h = INOHASH(inode);
+ for (rp = Rawsin[h]; rp; rp = rp->next) {
+ if (inode == rp->inode)
+ break;
+ }
+ if (rp)
+ continue;
+ /*
+ * Save the local address, remote address, and state.
+ */
+ if (!fp[1] || !*fp[1] || (lal = strlen(fp[1])) < 1) {
+ la = (char *)NULL;
+ lal = (MALLOC_S)0;
+ } else {
+ if (!(la = (char *)malloc(lal + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d local raw address bytes: %s\n",
+ Pn, lal + 1, fp[1]);
+ Exit(1);
+ }
+ (void) snpf(la, lal + 1, "%s", fp[1]);
+ }
+ if (!fp[2] || !*fp[2] || (ral = strlen(fp[2])) < 1) {
+ ra = (char *)NULL;
+ ral = (MALLOC_S)0;
+ } else {
+ if (!(ra = (char *)malloc(ral + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d remote raw address bytes: %s\n",
+ Pn, ral + 1, fp[2]);
+ Exit(1);
+ }
+ (void) snpf(ra, ral + 1, "%s", fp[2]);
+ }
+ if (!fp[3] || !*fp[3] || (spl = strlen(fp[3])) < 1) {
+ sp = (char *)NULL;
+ spl = (MALLOC_S)0;
+ } else {
+ if (!(sp = (char *)malloc(spl + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d remote raw state bytes: %s\n",
+ Pn, spl + 1, fp[2]);
+ Exit(1);
+ }
+ (void) snpf(sp, spl + 1, "%s", fp[3]);
+ }
+ /*
+ * Allocate space for an rawsin entry, fill it, and link it to its
+ * hash bucket.
+ */
+ if (!(rp = (struct rawsin *)malloc(sizeof(struct rawsin)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d byte rawsin structure\n",
+ Pn, sizeof(struct rawsin));
+ Exit(1);
+ }
+ rp->inode = inode;
+ rp->la = la;
+ rp->lal = lal;
+ rp->ra = ra;
+ rp->ral = ral;
+ rp->sp = sp;
+ rp->spl = spl;
+ rp->next = Rawsin[h];
+ Rawsin[h] = rp;
+ }
+ (void) fclose(xs);
+}
+
+
+/*
+ * get_tcpudp() - get IPv4 TCP, UDP or UDPLITE net info
+ */
+
+static void
+get_tcpudp(p, pr, clr)
+ char *p; /* /proc/net/{tcp,udp} path */
+ int pr; /* protocol: 0 = TCP, 1 = UDP,
+ * 2 = UDPLITE */
+ int clr; /* 1 == clear the table */
+{
+ char buf[MAXPATHLEN], *ep, **fp;
+ unsigned long faddr, fport, laddr, lport, rxq, state, txq;
+ FILE *fs;
+ int h, nf;
+ INODETYPE inode;
+ struct tcp_udp *np, *tp;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+/*
+ * Delete previous table contents.
+ */
+ if (TcpUdp) {
+ if (clr) {
+ for (h = 0; h < TcpUdp_bucks; h++) {
+ for (tp = TcpUdp[h]; tp; tp = np) {
+ np = tp->next;
+ (void) free((FREE_P *)tp);
+ }
+ TcpUdp[h] = (struct tcp_udp *)NULL;
+ }
+ }
+/*
+ * If no hash buckets have been allocated, do so now.
+ */
+ } else {
+
+ /*
+ * Open the /proc/net/sockstat file and establish the hash bucket
+ * count from its "sockets: used" line.
+ */
+ TcpUdp_bucks = INOBUCKS;
+ if ((fs = fopen(SockStatPath, "r"))) {
+ while(fgets(buf, sizeof(buf) - 1, fs)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) != 3)
+ continue;
+ if (!fp[0] || strcmp(fp[0], "sockets:")
+ || !fp[1] || strcmp(fp[1], "used")
+ || !fp[2] || !*fp[2])
+ continue;
+ if ((h = atoi(fp[2])) < 1)
+ h = INOBUCKS;
+ while (TcpUdp_bucks < h)
+ TcpUdp_bucks *= 2;
+ break;
+ }
+ (void) fclose(fs);
+ }
+ if (!(TcpUdp = (struct tcp_udp **)calloc(TcpUdp_bucks,
+ sizeof(struct tcp_udp *))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for TCP&UDP hash buckets\n",
+ Pn, (int)(TcpUdp_bucks * sizeof(struct tcp_udp *)));
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc/net file, assign a page size buffer to the stream, and
+ * read it.
+ */
+ if (!(fs = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ nf = 12;
+ while(fgets(buf, sizeof(buf) - 1, fs)) {
+ if (get_fields(buf,
+ (nf == 12) ? (char *)NULL : ":",
+ &fp, (int *)NULL, 0)
+ < nf)
+ continue;
+ if (nf == 12) {
+ if (!fp[1] || strcmp(fp[1], "local_address")
+ || !fp[2] || strcmp(fp[2], "rem_address")
+ || !fp[3] || strcmp(fp[3], "st")
+ || !fp[4] || strcmp(fp[4], "tx_queue")
+ || !fp[5] || strcmp(fp[5], "rx_queue")
+ || !fp[11] || strcmp(fp[11], "inode"))
+ {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: unsupported format: %s\n",
+ Pn, p);
+ }
+ break;
+ }
+ nf = 14;
+ continue;
+ }
+ /*
+ * Get the local and remote addresses.
+ */
+ ep = (char *)NULL;
+ if (!fp[1] || !*fp[1]
+ || (laddr = strtoul(fp[1], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[2] || !*fp[2]
+ || (lport = strtoul(fp[2], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[3] || !*fp[3]
+ || (faddr = strtoul(fp[3], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[4] || !*fp[4]
+ || (fport = strtoul(fp[4], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ /*
+ * Get the state and queue sizes.
+ */
+ ep = (char *)NULL;
+ if (!fp[5] || !*fp[5]
+ || (state = strtoul(fp[5], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[6] || !*fp[6]
+ || (txq = strtoul(fp[6], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[7] || !*fp[7]
+ || (rxq = strtoul(fp[7], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ /*
+ * Get the inode and use it for hashing and searching.
+ */
+ ep = (char *)NULL;
+ if (!fp[13] || !*fp[13]
+ || (inode = strtoull(fp[13], &ep, 0)) == ULONG_MAX || !ep || *ep)
+ continue;
+ h = TCPUDPHASH(inode);
+ for (tp = TcpUdp[h]; tp; tp = tp->next) {
+ if (tp->inode == inode)
+ break;
+ }
+ if (tp)
+ continue;
+ /*
+ * Create a new entry and link it to its hash bucket.
+ */
+ if (!(tp = (struct tcp_udp *)malloc(sizeof(struct tcp_udp)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for tcp_udp struct\n",
+ Pn, sizeof(struct tcp_udp));
+ Exit(1);
+ }
+ tp->inode = inode;
+ tp->faddr = faddr;
+ tp->fport = (int)(fport & 0xffff);
+ tp->laddr = laddr;
+ tp->lport = (int)(lport & 0xffff);
+ tp->txq = txq;
+ tp->rxq = rxq;
+ tp->proto = pr;
+ tp->state = (int)state;
+ tp->next = TcpUdp[h];
+ TcpUdp[h] = tp;
+ }
+ (void) fclose(fs);
+}
+
+
+#if defined(HASIPv6)
+/*
+ * get_raw6() - get /proc/net/raw6 info
+ */
+
+static void
+get_raw6(p)
+ char *p; /* /proc/net/raw path */
+{
+ char buf[MAXPATHLEN], *ep, **fp, *la, *ra, *sp;
+ int h;
+ INODETYPE inode;
+ int nf = 12;
+ struct rawsin *np, *rp;
+ MALLOC_S lal, ral, spl;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+ FILE *xs;
+/*
+ * Do second time cleanup or first time setup.
+ */
+ if (Rawsin6) {
+ for (h = 0; h < INOBUCKS; h++) {
+ for (rp = Rawsin6[h]; rp; rp = np) {
+ np = rp->next;
+ if (rp->la)
+ (void) free((FREE_P *)rp->la);
+ if (rp->ra)
+ (void) free((FREE_P *)rp->ra);
+ (void) free((FREE_P *)rp);
+ }
+ Rawsin6[h] = (struct rawsin *)NULL;
+ }
+ } else {
+ Rawsin6 = (struct rawsin **)calloc(INOBUCKS,
+ sizeof(struct rawsin *));
+ if (!Rawsin6) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d raw6 hash pointer bytes\n",
+ Pn, INOBUCKS * sizeof(struct rawsin *));
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc/net/raw6 file, assign a page size buffer to the stream,
+ * and read it. Store raw6 socket info in the Rawsin6[] hash buckets.
+ */
+ if (!(xs = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ while (fgets(buf, sizeof(buf) - 1, xs)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) < nf)
+ continue;
+ if (nf == 12) {
+
+ /*
+ * Check the column labels in the first line.
+ */
+ if (!fp[1] || strcmp(fp[1], "local_address")
+ || !fp[2] || strcmp(fp[2], "remote_address")
+ || !fp[3] || strcmp(fp[3], "st")
+ || !fp[11] || strcmp(fp[11], "inode"))
+ {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: unsupported format: %s\n",
+ Pn, p);
+ }
+ break;
+ }
+ nf = 10;
+ continue;
+ }
+ /*
+ * Assemble the inode number and see if the inode is already
+ * recorded.
+ */
+ ep = (char *)NULL;
+ if (!fp[9] || !*fp[9]
+ || (inode = strtoull(fp[9], &ep, 0)) == ULONG_MAX
+ || !ep || *ep)
+ continue;
+ h = INOHASH(inode);
+ for (rp = Rawsin6[h]; rp; rp = rp->next) {
+ if (inode == rp->inode)
+ break;
+ }
+ if (rp)
+ continue;
+ /*
+ * Save the local address, remote address, and state.
+ */
+ if (!fp[1] || !*fp[1] || (lal = strlen(fp[1])) < 1) {
+ la = (char *)NULL;
+ lal = (MALLOC_S)0;
+ } else {
+ if (!(la = (char *)malloc(lal + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d local raw6 address bytes: %s\n",
+ Pn, lal + 1, fp[1]);
+ Exit(1);
+ }
+ (void) snpf(la, lal + 1, "%s", fp[1]);
+ }
+ if (!fp[2] || !*fp[2] || (ral = strlen(fp[2])) < 1) {
+ ra = (char *)NULL;
+ ral = (MALLOC_S)0;
+ } else {
+ if (!(ra = (char *)malloc(ral + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d remote raw6 address bytes: %s\n",
+ Pn, ral + 1, fp[2]);
+ Exit(1);
+ }
+ (void) snpf(ra, ral + 1, "%s", fp[2]);
+ }
+ if (!fp[3] || !*fp[3] || (spl = strlen(fp[3])) < 1) {
+ sp = (char *)NULL;
+ spl = (MALLOC_S)0;
+ } else {
+ if (!(sp = (char *)malloc(spl + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d remote raw6 state bytes: %s\n",
+ Pn, spl + 1, fp[2]);
+ Exit(1);
+ }
+ (void) snpf(sp, spl + 1, "%s", fp[3]);
+ }
+ /*
+ * Allocate space for an rawsin entry, fill it, and link it to its
+ * hash bucket.
+ */
+ if (!(rp = (struct rawsin *)malloc(sizeof(struct rawsin)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d byte rawsin structure for IPv6\n",
+ Pn, sizeof(struct rawsin));
+ Exit(1);
+ }
+ rp->inode = inode;
+ rp->la = la;
+ rp->lal = lal;
+ rp->ra = ra;
+ rp->ral = ral;
+ rp->sp = sp;
+ rp->spl = spl;
+ rp->next = Rawsin6[h];
+ Rawsin6[h] = rp;
+ }
+ (void) fclose(xs);
+}
+
+
+/*
+ * get_tcpudp6() - get IPv6 TCP, UDP or UDPLITE net info
+ */
+
+static void
+get_tcpudp6(p, pr, clr)
+ char *p; /* /proc/net/{tcp,udp} path */
+ int pr; /* protocol: 0 = TCP, 1 = UDP */
+ int clr; /* 1 == clear the table */
+{
+ char buf[MAXPATHLEN], *ep, **fp;
+ struct in6_addr faddr, laddr;
+ unsigned long fport, lport, rxq, state, txq;
+ FILE *fs;
+ int h, i, nf;
+ INODETYPE inode;
+ struct tcp_udp6 *np6, *tp6;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+/*
+ * Delete previous table contents. Allocate a table for the first time.
+ */
+ if (TcpUdp6) {
+ if (clr) {
+ for (h = 0; h < TcpUdp6_bucks; h++) {
+ for (tp6 = TcpUdp6[h]; tp6; tp6 = np6) {
+ np6 = tp6->next;
+ (void) free((FREE_P *)tp6);
+ }
+ TcpUdp6[h] = (struct tcp_udp6 *)NULL;
+ }
+ }
+ } else {
+
+ /*
+ * Open the /proc/net/sockstat6 file and establish the hash bucket
+ * count from its "TCP6: inuse" and "UDP6: inuse" lines.
+ */
+ TcpUdp6_bucks = INOBUCKS;
+ h = i = nf = 0;
+ if ((fs = fopen(SockStatPath6, "r"))) {
+ while(fgets(buf, sizeof(buf) - 1, fs)) {
+ if (get_fields(buf, (char *)NULL, &fp, (int *)NULL, 0) != 3)
+ continue;
+ if (!fp[0]
+ || !fp[1] || strcmp(fp[1], "inuse")
+ || !fp[2] || !*fp[2])
+ continue;
+ if (!strcmp(fp[0], "TCP6:")) {
+ nf |= 1;
+ if ((h = atoi(fp[2])) < 1)
+ h = INOBUCKS;
+ i += h;
+ } else if (!strcmp(fp[0], "UDP6:")) {
+ nf |= 2;
+ if ((h = atoi(fp[2])) < 1)
+ h = INOBUCKS;
+ i += h;
+ } else
+ continue;
+ if (nf == 3) {
+ while (TcpUdp6_bucks < i)
+ TcpUdp6_bucks *= 2;
+ break;
+ }
+ }
+ (void) fclose(fs);
+ }
+ if (!(TcpUdp6 = (struct tcp_udp6 **)calloc(TcpUdp6_bucks,
+ sizeof(struct tcp_udp6 *))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for TCP6&UDP6 hash buckets\n",
+ Pn, (int)(TcpUdp6_bucks * sizeof(struct tcp_udp6 *)));
+ Exit(1);
+ }
+ }
+/*
+ * Open the /proc/net file, assign a page size buffer to the stream,
+ * and read it.
+ */
+ if (!(fs = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ nf = 12;
+ while(fgets(buf, sizeof(buf) - 1, fs)) {
+ if (get_fields(buf,
+ (nf == 12) ? (char *)NULL : ":",
+ &fp, (int *)NULL, 0)
+ < nf)
+ continue;
+ if (nf == 12) {
+ if (!fp[1] || strcmp(fp[1], "local_address")
+ || !fp[2] || strcmp(fp[2], "remote_address")
+ || !fp[3] || strcmp(fp[3], "st")
+ || !fp[4] || strcmp(fp[4], "tx_queue")
+ || !fp[5] || strcmp(fp[5], "rx_queue")
+ || !fp[11] || strcmp(fp[11], "inode"))
+ {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: unsupported format: %s\n",
+ Pn, p);
+ }
+ break;
+ }
+ nf = 14;
+ continue;
+ }
+ /*
+ * Get the local and remote addresses.
+ */
+ if (!fp[1] || !*fp[1] || net6a2in6(fp[1], &laddr))
+ continue;
+ ep = (char *)NULL;
+ if (!fp[2] || !*fp[2]
+ || (lport = strtoul(fp[2], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ if (!fp[3] || !*fp[3] || net6a2in6(fp[3], &faddr))
+ continue;
+ ep = (char *)NULL;
+ if (!fp[4] || !*fp[4]
+ || (fport = strtoul(fp[4], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ /*
+ * Get the state and queue sizes.
+ */
+ ep = (char *)NULL;
+ if (!fp[5] || !*fp[5]
+ || (state = strtoul(fp[5], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[6] || !*fp[6]
+ || (txq = strtoul(fp[6], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ ep = (char *)NULL;
+ if (!fp[7] || !*fp[7]
+ || (rxq = strtoul(fp[7], &ep, 16)) == ULONG_MAX || !ep || *ep)
+ continue;
+ /*
+ * Get the inode and use it for hashing and searching.
+ */
+ ep = (char *)NULL;
+ if (!fp[13] || !*fp[13]
+ || (inode = strtoull(fp[13], &ep, 0)) == ULONG_MAX || !ep || *ep)
+ continue;
+ h = TCPUDP6HASH(inode);
+ for (tp6 = TcpUdp6[h]; tp6; tp6 = tp6->next) {
+ if (tp6->inode == inode)
+ break;
+ }
+ if (tp6)
+ continue;
+ /*
+ * Create a new entry and link it to its hash bucket.
+ */
+ if (!(tp6 = (struct tcp_udp6 *)malloc(sizeof(struct tcp_udp6)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for tcp_udp6 struct\n",
+ Pn, sizeof(struct tcp_udp6));
+ Exit(1);
+ }
+ tp6->inode = inode;
+ tp6->faddr = faddr;
+ tp6->fport = (int)(fport & 0xffff);
+ tp6->laddr = laddr;
+ tp6->lport = (int)(lport & 0xffff);
+ tp6->txq = txq;
+ tp6->rxq = rxq;
+ tp6->proto = pr;
+ tp6->state = (int)state;
+ tp6->next = TcpUdp6[h];
+ TcpUdp6[h] = tp6;
+ }
+ (void) fclose(fs);
+}
+#endif /* defined(HASIPv6) */
+
+
+/*
+ * get_unix() - get UNIX net info
+ */
+
+static void
+get_unix(p)
+ char *p; /* /proc/net/unix path */
+{
+ char buf[MAXPATHLEN], *ep, **fp, *path, *pcb;
+ int fl = 1;
+ int h, nf;
+ INODETYPE inode;
+ MALLOC_S len;
+ struct uxsin *np, *up;
+ FILE *us;
+ static char *vbuf = (char *)NULL;
+ static size_t vsz = (size_t)0;
+/*
+ * Do second time cleanup or first time setup.
+ */
+ if (Uxsin) {
+ for (h = 0; h < INOBUCKS; h++) {
+ for (up = Uxsin[h]; up; up = np) {
+ np = up->next;
+ if (up->path)
+ (void) free((FREE_P *)up->path);
+ if (up->pcb)
+ (void) free((FREE_P *)up->pcb);
+ (void) free((FREE_P *)up);
+ }
+ Uxsin[h] = (struct uxsin *)NULL;
+ }
+ } else {
+ Uxsin = (struct uxsin **)calloc(INOBUCKS, sizeof(struct uxsin *));
+ if (!Uxsin) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for Unix socket info\n",
+ Pn, INOBUCKS * sizeof(struct uxsin *));
+ }
+ }
+/*
+ * Open the /proc/net/unix file, assign a page size buffer to the stream,
+ * read the file's contents, and add them to the Uxsin hash buckets.
+ */
+ if (!(us = open_proc_stream(p, "r", &vbuf, &vsz, 0)))
+ return;
+ while (fgets(buf, sizeof(buf) - 1, us)) {
+ if ((nf = get_fields(buf, ":", &fp, (int *)NULL, 0)) < 7)
+ continue;
+ if (fl) {
+
+ /*
+ * Check the first line for header words.
+ */
+ if (!fp[0] || strcmp(fp[0], "Num")
+ || !fp[1] || strcmp(fp[1], "RefCount")
+ || !fp[2] || strcmp(fp[2], "Protocol")
+ || !fp[3] || strcmp(fp[3], "Flags")
+ || !fp[4] || strcmp(fp[4], "Type")
+ || !fp[5] || strcmp(fp[5], "St")
+ || !fp[6] || strcmp(fp[6], "Inode")
+ || nf < 8
+ || !fp[7] || strcmp(fp[7], "Path"))
+ {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: unsupported format: %s\n",
+ Pn, p);
+ }
+ break;
+ }
+ fl = 0;
+ continue;
+ }
+ /*
+ * Assemble PCB address, inode number, and path name. If this
+ * inode is already represented in Uxsin, skip it.
+ */
+ ep = (char *)NULL;
+ if (!fp[6] || !*fp[6]
+ || (inode = strtoull(fp[6], &ep, 0)) == ULONG_MAX || !ep || *ep)
+ continue;
+ h = INOHASH(inode);
+ for (up = Uxsin[h]; up; up = up->next) {
+ if (inode == up->inode)
+ break;
+ }
+ if (up)
+ continue;
+ if (!fp[0] || !*fp[0])
+ pcb = (char *)NULL;
+ else {
+ len = strlen(fp[0]) + 2;
+ if (!(pcb = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for UNIX PCB: %s\n",
+ Pn, len + 1, fp[0]);
+ Exit(1);
+ }
+ (void) snpf(pcb, len + 1, "0x%s", fp[0]);
+ }
+ if (nf >= 8
+ && fp[7] && *fp[7] && *fp[7] != '@'
+ && (len = strlen(fp[7]))) {
+ if (!(path = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for UNIX path \"%s\"\n",
+ Pn, len + 1, fp[7]);
+ Exit(1);
+ }
+ (void) snpf(path, len + 1, "%s", fp[7]);
+ } else
+ path = (char *)NULL;
+ /*
+ * Allocate and fill a Unix socket info structure; link it to its
+ * hash bucket.
+ */
+ if (!(up = (struct uxsin *)malloc(sizeof(struct uxsin)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for uxsin struct\n",
+ Pn, sizeof(struct uxsin));
+ Exit(1);
+ }
+ up->inode = inode;
+ up->pcb = pcb;
+ up->sb_def = 0;
+ if ((up->path = path) && (*path == '/')) {
+
+ /*
+ * If an absolute path (i.e., one that begins with a '/') exists
+ * for the line, attempt to stat(2) it and save the device and
+ * node numbers reported in the stat buffer.
+ */
+ struct stat sb;
+ int sr;
+
+ if (HasNFS)
+ sr = statsafely(path, &sb);
+ else
+ sr = stat(path, &sb);
+ if (sr && ((sb.st_mode & S_IFMT) == S_IFSOCK)) {
+ up->sb_def = 1;
+ up->sb_dev = sb.st_dev;
+ up->sb_ino = (INODETYPE)sb.st_ino;
+ up->sb_rdev = sb.st_rdev;
+ }
+ }
+ up->next = Uxsin[h];
+ Uxsin[h] = up;
+ }
+ (void) fclose(us);
+}
+
+
+#if defined(HASIPv6)
+/*
+ * net6a2in6() - convert ASCII IPv6 address in /proc/net/{tcp,udp} form to
+ * an in6_addr
+ */
+
+static int
+net6a2in6(as, ad)
+ char *as; /* address source */
+ struct in6_addr *ad; /* address destination */
+{
+ char buf[9], *ep;
+ int i;
+ size_t len;
+/*
+ * Assemble four uint32_t's from 4 X 8 hex digits into s6_addr32[].
+ */
+ for (i = 0, len = strlen(as);
+ (i < 4) && (len >= 8);
+ as += 8, i++, len -= 8)
+ {
+ (void) strncpy(buf, as, 8);
+ buf[8] = '\0';
+ ep = (char *)NULL;
+ if ((ad->s6_addr32[i] = (uint32_t)strtoul(buf, &ep, 16))
+ == (uint32_t)UINT32_MAX || !ep || *ep)
+ break;
+ }
+ return((*as || (i != 4) || len) ? 1 : 0);
+}
+#endif /* defined(HASIPv6) */
+
+
+/*
+ * print_ax25info() - print AX25 socket info
+ */
+
+static void
+print_ax25info(ap)
+ struct ax25sin *ap; /* AX25 socket info */
+{
+ char *cp, pbuf[1024];
+ int ds;
+ MALLOC_S pl = (MALLOC_S)0;
+
+ if (Lf->nma)
+ return;
+ if (ap->sa) {
+ ds = (ap->da && strcmp(ap->da, "*")) ? 1 : 0;
+ (void) snpf(&pbuf[pl], sizeof(pbuf) - pl, "%s%s%s ", ap->sa,
+ ds ? "->" : "",
+ ds ? ap->da : "");
+ pl = strlen(pbuf);
+ }
+ if (ap->sqs) {
+ (void) snpf(&pbuf[pl], sizeof(pbuf) - pl, "(Sq=%lu ", ap->sq);
+ pl = strlen(pbuf);
+ cp = "";
+ } else
+ cp = "(";
+ if (ap->rqs) {
+ (void) snpf(&pbuf[pl], sizeof(pbuf) - pl, "%sRq=%lu ", cp, ap->rq);
+ pl = strlen(pbuf);
+ cp = "";
+ }
+ (void) snpf(&pbuf[pl], sizeof(pbuf) - pl, "%sState=%d", cp, ap->state);
+ pl = strlen(pbuf);
+ if ((ap->state >= 0) && (ap->state < NAX25ST))
+ cp = ax25st[ap->state];
+ else
+ cp = NULL;
+ (void) snpf(&pbuf[pl], sizeof(pbuf) - pl, "%s%s)",
+ cp ? ", " : "",
+ cp ? cp : "");
+ pl = strlen(pbuf);
+ if (!(cp = (char *)malloc(pl + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for AX25 sock state, PID: %d\n",
+ Pn, pl + 1, Lp->pid);
+ Exit(1);
+ }
+ (void) snpf(cp, pl + 1, "%s", pbuf);
+ Lf->nma = cp;
+}
+
+
+/*
+ * print_ipxinfo() - print IPX socket info
+ */
+
+static void
+print_ipxinfo(ip)
+ struct ipxsin *ip; /* IPX socket info */
+{
+ char *cp, pbuf[256];
+ MALLOC_S pl;
+
+ if (Lf->nma)
+ return;
+ (void) snpf(pbuf, sizeof(pbuf), "(Tx=%lx Rx=%lx State=%02x)",
+ ip->txq, ip->rxq, ip->state);
+ pl = strlen(pbuf);
+ if (!(cp = (char *)malloc(pl + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for IPX sock state, PID: %d\n",
+ Pn, pl + 1, Lp->pid);
+ Exit(1);
+ }
+ (void) snpf(cp, pl + 1, "%s", pbuf);
+ Lf->nma = cp;
+}
+
+
+/*
+ * print_tcptpi() - print TCP/TPI state
+ */
+
+void
+print_tcptpi(nl)
+ int nl; /* 1 == '\n' required */
+{
+ char buf[128];
+ char *cp = (char *)NULL;
+ int ps = 0;
+ int s;
+
+ if ((Ftcptpi & TCPTPI_STATE) && Lf->lts.type == 0) {
+ if (!TcpSt)
+ (void) build_IPstates();
+ if ((s = Lf->lts.state.i + TcpStOff) < 0 || s >= TcpNstates) {
+ (void) snpf(buf, sizeof(buf), "UNKNOWN_TCP_STATE_%d",
+ Lf->lts.state.i);
+ cp = buf;
+ } else
+ cp = TcpSt[s];
+ if (cp) {
+ if (Ffield)
+ (void) printf("%cST=%s%c", LSOF_FID_TCPTPI, cp, Terminator);
+ else {
+ putchar('(');
+ (void) fputs(cp, stdout);
+ }
+ ps++;
+ }
+ }
+
+# if defined(HASTCPTPIQ)
+ if (Ftcptpi & TCPTPI_QUEUES) {
+ if (Lf->lts.rqs) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("QR=%lu", Lf->lts.rq);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ if (Lf->lts.sqs) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("QS=%lu", Lf->lts.sq);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ }
+# endif /* defined(HASTCPTPIQ) */
+
+# if defined(HASTCPTPIW)
+ if (Ftcptpi & TCPTPI_WINDOWS) {
+ if (Lf->lts.rws) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("WR=%lu", Lf->lts.rw);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ if (Lf->lts.wws) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("WW=%lu", Lf->lts.ww);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ }
+# endif /* defined(HASTCPTPIW) */
+
+ if (!Ffield && ps)
+ putchar(')');
+ if (nl)
+ putchar('\n');
+}
+
+
+/*
+ * process_proc_sock() - process /proc-based socket
+ */
+
+void
+process_proc_sock(p, s, ss, l, lss)
+ char *p; /* node's readlink() path */
+ struct stat *s; /* stat() result for path */
+ int ss; /* *s status -- i.e, SB_* values */
+ struct stat *l; /* lstat() result for FD (NULL for
+ * others) */
+ int lss; /* *l status -- i.e, SB_* values */
+{
+ struct ax25sin *ap;
+ char *cp, dev_ch[32], *path;
+ unsigned char *fa, *la;
+ struct in_addr fs, ls;
+ struct ipxsin *ip;
+ int i, len, nl;
+ struct packin *pp;
+ char *pr;
+ struct rawsin *rp;
+ struct tcp_udp *tp;
+ struct uxsin *up;
+
+#if defined(HASIPv6)
+ int af;
+ struct tcp_udp6 *tp6;
+#endif /* defined(HASIPv6) */
+
+/*
+ * Enter offset, if possible.
+ */
+ if (Foffset || !Fsize) {
+ if (l && (lss & SB_SIZE) && OffType) {
+ Lf->off = (SZOFFTYPE)l->st_size;
+ Lf->off_def = 1;
+ }
+ }
+/*
+ * Check for socket's inode presence in the protocol info caches.
+ */
+ if (AX25path) {
+ (void) get_ax25(AX25path);
+ (void) free((FREE_P *)AX25path);
+ AX25path = (char *)NULL;
+ }
+ if ((ss & SB_INO)
+ && (ap = check_ax25((INODETYPE)s->st_ino))
+ ) {
+
+ /*
+ * The inode is connected to an AX25 /proc record.
+ *
+ * Set the type to "ax25"; save the device name; save the inode number;
+ * save the destination and source addresses; save the send and receive
+ * queue sizes; and save the connection state.
+ */
+ (void) snpf(Lf->type, sizeof(Lf->type), "ax25");
+ if (ap->dev_ch)
+ (void) enter_dev_ch(ap->dev_ch);
+ Lf->inode = ap->inode;
+ Lf->inp_ty = 1;
+ print_ax25info(ap);
+ return;
+ }
+ if (Ipxpath) {
+ (void) get_ipx(Ipxpath);
+ (void) free((FREE_P *)Ipxpath);
+ Ipxpath = (char *)NULL;
+ }
+ if ((ss & SB_INO)
+ && (ip = check_ipx((INODETYPE)s->st_ino))
+ ) {
+
+ /*
+ * The inode is connected to an IPX /proc record.
+ *
+ * Set the type to "ipx"; enter the inode and device numbers; store
+ * the addresses, queue sizes, and state in the NAME column.
+ */
+ (void) snpf(Lf->type, sizeof(Lf->type), "ipx");
+ if (ss & SB_INO) {
+ Lf->inode = (INODETYPE)s->st_ino;
+ Lf->inp_ty = 1;
+ }
+ if (ss & SB_DEV) {
+ Lf->dev = s->st_dev;
+ Lf->dev_def = 1;
+ }
+ cp = Namech;
+ nl = Namechl;
+ *cp = '\0';
+ if (ip->la && nl) {
+
+ /*
+ * Store the local IPX address.
+ */
+ len = strlen(ip->la);
+ if (len > nl)
+ len = nl;
+ (void) strncpy(cp, ip->la, len);
+ cp += len;
+ *cp = '\0';
+ nl -= len;
+ }
+ if (ip->ra && nl) {
+
+ /*
+ * Store the remote IPX address, prefixed with "->".
+ */
+ if (nl > 2) {
+ (void) snpf(cp, nl, "->");
+ cp += 2;
+ nl -= 2;
+ }
+ if (nl) {
+ (void) snpf(cp, nl, "%s", ip->ra);
+ cp += len;
+ nl -= len;
+ }
+ }
+ (void) print_ipxinfo(ip);
+ if (Namech[0])
+ enter_nm(Namech);
+ return;
+ }
+ if (Rawpath) {
+ (void) get_raw(Rawpath);
+ (void) free((FREE_P *)Rawpath);
+ Rawpath = (char *)NULL;
+ }
+ if ((ss & SB_INO)
+ && (rp = check_raw((INODETYPE)s->st_ino))
+ ) {
+
+ /*
+ * The inode is connected to a raw /proc record.
+ *
+ * Set the type to "raw"; enter the inode number; store the local
+ * address, remote address, and state in the NAME column.
+ */
+ (void) snpf(Lf->type, sizeof(Lf->type), "raw");
+ if (ss & SB_INO) {
+ Lf->inode = (INODETYPE)s->st_ino;
+ Lf->inp_ty = 1;
+ }
+ cp = Namech;
+ nl = Namechl - 2;
+ *cp = '\0';
+ if (rp->la && rp->lal) {
+
+ /*
+ * Store the local raw address.
+ */
+ if (nl > rp->lal) {
+ (void) snpf(cp, nl, "%s", rp->la);
+ cp += rp->lal;
+ *cp = '\0';
+ nl -= rp->lal;
+ }
+ }
+ if (rp->ra && rp->ral) {
+
+ /*
+ * Store the remote raw address, prefixed with "->".
+ */
+ if (nl > (rp->ral + 2)) {
+ (void) snpf(cp, nl, "->%s", rp->ra);
+ cp += (rp->ral + 2);
+ nl -= (rp->ral + 2);
+ }
+ }
+ if (rp->sp && rp->spl) {
+
+ /*
+ * Store the state, optionally prefixed by a space, in the
+ * form "st=x...x".
+ */
+
+ if (nl > (len = ((cp == Namech) ? 0 : 1) + 3 + rp->spl)) {
+ (void) snpf(cp, nl, "%sst=%s",
+ (cp == Namech) ? "" : " ", rp->sp);
+ cp += len;
+ *cp = '\0';
+ nl -= len;
+ }
+ }
+ if (Namech[0])
+ enter_nm(Namech);
+ return;
+ }
+ if (Packpath) {
+ (void) get_pack(Packpath);
+ (void) free((FREE_P *)Packpath);
+ Packpath = (char *)NULL;
+ }
+ if ((ss & SB_INO)
+ && (pp = check_pack((INODETYPE)s->st_ino))
+ ) {
+
+ /*
+ * The inode is connected to a packet /proc record.
+ *
+ * Set the type to "pack" and store the socket type in the NAME
+ * column. Put the protocol name in the NODE column and the inode
+ * number in the DEVICE column.
+ */
+ (void) snpf(Lf->type, sizeof(Lf->type), "pack");
+ for (i = 0; i < NPACKTY; i++) {
+ if (Packty[i].ty == pp->ty) {
+ (void) snpf(Namech, Namechl, "type=SOCK_%s",
+ Packty[i].nm);
+ break;
+ }
+ }
+ for (i = 0; i < NPACKPR; i++) {
+ if (Packpr[i].pr == pp->pr)
+ break;
+ }
+ cp = (i < NPACKPR) ? Packpr[i].nm : "unknown";
+ (void) snpf(Lf->iproto, sizeof(Lf->iproto), "%.*s", IPROTOL-1, cp);
+ Lf->inp_ty = 2;
+ if (ss & SB_INO) {
+ (void) snpf(dev_ch, sizeof(dev_ch), InodeFmt_d,
+ (INODETYPE)s->st_ino);
+ enter_dev_ch(dev_ch);
+ }
+ if (Namech[0])
+ enter_nm(Namech);
+ return;
+ }
+ if (UNIXpath) {
+ (void) get_unix(UNIXpath);
+ (void) free((FREE_P *)UNIXpath);
+ UNIXpath = (char *)NULL;
+ }
+ if ((ss & SB_INO)
+ && (up = check_unix((INODETYPE)s->st_ino))
+ ) {
+
+ /*
+ * The inode is connected to a UNIX /proc record.
+ *
+ * Set the type to "unix"; enter the PCB address in the DEVICE column;
+ * enter the inode number; and save the optional path.
+ */
+ if (Funix)
+ Lf->sf |= SELUNX;
+ (void) snpf(Lf->type, sizeof(Lf->type), "unix");
+ if (up->pcb)
+ enter_dev_ch(up->pcb);
+ if (ss & SB_INO) {
+ Lf->inode = (INODETYPE)s->st_ino;
+ Lf->inp_ty = 1;
+ }
+ path = up->path ? up->path : p;
+ (void) enter_nm(path);
+ if (Sfile) {
+
+ /*
+ * See if this UNIX domain socket was specified as a search
+ * argument.
+ *
+ * Search first by device and node numbers, if that is possible;
+ * then search by name.
+ */
+ unsigned char f = 0; /* file-found flag */
+
+ if (up->sb_def) {
+
+ /*
+ * If the UNIX socket information includes stat(2) results, do
+ * a device and node number search.
+ *
+ * Note: that requires the saving, temporary modification and
+ * restoration of some *Lf values.
+ */
+ unsigned char sv_dev_def; /* saved dev_def */
+ unsigned char sv_inp_ty; /* saved inp_ty */
+ unsigned char sv_rdev_def; /* saved rdev_def */
+ dev_t sv_dev; /* saved dev */
+ INODETYPE sv_inode; /* saved inode */
+ dev_t sv_rdev; /* saved rdev */
+
+ sv_dev_def = Lf->dev_def;
+ sv_dev = Lf->dev;
+ sv_inode = Lf->inode;
+ sv_inp_ty = Lf->inp_ty;
+ sv_rdev_def = Lf->rdev_def;
+ sv_rdev = Lf->rdev;
+ Lf->dev_def = Lf->inp_ty = Lf->rdev_def = 1;
+ Lf->dev = up->sb_dev;
+ Lf->inode = up->sb_ino;
+ Lf->rdev = up->sb_rdev;
+ if (is_file_named((char *)NULL, 0)) {
+ f = 1;
+ Lf->sf |= SELNM;
+ }
+ Lf->dev_def = sv_dev_def;
+ Lf->dev = sv_dev;
+ Lf->inode = sv_inode;
+ Lf->inp_ty = sv_inp_ty;
+ Lf->rdev_def = sv_rdev_def;
+ Lf->rdev = sv_rdev;
+ }
+ if (!f && (ss & SB_MODE)) {
+
+ /*
+ * If the file has not yet been found and the stat buffer has
+ * st_mode, search for the file by full path.
+ */
+ if (is_file_named(path,
+ ((s->st_mode & S_IFMT) == S_IFCHR)) ? 1 : 0)
+ {
+ Lf->sf |= SELNM;
+ }
+ }
+ }
+ return;
+ }
+
+#if defined(HASIPv6)
+ if (Raw6path) {
+ if (!Fxopt)
+ (void) get_raw6(Raw6path);
+ (void) free((FREE_P *)Raw6path);
+ Raw6path = (char *)NULL;
+ }
+ if (!Fxopt && (ss & SB_INO)
+ && (rp = check_raw6((INODETYPE)s->st_ino))
+ ) {
+
+ /*
+ * The inode is connected to a raw IPv6 /proc record.
+ *
+ * Set the type to "raw6"; enter the inode number; store the local
+ * address, remote address, and state in the NAME column.
+ */
+ (void) snpf(Lf->type, sizeof(Lf->type), "raw6");
+ if (ss & SB_INO) {
+ Lf->inode = (INODETYPE)s->st_ino;
+ Lf->inp_ty = 1;
+ }
+ cp = Namech;
+ nl = MAXPATHLEN - 2;
+ if (rp->la && rp->lal) {
+
+ /*
+ * Store the local raw IPv6 address.
+ */
+ if (nl > rp->lal) {
+ (void) snpf(cp, nl, "%s", rp->la);
+ cp += rp->lal;
+ *cp = '\0';
+ nl -= rp->lal;
+ }
+ }
+ if (rp->ra && rp->ral) {
+
+ /*
+ * Store the remote raw address, prefixed with "->".
+ */
+ if (nl > (rp->ral + 2)) {
+ (void) snpf(cp, nl, "->%s", rp->ra);
+ cp += (rp->ral + 2);
+ nl -= (rp->ral + 2);
+ }
+ }
+ if (rp->sp && rp->spl) {
+
+ /*
+ * Store the state, optionally prefixed by a space, in the
+ * form "st=x...x".
+ */
+
+ if (nl > (len = ((cp == Namech) ? 0 : 1) + 3 + rp->spl)) {
+ (void) snpf(cp, nl, "%sst=%s",
+ (cp == Namech) ? "" : " ", rp->sp);
+ cp += len;
+ *cp = '\0';
+ nl -= len;
+ }
+ }
+ if (Namech[0])
+ enter_nm(Namech);
+ return;
+ }
+ if (TCP6path) {
+ if (!Fxopt)
+ (void) get_tcpudp6(TCP6path, 0, 1);
+ (void) free((FREE_P *)TCP6path);
+ TCP6path = (char *)NULL;
+ }
+ if (UDP6path) {
+ if (!Fxopt)
+ (void) get_tcpudp6(UDP6path, 1, 0);
+ (void) free((FREE_P *)UDP6path);
+ UDP6path = (char *)NULL;
+ }
+ if (UDP6LITEpath) {
+ if (!Fxopt)
+ (void) get_tcpudp6(UDP6LITEpath, 2, 0);
+ (void) free((FREE_P *)UDP6LITEpath);
+ UDP6LITEpath = (char *)NULL;
+ }
+ if (!Fxopt && (ss & SB_INO)
+ && (tp6 = check_tcpudp6((INODETYPE)s->st_ino, &pr))
+ ) {
+
+ /*
+ * The inode is connected to an IPv6 TCP or UDP /proc record.
+ *
+ * Set the type to "IPv6"; enter the protocol; put the inode number
+ * in the DEVICE column in lieu of the PCB address; save the local
+ * and foreign IPv6 addresses; save the type and protocol; and
+ * (optionally) save the queue sizes.
+ */
+ i = tp6->state + TcpStOff;
+ if (TcpStXn) {
+
+ /*
+ * Check for state exclusion.
+ */
+ if (i >= 0 && i < TcpNstates) {
+ if (TcpStX[i]) {
+ Lf->sf |= SELEXCLF;
+ return;
+ }
+ }
+ }
+ if (TcpStIn) {
+
+ /*
+ * Check for state inclusion.
+ */
+ if (i >= 0 && i < TcpNstates) {
+ if (TcpStI[i])
+ TcpStI[i] = 2;
+ else {
+ Lf->sf |= SELEXCLF;
+ return;
+ }
+ }
+ }
+ if (Fnet && (FnetTy != 4))
+ Lf->sf |= SELNET;
+ (void) snpf(Lf->type, sizeof(Lf->type), "IPv6");
+ (void) snpf(Lf->iproto, sizeof(Lf->iproto), "%.*s", IPROTOL-1, pr);
+ Lf->inp_ty = 2;
+ if (ss && SB_INO) {
+ (void) snpf(dev_ch, sizeof(dev_ch), InodeFmt_d,
+ (INODETYPE)s->st_ino);
+ enter_dev_ch(dev_ch);
+ }
+ af = AF_INET6;
+ if (!IN6_IS_ADDR_UNSPECIFIED(&tp6->faddr) || tp6->fport)
+ fa = (unsigned char *)&tp6->faddr;
+ else
+ fa = (unsigned char *)NULL;
+ if (!IN6_IS_ADDR_UNSPECIFIED(&tp6->laddr) || tp6->lport)
+ la = (unsigned char *)&tp6->laddr;
+ else
+ la = (unsigned char *)NULL;
+ if ((fa && IN6_IS_ADDR_V4MAPPED(&tp6->faddr))
+ || (la && IN6_IS_ADDR_V4MAPPED(&tp6->laddr))) {
+ af = AF_INET;
+ if (fa)
+ fa += 12;
+ if (la)
+ la += 12;
+ }
+ ent_inaddr(la, tp6->lport, fa, tp6->fport, af);
+ Lf->lts.type = tp6->proto;
+ Lf->lts.state.i = tp6->state;
+
+#if defined(HASTCPTPIQ)
+ Lf->lts.rq = tp6->rxq;
+ Lf->lts.sq = tp6->txq;
+ Lf->lts.rqs = Lf->lts.sqs = 1;
+#endif /* defined(HASTCPTPIQ) */
+
+ return;
+ }
+#endif /* defined(HASIPv6) */
+
+ if (TCPpath) {
+ if (!Fxopt)
+ (void) get_tcpudp(TCPpath, 0, 1);
+ (void) free((FREE_P *)TCPpath);
+ TCPpath = (char *)NULL;
+ }
+ if (UDPpath) {
+ if (!Fxopt)
+ (void) get_tcpudp(UDPpath, 1, 0);
+ (void) free((FREE_P *)UDPpath);
+ UDPpath = (char *)NULL;
+ }
+ if (UDPLITEpath) {
+ if (!Fxopt)
+ (void) get_tcpudp(UDPLITEpath, 2, 0);
+ (void) free((FREE_P *)UDPLITEpath);
+ UDPLITEpath = (char *)NULL;
+ }
+ if (!Fxopt && (ss & SB_INO)
+ && (tp = check_tcpudp((INODETYPE)s->st_ino, &pr))
+ ) {
+
+ /*
+ * The inode is connected to an IPv4 TCP or UDP /proc record.
+ *
+ * Set the type to "inet" or "IPv4"; enter the protocol; put the
+ * inode number in the DEVICE column in lieu of the PCB address;
+ * save the local and foreign IPv4 addresses; save the type and
+ * protocol; and (optionally) save the queue sizes.
+ */
+ i = tp->state + TcpStOff;
+ if (TcpStXn) {
+
+ /*
+ * Check for state exclusion.
+ */
+ if (i >= 0 && i < TcpNstates) {
+ if (TcpStX[i]) {
+ Lf->sf |= SELEXCLF;
+ return;
+ }
+ }
+ }
+ if (TcpStIn) {
+
+ /*
+ * Check for state inclusion.
+ */
+ if (i >= 0 && i < TcpNstates) {
+ if (TcpStI[i])
+ TcpStI[i] = 2;
+ else {
+ Lf->sf |= SELEXCLF;
+ return;
+ }
+ }
+ }
+ if (Fnet && (FnetTy != 6))
+ Lf->sf |= SELNET;
+
+#if defined(HASIPv6)
+ (void) snpf(Lf->type, sizeof(Lf->type), "IPv4");
+#else /* !defined(HASIPv6) */
+ (void) snpf(Lf->type, sizeof(Lf->type), "inet");
+#endif /* defined(HASIPv6) */
+
+ (void) snpf(Lf->iproto, sizeof(Lf->iproto), "%.*s", IPROTOL-1, pr);
+ Lf->inp_ty = 2;
+ if (ss & SB_INO) {
+ (void) snpf(dev_ch, sizeof(dev_ch), InodeFmt_d,
+ (INODETYPE)s->st_ino);
+ enter_dev_ch(dev_ch);
+ }
+ if (tp->faddr || tp->fport) {
+ fs.s_addr = tp->faddr;
+ fa = (unsigned char *)&fs;
+ } else
+ fa = (unsigned char *)NULL;
+ if (tp->laddr || tp->lport) {
+ ls.s_addr = tp->laddr;
+ la = (unsigned char *)&ls;
+ } else
+ la = (unsigned char *)NULL;
+ ent_inaddr(la, tp->lport, fa, tp->fport, AF_INET);
+ Lf->lts.type = tp->proto;
+ Lf->lts.state.i = tp->state;
+
+#if defined(HASTCPTPIQ)
+ Lf->lts.rq = tp->rxq;
+ Lf->lts.sq = tp->txq;
+ Lf->lts.rqs = Lf->lts.sqs = 1;
+#endif /* defined(HASTCPTPIQ) */
+
+ return;
+ }
+/*
+ * The socket's protocol can't be identified.
+ */
+ (void) snpf(Lf->type, sizeof(Lf->type), "sock");
+ if (ss & SB_INO) {
+ Lf->inode = (INODETYPE)s->st_ino;
+ Lf->inp_ty = 1;
+ }
+ if (ss & SB_DEV) {
+ Lf->dev = s->st_dev;
+ Lf->dev_def = 1;
+ }
+ enter_nm(Fxopt ? "can't identify protocol (-X specified)"
+ : "can't identify protocol");
+}
+
+
+/*
+ * set_net_paths() - set /proc/net paths
+ */
+
+void
+set_net_paths(p, pl)
+ char *p; /* path to /proc/net/ */
+ int pl; /* strlen(p) */
+{
+ int pathl;
+
+ pathl = 0;
+ (void) make_proc_path(p, pl, &AX25path, &pathl, "ax25");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &Ipxpath, &pathl, "ipx");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &Packpath, &pathl, "packet");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &Rawpath, &pathl, "raw");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &SockStatPath, &pathl, "sockstat");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &TCPpath, &pathl, "tcp");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &UDPpath, &pathl, "udp");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &UDPLITEpath, &pathl, "udplite");
+
+#if defined(HASIPv6)
+ pathl = 0;
+ (void) make_proc_path(p, pl, &Raw6path, &pathl, "raw6");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &SockStatPath6, &pathl, "sockstat6");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &TCP6path, &pathl, "tcp6");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &UDP6path, &pathl, "udp6");
+ pathl = 0;
+ (void) make_proc_path(p, pl, &UDP6LITEpath, &pathl, "udp6lite");
+#endif /* defined(HASIPv6) */
+
+ pathl = 0;
+ (void) make_proc_path(p, pl, &UNIXpath, &pathl, "unix");
+}
--- /dev/null
+/*
+ * dstore.c - Linux global storage for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: dstore.c,v 1.3 2008/04/15 13:32:26 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+int HasNFS = 0; /* NFS mount point status:
+ * 1 == there is an NFS mount point,
+ * but its device number is
+ * unknown
+ * 2 == there is an NFS mount point
+ * and its device number is
+ * known
+ */
+int OffType = 0; /* offset type:
+ * 0 == unknown
+ * 1 == lstat's st_size
+ * 2 == from /proc/<PID>/fdinfo */
+
+/*
+ * Pff_tab[] - table for printing file flags
+ */
+
+struct pff_tab Pff_tab[] = {
+ { (long)O_WRONLY, FF_WRITE },
+ { (long)O_RDWR, FF_RDWR },
+ { (long)O_CREAT, FF_CREAT },
+ { (long)O_EXCL, FF_EXCL },
+ { (long)O_NOCTTY, FF_NOCTTY },
+ { (long)O_TRUNC, FF_TRUNC },
+ { (long)O_APPEND, FF_APPEND },
+ { (long)O_NDELAY, FF_NDELAY },
+ { (long)O_SYNC, FF_SYNC },
+ { (long)O_ASYNC, FF_ASYNC },
+
+#if defined(O_DIRECT)
+ { (long)O_DIRECT, FF_DIRECT },
+#endif /* defined(O_DIRECT) */
+
+#if defined(O_DIRECTORY)
+ { (long)O_DIRECTORY, FF_DIRECTORY },
+#endif /* defined(O_DIRECTORY) */
+
+#if defined(O_NOFOLLOW)
+ { (long)O_NOFOLLOW, FF_NOFOLNK },
+#endif /* defined(O_NOFOLLOW) */
+
+#if defined(O_NOATIME)
+ { (long)O_NOATIME, FF_NOATM },
+#endif /* defined(O_NOATIME) */
+
+#if defined(O_DSYNC)
+ { (long)O_DSYNC, FF_DSYNC },
+#endif /* defined(O_DSYNC) */
+
+#if defined(O_RSYNC)
+ { (long)O_RSYNC, FF_RSYNC },
+#endif /* defined(O_RSYNC) */
+
+#if defined(O_LARGEFILE)
+ { (long)O_LARGEFILE, FF_LARGEFILE },
+#endif /* defined(O_LARGEFILE) */
+ { (long)0, NULL }
+};
+
+
+/*
+ * Pof_tab[] - table for print process open file flags
+ */
+
+struct pff_tab Pof_tab[] = {
+ { (long)0, NULL }
+};
--- /dev/null
+/*
+ * machine.h - Linux definitions for /proc-based lsof
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * $Id: machine.h,v 1.33 2008/10/21 16:17:21 abe Exp $
+ */
+
+
+#if !defined(LSOF_MACHINE_H)
+#define LSOF_MACHINE_H 1
+
+
+#include <sys/types.h>
+#include <sys/param.h>
+
+
+/*
+ * CAN_USE_CLNT_CREATE is defined for those dialects where RPC clnt_create()
+ * can be used to obtain a CLIENT handle in lieu of clnttcp_create().
+ */
+
+#define CAN_USE_CLNT_CREATE 1
+
+
+/*
+ * DEVDEV_PATH defines the path to the directory that contains device
+ * nodes.
+ */
+
+#define DEVDEV_PATH "/dev"
+
+
+/*
+ * GET_MAX_FD is defined for those dialects that provide a function other than
+ * getdtablesize() to obtain the maximum file descriptor number plus one.
+ */
+
+/* #define GET_MAX_FD ? */
+
+
+/*
+ * HASAOPT is defined for those dialects that have AFS support; it specifies
+ * that the default path to an alternate AFS kernel name list file may be
+ * supplied with the -A <path> option.
+ */
+
+/* #define HASAOPT 1 */
+
+
+/*
+ * HASBLKDEV is defined for those dialects that want block device information
+ * recorded in BDevtp[].
+ */
+
+/* #define HASBLKDEV 1 */
+
+
+/*
+ * HASDCACHE is defined for those dialects that support a device cache
+ * file.
+ *
+ * CAUTION!!! Do not enable HASDCACHE for /proc-based Linux lsof. The source
+ * code cannot support it.
+ *
+ * The presence of NEVER_HASDCACHE in this comment prevents the Customize
+ * script from offering to change HASDCACHE.
+ *
+ *
+ * HASENVDC defined the name of an environment variable that contains the
+ * device cache file path. The HASENVDC environment variable is ignored when
+ * the lsof process is setuid(root) or its real UID is 0.
+ *
+ * HASPERSDC defines the format for the last component of a personal device
+ * cache file path. The first will be the home directory of the real UID that
+ * executes lsof.
+ *
+ * HASPERSDCPATH defines the environment variable whose value is the middle
+ * component of the personal device cache file path. The middle component
+ * follows the home directory and precedes the results of applying HASPERSDC.
+ * The HASPERSDCPATH environment variable is ignored when the lsof process is
+ * setuid(root) or its real UID is 0.
+ *
+ * HASSYSDC defines a public device cache file path. When it's defined, it's
+ * used as the path from which to read the device cache.
+ *
+ * Consult the 00DCACHE and 00FAQ files of the lsof distribution for more
+ * information on device cache file path construction.
+ *
+ * CAUTION!!! Do not enable HASDCACHE for /proc-based Linux lsof. The source
+ * code cannot support it.
+ */
+
+/* #define HASDCACHE 1 !!!DON'T ENABLE!!! -- see above comment */
+/* #define HASENVDC "LSOFDEVCACHE" */
+/* #define HASPERSDC "%h/%p.lsof_%L" */
+/* #define HASPERSDCPATH "LSOFPERSDCPATH" */
+/* #define HASSYSDC "/your/choice/of/path" */
+
+
+/*
+ * HASCDRNODE is defined for those dialects that have CD-ROM nodes.
+ */
+
+/* #define HASCDRNODE 1 */
+
+
+/*
+ * HASFIFONODE is defined for those dialects that have FIFO nodes.
+ */
+
+/* #define HASFIFONODE 1 */
+
+
+/*
+ * HASFSINO is defined for those dialects that have the file system
+ * inode element, fs_ino, in the lfile structure definition in lsof.h.
+ */
+
+/* #define HASFSINO 1 */
+
+
+/*
+ * HASFSTRUCT is defined if the dialect has a file structure.
+ *
+ * FSV_DEFAULT defines the default set of file structure values to list.
+ * It defaults to zero (0), but may be made up of a combination of the
+ * FSV_* symbols from lsof.h.
+ *
+ * HASNOFSADDR -- has no file structure address
+ * HASNOFSFLAGS -- has no file structure flags
+ * HASNOFSCOUNT -- has no file structure count
+ * HASNOFSNADDR -- has no file structure node address
+ */
+
+#define HASFSTRUCT 1
+/* #define FSV_DEFAULT FSV_? | FSV_? | FSV_? */
+#define HASNOFSADDR 1 /* has no file structure address */
+/* #define HASNOFSFLAGS 1 has no file structure flags */
+#define HASNOFSCOUNT 1 /* has no file structure count */
+#define HASNOFSNADDR 1 /* has no file structure node address */
+
+
+/*
+ * HASGNODE is defined for those dialects that have gnodes.
+ */
+
+/* #define HASGNODE 1 */
+
+
+/*
+ * HASHSNODE is defined for those dialects that have High Sierra nodes.
+ */
+
+/* #define HASHSNODE 1 */
+
+
+/*
+ * HASINODE is defined for those dialects that have inodes and wish to
+ * use readinode() from node.c.
+ */
+
+/* #define HASINODE 1 */
+
+
+/*
+ * HASINTSIGNAL is defined for those dialects whose signal function returns
+ * an int.
+ */
+
+/* #define HASINTSIGNAL 1 */
+
+
+/*
+ * HASKERNIDCK is defined for those dialects that support the comparison of
+ * the build to running kernel identity.
+ */
+
+/* #define HASKERNIDCK 1 */
+
+
+/*
+ * HASKOPT is defined for those dialects that support the -k option of
+ * reading the kernel's name list from an optional file.
+ */
+
+/* #define HASKOPT 1 */
+
+
+/*
+ * HASLFILEADD is defined for those dialects that need additional elements
+ * in struct lfile. The HASLFILEADD definition is a macro that defines
+ * them. If any of the additional elements need to be preset in the
+ * alloc_lfile() function of proc.c, the SETLFILEADD macro may be defined
+ * to do that.
+ *
+ * If any additional elements need to be cleared in alloc_lfile() or in the
+ * free_proc() function of proc.c, the CLRLFILEADD macro may be defined to
+ * do that. Note that CLRLFILEADD takes one argument, the pointer to the
+ * lfile struct. The CLRLFILEADD macro is expected to expand to statements
+ * that are complete -- i.e., have terminating semi-colons -- so the macro is
+ * called without a terminating semicolon by proc.c.
+ *
+ * The HASXOPT definition may be used to select the conditions under which
+ * private lfile elements are used.
+ */
+
+/* #define HASLFILEADD int ... */
+/* #define CLRLFILEADD(lf) (lf)->... = (type)NULL; */
+/* #define SETLFILEADD Lf->... */
+
+
+/*
+ * HASLWP is defined for dialects that have LWP support inside processes.
+ */
+
+#define HASLWP 1
+
+
+/*
+ * HASMNTSTAT indicates the dialect supports the mount stat(2) result option
+ * in its l_vfs and mounts structures.
+ */
+
+/* #define HASMNTSTAT 1 */
+
+
+/*
+ * HASMNTSUP is defined for those dialects that support the mount supplement
+ * option.
+ */
+
+#define HASMNTSUP 1
+
+
+/*
+ * HASMOPT is defined for those dialects that support the reading of
+ * kernel memory from an alternate file.
+ */
+
+/* #define HASMOPT 1 */
+
+
+/*
+ * HASNCACHE is defined for those dialects that have a kernel name cache
+ * that lsof can search. A value of 1 directs printname() to prefix the
+ * cache value with the file system directory name; 2, avoid the prefix.
+ *
+ * NCACHELDPFX is a set of C commands to execute before calling ncache_load().
+ *
+ * NCACHELDSFX is a set of C commands to execute after calling ncache_load().
+ */
+
+/* #define HASNCACHE 1 */
+/* #define NCACHELDPFX ??? */
+/* #define NCACHELDSFX ??? */
+
+
+/*
+ * HASNLIST is defined for those dialects that use nlist() to acccess
+ * kernel symbols.
+ */
+
+/* #define HASNLIST 1 */
+
+
+/*
+ * HASPIPEFN is defined for those dialects that have a special function to
+ * process DTYPE_PIPE file structure entries. Its value is the name of the
+ * function.
+ *
+ * NOTE: don't forget to define a prototype for this function in dproto.h.
+ */
+
+/* #define HASPIPEFN process_pipe? */
+
+
+/*
+ * HASPIPENODE is defined for those dialects that have pipe nodes.
+ */
+
+/* #define HASPIPENODE 1 */
+
+
+/*
+ * HASPMAPENABLED is defined when the reporting of portmapper registration
+ * info is enabled by default.
+ */
+
+/* #define HASPMAPENABLED 1 */
+
+
+/*
+ * HASPPID is defined for those dialects that support identification of
+ * the parent process IDentifier (PPID) of a process.
+ */
+
+#define HASPPID 1
+
+
+/*
+ * HASPRINTDEV, HASPRINTINO, HASPRINTNM, HASPRINTOFF, and HASPRINTSZ
+ * define private dialect-specific functions for printing DEVice numbers,
+ * INOde numbers, NaMes, file OFFsets, and file SiZes. The functions are
+ * called from print_file().
+ */
+
+/* #define HASPRINTDEV print_dev? */
+/* #define HASPRINTINO print_ino? */
+/* #define HASPRINTNM print_nm? */
+/* #define HASPRINTOFF print_off? */
+/* #define HASPRINTSZ print_sz? */
+
+
+/*
+ * HASPRIVFILETYPE and PRIVFILETYPE are defined for dialects that have a
+ * file structure type that isn't defined by a DTYPE_* symbol. They are
+ * used in lib/prfp.c to select the type's processing.
+ *
+ * PRIVFILETYPE is the definition of the f_type value in the file struct.
+ *
+ * HASPRIVFILETYPE is the name of the processing function.
+ */
+
+/* #define HASPRIVFILETYPE process_shmf? */
+/* #define PRIVFILETYPE ?? */
+
+
+/*
+ * HASPRIVNMCACHE is defined for dialects that have a private method for
+ * printing cached NAME column values for some files. HASPRIVNAMECACHE
+ * is defined to be the name of the function.
+ *
+ * The function takes one argument, a struct lfile pointer to the file, and
+ * returns non-zero if it prints a name to stdout.
+ */
+
+/* #define HASPRIVNMCACHE <function name> */
+
+
+/*
+ * HASPRIVPRIPP is defined for dialects that have a private function for
+ * printing IP protocol names. When HASPRIVPRIPP isn't defined, the
+ * IP protocol name printing function defaults to printiprto().
+ */
+
+/* #define HASPRIVPRIPP 1 */
+
+
+/*
+ * HASPROCFS is defined for those dialects that have a proc file system --
+ * usually /proc and usually in SYSV4 derivatives.
+ *
+ * HASFSTYPE is defined as 1 for those systems that have a file system type
+ * string, st_fstype, in the stat() buffer; 2, for those systems that have a
+ * file system type integer in the stat() buffer, named MOUNTS_STAT_FSTYPE;
+ * 0, for systems whose stat(2) structure has no file system type member. The
+ * additional symbols MOUNTS_FSTYPE, RMNT_FSTYPE, and RMNT_STAT_FSTYPE may be
+ * defined in dlsof.h to direct how the readmnt() function in lib/rmnt.c
+ * preserves these stat(2) and getmntent(3) buffer values in the local mounts
+ * structure.
+ *
+ * The defined value is the string that names the file system type.
+ *
+ * The HASPROCFS definition usually must be accompanied by the HASFSTYPE
+ * definition and the providing of an fstype element in the local mounts
+ * structure (defined in dlsof.h).
+ *
+ * The HASPROCFS definition may be accompanied by the HASPINODEN definition.
+ * HASPINODEN specifies that searching for files in HASPROCFS is to be done
+ * by inode number.
+ */
+
+/* #define HASPROCFS "proc?" */
+/* #define HASFSTYPE 1 */
+/* #define HASPINODEN 1 */
+
+
+/*
+ * HASRNODE is defined for those dialects that have rnodes.
+ */
+
+/* #define HASRNODE 1 */
+
+
+/*
+ * Define HASSECURITY to restrict the listing of all open files to the
+ * root user. When HASSECURITY is defined, the non-root user may list
+ * only files whose processes have the same user ID as the real user ID
+ * (the one that its user logged on with) of the lsof process.
+ */
+
+/* #define HASSECURITY 1 */
+
+
+/*
+ * If HASSECURITY is defined, define HASNOSOCKSECURITY to allow users
+ * restricted by HASSECURITY to list any open socket files, provide their
+ * listing is selected by the "-i" option.
+ */
+
+/* #define HASNOSOCKSECURITY 1 */
+
+
+/*
+ * HASSETLOCALE is defined for those dialects that have <locale.h> and
+ * setlocale().
+ *
+ * If the dialect also has wide character support for language locales,
+ * HASWIDECHAR activates lsof's wide character support and WIDECHARINCL
+ * defines the header file (if any) that must be #include'd to use the
+ * mblen() and mbtowc() functions.
+ */
+
+#define HASSETLOCALE 1
+#define HASWIDECHAR 1
+#define WIDECHARINCL <wctype.h>
+
+
+/*
+ * HASSNODE is defined for those dialects that have snodes.
+ */
+
+/* #define HASSNODE 1 */
+
+
+/*
+ * HASSOOPT, HASSOSTATE and HASTCPOPT define the availability of information
+ * on socket options (SO_* symbols), socket states (SS_* symbols) and TCP
+ * options.
+ */
+
+/* #define HASSOOPT 1 has socket option information */
+/* #define HASSOSTATE 1 has socket state information */
+/* #define HASTCPOPT 1 has TCP options or flags */
+
+
+/*
+ * Define HASSPECDEVD to be the name of a function that handles the results
+ * of a successful stat(2) of a file name argument.
+ *
+ * For example, HASSPECDEVD() for Darwin makes sure that st_dev is set to
+ * what stat("/dev") returns -- i.e., what's in DevDev.
+ *
+ * The function takes two arguments:
+ *
+ * 1: pointer to the full path name of file
+ * 2: pointer to the stat(2) result
+ *
+ * The function returns void.
+ */
+
+/* #define HASSPECDEVD process_dev_stat */
+
+
+/*
+ * HASSTREAMS is defined for those dialects that support streams.
+ */
+
+/* #define HASSTREAMS 1 */
+
+
+/*
+ * HASTCPTPIQ is defined for dialects where it is possible to report the
+ * TCP/TPI Recv-Q and Send-Q values produced by netstat.
+ */
+
+#define HASTCPTPIQ 1
+
+
+/*
+ * HASTCPTPIW is defined for dialects where it is possible to report the
+ * TCP/TPI send and receive window sizes produced by netstat.
+ */
+
+/* #define HASTCPTPIW 1 */
+
+
+/*
+ * HASTCPUDPSTATE is defined for dialects that have TCP and UDP state
+ * support -- i.e., for the "-stcp|udp:state" option and its associated
+ * speed improvements.
+ */
+
+#define HASTCPUDPSTATE 1
+
+
+/*
+ * HASTMPNODE is defined for those dialects that have tmpnodes.
+ */
+
+/* #define HASTMPNODE 1 */
+
+
+/*
+ * HASVNODE is defined for those dialects that use the Sun virtual file system
+ * node, the vnode. BSD derivatives usually do; System V derivatives prior to
+ * R4 usually don't.
+ * doesn't.
+ */
+
+/* #define HASVNODE 1 */
+
+
+/*
+ * HASXOPT is defined for those dialects that have an X option. It
+ * defines the text for the usage display. HASXOPT_VALUE defines the
+ * option's default binary value -- 0 or 1.
+ */
+
+#define HASXOPT "skip TCP&UDP* files"
+#define HASXOPT_VALUE 0
+
+
+/*
+ * INODETYPE and INODEPSPEC define the internal node number type and its
+ * printf specification modifier. These need not be defined and lsof.h
+ * can be allowed to define defaults.
+ *
+ * These are defined here, because they must be used in dlsof.h.
+ */
+
+#define INODETYPE unsigned long long
+ /* inode number internal storage type */
+#define INODEPSPEC "ll" /* INODETYPE printf specification
+ * modifier */
+
+
+/*
+ * UID_ARG defines the size of a User ID number when it is passed
+ * as a function argument.
+ */
+
+#define UID_ARG u_int
+
+
+/*
+ * Each USE_LIB_<function_name> is defined for dialects that use the
+ * <function_name> in the lsof library.
+ *
+ * Note: other definitions and operations may be required to condition the
+ * library function source code. They may be found in the dialect dlsof.h
+ * header files.
+ */
+
+/* #define USE_LIB_CKKV 1 ckkv.c */
+/* #define USE_LIB_COMPLETEVFS 1 cvfs.c */
+/* #define USE_LIB_FIND_CH_INO 1 fino.c */
+#define USE_LIB_IS_FILE_NAMED 1 /* isfn.c */
+/* #define USE_LIB_LKUPDEV 1 lkud.c */
+/* #define USE_LIB_PRINTDEVNAME 1 pdvn.c */
+/* #define USE_LIB_PROCESS_FILE 1 prfp.c */
+/* #define USE_LIB_PRINT_TCPTPI 1 ptti.c */
+/* #define USE_LIB_READDEV 1 rdev.c */
+/* #define USE_LIB_READMNT 1 rmnt.c */
+/* #define USE_LIB_REGEX 1 regex.c */
+/* #define USE_LIB_RNAM 1 rnam.c */
+/* #define USE_LIB_RNCH 1 rnch.c */
+/* #define USE_LIB_RNMH 1 rnmh.c */
+/* #define USE_LIB_SNPF 1 snpf.c */
+#define snpf snprintf /* use the system's snprintf() */
+
+
+/*
+ * WARNDEVACCESS is defined for those dialects that should issue a warning
+ * when lsof can't access /dev (or /device) or one of its sub-directories.
+ * The warning can be inhibited by the lsof caller with the -w option.
+ *
+ * CAUTION!!! Don't enable the WARNDEVACCESS definiton for /proc-based Linux
+ * lsof; it doesn't process /dev at all.
+ *
+ * The presence of NEVER_WARNDEVACCESS in this comment prevents the Customize
+ * script from offering to change WARNDEVACCESS.
+ */
+
+/* #define WARNDEVACCESS 1 DON'T ENABLE!!! -- see above comment */
+
+
+/*
+ * WARNINGSTATE is defined for those dialects that want to suppress all lsof
+ * warning messages.
+ */
+
+/* #define WARNINGSTATE 1 warnings are enabled by default */
+
+
+/*
+ * WILLDROPGID is defined for those dialects whose lsof executable runs
+ * setgid(not_real_GID) and whose setgid power can be relinquished after
+ * the dialect's initialize() function has been executed.
+ */
+
+/* #define WILLDROPGID 1 */
+
+
+/*
+ * zeromem is a macro that uses bzero or memset.
+ */
+
+#define zeromem(a, l) bzero(a, l)
+
+#endif /* !defined(LSOF_MACHINE_H) */
--- /dev/null
+# Lsof library Makefile skeleton
+#
+# This skeleton is added to definitions established by Configure.
+#
+# $Id: Makefile.skel,v 1.13 2001/02/13 02:12:16 abe Exp $
+
+LIB= liblsof.a
+
+CDEF= ${RC_CFLAGS}
+CDEFS= ${CDEF} ${CFGF}
+INCL= ${DINC}
+
+HDR= ../lsof.h ../proto.h ../dlsof.h ../dproto.h ../machine.h
+
+SRC= ckkv.c cvfs.c dvch.c fino.c isfn.c lkud.c pdvn.c prfp.c \
+ ptti.c rdev.c regex.c rmnt.c rnam.c rnch.c rnmh.c snpf.c
+
+OBJ= ckkv.o cvfs.o dvch.o fino.o isfn.o lkud.o pdvn.o prfp.o \
+ ptti.o rdev.o regex.o rmnt.o rnam.o rnch.o rnmh.o snpf.o
+
+all: ${LIB}
+
+${LIB}: ${OBJ}
+ ${AR}
+ ${RANLIB}
+
+clean: FRC
+ rm -f ${LIB} ${OBJ} errs Makefile.bak a.out core
+
+FRC:
+
+ckkv.o: ${HDR} ckkv.c
+
+cvfs.o: ${HDR} cvfs.c
+
+dvch.o: ${HDR} dvch.c
+
+fino.o: ${HDR} fino.c
+
+isfn.o: ${HDR} isfn.c
+
+lkud.o: ${HDR} lkud.c
+
+pdvn.o: ${HDR} pdvn.c
+
+prfp.o: ${HDR} prfp.c
+
+ptti.o: ${HDR} ptti.c
+
+rdev.o: ${HDR} rdev.c
+
+regex.o: ${HDR} ../regex.h regex.c
+
+rmnt.o: ${HDR} rmnt.c
+
+rnam.o: ${HDR} rnam.c
+
+rnch.o: ${HDR} rnch.c
+
+rnmh.o: ${HDR} rnmh.c
+
+snpf.o: ${HDR} snpf.c
--- /dev/null
+/*
+ * cvfs.c -- ckkv() function for lsof library
+ */
+
+
+/*
+ * Copyright 1998 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_CKKV)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1998 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: ckkv.c,v 1.3 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+#include <sys/utsname.h>
+
+
+/*
+ * ckkv() - check kernel version
+ */
+
+void
+ckkv(d, er, ev, ea)
+ char *d; /* dialect */
+ char *er; /* expected revision; NULL, no test */
+ char *ev; /* expected version; NULL, no test */
+ char *ea; /* expected architecture; NULL, no
+ * test */
+{
+
+# if defined(HASKERNIDCK)
+ struct utsname u;
+
+ if (Fwarn)
+ return;
+/*
+ * Read the system information via uname(2).
+ */
+ if (uname(&u) < 0) {
+ (void) fprintf(stderr, "%s: uname error: %s\n",
+ Pn, strerror(errno));
+ Exit(1);
+ }
+ if (er && strcmp(er, u.release)) {
+ (void) fprintf(stderr,
+ "%s: WARNING: compiled for %s release %s; this is %s.\n",
+ Pn, d, er, u.release);
+ }
+ if (ev && strcmp(ev, u.version)) {
+ (void) fprintf(stderr,
+ "%s: WARNING: compiled for %s version %s; this is %s.\n",
+ Pn, d, ev, u.version);
+ }
+ if (ea && strcmp(ea, u.machine)) {
+ (void) fprintf(stderr,
+ "%s: WARNING: compiled for %s architecture %s; this is %s.\n",
+ Pn, d, ea, u.machine);
+ }
+# endif /* defined(HASKERNIDCK) */
+
+}
+#else /* !defined(USE_LIB_CKKV) */
+char ckkv_d1[] = "d"; char *ckkv_d2 = ckkv_d1;
+#endif /* defined(USE_LIB_CKKV) */
--- /dev/null
+/*
+ * cvfs.c -- completevfs() function for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * The caller must define CVFS_DEVSAVE to have the device number moved
+ * from the mounts entry to the local vfs structure.
+ *
+ * The caller must define CVFS_NLKSAVE to have the link count moved from
+ * the mounts entry to the local vfs structure.
+ *
+ * The caller must define CVFS_SZSAVE to have the size moved from the
+ * mounts entry to the local vfs structure.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_COMPLETEVFS)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: cvfs.c,v 1.6 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+/*
+ * completevfs() - complete local vfs structure
+ */
+
+void
+completevfs(vfs, dev)
+ struct l_vfs *vfs; /* local vfs structure pointer */
+ dev_t *dev; /* device */
+{
+ struct mounts *mp;
+/*
+ * If only Internet socket files are selected, don't bother completing the
+ * local vfs structure.
+ */
+ if (Selinet)
+ return;
+/*
+ * Search for a match on device number.
+ */
+ for (mp = readmnt(); mp; mp = mp->next) {
+ if (mp->dev == *dev) {
+
+# if defined(CVFS_DEVSAVE)
+ vfs->dev = mp->dev;
+# endif /* defined(CVFS_DEVSAVE) */
+
+# if defined(CVFS_NLKSAVE)
+ vfs->nlink = mp->nlink;
+# endif /* defined(CVFS_NLKSAVE) */
+
+# if defined(CVFS_SZSAVE)
+ vfs->size = mp->size;
+# endif /* defined(CVFS_SZSAVE) */
+
+ vfs->dir = mp->dir;
+ vfs->fsname = mp->fsname;
+
+# if defined(HASFSINO)
+ vfs->fs_ino = mp->inode;
+# endif /* defined(HASFSINO) */
+
+# if defined(HASMNTSTAT)
+ vfs->mnt_stat = mp->stat;
+# endif /* defined(HASMNTSTAT) */
+
+
+ return;
+ }
+ }
+}
+#else /* !defined(USE_LIB_COMPLETEVFS) */
+char cvfs_d1[] = "d"; char *cvfs_d2 = cvfs_d1;
+#endif /* defined(USE_LIB_COMPLETEVFS) */
--- /dev/null
+/*
+ * dvch.c -- device cache functions for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(HASDCACHE)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: dvch.c,v 1.16 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+/*
+ * dvch.c - module that contains common device cache functions
+ *
+ * The caller may define the following:
+ *
+ * DCACHE_CLONE is the name of the function that reads and writes the
+ * clone section of the device cache file. The clone
+ * section follows the device section. If DCACHE_CLONE
+ * isn't defined, but HAS_STD_CLONE is defined to be 1,
+ * DCACHE_CLONE defaults to the local static function
+ * rw_clone_sect() that reads and writes a standard
+ * clone cache.
+ *
+ * DCACHE_CLR is the name of the function that clears the clone and
+ * pseudo caches when reading the device cache fails. If
+ * DCACHE_CLR isn't defined, but HAS_STD_CLONE is defined
+ * to be 1, DCACHE_CLR defaults to the local static
+ * function clr_sect() that clears a standard clone cache.
+ *
+ * DCACHE_PSEUDO is the name of the function that reads and writes
+ * the pseudo section of the device cache file. The
+ * pseudo section follows the device section and the
+ * clone section, if there is one.
+ *
+ * DVCH_CHOWN if the dialect has no fchown() function, so
+ * chown() must be used instead.
+ *
+ * DVCH_DEVPATH if the path to the device directory isn't "/dev".
+ *
+ * DVCH_EXPDEV if st_rdev must be expanded with the expdev()
+ * macro before use. (This is an EP/IX artifact.)
+ *
+ * HASBLKDEV if block device information is stored in BDevtp[].
+ */
+
+
+/*
+ * Local definitions
+ */
+
+# if !defined(DVCH_DEVPATH)
+#define DVCH_DEVPATH "/dev"
+# endif /* !defined(DVCH_DEVPATH) */
+
+/*
+ * Local storage
+ */
+
+static int crctbl[CRC_TBLL]; /* crc partial results table */
+
+
+/*
+ * Local function prototypes
+ */
+
+#undef DCACHE_CLR_LOCAL
+# if !defined(DCACHE_CLR)
+# if defined(HAS_STD_CLONE) && HAS_STD_CLONE==1
+#define DCACHE_CLR clr_sect
+#define DCACHE_CLR_LOCAL 1
+_PROTOTYPE(static void clr_sect,(void));
+# endif /* defined(HAS_STD_CLONE) && HAS_STD_CLONE==1 */
+# endif /* !defined(DCACHE_CLR) */
+
+#undef DCACHE_CLONE_LOCAL
+# if !defined(DCACHE_CLONE)
+# if defined(HAS_STD_CLONE) && HAS_STD_CLONE==1
+#define DCACHE_CLONE rw_clone_sect
+#define DCACHE_CLONE_LOCAL 1
+_PROTOTYPE(static int rw_clone_sect,(int m));
+# endif /* defined(HAS_STD_CLONE) && HAS_STD_CLONE==1 */
+# endif /*!defined(DCACHE_CLONE) */
+
+
+# if defined(HASBLKDEV)
+/*
+ * alloc_bdcache() - allocate block device cache
+ */
+
+void
+alloc_bdcache()
+{
+ if (!(BDevtp = (struct l_dev *)calloc((MALLOC_S)BNdev,
+ sizeof(struct l_dev))))
+ {
+ (void) fprintf(stderr, "%s: no space for block devices\n", Pn);
+ Exit(1);
+ }
+ if (!(BSdev = (struct l_dev **)malloc((MALLOC_S)(sizeof(struct l_dev *)
+ * BNdev))))
+ {
+ (void) fprintf(stderr, "%s: no space for block device pointers\n",
+ Pn);
+ Exit(1);
+ }
+}
+# endif /* defined(HASBLKDEV) */
+
+
+/*
+ * alloc_dcache() - allocate device cache
+ */
+
+void
+alloc_dcache()
+{
+ if (!(Devtp = (struct l_dev *)calloc((MALLOC_S)Ndev,
+ sizeof(struct l_dev))))
+ {
+ (void) fprintf(stderr, "%s: no space for devices\n", Pn);
+ Exit(1);
+ }
+ if (!(Sdev = (struct l_dev **)malloc((MALLOC_S)(sizeof(struct l_dev *)
+ * Ndev))))
+ {
+ (void) fprintf(stderr, "%s: no space for device pointers\n",
+ Pn);
+ Exit(1);
+ }
+}
+
+
+/*
+ * clr_devtab() - clear the device tables and free their space
+ */
+
+void
+clr_devtab()
+{
+ int i;
+
+ if (Devtp) {
+ for (i = 0; i < Ndev; i++) {
+ if (Devtp[i].name) {
+ (void) free((FREE_P *)Devtp[i].name);
+ Devtp[i].name = (char *)NULL;
+ }
+ }
+ (void) free((FREE_P *)Devtp);
+ Devtp = (struct l_dev *)NULL;
+ }
+ if (Sdev) {
+ (void) free((FREE_P *)Sdev);
+ Sdev = (struct l_dev **)NULL;
+ }
+ Ndev = 0;
+
+# if defined(HASBLKDEV)
+ if (BDevtp) {
+ for (i = 0; i < BNdev; i++) {
+ if (BDevtp[i].name) {
+ (void) free((FREE_P *)BDevtp[i].name);
+ BDevtp[i].name = (char *)NULL;
+ }
+ }
+ (void) free((FREE_P *)BDevtp);
+ BDevtp = (struct l_dev *)NULL;
+ }
+ if (BSdev) {
+ (void) free((FREE_P *)BSdev);
+ BSdev = (struct l_dev **)NULL;
+ }
+ BNdev = 0;
+# endif /* defined(HASBLKDEV) */
+
+}
+
+
+# if defined(DCACHE_CLR_LOCAL)
+/*
+ * clr_sect() - clear cached standard clone sections
+ */
+
+static void
+clr_sect()
+{
+ struct clone *c, *c1;
+
+ if (Clone) {
+ for (c = Clone; c; c = c1) {
+ c1 = c->next;
+ (void) free((FREE_P *)c);
+ }
+ Clone = (struct clone *)NULL;
+ }
+}
+# endif /* defined(DCACHE_CLR_LOCAL) */
+
+
+/*
+ * crc(b, l, s) - compute a crc for a block of bytes
+ */
+
+void
+crc(b, l, s)
+ char *b; /* block address */
+ int l; /* length */
+ unsigned *s; /* sum */
+{
+ char *cp; /* character pointer */
+ char *lm; /* character limit pointer */
+ unsigned sum; /* check sum */
+
+ cp = b;
+ lm = cp + l;
+ sum = *s;
+ do {
+ sum ^= ((int) *cp++) & 0xff;
+ sum = (sum >> 8) ^ crctbl[sum & 0xff];
+ } while (cp < lm);
+ *s = sum;
+}
+
+
+/*
+ * crcbld - build the CRC-16 partial results table
+ */
+
+void
+crcbld()
+{
+ int bit; /* temporary bit value */
+ unsigned entry; /* entry under construction */
+ int i; /* polynomial table index */
+ int j; /* bit shift count */
+
+ for(i = 0; i < CRC_TBLL; i++) {
+ entry = i;
+ for (j = 1; j <= CRC_BITS; j++) {
+ bit = entry & 1;
+ entry >>= 1;
+ if (bit)
+ entry ^= CRC_POLY;
+ }
+ crctbl[i] = entry;
+ }
+}
+
+
+/*
+ * dcpath() - define device cache file paths
+ */
+
+int
+dcpath(rw, npw)
+ int rw; /* read (1) or write (2) mode */
+ int npw; /* inhibit (0) or enable (1) no
+ * path warning message */
+{
+ char buf[MAXPATHLEN+1], *cp1, *cp2, hn[MAXPATHLEN+1];
+ int endf;
+ int i, j;
+ int l = 0;
+ int ierr = 0; /* intermediate error state */
+ int merr = 0; /* malloc error state */
+ struct passwd *p = (struct passwd *)NULL;
+ static short wenv = 1; /* HASENVDC warning state */
+ static short wpp = 1; /* HASPERSDCPATH warning state */
+/*
+ * Release any space reserved by previous path calls to dcpath().
+ */
+ if (DCpath[1]) {
+ (void) free((FREE_P *)DCpath[1]);
+ DCpath[1] = (char *)NULL;
+ }
+ if (DCpath[3]) {
+ (void) free((FREE_P *)DCpath[3]);
+ DCpath[3] = (char *)NULL;
+ }
+/*
+ * If a path was specified via -D, it's character address will have been
+ * stored in DCpathArg by ctrl_dcache(). Use that address if the real UID
+ * of this process is root, or the mode is read, or the process is neither
+ * setuid-root nor setgid.
+ */
+ if (Myuid == 0 || rw == 1 || (!Setuidroot && !Setgid))
+ DCpath[0] = DCpathArg;
+ else
+ DCpath[0] = (char *)NULL;
+
+# if defined(HASENVDC)
+/*
+ * If HASENVDC is defined, get its value from the environment, unless this
+ * is a setuid-root process, or the real UID of the process is 0, or the
+ * mode is write and the process is setgid.
+ */
+ if ((cp1 = getenv(HASENVDC)) && (l = strlen(cp1)) > 0
+ && !Setuidroot && Myuid && (rw == 1 || !Setgid)) {
+ if (!(cp2 = mkstrcpy(cp1, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for device cache path: %s=", Pn, HASENVDC);
+ safestrprt(cp1, stderr, 1);
+ merr = 1;
+ } else
+ DCpath[1] = cp2;
+ } else if (cp1 && l > 0) {
+ if (!Fwarn && wenv) {
+ (void) fprintf(stderr,
+ "%s: WARNING: ignoring environment: %s=", Pn, HASENVDC);
+ safestrprt(cp1, stderr, 1);
+ }
+ wenv = 0;
+ }
+# endif /* defined(HASENVDC) */
+
+# if defined(HASSYSDC)
+/*
+ * If HASSYSDC is defined, record the path of the system-wide device
+ * cache file, unless the mode is write.
+ */
+ if (rw != 2)
+ DCpath[2] = HASSYSDC;
+ else
+ DCpath[2] = (char *)NULL;
+# endif /* defined(HASSYSDC) */
+
+# if defined(HASPERSDC)
+/*
+ * If HASPERSDC is defined, form a personal device cache path by
+ * interpreting the conversions specified in it.
+ *
+ * Get (HASPERSDCPATH) from the environment and add it to the home directory
+ * path, if possible.
+ */
+ for (cp1 = HASPERSDC, endf = i = 0; *cp1 && !endf; cp1++) {
+ if (*cp1 != '%') {
+
+ /*
+ * If the format character isn't a `%', copy it.
+ */
+ if (i < (int)sizeof(buf)) {
+ buf[i++] = *cp1;
+ continue;
+ } else {
+ ierr = 2;
+ break;
+ }
+ }
+ /*
+ * `%' starts a conversion; the next character specifies
+ * the conversion type.
+ */
+ cp1++;
+ switch (*cp1) {
+
+ /*
+ * Two consecutive `%' characters convert to one `%'
+ * character in the output.
+ */
+
+ case '%':
+ if (i < (int)sizeof(buf))
+ buf[i++] = '%';
+ else
+ ierr = 2;
+ break;
+
+ /*
+ * ``%0'' defines a root boundary. If the effective
+ * (setuid-root) or real UID of the process is root, any
+ * path formed to this point is discarded and path formation
+ * begins with the next character.
+ *
+ * If neither the effective nor the real UID is root, path
+ * formation ends.
+ *
+ * This allows HASPERSDC to specify one path for non-root
+ * UIDs and another for the root (effective or real) UID.
+ */
+
+ case '0':
+ if (Setuidroot || !Myuid)
+ i = 0;
+ else
+ endf = 1;
+ break;
+
+ /*
+ * ``%h'' converts to the home directory.
+ */
+
+ case 'h':
+ if (!p && !(p = getpwuid(Myuid))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't get home dir for UID: %d\n",
+ Pn, (int)Myuid);
+ ierr = 1;
+ break;
+ }
+ if ((i + (l = strlen(p->pw_dir))) >= (int)sizeof(buf)) {
+ ierr = 2;
+ break;
+ }
+ (void) strcpy(&buf[i], p->pw_dir);
+ i += l;
+ if (i > 0 && buf[i - 1] == '/' && *(cp1 + 1)) {
+
+ /*
+ * If the home directory ends in a '/' and the next format
+ * character is a '/', delete the '/' at the end of the home
+ * directory.
+ */
+ i--;
+ buf[i] = '\0';
+ }
+ break;
+
+ /*
+ * ``%l'' converts to the full host name.
+ *
+ * ``%L'' converts to the first component (characters up
+ * to the first `.') of the host name.
+ */
+
+ case 'l':
+ case 'L':
+ if (gethostname(hn, sizeof(hn) - 1) < 0) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: no gethostname for %%l or %%L: %s\n",
+ Pn, strerror(errno));
+ ierr = 1;
+ break;
+ }
+ hn[sizeof(hn) - 1] = '\0';
+ if (*cp1 == 'L' && (cp2 = strchr(hn, '.')) && cp2 > hn)
+ *cp2 = '\0';
+ j = strlen(hn);
+ if ((j + i) < (int)sizeof(buf)) {
+ (void) strcpy(&buf[i], hn);
+ i += j;
+ } else
+ ierr = 2;
+ break;
+
+ /*
+ * ``%p'' converts to the contents of LSOFPERSDCPATH, followed
+ * by a '/'.
+ *
+ * It is ignored when:
+ *
+ * The lsof process is setuid-root;
+ * The real UID of the lsof process is 0;
+ * The mode is write and the process is setgid.
+ */
+
+ case 'p':
+
+# if defined(HASPERSDCPATH)
+ if ((cp2 = getenv(HASPERSDCPATH))
+ && (l = strlen(cp2)) > 0
+ && !Setuidroot
+ && Myuid
+ && (rw == 1 || !Setgid))
+ {
+ if (i && buf[i - 1] == '/' && *cp2 == '/') {
+ cp2++;
+ l--;
+ }
+ if ((i + l) < ((int)sizeof(buf) - 1)) {
+ (void) strcpy(&buf[i], cp2);
+ i += l;
+ if (buf[i - 1] != '/') {
+ if (i < ((int)sizeof(buf) - 2)) {
+ buf[i++] = '/';
+ buf[i] = '\0';
+ } else
+ ierr = 2;
+ }
+ } else
+ ierr = 2;
+ } else {
+ if (cp2 && l > 0) {
+ if (!Fwarn && wpp) {
+ (void) fprintf(stderr,
+ "%s: WARNING: ignoring environment: %s",
+ Pn, HASPERSDCPATH);
+ safestrprt(cp2, stderr, 1);
+ }
+ wpp = 0;
+ }
+ }
+# else /* !defined(HASPERSDCPATH) */
+ if (!Fwarn && wpp)
+ (void) fprintf(stderr,
+ "%s: WARNING: HASPERSDCPATH disabled: %s\n",
+ Pn, HASPERSDC);
+ ierr = 1;
+ wpp = 0;
+# endif /* defined(HASPERSDCPATH) */
+
+ break;
+
+ /*
+ * ``%u'' converts to the login name of the real UID of the
+ * lsof process.
+ */
+
+ case 'u':
+ if (!p && !(p = getpwuid(Myuid))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't get login name for UID: %d\n",
+ Pn, (int)Myuid);
+ ierr = 1;
+ break;
+ }
+ if ((i + (l = strlen(p->pw_name))) >= (int)sizeof(buf)) {
+ ierr = 2;
+ break;
+ }
+ (void) strcpy(&buf[i], p->pw_name);
+ i += l;
+ break;
+
+ /*
+ * ``%U'' converts to the real UID of the lsof process.
+ */
+
+ case 'U':
+ (void) snpf(hn, sizeof(hn), "%d", (int)Myuid);
+ if ((i + (l = strlen(hn))) >= (int)sizeof(buf))
+ ierr = 2;
+ else {
+ (void) strcpy(&buf[i], hn);
+ i += l;
+ }
+ break;
+ default:
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: bad conversion (%%%c): %s\n",
+ Pn, *cp1, HASPERSDC);
+ ierr = 1;
+ }
+ if (endf || ierr > 1)
+ break;
+ }
+ if (ierr) {
+
+ /*
+ * If there was an intermediate error of some type, handle it.
+ * A type 1 intermediate error has already been noted with a
+ * warning message. A type 2 intermediate error requires the
+ * issuing of a buffer overlow warning message.
+ */
+ if (ierr == 2 && !Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: device cache path too large: %s\n",
+ Pn, HASPERSDC);
+ i = 0;
+ }
+ buf[i] = '\0';
+/*
+ * If there is one, allocate space for the personal device cache path,
+ * copy buf[] to it, and store its pointer in DCpath[3].
+ */
+ if (i) {
+ if (!(cp1 = mkstrcpy(buf, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for device cache path: ", Pn);
+ safestrprt(buf, stderr, 1);
+ merr = 1;
+ } else
+ DCpath[3] = cp1;
+ }
+# endif /* defined(HASPERSDC) */
+
+/*
+ * Quit if there was a malloc() error. The appropriate error message
+ * will have been issued to stderr.
+ */
+ if (merr)
+ Exit(1);
+/*
+ * Return the index of the first defined path. Since DCpath[] is arranged
+ * in priority order, searching it beginning to end follows priority.
+ * Return an error indication if the search discloses no path name.
+ */
+ for (i = 0; i < MAXDCPATH; i++) {
+ if (DCpath[i])
+ return(i);
+ }
+ if (!Fwarn && npw)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't form any device cache path\n", Pn);
+ return(-1);
+}
+
+
+/*
+ * open_dcache() - open device cache file
+ */
+
+int
+open_dcache(m, r, s)
+ int m; /* mode: 1 = read; 2 = write */
+ int r; /* create DCpath[] if 0, reuse if 1 */
+ struct stat *s; /* stat() receiver */
+{
+ char buf[128];
+ char *w = (char *)NULL;
+/*
+ * Get the device cache file paths.
+ */
+ if (!r) {
+ if ((DCpathX = dcpath(m, 1)) < 0)
+ return(1);
+ }
+/*
+ * Switch to the requested open() action.
+ */
+ switch (m) {
+ case 1:
+
+ /*
+ * Check for access permission.
+ */
+ if (!is_readable(DCpath[DCpathX], 0)) {
+ if (DCpathX == 2 && errno == ENOENT)
+ return(2);
+ if (!Fwarn)
+ (void) fprintf(stderr, ACCESSERRFMT,
+ Pn, DCpath[DCpathX], strerror(errno));
+ return(1);
+ }
+ /*
+ * Open for reading.
+ */
+ if ((DCfd = open(DCpath[DCpathX], O_RDONLY, 0)) < 0) {
+ if (DCstate == 3 && errno == ENOENT)
+ return(1);
+
+cant_open:
+ (void) fprintf(stderr,
+ "%s: WARNING: can't open %s: %s\n",
+ Pn, DCpath[DCpathX], strerror(errno));
+ return(1);
+ }
+ if (stat(DCpath[DCpathX], s) != 0) {
+
+cant_stat:
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't stat(%s): %s\n",
+ Pn, DCpath[DCpathX], strerror(errno));
+close_exit:
+ (void) close(DCfd);
+ DCfd = -1;
+ return(1);
+ }
+ if ((int)(s->st_mode & 07777) != ((DCpathX == 2) ? 0644 : 0600)) {
+ (void) snpf(buf, sizeof(buf), "doesn't have %04o modes",
+ (DCpathX == 2) ? 0644 : 0600);
+ w = buf;
+ } else if ((s->st_mode & S_IFMT) != S_IFREG)
+ w = "isn't a regular file";
+ else if (!s->st_size)
+ w = "is empty";
+ if (w) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: %s %s.\n", Pn, DCpath[DCpathX], w);
+ goto close_exit;
+ }
+ return(0);
+ case 2:
+
+ /*
+ * Open for writing: first unlink any previous version; then
+ * open exclusively, specifying it's an error if the file exists.
+ */
+ if (unlink(DCpath[DCpathX]) < 0) {
+ if (errno != ENOENT) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't unlink %s: %s\n",
+ Pn, DCpath[DCpathX], strerror(errno));
+ return(1);
+ }
+ }
+ if ((DCfd = open(DCpath[DCpathX], O_RDWR|O_CREAT|O_EXCL, 0600)) < 0)
+ goto cant_open;
+ /*
+ * If the real user is not root, but the process is setuid-root,
+ * change the ownerships of the file to the real ones.
+ */
+ if (Myuid && Setuidroot) {
+
+# if defined(DVCH_CHOWN)
+ if (chown(DCpath[DCpathX], Myuid, getgid()) < 0)
+# else /* !defined(DVCH_CHOWN) */
+ if (fchown(DCfd, Myuid, getgid()) < 0)
+# endif /* defined(DVCH_CHOWN) */
+
+ {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't change ownerships of %s: %s\n",
+ Pn, DCpath[DCpathX], strerror(errno));
+ }
+ }
+ if (!Fwarn && DCstate != 1 && !DCunsafe)
+ (void) fprintf(stderr,
+ "%s: WARNING: created device cache file: %s\n",
+ Pn, DCpath[DCpathX]);
+ if (stat(DCpath[DCpathX], s) != 0) {
+ (void) unlink(DCpath[DCpathX]);
+ goto cant_stat;
+ }
+ return(0);
+ default:
+
+ /*
+ * Oops!
+ */
+ (void) fprintf(stderr, "%s: internal error: open_dcache=%d\n",
+ Pn, m);
+ Exit(1);
+ }
+ return(1);
+}
+
+
+/*
+ * read_dcache() - read device cache file
+ */
+
+int
+read_dcache()
+{
+ char buf[MAXPATHLEN*2], cbuf[64], *cp;
+ int i, len, ov;
+ struct stat sb, devsb;
+/*
+ * Open the device cache file.
+ *
+ * If the open at HASSYSDC fails because the file doesn't exist, and
+ * the real UID of this process is not zero, try to open a device cache
+ * file at HASPERSDC.
+ */
+ if ((ov = open_dcache(1, 0, &sb)) != 0) {
+ if (DCpathX == 2) {
+ if (ov == 2 && DCpath[3]) {
+ DCpathX = 3;
+ if (open_dcache(1, 1, &sb) != 0)
+ return(1);
+ } else
+ return(1);
+ } else
+ return(1);
+ }
+/*
+ * If the open device cache file's last mtime/ctime isn't greater than
+ * DVCH_DEVPATH's mtime/ctime, ignore it, unless -Dr was specified.
+ */
+ if (stat(DVCH_DEVPATH, &devsb) != 0) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't stat(%s): %s\n",
+ Pn, DVCH_DEVPATH, strerror(errno));
+ } else {
+ if (sb.st_mtime <= devsb.st_mtime || sb.st_ctime <= devsb.st_ctime)
+ DCunsafe = 1;
+ }
+ if (!(DCfs = fdopen(DCfd, "r"))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't fdopen(%s)\n", Pn, DCpath[DCpathX]);
+ (void) close(DCfd);
+ DCfd = -1;
+ return(1);
+ }
+/*
+ * Read the section count line; initialize the CRC table;
+ * validate the section count line.
+ */
+ if (!fgets(buf, sizeof(buf), DCfs)) {
+
+cant_read:
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't fread %s: %s\n", Pn, DCpath[DCpathX],
+ strerror(errno));
+read_close:
+ (void) fclose(DCfs);
+ DCfd = -1;
+ DCfs = (FILE *)NULL;
+ (void) clr_devtab();
+
+# if defined(DCACHE_CLR)
+ (void) DCACHE_CLR();
+# endif /* defined(DCACHE_CLR) */
+
+ return(1);
+ }
+ (void) crcbld();
+ DCcksum = 0;
+ (void) crc(buf, strlen(buf), &DCcksum);
+ i = 1;
+ cp = "";
+
+# if defined(HASBLKDEV)
+ i++;
+ cp = "s";
+# endif /* defined(HASBLKDEV) */
+
+# if defined(DCACHE_CLONE)
+ i++;
+ cp = "s";
+# endif /* defined(DCACHE_CLONE) */
+
+# if defined(DCACHE_PSEUDO)
+ i++;
+ cp = "s";
+# endif /* defined(DCACHE_PSEUDO) */
+
+ (void) snpf(cbuf, sizeof(cbuf), "%d section%s", i, cp);
+ len = strlen(cbuf);
+ (void) snpf(&cbuf[len], sizeof(cbuf) - len, ", dev=%lx\n",
+ (long)DevDev);
+ if (!strncmp(buf, cbuf, len) && (buf[len] == '\n')) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: no /dev device in %s: line ", Pn,
+ DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ if (strcmp(buf, cbuf)) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: bad section count line in %s: line ",
+ Pn, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+/*
+ * Read device section header and validate it.
+ */
+ if (!fgets(buf, sizeof(buf), DCfs))
+ goto cant_read;
+ (void) crc(buf, strlen(buf), &DCcksum);
+ len = strlen("device section: ");
+ if (strncmp(buf, "device section: ", len) != 0) {
+
+read_dhdr:
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: bad device section header in %s: line ",
+ Pn, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+/*
+ * Compute the device count; allocate Sdev[] and Devtp[] space.
+ */
+ if ((Ndev = atoi(&buf[len])) < 1)
+ goto read_dhdr;
+ alloc_dcache();
+/*
+ * Read the device lines and store their information in Devtp[].
+ * Construct the Sdev[] pointers to Devtp[].
+ */
+ for (i = 0; i < Ndev; i++) {
+ if (!fgets(buf, sizeof(buf), DCfs)) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read device %d from %s\n",
+ Pn, i + 1, DCpath[DCpathX]);
+ goto read_close;
+ }
+ (void) crc(buf, strlen(buf), &DCcksum);
+ /*
+ * Convert hexadecimal device number.
+ */
+ if (!(cp = x2dev(buf, &Devtp[i].rdev)) || *cp != ' ') {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: device %d: bad device in %s: line ",
+ Pn, i + 1, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ /*
+ * Convert inode number.
+ */
+ for (cp++, Devtp[i].inode = (INODETYPE)0; *cp != ' '; cp++) {
+ if (*cp < '0' || *cp > '9') {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: device %d: bad inode # in %s: line ",
+ Pn, i + 1, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ Devtp[i].inode = (INODETYPE)((Devtp[i].inode * 10)
+ + (int)(*cp - '0'));
+ }
+ /*
+ * Get path name; allocate space for it; copy it; store the
+ * pointer in Devtp[]; clear verify status; construct the Sdev[]
+ * pointer to Devtp[].
+ */
+ if ((len = strlen(++cp)) < 2 || *(cp + len - 1) != '\n') {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: device %d: bad path in %s: line ",
+ Pn, i + 1, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ *(cp + len - 1) = '\0';
+ if (!(Devtp[i].name = mkstrcpy(cp, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: device %d: no space for path: line ", Pn, i + 1);
+ safestrprt(buf, stderr, 1+4+8);
+ Exit(1);
+ }
+ Devtp[i].v = 0;
+ Sdev[i] = &Devtp[i];
+ }
+
+# if defined(HASBLKDEV)
+/*
+ * Read block device section header and validate it.
+ */
+ if (!fgets(buf, sizeof(buf), DCfs))
+ goto cant_read;
+ (void) crc(buf, strlen(buf), &DCcksum);
+ len = strlen("block device section: ");
+ if (strncmp(buf, "block device section: ", len) != 0) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: bad block device section header in %s: line ",
+ Pn, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+/*
+ * Compute the block device count; allocate BSdev[] and BDevtp[] space.
+ */
+ if ((BNdev = atoi(&buf[len])) > 0) {
+ alloc_bdcache();
+ /*
+ * Read the block device lines and store their information in BDevtp[].
+ * Construct the BSdev[] pointers to BDevtp[].
+ */
+ for (i = 0; i < BNdev; i++) {
+ if (!fgets(buf, sizeof(buf), DCfs)) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read block device %d from %s\n",
+ Pn, i + 1, DCpath[DCpathX]);
+ goto read_close;
+ }
+ (void) crc(buf, strlen(buf), &DCcksum);
+ /*
+ * Convert hexadecimal device number.
+ */
+ if (!(cp = x2dev(buf, &BDevtp[i].rdev)) || *cp != ' ') {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: block dev %d: bad device in %s: line ",
+ Pn, i + 1, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ /*
+ * Convert inode number.
+ */
+ for (cp++, BDevtp[i].inode = (INODETYPE)0; *cp != ' '; cp++) {
+ if (*cp < '0' || *cp > '9') {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: block dev %d: bad inode # in %s: line ",
+ Pn, i + 1, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ BDevtp[i].inode = (INODETYPE)((BDevtp[i].inode * 10)
+ + (int)(*cp - '0'));
+ }
+ /*
+ * Get path name; allocate space for it; copy it; store the
+ * pointer in BDevtp[]; clear verify status; construct the BSdev[]
+ * pointer to BDevtp[].
+ */
+ if ((len = strlen(++cp)) < 2 || *(cp + len - 1) != '\n') {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: block dev %d: bad path in %s: line",
+ Pn, i + 1, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ *(cp + len - 1) = '\0';
+ if (!(BDevtp[i].name = mkstrcpy(cp, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: block dev %d: no space for path: line", Pn, i + 1);
+ safestrprt(buf, stderr, 1+4+8);
+ Exit(1);
+ }
+ BDevtp[i].v = 0;
+ BSdev[i] = &BDevtp[i];
+ }
+ }
+# endif /* defined(HASBLKDEV) */
+
+# if defined(DCACHE_CLONE)
+/*
+ * Read the clone section.
+ */
+ if (DCACHE_CLONE(1))
+ goto read_close;
+# endif /* defined(DCACHE_CLONE) */
+
+# if defined(DCACHE_PSEUDO)
+/*
+ * Read the pseudo section.
+ */
+ if (DCACHE_PSEUDO(1))
+ goto read_close;
+# endif /* defined(DCACHE_PSEUDO) */
+
+/*
+ * Read and check the CRC section; it must be the last thing in the file.
+ */
+ (void) snpf(cbuf, sizeof(cbuf), "CRC section: %x\n", DCcksum);
+ if (!fgets(buf, sizeof(buf), DCfs) || strcmp(buf, cbuf) != 0) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: bad CRC section in %s: line ",
+ Pn, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+ if (fgets(buf, sizeof(buf), DCfs)) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: data follows CRC section in %s: line ",
+ Pn, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ goto read_close;
+ }
+/*
+ * Check one device entry at random -- the randomness based on our
+ * PID.
+ */
+ i = (int)(Mypid % Ndev);
+ if (stat(Devtp[i].name, &sb) != 0
+
+# if defined(DVCH_EXPDEV)
+ || expdev(sb.st_rdev) != Devtp[i].rdev
+# else /* !defined(DVCH_EXPDEV) */
+ || sb.st_rdev != Devtp[i].rdev
+# endif /* defined(DVCH_EXPDEV) */
+
+ || sb.st_ino != Devtp[i].inode) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: device cache mismatch: %s\n",
+ Pn, Devtp[i].name);
+ goto read_close;
+ }
+/*
+ * Close the device cache file and return OK.
+ */
+ (void) fclose(DCfs);
+ DCfd = -1;
+ DCfs = (FILE *)NULL;
+ return(0);
+}
+
+
+# if defined(DCACHE_CLONE_LOCAL)
+/*
+ * rw_clone_sect() - read/write the device cache file clone section
+ */
+
+static int
+rw_clone_sect(m)
+ int m; /* mode: 1 = read; 2 = write */
+{
+ char buf[MAXPATHLEN*2], *cp, *cp1;
+ struct clone *c;
+ struct l_dev *dp;
+ int i, j, len, n;
+
+ if (m == 1) {
+
+ /*
+ * Read the clone section header and validate it.
+ */
+ if (!fgets(buf, sizeof(buf), DCfs)) {
+
+bad_clone_sect:
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: bad clone section header in %s: line ",
+ Pn, DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ return(1);
+ }
+ (void) crc(buf, strlen(buf), &DCcksum);
+ len = strlen("clone section: ");
+ if (strncmp(buf, "clone section: ", len) != 0)
+ goto bad_clone_sect;
+ if ((n = atoi(&buf[len])) < 0)
+ goto bad_clone_sect;
+ /*
+ * Read the clone section lines and create the Clone list.
+ */
+ for (i = 0; i < n; i++) {
+ if (fgets(buf, sizeof(buf), DCfs) == NULL) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: no %d clone line in %s: line ", Pn, i + 1,
+ DCpath[DCpathX]);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ return(1);
+ }
+ (void) crc(buf, strlen(buf), &DCcksum);
+ /*
+ * Assemble Devtp[] index and make sure it's correct.
+ */
+ for (cp = buf, j = 0; *cp != ' '; cp++) {
+ if (*cp < '0' || *cp > '9') {
+
+bad_clone_index:
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: clone %d: bad cached device index: line ",
+ Pn, i + 1);
+ safestrprt(buf, stderr, 1+4+8);
+ }
+ return(1);
+ }
+ j = (j * 10) + (int)(*cp - '0');
+ }
+ if (j < 0 || j >= Ndev || (cp1 = strchr(++cp, '\n')) == NULL)
+ goto bad_clone_index;
+ if (strncmp(cp, Devtp[j].name, (cp1 - cp)) != 0)
+ goto bad_clone_index;
+ /*
+ * Allocate and complete a clone structure.
+ */
+ if (!(c = (struct clone *)malloc(sizeof(struct clone)))) {
+ (void) fprintf(stderr,
+ "%s: clone %d: no space for cached clone: line ", Pn,
+ i + 1);
+ safestrprt(buf, stderr, 1+4+8);
+ Exit(1);
+ }
+ c->dx = j;
+ c->next = Clone;
+ Clone = c;
+ }
+ return(0);
+ } else if (m == 2) {
+
+ /*
+ * Write the clone section header.
+ */
+ for (c = Clone, n = 0; c; c = c->next, n++)
+ ;
+ (void) snpf(buf, sizeof(buf), "clone section: %d\n", n);
+ if (wr2DCfd(buf, &DCcksum))
+ return(1);
+ /*
+ * Write the clone section lines.
+ */
+ for (c = Clone; c; c = c->next) {
+ for (dp = &Devtp[c->dx], j = 0; j < Ndev; j++) {
+ if (dp == Sdev[j])
+ break;
+ }
+ if (j >= Ndev) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: can't make index for clone: ", Pn);
+ safestrprt(dp->name, stderr, 1);
+ }
+ (void) unlink(DCpath[DCpathX]);
+ (void) close(DCfd);
+ DCfd = -1;
+ return(1);
+ }
+ (void) snpf(buf, sizeof(buf), "%d %s\n", j, dp->name);
+ if (wr2DCfd(buf, &DCcksum))
+ return(1);
+ }
+ return(0);
+ }
+/*
+ * A shouldn't-happen case: mode neither 1 nor 2.
+ */
+ (void) fprintf(stderr, "%s: internal rw_clone_sect error: %d\n",
+ Pn, m);
+ Exit(1);
+ return(1); /* This useless return(1) keeps some
+ * compilers happy. */
+}
+# endif /* defined(DCACHE_CLONE_LOCAL) */
+
+
+/*
+ * write_dcache() - write device cache file
+ */
+
+void
+write_dcache()
+{
+ char buf[MAXPATHLEN*2], *cp;
+ struct l_dev *dp;
+ int i;
+ struct stat sb;
+/*
+ * Open the cache file; set up the CRC table; write the section count.
+ */
+ if (open_dcache(2, 0, &sb))
+ return;
+ i = 1;
+ cp = "";
+
+# if defined(HASBLKDEV)
+ i++;
+ cp = "s";
+# endif /* defined(HASBLKDEV) */
+
+# if defined(DCACHE_CLONE)
+ i++;
+ cp = "s";
+# endif /* defined(DCACHE_CLONE) */
+
+# if defined(DCACHE_PSEUDO)
+ i++;
+ cp = "s";
+# endif /* defined(DCACHE_PSEUDO) */
+
+ (void) snpf(buf, sizeof(buf), "%d section%s, dev=%lx\n", i, cp,
+ (long)DevDev);
+ (void) crcbld();
+ DCcksum = 0;
+ if (wr2DCfd(buf, &DCcksum))
+ return;
+/*
+ * Write the device section from the contents of Sdev[] and Devtp[].
+ */
+ (void) snpf(buf, sizeof(buf), "device section: %d\n", Ndev);
+ if (wr2DCfd(buf, &DCcksum))
+ return;
+ for (i = 0; i < Ndev; i++) {
+ dp = Sdev[i];
+ (void) snpf(buf, sizeof(buf), "%lx %ld %s\n", (long)dp->rdev,
+ (long)dp->inode, dp->name);
+ if (wr2DCfd(buf, &DCcksum))
+ return;
+ }
+
+# if defined(HASBLKDEV)
+/*
+ * Write the block device section from the contents of BSdev[] and BDevtp[].
+ */
+ (void) snpf(buf, sizeof(buf), "block device section: %d\n", BNdev);
+ if (wr2DCfd(buf, &DCcksum))
+ return;
+ if (BNdev) {
+ for (i = 0; i < BNdev; i++) {
+ dp = BSdev[i];
+ (void) snpf(buf, sizeof(buf), "%lx %ld %s\n", (long)dp->rdev,
+ (long)dp->inode, dp->name);
+ if (wr2DCfd(buf, &DCcksum))
+ return;
+ }
+ }
+# endif /* defined(HASBLKDEV) */
+
+# if defined(DCACHE_CLONE)
+/*
+ * Write the clone section.
+ */
+ if (DCACHE_CLONE(2))
+ return;
+# endif /* defined(DCACHE_CLONE) */
+
+# if defined(DCACHE_PSEUDO)
+/*
+ * Write the pseudo section.
+ */
+ if (DCACHE_PSEUDO(2))
+ return;
+# endif /* defined(DCACHE_PSEUDO) */
+
+/*
+ * Write the CRC section and close the file.
+ */
+ (void) snpf(buf, sizeof(buf), "CRC section: %x\n", DCcksum);
+ if (wr2DCfd(buf, (unsigned *)NULL))
+ return;
+ if (close(DCfd) != 0) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't close %s: %s\n",
+ Pn, DCpath[DCpathX], strerror(errno));
+ (void) unlink(DCpath[DCpathX]);
+ DCfd = -1;
+ }
+ DCfd = -1;
+/*
+ * If the previous reading of the previous device cache file marked it
+ * "unsafe," drop that marking and record that the device cache file was
+ * rebuilt.
+ */
+ if (DCunsafe) {
+ DCunsafe = 0;
+ DCrebuilt = 1;
+ }
+}
+
+
+/*
+ * wr2DCfd() - write to the DCfd file descriptor
+ */
+
+int
+wr2DCfd(b, c)
+ char *b; /* buffer */
+ unsigned *c; /* checksum receiver */
+{
+ int bl, bw;
+
+ bl = strlen(b);
+ if (c)
+ (void) crc(b, bl, c);
+ while (bl > 0) {
+ if ((bw = write(DCfd, b, bl)) < 0) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't write to %s: %s\n",
+ Pn, DCpath[DCpathX], strerror(errno));
+ (void) unlink(DCpath[DCpathX]);
+ (void) close(DCfd);
+ DCfd = -1;
+ return(1);
+ }
+ b += bw;
+ bl -= bw;
+ }
+ return(0);
+}
+#else /* !defined(HASDCACHE) */
+char dvch_d1[] = "d"; char *dvch_d2 = dvch_d1;
+#endif /* defined(HASDCACHE) */
--- /dev/null
+/*
+ * fino.c -- find inode functions for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * fino.c -- find block (optional) and character device file inode numbers
+ *
+ * The caller must define:
+ *
+ * HASBLKDEV to activate the block device inode lookup
+ */
+
+
+#include "../machine.h"
+
+#if defined(HASBLKDEV) || defined(USE_LIB_FIND_CH_INO)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: fino.c,v 1.5 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+#else /* !defined(HASBLKDEV) && !defined(USE_LIB_FIND_CH_INO) */
+char fino_d1[] = "d"; char *fino_d2 = fino_d1;
+#endif /* defined(HASBLKDEV) || defined(USE_LIB_FIND_CH_INO) */
+
+
+#if defined(HASBLKDEV)
+/*
+ * find_bl_ino() - find the inode number for a block device file
+ */
+
+void
+find_bl_ino()
+{
+ dev_t ldev, tdev;
+ int low, hi, mid;
+
+ readdev(0);
+
+# if defined(HASDCACHE)
+find_bl_ino_again:
+# endif /* defined(HASDCACHE) */
+
+ low = mid = 0;
+ hi = BNdev - 1;
+ if (!Lf->dev_def || (Lf->dev != DevDev) || !Lf->rdev_def)
+ return;
+ ldev = Lf->rdev;
+ while (low <= hi) {
+ mid = (low + hi) / 2;
+ tdev = BSdev[mid]->rdev;
+ if (ldev < tdev)
+ hi = mid - 1;
+ else if (ldev > tdev)
+ low = mid + 1;
+ else {
+
+# if defined(HASDCACHE)
+ if (DCunsafe && !BSdev[mid]->v && !vfy_dev(BSdev[mid]))
+ goto find_bl_ino_again;
+# endif /* defined(HASDCACHE) */
+
+ Lf->inode = BSdev[mid]->inode;
+ if (Lf->inp_ty == 0)
+ Lf->inp_ty = 1;
+ return;
+ }
+ }
+}
+#endif /* defined(HASBLKDEV) */
+
+
+#if defined(USE_LIB_FIND_CH_INO)
+/*
+ * find_ch_ino() - find the inode number for a character device file
+ */
+
+void
+find_ch_ino()
+{
+ dev_t ldev, tdev;
+ int low, hi, mid;
+
+ readdev(0);
+
+# if defined(HASDCACHE)
+find_ch_ino_again:
+# endif /* defined(HASDCACHE) */
+
+ low = mid = 0;
+ hi = Ndev - 1;
+ if (!Lf->dev_def || (Lf->dev != DevDev) || !Lf->rdev_def)
+ return;
+ ldev = Lf->rdev;
+ while (low <= hi) {
+ mid = (low + hi) / 2;
+ tdev = Sdev[mid]->rdev;
+ if (ldev < tdev)
+ hi = mid - 1;
+ else if (ldev > tdev)
+ low = mid + 1;
+ else {
+
+# if defined(HASDCACHE)
+ if (DCunsafe && !Sdev[mid]->v && !vfy_dev(Sdev[mid]))
+ goto find_ch_ino_again;
+# endif /* defined(HASDCACHE) */
+
+ Lf->inode = Sdev[mid]->inode;
+ if (Lf->inp_ty == 0)
+ Lf->inp_ty = 1;
+ return;
+ }
+ }
+}
+#endif /* defined(USE_LIB_FIND_CH_INO) */
--- /dev/null
+/*
+ * isfn.c -- is_file_named() function for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * To use this source file:
+ *
+ * 1. Define USE_LIB_IS_FILE_NAMED.
+ *
+ * 2. If clone support is required:
+ *
+ * a. Define HAVECLONEMAJ to be the name of the variable that
+ * contains the status of the clone major device -- e.g.,
+ *
+ * #define HAVECLONEMAJ HaveCloneMaj
+ *
+ * b. Define CLONEMAJ to be the name of the constant or
+ * variable that defines the clone major device -- e.g.,
+ *
+ * #define CLONEMAJ CloneMaj
+ *
+ * c. Make sure that clone devices are identified by an lfile
+ * element is_stream value of 1.
+ *
+ * d. Accept clone searching by device number only.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_IS_FILE_NAMED)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: isfn.c,v 1.10 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+/*
+ * Local structures
+ */
+
+struct hsfile {
+ struct sfile *s; /* the Sfile table address */
+ struct hsfile *next; /* the next hash bucket entry */
+};
+
+/*
+ * Local static variables
+ */
+
+# if defined(HAVECLONEMAJ)
+static struct hsfile *HbyCd = /* hash by clone buckets */
+ (struct hsfile *)NULL;
+static int HbyCdCt = 0; /* HbyCd entry count */
+# endif /* defined(HAVECLONEMAJ) */
+
+static struct hsfile *HbyFdi = /* hash by file (dev,ino) buckets */
+ (struct hsfile *)NULL;
+static int HbyFdiCt = 0; /* HbyFdi entry count */
+static struct hsfile *HbyFrd = /* hash by file raw device buckets */
+ (struct hsfile *)NULL;
+static int HbyFrdCt = 0; /* HbyFrd entry count */
+static struct hsfile *HbyFsd = /* hash by file system buckets */
+ (struct hsfile *)NULL;
+static int HbyFsdCt = 0; /* HbyFsd entry count */
+static struct hsfile *HbyNm = /* hash by name buckets */
+ (struct hsfile *)NULL;
+static int HbyNmCt = 0; /* HbyNm entry count */
+
+
+/*
+ * Local definitions
+ */
+
+# if defined(HAVECLONEMAJ)
+#define SFCDHASH 1024 /* Sfile hash by clone device (power
+ * of 2!) */
+# endif /* defined(HAVECLONEMAJ) */
+
+#define SFDIHASH 4094 /* Sfile hash by (device,inode) number
+ * pair bucket count (power of 2!) */
+#define SFFSHASH 1024 /* Sfile hash by file system device
+ * number bucket count (power of 2!) */
+#define SFHASHDEVINO(maj, min, ino, mod) ((int)(((int)((((int)(maj+1))*((int)((min+1))))+ino)*31415)&(mod-1)))
+ /* hash for Sfile by major device,
+ * minor device, and inode, modulo mod
+ * (mod must be a power of 2) */
+#define SFRDHASH 1024 /* Sfile hash by raw device number
+ * bucket count (power of 2!) */
+#define SFHASHRDEVI(maj, min, rmaj, rmin, ino, mod) ((int)(((int)((((int)(maj+1))*((int)((min+1))))+((int)(rmaj+1)*(int)(rmin+1))+ino)*31415)&(mod-1)))
+ /* hash for Sfile by major device,
+ * minor device, major raw device,
+ * minor raw device, and inode, modulo
+ * mod (mod must be a power of 2) */
+#define SFNMHASH 4096 /* Sfile hash by name bucket count
+ * (must be a power of 2!) */
+
+
+
+/*
+ * hashSfile() - hash Sfile entries for use in is_file_named() searches
+ */
+
+void
+hashSfile()
+{
+ static int hs = 0;
+ int i;
+ int sfplm = 3;
+ struct sfile *s;
+ struct hsfile *sh, *sn;
+/*
+ * Do nothing if there are no file search arguments cached or if the
+ * hashes have already been constructed.
+ */
+ if (!Sfile || hs)
+ return;
+/*
+ * Allocate hash buckets by (device,inode), file system device, and file name.
+ */
+
+# if defined(HAVECLONEMAJ)
+ if (HAVECLONEMAJ) {
+ if (!(HbyCd = (struct hsfile *)calloc((MALLOC_S)SFCDHASH,
+ sizeof(struct hsfile))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate space for %d clone hash buckets\n",
+ Pn, SFCDHASH);
+ Exit(1);
+ }
+ sfplm++;
+ }
+# endif /* defined(HAVECLONEMAJ) */
+
+ if (!(HbyFdi = (struct hsfile *)calloc((MALLOC_S)SFDIHASH,
+ sizeof(struct hsfile))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate space for %d (dev,ino) hash buckets\n",
+ Pn, SFDIHASH);
+ Exit(1);
+ }
+ if (!(HbyFrd = (struct hsfile *)calloc((MALLOC_S)SFRDHASH,
+ sizeof(struct hsfile))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate space for %d rdev hash buckets\n",
+ Pn, SFRDHASH);
+ Exit(1);
+ }
+ if (!(HbyFsd = (struct hsfile *)calloc((MALLOC_S)SFFSHASH,
+ sizeof(struct hsfile))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate space for %d file sys hash buckets\n",
+ Pn, SFFSHASH);
+ Exit(1);
+ }
+ if (!(HbyNm = (struct hsfile *)calloc((MALLOC_S)SFNMHASH,
+ sizeof(struct hsfile))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate space for %d name hash buckets\n",
+ Pn, SFNMHASH);
+ Exit(1);
+ }
+ hs++;
+/*
+ * Scan the Sfile chain, building file, file system, raw device, and file
+ * name hash bucket chains.
+ */
+ for (s = Sfile; s; s = s->next) {
+ for (i = 0; i < sfplm; i++) {
+ if (i == 0) {
+ if (!s->aname)
+ continue;
+ sh = &HbyNm[hashbyname(s->aname, SFNMHASH)];
+ HbyNmCt++;
+ } else if (i == 1) {
+ if (s->type) {
+ sh = &HbyFdi[SFHASHDEVINO(GET_MAJ_DEV(s->dev),
+ GET_MIN_DEV(s->dev), s->i,
+ SFDIHASH)];
+ HbyFdiCt++;
+ } else {
+ sh = &HbyFsd[SFHASHDEVINO(GET_MAJ_DEV(s->dev),
+ GET_MIN_DEV(s->dev),
+ 0,
+ SFFSHASH)];
+ HbyFsdCt++;
+ }
+ } else if (i == 2) {
+ if ((s->mode == S_IFCHR) || (s->mode == S_IFBLK)) {
+ sh = &HbyFrd[SFHASHRDEVI(GET_MAJ_DEV(s->dev),
+ GET_MIN_DEV(s->dev),
+ GET_MAJ_DEV(s->rdev),
+ GET_MIN_DEV(s->rdev),
+ s->i,
+ SFRDHASH)];
+ HbyFrdCt++;
+ } else
+ continue;
+ }
+
+# if defined(HAVECLONEMAJ)
+ else {
+ if (!HAVECLONEMAJ || (GET_MAJ_DEV(s->rdev) != CLONEMAJ))
+ continue;
+ sh = &HbyCd[SFHASHDEVINO(0, GET_MIN_DEV(s->rdev), 0,
+ SFCDHASH)];
+ HbyCdCt++;
+ }
+# else /* ! defined(HAVECLONEMAJ) */
+ else
+ continue;
+# endif /* defined(HAVECLONEMAJ) */
+
+ if (!sh->s) {
+ sh->s = s;
+ sh->next = (struct hsfile *)NULL;
+ continue;
+ } else {
+ if (!(sn = (struct hsfile *)malloc(
+ (MALLOC_S)sizeof(struct hsfile))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate hsfile bucket for: %s\n",
+ Pn, s->aname);
+ Exit(1);
+ }
+ sn->s = s;
+ sn->next = sh->next;
+ sh->next = sn;
+ }
+ }
+ }
+}
+
+
+/*
+ * is_file_named() - is this file named?
+ */
+
+int
+is_file_named(p, cd)
+ char *p; /* path name; NULL = search by device
+ * and inode (from *Lf) */
+ int cd; /* character or block type file --
+ * VCHR or VBLK vnode, or S_IFCHR
+ * or S_IFBLK inode */
+{
+ char *ep;
+ int f = 0;
+ struct sfile *s = (struct sfile *)NULL;
+ struct hsfile *sh;
+ size_t sz;
+/*
+ * Check for a path name match, as requested.
+ */
+ if (p && HbyNmCt) {
+ for (sh = &HbyNm[hashbyname(p, SFNMHASH)]; sh; sh = sh->next) {
+ if ((s = sh->s) && strcmp(p, s->aname) == 0) {
+ f = 2;
+ break;
+ }
+ }
+ }
+
+# if defined(HAVECLONEMAJ)
+/*
+ * If this is a stream, check for a clone device match.
+ */
+ if (!f && HbyCdCt && Lf->is_stream && Lf->dev_def && Lf->rdev_def
+ && (Lf->dev == DevDev))
+ {
+ for (sh = &HbyCd[SFHASHDEVINO(0, GET_MAJ_DEV(Lf->rdev), 0,
+ SFCDHASH)];
+ sh;
+ sh = sh->next)
+ {
+ if ((s = sh->s) && (GET_MAJ_DEV(Lf->rdev)
+ == GET_MIN_DEV(s->rdev))) {
+ f = 3;
+ break;
+ }
+ }
+ }
+# endif /* defined(HAVECLONEMAJ) */
+
+/*
+ * Check for a regular file.
+ */
+ if (!f && HbyFdiCt && Lf->dev_def
+ && (Lf->inp_ty == 1 || Lf->inp_ty == 3))
+ {
+ for (sh = &HbyFdi[SFHASHDEVINO(GET_MAJ_DEV(Lf->dev),
+ GET_MIN_DEV(Lf->dev),
+ Lf->inode,
+ SFDIHASH)];
+ sh;
+ sh = sh->next)
+ {
+ if ((s = sh->s) && (Lf->dev == s->dev)
+ && (Lf->inode == s->i)) {
+ f = 1;
+ break;
+ }
+ }
+ }
+/*
+ * Check for a file system match.
+ */
+ if (!f && HbyFsdCt && Lf->dev_def) {
+ for (sh = &HbyFsd[SFHASHDEVINO(GET_MAJ_DEV(Lf->dev),
+ GET_MIN_DEV(Lf->dev), 0,
+ SFFSHASH)];
+ sh;
+ sh = sh->next)
+ {
+ if ((s = sh->s) && (s->dev == Lf->dev)) {
+ f = 1;
+ break;
+ }
+ }
+ }
+/*
+ * Check for a character or block device match.
+ */
+ if (!f && HbyFrdCt && cd
+ && Lf->dev_def && (Lf->dev == DevDev)
+ && Lf->rdev_def
+ && (Lf->inp_ty == 1 || Lf->inp_ty == 3))
+ {
+ for (sh = &HbyFrd[SFHASHRDEVI(GET_MAJ_DEV(Lf->dev),
+ GET_MIN_DEV(Lf->dev),
+ GET_MAJ_DEV(Lf->rdev),
+ GET_MIN_DEV(Lf->rdev),
+ Lf->inode, SFRDHASH)];
+ sh;
+ sh = sh->next)
+ {
+ if ((s = sh->s) && (s->dev == Lf->dev)
+ && (s->rdev == Lf->rdev) && (s->i == Lf->inode))
+ {
+ f = 1;
+ break;
+ }
+ }
+ }
+/*
+ * Convert the name if a match occurred.
+ */
+ switch (f) {
+ case 0:
+ return(0);
+ case 1:
+ if (s->type) {
+
+ /*
+ * If the search argument isn't a file system, propagate it
+ * to Namech[]; otherwise, let printname() compose the name.
+ */
+ (void) snpf(Namech, Namechl, "%s", s->name);
+ if (s->devnm) {
+ ep = endnm(&sz);
+ (void) snpf(ep, sz, " (%s)", s->devnm);
+ }
+ }
+ break;
+ case 2:
+ (void) strcpy(Namech, p);
+ break;
+
+# if defined(HAVECLONEMAJ)
+ /* case 3: do nothing for stream clone matches */
+# endif /* defined(HAVECLONEMAJ) */
+
+ }
+ if (s)
+ s->f = 1;
+ return(1);
+}
+#else /* !defined(USE_LIB_IS_FILE_NAMED) */
+char isfn_d1[] = "d"; char *isfn_d2 = isfn_d1;
+#endif /* defined(USE_LIB_IS_FILE_NAMED) */
--- /dev/null
+/*
+ * lkud.c -- device lookup functions for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * lkud.c -- lookup device
+ *
+ * The caller may define:
+ *
+ * HASBLKDEV to activate block device lookup
+ */
+
+
+#include "../machine.h"
+
+#if defined(HASBLKDEV) || defined(USE_LIB_LKUPDEV)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: lkud.c,v 1.7 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+#else /* !defined(HASBLKDEV) && !defined(USE_LIB_LKUPDEV) */
+char lkud_d1[] = "d"; char *lkud_d2 = lkud_d1;
+#endif /* defined(HASBLKDEV) || defined(USE_LIB_LKUPDEV) */
+
+
+
+#if defined(HASBLKDEV)
+/*
+ * lkupbdev() - look up a block device
+ */
+
+struct l_dev *
+lkupbdev(dev, rdev, i, r)
+ dev_t *dev; /* pointer to device number */
+ dev_t *rdev; /* pointer to raw device number */
+ int i; /* inode match status */
+ int r; /* if 1, rebuild the device cache with
+ * rereaddev() when no match is found
+ * and HASDCACHE is defined and
+ * DCunsafe is one */
+{
+ INODETYPE inode = (INODETYPE)0;
+ int low, hi, mid;
+ struct l_dev *dp;
+ int ty = 0;
+
+ if (*dev != DevDev)
+ return((struct l_dev *)NULL);
+ readdev(0);
+ if (i) {
+ inode = Lf->inode;
+ ty = Lf->inp_ty;
+ }
+/*
+ * Search block device table for match.
+ */
+
+# if defined(HASDCACHE)
+
+lkupbdev_again:
+
+# endif /* defined(HASDCACHE) */
+
+ low = mid = 0;
+ hi = BNdev - 1;
+ while (low <= hi) {
+ mid = (low + hi) / 2;
+ dp = BSdev[mid];
+ if (*rdev < dp->rdev)
+ hi = mid - 1;
+ else if (*rdev > dp->rdev)
+ low = mid + 1;
+ else {
+ if ((i == 0) || (ty != 1) || (inode == dp->inode)) {
+
+# if defined(HASDCACHE)
+ if (DCunsafe && !dp->v && !vfy_dev(dp))
+ goto lkupbdev_again;
+# endif /* defined(HASDCACHE) */
+
+ return(dp);
+ }
+ if (inode < dp->inode)
+ hi = mid - 1;
+ else
+ low = mid + 1;
+ }
+ }
+
+# if defined(HASDCACHE)
+ if (DCunsafe && r) {
+ (void) rereaddev();
+ goto lkupbdev_again;
+ }
+# endif /* defined(HASDCACHE) */
+
+ return((struct l_dev *)NULL);
+}
+#endif /* defined(HASBLKDEV) */
+
+
+#if defined(USE_LIB_LKUPDEV)
+/*
+ * lkupdev() - look up a character device
+ */
+
+struct l_dev *
+lkupdev(dev, rdev, i, r)
+ dev_t *dev; /* pointer to device number */
+ dev_t *rdev; /* pointer to raw device number */
+ int i; /* inode match status */
+ int r; /* if 1, rebuild the device cache with
+ * rereaddev() when no match is found
+ * and HASDCACHE is defined and
+ * DCunsafe is one */
+{
+ INODETYPE inode = (INODETYPE)0;
+ int low, hi, mid;
+ struct l_dev *dp;
+ int ty = 0;
+
+ if (*dev != DevDev)
+ return((struct l_dev *)NULL);
+ readdev(0);
+ if (i) {
+ inode = Lf->inode;
+ ty = Lf->inp_ty;
+ }
+/*
+ * Search device table for match.
+ */
+
+# if defined(HASDCACHE)
+
+lkupdev_again:
+
+# endif /* defined(HASDCACHE) */
+
+ low = mid = 0;
+ hi = Ndev - 1;
+ while (low <= hi) {
+ mid = (low + hi) / 2;
+ dp = Sdev[mid];
+ if (*rdev < dp->rdev)
+ hi = mid - 1;
+ else if (*rdev > dp->rdev)
+ low = mid + 1;
+ else {
+ if ((i == 0) || (ty != 1) || (inode == dp->inode)) {
+
+# if defined(HASDCACHE)
+ if (DCunsafe && !dp->v && !vfy_dev(dp))
+ goto lkupdev_again;
+# endif /* defined(HASDCACHE) */
+
+ return(dp);
+ }
+ if (inode < dp->inode)
+ hi = mid - 1;
+ else
+ low = mid + 1;
+ }
+ }
+
+# if defined(HASDCACHE)
+ if (DCunsafe && r) {
+ (void) rereaddev();
+ goto lkupdev_again;
+ }
+# endif /* defined(HASDCACHE) */
+
+ return((struct l_dev *)NULL);
+}
+#endif /* defined(USE_LIB_LKUPDEV) */
--- /dev/null
+/*
+ * pdvn.c -- print device name functions for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_PRINTDEVNAME)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: pdvn.c,v 1.8 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+#else /* !defined(USE_LIB_PRINTDEVNAME) */
+char pdvn_d1[] = "d"; char *pdvn_d2 = pdvn_d1;
+#endif /* defined(USE_LIB_PRINTDEVNAME) */
+
+
+/*
+ * To use this source file:
+ *
+ * 1. Define USE_LIB_PRINTDEVNAME, or both.
+ *
+ * 2. Define HAS_STD_CLONE to enable standard clone searches in
+ * printdevname().
+ *
+ * 3. Define HASBLDKDEV to enable block device processing.
+ */
+
+
+/*
+ * Local definitions
+ */
+
+#define LIKE_BLK_SPEC "like block special"
+#define LIKE_CHR_SPEC "like character special"
+
+
+# if defined(USE_LIB_PRINTDEVNAME)
+/*
+ * printdevname() - print block or character device name
+ */
+
+int
+printdevname(dev, rdev, f, nty)
+ dev_t *dev; /* device */
+ dev_t *rdev; /* raw device */
+ int f; /* 1 = print trailing '\n' */
+ int nty; /* node type: N_BLK or N_CHR */
+{
+
+# if defined(HAS_STD_CLONE)
+ struct clone *c;
+# endif /* defined(HAS_STD_CLONE) */
+
+ struct l_dev *dp;
+ int r = 1;
+
+# if defined(HASDCACHE)
+
+printdevname_again:
+
+# endif /* defined(HASDCACHE) */
+
+# if defined(HAS_STD_CLONE)
+/*
+ * Search for clone if this is a character device on the same device as
+ * /dev (or /devices).
+ */
+ if ((nty == N_CHR) && Lf->is_stream && Clone && (*dev == DevDev)) {
+ r = 0; /* Don't let lkupdev() rebuild the device cache,
+ * because when it has been rebuilt we want to
+ * search again for clones. */
+ readdev(0);
+ for (c = Clone; c; c = c->next) {
+ if (GET_MAJ_DEV(*rdev) == GET_MIN_DEV(Devtp[c->dx].rdev)) {
+
+# if defined(HASDCACHE)
+ if (DCunsafe && !Devtp[c->dx].v && !vfy_dev(&Devtp[c->dx]))
+ goto printdevname_again;
+# endif /* defined(HASDCACHE) */
+
+ safestrprt(Devtp[c->dx].name, stdout, f);
+ return(1);
+ }
+ }
+ }
+# endif /* defined(HAS_STD_CLONE) */
+
+/*
+ * Search appropriate device table for a full match.
+ */
+
+# if defined(HASBLKDEV)
+ if (nty == N_BLK)
+ dp = lkupbdev(dev, rdev, 1, r);
+ else
+# endif /* defined(HASBLKDEV) */
+
+ dp = lkupdev(dev, rdev, 1, r);
+ if (dp) {
+ safestrprt(dp->name, stdout, f);
+ return(1);
+ }
+/*
+ * Search device table for a match without inode number and dev.
+ */
+
+# if defined(HASBLKDEV)
+ if (nty == N_BLK)
+ dp = lkupbdev(&DevDev, rdev, 0, r);
+ else
+# endif /* defined(HASBLKDEV) */
+
+ dp = lkupdev(&DevDev, rdev, 0, r);
+ if (dp) {
+ /*
+ * A match was found. Record it as a name column addition.
+ */
+ char *cp, *ttl;
+ int len;
+
+ ttl = (nty == N_BLK) ? LIKE_BLK_SPEC : LIKE_CHR_SPEC;
+ len = (int)(1 + strlen(ttl) + 1 + strlen(dp->name) + 1);
+ if (!(cp = (char *)malloc((MALLOC_S)(len + 1)))) {
+ (void) fprintf(stderr, "%s: no nma space for: (%s %s)\n",
+ Pn, ttl, dp->name);
+ Exit(1);
+ }
+ (void) snpf(cp, len + 1, "(%s %s)", ttl, dp->name);
+ (void) add_nma(cp, len);
+ (void) free((MALLOC_P *)cp);
+ return(0);
+ }
+
+# if defined(HASDCACHE)
+/*
+ * We haven't found a match.
+ *
+ * If rebuilding the device cache was suppressed and the device cache is
+ * "unsafe," rebuild it.
+ */
+ if (!r && DCunsafe) {
+ (void) rereaddev();
+ goto printdevname_again;
+ }
+# endif /* defined(HASDCACHE) */
+
+ return(0);
+}
+#endif /* defined(USE_LIB_PRINTDEVNAME) */
--- /dev/null
+/*
+ * prfp.c -- process_file() function for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_PROCESS_FILE)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: prfp.c,v 1.14 2008/10/21 16:12:36 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+/*
+ * process_file() - process file
+ */
+
+/*
+ * The caller may define:
+ *
+ * FILEPTR as the name of the location to store a pointer
+ * to the current file struct -- e.g.,
+ *
+ * struct file *foobar;
+ * #define FILEPTR foobar
+ */
+
+void
+process_file(fp)
+ KA_T fp; /* kernel file structure address */
+{
+ struct file f;
+ int flag;
+ char tbuf[32];
+
+#if defined(FILEPTR)
+/*
+ * Save file structure address for process_node().
+ */
+ FILEPTR = &f;
+#endif /* defined(FILEPTR) */
+
+/*
+ * Read file structure.
+ */
+ if (kread((KA_T)fp, (char *)&f, sizeof(f))) {
+ (void) snpf(Namech, Namechl, "can't read file struct from %s",
+ print_kptr(fp, (char *)NULL, 0));
+ enter_nm(Namech);
+ return;
+ }
+ Lf->off = (SZOFFTYPE)f.f_offset;
+ if (f.f_count) {
+
+ /*
+ * Construct access code.
+ */
+ if ((flag = (f.f_flag & (FREAD | FWRITE))) == FREAD)
+ Lf->access = 'r';
+ else if (flag == FWRITE)
+ Lf->access = 'w';
+ else if (flag == (FREAD | FWRITE))
+ Lf->access = 'u';
+
+#if defined(HASFSTRUCT)
+ /*
+ * Save file structure values.
+ */
+
+# if !defined(HASNOFSCOUNT)
+ if (Fsv & FSV_CT) {
+ Lf->fct = (long)f.f_count;
+ Lf->fsv |= FSV_CT;
+ }
+# endif /* !defined(HASNOFSCOUNT) */
+
+# if !defined(HASNOFSADDR)
+ if (Fsv & FSV_FA) {
+ Lf->fsa = fp;
+ Lf->fsv |= FSV_FA;
+ }
+# endif /* !defined(HASNOFSADDR) */
+
+# if !defined(HASNOFSFLAGS)
+ if (Fsv & FSV_FG) {
+ Lf->ffg = (long)f.f_flag;
+ Lf->fsv |= FSV_FG;
+ }
+# endif /* !defined(HASNOFSFLAGS) */
+
+# if !defined(HASNOFSNADDR)
+ if (Fsv & FSV_NI) {
+ Lf->fna = (KA_T)f.f_data;
+ Lf->fsv |= FSV_NI;
+ }
+# endif /* !defined(HASNOFSNADDR) */
+#endif /* defined(HASFSTRUCT) */
+
+ /*
+ * Process structure by its type.
+ */
+ switch (f.f_type) {
+
+
+#if defined(DTYPE_PIPE)
+ case DTYPE_PIPE:
+# if defined(HASPIPEFN)
+ if (!Selinet)
+ HASPIPEFN((KA_T)f.f_data);
+# endif /* defined(HASPIPEFN) */
+ return;
+#endif /* defined(DTYPE_PIPE) */
+
+#if defined(DTYPE_GNODE)
+ case DTYPE_GNODE:
+#endif /* defined(DTYPE_GNODE) */
+
+#if defined(DTYPE_INODE)
+ case DTYPE_INODE:
+#endif /* defined(DTYPE_INODE) */
+
+#if defined(DTYPE_PORT)
+ case DTYPE_PORT:
+#endif /* defined(DTYPE_PORT) */
+
+#if defined(DTYPE_VNODE)
+ case DTYPE_VNODE:
+#endif /* defined(DTYPE_VNODE) */
+
+#if defined(HASF_VNODE)
+ process_node((KA_T)f.f_vnode);
+#else /* !defined(HASF_VNODE) */
+ process_node((KA_T)f.f_data);
+#endif /* defined(HASF_VNODE) */
+
+ return;
+ case DTYPE_SOCKET:
+ process_socket((KA_T)f.f_data);
+ return;
+
+#if defined(HASKQUEUE)
+ case DTYPE_KQUEUE:
+ process_kqueue((KA_T)f.f_data);
+ return;
+#endif /* defined(HASKQUEUE) */
+
+#if defined(HASPSXSEM)
+ case DTYPE_PSXSEM:
+ process_psxsem((KA_T)f.f_data);
+ return;
+#endif /* defined(HASPSXSEM) */
+
+#if defined(HASPSXSHM)
+ case DTYPE_PSXSHM:
+ process_psxshm((KA_T)f.f_data);
+ return;
+#endif /* defined(HASPSXSHM) */
+
+#if defined(HASPRIVFILETYPE)
+ case PRIVFILETYPE:
+ HASPRIVFILETYPE((KA_T)f.f_data);
+ return;
+#endif /* defined(HASPRIVFILETYPE) */
+
+ default:
+ if (f.f_type || f.f_ops) {
+ (void) snpf(Namech, Namechl,
+ "%s file struct, ty=%#x, op=%s",
+ print_kptr(fp, tbuf, sizeof(tbuf)), (int)f.f_type,
+ print_kptr((KA_T)f.f_ops, (char *)NULL, 0));
+ enter_nm(Namech);
+ return;
+ }
+ }
+ }
+ enter_nm("no more information");
+}
+#else /* !defined(USE_LIB_PROCESS_FILE) */
+char prfp_d1[] = "d"; char *prfp_d2 = prfp_d1;
+#endif /* defined(USE_LIB_PROCESS_FILE) */
--- /dev/null
+/*
+ * ptti.c -- BSD style print_tcptpi() function for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_PRINT_TCPTPI)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: ptti.c,v 1.6 2008/10/21 16:13:23 abe Exp $";
+# endif /* !defined(lint) */
+
+#define TCPSTATES /* activate tcpstates[] */
+#include "../lsof.h"
+
+
+/*
+ * build_IPstates() -- build the TCP and UDP state tables
+ *
+ * Note: this module does not support a UDP state table.
+ */
+
+void
+build_IPstates()
+{
+
+/*
+ * Set the TcpNstates global variable.
+ */
+ TcpNstates = TCP_NSTATES;
+ TcpSt = (char **)&tcpstates;
+}
+
+
+/*
+ * print_tcptpi() - print TCP/TPI info
+ */
+
+void
+print_tcptpi(nl)
+ int nl; /* 1 == '\n' required */
+{
+ int ps = 0;
+ int s;
+
+ if ((Ftcptpi & TCPTPI_STATE) && Lf->lts.type == 0) {
+ if (Ffield)
+ (void) printf("%cST=", LSOF_FID_TCPTPI);
+ else
+ putchar('(');
+ if (!TcpNstates)
+ (void) build_IPstates();
+ if ((s = Lf->lts.state.i) < 0 || s >= TcpNstates)
+ (void) printf("UNKNOWN_TCP_STATE_%d", s);
+ else
+ (void) fputs(TcpSt[s], stdout);
+ ps++;
+ if (Ffield)
+ putchar(Terminator);
+ }
+
+#if defined(HASTCPTPIQ)
+ if (Ftcptpi & TCPTPI_QUEUES) {
+ if (Lf->lts.rqs) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("QR=%lu", Lf->lts.rq);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ if (Lf->lts.sqs) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("QS=%lu", Lf->lts.sq);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ }
+#endif /* defined(HASTCPTPIQ) */
+
+#if defined(HASSOOPT)
+ if (Ftcptpi & TCPTPI_FLAGS) {
+ int opt;
+
+ if ((opt = Lf->lts.opt)
+ || Lf->lts.pqlens || Lf->lts.qlens || Lf->lts.qlims
+ || Lf->lts.rbszs || Lf->lts.sbsz
+ ) {
+ char sep = ' ';
+
+ if (Ffield)
+ sep = LSOF_FID_TCPTPI;
+ else if (!ps)
+ sep = '(';
+ (void) printf("%cSO", sep);
+ ps++;
+ sep = '=';
+
+# if defined(SO_ACCEPTCONN)
+ if (opt & SO_ACCEPTCONN) {
+ (void) printf("%cACCEPTCONN", sep);
+ opt &= ~SO_ACCEPTCONN;
+ sep = ',';
+ }
+# endif /* defined(SO_ACCEPTCONN) */
+
+# if defined(SO_ACCEPTFILTER)
+ if (opt & SO_ACCEPTFILTER) {
+ (void) printf("%cACCEPTFILTER", sep);
+ opt &= ~SO_ACCEPTFILTER;
+ sep = ',';
+ }
+# endif /* defined(SO_ACCEPTFILTER) */
+
+# if defined(SO_AUDIT)
+ if (opt & SO_AUDIT) {
+ (void) printf("%cAUDIT", sep);
+ opt &= ~SO_AUDIT;
+ sep = ',';
+ }
+# endif /* defined(SO_AUDIT) */
+
+# if defined(SO_BINDANY)
+ if (opt & SO_BINDANY) {
+ (void) printf("%cBINDANY", sep);
+ opt &= ~SO_BINDANY;
+ sep = ',';
+ }
+# endif /* defined(SO_BINDANY) */
+
+# if defined(SO_BINTIME)
+ if (opt & SO_BINTIME) {
+ (void) printf("%cBINTIME", sep);
+ opt &= ~SO_BINTIME;
+ sep = ',';
+ }
+# endif /* defined(SO_BINTIME) */
+
+# if defined(SO_BROADCAST)
+ if (opt & SO_BROADCAST) {
+ (void) printf("%cBROADCAST", sep);
+ opt &= ~SO_BROADCAST;
+ sep = ',';
+ }
+# endif /* defined(SO_BROADCAST) */
+
+# if defined(SO_CKSUMRECV)
+ if (opt & SO_CKSUMRECV) {
+ (void) printf("%cCKSUMRECV", sep);
+ opt &= ~SO_CKSUMRECV;
+ sep = ',';
+ }
+# endif /* defined(SO_CKSUMRECV) */
+
+# if defined(SO_CLUA_IN_NOALIAS)
+ if (opt & SO_CLUA_IN_NOALIAS) {
+ (void) printf("%cCLUA_IN_NOALIAS", sep);
+ opt &= ~SO_CLUA_IN_NOALIAS;
+ sep = ',';
+ }
+# endif /* defined(SO_CLUA_IN_NOALIAS) */
+
+# if defined(SO_CLUA_IN_NOLOCAL)
+ if (opt & SO_CLUA_IN_NOLOCAL) {
+ (void) printf("%cCLUA_IN_NOLOCAL", sep);
+ opt &= ~SO_CLUA_IN_NOLOCAL;
+ sep = ',';
+ }
+# endif /* defined(SO_CLUA_IN_NOLOCAL) */
+
+# if defined(SO_DEBUG)
+ if (opt & SO_DEBUG) {
+ (void) printf("%cDEBUG", sep);
+ opt &= ~ SO_DEBUG;
+ sep = ',';
+ }
+# endif /* defined(SO_DEBUG) */
+
+# if defined(SO_DGRAM_ERRIND)
+ if (opt & SO_DGRAM_ERRIND) {
+ (void) printf("%cDGRAM_ERRIND", sep);
+ opt &= ~SO_DGRAM_ERRIND;
+ sep = ',';
+ }
+# endif /* defined(SO_DGRAM_ERRIND) */
+
+# if defined(SO_DONTROUTE)
+ if (opt & SO_DONTROUTE) {
+ (void) printf("%cDONTROUTE", sep);
+ opt &= ~SO_DONTROUTE;
+ sep = ',';
+ }
+# endif /* defined(SO_DONTROUTE) */
+
+# if defined(SO_DONTTRUNC)
+ if (opt & SO_DONTTRUNC) {
+ (void) printf("%cDONTTRUNC", sep);
+ opt &= ~SO_DONTTRUNC;
+ sep = ',';
+ }
+# endif /* defined(SO_DONTTRUNC) */
+
+# if defined(SO_EXPANDED_RIGHTS)
+ if (opt & SO_EXPANDED_RIGHTS) {
+ (void) printf("%cEXPANDED_RIGHTS", sep);
+ opt &= ~SO_EXPANDED_RIGHTS;
+ sep = ',';
+ }
+# endif /* defined(SO_EXPANDED_RIGHTS) */
+
+# if defined(SO_KEEPALIVE)
+ if (opt & SO_KEEPALIVE) {
+ (void) printf("%cKEEPALIVE", sep);
+ if (Lf->lts.kai)
+ (void) printf("=%d", Lf->lts.kai);
+ opt &= ~SO_KEEPALIVE;
+ sep = ',';
+ }
+# endif /* defined(SO_KEEPALIVE) */
+
+# if defined(SO_KERNACCEPT)
+ if (opt & SO_KERNACCEPT) {
+ (void) printf("%cKERNACCEPT", sep);
+ opt &= ~SO_KERNACCEPT;
+ sep = ',';
+ }
+# endif /* defined(SO_KERNACCEPT) */
+
+# if defined(SO_IMASOCKET)
+ if (opt & SO_IMASOCKET) {
+ (void) printf("%cIMASOCKET", sep);
+ opt &= ~SO_IMASOCKET;
+ sep = ',';
+ }
+# endif /* defined(SO_IMASOCKET) */
+
+# if defined(SO_LINGER)
+ if (opt & SO_LINGER) {
+ (void) printf("%cLINGER", sep);
+ if (Lf->lts.ltm)
+ (void) printf("=%d", Lf->lts.ltm);
+ opt &= ~SO_LINGER;
+ sep = ',';
+ }
+# endif /* defined(SO_LINGER) */
+
+# if defined(SO_LISTENING)
+ if (opt & SO_LISTENING) {
+ (void) printf("%cLISTENING", sep);
+ opt &= ~SO_LISTENING;
+ sep = ',';
+ }
+# endif /* defined(SO_LISTENING) */
+
+# if defined(SO_MGMT)
+ if (opt & SO_MGMT) {
+ (void) printf("%cMGMT", sep);
+ opt &= ~SO_MGMT;
+ sep = ',';
+ }
+# endif /* defined(SO_MGMT) */
+
+# if defined(SO_PAIRABLE)
+ if (opt & SO_PAIRABLE) {
+ (void) printf("%cPAIRABLE", sep);
+ opt &= ~SO_PAIRABLE;
+ sep = ',';
+ }
+# endif /* defined(SO_PAIRABLE) */
+
+# if defined(SO_RESVPORT)
+ if (opt & SO_RESVPORT) {
+ (void) printf("%cRESVPORT", sep);
+ opt &= ~SO_RESVPORT;
+ sep = ',';
+ }
+# endif /* defined(SO_RESVPORT) */
+
+# if defined(SO_NOREUSEADDR)
+ if (opt & SO_NOREUSEADDR) {
+ (void) printf("%cNOREUSEADDR", sep);
+ opt &= ~SO_NOREUSEADDR;
+ sep = ',';
+ }
+# endif /* defined(SO_NOREUSEADDR) */
+
+# if defined(SO_NOSIGPIPE)
+ if (opt & SO_NOSIGPIPE) {
+ (void) printf("%cNOSIGPIPE", sep);
+ opt &= ~SO_NOSIGPIPE;
+ sep = ',';
+ }
+# endif /* defined(SO_NOSIGPIPE) */
+
+# if defined(SO_OOBINLINE)
+ if (opt & SO_OOBINLINE) {
+ (void) printf("%cOOBINLINE", sep);
+ opt &= ~SO_OOBINLINE;
+ sep = ',';
+ }
+# endif /* defined(SO_OOBINLINE) */
+
+# if defined(SO_ORDREL)
+ if (opt & SO_ORDREL) {
+ (void) printf("%cORDREL", sep);
+ opt &= ~SO_ORDREL;
+ sep = ',';
+ }
+# endif /* defined(SO_ORDREL) */
+
+ if (Lf->lts.pqlens) {
+ (void) printf("%cPQLEN=%u", sep, Lf->lts.pqlen);
+ sep = ',';
+ }
+ if (Lf->lts.qlens) {
+ (void) printf("%cQLEN=%u", sep, Lf->lts.qlen);
+ sep = ',';
+ }
+ if (Lf->lts.qlims) {
+ (void) printf("%cQLIM=%u", sep, Lf->lts.qlim);
+ sep = ',';
+ }
+ if (Lf->lts.rbszs) {
+ (void) printf("%cRCVBUF=%lu", sep, Lf->lts.rbsz);
+ sep = ',';
+ }
+
+# if defined(SO_REUSEADDR)
+ if (opt & SO_REUSEADDR) {
+ (void) printf("%cREUSEADDR", sep);
+ opt &= ~SO_REUSEADDR;
+ sep = ',';
+ }
+# endif /* defined(SO_REUSEADDR) */
+
+# if defined(SO_REUSEALIASPORT)
+ if (opt & SO_REUSEALIASPORT) {
+ (void) printf("%cREUSEALIASPORT", sep);
+ opt &= ~SO_REUSEALIASPORT;
+ sep = ',';
+ }
+# endif /* defined(SO_REUSEALIASPORT) */
+
+# if defined(SO_REUSEPORT)
+ if (opt & SO_REUSEPORT) {
+ (void) printf("%cREUSEPORT", sep);
+ opt &= ~SO_REUSEPORT;
+ sep = ',';
+ }
+# endif /* defined(SO_REUSEPORT) */
+
+# if defined(SO_REUSERAD)
+ if (opt & SO_REUSERAD) {
+ (void) printf("%cREUSERAD", sep);
+ opt &= ~SO_REUSERAD;
+ sep = ',';
+ }
+# endif /* defined(SO_REUSERAD) */
+
+# if defined(SO_SECURITY_REQUEST)
+ if (opt & SO_SECURITY_REQUEST) {
+ (void) printf("%cSECURITY_REQUEST", sep);
+ opt &= ~SO_SECURITY_REQUEST;
+ sep = ',';
+ }
+# endif /* defined(SO_SECURITY_REQUEST) */
+
+ if (Lf->lts.sbszs) {
+ (void) printf("%cSNDBUF=%lu", sep, Lf->lts.sbsz);
+ sep = ',';
+ }
+
+# if defined(SO_TIMESTAMP)
+ if (opt & SO_TIMESTAMP) {
+ (void) printf("%cTIMESTAMP", sep);
+ opt &= ~SO_TIMESTAMP;
+ sep = ',';
+ }
+# endif /* defined(SO_TIMESTAMP) */
+
+# if defined(SO_UMC)
+ if (opt & SO_UMC) {
+ (void) printf("%cUMC", sep);
+ opt &= ~SO_UMC;
+ sep = ',';
+ }
+# endif /* defined(SO_UMC) */
+
+# if defined(SO_USE_IFBUFS)
+ if (opt & SO_USE_IFBUFS) {
+ (void) printf("%cUSE_IFBUFS", sep);
+ opt &= ~SO_USE_IFBUFS;
+ sep = ',';
+ }
+# endif /* defined(SO_USE_IFBUFS) */
+
+# if defined(SO_USELOOPBACK)
+ if (opt & SO_USELOOPBACK) {
+ (void) printf("%cUSELOOPBACK", sep);
+ opt &= ~SO_USELOOPBACK;
+ sep = ',';
+ }
+# endif /* defined(SO_USELOOPBACK) */
+
+# if defined(SO_WANTMORE)
+ if (opt & SO_WANTMORE) {
+ (void) printf("%cWANTMORE", sep);
+ opt &= ~SO_WANTMORE;
+ sep = ',';
+ }
+# endif /* defined(SO_WANTMORE) */
+
+# if defined(SO_WANTOOBFLAG)
+ if (opt & SO_WANTOOBFLAG) {
+ (void) printf("%cWANTOOBFLAG", sep);
+ opt &= ~SO_WANTOOBFLAG;
+ sep = ',';
+ }
+# endif /* defined(SO_WANTOOBFLAG) */
+
+ if (opt)
+ (void) printf("%cUNKNOWN=%#x", sep, opt);
+ if (Ffield)
+ putchar(Terminator);
+ }
+ }
+#endif /* defined(HASSOOPT) */
+
+#if defined(HASSOSTATE)
+ if (Ftcptpi & TCPTPI_FLAGS) {
+ unsigned int ss;
+
+ if ((ss = Lf->lts.ss)) {
+ char sep = ' ';
+
+ if (Ffield)
+ sep = LSOF_FID_TCPTPI;
+ else if (!ps)
+ sep = '(';
+ (void) printf("%cSS", sep);
+ ps++;
+ sep = '=';
+
+# if defined(SS_ASYNC)
+ if (ss & SS_ASYNC) {
+ (void) printf("%cASYNC", sep);
+ ss &= ~SS_ASYNC;
+ sep = ',';
+ }
+# endif /* defined(SS_ASYNC) */
+
+# if defined(SS_BOUND)
+ if (ss & SS_BOUND) {
+ (void) printf("%cBOUND", sep);
+ ss &= ~SS_BOUND;
+ sep = ',';
+ }
+# endif /* defined(SS_BOUND) */
+
+# if defined(HASSBSTATE)
+# if defined(SBS_CANTRCVMORE)
+ if (Lf->lts.sbs_rcv & SBS_CANTRCVMORE) {
+ (void) printf("%cCANTRCVMORE", sep);
+ Lf->lts.sbs_rcv &= ~SBS_CANTRCVMORE;
+ sep = ',';
+ }
+# endif /* defined(SBS_CANTRCVMORE) */
+
+# if defined(SBS_CANTSENDMORE)
+ if (Lf->lts.sbs_snd & SBS_CANTSENDMORE) {
+ (void) printf("%cCANTSENDMORE", sep);
+ Lf->lts.sbs_snd &= ~SBS_CANTSENDMORE;
+ sep = ',';
+ }
+# endif /* defined(SS_CANTSENDMORE) */
+# else /* !defined(HASSBSTATE) */
+
+# if defined(SS_CANTRCVMORE)
+ if (ss & SS_CANTRCVMORE) {
+ (void) printf("%cCANTRCVMORE", sep);
+ ss &= ~SS_CANTRCVMORE;
+ sep = ',';
+ }
+# endif /* defined(SS_CANTRCVMORE) */
+
+# if defined(SS_CANTSENDMORE)
+ if (ss & SS_CANTSENDMORE) {
+ (void) printf("%cCANTSENDMORE", sep);
+ ss &= ~SS_CANTSENDMORE;
+ sep = ',';
+ }
+# endif /* defined(SS_CANTSENDMORE) */
+# endif /* defined(HASSBSTATE) */
+
+# if defined(SS_COMP)
+ if (ss & SS_COMP) {
+ (void) printf("%cCOMP", sep);
+ ss &= ~SS_COMP;
+ sep = ',';
+ }
+# endif /* defined(SS_COMP) */
+
+# if defined(SS_CONNECTOUT)
+ if (ss & SS_CONNECTOUT) {
+ (void) printf("%cCONNECTOUT", sep);
+ ss &= ~SS_CONNECTOUT;
+ sep = ',';
+ }
+# endif /* defined(SS_CONNECTOUT) */
+
+# if defined(SS_HIPRI)
+ if (ss & SS_HIPRI) {
+ (void) printf("%cHIPRI", sep);
+ ss &= ~SS_HIPRI;
+ sep = ',';
+ }
+# endif /* defined(SS_HIPRI) */
+
+# if defined(SS_IGNERR)
+ if (ss & SS_IGNERR) {
+ (void) printf("%cIGNERR", sep);
+ ss &= ~SS_IGNERR;
+ sep = ',';
+ }
+# endif /* defined(SS_IGNERR) */
+
+# if defined(SS_INCOMP)
+ if (ss & SS_INCOMP) {
+ (void) printf("%cINCOMP", sep);
+ ss &= ~SS_INCOMP;
+ sep = ',';
+ }
+# endif /* defined(SS_INCOMP) */
+
+# if defined(SS_IOCWAIT)
+ if (ss & SS_IOCWAIT) {
+ (void) printf("%cIOCWAIT", sep);
+ ss &= ~SS_IOCWAIT;
+ sep = ',';
+ }
+# endif /* defined(SS_IOCWAIT) */
+
+# if defined(SS_ISCONFIRMING)
+ if (ss & SS_ISCONFIRMING) {
+ (void) printf("%cISCONFIRMING", sep);
+ ss &= ~SS_ISCONFIRMING;
+ sep = ',';
+ }
+# endif /* defined(SS_ISCONFIRMING) */
+
+# if defined(SS_ISCONNECTED)
+ if (ss & SS_ISCONNECTED) {
+ (void) printf("%cISCONNECTED", sep);
+ ss &= ~SS_ISCONNECTED;
+ sep = ',';
+ }
+# endif /* defined(SS_ISCONNECTED) */
+
+# if defined(SS_ISCONNECTING)
+ if (ss & SS_ISCONNECTING) {
+ (void) printf("%cISCONNECTING", sep);
+ ss &= ~SS_ISCONNECTING;
+ sep = ',';
+ }
+# endif /* defined(SS_ISCONNECTING) */
+
+# if defined(SS_ISDISCONNECTING)
+ if (ss & SS_ISDISCONNECTING) {
+ (void) printf("%cISDISCONNECTING", sep);
+ ss &= ~SS_ISDISCONNECTING;
+ sep = ',';
+ }
+# endif /* defined(SS_ISDISCONNECTING) */
+
+# if defined(SS_MORETOSEND)
+ if (ss & SS_MORETOSEND) {
+ (void) printf("%cMORETOSEND", sep);
+ ss &= ~SS_MORETOSEND;
+ sep = ',';
+ }
+# endif /* defined(SS_MORETOSEND) */
+
+# if defined(SS_NBIO)
+ if (ss & SS_NBIO) {
+ (void) printf("%cNBIO", sep);
+ ss &= ~SS_NBIO;
+ sep = ',';
+ }
+# endif /* defined(SS_NBIO) */
+
+# if defined(SS_NOCONN)
+ if (ss & SS_NOCONN) {
+ (void) printf("%cNOCONN", sep);
+ ss &= ~SS_NOCONN;
+ sep = ',';
+ }
+# endif /* defined(SS_NOCONN) */
+
+# if defined(SS_NODELETE)
+ if (ss & SS_NODELETE) {
+ (void) printf("%cNODELETE", sep);
+ ss &= ~SS_NODELETE;
+ sep = ',';
+ }
+# endif /* defined(SS_NODELETE) */
+
+# if defined(SS_NOFDREF)
+ if (ss & SS_NOFDREF) {
+ (void) printf("%cNOFDREF", sep);
+ ss &= ~SS_NOFDREF;
+ sep = ',';
+ }
+# endif /* defined(SS_NOFDREF) */
+
+# if defined(SS_NOGHOST)
+ if (ss & SS_NOGHOST) {
+ (void) printf("%cNOGHOST", sep);
+ ss &= ~SS_NOGHOST;
+ sep = ',';
+ }
+# endif /* defined(SS_NOGHOST) */
+
+# if defined(SS_NOINPUT)
+ if (ss & SS_NOINPUT) {
+ (void) printf("%cNOINPUT", sep);
+ ss &= ~SS_NOINPUT;
+ sep = ',';
+ }
+# endif /* defined(SS_NOINPUT) */
+
+# if defined(SS_PRIV)
+ if (ss & SS_PRIV) {
+ (void) printf("%cPRIV", sep);
+ ss &= ~SS_PRIV;
+ sep = ',';
+ }
+# endif /* defined(SS_PRIV) */
+
+# if defined(SS_QUEUE)
+ if (ss & SS_QUEUE) {
+ (void) printf("%cQUEUE", sep);
+ ss &= ~SS_QUEUE;
+ sep = ',';
+ }
+# endif /* defined(SS_QUEUE) */
+
+# if defined(HASSBSTATE)
+# if defined(SBS_RCVATMARK)
+ if (Lf->lts.sbs_rcv & SBS_RCVATMARK) {
+ (void) printf("%cRCVATMARK", sep);
+ Lf->lts.sbs_rcv &= ~SBS_RCVATMARK;
+ sep = ',';
+ }
+# endif /* defined(SBS_RCVATMARK) */
+
+# else /* !defined(HASSBSTATE) */
+# if defined(SS_RCVATMARK)
+ if (ss & SS_RCVATMARK) {
+ (void) printf("%cRCVATMARK", sep);
+ ss &= ~SS_RCVATMARK;
+ sep = ',';
+ }
+# endif /* defined(SS_RCVATMARK) */
+# endif /* defined(HASSBSTATE) */
+
+# if defined(SS_READWAIT)
+ if (ss & SS_READWAIT) {
+ (void) printf("%cREADWAIT", sep);
+ ss &= ~SS_READWAIT;
+ sep = ',';
+ }
+# endif /* defined(SS_READWAIT) */
+
+# if defined(SS_SETRCV)
+ if (ss & SS_SETRCV) {
+ (void) printf("%cSETRCV", sep);
+ ss &= ~SS_SETRCV;
+ sep = ',';
+ }
+# endif /* defined(SS_SETRCV) */
+
+# if defined(SS_SETSND)
+ if (ss & SS_SETSND) {
+ (void) printf("%cSETSND", sep);
+ ss &= ~SS_SETSND;
+ sep = ',';
+ }
+# endif /* defined(SS_SETSND) */
+
+# if defined(SS_SIGREAD)
+ if (ss & SS_SIGREAD) {
+ (void) printf("%cSIGREAD", sep);
+ ss &= ~SS_SIGREAD;
+ sep = ',';
+ }
+# endif /* defined(SS_SIGREAD) */
+
+# if defined(SS_SIGWRITE)
+ if (ss & SS_SIGWRITE) {
+ (void) printf("%cSIGWRITE", sep);
+ ss &= ~SS_SIGWRITE;
+ sep = ',';
+ }
+# endif /* defined(SS_SIGWRITE) */
+
+# if defined(SS_SPLICED)
+ if (ss & SS_SPLICED) {
+ (void) printf("%cSPLICED", sep);
+ ss &= ~SS_SPLICED;
+ sep = ',';
+ }
+# endif /* defined(SS_SPLICED) */
+
+# if defined(SS_WRITEWAIT)
+ if (ss & SS_WRITEWAIT) {
+ (void) printf("%cWRITEWAIT", sep);
+ ss &= ~SS_WRITEWAIT;
+ sep = ',';
+ }
+# endif /* defined(SS_WRITEWAIT) */
+
+# if defined(SS_ZOMBIE)
+ if (ss & SS_ZOMBIE) {
+ (void) printf("%cZOMBIE", sep);
+ ss &= ~SS_ZOMBIE;
+ sep = ',';
+ }
+# endif /* defined(SS_ZOMBIE) */
+
+ if (ss)
+ (void) printf("%cUNKNOWN=%#x", sep, ss);
+ if (Ffield)
+ putchar(Terminator);
+ }
+ }
+#endif /* defined(HASSOSTATE) */
+
+#if defined(HASTCPOPT)
+ if (Ftcptpi & TCPTPI_FLAGS) {
+ int topt;
+
+ if ((topt = Lf->lts.topt) || Lf->lts.msss) {
+ char sep = ' ';
+
+ if (Ffield)
+ sep = LSOF_FID_TCPTPI;
+ else if (!ps)
+ sep = '(';
+ (void) printf("%cTF", sep);
+ ps++;
+ sep = '=';
+
+# if defined(TF_ACKNOW)
+ if (topt & TF_ACKNOW) {
+ (void) printf("%cACKNOW", sep);
+ topt &= ~TF_ACKNOW;
+ sep = ',';
+ }
+# endif /* defined(TF_ACKNOW) */
+
+# if defined(TF_CANT_TXSACK)
+ if (topt & TF_CANT_TXSACK) {
+ (void) printf("%cCANT_TXSACK", sep);
+ topt &= ~TF_CANT_TXSACK;
+ sep = ',';
+ }
+# endif /* defined(TF_CANT_TXSACK) */
+
+# if defined(TF_DEAD)
+ if (topt & TF_DEAD) {
+ (void) printf("%cDEAD", sep);
+ topt &= ~TF_DEAD;
+ sep = ',';
+ }
+# endif /* defined(TF_DEAD) */
+
+# if defined(TF_DELACK)
+ if (topt & TF_DELACK) {
+ (void) printf("%cDELACK", sep);
+ topt &= ~TF_DELACK;
+ sep = ',';
+ }
+# endif /* defined(TF_DELACK) */
+
+# if defined(TF_DELAY_ACK)
+ if (topt & TF_DELAY_ACK) {
+ (void) printf("%cDELAY_ACK", sep);
+ topt &= ~TF_DELAY_ACK;
+ sep = ',';
+ }
+# endif /* defined(TF_DELAY_ACK) */
+
+# if defined(TF_DISABLE_ECN)
+ if (topt & TF_DISABLE_ECN) {
+ (void) printf("%cDISABLE_ECN", sep);
+ topt &= ~TF_DISABLE_ECN;
+ sep = ',';
+ }
+# endif /* defined(TF_DISABLE_ECN) */
+
+# if defined(TF_ECN)
+ if (topt & TF_ECN) {
+ (void) printf("%cECN", sep);
+ topt &= ~TF_ECN;
+ sep = ',';
+ }
+# endif /* defined(TF_ECN) */
+
+# if defined(TF_ECN_PERMIT)
+ if (topt & TF_ECN_PERMIT) {
+ (void) printf("%cECN_PERMIT", sep);
+ topt &= ~TF_ECN_PERMIT;
+ sep = ',';
+ }
+# endif /* defined(TF_ECN_PERMIT) */
+
+# if defined(TF_FASTRECOVERY)
+ if (topt & TF_FASTRECOVERY) {
+ (void) printf("%cFASTRECOVERY", sep);
+ topt &= ~TF_FASTRECOVERY;
+ sep = ',';
+ }
+# endif /* defined(TF_FASTRECOVERY) */
+
+# if defined(TF_FASTRXMT_PHASE)
+ if (topt & TF_FASTRXMT_PHASE) {
+ (void) printf("%cFASTRXMT_PHASE", sep);
+ topt &= ~TF_FASTRXMT_PHASE;
+ sep = ',';
+ }
+# endif /* defined(TF_FASTRXMT_PHASE) */
+
+# if defined(TF_HAVEACKED)
+ if (topt & TF_HAVEACKED) {
+ (void) printf("%cHAVEACKED", sep);
+ topt &= ~TF_HAVEACKED;
+ sep = ',';
+ }
+# endif /* defined(TF_HAVEACKED) */
+
+# if defined(TF_HAVECLOSED)
+ if (topt & TF_HAVECLOSED) {
+ (void) printf("%cHAVECLOSED", sep);
+ topt &= ~TF_HAVECLOSED;
+ sep = ',';
+ }
+# endif /* defined(TF_HAVECLOSED) */
+
+# if defined(TF_IGNR_RXSACK)
+ if (topt & TF_IGNR_RXSACK) {
+ (void) printf("%cIGNR_RXSACK", sep);
+ topt &= ~TF_IGNR_RXSACK;
+ sep = ',';
+ }
+# endif /* defined(TF_IGNR_RXSACK) */
+
+# if defined(TF_IOLOCK)
+ if (topt & TF_IOLOCK) {
+ (void) printf("%cIOLOCK", sep);
+ topt &= ~TF_IOLOCK;
+ sep = ',';
+ }
+# endif /* defined(TF_IOLOCK) */
+
+# if defined(TF_LARGESEND)
+ if (topt & TF_LARGESEND) {
+ (void) printf("%cLARGESEND", sep);
+ topt &= ~TF_LARGESEND;
+ sep = ',';
+ }
+# endif /* defined(TF_LARGESEND) */
+
+# if defined(TF_LASTIDLE)
+ if (topt & TF_LASTIDLE) {
+ (void) printf("%cLASTIDLE", sep);
+ topt &= ~TF_LASTIDLE;
+ sep = ',';
+ }
+# endif /* defined(TF_LASTIDLE) */
+
+# if defined(TF_LQ_OVERFLOW)
+ if (topt & TF_LQ_OVERFLOW) {
+ (void) printf("%cLQ_OVERFLOW", sep);
+ topt &= ~TF_LQ_OVERFLOW;
+ sep = ',';
+ }
+# endif /* defined(TF_LQ_OVERFLOW) */
+
+ if (Lf->lts.msss) {
+ (void) printf("%cMSS=%lu", sep, Lf->lts.mss);
+ sep = ',';
+ }
+
+# if defined(TF_MORETOCOME)
+ if (topt & TF_MORETOCOME) {
+ (void) printf("%cMORETOCOME", sep);
+ topt &= ~TF_MORETOCOME;
+ sep = ',';
+ }
+# endif /* defined(TF_MORETOCOME) */
+
+# if defined(TF_NEEDACK)
+ if (topt & TF_NEEDACK) {
+ (void) printf("%cNEEDACK", sep);
+ topt &= ~TF_NEEDACK;
+ sep = ',';
+ }
+# endif /* defined(TF_NEEDACK) */
+
+# if defined(TF_NEEDCLOSE)
+ if (topt & TF_NEEDCLOSE) {
+ (void) printf("%cNEEDCLOSE", sep);
+ topt &= ~TF_NEEDCLOSE;
+ sep = ',';
+ }
+# endif /* defined(TF_NEEDCLOSE) */
+
+# if defined(TF_NEEDFIN)
+ if (topt & TF_NEEDFIN) {
+ (void) printf("%cNEEDFIN", sep);
+ topt &= ~TF_NEEDFIN;
+ sep = ',';
+ }
+# endif /* defined(TF_NEEDFIN) */
+
+# if defined(TF_NEEDIN)
+ if (topt & TF_NEEDIN) {
+ (void) printf("%cNEEDIN", sep);
+ topt &= ~TF_NEEDIN;
+ sep = ',';
+ }
+# endif /* defined(TF_NEEDIN) */
+
+# if defined(TF_NEEDOUT)
+ if (topt & TF_NEEDOUT) {
+ (void) printf("%cNEEDOUT", sep);
+ topt &= ~TF_NEEDOUT;
+ sep = ',';
+ }
+# endif /* defined(TF_NEEDOUT) */
+
+# if defined(TF_NEEDSYN)
+ if (topt & TF_NEEDSYN) {
+ (void) printf("%cNEEDSYN", sep);
+ topt &= ~TF_NEEDSYN;
+ sep = ',';
+ }
+# endif /* defined(TF_NEEDSYN) */
+
+# if defined(TF_NEEDTIMER)
+ if (topt & TF_NEEDTIMER) {
+ (void) printf("%cNEEDTIMER", sep);
+ topt &= ~TF_NEEDTIMER;
+ sep = ',';
+ }
+# endif /* defined(TF_NEEDTIMER) */
+
+# if defined(TF_NEWRENO_RXMT)
+ if (topt & TF_NEWRENO_RXMT) {
+ (void) printf("%cNEWRENO_RXMT", sep);
+ topt &= ~TF_NEWRENO_RXMT;
+ sep = ',';
+ }
+# endif /* defined(TF_NEWRENO_RXMT) */
+
+# if defined(TF_NODELACK)
+ if (topt & TF_NODELACK) {
+ (void) printf("%cNODELACK", sep);
+ topt &= ~TF_NODELACK;
+ sep = ',';
+ }
+# endif /* defined(TF_NODELACK) */
+
+# if defined(TF_NODELAY)
+ if (topt & TF_NODELAY) {
+ (void) printf("%cNODELAY", sep);
+ topt &= ~TF_NODELAY;
+ sep = ',';
+ }
+# endif /* defined(TF_NODELAY) */
+
+# if defined(TF_NOOPT)
+ if (topt & TF_NOOPT) {
+ (void) printf("%cNOOPT", sep);
+ topt &= ~TF_NOOPT;
+ sep = ',';
+ }
+# endif /* defined(TF_NOOPT) */
+
+# if defined(TF_NOPUSH)
+ if (topt & TF_NOPUSH) {
+ (void) printf("%cNOPUSH", sep);
+ topt &= ~TF_NOPUSH;
+ sep = ',';
+ }
+# endif /* defined(TF_NOPUSH) */
+
+# if defined(TF_NO_PMTU)
+ if (topt & TF_NO_PMTU) {
+ (void) printf("%cNO_PMTU", sep);
+ topt &= ~TF_NO_PMTU;
+ sep = ',';
+ }
+# endif /* defined(TF_NO_PMTU) */
+
+# if defined(TF_RAW)
+ if (topt & TF_RAW) {
+ (void) printf("%cRAW", sep);
+ topt &= ~TF_RAW;
+ sep = ',';
+ }
+# endif /* defined(TF_RAW) */
+
+# if defined(TF_RCVD_CC)
+ if (topt & TF_RCVD_CC) {
+ (void) printf("%cRCVD_CC", sep);
+ topt &= ~TF_RCVD_CC;
+ sep = ',';
+ }
+# endif /* defined(TF_RCVD_CC) */
+
+# if defined(TF_RCVD_SCALE)
+ if (topt & TF_RCVD_SCALE) {
+ (void) printf("%cRCVD_SCALE", sep);
+ topt &= ~TF_RCVD_SCALE;
+ sep = ',';
+ }
+# endif /* defined(TF_RCVD_SCALE) */
+
+# if defined(TF_RCVD_CE)
+ if (topt & TF_RCVD_CE) {
+ (void) printf("%cRCVD_CE", sep);
+ topt &= ~TF_RCVD_CE;
+ sep = ',';
+ }
+# endif /* defined(TF_RCVD_CE) */
+
+# if defined(TF_RCVD_TS)
+ if (topt & TF_RCVD_TS) {
+ (void) printf("%cRCVD_TS", sep);
+ topt &= ~TF_RCVD_TS;
+ sep = ',';
+ }
+# endif /* defined(TF_RCVD_TS) */
+
+# if defined(TF_RCVD_TSTMP)
+ if (topt & TF_RCVD_TSTMP) {
+ (void) printf("%cRCVD_TSTMP", sep);
+ topt &= ~TF_RCVD_TSTMP;
+ sep = ',';
+ }
+# endif /* defined(TF_RCVD_TSTMP) */
+
+# if defined(TF_RCVD_WS)
+ if (topt & TF_RCVD_WS) {
+ (void) printf("%cRCVD_WS", sep);
+ topt &= ~TF_RCVD_WS;
+ sep = ',';
+ }
+# endif /* defined(TF_RCVD_WS) */
+
+# if defined(TF_REASSEMBLING)
+ if (topt & TF_REASSEMBLING) {
+ (void) printf("%cREASSEMBLING", sep);
+ topt &= ~TF_REASSEMBLING;
+ sep = ',';
+ }
+# endif /* defined(TF_REASSEMBLING) */
+
+# if defined(TF_REQ_CC)
+ if (topt & TF_REQ_CC) {
+ (void) printf("%cREQ_CC", sep);
+ topt &= ~TF_REQ_CC;
+ sep = ',';
+ }
+# endif /* defined(TF_REQ_CC) */
+
+# if defined(TF_REQ_SCALE)
+ if (topt & TF_REQ_SCALE) {
+ (void) printf("%cREQ_SCALE", sep);
+ topt &= ~TF_REQ_SCALE;
+ sep = ',';
+ }
+# endif /* defined(TF_REQ_SCALE) */
+
+# if defined(TF_REQ_TSTMP)
+ if (topt & TF_REQ_TSTMP) {
+ (void) printf("%cREQ_TSTMP", sep);
+ topt &= ~TF_REQ_TSTMP;
+ sep = ',';
+ }
+# endif /* defined(TF_REQ_TSTMP) */
+
+# if defined(TF_RFC1323)
+ if (topt & TF_RFC1323) {
+ (void) printf("%cRFC1323", sep);
+ topt &= ~TF_RFC1323;
+ sep = ',';
+ }
+# endif /* defined(TF_RFC1323) */
+
+# if defined(TF_RXWIN0SENT)
+ if (topt & TF_RXWIN0SENT) {
+ (void) printf("%cRXWIN0SENT", sep);
+ topt &= ~TF_RXWIN0SENT;
+ sep = ',';
+ }
+# endif /* defined(TF_RXWIN0SENT) */
+
+# if defined(TF_SACK_GENERATE)
+ if (topt & TF_SACK_GENERATE) {
+ (void) printf("%cSACK_GENERATE", sep);
+ topt &= ~TF_SACK_GENERATE;
+ sep = ',';
+ }
+# endif /* defined(TF_SACK_GENERATE) */
+
+# if defined(TF_SACK_PERMIT)
+ if (topt & TF_SACK_PERMIT) {
+ (void) printf("%cSACK_PERMIT", sep);
+ topt &= ~TF_SACK_PERMIT;
+ sep = ',';
+ }
+# endif /* defined(TF_SACK_PERMIT) */
+
+# if defined(TF_SACK_PROCESS)
+ if (topt & TF_SACK_PROCESS) {
+ (void) printf("%cSACK_PROCESS", sep);
+ topt &= ~TF_SACK_PROCESS;
+ sep = ',';
+ }
+# endif /* defined(TF_SACK_PROCESS) */
+
+# if defined(TF_SEND)
+ if (topt & TF_SEND) {
+ (void) printf("%cSEND", sep);
+ topt &= ~TF_SEND;
+ sep = ',';
+ }
+# endif /* defined(TF_SEND) */
+
+# if defined(TF_SEND_AND_DISCONNECT)
+ if (topt & TF_SEND_AND_DISCONNECT) {
+ (void) printf("%cSEND_AND_DISCONNECT", sep);
+ topt &= ~TF_SEND_AND_DISCONNECT;
+ sep = ',';
+ }
+# endif /* defined(TF_SEND_AND_DISCONNECT) */
+
+# if defined(TF_SENDCCNEW)
+ if (topt & TF_SENDCCNEW) {
+ (void) printf("%cSENDCCNEW", sep);
+ topt &= ~TF_SENDCCNEW;
+ sep = ',';
+ }
+# endif /* defined(TF_SENDCCNEW) */
+
+# if defined(TF_SEND_CWR)
+ if (topt & TF_SEND_CWR) {
+ (void) printf("%cSEND_CWR", sep);
+ topt &= ~TF_SEND_CWR;
+ sep = ',';
+ }
+# endif /* defined(TF_SEND_CWR) */
+
+# if defined(TF_SEND_ECHO)
+ if (topt & TF_SEND_ECHO) {
+ (void) printf("%cSEND_ECHO", sep);
+ topt &= ~TF_SEND_ECHO;
+ sep = ',';
+ }
+# endif /* defined(TF_SEND_ECHO) */
+
+# if defined(TF_SEND_TSTMP)
+ if (topt & TF_SEND_TSTMP) {
+ (void) printf("%cSEND_TSTMP", sep);
+ topt &= ~TF_SEND_TSTMP;
+ sep = ',';
+ }
+# endif /* defined(TF_SEND_TSTMP) */
+
+# if defined(TF_SENTFIN)
+ if (topt & TF_SENTFIN) {
+ (void) printf("%cSENTFIN", sep);
+ topt &= ~TF_SENTFIN;
+ sep = ',';
+ }
+# endif /* defined(TF_SENTFIN) */
+
+# if defined(TF_SENT_TS)
+ if (topt & TF_SENT_TS) {
+ (void) printf("%cSENT_TS", sep);
+ topt &= ~TF_SENT_TS;
+ sep = ',';
+ }
+# endif /* defined(TF_SENT_TS) */
+
+# if defined(TF_SENT_WS)
+ if (topt & TF_SENT_WS) {
+ (void) printf("%cSENT_WS", sep);
+ topt &= ~TF_SENT_WS;
+ sep = ',';
+ }
+# endif /* defined(TF_SENT_WS) */
+
+# if defined(TF_SIGNATURE)
+ if (topt & TF_SIGNATURE) {
+ (void) printf("%cSIGNATURE", sep);
+ topt &= ~TF_SIGNATURE;
+ sep = ',';
+ }
+# endif /* defined(TF_SIGNATURE) */
+
+# if defined(TF_SLOWLINK)
+ if (topt & TF_SLOWLINK) {
+ (void) printf("%cSLOWLINK", sep);
+ topt &= ~TF_SLOWLINK;
+ sep = ',';
+ }
+# endif /* defined(TF_SLOWLINK) */
+
+# if defined(TF_STDURG)
+ if (topt & TF_STDURG) {
+ (void) printf("%cSTDURG", sep);
+ topt &= ~TF_STDURG;
+ sep = ',';
+ }
+# endif /* defined(TF_STDURG) */
+
+# if defined(TF_SYN_REXMT)
+ if (topt & TF_SYN_REXMT) {
+ (void) printf("%cSYN_REXMT", sep);
+ topt &= ~TF_SYN_REXMT;
+ sep = ',';
+ }
+# endif /* defined(TF_SYN_REXMT) */
+
+# if defined(TF_UIOMOVED)
+ if (topt & TF_UIOMOVED) {
+ (void) printf("%cUIOMOVED", sep);
+ topt &= ~TF_UIOMOVED;
+ sep = ',';
+ }
+# endif /* defined(TF_UIOMOVED) */
+
+# if defined(TF_USE_SCALE)
+ if (topt & TF_USE_SCALE) {
+ (void) printf("%cUSE_SCALE", sep);
+ topt &= ~TF_USE_SCALE;
+ sep = ',';
+ }
+# endif /* defined(TF_USE_SCALE) */
+
+# if defined(TF_WASIDLE)
+ if (topt & TF_WASIDLE) {
+ (void) printf("%cWASIDLE", sep);
+ topt &= ~TF_WASIDLE;
+ sep = ',';
+ }
+# endif /* defined(TF_WASIDLE) */
+
+# if defined(TF_WASFRECOVERY)
+ if (topt & TF_WASFRECOVERY) {
+ (void) printf("%cWASFRECOVERY", sep);
+ topt &= ~TF_WASFRECOVERY;
+ sep = ',';
+ }
+# endif /* defined(TF_WASFRECOVERY) */
+
+# if defined(TF_WILL_SACK)
+ if (topt & TF_WILL_SACK) {
+ (void) printf("%cWILL_SACK", sep);
+ topt &= ~TF_WILL_SACK;
+ sep = ',';
+ }
+# endif /* defined(TF_WILL_SACK) */
+
+ if (topt)
+ (void) printf("%cUNKNOWN=%#x", sep, topt);
+ if (Ffield)
+ putchar(Terminator);
+ }
+ }
+#endif /* defined(HASTCPOPT) */
+
+#if defined(HASTCPTPIW)
+ if (Ftcptpi & TCPTPI_WINDOWS) {
+ if (Lf->lts.rws) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("WR=%lu", Lf->lts.rw);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ if (Lf->lts.wws) {
+ if (Ffield)
+ putchar(LSOF_FID_TCPTPI);
+ else {
+ if (ps)
+ putchar(' ');
+ else
+ putchar('(');
+ }
+ (void) printf("WW=%lu", Lf->lts.ww);
+ if (Ffield)
+ putchar(Terminator);
+ ps++;
+ }
+ }
+#endif /* defined(HASTCPTPIW) */
+
+ if (ps && !Ffield)
+ putchar(')');
+ if (nl)
+ putchar('\n');
+}
+#else /* !defined(USE_LIB_PRINT_TCPTPI) */
+char ptti_d1[] = "d"; char *ptti_d2 = ptti_d1;
+#endif /* defined(USE_LIB_PRINT_TCPTPI) */
--- /dev/null
+/*
+ * rdev.c -- readdev() function for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_READDEV)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: rdev.c,v 1.12 2008/10/21 16:13:23 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+_PROTOTYPE(static int rmdupdev,(struct l_dev ***dp, int n, char *nm));
+
+
+/*
+ * To use this source file:
+ *
+ * 1. Define DIRTYPE as:
+ *
+ * #define DIRTYPE direct
+ * or #define DIRTYPE dirent
+ *
+ * 2. Define HASDNAMLEN if struct DIRTYPE has a d_namlen element, giving
+ * the length of d_name.
+ *
+ * 3. Define the RDEV_EXPDEV macro to apply special handling to device
+ * numbers, as required. For example, for EP/IX 2.1.1:
+ *
+ * #define RDEV_EXPDEV(n) expdev(n)
+ *
+ * to use the expdev() function to expand device numbers. If
+ * no RDEV_EXPDEV macro is defined, it defaults to:
+ *
+ * #define RDEV_EXPDEV(n) (n)
+ *
+ * 4. Define HASBLKDEV to request that information on S_IFBLK devices be
+ * recorded in BDevtp[].
+ *
+ * Define NOWARNBLKDEV to suppress the issuance of a warning when no
+ * block devices are found.
+ *
+ * 5. Define RDEV_STATFN to be a stat function other than stat() or lstat()
+ * -- e.g.,
+ *
+ * #define RDEV_STATFN private_stat
+ *
+ * 6. Define HAS_STD_CLONE to request that clone device information be stored
+ * in standard clone structures (defined in lsof.h and addressed via
+ * Clone). If HAS_STD_CLONE is defined, these must also be defined:
+ *
+ * a. Define CLONEMAJ to be the name of the constant or
+ * variable that defines the clone major device -- e.g.,
+ *
+ * #define CLONEMAJ CloneMaj
+ *
+ * b. Define HAVECLONEMAJ to be the name of the variable that
+ * contains the status of the clone major device -- e.g.,
+ *
+ * #define HAVECLONEMAJ HaveCloneMaj
+ *
+ * Define HAS_STD_CLONE to be 1 if readdev() is expected to build the
+ * clone table, the clone table is cached (if HASDCACHE is defined), and
+ * there is a function to clear the cache table when the device table must
+ * be reloaded. (See dvch.c for naming the clone cache build and clear
+ * functions.)
+ */
+
+
+# if !defined(RDEV_EXPDEV)
+#define RDEV_EXPDEV(n) (n)
+# endif /* !defined(RDEV_EXPDEV) */
+
+# if !defined(RDEV_STATFN)
+# if defined(USE_STAT)
+#define RDEV_STATFN stat
+# else /* !defined(USE_STAT) */
+#define RDEV_STATFN lstat
+# endif /* defined(USE_STAT) */
+# endif /* !defined(RDEV_STATFN) */
+
+
+/*
+ * readdev() - read device names, modes and types
+ */
+
+void
+readdev(skip)
+ int skip; /* skip device cache read if 1 */
+{
+
+# if defined(HAS_STD_CLONE) && HAS_STD_CLONE==1
+ struct clone *c;
+# endif /* defined(HAS_STD_CLONE) && HAS_STD_CLONE==1 */
+
+# if defined(HASDCACHE)
+ int dcrd;
+# endif /* defined(HASDCACHE) */
+
+ DIR *dfp;
+ int dnamlen;
+ struct DIRTYPE *dp;
+ char *fp = (char *)NULL;
+ int i = 0;
+
+# if defined(HASBLKDEV)
+ int j = 0;
+# endif /* defined(HASBLKDEV) */
+
+ char *path = (char *)NULL;
+ MALLOC_S pl;
+ struct stat sb;
+
+ if (Sdev)
+ return;
+
+# if defined(HASDCACHE)
+/*
+ * Read device cache, as directed.
+ */
+ if (!skip) {
+ if (DCstate == 2 || DCstate == 3) {
+ if ((dcrd = read_dcache()) == 0)
+ return;
+ }
+ } else
+ dcrd = 1;
+# endif /* defined(HASDCACHE) */
+
+ Dstkn = Dstkx = 0;
+ Dstk = (char **)NULL;
+ (void) stkdir("/dev");
+/*
+ * Unstack the next /dev or /dev/<subdirectory> directory.
+ */
+ while (--Dstkx >= 0) {
+ if (!(dfp = OpenDir(Dstk[Dstkx]))) {
+
+# if defined(WARNDEVACCESS)
+ if (!Fwarn) {
+ (void) fprintf(stderr, "%s: WARNING: can't open: ", Pn);
+ safestrprt(Dstk[Dstkx], stderr, 1);
+ }
+# endif /* defined(WARNDEVACCESS) */
+
+ (void) free((FREE_P *)Dstk[Dstkx]);
+ Dstk[Dstkx] = (char *)NULL;
+ continue;
+ }
+ if (path) {
+ (void) free((FREE_P *)path);
+ path = (char *)NULL;
+ }
+ if (!(path = mkstrcat(Dstk[Dstkx], -1, "/", 1, (char *)NULL, -1,
+ &pl)))
+ {
+ (void) fprintf(stderr, "%s: no space for: ", Pn);
+ safestrprt(Dstk[Dstkx], stderr, 1);
+ Exit(1);
+ }
+ (void) free((FREE_P *)Dstk[Dstkx]);
+ Dstk[Dstkx] = (char *)NULL;
+ /*
+ * Scan the directory.
+ */
+ for (dp = ReadDir(dfp); dp; dp = ReadDir(dfp)) {
+ if (dp->d_ino == 0 || dp->d_name[0] == '.')
+ continue;
+ /*
+ * Form the full path name and get its status.
+ */
+
+# if defined(HASDNAMLEN)
+ dnamlen = (int)dp->d_namlen;
+# else /* !defined(HASDNAMLEN) */
+ dnamlen = (int)strlen(dp->d_name);
+# endif /* defined(HASDNAMLEN) */
+
+ if (fp) {
+ (void) free((FREE_P *)fp);
+ fp = (char *)NULL;
+ }
+ if (!(fp = mkstrcat(path, pl, dp->d_name, dnamlen,
+ (char *)NULL, -1, (MALLOC_S *)NULL)))
+ {
+ (void) fprintf(stderr, "%s: no space for: ", Pn);
+ safestrprt(path, stderr, 0);
+ safestrprtn(dp->d_name, dnamlen, stderr, 1);
+ Exit(1);
+ }
+ if (RDEV_STATFN(fp, &sb) != 0) {
+ if (errno == ENOENT) /* a sym link to nowhere? */
+ continue;
+
+# if defined(WARNDEVACCESS)
+ if (!Fwarn) {
+ int errno_save = errno;
+
+ (void) fprintf(stderr, "%s: can't stat ", Pn);
+ safestrprt(fp, stderr, 0);
+ (void) fprintf(stderr, ": %s\n", strerror(errno_save));
+ }
+# endif /* defined(WARNDEVACCESS) */
+
+ continue;
+ }
+ /*
+ * If it's a subdirectory, stack its name for later
+ * processing.
+ */
+ if ((sb.st_mode & S_IFMT) == S_IFDIR) {
+ (void) stkdir(fp);
+ continue;
+ }
+ if ((sb.st_mode & S_IFMT) == S_IFCHR) {
+
+ /*
+ * Save character device information in Devtp[].
+ */
+ if (i >= Ndev) {
+ Ndev += DEVINCR;
+ if (!Devtp)
+ Devtp = (struct l_dev *)malloc(
+ (MALLOC_S)(sizeof(struct l_dev)*Ndev));
+ else
+ Devtp = (struct l_dev *)realloc((MALLOC_P *)Devtp,
+ (MALLOC_S)(sizeof(struct l_dev)*Ndev));
+ if (!Devtp) {
+ (void) fprintf(stderr,
+ "%s: no space for character device\n", Pn);
+ Exit(1);
+ }
+ }
+ Devtp[i].rdev = RDEV_EXPDEV(sb.st_rdev);
+ Devtp[i].inode = (INODETYPE)sb.st_ino;
+ if (!(Devtp[i].name = mkstrcpy(fp, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: no space for device name: ", Pn);
+ safestrprt(fp, stderr, 1);
+ Exit(1);
+ }
+ Devtp[i].v = 0;
+
+# if defined(HAS_STD_CLONE) && HAS_STD_CLONE==1
+ if (HAVECLONEMAJ && GET_MAJ_DEV(Devtp[i].rdev) == CLONEMAJ)
+ {
+
+ /*
+ * Record clone device information.
+ */
+ if (!(c = (struct clone *)malloc(sizeof(struct clone))))
+ {
+ (void) fprintf(stderr,
+ "%s: no space for clone device: ", Pn);
+ safestrprt(fp, stderr, 1);
+ Exit(1);
+ }
+ c->dx = i;
+ c->next = Clone;
+ Clone = c;
+ }
+# endif /* defined(HAS_STD_CLONE) && HAS_STD_CLONE==1 */
+
+ i++;
+ }
+
+# if defined(HASBLKDEV)
+ if ((sb.st_mode & S_IFMT) == S_IFBLK) {
+
+ /*
+ * Save block device information in BDevtp[].
+ */
+ if (j >= BNdev) {
+ BNdev += DEVINCR;
+ if (!BDevtp)
+ BDevtp = (struct l_dev *)malloc(
+ (MALLOC_S)(sizeof(struct l_dev)*BNdev));
+ else
+ BDevtp = (struct l_dev *)realloc((MALLOC_P *)BDevtp,
+ (MALLOC_S)(sizeof(struct l_dev)*BNdev));
+ if (!BDevtp) {
+ (void) fprintf(stderr,
+ "%s: no space for block device\n", Pn);
+ Exit(1);
+ }
+ }
+ BDevtp[j].name = fp;
+ fp = (char *)NULL;
+ BDevtp[j].inode = (INODETYPE)sb.st_ino;
+ BDevtp[j].rdev = RDEV_EXPDEV(sb.st_rdev);
+ BDevtp[j].v = 0;
+ j++;
+ }
+# endif /* defined(HASBLKDEV) */
+
+ }
+ (void) CloseDir(dfp);
+ }
+/*
+ * Free any allocated space.
+ */
+ if (!Dstk) {
+ (void) free((FREE_P *)Dstk);
+ Dstk = (char **)NULL;
+ }
+ if (fp)
+ (void) free((FREE_P *)fp);
+ if (path)
+ (void) free((FREE_P *)path);
+
+# if defined(HASBLKDEV)
+/*
+ * Reduce the BDevtp[] (optional) and Devtp[] tables to their minimum
+ * sizes; allocate and build sort pointer lists; and sort the tables by
+ * device number.
+ */
+ if (BNdev) {
+ if (BNdev > j) {
+ BNdev = j;
+ BDevtp = (struct l_dev *)realloc((MALLOC_P *)BDevtp,
+ (MALLOC_S)(sizeof(struct l_dev) * BNdev));
+ }
+ if (!(BSdev = (struct l_dev **)malloc(
+ (MALLOC_S)(sizeof(struct l_dev *) * BNdev))))
+ {
+ (void) fprintf(stderr,
+ "%s: no space for block device sort pointers\n", Pn);
+ Exit(1);
+ }
+ for (j = 0; j < BNdev; j++) {
+ BSdev[j] = &BDevtp[j];
+ }
+ (void) qsort((QSORT_P *)BSdev, (size_t)BNdev,
+ (size_t)sizeof(struct l_dev *), compdev);
+ BNdev = rmdupdev(&BSdev, BNdev, "block");
+ }
+
+# if !defined(NOWARNBLKDEV)
+ else {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: no block devices found\n", Pn);
+ }
+# endif /* !defined(NOWARNBLKDEV) */
+# endif /* defined(HASBLKDEV) */
+
+ if (Ndev) {
+ if (Ndev > i) {
+ Ndev = i;
+ Devtp = (struct l_dev *)realloc((MALLOC_P *)Devtp,
+ (MALLOC_S)(sizeof(struct l_dev) * Ndev));
+ }
+ if (!(Sdev = (struct l_dev **)malloc(
+ (MALLOC_S)(sizeof(struct l_dev *) * Ndev))))
+ {
+ (void) fprintf(stderr,
+ "%s: no space for character device sort pointers\n", Pn);
+ Exit(1);
+ }
+ for (i = 0; i < Ndev; i++) {
+ Sdev[i] = &Devtp[i];
+ }
+ (void) qsort((QSORT_P *)Sdev, (size_t)Ndev,
+ (size_t)sizeof(struct l_dev *), compdev);
+ Ndev = rmdupdev(&Sdev, Ndev, "char");
+ } else {
+ (void) fprintf(stderr, "%s: no character devices found\n", Pn);
+ Exit(1);
+ }
+
+# if defined(HASDCACHE)
+/*
+ * Write device cache file, as required.
+ */
+ if (DCstate == 1 || (DCstate == 3 && dcrd))
+ write_dcache();
+# endif /* defined(HASDCACHE) */
+
+}
+
+
+# if defined(HASDCACHE)
+/*
+ * rereaddev() - reread device names, modes and types
+ */
+
+void
+rereaddev()
+{
+ (void) clr_devtab();
+
+# if defined(DCACHE_CLR)
+ (void) DCACHE_CLR();
+# endif /* defined(DCACHE_CLR) */
+
+ readdev(1);
+ DCunsafe = 0;
+}
+#endif /* defined(HASDCACHE) */
+
+
+/*
+ * rmdupdev() - remove duplicate (major/minor/inode) devices
+ */
+
+static int
+rmdupdev(dp, n, nm)
+ struct l_dev ***dp; /* device table pointers address */
+ int n; /* number of pointers */
+ char *nm; /* device table name for error message */
+{
+
+# if defined(HAS_STD_CLONE) && HAS_STD_CLONE==1
+ struct clone *c, *cp;
+# endif /* defined(HAS_STD_CLONE) && HAS_STD_CLONE==1 */
+
+ int i, j, k;
+ struct l_dev **p;
+
+ for (i = j = 0, p = *dp; i < n ;) {
+ for (k = i + 1; k < n; k++) {
+ if (p[i]->rdev != p[k]->rdev || p[i]->inode != p[k]->inode)
+ break;
+
+# if defined(HAS_STD_CLONE) && HAS_STD_CLONE==1
+ /*
+ * See if we're deleting a duplicate clone device. If so,
+ * delete its clone table entry.
+ */
+ for (c = Clone, cp = (struct clone *)NULL;
+ c;
+ cp = c, c = c->next)
+ {
+ if (&Devtp[c->dx] != p[k])
+ continue;
+ if (!cp)
+ Clone = c->next;
+ else
+ cp->next = c->next;
+ (void) free((FREE_P *)c);
+ break;
+ }
+# endif /* defined(HAS_STD_CLONE) && HAS_STD_CLONE==1 */
+
+ }
+ if (i != j)
+ p[j] = p[i];
+ j++;
+ i = k;
+ }
+ if (n == j)
+ return(n);
+ if (!(*dp = (struct l_dev **)realloc((MALLOC_P *)*dp,
+ (MALLOC_S)(j * sizeof(struct l_dev *)))))
+ {
+ (void) fprintf(stderr, "%s: can't realloc %s device pointers\n",
+ Pn, nm);
+ Exit(1);
+ }
+ return(j);
+}
+
+
+# if defined(HASDCACHE)
+/*
+ * vfy_dev() - verify a device table entry (usually when DCunsafe == 1)
+ *
+ * Note: rereads entire device table when an entry can't be verified.
+ */
+
+int
+vfy_dev(dp)
+ struct l_dev *dp; /* device table pointer */
+{
+ struct stat sb;
+
+ if (!DCunsafe || dp->v)
+ return(1);
+ if (RDEV_STATFN(dp->name, &sb) != 0
+ || dp->rdev != RDEV_EXPDEV(sb.st_rdev)
+ || dp->inode != sb.st_ino) {
+ (void) rereaddev();
+ return(0);
+ }
+ dp->v = 1;
+ return(1);
+}
+# endif /* defined(HASDCACHE) */
+#else /* !defined(USE_LIB_READDEV) */
+char rdev_d1[] = "d"; char *rdev_d2 = rdev_d1;
+#endif /* defined(USE_LIB_READDEV) */
--- /dev/null
+/*
+ * regex.c -- POSIX-conformant regular expression function set for the lsof
+ * library
+ *
+ * This file is used when the UNIX dialect does not have a POSIX-conformant
+ * regular expression function set. In that case USE_LIB_REGEX is defined.
+ *
+ * V. Abell <abe@purdue.edu>
+ * Purdue University Computing Center
+ */
+
+
+/*
+ * Copyright 2000 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * This software has been adapted from snprintf.c in sendmail 8.9.3. It
+ * is subject to the sendmail copyright statements listed below, and the
+ * sendmail licensing terms stated in the sendmail LICENSE file comment
+ * section of this file.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#ifdef USE_LIB_REGEX
+/*
+ * This file comes from GLIBC 2.2. It is used when the UNIX dialect does not
+ * have a POSIX-conformant regular expression function set. In that case
+ * USE_LIB_REGEX is defined.
+ */
+
+/* Extended regular expression matching and search library,
+ version 0.12.
+ (Implements POSIX draft P1003.2/D11.2, except for some of the
+ internationalization features.)
+ Copyright (C) 1993-1999, 2000 Free Software Foundation, Inc.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Library General Public License as
+ published by the Free Software Foundation; either version 2 of the
+ License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Library General Public License for more details.
+
+ You should have received a copy of the GNU Library General Public
+ License along with the GNU C Library; see the file COPYING.LIB. If not,
+ write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ Boston, MA 02111-1307, USA. */
+
+/* AIX requires this to be the first thing in the file. */
+#if defined _AIX && !defined REGEX_MALLOC
+ #pragma alloca
+#endif
+
+#undef _GNU_SOURCE
+#define _GNU_SOURCE
+
+#ifdef HAVE_CONFIG_H
+# include <config.h>
+#endif
+
+#ifndef PARAMS
+# if defined __GNUC__ || (defined __STDC__ && __STDC__)
+# define PARAMS(args) args
+# else
+# define PARAMS(args) ()
+# endif /* GCC. */
+#endif /* Not PARAMS. */
+
+#if defined STDC_HEADERS && !defined emacs
+# include <stddef.h>
+#else
+/* We need this for `regex.h', and perhaps for the Emacs include files. */
+# include <sys/types.h>
+#endif
+
+#define WIDE_CHAR_SUPPORT (HAVE_WCTYPE_H && HAVE_WCHAR_H && HAVE_BTOWC)
+
+/* For platform which support the ISO C amendement 1 functionality we
+ support user defined character classes. */
+#if defined _LIBC || WIDE_CHAR_SUPPORT
+/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
+# include <wchar.h>
+# include <wctype.h>
+#endif
+
+#ifdef _LIBC
+/* We have to keep the namespace clean. */
+# define regfree(preg) __regfree (preg)
+# define regexec(pr, st, nm, pm, ef) __regexec (pr, st, nm, pm, ef)
+# define regcomp(preg, pattern, cflags) __regcomp (preg, pattern, cflags)
+# define regerror(errcode, preg, errbuf, errbuf_size) \
+ __regerror(errcode, preg, errbuf, errbuf_size)
+# define re_set_registers(bu, re, nu, st, en) \
+ __re_set_registers (bu, re, nu, st, en)
+# define re_match_2(bufp, string1, size1, string2, size2, pos, regs, stop) \
+ __re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
+# define re_match(bufp, string, size, pos, regs) \
+ __re_match (bufp, string, size, pos, regs)
+# define re_search(bufp, string, size, startpos, range, regs) \
+ __re_search (bufp, string, size, startpos, range, regs)
+# define re_compile_pattern(pattern, length, bufp) \
+ __re_compile_pattern (pattern, length, bufp)
+# define re_set_syntax(syntax) __re_set_syntax (syntax)
+# define re_search_2(bufp, st1, s1, st2, s2, startpos, range, regs, stop) \
+ __re_search_2 (bufp, st1, s1, st2, s2, startpos, range, regs, stop)
+# define re_compile_fastmap(bufp) __re_compile_fastmap (bufp)
+
+# define btowc __btowc
+
+/* We are also using some library internals. */
+# include <locale/localeinfo.h>
+# include <locale/elem-hash.h>
+# include <langinfo.h>
+#endif
+
+/* This is for other GNU distributions with internationalized messages. */
+#if HAVE_LIBINTL_H || defined _LIBC
+# include <libintl.h>
+# ifdef _LIBC
+# undef gettext
+# define gettext(msgid) __dcgettext ("libc", msgid, LC_MESSAGES)
+# endif
+#else
+# define gettext(msgid) (msgid)
+#endif
+
+#ifndef gettext_noop
+/* This define is so xgettext can find the internationalizable
+ strings. */
+# define gettext_noop(String) String
+#endif
+
+/* The `emacs' switch turns on certain matching commands
+ that make sense only in Emacs. */
+#ifdef emacs
+
+# include "lisp.h"
+# include "buffer.h"
+# include "syntax.h"
+
+#else /* not emacs */
+
+/* If we are not linking with Emacs proper,
+ we can't use the relocating allocator
+ even if config.h says that we can. */
+# undef REL_ALLOC
+
+# if defined STDC_HEADERS || defined _LIBC
+# include <stdlib.h>
+# else
+char *malloc ();
+char *realloc ();
+# endif
+
+/* When used in Emacs's lib-src, we need to get bzero and bcopy somehow.
+ If nothing else has been done, use the method below. */
+# ifdef INHIBIT_STRING_HEADER
+# if !(defined HAVE_BZERO && defined HAVE_BCOPY)
+# if !defined bzero && !defined bcopy
+# undef INHIBIT_STRING_HEADER
+# endif
+# endif
+# endif
+
+/* This is the normal way of making sure we have a bcopy and a bzero.
+ This is used in most programs--a few other programs avoid this
+ by defining INHIBIT_STRING_HEADER. */
+# ifndef INHIBIT_STRING_HEADER
+# if defined HAVE_STRING_H || defined STDC_HEADERS || defined _LIBC
+# include <string.h>
+# ifndef bzero
+# ifndef _LIBC
+# define bzero(s, n) (memset (s, '\0', n), (s))
+# else
+# define bzero(s, n) __bzero (s, n)
+# endif
+# endif
+# else
+# include <strings.h>
+# ifndef memcmp
+# define memcmp(s1, s2, n) bcmp (s1, s2, n)
+# endif
+# ifndef memcpy
+# define memcpy(d, s, n) (bcopy (s, d, n), (d))
+# endif
+# endif
+# endif
+
+/* Define the syntax stuff for \<, \>, etc. */
+
+/* This must be nonzero for the wordchar and notwordchar pattern
+ commands in re_match_2. */
+# ifndef Sword
+# define Sword 1
+# endif
+
+# ifdef SWITCH_ENUM_BUG
+# define SWITCH_ENUM_CAST(x) ((int)(x))
+# else
+# define SWITCH_ENUM_CAST(x) (x)
+# endif
+
+#endif /* not emacs */
+
+#if defined _LIBC || HAVE_LIMITS_H
+# include <limits.h>
+#endif
+
+#ifndef MB_LEN_MAX
+# define MB_LEN_MAX 1
+#endif
+\f
+/* Get the interface, including the syntax bits. */
+/* Disabled by V. Abell on January 29, 2001: #include <regex.h> */
+#include "../regex.h"
+
+/* isalpha etc. are used for the character classes. */
+#include <ctype.h>
+
+/* Jim Meyering writes:
+
+ "... Some ctype macros are valid only for character codes that
+ isascii says are ASCII (SGI's IRIX-4.0.5 is one such system --when
+ using /bin/cc or gcc but without giving an ansi option). So, all
+ ctype uses should be through macros like ISPRINT... If
+ STDC_HEADERS is defined, then autoconf has verified that the ctype
+ macros don't need to be guarded with references to isascii. ...
+ Defining isascii to 1 should let any compiler worth its salt
+ eliminate the && through constant folding."
+ Solaris defines some of these symbols so we must undefine them first. */
+
+#undef ISASCII
+#if defined STDC_HEADERS || (!defined isascii && !defined HAVE_ISASCII)
+# define ISASCII(c) 1
+#else
+# define ISASCII(c) isascii(c)
+#endif
+
+#ifdef isblank
+# define ISBLANK(c) (ISASCII (c) && isblank (c))
+#else
+# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
+#endif
+#ifdef isgraph
+# define ISGRAPH(c) (ISASCII (c) && isgraph (c))
+#else
+# define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c))
+#endif
+
+#undef ISPRINT
+#define ISPRINT(c) (ISASCII (c) && isprint (c))
+#define ISDIGIT(c) (ISASCII (c) && isdigit (c))
+#define ISALNUM(c) (ISASCII (c) && isalnum (c))
+#define ISALPHA(c) (ISASCII (c) && isalpha (c))
+#define ISCNTRL(c) (ISASCII (c) && iscntrl (c))
+#define ISLOWER(c) (ISASCII (c) && islower (c))
+#define ISPUNCT(c) (ISASCII (c) && ispunct (c))
+#define ISSPACE(c) (ISASCII (c) && isspace (c))
+#define ISUPPER(c) (ISASCII (c) && isupper (c))
+#define ISXDIGIT(c) (ISASCII (c) && isxdigit (c))
+
+#ifdef _tolower
+# define TOLOWER(c) _tolower(c)
+#else
+# define TOLOWER(c) tolower(c)
+#endif
+
+#ifndef NULL
+# define NULL (void *)0
+#endif
+
+/* We remove any previous definition of `SIGN_EXTEND_CHAR',
+ since ours (we hope) works properly with all combinations of
+ machines, compilers, `char' and `unsigned char' argument types.
+ (Per Bothner suggested the basic approach.) */
+#undef SIGN_EXTEND_CHAR
+#if __STDC__
+# define SIGN_EXTEND_CHAR(c) ((signed char) (c))
+#else /* not __STDC__ */
+/* As in Harbison and Steele. */
+# define SIGN_EXTEND_CHAR(c) ((((unsigned char) (c)) ^ 128) - 128)
+#endif
+\f
+#ifndef emacs
+/* How many characters in the character set. */
+# define CHAR_SET_SIZE 256
+
+# ifdef SYNTAX_TABLE
+
+extern char *re_syntax_table;
+
+# else /* not SYNTAX_TABLE */
+
+static char re_syntax_table[CHAR_SET_SIZE];
+
+static void
+init_syntax_once ()
+{
+ register int c;
+ static int done = 0;
+
+ if (done)
+ return;
+ bzero (re_syntax_table, sizeof re_syntax_table);
+
+ for (c = 0; c < CHAR_SET_SIZE; ++c)
+ if (ISALNUM (c))
+ re_syntax_table[c] = Sword;
+
+ re_syntax_table['_'] = Sword;
+
+ done = 1;
+}
+
+# endif /* not SYNTAX_TABLE */
+
+# define SYNTAX(c) re_syntax_table[(unsigned char) (c)]
+
+#endif /* emacs */
+\f
+/* Should we use malloc or alloca? If REGEX_MALLOC is not defined, we
+ use `alloca' instead of `malloc'. This is because using malloc in
+ re_search* or re_match* could cause memory leaks when C-g is used in
+ Emacs; also, malloc is slower and causes storage fragmentation. On
+ the other hand, malloc is more portable, and easier to debug.
+
+ Because we sometimes use alloca, some routines have to be macros,
+ not functions -- `alloca'-allocated space disappears at the end of the
+ function it is called in. */
+
+#ifdef REGEX_MALLOC
+
+# define REGEX_ALLOCATE malloc
+# define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize)
+# define REGEX_FREE free
+
+#else /* not REGEX_MALLOC */
+
+/* Emacs already defines alloca, sometimes. */
+# ifndef alloca
+
+/* Make alloca work the best possible way. */
+# ifdef __GNUC__
+# define alloca __builtin_alloca
+# else /* not __GNUC__ */
+# if HAVE_ALLOCA_H
+# include <alloca.h>
+# endif /* HAVE_ALLOCA_H */
+# endif /* not __GNUC__ */
+
+# endif /* not alloca */
+
+# define REGEX_ALLOCATE alloca
+
+/* Assumes a `char *destination' variable. */
+# define REGEX_REALLOCATE(source, osize, nsize) \
+ (destination = (char *) alloca (nsize), \
+ memcpy (destination, source, osize))
+
+/* No need to do anything to free, after alloca. */
+# define REGEX_FREE(arg) ((void)0) /* Do nothing! But inhibit gcc warning. */
+
+#endif /* not REGEX_MALLOC */
+
+/* Define how to allocate the failure stack. */
+
+#if defined REL_ALLOC && defined REGEX_MALLOC
+
+# define REGEX_ALLOCATE_STACK(size) \
+ r_alloc (&failure_stack_ptr, (size))
+# define REGEX_REALLOCATE_STACK(source, osize, nsize) \
+ r_re_alloc (&failure_stack_ptr, (nsize))
+# define REGEX_FREE_STACK(ptr) \
+ r_alloc_free (&failure_stack_ptr)
+
+#else /* not using relocating allocator */
+
+# ifdef REGEX_MALLOC
+
+# define REGEX_ALLOCATE_STACK malloc
+# define REGEX_REALLOCATE_STACK(source, osize, nsize) realloc (source, nsize)
+# define REGEX_FREE_STACK free
+
+# else /* not REGEX_MALLOC */
+
+# define REGEX_ALLOCATE_STACK alloca
+
+# define REGEX_REALLOCATE_STACK(source, osize, nsize) \
+ REGEX_REALLOCATE (source, osize, nsize)
+/* No need to explicitly free anything. */
+# define REGEX_FREE_STACK(arg)
+
+# endif /* not REGEX_MALLOC */
+#endif /* not using relocating allocator */
+
+
+/* True if `size1' is non-NULL and PTR is pointing anywhere inside
+ `string1' or just past its end. This works if PTR is NULL, which is
+ a good thing. */
+#define FIRST_STRING_P(ptr) \
+ (size1 && string1 <= (ptr) && (ptr) <= string1 + size1)
+
+/* (Re)Allocate N items of type T using malloc, or fail. */
+#define TALLOC(n, t) ((t *) malloc ((n) * sizeof (t)))
+#define RETALLOC(addr, n, t) ((addr) = (t *) realloc (addr, (n) * sizeof (t)))
+#define RETALLOC_IF(addr, n, t) \
+ if (addr) RETALLOC((addr), (n), t); else (addr) = TALLOC ((n), t)
+#define REGEX_TALLOC(n, t) ((t *) REGEX_ALLOCATE ((n) * sizeof (t)))
+
+#define BYTEWIDTH 8 /* In bits. */
+
+#define STREQ(s1, s2) ((strcmp (s1, s2) == 0))
+
+#undef MAX
+#undef MIN
+#define MAX(a, b) ((a) > (b) ? (a) : (b))
+#define MIN(a, b) ((a) < (b) ? (a) : (b))
+
+typedef char boolean;
+#define false 0
+#define true 1
+
+static int re_match_2_internal PARAMS ((struct re_pattern_buffer *bufp,
+ const char *string1, int size1,
+ const char *string2, int size2,
+ int pos,
+ struct re_registers *regs,
+ int stop));
+\f
+/* These are the command codes that appear in compiled regular
+ expressions. Some opcodes are followed by argument bytes. A
+ command code can specify any interpretation whatsoever for its
+ arguments. Zero bytes may appear in the compiled regular expression. */
+
+typedef enum
+{
+ no_op = 0,
+
+ /* Succeed right away--no more backtracking. */
+ succeed,
+
+ /* Followed by one byte giving n, then by n literal bytes. */
+ exactn,
+
+ /* Matches any (more or less) character. */
+ anychar,
+
+ /* Matches any one char belonging to specified set. First
+ following byte is number of bitmap bytes. Then come bytes
+ for a bitmap saying which chars are in. Bits in each byte
+ are ordered low-bit-first. A character is in the set if its
+ bit is 1. A character too large to have a bit in the map is
+ automatically not in the set. */
+ charset,
+
+ /* Same parameters as charset, but match any character that is
+ not one of those specified. */
+ charset_not,
+
+ /* Start remembering the text that is matched, for storing in a
+ register. Followed by one byte with the register number, in
+ the range 0 to one less than the pattern buffer's re_nsub
+ field. Then followed by one byte with the number of groups
+ inner to this one. (This last has to be part of the
+ start_memory only because we need it in the on_failure_jump
+ of re_match_2.) */
+ start_memory,
+
+ /* Stop remembering the text that is matched and store it in a
+ memory register. Followed by one byte with the register
+ number, in the range 0 to one less than `re_nsub' in the
+ pattern buffer, and one byte with the number of inner groups,
+ just like `start_memory'. (We need the number of inner
+ groups here because we don't have any easy way of finding the
+ corresponding start_memory when we're at a stop_memory.) */
+ stop_memory,
+
+ /* Match a duplicate of something remembered. Followed by one
+ byte containing the register number. */
+ duplicate,
+
+ /* Fail unless at beginning of line. */
+ begline,
+
+ /* Fail unless at end of line. */
+ endline,
+
+ /* Succeeds if at beginning of buffer (if emacs) or at beginning
+ of string to be matched (if not). */
+ begbuf,
+
+ /* Analogously, for end of buffer/string. */
+ endbuf,
+
+ /* Followed by two byte relative address to which to jump. */
+ jump,
+
+ /* Same as jump, but marks the end of an alternative. */
+ jump_past_alt,
+
+ /* Followed by two-byte relative address of place to resume at
+ in case of failure. */
+ on_failure_jump,
+
+ /* Like on_failure_jump, but pushes a placeholder instead of the
+ current string position when executed. */
+ on_failure_keep_string_jump,
+
+ /* Throw away latest failure point and then jump to following
+ two-byte relative address. */
+ pop_failure_jump,
+
+ /* Change to pop_failure_jump if know won't have to backtrack to
+ match; otherwise change to jump. This is used to jump
+ back to the beginning of a repeat. If what follows this jump
+ clearly won't match what the repeat does, such that we can be
+ sure that there is no use backtracking out of repetitions
+ already matched, then we change it to a pop_failure_jump.
+ Followed by two-byte address. */
+ maybe_pop_jump,
+
+ /* Jump to following two-byte address, and push a dummy failure
+ point. This failure point will be thrown away if an attempt
+ is made to use it for a failure. A `+' construct makes this
+ before the first repeat. Also used as an intermediary kind
+ of jump when compiling an alternative. */
+ dummy_failure_jump,
+
+ /* Push a dummy failure point and continue. Used at the end of
+ alternatives. */
+ push_dummy_failure,
+
+ /* Followed by two-byte relative address and two-byte number n.
+ After matching N times, jump to the address upon failure. */
+ succeed_n,
+
+ /* Followed by two-byte relative address, and two-byte number n.
+ Jump to the address N times, then fail. */
+ jump_n,
+
+ /* Set the following two-byte relative address to the
+ subsequent two-byte number. The address *includes* the two
+ bytes of number. */
+ set_number_at,
+
+ wordchar, /* Matches any word-constituent character. */
+ notwordchar, /* Matches any char that is not a word-constituent. */
+
+ wordbeg, /* Succeeds if at word beginning. */
+ wordend, /* Succeeds if at word end. */
+
+ wordbound, /* Succeeds if at a word boundary. */
+ notwordbound /* Succeeds if not at a word boundary. */
+
+#ifdef emacs
+ ,before_dot, /* Succeeds if before point. */
+ at_dot, /* Succeeds if at point. */
+ after_dot, /* Succeeds if after point. */
+
+ /* Matches any character whose syntax is specified. Followed by
+ a byte which contains a syntax code, e.g., Sword. */
+ syntaxspec,
+
+ /* Matches any character whose syntax is not that specified. */
+ notsyntaxspec
+#endif /* emacs */
+} re_opcode_t;
+\f
+/* Common operations on the compiled pattern. */
+
+/* Store NUMBER in two contiguous bytes starting at DESTINATION. */
+
+#define STORE_NUMBER(destination, number) \
+ do { \
+ (destination)[0] = (number) & 0377; \
+ (destination)[1] = (number) >> 8; \
+ } while (0)
+
+/* Same as STORE_NUMBER, except increment DESTINATION to
+ the byte after where the number is stored. Therefore, DESTINATION
+ must be an lvalue. */
+
+#define STORE_NUMBER_AND_INCR(destination, number) \
+ do { \
+ STORE_NUMBER (destination, number); \
+ (destination) += 2; \
+ } while (0)
+
+/* Put into DESTINATION a number stored in two contiguous bytes starting
+ at SOURCE. */
+
+#define EXTRACT_NUMBER(destination, source) \
+ do { \
+ (destination) = *(source) & 0377; \
+ (destination) += SIGN_EXTEND_CHAR (*((source) + 1)) << 8; \
+ } while (0)
+
+#ifdef DEBUG
+static void extract_number _RE_ARGS ((int *dest, unsigned char *source));
+static void
+extract_number (dest, source)
+ int *dest;
+ unsigned char *source;
+{
+ int temp = SIGN_EXTEND_CHAR (*(source + 1));
+ *dest = *source & 0377;
+ *dest += temp << 8;
+}
+
+# ifndef EXTRACT_MACROS /* To debug the macros. */
+# undef EXTRACT_NUMBER
+# define EXTRACT_NUMBER(dest, src) extract_number (&dest, src)
+# endif /* not EXTRACT_MACROS */
+
+#endif /* DEBUG */
+
+/* Same as EXTRACT_NUMBER, except increment SOURCE to after the number.
+ SOURCE must be an lvalue. */
+
+#define EXTRACT_NUMBER_AND_INCR(destination, source) \
+ do { \
+ EXTRACT_NUMBER (destination, source); \
+ (source) += 2; \
+ } while (0)
+
+#ifdef DEBUG
+static void extract_number_and_incr _RE_ARGS ((int *destination,
+ unsigned char **source));
+static void
+extract_number_and_incr (destination, source)
+ int *destination;
+ unsigned char **source;
+{
+ extract_number (destination, *source);
+ *source += 2;
+}
+
+# ifndef EXTRACT_MACROS
+# undef EXTRACT_NUMBER_AND_INCR
+# define EXTRACT_NUMBER_AND_INCR(dest, src) \
+ extract_number_and_incr (&dest, &src)
+# endif /* not EXTRACT_MACROS */
+
+#endif /* DEBUG */
+\f
+/* If DEBUG is defined, Regex prints many voluminous messages about what
+ it is doing (if the variable `debug' is nonzero). If linked with the
+ main program in `iregex.c', you can enter patterns and strings
+ interactively. And if linked with the main program in `main.c' and
+ the other test files, you can run the already-written tests. */
+
+#ifdef DEBUG
+
+/* We use standard I/O for debugging. */
+# include <stdio.h>
+
+/* It is useful to test things that ``must'' be true when debugging. */
+# include <assert.h>
+
+static int debug;
+
+# define DEBUG_STATEMENT(e) e
+# define DEBUG_PRINT1(x) if (debug) printf (x)
+# define DEBUG_PRINT2(x1, x2) if (debug) printf (x1, x2)
+# define DEBUG_PRINT3(x1, x2, x3) if (debug) printf (x1, x2, x3)
+# define DEBUG_PRINT4(x1, x2, x3, x4) if (debug) printf (x1, x2, x3, x4)
+# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \
+ if (debug) print_partial_compiled_pattern (s, e)
+# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) \
+ if (debug) print_double_string (w, s1, sz1, s2, sz2)
+
+
+/* Print the fastmap in human-readable form. */
+
+void
+print_fastmap (fastmap)
+ char *fastmap;
+{
+ unsigned was_a_range = 0;
+ unsigned i = 0;
+
+ while (i < (1 << BYTEWIDTH))
+ {
+ if (fastmap[i++])
+ {
+ was_a_range = 0;
+ putchar (i - 1);
+ while (i < (1 << BYTEWIDTH) && fastmap[i])
+ {
+ was_a_range = 1;
+ i++;
+ }
+ if (was_a_range)
+ {
+ printf ("-");
+ putchar (i - 1);
+ }
+ }
+ }
+ putchar ('\n');
+}
+
+
+/* Print a compiled pattern string in human-readable form, starting at
+ the START pointer into it and ending just before the pointer END. */
+
+void
+print_partial_compiled_pattern (start, end)
+ unsigned char *start;
+ unsigned char *end;
+{
+ int mcnt, mcnt2;
+ unsigned char *p1;
+ unsigned char *p = start;
+ unsigned char *pend = end;
+
+ if (start == NULL)
+ {
+ printf ("(null)\n");
+ return;
+ }
+
+ /* Loop over pattern commands. */
+ while (p < pend)
+ {
+#ifdef _LIBC
+ printf ("%t:\t", p - start);
+#else
+ printf ("%ld:\t", (long int) (p - start));
+#endif
+
+ switch ((re_opcode_t) *p++)
+ {
+ case no_op:
+ printf ("/no_op");
+ break;
+
+ case exactn:
+ mcnt = *p++;
+ printf ("/exactn/%d", mcnt);
+ do
+ {
+ putchar ('/');
+ putchar (*p++);
+ }
+ while (--mcnt);
+ break;
+
+ case start_memory:
+ mcnt = *p++;
+ printf ("/start_memory/%d/%d", mcnt, *p++);
+ break;
+
+ case stop_memory:
+ mcnt = *p++;
+ printf ("/stop_memory/%d/%d", mcnt, *p++);
+ break;
+
+ case duplicate:
+ printf ("/duplicate/%d", *p++);
+ break;
+
+ case anychar:
+ printf ("/anychar");
+ break;
+
+ case charset:
+ case charset_not:
+ {
+ register int c, last = -100;
+ register int in_range = 0;
+
+ printf ("/charset [%s",
+ (re_opcode_t) *(p - 1) == charset_not ? "^" : "");
+
+ assert (p + *p < pend);
+
+ for (c = 0; c < 256; c++)
+ if (c / 8 < *p
+ && (p[1 + (c/8)] & (1 << (c % 8))))
+ {
+ /* Are we starting a range? */
+ if (last + 1 == c && ! in_range)
+ {
+ putchar ('-');
+ in_range = 1;
+ }
+ /* Have we broken a range? */
+ else if (last + 1 != c && in_range)
+ {
+ putchar (last);
+ in_range = 0;
+ }
+
+ if (! in_range)
+ putchar (c);
+
+ last = c;
+ }
+
+ if (in_range)
+ putchar (last);
+
+ putchar (']');
+
+ p += 1 + *p;
+ }
+ break;
+
+ case begline:
+ printf ("/begline");
+ break;
+
+ case endline:
+ printf ("/endline");
+ break;
+
+ case on_failure_jump:
+ extract_number_and_incr (&mcnt, &p);
+#ifdef _LIBC
+ printf ("/on_failure_jump to %t", p + mcnt - start);
+#else
+ printf ("/on_failure_jump to %ld", (long int) (p + mcnt - start));
+#endif
+ break;
+
+ case on_failure_keep_string_jump:
+ extract_number_and_incr (&mcnt, &p);
+#ifdef _LIBC
+ printf ("/on_failure_keep_string_jump to %t", p + mcnt - start);
+#else
+ printf ("/on_failure_keep_string_jump to %ld",
+ (long int) (p + mcnt - start));
+#endif
+ break;
+
+ case dummy_failure_jump:
+ extract_number_and_incr (&mcnt, &p);
+#ifdef _LIBC
+ printf ("/dummy_failure_jump to %t", p + mcnt - start);
+#else
+ printf ("/dummy_failure_jump to %ld", (long int) (p + mcnt - start));
+#endif
+ break;
+
+ case push_dummy_failure:
+ printf ("/push_dummy_failure");
+ break;
+
+ case maybe_pop_jump:
+ extract_number_and_incr (&mcnt, &p);
+#ifdef _LIBC
+ printf ("/maybe_pop_jump to %t", p + mcnt - start);
+#else
+ printf ("/maybe_pop_jump to %ld", (long int) (p + mcnt - start));
+#endif
+ break;
+
+ case pop_failure_jump:
+ extract_number_and_incr (&mcnt, &p);
+#ifdef _LIBC
+ printf ("/pop_failure_jump to %t", p + mcnt - start);
+#else
+ printf ("/pop_failure_jump to %ld", (long int) (p + mcnt - start));
+#endif
+ break;
+
+ case jump_past_alt:
+ extract_number_and_incr (&mcnt, &p);
+#ifdef _LIBC
+ printf ("/jump_past_alt to %t", p + mcnt - start);
+#else
+ printf ("/jump_past_alt to %ld", (long int) (p + mcnt - start));
+#endif
+ break;
+
+ case jump:
+ extract_number_and_incr (&mcnt, &p);
+#ifdef _LIBC
+ printf ("/jump to %t", p + mcnt - start);
+#else
+ printf ("/jump to %ld", (long int) (p + mcnt - start));
+#endif
+ break;
+
+ case succeed_n:
+ extract_number_and_incr (&mcnt, &p);
+ p1 = p + mcnt;
+ extract_number_and_incr (&mcnt2, &p);
+#ifdef _LIBC
+ printf ("/succeed_n to %t, %d times", p1 - start, mcnt2);
+#else
+ printf ("/succeed_n to %ld, %d times",
+ (long int) (p1 - start), mcnt2);
+#endif
+ break;
+
+ case jump_n:
+ extract_number_and_incr (&mcnt, &p);
+ p1 = p + mcnt;
+ extract_number_and_incr (&mcnt2, &p);
+ printf ("/jump_n to %d, %d times", p1 - start, mcnt2);
+ break;
+
+ case set_number_at:
+ extract_number_and_incr (&mcnt, &p);
+ p1 = p + mcnt;
+ extract_number_and_incr (&mcnt2, &p);
+#ifdef _LIBC
+ printf ("/set_number_at location %t to %d", p1 - start, mcnt2);
+#else
+ printf ("/set_number_at location %ld to %d",
+ (long int) (p1 - start), mcnt2);
+#endif
+ break;
+
+ case wordbound:
+ printf ("/wordbound");
+ break;
+
+ case notwordbound:
+ printf ("/notwordbound");
+ break;
+
+ case wordbeg:
+ printf ("/wordbeg");
+ break;
+
+ case wordend:
+ printf ("/wordend");
+
+# ifdef emacs
+ case before_dot:
+ printf ("/before_dot");
+ break;
+
+ case at_dot:
+ printf ("/at_dot");
+ break;
+
+ case after_dot:
+ printf ("/after_dot");
+ break;
+
+ case syntaxspec:
+ printf ("/syntaxspec");
+ mcnt = *p++;
+ printf ("/%d", mcnt);
+ break;
+
+ case notsyntaxspec:
+ printf ("/notsyntaxspec");
+ mcnt = *p++;
+ printf ("/%d", mcnt);
+ break;
+# endif /* emacs */
+
+ case wordchar:
+ printf ("/wordchar");
+ break;
+
+ case notwordchar:
+ printf ("/notwordchar");
+ break;
+
+ case begbuf:
+ printf ("/begbuf");
+ break;
+
+ case endbuf:
+ printf ("/endbuf");
+ break;
+
+ default:
+ printf ("?%d", *(p-1));
+ }
+
+ putchar ('\n');
+ }
+
+#ifdef _LIBC
+ printf ("%t:\tend of pattern.\n", p - start);
+#else
+ printf ("%ld:\tend of pattern.\n", (long int) (p - start));
+#endif
+}
+
+
+void
+print_compiled_pattern (bufp)
+ struct re_pattern_buffer *bufp;
+{
+ unsigned char *buffer = bufp->buffer;
+
+ print_partial_compiled_pattern (buffer, buffer + bufp->used);
+ printf ("%ld bytes used/%ld bytes allocated.\n",
+ bufp->used, bufp->allocated);
+
+ if (bufp->fastmap_accurate && bufp->fastmap)
+ {
+ printf ("fastmap: ");
+ print_fastmap (bufp->fastmap);
+ }
+
+#ifdef _LIBC
+ printf ("re_nsub: %Zd\t", bufp->re_nsub);
+#else
+ printf ("re_nsub: %ld\t", (long int) bufp->re_nsub);
+#endif
+ printf ("regs_alloc: %d\t", bufp->regs_allocated);
+ printf ("can_be_null: %d\t", bufp->can_be_null);
+ printf ("newline_anchor: %d\n", bufp->newline_anchor);
+ printf ("no_sub: %d\t", bufp->no_sub);
+ printf ("not_bol: %d\t", bufp->not_bol);
+ printf ("not_eol: %d\t", bufp->not_eol);
+ printf ("syntax: %lx\n", bufp->syntax);
+ /* Perhaps we should print the translate table? */
+}
+
+
+void
+print_double_string (where, string1, size1, string2, size2)
+ const char *where;
+ const char *string1;
+ const char *string2;
+ int size1;
+ int size2;
+{
+ int this_char;
+
+ if (where == NULL)
+ printf ("(null)");
+ else
+ {
+ if (FIRST_STRING_P (where))
+ {
+ for (this_char = where - string1; this_char < size1; this_char++)
+ putchar (string1[this_char]);
+
+ where = string2;
+ }
+
+ for (this_char = where - string2; this_char < size2; this_char++)
+ putchar (string2[this_char]);
+ }
+}
+
+void
+printchar (c)
+ int c;
+{
+ putc (c, stderr);
+}
+
+#else /* not DEBUG */
+
+# undef assert
+# define assert(e)
+
+# define DEBUG_STATEMENT(e)
+# define DEBUG_PRINT1(x)
+# define DEBUG_PRINT2(x1, x2)
+# define DEBUG_PRINT3(x1, x2, x3)
+# define DEBUG_PRINT4(x1, x2, x3, x4)
+# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
+# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
+
+#endif /* not DEBUG */
+\f
+/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can
+ also be assigned to arbitrarily: each pattern buffer stores its own
+ syntax, so it can be changed between regex compilations. */
+/* This has no initializer because initialized variables in Emacs
+ become read-only after dumping. */
+reg_syntax_t re_syntax_options;
+
+
+/* Specify the precise syntax of regexps for compilation. This provides
+ for compatibility for various utilities which historically have
+ different, incompatible syntaxes.
+
+ The argument SYNTAX is a bit mask comprised of the various bits
+ defined in regex.h. We return the old syntax. */
+
+reg_syntax_t
+re_set_syntax (syntax)
+ reg_syntax_t syntax;
+{
+ reg_syntax_t ret = re_syntax_options;
+
+ re_syntax_options = syntax;
+#ifdef DEBUG
+ if (syntax & RE_DEBUG)
+ debug = 1;
+ else if (debug) /* was on but now is not */
+ debug = 0;
+#endif /* DEBUG */
+ return ret;
+}
+#ifdef _LIBC
+weak_alias (__re_set_syntax, re_set_syntax)
+#endif
+\f
+/* This table gives an error message for each of the error codes listed
+ in regex.h. Obviously the order here has to be same as there.
+ POSIX doesn't require that we do anything for REG_NOERROR,
+ but why not be nice? */
+
+static const char re_error_msgid[] =
+ {
+#define REG_NOERROR_IDX 0
+ gettext_noop ("Success") /* REG_NOERROR */
+ "\0"
+#define REG_NOMATCH_IDX (REG_NOERROR_IDX + sizeof "Success")
+ gettext_noop ("No match") /* REG_NOMATCH */
+ "\0"
+#define REG_BADPAT_IDX (REG_NOMATCH_IDX + sizeof "No match")
+ gettext_noop ("Invalid regular expression") /* REG_BADPAT */
+ "\0"
+#define REG_ECOLLATE_IDX (REG_BADPAT_IDX + sizeof "Invalid regular expression")
+ gettext_noop ("Invalid collation character") /* REG_ECOLLATE */
+ "\0"
+#define REG_ECTYPE_IDX (REG_ECOLLATE_IDX + sizeof "Invalid collation character")
+ gettext_noop ("Invalid character class name") /* REG_ECTYPE */
+ "\0"
+#define REG_EESCAPE_IDX (REG_ECTYPE_IDX + sizeof "Invalid character class name")
+ gettext_noop ("Trailing backslash") /* REG_EESCAPE */
+ "\0"
+#define REG_ESUBREG_IDX (REG_EESCAPE_IDX + sizeof "Trailing backslash")
+ gettext_noop ("Invalid back reference") /* REG_ESUBREG */
+ "\0"
+#define REG_EBRACK_IDX (REG_ESUBREG_IDX + sizeof "Invalid back reference")
+ gettext_noop ("Unmatched [ or [^") /* REG_EBRACK */
+ "\0"
+#define REG_EPAREN_IDX (REG_EBRACK_IDX + sizeof "Unmatched [ or [^")
+ gettext_noop ("Unmatched ( or \\(") /* REG_EPAREN */
+ "\0"
+#define REG_EBRACE_IDX (REG_EPAREN_IDX + sizeof "Unmatched ( or \\(")
+ gettext_noop ("Unmatched \\{") /* REG_EBRACE */
+ "\0"
+#define REG_BADBR_IDX (REG_EBRACE_IDX + sizeof "Unmatched \\{")
+ gettext_noop ("Invalid content of \\{\\}") /* REG_BADBR */
+ "\0"
+#define REG_ERANGE_IDX (REG_BADBR_IDX + sizeof "Invalid content of \\{\\}")
+ gettext_noop ("Invalid range end") /* REG_ERANGE */
+ "\0"
+#define REG_ESPACE_IDX (REG_ERANGE_IDX + sizeof "Invalid range end")
+ gettext_noop ("Memory exhausted") /* REG_ESPACE */
+ "\0"
+#define REG_BADRPT_IDX (REG_ESPACE_IDX + sizeof "Memory exhausted")
+ gettext_noop ("Invalid preceding regular expression") /* REG_BADRPT */
+ "\0"
+#define REG_EEND_IDX (REG_BADRPT_IDX + sizeof "Invalid preceding regular expression")
+ gettext_noop ("Premature end of regular expression") /* REG_EEND */
+ "\0"
+#define REG_ESIZE_IDX (REG_EEND_IDX + sizeof "Premature end of regular expression")
+ gettext_noop ("Regular expression too big") /* REG_ESIZE */
+ "\0"
+#define REG_ERPAREN_IDX (REG_ESIZE_IDX + sizeof "Regular expression too big")
+ gettext_noop ("Unmatched ) or \\)") /* REG_ERPAREN */
+ };
+
+static const size_t re_error_msgid_idx[] =
+ {
+ REG_NOERROR_IDX,
+ REG_NOMATCH_IDX,
+ REG_BADPAT_IDX,
+ REG_ECOLLATE_IDX,
+ REG_ECTYPE_IDX,
+ REG_EESCAPE_IDX,
+ REG_ESUBREG_IDX,
+ REG_EBRACK_IDX,
+ REG_EPAREN_IDX,
+ REG_EBRACE_IDX,
+ REG_BADBR_IDX,
+ REG_ERANGE_IDX,
+ REG_ESPACE_IDX,
+ REG_BADRPT_IDX,
+ REG_EEND_IDX,
+ REG_ESIZE_IDX,
+ REG_ERPAREN_IDX
+ };
+\f
+/* Avoiding alloca during matching, to placate r_alloc. */
+
+/* Define MATCH_MAY_ALLOCATE unless we need to make sure that the
+ searching and matching functions should not call alloca. On some
+ systems, alloca is implemented in terms of malloc, and if we're
+ using the relocating allocator routines, then malloc could cause a
+ relocation, which might (if the strings being searched are in the
+ ralloc heap) shift the data out from underneath the regexp
+ routines.
+
+ Here's another reason to avoid allocation: Emacs
+ processes input from X in a signal handler; processing X input may
+ call malloc; if input arrives while a matching routine is calling
+ malloc, then we're scrod. But Emacs can't just block input while
+ calling matching routines; then we don't notice interrupts when
+ they come in. So, Emacs blocks input around all regexp calls
+ except the matching calls, which it leaves unprotected, in the
+ faith that they will not malloc. */
+
+/* Normally, this is fine. */
+#define MATCH_MAY_ALLOCATE
+
+/* When using GNU C, we are not REALLY using the C alloca, no matter
+ what config.h may say. So don't take precautions for it. */
+#ifdef __GNUC__
+# undef C_ALLOCA
+#endif
+
+/* The match routines may not allocate if (1) they would do it with malloc
+ and (2) it's not safe for them to use malloc.
+ Note that if REL_ALLOC is defined, matching would not use malloc for the
+ failure stack, but we would still use it for the register vectors;
+ so REL_ALLOC should not affect this. */
+#if (defined C_ALLOCA || defined REGEX_MALLOC) && defined emacs
+# undef MATCH_MAY_ALLOCATE
+#endif
+
+\f
+/* Failure stack declarations and macros; both re_compile_fastmap and
+ re_match_2 use a failure stack. These have to be macros because of
+ REGEX_ALLOCATE_STACK. */
+
+
+/* Number of failure points for which to initially allocate space
+ when matching. If this number is exceeded, we allocate more
+ space, so it is not a hard limit. */
+#ifndef INIT_FAILURE_ALLOC
+# define INIT_FAILURE_ALLOC 5
+#endif
+
+/* Roughly the maximum number of failure points on the stack. Would be
+ exactly that if always used MAX_FAILURE_ITEMS items each time we failed.
+ This is a variable only so users of regex can assign to it; we never
+ change it ourselves. */
+
+#ifdef INT_IS_16BIT
+
+# if defined MATCH_MAY_ALLOCATE
+/* 4400 was enough to cause a crash on Alpha OSF/1,
+ whose default stack limit is 2mb. */
+long int re_max_failures = 4000;
+# else
+long int re_max_failures = 2000;
+# endif
+
+union fail_stack_elt
+{
+ unsigned char *pointer;
+ long int integer;
+};
+
+typedef union fail_stack_elt fail_stack_elt_t;
+
+typedef struct
+{
+ fail_stack_elt_t *stack;
+ unsigned long int size;
+ unsigned long int avail; /* Offset of next open position. */
+} fail_stack_type;
+
+#else /* not INT_IS_16BIT */
+
+# if defined MATCH_MAY_ALLOCATE
+/* 4400 was enough to cause a crash on Alpha OSF/1,
+ whose default stack limit is 2mb. */
+int re_max_failures = 4000;
+# else
+int re_max_failures = 2000;
+# endif
+
+union fail_stack_elt
+{
+ unsigned char *pointer;
+ int integer;
+};
+
+typedef union fail_stack_elt fail_stack_elt_t;
+
+typedef struct
+{
+ fail_stack_elt_t *stack;
+ unsigned size;
+ unsigned avail; /* Offset of next open position. */
+} fail_stack_type;
+
+#endif /* INT_IS_16BIT */
+
+#define FAIL_STACK_EMPTY() (fail_stack.avail == 0)
+#define FAIL_STACK_PTR_EMPTY() (fail_stack_ptr->avail == 0)
+#define FAIL_STACK_FULL() (fail_stack.avail == fail_stack.size)
+
+
+/* Define macros to initialize and free the failure stack.
+ Do `return -2' if the alloc fails. */
+
+#ifdef MATCH_MAY_ALLOCATE
+# define INIT_FAIL_STACK() \
+ do { \
+ fail_stack.stack = (fail_stack_elt_t *) \
+ REGEX_ALLOCATE_STACK (INIT_FAILURE_ALLOC * sizeof (fail_stack_elt_t)); \
+ \
+ if (fail_stack.stack == NULL) \
+ return -2; \
+ \
+ fail_stack.size = INIT_FAILURE_ALLOC; \
+ fail_stack.avail = 0; \
+ } while (0)
+
+# define RESET_FAIL_STACK() REGEX_FREE_STACK (fail_stack.stack)
+#else
+# define INIT_FAIL_STACK() \
+ do { \
+ fail_stack.avail = 0; \
+ } while (0)
+
+# define RESET_FAIL_STACK()
+#endif
+
+
+/* Double the size of FAIL_STACK, up to approximately `re_max_failures' items.
+
+ Return 1 if succeeds, and 0 if either ran out of memory
+ allocating space for it or it was already too large.
+
+ REGEX_REALLOCATE_STACK requires `destination' be declared. */
+
+#define DOUBLE_FAIL_STACK(fail_stack) \
+ ((fail_stack).size > (unsigned) (re_max_failures * MAX_FAILURE_ITEMS) \
+ ? 0 \
+ : ((fail_stack).stack = (fail_stack_elt_t *) \
+ REGEX_REALLOCATE_STACK ((fail_stack).stack, \
+ (fail_stack).size * sizeof (fail_stack_elt_t), \
+ ((fail_stack).size << 1) * sizeof (fail_stack_elt_t)), \
+ \
+ (fail_stack).stack == NULL \
+ ? 0 \
+ : ((fail_stack).size <<= 1, \
+ 1)))
+
+
+/* Push pointer POINTER on FAIL_STACK.
+ Return 1 if was able to do so and 0 if ran out of memory allocating
+ space to do so. */
+#define PUSH_PATTERN_OP(POINTER, FAIL_STACK) \
+ ((FAIL_STACK_FULL () \
+ && !DOUBLE_FAIL_STACK (FAIL_STACK)) \
+ ? 0 \
+ : ((FAIL_STACK).stack[(FAIL_STACK).avail++].pointer = POINTER, \
+ 1))
+
+/* Push a pointer value onto the failure stack.
+ Assumes the variable `fail_stack'. Probably should only
+ be called from within `PUSH_FAILURE_POINT'. */
+#define PUSH_FAILURE_POINTER(item) \
+ fail_stack.stack[fail_stack.avail++].pointer = (unsigned char *) (item)
+
+/* This pushes an integer-valued item onto the failure stack.
+ Assumes the variable `fail_stack'. Probably should only
+ be called from within `PUSH_FAILURE_POINT'. */
+#define PUSH_FAILURE_INT(item) \
+ fail_stack.stack[fail_stack.avail++].integer = (item)
+
+/* Push a fail_stack_elt_t value onto the failure stack.
+ Assumes the variable `fail_stack'. Probably should only
+ be called from within `PUSH_FAILURE_POINT'. */
+#define PUSH_FAILURE_ELT(item) \
+ fail_stack.stack[fail_stack.avail++] = (item)
+
+/* These three POP... operations complement the three PUSH... operations.
+ All assume that `fail_stack' is nonempty. */
+#define POP_FAILURE_POINTER() fail_stack.stack[--fail_stack.avail].pointer
+#define POP_FAILURE_INT() fail_stack.stack[--fail_stack.avail].integer
+#define POP_FAILURE_ELT() fail_stack.stack[--fail_stack.avail]
+
+/* Used to omit pushing failure point id's when we're not debugging. */
+#ifdef DEBUG
+# define DEBUG_PUSH PUSH_FAILURE_INT
+# define DEBUG_POP(item_addr) *(item_addr) = POP_FAILURE_INT ()
+#else
+# define DEBUG_PUSH(item)
+# define DEBUG_POP(item_addr)
+#endif
+
+
+/* Push the information about the state we will need
+ if we ever fail back to it.
+
+ Requires variables fail_stack, regstart, regend, reg_info, and
+ num_regs_pushed be declared. DOUBLE_FAIL_STACK requires `destination'
+ be declared.
+
+ Does `return FAILURE_CODE' if runs out of memory. */
+
+#define PUSH_FAILURE_POINT(pattern_place, string_place, failure_code) \
+ do { \
+ char *destination; \
+ /* Must be int, so when we don't save any registers, the arithmetic \
+ of 0 + -1 isn't done as unsigned. */ \
+ /* Can't be int, since there is not a shred of a guarantee that int \
+ is wide enough to hold a value of something to which pointer can \
+ be assigned */ \
+ active_reg_t this_reg; \
+ \
+ DEBUG_STATEMENT (failure_id++); \
+ DEBUG_STATEMENT (nfailure_points_pushed++); \
+ DEBUG_PRINT2 ("\nPUSH_FAILURE_POINT #%u:\n", failure_id); \
+ DEBUG_PRINT2 (" Before push, next avail: %d\n", (fail_stack).avail);\
+ DEBUG_PRINT2 (" size: %d\n", (fail_stack).size);\
+ \
+ DEBUG_PRINT2 (" slots needed: %ld\n", NUM_FAILURE_ITEMS); \
+ DEBUG_PRINT2 (" available: %d\n", REMAINING_AVAIL_SLOTS); \
+ \
+ /* Ensure we have enough space allocated for what we will push. */ \
+ while (REMAINING_AVAIL_SLOTS < NUM_FAILURE_ITEMS) \
+ { \
+ if (!DOUBLE_FAIL_STACK (fail_stack)) \
+ return failure_code; \
+ \
+ DEBUG_PRINT2 ("\n Doubled stack; size now: %d\n", \
+ (fail_stack).size); \
+ DEBUG_PRINT2 (" slots available: %d\n", REMAINING_AVAIL_SLOTS);\
+ } \
+ \
+ /* Push the info, starting with the registers. */ \
+ DEBUG_PRINT1 ("\n"); \
+ \
+ if (1) \
+ for (this_reg = lowest_active_reg; this_reg <= highest_active_reg; \
+ this_reg++) \
+ { \
+ DEBUG_PRINT2 (" Pushing reg: %lu\n", this_reg); \
+ DEBUG_STATEMENT (num_regs_pushed++); \
+ \
+ DEBUG_PRINT2 (" start: %p\n", regstart[this_reg]); \
+ PUSH_FAILURE_POINTER (regstart[this_reg]); \
+ \
+ DEBUG_PRINT2 (" end: %p\n", regend[this_reg]); \
+ PUSH_FAILURE_POINTER (regend[this_reg]); \
+ \
+ DEBUG_PRINT2 (" info: %p\n ", \
+ reg_info[this_reg].word.pointer); \
+ DEBUG_PRINT2 (" match_null=%d", \
+ REG_MATCH_NULL_STRING_P (reg_info[this_reg])); \
+ DEBUG_PRINT2 (" active=%d", IS_ACTIVE (reg_info[this_reg])); \
+ DEBUG_PRINT2 (" matched_something=%d", \
+ MATCHED_SOMETHING (reg_info[this_reg])); \
+ DEBUG_PRINT2 (" ever_matched=%d", \
+ EVER_MATCHED_SOMETHING (reg_info[this_reg])); \
+ DEBUG_PRINT1 ("\n"); \
+ PUSH_FAILURE_ELT (reg_info[this_reg].word); \
+ } \
+ \
+ DEBUG_PRINT2 (" Pushing low active reg: %ld\n", lowest_active_reg);\
+ PUSH_FAILURE_INT (lowest_active_reg); \
+ \
+ DEBUG_PRINT2 (" Pushing high active reg: %ld\n", highest_active_reg);\
+ PUSH_FAILURE_INT (highest_active_reg); \
+ \
+ DEBUG_PRINT2 (" Pushing pattern %p:\n", pattern_place); \
+ DEBUG_PRINT_COMPILED_PATTERN (bufp, pattern_place, pend); \
+ PUSH_FAILURE_POINTER (pattern_place); \
+ \
+ DEBUG_PRINT2 (" Pushing string %p: `", string_place); \
+ DEBUG_PRINT_DOUBLE_STRING (string_place, string1, size1, string2, \
+ size2); \
+ DEBUG_PRINT1 ("'\n"); \
+ PUSH_FAILURE_POINTER (string_place); \
+ \
+ DEBUG_PRINT2 (" Pushing failure id: %u\n", failure_id); \
+ DEBUG_PUSH (failure_id); \
+ } while (0)
+
+/* This is the number of items that are pushed and popped on the stack
+ for each register. */
+#define NUM_REG_ITEMS 3
+
+/* Individual items aside from the registers. */
+#ifdef DEBUG
+# define NUM_NONREG_ITEMS 5 /* Includes failure point id. */
+#else
+# define NUM_NONREG_ITEMS 4
+#endif
+
+/* We push at most this many items on the stack. */
+/* We used to use (num_regs - 1), which is the number of registers
+ this regexp will save; but that was changed to 5
+ to avoid stack overflow for a regexp with lots of parens. */
+#define MAX_FAILURE_ITEMS (5 * NUM_REG_ITEMS + NUM_NONREG_ITEMS)
+
+/* We actually push this many items. */
+#define NUM_FAILURE_ITEMS \
+ (((0 \
+ ? 0 : highest_active_reg - lowest_active_reg + 1) \
+ * NUM_REG_ITEMS) \
+ + NUM_NONREG_ITEMS)
+
+/* How many items can still be added to the stack without overflowing it. */
+#define REMAINING_AVAIL_SLOTS ((fail_stack).size - (fail_stack).avail)
+
+
+/* Pops what PUSH_FAIL_STACK pushes.
+
+ We restore into the parameters, all of which should be lvalues:
+ STR -- the saved data position.
+ PAT -- the saved pattern position.
+ LOW_REG, HIGH_REG -- the highest and lowest active registers.
+ REGSTART, REGEND -- arrays of string positions.
+ REG_INFO -- array of information about each subexpression.
+
+ Also assumes the variables `fail_stack' and (if debugging), `bufp',
+ `pend', `string1', `size1', `string2', and `size2'. */
+
+#define POP_FAILURE_POINT(str, pat, low_reg, high_reg, regstart, regend, reg_info)\
+{ \
+ DEBUG_STATEMENT (unsigned failure_id;) \
+ active_reg_t this_reg; \
+ const unsigned char *string_temp; \
+ \
+ assert (!FAIL_STACK_EMPTY ()); \
+ \
+ /* Remove failure points and point to how many regs pushed. */ \
+ DEBUG_PRINT1 ("POP_FAILURE_POINT:\n"); \
+ DEBUG_PRINT2 (" Before pop, next avail: %d\n", fail_stack.avail); \
+ DEBUG_PRINT2 (" size: %d\n", fail_stack.size); \
+ \
+ assert (fail_stack.avail >= NUM_NONREG_ITEMS); \
+ \
+ DEBUG_POP (&failure_id); \
+ DEBUG_PRINT2 (" Popping failure id: %u\n", failure_id); \
+ \
+ /* If the saved string location is NULL, it came from an \
+ on_failure_keep_string_jump opcode, and we want to throw away the \
+ saved NULL, thus retaining our current position in the string. */ \
+ string_temp = POP_FAILURE_POINTER (); \
+ if (string_temp != NULL) \
+ str = (const char *) string_temp; \
+ \
+ DEBUG_PRINT2 (" Popping string %p: `", str); \
+ DEBUG_PRINT_DOUBLE_STRING (str, string1, size1, string2, size2); \
+ DEBUG_PRINT1 ("'\n"); \
+ \
+ pat = (unsigned char *) POP_FAILURE_POINTER (); \
+ DEBUG_PRINT2 (" Popping pattern %p:\n", pat); \
+ DEBUG_PRINT_COMPILED_PATTERN (bufp, pat, pend); \
+ \
+ /* Restore register info. */ \
+ high_reg = (active_reg_t) POP_FAILURE_INT (); \
+ DEBUG_PRINT2 (" Popping high active reg: %ld\n", high_reg); \
+ \
+ low_reg = (active_reg_t) POP_FAILURE_INT (); \
+ DEBUG_PRINT2 (" Popping low active reg: %ld\n", low_reg); \
+ \
+ if (1) \
+ for (this_reg = high_reg; this_reg >= low_reg; this_reg--) \
+ { \
+ DEBUG_PRINT2 (" Popping reg: %ld\n", this_reg); \
+ \
+ reg_info[this_reg].word = POP_FAILURE_ELT (); \
+ DEBUG_PRINT2 (" info: %p\n", \
+ reg_info[this_reg].word.pointer); \
+ \
+ regend[this_reg] = (const char *) POP_FAILURE_POINTER (); \
+ DEBUG_PRINT2 (" end: %p\n", regend[this_reg]); \
+ \
+ regstart[this_reg] = (const char *) POP_FAILURE_POINTER (); \
+ DEBUG_PRINT2 (" start: %p\n", regstart[this_reg]); \
+ } \
+ else \
+ { \
+ for (this_reg = highest_active_reg; this_reg > high_reg; this_reg--) \
+ { \
+ reg_info[this_reg].word.integer = 0; \
+ regend[this_reg] = 0; \
+ regstart[this_reg] = 0; \
+ } \
+ highest_active_reg = high_reg; \
+ } \
+ \
+ set_regs_matched_done = 0; \
+ DEBUG_STATEMENT (nfailure_points_popped++); \
+} /* POP_FAILURE_POINT */
+
+
+\f
+/* Structure for per-register (a.k.a. per-group) information.
+ Other register information, such as the
+ starting and ending positions (which are addresses), and the list of
+ inner groups (which is a bits list) are maintained in separate
+ variables.
+
+ We are making a (strictly speaking) nonportable assumption here: that
+ the compiler will pack our bit fields into something that fits into
+ the type of `word', i.e., is something that fits into one item on the
+ failure stack. */
+
+
+/* Declarations and macros for re_match_2. */
+
+typedef union
+{
+ fail_stack_elt_t word;
+ struct
+ {
+ /* This field is one if this group can match the empty string,
+ zero if not. If not yet determined, `MATCH_NULL_UNSET_VALUE'. */
+#define MATCH_NULL_UNSET_VALUE 3
+ unsigned match_null_string_p : 2;
+ unsigned is_active : 1;
+ unsigned matched_something : 1;
+ unsigned ever_matched_something : 1;
+ } bits;
+} register_info_type;
+
+#define REG_MATCH_NULL_STRING_P(R) ((R).bits.match_null_string_p)
+#define IS_ACTIVE(R) ((R).bits.is_active)
+#define MATCHED_SOMETHING(R) ((R).bits.matched_something)
+#define EVER_MATCHED_SOMETHING(R) ((R).bits.ever_matched_something)
+
+
+/* Call this when have matched a real character; it sets `matched' flags
+ for the subexpressions which we are currently inside. Also records
+ that those subexprs have matched. */
+#define SET_REGS_MATCHED() \
+ do \
+ { \
+ if (!set_regs_matched_done) \
+ { \
+ active_reg_t r; \
+ set_regs_matched_done = 1; \
+ for (r = lowest_active_reg; r <= highest_active_reg; r++) \
+ { \
+ MATCHED_SOMETHING (reg_info[r]) \
+ = EVER_MATCHED_SOMETHING (reg_info[r]) \
+ = 1; \
+ } \
+ } \
+ } \
+ while (0)
+
+/* Registers are set to a sentinel when they haven't yet matched. */
+static char reg_unset_dummy;
+#define REG_UNSET_VALUE (®_unset_dummy)
+#define REG_UNSET(e) ((e) == REG_UNSET_VALUE)
+\f
+/* Subroutine declarations and macros for regex_compile. */
+
+static reg_errcode_t regex_compile _RE_ARGS ((const char *pattern, size_t size,
+ reg_syntax_t syntax,
+ struct re_pattern_buffer *bufp));
+static void store_op1 _RE_ARGS ((re_opcode_t op, unsigned char *loc, int arg));
+static void store_op2 _RE_ARGS ((re_opcode_t op, unsigned char *loc,
+ int arg1, int arg2));
+static void insert_op1 _RE_ARGS ((re_opcode_t op, unsigned char *loc,
+ int arg, unsigned char *end));
+static void insert_op2 _RE_ARGS ((re_opcode_t op, unsigned char *loc,
+ int arg1, int arg2, unsigned char *end));
+static boolean at_begline_loc_p _RE_ARGS ((const char *pattern, const char *p,
+ reg_syntax_t syntax));
+static boolean at_endline_loc_p _RE_ARGS ((const char *p, const char *pend,
+ reg_syntax_t syntax));
+static reg_errcode_t compile_range _RE_ARGS ((unsigned int range_start,
+ const char **p_ptr,
+ const char *pend,
+ char *translate,
+ reg_syntax_t syntax,
+ unsigned char *b));
+
+/* Fetch the next character in the uncompiled pattern---translating it
+ if necessary. Also cast from a signed character in the constant
+ string passed to us by the user to an unsigned char that we can use
+ as an array index (in, e.g., `translate'). */
+#ifndef PATFETCH
+# define PATFETCH(c) \
+ do {if (p == pend) return REG_EEND; \
+ c = (unsigned char) *p++; \
+ if (translate) c = (unsigned char) translate[c]; \
+ } while (0)
+#endif
+
+/* Fetch the next character in the uncompiled pattern, with no
+ translation. */
+#define PATFETCH_RAW(c) \
+ do {if (p == pend) return REG_EEND; \
+ c = (unsigned char) *p++; \
+ } while (0)
+
+/* Go backwards one character in the pattern. */
+#define PATUNFETCH p--
+
+
+/* If `translate' is non-null, return translate[D], else just D. We
+ cast the subscript to translate because some data is declared as
+ `char *', to avoid warnings when a string constant is passed. But
+ when we use a character as a subscript we must make it unsigned. */
+#ifndef TRANSLATE
+# define TRANSLATE(d) \
+ (translate ? (char) translate[(unsigned char) (d)] : (d))
+#endif
+
+
+/* Macros for outputting the compiled pattern into `buffer'. */
+
+/* If the buffer isn't allocated when it comes in, use this. */
+#define INIT_BUF_SIZE 32
+
+/* Make sure we have at least N more bytes of space in buffer. */
+#define GET_BUFFER_SPACE(n) \
+ while ((unsigned long) (b - bufp->buffer + (n)) > bufp->allocated) \
+ EXTEND_BUFFER ()
+
+/* Make sure we have one more byte of buffer space and then add C to it. */
+#define BUF_PUSH(c) \
+ do { \
+ GET_BUFFER_SPACE (1); \
+ *b++ = (unsigned char) (c); \
+ } while (0)
+
+
+/* Ensure we have two more bytes of buffer space and then append C1 and C2. */
+#define BUF_PUSH_2(c1, c2) \
+ do { \
+ GET_BUFFER_SPACE (2); \
+ *b++ = (unsigned char) (c1); \
+ *b++ = (unsigned char) (c2); \
+ } while (0)
+
+
+/* As with BUF_PUSH_2, except for three bytes. */
+#define BUF_PUSH_3(c1, c2, c3) \
+ do { \
+ GET_BUFFER_SPACE (3); \
+ *b++ = (unsigned char) (c1); \
+ *b++ = (unsigned char) (c2); \
+ *b++ = (unsigned char) (c3); \
+ } while (0)
+
+
+/* Store a jump with opcode OP at LOC to location TO. We store a
+ relative address offset by the three bytes the jump itself occupies. */
+#define STORE_JUMP(op, loc, to) \
+ store_op1 (op, loc, (int) ((to) - (loc) - 3))
+
+/* Likewise, for a two-argument jump. */
+#define STORE_JUMP2(op, loc, to, arg) \
+ store_op2 (op, loc, (int) ((to) - (loc) - 3), arg)
+
+/* Like `STORE_JUMP', but for inserting. Assume `b' is the buffer end. */
+#define INSERT_JUMP(op, loc, to) \
+ insert_op1 (op, loc, (int) ((to) - (loc) - 3), b)
+
+/* Like `STORE_JUMP2', but for inserting. Assume `b' is the buffer end. */
+#define INSERT_JUMP2(op, loc, to, arg) \
+ insert_op2 (op, loc, (int) ((to) - (loc) - 3), arg, b)
+
+
+/* This is not an arbitrary limit: the arguments which represent offsets
+ into the pattern are two bytes long. So if 2^16 bytes turns out to
+ be too small, many things would have to change. */
+/* Any other compiler which, like MSC, has allocation limit below 2^16
+ bytes will have to use approach similar to what was done below for
+ MSC and drop MAX_BUF_SIZE a bit. Otherwise you may end up
+ reallocating to 0 bytes. Such thing is not going to work too well.
+ You have been warned!! */
+#if defined _MSC_VER && !defined WIN32
+/* Microsoft C 16-bit versions limit malloc to approx 65512 bytes.
+ The REALLOC define eliminates a flurry of conversion warnings,
+ but is not required. */
+# define MAX_BUF_SIZE 65500L
+# define REALLOC(p,s) realloc ((p), (size_t) (s))
+#else
+# define MAX_BUF_SIZE (1L << 16)
+# define REALLOC(p,s) realloc ((p), (s))
+#endif
+
+/* Extend the buffer by twice its current size via realloc and
+ reset the pointers that pointed into the old block to point to the
+ correct places in the new one. If extending the buffer results in it
+ being larger than MAX_BUF_SIZE, then flag memory exhausted. */
+#if __BOUNDED_POINTERS__
+# define SET_HIGH_BOUND(P) (__ptrhigh (P) = __ptrlow (P) + bufp->allocated)
+# define MOVE_BUFFER_POINTER(P) \
+ (__ptrlow (P) += incr, SET_HIGH_BOUND (P), __ptrvalue (P) += incr)
+# define ELSE_EXTEND_BUFFER_HIGH_BOUND \
+ else \
+ { \
+ SET_HIGH_BOUND (b); \
+ SET_HIGH_BOUND (begalt); \
+ if (fixup_alt_jump) \
+ SET_HIGH_BOUND (fixup_alt_jump); \
+ if (laststart) \
+ SET_HIGH_BOUND (laststart); \
+ if (pending_exact) \
+ SET_HIGH_BOUND (pending_exact); \
+ }
+#else
+# define MOVE_BUFFER_POINTER(P) (P) += incr
+# define ELSE_EXTEND_BUFFER_HIGH_BOUND
+#endif
+#define EXTEND_BUFFER() \
+ do { \
+ unsigned char *old_buffer = bufp->buffer; \
+ if (bufp->allocated == MAX_BUF_SIZE) \
+ return REG_ESIZE; \
+ bufp->allocated <<= 1; \
+ if (bufp->allocated > MAX_BUF_SIZE) \
+ bufp->allocated = MAX_BUF_SIZE; \
+ bufp->buffer = (unsigned char *) REALLOC (bufp->buffer, bufp->allocated);\
+ if (bufp->buffer == NULL) \
+ return REG_ESPACE; \
+ /* If the buffer moved, move all the pointers into it. */ \
+ if (old_buffer != bufp->buffer) \
+ { \
+ int incr = bufp->buffer - old_buffer; \
+ MOVE_BUFFER_POINTER (b); \
+ MOVE_BUFFER_POINTER (begalt); \
+ if (fixup_alt_jump) \
+ MOVE_BUFFER_POINTER (fixup_alt_jump); \
+ if (laststart) \
+ MOVE_BUFFER_POINTER (laststart); \
+ if (pending_exact) \
+ MOVE_BUFFER_POINTER (pending_exact); \
+ } \
+ ELSE_EXTEND_BUFFER_HIGH_BOUND \
+ } while (0)
+
+
+/* Since we have one byte reserved for the register number argument to
+ {start,stop}_memory, the maximum number of groups we can report
+ things about is what fits in that byte. */
+#define MAX_REGNUM 255
+
+/* But patterns can have more than `MAX_REGNUM' registers. We just
+ ignore the excess. */
+typedef unsigned regnum_t;
+
+
+/* Macros for the compile stack. */
+
+/* Since offsets can go either forwards or backwards, this type needs to
+ be able to hold values from -(MAX_BUF_SIZE - 1) to MAX_BUF_SIZE - 1. */
+/* int may be not enough when sizeof(int) == 2. */
+typedef long pattern_offset_t;
+
+typedef struct
+{
+ pattern_offset_t begalt_offset;
+ pattern_offset_t fixup_alt_jump;
+ pattern_offset_t inner_group_offset;
+ pattern_offset_t laststart_offset;
+ regnum_t regnum;
+} compile_stack_elt_t;
+
+
+typedef struct
+{
+ compile_stack_elt_t *stack;
+ unsigned size;
+ unsigned avail; /* Offset of next open position. */
+} compile_stack_type;
+
+
+#define INIT_COMPILE_STACK_SIZE 32
+
+#define COMPILE_STACK_EMPTY (compile_stack.avail == 0)
+#define COMPILE_STACK_FULL (compile_stack.avail == compile_stack.size)
+
+/* The next available element. */
+#define COMPILE_STACK_TOP (compile_stack.stack[compile_stack.avail])
+
+
+/* Set the bit for character C in a list. */
+#define SET_LIST_BIT(c) \
+ (b[((unsigned char) (c)) / BYTEWIDTH] \
+ |= 1 << (((unsigned char) c) % BYTEWIDTH))
+
+
+/* Get the next unsigned number in the uncompiled pattern. */
+#define GET_UNSIGNED_NUMBER(num) \
+ { if (p != pend) \
+ { \
+ PATFETCH (c); \
+ while ('0' <= c && c <= '9') \
+ { \
+ if (num < 0) \
+ num = 0; \
+ num = num * 10 + c - '0'; \
+ if (p == pend) \
+ break; \
+ PATFETCH (c); \
+ } \
+ } \
+ }
+
+#if defined _LIBC || WIDE_CHAR_SUPPORT
+/* The GNU C library provides support for user-defined character classes
+ and the functions from ISO C amendement 1. */
+# ifdef CHARCLASS_NAME_MAX
+# define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX
+# else
+/* This shouldn't happen but some implementation might still have this
+ problem. Use a reasonable default value. */
+# define CHAR_CLASS_MAX_LENGTH 256
+# endif
+
+# ifdef _LIBC
+# define IS_CHAR_CLASS(string) __wctype (string)
+# else
+# define IS_CHAR_CLASS(string) wctype (string)
+# endif
+#else
+# define CHAR_CLASS_MAX_LENGTH 6 /* Namely, `xdigit'. */
+
+# define IS_CHAR_CLASS(string) \
+ (STREQ (string, "alpha") || STREQ (string, "upper") \
+ || STREQ (string, "lower") || STREQ (string, "digit") \
+ || STREQ (string, "alnum") || STREQ (string, "xdigit") \
+ || STREQ (string, "space") || STREQ (string, "print") \
+ || STREQ (string, "punct") || STREQ (string, "graph") \
+ || STREQ (string, "cntrl") || STREQ (string, "blank"))
+#endif
+\f
+#ifndef MATCH_MAY_ALLOCATE
+
+/* If we cannot allocate large objects within re_match_2_internal,
+ we make the fail stack and register vectors global.
+ The fail stack, we grow to the maximum size when a regexp
+ is compiled.
+ The register vectors, we adjust in size each time we
+ compile a regexp, according to the number of registers it needs. */
+
+static fail_stack_type fail_stack;
+
+/* Size with which the following vectors are currently allocated.
+ That is so we can make them bigger as needed,
+ but never make them smaller. */
+static int regs_allocated_size;
+
+static const char ** regstart, ** regend;
+static const char ** old_regstart, ** old_regend;
+static const char **best_regstart, **best_regend;
+static register_info_type *reg_info;
+static const char **reg_dummy;
+static register_info_type *reg_info_dummy;
+
+/* Make the register vectors big enough for NUM_REGS registers,
+ but don't make them smaller. */
+
+static
+regex_grow_registers (num_regs)
+ int num_regs;
+{
+ if (num_regs > regs_allocated_size)
+ {
+ RETALLOC_IF (regstart, num_regs, const char *);
+ RETALLOC_IF (regend, num_regs, const char *);
+ RETALLOC_IF (old_regstart, num_regs, const char *);
+ RETALLOC_IF (old_regend, num_regs, const char *);
+ RETALLOC_IF (best_regstart, num_regs, const char *);
+ RETALLOC_IF (best_regend, num_regs, const char *);
+ RETALLOC_IF (reg_info, num_regs, register_info_type);
+ RETALLOC_IF (reg_dummy, num_regs, const char *);
+ RETALLOC_IF (reg_info_dummy, num_regs, register_info_type);
+
+ regs_allocated_size = num_regs;
+ }
+}
+
+#endif /* not MATCH_MAY_ALLOCATE */
+\f
+static boolean group_in_compile_stack _RE_ARGS ((compile_stack_type
+ compile_stack,
+ regnum_t regnum));
+
+/* `regex_compile' compiles PATTERN (of length SIZE) according to SYNTAX.
+ Returns one of error codes defined in `regex.h', or zero for success.
+
+ Assumes the `allocated' (and perhaps `buffer') and `translate'
+ fields are set in BUFP on entry.
+
+ If it succeeds, results are put in BUFP (if it returns an error, the
+ contents of BUFP are undefined):
+ `buffer' is the compiled pattern;
+ `syntax' is set to SYNTAX;
+ `used' is set to the length of the compiled pattern;
+ `fastmap_accurate' is zero;
+ `re_nsub' is the number of subexpressions in PATTERN;
+ `not_bol' and `not_eol' are zero;
+
+ The `fastmap' and `newline_anchor' fields are neither
+ examined nor set. */
+
+/* Return, freeing storage we allocated. */
+#define FREE_STACK_RETURN(value) \
+ return (free (compile_stack.stack), value)
+
+static reg_errcode_t
+regex_compile (pattern, size, syntax, bufp)
+ const char *pattern;
+ size_t size;
+ reg_syntax_t syntax;
+ struct re_pattern_buffer *bufp;
+{
+ /* We fetch characters from PATTERN here. Even though PATTERN is
+ `char *' (i.e., signed), we declare these variables as unsigned, so
+ they can be reliably used as array indices. */
+ register unsigned char c, c1;
+
+ /* A random temporary spot in PATTERN. */
+ const char *p1;
+
+ /* Points to the end of the buffer, where we should append. */
+ register unsigned char *b;
+
+ /* Keeps track of unclosed groups. */
+ compile_stack_type compile_stack;
+
+ /* Points to the current (ending) position in the pattern. */
+ const char *p = pattern;
+ const char *pend = pattern + size;
+
+ /* How to translate the characters in the pattern. */
+ RE_TRANSLATE_TYPE translate = bufp->translate;
+
+ /* Address of the count-byte of the most recently inserted `exactn'
+ command. This makes it possible to tell if a new exact-match
+ character can be added to that command or if the character requires
+ a new `exactn' command. */
+ unsigned char *pending_exact = 0;
+
+ /* Address of start of the most recently finished expression.
+ This tells, e.g., postfix * where to find the start of its
+ operand. Reset at the beginning of groups and alternatives. */
+ unsigned char *laststart = 0;
+
+ /* Address of beginning of regexp, or inside of last group. */
+ unsigned char *begalt;
+
+ /* Place in the uncompiled pattern (i.e., the {) to
+ which to go back if the interval is invalid. */
+ const char *beg_interval;
+
+ /* Address of the place where a forward jump should go to the end of
+ the containing expression. Each alternative of an `or' -- except the
+ last -- ends with a forward jump of this sort. */
+ unsigned char *fixup_alt_jump = 0;
+
+ /* Counts open-groups as they are encountered. Remembered for the
+ matching close-group on the compile stack, so the same register
+ number is put in the stop_memory as the start_memory. */
+ regnum_t regnum = 0;
+
+#ifdef DEBUG
+ DEBUG_PRINT1 ("\nCompiling pattern: ");
+ if (debug)
+ {
+ unsigned debug_count;
+
+ for (debug_count = 0; debug_count < size; debug_count++)
+ putchar (pattern[debug_count]);
+ putchar ('\n');
+ }
+#endif /* DEBUG */
+
+ /* Initialize the compile stack. */
+ compile_stack.stack = TALLOC (INIT_COMPILE_STACK_SIZE, compile_stack_elt_t);
+ if (compile_stack.stack == NULL)
+ return REG_ESPACE;
+
+ compile_stack.size = INIT_COMPILE_STACK_SIZE;
+ compile_stack.avail = 0;
+
+ /* Initialize the pattern buffer. */
+ bufp->syntax = syntax;
+ bufp->fastmap_accurate = 0;
+ bufp->not_bol = bufp->not_eol = 0;
+
+ /* Set `used' to zero, so that if we return an error, the pattern
+ printer (for debugging) will think there's no pattern. We reset it
+ at the end. */
+ bufp->used = 0;
+
+ /* Always count groups, whether or not bufp->no_sub is set. */
+ bufp->re_nsub = 0;
+
+#if !defined emacs && !defined SYNTAX_TABLE
+ /* Initialize the syntax table. */
+ init_syntax_once ();
+#endif
+
+ if (bufp->allocated == 0)
+ {
+ if (bufp->buffer)
+ { /* If zero allocated, but buffer is non-null, try to realloc
+ enough space. This loses if buffer's address is bogus, but
+ that is the user's responsibility. */
+ RETALLOC (bufp->buffer, INIT_BUF_SIZE, unsigned char);
+ }
+ else
+ { /* Caller did not allocate a buffer. Do it for them. */
+ bufp->buffer = TALLOC (INIT_BUF_SIZE, unsigned char);
+ }
+ if (!bufp->buffer) FREE_STACK_RETURN (REG_ESPACE);
+
+ bufp->allocated = INIT_BUF_SIZE;
+ }
+
+ begalt = b = bufp->buffer;
+
+ /* Loop through the uncompiled pattern until we're at the end. */
+ while (p != pend)
+ {
+ PATFETCH (c);
+
+ switch (c)
+ {
+ case '^':
+ {
+ if ( /* If at start of pattern, it's an operator. */
+ p == pattern + 1
+ /* If context independent, it's an operator. */
+ || syntax & RE_CONTEXT_INDEP_ANCHORS
+ /* Otherwise, depends on what's come before. */
+ || at_begline_loc_p (pattern, p, syntax))
+ BUF_PUSH (begline);
+ else
+ goto normal_char;
+ }
+ break;
+
+
+ case '$':
+ {
+ if ( /* If at end of pattern, it's an operator. */
+ p == pend
+ /* If context independent, it's an operator. */
+ || syntax & RE_CONTEXT_INDEP_ANCHORS
+ /* Otherwise, depends on what's next. */
+ || at_endline_loc_p (p, pend, syntax))
+ BUF_PUSH (endline);
+ else
+ goto normal_char;
+ }
+ break;
+
+
+ case '+':
+ case '?':
+ if ((syntax & RE_BK_PLUS_QM)
+ || (syntax & RE_LIMITED_OPS))
+ goto normal_char;
+ handle_plus:
+ case '*':
+ /* If there is no previous pattern... */
+ if (!laststart)
+ {
+ if (syntax & RE_CONTEXT_INVALID_OPS)
+ FREE_STACK_RETURN (REG_BADRPT);
+ else if (!(syntax & RE_CONTEXT_INDEP_OPS))
+ goto normal_char;
+ }
+
+ {
+ /* Are we optimizing this jump? */
+ boolean keep_string_p = false;
+
+ /* 1 means zero (many) matches is allowed. */
+ char zero_times_ok = 0, many_times_ok = 0;
+
+ /* If there is a sequence of repetition chars, collapse it
+ down to just one (the right one). We can't combine
+ interval operators with these because of, e.g., `a{2}*',
+ which should only match an even number of `a's. */
+
+ for (;;)
+ {
+ zero_times_ok |= c != '+';
+ many_times_ok |= c != '?';
+
+ if (p == pend)
+ break;
+
+ PATFETCH (c);
+
+ if (c == '*'
+ || (!(syntax & RE_BK_PLUS_QM) && (c == '+' || c == '?')))
+ ;
+
+ else if (syntax & RE_BK_PLUS_QM && c == '\\')
+ {
+ if (p == pend) FREE_STACK_RETURN (REG_EESCAPE);
+
+ PATFETCH (c1);
+ if (!(c1 == '+' || c1 == '?'))
+ {
+ PATUNFETCH;
+ PATUNFETCH;
+ break;
+ }
+
+ c = c1;
+ }
+ else
+ {
+ PATUNFETCH;
+ break;
+ }
+
+ /* If we get here, we found another repeat character. */
+ }
+
+ /* Star, etc. applied to an empty pattern is equivalent
+ to an empty pattern. */
+ if (!laststart)
+ break;
+
+ /* Now we know whether or not zero matches is allowed
+ and also whether or not two or more matches is allowed. */
+ if (many_times_ok)
+ { /* More than one repetition is allowed, so put in at the
+ end a backward relative jump from `b' to before the next
+ jump we're going to put in below (which jumps from
+ laststart to after this jump).
+
+ But if we are at the `*' in the exact sequence `.*\n',
+ insert an unconditional jump backwards to the .,
+ instead of the beginning of the loop. This way we only
+ push a failure point once, instead of every time
+ through the loop. */
+ assert (p - 1 > pattern);
+
+ /* Allocate the space for the jump. */
+ GET_BUFFER_SPACE (3);
+
+ /* We know we are not at the first character of the pattern,
+ because laststart was nonzero. And we've already
+ incremented `p', by the way, to be the character after
+ the `*'. Do we have to do something analogous here
+ for null bytes, because of RE_DOT_NOT_NULL? */
+ if (TRANSLATE (*(p - 2)) == TRANSLATE ('.')
+ && zero_times_ok
+ && p < pend && TRANSLATE (*p) == TRANSLATE ('\n')
+ && !(syntax & RE_DOT_NEWLINE))
+ { /* We have .*\n. */
+ STORE_JUMP (jump, b, laststart);
+ keep_string_p = true;
+ }
+ else
+ /* Anything else. */
+ STORE_JUMP (maybe_pop_jump, b, laststart - 3);
+
+ /* We've added more stuff to the buffer. */
+ b += 3;
+ }
+
+ /* On failure, jump from laststart to b + 3, which will be the
+ end of the buffer after this jump is inserted. */
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (keep_string_p ? on_failure_keep_string_jump
+ : on_failure_jump,
+ laststart, b + 3);
+ pending_exact = 0;
+ b += 3;
+
+ if (!zero_times_ok)
+ {
+ /* At least one repetition is required, so insert a
+ `dummy_failure_jump' before the initial
+ `on_failure_jump' instruction of the loop. This
+ effects a skip over that instruction the first time
+ we hit that loop. */
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (dummy_failure_jump, laststart, laststart + 6);
+ b += 3;
+ }
+ }
+ break;
+
+
+ case '.':
+ laststart = b;
+ BUF_PUSH (anychar);
+ break;
+
+
+ case '[':
+ {
+ boolean had_char_class = false;
+ unsigned int range_start = 0xffffffff;
+
+ if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
+
+ /* Ensure that we have enough space to push a charset: the
+ opcode, the length count, and the bitset; 34 bytes in all. */
+ GET_BUFFER_SPACE (34);
+
+ laststart = b;
+
+ /* We test `*p == '^' twice, instead of using an if
+ statement, so we only need one BUF_PUSH. */
+ BUF_PUSH (*p == '^' ? charset_not : charset);
+ if (*p == '^')
+ p++;
+
+ /* Remember the first position in the bracket expression. */
+ p1 = p;
+
+ /* Push the number of bytes in the bitmap. */
+ BUF_PUSH ((1 << BYTEWIDTH) / BYTEWIDTH);
+
+ /* Clear the whole map. */
+ bzero (b, (1 << BYTEWIDTH) / BYTEWIDTH);
+
+ /* charset_not matches newline according to a syntax bit. */
+ if ((re_opcode_t) b[-2] == charset_not
+ && (syntax & RE_HAT_LISTS_NOT_NEWLINE))
+ SET_LIST_BIT ('\n');
+
+ /* Read in characters and ranges, setting map bits. */
+ for (;;)
+ {
+ if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
+
+ PATFETCH (c);
+
+ /* \ might escape characters inside [...] and [^...]. */
+ if ((syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) && c == '\\')
+ {
+ if (p == pend) FREE_STACK_RETURN (REG_EESCAPE);
+
+ PATFETCH (c1);
+ SET_LIST_BIT (c1);
+ range_start = c1;
+ continue;
+ }
+
+ /* Could be the end of the bracket expression. If it's
+ not (i.e., when the bracket expression is `[]' so
+ far), the ']' character bit gets set way below. */
+ if (c == ']' && p != p1 + 1)
+ break;
+
+ /* Look ahead to see if it's a range when the last thing
+ was a character class. */
+ if (had_char_class && c == '-' && *p != ']')
+ FREE_STACK_RETURN (REG_ERANGE);
+
+ /* Look ahead to see if it's a range when the last thing
+ was a character: if this is a hyphen not at the
+ beginning or the end of a list, then it's the range
+ operator. */
+ if (c == '-'
+ && !(p - 2 >= pattern && p[-2] == '[')
+ && !(p - 3 >= pattern && p[-3] == '[' && p[-2] == '^')
+ && *p != ']')
+ {
+ reg_errcode_t ret
+ = compile_range (range_start, &p, pend, translate,
+ syntax, b);
+ if (ret != REG_NOERROR) FREE_STACK_RETURN (ret);
+ range_start = 0xffffffff;
+ }
+
+ else if (p[0] == '-' && p[1] != ']')
+ { /* This handles ranges made up of characters only. */
+ reg_errcode_t ret;
+
+ /* Move past the `-'. */
+ PATFETCH (c1);
+
+ ret = compile_range (c, &p, pend, translate, syntax, b);
+ if (ret != REG_NOERROR) FREE_STACK_RETURN (ret);
+ range_start = 0xffffffff;
+ }
+
+ /* See if we're at the beginning of a possible character
+ class. */
+
+ else if (syntax & RE_CHAR_CLASSES && c == '[' && *p == ':')
+ { /* Leave room for the null. */
+ char str[CHAR_CLASS_MAX_LENGTH + 1];
+
+ PATFETCH (c);
+ c1 = 0;
+
+ /* If pattern is `[[:'. */
+ if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
+
+ for (;;)
+ {
+ PATFETCH (c);
+ if ((c == ':' && *p == ']') || p == pend)
+ break;
+ if (c1 < CHAR_CLASS_MAX_LENGTH)
+ str[c1++] = c;
+ else
+ /* This is in any case an invalid class name. */
+ str[0] = '\0';
+ }
+ str[c1] = '\0';
+
+ /* If isn't a word bracketed by `[:' and `:]':
+ undo the ending character, the letters, and leave
+ the leading `:' and `[' (but set bits for them). */
+ if (c == ':' && *p == ']')
+ {
+#if defined _LIBC || WIDE_CHAR_SUPPORT
+ boolean is_lower = STREQ (str, "lower");
+ boolean is_upper = STREQ (str, "upper");
+ wctype_t wt;
+ int ch;
+
+ wt = IS_CHAR_CLASS (str);
+ if (wt == 0)
+ FREE_STACK_RETURN (REG_ECTYPE);
+
+ /* Throw away the ] at the end of the character
+ class. */
+ PATFETCH (c);
+
+ if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
+
+ for (ch = 0; ch < 1 << BYTEWIDTH; ++ch)
+ {
+# ifdef _LIBC
+ if (__iswctype (__btowc (ch), wt))
+ SET_LIST_BIT (ch);
+# else
+ if (iswctype (btowc (ch), wt))
+ SET_LIST_BIT (ch);
+# endif
+
+ if (translate && (is_upper || is_lower)
+ && (ISUPPER (ch) || ISLOWER (ch)))
+ SET_LIST_BIT (ch);
+ }
+
+ had_char_class = true;
+#else
+ int ch;
+ boolean is_alnum = STREQ (str, "alnum");
+ boolean is_alpha = STREQ (str, "alpha");
+ boolean is_blank = STREQ (str, "blank");
+ boolean is_cntrl = STREQ (str, "cntrl");
+ boolean is_digit = STREQ (str, "digit");
+ boolean is_graph = STREQ (str, "graph");
+ boolean is_lower = STREQ (str, "lower");
+ boolean is_print = STREQ (str, "print");
+ boolean is_punct = STREQ (str, "punct");
+ boolean is_space = STREQ (str, "space");
+ boolean is_upper = STREQ (str, "upper");
+ boolean is_xdigit = STREQ (str, "xdigit");
+
+ if (!IS_CHAR_CLASS (str))
+ FREE_STACK_RETURN (REG_ECTYPE);
+
+ /* Throw away the ] at the end of the character
+ class. */
+ PATFETCH (c);
+
+ if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
+
+ for (ch = 0; ch < 1 << BYTEWIDTH; ch++)
+ {
+ /* This was split into 3 if's to
+ avoid an arbitrary limit in some compiler. */
+ if ( (is_alnum && ISALNUM (ch))
+ || (is_alpha && ISALPHA (ch))
+ || (is_blank && ISBLANK (ch))
+ || (is_cntrl && ISCNTRL (ch)))
+ SET_LIST_BIT (ch);
+ if ( (is_digit && ISDIGIT (ch))
+ || (is_graph && ISGRAPH (ch))
+ || (is_lower && ISLOWER (ch))
+ || (is_print && ISPRINT (ch)))
+ SET_LIST_BIT (ch);
+ if ( (is_punct && ISPUNCT (ch))
+ || (is_space && ISSPACE (ch))
+ || (is_upper && ISUPPER (ch))
+ || (is_xdigit && ISXDIGIT (ch)))
+ SET_LIST_BIT (ch);
+ if ( translate && (is_upper || is_lower)
+ && (ISUPPER (ch) || ISLOWER (ch)))
+ SET_LIST_BIT (ch);
+ }
+ had_char_class = true;
+#endif /* libc || wctype.h */
+ }
+ else
+ {
+ c1++;
+ while (c1--)
+ PATUNFETCH;
+ SET_LIST_BIT ('[');
+ SET_LIST_BIT (':');
+ range_start = ':';
+ had_char_class = false;
+ }
+ }
+ else if (syntax & RE_CHAR_CLASSES && c == '[' && *p == '=')
+ {
+ unsigned char str[MB_LEN_MAX + 1];
+#ifdef _LIBC
+ uint32_t nrules =
+ _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
+#endif
+
+ PATFETCH (c);
+ c1 = 0;
+
+ /* If pattern is `[[='. */
+ if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
+
+ for (;;)
+ {
+ PATFETCH (c);
+ if ((c == '=' && *p == ']') || p == pend)
+ break;
+ if (c1 < MB_LEN_MAX)
+ str[c1++] = c;
+ else
+ /* This is in any case an invalid class name. */
+ str[0] = '\0';
+ }
+ str[c1] = '\0';
+
+ if (c == '=' && *p == ']' && str[0] != '\0')
+ {
+ /* If we have no collation data we use the default
+ collation in which each character is in a class
+ by itself. It also means that ASCII is the
+ character set and therefore we cannot have character
+ with more than one byte in the multibyte
+ representation. */
+#ifdef _LIBC
+ if (nrules == 0)
+#endif
+ {
+ if (c1 != 1)
+ FREE_STACK_RETURN (REG_ECOLLATE);
+
+ /* Throw away the ] at the end of the equivalence
+ class. */
+ PATFETCH (c);
+
+ /* Set the bit for the character. */
+ SET_LIST_BIT (str[0]);
+ }
+#ifdef _LIBC
+ else
+ {
+ /* Try to match the byte sequence in `str' against
+ those known to the collate implementation.
+ First find out whether the bytes in `str' are
+ actually from exactly one character. */
+ const int32_t *table;
+ const unsigned char *weights;
+ const unsigned char *extra;
+ const int32_t *indirect;
+ int32_t idx;
+ const unsigned char *cp = str;
+ int ch;
+
+ /* This #include defines a local function! */
+# include <locale/weight.h>
+
+ table = (const int32_t *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
+ weights = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_WEIGHTMB);
+ extra = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB);
+ indirect = (const int32_t *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_INDIRECTMB);
+
+ idx = findidx (&cp);
+ if (idx == 0 || cp < str + c1)
+ /* This is no valid character. */
+ FREE_STACK_RETURN (REG_ECOLLATE);
+
+ /* Throw away the ] at the end of the equivalence
+ class. */
+ PATFETCH (c);
+
+ /* Now we have to go throught the whole table
+ and find all characters which have the same
+ first level weight.
+
+ XXX Note that this is not entirely correct.
+ we would have to match multibyte sequences
+ but this is not possible with the current
+ implementation. */
+ for (ch = 1; ch < 256; ++ch)
+ /* XXX This test would have to be changed if we
+ would allow matching multibyte sequences. */
+ if (table[ch] > 0)
+ {
+ int32_t idx2 = table[ch];
+ size_t len = weights[idx2];
+
+ /* Test whether the lenghts match. */
+ if (weights[idx] == len)
+ {
+ /* They do. New compare the bytes of
+ the weight. */
+ size_t cnt = 0;
+
+ while (cnt < len
+ && (weights[idx + 1 + cnt]
+ == weights[idx2 + 1 + cnt]))
+ ++len;
+
+ if (cnt == len)
+ /* They match. Mark the character as
+ acceptable. */
+ SET_LIST_BIT (ch);
+ }
+ }
+ }
+#endif
+ had_char_class = true;
+ }
+ else
+ {
+ c1++;
+ while (c1--)
+ PATUNFETCH;
+ SET_LIST_BIT ('[');
+ SET_LIST_BIT ('=');
+ range_start = '=';
+ had_char_class = false;
+ }
+ }
+ else if (syntax & RE_CHAR_CLASSES && c == '[' && *p == '.')
+ {
+ unsigned char str[128]; /* Should be large enough. */
+#ifdef _LIBC
+ uint32_t nrules =
+ _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
+#endif
+
+ PATFETCH (c);
+ c1 = 0;
+
+ /* If pattern is `[[='. */
+ if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
+
+ for (;;)
+ {
+ PATFETCH (c);
+ if ((c == '.' && *p == ']') || p == pend)
+ break;
+ if (c1 < sizeof (str))
+ str[c1++] = c;
+ else
+ /* This is in any case an invalid class name. */
+ str[0] = '\0';
+ }
+ str[c1] = '\0';
+
+ if (c == '.' && *p == ']' && str[0] != '\0')
+ {
+ /* If we have no collation data we use the default
+ collation in which each character is the name
+ for its own class which contains only the one
+ character. It also means that ASCII is the
+ character set and therefore we cannot have character
+ with more than one byte in the multibyte
+ representation. */
+#ifdef _LIBC
+ if (nrules == 0)
+#endif
+ {
+ if (c1 != 1)
+ FREE_STACK_RETURN (REG_ECOLLATE);
+
+ /* Throw away the ] at the end of the equivalence
+ class. */
+ PATFETCH (c);
+
+ /* Set the bit for the character. */
+ SET_LIST_BIT (str[0]);
+ range_start = ((const unsigned char *) str)[0];
+ }
+#ifdef _LIBC
+ else
+ {
+ /* Try to match the byte sequence in `str' against
+ those known to the collate implementation.
+ First find out whether the bytes in `str' are
+ actually from exactly one character. */
+ int32_t table_size;
+ const int32_t *symb_table;
+ const unsigned char *extra;
+ int32_t idx;
+ int32_t elem;
+ int32_t second;
+ int32_t hash;
+
+ table_size =
+ _NL_CURRENT_WORD (LC_COLLATE,
+ _NL_COLLATE_SYMB_HASH_SIZEMB);
+ symb_table = (const int32_t *)
+ _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_SYMB_TABLEMB);
+ extra = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_SYMB_EXTRAMB);
+
+ /* Locate the character in the hashing table. */
+ hash = elem_hash (str, c1);
+
+ idx = 0;
+ elem = hash % table_size;
+ second = hash % (table_size - 2);
+ while (symb_table[2 * elem] != 0)
+ {
+ /* First compare the hashing value. */
+ if (symb_table[2 * elem] == hash
+ && c1 == extra[symb_table[2 * elem + 1]]
+ && memcmp (str,
+ &extra[symb_table[2 * elem + 1]
+ + 1],
+ c1) == 0)
+ {
+ /* Yep, this is the entry. */
+ idx = symb_table[2 * elem + 1];
+ idx += 1 + extra[idx];
+ break;
+ }
+
+ /* Next entry. */
+ elem += second;
+ }
+
+ if (symb_table[2 * elem] == 0)
+ /* This is no valid character. */
+ FREE_STACK_RETURN (REG_ECOLLATE);
+
+ /* Throw away the ] at the end of the equivalence
+ class. */
+ PATFETCH (c);
+
+ /* Now add the multibyte character(s) we found
+ to the accept list.
+
+ XXX Note that this is not entirely correct.
+ we would have to match multibyte sequences
+ but this is not possible with the current
+ implementation. Also, we have to match
+ collating symbols, which expand to more than
+ one file, as a whole and not allow the
+ individual bytes. */
+ c1 = extra[idx++];
+ if (c1 == 1)
+ range_start = extra[idx];
+ while (c1-- > 0)
+ {
+ SET_LIST_BIT (extra[idx]);
+ ++idx;
+ }
+ }
+#endif
+ had_char_class = false;
+ }
+ else
+ {
+ c1++;
+ while (c1--)
+ PATUNFETCH;
+ SET_LIST_BIT ('[');
+ SET_LIST_BIT ('.');
+ range_start = '.';
+ had_char_class = false;
+ }
+ }
+ else
+ {
+ had_char_class = false;
+ SET_LIST_BIT (c);
+ range_start = c;
+ }
+ }
+
+ /* Discard any (non)matching list bytes that are all 0 at the
+ end of the map. Decrease the map-length byte too. */
+ while ((int) b[-1] > 0 && b[b[-1] - 1] == 0)
+ b[-1]--;
+ b += b[-1];
+ }
+ break;
+
+
+ case '(':
+ if (syntax & RE_NO_BK_PARENS)
+ goto handle_open;
+ else
+ goto normal_char;
+
+
+ case ')':
+ if (syntax & RE_NO_BK_PARENS)
+ goto handle_close;
+ else
+ goto normal_char;
+
+
+ case '\n':
+ if (syntax & RE_NEWLINE_ALT)
+ goto handle_alt;
+ else
+ goto normal_char;
+
+
+ case '|':
+ if (syntax & RE_NO_BK_VBAR)
+ goto handle_alt;
+ else
+ goto normal_char;
+
+
+ case '{':
+ if (syntax & RE_INTERVALS && syntax & RE_NO_BK_BRACES)
+ goto handle_interval;
+ else
+ goto normal_char;
+
+
+ case '\\':
+ if (p == pend) FREE_STACK_RETURN (REG_EESCAPE);
+
+ /* Do not translate the character after the \, so that we can
+ distinguish, e.g., \B from \b, even if we normally would
+ translate, e.g., B to b. */
+ PATFETCH_RAW (c);
+
+ switch (c)
+ {
+ case '(':
+ if (syntax & RE_NO_BK_PARENS)
+ goto normal_backslash;
+
+ handle_open:
+ bufp->re_nsub++;
+ regnum++;
+
+ if (COMPILE_STACK_FULL)
+ {
+ RETALLOC (compile_stack.stack, compile_stack.size << 1,
+ compile_stack_elt_t);
+ if (compile_stack.stack == NULL) return REG_ESPACE;
+
+ compile_stack.size <<= 1;
+ }
+
+ /* These are the values to restore when we hit end of this
+ group. They are all relative offsets, so that if the
+ whole pattern moves because of realloc, they will still
+ be valid. */
+ COMPILE_STACK_TOP.begalt_offset = begalt - bufp->buffer;
+ COMPILE_STACK_TOP.fixup_alt_jump
+ = fixup_alt_jump ? fixup_alt_jump - bufp->buffer + 1 : 0;
+ COMPILE_STACK_TOP.laststart_offset = b - bufp->buffer;
+ COMPILE_STACK_TOP.regnum = regnum;
+
+ /* We will eventually replace the 0 with the number of
+ groups inner to this one. But do not push a
+ start_memory for groups beyond the last one we can
+ represent in the compiled pattern. */
+ if (regnum <= MAX_REGNUM)
+ {
+ COMPILE_STACK_TOP.inner_group_offset = b - bufp->buffer + 2;
+ BUF_PUSH_3 (start_memory, regnum, 0);
+ }
+
+ compile_stack.avail++;
+
+ fixup_alt_jump = 0;
+ laststart = 0;
+ begalt = b;
+ /* If we've reached MAX_REGNUM groups, then this open
+ won't actually generate any code, so we'll have to
+ clear pending_exact explicitly. */
+ pending_exact = 0;
+ break;
+
+
+ case ')':
+ if (syntax & RE_NO_BK_PARENS) goto normal_backslash;
+
+ if (COMPILE_STACK_EMPTY)
+ {
+ if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
+ goto normal_backslash;
+ else
+ FREE_STACK_RETURN (REG_ERPAREN);
+ }
+
+ handle_close:
+ if (fixup_alt_jump)
+ { /* Push a dummy failure point at the end of the
+ alternative for a possible future
+ `pop_failure_jump' to pop. See comments at
+ `push_dummy_failure' in `re_match_2'. */
+ BUF_PUSH (push_dummy_failure);
+
+ /* We allocated space for this jump when we assigned
+ to `fixup_alt_jump', in the `handle_alt' case below. */
+ STORE_JUMP (jump_past_alt, fixup_alt_jump, b - 1);
+ }
+
+ /* See similar code for backslashed left paren above. */
+ if (COMPILE_STACK_EMPTY)
+ {
+ if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
+ goto normal_char;
+ else
+ FREE_STACK_RETURN (REG_ERPAREN);
+ }
+
+ /* Since we just checked for an empty stack above, this
+ ``can't happen''. */
+ assert (compile_stack.avail != 0);
+ {
+ /* We don't just want to restore into `regnum', because
+ later groups should continue to be numbered higher,
+ as in `(ab)c(de)' -- the second group is #2. */
+ regnum_t this_group_regnum;
+
+ compile_stack.avail--;
+ begalt = bufp->buffer + COMPILE_STACK_TOP.begalt_offset;
+ fixup_alt_jump
+ = COMPILE_STACK_TOP.fixup_alt_jump
+ ? bufp->buffer + COMPILE_STACK_TOP.fixup_alt_jump - 1
+ : 0;
+ laststart = bufp->buffer + COMPILE_STACK_TOP.laststart_offset;
+ this_group_regnum = COMPILE_STACK_TOP.regnum;
+ /* If we've reached MAX_REGNUM groups, then this open
+ won't actually generate any code, so we'll have to
+ clear pending_exact explicitly. */
+ pending_exact = 0;
+
+ /* We're at the end of the group, so now we know how many
+ groups were inside this one. */
+ if (this_group_regnum <= MAX_REGNUM)
+ {
+ unsigned char *inner_group_loc
+ = bufp->buffer + COMPILE_STACK_TOP.inner_group_offset;
+
+ *inner_group_loc = regnum - this_group_regnum;
+ BUF_PUSH_3 (stop_memory, this_group_regnum,
+ regnum - this_group_regnum);
+ }
+ }
+ break;
+
+
+ case '|': /* `\|'. */
+ if (syntax & RE_LIMITED_OPS || syntax & RE_NO_BK_VBAR)
+ goto normal_backslash;
+ handle_alt:
+ if (syntax & RE_LIMITED_OPS)
+ goto normal_char;
+
+ /* Insert before the previous alternative a jump which
+ jumps to this alternative if the former fails. */
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (on_failure_jump, begalt, b + 6);
+ pending_exact = 0;
+ b += 3;
+
+ /* The alternative before this one has a jump after it
+ which gets executed if it gets matched. Adjust that
+ jump so it will jump to this alternative's analogous
+ jump (put in below, which in turn will jump to the next
+ (if any) alternative's such jump, etc.). The last such
+ jump jumps to the correct final destination. A picture:
+ _____ _____
+ | | | |
+ | v | v
+ a | b | c
+
+ If we are at `b', then fixup_alt_jump right now points to a
+ three-byte space after `a'. We'll put in the jump, set
+ fixup_alt_jump to right after `b', and leave behind three
+ bytes which we'll fill in when we get to after `c'. */
+
+ if (fixup_alt_jump)
+ STORE_JUMP (jump_past_alt, fixup_alt_jump, b);
+
+ /* Mark and leave space for a jump after this alternative,
+ to be filled in later either by next alternative or
+ when know we're at the end of a series of alternatives. */
+ fixup_alt_jump = b;
+ GET_BUFFER_SPACE (3);
+ b += 3;
+
+ laststart = 0;
+ begalt = b;
+ break;
+
+
+ case '{':
+ /* If \{ is a literal. */
+ if (!(syntax & RE_INTERVALS)
+ /* If we're at `\{' and it's not the open-interval
+ operator. */
+ || (syntax & RE_NO_BK_BRACES))
+ goto normal_backslash;
+
+ handle_interval:
+ {
+ /* If got here, then the syntax allows intervals. */
+
+ /* At least (most) this many matches must be made. */
+ int lower_bound = -1, upper_bound = -1;
+
+ beg_interval = p - 1;
+
+ if (p == pend)
+ {
+ if (!(syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES))
+ goto unfetch_interval;
+ else
+ FREE_STACK_RETURN (REG_EBRACE);
+ }
+
+ GET_UNSIGNED_NUMBER (lower_bound);
+
+ if (c == ',')
+ {
+ GET_UNSIGNED_NUMBER (upper_bound);
+ if ((!(syntax & RE_NO_BK_BRACES) && c != '\\')
+ || ((syntax & RE_NO_BK_BRACES) && c != '}'))
+ FREE_STACK_RETURN (REG_BADBR);
+
+ if (upper_bound < 0)
+ upper_bound = RE_DUP_MAX;
+ }
+ else
+ /* Interval such as `{1}' => match exactly once. */
+ upper_bound = lower_bound;
+
+ if (lower_bound < 0 || upper_bound > RE_DUP_MAX
+ || lower_bound > upper_bound)
+ {
+ if (!(syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES))
+ goto unfetch_interval;
+ else
+ FREE_STACK_RETURN (REG_BADBR);
+ }
+
+ if (!(syntax & RE_NO_BK_BRACES))
+ {
+ if (c != '\\') FREE_STACK_RETURN (REG_EBRACE);
+
+ PATFETCH (c);
+ }
+
+ if (c != '}')
+ {
+ if (!(syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES))
+ goto unfetch_interval;
+ else
+ FREE_STACK_RETURN (REG_BADBR);
+ }
+
+ /* We just parsed a valid interval. */
+
+ /* If it's invalid to have no preceding re. */
+ if (!laststart)
+ {
+ if (syntax & RE_CONTEXT_INVALID_OPS)
+ FREE_STACK_RETURN (REG_BADRPT);
+ else if (syntax & RE_CONTEXT_INDEP_OPS)
+ laststart = b;
+ else
+ goto unfetch_interval;
+ }
+
+ /* If the upper bound is zero, don't want to succeed at
+ all; jump from `laststart' to `b + 3', which will be
+ the end of the buffer after we insert the jump. */
+ if (upper_bound == 0)
+ {
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (jump, laststart, b + 3);
+ b += 3;
+ }
+
+ /* Otherwise, we have a nontrivial interval. When
+ we're all done, the pattern will look like:
+ set_number_at <jump count> <upper bound>
+ set_number_at <succeed_n count> <lower bound>
+ succeed_n <after jump addr> <succeed_n count>
+ <body of loop>
+ jump_n <succeed_n addr> <jump count>
+ (The upper bound and `jump_n' are omitted if
+ `upper_bound' is 1, though.) */
+ else
+ { /* If the upper bound is > 1, we need to insert
+ more at the end of the loop. */
+ unsigned nbytes = 10 + (upper_bound > 1) * 10;
+
+ GET_BUFFER_SPACE (nbytes);
+
+ /* Initialize lower bound of the `succeed_n', even
+ though it will be set during matching by its
+ attendant `set_number_at' (inserted next),
+ because `re_compile_fastmap' needs to know.
+ Jump to the `jump_n' we might insert below. */
+ INSERT_JUMP2 (succeed_n, laststart,
+ b + 5 + (upper_bound > 1) * 5,
+ lower_bound);
+ b += 5;
+
+ /* Code to initialize the lower bound. Insert
+ before the `succeed_n'. The `5' is the last two
+ bytes of this `set_number_at', plus 3 bytes of
+ the following `succeed_n'. */
+ insert_op2 (set_number_at, laststart, 5, lower_bound, b);
+ b += 5;
+
+ if (upper_bound > 1)
+ { /* More than one repetition is allowed, so
+ append a backward jump to the `succeed_n'
+ that starts this interval.
+
+ When we've reached this during matching,
+ we'll have matched the interval once, so
+ jump back only `upper_bound - 1' times. */
+ STORE_JUMP2 (jump_n, b, laststart + 5,
+ upper_bound - 1);
+ b += 5;
+
+ /* The location we want to set is the second
+ parameter of the `jump_n'; that is `b-2' as
+ an absolute address. `laststart' will be
+ the `set_number_at' we're about to insert;
+ `laststart+3' the number to set, the source
+ for the relative address. But we are
+ inserting into the middle of the pattern --
+ so everything is getting moved up by 5.
+ Conclusion: (b - 2) - (laststart + 3) + 5,
+ i.e., b - laststart.
+
+ We insert this at the beginning of the loop
+ so that if we fail during matching, we'll
+ reinitialize the bounds. */
+ insert_op2 (set_number_at, laststart, b - laststart,
+ upper_bound - 1, b);
+ b += 5;
+ }
+ }
+ pending_exact = 0;
+ beg_interval = NULL;
+ }
+ break;
+
+ unfetch_interval:
+ /* If an invalid interval, match the characters as literals. */
+ assert (beg_interval);
+ p = beg_interval;
+ beg_interval = NULL;
+
+ /* normal_char and normal_backslash need `c'. */
+ PATFETCH (c);
+
+ if (!(syntax & RE_NO_BK_BRACES))
+ {
+ if (p > pattern && p[-1] == '\\')
+ goto normal_backslash;
+ }
+ goto normal_char;
+
+#ifdef emacs
+ /* There is no way to specify the before_dot and after_dot
+ operators. rms says this is ok. --karl */
+ case '=':
+ BUF_PUSH (at_dot);
+ break;
+
+ case 's':
+ laststart = b;
+ PATFETCH (c);
+ BUF_PUSH_2 (syntaxspec, syntax_spec_code[c]);
+ break;
+
+ case 'S':
+ laststart = b;
+ PATFETCH (c);
+ BUF_PUSH_2 (notsyntaxspec, syntax_spec_code[c]);
+ break;
+#endif /* emacs */
+
+
+ case 'w':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ laststart = b;
+ BUF_PUSH (wordchar);
+ break;
+
+
+ case 'W':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ laststart = b;
+ BUF_PUSH (notwordchar);
+ break;
+
+
+ case '<':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ BUF_PUSH (wordbeg);
+ break;
+
+ case '>':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ BUF_PUSH (wordend);
+ break;
+
+ case 'b':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ BUF_PUSH (wordbound);
+ break;
+
+ case 'B':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ BUF_PUSH (notwordbound);
+ break;
+
+ case '`':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ BUF_PUSH (begbuf);
+ break;
+
+ case '\'':
+ if (syntax & RE_NO_GNU_OPS)
+ goto normal_char;
+ BUF_PUSH (endbuf);
+ break;
+
+ case '1': case '2': case '3': case '4': case '5':
+ case '6': case '7': case '8': case '9':
+ if (syntax & RE_NO_BK_REFS)
+ goto normal_char;
+
+ c1 = c - '0';
+
+ if (c1 > regnum)
+ FREE_STACK_RETURN (REG_ESUBREG);
+
+ /* Can't back reference to a subexpression if inside of it. */
+ if (group_in_compile_stack (compile_stack, (regnum_t) c1))
+ goto normal_char;
+
+ laststart = b;
+ BUF_PUSH_2 (duplicate, c1);
+ break;
+
+
+ case '+':
+ case '?':
+ if (syntax & RE_BK_PLUS_QM)
+ goto handle_plus;
+ else
+ goto normal_backslash;
+
+ default:
+ normal_backslash:
+ /* You might think it would be useful for \ to mean
+ not to translate; but if we don't translate it
+ it will never match anything. */
+ c = TRANSLATE (c);
+ goto normal_char;
+ }
+ break;
+
+
+ default:
+ /* Expects the character in `c'. */
+ normal_char:
+ /* If no exactn currently being built. */
+ if (!pending_exact
+
+ /* If last exactn not at current position. */
+ || pending_exact + *pending_exact + 1 != b
+
+ /* We have only one byte following the exactn for the count. */
+ || *pending_exact == (1 << BYTEWIDTH) - 1
+
+ /* If followed by a repetition operator. */
+ || *p == '*' || *p == '^'
+ || ((syntax & RE_BK_PLUS_QM)
+ ? *p == '\\' && (p[1] == '+' || p[1] == '?')
+ : (*p == '+' || *p == '?'))
+ || ((syntax & RE_INTERVALS)
+ && ((syntax & RE_NO_BK_BRACES)
+ ? *p == '{'
+ : (p[0] == '\\' && p[1] == '{'))))
+ {
+ /* Start building a new exactn. */
+
+ laststart = b;
+
+ BUF_PUSH_2 (exactn, 0);
+ pending_exact = b - 1;
+ }
+
+ BUF_PUSH (c);
+ (*pending_exact)++;
+ break;
+ } /* switch (c) */
+ } /* while p != pend */
+
+
+ /* Through the pattern now. */
+
+ if (fixup_alt_jump)
+ STORE_JUMP (jump_past_alt, fixup_alt_jump, b);
+
+ if (!COMPILE_STACK_EMPTY)
+ FREE_STACK_RETURN (REG_EPAREN);
+
+ /* If we don't want backtracking, force success
+ the first time we reach the end of the compiled pattern. */
+ if (syntax & RE_NO_POSIX_BACKTRACKING)
+ BUF_PUSH (succeed);
+
+ free (compile_stack.stack);
+
+ /* We have succeeded; set the length of the buffer. */
+ bufp->used = b - bufp->buffer;
+
+#ifdef DEBUG
+ if (debug)
+ {
+ DEBUG_PRINT1 ("\nCompiled pattern: \n");
+ print_compiled_pattern (bufp);
+ }
+#endif /* DEBUG */
+
+#ifndef MATCH_MAY_ALLOCATE
+ /* Initialize the failure stack to the largest possible stack. This
+ isn't necessary unless we're trying to avoid calling alloca in
+ the search and match routines. */
+ {
+ int num_regs = bufp->re_nsub + 1;
+
+ /* Since DOUBLE_FAIL_STACK refuses to double only if the current size
+ is strictly greater than re_max_failures, the largest possible stack
+ is 2 * re_max_failures failure points. */
+ if (fail_stack.size < (2 * re_max_failures * MAX_FAILURE_ITEMS))
+ {
+ fail_stack.size = (2 * re_max_failures * MAX_FAILURE_ITEMS);
+
+# ifdef emacs
+ if (! fail_stack.stack)
+ fail_stack.stack
+ = (fail_stack_elt_t *) xmalloc (fail_stack.size
+ * sizeof (fail_stack_elt_t));
+ else
+ fail_stack.stack
+ = (fail_stack_elt_t *) xrealloc (fail_stack.stack,
+ (fail_stack.size
+ * sizeof (fail_stack_elt_t)));
+# else /* not emacs */
+ if (! fail_stack.stack)
+ fail_stack.stack
+ = (fail_stack_elt_t *) malloc (fail_stack.size
+ * sizeof (fail_stack_elt_t));
+ else
+ fail_stack.stack
+ = (fail_stack_elt_t *) realloc (fail_stack.stack,
+ (fail_stack.size
+ * sizeof (fail_stack_elt_t)));
+# endif /* not emacs */
+ }
+
+ regex_grow_registers (num_regs);
+ }
+#endif /* not MATCH_MAY_ALLOCATE */
+
+ return REG_NOERROR;
+} /* regex_compile */
+\f
+/* Subroutines for `regex_compile'. */
+
+/* Store OP at LOC followed by two-byte integer parameter ARG. */
+
+static void
+store_op1 (op, loc, arg)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg;
+{
+ *loc = (unsigned char) op;
+ STORE_NUMBER (loc + 1, arg);
+}
+
+
+/* Like `store_op1', but for two two-byte parameters ARG1 and ARG2. */
+
+static void
+store_op2 (op, loc, arg1, arg2)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg1, arg2;
+{
+ *loc = (unsigned char) op;
+ STORE_NUMBER (loc + 1, arg1);
+ STORE_NUMBER (loc + 3, arg2);
+}
+
+
+/* Copy the bytes from LOC to END to open up three bytes of space at LOC
+ for OP followed by two-byte integer parameter ARG. */
+
+static void
+insert_op1 (op, loc, arg, end)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg;
+ unsigned char *end;
+{
+ register unsigned char *pfrom = end;
+ register unsigned char *pto = end + 3;
+
+ while (pfrom != loc)
+ *--pto = *--pfrom;
+
+ store_op1 (op, loc, arg);
+}
+
+
+/* Like `insert_op1', but for two two-byte parameters ARG1 and ARG2. */
+
+static void
+insert_op2 (op, loc, arg1, arg2, end)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg1, arg2;
+ unsigned char *end;
+{
+ register unsigned char *pfrom = end;
+ register unsigned char *pto = end + 5;
+
+ while (pfrom != loc)
+ *--pto = *--pfrom;
+
+ store_op2 (op, loc, arg1, arg2);
+}
+
+
+/* P points to just after a ^ in PATTERN. Return true if that ^ comes
+ after an alternative or a begin-subexpression. We assume there is at
+ least one character before the ^. */
+
+static boolean
+at_begline_loc_p (pattern, p, syntax)
+ const char *pattern, *p;
+ reg_syntax_t syntax;
+{
+ const char *prev = p - 2;
+ boolean prev_prev_backslash = prev > pattern && prev[-1] == '\\';
+
+ return
+ /* After a subexpression? */
+ (*prev == '(' && (syntax & RE_NO_BK_PARENS || prev_prev_backslash))
+ /* After an alternative? */
+ || (*prev == '|' && (syntax & RE_NO_BK_VBAR || prev_prev_backslash));
+}
+
+
+/* The dual of at_begline_loc_p. This one is for $. We assume there is
+ at least one character after the $, i.e., `P < PEND'. */
+
+static boolean
+at_endline_loc_p (p, pend, syntax)
+ const char *p, *pend;
+ reg_syntax_t syntax;
+{
+ const char *next = p;
+ boolean next_backslash = *next == '\\';
+ const char *next_next = p + 1 < pend ? p + 1 : 0;
+
+ return
+ /* Before a subexpression? */
+ (syntax & RE_NO_BK_PARENS ? *next == ')'
+ : next_backslash && next_next && *next_next == ')')
+ /* Before an alternative? */
+ || (syntax & RE_NO_BK_VBAR ? *next == '|'
+ : next_backslash && next_next && *next_next == '|');
+}
+
+
+/* Returns true if REGNUM is in one of COMPILE_STACK's elements and
+ false if it's not. */
+
+static boolean
+group_in_compile_stack (compile_stack, regnum)
+ compile_stack_type compile_stack;
+ regnum_t regnum;
+{
+ int this_element;
+
+ for (this_element = compile_stack.avail - 1;
+ this_element >= 0;
+ this_element--)
+ if (compile_stack.stack[this_element].regnum == regnum)
+ return true;
+
+ return false;
+}
+
+
+/* Read the ending character of a range (in a bracket expression) from the
+ uncompiled pattern *P_PTR (which ends at PEND). We assume the
+ starting character is in `P[-2]'. (`P[-1]' is the character `-'.)
+ Then we set the translation of all bits between the starting and
+ ending characters (inclusive) in the compiled pattern B.
+
+ Return an error code.
+
+ We use these short variable names so we can use the same macros as
+ `regex_compile' itself. */
+
+static reg_errcode_t
+compile_range (range_start_char, p_ptr, pend, translate, syntax, b)
+ unsigned int range_start_char;
+ const char **p_ptr, *pend;
+ RE_TRANSLATE_TYPE translate;
+ reg_syntax_t syntax;
+ unsigned char *b;
+{
+ unsigned this_char;
+ const char *p = *p_ptr;
+ reg_errcode_t ret;
+#if _LIBC
+ const unsigned char *collseq;
+ unsigned int start_colseq;
+ unsigned int end_colseq;
+#else
+ unsigned end_char;
+#endif
+
+ if (p == pend)
+ return REG_ERANGE;
+
+ /* Have to increment the pointer into the pattern string, so the
+ caller isn't still at the ending character. */
+ (*p_ptr)++;
+
+ /* Report an error if the range is empty and the syntax prohibits this. */
+ ret = syntax & RE_NO_EMPTY_RANGES ? REG_ERANGE : REG_NOERROR;
+
+#if _LIBC
+ collseq = (const unsigned char *) _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_COLLSEQMB);
+
+ start_colseq = collseq[(unsigned char) TRANSLATE (range_start_char)];
+ end_colseq = collseq[(unsigned char) TRANSLATE (p[0])];
+ for (this_char = 0; this_char <= (unsigned char) -1; ++this_char)
+ {
+ unsigned int this_colseq = collseq[(unsigned char) TRANSLATE (this_char)];
+
+ if (start_colseq <= this_colseq && this_colseq <= end_colseq)
+ {
+ SET_LIST_BIT (TRANSLATE (this_char));
+ ret = REG_NOERROR;
+ }
+ }
+#else
+ /* Here we see why `this_char' has to be larger than an `unsigned
+ char' -- we would otherwise go into an infinite loop, since all
+ characters <= 0xff. */
+ range_start_char = TRANSLATE (range_start_char);
+ end_char = TRANSLATE (p[0]);
+ for (this_char = range_start_char; this_char <= end_char; ++this_char)
+ {
+ SET_LIST_BIT (TRANSLATE (this_char));
+ ret = REG_NOERROR;
+ }
+#endif
+
+ return ret;
+}
+\f
+/* re_compile_fastmap computes a ``fastmap'' for the compiled pattern in
+ BUFP. A fastmap records which of the (1 << BYTEWIDTH) possible
+ characters can start a string that matches the pattern. This fastmap
+ is used by re_search to skip quickly over impossible starting points.
+
+ The caller must supply the address of a (1 << BYTEWIDTH)-byte data
+ area as BUFP->fastmap.
+
+ We set the `fastmap', `fastmap_accurate', and `can_be_null' fields in
+ the pattern buffer.
+
+ Returns 0 if we succeed, -2 if an internal error. */
+
+int
+re_compile_fastmap (bufp)
+ struct re_pattern_buffer *bufp;
+{
+ int j, k;
+#ifdef MATCH_MAY_ALLOCATE
+ fail_stack_type fail_stack;
+#endif
+#ifndef REGEX_MALLOC
+ char *destination;
+#endif
+
+ register char *fastmap = bufp->fastmap;
+ unsigned char *pattern = bufp->buffer;
+ unsigned char *p = pattern;
+ register unsigned char *pend = pattern + bufp->used;
+
+#ifdef REL_ALLOC
+ /* This holds the pointer to the failure stack, when
+ it is allocated relocatably. */
+ fail_stack_elt_t *failure_stack_ptr;
+#endif
+
+ /* Assume that each path through the pattern can be null until
+ proven otherwise. We set this false at the bottom of switch
+ statement, to which we get only if a particular path doesn't
+ match the empty string. */
+ boolean path_can_be_null = true;
+
+ /* We aren't doing a `succeed_n' to begin with. */
+ boolean succeed_n_p = false;
+
+ assert (fastmap != NULL && p != NULL);
+
+ INIT_FAIL_STACK ();
+ bzero (fastmap, 1 << BYTEWIDTH); /* Assume nothing's valid. */
+ bufp->fastmap_accurate = 1; /* It will be when we're done. */
+ bufp->can_be_null = 0;
+
+ while (1)
+ {
+ if (p == pend || *p == succeed)
+ {
+ /* We have reached the (effective) end of pattern. */
+ if (!FAIL_STACK_EMPTY ())
+ {
+ bufp->can_be_null |= path_can_be_null;
+
+ /* Reset for next path. */
+ path_can_be_null = true;
+
+ p = fail_stack.stack[--fail_stack.avail].pointer;
+
+ continue;
+ }
+ else
+ break;
+ }
+
+ /* We should never be about to go beyond the end of the pattern. */
+ assert (p < pend);
+
+ switch (SWITCH_ENUM_CAST ((re_opcode_t) *p++))
+ {
+
+ /* I guess the idea here is to simply not bother with a fastmap
+ if a backreference is used, since it's too hard to figure out
+ the fastmap for the corresponding group. Setting
+ `can_be_null' stops `re_search_2' from using the fastmap, so
+ that is all we do. */
+ case duplicate:
+ bufp->can_be_null = 1;
+ goto done;
+
+
+ /* Following are the cases which match a character. These end
+ with `break'. */
+
+ case exactn:
+ fastmap[p[1]] = 1;
+ break;
+
+
+ case charset:
+ for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--)
+ if (p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH)))
+ fastmap[j] = 1;
+ break;
+
+
+ case charset_not:
+ /* Chars beyond end of map must be allowed. */
+ for (j = *p * BYTEWIDTH; j < (1 << BYTEWIDTH); j++)
+ fastmap[j] = 1;
+
+ for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--)
+ if (!(p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH))))
+ fastmap[j] = 1;
+ break;
+
+
+ case wordchar:
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) == Sword)
+ fastmap[j] = 1;
+ break;
+
+
+ case notwordchar:
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) != Sword)
+ fastmap[j] = 1;
+ break;
+
+
+ case anychar:
+ {
+ int fastmap_newline = fastmap['\n'];
+
+ /* `.' matches anything ... */
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ fastmap[j] = 1;
+
+ /* ... except perhaps newline. */
+ if (!(bufp->syntax & RE_DOT_NEWLINE))
+ fastmap['\n'] = fastmap_newline;
+
+ /* Return if we have already set `can_be_null'; if we have,
+ then the fastmap is irrelevant. Something's wrong here. */
+ else if (bufp->can_be_null)
+ goto done;
+
+ /* Otherwise, have to check alternative paths. */
+ break;
+ }
+
+#ifdef emacs
+ case syntaxspec:
+ k = *p++;
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) == (enum syntaxcode) k)
+ fastmap[j] = 1;
+ break;
+
+
+ case notsyntaxspec:
+ k = *p++;
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) != (enum syntaxcode) k)
+ fastmap[j] = 1;
+ break;
+
+
+ /* All cases after this match the empty string. These end with
+ `continue'. */
+
+
+ case before_dot:
+ case at_dot:
+ case after_dot:
+ continue;
+#endif /* emacs */
+
+
+ case no_op:
+ case begline:
+ case endline:
+ case begbuf:
+ case endbuf:
+ case wordbound:
+ case notwordbound:
+ case wordbeg:
+ case wordend:
+ case push_dummy_failure:
+ continue;
+
+
+ case jump_n:
+ case pop_failure_jump:
+ case maybe_pop_jump:
+ case jump:
+ case jump_past_alt:
+ case dummy_failure_jump:
+ EXTRACT_NUMBER_AND_INCR (j, p);
+ p += j;
+ if (j > 0)
+ continue;
+
+ /* Jump backward implies we just went through the body of a
+ loop and matched nothing. Opcode jumped to should be
+ `on_failure_jump' or `succeed_n'. Just treat it like an
+ ordinary jump. For a * loop, it has pushed its failure
+ point already; if so, discard that as redundant. */
+ if ((re_opcode_t) *p != on_failure_jump
+ && (re_opcode_t) *p != succeed_n)
+ continue;
+
+ p++;
+ EXTRACT_NUMBER_AND_INCR (j, p);
+ p += j;
+
+ /* If what's on the stack is where we are now, pop it. */
+ if (!FAIL_STACK_EMPTY ()
+ && fail_stack.stack[fail_stack.avail - 1].pointer == p)
+ fail_stack.avail--;
+
+ continue;
+
+
+ case on_failure_jump:
+ case on_failure_keep_string_jump:
+ handle_on_failure_jump:
+ EXTRACT_NUMBER_AND_INCR (j, p);
+
+ /* For some patterns, e.g., `(a?)?', `p+j' here points to the
+ end of the pattern. We don't want to push such a point,
+ since when we restore it above, entering the switch will
+ increment `p' past the end of the pattern. We don't need
+ to push such a point since we obviously won't find any more
+ fastmap entries beyond `pend'. Such a pattern can match
+ the null string, though. */
+ if (p + j < pend)
+ {
+ if (!PUSH_PATTERN_OP (p + j, fail_stack))
+ {
+ RESET_FAIL_STACK ();
+ return -2;
+ }
+ }
+ else
+ bufp->can_be_null = 1;
+
+ if (succeed_n_p)
+ {
+ EXTRACT_NUMBER_AND_INCR (k, p); /* Skip the n. */
+ succeed_n_p = false;
+ }
+
+ continue;
+
+
+ case succeed_n:
+ /* Get to the number of times to succeed. */
+ p += 2;
+
+ /* Increment p past the n for when k != 0. */
+ EXTRACT_NUMBER_AND_INCR (k, p);
+ if (k == 0)
+ {
+ p -= 4;
+ succeed_n_p = true; /* Spaghetti code alert. */
+ goto handle_on_failure_jump;
+ }
+ continue;
+
+
+ case set_number_at:
+ p += 4;
+ continue;
+
+
+ case start_memory:
+ case stop_memory:
+ p += 2;
+ continue;
+
+
+ default:
+ abort (); /* We have listed all the cases. */
+ } /* switch *p++ */
+
+ /* Getting here means we have found the possible starting
+ characters for one path of the pattern -- and that the empty
+ string does not match. We need not follow this path further.
+ Instead, look at the next alternative (remembered on the
+ stack), or quit if no more. The test at the top of the loop
+ does these things. */
+ path_can_be_null = false;
+ p = pend;
+ } /* while p */
+
+ /* Set `can_be_null' for the last path (also the first path, if the
+ pattern is empty). */
+ bufp->can_be_null |= path_can_be_null;
+
+ done:
+ RESET_FAIL_STACK ();
+ return 0;
+} /* re_compile_fastmap */
+#ifdef _LIBC
+weak_alias (__re_compile_fastmap, re_compile_fastmap)
+#endif
+\f
+/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
+ ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use
+ this memory for recording register information. STARTS and ENDS
+ must be allocated using the malloc library routine, and must each
+ be at least NUM_REGS * sizeof (regoff_t) bytes long.
+
+ If NUM_REGS == 0, then subsequent matches should allocate their own
+ register data.
+
+ Unless this function is called, the first search or match using
+ PATTERN_BUFFER will allocate its own register data, without
+ freeing the old data. */
+
+void
+re_set_registers (bufp, regs, num_regs, starts, ends)
+ struct re_pattern_buffer *bufp;
+ struct re_registers *regs;
+ unsigned num_regs;
+ regoff_t *starts, *ends;
+{
+ if (num_regs)
+ {
+ bufp->regs_allocated = REGS_REALLOCATE;
+ regs->num_regs = num_regs;
+ regs->start = starts;
+ regs->end = ends;
+ }
+ else
+ {
+ bufp->regs_allocated = REGS_UNALLOCATED;
+ regs->num_regs = 0;
+ regs->start = regs->end = (regoff_t *) 0;
+ }
+}
+#ifdef _LIBC
+weak_alias (__re_set_registers, re_set_registers)
+#endif
+\f
+/* Searching routines. */
+
+/* Like re_search_2, below, but only one string is specified, and
+ doesn't let you say where to stop matching. */
+
+int
+re_search (bufp, string, size, startpos, range, regs)
+ struct re_pattern_buffer *bufp;
+ const char *string;
+ int size, startpos, range;
+ struct re_registers *regs;
+{
+ return re_search_2 (bufp, NULL, 0, string, size, startpos, range,
+ regs, size);
+}
+#ifdef _LIBC
+weak_alias (__re_search, re_search)
+#endif
+
+
+/* Using the compiled pattern in BUFP->buffer, first tries to match the
+ virtual concatenation of STRING1 and STRING2, starting first at index
+ STARTPOS, then at STARTPOS + 1, and so on.
+
+ STRING1 and STRING2 have length SIZE1 and SIZE2, respectively.
+
+ RANGE is how far to scan while trying to match. RANGE = 0 means try
+ only at STARTPOS; in general, the last start tried is STARTPOS +
+ RANGE.
+
+ In REGS, return the indices of the virtual concatenation of STRING1
+ and STRING2 that matched the entire BUFP->buffer and its contained
+ subexpressions.
+
+ Do not consider matching one past the index STOP in the virtual
+ concatenation of STRING1 and STRING2.
+
+ We return either the position in the strings at which the match was
+ found, -1 if no match, or -2 if error (such as failure
+ stack overflow). */
+
+int
+re_search_2 (bufp, string1, size1, string2, size2, startpos, range, regs, stop)
+ struct re_pattern_buffer *bufp;
+ const char *string1, *string2;
+ int size1, size2;
+ int startpos;
+ int range;
+ struct re_registers *regs;
+ int stop;
+{
+ int val;
+ register char *fastmap = bufp->fastmap;
+ register RE_TRANSLATE_TYPE translate = bufp->translate;
+ int total_size = size1 + size2;
+ int endpos = startpos + range;
+
+ /* Check for out-of-range STARTPOS. */
+ if (startpos < 0 || startpos > total_size)
+ return -1;
+
+ /* Fix up RANGE if it might eventually take us outside
+ the virtual concatenation of STRING1 and STRING2.
+ Make sure we won't move STARTPOS below 0 or above TOTAL_SIZE. */
+ if (endpos < 0)
+ range = 0 - startpos;
+ else if (endpos > total_size)
+ range = total_size - startpos;
+
+ /* If the search isn't to be a backwards one, don't waste time in a
+ search for a pattern that must be anchored. */
+ if (bufp->used > 0 && range > 0
+ && ((re_opcode_t) bufp->buffer[0] == begbuf
+ /* `begline' is like `begbuf' if it cannot match at newlines. */
+ || ((re_opcode_t) bufp->buffer[0] == begline
+ && !bufp->newline_anchor)))
+ {
+ if (startpos > 0)
+ return -1;
+ else
+ range = 1;
+ }
+
+#ifdef emacs
+ /* In a forward search for something that starts with \=.
+ don't keep searching past point. */
+ if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == at_dot && range > 0)
+ {
+ range = PT - startpos;
+ if (range <= 0)
+ return -1;
+ }
+#endif /* emacs */
+
+ /* Update the fastmap now if not correct already. */
+ if (fastmap && !bufp->fastmap_accurate)
+ if (re_compile_fastmap (bufp) == -2)
+ return -2;
+
+ /* Loop through the string, looking for a place to start matching. */
+ for (;;)
+ {
+ /* If a fastmap is supplied, skip quickly over characters that
+ cannot be the start of a match. If the pattern can match the
+ null string, however, we don't need to skip characters; we want
+ the first null string. */
+ if (fastmap && startpos < total_size && !bufp->can_be_null)
+ {
+ if (range > 0) /* Searching forwards. */
+ {
+ register const char *d;
+ register int lim = 0;
+ int irange = range;
+
+ if (startpos < size1 && startpos + range >= size1)
+ lim = range - (size1 - startpos);
+
+ d = (startpos >= size1 ? string2 - size1 : string1) + startpos;
+
+ /* Written out as an if-else to avoid testing `translate'
+ inside the loop. */
+ if (translate)
+ while (range > lim
+ && !fastmap[(unsigned char)
+ translate[(unsigned char) *d++]])
+ range--;
+ else
+ while (range > lim && !fastmap[(unsigned char) *d++])
+ range--;
+
+ startpos += irange - range;
+ }
+ else /* Searching backwards. */
+ {
+ register char c = (size1 == 0 || startpos >= size1
+ ? string2[startpos - size1]
+ : string1[startpos]);
+
+ if (!fastmap[(unsigned char) TRANSLATE (c)])
+ goto advance;
+ }
+ }
+
+ /* If can't match the null string, and that's all we have left, fail. */
+ if (range >= 0 && startpos == total_size && fastmap
+ && !bufp->can_be_null)
+ return -1;
+
+ val = re_match_2_internal (bufp, string1, size1, string2, size2,
+ startpos, regs, stop);
+#ifndef REGEX_MALLOC
+# ifdef C_ALLOCA
+ alloca (0);
+# endif
+#endif
+
+ if (val >= 0)
+ return startpos;
+
+ if (val == -2)
+ return -2;
+
+ advance:
+ if (!range)
+ break;
+ else if (range > 0)
+ {
+ range--;
+ startpos++;
+ }
+ else
+ {
+ range++;
+ startpos--;
+ }
+ }
+ return -1;
+} /* re_search_2 */
+#ifdef _LIBC
+weak_alias (__re_search_2, re_search_2)
+#endif
+\f
+/* This converts PTR, a pointer into one of the search strings `string1'
+ and `string2' into an offset from the beginning of that string. */
+#define POINTER_TO_OFFSET(ptr) \
+ (FIRST_STRING_P (ptr) \
+ ? ((regoff_t) ((ptr) - string1)) \
+ : ((regoff_t) ((ptr) - string2 + size1)))
+
+/* Macros for dealing with the split strings in re_match_2. */
+
+#define MATCHING_IN_FIRST_STRING (dend == end_match_1)
+
+/* Call before fetching a character with *d. This switches over to
+ string2 if necessary. */
+#define PREFETCH() \
+ while (d == dend) \
+ { \
+ /* End of string2 => fail. */ \
+ if (dend == end_match_2) \
+ goto fail; \
+ /* End of string1 => advance to string2. */ \
+ d = string2; \
+ dend = end_match_2; \
+ }
+
+
+/* Test if at very beginning or at very end of the virtual concatenation
+ of `string1' and `string2'. If only one string, it's `string2'. */
+#define AT_STRINGS_BEG(d) ((d) == (size1 ? string1 : string2) || !size2)
+#define AT_STRINGS_END(d) ((d) == end2)
+
+
+/* Test if D points to a character which is word-constituent. We have
+ two special cases to check for: if past the end of string1, look at
+ the first character in string2; and if before the beginning of
+ string2, look at the last character in string1. */
+#define WORDCHAR_P(d) \
+ (SYNTAX ((d) == end1 ? *string2 \
+ : (d) == string2 - 1 ? *(end1 - 1) : *(d)) \
+ == Sword)
+
+/* Disabled due to a compiler bug -- see comment at case wordbound */
+#if 0
+/* Test if the character before D and the one at D differ with respect
+ to being word-constituent. */
+#define AT_WORD_BOUNDARY(d) \
+ (AT_STRINGS_BEG (d) || AT_STRINGS_END (d) \
+ || WORDCHAR_P (d - 1) != WORDCHAR_P (d))
+#endif
+
+/* Free everything we malloc. */
+#ifdef MATCH_MAY_ALLOCATE
+# define FREE_VAR(var) if (var) REGEX_FREE (var); var = NULL
+# define FREE_VARIABLES() \
+ do { \
+ REGEX_FREE_STACK (fail_stack.stack); \
+ FREE_VAR (regstart); \
+ FREE_VAR (regend); \
+ FREE_VAR (old_regstart); \
+ FREE_VAR (old_regend); \
+ FREE_VAR (best_regstart); \
+ FREE_VAR (best_regend); \
+ FREE_VAR (reg_info); \
+ FREE_VAR (reg_dummy); \
+ FREE_VAR (reg_info_dummy); \
+ } while (0)
+#else
+# define FREE_VARIABLES() ((void)0) /* Do nothing! But inhibit gcc warning. */
+#endif /* not MATCH_MAY_ALLOCATE */
+
+/* These values must meet several constraints. They must not be valid
+ register values; since we have a limit of 255 registers (because
+ we use only one byte in the pattern for the register number), we can
+ use numbers larger than 255. They must differ by 1, because of
+ NUM_FAILURE_ITEMS above. And the value for the lowest register must
+ be larger than the value for the highest register, so we do not try
+ to actually save any registers when none are active. */
+#define NO_HIGHEST_ACTIVE_REG (1 << BYTEWIDTH)
+#define NO_LOWEST_ACTIVE_REG (NO_HIGHEST_ACTIVE_REG + 1)
+\f
+/* Matching routines. */
+
+#ifndef emacs /* Emacs never uses this. */
+/* re_match is like re_match_2 except it takes only a single string. */
+
+int
+re_match (bufp, string, size, pos, regs)
+ struct re_pattern_buffer *bufp;
+ const char *string;
+ int size, pos;
+ struct re_registers *regs;
+{
+ int result = re_match_2_internal (bufp, NULL, 0, string, size,
+ pos, regs, size);
+# ifndef REGEX_MALLOC
+# ifdef C_ALLOCA
+ alloca (0);
+# endif
+# endif
+ return result;
+}
+# ifdef _LIBC
+weak_alias (__re_match, re_match)
+# endif
+#endif /* not emacs */
+
+static boolean group_match_null_string_p _RE_ARGS ((unsigned char **p,
+ unsigned char *end,
+ register_info_type *reg_info));
+static boolean alt_match_null_string_p _RE_ARGS ((unsigned char *p,
+ unsigned char *end,
+ register_info_type *reg_info));
+static boolean common_op_match_null_string_p _RE_ARGS ((unsigned char **p,
+ unsigned char *end,
+ register_info_type *reg_info));
+static int bcmp_translate _RE_ARGS ((const char *s1, const char *s2,
+ int len, char *translate));
+
+/* re_match_2 matches the compiled pattern in BUFP against the
+ the (virtual) concatenation of STRING1 and STRING2 (of length SIZE1
+ and SIZE2, respectively). We start matching at POS, and stop
+ matching at STOP.
+
+ If REGS is non-null and the `no_sub' field of BUFP is nonzero, we
+ store offsets for the substring each group matched in REGS. See the
+ documentation for exactly how many groups we fill.
+
+ We return -1 if no match, -2 if an internal error (such as the
+ failure stack overflowing). Otherwise, we return the length of the
+ matched substring. */
+
+int
+re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
+ struct re_pattern_buffer *bufp;
+ const char *string1, *string2;
+ int size1, size2;
+ int pos;
+ struct re_registers *regs;
+ int stop;
+{
+ int result = re_match_2_internal (bufp, string1, size1, string2, size2,
+ pos, regs, stop);
+#ifndef REGEX_MALLOC
+# ifdef C_ALLOCA
+ alloca (0);
+# endif
+#endif
+ return result;
+}
+#ifdef _LIBC
+weak_alias (__re_match_2, re_match_2)
+#endif
+
+/* This is a separate function so that we can force an alloca cleanup
+ afterwards. */
+static int
+re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop)
+ struct re_pattern_buffer *bufp;
+ const char *string1, *string2;
+ int size1, size2;
+ int pos;
+ struct re_registers *regs;
+ int stop;
+{
+ /* General temporaries. */
+ int mcnt;
+ unsigned char *p1;
+
+ /* Just past the end of the corresponding string. */
+ const char *end1, *end2;
+
+ /* Pointers into string1 and string2, just past the last characters in
+ each to consider matching. */
+ const char *end_match_1, *end_match_2;
+
+ /* Where we are in the data, and the end of the current string. */
+ const char *d, *dend;
+
+ /* Where we are in the pattern, and the end of the pattern. */
+ unsigned char *p = bufp->buffer;
+ register unsigned char *pend = p + bufp->used;
+
+ /* Mark the opcode just after a start_memory, so we can test for an
+ empty subpattern when we get to the stop_memory. */
+ unsigned char *just_past_start_mem = 0;
+
+ /* We use this to map every character in the string. */
+ RE_TRANSLATE_TYPE translate = bufp->translate;
+
+ /* Failure point stack. Each place that can handle a failure further
+ down the line pushes a failure point on this stack. It consists of
+ restart, regend, and reg_info for all registers corresponding to
+ the subexpressions we're currently inside, plus the number of such
+ registers, and, finally, two char *'s. The first char * is where
+ to resume scanning the pattern; the second one is where to resume
+ scanning the strings. If the latter is zero, the failure point is
+ a ``dummy''; if a failure happens and the failure point is a dummy,
+ it gets discarded and the next next one is tried. */
+#ifdef MATCH_MAY_ALLOCATE /* otherwise, this is global. */
+ fail_stack_type fail_stack;
+#endif
+#ifdef DEBUG
+ static unsigned failure_id;
+ unsigned nfailure_points_pushed = 0, nfailure_points_popped = 0;
+#endif
+
+#ifdef REL_ALLOC
+ /* This holds the pointer to the failure stack, when
+ it is allocated relocatably. */
+ fail_stack_elt_t *failure_stack_ptr;
+#endif
+
+ /* We fill all the registers internally, independent of what we
+ return, for use in backreferences. The number here includes
+ an element for register zero. */
+ size_t num_regs = bufp->re_nsub + 1;
+
+ /* The currently active registers. */
+ active_reg_t lowest_active_reg = NO_LOWEST_ACTIVE_REG;
+ active_reg_t highest_active_reg = NO_HIGHEST_ACTIVE_REG;
+
+ /* Information on the contents of registers. These are pointers into
+ the input strings; they record just what was matched (on this
+ attempt) by a subexpression part of the pattern, that is, the
+ regnum-th regstart pointer points to where in the pattern we began
+ matching and the regnum-th regend points to right after where we
+ stopped matching the regnum-th subexpression. (The zeroth register
+ keeps track of what the whole pattern matches.) */
+#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
+ const char **regstart, **regend;
+#endif
+
+ /* If a group that's operated upon by a repetition operator fails to
+ match anything, then the register for its start will need to be
+ restored because it will have been set to wherever in the string we
+ are when we last see its open-group operator. Similarly for a
+ register's end. */
+#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
+ const char **old_regstart, **old_regend;
+#endif
+
+ /* The is_active field of reg_info helps us keep track of which (possibly
+ nested) subexpressions we are currently in. The matched_something
+ field of reg_info[reg_num] helps us tell whether or not we have
+ matched any of the pattern so far this time through the reg_num-th
+ subexpression. These two fields get reset each time through any
+ loop their register is in. */
+#ifdef MATCH_MAY_ALLOCATE /* otherwise, this is global. */
+ register_info_type *reg_info;
+#endif
+
+ /* The following record the register info as found in the above
+ variables when we find a match better than any we've seen before.
+ This happens as we backtrack through the failure points, which in
+ turn happens only if we have not yet matched the entire string. */
+ unsigned best_regs_set = false;
+#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
+ const char **best_regstart, **best_regend;
+#endif
+
+ /* Logically, this is `best_regend[0]'. But we don't want to have to
+ allocate space for that if we're not allocating space for anything
+ else (see below). Also, we never need info about register 0 for
+ any of the other register vectors, and it seems rather a kludge to
+ treat `best_regend' differently than the rest. So we keep track of
+ the end of the best match so far in a separate variable. We
+ initialize this to NULL so that when we backtrack the first time
+ and need to test it, it's not garbage. */
+ const char *match_end = NULL;
+
+ /* This helps SET_REGS_MATCHED avoid doing redundant work. */
+ int set_regs_matched_done = 0;
+
+ /* Used when we pop values we don't care about. */
+#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
+ const char **reg_dummy;
+ register_info_type *reg_info_dummy;
+#endif
+
+#ifdef DEBUG
+ /* Counts the total number of registers pushed. */
+ unsigned num_regs_pushed = 0;
+#endif
+
+ DEBUG_PRINT1 ("\n\nEntering re_match_2.\n");
+
+ INIT_FAIL_STACK ();
+
+#ifdef MATCH_MAY_ALLOCATE
+ /* Do not bother to initialize all the register variables if there are
+ no groups in the pattern, as it takes a fair amount of time. If
+ there are groups, we include space for register 0 (the whole
+ pattern), even though we never use it, since it simplifies the
+ array indexing. We should fix this. */
+ if (bufp->re_nsub)
+ {
+ regstart = REGEX_TALLOC (num_regs, const char *);
+ regend = REGEX_TALLOC (num_regs, const char *);
+ old_regstart = REGEX_TALLOC (num_regs, const char *);
+ old_regend = REGEX_TALLOC (num_regs, const char *);
+ best_regstart = REGEX_TALLOC (num_regs, const char *);
+ best_regend = REGEX_TALLOC (num_regs, const char *);
+ reg_info = REGEX_TALLOC (num_regs, register_info_type);
+ reg_dummy = REGEX_TALLOC (num_regs, const char *);
+ reg_info_dummy = REGEX_TALLOC (num_regs, register_info_type);
+
+ if (!(regstart && regend && old_regstart && old_regend && reg_info
+ && best_regstart && best_regend && reg_dummy && reg_info_dummy))
+ {
+ FREE_VARIABLES ();
+ return -2;
+ }
+ }
+ else
+ {
+ /* We must initialize all our variables to NULL, so that
+ `FREE_VARIABLES' doesn't try to free them. */
+ regstart = regend = old_regstart = old_regend = best_regstart
+ = best_regend = reg_dummy = NULL;
+ reg_info = reg_info_dummy = (register_info_type *) NULL;
+ }
+#endif /* MATCH_MAY_ALLOCATE */
+
+ /* The starting position is bogus. */
+ if (pos < 0 || pos > size1 + size2)
+ {
+ FREE_VARIABLES ();
+ return -1;
+ }
+
+ /* Initialize subexpression text positions to -1 to mark ones that no
+ start_memory/stop_memory has been seen for. Also initialize the
+ register information struct. */
+ for (mcnt = 1; (unsigned) mcnt < num_regs; mcnt++)
+ {
+ regstart[mcnt] = regend[mcnt]
+ = old_regstart[mcnt] = old_regend[mcnt] = REG_UNSET_VALUE;
+
+ REG_MATCH_NULL_STRING_P (reg_info[mcnt]) = MATCH_NULL_UNSET_VALUE;
+ IS_ACTIVE (reg_info[mcnt]) = 0;
+ MATCHED_SOMETHING (reg_info[mcnt]) = 0;
+ EVER_MATCHED_SOMETHING (reg_info[mcnt]) = 0;
+ }
+
+ /* We move `string1' into `string2' if the latter's empty -- but not if
+ `string1' is null. */
+ if (size2 == 0 && string1 != NULL)
+ {
+ string2 = string1;
+ size2 = size1;
+ string1 = 0;
+ size1 = 0;
+ }
+ end1 = string1 + size1;
+ end2 = string2 + size2;
+
+ /* Compute where to stop matching, within the two strings. */
+ if (stop <= size1)
+ {
+ end_match_1 = string1 + stop;
+ end_match_2 = string2;
+ }
+ else
+ {
+ end_match_1 = end1;
+ end_match_2 = string2 + stop - size1;
+ }
+
+ /* `p' scans through the pattern as `d' scans through the data.
+ `dend' is the end of the input string that `d' points within. `d'
+ is advanced into the following input string whenever necessary, but
+ this happens before fetching; therefore, at the beginning of the
+ loop, `d' can be pointing at the end of a string, but it cannot
+ equal `string2'. */
+ if (size1 > 0 && pos <= size1)
+ {
+ d = string1 + pos;
+ dend = end_match_1;
+ }
+ else
+ {
+ d = string2 + pos - size1;
+ dend = end_match_2;
+ }
+
+ DEBUG_PRINT1 ("The compiled pattern is:\n");
+ DEBUG_PRINT_COMPILED_PATTERN (bufp, p, pend);
+ DEBUG_PRINT1 ("The string to match is: `");
+ DEBUG_PRINT_DOUBLE_STRING (d, string1, size1, string2, size2);
+ DEBUG_PRINT1 ("'\n");
+
+ /* This loops over pattern commands. It exits by returning from the
+ function if the match is complete, or it drops through if the match
+ fails at this starting point in the input data. */
+ for (;;)
+ {
+#ifdef _LIBC
+ DEBUG_PRINT2 ("\n%p: ", p);
+#else
+ DEBUG_PRINT2 ("\n0x%x: ", p);
+#endif
+
+ if (p == pend)
+ { /* End of pattern means we might have succeeded. */
+ DEBUG_PRINT1 ("end of pattern ... ");
+
+ /* If we haven't matched the entire string, and we want the
+ longest match, try backtracking. */
+ if (d != end_match_2)
+ {
+ /* 1 if this match ends in the same string (string1 or string2)
+ as the best previous match. */
+ boolean same_str_p = (FIRST_STRING_P (match_end)
+ == MATCHING_IN_FIRST_STRING);
+ /* 1 if this match is the best seen so far. */
+ boolean best_match_p;
+
+ /* AIX compiler got confused when this was combined
+ with the previous declaration. */
+ if (same_str_p)
+ best_match_p = d > match_end;
+ else
+ best_match_p = !MATCHING_IN_FIRST_STRING;
+
+ DEBUG_PRINT1 ("backtracking.\n");
+
+ if (!FAIL_STACK_EMPTY ())
+ { /* More failure points to try. */
+
+ /* If exceeds best match so far, save it. */
+ if (!best_regs_set || best_match_p)
+ {
+ best_regs_set = true;
+ match_end = d;
+
+ DEBUG_PRINT1 ("\nSAVING match as best so far.\n");
+
+ for (mcnt = 1; (unsigned) mcnt < num_regs; mcnt++)
+ {
+ best_regstart[mcnt] = regstart[mcnt];
+ best_regend[mcnt] = regend[mcnt];
+ }
+ }
+ goto fail;
+ }
+
+ /* If no failure points, don't restore garbage. And if
+ last match is real best match, don't restore second
+ best one. */
+ else if (best_regs_set && !best_match_p)
+ {
+ restore_best_regs:
+ /* Restore best match. It may happen that `dend ==
+ end_match_1' while the restored d is in string2.
+ For example, the pattern `x.*y.*z' against the
+ strings `x-' and `y-z-', if the two strings are
+ not consecutive in memory. */
+ DEBUG_PRINT1 ("Restoring best registers.\n");
+
+ d = match_end;
+ dend = ((d >= string1 && d <= end1)
+ ? end_match_1 : end_match_2);
+
+ for (mcnt = 1; (unsigned) mcnt < num_regs; mcnt++)
+ {
+ regstart[mcnt] = best_regstart[mcnt];
+ regend[mcnt] = best_regend[mcnt];
+ }
+ }
+ } /* d != end_match_2 */
+
+ succeed_label:
+ DEBUG_PRINT1 ("Accepting match.\n");
+
+ /* If caller wants register contents data back, do it. */
+ if (regs && !bufp->no_sub)
+ {
+ /* Have the register data arrays been allocated? */
+ if (bufp->regs_allocated == REGS_UNALLOCATED)
+ { /* No. So allocate them with malloc. We need one
+ extra element beyond `num_regs' for the `-1' marker
+ GNU code uses. */
+ regs->num_regs = MAX (RE_NREGS, num_regs + 1);
+ regs->start = TALLOC (regs->num_regs, regoff_t);
+ regs->end = TALLOC (regs->num_regs, regoff_t);
+ if (regs->start == NULL || regs->end == NULL)
+ {
+ FREE_VARIABLES ();
+ return -2;
+ }
+ bufp->regs_allocated = REGS_REALLOCATE;
+ }
+ else if (bufp->regs_allocated == REGS_REALLOCATE)
+ { /* Yes. If we need more elements than were already
+ allocated, reallocate them. If we need fewer, just
+ leave it alone. */
+ if (regs->num_regs < num_regs + 1)
+ {
+ regs->num_regs = num_regs + 1;
+ RETALLOC (regs->start, regs->num_regs, regoff_t);
+ RETALLOC (regs->end, regs->num_regs, regoff_t);
+ if (regs->start == NULL || regs->end == NULL)
+ {
+ FREE_VARIABLES ();
+ return -2;
+ }
+ }
+ }
+ else
+ {
+ /* These braces fend off a "empty body in an else-statement"
+ warning under GCC when assert expands to nothing. */
+ assert (bufp->regs_allocated == REGS_FIXED);
+ }
+
+ /* Convert the pointer data in `regstart' and `regend' to
+ indices. Register zero has to be set differently,
+ since we haven't kept track of any info for it. */
+ if (regs->num_regs > 0)
+ {
+ regs->start[0] = pos;
+ regs->end[0] = (MATCHING_IN_FIRST_STRING
+ ? ((regoff_t) (d - string1))
+ : ((regoff_t) (d - string2 + size1)));
+ }
+
+ /* Go through the first `min (num_regs, regs->num_regs)'
+ registers, since that is all we initialized. */
+ for (mcnt = 1; (unsigned) mcnt < MIN (num_regs, regs->num_regs);
+ mcnt++)
+ {
+ if (REG_UNSET (regstart[mcnt]) || REG_UNSET (regend[mcnt]))
+ regs->start[mcnt] = regs->end[mcnt] = -1;
+ else
+ {
+ regs->start[mcnt]
+ = (regoff_t) POINTER_TO_OFFSET (regstart[mcnt]);
+ regs->end[mcnt]
+ = (regoff_t) POINTER_TO_OFFSET (regend[mcnt]);
+ }
+ }
+
+ /* If the regs structure we return has more elements than
+ were in the pattern, set the extra elements to -1. If
+ we (re)allocated the registers, this is the case,
+ because we always allocate enough to have at least one
+ -1 at the end. */
+ for (mcnt = num_regs; (unsigned) mcnt < regs->num_regs; mcnt++)
+ regs->start[mcnt] = regs->end[mcnt] = -1;
+ } /* regs && !bufp->no_sub */
+
+ DEBUG_PRINT4 ("%u failure points pushed, %u popped (%u remain).\n",
+ nfailure_points_pushed, nfailure_points_popped,
+ nfailure_points_pushed - nfailure_points_popped);
+ DEBUG_PRINT2 ("%u registers pushed.\n", num_regs_pushed);
+
+ mcnt = d - pos - (MATCHING_IN_FIRST_STRING
+ ? string1
+ : string2 - size1);
+
+ DEBUG_PRINT2 ("Returning %d from re_match_2.\n", mcnt);
+
+ FREE_VARIABLES ();
+ return mcnt;
+ }
+
+ /* Otherwise match next pattern command. */
+ switch (SWITCH_ENUM_CAST ((re_opcode_t) *p++))
+ {
+ /* Ignore these. Used to ignore the n of succeed_n's which
+ currently have n == 0. */
+ case no_op:
+ DEBUG_PRINT1 ("EXECUTING no_op.\n");
+ break;
+
+ case succeed:
+ DEBUG_PRINT1 ("EXECUTING succeed.\n");
+ goto succeed_label;
+
+ /* Match the next n pattern characters exactly. The following
+ byte in the pattern defines n, and the n bytes after that
+ are the characters to match. */
+ case exactn:
+ mcnt = *p++;
+ DEBUG_PRINT2 ("EXECUTING exactn %d.\n", mcnt);
+
+ /* This is written out as an if-else so we don't waste time
+ testing `translate' inside the loop. */
+ if (translate)
+ {
+ do
+ {
+ PREFETCH ();
+ if ((unsigned char) translate[(unsigned char) *d++]
+ != (unsigned char) *p++)
+ goto fail;
+ }
+ while (--mcnt);
+ }
+ else
+ {
+ do
+ {
+ PREFETCH ();
+ if (*d++ != (char) *p++) goto fail;
+ }
+ while (--mcnt);
+ }
+ SET_REGS_MATCHED ();
+ break;
+
+
+ /* Match any character except possibly a newline or a null. */
+ case anychar:
+ DEBUG_PRINT1 ("EXECUTING anychar.\n");
+
+ PREFETCH ();
+
+ if ((!(bufp->syntax & RE_DOT_NEWLINE) && TRANSLATE (*d) == '\n')
+ || (bufp->syntax & RE_DOT_NOT_NULL && TRANSLATE (*d) == '\000'))
+ goto fail;
+
+ SET_REGS_MATCHED ();
+ DEBUG_PRINT2 (" Matched `%d'.\n", *d);
+ d++;
+ break;
+
+
+ case charset:
+ case charset_not:
+ {
+ register unsigned char c;
+ boolean not = (re_opcode_t) *(p - 1) == charset_not;
+
+ DEBUG_PRINT2 ("EXECUTING charset%s.\n", not ? "_not" : "");
+
+ PREFETCH ();
+ c = TRANSLATE (*d); /* The character to match. */
+
+ /* Cast to `unsigned' instead of `unsigned char' in case the
+ bit list is a full 32 bytes long. */
+ if (c < (unsigned) (*p * BYTEWIDTH)
+ && p[1 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
+ not = !not;
+
+ p += 1 + *p;
+
+ if (!not) goto fail;
+
+ SET_REGS_MATCHED ();
+ d++;
+ break;
+ }
+
+
+ /* The beginning of a group is represented by start_memory.
+ The arguments are the register number in the next byte, and the
+ number of groups inner to this one in the next. The text
+ matched within the group is recorded (in the internal
+ registers data structure) under the register number. */
+ case start_memory:
+ DEBUG_PRINT3 ("EXECUTING start_memory %d (%d):\n", *p, p[1]);
+
+ /* Find out if this group can match the empty string. */
+ p1 = p; /* To send to group_match_null_string_p. */
+
+ if (REG_MATCH_NULL_STRING_P (reg_info[*p]) == MATCH_NULL_UNSET_VALUE)
+ REG_MATCH_NULL_STRING_P (reg_info[*p])
+ = group_match_null_string_p (&p1, pend, reg_info);
+
+ /* Save the position in the string where we were the last time
+ we were at this open-group operator in case the group is
+ operated upon by a repetition operator, e.g., with `(a*)*b'
+ against `ab'; then we want to ignore where we are now in
+ the string in case this attempt to match fails. */
+ old_regstart[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p])
+ ? REG_UNSET (regstart[*p]) ? d : regstart[*p]
+ : regstart[*p];
+ DEBUG_PRINT2 (" old_regstart: %d\n",
+ POINTER_TO_OFFSET (old_regstart[*p]));
+
+ regstart[*p] = d;
+ DEBUG_PRINT2 (" regstart: %d\n", POINTER_TO_OFFSET (regstart[*p]));
+
+ IS_ACTIVE (reg_info[*p]) = 1;
+ MATCHED_SOMETHING (reg_info[*p]) = 0;
+
+ /* Clear this whenever we change the register activity status. */
+ set_regs_matched_done = 0;
+
+ /* This is the new highest active register. */
+ highest_active_reg = *p;
+
+ /* If nothing was active before, this is the new lowest active
+ register. */
+ if (lowest_active_reg == NO_LOWEST_ACTIVE_REG)
+ lowest_active_reg = *p;
+
+ /* Move past the register number and inner group count. */
+ p += 2;
+ just_past_start_mem = p;
+
+ break;
+
+
+ /* The stop_memory opcode represents the end of a group. Its
+ arguments are the same as start_memory's: the register
+ number, and the number of inner groups. */
+ case stop_memory:
+ DEBUG_PRINT3 ("EXECUTING stop_memory %d (%d):\n", *p, p[1]);
+
+ /* We need to save the string position the last time we were at
+ this close-group operator in case the group is operated
+ upon by a repetition operator, e.g., with `((a*)*(b*)*)*'
+ against `aba'; then we want to ignore where we are now in
+ the string in case this attempt to match fails. */
+ old_regend[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p])
+ ? REG_UNSET (regend[*p]) ? d : regend[*p]
+ : regend[*p];
+ DEBUG_PRINT2 (" old_regend: %d\n",
+ POINTER_TO_OFFSET (old_regend[*p]));
+
+ regend[*p] = d;
+ DEBUG_PRINT2 (" regend: %d\n", POINTER_TO_OFFSET (regend[*p]));
+
+ /* This register isn't active anymore. */
+ IS_ACTIVE (reg_info[*p]) = 0;
+
+ /* Clear this whenever we change the register activity status. */
+ set_regs_matched_done = 0;
+
+ /* If this was the only register active, nothing is active
+ anymore. */
+ if (lowest_active_reg == highest_active_reg)
+ {
+ lowest_active_reg = NO_LOWEST_ACTIVE_REG;
+ highest_active_reg = NO_HIGHEST_ACTIVE_REG;
+ }
+ else
+ { /* We must scan for the new highest active register, since
+ it isn't necessarily one less than now: consider
+ (a(b)c(d(e)f)g). When group 3 ends, after the f), the
+ new highest active register is 1. */
+ unsigned char r = *p - 1;
+ while (r > 0 && !IS_ACTIVE (reg_info[r]))
+ r--;
+
+ /* If we end up at register zero, that means that we saved
+ the registers as the result of an `on_failure_jump', not
+ a `start_memory', and we jumped to past the innermost
+ `stop_memory'. For example, in ((.)*) we save
+ registers 1 and 2 as a result of the *, but when we pop
+ back to the second ), we are at the stop_memory 1.
+ Thus, nothing is active. */
+ if (r == 0)
+ {
+ lowest_active_reg = NO_LOWEST_ACTIVE_REG;
+ highest_active_reg = NO_HIGHEST_ACTIVE_REG;
+ }
+ else
+ highest_active_reg = r;
+ }
+
+ /* If just failed to match something this time around with a
+ group that's operated on by a repetition operator, try to
+ force exit from the ``loop'', and restore the register
+ information for this group that we had before trying this
+ last match. */
+ if ((!MATCHED_SOMETHING (reg_info[*p])
+ || just_past_start_mem == p - 1)
+ && (p + 2) < pend)
+ {
+ boolean is_a_jump_n = false;
+
+ p1 = p + 2;
+ mcnt = 0;
+ switch ((re_opcode_t) *p1++)
+ {
+ case jump_n:
+ is_a_jump_n = true;
+ case pop_failure_jump:
+ case maybe_pop_jump:
+ case jump:
+ case dummy_failure_jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ if (is_a_jump_n)
+ p1 += 2;
+ break;
+
+ default:
+ /* do nothing */ ;
+ }
+ p1 += mcnt;
+
+ /* If the next operation is a jump backwards in the pattern
+ to an on_failure_jump right before the start_memory
+ corresponding to this stop_memory, exit from the loop
+ by forcing a failure after pushing on the stack the
+ on_failure_jump's jump in the pattern, and d. */
+ if (mcnt < 0 && (re_opcode_t) *p1 == on_failure_jump
+ && (re_opcode_t) p1[3] == start_memory && p1[4] == *p)
+ {
+ /* If this group ever matched anything, then restore
+ what its registers were before trying this last
+ failed match, e.g., with `(a*)*b' against `ab' for
+ regstart[1], and, e.g., with `((a*)*(b*)*)*'
+ against `aba' for regend[3].
+
+ Also restore the registers for inner groups for,
+ e.g., `((a*)(b*))*' against `aba' (register 3 would
+ otherwise get trashed). */
+
+ if (EVER_MATCHED_SOMETHING (reg_info[*p]))
+ {
+ unsigned r;
+
+ EVER_MATCHED_SOMETHING (reg_info[*p]) = 0;
+
+ /* Restore this and inner groups' (if any) registers. */
+ for (r = *p; r < (unsigned) *p + (unsigned) *(p + 1);
+ r++)
+ {
+ regstart[r] = old_regstart[r];
+
+ /* xx why this test? */
+ if (old_regend[r] >= regstart[r])
+ regend[r] = old_regend[r];
+ }
+ }
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ PUSH_FAILURE_POINT (p1 + mcnt, d, -2);
+
+ goto fail;
+ }
+ }
+
+ /* Move past the register number and the inner group count. */
+ p += 2;
+ break;
+
+
+ /* \<digit> has been turned into a `duplicate' command which is
+ followed by the numeric value of <digit> as the register number. */
+ case duplicate:
+ {
+ register const char *d2, *dend2;
+ int regno = *p++; /* Get which register to match against. */
+ DEBUG_PRINT2 ("EXECUTING duplicate %d.\n", regno);
+
+ /* Can't back reference a group which we've never matched. */
+ if (REG_UNSET (regstart[regno]) || REG_UNSET (regend[regno]))
+ goto fail;
+
+ /* Where in input to try to start matching. */
+ d2 = regstart[regno];
+
+ /* Where to stop matching; if both the place to start and
+ the place to stop matching are in the same string, then
+ set to the place to stop, otherwise, for now have to use
+ the end of the first string. */
+
+ dend2 = ((FIRST_STRING_P (regstart[regno])
+ == FIRST_STRING_P (regend[regno]))
+ ? regend[regno] : end_match_1);
+ for (;;)
+ {
+ /* If necessary, advance to next segment in register
+ contents. */
+ while (d2 == dend2)
+ {
+ if (dend2 == end_match_2) break;
+ if (dend2 == regend[regno]) break;
+
+ /* End of string1 => advance to string2. */
+ d2 = string2;
+ dend2 = regend[regno];
+ }
+ /* At end of register contents => success */
+ if (d2 == dend2) break;
+
+ /* If necessary, advance to next segment in data. */
+ PREFETCH ();
+
+ /* How many characters left in this segment to match. */
+ mcnt = dend - d;
+
+ /* Want how many consecutive characters we can match in
+ one shot, so, if necessary, adjust the count. */
+ if (mcnt > dend2 - d2)
+ mcnt = dend2 - d2;
+
+ /* Compare that many; failure if mismatch, else move
+ past them. */
+ if (translate
+ ? bcmp_translate (d, d2, mcnt, translate)
+ : memcmp (d, d2, mcnt))
+ goto fail;
+ d += mcnt, d2 += mcnt;
+
+ /* Do this because we've match some characters. */
+ SET_REGS_MATCHED ();
+ }
+ }
+ break;
+
+
+ /* begline matches the empty string at the beginning of the string
+ (unless `not_bol' is set in `bufp'), and, if
+ `newline_anchor' is set, after newlines. */
+ case begline:
+ DEBUG_PRINT1 ("EXECUTING begline.\n");
+
+ if (AT_STRINGS_BEG (d))
+ {
+ if (!bufp->not_bol) break;
+ }
+ else if (d[-1] == '\n' && bufp->newline_anchor)
+ {
+ break;
+ }
+ /* In all other cases, we fail. */
+ goto fail;
+
+
+ /* endline is the dual of begline. */
+ case endline:
+ DEBUG_PRINT1 ("EXECUTING endline.\n");
+
+ if (AT_STRINGS_END (d))
+ {
+ if (!bufp->not_eol) break;
+ }
+
+ /* We have to ``prefetch'' the next character. */
+ else if ((d == end1 ? *string2 : *d) == '\n'
+ && bufp->newline_anchor)
+ {
+ break;
+ }
+ goto fail;
+
+
+ /* Match at the very beginning of the data. */
+ case begbuf:
+ DEBUG_PRINT1 ("EXECUTING begbuf.\n");
+ if (AT_STRINGS_BEG (d))
+ break;
+ goto fail;
+
+
+ /* Match at the very end of the data. */
+ case endbuf:
+ DEBUG_PRINT1 ("EXECUTING endbuf.\n");
+ if (AT_STRINGS_END (d))
+ break;
+ goto fail;
+
+
+ /* on_failure_keep_string_jump is used to optimize `.*\n'. It
+ pushes NULL as the value for the string on the stack. Then
+ `pop_failure_point' will keep the current value for the
+ string, instead of restoring it. To see why, consider
+ matching `foo\nbar' against `.*\n'. The .* matches the foo;
+ then the . fails against the \n. But the next thing we want
+ to do is match the \n against the \n; if we restored the
+ string value, we would be back at the foo.
+
+ Because this is used only in specific cases, we don't need to
+ check all the things that `on_failure_jump' does, to make
+ sure the right things get saved on the stack. Hence we don't
+ share its code. The only reason to push anything on the
+ stack at all is that otherwise we would have to change
+ `anychar's code to do something besides goto fail in this
+ case; that seems worse than this. */
+ case on_failure_keep_string_jump:
+ DEBUG_PRINT1 ("EXECUTING on_failure_keep_string_jump");
+
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+#ifdef _LIBC
+ DEBUG_PRINT3 (" %d (to %p):\n", mcnt, p + mcnt);
+#else
+ DEBUG_PRINT3 (" %d (to 0x%x):\n", mcnt, p + mcnt);
+#endif
+
+ PUSH_FAILURE_POINT (p + mcnt, NULL, -2);
+ break;
+
+
+ /* Uses of on_failure_jump:
+
+ Each alternative starts with an on_failure_jump that points
+ to the beginning of the next alternative. Each alternative
+ except the last ends with a jump that in effect jumps past
+ the rest of the alternatives. (They really jump to the
+ ending jump of the following alternative, because tensioning
+ these jumps is a hassle.)
+
+ Repeats start with an on_failure_jump that points past both
+ the repetition text and either the following jump or
+ pop_failure_jump back to this on_failure_jump. */
+ case on_failure_jump:
+ on_failure:
+ DEBUG_PRINT1 ("EXECUTING on_failure_jump");
+
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+#ifdef _LIBC
+ DEBUG_PRINT3 (" %d (to %p)", mcnt, p + mcnt);
+#else
+ DEBUG_PRINT3 (" %d (to 0x%x)", mcnt, p + mcnt);
+#endif
+
+ /* If this on_failure_jump comes right before a group (i.e.,
+ the original * applied to a group), save the information
+ for that group and all inner ones, so that if we fail back
+ to this point, the group's information will be correct.
+ For example, in \(a*\)*\1, we need the preceding group,
+ and in \(zz\(a*\)b*\)\2, we need the inner group. */
+
+ /* We can't use `p' to check ahead because we push
+ a failure point to `p + mcnt' after we do this. */
+ p1 = p;
+
+ /* We need to skip no_op's before we look for the
+ start_memory in case this on_failure_jump is happening as
+ the result of a completed succeed_n, as in \(a\)\{1,3\}b\1
+ against aba. */
+ while (p1 < pend && (re_opcode_t) *p1 == no_op)
+ p1++;
+
+ if (p1 < pend && (re_opcode_t) *p1 == start_memory)
+ {
+ /* We have a new highest active register now. This will
+ get reset at the start_memory we are about to get to,
+ but we will have saved all the registers relevant to
+ this repetition op, as described above. */
+ highest_active_reg = *(p1 + 1) + *(p1 + 2);
+ if (lowest_active_reg == NO_LOWEST_ACTIVE_REG)
+ lowest_active_reg = *(p1 + 1);
+ }
+
+ DEBUG_PRINT1 (":\n");
+ PUSH_FAILURE_POINT (p + mcnt, d, -2);
+ break;
+
+
+ /* A smart repeat ends with `maybe_pop_jump'.
+ We change it to either `pop_failure_jump' or `jump'. */
+ case maybe_pop_jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+ DEBUG_PRINT2 ("EXECUTING maybe_pop_jump %d.\n", mcnt);
+ {
+ register unsigned char *p2 = p;
+
+ /* Compare the beginning of the repeat with what in the
+ pattern follows its end. If we can establish that there
+ is nothing that they would both match, i.e., that we
+ would have to backtrack because of (as in, e.g., `a*a')
+ then we can change to pop_failure_jump, because we'll
+ never have to backtrack.
+
+ This is not true in the case of alternatives: in
+ `(a|ab)*' we do need to backtrack to the `ab' alternative
+ (e.g., if the string was `ab'). But instead of trying to
+ detect that here, the alternative has put on a dummy
+ failure point which is what we will end up popping. */
+
+ /* Skip over open/close-group commands.
+ If what follows this loop is a ...+ construct,
+ look at what begins its body, since we will have to
+ match at least one of that. */
+ while (1)
+ {
+ if (p2 + 2 < pend
+ && ((re_opcode_t) *p2 == stop_memory
+ || (re_opcode_t) *p2 == start_memory))
+ p2 += 3;
+ else if (p2 + 6 < pend
+ && (re_opcode_t) *p2 == dummy_failure_jump)
+ p2 += 6;
+ else
+ break;
+ }
+
+ p1 = p + mcnt;
+ /* p1[0] ... p1[2] are the `on_failure_jump' corresponding
+ to the `maybe_finalize_jump' of this case. Examine what
+ follows. */
+
+ /* If we're at the end of the pattern, we can change. */
+ if (p2 == pend)
+ {
+ /* Consider what happens when matching ":\(.*\)"
+ against ":/". I don't really understand this code
+ yet. */
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT1
+ (" End of pattern: change to `pop_failure_jump'.\n");
+ }
+
+ else if ((re_opcode_t) *p2 == exactn
+ || (bufp->newline_anchor && (re_opcode_t) *p2 == endline))
+ {
+ register unsigned char c
+ = *p2 == (unsigned char) endline ? '\n' : p2[2];
+
+ if ((re_opcode_t) p1[3] == exactn && p1[5] != c)
+ {
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT3 (" %c != %c => pop_failure_jump.\n",
+ c, p1[5]);
+ }
+
+ else if ((re_opcode_t) p1[3] == charset
+ || (re_opcode_t) p1[3] == charset_not)
+ {
+ int not = (re_opcode_t) p1[3] == charset_not;
+
+ if (c < (unsigned char) (p1[4] * BYTEWIDTH)
+ && p1[5 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
+ not = !not;
+
+ /* `not' is equal to 1 if c would match, which means
+ that we can't change to pop_failure_jump. */
+ if (!not)
+ {
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT1 (" No match => pop_failure_jump.\n");
+ }
+ }
+ }
+ else if ((re_opcode_t) *p2 == charset)
+ {
+ /* We win if the first character of the loop is not part
+ of the charset. */
+ if ((re_opcode_t) p1[3] == exactn
+ && ! ((int) p2[1] * BYTEWIDTH > (int) p1[5]
+ && (p2[2 + p1[5] / BYTEWIDTH]
+ & (1 << (p1[5] % BYTEWIDTH)))))
+ {
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT1 (" No match => pop_failure_jump.\n");
+ }
+
+ else if ((re_opcode_t) p1[3] == charset_not)
+ {
+ int idx;
+ /* We win if the charset_not inside the loop
+ lists every character listed in the charset after. */
+ for (idx = 0; idx < (int) p2[1]; idx++)
+ if (! (p2[2 + idx] == 0
+ || (idx < (int) p1[4]
+ && ((p2[2 + idx] & ~ p1[5 + idx]) == 0))))
+ break;
+
+ if (idx == p2[1])
+ {
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT1 (" No match => pop_failure_jump.\n");
+ }
+ }
+ else if ((re_opcode_t) p1[3] == charset)
+ {
+ int idx;
+ /* We win if the charset inside the loop
+ has no overlap with the one after the loop. */
+ for (idx = 0;
+ idx < (int) p2[1] && idx < (int) p1[4];
+ idx++)
+ if ((p2[2 + idx] & p1[5 + idx]) != 0)
+ break;
+
+ if (idx == p2[1] || idx == p1[4])
+ {
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT1 (" No match => pop_failure_jump.\n");
+ }
+ }
+ }
+ }
+ p -= 2; /* Point at relative address again. */
+ if ((re_opcode_t) p[-1] != pop_failure_jump)
+ {
+ p[-1] = (unsigned char) jump;
+ DEBUG_PRINT1 (" Match => jump.\n");
+ goto unconditional_jump;
+ }
+ /* Note fall through. */
+
+
+ /* The end of a simple repeat has a pop_failure_jump back to
+ its matching on_failure_jump, where the latter will push a
+ failure point. The pop_failure_jump takes off failure
+ points put on by this pop_failure_jump's matching
+ on_failure_jump; we got through the pattern to here from the
+ matching on_failure_jump, so didn't fail. */
+ case pop_failure_jump:
+ {
+ /* We need to pass separate storage for the lowest and
+ highest registers, even though we don't care about the
+ actual values. Otherwise, we will restore only one
+ register from the stack, since lowest will == highest in
+ `pop_failure_point'. */
+ active_reg_t dummy_low_reg, dummy_high_reg;
+ unsigned char *pdummy;
+ const char *sdummy;
+
+ DEBUG_PRINT1 ("EXECUTING pop_failure_jump.\n");
+ POP_FAILURE_POINT (sdummy, pdummy,
+ dummy_low_reg, dummy_high_reg,
+ reg_dummy, reg_dummy, reg_info_dummy);
+ }
+ /* Note fall through. */
+
+ unconditional_jump:
+#ifdef _LIBC
+ DEBUG_PRINT2 ("\n%p: ", p);
+#else
+ DEBUG_PRINT2 ("\n0x%x: ", p);
+#endif
+ /* Note fall through. */
+
+ /* Unconditionally jump (without popping any failure points). */
+ case jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p); /* Get the amount to jump. */
+ DEBUG_PRINT2 ("EXECUTING jump %d ", mcnt);
+ p += mcnt; /* Do the jump. */
+#ifdef _LIBC
+ DEBUG_PRINT2 ("(to %p).\n", p);
+#else
+ DEBUG_PRINT2 ("(to 0x%x).\n", p);
+#endif
+ break;
+
+
+ /* We need this opcode so we can detect where alternatives end
+ in `group_match_null_string_p' et al. */
+ case jump_past_alt:
+ DEBUG_PRINT1 ("EXECUTING jump_past_alt.\n");
+ goto unconditional_jump;
+
+
+ /* Normally, the on_failure_jump pushes a failure point, which
+ then gets popped at pop_failure_jump. We will end up at
+ pop_failure_jump, also, and with a pattern of, say, `a+', we
+ are skipping over the on_failure_jump, so we have to push
+ something meaningless for pop_failure_jump to pop. */
+ case dummy_failure_jump:
+ DEBUG_PRINT1 ("EXECUTING dummy_failure_jump.\n");
+ /* It doesn't matter what we push for the string here. What
+ the code at `fail' tests is the value for the pattern. */
+ PUSH_FAILURE_POINT (NULL, NULL, -2);
+ goto unconditional_jump;
+
+
+ /* At the end of an alternative, we need to push a dummy failure
+ point in case we are followed by a `pop_failure_jump', because
+ we don't want the failure point for the alternative to be
+ popped. For example, matching `(a|ab)*' against `aab'
+ requires that we match the `ab' alternative. */
+ case push_dummy_failure:
+ DEBUG_PRINT1 ("EXECUTING push_dummy_failure.\n");
+ /* See comments just above at `dummy_failure_jump' about the
+ two zeroes. */
+ PUSH_FAILURE_POINT (NULL, NULL, -2);
+ break;
+
+ /* Have to succeed matching what follows at least n times.
+ After that, handle like `on_failure_jump'. */
+ case succeed_n:
+ EXTRACT_NUMBER (mcnt, p + 2);
+ DEBUG_PRINT2 ("EXECUTING succeed_n %d.\n", mcnt);
+
+ assert (mcnt >= 0);
+ /* Originally, this is how many times we HAVE to succeed. */
+ if (mcnt > 0)
+ {
+ mcnt--;
+ p += 2;
+ STORE_NUMBER_AND_INCR (p, mcnt);
+#ifdef _LIBC
+ DEBUG_PRINT3 (" Setting %p to %d.\n", p - 2, mcnt);
+#else
+ DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p - 2, mcnt);
+#endif
+ }
+ else if (mcnt == 0)
+ {
+#ifdef _LIBC
+ DEBUG_PRINT2 (" Setting two bytes from %p to no_op.\n", p+2);
+#else
+ DEBUG_PRINT2 (" Setting two bytes from 0x%x to no_op.\n", p+2);
+#endif
+ p[2] = (unsigned char) no_op;
+ p[3] = (unsigned char) no_op;
+ goto on_failure;
+ }
+ break;
+
+ case jump_n:
+ EXTRACT_NUMBER (mcnt, p + 2);
+ DEBUG_PRINT2 ("EXECUTING jump_n %d.\n", mcnt);
+
+ /* Originally, this is how many times we CAN jump. */
+ if (mcnt)
+ {
+ mcnt--;
+ STORE_NUMBER (p + 2, mcnt);
+#ifdef _LIBC
+ DEBUG_PRINT3 (" Setting %p to %d.\n", p + 2, mcnt);
+#else
+ DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p + 2, mcnt);
+#endif
+ goto unconditional_jump;
+ }
+ /* If don't have to jump any more, skip over the rest of command. */
+ else
+ p += 4;
+ break;
+
+ case set_number_at:
+ {
+ DEBUG_PRINT1 ("EXECUTING set_number_at.\n");
+
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+ p1 = p + mcnt;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+#ifdef _LIBC
+ DEBUG_PRINT3 (" Setting %p to %d.\n", p1, mcnt);
+#else
+ DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p1, mcnt);
+#endif
+ STORE_NUMBER (p1, mcnt);
+ break;
+ }
+
+#if 0
+ /* The DEC Alpha C compiler 3.x generates incorrect code for the
+ test WORDCHAR_P (d - 1) != WORDCHAR_P (d) in the expansion of
+ AT_WORD_BOUNDARY, so this code is disabled. Expanding the
+ macro and introducing temporary variables works around the bug. */
+
+ case wordbound:
+ DEBUG_PRINT1 ("EXECUTING wordbound.\n");
+ if (AT_WORD_BOUNDARY (d))
+ break;
+ goto fail;
+
+ case notwordbound:
+ DEBUG_PRINT1 ("EXECUTING notwordbound.\n");
+ if (AT_WORD_BOUNDARY (d))
+ goto fail;
+ break;
+#else
+ case wordbound:
+ {
+ boolean prevchar, thischar;
+
+ DEBUG_PRINT1 ("EXECUTING wordbound.\n");
+ if (AT_STRINGS_BEG (d) || AT_STRINGS_END (d))
+ break;
+
+ prevchar = WORDCHAR_P (d - 1);
+ thischar = WORDCHAR_P (d);
+ if (prevchar != thischar)
+ break;
+ goto fail;
+ }
+
+ case notwordbound:
+ {
+ boolean prevchar, thischar;
+
+ DEBUG_PRINT1 ("EXECUTING notwordbound.\n");
+ if (AT_STRINGS_BEG (d) || AT_STRINGS_END (d))
+ goto fail;
+
+ prevchar = WORDCHAR_P (d - 1);
+ thischar = WORDCHAR_P (d);
+ if (prevchar != thischar)
+ goto fail;
+ break;
+ }
+#endif
+
+ case wordbeg:
+ DEBUG_PRINT1 ("EXECUTING wordbeg.\n");
+ if (WORDCHAR_P (d) && (AT_STRINGS_BEG (d) || !WORDCHAR_P (d - 1)))
+ break;
+ goto fail;
+
+ case wordend:
+ DEBUG_PRINT1 ("EXECUTING wordend.\n");
+ if (!AT_STRINGS_BEG (d) && WORDCHAR_P (d - 1)
+ && (!WORDCHAR_P (d) || AT_STRINGS_END (d)))
+ break;
+ goto fail;
+
+#ifdef emacs
+ case before_dot:
+ DEBUG_PRINT1 ("EXECUTING before_dot.\n");
+ if (PTR_CHAR_POS ((unsigned char *) d) >= point)
+ goto fail;
+ break;
+
+ case at_dot:
+ DEBUG_PRINT1 ("EXECUTING at_dot.\n");
+ if (PTR_CHAR_POS ((unsigned char *) d) != point)
+ goto fail;
+ break;
+
+ case after_dot:
+ DEBUG_PRINT1 ("EXECUTING after_dot.\n");
+ if (PTR_CHAR_POS ((unsigned char *) d) <= point)
+ goto fail;
+ break;
+
+ case syntaxspec:
+ DEBUG_PRINT2 ("EXECUTING syntaxspec %d.\n", mcnt);
+ mcnt = *p++;
+ goto matchsyntax;
+
+ case wordchar:
+ DEBUG_PRINT1 ("EXECUTING Emacs wordchar.\n");
+ mcnt = (int) Sword;
+ matchsyntax:
+ PREFETCH ();
+ /* Can't use *d++ here; SYNTAX may be an unsafe macro. */
+ d++;
+ if (SYNTAX (d[-1]) != (enum syntaxcode) mcnt)
+ goto fail;
+ SET_REGS_MATCHED ();
+ break;
+
+ case notsyntaxspec:
+ DEBUG_PRINT2 ("EXECUTING notsyntaxspec %d.\n", mcnt);
+ mcnt = *p++;
+ goto matchnotsyntax;
+
+ case notwordchar:
+ DEBUG_PRINT1 ("EXECUTING Emacs notwordchar.\n");
+ mcnt = (int) Sword;
+ matchnotsyntax:
+ PREFETCH ();
+ /* Can't use *d++ here; SYNTAX may be an unsafe macro. */
+ d++;
+ if (SYNTAX (d[-1]) == (enum syntaxcode) mcnt)
+ goto fail;
+ SET_REGS_MATCHED ();
+ break;
+
+#else /* not emacs */
+ case wordchar:
+ DEBUG_PRINT1 ("EXECUTING non-Emacs wordchar.\n");
+ PREFETCH ();
+ if (!WORDCHAR_P (d))
+ goto fail;
+ SET_REGS_MATCHED ();
+ d++;
+ break;
+
+ case notwordchar:
+ DEBUG_PRINT1 ("EXECUTING non-Emacs notwordchar.\n");
+ PREFETCH ();
+ if (WORDCHAR_P (d))
+ goto fail;
+ SET_REGS_MATCHED ();
+ d++;
+ break;
+#endif /* not emacs */
+
+ default:
+ abort ();
+ }
+ continue; /* Successfully executed one pattern command; keep going. */
+
+
+ /* We goto here if a matching operation fails. */
+ fail:
+ if (!FAIL_STACK_EMPTY ())
+ { /* A restart point is known. Restore to that state. */
+ DEBUG_PRINT1 ("\nFAIL:\n");
+ POP_FAILURE_POINT (d, p,
+ lowest_active_reg, highest_active_reg,
+ regstart, regend, reg_info);
+
+ /* If this failure point is a dummy, try the next one. */
+ if (!p)
+ goto fail;
+
+ /* If we failed to the end of the pattern, don't examine *p. */
+ assert (p <= pend);
+ if (p < pend)
+ {
+ boolean is_a_jump_n = false;
+
+ /* If failed to a backwards jump that's part of a repetition
+ loop, need to pop this failure point and use the next one. */
+ switch ((re_opcode_t) *p)
+ {
+ case jump_n:
+ is_a_jump_n = true;
+ case maybe_pop_jump:
+ case pop_failure_jump:
+ case jump:
+ p1 = p + 1;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ p1 += mcnt;
+
+ if ((is_a_jump_n && (re_opcode_t) *p1 == succeed_n)
+ || (!is_a_jump_n
+ && (re_opcode_t) *p1 == on_failure_jump))
+ goto fail;
+ break;
+ default:
+ /* do nothing */ ;
+ }
+ }
+
+ if (d >= string1 && d <= end1)
+ dend = end_match_1;
+ }
+ else
+ break; /* Matching at this starting point really fails. */
+ } /* for (;;) */
+
+ if (best_regs_set)
+ goto restore_best_regs;
+
+ FREE_VARIABLES ();
+
+ return -1; /* Failure to match. */
+} /* re_match_2 */
+\f
+/* Subroutine definitions for re_match_2. */
+
+
+/* We are passed P pointing to a register number after a start_memory.
+
+ Return true if the pattern up to the corresponding stop_memory can
+ match the empty string, and false otherwise.
+
+ If we find the matching stop_memory, sets P to point to one past its number.
+ Otherwise, sets P to an undefined byte less than or equal to END.
+
+ We don't handle duplicates properly (yet). */
+
+static boolean
+group_match_null_string_p (p, end, reg_info)
+ unsigned char **p, *end;
+ register_info_type *reg_info;
+{
+ int mcnt;
+ /* Point to after the args to the start_memory. */
+ unsigned char *p1 = *p + 2;
+
+ while (p1 < end)
+ {
+ /* Skip over opcodes that can match nothing, and return true or
+ false, as appropriate, when we get to one that can't, or to the
+ matching stop_memory. */
+
+ switch ((re_opcode_t) *p1)
+ {
+ /* Could be either a loop or a series of alternatives. */
+ case on_failure_jump:
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+
+ /* If the next operation is not a jump backwards in the
+ pattern. */
+
+ if (mcnt >= 0)
+ {
+ /* Go through the on_failure_jumps of the alternatives,
+ seeing if any of the alternatives cannot match nothing.
+ The last alternative starts with only a jump,
+ whereas the rest start with on_failure_jump and end
+ with a jump, e.g., here is the pattern for `a|b|c':
+
+ /on_failure_jump/0/6/exactn/1/a/jump_past_alt/0/6
+ /on_failure_jump/0/6/exactn/1/b/jump_past_alt/0/3
+ /exactn/1/c
+
+ So, we have to first go through the first (n-1)
+ alternatives and then deal with the last one separately. */
+
+
+ /* Deal with the first (n-1) alternatives, which start
+ with an on_failure_jump (see above) that jumps to right
+ past a jump_past_alt. */
+
+ while ((re_opcode_t) p1[mcnt-3] == jump_past_alt)
+ {
+ /* `mcnt' holds how many bytes long the alternative
+ is, including the ending `jump_past_alt' and
+ its number. */
+
+ if (!alt_match_null_string_p (p1, p1 + mcnt - 3,
+ reg_info))
+ return false;
+
+ /* Move to right after this alternative, including the
+ jump_past_alt. */
+ p1 += mcnt;
+
+ /* Break if it's the beginning of an n-th alternative
+ that doesn't begin with an on_failure_jump. */
+ if ((re_opcode_t) *p1 != on_failure_jump)
+ break;
+
+ /* Still have to check that it's not an n-th
+ alternative that starts with an on_failure_jump. */
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ if ((re_opcode_t) p1[mcnt-3] != jump_past_alt)
+ {
+ /* Get to the beginning of the n-th alternative. */
+ p1 -= 3;
+ break;
+ }
+ }
+
+ /* Deal with the last alternative: go back and get number
+ of the `jump_past_alt' just before it. `mcnt' contains
+ the length of the alternative. */
+ EXTRACT_NUMBER (mcnt, p1 - 2);
+
+ if (!alt_match_null_string_p (p1, p1 + mcnt, reg_info))
+ return false;
+
+ p1 += mcnt; /* Get past the n-th alternative. */
+ } /* if mcnt > 0 */
+ break;
+
+
+ case stop_memory:
+ assert (p1[1] == **p);
+ *p = p1 + 2;
+ return true;
+
+
+ default:
+ if (!common_op_match_null_string_p (&p1, end, reg_info))
+ return false;
+ }
+ } /* while p1 < end */
+
+ return false;
+} /* group_match_null_string_p */
+
+
+/* Similar to group_match_null_string_p, but doesn't deal with alternatives:
+ It expects P to be the first byte of a single alternative and END one
+ byte past the last. The alternative can contain groups. */
+
+static boolean
+alt_match_null_string_p (p, end, reg_info)
+ unsigned char *p, *end;
+ register_info_type *reg_info;
+{
+ int mcnt;
+ unsigned char *p1 = p;
+
+ while (p1 < end)
+ {
+ /* Skip over opcodes that can match nothing, and break when we get
+ to one that can't. */
+
+ switch ((re_opcode_t) *p1)
+ {
+ /* It's a loop. */
+ case on_failure_jump:
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ p1 += mcnt;
+ break;
+
+ default:
+ if (!common_op_match_null_string_p (&p1, end, reg_info))
+ return false;
+ }
+ } /* while p1 < end */
+
+ return true;
+} /* alt_match_null_string_p */
+
+
+/* Deals with the ops common to group_match_null_string_p and
+ alt_match_null_string_p.
+
+ Sets P to one after the op and its arguments, if any. */
+
+static boolean
+common_op_match_null_string_p (p, end, reg_info)
+ unsigned char **p, *end;
+ register_info_type *reg_info;
+{
+ int mcnt;
+ boolean ret;
+ int reg_no;
+ unsigned char *p1 = *p;
+
+ switch ((re_opcode_t) *p1++)
+ {
+ case no_op:
+ case begline:
+ case endline:
+ case begbuf:
+ case endbuf:
+ case wordbeg:
+ case wordend:
+ case wordbound:
+ case notwordbound:
+#ifdef emacs
+ case before_dot:
+ case at_dot:
+ case after_dot:
+#endif
+ break;
+
+ case start_memory:
+ reg_no = *p1;
+ assert (reg_no > 0 && reg_no <= MAX_REGNUM);
+ ret = group_match_null_string_p (&p1, end, reg_info);
+
+ /* Have to set this here in case we're checking a group which
+ contains a group and a back reference to it. */
+
+ if (REG_MATCH_NULL_STRING_P (reg_info[reg_no]) == MATCH_NULL_UNSET_VALUE)
+ REG_MATCH_NULL_STRING_P (reg_info[reg_no]) = ret;
+
+ if (!ret)
+ return false;
+ break;
+
+ /* If this is an optimized succeed_n for zero times, make the jump. */
+ case jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ if (mcnt >= 0)
+ p1 += mcnt;
+ else
+ return false;
+ break;
+
+ case succeed_n:
+ /* Get to the number of times to succeed. */
+ p1 += 2;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+
+ if (mcnt == 0)
+ {
+ p1 -= 4;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ p1 += mcnt;
+ }
+ else
+ return false;
+ break;
+
+ case duplicate:
+ if (!REG_MATCH_NULL_STRING_P (reg_info[*p1]))
+ return false;
+ break;
+
+ case set_number_at:
+ p1 += 4;
+
+ default:
+ /* All other opcodes mean we cannot match the empty string. */
+ return false;
+ }
+
+ *p = p1;
+ return true;
+} /* common_op_match_null_string_p */
+
+
+/* Return zero if TRANSLATE[S1] and TRANSLATE[S2] are identical for LEN
+ bytes; nonzero otherwise. */
+
+static int
+bcmp_translate (s1, s2, len, translate)
+ const char *s1, *s2;
+ register int len;
+ RE_TRANSLATE_TYPE translate;
+{
+ register const unsigned char *p1 = (const unsigned char *) s1;
+ register const unsigned char *p2 = (const unsigned char *) s2;
+ while (len)
+ {
+ if (translate[*p1++] != translate[*p2++]) return 1;
+ len--;
+ }
+ return 0;
+}
+\f
+/* Entry points for GNU code. */
+
+/* re_compile_pattern is the GNU regular expression compiler: it
+ compiles PATTERN (of length SIZE) and puts the result in BUFP.
+ Returns 0 if the pattern was valid, otherwise an error string.
+
+ Assumes the `allocated' (and perhaps `buffer') and `translate' fields
+ are set in BUFP on entry.
+
+ We call regex_compile to do the actual compilation. */
+
+const char *
+re_compile_pattern (pattern, length, bufp)
+ const char *pattern;
+ size_t length;
+ struct re_pattern_buffer *bufp;
+{
+ reg_errcode_t ret;
+
+ /* GNU code is written to assume at least RE_NREGS registers will be set
+ (and at least one extra will be -1). */
+ bufp->regs_allocated = REGS_UNALLOCATED;
+
+ /* And GNU code determines whether or not to get register information
+ by passing null for the REGS argument to re_match, etc., not by
+ setting no_sub. */
+ bufp->no_sub = 0;
+
+ /* Match anchors at newline. */
+ bufp->newline_anchor = 1;
+
+ ret = regex_compile (pattern, length, re_syntax_options, bufp);
+
+ if (!ret)
+ return NULL;
+ return gettext (re_error_msgid + re_error_msgid_idx[(int) ret]);
+}
+#ifdef _LIBC
+weak_alias (__re_compile_pattern, re_compile_pattern)
+#endif
+\f
+/* Entry points compatible with 4.2 BSD regex library. We don't define
+ them unless specifically requested. */
+
+#if defined _REGEX_RE_COMP || defined _LIBC
+
+/* BSD has one and only one pattern buffer. */
+static struct re_pattern_buffer re_comp_buf;
+
+char *
+#ifdef _LIBC
+/* Make these definitions weak in libc, so POSIX programs can redefine
+ these names if they don't use our functions, and still use
+ regcomp/regexec below without link errors. */
+weak_function
+#endif
+re_comp (s)
+ const char *s;
+{
+ reg_errcode_t ret;
+
+ if (!s)
+ {
+ if (!re_comp_buf.buffer)
+ return gettext ("No previous regular expression");
+ return 0;
+ }
+
+ if (!re_comp_buf.buffer)
+ {
+ re_comp_buf.buffer = (unsigned char *) malloc (200);
+ if (re_comp_buf.buffer == NULL)
+ return (char *) gettext (re_error_msgid
+ + re_error_msgid_idx[(int) REG_ESPACE]);
+ re_comp_buf.allocated = 200;
+
+ re_comp_buf.fastmap = (char *) malloc (1 << BYTEWIDTH);
+ if (re_comp_buf.fastmap == NULL)
+ return (char *) gettext (re_error_msgid
+ + re_error_msgid_idx[(int) REG_ESPACE]);
+ }
+
+ /* Since `re_exec' always passes NULL for the `regs' argument, we
+ don't need to initialize the pattern buffer fields which affect it. */
+
+ /* Match anchors at newlines. */
+ re_comp_buf.newline_anchor = 1;
+
+ ret = regex_compile (s, strlen (s), re_syntax_options, &re_comp_buf);
+
+ if (!ret)
+ return NULL;
+
+ /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
+ return (char *) gettext (re_error_msgid + re_error_msgid_idx[(int) ret]);
+}
+
+
+int
+#ifdef _LIBC
+weak_function
+#endif
+re_exec (s)
+ const char *s;
+{
+ const int len = strlen (s);
+ return
+ 0 <= re_search (&re_comp_buf, s, len, 0, len, (struct re_registers *) 0);
+}
+
+#endif /* _REGEX_RE_COMP */
+\f
+/* POSIX.2 functions. Don't define these for Emacs. */
+
+#ifndef emacs
+
+/* regcomp takes a regular expression as a string and compiles it.
+
+ PREG is a regex_t *. We do not expect any fields to be initialized,
+ since POSIX says we shouldn't. Thus, we set
+
+ `buffer' to the compiled pattern;
+ `used' to the length of the compiled pattern;
+ `syntax' to RE_SYNTAX_POSIX_EXTENDED if the
+ REG_EXTENDED bit in CFLAGS is set; otherwise, to
+ RE_SYNTAX_POSIX_BASIC;
+ `newline_anchor' to REG_NEWLINE being set in CFLAGS;
+ `fastmap' to an allocated space for the fastmap;
+ `fastmap_accurate' to zero;
+ `re_nsub' to the number of subexpressions in PATTERN.
+
+ PATTERN is the address of the pattern string.
+
+ CFLAGS is a series of bits which affect compilation.
+
+ If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we
+ use POSIX basic syntax.
+
+ If REG_NEWLINE is set, then . and [^...] don't match newline.
+ Also, regexec will try a match beginning after every newline.
+
+ If REG_ICASE is set, then we considers upper- and lowercase
+ versions of letters to be equivalent when matching.
+
+ If REG_NOSUB is set, then when PREG is passed to regexec, that
+ routine will report only success or failure, and nothing about the
+ registers.
+
+ It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for
+ the return codes and their meanings.) */
+
+int
+regcomp (preg, pattern, cflags)
+ regex_t *preg;
+ const char *pattern;
+ int cflags;
+{
+ reg_errcode_t ret;
+ reg_syntax_t syntax
+ = (cflags & REG_EXTENDED) ?
+ RE_SYNTAX_POSIX_EXTENDED : RE_SYNTAX_POSIX_BASIC;
+
+ /* regex_compile will allocate the space for the compiled pattern. */
+ preg->buffer = 0;
+ preg->allocated = 0;
+ preg->used = 0;
+
+ /* Try to allocate space for the fastmap. */
+ preg->fastmap = (char *) malloc (1 << BYTEWIDTH);
+
+ if (cflags & REG_ICASE)
+ {
+ unsigned i;
+
+ preg->translate
+ = (RE_TRANSLATE_TYPE) malloc (CHAR_SET_SIZE
+ * sizeof (*(RE_TRANSLATE_TYPE)0));
+ if (preg->translate == NULL)
+ return (int) REG_ESPACE;
+
+ /* Map uppercase characters to corresponding lowercase ones. */
+ for (i = 0; i < CHAR_SET_SIZE; i++)
+ preg->translate[i] = ISUPPER (i) ? TOLOWER (i) : i;
+ }
+ else
+ preg->translate = NULL;
+
+ /* If REG_NEWLINE is set, newlines are treated differently. */
+ if (cflags & REG_NEWLINE)
+ { /* REG_NEWLINE implies neither . nor [^...] match newline. */
+ syntax &= ~RE_DOT_NEWLINE;
+ syntax |= RE_HAT_LISTS_NOT_NEWLINE;
+ /* It also changes the matching behavior. */
+ preg->newline_anchor = 1;
+ }
+ else
+ preg->newline_anchor = 0;
+
+ preg->no_sub = !!(cflags & REG_NOSUB);
+
+ /* POSIX says a null character in the pattern terminates it, so we
+ can use strlen here in compiling the pattern. */
+ ret = regex_compile (pattern, strlen (pattern), syntax, preg);
+
+ /* POSIX doesn't distinguish between an unmatched open-group and an
+ unmatched close-group: both are REG_EPAREN. */
+ if (ret == REG_ERPAREN) ret = REG_EPAREN;
+
+ if (ret == REG_NOERROR && preg->fastmap)
+ {
+ /* Compute the fastmap now, since regexec cannot modify the pattern
+ buffer. */
+ if (re_compile_fastmap (preg) == -2)
+ {
+ /* Some error occurred while computing the fastmap, just forget
+ about it. */
+ free (preg->fastmap);
+ preg->fastmap = NULL;
+ }
+ }
+
+ return (int) ret;
+}
+#ifdef _LIBC
+weak_alias (__regcomp, regcomp)
+#endif
+
+
+/* regexec searches for a given pattern, specified by PREG, in the
+ string STRING.
+
+ If NMATCH is zero or REG_NOSUB was set in the cflags argument to
+ `regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at
+ least NMATCH elements, and we set them to the offsets of the
+ corresponding matched substrings.
+
+ EFLAGS specifies `execution flags' which affect matching: if
+ REG_NOTBOL is set, then ^ does not match at the beginning of the
+ string; if REG_NOTEOL is set, then $ does not match at the end.
+
+ We return 0 if we find a match and REG_NOMATCH if not. */
+
+int
+regexec (preg, string, nmatch, pmatch, eflags)
+ const regex_t *preg;
+ const char *string;
+ size_t nmatch;
+ regmatch_t pmatch[];
+ int eflags;
+{
+ int ret;
+ struct re_registers regs;
+ regex_t private_preg;
+ int len = strlen (string);
+ boolean want_reg_info = !preg->no_sub && nmatch > 0;
+
+ private_preg = *preg;
+
+ private_preg.not_bol = !!(eflags & REG_NOTBOL);
+ private_preg.not_eol = !!(eflags & REG_NOTEOL);
+
+ /* The user has told us exactly how many registers to return
+ information about, via `nmatch'. We have to pass that on to the
+ matching routines. */
+ private_preg.regs_allocated = REGS_FIXED;
+
+ if (want_reg_info)
+ {
+ regs.num_regs = nmatch;
+ regs.start = TALLOC (nmatch * 2, regoff_t);
+ if (regs.start == NULL)
+ return (int) REG_NOMATCH;
+ regs.end = regs.start + nmatch;
+ }
+
+ /* Perform the searching operation. */
+ ret = re_search (&private_preg, string, len,
+ /* start: */ 0, /* range: */ len,
+ want_reg_info ? ®s : (struct re_registers *) 0);
+
+ /* Copy the register information to the POSIX structure. */
+ if (want_reg_info)
+ {
+ if (ret >= 0)
+ {
+ unsigned r;
+
+ for (r = 0; r < nmatch; r++)
+ {
+ pmatch[r].rm_so = regs.start[r];
+ pmatch[r].rm_eo = regs.end[r];
+ }
+ }
+
+ /* If we needed the temporary register info, free the space now. */
+ free (regs.start);
+ }
+
+ /* We want zero return to mean success, unlike `re_search'. */
+ return ret >= 0 ? (int) REG_NOERROR : (int) REG_NOMATCH;
+}
+#ifdef _LIBC
+weak_alias (__regexec, regexec)
+#endif
+
+
+/* Returns a message corresponding to an error code, ERRCODE, returned
+ from either regcomp or regexec. We don't use PREG here. */
+
+size_t
+regerror (errcode, preg, errbuf, errbuf_size)
+ int errcode;
+ const regex_t *preg;
+ char *errbuf;
+ size_t errbuf_size;
+{
+ const char *msg;
+ size_t msg_size;
+
+ if (errcode < 0
+ || errcode >= (int) (sizeof (re_error_msgid_idx)
+ / sizeof (re_error_msgid_idx[0])))
+ /* Only error codes returned by the rest of the code should be passed
+ to this routine. If we are given anything else, or if other regex
+ code generates an invalid error code, then the program has a bug.
+ Dump core so we can fix it. */
+ abort ();
+
+ msg = gettext (re_error_msgid + re_error_msgid_idx[errcode]);
+
+ msg_size = strlen (msg) + 1; /* Includes the null. */
+
+ if (errbuf_size != 0)
+ {
+ if (msg_size > errbuf_size)
+ {
+#if defined HAVE_MEMPCPY || defined _LIBC
+ *((char *) __mempcpy (errbuf, msg, errbuf_size - 1)) = '\0';
+#else
+ memcpy (errbuf, msg, errbuf_size - 1);
+ errbuf[errbuf_size - 1] = 0;
+#endif
+ }
+ else
+ memcpy (errbuf, msg, msg_size);
+ }
+
+ return msg_size;
+}
+#ifdef _LIBC
+weak_alias (__regerror, regerror)
+#endif
+
+
+/* Free dynamically allocated space used by PREG. */
+
+void
+regfree (preg)
+ regex_t *preg;
+{
+ if (preg->buffer != NULL)
+ free (preg->buffer);
+ preg->buffer = NULL;
+
+ preg->allocated = 0;
+ preg->used = 0;
+
+ if (preg->fastmap != NULL)
+ free (preg->fastmap);
+ preg->fastmap = NULL;
+ preg->fastmap_accurate = 0;
+
+ if (preg->translate != NULL)
+ free (preg->translate);
+ preg->translate = NULL;
+}
+#ifdef _LIBC
+weak_alias (__regfree, regfree)
+#endif
+
+#endif /* not emacs */
+#else /* !defined(USE_LIB_REGEX) */
+char regex_d1[] = "d"; char *regex_d2 = regex_d1;
+#endif /* defined(USE_LIB_REGEX) */
--- /dev/null
+/*
+ * rmnt.c -- readmnt() function for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(USE_LIB_READMNT)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: rmnt.c,v 1.12 2008/10/21 16:13:23 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+
+/*
+ * The caller may define:
+ *
+ * 1. An RMNT_EXPDEV macro to expand (ala EP/IX) device numbers;
+ *
+ * EP/IX, for example, uses:
+ *
+ * #define RMNT_EXPDEV(n) expdev(n)
+ *
+ * 2. A custom macro, MNTSKIP, for making decisions to skip entries
+ * -- e.g., ones whose mnt_type is MNTTYPE_IGNORE.
+ *
+ * 3. RMNT_FSTYPE to specify the member name of the character string of the
+ * mntent structure containing the file system type, and MOUNTS_FSTYPE to
+ * specify the member name of the character string pointer of the local
+ * mounts structure where RMNT_FSTYPE is to be copied.
+ *
+ * 4. RMNT_STAT_FSTYPE to specify the member name of the stat structure
+ * containing an integer file system type, and MOUNTS_STAT_FSTYPE to
+ * specify the member name of the integer in the local mounts structure
+ * where RMNT_STAT_FSTYPE is to be copied.
+ *
+ */
+
+#if !defined(RMNT_EXPDEV)
+#define RMNT_EXPDEV(n) n
+#endif /* !defined(RMNT_EXPDEV) */
+
+
+/*
+ * Local static definitions
+ */
+
+static struct mounts *Lmi = (struct mounts *)NULL; /* local mount info */
+static int Lmist = 0; /* Lmi status */
+
+
+/*
+ * readmnt() - read mount table
+ */
+
+struct mounts *
+readmnt()
+{
+ char *dn = (char *)NULL;
+ char *ln;
+ FILE *mfp;
+ struct mntent *mp;
+ struct mounts *mtp;
+ char *opt, *opte;
+ struct stat sb;
+
+ if (Lmi || Lmist)
+ return(Lmi);
+/*
+ * Open access to the mount table.
+ */
+ if (!(mfp = setmntent(MOUNTED, "r"))) {
+ (void) fprintf(stderr, "%s: can't access %s\n", Pn, MOUNTED);
+ Exit(1);
+ }
+/*
+ * Read mount table entries.
+ */
+ while ((mp = getmntent(mfp))) {
+
+#if defined(MNTSKIP)
+ /*
+ * Specfy in the MNTSKIP macro the decisions needed to determine
+ * that this entry should be skipped.
+ *
+ * Typically entries whose mnt_type is MNTTYPE_IGNORE are skipped.
+ *
+ * The MNTSKIP macro allows the caller to use other tests.
+ */
+ MNTSKIP
+#endif /* MNTSKIP */
+
+ /*
+ * Interpolate a possible symbolic directory link.
+ */
+ if (dn)
+ (void) free((FREE_P *)dn);
+ if (!(dn = mkstrcpy(mp->mnt_dir, (MALLOC_S *)NULL)))
+ goto no_space_for_mount;
+ if (!(ln = Readlink(dn))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ " Output information may be incomplete.\n");
+ continue;
+ }
+ if (ln != dn) {
+ (void) free((FREE_P *)dn);
+ dn = ln;
+ }
+ if (*dn != '/')
+ continue;
+ /*
+ * Stat() the directory.
+ */
+ if (statsafely(dn, &sb)) {
+ if (!Fwarn) {
+ (void) fprintf(stderr, "%s: WARNING: can't stat() ", Pn);
+ safestrprt(mp->mnt_type, stderr, 0);
+ (void) fprintf(stderr, " file system ");
+ safestrprt(mp->mnt_dir, stderr, 1);
+ (void) fprintf(stderr,
+ " Output information may be incomplete.\n");
+ }
+ if ((opt = strstr(mp->mnt_opts, "dev="))) {
+ (void) zeromem(&sb, sizeof(sb));
+ if ((opte = x2dev(opt + 4, (dev_t *)&sb.st_dev))) {
+ sb.st_mode = S_IFDIR | 0777;
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ " assuming \"%.*s\" from %s\n",
+ (int)(opte - opt), opt, MOUNTED);
+ } else
+ opt = (char *)NULL;
+ }
+ if (!opt)
+ continue;
+ }
+ /*
+ * Allocate and fill a local mounts structure with the directory
+ * (mounted) information.
+ */
+ if (!(mtp = (struct mounts *)malloc(sizeof(struct mounts)))) {
+
+no_space_for_mount:
+
+ (void) fprintf(stderr, "%s: no space for mount at ", Pn);
+ safestrprt(mp->mnt_fsname, stderr, 0);
+ (void) fprintf(stderr, " (");
+ safestrprt(mp->mnt_dir, stderr, 0);
+ (void) fprintf(stderr, ")\n");
+ Exit(1);
+ }
+ mtp->dir = dn;
+ dn = (char *)NULL;
+ mtp->next = Lmi;
+ mtp->dev = RMNT_EXPDEV(sb.st_dev);
+ mtp->rdev = RMNT_EXPDEV(sb.st_rdev);
+ mtp->inode = (INODETYPE)sb.st_ino;
+ mtp->mode = sb.st_mode;
+
+# if defined(RMNT_FSTYPE) && defined(MOUNTS_FSTYPE)
+ /*
+ * Make a copy of RMNT_FSTYPE in MOUNTS_FSTYPE.
+ */
+ if (!(mtp->MOUNTS_FSTYPE = mkstrcpy(mp->RMNT_FSTYPE,
+ (MALLOC_S *)NULL)))
+ {
+ (void) fprintf(stderr, "%s: no space for fstype (%s): %s\n",
+ Pn, mtp->dir, mp->RMNT_FSTYPE);
+ Exit(1);
+ }
+ (void) strcpy(mtp->MOUNTS_FSTYPE, mp->RMNT_FSTYPE);
+# endif /* defined(RMNT_FSTYP) && defined(MOUNTS_FSTYP) */
+
+# if defined(RMNT_STAT_FSTYPE) && defined(MOUNTS_STAT_FSTYPE)
+ /*
+ * Make a copy of RMNT_STAT_FSTYPE in MOUNTS_STAT_FSTYPE.
+ */
+ mtp->MOUNTS_STAT_FSTYPE = (int)sb.RMNT_STAT_FSTYPE;
+# endif /* defined(RMNT_STAT_FSTYP) && defined(MOUNTS_STAT_FSTYP) */
+
+ /*
+ * Interpolate a possible file system (mounted-on device) name link.
+ */
+ if (!(dn = mkstrcpy(mp->mnt_fsname, (MALLOC_S *)NULL)))
+ goto no_space_for_mount;
+ mtp->fsname = dn;
+ ln = Readlink(dn);
+ dn = (char *)NULL;
+ /*
+ * Stat() the file system (mounted-on) name and add file system
+ * information to the local mounts structure.
+ */
+ if (!ln || statsafely(ln, &sb))
+ sb.st_mode = 0;
+ mtp->fsnmres = ln;
+ mtp->fs_mode = sb.st_mode;
+ Lmi = mtp;
+ }
+ (void) endmntent(mfp);
+/*
+ * Clean up and return the local nount info table address.
+ */
+ if (dn)
+ (void) free((FREE_P *)dn);
+ Lmist = 1;
+ return(Lmi);
+}
+#else /* !defined(USE_LIB_READMNT) */
+char rmnt_d1[] = "d"; char *rmnt_d2 = rmnt_d1;
+#endif /* defined(USE_LIB_READMNT) */
--- /dev/null
+/*
+ * rnam.c -- BSD format name cache functions for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(HASNCACHE) && defined(USE_LIB_RNAM)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: rnam.c,v 1.11 2008/10/21 16:13:23 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+/*
+ * rnam.c - read BSD format (struct namecache or nch) name cache
+ *
+ * This code is effective only when HASNCACHE is defined.
+ */
+
+/*
+ * The caller must:
+ *
+ * #include the relevant header file -- e.g., <sys/namei.h>.
+ *
+ * Define X_NCACHE as the nickname for the kernel cache address.
+ *
+ * Define X_NCSIZE as the nickname for the size of the kernel cache.
+ *
+ * Define NCACHE_NXT if the kernel's name cache is a linked list, starting
+ * at the X_NCACHE address, rather than a table, starting at that address.
+ *
+ * Define NCACHE_NO_ROOT if the calling dialect doesn't support
+ * the locating of the root node of a file system.
+ *
+ * Define the name of the name cache structure -- e.g.,
+ *
+ * #define NCACHE <structure name>
+ *
+ * Define the following casts, if they differ from the defaults:
+ *
+ * NCACHE_SZ_CAST cast for X_NCSIZE (default int)
+ *
+ * e.g.,
+ * #define NCACHE_SZ_CAST unsigned long
+ *
+ * Define the names of these elements of struct NCACHE:
+ *
+ * must #define NCACHE_NM <name>
+ * must #define NCACHE_NMLEN <name length
+ * optional #define NCACHE_NXT <link to next entry>
+ * must #define NCACHE_NODEADDR <node address>
+ * must #define NCACHE_NODEID <node capability ID)
+ * optional #define NCACHE_PARADDR <parent node address>
+ * optional #define NCACHE_PARID <parent node capability ID)
+ *
+ * The caller may need to:
+ *
+ * Define NCHNAMLEN as the length of the name element of NCACHE, if it's
+ * not defined in the header file that defines the NCACHE structure.
+ *
+ * Define this prototype for ncache_load():
+ *
+ * _PROTOTYPE(static void ncache_load,(void));
+ */
+
+
+/*
+ * Local static values
+ */
+
+static int Mch; /* name cache hash mask */
+
+# if !defined(NCACHE_NC_CAST)
+#define NCACHE_SZ_CAST int
+# endif /* !defined(NCACHE_NC_CAST) */
+
+static NCACHE_SZ_CAST Nc = 0; /* size of name cache */
+static int Nch = 0; /* size of name cache hash pointer
+ * table */
+struct l_nch {
+ KA_T na; /* node address */
+
+# if defined(NCACHE_NODEID)
+ unsigned long id; /* capability ID */
+# endif /* defined(NCACHE_NODEID) */
+
+# if defined(NCACHE_PARADDR) && defined(NCACHE_PARID)
+ KA_T pa; /* parent node address */
+ struct l_nch *pla; /* parent local node address */
+ unsigned long did; /* parent capability ID */
+# endif /* defined(NCACHE_PARADDR) && defined(NCACHE_PARID) */
+
+ char nm[NCHNAMLEN+1]; /* name */
+ int nl; /* name length */
+};
+
+static struct l_nch *Ncache = (struct l_nch*)NULL;
+ /* the local name cache */
+static struct l_nch **Nchash = (struct l_nch **)NULL;
+ /* Ncache hash pointers */
+static int Ncfirst = 1; /* first-call status */
+
+# if defined(NCACHE_NODEID)
+#define ncachehash(i,n) Nchash+(((((int)(n)>>2)+((int)(i)))*31415)&Mch)
+_PROTOTYPE(static struct l_nch *ncache_addr,(unsigned long i, KA_T na));
+# else /* !defined(NCACHE_NODEID) */
+#define ncachehash(n) Nchash+((((int)(n)>>2)*31415)&Mch)
+_PROTOTYPE(static struct l_nch *ncache_addr,(KA_T na));
+# endif /* defined(NCACHE_NODEID) */
+
+#define DEFNCACHESZ 1024 /* local size if X_NCSIZE kernel value < 1 */
+#define LNCHINCRSZ 64 /* local size increment */
+
+# if !defined(NCACHE_NO_ROOT)
+_PROTOTYPE(static int ncache_isroot,(KA_T na, char *cp));
+# endif /* !defined(NCACHE_NO_ROOT) */
+
+
+/*
+ * ncache_addr() - look up a node's local ncache address
+ */
+
+static struct l_nch *
+
+# if defined(NCACHE_NODEID)
+ncache_addr(i, na)
+ unsigned long i; /* node's capability ID */
+# else /* !defined(NCACHE_NODEID) */
+ncache_addr(na)
+# endif /* defined(NCACHE_NODEID) */
+
+ KA_T na; /* node's address */
+{
+ struct l_nch **hp;
+
+# if defined(NCACHE_NODEID)
+ for (hp = ncachehash(i, na); *hp; hp++)
+# else /* !defined(NCACHE_NODEID) */
+ for (hp = ncachehash(na); *hp; hp++)
+# endif /* defined(NCACHE_NODEID) */
+
+ {
+
+# if defined(NCACHE_NODEID)
+ if ((*hp)->id == i && (*hp)->na == na)
+# else /* !defined(NCACHE_NODEID) */
+ if ((*hp)->na == na)
+# endif /* defined(NCACHE_NODEID) */
+
+ return(*hp);
+ }
+ return((struct l_nch *)NULL);
+}
+
+
+# if !defined(NCACHE_NO_ROOT)
+/*
+ * ncache_isroot() - is head of name cache path a file system root?
+ */
+
+static int
+ncache_isroot(na, cp)
+ KA_T na; /* kernel node address */
+ char *cp; /* partial path */
+{
+ char buf[MAXPATHLEN];
+ int i;
+ MALLOC_S len;
+ struct mounts *mtp;
+ static int nca = 0;
+ static int ncn = 0;
+ static KA_T *nc = (KA_T *)NULL;
+ struct stat sb;
+ struct vnode v;
+
+ if (!na)
+ return(0);
+/*
+ * Search the root vnode cache.
+ */
+ for (i = 0; i < ncn; i++) {
+ if (na == nc[i])
+ return(1);
+ }
+/*
+ * Read the vnode and see if it's a VDIR node with the VROOT flag set. If
+ * it is, then the path is complete.
+ *
+ * If it isn't, and if the file has an inode number, search the mount table
+ * and see if the file system's inode number is known. If it is, form the
+ * possible full path, safely stat() it, and see if it's inode number matches
+ * the one we have for this file. If it does, then the path is complete.
+ */
+ if (kread((KA_T)na, (char *)&v, sizeof(v))
+ || v.v_type != VDIR || !(v.v_flag & VROOT)) {
+
+ /*
+ * The vnode tests failed. Try the inode tests.
+ */
+ if (Lf->inp_ty != 1 || !Lf->inode
+ || !Lf->fsdir || (len = strlen(Lf->fsdir)) < 1)
+ return(0);
+ if ((len + 1 + strlen(cp) + 1) > sizeof(buf))
+ return(0);
+ for (mtp = readmnt(); mtp; mtp = mtp->next) {
+ if (!mtp->dir || !mtp->inode)
+ continue;
+ if (strcmp(Lf->fsdir, mtp->dir) == 0)
+ break;
+ }
+ if (!mtp)
+ return(0);
+ (void) strcpy(buf, Lf->fsdir);
+ if (buf[len - 1] != '/')
+ buf[len++] = '/';
+ (void) strcpy(&buf[len], cp);
+ if (statsafely(buf, &sb) != 0
+ || (unsigned long)sb.st_ino != Lf->inode)
+ return(0);
+ }
+/*
+ * Add the node address to the root node cache.
+ */
+ if (ncn >= nca) {
+ if (!nca) {
+ len = (MALLOC_S)(10 * sizeof(KA_T));
+ nc = (KA_T *)malloc(len);
+ } else {
+ len = (MALLOC_S)((nca + 10) * sizeof(KA_T));
+ nc = (KA_T *)realloc(nc, len);
+ }
+ if (!nc) {
+ (void) fprintf(stderr, "%s: no space for root node table\n",
+ Pn);
+ Exit(1);
+ }
+ nca += 10;
+ }
+ nc[ncn++] = na;
+ return(1);
+}
+# endif /* !defined(NCACHE_NO_ROOT) */
+
+
+/*
+ * ncache_load() - load the kernel's name cache
+ */
+
+void
+ncache_load()
+{
+ struct l_nch **hp, *lc;
+ int i, len, n;
+ static int iNc = 0;
+ struct NCACHE *kc;
+ static KA_T kp = (KA_T)NULL;
+ KA_T v;
+
+# if defined(NCACHE_NXT)
+ static KA_T kf;
+ struct NCACHE nc;
+# else /* !defined NCACHE_NXT) */
+ static struct NCACHE *kca = (struct NCACHE *)NULL;
+# endif /* defined(NCACHE_NXT) */
+
+ if (!Fncache)
+ return;
+ if (Ncfirst) {
+
+ /*
+ * Do startup (first-time) functions.
+ */
+ Ncfirst = 0;
+ /*
+ * Establish kernel cache size.
+ */
+ v = (KA_T)0;
+ if (get_Nl_value(X_NCSIZE, (struct drive_Nl *)NULL, &v) < 0
+ || !v
+ || kread((KA_T)v, (char *)&Nc, sizeof(Nc)))
+ {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read name cache size: %s\n",
+ Pn, print_kptr(v, (char *)NULL, 0));
+ iNc = Nc = 0;
+ return;
+ }
+ iNc = Nc;
+ if (Nc < 1) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: kernel name cache size: %d\n", Pn, Nc);
+ (void) fprintf(stderr,
+ " Cache size assumed to be: %d\n", DEFNCACHESZ);
+ }
+ iNc = Nc = DEFNCACHESZ;
+ }
+ /*
+ * Establish kernel cache address.
+ */
+ v = (KA_T)0;
+ if (get_Nl_value(X_NCACHE, (struct drive_Nl *)NULL, &v) < 0
+ || !v
+ || kread((KA_T)v, (char *)&kp, sizeof(kp))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read name cache address: %s\n",
+ Pn, print_kptr(v, (char *)NULL, 0));
+ iNc = Nc = 0;
+ return;
+ }
+
+# if defined(NCACHE_NXT)
+ kf = kp;
+
+# else /* !defined(NCACHE_NXT) */
+ /*
+ * Allocate space for a local copy of the kernel's cache.
+ */
+ len = Nc * sizeof(struct NCACHE);
+ if (!(kca = (struct NCACHE *)malloc((MALLOC_S)len))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: can't allocate name cache space: %d\n", Pn, len);
+ Exit(1);
+ }
+# endif /* defined(NCACHE_NXT) */
+
+ /*
+ * Allocate space for the local cache.
+ */
+ len = Nc * sizeof(struct l_nch);
+ if (!(Ncache = (struct l_nch *)malloc((MALLOC_S)len))) {
+
+no_local_space:
+
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: no space for %d byte local name cache\n", Pn, len);
+ Exit(1);
+ }
+ } else {
+
+ /*
+ * Do setup for repeat calls.
+ */
+ if ((Nc = iNc) == 0)
+ return;
+ if (Nchash) {
+ (void) free((FREE_P *)Nchash);
+ Nchash = (struct l_nch **)NULL;
+ }
+
+# if defined(NCACHE_NXT)
+ kp = kf;
+# endif /* defined(NCACHE_NXT) */
+
+ }
+
+# if !defined(NCACHE_NXT)
+
+/*
+ * Read the kernel's name cache.
+ */
+ if (kread(kp, (char *)kca, (Nc * sizeof(struct NCACHE)))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read kernel's name cache: %s\n",
+ Pn, print_kptr(kp, (char *)NULL, 0));
+ Nc = 0;
+ return;
+ }
+# endif /* !defined(NCACHE_NXT) */
+
+/*
+ * Build a local copy of the kernel name cache.
+ */
+
+# if defined(NCACHE_NXT)
+ for (i = iNc * 16, kc = &nc, lc = Ncache, n = 0; kp; )
+# else /* !defined(NCACHE_NXT) */
+ for (i = n = 0, kc = kca, lc = Ncache; i < Nc; i++, kc++)
+# endif /* defined(NCACHE_NXT) */
+
+ {
+
+# if defined(NCACHE_NXT)
+ if (kread(kp, (char *)kc, sizeof(nc)))
+ break;
+ if ((kp = (KA_T)kc->NCACHE_NXT) == kf)
+ kp = (KA_T)NULL;
+# endif /* defined(NCACHE_NXT) */
+
+ if (!kc->NCACHE_NODEADDR)
+ continue;
+ if ((len = kc->NCACHE_NMLEN) < 1 || len > NCHNAMLEN)
+ continue;
+ if (len < 3 && kc->NCACHE_NM[0] == '.') {
+ if (len == 1 || (len == 2 && kc->NCACHE_NM[1] == '.'))
+ continue;
+ }
+
+# if defined(NCACHE_NXT)
+ if (n >= Nc) {
+ Nc += LNCHINCRSZ;
+ if (!(Ncache = (struct l_nch *)realloc(Ncache,
+ (MALLOC_S)(Nc * sizeof(struct l_nch)))))
+ {
+ (void) fprintf(stderr,
+ "%s: no more space for %d entry local name cache\n",
+ Pn, Nc);
+ Exit(1);
+ }
+ lc = &Ncache[n];
+ }
+# endif /* defined(NCACHE_NXT) */
+
+# if defined(NCACHE_NODEID)
+ lc->na = (KA_T)kc->NCACHE_NODEADDR;
+ lc->id = kc->NCACHE_NODEID;
+# endif /* defined(NCACHE_NODEID) */
+
+# if defined(NCACHE_PARADDR)
+ lc->pa = (KA_T)kc->NCACHE_PARADDR;
+ lc->pla = (struct l_nch *)NULL;
+# endif /* defined(NCACHE_PARADDR) */
+
+# if defined(NCACHE_PARID)
+ lc->did = kc->NCACHE_PARID;
+# endif /* defined(NCACHE_PARID) */
+
+ (void) strncpy(lc->nm, kc->NCACHE_NM, len);
+ lc->nm[len] = '\0';
+ lc->nl = strlen(lc->nm);
+ n++;
+ lc++;
+
+# if defined(NCACHE_NXT)
+ if (n >= i) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: name cache truncated at %d entries\n",
+ Pn, n);
+ break;
+ }
+# endif /* defined(NCACHE_NXT) */
+
+ }
+/*
+ * Reduce memory usage, as required.
+ */
+
+# if !defined(NCACHE_NXT)
+ if (!RptTm)
+ (void) free((FREE_P *)kca);
+# endif /* !defined(NCACHE_NXT) */
+
+ if (n < 1) {
+ Nc = 0;
+ if (!RptTm) {
+ (void) free((FREE_P *)Ncache);
+ Ncache = (struct l_nch *)NULL;
+ }
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: unusable name cache size: %d\n", Pn, n);
+ return;
+ }
+ if (n < Nc) {
+ Nc = n;
+ if (!RptTm) {
+ len = Nc * sizeof(struct l_nch);
+ if (!(Ncache = (struct l_nch *)realloc(Ncache, len)))
+ goto no_local_space;
+ }
+ }
+/*
+ * Build a hash table to locate Ncache entries.
+ */
+ for (Nch = 1; Nch < Nc; Nch <<= 1)
+ ;
+ Nch <<= 1;
+ Mch = Nch - 1;
+ if (!(Nchash = (struct l_nch **)calloc(Nch+Nc, sizeof(struct l_nch *))))
+ {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: no space for %d name cache hash pointers\n",
+ Pn, Nch + Nc);
+ Exit(1);
+ }
+ for (i = 0, lc = Ncache; i < Nc; i++, lc++) {
+
+# if defined(NCACHE_NODEID)
+ for (hp = ncachehash(lc->id, lc->na), n = 1; *hp; hp++)
+# else /* defined(NCACHE_NODEID) */
+ for (hp = ncachehash(lc->na), n = 1; *hp; hp++)
+# endif /* defined(NCACHE_NODEID) */
+
+ {
+
+# if defined(NCACHE_NODEID)
+ if ((*hp)->na == lc->na && (*hp)->id == lc->id
+# else /* defined(NCACHE_NODEID) */
+ if ((*hp)->na == lc->na
+# endif /* defined(NCACHE_NODEID) */
+
+ && strcmp((*hp)->nm, lc->nm) == 0
+
+# if defined(NCACHE_PARADDR) && defined(NCACHE_PARID)
+ && (*hp)->pa == lc->pa && (*hp)->did == lc->did
+# endif /* defined(NCACHE_PARADDR) && defined(NCACHE_PARID) */
+
+ ) {
+ n = 0;
+ break;
+ }
+ }
+ if (n)
+ *hp = lc;
+ }
+
+# if defined(NCACHE_PARADDR) && defined(NCACHE_PARID)
+/*
+ * Make a final pass through the local cache and convert parent node
+ * addresses to local name cache pointers.
+ */
+ for (i = 0, lc = Ncache; i < Nc; i++, lc++) {
+ if (!lc->pa)
+ continue;
+ lc->pla = ncache_addr(lc->did, lc->pa);
+ }
+# endif /* defined(NCACHE_PARADDR) && defined(NCACHE_PARID) */
+}
+
+
+/*
+ * ncache_lookup() - look up a node's name in the kernel's name cache
+ */
+
+char *
+ncache_lookup(buf, blen, fp)
+ char *buf; /* receiving name buffer */
+ int blen; /* receiving buffer length */
+ int *fp; /* full path reply */
+{
+ char *cp = buf;
+ struct l_nch *lc;
+ struct mounts *mtp;
+ int nl, rlen;
+
+ *cp = '\0';
+ *fp = 0;
+
+# if defined(HASFSINO)
+/*
+ * If the entry has an inode number that matches the inode number of the
+ * file system mount point, return an empty path reply. That tells the
+ * caller to print the file system mount point name only.
+ */
+ if ((Lf->inp_ty == 1) && Lf->fs_ino && (Lf->inode == Lf->fs_ino))
+ return(cp);
+# endif /* defined(HASFSINO) */
+
+/*
+ * Look up the name cache entry for the node address.
+ */
+
+# if defined(NCACHE_NODEID)
+ if (Nc == 0 || !(lc = ncache_addr(Lf->id, Lf->na)))
+# else /* defined(NCACHE_NODEID) */
+ if (Nc == 0 || !(lc = ncache_addr(Lf->na)))
+# endif /* defined(NCACHE_NODEID) */
+
+
+ {
+
+ /*
+ * If the node has no cache entry, see if it's the mount
+ * point of a known file system.
+ */
+ if (!Lf->fsdir || !Lf->dev_def || Lf->inp_ty != 1)
+ return((char *)NULL);
+ for (mtp = readmnt(); mtp; mtp = mtp->next) {
+ if (!mtp->dir || !mtp->inode)
+ continue;
+ if (Lf->dev == mtp->dev
+ && mtp->inode == Lf->inode
+ && strcmp(mtp->dir, Lf->fsdir) == 0)
+ return(cp);
+ }
+ return((char *)NULL);
+ }
+/*
+ * Start the path assembly.
+ */
+ if ((nl = lc->nl) > (blen - 1))
+ return((char *)NULL);
+ cp = buf + blen - nl - 1;
+ rlen = blen - nl - 1;
+ (void) strcpy(cp, lc->nm);
+
+# if defined(NCACHE_PARADDR) && defined(NCACHE_PARID)
+/*
+ * Look up the name cache entries that are parents of the node address.
+ * Quit when:
+ *
+ * there's no parent;
+ * the name length is too large to fit in the receiving buffer.
+ */
+ for (;;) {
+ if (!lc->pla) {
+
+# if !defined(NCACHE_NO_ROOT)
+ if (ncache_isroot(lc->pa, cp))
+ *fp = 1;
+# endif /* !defined(NCACHE_NO_ROOT) */
+
+ break;
+ }
+ lc = lc->pla;
+ if (((nl = lc->nl) + 1) > rlen)
+ break;
+ *(cp - 1) = '/';
+ cp--;
+ rlen--;
+ (void) strncpy((cp - nl), lc->nm, nl);
+ cp -= nl;
+ rlen -= nl;
+ }
+# endif /* defined(NCACHE_PARADDR) && defined(NCACHE_PARID) */
+ return(cp);
+}
+#else /* !defined(HASNCACHE) || !defined(USE_LIB_RNAM) */
+char rnam_d1[] = "d"; char *rnam_d2 = rnam_d1;
+#endif /* defined(HASNCACHE) && defined(USE_LIB_RNAM) */
--- /dev/null
+/*
+ * rnch.c -- Sun format name cache functions for lsof library
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(HASNCACHE) && defined(USE_LIB_RNCH)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: rnch.c,v 1.11 2008/10/21 16:13:23 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+/*
+ * rnch.c - read Sun format (struct ncache) name cache
+ *
+ * This code is effective only when HASNCACHE is defined.
+ */
+
+/*
+ * The caller must:
+ *
+ * #include the relevant header file -- e.g., <sys/dnlc.h>.
+ *
+ * Define X_NCSIZE as the nickname for the kernel cache size variable,
+ * or, if X_NCSIZE is undefined, define FIXED_NCSIZE as the size of the
+ * kernel cache.
+ *
+ * Define X_NCACHE as the nickname for the kernel cache address and
+ * define ADDR_NCACHE if the address is the address of the cache,
+ * rather than the address of a pointer to it.
+ *
+ * Define NCACHE_NXT if the kernel's name cache is a linked list, starting
+ * at the X_NCACHE address, rather than a table, starting at that address.
+ *
+ * Define any of the following casts that differ from their defaults:
+ *
+ * NCACHE_SZ_CAST cast for X_NCACHE (default int)
+ *
+ * The caller may:
+ *
+ * Define NCACHE_DP as the name of the element in the
+ * ncache structure that contains the
+ * parent vnode pointer.
+ *
+ * Default: dp
+ *
+ * Define NCACHE_NAME as the name of the element in the
+ * ncache structure that contains the
+ * name.
+ *
+ * Default: name
+ *
+ * Define NCACHE_NAMLEN as the name of the element in the
+ * ncache structure that contains the
+ * name length.
+ *
+ * Deafult: namlen
+ *
+ * Define NCACHE_NEGVN as the name of the name list element
+ * whose value is a vnode address to
+ * ignore when loading the kernel name
+ * cache.
+ *
+ * Define NCACHE_NODEID as the name of the element in the
+ * ncache structure that contains the
+ * vnode's capability ID.
+ *
+ * Define NCACHE_PARID as the name of the element in the
+ * ncache structure that contains the
+ * parent vnode's capability ID.
+ *
+ * Define NCACHE_VP as the name of the element in the
+ * ncache structure that contains the
+ * vnode pointer.
+ *
+ * Default: vp
+ *
+ * Note: if NCACHE_NODEID is defined, then NCACHE_PARID must be defined.
+ *
+ *
+ * The caller must:
+ *
+ * Define this prototype for ncache_load():
+ *
+ * _PROTOTYPE(void ncache_load,(void));
+ */
+
+
+/*
+ * Local static values
+ */
+
+static int Mch; /* name cache hash mask */
+
+# if !defined(NCACHE_NC_CAST)
+#define NCACHE_SZ_CAST int
+# endif /* !defined(NCACHE_NC_CAST) */
+
+static NCACHE_SZ_CAST Nc = 0; /* size of name cache */
+static int Nch = 0; /* size of name cache hash pointer
+ * table */
+struct l_nch {
+ KA_T vp; /* vnode address */
+ KA_T dp; /* parent vnode address */
+ struct l_nch *pa; /* parent Ncache address */
+
+# if defined(NCACHE_NODEID)
+ unsigned long id; /* node's capability ID */
+ unsigned long did; /* parent node's capability ID */
+# endif /* defined(NCACHE_NODEID) */
+
+ char *nm; /* name */
+ int nl; /* name length */
+};
+
+static struct l_nch *Ncache = (struct l_nch *)NULL;
+ /* the local name cache */
+static struct l_nch **Nchash = (struct l_nch **)NULL;
+ /* Ncache hash pointers */
+static int Ncfirst = 1; /* first-call status */
+
+# if defined(NCACHE_NEGVN)
+static KA_T NegVN = (KA_T)NULL; /* negative vnode address */
+static int NegVNSt = 0; /* NegVN status: 0 = not loaded */
+# endif /* defined(NCACHE_NEGVN) */
+
+# if defined(NCACHE_NODEID)
+_PROTOTYPE(static struct l_nch *ncache_addr,(unsigned long i, KA_T v));
+#define ncachehash(i,v) Nchash+(((((int)(v)>>2)+((int)(i)))*31415)&Mch)
+# else /* !defined(NCACHE_NODEID) */
+_PROTOTYPE(static struct l_nch *ncache_addr,(KA_T v));
+#define ncachehash(v) Nchash+((((int)(v)>>2)*31415)&Mch)
+# endif /* defined(NCACHE_NODEID) */
+
+_PROTOTYPE(static int ncache_isroot,(KA_T va, char *cp));
+
+#define DEFNCACHESZ 1024 /* local size if X_NCSIZE kernel value < 1 */
+#define LNCHINCRSZ 64 /* local size increment */
+
+# if !defined(NCACHE_DP)
+#define NCACHE_DP dp
+# endif /* !defined(NCACHE_DP) */
+
+# if !defined(NCACHE_NAME)
+#define NCACHE_NAME name
+# endif /* !defined(NCACHE_NAME) */
+
+# if !defined(NCACHE_NAMLEN)
+#define NCACHE_NAMLEN namlen
+# endif /* !defined(NCACHE_NAMLEN) */
+
+# if !defined(NCACHE_VP)
+#define NCACHE_VP vp
+# endif /* !defined(NCACHE_VP) */
+
+
+/*
+ * ncache_addr() - look up a node's local ncache address
+ */
+
+static struct l_nch *
+
+# if defined(NCACHE_NODEID)
+ncache_addr(i, v)
+# else /* !defined(NCACHE_NODEID) */
+ncache_addr(v)
+# endif /* defined(NCACHE_NODEID) */
+
+# if defined(NCACHE_NODEID)
+ unsigned long i; /* capability ID */
+# endif /* defined(NCACHE_NODEID) */
+
+ KA_T v; /* vnode's address */
+{
+ struct l_nch **hp;
+
+# if defined(NCACHE_NODEID)
+ for (hp = ncachehash(i, v); *hp; hp++)
+# else /* !defined(NCACHE_NODEID) */
+ for (hp = ncachehash(v); *hp; hp++)
+# endif /* defined(NCACHE_NODEID) */
+
+ {
+
+# if defined(NCACHE_NODEID)
+ if ((*hp)->vp == v && (*hp)->id == i)
+# else /* !defined(NCACHE_NODEID) */
+ if ((*hp)->vp == v)
+# endif /* defined(NCACHE_NODEID) */
+
+ return(*hp);
+ }
+ return((struct l_nch *)NULL);
+}
+
+
+/*
+ * ncache_isroot() - is head of name cache path a file system root?
+ */
+
+static int
+ncache_isroot(va, cp)
+ KA_T va; /* kernel vnode address */
+ char *cp; /* partial path */
+{
+ char buf[MAXPATHLEN];
+ int i;
+ MALLOC_S len;
+ struct mounts *mtp;
+ struct stat sb;
+ struct vnode v;
+ static int vca = 0;
+ static int vcn = 0;
+ static KA_T *vc = (KA_T *)NULL;
+
+ if (!va)
+ return(0);
+/*
+ * Search the root vnode cache.
+ */
+ for (i = 0; i < vcn; i++) {
+ if (va == vc[i])
+ return(1);
+ }
+/*
+ * Read the vnode and see if it's a VDIR node with the VROOT flag set. If
+ * it is, then the path is complete.
+ *
+ * If it isn't, and if the file has an inode number, search the mount table
+ * and see if the file system's inode number is known. If it is, form the
+ * possible full path, safely stat() it, and see if it's inode number matches
+ * the one we have for this file. If it does, then the path is complete.
+ */
+ if (kread((KA_T)va, (char *)&v, sizeof(v))
+ || v.v_type != VDIR || !(v.v_flag & VROOT)) {
+
+ /*
+ * The vnode tests failed. Try the inode tests.
+ */
+ if (Lf->inp_ty != 1 || !Lf->inode
+ || !Lf->fsdir || (len = strlen(Lf->fsdir)) < 1)
+ return(0);
+ if ((len + 1 + strlen(cp) + 1) > sizeof(buf))
+ return(0);
+ for (mtp = readmnt(); mtp; mtp = mtp->next) {
+ if (!mtp->dir || !mtp->inode)
+ continue;
+ if (strcmp(Lf->fsdir, mtp->dir) == 0)
+ break;
+ }
+ if (!mtp)
+ return(0);
+ (void) strcpy(buf, Lf->fsdir);
+ if (buf[len - 1] != '/')
+ buf[len++] = '/';
+ (void) strcpy(&buf[len], cp);
+ if (statsafely(buf, &sb) != 0
+ || (unsigned long)sb.st_ino != Lf->inode)
+ return(0);
+ }
+/*
+ * Add the vnode address to the root vnode cache.
+ */
+ if (vcn >= vca) {
+ vca += 10;
+ len = (MALLOC_S)(vca * sizeof(KA_T));
+ if (!vc)
+ vc = (KA_T *)malloc(len);
+ else
+ vc = (KA_T *)realloc(vc, len);
+ if (!vc) {
+ (void) fprintf(stderr, "%s: no space for root vnode table\n",
+ Pn);
+ Exit(1);
+ }
+ }
+ vc[vcn++] = va;
+ return(1);
+}
+
+
+/*
+ * ncache_load() - load the kernel's name cache
+ */
+
+void
+ncache_load()
+{
+ char *cp, *np;
+ struct l_nch **hp, *lc;
+ int i, len, n;
+ static int iNc = 0;
+ struct ncache *kc;
+ static KA_T kp = (KA_T)NULL;
+ KA_T v;
+
+# if defined(HASDNLCPTR)
+ static int na = 0;
+ static char *nb = (char *)NULL;
+# endif /* defined(HASDNLCPTR) */
+
+# if defined(NCACHE_NXT)
+ static KA_T kf;
+ struct ncache nc;
+# else /* !defined(NCACHE_NXT) */
+ static struct ncache *kca = (struct ncache *)NULL;
+# endif /* defined(NCACHE_NXT) */
+
+ if (!Fncache)
+ return;
+ if (Ncfirst) {
+
+ /*
+ * Do startup (first-time) functions.
+ */
+ Ncfirst = 0;
+ /*
+ * Establish kernel cache size.
+ */
+
+# if defined(X_NCSIZE)
+ v = (KA_T)0;
+ if (get_Nl_value(X_NCSIZE, (struct drive_Nl *)NULL, &v) < 0
+ || !v
+ || kread((KA_T)v, (char *)&Nc, sizeof(Nc)))
+ {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read name cache size: %s\n",
+ Pn, print_kptr(v, (char *)NULL, 0));
+ iNc = Nc = 0;
+ return;
+ }
+ iNc = Nc;
+# else /* !defined(X_NCSIZE) */
+ iNc = Nc = FIXED_NCSIZE;
+# endif /* defined(X_NCSIZE) */
+
+ if (Nc < 1) {
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: WARNING: kernel name cache size: %d\n", Pn, Nc);
+ (void) fprintf(stderr,
+ " Cache size assumed to be: %d\n", DEFNCACHESZ);
+ }
+ iNc = Nc = DEFNCACHESZ;
+ }
+
+# if defined(NCACHE_NEGVN)
+ /*
+ * Get negative vnode address.
+ */
+ if (!NegVNSt) {
+ if (get_Nl_value(NCACHE_NEGVN, (struct drive_Nl *)NULL, &NegVN)
+ < 0)
+ NegVN = (KA_T)NULL;
+ NegVNSt = 1;
+ }
+# endif /* defined(NCACHE_NEGVN) */
+
+ /*
+ * Establish kernel cache address.
+ */
+
+# if defined(ADDR_NCACHE)
+ kp = (KA_T)0;
+ if (get_Nl_value(X_NCACHE,(struct drive_Nl *)NULL,(KA_T *)&kp) < 0
+ || !kp) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: no name cache address\n", Pn);
+ iNc = Nc = 0;
+ return;
+ }
+# else /* !defined(ADDR_NCACHE) */
+ v = (KA_T)0;
+ if (get_Nl_value(X_NCACHE, (struct drive_Nl *)NULL, &v) < 0
+ || !v
+ || kread((KA_T)v, (char *)&kp, sizeof(kp))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read name cache ptr: %s\n",
+ Pn, print_kptr(v, (char *)NULL, 0));
+ iNc = Nc = 0;
+ return;
+ }
+# endif /* defined(ADDR_NCACHE) */
+
+ /*
+ * Allocate space for a local copy of the kernel's cache.
+ */
+
+# if !defined(NCACHE_NXT)
+ len = Nc * sizeof(struct ncache);
+ if (!(kca = (struct ncache *)malloc((MALLOC_S)len))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: can't allocate name cache space: %d\n", Pn, len);
+ Exit(1);
+ }
+# endif /* !defined(NCACHE_NXT) */
+
+ /*
+ * Allocate space for the local cache.
+ */
+ len = Nc * sizeof(struct l_nch);
+ if (!(Ncache = (struct l_nch *)calloc(Nc, sizeof(struct l_nch)))) {
+
+no_local_space:
+
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: no space for %d byte local name cache\n", Pn, len);
+ Exit(1);
+ }
+ } else {
+
+ /*
+ * Do setup for repeat calls.
+ */
+ if (!iNc)
+ return;
+ if (Nchash) {
+ (void) free((FREE_P *)Nchash);
+ Nchash = (struct l_nch **)NULL;
+ }
+ if (Ncache) {
+
+ /*
+ * Free space malloc'd to names in local name cache.
+ */
+ for (i = 0, lc = Ncache; i < Nc; i++, lc++) {
+ if (lc->nm) {
+ (void) free((FREE_P *)lc->nm);
+ lc->nm = (char *)NULL;
+ }
+ }
+ }
+ Nc = iNc;
+
+# if defined(NCACHE_NXT)
+ kp = kf;
+# endif /* defined(NCACHE_NXT) */
+
+ }
+
+# if !defined(NCACHE_NXT)
+
+/*
+ * Read the kernel's name cache.
+ */
+ if (kread(kp, (char *)kca, (Nc * sizeof(struct ncache)))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read kernel's name cache: %s\n",
+ Pn, print_kptr(kp, (char *)NULL, 0));
+ Nc = 0;
+ return;
+ }
+# endif /* !defined(NCACHE_NXT) */
+
+/*
+ * Build a local copy of the kernel name cache.
+ */
+
+# if defined(NCACHE_NXT)
+ for (i = iNc * 16, kc = &nc, kf = kp, lc = Ncache, n = 0; kp; )
+# else /* !defined(NCACHE_NXT) */
+ for (i = n = 0, kc = kca, lc = Ncache; i < Nc; i++, kc++)
+# endif /* defined(NCACHE_NXT) */
+
+ {
+
+# if defined(NCACHE_NXT)
+ if (kread(kp, (char *)kc, sizeof(nc)))
+ break;
+ if ((kp = (KA_T)kc->NCACHE_NXT) == kf)
+ kp = (KA_T)NULL;
+# endif /* defined(NCACHE_NXT) */
+
+ if (!kc->NCACHE_VP || (len = kc->NCACHE_NAMLEN) < 1)
+ continue;
+
+# if defined(NCACHE_NEGVN)
+ if (NegVN && ((KA_T)kc->NCACHE_VP == NegVN))
+ continue;
+# endif /* defined(NCACHE_NEGVN) */
+
+# if defined(HASDNLCPTR)
+ /*
+ * Read name from kernel to a temporary buffer.
+ */
+ if (len > na) {
+ na = len;
+ if (!nb)
+ nb = (char *)malloc(na);
+ else
+ nb = (char *)realloc((MALLOC_P *)nb, na);
+ if (!nb) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d byte temporary name buffer\n",
+ Pn, na);
+ Exit(1);
+ }
+ }
+ if (!kc->NCACHE_NAME || kread((KA_T)kc->NCACHE_NAME, nb, len))
+ continue;
+ np = nb;
+# else /* !defined(HASDNLCPTR) */
+ /*
+ * Use name that is in the kernel cache entry.
+ */
+ if (len > NC_NAMLEN)
+ continue;
+ np = kc->NCACHE_NAME;
+# endif /* defined(HASDNLCPTR) */
+
+ if (len < 3 && *np == '.') {
+ if (len == 1 || (len == 2 && np[1] == '.'))
+ continue;
+ }
+ /*
+ * Allocate space for name in local cache entry.
+ */
+ if (!(cp = (char *)malloc(len + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for name cache name: %s\n",
+ Pn, len + 1, np);
+ Exit(1);
+ }
+ (void) strncpy(cp, np, len);
+ cp[len] = '\0';
+
+# if defined(NCACHE_NXT)
+ if (n >= Nc) {
+
+ /*
+ * Allocate more local space to receive the kernel's linked
+ * entries.
+ */
+ Nc += LNCHINCRSZ;
+ if (!(Ncache = (struct l_nch *)realloc(Ncache,
+ (MALLOC_S)(Nc * sizeof(struct l_nch)))))
+ {
+ (void) fprintf(stderr,
+ "%s: no more space for %d entry local name cache\n",
+ Pn, Nc);
+ Exit(1);
+ }
+ lc = &Ncache[n];
+ iNc = Nc;
+ }
+# endif /* defined(NCACHE_NXT) */
+
+ /*
+ * Complete the local cache entry.
+ */
+ lc->vp = (KA_T)kc->NCACHE_VP;
+ lc->dp = (KA_T)kc->NCACHE_DP;
+ lc->pa = (struct l_nch *)NULL;
+ lc->nm = cp;
+ lc->nl = len;
+
+# if defined(NCACHE_NODEID)
+ lc->id = (unsigned long)kc->NCACHE_NODEID;
+ lc->did = (unsigned long)kc->NCACHE_PARID;
+# endif /* defined(NCACHE_NODEID) */
+
+ n++;
+ lc++;
+
+# if defined(NCACHE_NXT)
+ if (n >= i) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: name cache truncated at %d entries\n",
+ Pn, n);
+ break;
+ }
+# endif /* defined(NCACHE_NXT) */
+
+ }
+/*
+ * Reduce memory usage, as required.
+ */
+
+# if !defined(NCACHE_NXT)
+ if (!RptTm)
+ (void) free((FREE_P *)kca);
+# endif /* !defined(NCACHE_NXT) */
+
+ if (n < 1) {
+ if (!RptTm && Ncache) {
+
+ /*
+ * If not in repeat mode, free the space that has been malloc'd
+ * to the local name cache.
+ */
+ for (i = 0, lc = Ncache; i < Nc; i++, lc++) {
+ if (lc->nm) {
+ (void) free((FREE_P *)lc->nm);
+ lc->nm = (char *)NULL;
+ }
+ }
+ (void) free((FREE_P *)Ncache);
+ Ncache = (struct l_nch *)NULL;
+ Nc = 0;
+ }
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: unusable name cache size: %d\n", Pn, n);
+ return;
+ }
+ if (n < Nc) {
+ Nc = n;
+ if (!RptTm) {
+ len = Nc * sizeof(struct l_nch);
+ if (!(Ncache = (struct l_nch *)realloc(Ncache, len)))
+ goto no_local_space;
+ }
+ }
+/*
+ * Build a hash table to locate Ncache entries.
+ */
+ for (Nch = 1; Nch < Nc; Nch <<= 1)
+ ;
+ Nch <<= 1;
+ Mch = Nch - 1;
+ if (!(Nchash = (struct l_nch **)calloc(Nch+Nc, sizeof(struct l_nch *))))
+ {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: no space for %d name cache hash pointers\n",
+ Pn, Nch + Nc);
+ Exit(1);
+ }
+ for (i = 0, lc = Ncache; i < Nc; i++, lc++) {
+
+# if defined(NCACHE_NODEID)
+ for (hp = ncachehash(lc->id, lc->vp), n = 1; *hp; hp++)
+# else /* !defined(NCACHE_NODEID) */
+ for (hp = ncachehash(lc->vp), n = 1; *hp; hp++)
+# endif /* defined(NCACHE_NODEID) */
+
+ {
+ if ((*hp)->vp == lc->vp && strcmp((*hp)->nm, lc->nm) == 0
+ && (*hp)->dp == lc->dp
+
+# if defined(NCACHE_NODEID)
+ && (*hp)->id == lc->id && (*hp)->did == lc->did
+# endif /* defined(NCACHE_NODEID) */
+
+ ) {
+ n = 0;
+ break;
+ }
+ }
+ if (n)
+ *hp = lc;
+ }
+/*
+ * Make a final pass through the local cache and convert parent vnode
+ * addresses to local name cache pointers.
+ */
+ for (i = 0, lc = Ncache; i < Nc; i++, lc++) {
+ if (!lc->dp)
+ continue;
+
+# if defined(NCACHE_NEGVN)
+ if (NegVN && (lc->dp == NegVN)) {
+ lc->pa = (struct l_nch *)NULL;
+ continue;
+ }
+# endif /* defined(NCACHE_NEGVN) */
+
+# if defined(NCACHE_NODEID)
+ lc->pa = ncache_addr(lc->did, lc->dp);
+# else /* !defined(NCACHE_NODEID) */
+ lc->pa = ncache_addr(lc->dp);
+# endif /* defined(NCACHE_NODEID) */
+
+ }
+}
+
+
+/*
+ * ncache_lookup() - look up a node's name in the kernel's name cache
+ */
+
+char *
+ncache_lookup(buf, blen, fp)
+ char *buf; /* receiving name buffer */
+ int blen; /* receiving buffer length */
+ int *fp; /* full path reply */
+{
+ char *cp = buf;
+ struct l_nch *lc;
+ struct mounts *mtp;
+ int nl, rlen;
+
+ *cp = '\0';
+ *fp = 0;
+
+# if defined(HASFSINO)
+/*
+ * If the entry has an inode number that matches the inode number of the
+ * file system mount point, return an empty path reply. That tells the
+ * caller to print the file system mount point name only.
+ */
+ if ((Lf->inp_ty == 1) && Lf->fs_ino && (Lf->inode == Lf->fs_ino))
+ return(cp);
+# endif /* defined(HASFSINO) */
+
+/*
+ * Look up the name cache entry for the node address.
+ */
+ if (!Nc
+
+# if defined(NCACHE_NODEID)
+ || !(lc = ncache_addr(Lf->id, Lf->na))
+# else /* !defined(NCACHE_NODEID) */
+ || !(lc = ncache_addr(Lf->na))
+# endif /* defined(NCACHE_NODEID) */
+
+ ) {
+
+ /*
+ * If the node has no cache entry, see if it's the mount
+ * point of a known file system.
+ */
+ if (!Lf->fsdir || !Lf->dev_def || Lf->inp_ty != 1)
+ return((char *)NULL);
+ for (mtp = readmnt(); mtp; mtp = mtp->next) {
+ if (!mtp->dir || !mtp->inode)
+ continue;
+ if (Lf->dev == mtp->dev
+ && mtp->inode == Lf->inode
+ && strcmp(mtp->dir, Lf->fsdir) == 0)
+ return(cp);
+ }
+ return((char *)NULL);
+ }
+/*
+ * Begin the path assembly.
+ */
+ if ((nl = lc->nl) > (blen - 1))
+ return((char *)NULL);
+ cp = buf + blen - nl - 1;
+ rlen = blen - nl - 1;
+ (void) strcpy(cp, lc->nm);
+/*
+ * Look up the name cache entries that are parents of the node address.
+ * Quit when:
+ *
+ * there's no parent;
+ * the name is too large to fit in the receiving buffer.
+ */
+ for (;;) {
+ if (!lc->pa) {
+ if (ncache_isroot(lc->dp, cp))
+ *fp = 1;
+ break;
+ }
+ lc = lc->pa;
+ if (((nl = lc->nl) + 1) > rlen)
+ break;
+ *(cp - 1) = '/';
+ cp--;
+ rlen--;
+ (void) strncpy((cp - nl), lc->nm, nl);
+ cp -= nl;
+ rlen -= nl;
+ }
+ return(cp);
+}
+#else /* !defined(HASNCACHE) || !defined(USE_LIB_RNCH) */
+char rnch_d1[] = "d"; char *rnch_d2 = rnch_d1;
+#endif /* defined(HASNCACHE) && defined(USE_LIB_RNCH) */
--- /dev/null
+/*
+ * rnmh.c -- functions to read BSD format name cache information from a
+ * kernel hash table
+ */
+
+
+/*
+ * Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#include "../machine.h"
+
+#if defined(HASNCACHE) && defined(USE_LIB_RNMH)
+
+# if !defined(lint)
+static char copyright[] =
+"@(#) Copyright 1997 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: rnmh.c,v 1.13 2008/10/21 16:13:23 abe Exp $";
+# endif /* !defined(lint) */
+
+#include "../lsof.h"
+
+
+/*
+ * rnmh.c - read BSD format hashed kernel name cache
+ */
+
+/*
+ * The caller must:
+ *
+ * #include the relevant header file -- e.g., <sys/namei.h>.
+ *
+ * Define X_NCACHE as the nickname for the kernel cache hash tables
+ * address.
+ *
+ * Define X_NCSIZE as the nickname for the size of the kernel cache has
+ * table length.
+ *
+ * Define NCACHE_NO_ROOT if the calling dialect doesn't support
+ * the locating of the root node of a file system.
+ *
+ * Define the name of the name cache structure -- e.g.,
+ *
+ * #define NCACHE <structure name>
+ *
+ *
+ * Define the following casts, if they differ from the defaults:
+ *
+ * NCACHE_SZ_CAST case for X_NCSIZE (default unsigned long)
+ *
+ * Define the names of these elements of struct NCACHE:
+ *
+ * #define NCACHE_NM <name>
+ * #define NCACHE_NXT <link to next entry>
+ * #define NCACHE_NODEADDR <node address>
+ * #define NCACHE_PARADDR <parent node address>
+ *
+ * Optionally define:
+ *
+ * #define NCACHE_NMLEN <name length>
+ *
+ * Optionally define *both*:
+ *
+ * #define NCACHE_NODEID <node capability ID>
+ * #define NCACHE_PARID <parent node capability ID>
+ *
+ * The caller may need to:
+ *
+ * Define this prototype for ncache_load():
+ *
+ * _PROTOTYPE(static void ncache_load,(void));
+ *
+ * Define NCACHE_VROOT to be the value of the flag that signifies that
+ * the vnode is the root of its file system.
+ *
+ * E.g., for BSDI >= 5:
+ *
+ * #define NCACHE_VROOT VV_ROOT
+ *
+ * If not defined, NCACHE_VROOT is defined as "VROOT".
+ *
+ * Define VNODE_VFLAG if the vnode's flag member's name isn't v_flag.
+ *
+ * Note: if NCHNAMLEN is defined, the name is assumed to be in
+ * NCACHE_NM[NCHNAMLEN]; if it isn't defined, the name is assumed to be in an
+ * extension that begins at NCACHE_NM[0].
+ *
+ * Note: if NCACHE_NMLEN is not defined, then NCACHE_NM must be a pointer to
+ * a kernel allocated, NUL-terminated, string buffer.
+ */
+
+
+/*
+ * Casts
+ */
+
+# if !defined(NCACHE_NC_CAST)
+#define NCACHE_SZ_CAST unsigned long
+# endif /* !defined(NCACHE_NC_CAST) */
+
+
+/*
+ * Flags
+ */
+
+# if !defined(NCACHE_NMLEN)
+#undef NCHNAMLEN
+# endif /* !defined(NCACHE_NMLEN) */
+
+# if !defined(NCACHE_VROOT)
+#define NCACHE_VROOT VROOT /* vnode is root of its file system */
+# endif /* !defined(NCACHE_VROOT) */
+
+# if !defined(VNODE_VFLAG)
+#define VNODE_VFLAG v_flag
+# endif /* !defined(VNODE_VFLAG) */
+
+
+/*
+ * Local static values
+ */
+
+static int Mch; /* name cache hash mask */
+
+struct l_nch {
+ KA_T na; /* node address */
+ KA_T pa; /* parent node address */
+ struct l_nch *pla; /* parent local node address */
+ int nl; /* name length */
+ struct l_nch *next; /* next entry */
+
+# if defined(NCACHE_NODEID)
+ unsigned long id; /* capability ID */
+ unsigned long did; /* parent capability ID */
+# endif /* defined(NCACHE_NODEID) */
+
+# if defined(NCHNAMLEN)
+ char nm[NCHNAMLEN + 1]; /* name */
+# else /* !defined(NCHNAMLEN) */
+ char nm[1]; /* variable length name */
+# endif /* defined(NCHNAMLEN) */
+
+};
+
+static struct l_nch *Ncache = (struct l_nch *)NULL;
+ /* the head of the local name cache */
+static struct l_nch **Nchash = (struct l_nch **)NULL;
+ /* Ncache hash pointers */
+
+# if defined(NCACHE_NODEID)
+#define ncachehash(i,n) Nchash+(((((int)(n)>>2)+((int)(i)))*31415)&Mch)
+_PROTOTYPE(static struct l_nch *ncache_addr,(unsigned long i, KA_T na));
+# else /* !defined(NCACHE_NODEID) */
+#define ncachehash(n) Nchash+((((int)(n)>>2)*31415)&Mch)
+_PROTOTYPE(static struct l_nch *ncache_addr,(KA_T na));
+# endif /* defined(NCACHE_NODEID) */
+
+# if !defined(NCACHE_NO_ROOT)
+_PROTOTYPE(static int ncache_isroot,(KA_T na, char *cp));
+# endif /* !defined(NCACHE_NO_ROOT) */
+
+
+/*
+ * ncache_addr() - look up a node's local ncache address
+ */
+
+static struct l_nch *
+
+# if defined(NCACHE_NODEID)
+ncache_addr(i, na)
+ unsigned long i; /* node's capability ID */
+# else /* !defined(NCACHE_NODEID) */
+ncache_addr(na)
+# endif /* defined(NCACHE_NODEID) */
+
+ KA_T na; /* node's address */
+{
+ struct l_nch **hp;
+
+# if defined(NCACHE_NODEID)
+ for (hp = ncachehash(i, na); *hp; hp++)
+# else /* !defined(NCACHE_NODEID) */
+ for (hp = ncachehash(na); *hp; hp++)
+# endif /* defined(NCACHE_NODEID) */
+
+ {
+
+# if defined(NCACHE_NODEID)
+ if ((*hp)->id == i && (*hp)->na == na)
+# else /* !defined(NCACHE_NODEID) */
+ if ((*hp)->na == na)
+# endif /* defined(NCACHE_NODEID) */
+
+ return(*hp);
+ }
+ return((struct l_nch *)NULL);
+}
+
+
+# if !defined(NCACHE_NO_ROOT)
+/*
+ * ncache_isroot() - is head of name cache path a file system root?
+ */
+
+static int
+ncache_isroot(na, cp)
+ KA_T na; /* kernel node address */
+ char *cp; /* partial path */
+{
+ char buf[MAXPATHLEN];
+ int i;
+ MALLOC_S len;
+ struct mounts *mtp;
+ static int nca = 0;
+ static int ncn = 0;
+ static KA_T *nc = (KA_T *)NULL;
+ struct stat sb;
+ struct vnode v;
+
+ if (!na)
+ return(0);
+/*
+ * Search the root vnode cache.
+ */
+ for (i = 0; i < ncn; i++) {
+ if (na == nc[i])
+ return(1);
+ }
+/*
+ * Read the vnode and see if it's a VDIR node with the NCACHE_VROOT flag set.
+ * If it is, then the path is complete.
+ *
+ * If it isn't, and if the file has an inode number, search the mount table
+ * and see if the file system's inode number is known. If it is, form the
+ * possible full path, safely stat() it, and see if it's inode number matches
+ * the one we have for this file. If it does, then the path is complete.
+ */
+ if (kread((KA_T)na, (char *)&v, sizeof(v))
+ || v.v_type != VDIR || !(v.VNODE_VFLAG & NCACHE_VROOT)) {
+
+ /*
+ * The vnode tests failed. Try the inode tests.
+ */
+ if (Lf->inp_ty != 1 || !Lf->inode
+ || !Lf->fsdir || (len = strlen(Lf->fsdir)) < 1)
+ return(0);
+ if ((len + 1 + strlen(cp) + 1) > sizeof(buf))
+ return(0);
+ for (mtp = readmnt(); mtp; mtp = mtp->next) {
+ if (!mtp->dir || !mtp->inode)
+ continue;
+ if (strcmp(Lf->fsdir, mtp->dir) == 0)
+ break;
+ }
+ if (!mtp)
+ return(0);
+ (void) strcpy(buf, Lf->fsdir);
+ if (buf[len - 1] != '/')
+ buf[len++] = '/';
+ (void) strcpy(&buf[len], cp);
+ if (statsafely(buf, &sb) != 0
+ || (unsigned long)sb.st_ino != Lf->inode)
+ return(0);
+ }
+/*
+ * Add the node address to the root node cache.
+ */
+ if (ncn >= nca) {
+ if (!nca) {
+ len = (MALLOC_S)(10 * sizeof(KA_T));
+ nc = (KA_T *)malloc(len);
+ } else {
+ len = (MALLOC_S)((nca + 10) * sizeof(KA_T));
+ nc = (KA_T *)realloc(nc, len);
+ }
+ if (!nc) {
+ (void) fprintf(stderr, "%s: no space for root node table\n",
+ Pn);
+ Exit(1);
+ }
+ nca += 10;
+ }
+ nc[ncn++] = na;
+ return(1);
+}
+# endif /* !defined(NCACHE_NO_ROOT) */
+
+
+/*
+ * ncache_load() - load the kernel's name cache
+ */
+
+void
+ncache_load()
+{
+ struct NCACHE c;
+ struct l_nch **hp, *ln;
+ KA_T ka, knx;
+ static struct NCACHE **khp = (struct namecache **)NULL;
+ static int khpl = 0;
+ NCACHE_SZ_CAST khsz;
+ unsigned long kx;
+ static struct l_nch *lc = (struct l_nch *)NULL;
+ static int lcl = 0;
+ int len, lim, n, nch, nchl, nlcl;
+ char tbuf[32];
+ KA_T v;
+
+# if !defined(NCHNAMLEN)
+ int cin = sizeof(c.NCACHE_NM);
+ KA_T nmo = (KA_T)offsetof(struct NCACHE, NCACHE_NM);
+# endif /* !defined(NCHNAMLEN) */
+
+# if !defined(NCACHE_NMLEN)
+ char nbf[MAXPATHLEN + 1];
+ int nbfl = (int)(sizeof(nbf) - 1);
+ KA_T nk;
+ char *np;
+ int rl;
+
+ nbf[nbfl] = '\0';
+# endif /* !defined(NCACHE_NMLEN) */
+
+ if (!Fncache)
+ return;
+/*
+ * Free previously allocated space.
+ */
+ for (lc = Ncache; lc; lc = ln) {
+ ln = lc->next;
+ (void) free((FREE_P *)lc);
+ }
+ Ncache = (struct l_nch *)NULL;
+ if (Nchash)
+ (void) free((FREE_P *)Nchash);
+ Nchash = (struct l_nch **)NULL;
+/*
+ * Get kernel cache hash table size
+ */
+ v = (KA_T)0;
+ if (get_Nl_value(X_NCSIZE, (struct drive_Nl *)NULL, &v) < 0
+ || !v
+ || kread((KA_T)v, (char *)&khsz, sizeof(khsz)))
+ {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: can't read name cache hash size: %s\n",
+ Pn, print_kptr(v, (char *)NULL, 0));
+ return;
+ }
+ if (khsz < 1) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: name cache hash size length error: %#lx\n",
+ Pn, khsz);
+ return;
+ }
+/*
+ * Get kernel cache hash table address.
+ */
+ ka = (KA_T)0;
+ v = (KA_T)0;
+ if (get_Nl_value(X_NCACHE, (struct drive_Nl *)NULL, &v) < 0
+ || !v
+ || kread((KA_T)v, (char *)&ka, sizeof(ka))
+ || !ka)
+ {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: unusable name cache hash pointer: (%s)=%s\n",
+ Pn, print_kptr(v, tbuf, sizeof(tbuf)),
+ print_kptr(ka, (char *)NULL, 0));
+ return;
+ }
+/*
+ * Allocate space for the hash table pointers and read them.
+ */
+ len = (MALLOC_S)(khsz * sizeof(struct NCACHE *));
+ if (len > khpl) {
+ if (khp)
+ khp = (struct NCACHE **)realloc((MALLOC_P *)khp, len);
+ else
+ khp = (struct NCACHE **)malloc(len);
+ if (!khp) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for name cache hash table\n",
+ Pn, len);
+ Exit(1);
+ }
+ khpl = len;
+ }
+ if (kread((KA_T)ka, (char *)khp, len)) {
+ (void) fprintf(stderr,
+ "%s: can't read name cache hash pointers from: %s\n",
+ Pn, print_kptr(ka, (char *)NULL, 0));
+ return;
+ }
+/*
+ * Process the kernel's name cache hash table buckets.
+ */
+ lim = khsz * 10;
+ for (kx = nch = 0; kx < khsz; kx++) {
+
+ /*
+ * Loop through the entries for a hash bucket.
+ */
+ for (ka = (KA_T)khp[kx], n = 0; ka; ka = knx, n++) {
+ if (n > lim) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: name cache hash chain too long\n",
+ Pn);
+ break;
+ }
+ if (kread(ka, (char *)&c, sizeof(c)))
+ break;
+ knx = (KA_T)c.NCACHE_NXT;
+ if (!c.NCACHE_NODEADDR)
+ continue;
+
+# if defined(NCACHE_NMLEN)
+ if ((len = c.NCACHE_NMLEN) < 1)
+ continue;
+# else /* !defined(NCACHE_NMLEN) */
+ /*
+ * If it's possible to read the first four characters of the name,
+ * do so and check for "." and "..".
+ */
+ if (!c.NCACHE_NM
+ || kread((KA_T)c.NCACHE_NM, nbf, 4))
+ continue;
+ if (nbf[0] == '.') {
+ if (!nbf[1]
+ || ((nbf[1] == '.') && !nbf[2]))
+ continue;
+ }
+ /*
+ * Read the rest of the name, 32 characters at a time, until a NUL
+ * character has been read or nbfl characters have been read.
+ */
+ nbf[4] = '\0';
+ if ((len = (int)strlen(nbf)) < 4) {
+ if (!len)
+ continue;
+ } else {
+ for (np = &nbf[4]; len < nbfl; np += rl) {
+ if ((rl = nbfl - len) > 32) {
+ rl = 32;
+ nbf[len + rl] = '\0';
+ }
+ nk = (KA_T)((char *)c.NCACHE_NM + len);
+ if (kread(nk, np, rl)) {
+ rl = -1;
+ break;
+ }
+ rl = (int)strlen(np);
+ len += rl;
+ if (rl < 32)
+ break;
+ }
+ if (rl < 0)
+ continue;
+ }
+# endif /* defined(NCACHE_NMLEN) */
+
+ /*
+ * Allocate a cache entry long enough to contain the name and
+ * move the name to it.
+ */
+
+# if defined(NCHNAMLEN)
+ if (len > NCHNAMLEN)
+ continue;
+ if (len < 3 && c.NCACHE_NM[0] == '.') {
+ if (len == 1 || (len == 2 && c.NCACHE_NM[1] == '.'))
+ continue;
+ }
+ if ((nlcl = sizeof(struct l_nch)) > lcl)
+# else /* !defined(NCHNAMLEN) */
+ if ((nlcl = sizeof(struct l_nch) + len) > lcl)
+# endif /* defined(NCHNAMLEN) */
+
+ {
+ if (lc)
+ lc = (struct l_nch *)realloc(lc, nlcl);
+ else
+ lc = (struct l_nch *)malloc(nlcl);
+ if (!lc) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d local name cache bytes\n",
+ Pn, nlcl);
+ Exit(1);
+ }
+ lcl = nlcl;
+ }
+
+# if defined(NCHNAMLEN)
+ (void) strncpy(lc->nm, c.NCACHE_NM, len);
+# else /* !defined(NCHNAMLEN) */
+# if defined(NCACHE_NMLEN)
+ if ((len < 3) && (cin > 1)) {
+
+ /*
+ * If this is a one or two character name, and if NCACHE_NM[]
+ * in c has room for at least two characters, check for "."
+ * and ".." first, ignoring this entry if the name is either.
+ */
+ if (len < 3 && c.NCACHE_NM[0] == '.') {
+ if (len == 1 || (len == 2 && c.NCACHE_NM[1] == '.'))
+ continue;
+ }
+ }
+ if (len > cin) {
+
+ /*
+ * If not all (possibly not any, depending on the value in
+ * cin) of the name has yet been read to lc->nm[], read it
+ * or the rest of it. If it wasn't possible before to check
+ * for "." or "..", do that. too.
+ */
+ if (cin > 0)
+ (void) strncpy(lc->nm, c.NCACHE_NM, cin);
+ if (kread(ka + (KA_T)(nmo + cin), &lc->nm[cin], len - cin))
+ continue;
+ if ((cin < 2) && (len < 3) && (lc->nm[0] == '.')) {
+ if (len == 1 || (len == 2 && lc->nm[1] == '.'))
+ continue;
+ }
+ } else
+ (void) strncpy(lc->nm, c.NCACHE_NM, len);
+# else /* !defined(NCACHE_NMLEN) */
+ (void) strncpy(lc->nm, nbf, len);
+# endif /* defined(NCACHE_NMLEN) */
+
+# endif /* defined(NCHNAMLEN) */
+ lc->nm[len] = '\0';
+ /*
+ * Complete the new local cache entry and link it to the previous
+ * local cache chain.
+ */
+ lc->next = Ncache;
+ Ncache = lc;
+ lc->na = (KA_T)c.NCACHE_NODEADDR;
+ lc->nl = len;
+ lc->pa = (KA_T)c.NCACHE_PARADDR;
+ lc->pla = (struct l_nch *)NULL;
+
+# if defined(NCACHE_NODEID)
+ lc->id = c.NCACHE_NODEID;
+ lc->did = c.NCACHE_PARID;
+# endif /* defined(NCACHE_NODEID) */
+
+ lcl = 0;
+ lc = (struct l_nch *)NULL;
+ nch++;
+ }
+ }
+/*
+ * Reduce memory usage, as required.
+ */
+ if (!RptTm) {
+ (void) free((FREE_P *)khp);
+ khp = (struct NCACHE **)NULL;
+ khpl = 0;
+ }
+ if (nch < 1) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING: unusable name cache size: %d\n", Pn, nch);
+ return;
+ }
+/*
+ * Build a hash table to locate Ncache entries.
+ */
+ for (nchl = 1; nchl < nch; nchl <<= 1)
+ ;
+ nchl <<= 1;
+ Mch = nchl - 1;
+ len = nchl + nch;
+ if (!(Nchash = (struct l_nch **)calloc(len, sizeof(struct l_nch *)))) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: no space for %d local name cache hash pointers\n",
+ Pn, len);
+ Exit(1);
+ }
+ for (lc = Ncache; lc; lc = lc->next) {
+
+# if defined(NCACHE_NODEID)
+ for (hp = ncachehash(lc->id, lc->na),
+# else /* !defined(NCACHE_NODEID) */
+ for (hp = ncachehash(lc->na),
+# endif /* defined(NCACHE_NODEID) */
+
+ n = 1; *hp; hp++)
+ {
+ if ((*hp)->na == lc->na && strcmp((*hp)->nm, lc->nm) == 0) {
+ n = 0;
+ break;
+ }
+ }
+ if (n)
+ *hp = lc;
+ else
+ lc->pa = (KA_T)0;
+ }
+/*
+ * Make a final pass through the local cache and convert parent node
+ * addresses to local name cache pointers.
+ */
+ for (lc = Ncache; lc; lc = lc->next) {
+ if (!lc->pa)
+ continue;
+
+# if defined(NCACHE_NODEID)
+ lc->pla = ncache_addr(lc->did, lc->pa);
+# else /* !defined(NCACHE_NODEID) */
+ lc->pla = ncache_addr(lc->pa);
+# endif /* defined(NCACHE_NODEID) */
+
+ }
+}
+
+
+/*
+ * ncache_lookup() - look up a node's name in the kernel's name cache
+ */
+
+char *
+ncache_lookup(buf, blen, fp)
+ char *buf; /* receiving name buffer */
+ int blen; /* receiving buffer length */
+ int *fp; /* full path reply */
+{
+ char *cp = buf;
+ struct l_nch *lc;
+ struct mounts *mtp;
+ int nl, rlen;
+
+ *cp = '\0';
+ *fp = 0;
+
+# if defined(HASFSINO)
+/*
+ * If the entry has an inode number that matches the inode number of the
+ * file system mount point, return an empty path reply. That tells the
+ * caller to print the file system mount point name only.
+ */
+ if ((Lf->inp_ty == 1) && Lf->fs_ino && (Lf->inode == Lf->fs_ino))
+ return(cp);
+# endif /* defined(HASFSINO) */
+
+/*
+ * Look up the name cache entry for the node address.
+ */
+
+# if defined(NCACHE_NODEID)
+ if (!Nchash || !(lc = ncache_addr(Lf->id, Lf->na)))
+# else /* !defined(NCACHE_NODEID) */
+ if (!Nchash || !(lc = ncache_addr(Lf->na)))
+# endif /* defined(NCACHE_NODEID) */
+
+ {
+
+ /*
+ * If the node has no cache entry, see if it's the mount
+ * point of a known file system.
+ */
+ if (!Lf->fsdir || !Lf->dev_def || Lf->inp_ty != 1)
+ return((char *)NULL);
+ for (mtp = readmnt(); mtp; mtp = mtp->next) {
+ if (!mtp->dir || !mtp->inode)
+ continue;
+ if (Lf->dev == mtp->dev
+ && mtp->inode == Lf->inode
+ && (strcmp(mtp->dir, Lf->fsdir) == 0))
+ return(cp);
+ }
+ return((char *)NULL);
+ }
+/*
+ * Start the path assembly.
+ */
+ if ((nl = lc->nl) > (blen - 1))
+ return((char *)NULL);
+ cp = buf + blen - nl - 1;
+ rlen = blen - nl - 1;
+ (void) strcpy(cp, lc->nm);
+/*
+ * Look up the name cache entries that are parents of the node address.
+ * Quit when:
+ *
+ * there's no parent;
+ * the name length is too large to fit in the receiving buffer.
+ */
+ for (;;) {
+ if (!lc->pla) {
+
+# if !defined(NCACHE_NO_ROOT)
+ if (ncache_isroot(lc->pa, cp))
+ *fp = 1;
+# endif /* !defined(NCACHE_NO_ROOT) */
+
+ break;
+ }
+ lc = lc->pla;
+ if (((nl = lc->nl) + 1) > rlen)
+ break;
+ *(cp - 1) = '/';
+ cp--;
+ rlen--;
+ (void) strncpy((cp - nl), lc->nm, nl);
+ cp -= nl;
+ rlen -= nl;
+ }
+ return(cp);
+}
+#else /* !defined(HASNCACHE) || !defined(USE_LIB_RNMH) */
+char rnmh_d1[] = "d"; char *rnmh_d2 = rnmh_d1;
+#endif /* defined(HASNCACHE) && defined(USE_LIB_RNMH) */
--- /dev/null
+/*
+ * snpf.c -- snprintf() empulation functions for lsof library
+ *
+ * V. Abell
+ * Purdue University Computing Center
+ */
+
+/*
+ * Copyright 2000 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * This software has been adapted from snprintf.c in sendmail 8.9.3. It
+ * is subject to the sendmail copyright statements listed below, and the
+ * sendmail licensing terms stated in the sendmail LICENSE file comment
+ * section of this file.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#include "../machine.h"
+
+#ifdef USE_LIB_SNPF
+
+/*
+ * Sendmail copyright statements:
+ *
+ * Copyright (c) 1998 Sendmail, Inc. All rights reserved.
+ * Copyright (c) 1997 Eric P. Allman. All rights reserved.
+ * Copyright (c) 1988, 1993
+ * The Regents of the University of California. All rights reserved.
+ *
+ * By using this file, you agree to the terms and conditions set
+ * forth in the LICENSE file which can be found at the top level of
+ * the sendmail distribution.
+ *
+ * The LICENSE file may be found in the following comment section.
+ */
+
+
+/*
+ * Begin endmail LICENSE file.
+
+ SENDMAIL LICENSE
+
+The following license terms and conditions apply, unless a different
+license is obtained from Sendmail, Inc., 1401 Park Avenue, Emeryville, CA
+94608, or by electronic mail at license@sendmail.com.
+
+License Terms:
+
+Use, Modification and Redistribution (including distribution of any
+modified or derived work) in source and binary forms is permitted only if
+each of the following conditions is met:
+
+1. Redistributions qualify as "freeware" or "Open Source Software" under
+ one of the following terms:
+
+ (a) Redistributions are made at no charge beyond the reasonable cost of
+ materials and delivery.
+
+ (b) Redistributions are accompanied by a copy of the Source Code or by an
+ irrevocable offer to provide a copy of the Source Code for up to three
+ years at the cost of materials and delivery. Such redistributions
+ must allow further use, modification, and redistribution of the Source
+ Code under substantially the same terms as this license. For the
+ purposes of redistribution "Source Code" means the complete source
+ code of sendmail including all modifications.
+
+ Other forms of redistribution are allowed only under a separate royalty-
+ free agreement permitting such redistribution subject to standard
+ commercial terms and conditions. A copy of such agreement may be
+ obtained from Sendmail, Inc. at the above address.
+
+2. Redistributions of source code must retain the copyright notices as they
+ appear in each source code file, these license terms, and the
+ disclaimer/limitation of liability set forth as paragraph 6 below.
+
+3. Redistributions in binary form must reproduce the Copyright Notice,
+ these license terms, and the disclaimer/limitation of liability set
+ forth as paragraph 6 below, in the documentation and/or other materials
+ provided with the distribution. For the purposes of binary distribution
+ the "Copyright Notice" refers to the following language:
+ "Copyright (c) 1998 Sendmail, Inc. All rights reserved."
+
+4. Neither the name of Sendmail, Inc. nor the University of California nor
+ the names of their contributors may be used to endorse or promote
+ products derived from this software without specific prior written
+ permission. The name "sendmail" is a trademark of Sendmail, Inc.
+
+5. All redistributions must comply with the conditions imposed by the
+ University of California on certain embedded code, whose copyright
+ notice and conditions for redistribution are as follows:
+
+ (a) Copyright (c) 1988, 1993 The Regents of the University of
+ California. All rights reserved.
+
+ (b) Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions
+ are met:
+
+ (i) Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+
+ (ii) Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following
+ disclaimer in the documentation and/or other materials provided
+ with the distribution.
+
+ (iii) All advertising materials mentioning features or use of this
+ software must display the following acknowledgement: "This
+ product includes software developed by the University of
+ California, Berkeley and its contributors."
+
+ (iv) Neither the name of the University nor the names of its
+ contributors may be used to endorse or promote products derived
+ from this software without specific prior written permission.
+
+6. Disclaimer/Limitation of Liability: THIS SOFTWARE IS PROVIDED BY
+ SENDMAIL, INC. AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED
+ WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN
+ NO EVENT SHALL SENDMAIL, INC., THE REGENTS OF THE UNIVERSITY OF
+ CALIFORNIA OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+ INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+ NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
+ USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+
+(Version 8.6, last updated 6/24/1998)
+
+ * End endmail LICENSE file.
+ */
+
+
+/*
+ * If "ll" format support is not possible -- e.g., the long long type isn't
+ * supported -- define HAS_NO_LONG_LONG.
+ */
+
+# ifndef lint
+static char copyright[] =
+"@(#) Copyright 2000 Purdue Research Foundation.\nAll rights reserved.\n";
+# endif /* !defined(lint) */
+
+#include <varargs.h>
+
+#if defined(__STDC__)
+#define _PROTOTYPE(function, params) function params
+#else /* !defined(__STDC__) */
+#define _PROTOTYPE(function, params) function()
+#endif /* defined(__STDC__) */
+
+
+/*
+** SNPRINTF, VSNPRINT -- counted versions of printf
+**
+** These versions have been grabbed off the net. They have been
+** cleaned up to compile properly and support for .precision and
+** %lx has been added.
+*/
+
+/**************************************************************
+ * Original:
+ * Patrick Powell Tue Apr 11 09:48:21 PDT 1995
+ * A bombproof version of doprnt (dopr) included.
+ * Sigh. This sort of thing is always nasty do deal with. Note that
+ * the version here does not include floating point...
+ *
+ * snprintf() is used instead of sprintf() as it does limit checks
+ * for string length. This covers a nasty loophole.
+ *
+ * The other functions are there to prevent NULL pointers from
+ * causing nast effects.
+ **************************************************************/
+
+/*static char _id[] = "$Id: snpf.c,v 1.5 2008/10/21 16:13:23 abe Exp $";*/
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void dopr,(char *bp, char *ep, char *fmt, va_list args));
+_PROTOTYPE(static void dopr_outch,(char **bp, char *ep, int c));
+_PROTOTYPE(static void dostr,(char **bp, char *ep, char *str, int));
+
+# if !defined(HAS_NO_LONG_LONG)
+_PROTOTYPE(static void fmtllnum,(char **bp, char *ep, long long value,
+ int base, int dosign, int ljust, int len,
+ int zpad));
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+_PROTOTYPE(static void fmtnum,(char **bp, char *ep, long value, int base,
+ int dosign, int ljust, int len, int zpad));
+_PROTOTYPE(static void fmtstr,(char **bp, char *ep, char *value, int ljust,
+ int len, int zpad,
+ int maxwidth));
+
+
+/*
+ * Local variables
+ */
+
+static int Length;
+
+
+/*
+ * snpf() -- count-controlled sprintf()
+ */
+
+int
+snpf(va_alist)
+ va_dcl /* requires at least three arguments:
+ * bp = receiving buffer pointer
+ * ct = length of buffer
+ * fmt = format string
+ */
+{
+ va_list args;
+ char *bp, *fmt;
+ int ct, len;
+
+ va_start(args);
+ bp = va_arg(args, char *);
+ ct = va_arg(args, int);
+ fmt = va_arg(args, char *);
+ len = vsnpf(bp, ct, fmt, args);
+ va_end(args);
+ return(len);
+}
+
+
+/*
+ * vsnpf() -- count-controlled vsprintf()
+ */
+
+int
+vsnpf(str, count, fmt, args)
+ char *str; /* result buffer */
+ int count; /* size of buffer */
+ char *fmt; /* format */
+ va_list args; /* variable length argument list */
+{
+ char *ep = str + count - 1;
+
+ *str = '\0';
+ (void) dopr(str, ep, fmt, args);
+ if (count > 0)
+ *ep = '\0';
+ return(Length);
+}
+
+
+/*
+ * dopr() -- poor man's version of doprintf
+ */
+
+
+static void
+dopr(bp, ep, fmt, args)
+ char *bp; /* buffer start */
+ char *ep; /* buffer end (start + length - 1) */
+ char *fmt; /* format */
+ va_list args; /* variable length argument list */
+{
+ int ch;
+ char ebuf[64];
+ int ebufl = (int)(sizeof(ebuf) - 1);
+ long value;
+ int longflag = 0;
+ int longlongflag = 0;
+ int pointflag = 0;
+ int maxwidth = 0;
+ char *strvalue;
+ int ljust;
+ int len;
+ int zpad;
+ int zxflag = 0;
+
+# if !defined(HAS_NO_LONG_LONG)
+ long long llvalue;
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+ Length = 0;
+ while((ch = *fmt++)) {
+ switch (ch) {
+ case '%':
+ ljust = len = zpad = zxflag = maxwidth = 0;
+ longflag = longlongflag = pointflag = 0;
+
+nextch:
+
+ ch = *fmt++;
+ switch (ch) {
+ case '\0':
+ dostr(&bp, ep, "**end of format**" , 0);
+ return;
+ case '-':
+ ljust = 1;
+ goto nextch;
+ case '0': /* set zero padding if len not set */
+ if ((len == 0) && !pointflag)
+ zpad = '0';
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ if (pointflag)
+ maxwidth = (maxwidth * 10) + (int)(ch - '0');
+ else
+ len = (len * 10) + (int)(ch - '0');
+ goto nextch;
+ case '*':
+ if (pointflag)
+ maxwidth = va_arg(args, int);
+ else
+ len = va_arg(args, int);
+ goto nextch;
+ case '#':
+ zxflag = 1;
+ goto nextch;
+ case '.':
+ pointflag = 1;
+ goto nextch;
+ case 'l':
+ if (longflag) {
+ longflag = 0;
+ longlongflag = 1;
+ goto nextch;
+ }
+ longflag = 1;
+ goto nextch;
+ case 'u':
+ case 'U':
+ if (longlongflag) {
+
+# if !defined(HAS_NO_LONG_LONG)
+ llvalue = va_arg(args, long long);
+ (void) fmtllnum(&bp,ep,llvalue,10,0,ljust,len,zpad);
+# else /* defined(HAS_NO_LONG_LONG) */
+ (void) strncpy(ebuf, "ll is unsupported", ebufl);
+ ebuf[(int)ebufl] = '\0';
+ (void) dostr(&bp, ep, ebuf, 0);
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+ break;
+ }
+ if (longflag)
+ value = va_arg(args, long);
+ else
+ value = va_arg(args, int);
+ (void) fmtnum(&bp, ep, value, 10,0, ljust, len, zpad);
+ break;
+ case 'o':
+ case 'O':
+ if (longlongflag) {
+
+# if !defined(HAS_NO_LONG_LONG)
+ llvalue = va_arg(args, long long);
+ (void) fmtllnum(&bp,ep,llvalue,8,0,ljust,len,zpad);
+# else /* defined(HAS_NO_LONG_LONG) */
+ (void) strncpy(ebuf, "ll is unsupported", ebufl);
+ ebuf[(int)ebufl] = '\0';
+ (void) dostr(&bp, ep, ebuf, 0);
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+ break;
+ }
+ if (longflag)
+ value = va_arg(args, long);
+ else
+ value = va_arg(args, int);
+ (void) fmtnum(&bp, ep, value, 8,0, ljust, len, zpad);
+ break;
+ case 'd':
+ case 'D':
+ if (longlongflag) {
+
+# if !defined(HAS_NO_LONG_LONG)
+ llvalue = va_arg(args, long long);
+ (void) fmtllnum(&bp,ep,llvalue,10,1,ljust,len,zpad);
+# else /* defined(HAS_NO_LONG_LONG) */
+ (void) strncpy(ebuf, "ll is unsupported", ebufl);
+ ebuf[(int)ebufl] = '\0';
+ (void) dostr(&bp, ep, ebuf, 0);
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+ break;
+ }
+ if (longflag)
+ value = va_arg(args, long);
+ else
+ value = va_arg(args, int);
+ (void) fmtnum(&bp, ep, value, 10,1, ljust, len, zpad);
+ break;
+ case 'x':
+ if (longlongflag) {
+
+# if !defined(HAS_NO_LONG_LONG)
+ llvalue = va_arg(args, long long);
+ if (zxflag && llvalue) {
+ (void) dostr(&bp, ep, "0x", 0);
+ if (len >= 2)
+ len -= 2;
+ }
+ (void) fmtllnum(&bp,ep,llvalue,16,0,ljust,len,zpad);
+# else /* defined(HAS_NO_LONG_LONG) */
+ (void) strncpy(ebuf, "ll is unsupported", ebufl);
+ ebuf[(int)ebufl] = '\0';
+ (void) dostr(&bp, ep, ebuf, 0);
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+ break;
+ }
+ if (longflag)
+ value = va_arg(args, long);
+ else
+ value = va_arg(args, int);
+ if (zxflag && value) {
+ (void) dostr(&bp, ep, "0x", 0);
+ if (len >= 2)
+ len -= 2;
+ }
+ (void) fmtnum(&bp, ep, value, 16,0, ljust, len, zpad);
+ break;
+ case 'X':
+ if (longlongflag) {
+
+# if !defined(HAS_NO_LONG_LONG)
+ llvalue = va_arg(args, long long);
+ if (zxflag && llvalue) {
+ (void) dostr(&bp, ep, "0x", 0);
+ if (len >= 2)
+ len -= 2;
+ }
+ (void) fmtllnum(&bp,ep,llvalue,-16,0,ljust,len,zpad);
+# else /* defined(HAS_NO_LONG_LONG) */
+ (void) strncpy(ebuf, "ll is unsupported", ebufl);
+ ebuf[(int)ebufl] = '\0';
+ (void) dostr(&bp, ep, ebuf, 0);
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+ break;
+ }
+ if (longflag)
+ value = va_arg(args, long);
+ else
+ value = va_arg(args, int);
+ if (zxflag && value) {
+ (void) dostr(&bp, ep, "0x", 0);
+ if (len >= 2)
+ len -= 2;
+ }
+ (void) fmtnum(&bp, ep, value,-16,0, ljust, len, zpad);
+ break;
+ case 's':
+ strvalue = va_arg(args, char *);
+ if (maxwidth > 0 || !pointflag) {
+ if (pointflag && len > maxwidth)
+ len = maxwidth; /* Adjust padding */
+ (void) fmtstr(&bp, ep, strvalue, ljust, len, zpad,
+ maxwidth);
+ }
+ break;
+ case 'c':
+ ch = va_arg(args, int);
+ dopr_outch(&bp, ep, ch);
+ break;
+ case '%':
+ (void) dopr_outch(&bp, ep, ch);
+ continue;
+ default:
+ ebuf[0] = ch;
+ (void) strncpy(&ebuf[1], " is unsupported", ebufl);
+ ebuf[(int)ebufl] = '\0';
+ (void) dostr(&bp, ep, ebuf, 0);
+ }
+ break;
+ default:
+ (void) dopr_outch(&bp, ep, ch);
+ break;
+ }
+ }
+ *bp = '\0';
+}
+
+
+# if !defined(HAS_NO_LONG_LONG)
+/*
+ * fmtllnum() -- format long long number for output
+ */
+
+static void
+fmtllnum(bp, ep, value, base, dosign, ljust, len, zpad)
+ char **bp; /* current buffer pointer */
+ char *ep; /* end of buffer (-1) */
+ long long value; /* number to format */
+ int base; /* number base */
+ int dosign; /* sign request */
+ int ljust; /* left justfication request */
+ int len; /* length request */
+ int zpad; /* zero padding request */
+{
+ int signvalue = 0;
+ unsigned long long uvalue;
+ char convert[20];
+ int place = 0;
+ int padlen = 0; /* amount to pad */
+ int caps = 0;
+
+ uvalue = value;
+ if (dosign) {
+ if (value < 0) {
+ signvalue = '-';
+ uvalue = -value;
+ }
+ }
+ if (base < 0) {
+ caps = 1;
+ base = -base;
+ }
+ do {
+ convert[place++] =
+ (caps ? "0123456789ABCDEF" : "0123456789abcdef")
+ [uvalue % (unsigned)base];
+ uvalue = (uvalue / (unsigned)base);
+ } while (uvalue && (place < (int)(sizeof(convert) - 1)));
+ convert[place] = 0;
+ padlen = len - place;
+ if (padlen < 0)
+ padlen = 0;
+ if(ljust)
+ padlen = -padlen;
+ if (zpad && padlen > 0) {
+ if (signvalue) {
+ (void) dopr_outch(bp, ep, signvalue);
+ --padlen;
+ signvalue = 0;
+ }
+ while (padlen > 0) {
+ (void) dopr_outch(bp, ep, zpad);
+ --padlen;
+ }
+ }
+ while (padlen > 0) {
+ (void) dopr_outch(bp, ep, ' ');
+ --padlen;
+ }
+ if (signvalue)
+ (void) dopr_outch(bp, ep, signvalue);
+ while (place > 0)
+ (void) dopr_outch(bp, ep, convert[--place]);
+ while (padlen < 0) {
+ (void) dopr_outch(bp, ep, ' ');
+ ++padlen;
+ }
+}
+# endif /* !defined(HAS_NO_LONG_LONG) */
+
+
+/*
+ * fmtnum() -- format number for output
+ */
+
+static void
+fmtnum(bp, ep, value, base, dosign, ljust, len, zpad)
+ char **bp; /* current buffer pointer */
+ char *ep; /* end of buffer (-1) */
+ long value; /* number to format */
+ int base; /* number base */
+ int dosign; /* sign request */
+ int ljust; /* left justfication request */
+ int len; /* length request */
+ int zpad; /* zero padding request */
+{
+ int signvalue = 0;
+ unsigned long uvalue;
+ char convert[20];
+ int place = 0;
+ int padlen = 0; /* amount to pad */
+ int caps = 0;
+
+ uvalue = value;
+ if (dosign) {
+ if (value < 0) {
+ signvalue = '-';
+ uvalue = -value;
+ }
+ }
+ if (base < 0) {
+ caps = 1;
+ base = -base;
+ }
+ do {
+ convert[place++] =
+ (caps ? "0123456789ABCDEF" : "0123456789abcdef")
+ [uvalue % (unsigned)base];
+ uvalue = (uvalue / (unsigned)base);
+ } while (uvalue && (place < (int)(sizeof(convert) - 1)));
+ convert[place] = 0;
+ padlen = len - place;
+ if (padlen < 0)
+ padlen = 0;
+ if(ljust)
+ padlen = -padlen;
+ if (zpad && padlen > 0) {
+ if (signvalue) {
+ (void) dopr_outch(bp, ep, signvalue);
+ --padlen;
+ signvalue = 0;
+ }
+ while (padlen > 0) {
+ (void) dopr_outch(bp, ep, zpad);
+ --padlen;
+ }
+ }
+ while (padlen > 0) {
+ (void) dopr_outch(bp, ep, ' ');
+ --padlen;
+ }
+ if (signvalue)
+ (void) dopr_outch(bp, ep, signvalue);
+ while (place > 0)
+ (void) dopr_outch(bp, ep, convert[--place]);
+ while (padlen < 0) {
+ (void) dopr_outch(bp, ep, ' ');
+ ++padlen;
+ }
+}
+
+
+/*
+ * fmtstr() -- format string for output
+ */
+
+static void
+fmtstr(bp, ep, value, ljust, len, zpad, maxwidth)
+ char **bp; /* current buffer pointer */
+ char *ep; /* end of buffer (-1) */
+ char *value; /* string to format */
+ int ljust; /* left justification request */
+ int len; /* length request */
+ int zpad; /* zero padding request */
+ int maxwidth; /* maximum width request */
+{
+ int padlen, strlen; /* amount to pad */
+
+ if (value == 0)
+ value = "<NULL>";
+ for (strlen = 0; value[strlen]; ++ strlen) /* strlen() */
+ ;
+ if ((strlen > maxwidth) && maxwidth)
+ strlen = maxwidth;
+ padlen = len - strlen;
+ if (padlen < 0)
+ padlen = 0;
+ if (ljust)
+ padlen = -padlen;
+ while (padlen > 0) {
+ (void) dopr_outch(bp, ep, ' ');
+ --padlen;
+ }
+ (void) dostr(bp, ep, value, maxwidth);
+ while (padlen < 0) {
+ (void) dopr_outch(bp, ep, ' ');
+ ++padlen;
+ }
+}
+
+
+/*
+ * dostr() -- do string output
+ */
+
+static void
+dostr(bp, ep, str, cut)
+ char **bp; /* current buffer pointer */
+ char *ep; /* end of buffer (-1) */
+ char *str; /* string to output */
+ int cut; /* limit on amount of string to output:
+ * 0 == no limit */
+{
+ int f;
+
+ f = cut ? 1 : 0;
+ while (*str) {
+ if (f) {
+ if (cut-- > 0)
+ (void) dopr_outch(bp, ep, *str);
+ } else
+ (void) dopr_outch(bp, ep, *str);
+ str++;
+ }
+}
+
+
+/*
+ * dopr_outch() -- output a character (or two)
+ */
+
+static void
+dopr_outch(bp, ep, c)
+ char **bp; /* current buffer pointer */
+ char *ep; /* end of buffer (-1) */
+ int c; /* character to output */
+{
+ register char *cp = *bp;
+
+ if (iscntrl(c) && c != '\n' && c != '\t') {
+ c = '@' + (c & 0x1F);
+ if (cp < ep)
+ *cp++ = '^';
+ Length++;
+ }
+ if (cp < ep)
+ *cp++ = c;
+ *bp = cp;
+ Length++;
+}
+
+#else /* !defined(USE_LIB_SNPF) */
+char snpf_d1[] = "d"; char *snpf_d2 = snpf_d1;
+#endif /* defined(USE_LIB_SNPF) */
--- /dev/null
+.ds VN 4.82
+.TH LSOF 8 Revision-\*(VN
+.if !\n()P .nr )P 1v
+.SH NAME
+lsof \- list open files
+.SH SYNOPSIS
+.B lsof
+[
+.B \-?abChlnNOPRtUvVX
+] [
+.BI -A " A"
+] [
+.BI \-c " c"
+] [
+.BI +c " c"
+] [
+.BI +|\-d " d"
+] [
+.BI +|\-D " D"
+] [
+.B +|\-f [cfgGn]
+] [
+.BI \-F " [f]"
+] [
+.BI \-g " [s]"
+] [
+.BI \-i " [i]"
+] [
+.BI \-k " k"
+] [
+.BI +|\-L " [l]"
+] [
+.BI +|\-m " m"
+] [
+.B +|\-M
+] [
+.BI \-o " [o]"
+] [
+.BI \-p " s"
+] [
+.BI +|\-r " [t[m<fmt>]]"
+] [
+.BI \-s " [p:s]"
+] [
+.BI \-S " [t]"
+] [
+.BI \-T " [t]"
+] [
+.BI \-u " s"
+] [
+.B +|\-w
+] [
+.BI \-x " [fl]"
+] [
+.BI \-z " [z]"
+] [
+.BI \-Z " [Z]"
+] [
+.B --
+] [\fInames\fP]
+.SH DESCRIPTION
+.I Lsof
+revision \*(VN lists on its standard output file information about files
+opened by processes for the following UNIX dialects:
+.PP
+.nf
+ AIX 5.3
+ Apple Darwin 9 (Mac OS X 10.5)
+ FreeBSD 4.9 for x86-based systems
+ FreeBSD 7.[012] and 8.0 for AMD64-based systems
+ Linux 2.1.72 and above for x86-based systems
+ Solaris 9 and 10
+.fi
+.PP
+(See the
+.B DISTRIBUTION
+section of this manual page for information on how to obtain the
+latest
+.I lsof
+revision.)
+.PP
+An open file may be a regular file, a directory, a block special file,
+a character special file, an executing text reference, a library,
+a stream or a network file (Internet socket, NFS file or UNIX domain socket.)
+A specific file or all the files in a file system may be selected by path.
+.PP
+Instead of a formatted display,
+.I lsof
+will produce output that can be parsed by other programs.
+See the
+.BR \-F ,
+option description, and the
+.B "OUTPUT FOR OTHER PROGRAMS"
+section for more information.
+.PP
+In addition to producing a single output list,
+.I lsof
+will run in repeat mode.
+In repeat mode it will produce output, delay, then repeat the output
+operation until stopped with an interrupt or quit signal.
+See the
+.BI +|\-r " [t[m<fmt>]]"
+option description for more information.
+.SH OPTIONS
+In the absence of any options,
+.I lsof
+lists all open files belonging to all active processes.
+.PP
+If any list request option is specified, other list requests must be
+specifically requested \- e.g., if
+.B \-U
+is specified for the listing of UNIX socket files, NFS files won't be
+listed unless
+.B \-N
+is also specified;
+or if a user list is specified with the
+.B \-u
+option, UNIX domain socket files, belonging to users not in the list,
+won't be listed unless the
+.B \-U
+option is also specified.
+.PP
+Normally list options that are specifically stated are ORed \- i.e.,
+specifying the
+.B \-i
+option without an address and the \fB\-u\fPfoo option produces a
+listing of all network files OR files belonging to processes owned
+by user ``foo''.
+The exceptions are:
+.TP \w'1)\ 'u
+1)
+the `^' (negated) login name or user ID (UID), specified with the
+.B \-u
+option;
+.TP \w'1)\ 'u
+2)
+the `^' (negated) process ID (PID), specified with the
+.B \-p
+option;
+.TP \w'1)\ 'u
+3)
+the `^' (negated) process group ID (PGID), specified with the
+.B \-g
+option;
+.TP \w'1)\ 'u
+4)
+the `^' (negated) command, specified with the
+.B \-c
+option;
+.TP \w'1)\ 'u
+5)
+the (`^') negated TCP or UDP protocol state names, specified with the
+.BI \-s " [p:s]"
+option.
+.PP
+Since they represent exclusions, they are applied without ORing or ANDing
+and take effect before any other selection criteria are applied.
+.PP
+The
+.B \-a
+option may be used to AND the selections.
+For example, specifying
+.BR \-a ,
+.BR \-U ,
+and \fB\-u\fPfoo produces a listing of only UNIX socket files that
+belong to processes owned by user ``foo''.
+.PP
+Caution: the
+.B \-a
+option causes all list selection options to be ANDed; it can't
+be used to cause ANDing of selected pairs of selection options
+by placing it between them, even though its placement there is
+acceptable.
+Wherever
+.B \-a
+is placed, it causes the ANDing of all selection options.
+.PP
+Items of the same selection set \- command names, file descriptors,
+network addresses, process identifiers, user identifiers, zone names,
+security contexts \- are joined in a single ORed set and applied
+before the result participates in ANDing.
+Thus, for example, specifying \fB\-i\fP@aaa.bbb, \fB\-i\fP@ccc.ddd,
+.BR \-a ,
+and \fB\-u\fPfff,ggg will select the listing of files that belong to
+either login ``fff'' OR ``ggg'' AND have network connections to either
+host aaa.bbb OR ccc.ddd.
+.PP
+Options may be grouped together following a single prefix -- e.g.,
+the option set ``\fB\-a \-b \-C\fP'' may be stated as
+.BR \-abC .
+However, since values are optional following
+.BR +|\-f ,
+.BR \-F ,
+.BR \-g ,
+.BR \-i ,
+.BR +|\-L ,
+.BR \-o ,
+.BR +|\-r ,
+.BR \-s ,
+.BR \-S ,
+.BR \-T ,
+.B \-x
+and
+.BR \-z .
+when you have no values for them be careful that the
+following character isn't ambiguous.
+For example,
+.B \-Fn
+might represent the
+.B \-F
+and
+.B \-n
+options, or it might represent the
+.B n
+field identifier character following the
+.B \-F
+option.
+When ambiguity is possible, start a new option with a `-'
+character \- e.g., ``\fB\-F \-n\fP''.
+If the next option is a file name, follow the possibly ambiguous
+option with ``--'' \- e.g., ``\fB\-F -- \fIname\fR''.
+.PP
+Either the `+' or the `\-' prefix may be applied to a group of options.
+Options that don't take on separate meanings for each
+prefix \- e.g., \fB\-i\fP \- may be grouped under either prefix.
+Thus, for example, ``+M -i'' may be stated as ``+Mi'' and the group
+means the same as the separate options.
+Be careful of prefix grouping when one or more options in the group
+does take on separate meanings under different prefixes \-
+e.g., \fB+|\-M\fP; ``-iM'' is not the same request as ``\-i +M''.
+When in doubt, use separate options with appropriate prefixes.
+.TP \w'names'u+4
+.B \-? \-h
+These two equivalent options select a usage (help) output list.
+.I Lsof
+displays a shortened form of this output when it detects an error
+in the options supplied to it, after it has displayed messages
+explaining each error.
+(Escape the `?' character as your shell requires.)
+.TP \w'names'u+4
+.B \-a
+This option causes list selection options to be ANDed, as described
+above.
+.TP \w'names'u+4
+.BI \-A " A"
+This option is available on systems configured for AFS whose AFS
+kernel code is implemented via dynamic modules.
+It allows the
+.I lsof
+user to specify
+.I A
+as an alternate name list file where the kernel addresses of the dynamic
+modules might be found.
+See the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information about dynamic modules, their
+symbols, and how they affect
+.IR lsof .
+.TP \w'names'u+4
+.B \-b
+This option causes
+.I lsof
+to avoid kernel functions that might block \-
+.IR lstat (2),
+.IR readlink (2),
+and
+.IR stat (2).
+.IP
+See the
+.B "BLOCKS AND TIMEOUTS"
+and
+.B "AVOIDING KERNEL BLOCKS"
+sections for information on using this option.
+.TP \w'names'u+4
+.BI \-c " c"
+This option selects the listing of files for processes executing the
+command that begins with the characters of
+.IR c .
+Multiple commands may be specified, using multiple
+.B \-c
+options.
+They are joined in a single ORed set before participating in
+AND option selection.
+.IP
+If
+.I c
+begins with a `^', then the following characters specify a command
+name whose processes are to be ignored (excluded.)
+.IP
+If
+.I c
+begins and ends with a slash ('/'), the characters between the slashes
+are interpreted as a regular expression.
+Shell meta\-characters in the regular expression must be quoted to prevent
+their interpretation by the shell.
+The closing slash may be followed by these modifiers:
+.IP
+.nf
+ b the regular expression is a basic one.
+.br
+ i ignore the case of letters.
+.br
+ x the regular expression is an extended one
+.br
+ (default).
+.fi
+.IP
+See the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information on basic and extended regular
+expressions.
+.IP
+The simple command specification is tested first.
+If that test fails, the command regular expression is applied.
+If the simple command test succeeds, the command regular expression
+test isn't made.
+This may result in ``no command found for regex:'' messages
+when lsof's
+.B \-V
+option is specified.
+.TP \w'names'u+4
+.BI +c " w"
+This option defines the maximum number of initial characters of the name,
+supplied by the UNIX dialect, of the UNIX command associated with a process
+to be printed in the COMMAND column.
+(The
+.I lsof
+default is nine.)
+.IP
+Note that many UNIX dialects do not supply all command name characters
+to
+.I lsof
+in the files and structures from which
+.I lsof
+obtains command name.
+Often dialects limit the number of characters supplied in those sources.
+For example, Linux 2.4.27 and Solaris 9 both limit command name length to
+16 characters.
+.IP
+If
+.I w
+is zero ('0'), all command characters supplied to
+.I lsof
+by the UNIX dialect will be printed.
+.IP
+If
+.I w
+is less than the length of the column title, ``COMMAND'', it will
+be raised to that length.
+.TP \w'names'u+4
+.B \-C
+This option disables the reporting of any path name
+components from the kernel's name cache.
+See the
+.B "KERNEL NAME CACHE"
+section for more information.
+.TP \w'names'u+4
+.BI +d " s"
+This option causes
+.I lsof
+to search for all open instances of directory
+.I s
+and the files and directories it contains at its top level.
+This option does NOT descend the directory tree, rooted at
+.IR s .
+The
+.BI +D " D"
+option may be used to request a full\-descent directory tree search,
+rooted at directory
+.IR D .
+.IP
+Processing of the
+.B +d
+option does not follow symbolic links within
+.I s
+unless the
+.B \-x
+or
+.B \-x " l"
+option is also specified.
+Nor does it
+search for open files on file system mount points on subdirectories of
+.I s
+unless the
+.B \-x
+or
+.B \-x " f"
+option is also specified.
+.IP
+Note: the authority of the user of this option limits it to searching for
+files that the user has permission to examine with the system
+.IR stat (2)
+function.
+.TP \w'names'u+4
+.BI \-d " s"
+This option specifies a list of file descriptors (FDs) to exclude from
+or include in the output listing.
+The file descriptors are specified in the comma\-separated set
+.I s
+\&\- e.g., ``cwd,1,3'', ``^6,^2''.
+(There should be no spaces in the set.)
+.IP
+The list is an exclusion list if all entries of the set begin with `^'.
+It is an inclusion list if no entry begins with `^'.
+Mixed lists are not permitted.
+.IP
+A file descriptor number range may be in the set as long as
+neither member is empty, both members are numbers, and the ending
+member is larger than the starting one \- e.g., ``0-7'' or ``3-10''.
+Ranges may be specified for exclusion if they have the `^' prefix \-
+e.g., ``^0-7'' excludes all file descriptors 0 through 7.
+.IP
+Multiple file descriptor numbers are joined in a single ORed set before
+participating in AND option selection.
+.IP
+When there are exclusion and inclusion members in the set,
+.I lsof
+reports them as errors and exits with a non\-zero return code.
+.IP
+See the description of File Descriptor (FD) output values in the
+.B OUTPUT
+section for more information on file descriptor names.
+.TP \w'names'u+4
+.BI +D " D"
+This option causes
+.I lsof
+to search for all open instances of directory
+.I D
+and all the files and directories it contains to its complete depth.
+.IP
+Processing of the
+.B +D
+option does not follow symbolic links within
+.I D
+unless the
+.B \-x
+or
+.B \-x " l"
+option is also specified.
+Nor does it
+search for open files on file system mount points on subdirectories of
+.I D
+unless the
+.B \-x
+or
+.B \-x " f"
+option is also specified.
+.IP
+Note: the authority of the user of this option limits it to searching for
+files that the user has permission to examine with the system
+.IR stat (2)
+function.
+.IP
+Further note:
+.I lsof
+may process this option slowly and require a large amount of dynamic memory
+to do it.
+This is because it must descend the entire directory tree, rooted at
+.IR D ,
+calling
+.IR stat (2)
+for each file and directory, building a list of all the files it finds, and
+searching that list for a match with every open file.
+When directory
+.I D
+is large, these steps can take a long time, so use this option prudently.
+.TP \w'names'u+4
+.BI \-D " D"
+This option directs
+.I lsof's
+use of the device cache file.
+The use of this option is sometimes restricted.
+See the
+.B "DEVICE CACHE FILE"
+section and the sections that follow it for more information on this
+option.
+.IP
+.B -D
+must be followed by a function letter; the function letter may optionally
+be followed by a path name.
+.I Lsof
+recognizes these function letters:
+.IP
+.nf
+ \fB?\fP \- report device cache file paths
+ \fBb\fP \- build the device cache file
+ \fBi\fP \- ignore the device cache file
+ \fBr\fP \- read the device cache file
+ \fBu\fP \- read and update the device cache file
+.fi
+.IP
+The
+.BR b ,
+.BR r ,
+and
+.B u
+functions, accompanied by a path name, are sometimes restricted.
+When these functions are restricted, they will not appear in
+the description of the
+.B \-D
+option that accompanies
+.B \-h
+or
+.B \-?
+option output.
+See the
+.B "DEVICE CACHE FILE"
+section and the sections that follow it for more information on these
+functions and when they're restricted.
+.IP
+The
+.B ?
+function reports the read\-only and write paths that lsof can
+use for the device cache file,
+the names of any environment variables whose values
+.I lsof
+will examine when forming the device cache file path,
+and the format for the personal device cache file path.
+(Escape the `?' character as your shell requires.)
+.IP
+When available, the
+.BR b ,
+.BR r ,
+and
+.B u
+functions may be followed by the device cache file's path.
+The standard default is
+.I .lsof_hostname
+in the home directory of the real user ID that executes
+.IR lsof ,
+but this could have been changed when
+.I lsof
+was configured and compiled.
+(The output of the
+.B \-h
+and
+.B \-?
+options show the current default prefix \- e.g., ``.lsof''.)
+The suffix,
+.IR hostname ,
+is the first component of the host's name returned by
+.IR gethostname (2) .
+.IP
+When available, the
+.B b
+function directs
+.I lsof
+to build a new device cache file at the default or specified path.
+.IP
+The
+.B i
+function directs
+.I lsof
+to ignore the default device cache file and obtain its information
+about devices via direct calls to the kernel.
+.IP
+The
+.B r
+function directs
+.I lsof
+to read the device cache at the default or specified path, but
+prevents it from creating a new device cache file when none
+exists or the existing one is improperly structured.
+The
+.B r
+function, when specified without a path name, prevents
+.I lsof
+from updating an incorrect or outdated device cache file,
+or creating a new one in its place.
+The
+.B r
+function is always available when it is specified without a
+path name argument; it may be restricted by the permissions of the
+.I lsof
+process.
+.IP
+When available, the
+.B u
+function directs
+.I lsof
+to read the device cache file at the default or specified path,
+if possible, and to rebuild it, if necessary.
+This is the default device cache file function when no
+.B \-D
+option has been specified.
+.TP \w'names'u+4
+.B +|\-f [cfgGn]
+.B f
+by itself clarifies how path name arguments are to be interpreted.
+When followed by
+.BR c ,
+.BR f ,
+.BR g ,
+.BR G ,
+or
+.B n
+in any combination it specifies
+that the listing of kernel file structure information is to be enabled
+(`+') or inhibited (`\-').
+.IP
+Normally a path name argument is taken to be a file system name if
+it matches a mounted\-on directory name reported by
+.IR mount (8),
+or if it represents a block device, named in the
+.I mount
+output and associated with a mounted directory name.
+When
+.B +f
+is specified, all path name arguments will be taken to be file
+system names, and
+.I lsof
+will complain if any are not.
+This can be useful, for example, when the file system name
+(mounted\-on device) isn't a block device.
+This happens for some CD-ROM file systems.
+.IP
+When
+.B \-f
+is specified by itself, all path name arguments will be taken to be
+simple files.
+Thus, for example, the ``\fB\-f\fP\ -- /'' arguments direct lsof to search
+for open files with a `/' path name, not all open files in the `/'
+(root) file system.
+.IP
+Be careful to make sure
+.B +f
+and
+.B \-f
+are properly terminated and aren't followed by a character (e.g., of
+the file or file system name) that might be taken as a parameter.
+For example, use ``--'' after
+.B +f
+and
+.B \-f
+as in these examples.
+.IP
+.nf
+ $ lsof +f -- /file/system/name
+ $ lsof -f -- /file/name
+.fi
+.IP
+The listing of information from kernel file structures, requested with the
+.B +f [cfgGn]
+option form, is normally
+inhibited, and is not available in whole or part for some dialects \- e.g.,
+/proc\-based Linux kernels below 2.6.22.
+When the prefix to
+.B f
+is a plus sign (`+'), these characters request file structure information:
+.IP
+.nf
+ \fBc\fR file structure use count (not Linux)
+ \fBf\fR file structure address (not Linux)
+ \fBg\fR file flag abbreviations (Linux 2.6.22 and up)
+ \fBG\fR file flags in hexadecimal (Linux 2.6.22 and up)
+ \fBn\fR file structure node address (not Linux)
+.fi
+.IP
+When the prefix is minus (`\-') the same characters disable the
+listing of the indicated values.
+.IP
+File structure addresses, use counts, flags, and node addresses may be
+used to detect more readily identical files inherited by child
+processes and identical files in use by different processes.
+.I Lsof
+column output can be sorted by output columns holding the values
+and listed to identify identical file use, or
+.I lsof
+field output can be parsed by an AWK or Perl post\-filter script,
+or by a C program.
+.TP \w'names'u+4
+.BI \-F " f"
+This option specifies a character list,
+.IR f ,
+that selects the fields to be output for processing by another program,
+and the character that terminates each output field.
+Each field to be output is specified with a single character in
+.IR f .
+The field terminator defaults to NL, but may be changed to NUL (000).
+See the
+.B "OUTPUT FOR OTHER PROGRAMS"
+section for a description of the field identification characters and
+the field output process.
+.IP
+When the field selection character list is empty, all standard fields are
+selected (except the raw device field, security context and zone field for
+compatibility reasons)
+and the NL field terminator is used.
+.IP
+When the field selection character list contains only a zero (`0'),
+all fields are selected (except the raw device field for compatibility
+reasons) and the NUL terminator character is used.
+.IP
+Other combinations of fields and their associated field terminator
+character must be set with explicit entries in
+.IR f ,
+as described in the
+.B "OUTPUT FOR OTHER PROGRAMS"
+section.
+.IP
+When a field selection character identifies an item
+.I lsof
+does not normally list \- e.g., PPID, selected with
+.BR \-R " \-"
+specification of the field character \- e.g., ``\fB\-FR\fP'' \-
+also selects the listing of the item.
+.IP
+When the field selection character list contains the single
+character `?',
+.I lsof
+will display a help list of the field identification characters.
+(Escape the `?' character as your shell requires.)
+.TP \w'names'u+4
+.BI \-g " [s]"
+This option excludes or selects the listing of files for the processes
+whose optional process group IDentification (PGID) numbers are in the
+comma\-separated set
+.I s
+\&\- e.g., ``123'' or ``123,^456''.
+(There should be no spaces in the set.)
+.IP
+PGID numbers that begin with `^' (negation) represent exclusions.
+.IP
+Multiple PGID numbers are joined in a single ORed set before participating
+in AND option selection.
+However, PGID exclusions are applied without ORing or ANDing
+and take effect before other selection criteria are applied.
+.IP
+The
+.B \-g
+option also enables the output display of PGID numbers.
+When specified without a PGID set that's all it does.
+.TP \w'names'u+4
+.BI \-i " [i]"
+This option selects the listing of files any of whose Internet address
+matches the address specified in \fIi\fP.
+If no address is specified, this option selects the listing of all
+Internet and x.25 (HP\-UX) network files.
+.IP
+If
+.BI \-i 4
+or
+.BI \-i 6
+is specified with no following address, only files of the indicated
+IP version, IPv4 or IPv6, are displayed.
+(An IPv6 specification may be used only if the dialects supports IPv6,
+as indicated by ``[46]'' and ``IPv[46]'' in
+.I lsof's
+.B \-h
+or
+.B \-?
+output.)
+Sequentially specifying
+.BR \-i 4,
+followed by
+.BR \-i 6
+is the same as specifying
+.BR \-i ,
+and vice-versa.
+Specifying
+.BR \-i 4,
+or
+.BR \-i 6
+after
+.B \-i
+is the same as specifying
+.BR \-i 4
+or
+.BR \-i 6
+by itself.
+.IP
+Multiple addresses (up to a limit of 100) may be specified with multiple
+.B \-i
+options.
+(A port number or service name range is counted as one address.)
+They are joined in a single ORed set before participating in
+AND option selection.
+.IP
+An Internet address is specified in the form (Items in square
+brackets are optional.):
+.IP
+[\fI46\fP][\fIprotocol\fP][@\fIhostname\fP\||\|\fIhostaddr\fP][:\fIservice\fP\||\|\fIport\fP]
+.IP
+where:
+.nf
+.br
+ \fI46\fP specifies the IP version, IPv4 or IPv6
+.br
+ that applies to the following address.
+.br
+ '6' may be be specified only if the UNIX
+.br
+ dialect supports IPv6. If neither '4' nor
+.br
+ '6' is specified, the following address
+.br
+ applies to all IP versions.
+.br
+ \fIprotocol\fP is a protocol name \- \fBTCP\fP, \fBUDP\fP
+.br or \fBUDPLITE\fP.
+.br
+ \fIhostname\fP is an Internet host name. Unless a
+.br
+ specific IP version is specified, open
+.br
+ network files associated with host names
+.br
+ of all versions will be selected.
+.br
+ \fIhostaddr\fP is a numeric Internet IPv4 address in
+.br
+ dot form; or an IPv6 numeric address in
+.br
+ colon form, enclosed in brackets, if the
+.br
+ UNIX dialect supports IPv6. When an IP
+.br
+ version is selected, only its numeric
+.br
+ addresses may be specified.
+.br
+ \fIservice\fP is an \fI/etc/services\fP name \- e.g., \fBsmtp\fP \-
+ or a list of them.
+.br
+ \fIport\fP is a port number, or a list of them.
+.fi
+.IP
+IPv6 options may be used only if the UNIX dialect supports IPv6.
+To see if the dialect supports IPv6, run
+.I lsof
+and specify the
+.B \-h
+or
+.B \-?
+(help) option.
+If the displayed description of the
+.B \-i
+option contains ``[46]'' and ``IPv[46]'', IPv6 is supported.
+.IP
+IPv4 host names and addresses may not be specified if network file selection
+is limited to IPv6 with
+.BR \-i " 6."
+IPv6 host names and addresses may not be specified if network file selection
+is limited to IPv4 with
+.BR \-i " 4."
+When an open IPv4 network file's address is mapped in an IPv6 address,
+the open file's type will be IPv6, not IPv4, and its display will be
+selected by '6', not '4'.
+.IP
+At least one address component \-
+.BR 4,
+.BR 6,
+.IR protocol ,
+,IR hostname ,
+.IR hostaddr ,
+or
+.I service
+\&\- must be supplied.
+The `@' character, leading the host specification, is always required;
+as is the `:', leading the port specification.
+Specify either
+.I hostname
+or
+.IR hostaddr .
+Specify either
+.I service
+name list or
+.I port
+number list.
+If a
+.I service
+name list is specified, the
+.I protocol
+may also need to be specified if the TCP, UDP and UDPLITE port numbers for
+the service name are different.
+Use any case \- lower or upper \- for
+.IR protocol .
+.IP
+.I Service
+names and
+.I port
+numbers may be combined in a list whose entries are separated by commas
+and whose numeric range entries are separated by minus signs.
+There may be no embedded spaces, and all service names must belong to
+the specified
+.IR protocol .
+Since service names may contain embedded minus signs, the starting entry
+of a range can't be a service name; it can be a port number, however.
+.IP
+Here are some sample addresses:
+.nf
+
+.br
+ -i6 \- IPv6 only
+.br
+ TCP:25 \- TCP and port 25
+.br
+ @1.2.3.4 \- Internet IPv4 host address 1.2.3.4
+.br
+ @[3ffe:1ebc::1]:1234 \- Internet IPv6 host address
+ 3ffe:1ebc::1, port 1234
+.br
+ UDP:who \- UDP who service port
+.br
+ TCP@lsof.itap:513 \- TCP, port 513 and host name lsof.itap
+.br
+ tcp@foo:1-10,smtp,99 \- TCP, ports 1 through 10,
+ service name \fIsmtp\fP, port 99, host name foo
+.br
+ tcp@bar:1-smtp \- TCP, ports 1 through \fIsmtp\fP, host bar
+.br
+ :time \- either TCP, UDP or UDPLITE time service port
+.fi
+.TP \w'names'u+4
+.BI \-k " k"
+This option specifies a kernel name list file,
+.IR k ,
+in place of /vmunix, /mach, etc.
+This option is not available under AIX on the IBM RISC/System 6000.
+.TP \w'names'u+4
+.B \-l
+This option inhibits the conversion of user ID numbers to login names.
+It is also useful when login name lookup is working improperly or slowly.
+.TP \w'names'u+4
+.BI +|\-L " [l]"
+This option enables (`+') or disables (`-') the listing of file link
+counts, where they are available \- e.g., they aren't available
+for sockets, or most FIFOs and pipes.
+.IP
+When
+.B +L
+is specified without a following number, all link counts will be listed.
+When
+.B \-L
+is specified (the default), no link counts will be listed.
+.IP
+When
+.B +L
+is followed by a number, only files having a link count less than
+that number will be listed.
+(No number may follow
+.BR \-L .)
+A specification of the form ``\fB+L1\fP'' will select open files that
+have been unlinked.
+A specification of the form ``\fB+aL1\ \fI<file_system>\fR'' will select
+unlinked open files on the specified file system.
+.IP
+For other link count comparisons, use field output (\fB\-F\fP)
+and a post\-processing script or program.
+.TP \w'names'u+4
+.BI +|\-m " m"
+This option specifies an alternate kernel memory file or activates
+mount table supplement processing.
+.IP
+The option form
+.BI \-m " m"
+specifies a kernel memory file,
+.IR m ,
+in place of
+.I /dev/kmem
+or
+.I /dev/mem
+\&\- e.g., a crash dump file.
+.IP
+The option form
+.B +m
+requests that a mount supplement file be written to the standard output
+file.
+All other options are silently ignored.
+.IP
+There will be a line in the mount supplement file for each mounted file
+system, containing the mounted file system directory, followed by a single
+space, followed by the device number in hexadecimal "0x" format \- e.g.,
+.IP
+.nf
+ / 0x801
+.fi
+.IP
+.I Lsof
+can use the mount supplement file to get device numbers for file systems
+when it can't get them via
+.IR stat (2)
+or
+.IR lstat (2).
+.IP
+The option form
+.BI +m " m"
+identifies
+.I m
+as a mount supplement file.
+.IP
+Note: the
+.B +m
+and
+.BI +m " m"
+options are not available for all supported dialects.
+Check the output of
+.I lsof's
+.B \-h
+or
+.B \-?
+options to see if the
+.B +m
+and
+.BI +m " m"
+options are available.
+.TP \w'names'u+4
+.B +|\-M
+Enables (\fB+\fP) or disables (\fB-\fP) the
+reporting of portmapper registrations for local TCP, UDP and UDPLITE ports.
+The default reporting mode is set by the
+.I lsof
+builder with the HASPMAPENABLED #define in the dialect's machine.h
+header file;
+.I lsof
+is distributed with the HASPMAPENABLED #define deactivated, so
+portmapper reporting is disabled by default and must be requested
+with
+.BR \+M .
+Specifying
+.I lsof's
+.B \-h
+or
+.B \-?
+option will report the default mode.
+Disabling portmapper registration when it is already disabled or
+enabling it when already enabled is acceptable.
+.IP
+When portmapper registration reporting is enabled,
+.I lsof
+displays the portmapper registration (if any) for local TCP, UDP or
+UDPLITE ports
+in square brackets immediately following the port numbers or service
+names \- e.g., ``:1234[name]'' or ``:name[100083]''.
+The registration information may be a name or number, depending
+on what the registering program supplied to the portmapper when
+it registered the port.
+.IP
+When portmapper registration reporting is enabled,
+.I lsof
+may run a little more slowly or even become blocked when access to the
+portmapper becomes congested or stopped.
+Reverse the reporting mode to determine if portmapper registration
+reporting is slowing or blocking
+.IR lsof .
+.IP
+For purposes of portmapper registration reporting
+.I lsof
+considers a TCP, UDP or UDPLITE port local if: it is found in the local part
+of its containing kernel structure;
+or if it is located in the foreign part of its containing kernel
+structure and the local and foreign Internet addresses are the same;
+or if it is located in the foreign part of its containing kernel
+structure and the foreign Internet address is INADDR_LOOPBACK (127.0.0.1).
+This rule may make
+.I lsof
+ignore some foreign ports on machines with multiple interfaces
+when the foreign Internet address is on a different interface
+from the local one.
+.IP
+See the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for further discussion of portmapper registration
+reporting issues.
+.TP \w'names'u+4
+.B \-n
+This option inhibits the conversion of network numbers to
+host names for network files.
+Inhibiting conversion may make
+.I lsof
+run faster.
+It is also useful when host name lookup is not working properly.
+.TP \w'names'u+4
+.B \-N
+This option selects the listing of NFS files.
+.TP \w'names'u+4
+.BI \-o
+This option directs
+.I lsof
+to display file offset at all times.
+It causes the SIZE/OFF output column title to be changed to OFFSET.
+Note: on some UNIX dialects
+.I lsof
+can't obtain accurate or consistent file offset information from its
+kernel data sources, sometimes just for particular kinds of files
+(e.g., socket files.)
+Consult the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information.
+.IP
+The
+.B \-o
+and
+.B \-s
+options are mutually exclusive; they can't both be specified.
+When neither is specified,
+.I lsof
+displays whatever value \- size or offset \- is appropriate and
+available for the type of the file.
+.TP \w'names'u+4
+.BI \-o " o"
+This option defines the number of decimal digits (\fIo\fP) to be
+printed after the ``0t'' for a file offset before the form is switched
+to ``0x...''.
+An
+.I o
+value of zero (unlimited) directs
+.I lsof
+to use the ``0t'' form for all offset output.
+.IP
+This option does NOT direct
+.I lsof
+to display offset at all times; specify
+.B \-o
+(without a trailing number) to do that.
+This option only specifies the number of digits after ``0t'' in
+either mixed size and offset or offset\-only output.
+Thus, for example, to direct
+.I lsof
+to display offset at all times with a decimal digit count of 10, use:
+.IP
+.nf
+ -o -o 10
+or
+ -oo10
+.fi
+.IP
+The default number of digits allowed after ``0t'' is normally 8,
+but may have been changed by the lsof builder.
+Consult the description of the
+.BI \-o " o"
+option in the output of the
+.B \-h
+or
+.B \-?
+option to determine the default that is in effect.
+.TP \w'names'u+4
+.B \-O
+This option directs
+.I lsof
+to bypass the strategy it uses to avoid being blocked by some
+kernel operations \- i.e., doing them in forked child processes.
+See the
+.B "BLOCKS AND TIMEOUTS"
+and
+.B "AVOIDING KERNEL BLOCKS"
+sections for more information on kernel operations that may block
+.IR lsof .
+.IP
+While use of this option will reduce
+.I lsof
+startup overhead, it may also cause
+.I lsof
+to hang when the kernel doesn't respond to a function.
+Use this option cautiously.
+.TP \w'names'u+4
+.BI \-p " s"
+This option excludes or selects the listing of files for the processes
+whose optional process IDentification (PID) numbers are in the
+comma\-separated set
+.I s
+\&\- e.g., ``123'' or ``123,^456''.
+(There should be no spaces in the set.)
+.IP
+PID numbers that begin with `^' (negation) represent exclusions.
+.IP
+Multiple process ID numbers are joined in a single ORed set before
+participating in AND option selection.
+However, PID exclusions are applied without ORing or ANDing
+and take effect before other selection criteria are applied.
+.TP \w'names'u+4
+.B \-P
+This option inhibits the conversion of port numbers to port
+names for network files.
+Inhibiting the conversion may make
+.I lsof
+run a little faster.
+It is also useful when port name lookup is not working properly.
+.TP \w'names'u+4
+.BI +|\-r " [t[m<fmt>]]"
+This option puts
+.I lsof
+in repeat mode.
+There
+.I lsof
+lists open files as selected by other options, delays
+.I t
+seconds (default fifteen), then repeats the listing, delaying
+and listing repetitively until stopped by a condition defined by
+the prefix to the option.
+.IP
+If the prefix is a `\-', repeat mode is endless.
+.I Lsof
+must be terminated with an interrupt or quit signal.
+.IP
+If the prefix is `+', repeat mode will end the first cycle no open files
+are listed \- and of course when
+.I lsof
+is stopped with an interrupt or quit signal.
+When repeat mode ends because no files are listed, the process exit code
+will be zero if any open files were ever listed; one, if none were ever
+listed.
+.IP
+.I Lsof
+marks the end of each listing:
+if field output is in progress (the
+.BR \-F ,
+option has been specified), the default marker is `m'; otherwise the
+default marker is ``========''.
+The marker is followed by a NL character.
+.IP
+The optional "m<fmt>" argument specifies a format for the marker line.
+The <fmt> characters following `m' are interpreted as a format
+specification to the
+.IR strftime (3)
+function, when both it and the
+.IR localtime (3)
+function are available in the dialect's C library.
+Consult the
+.IR strftime (3)
+documentation for what may appear in its format specification.
+Note that when field output is requested with the
+.B \-F
+option, <fmt> cannot contain the NL format, ``%n''.
+Note also that when <fmt> contains spaces or other characters that
+affect the shell's interpretation of arguments, <fmt> must be
+quoted appropriately.
+.IP
+Repeat mode reduces
+.I lsof
+startup overhead, so it is more efficient to use this mode
+than to call
+.I lsof
+repetitively from a shell script, for example.
+.IP
+To use repeat mode most efficiently, accompany
+.B +|\-r
+with specification of other
+.I lsof
+selection options, so the amount of kernel memory access
+.I lsof
+does will be kept to a minimum.
+Options that filter at the process level \- e.g.,
+.BR \-c ,
+.BR \-g ,
+.BR \-p ,
+.B \-u
+\&\- are the most efficient selectors.
+.IP
+Repeat mode is useful when coupled with field output (see the
+.BR \-F ,
+option description) and a supervising
+.I awk
+or
+.I Perl
+script, or a C program.
+.TP \w'names'u+4
+.B \-R
+This option directs lsof to list the Parent Process IDentification
+number in the PPID column.
+.TP \w'names'u+4
+.BI \-s " [p:s]"
+.B s
+alone directs
+.I lsof
+to display file size at all times.
+It causes the SIZE/OFF output column title to be changed to SIZE.
+If the file does not have a size, nothing is displayed.
+.IP
+When followed by a protocol name (\fIp\fR), either TCP or UDP,
+a colon (`:') and a comma\-separated protocol state name list,
+the option causes open TCP and UDP files to be excluded if their
+state name(s) are in the list (\fIs\fP) preceded by a `^'; or
+included if their name(s) are not preceded by a `^'.
+.IP
+When an inclusion list is defined, only network files with state
+names in the list will be present in the
+.I lsof
+output.
+Thus, specifying one state name means that only network files
+with that lone state name will be listed.
+.IP
+Case is unimportant in the protocol or state names, but there may
+be no spaces and the colon (`:') separating the protocol
+name (\fIp\fP) and the state name list (\fIs\fP) is required.
+.IP
+If only TCP and UDP files are to be listed, as controlled by
+the specified exclusions and inclusions, the
+.B \-i
+option must be specified, too.
+If only a single protocol's files are to be listed, add its name
+as an argument to the
+.B \-i
+option.
+.IP
+For example, to list only network files with TCP state LISTEN, use:
+.IP
+.nf
+ \-iTCP \-sTCP:LISTEN
+.fi
+.IP
+Or, for example, to list network files with all UDP states except
+Idle, use:
+.IP
+.nf
+ \-iUDP -sUDP:Idle
+.fi
+.IP
+State names vary with UNIX dialects, so it's not possible to
+provide a complete list. Some common TCP state names are:
+CLOSED, IDLE, BOUND, LISTEN, ESTABLISHED, SYN_SENT, SYN_RCDV,
+ESTABLISHED, CLOSE_WAIT, FIN_WAIT1, CLOSING, LAST_ACK, FIN_WAIT_2,
+and TIME_WAIT.
+Two common UDP state names are Unbound and Idle.
+.IP
+See the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information on how to use protocol state exclusion and
+inclusion, including examples.
+.IP
+The
+.B \-o
+(without a following decimal digit count) and
+.B \-s
+option (without a following protocol and state name list)
+are mutually exclusive; they can't both be specified.
+When neither is specified,
+.I lsof
+displays whatever value \- size or offset \- is appropriate and
+available for the type of file.
+.IP
+Since some types of files don't have true sizes \- sockets, FIFOs,
+pipes, etc. \- lsof displays for their sizes the content amounts in
+their associated kernel buffers, if possible.
+.TP \w'names'u+4
+.BI \-S " [t]"
+This option specifies an optional time-out seconds value for kernel functions \-
+.IR lstat (2),
+.IR readlink (2),
+and
+.IR stat (2)
+\- that might otherwise deadlock.
+The minimum for
+.I t
+is two;
+the default, fifteen; when no value is specified, the default is used.
+.IP
+See the
+.B "BLOCKS AND TIMEOUTS"
+section for more information.
+.TP \w'names'u+4
+.BI \-T " [t]"
+This option controls the reporting of some TCP/TPI information, also
+reported by
+.IR netstat (1),
+following the network addresses.
+In normal output the information appears in parentheses, each item
+except TCP or TPI state name identified by a keyword, followed by `=',
+separated from others by a single space:
+.IP
+.nf
+ <TCP or TPI state name>
+ QR=<read queue length>
+ QS=<send queue length>
+ SO=<socket options and values>
+ SS=<socket states>
+ TF=<TCP flags and values>
+ WR=<window read length>
+ WW=<window write length>
+.fi
+.IP
+Not all values are reported for all UNIX dialects.
+Items values (when available) are reported after the item name and '='.
+.IP
+When the field output mode is in effect (See
+.BR "OUTPUT FOR OTHER PROGRAMS" .)
+each item appears as a field with a `T' leading character.
+.IP
+.B \-T
+with no following key characters disables TCP/TPI information reporting.
+.IP
+.B \-T
+with following characters selects the reporting of specific TCP/TPI
+information:
+.IP
+.nf
+ \fBf\fP selects reporting of socket options,
+ states and values, and TCP flags and
+ values.
+ \fBq\fP selects queue length reporting.
+ \fBs\fP selects connection state reporting.
+ \fBw\fP selects window size reporting.
+.fi
+.IP
+Not all selections are enabled for some UNIX dialects.
+State may be selected for all dialects and is reported by default.
+The
+.B \-h
+or
+.B \-?
+help output for the
+.B \-T
+option will show what selections may be used with the UNIX dialect.
+.IP
+When
+.B \-T
+is used to select information \- i.e., it is followed by one or more
+selection characters \- the displaying of state is disabled by default,
+and it must be explicitly selected again in the characters following
+.BR \-T .
+(In effect, then, the default is equivalent to
+.BR -Ts .)
+For example, if queue lengths and state are desired, use
+.BR \-Tqs .
+.IP
+Socket options, socket states, some socket values, TCP flags and
+one TCP value may be reported (when available in the UNIX dialect)
+in the form of the names that commonly appear after SO_, so_, SS_,
+TCP_ and TF_ in the dialect's header files \-
+most often <sys/socket.h>, <sys/socketvar.h> and <netinet/tcp_var.h>.
+Consult those header files for the meaning of the flags, options,
+states and values.
+.IP
+``SO='' precedes socket options and values; ``SS='', socket states;
+and ``TF='', TCP flags and values.
+.IP
+If a flag or option has a value, the value will follow an '=' and
+the name -- e.g., ``SO=LINGER=5'', ``SO=QLIM=5'', ``TF=MSS=512''.
+The following seven values may be reported:
+.IP
+.nf
+ Name
+ Reported Description (Common Symbol)
+
+ KEEPALIVE keep alive time (SO_KEEPALIVE)
+ LINGER linger time (SO_LINGER)
+ MSS maximum segment size (TCP_MAXSEG)
+ PQLEN partial listen queue connections
+ QLEN established listen queue connections
+ QLIM established listen queue limit
+ RCVBUF receive buffer length (SO_RCVBUF)
+ SNDBUF send buffer length (SO_SNDBUF)
+.fi
+.IP
+Details on what socket options and values, socket states, and TCP flags
+and values may be displayed for particular UNIX dialects may be found in
+the answer to the ``Why doesn't lsof report socket options, socket states,
+and TCP flags and values for my dialect?'' and ``Why doesn't lsof report
+the partial listen queue connection count for my dialect?''
+questions in the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+.TP \w'names'u+4
+.B \-t
+This option specifies that
+.I lsof
+should produce terse output with process identifiers only and no header \-
+e.g., so that the output may be piped to
+.IR kill (1).
+This option selects the
+.B \-w
+option.
+.TP \w'names'u+4
+.BI \-u " s"
+This option selects the listing of files for the user whose login names
+or user ID numbers are in the comma\-separated set
+.I s
+\&\- e.g., ``abe'',
+or ``548,root''.
+(There should be no spaces in the set.)
+.IP
+Multiple login names or user ID numbers are joined in a single ORed set
+before participating in AND option selection.
+.IP
+If a login name or user ID is preceded by a `^', it becomes a negation \-
+i.e., files of processes owned by the login name or user ID will never
+be listed.
+A negated login name or user ID selection is neither ANDed nor ORed
+with other selections; it is applied before all other selections and
+absolutely excludes the listing of the files of the process.
+For example, to direct
+.I lsof
+to exclude the listing of files belonging to root processes,
+specify ``\-u^root'' or ``\-u^0''.
+.TP \w'names'u+4
+.B \-U
+This option selects the listing of UNIX domain socket files.
+.TP \w'names'u+4
+.B \-v
+This option selects the listing of
+.I lsof
+version information, including: revision number;
+when the
+.I lsof
+binary was constructed;
+who constructed the binary and where;
+the name of the compiler used to construct the
+.I lsof binary;
+the version number of the compiler when readily available;
+the compiler and loader flags used to construct the
+.I lsof
+binary;
+and system information, typically the output of
+.IR uname 's
+.B \-a
+option.
+.TP \w'names'u+4
+.B \-V
+This option directs
+.I lsof
+to indicate the items it was asked to list and failed to find \- command
+names, file names, Internet addresses or files, login names, NFS files,
+PIDs, PGIDs, and UIDs.
+.IP
+When other options are ANDed to search options, or compile\-time
+options restrict the listing of some files,
+.I lsof
+may not report that it failed to find a search item when an ANDed
+option or compile\-time option prevents the listing of the open file
+containing the located search item.
+.IP
+For example, ``lsof -V -iTCP@foobar -a -d 999'' may not report a
+failure to locate open files at ``TCP@foobar'' and may not list
+any, if none have a file descriptor number of 999.
+A similar situation arises when HASSECURITY and HASNOSOCKSECURITY are
+defined at compile time and they prevent the listing of open files.
+.TP \w'names'u+4
+.B +|\-w
+Enables (\fB+\fP) or disables (\fB-\fP) the suppression of warning messages.
+.IP
+The
+.I lsof
+builder may choose to have warning messages disabled or enabled by
+default.
+The default warning message state is indicated in the output of the
+.B \-h
+or
+.B \-?
+option.
+Disabling warning messages when they are already disabled or enabling
+them when already enabled is acceptable.
+.IP
+The
+.B \-t
+option selects the
+.B \-w
+option.
+.TP \w'names'u+4
+.B \-x " [fl]"
+This option may accompany the
+.B +d
+and
+.B +D
+options to direct their processing to cross over symbolic links
+and|or file system mount points encountered when scanning the
+directory (\fB+d\fP) or directory tree (\fB+D\fP).
+.IP
+If
+.B -x
+is specified by itself without a following parameter, cross\-over
+processing of both symbolic links and file system mount points is
+enabled.
+Note that when
+.B \-x
+is specified without a parameter, the next argument must begin with '-'
+or '+'.
+.IP
+The optional 'f' parameter enables file system mount point cross\-over
+processing; 'l', symbolic link cross\-over processing.
+.IP
+The
+.B \-x
+option may not be supplied without also supplying a
+.B +d
+or
+.B +D
+option.
+.TP \w'names'u+4
+.B \-X
+This is a dialect\-specific option.
+.HP \w'names'u+4
+\ \ \ \ AIX:
+.br
+This IBM AIX RISC/System 6000 option requests the reporting
+of executed text file and shared library references.
+.IP
+.B WARNING:
+because this option uses the kernel readx() function, its use on
+a busy AIX system might cause an application process to hang so
+completely that it can neither be killed nor stopped.
+I have never seen this happen or had a report of its happening,
+but I think there is a remote possibility it could happen.
+.IP
+By default use of readx() is disabled.
+On AIX 5L and above
+.I lsof
+may need setuid\-root permission to perform the actions this
+option requests.
+.IP
+The
+.I lsof
+builder may specify that the
+.B \-X
+option be restricted to processes whose real UID is root.
+If that has been done, the
+.B \-X
+option will not appear in the
+.B \-h
+or
+.B \-?
+help output unless the real UID of the
+.I lsof
+process is root.
+The default
+.I lsof
+distribution allows any UID to specify
+.BR \-X,
+so by default it will appear in the help output.
+.IP
+When AIX readx() use
+is disabled,
+.I lsof
+may not be able to report information for all text and loader file
+references, but it may also avoid exacerbating an AIX
+kernel directory search kernel error, known as the Stale Segment
+ID bug.
+.IP
+The readx() function, used by
+.I lsof
+or any other program to access some sections of kernel virtual
+memory, can trigger the Stale Segment ID bug.
+It can cause the kernel's dir_search() function to believe erroneously
+that part of an in\-memory copy of a file system directory has been
+zeroed.
+Another application process, distinct from
+.IR lsof ,
+asking the kernel to search the directory \- e.g., by using
+.IR open "(2) \-"
+can cause dir_search() to loop forever, thus hanging the application process.
+.IP
+Consult the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+and the
+.I 00README
+file of the
+.I lsof
+distribution for a more complete description of the Stale Segment ID bug,
+its APAR, and methods for defining readx() use when compiling
+.IR lsof .
+.HP \w'names'u+4
+\ \ \ \ Linux:
+.br
+This Linux option requests that
+.I lsof
+skip the reporting of information on all open TCP, UDP and UDPLITE IPv4
+and IPv6 files.
+.IP
+This Linux option is most useful when the system has an extremely
+large number of open TCP, UDP and UDPLITE files, the processing of whose
+information in the
+.I /proc/net/tcp*
+and
+.I /proc/net/udp*
+files would take
+.I lsof
+a long time, and whose reporting is not of interest.
+.IP
+Use this option with care and only when you are sure that the
+information you want
+.I lsof
+to display isn't associated with open TCP, UDP or UDPLITE socket files.
+.HP \w'names'u+4
+\ \ \ \ Solaris 10 and above:
+.br
+This Solaris 10 and above option requests the reporting of cached
+paths for files that have been deleted \- i.e., removed with
+.IR rm (1)
+or
+.IR unlink (2).
+.IP
+The cached path is followed by the string ``\ (deleted)'' to indicate
+that the path by which the file was opened has been deleted.
+.IP
+Because intervening changes made to the path \- i.e., renames with
+.IR mv (1)
+or
+.IR rename (2)
+\- are not recorded in the cached path, what
+.I lsof
+reports is only the path by which the file was opened, not its
+possibly different final path.
+.TP \w'names'u+4
+.BI \-z " [z]"
+specifies how Solaris 10 and higher zone information is to be handled.
+.IP
+Without a following argument \- e.g., NO
+.IR z " \-"
+the option specifies that zone names are to be listed in the ZONE
+output column.
+.IP
+The
+.B \-z
+option may be followed by a zone name,
+.BI z .
+That causes lsof to list only open files for processes in that zone.
+Multiple
+.BI \-z " z"
+option and argument pairs may be specified to form a list of named zones.
+Any open file of any process in any of the zones will be listed, subject
+to other conditions specified by other options and arguments.
+.TP \w'names'u+4
+.BI \-Z " [Z]"
+specifies how SELinux security contexts are to be handled.
+This option and 'Z' field output character support are inhibited
+when SELinux is disabled in the running Linux kernel.
+See
+.B "OUTPUT FOR OTHER PROGRAMS"
+for more information on the 'Z' field output character.
+.IP
+Without a following argument \- e.g., NO
+.IR Z " \-"
+the option specifies that security contexts are to be listed in the
+SECURITY\-CONTEXT output column.
+.IP
+The
+.B \-Z
+option may be followed by a wildcard security context name,
+.BI Z .
+That causes lsof to list only open files for processes in that security
+context.
+Multiple
+.BI \-Z " Z"
+option and argument pairs may be specified to form a list of security
+contexts.
+Any open file of any process in any of the security contexts will be listed,
+subject to other conditions specified by other options and arguments.
+Note that
+.I Z
+can be A:B:C or *:B:C or A:B:* or *:*:C to match against the A:B:C context.
+.TP \w'names'u+4
+.B --
+The double minus sign option is a marker that signals the end of
+the keyed options.
+It may be used, for example, when the first file name begins with
+a minus sign.
+It may also be used when the absence of a value for the last keyed
+option must be signified by the presence of a minus sign in the following
+option and before the start of the file names.
+.TP \w'names'u+4
+.I names
+These are path names of specific files to list.
+Symbolic links are resolved before use.
+The first name may be separated from the preceding options with
+the ``--'' option.
+.IP
+If a
+.I name
+is the mounted\-on directory of a file system or the device of the
+file system,
+.I lsof
+will list all the files open on the file system.
+To be considered a file system, the
+.I name
+must match a mounted\-on directory name in
+.IR mount (8)
+output, or match the name of a block device associated with a mounted\-on
+directory name.
+The
+.B +|\-f
+option may be used to force
+.I lsof
+to consider a
+.I name
+a file system identifier (\fB+f\fP) or a simple file (\fB\-f\fP).
+.IP
+If
+.I name
+is a path to a directory that is not the mounted\-on directory name of
+a file system, it is treated just as a regular file is treated \- i.e.,
+its listing is restricted to processes that have it open as a file or
+as a process\-specific directory, such as the root or current working
+directory.
+To request that
+.I lsof
+look for open files inside a directory name, use the
+.BI +d " s"
+and
+.BI +D " D"
+options.
+.IP
+If a
+.I name
+is the base name of a family of multiplexed files \- e. g, AIX's
+.IR /dev/pt[cs] " \-"
+.I lsof
+will list all the associated multiplexed files on the device that
+are open \- e.g.,
+.IR /dev/pt[cs]/1 ,
+.IR /dev/pt[cs]/2 ,
+etc.
+.IP
+If a
+.I name
+is a UNIX domain socket name,
+.I lsof
+will usually search for it by the characters of the name alone \- exactly as
+it is specified and is recorded in the kernel socket structure.
+(See the next paragraph for an exception to that rule for Linux.)
+Specifying a relative path \- e.g.,
+.I ./file
+\&\- in place of the
+file's absolute path \- e.g.,
+.I /tmp/file
+\&\- won't work because
+.I lsof
+must match the characters you specify with what it finds in the
+kernel UNIX domain socket structures.
+.IP
+If a
+.I name
+is a Linux UNIX domain socket name, in one case
+.I lsof
+is able to search for it by its device and inode number, allowing
+.I name
+to be a relative path.
+The case requires that the absolute path -- i.e., one beginning with a
+slash ('/') be used by the process that created the socket, and hence be
+stored in the
+.I /proc/net/unix
+file; and it requires that
+.I lsof
+be able to obtain the device and node numbers of both the absolute path in
+.I /proc/net/unix
+and
+.I name
+via successful
+.IR stat (2)
+system calls.
+When those conditions are met,
+.I lsof
+will be able to search for the UNIX domain socket when some path to it is
+is specified in
+.IR name .
+Thus, for example, if the path is
+.IR /dev/log ,
+and an
+.I lsof
+search is initiated when the working directory is
+.IR /dev ,
+then
+.I name
+could be
+.IR ./log .
+.IP
+If a
+.I name
+is none of the above,
+.I lsof
+will list any open files whose device and inode match that of the
+specified path
+.IR name .
+.IP
+If you have also specified the
+.B \-b
+option,
+the only
+.I names
+you may safely specify are file systems for which your mount table
+supplies alternate device numbers.
+See the
+.B "AVOIDING KERNEL BLOCKS"
+and
+.B "ALTERNATE DEVICE NUMBERS"
+sections for more information.
+.IP
+Multiple file names are joined in a single ORed set before
+participating in AND option selection.
+.SH AFS
+.I Lsof
+supports the recognition of AFS files for these dialects (and AFS
+versions):
+.PP
+.nf
+ AIX 4.1.4 (AFS 3.4a)
+ HP\-UX 9.0.5 (AFS 3.4a)
+ Linux 1.2.13 (AFS 3.3)
+ Solaris 2.[56] (AFS 3.4a)
+.fi
+.PP
+It may recognize AFS files on other versions of these dialects,
+but has not been tested there.
+Depending on how AFS is implemented,
+.I lsof
+may recognize AFS files in other dialects, or may have difficulties
+recognizing AFS files in the supported dialects.
+.PP
+.I Lsof
+may have trouble identifying all aspects of AFS files in
+supported dialects when AFS kernel support is implemented via
+dynamic modules whose addresses do not appear in the kernel's
+variable name list.
+In that case,
+.I lsof
+may have to guess at the identity of AFS files, and might not be able to
+obtain volume information from the kernel that is needed for calculating
+AFS volume node numbers.
+When
+.I lsof
+can't compute volume node numbers, it reports blank in the NODE column.
+.PP
+The
+.BI \-A " A"
+option is available in some dialect implementations of
+.I lsof
+for specifying the name list file where dynamic module kernel
+addresses may be found.
+When this option is available, it will be listed in the
+.I lsof
+help output, presented in response to the
+.B \-h
+or
+.B \-?
+.PP
+See the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information about dynamic modules, their
+symbols, and how they affect
+.I lsof
+options.
+.PP
+Because AFS path lookups don't seem to participate in the
+kernel's name cache operations,
+.I lsof
+can't identify path name components for AFS files.
+.SH SECURITY
+.I Lsof
+has three features that may cause security concerns.
+First, its default compilation mode allows anyone to list all
+open files with it.
+Second, by default it creates a user\-readable and user\-writable device
+cache file in the home directory of the real user ID that executes
+.IR lsof .
+(The list\-all\-open\-files and device cache features may be disabled when
+.I lsof
+is compiled.)
+Third, its
+.B \-k
+and
+.B \-m
+options name alternate kernel name list or memory files.
+.PP
+Restricting the listing of all open files is controlled by the
+compile\-time HASSECURITY and HASNOSOCKSECURITY options.
+When HASSECURITY is defined,
+.I lsof
+will allow only the root user to list all open files.
+The non\-root user may list only open files of processes with the same user
+IDentification number as the real user ID number of the
+.I lsof
+process (the one that its user logged on with).
+.PP
+However, if HASSECURITY and HASNOSOCKSECURITY are both defined,
+anyone may list open socket files, provided they are selected
+with the
+.B \-i
+option.
+.PP
+When HASSECURITY is not defined, anyone may list all open files.
+.PP
+Help output, presented in response to the
+.B \-h
+or
+.B \-?
+option, gives the status of the HASSECURITY and HASNOSOCKSECURITY definitions.
+.PP
+See the
+.B Security
+section of the
+.I 00README
+file of the
+.I lsof
+distribution for information on building
+.I lsof
+with the HASSECURITY and HASNOSOCKSECURITY options enabled.
+.PP
+Creation and use of a user\-readable and user\-writable device
+cache file is controlled by the compile\-time HASDCACHE option.
+See the
+.B "DEVICE CACHE FILE"
+section and the sections that follow it for details on how its path
+is formed.
+For security considerations it is important to note that in the default
+.I lsof
+distribution, if the real user ID under which
+.I lsof
+is executed is root, the device cache file will be written in root's
+home directory \- e.g.,
+.I /
+or
+.IR /root .
+When HASDCACHE is not defined,
+.I lsof
+does not write or attempt to read a device cache file.
+.PP
+When HASDCACHE is defined, the
+.I lsof
+help output, presented in response to the
+.BR \-h ,
+.BR \-D? ,
+or
+.B \-?
+options, will provide device cache file handling information.
+When HASDCACHE is not defined, the
+.B \-h
+or
+.B \-?
+output will have no
+.B \-D
+option description.
+.PP
+Before you decide to disable the device cache file feature \- enabling
+it improves the performance of
+.I lsof
+by reducing the startup overhead of examining all the nodes in
+.I /dev
+(or
+.IR /devices )
+\&\- read the discussion of it in the
+.I 00DCACHE
+file of the
+.I lsof
+distribution and the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+.PP
+WHEN IN DOUBT, YOU CAN TEMPORARILY DISABLE THE USE OF THE DEVICE CACHE FILE
+WITH THE
+.B \-Di
+OPTION.
+.PP
+When
+.I lsof
+user declares alternate kernel name list or memory files with the
+.B \-k
+and
+.B \-m
+options,
+.I lsof
+checks the user's authority to read them with
+.IR access (2).
+This is intended to prevent whatever special power
+.I lsof's
+modes might confer on it from letting it read files not normally
+accessible via the authority of the real user ID.
+.SH OUTPUT
+This section describes the information
+.I lsof
+lists for each open file.
+See the
+.B "OUTPUT FOR OTHER PROGRAMS"
+section for additional information on output that can be processed
+by another program.
+.PP
+.I Lsof
+only outputs printable (declared so by
+.IR isprint (3))
+8 bit characters.
+Non\-printable characters are printed in one of three forms:
+the C ``\\[bfrnt]'' form;
+the control character `^' form (e.g., ``^@'');
+or hexadecimal leading ``\\x'' form (e.g., ``\\xab'').
+Space is non\-printable in the COMMAND column (``\\x20'')
+and printable elsewhere.
+.PP
+For some dialects \- if HASSETLOCALE is defined in the dialect's
+machine.h header file \-
+.I lsof
+will print the extended 8 bit characters of a language locale.
+The
+.I lsof
+process must be supplied a language locale environment variable
+(e.g., LANG) whose value represents a known language locale
+in which the extended characters are considered printable by
+.IR isprint (3).
+Otherwise
+.I lsof
+considers the extended characters non\-printable and prints them according
+to its rules for non\-printable characters, stated above.
+Consult your dialect's
+.IR setlocale (3)
+man page for the names of other environment variables that may
+be used in place of LANG \- e.g., LC_ALL, LC_CTYPE, etc.
+.PP
+.I Lsof's
+language locale support for a dialect also covers wide characters \- e.g.,
+UTF-8 \- when HASSETLOCALE and HASWIDECHAR are defined in the dialect's
+machine.h header file, and when a suitable language locale has been defined
+in the appropriate environment variable for the
+.I lsof
+process.
+Wide characters are printable under those conditions if
+.IR iswprint (3)
+reports them to be.
+If HASSETLOCALE, HASWIDECHAR and a suitable language locale aren't defined,
+or if
+.IR iswprint (3)
+reports wide characters that aren't printable,
+.I lsof
+considers the wide characters non\-printable and prints each of their
+8 bits according to its rules for non\-printable characters, stated above.
+.PP
+Consult the answers to the "Language locale support" questions in the
+lsof FAQ (The \fBFAQ\fP section gives its location.) for more information.
+.PP
+.I Lsof
+dynamically sizes the output columns each time it runs, guaranteeing
+that each column is a minimum size.
+It also guarantees that each column is separated from its predecessor
+by at least one space.
+.TP \w'COMMAND'u+4
+COMMAND
+contains the first nine characters of the name of the UNIX command
+associated with the process.
+If a non\-zero
+.I w
+value is specified to the
+.BI +c " w"
+option, the column contains the first
+.I w
+characters of the name of the UNIX command associated with the process
+up to the limit of characters supplied to
+.I lsof
+by the UNIX dialect.
+(See the description of the
+.BI +c " w"
+command or the
+.I lsof
+FAQ for more information.
+The \fBFAQ\fP section gives its location.)
+.IP
+If
+.I w
+is less than the length of the column title, ``COMMAND'', it will
+be raised to that length.
+.IP
+If a zero
+.I w
+value is specified to the
+.BI +c " w"
+option, the column contains all the characters of the name of the UNIX command
+associated with the process.
+.IP
+All command name characters maintained by the kernel in its structures
+are displayed in field output when the command name descriptor (`c')
+is specified.
+See the
+.B "OUTPUT FOR OTHER COMMANDS"
+section for information on selecting field output and the associated
+command name descriptor.
+.TP
+PID
+is the Process IDentification number of the process.
+.TP
+ZONE
+is the Solaris 10 and higher zone name.
+This column must be selected with the
+.B \-z
+option.
+.TP
+SECURITY\-CONTEXT
+is the SELinux security context.
+This column must be selected with the
+.B -Z
+option.
+Note that the
+.B -Z
+option is inhibited when SELinux is disabled in the running Linux
+kernel.
+.TP
+PPID
+is the Parent Process IDentification number of the process.
+It is only displayed when the
+.B \-R
+option has been specified.
+.TP
+PGID
+is the process group IDentification number associated with
+the process.
+It is only displayed when the
+.B \-g
+option has been specified.
+.TP
+USER
+is the user ID number or login name of the user to whom
+the process belongs, usually the same as reported by
+.IR ps (1).
+However, on Linux USER is the user ID number or login that owns
+the directory in /proc where
+.I lsof
+finds information about the process.
+Usually that is the same value reported by
+.IR ps (1),
+but may differ when the process has changed its effective user ID.
+(See the
+.B \-l
+option description for information on when a user ID number or
+login name is displayed.)
+.TP
+FD
+is the File Descriptor number of the file or:
+.IP
+.nf
+ \fBcwd\fP current working directory;
+.br
+ \fBL\fInn\fR library references (AIX);
+.br
+ \fBerr\fR FD information error (see NAME column);
+.br
+ \fBjld\fR jail directory (FreeBSD);
+.br
+ \fBltx\fP shared library text (code and data);
+.br
+ \fBMxx\fP hex memory\-mapped type number xx.
+.br
+ \fBm86\fP DOS Merge mapped file;
+.br
+ \fBmem\fP memory\-mapped file;
+.br
+ \fBmmap\fP memory\-mapped device;
+.br
+ \fBpd\fP parent directory;
+.br
+ \fBrtd\fP root directory;
+.br
+ \fBtr\fR kernel trace file (OpenBSD);
+.br
+ \fBtxt\fP program text (code and data);
+.br
+ \fBv86\fP VP/ix mapped file;
+.fi
+.IP
+FD is followed by one of these characters, describing the mode under which
+the file is open:
+.IP
+ \fBr\fP for read access;
+.br
+ \fBw\fP for write access;
+.br
+ \fBu\fP for read and write access;
+.br
+ space if mode unknown and no lock
+.br
+ character follows;
+.br
+ `\-' if mode unknown and lock
+.br
+ character follows.
+.IP
+The mode character is followed by one of these lock characters, describing
+the type of lock applied to the file:
+.IP
+ \fBN\fP for a Solaris NFS lock of unknown type;
+.br
+ \fBr\fP for read lock on part of the file;
+.br
+ \fBR\fP for a read lock on the entire file;
+.br
+ \fBw\fP for a write lock on part of the file;
+.br
+ \fBW\fP for a write lock on the entire file;
+.br
+ \fBu\fP for a read and write lock of any length;
+.br
+ \fBU\fP for a lock of unknown type;
+.br
+ \fBx\fP for an SCO OpenServer Xenix lock on part
+ of the file;
+.br
+ \fBX\fP for an SCO OpenServer Xenix lock on the
+ entire file;
+.br
+ space if there is no lock.
+.IP
+See the
+.B LOCKS
+section for more information on the lock information character.
+.IP
+The FD column contents constitutes a single field for parsing in
+post\-processing scripts.
+.TP
+TYPE
+is the type of the node associated with the file \- e.g., GDIR, GREG,
+VDIR, VREG, etc.
+.IP
+or ``IPv4'' for an IPv4 socket;
+.IP
+or ``IPv6'' for an open IPv6 network file \- even if its address is
+IPv4, mapped in an IPv6 address;
+.IP
+or ``ax25'' for a Linux AX.25 socket;
+.IP
+or ``inet'' for an Internet domain socket;
+.IP
+or ``lla'' for a HP\-UX link level access file;
+.IP
+or ``rte'' for an AF_ROUTE socket;
+.IP
+or ``sock'' for a socket of unknown domain;
+.IP
+or ``unix'' for a UNIX domain socket;
+.IP
+or ``x.25'' for an HP\-UX x.25 socket;
+.IP
+or ``BLK'' for a block special file;
+.IP
+or ``CHR'' for a character special file;
+.IP
+or ``DEL'' for a Linux map file that has been deleted;
+.IP
+or ``DIR'' for a directory;
+.IP
+or ``DOOR'' for a VDOOR file;
+.IP
+or ``FIFO'' for a FIFO special file;
+.IP
+or ``KQUEUE'' for a BSD style kernel event queue file;
+.IP
+or ``LINK'' for a symbolic link file;
+.IP
+or ``MPB'' for a multiplexed block file;
+.IP
+or ``MPC'' for a multiplexed character file;
+.IP
+or ``NOFD'' for a Linux /proc/<PID>/fd directory that can't be opened \--
+the directory path appears in the NAME column, followed by an error
+message;
+.IP
+or ``PAS'' for a
+.I /proc/as
+file;
+.IP
+or ``PAXV'' for a
+.I /proc/auxv
+file;
+.IP
+or ``PCRE'' for a
+.I /proc/cred
+file;
+.IP
+or ``PCTL'' for a
+.I /proc
+control file;
+.IP
+or ``PCUR'' for the current
+.I /proc
+process;
+.IP
+or ``PCWD'' for a
+.I /proc
+current working directory;
+.IP
+or ``PDIR'' for a
+.I /proc
+directory;
+.IP
+or ``PETY'' for a
+.I /proc
+executable type (\fIetype\fP);
+.IP
+or ``PFD'' for a
+.I /proc
+file descriptor;
+.IP
+or ``PFDR'' for a
+.I /proc
+file descriptor directory;
+.IP
+or ``PFIL'' for an executable
+.I /proc
+file;
+.IP
+or ``PFPR'' for a
+.I /proc
+FP register set;
+.IP
+or ``PGD'' for a
+.I /proc/pagedata
+file;
+.IP
+or ``PGID'' for a
+.I /proc
+group notifier file;
+.IP
+or ``PIPE'' for pipes;
+.IP
+or ``PLC'' for a
+.I /proc/lwpctl
+file;
+.IP
+or ``PLDR'' for a
+.I /proc/lpw
+directory;
+.IP
+or ``PLDT'' for a
+.I /proc/ldt
+file;
+.IP
+or ``PLPI'' for a
+.I /proc/lpsinfo
+file;
+.IP
+or ``PLST'' for a
+.I /proc/lstatus
+file;
+.IP
+or ``PLU'' for a
+.I /proc/lusage
+file;
+.IP
+or ``PLWG'' for a
+.I /proc/gwindows
+file;
+.IP
+or ``PLWI'' for a
+.I /proc/lwpsinfo
+file;
+.IP
+or ``PLWS'' for a
+.I /proc/lwpstatus
+file;
+.IP
+or ``PLWU'' for a
+.I /proc/lwpusage
+file;
+.IP
+or ``PLWX'' for a
+.I /proc/xregs
+file'
+.IP
+or ``PMAP'' for a
+.I /proc
+map file (\fImap\fP);
+.IP
+or ``PMEM'' for a
+.I /proc
+memory image file;
+.IP
+or ``PNTF'' for a
+.I /proc
+process notifier file;
+.IP
+or ``POBJ'' for a
+.I /proc/object
+file;
+.IP
+or ``PODR'' for a
+.I /proc/object
+directory;
+.IP
+or ``POLP'' for an old format
+.I /proc
+light weight process file;
+.IP
+or ``POPF'' for an old format
+.I /proc
+PID file;
+.IP
+or ``POPG'' for an old format
+.I /proc
+page data file;
+.IP
+or ``PORT'' for a SYSV named pipe;
+.IP
+or ``PREG'' for a
+.I /proc
+register file;
+.IP
+or ``PRMP'' for a
+.I /proc/rmap
+file;
+.IP
+or ``PRTD'' for a
+.I /proc
+root directory;
+.IP
+or ``PSGA'' for a
+.I /proc/sigact
+file;
+.IP
+or ``PSIN'' for a
+.I /proc/psinfo
+file;
+.IP
+or ``PSTA'' for a
+.I /proc
+status file;
+.IP
+or ``PSXSEM'' for a POSIX semaphore file;
+.IP
+or ``PSXSHM'' for a POSIX shared memory file;
+.IP
+or ``PUSG'' for a
+.I /proc/usage
+file;
+.IP
+or ``PW'' for a
+.I /proc/watch
+file;
+.IP
+or ``PXMP'' for a
+.I /proc/xmap
+file;
+.IP
+or ``REG'' for a regular file;
+.IP
+or ``SMT'' for a shared memory transport file;
+.IP
+or ``STSO'' for a stream socket;
+.IP
+or ``UNNM'' for an unnamed type file;
+.IP
+or ``XNAM'' for an OpenServer Xenix special file of unknown type;
+.IP
+or ``XSEM'' for an OpenServer Xenix semaphore file;
+.IP
+or ``XSD'' for an OpenServer Xenix shared data file;
+.IP
+or the four type number octets if the corresponding name isn't known.
+.TP
+FILE\-ADDR
+contains the kernel file structure address when
+.B f
+has been specified to
+.BR +f ;
+.TP
+FCT
+contains the file reference count from the kernel file structure when
+.B c
+has been specified to
+.BR +f ;
+.TP
+FILE\-FLAG
+when
+.B g
+or
+.B G
+has been specified to
+.BR +f ,
+this field contains the contents of the f_flag[s] member of the kernel
+file structure and the kernel's per\-process open file flags (if available);
+\&`G' causes them to be displayed in hexadecimal;
+\&`g', as short\-hand names;
+two lists may be displayed with entries separated by commas, the
+lists separated by a semicolon (`;');
+the first list may contain short\-hand names for f_flag[s] values from
+the following table:
+.IP
+.nf
+ AIO asynchronous I/O (e.g., FAIO)
+ AP append
+ ASYN asynchronous I/O (e.g., FASYNC)
+ BAS block, test, and set in use
+ BKIU block if in use
+ BL use block offsets
+ BSK block seek
+ CA copy avoid
+ CIO concurrent I/O
+ CLON clone
+ CLRD CL read
+ CR create
+ DF defer
+ DFI defer IND
+ DFLU data flush
+ DIR direct
+ DLY delay
+ DOCL do clone
+ DSYN data\-only integrity
+ DTY must be a directory
+ EVO event only
+ EX open for exec
+ EXCL exclusive open
+ FSYN synchronous writes
+ GCDF defer during unp_gc() (AIX)
+ GCMK mark during unp_gc() (AIX)
+ GTTY accessed via /dev/tty
+ HUP HUP in progress
+ KERN kernel
+ KIOC kernel\-issued ioctl
+ LCK has lock
+ LG large file
+ MBLK stream message block
+ MK mark
+ MNT mount
+ MSYN multiplex synchronization
+ NATM don't update atime
+ NB non\-blocking I/O
+ NBDR no BDRM check
+ NBIO SYSV non\-blocking I/O
+ NBF n\-buffering in effect
+ NC no cache
+ ND no delay
+ NDSY no data synchronization
+ NET network
+ NFLK don't follow links
+ NMFS NM file system
+ NOTO disable background stop
+ NSH no share
+ NTTY no controlling TTY
+ OLRM OLR mirror
+ PAIO POSIX asynchronous I/O
+ PP POSIX pipe
+ R read
+ RC file and record locking cache
+ REV revoked
+ RSH shared read
+ RSYN read synchronization
+ RW read and write access
+ SL shared lock
+ SNAP cooked snapshot
+ SOCK socket
+ SQSH Sequent shared set on open
+ SQSV Sequent SVM set on open
+ SQR Sequent set repair on open
+ SQS1 Sequent full shared open
+ SQS2 Sequent partial shared open
+ STPI stop I/O
+ SWR synchronous read
+ SYN file integrity while writing
+ TCPM avoid TCP collision
+ TR truncate
+ W write
+ WKUP parallel I/O synchronization
+ WTG parallel I/O synchronization
+ VH vhangup pending
+ VTXT virtual text
+ XL exclusive lock
+.fi
+.IP
+this list of names was derived from F* #define's in dialect header files
+<fcntl.h>, <linux</fs.h>, <sys/fcntl.c>, <sys/fcntlcom.h>, and <sys/file.h>;
+see the lsof.h header file for a list showing the correspondence
+between the above short\-hand names and the header file definitions;
+.IP
+the second list (after the semicolon) may contain short\-hand names
+for kernel per\-process open file flags from this table:
+.IP
+.nf
+ ALLC allocated
+ BR the file has been read
+ BHUP activity stopped by SIGHUP
+ BW the file has been written
+ CLSG closing
+ CX close\-on-exec (see fcntl(F_SETFD))
+ LCK lock was applied
+ MP memory\-mapped
+ OPIP open pending \- in progress
+ RSVW reserved wait
+ SHMT UF_FSHMAT set (AIX)
+ USE in use (multi\-threaded)
+.fi
+.TP
+NODE\-ID
+(or INODE\-ADDR for some dialects)
+contains a unique identifier for the file node (usually the kernel
+vnode or inode address, but also occasionally a concatenation of
+device and node number) when
+.B n
+has been specified to
+.BR +f ;
+.TP
+DEVICE
+contains the device numbers, separated by commas, for a character special,
+block special, regular, directory or NFS file;
+.IP
+or ``memory'' for a memory file system node under Tru64 UNIX;
+.IP
+or the address of the private data area of a Solaris socket
+stream;
+.IP
+or a kernel reference address that identifies the file
+(The kernel reference address may be used for FIFO's, for example.);
+.IP
+or
+the base address or device name of a Linux AX.25 socket device.
+.IP
+Usually only the lower thirty two bits of Tru64 UNIX kernel addresses
+are displayed.
+.TP
+SIZE, SIZE/OFF, or OFFSET
+is the size of the file or the file offset in bytes.
+A value is displayed in this column only if it is available.
+.I Lsof
+displays whatever value \- size or offset \- is appropriate for the type
+of the file and the version of
+.IR lsof .
+.IP
+On some UNIX dialects
+.I lsof
+can't obtain accurate or consistent file offset information from its
+kernel data sources, sometimes just for particular kinds of files
+(e.g., socket files.)
+In other cases, files don't have true sizes \- e.g., sockets, FIFOs,
+pipes \- so
+.I lsof
+displays for their sizes the content amounts it finds in their kernel
+buffer descriptors (e.g., socket buffer size counts or TCP/IP window
+sizes.)
+Consult the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information.
+.IP
+The file size is displayed in decimal;
+the offset is normally displayed in decimal with a leading ``0t'' if
+it contains 8 digits or less; in hexadecimal with a leading ``0x'' if
+it is longer than 8 digits.
+(Consult the
+.BI \-o " o"
+option description for information on when 8 might default to
+some other value.)
+.IP
+Thus the leading ``0t'' and ``0x'' identify an offset when the column
+may contain both a size and an offset (i.e., its title is SIZE/OFF).
+.IP
+If the
+.B \-o
+option is specified,
+.I lsof
+always displays the file offset (or nothing if no offset is available)
+and labels the column OFFSET.
+The offset always begins with ``0t'' or ``0x'' as described above.
+.IP
+The
+.I lsof
+user can control the switch from ``0t'' to ``0x'' with the
+.BI \-o " o"
+option.
+Consult its description for more information.
+.IP
+If the
+.B \-s
+option is specified,
+.I lsof
+always displays the file size (or nothing if no size is available)
+and labels the column SIZE.
+The
+.B \-o
+and
+.B \-s
+options are mutually exclusive; they can't both be specified.
+.IP
+For files that don't have a fixed size \- e.g., don't reside
+on a disk device \-
+.I lsof
+will display appropriate information about the current size or
+position of the file if it is available in the kernel structures
+that define the file.
+.TP
+NLINK
+contains the file link count when
+.B +L
+has been specified;
+.TP
+NODE
+is the node number of a local file;
+.IP
+or the inode number of an NFS file in the server host;
+.IP
+or the Internet protocol type \- e. g, ``TCP'';
+.IP
+or ``STR'' for a stream;
+.IP
+or ``CCITT'' for an HP\-UX x.25 socket;
+.IP
+or the IRQ or inode number of a Linux AX.25 socket device.
+.TP
+NAME
+is the name of the mount point and file system on which the file resides;
+.IP
+or the name of a file specified in the
+.I names
+option (after any symbolic links have been resolved);
+.IP
+or the name of a character special or block special device;
+.IP
+or the local and remote Internet addresses of a network file;
+the local host name or IP number is followed by a colon (':'), the
+port, ``->'', and the two\-part remote address;
+IP addresses may be reported as numbers or names, depending on the
+.BR +|\-M ,
+.BR \-n ,
+and
+.B \-P
+options;
+colon\-separated IPv6 numbers are enclosed in square brackets;
+IPv4 INADDR_ANY and IPv6 IN6_IS_ADDR_UNSPECIFIED addresses, and
+zero port numbers are represented by an asterisk ('*');
+a UDP destination address may be followed by the amount of time
+elapsed since the last packet was sent to the destination;
+TCP, UDP and UDPLITE remote addresses may be followed by TCP/TPI
+information in parentheses \- state (e.g., ``(ESTABLISHED)'', ``(Unbound)''),
+queue sizes, and window sizes (not all dialects) \- in a fashion
+similar to what
+.IR netstat (1)
+reports;
+see the
+.B \-T
+option description or the description of the TCP/TPI field in
+.B "OUTPUT FOR OTHER PROGRAMS"
+for more information on state, queue size, and window size;
+.IP
+or the address or name of a UNIX domain socket, possibly including
+a stream clone device name, a file system object's path name, local
+and foreign kernel addresses, socket pair information, and a bound
+vnode address;
+.IP
+or the local and remote mount point names of an NFS file;
+.IP
+or ``STR'', followed by the stream name;
+.IP
+or a stream character device name, followed by ``->'' and the stream name
+or a list of stream module names, separated by ``->'';
+.IP
+or ``STR:'' followed by the SCO OpenServer stream device and module
+names, separated by ``->'';
+.IP
+or system directory name, `` -- '', and as many components of the path
+name as
+.I lsof
+can find in the kernel's name cache for selected dialects
+(See the
+.B "KERNEL NAME CACHE"
+section for more information.);
+.IP
+or ``PIPE->'', followed by a Solaris kernel pipe destination address;
+.IP
+or ``COMMON:'', followed by the vnode device information structure's
+device name, for a Solaris common vnode;
+.IP
+or the address family, followed by a slash (`/'), followed by fourteen
+comma\-separated bytes of a non\-Internet raw socket address;
+.IP
+or the HP\-UX x.25 local address, followed by the virtual connection
+number (if any), followed by the remote address (if any);
+.IP
+or ``(dead)'' for disassociated Tru64 UNIX files \- typically terminal files
+that have been flagged with the TIOCNOTTY ioctl and closed by daemons;
+.IP
+or ``rd=<offset>'' and ``wr=<offset>'' for the values of the
+read and write offsets of a FIFO;
+.IP
+or ``clone \fIn\fP:/dev/event'' for SCO OpenServer file clones of the
+.I /dev/event
+device, where
+.I n
+is the minor device number of the file;
+.IP
+or ``(socketpair: n)'' for a Solaris 2.6, 8, 9 or 10
+UNIX domain socket, created by the
+.IR socketpair (3N)
+network function;
+.IP
+or ``no PCB'' for socket files that do not have a protocol block
+associated with them, optionally followed by ``, CANTSENDMORE'' if
+sending on the socket has been disabled, or ``, CANTRCVMORE'' if
+receiving on the socket has been disabled (e.g., by the
+.IR shutdown (2)
+function);
+.IP
+or the local and remote addresses of a Linux IPX socket file
+in the form <net>:[<node>:]<port>, followed in parentheses
+by the transmit and receive queue sizes, and the connection state;
+.IP
+or ``dgram'' or ``stream'' for the type UnixWare 7.1.1 and above in\-kernel
+UNIX domain sockets, followed by a colon (':') and the local path name
+when available, followed by ``->'' and the remote path name or kernel
+socket address in hexadecimal when available.
+.PP
+For dialects that support a ``namefs'' file system, allowing one
+file to be attached to another with
+.IR fattach (3C),
+.I lsof
+will add ``(FA:<address1><direction><address2>)'' to the NAME column.
+<address1> and <address2> are hexadecimal vnode addresses.
+<direction> will be ``<-'' if <address2> has been fattach'ed to
+this vnode whose address is <address1>;
+and ``->'' if <address1>, the vnode address of this vnode, has been
+fattach'ed to <address2>.
+<address1> may be omitted if it already appears in the DEVICE column.
+.PP
+.I
+Lsof
+may add two parenthetical notes to the NAME column for open Solaris 10 files:
+\&``(?)'' if
+.I lsof
+considers the path name of questionable accuracy;
+and ``(deleted)'' if the
+.B \-X
+option has been specified and
+.I lsof
+detects the open file's path name has been deleted.
+Consult the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information on these NAME column additions.
+.SH LOCKS
+.I Lsof
+can't adequately report the wide variety of UNIX dialect file locks
+in a single character.
+What it reports in a single character is a compromise between the
+information it finds in the kernel and the limitations of the reporting
+format.
+.PP
+Moreover, when a process holds several byte level locks on a file,
+.I lsof
+only reports the status of the first lock it encounters.
+If it is a byte level lock, then the lock character will be reported
+in lower case \- i.e., `r', `w', or `x' \- rather than the upper case
+equivalent reported for a full file lock.
+.PP
+Generally
+.I lsof
+can only report on locks held by local processes on local files.
+When a local process sets a lock on a remotely mounted (e.g., NFS)
+file, the remote server host usually records the lock state.
+One exception is Solaris \- at some patch levels of 2.3, and in all
+versions above 2.4, the Solaris kernel records information on remote
+locks in local structures.
+.PP
+.I Lsof
+has trouble reporting locks for some UNIX dialects.
+Consult the
+.B BUGS
+section of this manual page or the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for more information.
+.SH "OUTPUT FOR OTHER PROGRAMS"
+When the
+.B \-F
+option is specified,
+.I lsof
+produces output that is suitable for processing by another program \- e.g, an
+.I awk
+or
+.I Perl
+script, or a C program.
+.PP
+Each unit of information is output in a field that is identified
+with a leading character and terminated by a NL (012) (or a NUL
+(000) if the 0 (zero) field identifier character is specified.)
+The data of the field follows immediately after the field identification
+character and extends to the field terminator.
+.PP
+It is possible to think of field output as process and file sets.
+A process set begins with a field whose identifier is `p' (for
+process IDentifier (PID)).
+It extends to the beginning of the next PID field or the beginning
+of the first file set of the process, whichever comes first.
+Included in the process set are fields that identify the command,
+the process group IDentification (PGID) number, and the user ID (UID)
+number or login name.
+.PP
+A file set begins with a field whose identifier is `f' (for
+file descriptor).
+It is followed by lines that describe the file's access mode,
+lock state, type, device, size, offset, inode, protocol, name
+and stream module names.
+It extends to the beginning of the next file or process set,
+whichever comes first.
+.PP
+When the NUL (000) field terminator has been selected with the
+0 (zero) field identifier character,
+.I lsof
+ends each process and file set with a NL (012) character.
+.PP
+.I Lsof
+always produces one field, the PID (`p') field.
+All other fields may be declared optionally in the field identifier
+character list that follows the
+.B \-F
+option.
+When a field selection character identifies an item
+.I lsof
+does not normally list \- e.g., PPID, selected with
+.BR \-R " \-"
+specification of the field character \- e.g., ``\fB\-FR\fP'' \-
+also selects the listing of the item.
+.PP
+It is entirely possible to select a set of fields that cannot
+easily be parsed \- e.g., if the field descriptor field is not
+selected, it may be difficult to identify file sets.
+To help you avoid this difficulty,
+.I lsof
+supports the
+.B \-F
+option; it selects the output of all fields with NL terminators
+(the
+.B \-F0
+option pair selects the output of all fields with NUL terminators).
+For compatibility reasons neither
+.B \-F
+nor
+.B \-F0
+select the raw device field.
+.PP
+These are the fields that
+.I lsof
+will produce.
+The single character listed first is the field identifier.
+.PP
+.nf
+ a file access mode
+ c process command name (all characters from proc or
+ user structure)
+ C file structure share count
+ d file's device character code
+ D file's major/minor device number (0x<hexadecimal>)
+ f file descriptor
+ F file structure address (0x<hexadecimal>)
+ G file flaGs (0x<hexadecimal>; names if \fB+fg\fP follows)
+ i file's inode number
+ k link count
+ l file's lock status
+ L process login name
+ m marker between repeated output
+ n file name, comment, Internet address
+ N node identifier (ox<hexadecimal>
+ o file's offset (decimal)
+ p process ID (always selected)
+ g process group ID
+ P protocol name
+ r raw device number (0x<hexadecimal>)
+ R parent process ID
+ s file's size (decimal)
+ S file's stream identification
+ t file's type
+ T TCP/TPI information, identified by prefixes (the
+ `=' is part of the prefix):
+ QR=<read queue size>
+ QS=<send queue size>
+ SO=<socket options and values> (not all dialects)
+ SS=<socket states> (not all dialects)
+ ST=<connection state>
+ TF=<TCP flags and values> (not all dialects)
+ WR=<window read size> (not all dialects)
+ WW=<window write size> (not all dialects)
+ (TCP/TPI information isn't reported for all supported
+ UNIX dialects. The \fB\-h\fP or \fB\-?\fP help output for the
+ \fB\-T\fP option will show what TCP/TPI reporting can be
+ requested.)
+ u process user ID
+ z Solaris 10 and higher zone name
+ Z SELinux security context (inhibited when SELinux is disabled)
+ 0 use NUL field terminator character in place of NL
+ 1\-9 dialect\-specific field identifiers (The output
+ of \fB\-F?\fP identifies the information to be found
+ in dialect\-specific fields.)
+.fi
+.PP
+You can get on\-line help information on these characters and their
+descriptions by specifying the
+.B \-F?
+option pair.
+(Escape the `?' character as your shell requires.)
+Additional information on field content can be found in the
+.B OUTPUT
+section.
+.PP
+As an example, ``\fB\-F pcfn\fP'' will select the process ID (`p'),
+command name (`c'), file descriptor (`f') and file name (`n')
+fields with an NL field terminator character; ``\fB\-F pcfn0\fP''
+selects the same output with a NUL (000) field terminator character.
+.PP
+.I Lsof
+doesn't produce all fields for every process or file set, only
+those that are available.
+Some fields are mutually exclusive: file device characters and
+file major/minor device numbers; file inode number and protocol
+name; file name and stream identification; file size and offset.
+One or the other member of these mutually exclusive sets will appear
+in field output, but not both.
+.PP
+Normally
+.I lsof
+ends each field with a NL (012) character.
+The
+0 (zero) field identifier character may be specified to change the
+field terminator character
+to a NUL (000).
+A NUL terminator may be easier to process with
+.I xargs (1),
+for example, or with programs whose quoting mechanisms may not
+easily cope with the range of characters in the field output.
+When the NUL field terminator is in use,
+.I lsof
+ends each process and file set with a NL (012).
+.PP
+Three aids to producing programs that can process
+.I lsof
+field output are included in the
+.I lsof
+distribution.
+The first is a C header file,
+.IR lsof_fields.h ,
+that contains symbols for the field identification characters, indexes for
+storing them in a table, and explanation strings that may be compiled into
+programs.
+.I Lsof
+uses this header file.
+.PP
+The second aid is a set of sample scripts that process field output,
+written in
+.IR awk ,
+.I Perl
+4, and
+.I Perl
+5.
+They're located in the
+.I scripts
+subdirectory of the
+.I lsof
+distribution.
+.PP
+The third aid is the C library used for the
+.I lsof
+test suite.
+The test suite is written in C and uses field output to validate
+the correct operation of
+.IR lsof .
+The library can be found in the
+.I tests/LTlib.c
+file of the
+.I lsof
+distribution.
+The library uses the first aid, the
+.I lsof_fields.h
+header file.
+.SH "BLOCKS AND TIMEOUTS"
+.I Lsof
+can be blocked by some kernel functions that it uses \-
+.IR lstat (2),
+.IR readlink (2),
+and
+.IR stat (2).
+These functions are stalled in the kernel, for example, when the
+hosts where mounted NFS file systems reside become inaccessible.
+.PP
+.I Lsof
+attempts to break these blocks with timers and child processes,
+but the techniques are not wholly reliable.
+When
+.I lsof
+does manage to break a block, it will report the break with an error
+message.
+The messages may be suppressed with the
+.B \-t
+and
+.B \-w
+options.
+.PP
+The default timeout value may be displayed with the
+.B \-h
+or
+.B \-?
+option, and it may be changed with the
+.BI \-S " [t]"
+option.
+The minimum for
+.I t
+is two seconds, but you should avoid small values, since slow system
+responsiveness can cause short timeouts to expire unexpectedly and
+perhaps stop
+.I lsof
+before it can produce any output.
+.PP
+When
+.I lsof
+has to break a block during its access of mounted file system
+information, it normally continues, although with less information
+available to display about open files.
+.PP
+.I Lsof
+can also be directed to avoid the protection of timers and child processes
+when using the kernel functions that might block by specifying the
+.B \-O
+option.
+While this will allow
+.I lsof
+to start up with less overhead, it exposes
+.I lsof
+completely to the kernel situations that might block it.
+Use this option cautiously.
+.SH "AVOIDING KERNEL BLOCKS"
+.PP
+You can use the
+.B \-b
+option to tell
+.I lsof
+to avoid using kernel functions that would block.
+Some cautions apply.
+.PP
+First, using this option usually requires that your system supply
+alternate device numbers in place of the device numbers that
+.I lsof
+would normally obtain with the
+.IR lstat (2)
+and
+.IR stat (2)
+kernel functions.
+See the
+.B "ALTERNATE DEVICE NUMBERS"
+section for more information on alternate device numbers.
+.PP
+Second, you can't specify
+.I names
+for
+.I lsof
+to locate unless they're file system names.
+This is because
+.I lsof
+needs to know the device and inode numbers of files listed with
+.I names
+in the
+.I lsof
+options, and the
+.B \-b
+option prevents
+.I lsof
+from obtaining them.
+Moreover, since
+.I lsof
+only has device numbers for the file systems that have alternates,
+its ability to locate files on file systems depends completely on the
+availability and accuracy of the alternates.
+If no alternates are available, or if they're incorrect,
+.I lsof
+won't be able to locate files on the named file systems.
+.PP
+Third, if the names of your file system directories that
+.I lsof
+obtains from your system's mount table are symbolic links,
+.I lsof
+won't be able to resolve the links.
+This is because the
+.B \-b
+option causes
+.I lsof
+to avoid the kernel
+.IR readlink (2)
+function it uses to resolve symbolic links.
+.PP
+Finally, using the
+.B \-b
+option causes
+.I lsof
+to issue warning messages when it needs to use the kernel functions
+that the
+.B \-b
+option directs it to avoid.
+You can suppress these messages by specifying the
+.B \-w
+option, but if you do, you won't see the alternate device numbers
+reported in the warning messages.
+.SH "ALTERNATE DEVICE NUMBERS"
+.PP
+On some dialects, when
+.I lsof
+has to break a block because it can't get information about a
+mounted file system via the
+.IR lstat (2)
+and
+.IR stat (2)
+kernel functions, or because you specified the
+.B \-b
+option,
+.I lsof
+can obtain some of the information it needs \- the device number and
+possibly the file system type \- from the system mount table.
+When that is possible,
+.I lsof
+will report the device number it obtained.
+(You can suppress the report by specifying the
+.B \-w
+option.)
+.PP
+You can assist this process if your mount table is supported with an
+.I /etc/mtab
+or
+.I /etc/mnttab
+file that contains an options field by adding a ``dev=xxxx'' field for
+mount points that do not have one in their options strings.
+Note: you must be able to edit the file \- i.e., some mount tables
+like recent Solaris /etc/mnttab or Linux /proc/mounts are read\-only
+and can't be modified.
+.PP
+You may also be able to supply device numbers using the
+.B +m
+and
+.BI +m " m"
+options, provided they are supported by your dialect.
+Check the output of
+.I lsof's
+.B \-h
+or
+.B \-?
+options to see if the
+.B +m
+and
+.BI +m " m"
+options are available.
+.PP
+The ``xxxx'' portion of the field is the hexadecimal value
+of the file system's device number.
+(Consult the
+.I st_dev
+field of the output of the
+.IR lstat (2)
+and
+.IR stat (2)
+functions for the appropriate values for your file systems.)
+Here's an example from a Sun Solaris 2.6
+.I /etc/mnttab
+for a file system remotely mounted via NFS:
+.PP
+.nf
+ nfs ignore,noquota,dev=2a40001
+.fi
+.PP
+There's an advantage to having ``dev=xxxx'' entries in your mount
+table file, especially for file systems that are mounted from remote
+NFS servers.
+When a remote server crashes and you want to identify its users by running
+.I lsof
+on one of its clients,
+.I lsof
+probably won't be able to get output from the
+.IR lstat (2)
+and
+.IR stat (2)
+functions for the file system.
+If it can obtain the file system's device number from the mount table,
+it will be able to display the files open on the crashed NFS server.
+.PP
+Some dialects that do not use an ASCII
+.I /etc/mtab
+or
+.I /etc/mnttab
+file for the mount table may still provide an alternative device number
+in their internal mount tables.
+This includes AIX, Apple Darwin, FreeBSD, NetBSD, OpenBSD, and Tru64 UNIX.
+.I Lsof
+knows how to obtain the alternative device number for these dialects
+and uses it when its attempt to
+.IR lstat (2)
+or
+.IR stat (2)
+the file system is blocked.
+.PP
+If you're not sure your dialect supplies alternate device numbers
+for file systems from its mount table, use this
+.I lsof
+incantation to see if it reports any alternate device numbers:
+.PP
+.IP
+lsof -b
+.PP
+Look for standard error file warning messages that
+begin ``assuming "dev=xxxx" from ...''.
+.SH "KERNEL NAME CACHE"
+.PP
+.I Lsof
+is able to examine the kernel's name cache or use other kernel
+facilities (e.g., the ADVFS 4.x tag_to_path() function under
+Tru64 UNIX) on some dialects for most file system types,
+excluding AFS, and extract recently used path name components from it.
+(AFS file system path lookups don't use the kernel's name cache; some
+Solaris VxFS file system operations apparently don't use it, either.)
+.PP
+.I Lsof
+reports the complete paths it finds in the NAME column.
+If
+.I lsof
+can't report all components in a path, it reports in the NAME column
+the file system name, followed by a space, two `-' characters, another
+space, and the name components it has located, separated by
+the `/' character.
+.PP
+When
+.I lsof
+is run in repeat mode \- i.e., with the
+.B \-r
+option specified \- the extent to which it can report path name
+components for the same file may vary from cycle to cycle.
+That's because other running processes can cause the kernel to
+remove entries from its name cache and replace them with others.
+.PP
+.I Lsof's
+use of the kernel name cache to identify the paths of files
+can lead it to report incorrect components under some circumstances.
+This can happen when the kernel name cache uses device and node
+number as a key (e.g., SCO OpenServer) and a key on a rapidly
+changing file system is reused.
+If the UNIX dialect's kernel doesn't purge the name cache entry for
+a file when it is unlinked,
+.I lsof
+may find a reference to the wrong entry in the cache.
+The
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+has more information on this situation.
+.PP
+.I Lsof
+can report path name components for these dialects:
+.PP
+.nf
+ FreeBSD
+ HP\-UX
+ Linux
+ NetBSD
+ NEXTSTEP
+ OpenBSD
+ OPENSTEP
+ SCO OpenServer
+ SCO|Caldera UnixWare
+ Solaris
+ Tru64 UNIX
+.fi
+.PP
+.I Lsof
+can't report path name components for these dialects:
+.PP
+.nf
+ AIX
+.fi
+.PP
+If you want to know why
+.I lsof
+can't report path name components for some dialects, see the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+.SH "DEVICE CACHE FILE"
+.PP
+Examining all members of the
+.I /dev
+(or
+.IR /devices )
+node tree with
+.IR stat (2)
+functions can be time consuming.
+What's more, the information that
+.I lsof
+needs \- device number, inode number, and path \- rarely changes.
+.PP
+Consequently,
+.I lsof
+normally maintains an ASCII text file of cached
+.I /dev
+(or
+.IR /devices )
+information (exception: the /proc\-based Linux
+.I lsof
+where it's not needed.)
+The local system administrator who builds
+.I lsof
+can control the way the device cache file path is formed, selecting
+from these options:
+.PP
+.nf
+ Path from the \fB\-D\fP option;
+ Path from an environment variable;
+ System\-wide path;
+ Personal path (the default);
+ Personal path, modified by an environment variable.
+.fi
+.PP
+Consult the output of the
+.BR \-h ,
+.B \-D? ,
+or
+.B \-?
+help options for the current state of device cache support.
+The help output lists the default read\-mode device cache file path that
+is in effect for the current invocation of
+.IR lsof .
+The
+.B \-D?
+option output lists the read\-only and write device cache file paths,
+the names of any applicable environment variables, and the personal
+device cache path format.
+.PP
+.I Lsof
+can detect that the current device cache file has been accidentally
+or maliciously modified by integrity checks, including the computation
+and verification of a sixteen bit Cyclic Redundancy Check (CRC) sum on
+the file's contents.
+When
+.I lsof
+senses something wrong with the file, it issues a warning and attempts
+to remove the current cache file and create a new copy, but only to
+a path that the process can legitimately write.
+.PP
+The path from which a
+.I lsof
+process may attempt to read a device cache file may not be the same
+as the path to which it can legitimately write.
+Thus when
+.I lsof
+senses that it needs to update the device cache file, it may
+choose a different path for writing it from the path from which
+it read an incorrect or outdated version.
+.PP
+If available, the
+.B \-Dr
+option will inhibit the writing of a new device cache file.
+(It's always available when specified without a path name argument.)
+.PP
+When a new device is added to the system, the device cache file may
+need to be recreated.
+Since
+.I lsof
+compares the mtime of the device cache file with the mtime and ctime
+of the
+.I /dev
+(or
+.IR /devices )
+directory, it usually detects that a new device has been added;
+in that case
+.I lsof
+issues a warning message and attempts to rebuild the device cache file.
+.PP
+Whenever
+.I lsof
+writes a device cache file, it sets its ownership to the real UID
+of the executing process, and its permission modes to 0600, this
+restricting its reading and writing to the file's owner.
+.SH "LSOF PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS"
+.PP
+Two permissions of the
+.I lsof
+executable affect its ability to access device cache files.
+The permissions are set by the local system administrator when
+.I lsof
+is installed.
+.PP
+The first and rarer permission is setuid\-root.
+It comes into effect when
+.I lsof
+is executed; its effective UID is then
+root, while its real (i.e., that of the logged\-on user) UID is not.
+The
+.I lsof
+distribution recommends that versions for these dialects run setuid\-root.
+.PP
+.nf
+ HP-UX 11.11 and 11.23
+ Linux
+.fi
+.PP
+The second and more common permission is setgid.
+It comes into effect when the effective group IDentification number (GID)
+of the
+.I lsof
+process is set to one that can access kernel memory devices \-
+e.g., ``kmem'', ``sys'', or ``system''.
+.PP
+An
+.I lsof
+process that has setgid permission usually surrenders the permission
+after it has accessed the kernel memory devices.
+When it does that,
+.I lsof
+can allow more liberal device cache path formations.
+The
+.I lsof
+distribution recommends that versions for these dialects run setgid
+and be allowed to surrender setgid permission.
+.PP
+.nf
+ AIX 5.[12] and 5.3-ML1
+ Apple Darwin 7.x Power Macintosh systems
+ FreeBSD 4.x, 4.1x, 5.x and [67].x for x86-based systems
+ FreeBSD 5.x and [67].x for Alpha, AMD64 and Sparc64-based
+ systems
+ HP\-UX 11.00
+ NetBSD 1.[456], 2.x and 3.x for Alpha, x86, and SPARC-based
+ systems
+ NEXTSTEP 3.[13] for NEXTSTEP architectures
+ OpenBSD 2.[89] and 3.[0\-9] for x86-based systems
+ OPENSTEP 4.x
+ SCO OpenServer Release 5.0.6 for x86-based systems
+ SCO|Caldera UnixWare 7.1.4 for x86-based systems
+ Solaris 2.6, 8, 9 and 10
+ Tru64 UNIX 5.1
+.fi
+.PP
+(Note:
+.I lsof
+for AIX 5L and above needs setuid\-root permission if its
+.B \-X
+option is used.)
+.PP
+.I Lsof
+for these dialects does not support a device cache, so the permissions
+given to the executable don't apply to the device cache file.
+.PP
+.nf
+ Linux
+.fi
+.SH "DEVICE CACHE FILE PATH FROM THE \-D OPTION"
+.PP
+The
+.B \-D
+option provides limited means for specifying the device cache file path.
+Its
+.B ?
+function will report the read\-only and write device cache file paths that
+.I lsof
+will use.
+.PP
+When the
+.B \-D
+.BR b ,
+.BR r ,
+and
+.B u
+functions are available, you can use them to request that the cache file be
+built in a specific location (\fBb\fR[\fIpath\fR]);
+read but not rebuilt (\fBr\fR[\fIpath\fR]);
+or read and rebuilt (\fBu\fR[\fIpath\fR]).
+The
+.BR b ,
+.BR r ,
+and
+.B u
+functions are restricted under some conditions.
+They are restricted when the
+.I lsof
+process is setuid\-root.
+The path specified with the
+.B r
+function is always read\-only, even
+when it is available.
+.PP
+The
+.BR b ,
+.BR r ,
+and
+.B u
+functions are also restricted when the
+.I lsof
+process runs setgid and
+.I lsof
+doesn't surrender the setgid permission.
+(See the
+.B "LSOF PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS"
+section for a list of implementations that normally don't surrender
+their setgid permission.)
+.PP
+A further
+.B \-D
+function,
+.B i
+(for ignore), is always available.
+.PP
+When available, the
+.B b
+function tells
+.I lsof
+to read device information from the kernel with the
+.IR stat (2)
+function and build a device cache file at the indicated path.
+.PP
+When available, the
+.B r
+function tells
+.I lsof
+to read the device cache file, but not update it.
+When a path argument accompanies
+.BR \-Dr ,
+it names the device cache file path.
+The
+.B r
+function is always available when it is specified without a
+path name argument.
+If
+.I lsof
+is not running setuid\-root and surrenders its setgid permission,
+a path name argument may accompany the
+.B r
+function.
+.PP
+When available, the
+.B u
+function tells
+.I lsof
+to attempt to read and use the device cache file.
+If it can't read the file, or if it finds the contents of the
+file incorrect or outdated, it will read information from the kernel,
+and attempt to write an updated version of the device cache file,
+but only to a path it considers legitimate for the
+.I lsof
+process effective and real UIDs.
+.SH "DEVICE CACHE PATH FROM AN ENVIRONMENT VARIABLE"
+.PP
+.I Lsof's
+second choice for the device cache file is the contents of the
+LSOFDEVCACHE environment variable.
+It avoids this choice if the
+.I lsof
+process is setuid\-root, or the real UID of the process is root.
+.PP
+A further restriction applies to a device cache file path taken from
+the LSOFDEVCACHE environment variable:
+.I lsof
+will not write a device cache file to the path if the
+.I lsof
+process doesn't surrender its setgid permission.
+(See the
+.B "LSOF PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS"
+section for information on implementations that don't surrender
+their setgid permission.)
+.PP
+The local system administrator can disable the use of the LSOFDEVCACHE
+environment variable or change its name when building
+.IR lsof .
+Consult the output of
+.B \-D?
+for the environment variable's name.
+.SH "SYSTEM-WIDE DEVICE CACHE PATH"
+.PP
+The local system administrator may choose to have a system\-wide
+device cache file when building
+.IR lsof .
+That file will generally be constructed by a special system administration
+procedure when the system is booted or when the contents of
+.I /dev
+or
+.IR /devices )
+changes.
+If defined, it is
+.I lsof's
+third device cache file path choice.
+.PP
+You can tell that a system\-wide device cache file is in effect
+for your local installation by examining the
+.I lsof
+help option output \- i.e., the output from the
+.B \-h
+or
+.B \-?
+option.
+.PP
+.I Lsof
+will never write to the system\-wide device cache file path by
+default.
+It must be explicitly named with a
+.B \-D
+function in a root\-owned procedure.
+Once the file has been written, the procedure must change its permission
+modes to 0644 (owner\-read and owner\-write, group\-read, and other\-read).
+.SH "PERSONAL DEVICE CACHE PATH (DEFAULT)"
+.PP
+The default device cache file path of the
+.I lsof
+distribution is one recorded in the home directory of the real UID
+that executes
+.IR lsof .
+Added to the home directory is a second path component of the form
+.IR .lsof_hostname .
+.PP
+This is
+.I lsof's
+fourth device cache file path choice, and is
+usually the default.
+If a system\-wide device cache file path was defined when
+.I lsof
+was built,
+this fourth choice will be applied when
+.I lsof
+can't find the system\-wide device cache file.
+This is the
+.B only
+time
+.I lsof
+uses two paths when reading the device cache file.
+.PP
+The
+.I hostname
+part of the second component is the base
+name of the executing host, as returned by
+.IR gethostname (2).
+The base name is defined to be the characters preceding the first `.'
+in the
+.IR gethostname (2)
+output, or all the
+.IR gethostname (2)
+output if it contains no `.'.
+.PP
+The device cache file belongs to the user ID and is readable and
+writable by the user ID alone \- i.e., its modes are 0600.
+Each distinct real user ID on a given host that executes
+.I lsof
+has a distinct device cache file.
+The
+.I hostname
+part of the path distinguishes device cache files in an NFS\-mounted
+home directory into which device cache files are written from
+several different hosts.
+.PP
+The personal device cache file path formed by this method represents
+a device cache file that
+.I lsof
+will attempt to read, and will attempt to write should it not
+exist or should its contents be incorrect or outdated.
+.PP
+The
+.B \-Dr
+option without a path name argument will inhibit the writing of a new
+device cache file.
+.PP
+The
+.B \-D?
+option will list the format specification for constructing the
+personal device cache file.
+The conversions used in the format specification are described in the
+.I 00DCACHE
+file of the
+.I lsof
+distribution.
+.SH "MODIFIED PERSONAL DEVICE CACHE PATH"
+.PP
+If this option is defined by the local system administrator when
+.I lsof
+is built, the LSOFPERSDCPATH environment variable contents may
+be used to add a component of the personal device cache file path.
+.PP
+The LSOFPERSDCPATH variable contents are inserted in the path at the
+place marked by the local system administrator with the ``%p''
+conversion in the HASPERSDC format specification of the dialect's
+.I machine.h
+header file.
+(It's placed right after the home directory in the default
+.I lsof
+distribution.)
+.PP
+Thus, for example, if LSOFPERSDCPATH contains ``LSOF'', the home
+directory is ``/Homes/abe'', the host name is ``lsof.itap.purdue.edu'',
+and the HASPERSDC format is the default (``%h/%p.lsof_%L''), the
+modified personal device cache file path is:
+.PP
+.nf
+ /Homes/abe/LSOF/.lsof_vic
+.fi
+.PP
+The LSOFPERSDCPATH environment variable is ignored when the
+.I lsof
+process is setuid\-root or when the real UID of the process is root.
+.PP
+.I Lsof
+will not write to a modified personal device cache file path if the
+.I lsof
+process doesn't surrender setgid permission.
+(See the
+.B "LSOF PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS"
+section for a list of implementations that normally don't surrender
+their setgid permission.)
+.PP
+If, for example, you want to create a sub\-directory of personal
+device cache file paths by using the LSOFPERSDCPATH environment
+variable to name it, and
+.I lsof
+doesn't surrender its setgid permission, you will have to allow
+.I lsof
+to create device cache files at the standard personal path and
+move them to your subdirectory with shell commands.
+.PP
+The local system administrator may: disable this option when
+.I lsof
+is built; change the name of the environment variable from
+LSOFPERSDCPATH to something else; change the HASPERSDC
+format to include the personal path component in another place;
+or exclude the personal path component entirely.
+Consult the output of the
+.B \-D?
+option for the environment variable's name and the HASPERSDC
+format specification.
+.SH DIAGNOSTICS
+Errors are identified with messages on the standard error file.
+.PP
+.I Lsof
+returns a one (1) if any error was detected, including the failure to
+locate command names, file names, Internet addresses or files, login
+names, NFS files, PIDs, PGIDs, or UIDs it was asked to list.
+If the
+.B \-V
+option is specified,
+.I lsof
+will indicate the search items it failed to list.
+.PP
+It returns a zero (0) if no errors were detected and if it was able to
+list some information about all the specified search arguments.
+.PP
+.PP
+When
+.I lsof
+cannot open access to
+.I /dev
+(or
+.IR /devices )
+or one of its subdirectories, or get information on a file in them with
+.IR stat (2),
+it issues a warning message and continues.
+That
+.I lsof
+will issue warning messages about inaccessible files in
+.I /dev
+(or
+.IR /devices )
+is indicated in its help output \- requested with the
+.B \-h
+or
+>B \-?
+options \- with the message:
+.PP
+.nf
+ Inaccessible /dev warnings are enabled.
+.fi
+.PP
+The warning message may be suppressed with the
+.B \-w
+option.
+It may also have been suppressed by the system administrator when
+.I lsof
+was compiled by the setting of the WARNDEVACCESS definition.
+In this case, the output from the help options will include the message:
+.PP
+.nf
+ Inaccessible /dev warnings are disabled.
+.fi
+.PP
+Inaccessible device warning messages usually disappear after
+.I lsof
+has created a working device cache file.
+.SH EXAMPLES
+For a more extensive set of examples, documented more fully, see the
+.I 00QUICKSTART
+file of the
+.I lsof
+distribution.
+.PP
+To list all open files, use:
+.IP
+lsof
+.PP
+To list all open Internet, x.25 (HP\-UX), and UNIX domain files, use:
+.IP
+lsof -i -U
+.PP
+To list all open IPv4 network files in use by the process whose PID is
+1234, use:
+.IP
+lsof -i 4 -a -p 1234
+.PP
+Presuming the UNIX dialect supports IPv6, to list only open IPv6
+network files, use:
+.IP
+lsof -i 6
+.PP
+To list all files using any protocol on ports 513, 514, or 515 of host
+wonderland.cc.purdue.edu, use:
+.IP
+lsof -i @wonderland.cc.purdue.edu:513-515
+.PP
+To list all files using any protocol on any port of mace.cc.purdue.edu
+(cc.purdue.edu is the default domain), use:
+.IP
+lsof -i @mace
+.PP
+To list all open files for login name ``abe'', or user ID 1234, or
+process 456, or process 123, or process 789, use:
+.IP
+lsof -p 456,123,789 -u 1234,abe
+.PP
+To list all open files on device /dev/hd4, use:
+.IP
+lsof /dev/hd4
+.PP
+To find the process that has /u/abe/foo open, use:
+.IP
+lsof /u/abe/foo
+.PP
+To send a SIGHUP to the processes that have /u/abe/bar open, use:
+.IP
+kill -HUP `lsof -t /u/abe/bar`
+.PP
+To find any open file, including an open UNIX domain socket file,
+with the name
+.IR /dev/log ,
+use:
+.IP
+lsof /dev/log
+.PP
+To find processes with open files on the NFS file system named
+.I /nfs/mount/point
+whose server is inaccessible, and presuming your mount table supplies
+the device number for
+.IR /nfs/mount/point ,
+use:
+.IP
+lsof -b /nfs/mount/point
+.PP
+To do the preceding search with warning messages suppressed, use:
+.IP
+lsof -bw /nfs/mount/point
+.PP
+To ignore the device cache file, use:
+.IP
+lsof -Di
+.PP
+To obtain PID and command name field output for each process, file
+descriptor, file device number, and file inode number for each file
+of each process, use:
+.IP
+lsof -FpcfDi
+.PP
+To list the files at descriptors 1 and 3 of every process running the
+.I lsof
+command for login ID ``abe'' every 10 seconds, use:
+.IP
+lsof -c lsof -a -d 1 -d 3 -u abe -r10
+.PP
+To list the current working directory of processes running a command that
+is exactly four characters long and has an 'o' or 'O' in character three,
+use this regular expression form of the
+.BI \-c " c"
+option:
+.IP
+lsof -c /^..o.$/i -a -d cwd
+.PP
+To find an IP version 4 socket file by its associated numeric dot\-form
+address, use:
+.IP
+lsof -i@128.210.15.17
+.PP
+To find an IP version 6 socket file (when the UNIX dialect supports
+IPv6) by its associated numeric colon\-form address, use:
+.IP
+lsof -i@[0:1:2:3:4:5:6:7]
+.PP
+To find an IP version 6 socket file (when the UNIX dialect supports
+IPv6) by an associated numeric colon\-form address that has a run of
+zeroes in it \- e.g., the loop\-back address \- use:
+.IP
+lsof -i@[::1]
+.PP
+To obtain a repeat mode marker line that contains the current time, use:
+.IP
+lsof -rm====%T====
+.PP
+To add spaces to the previous marker line, use:
+.IP
+lsof -r "m==== %T ===="
+.SH BUGS
+Since
+.I lsof
+reads kernel memory in its search for open files, rapid changes in kernel
+memory may produce unpredictable results.
+.PP
+When a file has multiple record locks, the lock status character
+(following the file descriptor) is derived from a test of the first
+lock structure, not from any combination of the individual record
+locks that might be described by multiple lock structures.
+.PP
+.I Lsof
+can't search for files with restrictive access permissions by
+.I name
+unless it is installed with root set\-UID permission.
+Otherwise it is limited to searching for files to which its user
+or its set-GID group (if any) has access permission.
+.PP
+The display of the destination address of a raw socket (e.g., for
+.IR ping )
+depends on the UNIX operating system.
+Some dialects store the destination address in the raw socket's protocol
+control block, some do not.
+.PP
+.I Lsof
+can't always represent Solaris device numbers in the same way that
+.IR ls (1)
+does.
+For example, the major and minor device numbers that the
+.IR lstat (2)
+and
+.IR stat (2)
+functions report for the directory on which CD-ROM files are mounted
+(typically
+.IR /cdrom )
+are not the same as the ones that it reports for the device on which
+CD-ROM files are mounted (typically
+.IR /dev/sr0 ).
+(\fILsof\fP reports the directory numbers.)
+.PP
+The support for
+.I /proc
+file systems is available only for BSD and Tru64 UNIX dialects, Linux, and
+dialects derived from SYSV R4 \- e.g., FreeBSD, NetBSD, OpenBSD, Solaris,
+UnixWare.
+.PP
+Some
+.I /proc
+file items \- device number, inode number, and file size \-
+are unavailable in some dialects.
+Searching for files in a
+.I /proc
+file system may require that the full path name be specified.
+.PP
+No text (\fBtxt\fP) file descriptors are displayed for Linux
+processes.
+All entries for files other than the current working directory,
+the root directory, and numerical file descriptors are labeled
+.B mem
+descriptors.
+.PP
+.I Lsof
+can't search for Tru64 UNIX named pipes by name, because their kernel
+implementation of lstat(2) returns an improper device number for a
+named pipe.
+.PP
+.I Lsof
+can't report fully or correctly on HP\-UX 9.01, 10.20, and 11.00 locks
+because of insufficient access to kernel data or errors in the
+kernel data.
+See the
+.I lsof
+FAQ (The \fBFAQ\fP section gives its location.)
+for details.
+.PP
+The AIX SMT file type is a fabrication.
+It's made up for file structures whose type (15) isn't defined in the AIX
+.I /usr/include/sys/file.h
+header file.
+One way to create such file structures is to run X clients with the DISPLAY
+variable set to ``:0.0''.
+.PP
+The
+.BI +|\-f [cfgGn]
+option is not supported under /proc\-based Linux
+.IR lsof ,
+because it doesn't read kernel structures from kernel memory.
+.SH ENVIRONMENT
+.I Lsof
+may access these environment variables.
+.TP \w'LSOFPERSDCPATH'u+4
+LANG
+defines a language locale.
+See
+.IR setlocale (3)
+for the names of other variables that can be used in place
+of LANG \- e.g., LC_ALL, LC_TYPE, etc.
+.TP
+LSOFDEVCACHE
+defines the path to a device cache file.
+See the
+.B "DEVICE CACHE PATH FROM AN ENVIRONMENT VARIABLE"
+section for more information.
+.TP
+LSOFPERSDCPATH
+defines the middle component of a modified personal device cache
+file path.
+See the
+.B "MODIFIED PERSONAL DEVICE CACHE PATH"
+section for more information.
+.SH FAQ
+Frequently-asked questions and their answers (an FAQ) are
+available in the
+.I 00FAQ
+file of the
+.I lsof
+distribution.
+.PP
+That file is also available via anonymous ftp from
+.I lsof.itap.purdue.edu
+at
+.IR pub/tools/unix/lsof FAQ .
+The URL is:
+.IP
+ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/FAQ
+.SH FILES
+.TP \w'.lsof_hostname'u+4
+.I /dev/kmem
+kernel virtual memory device
+.TP
+.I /dev/mem
+physical memory device
+.TP
+.I /dev/swap
+system paging device
+.TP
+.I .lsof_hostname
+.I lsof's
+device cache file
+(The suffix,
+.IR hostname ,
+is the first component of the host's name returned by
+.IR gethostname (2) .)
+.SH AUTHORS
+.I Lsof
+was written by Victor A. Abell <abe@purdue.edu> of Purdue University.
+Many others have contributed to
+.IR lsof .
+They're listed in the
+.I 00CREDITS
+file of the
+.I lsof
+distribution.
+.SH DISTRIBUTION
+The latest distribution of
+.I lsof
+is available via anonymous ftp from the host
+.IR lsof.itap.purdue.edu .
+You'll find the
+.I lsof
+distribution in the
+.I pub/tools/unix/lsof
+directory.
+.PP
+You can also use this URL:
+.IP
+ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof
+.PP
+.I Lsof
+is also mirrored elsewhere.
+When you access
+.I lsof.itap.purdue.edu
+and change to its
+.I pub/tools/unix/lsof
+directory, you'll be given a list of some mirror sites.
+The
+.I pub/tools/unix/lsof
+directory also contains a more complete list in its
+.I mirrors
+file.
+Use mirrors with caution \- not all mirrors always have the latest
+.I lsof
+revision.
+.PP
+Some pre\-compiled
+.I Lsof
+executables are available on
+.IR lsof.itap.purdue.edu ,
+but their use is discouraged \- it's better that you build
+your own from the sources.
+If you feel you must use a pre\-compiled executable, please
+read the cautions that appear in the README files of the
+.I pub/tools/unix/lsof/binaries
+subdirectories and in the 00* files of the distribution.
+.PP
+More information on the
+.I lsof
+distribution can be found in its
+.I README.lsof_<version>
+file.
+If you intend to get the
+.I lsof
+distribution and build it, please read
+.I README.lsof_<version>
+and the other 00* files of the distribution before sending questions
+to the author.
+.SH SEE ALSO
+.PP
+Not all the following manual pages may exist in every UNIX
+dialect to which
+.I lsof
+has been ported.
+.PP
+access(2),
+awk(1),
+crash(1),
+fattach(3C),
+ff(1),
+fstat(8),
+fuser(1),
+gethostname(2),
+isprint(3),
+kill(1),
+localtime(3),
+lstat(2),
+modload(8),
+mount(8),
+netstat(1),
+ofiles(8L),
+perl(1),
+ps(1),
+readlink(2),
+setlocale(3),
+stat(2),
+strftime(3),
+time(2),
+uname(1).
--- /dev/null
+/*
+ * lsof.h - common header file for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * $Id: lsof.h,v 1.62 2009/03/25 19:20:38 abe Exp $
+ */
+
+
+#if !defined(LSOF_H)
+#define LSOF_H 1
+
+#include "machine.h"
+
+# if !defined(FSV_DEFAULT)
+#define FSV_DEFAULT 0
+# endif /* !defined(FSV_DEFAULT) */
+
+#include "lsof_fields.h"
+
+#include <ctype.h>
+#include <errno.h>
+
+# if defined(HASSETLOCALE)
+#include <locale.h>
+# endif /* defined(HASSETLOCALE) */
+
+#include <netdb.h>
+#include <pwd.h>
+#include <stdio.h>
+
+#include <sys/stat.h>
+#include <sys/wait.h>
+
+
+/*
+ * Definitions and structures that may be needed by dlsof.h
+ */
+
+# if !defined(INODETYPE)
+#define INODETYPE unsigned long /* node number storage type */
+#define INODEPSPEC "l" /* node number printf specification
+ * modifier */
+# endif /* !defined(INODETYPE) */
+
+struct l_dev {
+ dev_t rdev; /* device */
+ INODETYPE inode; /* inode number */
+ char *name; /* name */
+ int v; /* has been verified
+ * (when DCUnsafe == 1) */
+};
+
+
+/*
+ * FILE_FLAG column names
+ */
+
+#define FF_AIO "AIO"
+#define FF_APPEND "AP"
+#define FF_ASYNC "ASYN"
+#define FF_BLKANDSET "BAS"
+#define FF_BLKINUSE "BKIU"
+#define FF_BLKSEEK "BSK"
+#define FF_CIO "CIO"
+#define FF_CLONE "CLON"
+#define FF_CLREAD "CLRD"
+#define FF_COPYAVOID "CA"
+#define FF_CREAT "CR"
+#define FF_DATAFLUSH "DFLU"
+#define FF_DEFER "DF"
+#define FF_DEFERIND "DFI"
+#define FF_DELAY "DLY"
+#define FF_DIRECT "DIR"
+#define FF_DIRECTORY "DTY"
+#define FF_DOCLONE "DOCL"
+#define FF_DSYNC "DSYN"
+#define FF_EVTONLY "EVO"
+#define FF_EXCL "EXCL"
+#define FF_EXEC "EX"
+#define FF_EXLOCK "XL"
+#define FF_FILE_MBLK "MBLK"
+#define FF_FSYNC "FSYN"
+#define FF_GCFDEFER "GCDF"
+#define FF_GCFMARK "GCMK"
+#define FF_GENTTY "GTTY"
+#define FF_HASLOCK "LCK"
+#define FF_HUP "HUP"
+#define FF_KERNEL "KERN"
+#define FF_KIOCTL "KIOC"
+#define FF_LARGEFILE "LG"
+#define FF_MARK "MK"
+#define FF_MOUNT "MNT"
+#define FF_MSYNC "MSYN"
+#define FF_NBDRM "NBDR"
+#define FF_NBIO "NBIO"
+#define FF_NBLOCK "NB"
+#define FF_NBUF "NBF"
+#define FF_NMFS "NMFS"
+#define FF_NDELAY "ND"
+#define FF_NET "NET"
+#define FF_NOATM "NATM"
+#define FF_NOCACHE "NC"
+#define FF_NOCTTY "NTTY"
+#define FF_NODSYNC "NDSY"
+#define FF_NOFOLNK "NFLK"
+#define FF_NOTOSTOP "NOTO"
+#define FF_NSHARE "NSH"
+#define FF_OLRMIRROR "OLRM"
+#define FF_POSIX_AIO "PAIO"
+#define FF_POSIX_PIPE "PP"
+#define FF_RAIOSIG "RAIO"
+#define FF_RCACH "RC"
+#define FF_RDWR "RW"
+#define FF_READ "R"
+#define FF_REVOKED "REV"
+#define FF_RSHARE "RSH"
+#define FF_RSYNC "RSYN"
+#define FF_SETBLK "BL"
+#define FF_SHLOCK "SL"
+#define FF_SNAP "SNAP"
+#define FF_SOCKET "SOCK"
+#define FF_SQTSH1 "SQS1"
+#define FF_SQTSH2 "SQS2"
+#define FF_SQTREPAIR "SQR"
+#define FF_SQTSH "SQSH"
+#define FF_SQTSVM "SQSV"
+#define FF_STOPIO "STPI"
+#define FF_SYNC "SYN"
+#define FF_SYNCRON "SWR"
+#define FF_TCP_MDEVONLY "TCPM"
+#define FF_TERMIO "TIO"
+#define FF_TRUNC "TR"
+#define FF_VHANGUP "VH"
+#define FF_VTEXT "VTXT"
+#define FF_WAKEUP "WKUP"
+#define FF_WAITING "WTG"
+#define FF_WRITE "W"
+
+
+/*
+ * Process open file flag names
+ */
+
+#define POF_ALLOCATED "ALLC"
+#define POF_BNRD "BR"
+#define POF_BNWR "BW"
+#define POF_BNHUP "BHUP"
+#define POF_CLOEXEC "CX"
+#define POF_CLOSING "CLSG"
+#define POF_FDLOCK "LCK"
+#define POF_INUSE "USE"
+#define POF_MAPPED "MP"
+#define POF_FSHMAT "SHMT"
+#define POF_RESERVED "OPIP"
+#define POF_RSVWT "RSVW"
+
+
+/*
+ * Cross-over (-x) option values
+ */
+
+#define XO_FILESYS 0x1 /* file system mount points */
+#define XO_SYMLINK 0x2 /* symbolic links */
+#define XO_ALL (XO_FILESYS | XO_SYMLINK)
+
+#include "dlsof.h"
+
+#include <sys/types.h> /* just in case -- because utmp.h
+ * may need it */
+#include "./regex.h"
+
+# if defined(EMPTY)
+#undef EMPTY
+# endif /* defined(EMPTY) */
+
+# if defined(HASUTMPX)
+#include <utmpx.h>
+# else /* !defined(HASUTMPX) */
+#include <utmp.h>
+# endif /* defined(HASUTMPX) */
+
+extern int errno;
+extern char *optarg;
+extern int optind;
+
+#define ACCESSERRFMT "%s: WARNING: access %s: %s\n"
+
+# if defined(HASDCACHE)
+#define CRC_POLY 0120001 /* CRC-16 polynomial */
+#define CRC_TBLL 256 /* crc table length for software */
+#define CRC_BITS 8 /* number of bits contributing */
+# endif /* defined(HASDCACHE) */
+#define CMDL 9 /* maximum number of characters from
+ * command name to print in COMMAND
+ * column */
+#define CWD " cwd" /* current working directory fd name */
+#define FDLEN 8 /* fd printing array length */
+#define FSV_FA 0x1 /* file struct addr status */
+#define FSV_CT 0x2 /* file struct count status */
+#define FSV_FG 0x4 /* file struct flags */
+#define FSV_NI 0x8 /* file struct node ID status */
+
+# if !defined(GET_MAJ_DEV)
+#define GET_MAJ_DEV major /* if no dialect specific macro has
+ * been defined, use standard major()
+ * macro */
+# endif /* !defined(GET_MAJ_DEV) */
+
+# if !defined(GET_MIN_DEV)
+#define GET_MIN_DEV minor /* if no dialect specific macro has
+ * been defined, use standard minor()
+ * macro */
+# endif /* !defined(GET_MIN_DEV) */
+
+# if defined(HASSELINUX)
+#define HASHCNTX 128 /* security context hash bucket count
+ * -- MUST BE A POWER OF 2!!! */
+# endif /* defined(HASSELINUX) */
+
+# if defined(HASZONES)
+#define HASHZONE 128 /* zone hash bucket count -- MUST BE
+ * A POWER OF 2!!! */
+# endif /* defined(HASZONES) */
+
+#define IDINCR 10 /* PID/PGID table malloc() increment */
+
+# if !defined(INADDR_LOOPBACK)
+#define INADDR_LOOPBACK (u_long)0x7f000001
+# endif /* !defined(INADDR_LOOPBACK) */
+
+#define IPROTOL 8 /* Internet protocol length */
+
+# if !defined(KA_T_FMT_X)
+#define KA_T_FMT_X "0x%08lx" /* format for printing kernel
+ * addresses in 0x... format */
+# endif /* !defined(KA_T_FMT_X) */
+
+# if !defined(LOGINML)
+# if defined(HASUTMPX)
+static struct utmpx dummy_utmp; /* to get login name length */
+# else /* !defined(HASUTMPX) */
+static struct utmp dummy_utmp; /* to get login name length */
+# endif /* defined(HASUTMPX) */
+#define LOGINML sizeof(dummy_utmp.ut_name)
+ /* login name length */
+# endif /* !defined(LOGINML) */
+
+#define LPROCINCR 128 /* Lproc[] allocation increment */
+#define LSOF_URL "ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/"
+#define MIN_AF_ADDR sizeof(struct in_addr)
+ /* minimum AF_* address length */
+
+# if defined(HASIPv6)
+#define MAX_AF_ADDR sizeof(struct in6_addr)
+ /* maximum AF_* address length */
+# else /* !defined(HASIPv6) */
+#define MAX_AF_ADDR MIN_AF_ADDR /* maximum AF_* address length */
+# endif /* defined(HASIPv6) */
+
+#define MAXDCPATH 4 /* paths in DCpath[] */
+#define MAXNWAD 100 /* maximum network addresses */
+
+# if !defined(MEMMOVE)
+#define MEMMOVE memmove
+# endif /* !defined*MEMMOVE) */
+
+#define N_REGLR 0 /* regular file system node */
+#define N_AFS 1 /* AFS node */
+#define N_AFPFS 2 /* Apple Darwin AppleShare */
+#define N_AUSX 3 /* Auspex LFS node */
+#define N_AUTO 4 /* automount node */
+#define N_BLK 5 /* block device node */
+#define N_CACHE 6 /* cached file system node */
+#define N_CDFS 7 /* CD-ROM node */
+#define N_CFS 8 /* CFS node */
+#define N_CHR 9 /* character device node */
+#define N_COM 10 /* streams common device node */
+#define N_CTFSADIR 11 /* Solaris CTFS adir node */
+#define N_CTFSBUND 12 /* Solaris CTFS bundle node */
+#define N_CTFSCDIR 13 /* Solaris CTFS cdir node */
+#define N_CTFSCTL 14 /* Solaris CTFS ctl node */
+#define N_CTFSEVT 15 /* Solaris CTFS events node */
+#define N_CTFSLATE 16 /* Solaris CTFS latest node */
+#define N_CTFSROOT 17 /* Solaris CTFS root node */
+#define N_CTFSSTAT 18 /* Solaris CTFS status node */
+#define N_CTFSSYM 19 /* Solaris CTFS symbolic node */
+#define N_CTFSTDIR 20 /* Solaris CTFS type node */
+#define N_CTFSTMPL 21 /* Solaris CTFS template node */
+#define N_DEV 22 /* DEV FS node */
+#define N_DOOR 23 /* DOOR node */
+#define N_FD 24 /* FD node */
+#define N_FIFO 25 /* FIFO node */
+#define N_HSFS 26 /* High Sierra node */
+#define N_KERN 27 /* BSD /kern node */
+#define N_LOFS 28 /* loopback node */
+#define N_MNT 29 /* mount file system device node */
+#define N_MPC 30 /* multiplexed device node */
+#define N_MVFS 31 /* multi-volume file system node (?) */
+#define N_NFS 32 /* NFS node */
+#define N_NFS4 33 /* NFS version 4 node */
+#define N_NM 34 /* named file system node */
+#define N_OBJF 35 /* objfs file system node */
+#define N_PCFS 36 /* PC file system node */
+#define N_PIPE 37 /* pipe device node */
+#define N_PORT 38 /* port node */
+#define N_PROC 39 /* /proc node */
+#define N_PSEU 49 /* pseudofs node */
+#define N_SAMFS 41 /* Solaris SAM-FS */
+#define N_SANFS 42 /* AIX SANFS */
+#define N_SHARED 43 /* Solaris sharedfs */
+#define N_SOCK 44 /* sock_vnodeops node */
+#define N_SPEC 45 /* spec_vnodeops node */
+#define N_STREAM 46 /* stream node */
+#define N_TMP 47 /* tmpfs node */
+#define N_UFS 48 /* UNIX file system node */
+#define N_VXFS 49 /* Veritas file system node */
+#define N_XFS 50 /* XFS node */
+#define N_ZFS 51 /* ZFS node */
+
+# if !defined(OFFDECDIG)
+#define OFFDECDIG 8 /* maximum number of digits in the
+ * offset decimal form (0t...) */
+# endif /* !defined(OFFDECDIG) */
+
+# if !defined(USELOCALREADDIR)
+#define CloseDir closedir /* use standard closedir() */
+#define OpenDir opendir /* use standard opendir() */
+#define ReadDir readdir /* use standard readdir() */
+# endif /* !defined(USELOCALREADDIR) */
+
+#define RPTTM 15 /* default repeat seconds */
+#define RTD " rtd" /* root directory fd name */
+#define TCPTPI_FLAGS 0x0001 /* report TCP/TPI socket options and
+ * state, and TCP_NODELAY state */
+#define TCPTPI_QUEUES 0x0002 /* report TCP/TPI queue lengths */
+#define TCPTPI_STATE 0x0004 /* report TCP/TPI state */
+#define TCPTPI_WINDOWS 0x0008 /* report TCP/TPI window sizes */
+#define TCPTPI_ALL (TCPTPI_QUEUES | TCPTPI_STATE | TCPTPI_WINDOWS)
+ /* report all TCP/TPI info */
+#define TCPUDPALLOC 32 /* allocation amount for TCP and UDP
+ * state tables */
+#define TMLIMIT 15 /* readlink() & stat() timeout sec */
+#define TMLIMMIN 2 /* minimum timeout */
+#define TYPEL 8 /* type character length */
+#define UIDCACHEL 1024 /* UID cache length */
+#define UIDINCR 10 /* UID table malloc() increment */
+#define USERPRTL 8 /* UID/login print length limit */
+
+# if !defined(SZOFFTYPE)
+#define SZOFFTYPE unsigned long /* type for size and offset */
+#undef SZOFFPSPEC
+#define SZOFFPSPEC "l" /* SZOFFTYPE printf specification
+ * modifier */
+# endif /* !defined(SZOFFTYPE) */
+
+# if !defined(TIMEVAL_LSOF)
+#define TIMEVAL_LSOF timeval
+# endif /* !defined(TIMEVAL_LSOF) */
+
+# if !defined(XDR_PMAPLIST)
+#define XDR_PMAPLIST xdr_pmaplist
+# endif /* !defined(XDR_PMAPLIST) */
+
+# if !defined(XDR_VOID)
+#define XDR_VOID xdr_void
+# endif /* !defined(XDR_VOID) */
+
+
+/*
+ * Output title definitions
+ */
+
+#define CMDTTL "COMMAND"
+extern int CmdColW;
+#define CNTXTTL "SECURITY-CONTEXT"
+extern int CntxColW;
+#define DEVTTL "DEVICE"
+extern int DevColW;
+#define FCTTL "FCT"
+extern int FcColW;
+#define FDTTL "FD"
+extern int FdColW;
+#define FGTTL "FILE-FLAG"
+extern int FgColW;
+#define FSTTL "FILE-ADDR"
+extern int FsColW;
+#define NITTL "NODE-ID"
+extern int NiColW;
+extern char *NiTtl;
+#define NLTTL "NLINK"
+extern int NlColW;
+#define NMTTL "NAME"
+extern int NmColW;
+#define NODETTL "NODE"
+extern int NodeColW;
+#define OFFTTL "OFFSET"
+#define PGIDTTL "PGID"
+extern int PgidColW;
+#define PIDTTL "PID"
+extern int PidColW;
+#define PPIDTTL "PPID"
+extern int PpidColW;
+#define SZTTL "SIZE"
+#define SZOFFTTL "SIZE/OFF"
+extern int SzOffColW;
+#define TYPETTL "TYPE"
+extern int TypeColW;
+#define USERTTL "USER"
+extern int UserColW;
+#define ZONETTL "ZONE"
+extern int ZoneColW;
+
+
+/*
+ * Selection flags
+ */
+
+#define PS_PRI 1 /* primary process selection -- e.g.,
+ * by PID or UID */
+#define PS_SEC 2 /* secondary process selection -- e.g.,
+ * by directory or file */
+#define SELCMD 0x0001 /* select process by command name */
+#define SELCNTX 0x0002 /* select security context (-Z) */
+#define SELFD 0x0004 /* select file by descriptor name */
+#define SELNA 0x0008 /* select socket by address (-i@...) */
+#define SELNET 0x0010 /* select Internet socket files (-i) */
+#define SELNFS 0x0020 /* select NFS files (-N) */
+#define SELNLINK 0x0040 /* select based on link count */
+#define SELNM 0x0080 /* select by name */
+#define SELPGID 0x0100 /* select process group IDs (-g) */
+#define SELPID 0x0200 /* select PIDs (-p) */
+#define SELUID 0x0400 /* select UIDs (-u) */
+#define SELUNX 0x0800 /* select UNIX socket (-U) */
+#define SELZONE 0x1000 /* select zone (-z) */
+#define SELEXCLF 0x2000 /* file selection excluded */
+#define SELALL (SELCMD|SELCNTX|SELFD|SELNA|SELNET|SELNM|SELNFS|SELPID|SELUID|SELUNX|SELZONE)
+#define SELPROC (SELCMD|SELCNTX|SELPGID|SELPID|SELUID|SELZONE)
+ /* process selecters */
+#define SELFILE (SELFD|SELNFS|SELNLINK|SELNM) /* file selecters */
+#define SELNW (SELNA|SELNET|SELUNX) /* network selecters */
+
+/*
+ * Structure definitions
+ */
+
+# if defined(HAS_AFS)
+struct afsnode { /* AFS pseudo-node structure */
+ dev_t dev;
+ unsigned char ino_st; /* 1 if inode has a value */
+ unsigned char nlink_st; /* 1 if nlink has a value */
+ INODETYPE inode;
+ unsigned long size;
+ long nlink;
+};
+# endif /* defined(HAS_AFS) */
+
+# if defined(HAS_STD_CLONE)
+struct clone {
+ int dx; /* index of device entry in Devtp[] */
+ struct clone *next; /* forward link */
+};
+extern struct clone *Clone;
+# endif /* defined(HAS_STD_CLONE) */
+
+# if defined(HASNLIST)
+struct drive_Nl { /* data to drive build_Nl() */
+ char *nn; /* nickname for lookups */
+ char *knm; /* kernel variable for name list */
+};
+extern struct drive_Nl Drive_Nl[]; /* defined in dstore.c */
+# endif /* defined(HASNLIST) */
+
+
+/*
+ * Global storage definitions (including their structure definitions)
+ */
+
+struct int_lst {
+ int i; /* integer argument */
+ int f; /* find state -- meaningful only if
+ * x == 0 */
+ int x; /* excluded state */
+};
+
+typedef struct lsof_rx { /* regular expression table entry */
+ char *exp; /* original regular expression */
+ regex_t cx; /* compiled expression */
+ int mc; /* match count */
+} lsof_rx_t;
+extern lsof_rx_t *CmdRx;
+extern int NCmdRxU;
+
+# if defined(HASFSTRUCT)
+struct pff_tab { /* print file flags table structure */
+ long val; /* flag value */
+ char *nm; /* name to print for flag */
+};
+# endif /* defined(HASFSTRUCT) */
+
+struct seluid {
+ uid_t uid; /* User ID */
+ char *lnm; /* specified login name (NULL = none) */
+ unsigned char excl; /* excluded state */
+ unsigned char f; /* selected User ID find state
+ * (meaningful only if excl == 0) */
+};
+
+# if defined(HASBLKDEV)
+extern struct l_dev *BDevtp, **BSdev;
+extern int BNdev;
+# endif /* defined(HASBLKDEV) */
+
+extern int CkPasswd;
+
+struct str_lst {
+ char *str; /* string */
+ int len; /* string length */
+ short f; /* selected string find state */
+ short x; /* exclusion (if non-zero) */
+ struct str_lst *next; /* next list entry */
+};
+extern struct str_lst *Cmdl;
+extern int CmdLim;
+extern int Cmdni;
+extern int Cmdnx;
+
+# if defined(HASSELINUX)
+typedef struct cntxlist {
+ char *cntx; /* zone name */
+ int f; /* "find" flag (used only in CntxArg) */
+ struct cntxlist *next; /* next zone hash entry */
+} cntxlist_t;
+extern cntxlist_t *CntxArg;
+extern int CntxStatus;
+# endif /* defined(HASSELINUX) */
+
+# if defined(HASDCACHE)
+extern unsigned DCcksum;
+extern int DCfd;
+extern FILE *DCfs;
+extern char *DCpathArg;
+extern char *DCpath[];
+extern int DCpathX;
+extern int DCrebuilt;
+extern int DCstate;
+extern int DCunsafe;
+# endif /* defined(HASDCACHE) */
+
+extern int DChelp;
+extern dev_t DevDev;
+extern struct l_dev *Devtp;
+extern char **Dstk;
+extern int Dstkn;
+extern int Dstkx;
+extern int ErrStat;
+extern uid_t Euid;
+extern int Fand;
+extern int Fblock;
+extern int Fcntx;
+extern int Ffield;
+extern int Ffilesys;
+extern int Fhelp;
+extern int Fhost;
+
+# if defined(HASNCACHE)
+extern int Fncache;
+extern int NcacheReload;
+# endif /* defined(HASNCACHE) */
+
+extern int Fnet;
+extern int FnetTy;
+extern int Fnfs;
+extern int Fnlink;
+extern int Foffset;
+extern int Fovhd;
+extern int Fport;
+extern int FportMap;
+extern int Fpgid;
+extern int Fppid;
+extern int Fsize;
+extern int Fsv;
+extern int FsvByf;
+extern int FsvFlagX;
+extern int Ftcptpi;
+extern int Fterse;
+extern int Funix;
+extern int Futol;
+extern int Fverbose;
+extern int Fwarn;
+
+# if defined(HASXOPT_VALUE)
+extern int Fxopt;
+# endif /* defined(HASXOPT_VALUE) */
+
+extern int Fxover;
+extern int Fzone;
+
+struct fd_lst {
+ char *nm; /* file descriptor name -- range if
+ * NULL */
+ int lo; /* range start (if nm NULL) */
+ int hi; /* range end (if nm NULL) */
+ struct fd_lst *next;
+};
+extern struct fd_lst *Fdl;
+extern int FdlTy; /* Fdl[] type: -1 == none
+ * 0 == include
+ * 1 == exclude */
+
+struct fieldsel {
+ char id; /* field ID character */
+ unsigned char st; /* field status */
+ char *nm; /* field name */
+ int *opt; /* option variable address */
+ int ov; /* value to OR with option variable */
+};
+extern struct fieldsel FieldSel[];
+
+extern int Hdr;
+
+enum IDType {PGID, PID};
+extern char *InodeFmt_d;
+extern char *InodeFmt_x;
+
+struct lfile {
+ char access;
+ char lock;
+ unsigned char dev_def; /* device number definition status */
+ unsigned char inp_ty; /* inode/iproto type
+ * 0: neither inode nor iproto
+ * 1: print inode in decimal
+ * 2: iproto contains string
+ * 3: print inode in hex
+ */
+ unsigned char is_com; /* common stream status */
+ unsigned char is_nfs; /* NFS file status */
+ unsigned char is_stream; /* stream device status */
+
+# if defined(HASVXFS) && defined(HASVXFSDNLC)
+ unsigned char is_vxfs; /* VxFS file status */
+# endif /* defined(HASVXFS) && defined(HASVXFSDNLC) */
+
+ unsigned char lmi_srch; /* local mount info search status:
+ * 1 = printname() search required */
+
+# if defined(HASMNTSTAT)
+ unsigned char mnt_stat; /* mount point stat(2) status */
+# endif /* defined(HASMNTSTAT) */
+
+ unsigned char nlink_def; /* link count definition status */
+ unsigned char off_def; /* offset definition status */
+ unsigned char rdev_def; /* rdev definition status */
+ unsigned char sz_def; /* size definition status */
+
+# if defined(HASFSTRUCT)
+ unsigned char fsv; /* file struct value status */
+# endif /* defined(HASFSTRUCT) */
+
+ char fd[FDLEN];
+ char iproto[IPROTOL];
+ char type[TYPEL];
+ short sf; /* select flags -- SEL* symbols */
+ int ch; /* VMPC channel: -1 = none */
+ int ntype; /* node type -- N_* value */
+ SZOFFTYPE off;
+ SZOFFTYPE sz;
+ dev_t dev;
+ dev_t rdev;
+ INODETYPE inode;
+ long nlink; /* link count */
+ char *dev_ch;
+ char *fsdir; /* file system directory */
+ char *fsdev; /* file system device */
+
+# if defined(HASFSINO)
+ INODETYPE fs_ino; /* file system inode number */
+# endif /* defined HASFSINO) */
+
+ struct linaddr { /* local Internet address information */
+ int af; /* address family: 0 for none; AF_INET;
+ * or AF_INET6 */
+ int p; /* port */
+ union {
+ struct in_addr a4; /* AF_INET Internet address */
+
+# if defined(HASIPv6)
+ struct in6_addr a6; /* AF_INET6 Internet address */
+# endif /* defined(HASIPv6) */
+
+ } ia;
+ } li[2]; /* li[0]: local
+ * li[1]: foreign */
+ struct ltstate { /* local TCP/TPI state */
+ int type; /* state type:
+ * -1 == none
+ * 0 == TCP
+ * 1 == TPI or socket (SS_*) */
+ union {
+ int i; /* integer state */
+ unsigned int ui; /* unsigned integer state */
+ } state;
+
+# if defined(HASSOOPT)
+ unsigned char pqlens; /* pqlen status: 0 = none */
+ unsigned char qlens; /* qlen status: 0 = none */
+ unsigned char qlims; /* qlim status: 0 = none */
+ unsigned char rbszs; /* rbsz status: 0 = none */
+ unsigned char sbszs; /* sbsz status: 0 = none */
+ int kai; /* TCP keep-alive interval */
+ int ltm; /* TCP linger time */
+ unsigned int opt; /* socket options */
+ unsigned int pqlen; /* partial connection queue length */
+ unsigned int qlen; /* connection queue length */
+ unsigned int qlim; /* connection queue limit */
+ unsigned long rbsz; /* receive buffer size */
+ unsigned long sbsz; /* send buffer size */
+# endif /* defined(HASSOOPT) */
+
+# if defined(HASSOSTATE)
+ unsigned int ss; /* socket state */
+# if defined(HASSBSTATE)
+ unsigned int sbs_rcv; /* receive socket buffer state */
+ unsigned int sbs_snd; /* send socket buffer state */
+# endif /* defined(HASSBSTATE) */
+# endif /* defined(HASSOSTATE) */
+
+# if defined(HASTCPOPT)
+ unsigned int topt; /* TCP options */
+ unsigned char msss; /* mss status: 0 = none */
+ unsigned long mss; /* TCP maximum segment size */
+# endif /* defined(HASTCPOPT) */
+
+# if defined(HASTCPTPIQ)
+ unsigned long rq; /* receive queue length */
+ unsigned long sq; /* send queue length */
+ unsigned char rqs; /* rq status: 0 = none */
+ unsigned char sqs; /* sq status: 0 = none */
+# endif /* defined(HASTCPTPIQ) */
+
+# if defined(HASTCPTPIW)
+ unsigned char rws; /* rw status: 0 = none */
+ unsigned char wws; /* ww status: 0 = none */
+ unsigned long rw; /* read window size */
+ unsigned long ww; /* write window size */
+# endif /* defined(HASTCPTPIW) */
+
+ } lts;
+ char *nm;
+ char *nma; /* NAME column addition */
+
+# if defined(HASNCACHE) && HASNCACHE<2
+ KA_T na; /* file structure's node address */
+# endif /* defined(HASNCACHE) && HASNCACHE<2 */
+
+# if defined(HASNCACHE) && defined(HASNCVPID)
+ unsigned long id; /* capability ID */
+# endif /* defined(HASNCACHE) && defined(HASNCVPID) */
+
+# if defined(HASLFILEADD)
+ HASLFILEADD
+# endif /* defined(HASLFILEADD) */
+
+# if defined(HASFSTRUCT)
+ KA_T fsa; /* file structure address */
+ long fct; /* file structure's f_count */
+ long ffg; /* file structure's f_flag */
+ long pof; /* process open-file flags */
+ KA_T fna; /* file structure node address */
+# endif /* defined(HASFSTRUCT) */
+
+ struct lfile *next;
+};
+extern struct lfile *Lf, *Plf;
+
+
+struct lproc {
+ char *cmd; /* command name */
+
+# if defined(HASSELINUX)
+ char *cntx; /* security context */
+# endif /* defined(HASSELINUX) */
+
+ short sf; /* select flags -- SEL* symbols */
+ short pss; /* state: 0 = not selected
+ * 1 = wholly selected
+ * 2 = partially selected */
+ int pid; /* process ID */
+ int pgid; /* process group ID */
+ int ppid; /* parent process ID */
+ uid_t uid; /* user ID */
+
+# if defined(HASZONES)
+ char *zn; /* zone name */
+# endif /* defined(HASZONES) */
+
+ struct lfile *file; /* open files of process */
+};
+extern struct lproc *Lp, *Lproc;
+
+extern char *Memory;
+extern int MntSup;
+extern char *MntSupP;
+
+# if defined(HASPROCFS)
+extern struct mounts *Mtprocfs;
+# endif
+
+extern int Mxpgid;
+extern int Mxpid;
+extern int Mxuid;
+extern gid_t Mygid;
+extern int Mypid;
+extern uid_t Myuid;
+extern char *Namech;
+extern size_t Namechl;
+extern int Ndev;
+
+# if defined(HASNLIST)
+# if !defined(NLIST_TYPE)
+#define NLIST_TYPE nlist
+# endif /* !defined(NLIST_TYPE) */
+extern struct NLIST_TYPE *Nl;
+extern int Nll;
+# endif /* defined(HASNLIST) */
+extern long Nlink;
+extern int Nlproc;
+extern char *Nmlst;
+extern int Npgid;
+extern int Npgidi;
+extern int Npgidx;
+extern int Npid;
+extern int Npidi;
+extern int Npidx;
+extern int Npuns;
+extern int Ntype;
+extern int Nuid;
+extern int Nuidexcl;
+extern int Nuidincl;
+
+struct nwad {
+ char *arg; /* argument */
+ char *proto; /* protocol */
+ int af; /* address family -- e.g.,
+ * AF_INET, AF_INET6 */
+ unsigned char a[MAX_AF_ADDR]; /* address */
+ int sport; /* starting port */
+ int eport; /* ending port */
+ int f; /* find state */
+ struct nwad *next; /* forward link */
+};
+extern struct nwad *Nwad;
+
+extern int OffDecDig;
+extern char *Pn;
+
+# if defined(HASFSTRUCT)
+extern struct pff_tab Pff_tab[]; /* file flags table */
+extern struct pff_tab Pof_tab[]; /* process open file flags table */
+# endif /* defined(HASFSTRUCT) */
+
+# if defined(HASPROCFS)
+struct procfsid {
+ pid_t pid; /* search PID */
+ char *nm; /* search name */
+ unsigned char f; /* match found if == 1 */
+
+# if defined(HASPINODEN)
+ INODETYPE inode; /* search inode number */
+# endif /* defined(HASPINODEN) */
+
+ struct procfsid *next; /* forward link */
+};
+
+extern int Procfind;
+extern struct procfsid *Procfsid;
+extern int Procsrch;
+# endif /* defined(HASPROCFS) */
+
+extern int PrPass;
+extern int RptTm;
+extern struct l_dev **Sdev;
+extern int Selall;
+extern int Selflags;
+extern int Setgid;
+extern int Selinet;
+extern int Setuidroot;
+extern struct sfile *Sfile;
+extern struct int_lst *Spgid;
+extern struct int_lst *Spid;
+extern struct seluid *Suid;
+extern char *SzOffFmt_0t;
+extern char *SzOffFmt_d;
+extern char *SzOffFmt_dv;
+extern char *SzOffFmt_x;
+extern int TcpStAlloc;
+extern unsigned char *TcpStI;
+extern int TcpStIn;
+extern int TcpStOff;
+extern unsigned char *TcpStX;
+extern int TcpStXn;
+extern int TcpNstates;
+extern char **TcpSt;
+extern char Terminator;
+extern int TmLimit;
+extern int UdpStAlloc;
+extern unsigned char *UdpStI;
+extern int UdpStIn;
+extern int UdpStOff;
+extern unsigned char *UdpStX;
+extern int UdpStXn;
+extern int UdpNstates;
+extern char **UdpSt;
+
+# if defined(HASZONES)
+typedef struct znhash {
+ char *zn; /* zone name */
+ int f; /* "find" flag (used only in ZoneArg) */
+ struct znhash *next; /* next zone hash entry */
+} znhash_t;
+extern znhash_t **ZoneArg;
+# endif /* defined(HASZONES) */
+
+#include "proto.h"
+#include "dproto.h"
+
+#endif /* LSOF_H */
--- /dev/null
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+NAME
+ lsof - list open files
+
+SYNOPSIS
+ lsof [ -?abChlnNOPRtUvVX ] [ -A A ] [ -c c ] [ +c c ] [ +|-d
+ d ] [ +|-D D ] [ +|-f [cfgGn] ] [ -F [f] ] [ -g [s] ] [ -i
+ [i] ] [ -k k ] [ +|-L [l] ] [ +|-m m ] [ +|-M ] [ -o [o] ] [
+ -p s ] [ +|-r [t[m<fmt>]] ] [ -s [p:s] ] [ -S [t] ] [ -T [t]
+ ] [ -u s ] [ +|-w ] [ -x [fl] ] [ -z [z] ] [ -Z [Z] ] [ -- ]
+ [names]
+
+DESCRIPTION
+ Lsof revision 4.82 lists on its standard output file infor-
+ mation about files opened by processes for the following
+ UNIX dialects:
+
+ AIX 5.3
+ Apple Darwin 9 (Mac OS X 10.5)
+ FreeBSD 4.9 for x86-based systems
+ FreeBSD 7.[012] and 8.0 for AMD64-based systems
+ Linux 2.1.72 and above for x86-based systems
+ Solaris 9 and 10
+
+ (See the DISTRIBUTION section of this manual page for infor-
+ mation on how to obtain the latest lsof revision.)
+
+ An open file may be a regular file, a directory, a block
+ special file, a character special file, an executing text
+ reference, a library, a stream or a network file (Internet
+ socket, NFS file or UNIX domain socket.) A specific file or
+ all the files in a file system may be selected by path.
+
+ Instead of a formatted display, lsof will produce output
+ that can be parsed by other programs. See the -F, option
+ description, and the OUTPUT FOR OTHER PROGRAMS section for
+ more information.
+
+ In addition to producing a single output list, lsof will run
+ in repeat mode. In repeat mode it will produce output,
+ delay, then repeat the output operation until stopped with
+ an interrupt or quit signal. See the +|-r [t[m<fmt>]]
+ option description for more information.
+
+OPTIONS
+ In the absence of any options, lsof lists all open files
+ belonging to all active processes.
+
+ If any list request option is specified, other list requests
+ must be specifically requested - e.g., if -U is specified
+ for the listing of UNIX socket files, NFS files won't be
+ listed unless -N is also specified; or if a user list is
+ specified with the -u option, UNIX domain socket files,
+
+
+
+SunOS 5.9 Last change: Revision-4.82 1
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ belonging to users not in the list, won't be listed unless
+ the -U option is also specified.
+
+ Normally list options that are specifically stated are ORed
+ - i.e., specifying the -i option without an address and the
+ -ufoo option produces a listing of all network files OR
+ files belonging to processes owned by user ``foo''. The
+ exceptions are:
+
+ 1) the `^' (negated) login name or user ID (UID), specified
+ with the -u option;
+
+ 2) the `^' (negated) process ID (PID), specified with the -p
+ option;
+
+ 3) the `^' (negated) process group ID (PGID), specified with
+ the -g option;
+
+ 4) the `^' (negated) command, specified with the -c option;
+
+ 5) the (`^') negated TCP or UDP protocol state names, speci-
+ fied with the -s [p:s] option.
+
+ Since they represent exclusions, they are applied without
+ ORing or ANDing and take effect before any other selection
+ criteria are applied.
+
+ The -a option may be used to AND the selections. For exam-
+ ple, specifying -a, -U, and -ufoo produces a listing of only
+ UNIX socket files that belong to processes owned by user
+ ``foo''.
+
+ Caution: the -a option causes all list selection options to
+ be ANDed; it can't be used to cause ANDing of selected pairs
+ of selection options by placing it between them, even though
+ its placement there is acceptable. Wherever -a is placed,
+ it causes the ANDing of all selection options.
+
+ Items of the same selection set - command names, file
+ descriptors, network addresses, process identifiers, user
+ identifiers, zone names, security contexts - are joined in a
+ single ORed set and applied before the result participates
+ in ANDing. Thus, for example, specifying -i@aaa.bbb,
+ -i@ccc.ddd, -a, and -ufff,ggg will select the listing of
+ files that belong to either login ``fff'' OR ``ggg'' AND
+ have network connections to either host aaa.bbb OR ccc.ddd.
+
+ Options may be grouped together following a single prefix --
+ e.g., the option set ``-a -b -C'' may be stated as -abC.
+ However, since values are optional following +|-f, -F, -g,
+ -i, +|-L, -o, +|-r, -s, -S, -T, -x and -z. when you have no
+ values for them be careful that the following character
+
+
+
+SunOS 5.9 Last change: Revision-4.82 2
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ isn't ambiguous. For example, -Fn might represent the -F
+ and -n options, or it might represent the n field identifier
+ character following the -F option. When ambiguity is possi-
+ ble, start a new option with a `-' character - e.g., ``-F
+ -n''. If the next option is a file name, follow the possi-
+ bly ambiguous option with ``--'' - e.g., ``-F -- name''.
+
+ Either the `+' or the `-' prefix may be applied to a group
+ of options. Options that don't take on separate meanings
+ for each prefix - e.g., -i - may be grouped under either
+ prefix. Thus, for example, ``+M -i'' may be stated as
+ ``+Mi'' and the group means the same as the separate
+ options. Be careful of prefix grouping when one or more
+ options in the group does take on separate meanings under
+ different prefixes - e.g., +|-M; ``-iM'' is not the same
+ request as ``-i +M''. When in doubt, use separate options
+ with appropriate prefixes.
+
+ -? -h These two equivalent options select a usage (help)
+ output list. Lsof displays a shortened form of
+ this output when it detects an error in the options
+ supplied to it, after it has displayed messages
+ explaining each error. (Escape the `?' character
+ as your shell requires.)
+
+ -a This option causes list selection options to be
+ ANDed, as described above.
+
+ -A A This option is available on systems configured for
+ AFS whose AFS kernel code is implemented via
+ dynamic modules. It allows the lsof user to
+ specify A as an alternate name list file where the
+ kernel addresses of the dynamic modules might be
+ found. See the lsof FAQ (The FAQ section gives its
+ location.) for more information about dynamic
+ modules, their symbols, and how they affect lsof.
+
+ -b This option causes lsof to avoid kernel functions
+ that might block - lstat(2), readlink(2), and
+ stat(2).
+
+ See the BLOCKS AND TIMEOUTS and AVOIDING KERNEL
+ BLOCKS sections for information on using this
+ option.
+
+ -c c This option selects the listing of files for
+ processes executing the command that begins with
+ the characters of c. Multiple commands may be
+ specified, using multiple -c options. They are
+ joined in a single ORed set before participating in
+ AND option selection.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 3
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ If c begins with a `^', then the following charac-
+ ters specify a command name whose processes are to
+ be ignored (excluded.)
+
+ If c begins and ends with a slash ('/'), the char-
+ acters between the slashes are interpreted as a
+ regular expression. Shell meta-characters in the
+ regular expression must be quoted to prevent their
+ interpretation by the shell. The closing slash may
+ be followed by these modifiers:
+
+
+ b the regular expression is a basic one.
+ i ignore the case of letters.
+ x the regular expression is an extended one
+ (default).
+
+ See the lsof FAQ (The FAQ section gives its loca-
+ tion.) for more information on basic and extended
+ regular expressions.
+
+ The simple command specification is tested first.
+ If that test fails, the command regular expression
+ is applied. If the simple command test succeeds,
+ the command regular expression test isn't made.
+ This may result in ``no command found for regex:''
+ messages when lsof's -V option is specified.
+
+ +c w This option defines the maximum number of initial
+ characters of the name, supplied by the UNIX
+ dialect, of the UNIX command associated with a pro-
+ cess to be printed in the COMMAND column. (The
+ lsof default is nine.)
+
+ Note that many UNIX dialects do not supply all com-
+ mand name characters to lsof in the files and
+ structures from which lsof obtains command name.
+ Often dialects limit the number of characters sup-
+ plied in those sources. For example, Linux 2.4.27
+ and Solaris 9 both limit command name length to 16
+ characters.
+
+ If w is zero ('0'), all command characters supplied
+ to lsof by the UNIX dialect will be printed.
+
+ If w is less than the length of the column title,
+ ``COMMAND'', it will be raised to that length.
+
+ -C This option disables the reporting of any path name
+ components from the kernel's name cache. See the
+ KERNEL NAME CACHE section for more information.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 4
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ +d s This option causes lsof to search for all open
+ instances of directory s and the files and direc-
+ tories it contains at its top level. This option
+ does NOT descend the directory tree, rooted at s.
+ The +D D option may be used to request a
+ full-descent directory tree search, rooted at
+ directory D.
+
+ Processing of the +d option does not follow sym-
+ bolic links within s unless the -x or -x l option
+ is also specified. Nor does it search for open
+ files on file system mount points on subdirectories
+ of s unless the -x or -x f option is also speci-
+ fied.
+
+ Note: the authority of the user of this option lim-
+ its it to searching for files that the user has
+ permission to examine with the system stat(2) func-
+ tion.
+
+ -d s This option specifies a list of file descriptors
+ (FDs) to exclude from or include in the output
+ listing. The file descriptors are specified in the
+ comma-separated set s - e.g., ``cwd,1,3'',
+ ``^6,^2''. (There should be no spaces in the set.)
+
+ The list is an exclusion list if all entries of the
+ set begin with `^'. It is an inclusion list if no
+ entry begins with `^'. Mixed lists are not permit-
+ ted.
+
+ A file descriptor number range may be in the set as
+ long as neither member is empty, both members are
+ numbers, and the ending member is larger than the
+ starting one - e.g., ``0-7'' or ``3-10''. Ranges
+ may be specified for exclusion if they have the `^'
+ prefix - e.g., ``^0-7'' excludes all file descrip-
+ tors 0 through 7.
+
+ Multiple file descriptor numbers are joined in a
+ single ORed set before participating in AND option
+ selection.
+
+ When there are exclusion and inclusion members in
+ the set, lsof reports them as errors and exits with
+ a non-zero return code.
+
+ See the description of File Descriptor (FD) output
+ values in the OUTPUT section for more information
+ on file descriptor names.
+
+ +D D This option causes lsof to search for all open
+
+
+
+SunOS 5.9 Last change: Revision-4.82 5
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ instances of directory D and all the files and
+ directories it contains to its complete depth.
+
+ Processing of the +D option does not follow sym-
+ bolic links within D unless the -x or -x l option
+ is also specified. Nor does it search for open
+ files on file system mount points on subdirectories
+ of D unless the -x or -x f option is also speci-
+ fied.
+
+ Note: the authority of the user of this option lim-
+ its it to searching for files that the user has
+ permission to examine with the system stat(2) func-
+ tion.
+
+ Further note: lsof may process this option slowly
+ and require a large amount of dynamic memory to do
+ it. This is because it must descend the entire
+ directory tree, rooted at D, calling stat(2) for
+ each file and directory, building a list of all the
+ files it finds, and searching that list for a match
+ with every open file. When directory D is large,
+ these steps can take a long time, so use this
+ option prudently.
+
+ -D D This option directs lsof's use of the device cache
+ file. The use of this option is sometimes res-
+ tricted. See the DEVICE CACHE FILE section and the
+ sections that follow it for more information on
+ this option.
+
+ -D must be followed by a function letter; the func-
+ tion letter may optionally be followed by a path
+ name. Lsof recognizes these function letters:
+
+
+ ? - report device cache file paths
+ b - build the device cache file
+ i - ignore the device cache file
+ r - read the device cache file
+ u - read and update the device cache file
+
+ The b, r, and u functions, accompanied by a path
+ name, are sometimes restricted. When these func-
+ tions are restricted, they will not appear in the
+ description of the -D option that accompanies -h or
+ -? option output. See the DEVICE CACHE FILE sec-
+ tion and the sections that follow it for more
+ information on these functions and when they're
+ restricted.
+
+ The ? function reports the read-only and write
+
+
+
+SunOS 5.9 Last change: Revision-4.82 6
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ paths that lsof can use for the device cache file,
+ the names of any environment variables whose values
+ lsof will examine when forming the device cache
+ file path, and the format for the personal device
+ cache file path. (Escape the `?' character as your
+ shell requires.)
+
+ When available, the b, r, and u functions may be
+ followed by the device cache file's path. The
+ standard default is .lsof_hostname in the home
+ directory of the real user ID that executes lsof,
+ but this could have been changed when lsof was con-
+ figured and compiled. (The output of the -h and -?
+ options show the current default prefix - e.g.,
+ ``.lsof''.) The suffix, hostname, is the first
+ component of the host's name returned by gethost-
+ name(2).
+
+ When available, the b function directs lsof to
+ build a new device cache file at the default or
+ specified path.
+
+ The i function directs lsof to ignore the default
+ device cache file and obtain its information about
+ devices via direct calls to the kernel.
+
+ The r function directs lsof to read the device
+ cache at the default or specified path, but
+ prevents it from creating a new device cache file
+ when none exists or the existing one is improperly
+ structured. The r function, when specified without
+ a path name, prevents lsof from updating an
+ incorrect or outdated device cache file, or creat-
+ ing a new one in its place. The r function is
+ always available when it is specified without a
+ path name argument; it may be restricted by the
+ permissions of the lsof process.
+
+ When available, the u function directs lsof to read
+ the device cache file at the default or specified
+ path, if possible, and to rebuild it, if necessary.
+ This is the default device cache file function when
+ no -D option has been specified.
+
+ +|-f [cfgGn]
+ f by itself clarifies how path name arguments are
+ to be interpreted. When followed by c, f, g, G, or
+ n in any combination it specifies that the listing
+ of kernel file structure information is to be
+ enabled (`+') or inhibited (`-').
+
+ Normally a path name argument is taken to be a file
+
+
+
+SunOS 5.9 Last change: Revision-4.82 7
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ system name if it matches a mounted-on directory
+ name reported by mount(8), or if it represents a
+ block device, named in the mount output and associ-
+ ated with a mounted directory name. When +f is
+ specified, all path name arguments will be taken to
+ be file system names, and lsof will complain if any
+ are not. This can be useful, for example, when the
+ file system name (mounted-on device) isn't a block
+ device. This happens for some CD-ROM file systems.
+
+ When -f is specified by itself, all path name argu-
+ ments will be taken to be simple files. Thus, for
+ example, the ``-f -- /'' arguments direct lsof to
+ search for open files with a `/' path name, not all
+ open files in the `/' (root) file system.
+
+ Be careful to make sure +f and -f are properly ter-
+ minated and aren't followed by a character (e.g.,
+ of the file or file system name) that might be
+ taken as a parameter. For example, use ``--''
+ after +f and -f as in these examples.
+
+
+ $ lsof +f -- /file/system/name
+ $ lsof -f -- /file/name
+
+ The listing of information from kernel file struc-
+ tures, requested with the +f [cfgGn] option form,
+ is normally inhibited, and is not available in
+ whole or part for some dialects - e.g., /proc-based
+ Linux kernels below 2.6.22. When the prefix to f
+ is a plus sign (`+'), these characters request file
+ structure information:
+
+
+ c file structure use count (not Linux)
+ f file structure address (not Linux)
+ g file flag abbreviations (Linux 2.6.22 and up)
+ G file flags in hexadecimal (Linux 2.6.22 and up)
+ n file structure node address (not Linux)
+
+ When the prefix is minus (`-') the same characters
+ disable the listing of the indicated values.
+
+ File structure addresses, use counts, flags, and
+ node addresses may be used to detect more readily
+ identical files inherited by child processes and
+ identical files in use by different processes.
+ Lsof column output can be sorted by output columns
+ holding the values and listed to identify identical
+ file use, or lsof field output can be parsed by an
+ AWK or Perl post-filter script, or by a C program.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 8
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ -F f This option specifies a character list, f, that
+ selects the fields to be output for processing by
+ another program, and the character that terminates
+ each output field. Each field to be output is
+ specified with a single character in f. The field
+ terminator defaults to NL, but may be changed to
+ NUL (000). See the OUTPUT FOR OTHER PROGRAMS sec-
+ tion for a description of the field identification
+ characters and the field output process.
+
+ When the field selection character list is empty,
+ all standard fields are selected (except the raw
+ device field, security context and zone field for
+ compatibility reasons) and the NL field terminator
+ is used.
+
+ When the field selection character list contains
+ only a zero (`0'), all fields are selected (except
+ the raw device field for compatibility reasons) and
+ the NUL terminator character is used.
+
+ Other combinations of fields and their associated
+ field terminator character must be set with expli-
+ cit entries in f, as described in the OUTPUT FOR
+ OTHER PROGRAMS section.
+
+ When a field selection character identifies an item
+ lsof does not normally list - e.g., PPID, selected
+ with -R - specification of the field character -
+ e.g., ``-FR'' - also selects the listing of the
+ item.
+
+ When the field selection character list contains
+ the single character `?', lsof will display a help
+ list of the field identification characters.
+ (Escape the `?' character as your shell requires.)
+
+ -g [s] This option excludes or selects the listing of
+ files for the processes whose optional process
+ group IDentification (PGID) numbers are in the
+ comma-separated set s - e.g., ``123'' or
+ ``123,^456''. (There should be no spaces in the
+ set.)
+
+ PGID numbers that begin with `^' (negation)
+ represent exclusions.
+
+ Multiple PGID numbers are joined in a single ORed
+ set before participating in AND option selection.
+ However, PGID exclusions are applied without ORing
+ or ANDing and take effect before other selection
+ criteria are applied.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 9
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ The -g option also enables the output display of
+ PGID numbers. When specified without a PGID set
+ that's all it does.
+
+ -i [i] This option selects the listing of files any of
+ whose Internet address matches the address speci-
+ fied in i. If no address is specified, this option
+ selects the listing of all Internet and x.25
+ (HP-UX) network files.
+
+ If -i4 or -i6 is specified with no following
+ address, only files of the indicated IP version,
+ IPv4 or IPv6, are displayed. (An IPv6 specifica-
+ tion may be used only if the dialects supports
+ IPv6, as indicated by ``[46]'' and ``IPv[46]'' in
+ lsof's -h or -? output.) Sequentially specifying
+ -i4, followed by -i6 is the same as specifying -i,
+ and vice-versa. Specifying -i4, or -i6 after -i is
+ the same as specifying -i4 or -i6 by itself.
+
+ Multiple addresses (up to a limit of 100) may be
+ specified with multiple -i options. (A port number
+ or service name range is counted as one address.)
+ They are joined in a single ORed set before parti-
+ cipating in AND option selection.
+
+ An Internet address is specified in the form (Items
+ in square brackets are optional.):
+
+ [46][protocol][@hostname|hostaddr][:service|port]
+
+ where:
+ 46 specifies the IP version, IPv4 or IPv6
+ that applies to the following address.
+ '6' may be be specified only if the UNIX
+ dialect supports IPv6. If neither '4' nor
+ '6' is specified, the following address
+ applies to all IP versions.
+ protocol is a protocol name - TCP, UDP
+ hostname is an Internet host name. Unless a
+ specific IP version is specified, open
+ network files associated with host names
+ of all versions will be selected.
+ hostaddr is a numeric Internet IPv4 address in
+ dot form; or an IPv6 numeric address in
+ colon form, enclosed in brackets, if the
+ UNIX dialect supports IPv6. When an IP
+ version is selected, only its numeric
+ addresses may be specified.
+ service is an /etc/services name - e.g., smtp -
+ or a list of them.
+ port is a port number, or a list of them.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 10
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ IPv6 options may be used only if the UNIX dialect
+ supports IPv6. To see if the dialect supports
+ IPv6, run lsof and specify the -h or -? (help)
+ option. If the displayed description of the -i
+ option contains ``[46]'' and ``IPv[46]'', IPv6 is
+ supported.
+
+ IPv4 host names and addresses may not be specified
+ if network file selection is limited to IPv6 with
+ -i 6. IPv6 host names and addresses may not be
+ specified if network file selection is limited to
+ IPv4 with -i 4. When an open IPv4 network file's
+ address is mapped in an IPv6 address, the open
+ file's type will be IPv6, not IPv4, and its display
+ will be selected by '6', not '4'.
+
+ At least one address component - 4, 6, protocol,
+ ,IR hostname , hostaddr, or service - must be sup-
+ plied. The `@' character, leading the host specif-
+ ication, is always required; as is the `:', leading
+ the port specification. Specify either hostname or
+ hostaddr. Specify either service name list or port
+ number list. If a service name list is specified,
+ the protocol may also need to be specified if the
+ TCP, UDP and UDPLITE port numbers for the service
+ name are different. Use any case - lower or upper
+ - for protocol.
+
+ Service names and port numbers may be combined in a
+ list whose entries are separated by commas and
+ whose numeric range entries are separated by minus
+ signs. There may be no embedded spaces, and all
+ service names must belong to the specified proto-
+ col. Since service names may contain embedded
+ minus signs, the starting entry of a range can't be
+ a service name; it can be a port number, however.
+
+ Here are some sample addresses:
+
+ -i6 - IPv6 only
+ TCP:25 - TCP and port 25
+ @1.2.3.4 - Internet IPv4 host address 1.2.3.4
+ @[3ffe:1ebc::1]:1234 - Internet IPv6 host address
+ 3ffe:1ebc::1, port 1234
+ UDP:who - UDP who service port
+ TCP@lsof.itap:513 - TCP, port 513 and host name lsof.itap
+ tcp@foo:1-10,smtp,99 - TCP, ports 1 through 10,
+ service name smtp, port 99, host name foo
+ tcp@bar:1-smtp - TCP, ports 1 through smtp, host bar
+ :time - either TCP, UDP or UDPLITE time service port
+
+ -k k This option specifies a kernel name list file, k,
+
+
+
+SunOS 5.9 Last change: Revision-4.82 11
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ in place of /vmunix, /mach, etc. This option is
+ not available under AIX on the IBM RISC/System
+ 6000.
+
+ -l This option inhibits the conversion of user ID
+ numbers to login names. It is also useful when
+ login name lookup is working improperly or slowly.
+
+ +|-L [l] This option enables (`+') or disables (`-') the
+ listing of file link counts, where they are avail-
+ able - e.g., they aren't available for sockets, or
+ most FIFOs and pipes.
+
+ When +L is specified without a following number,
+ all link counts will be listed. When -L is speci-
+ fied (the default), no link counts will be listed.
+
+ When +L is followed by a number, only files having
+ a link count less than that number will be listed.
+ (No number may follow -L.) A specification of the
+ form ``+L1'' will select open files that have been
+ unlinked. A specification of the form
+ ``+aL1 <file_system>'' will select unlinked open
+ files on the specified file system.
+
+ For other link count comparisons, use field output
+ (-F) and a post-processing script or program.
+
+ +|-m m This option specifies an alternate kernel memory
+ file or activates mount table supplement process-
+ ing.
+
+ The option form -m m specifies a kernel memory
+ file, m, in place of /dev/kmem or /dev/mem - e.g.,
+ a crash dump file.
+
+ The option form +m requests that a mount supplement
+ file be written to the standard output file. All
+ other options are silently ignored.
+
+ There will be a line in the mount supplement file
+ for each mounted file system, containing the
+ mounted file system directory, followed by a single
+ space, followed by the device number in hexadecimal
+ "0x" format - e.g.,
+
+
+ / 0x801
+
+ Lsof can use the mount supplement file to get dev-
+ ice numbers for file systems when it can't get them
+ via stat(2) or lstat(2).
+
+
+
+SunOS 5.9 Last change: Revision-4.82 12
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ The option form +m m identifies m as a mount sup-
+ plement file.
+
+ Note: the +m and +m m options are not available for
+ all supported dialects. Check the output of lsof's
+ -h or -? options to see if the +m and +m m options
+ are available.
+
+ +|-M Enables (+) or disables (-) the reporting of port-
+ mapper registrations for local TCP, UDP and UDPLITE
+ ports. The default reporting mode is set by the
+ lsof builder with the HASPMAPENABLED #define in the
+ dialect's machine.h header file; lsof is distri-
+ buted with the HASPMAPENABLED #define deactivated,
+ so portmapper reporting is disabled by default and
+ must be requested with +M. Specifying lsof's -h or
+ -? option will report the default mode. Disabling
+ portmapper registration when it is already disabled
+ or enabling it when already enabled is acceptable.
+
+ When portmapper registration reporting is enabled,
+ lsof displays the portmapper registration (if any)
+ for local TCP, UDP or UDPLITE ports in square
+ brackets immediately following the port numbers or
+ service names - e.g., ``:1234[name]'' or
+ ``:name[100083]''. The registration information
+ may be a name or number, depending on what the
+ registering program supplied to the portmapper when
+ it registered the port.
+
+ When portmapper registration reporting is enabled,
+ lsof may run a little more slowly or even become
+ blocked when access to the portmapper becomes cong-
+ ested or stopped. Reverse the reporting mode to
+ determine if portmapper registration reporting is
+ slowing or blocking lsof.
+
+ For purposes of portmapper registration reporting
+ lsof considers a TCP, UDP or UDPLITE port local if:
+ it is found in the local part of its containing
+ kernel structure; or if it is located in the
+ foreign part of its containing kernel structure and
+ the local and foreign Internet addresses are the
+ same; or if it is located in the foreign part of
+ its containing kernel structure and the foreign
+ Internet address is INADDR_LOOPBACK (127.0.0.1).
+ This rule may make lsof ignore some foreign ports
+ on machines with multiple interfaces when the
+ foreign Internet address is on a different inter-
+ face from the local one.
+
+ See the lsof FAQ (The FAQ section gives its
+
+
+
+SunOS 5.9 Last change: Revision-4.82 13
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ location.) for further discussion of portmapper
+ registration reporting issues.
+
+ -n This option inhibits the conversion of network
+ numbers to host names for network files. Inhibit-
+ ing conversion may make lsof run faster. It is
+ also useful when host name lookup is not working
+ properly.
+
+ -N This option selects the listing of NFS files.
+
+ -o This option directs lsof to display file offset at
+ all times. It causes the SIZE/OFF output column
+ title to be changed to OFFSET. Note: on some UNIX
+ dialects lsof can't obtain accurate or consistent
+ file offset information from its kernel data
+ sources, sometimes just for particular kinds of
+ files (e.g., socket files.) Consult the lsof FAQ
+ (The FAQ section gives its location.) for more
+ information.
+
+ The -o and -s options are mutually exclusive; they
+ can't both be specified. When neither is speci-
+ fied, lsof displays whatever value - size or offset
+ - is appropriate and available for the type of the
+ file.
+
+ -o o This option defines the number of decimal digits
+ (o) to be printed after the ``0t'' for a file
+ offset before the form is switched to ``0x...''.
+ An o value of zero (unlimited) directs lsof to use
+ the ``0t'' form for all offset output.
+
+ This option does NOT direct lsof to display offset
+ at all times; specify -o (without a trailing
+ number) to do that. This option only specifies the
+ number of digits after ``0t'' in either mixed size
+ and offset or offset-only output. Thus, for exam-
+ ple, to direct lsof to display offset at all times
+ with a decimal digit count of 10, use:
+
+
+ -o -o 10
+ or
+ -oo10
+
+ The default number of digits allowed after ``0t''
+ is normally 8, but may have been changed by the
+ lsof builder. Consult the description of the -o o
+ option in the output of the -h or -? option to
+ determine the default that is in effect.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 14
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ -O This option directs lsof to bypass the strategy it
+ uses to avoid being blocked by some kernel opera-
+ tions - i.e., doing them in forked child processes.
+ See the BLOCKS AND TIMEOUTS and AVOIDING KERNEL
+ BLOCKS sections for more information on kernel
+ operations that may block lsof.
+
+ While use of this option will reduce lsof startup
+ overhead, it may also cause lsof to hang when the
+ kernel doesn't respond to a function. Use this
+ option cautiously.
+
+ -p s This option excludes or selects the listing of
+ files for the processes whose optional process
+ IDentification (PID) numbers are in the
+ comma-separated set s - e.g., ``123'' or
+ ``123,^456''. (There should be no spaces in the
+ set.)
+
+ PID numbers that begin with `^' (negation)
+ represent exclusions.
+
+ Multiple process ID numbers are joined in a single
+ ORed set before participating in AND option selec-
+ tion. However, PID exclusions are applied without
+ ORing or ANDing and take effect before other selec-
+ tion criteria are applied.
+
+ -P This option inhibits the conversion of port numbers
+ to port names for network files. Inhibiting the
+ conversion may make lsof run a little faster. It
+ is also useful when port name lookup is not working
+ properly.
+
+ +|-r [t[m<fmt>]]
+ This option puts lsof in repeat mode. There lsof
+ lists open files as selected by other options,
+ delays t seconds (default fifteen), then repeats
+ the listing, delaying and listing repetitively
+ until stopped by a condition defined by the prefix
+ to the option.
+
+ If the prefix is a `-', repeat mode is endless.
+ Lsof must be terminated with an interrupt or quit
+ signal.
+
+ If the prefix is `+', repeat mode will end the
+ first cycle no open files are listed - and of
+ course when lsof is stopped with an interrupt or
+ quit signal. When repeat mode ends because no
+ files are listed, the process exit code will be
+ zero if any open files were ever listed; one, if
+
+
+
+SunOS 5.9 Last change: Revision-4.82 15
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ none were ever listed.
+
+ Lsof marks the end of each listing: if field out-
+ put is in progress (the -F, option has been speci-
+ fied), the default marker is `m'; otherwise the
+ default marker is ``========''. The marker is fol-
+ lowed by a NL character.
+
+ The optional "m<fmt>" argument specifies a format
+ for the marker line. The <fmt> characters follow-
+ ing `m' are interpreted as a format specification
+ to the strftime(3) function, when both it and the
+ localtime(3) function are available in the
+ dialect's C library. Consult the strftime(3) docu-
+ mentation for what may appear in its format specif-
+ ication. Note that when field output is requested
+ with the -F option, <fmt> cannot contain the NL
+ format, ``%n''. Note also that when <fmt> contains
+ spaces or other characters that affect the shell's
+ interpretation of arguments, <fmt> must be quoted
+ appropriately.
+
+ Repeat mode reduces lsof startup overhead, so it is
+ more efficient to use this mode than to call lsof
+ repetitively from a shell script, for example.
+
+ To use repeat mode most efficiently, accompany +|-r
+ with specification of other lsof selection options,
+ so the amount of kernel memory access lsof does
+ will be kept to a minimum. Options that filter at
+ the process level - e.g., -c, -g, -p, -u - are the
+ most efficient selectors.
+
+ Repeat mode is useful when coupled with field out-
+ put (see the -F, option description) and a
+ supervising awk or Perl script, or a C program.
+
+ -R This option directs lsof to list the Parent Process
+ IDentification number in the PPID column.
+
+ -s [p:s] s alone directs lsof to display file size at all
+ times. It causes the SIZE/OFF output column title
+ to be changed to SIZE. If the file does not have a
+ size, nothing is displayed.
+
+ When followed by a protocol name (p), either TCP or
+ UDP, a colon (`:') and a comma-separated protocol
+ state name list, the option causes open TCP and UDP
+ files to be excluded if their state name(s) are in
+ the list (s) preceded by a `^'; or included if
+ their name(s) are not preceded by a `^'.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 16
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ When an inclusion list is defined, only network
+ files with state names in the list will be present
+ in the lsof output. Thus, specifying one state
+ name means that only network files with that lone
+ state name will be listed.
+
+ Case is unimportant in the protocol or state names,
+ but there may be no spaces and the colon (`:')
+ separating the protocol name (p) and the state name
+ list (s) is required.
+
+ If only TCP and UDP files are to be listed, as con-
+ trolled by the specified exclusions and inclusions,
+ the -i option must be specified, too. If only a
+ single protocol's files are to be listed, add its
+ name as an argument to the -i option.
+
+ For example, to list only network files with TCP
+ state LISTEN, use:
+
+
+ -iTCP -sTCP:LISTEN
+
+ Or, for example, to list network files with all UDP
+ states except Idle, use:
+
+
+ -iUDP -sUDP:Idle
+
+ State names vary with UNIX dialects, so it's not
+ possible to provide a complete list. Some common
+ TCP state names are: CLOSED, IDLE, BOUND, LISTEN,
+ ESTABLISHED, SYN_SENT, SYN_RCDV, ESTABLISHED,
+ CLOSE_WAIT, FIN_WAIT1, CLOSING, LAST_ACK,
+ FIN_WAIT_2, and TIME_WAIT. Two common UDP state
+ names are Unbound and Idle.
+
+ See the lsof FAQ (The FAQ section gives its loca-
+ tion.) for more information on how to use protocol
+ state exclusion and inclusion, including examples.
+
+ The -o (without a following decimal digit count)
+ and -s option (without a following protocol and
+ state name list) are mutually exclusive; they can't
+ both be specified. When neither is specified, lsof
+ displays whatever value - size or offset - is
+ appropriate and available for the type of file.
+
+ Since some types of files don't have true sizes -
+ sockets, FIFOs, pipes, etc. - lsof displays for
+ their sizes the content amounts in their associated
+ kernel buffers, if possible.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 17
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ -S [t] This option specifies an optional time-out seconds
+ value for kernel functions - lstat(2), readlink(2),
+ and stat(2) - that might otherwise deadlock. The
+ minimum for t is two; the default, fifteen; when no
+ value is specified, the default is used.
+
+ See the BLOCKS AND TIMEOUTS section for more infor-
+ mation.
+
+ -T [t] This option controls the reporting of some TCP/TPI
+ information, also reported by netstat(1), following
+ the network addresses. In normal output the infor-
+ mation appears in parentheses, each item except TCP
+ or TPI state name identified by a keyword, followed
+ by `=', separated from others by a single space:
+
+
+ <TCP or TPI state name>
+ QR=<read queue length>
+ QS=<send queue length>
+ SO=<socket options and values>
+ SS=<socket states>
+ TF=<TCP flags and values>
+ WR=<window read length>
+ WW=<window write length>
+
+ Not all values are reported for all UNIX dialects.
+ Items values (when available) are reported after
+ the item name and '='.
+
+ When the field output mode is in effect (See OUTPUT
+ FOR OTHER PROGRAMS.) each item appears as a field
+ with a `T' leading character.
+
+ -T with no following key characters disables
+ TCP/TPI information reporting.
+
+ -T with following characters selects the reporting
+ of specific TCP/TPI information:
+
+
+ f selects reporting of socket options,
+ states and values, and TCP flags and
+ values.
+ q selects queue length reporting.
+ s selects connection state reporting.
+ w selects window size reporting.
+
+ Not all selections are enabled for some UNIX
+ dialects. State may be selected for all dialects
+ and is reported by default. The -h or -? help out-
+ put for the -T option will show what selections may
+
+
+
+SunOS 5.9 Last change: Revision-4.82 18
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ be used with the UNIX dialect.
+
+ When -T is used to select information - i.e., it is
+ followed by one or more selection characters - the
+ displaying of state is disabled by default, and it
+ must be explicitly selected again in the characters
+ following -T. (In effect, then, the default is
+ equivalent to -Ts.) For example, if queue lengths
+ and state are desired, use -Tqs.
+
+ Socket options, socket states, some socket values,
+ TCP flags and one TCP value may be reported (when
+ available in the UNIX dialect) in the form of the
+ names that commonly appear after SO_, so_, SS_,
+ TCP_ and TF_ in the dialect's header files - most
+ often <sys/socket.h>, <sys/socketvar.h> and
+ <netinet/tcp_var.h>. Consult those header files
+ for the meaning of the flags, options, states and
+ values.
+
+ ``SO='' precedes socket options and values;
+ ``SS='', socket states; and ``TF='', TCP flags and
+ values.
+
+ If a flag or option has a value, the value will
+ follow an '=' and the name -- e.g.,
+ ``SO=LINGER=5'', ``SO=QLIM=5'', ``TF=MSS=512''.
+ The following seven values may be reported:
+
+
+ Name
+ Reported Description (Common Symbol)
+
+ KEEPALIVE keep alive time (SO_KEEPALIVE)
+ LINGER linger time (SO_LINGER)
+ MSS maximum segment size (TCP_MAXSEG)
+ PQLEN partial listen queue connections
+ QLEN established listen queue connections
+ QLIM established listen queue limit
+ RCVBUF receive buffer length (SO_RCVBUF)
+ SNDBUF send buffer length (SO_SNDBUF)
+
+ Details on what socket options and values, socket
+ states, and TCP flags and values may be displayed
+ for particular UNIX dialects may be found in the
+ answer to the ``Why doesn't lsof report socket
+ options, socket states, and TCP flags and values
+ for my dialect?'' and ``Why doesn't lsof report the
+ partial listen queue connection count for my
+ dialect?'' questions in the lsof FAQ (The FAQ sec-
+ tion gives its location.)
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 19
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ -t This option specifies that lsof should produce
+ terse output with process identifiers only and no
+ header - e.g., so that the output may be piped to
+ kill(1). This option selects the -w option.
+
+ -u s This option selects the listing of files for the
+ user whose login names or user ID numbers are in
+ the comma-separated set s - e.g., ``abe'', or
+ ``548,root''. (There should be no spaces in the
+ set.)
+
+ Multiple login names or user ID numbers are joined
+ in a single ORed set before participating in AND
+ option selection.
+
+ If a login name or user ID is preceded by a `^', it
+ becomes a negation - i.e., files of processes owned
+ by the login name or user ID will never be listed.
+ A negated login name or user ID selection is nei-
+ ther ANDed nor ORed with other selections; it is
+ applied before all other selections and absolutely
+ excludes the listing of the files of the process.
+ For example, to direct lsof to exclude the listing
+ of files belonging to root processes, specify
+ ``-u^root'' or ``-u^0''.
+
+ -U This option selects the listing of UNIX domain
+ socket files.
+
+ -v This option selects the listing of lsof version
+ information, including: revision number; when the
+ lsof binary was constructed; who constructed the
+ binary and where; the name of the compiler used to
+ construct the lsof binary; the version number of
+ the compiler when readily available; the compiler
+ and loader flags used to construct the lsof binary;
+ and system information, typically the output of
+ uname's -a option.
+
+ -V This option directs lsof to indicate the items it
+ was asked to list and failed to find - command
+ names, file names, Internet addresses or files,
+ login names, NFS files, PIDs, PGIDs, and UIDs.
+
+ When other options are ANDed to search options, or
+ compile-time options restrict the listing of some
+ files, lsof may not report that it failed to find a
+ search item when an ANDed option or compile-time
+ option prevents the listing of the open file con-
+ taining the located search item.
+
+ For example, ``lsof -V -iTCP@foobar -a -d 999'' may
+
+
+
+SunOS 5.9 Last change: Revision-4.82 20
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ not report a failure to locate open files at
+ ``TCP@foobar'' and may not list any, if none have a
+ file descriptor number of 999. A similar situation
+ arises when HASSECURITY and HASNOSOCKSECURITY are
+ defined at compile time and they prevent the list-
+ ing of open files.
+
+ +|-w Enables (+) or disables (-) the suppression of
+ warning messages.
+
+ The lsof builder may choose to have warning mes-
+ sages disabled or enabled by default. The default
+ warning message state is indicated in the output of
+ the -h or -? option. Disabling warning messages
+ when they are already disabled or enabling them
+ when already enabled is acceptable.
+
+ The -t option selects the -w option.
+
+ -x [fl] This option may accompany the +d and +D options to
+ direct their processing to cross over symbolic
+ links and|or file system mount points encountered
+ when scanning the directory (+d) or directory tree
+ (+D).
+
+ If -x is specified by itself without a following
+ parameter, cross-over processing of both symbolic
+ links and file system mount points is enabled.
+ Note that when -x is specified without a parameter,
+ the next argument must begin with '-' or '+'.
+
+ The optional 'f' parameter enables file system
+ mount point cross-over processing; 'l', symbolic
+ link cross-over processing.
+
+ The -x option may not be supplied without also sup-
+ plying a +d or +D option.
+
+ -X This is a dialect-specific option.
+
+ AIX:
+ This IBM AIX RISC/System 6000 option requests the
+ reporting of executed text file and shared library
+ references.
+
+ WARNING: because this option uses the kernel
+ readx() function, its use on a busy AIX system
+ might cause an application process to hang so com-
+ pletely that it can neither be killed nor stopped.
+ I have never seen this happen or had a report of
+ its happening, but I think there is a remote possi-
+ bility it could happen.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 21
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ By default use of readx() is disabled. On AIX 5L
+ and above lsof may need setuid-root permission to
+ perform the actions this option requests.
+
+ The lsof builder may specify that the -X option be
+ restricted to processes whose real UID is root. If
+ that has been done, the -X option will not appear
+ in the -h or -? help output unless the real UID of
+ the lsof process is root. The default lsof distri-
+ bution allows any UID to specify -X, so by default
+ it will appear in the help output.
+
+ When AIX readx() use is disabled, lsof may not be
+ able to report information for all text and loader
+ file references, but it may also avoid exacerbating
+ an AIX kernel directory search kernel error, known
+ as the Stale Segment ID bug.
+
+ The readx() function, used by lsof or any other
+ program to access some sections of kernel virtual
+ memory, can trigger the Stale Segment ID bug. It
+ can cause the kernel's dir_search() function to
+ believe erroneously that part of an in-memory copy
+ of a file system directory has been zeroed.
+ Another application process, distinct from lsof,
+ asking the kernel to search the directory - e.g.,
+ by using open(2) - can cause dir_search() to loop
+ forever, thus hanging the application process.
+
+ Consult the lsof FAQ (The FAQ section gives its
+ location.) and the 00README file of the lsof dis-
+ tribution for a more complete description of the
+ Stale Segment ID bug, its APAR, and methods for
+ defining readx() use when compiling lsof.
+
+ Linux:
+ This Linux option requests that lsof skip the
+ reporting of information on all open TCP, UDP and
+ UDPLITE IPv4 and IPv6 files.
+
+ This Linux option is most useful when the system
+ has an extremely large number of open TCP, UDP and
+ UDPLITE files, the processing of whose information
+ in the /proc/net/tcp* and /proc/net/udp* files
+ would take lsof a long time, and whose reporting is
+ not of interest.
+
+ Use this option with care and only when you are
+ sure that the information you want lsof to display
+ isn't associated with open TCP, UDP or UDPLITE
+ socket files.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 22
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ Solaris 10 and above:
+ This Solaris 10 and above option requests the
+ reporting of cached paths for files that have been
+ deleted - i.e., removed with rm(1) or unlink(2).
+
+ The cached path is followed by the string
+ `` (deleted)'' to indicate that the path by which
+ the file was opened has been deleted.
+
+ Because intervening changes made to the path -
+ i.e., renames with mv(1) or rename(2) - are not
+ recorded in the cached path, what lsof reports is
+ only the path by which the file was opened, not its
+ possibly different final path.
+
+ -z [z] specifies how Solaris 10 and higher zone informa-
+ tion is to be handled.
+
+ Without a following argument - e.g., NO z - the
+ option specifies that zone names are to be listed
+ in the ZONE output column.
+
+ The -z option may be followed by a zone name, z.
+ That causes lsof to list only open files for
+ processes in that zone. Multiple -z z option and
+ argument pairs may be specified to form a list of
+ named zones. Any open file of any process in any
+ of the zones will be listed, subject to other con-
+ ditions specified by other options and arguments.
+
+ -Z [Z] specifies how SELinux security contexts are to be
+ handled. This option and 'Z' field output charac-
+ ter support are inhibited when SELinux is disabled
+ in the running Linux kernel. See OUTPUT FOR OTHER
+ PROGRAMS for more information on the 'Z' field out-
+ put character.
+
+ Without a following argument - e.g., NO Z - the
+ option specifies that security contexts are to be
+ listed in the SECURITY-CONTEXT output column.
+
+ The -Z option may be followed by a wildcard secu-
+ rity context name, Z. That causes lsof to list
+ only open files for processes in that security con-
+ text. Multiple -Z Z option and argument pairs may
+ be specified to form a list of security contexts.
+ Any open file of any process in any of the security
+ contexts will be listed, subject to other condi-
+ tions specified by other options and arguments.
+ Note that Z can be A:B:C or *:B:C or A:B:* or *:*:C
+ to match against the A:B:C context.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 23
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ -- The double minus sign option is a marker that sig-
+ nals the end of the keyed options. It may be used,
+ for example, when the first file name begins with a
+ minus sign. It may also be used when the absence
+ of a value for the last keyed option must be signi-
+ fied by the presence of a minus sign in the follow-
+ ing option and before the start of the file names.
+
+ names These are path names of specific files to list.
+ Symbolic links are resolved before use. The first
+ name may be separated from the preceding options
+ with the ``--'' option.
+
+ If a name is the mounted-on directory of a file
+ system or the device of the file system, lsof will
+ list all the files open on the file system. To be
+ considered a file system, the name must match a
+ mounted-on directory name in mount(8) output, or
+ match the name of a block device associated with a
+ mounted-on directory name. The +|-f option may be
+ used to force lsof to consider a name a file system
+ identifier (+f) or a simple file (-f).
+
+ If name is a path to a directory that is not the
+ mounted-on directory name of a file system, it is
+ treated just as a regular file is treated - i.e.,
+ its listing is restricted to processes that have it
+ open as a file or as a process-specific directory,
+ such as the root or current working directory. To
+ request that lsof look for open files inside a
+ directory name, use the +d s and +D D options.
+
+ If a name is the base name of a family of multi-
+ plexed files - e. g, AIX's /dev/pt[cs] - lsof will
+ list all the associated multiplexed files on the
+ device that are open - e.g., /dev/pt[cs]/1,
+ /dev/pt[cs]/2, etc.
+
+ If a name is a UNIX domain socket name, lsof will
+ usually search for it by the characters of the name
+ alone - exactly as it is specified and is recorded
+ in the kernel socket structure. (See the next
+ paragraph for an exception to that rule for Linux.)
+ Specifying a relative path - e.g., ./file - in
+ place of the file's absolute path - e.g., /tmp/file
+ - won't work because lsof must match the characters
+ you specify with what it finds in the kernel UNIX
+ domain socket structures.
+
+ If a name is a Linux UNIX domain socket name, in
+ one case lsof is able to search for it by its dev-
+ ice and inode number, allowing name to be a
+
+
+
+SunOS 5.9 Last change: Revision-4.82 24
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ relative path. The case requires that the absolute
+ path -- i.e., one beginning with a slash ('/') be
+ used by the process that created the socket, and
+ hence be stored in the /proc/net/unix file; and it
+ requires that lsof be able to obtain the device and
+ node numbers of both the absolute path in
+ /proc/net/unix and name via successful stat(2) sys-
+ tem calls. When those conditions are met, lsof
+ will be able to search for the UNIX domain socket
+ when some path to it is is specified in name.
+ Thus, for example, if the path is /dev/log, and an
+ lsof search is initiated when the working directory
+ is /dev, then name could be ./log.
+
+ If a name is none of the above, lsof will list any
+ open files whose device and inode match that of the
+ specified path name.
+
+ If you have also specified the -b option, the only
+ names you may safely specify are file systems for
+ which your mount table supplies alternate device
+ numbers. See the AVOIDING KERNEL BLOCKS and ALTER-
+ NATE DEVICE NUMBERS sections for more information.
+
+ Multiple file names are joined in a single ORed set
+ before participating in AND option selection.
+
+AFS
+ Lsof supports the recognition of AFS files for these
+ dialects (and AFS versions):
+
+ AIX 4.1.4 (AFS 3.4a)
+ HP-UX 9.0.5 (AFS 3.4a)
+ Linux 1.2.13 (AFS 3.3)
+ Solaris 2.[56] (AFS 3.4a)
+
+ It may recognize AFS files on other versions of these
+ dialects, but has not been tested there. Depending on how
+ AFS is implemented, lsof may recognize AFS files in other
+ dialects, or may have difficulties recognizing AFS files in
+ the supported dialects.
+
+ Lsof may have trouble identifying all aspects of AFS files
+ in supported dialects when AFS kernel support is implemented
+ via dynamic modules whose addresses do not appear in the
+ kernel's variable name list. In that case, lsof may have to
+ guess at the identity of AFS files, and might not be able to
+ obtain volume information from the kernel that is needed for
+ calculating AFS volume node numbers. When lsof can't com-
+ pute volume node numbers, it reports blank in the NODE
+ column.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 25
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ The -A A option is available in some dialect implementations
+ of lsof for specifying the name list file where dynamic
+ module kernel addresses may be found. When this option is
+ available, it will be listed in the lsof help output,
+ presented in response to the -h or -?
+
+ See the lsof FAQ (The FAQ section gives its location.) for
+ more information about dynamic modules, their symbols, and
+ how they affect lsof options.
+
+ Because AFS path lookups don't seem to participate in the
+ kernel's name cache operations, lsof can't identify path
+ name components for AFS files.
+
+SECURITY
+ Lsof has three features that may cause security concerns.
+ First, its default compilation mode allows anyone to list
+ all open files with it. Second, by default it creates a
+ user-readable and user-writable device cache file in the
+ home directory of the real user ID that executes lsof. (The
+ list-all-open-files and device cache features may be dis-
+ abled when lsof is compiled.) Third, its -k and -m options
+ name alternate kernel name list or memory files.
+
+ Restricting the listing of all open files is controlled by
+ the compile-time HASSECURITY and HASNOSOCKSECURITY options.
+ When HASSECURITY is defined, lsof will allow only the root
+ user to list all open files. The non-root user may list
+ only open files of processes with the same user IDentifica-
+ tion number as the real user ID number of the lsof process
+ (the one that its user logged on with).
+
+ However, if HASSECURITY and HASNOSOCKSECURITY are both
+ defined, anyone may list open socket files, provided they
+ are selected with the -i option.
+
+ When HASSECURITY is not defined, anyone may list all open
+ files.
+
+ Help output, presented in response to the -h or -? option,
+ gives the status of the HASSECURITY and HASNOSOCKSECURITY
+ definitions.
+
+ See the Security section of the 00README file of the lsof
+ distribution for information on building lsof with the
+ HASSECURITY and HASNOSOCKSECURITY options enabled.
+
+ Creation and use of a user-readable and user-writable device
+ cache file is controlled by the compile-time HASDCACHE
+ option. See the DEVICE CACHE FILE section and the sections
+ that follow it for details on how its path is formed. For
+ security considerations it is important to note that in the
+
+
+
+SunOS 5.9 Last change: Revision-4.82 26
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ default lsof distribution, if the real user ID under which
+ lsof is executed is root, the device cache file will be
+ written in root's home directory - e.g., / or /root. When
+ HASDCACHE is not defined, lsof does not write or attempt to
+ read a device cache file.
+
+ When HASDCACHE is defined, the lsof help output, presented
+ in response to the -h, -D?, or -? options, will provide dev-
+ ice cache file handling information. When HASDCACHE is not
+ defined, the -h or -? output will have no -D option descrip-
+ tion.
+
+ Before you decide to disable the device cache file feature -
+ enabling it improves the performance of lsof by reducing the
+ startup overhead of examining all the nodes in /dev (or
+ /devices) - read the discussion of it in the 00DCACHE file
+ of the lsof distribution and the lsof FAQ (The FAQ section
+ gives its location.)
+
+ WHEN IN DOUBT, YOU CAN TEMPORARILY DISABLE THE USE OF THE
+ DEVICE CACHE FILE WITH THE -Di OPTION.
+
+ When lsof user declares alternate kernel name list or memory
+ files with the -k and -m options, lsof checks the user's
+ authority to read them with access(2). This is intended to
+ prevent whatever special power lsof's modes might confer on
+ it from letting it read files not normally accessible via
+ the authority of the real user ID.
+
+OUTPUT
+ This section describes the information lsof lists for each
+ open file. See the OUTPUT FOR OTHER PROGRAMS section for
+ additional information on output that can be processed by
+ another program.
+
+ Lsof only outputs printable (declared so by isprint(3)) 8
+ bit characters. Non-printable characters are printed in one
+ of three forms: the C ``\[bfrnt]'' form; the control char-
+ acter `^' form (e.g., ``^@''); or hexadecimal leading ``\x''
+ form (e.g., ``\xab''). Space is non-printable in the COM-
+ MAND column (``\x20'') and printable elsewhere.
+
+ For some dialects - if HASSETLOCALE is defined in the
+ dialect's machine.h header file - lsof will print the
+ extended 8 bit characters of a language locale. The lsof
+ process must be supplied a language locale environment vari-
+ able (e.g., LANG) whose value represents a known language
+ locale in which the extended characters are considered
+ printable by isprint(3). Otherwise lsof considers the
+ extended characters non-printable and prints them according
+ to its rules for non-printable characters, stated above.
+ Consult your dialect's setlocale(3) man page for the names
+
+
+
+SunOS 5.9 Last change: Revision-4.82 27
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ of other environment variables that may be used in place of
+ LANG - e.g., LC_ALL, LC_CTYPE, etc.
+
+ Lsof's language locale support for a dialect also covers
+ wide characters - e.g., UTF-8 - when HASSETLOCALE and
+ HASWIDECHAR are defined in the dialect's machine.h header
+ file, and when a suitable language locale has been defined
+ in the appropriate environment variable for the lsof pro-
+ cess. Wide characters are printable under those conditions
+ if iswprint(3) reports them to be. If HASSETLOCALE,
+ HASWIDECHAR and a suitable language locale aren't defined,
+ or if iswprint(3) reports wide characters that aren't print-
+ able, lsof considers the wide characters non-printable and
+ prints each of their 8 bits according to its rules for
+ non-printable characters, stated above.
+
+ Consult the answers to the "Language locale support" ques-
+ tions in the lsof FAQ (The FAQ section gives its location.)
+ for more information.
+
+ Lsof dynamically sizes the output columns each time it runs,
+ guaranteeing that each column is a minimum size. It also
+ guarantees that each column is separated from its predeces-
+ sor by at least one space.
+
+ COMMAND contains the first nine characters of the name of
+ the UNIX command associated with the process. If
+ a non-zero w value is specified to the +c w
+ option, the column contains the first w charac-
+ ters of the name of the UNIX command associated
+ with the process up to the limit of characters
+ supplied to lsof by the UNIX dialect. (See the
+ description of the +c w command or the lsof FAQ
+ for more information. The FAQ section gives its
+ location.)
+
+ If w is less than the length of the column title,
+ ``COMMAND'', it will be raised to that length.
+
+ If a zero w value is specified to the +c w
+ option, the column contains all the characters of
+ the name of the UNIX command associated with the
+ process.
+
+ All command name characters maintained by the
+ kernel in its structures are displayed in field
+ output when the command name descriptor (`c') is
+ specified. See the OUTPUT FOR OTHER COMMANDS
+ section for information on selecting field output
+ and the associated command name descriptor.
+
+ PID is the Process IDentification number of the
+
+
+
+SunOS 5.9 Last change: Revision-4.82 28
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ process.
+
+ ZONE is the Solaris 10 and higher zone name. This
+ column must be selected with the -z option.
+
+ SECURITY-CONTEXT
+ is the SELinux security context. This column
+ must be selected with the -Z option. Note that
+ the -Z option is inhibited when SELinux is dis-
+ abled in the running Linux kernel.
+
+ PPID is the Parent Process IDentification number of
+ the process. It is only displayed when the -R
+ option has been specified.
+
+ PGID is the process group IDentification number asso-
+ ciated with the process. It is only displayed
+ when the -g option has been specified.
+
+ USER is the user ID number or login name of the user
+ to whom the process belongs, usually the same as
+ reported by ps(1). However, on Linux USER is the
+ user ID number or login that owns the directory
+ in /proc where lsof finds information about the
+ process. Usually that is the same value reported
+ by ps(1), but may differ when the process has
+ changed its effective user ID. (See the -l
+ option description for information on when a user
+ ID number or login name is displayed.)
+
+ FD is the File Descriptor number of the file or:
+
+
+ cwd current working directory;
+ Lnn library references (AIX);
+ err FD information error (see NAME column);
+ jld jail directory (FreeBSD);
+ ltx shared library text (code and data);
+ Mxx hex memory-mapped type number xx.
+ m86 DOS Merge mapped file;
+ mem memory-mapped file;
+ mmap memory-mapped device;
+ pd parent directory;
+ rtd root directory;
+ tr kernel trace file (OpenBSD);
+ txt program text (code and data);
+ v86 VP/ix mapped file;
+
+ FD is followed by one of these characters,
+ describing the mode under which the file is open:
+
+ r for read access;
+
+
+
+SunOS 5.9 Last change: Revision-4.82 29
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ w for write access;
+ u for read and write access;
+ space if mode unknown and no lock
+ character follows;
+ `-' if mode unknown and lock
+ character follows.
+
+ The mode character is followed by one of these
+ lock characters, describing the type of lock
+ applied to the file:
+
+ N for a Solaris NFS lock of unknown type;
+ r for read lock on part of the file;
+ R for a read lock on the entire file;
+ w for a write lock on part of the file;
+ W for a write lock on the entire file;
+ u for a read and write lock of any length;
+ U for a lock of unknown type;
+ x for an SCO OpenServer Xenix lock on part
+ of the file;
+ X for an SCO OpenServer Xenix lock on the
+ entire file;
+ space if there is no lock.
+
+ See the LOCKS section for more information on the
+ lock information character.
+
+ The FD column contents constitutes a single field
+ for parsing in post-processing scripts.
+
+ TYPE is the type of the node associated with the file
+ - e.g., GDIR, GREG, VDIR, VREG, etc.
+
+ or ``IPv4'' for an IPv4 socket;
+
+ or ``IPv6'' for an open IPv6 network file - even
+ if its address is IPv4, mapped in an IPv6
+ address;
+
+ or ``ax25'' for a Linux AX.25 socket;
+
+ or ``inet'' for an Internet domain socket;
+
+ or ``lla'' for a HP-UX link level access file;
+
+ or ``rte'' for an AF_ROUTE socket;
+
+ or ``sock'' for a socket of unknown domain;
+
+ or ``unix'' for a UNIX domain socket;
+
+ or ``x.25'' for an HP-UX x.25 socket;
+
+
+
+SunOS 5.9 Last change: Revision-4.82 30
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ or ``BLK'' for a block special file;
+
+ or ``CHR'' for a character special file;
+
+ or ``DEL'' for a Linux map file that has been
+ deleted;
+
+ or ``DIR'' for a directory;
+
+ or ``DOOR'' for a VDOOR file;
+
+ or ``FIFO'' for a FIFO special file;
+
+ or ``KQUEUE'' for a BSD style kernel event queue
+ file;
+
+ or ``LINK'' for a symbolic link file;
+
+ or ``MPB'' for a multiplexed block file;
+
+ or ``MPC'' for a multiplexed character file;
+
+ or ``NOFD'' for a Linux /proc/<PID>/fd directory
+ that can't be opened -- the directory path
+ appears in the NAME column, followed by an error
+ message;
+
+ or ``PAS'' for a /proc/as file;
+
+ or ``PAXV'' for a /proc/auxv file;
+
+ or ``PCRE'' for a /proc/cred file;
+
+ or ``PCTL'' for a /proc control file;
+
+ or ``PCUR'' for the current /proc process;
+
+ or ``PCWD'' for a /proc current working direc-
+ tory;
+
+ or ``PDIR'' for a /proc directory;
+
+ or ``PETY'' for a /proc executable type (etype);
+
+ or ``PFD'' for a /proc file descriptor;
+
+ or ``PFDR'' for a /proc file descriptor direc-
+ tory;
+
+ or ``PFIL'' for an executable /proc file;
+
+ or ``PFPR'' for a /proc FP register set;
+
+
+
+SunOS 5.9 Last change: Revision-4.82 31
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ or ``PGD'' for a /proc/pagedata file;
+
+ or ``PGID'' for a /proc group notifier file;
+
+ or ``PIPE'' for pipes;
+
+ or ``PLC'' for a /proc/lwpctl file;
+
+ or ``PLDR'' for a /proc/lpw directory;
+
+ or ``PLDT'' for a /proc/ldt file;
+
+ or ``PLPI'' for a /proc/lpsinfo file;
+
+ or ``PLST'' for a /proc/lstatus file;
+
+ or ``PLU'' for a /proc/lusage file;
+
+ or ``PLWG'' for a /proc/gwindows file;
+
+ or ``PLWI'' for a /proc/lwpsinfo file;
+
+ or ``PLWS'' for a /proc/lwpstatus file;
+
+ or ``PLWU'' for a /proc/lwpusage file;
+
+ or ``PLWX'' for a /proc/xregs file'
+
+ or ``PMAP'' for a /proc map file (map);
+
+ or ``PMEM'' for a /proc memory image file;
+
+ or ``PNTF'' for a /proc process notifier file;
+
+ or ``POBJ'' for a /proc/object file;
+
+ or ``PODR'' for a /proc/object directory;
+
+ or ``POLP'' for an old format /proc light weight
+ process file;
+
+ or ``POPF'' for an old format /proc PID file;
+
+ or ``POPG'' for an old format /proc page data
+ file;
+
+ or ``PORT'' for a SYSV named pipe;
+
+ or ``PREG'' for a /proc register file;
+
+ or ``PRMP'' for a /proc/rmap file;
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 32
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ or ``PRTD'' for a /proc root directory;
+
+ or ``PSGA'' for a /proc/sigact file;
+
+ or ``PSIN'' for a /proc/psinfo file;
+
+ or ``PSTA'' for a /proc status file;
+
+ or ``PSXSEM'' for a POSIX semaphore file;
+
+ or ``PSXSHM'' for a POSIX shared memory file;
+
+ or ``PUSG'' for a /proc/usage file;
+
+ or ``PW'' for a /proc/watch file;
+
+ or ``PXMP'' for a /proc/xmap file;
+
+ or ``REG'' for a regular file;
+
+ or ``SMT'' for a shared memory transport file;
+
+ or ``STSO'' for a stream socket;
+
+ or ``UNNM'' for an unnamed type file;
+
+ or ``XNAM'' for an OpenServer Xenix special file
+ of unknown type;
+
+ or ``XSEM'' for an OpenServer Xenix semaphore
+ file;
+
+ or ``XSD'' for an OpenServer Xenix shared data
+ file;
+
+ or the four type number octets if the correspond-
+ ing name isn't known.
+
+ FILE-ADDR contains the kernel file structure address when f
+ has been specified to +f;
+
+ FCT contains the file reference count from the kernel
+ file structure when c has been specified to +f;
+
+ FILE-FLAG when g or G has been specified to +f, this field
+ contains the contents of the f_flag[s] member of
+ the kernel file structure and the kernel's
+ per-process open file flags (if available); `G'
+ causes them to be displayed in hexadecimal; `g',
+ as short-hand names; two lists may be displayed
+ with entries separated by commas, the lists
+ separated by a semicolon (`;'); the first list
+
+
+
+SunOS 5.9 Last change: Revision-4.82 33
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ may contain short-hand names for f_flag[s] values
+ from the following table:
+
+
+ AIO asynchronous I/O (e.g., FAIO)
+ AP append
+ ASYN asynchronous I/O (e.g., FASYNC)
+ BAS block, test, and set in use
+ BKIU block if in use
+ BL use block offsets
+ BSK block seek
+ CA copy avoid
+ CIO concurrent I/O
+ CLON clone
+ CLRD CL read
+ CR create
+ DF defer
+ DFI defer IND
+ DFLU data flush
+ DIR direct
+ DLY delay
+ DOCL do clone
+ DSYN data-only integrity
+ DTY must be a directory
+ EVO event only
+ EX open for exec
+ EXCL exclusive open
+ FSYN synchronous writes
+ GCDF defer during unp_gc() (AIX)
+ GCMK mark during unp_gc() (AIX)
+ GTTY accessed via /dev/tty
+ HUP HUP in progress
+ KERN kernel
+ KIOC kernel-issued ioctl
+ LCK has lock
+ LG large file
+ MBLK stream message block
+ MK mark
+ MNT mount
+ MSYN multiplex synchronization
+ NATM don't update atime
+ NB non-blocking I/O
+ NBDR no BDRM check
+ NBIO SYSV non-blocking I/O
+ NBF n-buffering in effect
+ NC no cache
+ ND no delay
+ NDSY no data synchronization
+ NET network
+ NFLK don't follow links
+ NMFS NM file system
+ NOTO disable background stop
+
+
+
+SunOS 5.9 Last change: Revision-4.82 34
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ NSH no share
+ NTTY no controlling TTY
+ OLRM OLR mirror
+ PAIO POSIX asynchronous I/O
+ PP POSIX pipe
+ R read
+ RC file and record locking cache
+ REV revoked
+ RSH shared read
+ RSYN read synchronization
+ RW read and write access
+ SL shared lock
+ SNAP cooked snapshot
+ SOCK socket
+ SQSH Sequent shared set on open
+ SQSV Sequent SVM set on open
+ SQR Sequent set repair on open
+ SQS1 Sequent full shared open
+ SQS2 Sequent partial shared open
+ STPI stop I/O
+ SWR synchronous read
+ SYN file integrity while writing
+ TCPM avoid TCP collision
+ TR truncate
+ W write
+ WKUP parallel I/O synchronization
+ WTG parallel I/O synchronization
+ VH vhangup pending
+ VTXT virtual text
+ XL exclusive lock
+
+ this list of names was derived from F* #define's
+ in dialect header files <fcntl.h>, <linux</fs.h>,
+ <sys/fcntl.c>, <sys/fcntlcom.h>, and
+ <sys/file.h>; see the lsof.h header file for a
+ list showing the correspondence between the above
+ short-hand names and the header file definitions;
+
+ the second list (after the semicolon) may contain
+ short-hand names for kernel per-process open file
+ flags from this table:
+
+
+ ALLC allocated
+ BR the file has been read
+ BHUP activity stopped by SIGHUP
+ BW the file has been written
+ CLSG closing
+ CX close-on-exec (see fcntl(F_SETFD))
+ LCK lock was applied
+ MP memory-mapped
+ OPIP open pending - in progress
+
+
+
+SunOS 5.9 Last change: Revision-4.82 35
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ RSVW reserved wait
+ SHMT UF_FSHMAT set (AIX)
+ USE in use (multi-threaded)
+
+ NODE-ID (or INODE-ADDR for some dialects) contains a
+ unique identifier for the file node (usually the
+ kernel vnode or inode address, but also occasion-
+ ally a concatenation of device and node number)
+ when n has been specified to +f;
+
+ DEVICE contains the device numbers, separated by commas,
+ for a character special, block special, regular,
+ directory or NFS file;
+
+ or ``memory'' for a memory file system node under
+ Tru64 UNIX;
+
+ or the address of the private data area of a
+ Solaris socket stream;
+
+ or a kernel reference address that identifies the
+ file (The kernel reference address may be used
+ for FIFO's, for example.);
+
+ or the base address or device name of a Linux
+ AX.25 socket device.
+
+ Usually only the lower thirty two bits of Tru64
+ UNIX kernel addresses are displayed.
+
+ SIZE, SIZE/OFF, or OFFSET
+ is the size of the file or the file offset in
+ bytes. A value is displayed in this column only
+ if it is available. Lsof displays whatever value
+ - size or offset - is appropriate for the type of
+ the file and the version of lsof.
+
+ On some UNIX dialects lsof can't obtain accurate
+ or consistent file offset information from its
+ kernel data sources, sometimes just for particu-
+ lar kinds of files (e.g., socket files.) In
+ other cases, files don't have true sizes - e.g.,
+ sockets, FIFOs, pipes - so lsof displays for
+ their sizes the content amounts it finds in their
+ kernel buffer descriptors (e.g., socket buffer
+ size counts or TCP/IP window sizes.) Consult the
+ lsof FAQ (The FAQ section gives its location.)
+ for more information.
+
+ The file size is displayed in decimal; the offset
+ is normally displayed in decimal with a leading
+ ``0t'' if it contains 8 digits or less; in
+
+
+
+SunOS 5.9 Last change: Revision-4.82 36
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ hexadecimal with a leading ``0x'' if it is longer
+ than 8 digits. (Consult the -o o option descrip-
+ tion for information on when 8 might default to
+ some other value.)
+
+ Thus the leading ``0t'' and ``0x'' identify an
+ offset when the column may contain both a size
+ and an offset (i.e., its title is SIZE/OFF).
+
+ If the -o option is specified, lsof always
+ displays the file offset (or nothing if no offset
+ is available) and labels the column OFFSET. The
+ offset always begins with ``0t'' or ``0x'' as
+ described above.
+
+ The lsof user can control the switch from ``0t''
+ to ``0x'' with the -o o option. Consult its
+ description for more information.
+
+ If the -s option is specified, lsof always
+ displays the file size (or nothing if no size is
+ available) and labels the column SIZE. The -o
+ and -s options are mutually exclusive; they can't
+ both be specified.
+
+ For files that don't have a fixed size - e.g.,
+ don't reside on a disk device - lsof will display
+ appropriate information about the current size or
+ position of the file if it is available in the
+ kernel structures that define the file.
+
+ NLINK contains the file link count when +L has been
+ specified;
+
+ NODE is the node number of a local file;
+
+ or the inode number of an NFS file in the server
+ host;
+
+ or the Internet protocol type - e. g, ``TCP'';
+
+ or ``STR'' for a stream;
+
+ or ``CCITT'' for an HP-UX x.25 socket;
+
+ or the IRQ or inode number of a Linux AX.25
+ socket device.
+
+ NAME is the name of the mount point and file system on
+ which the file resides;
+
+ or the name of a file specified in the names
+
+
+
+SunOS 5.9 Last change: Revision-4.82 37
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ option (after any symbolic links have been
+ resolved);
+
+ or the name of a character special or block spe-
+ cial device;
+
+ or the local and remote Internet addresses of a
+ network file; the local host name or IP number is
+ followed by a colon (':'), the port, ``->'', and
+ the two-part remote address; IP addresses may be
+ reported as numbers or names, depending on the
+ +|-M, -n, and -P options; colon-separated IPv6
+ numbers are enclosed in square brackets; IPv4
+ INADDR_ANY and IPv6 IN6_IS_ADDR_UNSPECIFIED
+ addresses, and zero port numbers are represented
+ by an asterisk ('*'); a UDP destination address
+ may be followed by the amount of time elapsed
+ since the last packet was sent to the destina-
+ tion; TCP, UDP and UDPLITE remote addresses may
+ be followed by TCP/TPI information in parentheses
+ - state (e.g., ``(ESTABLISHED)'', ``(Unbound)''),
+ queue sizes, and window sizes (not all dialects)
+ - in a fashion similar to what netstat(1)
+ reports; see the -T option description or the
+ description of the TCP/TPI field in OUTPUT FOR
+ OTHER PROGRAMS for more information on state,
+ queue size, and window size;
+
+ or the address or name of a UNIX domain socket,
+ possibly including a stream clone device name, a
+ file system object's path name, local and foreign
+ kernel addresses, socket pair information, and a
+ bound vnode address;
+
+ or the local and remote mount point names of an
+ NFS file;
+
+ or ``STR'', followed by the stream name;
+
+ or a stream character device name, followed by
+ ``->'' and the stream name or a list of stream
+ module names, separated by ``->'';
+
+ or ``STR:'' followed by the SCO OpenServer stream
+ device and module names, separated by ``->'';
+
+ or system directory name, `` -- '', and as many
+ components of the path name as lsof can find in
+ the kernel's name cache for selected dialects
+ (See the KERNEL NAME CACHE section for more
+ information.);
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 38
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ or ``PIPE->'', followed by a Solaris kernel pipe
+ destination address;
+
+ or ``COMMON:'', followed by the vnode device
+ information structure's device name, for a
+ Solaris common vnode;
+
+ or the address family, followed by a slash (`/'),
+ followed by fourteen comma-separated bytes of a
+ non-Internet raw socket address;
+
+ or the HP-UX x.25 local address, followed by the
+ virtual connection number (if any), followed by
+ the remote address (if any);
+
+ or ``(dead)'' for disassociated Tru64 UNIX files
+ - typically terminal files that have been flagged
+ with the TIOCNOTTY ioctl and closed by daemons;
+
+ or ``rd=<offset>'' and ``wr=<offset>'' for the
+ values of the read and write offsets of a FIFO;
+
+ or ``clone n:/dev/event'' for SCO OpenServer file
+ clones of the /dev/event device, where n is the
+ minor device number of the file;
+
+ or ``(socketpair: n)'' for a Solaris 2.6, 8, 9
+ or 10 UNIX domain socket, created by the
+ socketpair(3N) network function;
+
+ or ``no PCB'' for socket files that do not have a
+ protocol block associated with them, optionally
+ followed by ``, CANTSENDMORE'' if sending on the
+ socket has been disabled, or ``, CANTRCVMORE'' if
+ receiving on the socket has been disabled (e.g.,
+ by the shutdown(2) function);
+
+ or the local and remote addresses of a Linux IPX
+ socket file in the form <net>:[<node>:]<port>,
+ followed in parentheses by the transmit and
+ receive queue sizes, and the connection state;
+
+ or ``dgram'' or ``stream'' for the type UnixWare
+ 7.1.1 and above in-kernel UNIX domain sockets,
+ followed by a colon (':') and the local path name
+ when available, followed by ``->'' and the remote
+ path name or kernel socket address in hexadecimal
+ when available.
+
+ For dialects that support a ``namefs'' file system, allowing
+ one file to be attached to another with fattach(3C), lsof
+ will add ``(FA:<address1><direction><address2>)'' to the
+
+
+
+SunOS 5.9 Last change: Revision-4.82 39
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ NAME column. <address1> and <address2> are hexadecimal
+ vnode addresses. <direction> will be ``<-'' if <address2>
+ has been fattach'ed to this vnode whose address is
+ <address1>; and ``->'' if <address1>, the vnode address of
+ this vnode, has been fattach'ed to <address2>. <address1>
+ may be omitted if it already appears in the DEVICE column.
+
+ Lsof may add two parenthetical notes to the NAME column for
+ open Solaris 10 files: ``(?)'' if lsof considers the path
+ name of questionable accuracy; and ``(deleted)'' if the -X
+ option has been specified and lsof detects the open file's
+ path name has been deleted. Consult the lsof FAQ (The FAQ
+ section gives its location.) for more information on these
+ NAME column additions.
+
+LOCKS
+ Lsof can't adequately report the wide variety of UNIX
+ dialect file locks in a single character. What it reports
+ in a single character is a compromise between the informa-
+ tion it finds in the kernel and the limitations of the
+ reporting format.
+
+ Moreover, when a process holds several byte level locks on a
+ file, lsof only reports the status of the first lock it
+ encounters. If it is a byte level lock, then the lock char-
+ acter will be reported in lower case - i.e., `r', `w', or
+ `x' - rather than the upper case equivalent reported for a
+ full file lock.
+
+ Generally lsof can only report on locks held by local
+ processes on local files. When a local process sets a lock
+ on a remotely mounted (e.g., NFS) file, the remote server
+ host usually records the lock state. One exception is
+ Solaris - at some patch levels of 2.3, and in all versions
+ above 2.4, the Solaris kernel records information on remote
+ locks in local structures.
+
+ Lsof has trouble reporting locks for some UNIX dialects.
+ Consult the BUGS section of this manual page or the lsof FAQ
+ (The FAQ section gives its location.) for more information.
+
+OUTPUT FOR OTHER PROGRAMS
+ When the -F option is specified, lsof produces output that
+ is suitable for processing by another program - e.g, an awk
+ or Perl script, or a C program.
+
+ Each unit of information is output in a field that is iden-
+ tified with a leading character and terminated by a NL (012)
+ (or a NUL (000) if the 0 (zero) field identifier character
+ is specified.) The data of the field follows immediately
+ after the field identification character and extends to the
+ field terminator.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 40
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ It is possible to think of field output as process and file
+ sets. A process set begins with a field whose identifier is
+ `p' (for process IDentifier (PID)). It extends to the
+ beginning of the next PID field or the beginning of the
+ first file set of the process, whichever comes first.
+ Included in the process set are fields that identify the
+ command, the process group IDentification (PGID) number, and
+ the user ID (UID) number or login name.
+
+ A file set begins with a field whose identifier is `f' (for
+ file descriptor). It is followed by lines that describe the
+ file's access mode, lock state, type, device, size, offset,
+ inode, protocol, name and stream module names. It extends
+ to the beginning of the next file or process set, whichever
+ comes first.
+
+ When the NUL (000) field terminator has been selected with
+ the 0 (zero) field identifier character, lsof ends each pro-
+ cess and file set with a NL (012) character.
+
+ Lsof always produces one field, the PID (`p') field. All
+ other fields may be declared optionally in the field iden-
+ tifier character list that follows the -F option. When a
+ field selection character identifies an item lsof does not
+ normally list - e.g., PPID, selected with -R - specification
+ of the field character - e.g., ``-FR'' - also selects the
+ listing of the item.
+
+ It is entirely possible to select a set of fields that can-
+ not easily be parsed - e.g., if the field descriptor field
+ is not selected, it may be difficult to identify file sets.
+ To help you avoid this difficulty, lsof supports the -F
+ option; it selects the output of all fields with NL termina-
+ tors (the -F0 option pair selects the output of all fields
+ with NUL terminators). For compatibility reasons neither -F
+ nor -F0 select the raw device field.
+
+ These are the fields that lsof will produce. The single
+ character listed first is the field identifier.
+
+ a file access mode
+ c process command name (all characters from proc or
+ user structure)
+ C file structure share count
+ d file's device character code
+ D file's major/minor device number (0x<hexadecimal>)
+ f file descriptor
+ F file structure address (0x<hexadecimal>)
+ G file flaGs (0x<hexadecimal>; names if +fg follows)
+ i file's inode number
+ k link count
+ l file's lock status
+
+
+
+SunOS 5.9 Last change: Revision-4.82 41
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ L process login name
+ m marker between repeated output
+ n file name, comment, Internet address
+ N node identifier (ox<hexadecimal>
+ o file's offset (decimal)
+ p process ID (always selected)
+ g process group ID
+ P protocol name
+ r raw device number (0x<hexadecimal>)
+ R parent process ID
+ s file's size (decimal)
+ S file's stream identification
+ t file's type
+ T TCP/TPI information, identified by prefixes (the
+ `=' is part of the prefix):
+ QR=<read queue size>
+ QS=<send queue size>
+ SO=<socket options and values> (not all dialects)
+ SS=<socket states> (not all dialects)
+ ST=<connection state>
+ TF=<TCP flags and values> (not all dialects)
+ WR=<window read size> (not all dialects)
+ WW=<window write size> (not all dialects)
+ (TCP/TPI information isn't reported for all supported
+ UNIX dialects. The -h or -? help output for the
+ -T option will show what TCP/TPI reporting can be
+ requested.)
+ u process user ID
+ z Solaris 10 and higher zone name
+ Z SELinux security context (inhibited when SELinux is disabled)
+ 0 use NUL field terminator character in place of NL
+ 1-9 dialect-specific field identifiers (The output
+ of -F? identifies the information to be found
+ in dialect-specific fields.)
+
+ You can get on-line help information on these characters and
+ their descriptions by specifying the -F? option pair.
+ (Escape the `?' character as your shell requires.) Addi-
+ tional information on field content can be found in the OUT-
+ PUT section.
+
+ As an example, ``-F pcfn'' will select the process ID (`p'),
+ command name (`c'), file descriptor (`f') and file name
+ (`n') fields with an NL field terminator character; ``-F
+ pcfn0'' selects the same output with a NUL (000) field ter-
+ minator character.
+
+ Lsof doesn't produce all fields for every process or file
+ set, only those that are available. Some fields are mutu-
+ ally exclusive: file device characters and file major/minor
+ device numbers; file inode number and protocol name; file
+ name and stream identification; file size and offset. One
+
+
+
+SunOS 5.9 Last change: Revision-4.82 42
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ or the other member of these mutually exclusive sets will
+ appear in field output, but not both.
+
+ Normally lsof ends each field with a NL (012) character.
+ The 0 (zero) field identifier character may be specified to
+ change the field terminator character to a NUL (000). A NUL
+ terminator may be easier to process with xargs (1), for
+ example, or with programs whose quoting mechanisms may not
+ easily cope with the range of characters in the field out-
+ put. When the NUL field terminator is in use, lsof ends
+ each process and file set with a NL (012).
+
+ Three aids to producing programs that can process lsof field
+ output are included in the lsof distribution. The first is
+ a C header file, lsof_fields.h, that contains symbols for
+ the field identification characters, indexes for storing
+ them in a table, and explanation strings that may be com-
+ piled into programs. Lsof uses this header file.
+
+ The second aid is a set of sample scripts that process field
+ output, written in awk, Perl 4, and Perl 5. They're located
+ in the scripts subdirectory of the lsof distribution.
+
+ The third aid is the C library used for the lsof test suite.
+ The test suite is written in C and uses field output to
+ validate the correct operation of lsof. The library can be
+ found in the tests/LTlib.c file of the lsof distribution.
+ The library uses the first aid, the lsof_fields.h header
+ file.
+
+BLOCKS AND TIMEOUTS
+ Lsof can be blocked by some kernel functions that it uses -
+ lstat(2), readlink(2), and stat(2). These functions are
+ stalled in the kernel, for example, when the hosts where
+ mounted NFS file systems reside become inaccessible.
+
+ Lsof attempts to break these blocks with timers and child
+ processes, but the techniques are not wholly reliable. When
+ lsof does manage to break a block, it will report the break
+ with an error message. The messages may be suppressed with
+ the -t and -w options.
+
+ The default timeout value may be displayed with the -h or -?
+ option, and it may be changed with the -S [t] option. The
+ minimum for t is two seconds, but you should avoid small
+ values, since slow system responsiveness can cause short
+ timeouts to expire unexpectedly and perhaps stop lsof before
+ it can produce any output.
+
+ When lsof has to break a block during its access of mounted
+ file system information, it normally continues, although
+ with less information available to display about open files.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 43
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ Lsof can also be directed to avoid the protection of timers
+ and child processes when using the kernel functions that
+ might block by specifying the -O option. While this will
+ allow lsof to start up with less overhead, it exposes lsof
+ completely to the kernel situations that might block it.
+ Use this option cautiously.
+
+AVOIDING KERNEL BLOCKS
+ You can use the -b option to tell lsof to avoid using kernel
+ functions that would block. Some cautions apply.
+
+ First, using this option usually requires that your system
+ supply alternate device numbers in place of the device
+ numbers that lsof would normally obtain with the lstat(2)
+ and stat(2) kernel functions. See the ALTERNATE DEVICE
+ NUMBERS section for more information on alternate device
+ numbers.
+
+ Second, you can't specify names for lsof to locate unless
+ they're file system names. This is because lsof needs to
+ know the device and inode numbers of files listed with names
+ in the lsof options, and the -b option prevents lsof from
+ obtaining them. Moreover, since lsof only has device
+ numbers for the file systems that have alternates, its abil-
+ ity to locate files on file systems depends completely on
+ the availability and accuracy of the alternates. If no
+ alternates are available, or if they're incorrect, lsof
+ won't be able to locate files on the named file systems.
+
+ Third, if the names of your file system directories that
+ lsof obtains from your system's mount table are symbolic
+ links, lsof won't be able to resolve the links. This is
+ because the -b option causes lsof to avoid the kernel
+ readlink(2) function it uses to resolve symbolic links.
+
+ Finally, using the -b option causes lsof to issue warning
+ messages when it needs to use the kernel functions that the
+ -b option directs it to avoid. You can suppress these mes-
+ sages by specifying the -w option, but if you do, you won't
+ see the alternate device numbers reported in the warning
+ messages.
+
+ALTERNATE DEVICE NUMBERS
+ On some dialects, when lsof has to break a block because it
+ can't get information about a mounted file system via the
+ lstat(2) and stat(2) kernel functions, or because you speci-
+ fied the -b option, lsof can obtain some of the information
+ it needs - the device number and possibly the file system
+ type - from the system mount table. When that is possible,
+ lsof will report the device number it obtained. (You can
+ suppress the report by specifying the -w option.)
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 44
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ You can assist this process if your mount table is supported
+ with an /etc/mtab or /etc/mnttab file that contains an
+ options field by adding a ``dev=xxxx'' field for mount
+ points that do not have one in their options strings. Note:
+ you must be able to edit the file - i.e., some mount tables
+ like recent Solaris /etc/mnttab or Linux /proc/mounts are
+ read-only and can't be modified.
+
+ You may also be able to supply device numbers using the +m
+ and +m m options, provided they are supported by your
+ dialect. Check the output of lsof's -h or -? options to see
+ if the +m and +m m options are available.
+
+ The ``xxxx'' portion of the field is the hexadecimal value
+ of the file system's device number. (Consult the st_dev
+ field of the output of the lstat(2) and stat(2) functions
+ for the appropriate values for your file systems.) Here's
+ an example from a Sun Solaris 2.6 /etc/mnttab for a file
+ system remotely mounted via NFS:
+
+ nfs ignore,noquota,dev=2a40001
+
+ There's an advantage to having ``dev=xxxx'' entries in your
+ mount table file, especially for file systems that are
+ mounted from remote NFS servers. When a remote server
+ crashes and you want to identify its users by running lsof
+ on one of its clients, lsof probably won't be able to get
+ output from the lstat(2) and stat(2) functions for the file
+ system. If it can obtain the file system's device number
+ from the mount table, it will be able to display the files
+ open on the crashed NFS server.
+
+ Some dialects that do not use an ASCII /etc/mtab or
+ /etc/mnttab file for the mount table may still provide an
+ alternative device number in their internal mount tables.
+ This includes AIX, Apple Darwin, FreeBSD, NetBSD, OpenBSD,
+ and Tru64 UNIX. Lsof knows how to obtain the alternative
+ device number for these dialects and uses it when its
+ attempt to lstat(2) or stat(2) the file system is blocked.
+
+ If you're not sure your dialect supplies alternate device
+ numbers for file systems from its mount table, use this lsof
+ incantation to see if it reports any alternate device
+ numbers:
+
+ lsof -b
+
+ Look for standard error file warning messages that begin
+ ``assuming "dev=xxxx" from ...''.
+
+KERNEL NAME CACHE
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 45
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ Lsof is able to examine the kernel's name cache or use other
+ kernel facilities (e.g., the ADVFS 4.x tag_to_path() func-
+ tion under Tru64 UNIX) on some dialects for most file system
+ types, excluding AFS, and extract recently used path name
+ components from it. (AFS file system path lookups don't use
+ the kernel's name cache; some Solaris VxFS file system
+ operations apparently don't use it, either.)
+
+ Lsof reports the complete paths it finds in the NAME column.
+ If lsof can't report all components in a path, it reports in
+ the NAME column the file system name, followed by a space,
+ two `-' characters, another space, and the name components
+ it has located, separated by the `/' character.
+
+ When lsof is run in repeat mode - i.e., with the -r option
+ specified - the extent to which it can report path name com-
+ ponents for the same file may vary from cycle to cycle.
+ That's because other running processes can cause the kernel
+ to remove entries from its name cache and replace them with
+ others.
+
+ Lsof's use of the kernel name cache to identify the paths of
+ files can lead it to report incorrect components under some
+ circumstances. This can happen when the kernel name cache
+ uses device and node number as a key (e.g., SCO OpenServer)
+ and a key on a rapidly changing file system is reused. If
+ the UNIX dialect's kernel doesn't purge the name cache entry
+ for a file when it is unlinked, lsof may find a reference to
+ the wrong entry in the cache. The lsof FAQ (The FAQ section
+ gives its location.) has more information on this situa-
+ tion.
+
+ Lsof can report path name components for these dialects:
+
+ FreeBSD
+ HP-UX
+ Linux
+ NetBSD
+ NEXTSTEP
+ OpenBSD
+ OPENSTEP
+ SCO OpenServer
+ SCO|Caldera UnixWare
+ Solaris
+ Tru64 UNIX
+
+ Lsof can't report path name components for these dialects:
+
+ AIX
+
+ If you want to know why lsof can't report path name com-
+ ponents for some dialects, see the lsof FAQ (The FAQ section
+
+
+
+SunOS 5.9 Last change: Revision-4.82 46
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ gives its location.)
+
+DEVICE CACHE FILE
+ Examining all members of the /dev (or /devices) node tree
+ with stat(2) functions can be time consuming. What's more,
+ the information that lsof needs - device number, inode
+ number, and path - rarely changes.
+
+ Consequently, lsof normally maintains an ASCII text file of
+ cached /dev (or /devices) information (exception: the
+ /proc-based Linux lsof where it's not needed.) The local
+ system administrator who builds lsof can control the way the
+ device cache file path is formed, selecting from these
+ options:
+
+ Path from the -D option;
+ Path from an environment variable;
+ System-wide path;
+ Personal path (the default);
+ Personal path, modified by an environment variable.
+
+ Consult the output of the -h, -D? , or -? help options for
+ the current state of device cache support. The help output
+ lists the default read-mode device cache file path that is
+ in effect for the current invocation of lsof. The -D?
+ option output lists the read-only and write device cache
+ file paths, the names of any applicable environment vari-
+ ables, and the personal device cache path format.
+
+ Lsof can detect that the current device cache file has been
+ accidentally or maliciously modified by integrity checks,
+ including the computation and verification of a sixteen bit
+ Cyclic Redundancy Check (CRC) sum on the file's contents.
+ When lsof senses something wrong with the file, it issues a
+ warning and attempts to remove the current cache file and
+ create a new copy, but only to a path that the process can
+ legitimately write.
+
+ The path from which a lsof process may attempt to read a
+ device cache file may not be the same as the path to which
+ it can legitimately write. Thus when lsof senses that it
+ needs to update the device cache file, it may choose a dif-
+ ferent path for writing it from the path from which it read
+ an incorrect or outdated version.
+
+ If available, the -Dr option will inhibit the writing of a
+ new device cache file. (It's always available when speci-
+ fied without a path name argument.)
+
+ When a new device is added to the system, the device cache
+ file may need to be recreated. Since lsof compares the
+ mtime of the device cache file with the mtime and ctime of
+
+
+
+SunOS 5.9 Last change: Revision-4.82 47
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ the /dev (or /devices) directory, it usually detects that a
+ new device has been added; in that case lsof issues a warn-
+ ing message and attempts to rebuild the device cache file.
+
+ Whenever lsof writes a device cache file, it sets its owner-
+ ship to the real UID of the executing process, and its per-
+ mission modes to 0600, this restricting its reading and
+ writing to the file's owner.
+
+LSOF PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS
+ Two permissions of the lsof executable affect its ability to
+ access device cache files. The permissions are set by the
+ local system administrator when lsof is installed.
+
+ The first and rarer permission is setuid-root. It comes
+ into effect when lsof is executed; its effective UID is then
+ root, while its real (i.e., that of the logged-on user) UID
+ is not. The lsof distribution recommends that versions for
+ these dialects run setuid-root.
+
+ HP-UX 11.11 and 11.23
+ Linux
+
+ The second and more common permission is setgid. It comes
+ into effect when the effective group IDentification number
+ (GID) of the lsof process is set to one that can access ker-
+ nel memory devices - e.g., ``kmem'', ``sys'', or ``system''.
+
+ An lsof process that has setgid permission usually
+ surrenders the permission after it has accessed the kernel
+ memory devices. When it does that, lsof can allow more
+ liberal device cache path formations. The lsof distribution
+ recommends that versions for these dialects run setgid and
+ be allowed to surrender setgid permission.
+
+ AIX 5.[12] and 5.3-ML1
+ Apple Darwin 7.x Power Macintosh systems
+ FreeBSD 4.x, 4.1x, 5.x and [67].x for x86-based systems
+ FreeBSD 5.x and [67].x for Alpha, AMD64 and Sparc64-based
+ systems
+ HP-UX 11.00
+ NetBSD 1.[456], 2.x and 3.x for Alpha, x86, and SPARC-based
+ systems
+ NEXTSTEP 3.[13] for NEXTSTEP architectures
+ OpenBSD 2.[89] and 3.[0-9] for x86-based systems
+ OPENSTEP 4.x
+ SCO OpenServer Release 5.0.6 for x86-based systems
+ SCO|Caldera UnixWare 7.1.4 for x86-based systems
+ Solaris 2.6, 8, 9 and 10
+ Tru64 UNIX 5.1
+
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 48
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ (Note: lsof for AIX 5L and above needs setuid-root permis-
+ sion if its -X option is used.)
+
+ Lsof for these dialects does not support a device cache, so
+ the permissions given to the executable don't apply to the
+ device cache file.
+
+ Linux
+
+DEVICE CACHE FILE PATH FROM THE -D OPTION
+ The -D option provides limited means for specifying the dev-
+ ice cache file path. Its ? function will report the
+ read-only and write device cache file paths that lsof will
+ use.
+
+ When the -D b, r, and u functions are available, you can use
+ them to request that the cache file be built in a specific
+ location (b[path]); read but not rebuilt (r[path]); or read
+ and rebuilt (u[path]). The b, r, and u functions are res-
+ tricted under some conditions. They are restricted when the
+ lsof process is setuid-root. The path specified with the r
+ function is always read-only, even when it is available.
+
+ The b, r, and u functions are also restricted when the lsof
+ process runs setgid and lsof doesn't surrender the setgid
+ permission. (See the LSOF PERMISSIONS THAT AFFECT DEVICE
+ CACHE FILE ACCESS section for a list of implementations that
+ normally don't surrender their setgid permission.)
+
+ A further -D function, i (for ignore), is always available.
+
+ When available, the b function tells lsof to read device
+ information from the kernel with the stat(2) function and
+ build a device cache file at the indicated path.
+
+ When available, the r function tells lsof to read the device
+ cache file, but not update it. When a path argument accom-
+ panies -Dr, it names the device cache file path. The r
+ function is always available when it is specified without a
+ path name argument. If lsof is not running setuid-root and
+ surrenders its setgid permission, a path name argument may
+ accompany the r function.
+
+ When available, the u function tells lsof to attempt to read
+ and use the device cache file. If it can't read the file,
+ or if it finds the contents of the file incorrect or out-
+ dated, it will read information from the kernel, and attempt
+ to write an updated version of the device cache file, but
+ only to a path it considers legitimate for the lsof process
+ effective and real UIDs.
+
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 49
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+DEVICE CACHE PATH FROM AN ENVIRONMENT VARIABLE
+ Lsof's second choice for the device cache file is the con-
+ tents of the LSOFDEVCACHE environment variable. It avoids
+ this choice if the lsof process is setuid-root, or the real
+ UID of the process is root.
+
+ A further restriction applies to a device cache file path
+ taken from the LSOFDEVCACHE environment variable: lsof will
+ not write a device cache file to the path if the lsof pro-
+ cess doesn't surrender its setgid permission. (See the LSOF
+ PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS section for
+ information on implementations that don't surrender their
+ setgid permission.)
+
+ The local system administrator can disable the use of the
+ LSOFDEVCACHE environment variable or change its name when
+ building lsof. Consult the output of -D? for the environ-
+ ment variable's name.
+
+SYSTEM-WIDE DEVICE CACHE PATH
+ The local system administrator may choose to have a
+ system-wide device cache file when building lsof. That file
+ will generally be constructed by a special system adminis-
+ tration procedure when the system is booted or when the con-
+ tents of /dev or /devices) changes. If defined, it is
+ lsof's third device cache file path choice.
+
+ You can tell that a system-wide device cache file is in
+ effect for your local installation by examining the lsof
+ help option output - i.e., the output from the -h or -?
+ option.
+
+ Lsof will never write to the system-wide device cache file
+ path by default. It must be explicitly named with a -D
+ function in a root-owned procedure. Once the file has been
+ written, the procedure must change its permission modes to
+ 0644 (owner-read and owner-write, group-read, and
+ other-read).
+
+PERSONAL DEVICE CACHE PATH (DEFAULT)
+ The default device cache file path of the lsof distribution
+ is one recorded in the home directory of the real UID that
+ executes lsof. Added to the home directory is a second path
+ component of the form .lsof_hostname.
+
+ This is lsof's fourth device cache file path choice, and is
+ usually the default. If a system-wide device cache file
+ path was defined when lsof was built, this fourth choice
+ will be applied when lsof can't find the system-wide device
+ cache file. This is the only time lsof uses two paths when
+ reading the device cache file.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 50
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ The hostname part of the second component is the base name
+ of the executing host, as returned by gethostname(2). The
+ base name is defined to be the characters preceding the
+ first `.' in the gethostname(2) output, or all the gethost-
+ name(2) output if it contains no `.'.
+
+ The device cache file belongs to the user ID and is readable
+ and writable by the user ID alone - i.e., its modes are
+ 0600. Each distinct real user ID on a given host that exe-
+ cutes lsof has a distinct device cache file. The hostname
+ part of the path distinguishes device cache files in an
+ NFS-mounted home directory into which device cache files are
+ written from several different hosts.
+
+ The personal device cache file path formed by this method
+ represents a device cache file that lsof will attempt to
+ read, and will attempt to write should it not exist or
+ should its contents be incorrect or outdated.
+
+ The -Dr option without a path name argument will inhibit the
+ writing of a new device cache file.
+
+ The -D? option will list the format specification for con-
+ structing the personal device cache file. The conversions
+ used in the format specification are described in the
+ 00DCACHE file of the lsof distribution.
+
+MODIFIED PERSONAL DEVICE CACHE PATH
+ If this option is defined by the local system administrator
+ when lsof is built, the LSOFPERSDCPATH environment variable
+ contents may be used to add a component of the personal dev-
+ ice cache file path.
+
+ The LSOFPERSDCPATH variable contents are inserted in the
+ path at the place marked by the local system administrator
+ with the ``%p'' conversion in the HASPERSDC format specifi-
+ cation of the dialect's machine.h header file. (It's placed
+ right after the home directory in the default lsof distribu-
+ tion.)
+
+ Thus, for example, if LSOFPERSDCPATH contains ``LSOF'', the
+ home directory is ``/Homes/abe'', the host name is
+ ``lsof.itap.purdue.edu'', and the HASPERSDC format is the
+ default (``%h/%p.lsof_%L''), the modified personal device
+ cache file path is:
+
+ /Homes/abe/LSOF/.lsof_vic
+
+ The LSOFPERSDCPATH environment variable is ignored when the
+ lsof process is setuid-root or when the real UID of the pro-
+ cess is root.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 51
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ Lsof will not write to a modified personal device cache file
+ path if the lsof process doesn't surrender setgid permis-
+ sion. (See the LSOF PERMISSIONS THAT AFFECT DEVICE CACHE
+ FILE ACCESS section for a list of implementations that nor-
+ mally don't surrender their setgid permission.)
+
+ If, for example, you want to create a sub-directory of per-
+ sonal device cache file paths by using the LSOFPERSDCPATH
+ environment variable to name it, and lsof doesn't surrender
+ its setgid permission, you will have to allow lsof to create
+ device cache files at the standard personal path and move
+ them to your subdirectory with shell commands.
+
+ The local system administrator may: disable this option when
+ lsof is built; change the name of the environment variable
+ from LSOFPERSDCPATH to something else; change the HASPERSDC
+ format to include the personal path component in another
+ place; or exclude the personal path component entirely.
+ Consult the output of the -D? option for the environment
+ variable's name and the HASPERSDC format specification.
+
+DIAGNOSTICS
+ Errors are identified with messages on the standard error
+ file.
+
+ Lsof returns a one (1) if any error was detected, including
+ the failure to locate command names, file names, Internet
+ addresses or files, login names, NFS files, PIDs, PGIDs, or
+ UIDs it was asked to list. If the -V option is specified,
+ lsof will indicate the search items it failed to list.
+
+ It returns a zero (0) if no errors were detected and if it
+ was able to list some information about all the specified
+ search arguments.
+
+ When lsof cannot open access to /dev (or /devices) or one of
+ its subdirectories, or get information on a file in them
+ with stat(2), it issues a warning message and continues.
+ That lsof will issue warning messages about inaccessible
+ files in /dev (or /devices) is indicated in its help output
+ - requested with the -h or >B -? options - with the mes-
+ sage:
+
+ Inaccessible /dev warnings are enabled.
+
+ The warning message may be suppressed with the -w option.
+ It may also have been suppressed by the system administrator
+ when lsof was compiled by the setting of the WARNDEVACCESS
+ definition. In this case, the output from the help options
+ will include the message:
+
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 52
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ Inaccessible /dev warnings are disabled.
+
+ Inaccessible device warning messages usually disappear after
+ lsof has created a working device cache file.
+
+EXAMPLES
+ For a more extensive set of examples, documented more fully,
+ see the 00QUICKSTART file of the lsof distribution.
+
+ To list all open files, use:
+
+ lsof
+
+ To list all open Internet, x.25 (HP-UX), and UNIX domain
+ files, use:
+
+ lsof -i -U
+
+ To list all open IPv4 network files in use by the process
+ whose PID is 1234, use:
+
+ lsof -i 4 -a -p 1234
+
+ Presuming the UNIX dialect supports IPv6, to list only open
+ IPv6 network files, use:
+
+ lsof -i 6
+
+ To list all files using any protocol on ports 513, 514, or
+ 515 of host wonderland.cc.purdue.edu, use:
+
+ lsof -i @wonderland.cc.purdue.edu:513-515
+
+ To list all files using any protocol on any port of
+ mace.cc.purdue.edu (cc.purdue.edu is the default domain),
+ use:
+
+ lsof -i @mace
+
+ To list all open files for login name ``abe'', or user ID
+ 1234, or process 456, or process 123, or process 789, use:
+
+ lsof -p 456,123,789 -u 1234,abe
+
+ To list all open files on device /dev/hd4, use:
+
+ lsof /dev/hd4
+
+ To find the process that has /u/abe/foo open, use:
+
+ lsof /u/abe/foo
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 53
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ To send a SIGHUP to the processes that have /u/abe/bar open,
+ use:
+
+ kill -HUP `lsof -t /u/abe/bar`
+
+ To find any open file, including an open UNIX domain socket
+ file, with the name /dev/log, use:
+
+ lsof /dev/log
+
+ To find processes with open files on the NFS file system
+ named /nfs/mount/point whose server is inaccessible, and
+ presuming your mount table supplies the device number for
+ /nfs/mount/point, use:
+
+ lsof -b /nfs/mount/point
+
+ To do the preceding search with warning messages suppressed,
+ use:
+
+ lsof -bw /nfs/mount/point
+
+ To ignore the device cache file, use:
+
+ lsof -Di
+
+ To obtain PID and command name field output for each pro-
+ cess, file descriptor, file device number, and file inode
+ number for each file of each process, use:
+
+ lsof -FpcfDi
+
+ To list the files at descriptors 1 and 3 of every process
+ running the lsof command for login ID ``abe'' every 10
+ seconds, use:
+
+ lsof -c lsof -a -d 1 -d 3 -u abe -r10
+
+ To list the current working directory of processes running a
+ command that is exactly four characters long and has an 'o'
+ or 'O' in character three, use this regular expression form
+ of the -c c option:
+
+ lsof -c /^..o.$/i -a -d cwd
+
+ To find an IP version 4 socket file by its associated
+ numeric dot-form address, use:
+
+ lsof -i@128.210.15.17
+
+ To find an IP version 6 socket file (when the UNIX dialect
+ supports IPv6) by its associated numeric colon-form address,
+
+
+
+SunOS 5.9 Last change: Revision-4.82 54
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ use:
+
+ lsof -i@[0:1:2:3:4:5:6:7]
+
+ To find an IP version 6 socket file (when the UNIX dialect
+ supports IPv6) by an associated numeric colon-form address
+ that has a run of zeroes in it - e.g., the loop-back address
+ - use:
+
+ lsof -i@[::1]
+
+ To obtain a repeat mode marker line that contains the
+ current time, use:
+
+ lsof -rm====%T====
+
+ To add spaces to the previous marker line, use:
+
+ lsof -r "m==== %T ===="
+
+BUGS
+ Since lsof reads kernel memory in its search for open files,
+ rapid changes in kernel memory may produce unpredictable
+ results.
+
+ When a file has multiple record locks, the lock status char-
+ acter (following the file descriptor) is derived from a test
+ of the first lock structure, not from any combination of the
+ individual record locks that might be described by multiple
+ lock structures.
+
+ Lsof can't search for files with restrictive access permis-
+ sions by name unless it is installed with root set-UID per-
+ mission. Otherwise it is limited to searching for files to
+ which its user or its set-GID group (if any) has access per-
+ mission.
+
+ The display of the destination address of a raw socket
+ (e.g., for ping) depends on the UNIX operating system. Some
+ dialects store the destination address in the raw socket's
+ protocol control block, some do not.
+
+ Lsof can't always represent Solaris device numbers in the
+ same way that ls(1) does. For example, the major and minor
+ device numbers that the lstat(2) and stat(2) functions
+ report for the directory on which CD-ROM files are mounted
+ (typically /cdrom) are not the same as the ones that it
+ reports for the device on which CD-ROM files are mounted
+ (typically /dev/sr0). (Lsof reports the directory numbers.)
+
+ The support for /proc file systems is available only for BSD
+ and Tru64 UNIX dialects, Linux, and dialects derived from
+
+
+
+SunOS 5.9 Last change: Revision-4.82 55
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ SYSV R4 - e.g., FreeBSD, NetBSD, OpenBSD, Solaris, UnixWare.
+
+ Some /proc file items - device number, inode number, and
+ file size - are unavailable in some dialects. Searching for
+ files in a /proc file system may require that the full path
+ name be specified.
+
+ No text (txt) file descriptors are displayed for Linux
+ processes. All entries for files other than the current
+ working directory, the root directory, and numerical file
+ descriptors are labeled mem descriptors.
+
+ Lsof can't search for Tru64 UNIX named pipes by name,
+ because their kernel implementation of lstat(2) returns an
+ improper device number for a named pipe.
+
+ Lsof can't report fully or correctly on HP-UX 9.01, 10.20,
+ and 11.00 locks because of insufficient access to kernel
+ data or errors in the kernel data. See the lsof FAQ (The
+ FAQ section gives its location.) for details.
+
+ The AIX SMT file type is a fabrication. It's made up for
+ file structures whose type (15) isn't defined in the AIX
+ /usr/include/sys/file.h header file. One way to create such
+ file structures is to run X clients with the DISPLAY vari-
+ able set to ``:0.0''.
+
+ The +|-f[cfgGn] option is not supported under /proc-based
+ Linux lsof, because it doesn't read kernel structures from
+ kernel memory.
+
+ENVIRONMENT
+ Lsof may access these environment variables.
+
+ LANG defines a language locale. See setlo-
+ cale(3) for the names of other variables
+ that can be used in place of LANG - e.g.,
+ LC_ALL, LC_TYPE, etc.
+
+ LSOFDEVCACHE defines the path to a device cache file.
+ See the DEVICE CACHE PATH FROM AN ENVIRON-
+ MENT VARIABLE section for more informa-
+ tion.
+
+ LSOFPERSDCPATH defines the middle component of a modified
+ personal device cache file path. See the
+ MODIFIED PERSONAL DEVICE CACHE PATH sec-
+ tion for more information.
+
+FAQ
+ Frequently-asked questions and their answers (an FAQ) are
+ available in the 00FAQ file of the lsof distribution.
+
+
+
+SunOS 5.9 Last change: Revision-4.82 56
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+ That file is also available via anonymous ftp from
+ lsof.itap.purdue.edu at pub/tools/unix/lsofFAQ. The URL is:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/FAQ
+
+FILES
+ /dev/kmem kernel virtual memory device
+
+ /dev/mem physical memory device
+
+ /dev/swap system paging device
+
+ .lsof_hostname lsof's device cache file (The suffix,
+ hostname, is the first component of the
+ host's name returned by gethostname(2).)
+
+AUTHORS
+ Lsof was written by Victor A. Abell <abe@purdue.edu> of Pur-
+ due University. Many others have contributed to lsof.
+ They're listed in the 00CREDITS file of the lsof distribu-
+ tion.
+
+DISTRIBUTION
+ The latest distribution of lsof is available via anonymous
+ ftp from the host lsof.itap.purdue.edu. You'll find the
+ lsof distribution in the pub/tools/unix/lsof directory.
+
+ You can also use this URL:
+
+ ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof
+
+ Lsof is also mirrored elsewhere. When you access
+ lsof.itap.purdue.edu and change to its pub/tools/unix/lsof
+ directory, you'll be given a list of some mirror sites. The
+ pub/tools/unix/lsof directory also contains a more complete
+ list in its mirrors file. Use mirrors with caution - not
+ all mirrors always have the latest lsof revision.
+
+ Some pre-compiled Lsof executables are available on
+ lsof.itap.purdue.edu, but their use is discouraged - it's
+ better that you build your own from the sources. If you
+ feel you must use a pre-compiled executable, please read the
+ cautions that appear in the README files of the
+ pub/tools/unix/lsof/binaries subdirectories and in the 00*
+ files of the distribution.
+
+ More information on the lsof distribution can be found in
+ its README.lsof_<version> file. If you intend to get the
+ lsof distribution and build it, please read
+ README.lsof_<version> and the other 00* files of the distri-
+ bution before sending questions to the author.
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 57
+
+
+
+
+
+
+Maintenance Procedures LSOF(8)
+
+
+
+SEE ALSO
+ Not all the following manual pages may exist in every UNIX
+ dialect to which lsof has been ported.
+
+ access(2), awk(1), crash(1), fattach(3C), ff(1), fstat(8),
+ fuser(1), gethostname(2), isprint(3), kill(1), localtime(3),
+ lstat(2), modload(8), mount(8), netstat(1), ofiles(8L),
+ perl(1), ps(1), readlink(2), setlocale(3), stat(2),
+ strftime(3), time(2), uname(1).
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+SunOS 5.9 Last change: Revision-4.82 58
+
+
+
--- /dev/null
+/*
+ * lsof_field.h - field ID characters for lsof output that can be parsed
+ * (selected with -f or -F)
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * $Id: lsof_fields.h,v 1.11 2006/09/15 18:53:21 abe Exp $
+ */
+
+
+#if !defined(LSOF_FORMAT_H)
+#define LSOF_FORMAT_H 1
+
+/*
+ * Codes for output fields:
+ *
+ * LSOF_FID_* ID character
+ * LSOF_FIX_* ID index
+ * LSOF_FNM_* name
+ *
+ * A field is displayed in the form:
+ * <ID_character><data><field_terminator>
+ *
+ * Output fields are normally terminated with a NL ('\n'), but the field
+ * terminator can be set to NUL with the -0 (zero) option to lsof.
+ *
+ * Field sets -- process-specific information or information specific
+ * to a single file descriptor -- are terminated with NL when the field
+ * terminator is NUL.
+ */
+
+#define LSOF_FID_ACCESS 'a'
+#define LSOF_FIX_ACCESS 0
+#define LSOF_FNM_ACCESS "access: r = read; w = write; u = read/write"
+
+#define LSOF_FID_CMD 'c'
+#define LSOF_FIX_CMD 1
+#define LSOF_FNM_CMD "command name"
+
+#define LSOF_FID_CT 'C'
+#define LSOF_FIX_CT 2
+#define LSOF_FNM_CT "file struct share count"
+
+#define LSOF_FID_DEVCH 'd'
+#define LSOF_FIX_DEVCH 3
+#define LSOF_FNM_DEVCH "device character code"
+
+#define LSOF_FID_DEVN 'D'
+#define LSOF_FIX_DEVN 4
+#define LSOF_FNM_DEVN "major/minor device number as 0x<hex>"
+
+#define LSOF_FID_FD 'f'
+#define LSOF_FIX_FD 5
+#define LSOF_FNM_FD "file descriptor"
+
+#define LSOF_FID_FA 'F'
+#define LSOF_FIX_FA 6
+#define LSOF_FNM_FA "file struct address as 0x<hex>"
+
+#define LSOF_FID_FG 'G'
+#define LSOF_FIX_FG 7
+#define LSOF_FNM_FG "file flaGs"
+
+#define LSOF_FID_INODE 'i'
+#define LSOF_FIX_INODE 8
+#define LSOF_FNM_INODE "inode number"
+
+#define LSOF_FID_NLINK 'k'
+#define LSOF_FIX_NLINK 9
+#define LSOF_FNM_NLINK "link count"
+
+#define LSOF_FID_LOCK 'l'
+#define LSOF_FIX_LOCK 10
+#define LSOF_FNM_LOCK "lock: r/R = read; w/W = write; u = read/write"
+
+#define LSOF_FID_LOGIN 'L'
+#define LSOF_FIX_LOGIN 11
+#define LSOF_FNM_LOGIN "login name"
+
+#define LSOF_FID_MARK 'm'
+#define LSOF_FIX_MARK 12
+#define LSOF_FNM_MARK "marker between repeated output"
+
+#define LSOF_FID_NAME 'n'
+#define LSOF_FIX_NAME 13
+#define LSOF_FNM_NAME "comment, name, Internet addresses"
+
+#define LSOF_FID_NI 'N'
+#define LSOF_FIX_NI 14
+#define LSOF_FNM_NI "file struct node ID as 0x<hex>"
+
+#define LSOF_FID_OFFSET 'o'
+#define LSOF_FIX_OFFSET 15
+#define LSOF_FNM_OFFSET "file offset as 0t<dec> or 0x<hex>"
+
+#define LSOF_FID_PID 'p'
+#define LSOF_FIX_PID 16
+#define LSOF_FNM_PID "process ID (PID)"
+
+#define LSOF_FID_PGID 'g'
+#define LSOF_FIX_PGID 17
+#define LSOF_FNM_PGID "process group ID (PGID)"
+
+#define LSOF_FID_PROTO 'P'
+#define LSOF_FIX_PROTO 18
+#define LSOF_FNM_PROTO "protocol name"
+
+#define LSOF_FID_RDEV 'r'
+#define LSOF_FIX_RDEV 19
+#define LSOF_FNM_RDEV "raw device number as 0x<hex>"
+
+#define LSOF_FID_PPID 'R'
+#define LSOF_FIX_PPID 20
+#define LSOF_FNM_PPID "paRent PID"
+
+#define LSOF_FID_SIZE 's'
+#define LSOF_FIX_SIZE 21
+#define LSOF_FNM_SIZE "file size"
+
+#define LSOF_FID_STREAM 'S'
+#define LSOF_FIX_STREAM 22
+#define LSOF_FNM_STREAM "stream module and device names"
+
+#define LSOF_FID_TYPE 't'
+#define LSOF_FIX_TYPE 23
+#define LSOF_FNM_TYPE "file type"
+
+#define LSOF_FID_TCPTPI 'T'
+#define LSOF_FIX_TCPTPI 24
+#define LSOF_FNM_TCPTPI "TCP/TPI info"
+
+#define LSOF_FID_UID 'u'
+#define LSOF_FIX_UID 25
+#define LSOF_FNM_UID "user ID (UID)"
+
+#define LSOF_FID_ZONE 'z'
+#define LSOF_FIX_ZONE 26
+#define LSOF_FNM_ZONE "zone name"
+
+#define LSOF_FID_CNTX 'Z'
+#define LSOF_FIX_CNTX 27
+#define LSOF_FNM_CNTX "security context"
+
+#define LSOF_FID_TERM '0'
+#define LSOF_FIX_TERM 28
+#define LSOF_FNM_TERM "(zero) use NUL field terminator instead of NL"
+
+#endif /* !defined(LSOF_FORMAT_H) */
--- /dev/null
+/*
+ * main.c - common main function for lsof
+ *
+ * V. Abell, Purdue University
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1994 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: main.c,v 1.53 2008/10/21 16:21:41 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Local definitions
+ */
+
+static int GObk[] = { 1, 1 }; /* option backspace values */
+static char GOp; /* option prefix -- '+' or '-' */
+static char *GOv = (char *)NULL; /* option `:' value pointer */
+static int GOx1 = 1; /* first opt[][] index */
+static int GOx2 = 0; /* second opt[][] index */
+
+
+_PROTOTYPE(static int GetOpt,(int ct, char *opt[], char *rules, int *err));
+_PROTOTYPE(static char *sv_fmt_str,(char *f));
+
+
+/*
+ * main() - main function for lsof
+ */
+
+int
+main(argc, argv)
+ int argc;
+ char *argv[];
+{
+ int ad, c, i, n, rv, se1, se2, ss;
+ char *cp;
+ int err = 0;
+ int ev = 0;
+ int fh = 0;
+ char *fmtr = (char *)NULL;
+ long l;
+ MALLOC_S len;
+ struct lfile *lf;
+ struct nwad *np, *npn;
+ char options[128];
+ int rc = 0;
+ struct stat sb;
+ struct sfile *sfp;
+ struct lproc **slp = (struct lproc **)NULL;
+ int sp = 0;
+ struct str_lst *str, *strt;
+ int version = 0;
+ int xover = 0;
+
+#if defined(HAS_STRFTIME)
+ char *fmt = (char *)NULL;
+ size_t fmtl;
+#endif /* defined(HAS_STRFTIME) */
+
+#if defined(HASZONES)
+ znhash_t *zp;
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+/*
+ * This stanza must be immediately before the "Save progam name." code, since
+ * it contains code itself.
+ */
+ cntxlist_t *cntxp;
+
+ CntxStatus = is_selinux_enabled() ? 1 : 0;
+#endif /* defined(HASSELINUX) */
+
+/*
+ * Save program name.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+/*
+ * Close all file descriptors above 2.
+ *
+ * Make sure stderr, stdout, and stdin are open descriptors. Open /dev/null
+ * for ones that aren't. Be terse.
+ *
+ * Make sure umask allows lsof to define its own file permissions.
+ */
+ for (i = 3, n = GET_MAX_FD(); i < n; i++)
+ (void) close(i);
+ while (((i = open("/dev/null", O_RDWR, 0)) >= 0) && (i < 2))
+ ;
+ if (i < 0)
+ Exit(1);
+ if (i > 2)
+ (void) close(i);
+ (void) umask(0);
+
+#if defined(HASSETLOCALE)
+/*
+ * Set locale to environment's definition.
+ */
+ (void) setlocale(LC_CTYPE, "");
+#endif /* defined(HASSETLOCALE) */
+
+/*
+ * Common initialization.
+ */
+ Mypid = getpid();
+ if ((Mygid = (gid_t)getgid()) != getegid())
+ Setgid = 1;
+ Euid = geteuid();
+ if ((Myuid = (uid_t)getuid()) && !Euid)
+ Setuidroot = 1;
+ if (!(Namech = (char *)malloc(MAXPATHLEN + 1))) {
+ (void) fprintf(stderr, "%s: no space for name buffer\n", Pn);
+ Exit(1);
+ }
+ Namechl = (size_t)(MAXPATHLEN + 1);
+/*
+ * Create option mask.
+ */
+ (void) snpf(options, sizeof(options),
+ "?a%sbc:D:d:%sf:F:g:hi:%slL:%sMnNo:Op:Pr:%ss:S:tT:u:UvVwx:%s%s%s",
+
+#if defined(HAS_AFS) && defined(HASAOPT)
+ "A:",
+#else /* !defined(HAS_AFS) || !defined(HASAOPT) */
+ "",
+#endif /* defined(HAS_AFS) && defined(HASAOPT) */
+
+#if defined(HASNCACHE)
+ "C",
+#else /* !defined(HASNCACHE) */
+ "",
+#endif /* defined(HASNCACHE) */
+
+#if defined(HASKOPT)
+ "k:",
+#else /* !defined(HASKOPT) */
+ "",
+#endif /* defined(HASKOPT) */
+
+#if defined(HASMOPT) || defined(HASMNTSUP)
+ "m:",
+#else /* !defined(HASMOPT) && !defined(HASMNTSUP) */
+ "",
+#endif /* defined(HASMOPT) || defined(HASMNTSUP) */
+
+#if defined(HASPPID)
+ "R",
+#else /* !defined(HASPPID) */
+ "",
+#endif /* defined(HASPPID) */
+
+#if defined(HASXOPT)
+# if defined(HASXOPT_ROOT)
+ (Myuid == 0) ? "X" : "",
+# else /* !defined(HASXOPT_ROOT) */
+ "X",
+# endif /* defined(HASXOPT_ROOT) */
+#else /* !defined(HASXOPT) */
+ "",
+#endif /* defined(HASXOPT) */
+
+#if defined(HASZONES)
+ "z:",
+#else /* !defined(HASZONES) */
+ "",
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+ "Z:"
+#else /* !defined(HASSELINUX) */
+ ""
+#endif /* defined(HASSELINUX) */
+
+ );
+/*
+ * Loop through options.
+ */
+ while ((c = GetOpt(argc, argv, options, &rv)) != EOF) {
+ if (rv) {
+ err = 1;
+ continue;
+ }
+ switch (c) {
+ case 'a':
+ Fand = 1;
+ break;
+
+#if defined(HAS_AFS) && defined(HASAOPT)
+ case 'A':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ (void) fprintf(stderr, "%s: -A not followed by path\n", Pn);
+ err = 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ } else
+ AFSApath = GOv;
+ break;
+#endif /* defined(HAS_AFS) && defined(HASAOPT) */
+
+ case 'b':
+ Fblock = 1;
+ break;
+ case 'c':
+ if (GOp == '+') {
+ if (!GOv || (*GOv == '-') || (*GOv == '+')
+ || !isdigit((int)*GOv))
+ {
+ (void) fprintf(stderr,
+ "%s: +c not followed by width number\n", Pn);
+ err = 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ } else {
+ CmdLim = atoi(GOv);
+
+#if defined(MAXSYSCMDL)
+ if (CmdLim > MAXSYSCMDL) {
+ (void) fprintf(stderr,
+ "%s: +c %d > what system provides (%d)\n",
+ Pn, CmdLim, MAXSYSCMDL);
+ err = 1;
+ }
+#endif /* defined(MAXSYSCMDL) */
+
+ }
+ break;
+ }
+ if (GOv && (*GOv == '/')) {
+ if (enter_cmd_rx(GOv))
+ err = 1;
+ } else {
+ if (enter_str_lst("-c", GOv, &Cmdl, &Cmdni, &Cmdnx))
+ err = 1;
+
+#if defined(MAXSYSCMDL)
+ else if (Cmdl->len > MAXSYSCMDL) {
+ (void) fprintf(stderr, "%s: \"-c ", Pn);
+ (void) safestrprt(Cmdl->str, stderr, 2);
+ (void) fprintf(stderr, "\" length (%d) > what system",
+ Cmdl->len);
+ (void) fprintf(stderr, " provides (%d)\n",
+ MAXSYSCMDL);
+ Cmdl->len = 0; /* (to avoid later error report) */
+ err = 1;
+ }
+#endif /* defined(MAXSYSCMDL) */
+
+ }
+ break;
+
+#if defined(HASNCACHE)
+ case 'C':
+ Fncache = (GOp == '-') ? 0 : 1;
+ break;
+#endif /* defined(HASNCACHE) */
+
+ case 'd':
+ if (GOp == '+') {
+ if (enter_dir(GOv, 0))
+ err = 1;
+ else {
+ Selflags |= SELNM;
+ xover = 1;
+ }
+ } else {
+ if (enter_fd(GOv))
+ err = 1;
+ }
+ break;
+ case 'D':
+ if (GOp == '+') {
+ if (enter_dir(GOv, 1))
+ err = 1;
+ else {
+ Selflags |= SELNM;
+ xover = 1;
+ }
+ } else {
+
+#if defined(HASDCACHE)
+ if (ctrl_dcache(GOv))
+ err = 1;
+#else /* !defined(HASDCACHE) */
+ (void) fprintf(stderr, "%s: unsupported option: -D\n", Pn);
+ err = 1;
+#endif /* defined(HASDCACHE) */
+
+ }
+ break;
+ case 'f':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ Ffilesys = (GOp == '+') ? 2 : 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ }
+
+#if defined(HASFSTRUCT)
+ for (; *GOv; GOv++) {
+ switch (*GOv) {
+
+# if !defined(HASNOFSCOUNT)
+ case 'c':
+ case 'C':
+ if (GOp == '+') {
+ Fsv |= FSV_CT;
+ FsvByf = 1;
+ } else
+ Fsv &= (unsigned char)~FSV_CT;
+ break;
+# endif /* !defined(HASNOFSCOUNT) */
+
+# if !defined(HASNOFSADDR)
+ case 'f':
+ case 'F':
+ if (GOp == '+') {
+ Fsv |= FSV_FA;
+ FsvByf = 1;
+ } else
+ Fsv &= (unsigned char)~FSV_FA;
+ break;
+# endif /* !defined(HASNOFSADDR) */
+
+# if !defined(HASNOFSFLAGS)
+ case 'g':
+ case 'G':
+ if (GOp == '+') {
+ Fsv |= FSV_FG;
+ FsvByf = 1;
+ } else
+ Fsv &= (unsigned char)~FSV_FG;
+ FsvFlagX = (*GOv == 'G') ? 1 : 0;
+ break;
+# endif /* !defined(HASNOFSFLAGS) */
+
+# if !defined(HASNOFSNADDR)
+ case 'n':
+ case 'N':
+ if (GOp == '+') {
+ Fsv |= FSV_NI;
+ FsvByf = 1;
+ } else
+ Fsv &= (unsigned char)~FSV_NI;
+ break;
+# endif /* !defined(HASNOFSNADDR */
+
+ default:
+ (void) fprintf(stderr,
+ "%s: unknown file struct option: %c\n", Pn, *GOv);
+ err++;
+ }
+ }
+#else /* !defined(HASFSTRUCT) */
+ (void) fprintf(stderr,
+ "%s: unknown string for %cf: %s\n", Pn, GOp, GOv);
+ err++;
+#endif /* defined(HASFSTRUCT) */
+
+ break;
+ case 'F':
+ if (!GOv || *GOv == '-' || *GOv == '+'
+ || strcmp(GOv, "0") == 0) {
+ if (GOv) {
+ if (*GOv == '-' || *GOv == '+') {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ } else if (*GOv == '0')
+ Terminator = '\0';
+ }
+ for (i = 0; FieldSel[i].nm; i++) {
+
+#if !defined(HASPPID)
+ if (FieldSel[i].id == LSOF_FID_PPID)
+ continue;
+#endif /* !defined(HASPPID) */
+
+#if !defined(HASFSTRUCT)
+ if (FieldSel[i].id == LSOF_FID_CT
+ || FieldSel[i].id == LSOF_FID_FA
+ || FieldSel[i].id == LSOF_FID_FG
+ || FieldSel[i].id == LSOF_FID_NI)
+ continue;
+#endif /* !defined(HASFSTRUCT) */
+
+#if !defined(HASZONES)
+ if (FieldSel[i].id == LSOF_FID_ZONE)
+ continue;
+#endif /* !defined(HASZONES) */
+
+#if defined(HASSELINUX)
+ if ((FieldSel[i].id == LSOF_FID_CNTX) && !CntxStatus)
+ continue;
+#else /* !defined(HASSELINUX) */
+ if (FieldSel[i].id == LSOF_FID_CNTX)
+ continue;
+#endif /* !defined(HASSELINUX) */
+
+ if (FieldSel[i].id == LSOF_FID_RDEV)
+ continue; /* for compatibility */
+ FieldSel[i].st = 1;
+ if (FieldSel[i].opt && FieldSel[i].ov)
+ *(FieldSel[i].opt) |= FieldSel[i].ov;
+ }
+
+#if defined(HASFSTRUCT)
+ Ffield = FsvFlagX = 1;
+#else /* !defined(HASFSTRUCT) */
+ Ffield = 1;
+#endif /* defined(HASFSTRUCT) */
+
+ break;
+ }
+ if (strcmp(GOv, "?") == 0) {
+ fh = 1;
+ break;
+ }
+ for (; *GOv; GOv++) {
+ for (i = 0; FieldSel[i].nm; i++) {
+
+#if !defined(HASPPID)
+ if (FieldSel[i].id == LSOF_FID_PPID)
+ continue;
+#endif /* !defined(HASPPID) */
+
+#if !defined(HASFSTRUCT)
+ if (FieldSel[i].id == LSOF_FID_CT
+ || FieldSel[i].id == LSOF_FID_FA
+ || FieldSel[i].id == LSOF_FID_FG
+ || FieldSel[i].id == LSOF_FID_NI)
+ continue;
+#endif /* !defined(HASFSTRUCT) */
+
+ if (FieldSel[i].id == *GOv) {
+ FieldSel[i].st = 1;
+ if (FieldSel[i].opt && FieldSel[i].ov)
+ *(FieldSel[i].opt) |= FieldSel[i].ov;
+
+#if defined(HASFSTRUCT)
+ if (i == LSOF_FIX_FG)
+ FsvFlagX = 1;
+#endif /* defined(HASFSTRUCT) */
+
+ if (i == LSOF_FIX_TERM)
+ Terminator = '\0';
+ break;
+ }
+ }
+ if ( ! FieldSel[i].nm) {
+ (void) fprintf(stderr,
+ "%s: unknown field: %c\n", Pn, *GOv);
+ err++;
+ }
+ }
+ Ffield = 1;
+ break;
+ case 'g':
+ if (GOv) {
+ if (*GOv == '-' || *GOv == '+') {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ } else if (enter_id(PGID, GOv))
+ err = 1;
+ }
+ Fpgid = 1;
+ break;
+ case 'h':
+ case '?':
+ Fhelp = 1;
+ break;
+ case 'i':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ Fnet = 1;
+ FnetTy = 0;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ }
+ if (enter_network_address(GOv))
+ err = 1;
+ break;
+
+#if defined(HASKOPT)
+ case 'k':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ (void) fprintf(stderr, "%s: -k not followed by path\n", Pn);
+ err = 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ } else
+ Nmlst = GOv;
+ break;
+#endif /* defined(HASKOPT) */
+
+ case 'l':
+ Futol = 0;
+ break;
+ case 'L':
+ Fnlink = (GOp == '+') ? 1 : 0;
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ Nlink = 0l;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ }
+ for (cp = GOv, l = 0l, n = 0; *cp; cp++) {
+ if (!isdigit((unsigned char)*cp))
+ break;
+ l = (l * 10l) + ((long)*cp - (long)'0');
+ n++;
+ }
+ if (n) {
+ if (GOp != '+') {
+ (void) fprintf(stderr,
+ "%s: no number may follow -L\n", Pn);
+ err = 1;
+ } else {
+ Nlink = l;
+ Selflags |= SELNLINK;
+ }
+ } else
+ Nlink = 0l;
+ if (*cp) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1] + n;
+ }
+ break;
+
+#if defined(HASMOPT) || defined(HASMNTSUP)
+ case 'm':
+ if (GOp == '-') {
+
+# if defined(HASMOPT)
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ (void) fprintf(stderr,
+ "%s: -m not followed by path\n", Pn);
+ err = 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ } else
+ Memory = GOv;
+# else /* !defined(HASMOPT) */
+ (void) fprintf(stderr, "%s: -m not supported\n", Pn);
+ err = 1;
+# endif /* defined(HASMOPT) */
+
+ } else if (GOp == '+') {
+
+# if defined(HASMNTSUP)
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ MntSup = 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ } else {
+ MntSup = 2;
+ MntSupP = GOv;
+ }
+# else /* !defined(HASMNTSUP) */
+ (void) fprintf(stderr, "%s: +m not supported\n", Pn);
+ err = 1;
+# endif /* defined(HASMNTSUP) */
+
+ } else {
+ (void) fprintf(stderr, "%s: %cm not supported\n", Pn, GOp);
+ err = 1;
+ }
+ break;
+#endif /* defined(HASMOPT) || defined(HASMNTSUP) */
+
+ case 'M':
+ FportMap = (GOp == '+') ? 1 : 0;
+ break;
+ case 'n':
+ Fhost = (GOp == '-') ? 0 : 1;
+ break;
+ case 'N':
+ Fnfs = 1;
+ break;
+ case 'o':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ Foffset = 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ }
+ for (cp = GOv, i = n = 0; *cp; cp++) {
+ if (!isdigit((unsigned char)*cp))
+ break;
+ i = (i * 10) + ((int)*cp - '0');
+ n++;
+ }
+ if (n)
+ OffDecDig = i;
+ else
+ Foffset = 1;
+ if (*cp) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1] + n;
+ }
+ break;
+ case 'O':
+ Fovhd = (GOp == '-') ? 1 : 0;
+ break;
+ case 'p':
+ if (enter_id(PID, GOv))
+ err = 1;
+ break;
+ case 'P':
+ Fport = (GOp == '-') ? 0 : 1;
+ break;
+ case 'r':
+ if (GOp == '+')
+ ev = rc = 1;
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ RptTm = RPTTM;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ }
+ for (cp = GOv, i = n = 0; *cp; cp++) {
+ if (!isdigit((unsigned char)*cp))
+ break;
+ i = (i * 10) + ((int)*cp - '0');
+ n++;
+ }
+ if (n)
+ RptTm = i;
+ else
+ RptTm = RPTTM;
+ if (!*cp)
+ break;
+ while(*cp && (*cp == ' '))
+ cp++;
+ if (*cp != LSOF_FID_MARK) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1] + n;
+ break;
+ }
+
+#if defined(HAS_STRFTIME)
+
+ /*
+ * Collect the strftime(3) format and test it.
+ */
+ cp++;
+ if ((fmtl = strlen(cp) + 1) < 1) {
+ (void) fprintf(stderr, "%s: <fmt> too short: \"%s\"\n",
+ Pn, cp);
+ err = 1;
+ } else {
+ fmt = cp;
+ fmtl = (fmtl * 8) + 1;
+ if (!(fmtr = (char *)malloc((MALLOC_S)fmtl))) {
+ (void) fprintf(stderr,
+ "%s: no space (%d) for <fmt> result: \"%s\"\n",
+ Pn, (int)fmtl, cp);
+ Exit(1);
+ }
+ if (util_strftime(fmtr, fmtl - 1, fmt) < 1) {
+ (void) fprintf(stderr, "%s: illegal <fmt>: \"%s\"\n",
+ Pn, fmt);
+ err = 1;
+ }
+ }
+
+#else /* !defined(HAS_STRFTIME) */
+ (void) fprintf(stderr, "%s: m<fmt> not supported: \"%s\"\n",
+ Pn, cp);
+ err = 1;
+#endif /* defined(HAS_STRFTIME) */
+
+ break;
+
+#if defined(HASPPID)
+ case 'R':
+ Fppid = 1;
+ break;
+#endif /* defined(HASPPID) */
+
+ case 's':
+
+#if defined(HASTCPUDPSTATE)
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ Fsize = 1;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ } else {
+ if (enter_state_spec(GOv))
+ err = 1;
+ }
+#else /* !defined(HASTCPUDPSTATE) */
+ Fsize = 1;
+#endif /* defined(HASTCPUDPSTATE) */
+
+ break;
+ case 'S':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ TmLimit = TMLIMIT;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ }
+ for (cp = GOv, i = n = 0; *cp; cp++) {
+ if (!isdigit((unsigned char)*cp))
+ break;
+ i = (i * 10) + ((int)*cp - '0');
+ n++;
+ }
+ if (n)
+ TmLimit = i;
+ else
+ TmLimit = TMLIMIT;
+ if (*cp) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1] + n;
+ }
+ if (TmLimit < TMLIMMIN) {
+ (void) fprintf(stderr,
+ "%s: WARNING: -S time (%d) changed to %d\n",
+ Pn, TmLimit, TMLIMMIN);
+ TmLimit = TMLIMMIN;
+ }
+ break;
+ case 't':
+ Fterse = Fwarn = 1;
+ break;
+ case 'T':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ Ftcptpi = (GOp == '-') ? 0 : TCPTPI_STATE;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ }
+ for (Ftcptpi = 0; *GOv; GOv++) {
+ switch (*GOv) {
+
+#if defined(HASSOOPT) || defined(HASSOSTATE) || defined(HASTCPOPT)
+ case 'f':
+ Ftcptpi |= TCPTPI_FLAGS;
+ break;
+#endif /* defined(HASSOOPT) || defined(HASSOSTATE) || defined(HASTCPOPT) */
+
+#if defined(HASTCPTPIQ)
+ case 'q':
+ Ftcptpi |= TCPTPI_QUEUES;
+ break;
+#endif /* defined(HASTCPTPIQ) */
+
+ case 's':
+ Ftcptpi |= TCPTPI_STATE;
+ break;
+
+#if defined(HASTCPTPIW)
+ case 'w':
+ Ftcptpi |= TCPTPI_WINDOWS;
+ break;
+#endif /* defined(HASTCPTPIW) */
+
+ default:
+ (void) fprintf(stderr,
+ "%s: unsupported TCP/TPI info selection: %c\n",
+ Pn, *GOv);
+ err = 1;
+ }
+ }
+ break;
+ case 'u':
+ if (enter_uid(GOv))
+ err = 1;
+ break;
+ case 'U':
+ Funix = 1;
+ break;
+ case 'v':
+ version = 1;
+ break;
+ case 'V':
+ Fverbose = 1;
+ break;
+ case 'w':
+ Fwarn = (GOp == '+') ? 0 : 1;
+ break;
+ case 'x':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ Fxover = XO_ALL;
+ if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+ } else {
+ for (; *GOv; GOv++) {
+ switch (*GOv) {
+ case 'f':
+ Fxover |= XO_FILESYS;
+ break;
+ case 'l':
+ Fxover |= XO_SYMLINK;
+ break;
+ default:
+ (void) fprintf(stderr,
+ "%s: unknown cross-over option: %c\n",
+ Pn, *GOv);
+ err++;
+ }
+ }
+ }
+ break;
+
+#if defined(HASXOPT)
+ case 'X':
+ Fxopt = Fxopt ? 0 : 1;
+ break;
+#endif /* defined(HASXOPT) */
+
+#if defined(HASZONES)
+ case 'z':
+ Fzone = 1;
+ if (GOv && (*GOv != '-') && (*GOv != '+')) {
+
+ /*
+ * Add to the zone name argument hash.
+ */
+ if (enter_zone_arg(GOv))
+ err = 1;
+ } else if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ break;
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+ case 'Z':
+ if (!CntxStatus) {
+ (void) fprintf(stderr, "%s: -Z limited to SELinux\n", Pn);
+ err = 1;
+ } else {
+ Fcntx = 1;
+ if (GOv && (*GOv != '-') && (*GOv != '+')) {
+
+ /*
+ * Add to the context name argument hash.
+ */
+ if (enter_cntx_arg(GOv))
+ err = 1;
+ } else if (GOv) {
+ GOx1 = GObk[0];
+ GOx2 = GObk[1];
+ }
+ }
+ break;
+#endif /* defined(HASSELINUX) */
+
+ default:
+ (void) fprintf(stderr, "%s: unknown option (%c)\n", Pn, c);
+ err = 1;
+ }
+ }
+/*
+ * Check for argument consistency.
+ */
+ if (Cmdnx && Cmdni) {
+
+ /*
+ * Check for command inclusion/exclusion conflicts.
+ */
+ for (str = Cmdl; str; str = str->next) {
+ if (str->x) {
+ for (strt = Cmdl; strt; strt = strt->next) {
+ if (!strt->x) {
+ if (!strcmp(str->str, strt->str)) {
+ (void) fprintf(stderr,
+ "%s: -c^%s and -c%s conflict.\n",
+ Pn, str->str, strt->str);
+ err++;
+ }
+ }
+ }
+ }
+ }
+ }
+
+#if defined(HASTCPUDPSTATE)
+ if (TcpStXn && TcpStIn) {
+
+ /*
+ * Check for excluded and included TCP states.
+ */
+ for (i = 0; i < TcpNstates; i++) {
+ if (TcpStX[i] && TcpStI[i]) {
+ (void) fprintf(stderr,
+ "%s: can't include and exclude TCP state: %s\n",
+ Pn, TcpSt[i]);
+ err = 1;
+ }
+ }
+ }
+ if (UdpStXn && UdpStIn) {
+
+ /*
+ * Check for excluded and included UDP states.
+ */
+ for (i = 0; i < UdpNstates; i++) {
+ if (UdpStX[i] && UdpStI[i]) {
+ (void) fprintf(stderr,
+ "%s: can't include and exclude UDP state: %s\n",
+ Pn, UdpSt[i]);
+ err = 1;
+ }
+ }
+ }
+#endif /* defined(HASTCPUDPSTATE) */
+
+ if (Fsize && Foffset) {
+ (void) fprintf(stderr, "%s: -o and -s are mutually exclusive\n",
+ Pn);
+ err++;
+ }
+ if (Ffield) {
+ if (Fterse) {
+ (void) fprintf(stderr,
+ "%s: -F and -t are mutually exclusive\n", Pn);
+ err++;
+ }
+ FieldSel[LSOF_FIX_PID].st = 1;
+
+#if defined(HAS_STRFTIME)
+ if (fmtr) {
+
+ /*
+ * The field output marker format can't contain "%n" new line
+ * requests.
+ */
+ for (cp = strchr(fmt, '%'); cp; cp = strchr(cp, '%')) {
+ if (*++cp == 'n') {
+ (void) fprintf(stderr,
+ "%s: %%n illegal in -r m<fmt> when -F has", Pn);
+ (void) fprintf(stderr,
+ " been specified: \"%s\"\n", fmt);
+ err++;
+ break;
+ } else if (*cp == '%')
+ cp++;
+ }
+ }
+#endif /* defined(HAS_STRFTIME) */
+
+ }
+ if (Fxover && !xover) {
+ (void) fprintf(stderr, "%s: -x must accompany +d or +D\n", Pn);
+ err++;
+ }
+ if (DChelp || err || Fhelp || fh || version)
+ usage(err ? 1 : 0, fh, version);
+/*
+ * Reduce the size of Suid[], if necessary.
+ */
+ if (Suid && Nuid && Nuid < Mxuid) {
+ if (!(Suid = (struct seluid *)realloc((MALLOC_P *)Suid,
+ (MALLOC_S)(sizeof(struct seluid) * Nuid))))
+ {
+ (void) fprintf(stderr, "%s: can't realloc UID table\n", Pn);
+ Exit(1);
+ }
+ Mxuid = Nuid;
+ }
+/*
+ * Compute the selection flags.
+ */
+ if ((Cmdl && Cmdni) || CmdRx)
+ Selflags |= SELCMD;
+
+#if defined(HASSELINUX)
+ if (CntxArg)
+ Selflags |= SELCNTX;
+#endif /* defined(HASSELINUX) */
+
+ if (Fdl)
+ Selflags |= SELFD;
+ if (Fnet)
+ Selflags |= SELNET;
+ if (Fnfs)
+ Selflags |= SELNFS;
+ if (Funix)
+ Selflags |= SELUNX;
+ if (Npgid && Npgidi)
+ Selflags |= SELPGID;
+ if (Npid && Npidi)
+ Selflags |= SELPID;
+ if (Nuid && Nuidincl)
+ Selflags |= SELUID;
+ if (Nwad)
+ Selflags |= SELNA;
+
+#if defined(HASZONES)
+ if (ZoneArg)
+ Selflags |= SELZONE;
+#endif /* defined(HASZONES) */
+
+ if (GOx1 < argc)
+ Selflags |= SELNM;
+ if (Selflags == 0) {
+ if (Fand) {
+ (void) fprintf(stderr,
+ "%s: no select options to AND via -a\n", Pn);
+ usage(1, 0, 0);
+ }
+ Selflags = SELALL;
+ } else {
+ if (GOx1 >= argc && (Selflags & (SELNA|SELNET)) != 0
+ && (Selflags & ~(SELNA|SELNET)) == 0)
+ Selinet = 1;
+ Selall = 0;
+ }
+/*
+ * Get the device for DEVDEV_PATH.
+ */
+ if (stat(DEVDEV_PATH, &sb)) {
+ se1 = errno;
+ if ((ad = strcmp(DEVDEV_PATH, "/dev"))) {
+ if ((ss = stat("/dev", &sb)))
+ se2 = errno;
+ else
+ se2 = 0;
+ } else {
+ se2 = 0;
+ ss = 1;
+ }
+ if (ss) {
+ (void) fprintf(stderr, "%s: can't stat(%s): %s\n", Pn,
+ DEVDEV_PATH, strerror(se1));
+ if (ad) {
+ (void) fprintf(stderr, "%s: can't stat(/dev): %s\n", Pn,
+ strerror(se2));
+ }
+ Exit(1);
+ }
+ }
+ DevDev = sb.st_dev;
+/*
+ * Process the file arguments.
+ */
+ if (GOx1 < argc) {
+ if (ck_file_arg(GOx1, argc, argv, Ffilesys, 0, (struct stat *)NULL))
+ usage(1, 0, 0);
+ }
+/*
+ * Do dialect-specific initialization.
+ */
+ initialize();
+ if (Sfile)
+ (void) hashSfile();
+
+#if defined(WILLDROPGID)
+/*
+ * If this process isn't setuid(root), but it is setgid(not_real_gid),
+ * relinquish the setgid power. (If it hasn't already been done.)
+ */
+ (void) dropgid();
+#endif /* defined(WILLDROPGID) */
+
+
+#if defined(HASDCACHE)
+/*
+ * If there is a device cache, prepare the device table.
+ */
+ if (DCstate)
+ readdev(0);
+#endif /* defined(HASDCACHE) */
+
+/*
+ * Define the size and offset print formats.
+ */
+ (void) snpf(options, sizeof(options), "%%%su", INODEPSPEC);
+ InodeFmt_d = sv_fmt_str(options);
+ (void) snpf(options, sizeof(options), "%%#%sx", INODEPSPEC);
+ InodeFmt_x = sv_fmt_str(options);
+ (void) snpf(options, sizeof(options), "0t%%%su", SZOFFPSPEC);
+ SzOffFmt_0t = sv_fmt_str(options);
+ (void) snpf(options, sizeof(options), "%%%su", SZOFFPSPEC);
+ SzOffFmt_d = sv_fmt_str(options);
+ (void) snpf(options, sizeof(options), "%%*%su", SZOFFPSPEC);
+ SzOffFmt_dv = sv_fmt_str(options);
+ (void) snpf(options, sizeof(options), "%%#%sx", SZOFFPSPEC);
+ SzOffFmt_x = sv_fmt_str(options);
+
+#if defined(HASMNTSUP)
+/*
+ * Report mount supplement information, as requested.
+ */
+ if (MntSup == 1) {
+ (void) readmnt();
+ Exit(0);
+ }
+#endif /* defined(HASMNTSUP) */
+
+/*
+ * Gather and report process information every RptTm seconds.
+ */
+ if (RptTm)
+ CkPasswd = 1;
+ do {
+
+ /*
+ * Gather information about processes.
+ */
+ gather_proc_info();
+ /*
+ * If the local process table has more than one entry, sort it by PID.
+ */
+ if (Nlproc > 1) {
+ if (Nlproc > sp) {
+ len = (MALLOC_S)(Nlproc * sizeof(struct lproc *));
+ sp = Nlproc;
+ if (!slp)
+ slp = (struct lproc **)malloc(len);
+ else
+ slp = (struct lproc **)realloc((MALLOC_P *)slp, len);
+ if (!slp) {
+ (void) fprintf(stderr,
+ "%s: no space for %d sort pointers\n", Pn, Nlproc);
+ Exit(1);
+ }
+ }
+ for (i = 0; i < Nlproc; i++) {
+ slp[i] = &Lproc[i];
+ }
+ (void) qsort((QSORT_P *)slp, (size_t)Nlproc,
+ (size_t)sizeof(struct lproc *), comppid);
+ }
+ if ((n = Nlproc)) {
+
+#if defined(HASNCACHE)
+ /*
+ * If using the kernel name cache, force its reloading.
+ */
+ NcacheReload = 1;
+#endif /* defined(HASNCACHE) */
+
+ /*
+ * Print the selected processes and count them.
+ *
+ * Lf contents must be preserved, since they may point to a
+ * malloc()'d area, and since Lf is used throughout the print
+ * process.
+ */
+ for (lf = Lf, print_init(); PrPass < 2; PrPass++) {
+ for (i = n = 0; i < Nlproc; i++) {
+ Lp = (Nlproc > 1) ? slp[i] : &Lproc[i];
+ if (Lp->pss) {
+ if (print_proc())
+ n++;
+ }
+ if (RptTm && PrPass)
+ (void) free_lproc(Lp);
+ }
+ }
+ Lf = lf;
+ }
+ /*
+ * If a repeat time is set, sleep for the specified time.
+ *
+ * If conditional repeat mode is in effect, see if it's time to exit.
+ */
+ if (RptTm) {
+ if (rc) {
+ if (!n)
+ break;
+ else
+ ev = 0;
+ }
+
+#if defined(HAS_STRFTIME)
+ if (fmt && fmtr) {
+
+ /*
+ * Format the marker line.
+ */
+ (void) util_strftime(fmtr, fmtl - 1, fmt);
+ fmtr[fmtl - 1] = '\0';
+ }
+#endif /* defined(HAS_STRFTIME) */
+
+ if (Ffield) {
+ putchar(LSOF_FID_MARK);
+
+#if defined(HAS_STRFTIME)
+ if (fmtr)
+ (void) printf("%s", fmtr);
+#endif /* defined(HAS_STRFTIME) */
+
+ putchar(Terminator);
+ if (Terminator != '\n')
+ putchar('\n');
+ } else {
+
+#if defined(HAS_STRFTIME)
+ if (fmtr)
+ cp = fmtr;
+ else
+#endif /* defined(HAS_STRFTIME) */
+
+ cp = "=======";
+ puts(cp);
+ }
+ (void) fflush(stdout);
+ (void) childx();
+ (void) sleep(RptTm);
+ Hdr = Nlproc = 0;
+ CkPasswd = 1;
+ }
+ } while (RptTm);
+/*
+ * See if all requested information was displayed. Return zero if it
+ * was; one, if not. If -V was specified, report what was not displayed.
+ */
+ (void) childx();
+ rv = 0;
+ for (str = Cmdl; str; str = str->next) {
+
+ /*
+ * Check command specifications.
+ */
+ if (str->f)
+ continue;
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: command not located: ", Pn);
+ safestrprt(str->str, stdout, 1);
+ }
+ }
+ for (i = 0; i < NCmdRxU; i++) {
+
+ /*
+ * Check command regular expressions.
+ */
+ if (CmdRx[i].mc)
+ continue;
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: no command found for regex: ", Pn);
+ safestrprt(CmdRx[i].exp, stdout, 1);
+ }
+ }
+ for (sfp = Sfile; sfp; sfp = sfp->next) {
+
+ /*
+ * Check file specifications.
+ */
+ if (sfp->f)
+ continue;
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: no file%s use located: ", Pn,
+ sfp->type ? "" : " system");
+ safestrprt(sfp->aname, stdout, 1);
+ }
+ }
+
+#if defined(HASPROCFS)
+ /*
+ * Report on proc file system search results.
+ */
+ if (Procsrch && !Procfind) {
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: no file system use located: ", Pn);
+ safestrprt(Mtprocfs ? Mtprocfs->dir : HASPROCFS, stdout, 1);
+ }
+ }
+ {
+ struct procfsid *pfi;
+
+ for (pfi = Procfsid; pfi; pfi = pfi->next) {
+ if (!pfi->f) {
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: no file use located: ", Pn);
+ safestrprt(pfi->nm, stdout, 1);
+ }
+ }
+ }
+ }
+#endif /* defined(HASPROCFS) */
+
+ if ((np = Nwad)) {
+
+ /*
+ * Check Internet address specifications.
+ *
+ * If any Internet address derived from the same argument was found,
+ * consider all derivations found. If no derivation from the same
+ * argument was found, report only the first failure.
+ *
+ */
+ for (; np; np = np->next) {
+ if (!(cp = np->arg))
+ continue;
+ for (npn = np->next; npn; npn = npn->next) {
+ if (!npn->arg)
+ continue;
+ if (!strcmp(cp, npn->arg)) {
+
+ /*
+ * If either of the duplicate specifications was found,
+ * mark them both found. If neither was found, mark all
+ * but the first one found.
+ */
+ if (np->f)
+ npn->f = np->f;
+ else if (npn->f)
+ np->f = npn->f;
+ else
+ npn->f = 1;
+ }
+ }
+ }
+ for (np = Nwad; np; np = np->next) {
+ if (!np->f && (cp = np->arg)) {
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: Internet address not located: ", Pn);
+ safestrprt(cp ? cp : "(unknown)", stdout, 1);
+ }
+ }
+ }
+ }
+ if (Fnet && Fnet < 2) {
+
+ /*
+ * Report no Internet files located.
+ */
+ rv = 1;
+ if (Fverbose)
+ (void) printf("%s: no Internet files located\n", Pn);
+ }
+
+#if defined(HASTCPUDPSTATE)
+ if (TcpStIn) {
+
+ /*
+ * Check for included TCP states not located.
+ */
+ for (i = 0; i < TcpNstates; i++) {
+ if (TcpStI[i] == 1) {
+ rv = 1;
+ if (Fverbose)
+ (void) printf("%s: TCP state not located: %s\n",
+ Pn, TcpSt[i]);
+ }
+ }
+ }
+ if (UdpStIn) {
+
+ /*
+ * Check for included UDP states not located.
+ */
+ for (i = 0; i < UdpNstates; i++) {
+ if (UdpStI[i] == 1) {
+ rv = 1;
+ if (Fverbose)
+ (void) printf("%s: UDP state not located: %s\n",
+ Pn, UdpSt[i]);
+ }
+ }
+ }
+#endif /* defined(HASTCPUDPSTATE) */
+
+ if (Fnfs && Fnfs < 2) {
+
+ /*
+ * Report no NFS files located.
+ */
+ rv = 1;
+ if (Fverbose)
+ (void) printf("%s: no NFS files located\n", Pn);
+ }
+ for (i = 0; i < Npid; i++) {
+
+ /*
+ * Check inclusionary process ID specifications.
+ */
+ if (Spid[i].f || Spid[i].x)
+ continue;
+ rv = 1;
+ if (Fverbose)
+ (void) printf("%s: process ID not located: %d\n",
+ Pn, Spid[i].i);
+ }
+
+#if defined(HASZONES)
+ if (ZoneArg) {
+
+ /*
+ * Check zone argument results.
+ */
+ for (i = 0; i < HASHZONE; i++) {
+ for (zp = ZoneArg[i]; zp; zp = zp->next) {
+ if (!zp->f) {
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: zone not located: ", Pn);
+ safestrprt(zp->zn, stdout, 1);
+ }
+ }
+ }
+ }
+ }
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+ if (CntxArg) {
+
+ /*
+ * Check context argument results.
+ */
+ for (cntxp = CntxArg; cntxp; cntxp = cntxp->next) {
+ if (!cntxp->f) {
+ rv = 1;
+ if (Fverbose) {
+ (void) printf("%s: context not located: ", Pn);
+ safestrprt(cntxp->cntx, stdout, 1);
+ }
+ }
+ }
+ }
+#endif /* defined(HASSELINUX) */
+
+ for (i = 0; i < Npgid; i++) {
+
+ /*
+ * Check inclusionary process group ID specifications.
+ */
+ if (Spgid[i].f || Spgid[i].x)
+ continue;
+ rv = 1;
+ if (Fverbose)
+ (void) printf("%s: process group ID not located: %d\n",
+ Pn, Spgid[i].i);
+ }
+ for (i = 0; i < Nuid; i++) {
+
+ /*
+ * Check inclusionary user ID specifications.
+ */
+ if (Suid[i].excl || Suid[i].f)
+ continue;
+ rv = 1;
+ if (Fverbose) {
+ if (Suid[i].lnm) {
+ (void) printf("%s: login name (UID %lu) not located: ",
+ Pn, (unsigned long)Suid[i].uid);
+ safestrprt(Suid[i].lnm, stdout, 1);
+ } else
+ (void) printf("%s: user ID not located: %lu\n", Pn,
+ (unsigned long)Suid[i].uid);
+ }
+ }
+ if (!rv && rc)
+ rv = ev;
+ if (!rv && ErrStat)
+ rv = 1;
+ Exit(rv);
+ return(rv); /* to make code analyzers happy */
+}
+
+
+/*
+ * GetOpt() -- Local get option
+ *
+ * Liberally adapted from the public domain AT&T getopt() source,
+ * distributed at the 1985 UNIFORM conference in Dallas
+ *
+ * The modifications allow `?' to be an option character and allow
+ * the caller to decide that an option that may be followed by a
+ * value doesn't have one -- e.g., has a default instead.
+ */
+
+static int
+GetOpt(ct, opt, rules, err)
+ int ct; /* option count */
+ char *opt[]; /* options */
+ char *rules; /* option rules */
+ int *err; /* error return */
+{
+ register int c;
+ register char *cp = (char *)NULL;
+
+ if (GOx2 == 0) {
+
+ /*
+ * Move to a new entry of the option array.
+ *
+ * EOF if:
+ *
+ * Option list has been exhausted;
+ * Next option doesn't start with `-' or `+';
+ * Next option has nothing but `-' or `+';
+ * Next option is ``--'' or ``++''.
+ */
+ if (GOx1 >= ct
+ || (opt[GOx1][0] != '-' && opt[GOx1][0] != '+')
+ || !opt[GOx1][1])
+ return(EOF);
+ if (strcmp(opt[GOx1], "--") == 0 || strcmp(opt[GOx1], "++") == 0) {
+ GOx1++;
+ return(EOF);
+ }
+ GOp = opt[GOx1][0];
+ GOx2 = 1;
+ }
+/*
+ * Flag `:' option character as an error.
+ *
+ * Check for a rule on this option character.
+ */
+ *err = 0;
+ if ((c = opt[GOx1][GOx2]) == ':') {
+ (void) fprintf(stderr,
+ "%s: colon is an illegal option character.\n", Pn);
+ *err = 1;
+ } else if (!(cp = strchr(rules, c))) {
+ (void) fprintf(stderr, "%s: illegal option character: %c\n", Pn, c);
+ *err = 2;
+ }
+ if (*err) {
+
+ /*
+ * An error was detected.
+ *
+ * Advance to the next option character.
+ *
+ * Return the character causing the error.
+ */
+ if (opt[GOx1][++GOx2] == '\0') {
+ GOx1++;
+ GOx2 = 0;
+ }
+ return(c);
+ }
+ if (*(cp + 1) == ':') {
+
+ /*
+ * The option may have a following value. The caller decides
+ * if it does.
+ *
+ * Save the position of the possible value in case the caller
+ * decides it does not belong to the option and wants it
+ * reconsidered as an option character. The caller does that
+ * with:
+ * GOx1 = GObk[0]; GOx2 = GObk[1];
+ *
+ * Don't indicate that an option of ``--'' is a possible value.
+ *
+ * Finally, on the assumption that the caller will decide that
+ * the possible value belongs to the option, position to the
+ * option following the possible value, so that the next call
+ * to GetOpt() will find it.
+ */
+ if(opt[GOx1][GOx2 + 1] != '\0') {
+ GObk[0] = GOx1;
+ GObk[1] = ++GOx2;
+ GOv = &opt[GOx1++][GOx2];
+ } else if (++GOx1 >= ct)
+ GOv = (char *)NULL;
+ else {
+ GObk[0] = GOx1;
+ GObk[1] = 0;
+ GOv = opt[GOx1];
+ if (strcmp(GOv, "--") == 0)
+ GOv = (char *)NULL;
+ else
+ GOx1++;
+ }
+ GOx2 = 0;
+ } else {
+
+ /*
+ * The option character stands alone with no following value.
+ *
+ * Advance to the next option character.
+ */
+ if (opt[GOx1][++GOx2] == '\0') {
+ GOx2 = 0;
+ GOx1++;
+ }
+ GOv = (char *)NULL;
+ }
+/*
+ * Return the option character.
+ */
+ return(c);
+}
+
+
+/*
+ * sv_fmt_str() - save format string
+ */
+
+static char *
+sv_fmt_str(f)
+ char *f; /* format string */
+{
+ char *cp;
+ MALLOC_S l;
+
+ l = (MALLOC_S)(strlen(f) + 1);
+ if (!(cp = (char *)malloc(l))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for format: %s\n", Pn, (int)l, f);
+ Exit(1);
+ }
+ (void) snpf(cp, l, "%s", f);
+ return(cp);
+}
--- /dev/null
+/*
+ * misc.c - common miscellaneous functions for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1994 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: misc.c,v 1.26 2008/10/21 16:21:41 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+#if defined(HASWIDECHAR) && defined(WIDECHARINCL)
+#include WIDECHARINCL
+#endif /* defined(HASWIDECHAR) && defined(WIDECHARINCL) */
+
+
+/*
+ * Local definitions
+ */
+
+#if !defined(MAXSYMLINKS)
+#define MAXSYMLINKS 32
+#endif /* !defined(MAXSYMLINKS) */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void closePipes,(void));
+_PROTOTYPE(static int dolstat,(char *path, char *buf, int len));
+_PROTOTYPE(static int dostat,(char *path, char *buf, int len));
+_PROTOTYPE(static int doreadlink,(char *path, char *buf, int len));
+_PROTOTYPE(static int doinchild,(int (*fn)(), char *fp, char *rbuf, int rbln));
+
+#if defined(HASINTSIGNAL)
+_PROTOTYPE(static int handleint,(int sig));
+#else /* !defined(HASINTSIGNAL) */
+_PROTOTYPE(static void handleint,(int sig));
+#endif /* defined(HASINTSIGNAL) */
+
+_PROTOTYPE(static char *safepup,(unsigned int c, int *cl));
+
+
+/*
+ * Local variables
+ */
+
+static pid_t Cpid = 0; /* child PID */
+static jmp_buf Jmp_buf; /* jump buffer */
+static int Pipes[] = /* pipes for child process */
+ { -1, -1, -1, -1 };
+static int CtSigs[] = { 0, SIGINT, SIGKILL };
+ /* child termination signals (in order
+ * of application) -- the first is a
+ * dummy to allow pipe closure to
+ * cause the child to exit */
+#define NCTSIGS (sizeof(CtSigs) / sizeof(int))
+
+
+#if defined(HASNLIST)
+/*
+ * build-Nl() - build kernel name list table
+ */
+
+static struct drive_Nl *Build_Nl = (struct drive_Nl *)NULL;
+ /* the default Drive_Nl address */
+
+void
+build_Nl(d)
+ struct drive_Nl *d; /* data to drive the construction */
+{
+ struct drive_Nl *dp;
+ int i, n;
+
+ for (dp = d, n = 0; dp->nn; dp++, n++)
+ ;
+ if (n < 1) {
+ (void) fprintf(stderr,
+ "%s: can't calculate kernel name list length\n", Pn);
+ Exit(1);
+ }
+ if (!(Nl = (struct NLIST_TYPE *)calloc((n + 1),
+ sizeof(struct NLIST_TYPE))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes to kernel name list structure\n",
+ Pn, (int)((n + 1) * sizeof(struct NLIST_TYPE)));
+ Exit(1);
+ }
+ for (dp = d, i = 0; i < n; dp++, i++) {
+ Nl[i].NL_NAME = dp->knm;
+ }
+ Nll = (int)((n + 1) * sizeof(struct NLIST_TYPE));
+ Build_Nl = d;
+}
+#endif /* defined(HASNLIST) */
+
+
+/*
+ * childx() - make child process exit (if possible)
+ */
+
+void
+childx()
+{
+ static int at, sx;
+ pid_t wpid;
+
+ if (Cpid > 1) {
+
+ /*
+ * First close the pipes to and from the child. That should cause the
+ * child to exit. Compute alarm time shares.
+ */
+ (void) closePipes();
+ if ((at = TmLimit / NCTSIGS) < TMLIMMIN)
+ at = TMLIMMIN;
+ /*
+ * Loop, waiting for the child to exit. After the first pass, help
+ * the child exit by sending it signals.
+ */
+ for (sx = 0; sx < NCTSIGS; sx++) {
+ if (setjmp(Jmp_buf)) {
+
+ /*
+ * An alarm has rung. Disable further alarms.
+ *
+ * If there are more signals to send, continue the signal loop.
+ *
+ * If the last signal has been sent, issue a warning (unless
+ * warninge have been suppressed) and exit the signal loop.
+ */
+ (void) alarm(0);
+ (void) signal(SIGALRM, SIG_DFL);
+ if (sx < (NCTSIGS - 1))
+ continue;
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: WARNING -- child process %d may be hung.\n",
+ Pn, (int)Cpid);
+ break;
+ }
+ /*
+ * Send the next signal to the child process, after the first pass
+ * through the loop.
+ *
+ * Wrap the wait() with an alarm.
+ */
+ if (sx)
+ (void) kill(Cpid, CtSigs[sx]);
+ (void) signal(SIGALRM, handleint);
+ (void) alarm(at);
+ wpid = (pid_t) wait(NULL);
+ (void) alarm(0);
+ (void) signal(SIGALRM, SIG_DFL);
+ if (wpid == Cpid)
+ break;
+ }
+ Cpid = 0;
+ }
+}
+
+
+/*
+ * closePipes() - close open pipe file descriptors
+ */
+
+static void
+closePipes()
+{
+ int i;
+
+ for (i = 0; i < 4; i++) {
+ if (Pipes[i] >= 0) {
+ (void) close(Pipes[i]);
+ Pipes[i] = -1;
+ }
+ }
+}
+
+
+/*
+ * compdev() - compare Devtp[] entries
+ */
+
+int
+compdev(a1, a2)
+ COMP_P *a1, *a2;
+{
+ struct l_dev **p1 = (struct l_dev **)a1;
+ struct l_dev **p2 = (struct l_dev **)a2;
+
+ if ((dev_t)((*p1)->rdev) < (dev_t)((*p2)->rdev))
+ return(-1);
+ if ((dev_t)((*p1)->rdev) > (dev_t)((*p2)->rdev))
+ return(1);
+ if ((INODETYPE)((*p1)->inode) < (INODETYPE)((*p2)->inode))
+ return(-1);
+ if ((INODETYPE)((*p1)->inode) > (INODETYPE)((*p2)->inode))
+ return(1);
+ return(strcmp((*p1)->name, (*p2)->name));
+}
+
+
+/*
+ * doinchild() -- do a function in a child process
+ */
+
+static int
+doinchild(fn, fp, rbuf, rbln)
+ int (*fn)(); /* function to perform */
+ char *fp; /* function parameter */
+ char *rbuf; /* response buffer */
+ int rbln; /* response buffer length */
+{
+ int en, rv;
+/*
+ * Check reply buffer size.
+ */
+ if (!Fovhd && rbln > MAXPATHLEN) {
+ (void) fprintf(stderr,
+ "%s: doinchild error; response buffer too large: %d\n",
+ Pn, rbln);
+ Exit(1);
+ }
+/*
+ * Set up to handle an alarm signal; handle an alarm signal; build
+ * pipes for exchanging information with a child process; start the
+ * child process; and perform functions in the child process.
+ */
+ if (!Fovhd) {
+ if (setjmp(Jmp_buf)) {
+
+ /*
+ * Process an alarm that has rung.
+ */
+ (void) alarm(0);
+ (void) signal(SIGALRM, SIG_DFL);
+ (void) childx();
+ errno = ETIMEDOUT;
+ return(1);
+ } else if (!Cpid) {
+
+ /*
+ * Create pipes to exchange function information with a child
+ * process.
+ */
+ if (pipe(Pipes) < 0 || pipe(&Pipes[2]) < 0) {
+ (void) fprintf(stderr, "%s: can't open pipes: %s\n",
+ Pn, strerror(errno));
+ Exit(1);
+ }
+ /*
+ * Fork a child to execute functions.
+ */
+ if ((Cpid = fork()) == 0) {
+
+ /*
+ * Begin the child process.
+ */
+
+ int fd, nd, r_al, r_rbln;
+ char r_arg[MAXPATHLEN+1], r_rbuf[MAXPATHLEN+1];
+ int (*r_fn)();
+ /*
+ * Close all open file descriptors except Pipes[0] and
+ * Pipes[3].
+ */
+ for (fd = 0, nd = GET_MAX_FD(); fd < nd; fd++) {
+ if (fd == Pipes[0] || fd == Pipes[3])
+ continue;
+ (void) close(fd);
+ if (fd == Pipes[1])
+ Pipes[1] = -1;
+ else if (fd == Pipes[2])
+ Pipes[2] = -1;
+ }
+ if (Pipes[1] >= 0) {
+ (void) close(Pipes[1]);
+ Pipes[1] = -1;
+ }
+ if (Pipes[2] >= 0) {
+ (void) close(Pipes[2]);
+ Pipes[2] = -1;
+ }
+ /*
+ * Read function requests, process them, and return replies.
+ */
+ for (;;) {
+ if (read(Pipes[0], (char *)&r_fn, sizeof(r_fn))
+ != (int)sizeof(r_fn)
+ || read(Pipes[0], (char *)&r_al, sizeof(int))
+ != (int)sizeof(int)
+ || r_al < 1
+ || r_al > (int)sizeof(r_arg)
+ || read(Pipes[0], r_arg, r_al) != r_al
+ || read(Pipes[0], (char *)&r_rbln, sizeof(r_rbln))
+ != (int)sizeof(r_rbln)
+ || r_rbln < 1 || r_rbln > (int)sizeof(r_rbuf))
+ break;
+ rv = r_fn(r_arg, r_rbuf, r_rbln);
+ en = errno;
+ if (write(Pipes[3], (char *)&rv, sizeof(rv))
+ != sizeof(rv)
+ || write(Pipes[3], (char *)&en, sizeof(en))
+ != sizeof(en)
+ || write(Pipes[3], r_rbuf, r_rbln) != r_rbln)
+ break;
+ }
+ (void) _exit(0);
+ }
+ /*
+ * Continue in the parent process to finish the setup.
+ */
+ if (Cpid < 0) {
+ (void) fprintf(stderr, "%s: can't fork: %s\n",
+ Pn, strerror(errno));
+ Exit(1);
+ }
+ (void) close(Pipes[0]);
+ (void) close(Pipes[3]);
+ Pipes[0] = Pipes[3] = -1;
+ }
+ }
+ if (!Fovhd) {
+ int len;
+
+ /*
+ * Send a function to the child and wait for the response.
+ */
+ len = strlen(fp) + 1;
+ (void) signal(SIGALRM, handleint);
+ (void) alarm(TmLimit);
+ if (write(Pipes[1], (char *)&fn, sizeof(fn)) != sizeof(fn)
+ || write(Pipes[1], (char *)&len, sizeof(len)) != sizeof(len)
+ || write(Pipes[1], fp, len) != len
+ || write(Pipes[1], (char *)&rbln, sizeof(rbln)) != sizeof(rbln)
+ || read(Pipes[2], (char *)&rv, sizeof(rv)) != sizeof(rv)
+ || read(Pipes[2], (char *)&en, sizeof(en)) != sizeof(en)
+ || read(Pipes[2], rbuf, rbln) != rbln) {
+ (void) alarm(0);
+ (void) signal(SIGALRM, SIG_DFL);
+ (void) childx();
+ errno = ECHILD;
+ return(-1);
+ }
+ } else {
+
+ /*
+ * Do the operation directly -- not in a child.
+ */
+ (void) signal(SIGALRM, handleint);
+ (void) alarm(TmLimit);
+ rv = fn(fp, rbuf, rbln);
+ en = errno;
+ }
+/*
+ * Function completed, response collected -- complete the operation.
+ */
+ (void) alarm(0);
+ (void) signal(SIGALRM, SIG_DFL);
+ errno = en;
+ return(rv);
+}
+
+
+/*
+ * dolstat() - do an lstat() function
+ */
+
+static int
+dolstat(path, rbuf, rbln)
+ char *path; /* path */
+ char *rbuf; /* response buffer */
+ int rbln; /* response buffer length */
+
+/* ARGSUSED */
+
+{
+ return(lstat(path, (struct stat *)rbuf));
+}
+
+
+/*
+ * doreadlink() -- do a readlink() function
+ */
+
+static int
+doreadlink(path, rbuf, rbln)
+ char *path; /* path */
+ char *rbuf; /* response buffer */
+ int rbln; /* response buffer length */
+{
+ return(readlink(path, rbuf, rbln));
+}
+
+
+/*
+ * dostat() - do a stat() function
+ */
+
+static int
+dostat(path, rbuf, rbln)
+ char *path; /* path */
+ char *rbuf; /* response buffer */
+ int rbln; /* response buffer length */
+
+/* ARGSUSED */
+
+{
+ return(stat(path, (struct stat *)rbuf));
+}
+
+
+#if defined(WILLDROPGID)
+/*
+ * dropgid() - drop setgid permission
+ */
+
+void
+dropgid()
+{
+ if (!Setuidroot && Setgid) {
+ if (setgid(Mygid) < 0) {
+ (void) fprintf(stderr, "%s: can't setgid(%d): %s\n",
+ Pn, (int)Mygid, strerror(errno));
+ Exit(1);
+ }
+ Setgid = 0;
+ }
+}
+#endif /* defined(WILLDROPGID) */
+
+
+/*
+ * enter_dev_ch() - enter device characters in file structure
+ */
+
+void
+enter_dev_ch(m)
+ char *m;
+{
+ char *mp;
+
+ if (!m || *m == '\0')
+ return;
+ if (!(mp = mkstrcpy(m, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: no more dev_ch space at PID %d: \n",
+ Pn, Lp->pid);
+ safestrprt(m, stderr, 1);
+ Exit(1);
+ }
+ if (Lf->dev_ch)
+ (void) free((FREE_P *)Lf->dev_ch);
+ Lf->dev_ch = mp;
+}
+
+
+/*
+ * enter_IPstate() -- enter a TCP or UDP state
+ */
+
+void
+enter_IPstate(ty, nm, nr)
+ char *ty; /* type -- TCP or UDP */
+ char *nm; /* state name (may be NULL) */
+ int nr; /* state number */
+{
+
+#if defined(USE_LIB_PRINT_TCPTPI)
+ TcpNstates = nr;
+#else /* !defined(USE_LIB_PRINT_TCPTPI) */
+
+ int al, i, j, oc, nn, ns, off, tx;
+ char *cp;
+ MALLOC_S len;
+/*
+ * Check the type name and set the type index.
+ */
+ if (!ty) {
+ (void) fprintf(stderr,
+ "%s: no type specified to enter_IPstate()\n", Pn);
+ Exit(1);
+ }
+ if (!strcmp(ty, "TCP"))
+ tx = 0;
+ else if (!strcmp(ty, "UDP"))
+ tx = 1;
+ else {
+ (void) fprintf(stderr, "%s: unknown type for enter_IPstate: %s\n",
+ Pn, ty);
+ Exit(1);
+ }
+/*
+ * If the name argument is NULL, reduce the allocated table to its minimum
+ * size.
+ */
+ if (!nm) {
+ if (tx) {
+ if (UdpSt) {
+ if (!UdpNstates) {
+ (void) free((MALLOC_P *)UdpSt);
+ UdpSt = (char **)NULL;
+ }
+ if (UdpNstates < UdpStAlloc) {
+ len = (MALLOC_S)(UdpNstates * sizeof(char *));
+ if (!(UdpSt = (char **)realloc((MALLOC_P *)UdpSt, len)))
+ {
+ (void) fprintf(stderr,
+ "%s: can't reduce UdpSt[]\n", Pn);
+ Exit(1);
+ }
+ }
+ UdpStAlloc = UdpNstates;
+ }
+ } else {
+ if (TcpSt) {
+ if (!TcpNstates) {
+ (void) free((MALLOC_P *)TcpSt);
+ TcpSt = (char **)NULL;
+ }
+ if (TcpNstates < TcpStAlloc) {
+ len = (MALLOC_S)(TcpNstates * sizeof(char *));
+ if (!(TcpSt = (char **)realloc((MALLOC_P *)TcpSt, len)))
+ {
+ (void) fprintf(stderr,
+ "%s: can't reduce TcpSt[]\n", Pn);
+ Exit(1);
+ }
+ }
+ TcpStAlloc = TcpNstates;
+ }
+ }
+ return;
+ }
+/*
+ * Check the name and number.
+ */
+ if ((len = (size_t)strlen(nm)) < 1) {
+ (void) fprintf(stderr,
+ "%s: bad %s name (\"%s\"), number=%d\n", Pn, ty, nm, nr);
+ Exit(1);
+ }
+/*
+ * Make a copy of the name.
+ */
+ if (!(cp = mkstrcpy(nm, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr,
+ "%s: enter_IPstate(): no %s space for %s\n",
+ Pn, ty, nm);
+ Exit(1);
+ }
+/*
+ * Set the necessary offset for using nr as an index. If it is
+ * a new offset, adjust previous entries.
+ */
+ if ((nr < 0) && ((off = -nr) > (tx ? UdpStOff : TcpStOff))) {
+ if (tx ? UdpSt : TcpSt) {
+
+ /*
+ * A new, larger offset (smaller negative state number) could mean
+ * a previously allocated state table must be enlarged and its
+ * previous entries moved.
+ */
+ oc = off - (tx ? UdpStOff : TcpStOff);
+ al = tx ? UdpStAlloc : TcpStAlloc;
+ ns = tx ? UdpNstates : TcpNstates;
+ if ((nn = ns + oc) >= al) {
+ while ((nn + 5) > al) {
+ al += TCPUDPALLOC;
+ }
+ len = (MALLOC_S)(al * sizeof(char *));
+ if (tx) {
+ if (!(UdpSt = (char **)realloc((MALLOC_P *)UdpSt, len)))
+ goto no_IP_space;
+ UdpStAlloc = al;
+ } else {
+ if (!(TcpSt = (char **)realloc((MALLOC_P *)TcpSt, len)))
+ goto no_IP_space;
+ TcpStAlloc = al;
+ }
+ for (i = 0, j = oc; i < oc; i++, j++) {
+ if (tx) {
+ if (i < UdpNstates)
+ UdpSt[j] = UdpSt[i];
+ UdpSt[i] = (char *)NULL;
+ } else {
+ if (i < TcpNstates)
+ TcpSt[j] = TcpSt[i];
+ TcpSt[i] = (char *)NULL;
+ }
+ }
+ if (tx)
+ UdpNstates += oc;
+ else
+ TcpNstates += oc;
+ }
+ }
+ if (tx)
+ UdpStOff = off;
+ else
+ TcpStOff = off;
+ }
+/*
+ * Enter name as {Tc|Ud}pSt[nr + {Tc|Ud}pStOff].
+ *
+ * Allocate space, as required.
+ */
+ al = tx ? UdpStAlloc : TcpStAlloc;
+ off = tx ? UdpStOff : TcpStOff;
+ nn = nr + off + 1;
+ if (nn > al) {
+ i = tx ? UdpNstates : TcpNstates;
+ while ((nn + 5) > al) {
+ al += TCPUDPALLOC;
+ }
+ len = (MALLOC_S)(al * sizeof(char *));
+ if (tx) {
+ if (UdpSt)
+ UdpSt = (char **)realloc((MALLOC_P *)UdpSt, len);
+ else
+ UdpSt = (char **)malloc(len);
+ if (!UdpSt) {
+
+no_IP_space:
+
+ (void) fprintf(stderr, "%s: no %s state space\n", Pn, ty);
+ Exit(1);
+ }
+ UdpNstates = nn;
+ UdpStAlloc = al;
+ } else {
+ if (TcpSt)
+ TcpSt = (char **)realloc((MALLOC_P *)TcpSt, len);
+ else
+ TcpSt = (char **)malloc(len);
+ if (!TcpSt)
+ goto no_IP_space;
+ TcpNstates = nn;
+ TcpStAlloc = al;
+ }
+ while (i < al) {
+ if (tx)
+ UdpSt[i] = (char *)NULL;
+ else
+ TcpSt[i] = (char *)NULL;
+ i++;
+ }
+ } else {
+ if (tx) {
+ if (nn > UdpNstates)
+ UdpNstates = nn;
+ } else {
+ if (nn > TcpNstates)
+ TcpNstates = nn;
+ }
+ }
+ if (tx) {
+ if (UdpSt[nr + UdpStOff]) {
+
+dup_IP_state:
+
+ (void) fprintf(stderr,
+ "%s: duplicate %s state %d (already %s): %s\n",
+ Pn, ty, nr,
+ tx ? UdpSt[nr + UdpStOff] : TcpSt[nr + TcpStOff],
+ nm);
+ Exit(1);
+ }
+ UdpSt[nr + UdpStOff] = cp;
+ } else {
+ if (TcpSt[nr + TcpStOff])
+ goto dup_IP_state;
+ TcpSt[nr + TcpStOff] = cp;
+ }
+#endif /* defined(USE_LIB_PRINT_TCPTPI) */
+
+}
+
+
+/*
+ * enter_nm() - enter name in local file structure
+ */
+
+void
+enter_nm(m)
+ char *m;
+{
+ char *mp;
+
+ if (!m || *m == '\0')
+ return;
+ if (!(mp = mkstrcpy(m, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: no more nm space at PID %d for: ",
+ Pn, Lp->pid);
+ safestrprt(m, stderr, 1);
+ Exit(1);
+ }
+ if (Lf->nm)
+ (void) free((FREE_P *)Lf->nm);
+ Lf->nm = mp;
+}
+
+
+/*
+ * Exit() - do a clean exit()
+ */
+
+void
+Exit(xv)
+ int xv; /* exit() value */
+{
+ (void) childx();
+
+#if defined(HASDCACHE)
+ if (DCrebuilt && !Fwarn)
+ (void) fprintf(stderr, "%s: WARNING: %s was updated.\n",
+ Pn, DCpath[DCpathX]);
+#endif /* defined(HASDCACHE) */
+
+ exit(xv);
+}
+
+
+#if defined(HASNLIST)
+/*
+ * get_Nl_value() - get Nl value for nickname
+ */
+
+int
+get_Nl_value(nn, d, v)
+ char *nn; /* nickname of requested entry */
+ struct drive_Nl *d; /* drive_Nl table that built Nl
+ * (if NULL, use Build_Nl) */
+ KA_T *v; /* returned value (if NULL,
+ * return nothing) */
+{
+ int i;
+
+ if (!Nl || !Nll)
+ return(-1);
+ if (!d)
+ d = Build_Nl;
+ for (i = 0; d->nn; d++, i++) {
+ if (strcmp(d->nn, nn) == 0) {
+ if (v)
+ *v = (KA_T)Nl[i].n_value;
+ return(i);
+ }
+ }
+ return(-1);
+}
+#endif /* defined(HASNLIST) */
+
+
+/*
+ * handleint() - handle an interrupt
+ */
+
+#if defined(HASINTSIGNAL)
+static int
+#else
+static void
+#endif
+
+/* ARGSUSED */
+
+handleint(sig)
+ int sig;
+{
+ longjmp(Jmp_buf, 1);
+}
+
+
+/*
+ * hashbyname() - hash by name
+ */
+
+int
+hashbyname(nm, mod)
+ char *nm; /* pointer to NUL-terminated name */
+ int mod; /* hash modulus */
+{
+ int i, j;
+
+ for (i = j = 0; *nm; nm++) {
+ i ^= (int)*nm << j;
+ if (++j > 7)
+ j = 0;
+ }
+ return(((int)(i * 31415)) & (mod - 1));
+}
+
+
+/*
+ * is_nw_addr() - is this network address selected?
+ */
+
+int
+is_nw_addr(ia, p, af)
+ unsigned char *ia; /* Internet address */
+ int p; /* port */
+ int af; /* address family -- e.g., AF_INET,
+ * AF_INET6 */
+{
+ struct nwad *n;
+
+ if (!(n = Nwad))
+ return(0);
+ for (; n; n = n->next) {
+ if (n->proto) {
+ if (strcasecmp(n->proto, Lf->iproto) != 0)
+ continue;
+ }
+ if (af && n->af && af != n->af)
+ continue;
+
+#if defined(HASIPv6)
+ if (af == AF_INET6) {
+ if (n->a[15] || n->a[14] || n->a[13] || n->a[12]
+ || n->a[11] || n->a[10] || n->a[9] || n->a[8]
+ || n->a[7] || n->a[6] || n->a[5] || n->a[4]
+ || n->a[3] || n->a[2] || n->a[1] || n->a[0]) {
+ if (ia[15] != n->a[15] || ia[14] != n->a[14]
+ || ia[13] != n->a[13] || ia[12] != n->a[12]
+ || ia[11] != n->a[11] || ia[10] != n->a[10]
+ || ia[9] != n->a[9] || ia[8] != n->a[8]
+ || ia[7] != n->a[7] || ia[6] != n->a[6]
+ || ia[5] != n->a[5] || ia[4] != n->a[4]
+ || ia[3] != n->a[3] || ia[2] != n->a[2]
+ || ia[1] != n->a[1] || ia[0] != n->a[0])
+ continue;
+ }
+ } else if (af == AF_INET)
+#endif /* defined(HASIPv6) */
+
+ {
+ if (n->a[3] || n->a[2] || n->a[1] || n->a[0]) {
+ if (ia[3] != n->a[3] || ia[2] != n->a[2]
+ || ia[1] != n->a[1] || ia[0] != n->a[0])
+ continue;
+ }
+ }
+
+#if defined(HASIPv6)
+ else
+ continue;
+#endif /* defined(HASIPv6) */
+
+ if (n->sport == -1 || (p >= n->sport && p <= n->eport)) {
+ n->f = 1;
+ return(1);
+ }
+ }
+ return(0);
+}
+
+
+/*
+ * mkstrcpy() - make a string copy in malloc()'d space
+ *
+ * return: copy pointer
+ * copy length (optional)
+ */
+
+char *
+mkstrcpy(src, rlp)
+ char *src; /* source */
+ MALLOC_S *rlp; /* returned length pointer (optional)
+ * The returned length is an strlen()
+ * equivalent */
+{
+ MALLOC_S len;
+ char *ns;
+
+ len = (MALLOC_S)(src ? strlen(src) : 0);
+ ns = (char *)malloc(len + 1);
+ if (ns) {
+ if (src)
+ (void) snpf(ns, len + 1, "%s", src);
+ else
+ *ns = '\0';
+ }
+ if (rlp)
+ *rlp = len;
+ return(ns);
+}
+
+
+/*
+ * mkstrcat() - make a catenated copy of up to three strings under optional
+ * string-by-string count control
+ *
+ * return: copy pointer
+ * copy string length (optional)
+ */
+
+char *
+mkstrcat(s1, l1, s2, l2, s3, l3, clp)
+ char *s1; /* source string 1 */
+ int l1; /* length of string 1 (-1 if none) */
+ char *s2; /* source string 2 */
+ int l2; /* length of string 2 (-1 if none) */
+ char *s3; /* source string 3 (optional) */
+ int l3 ; /* length of string 3 (-1 if none) */
+ MALLOC_S *clp; /* pointer to return of copy length
+ * (optional) */
+{
+ MALLOC_S cl, len1, len2, len3;
+ char *cp;
+
+ if (s1)
+ len1 = (MALLOC_S)((l1 >= 0) ? l1 : strlen(s1));
+ else
+ len1 = (MALLOC_S)0;
+ if (s2)
+ len2 = (MALLOC_S)((l2 >= 0) ? l2 : strlen(s2));
+ else
+ len2 = (MALLOC_S)0;
+ if (s3)
+ len3 = (MALLOC_S)((l3 >= 0) ? l3 : strlen(s3));
+ else
+ len3 = (MALLOC_S)0;
+ cl = len1 + len2 + len3;
+ if ((cp = (char *)malloc(cl + 1))) {
+ char *tp = cp;
+
+ if (s1 && len1) {
+ (void) strncpy(tp, s1, len1);
+ tp += len1;
+ }
+ if (s2 && len2) {
+ (void) strncpy(tp, s2, len2);
+ tp += len2;
+ }
+ if (s3 && len3) {
+ (void) strncpy(tp, s3, len3);
+ tp += len3;
+ }
+ *tp = '\0';
+ }
+ if (clp)
+ *clp = cl;
+ return(cp);
+}
+
+
+/*
+ * is_readable() -- is file readable
+ */
+
+int
+is_readable(path, msg)
+ char *path; /* file path */
+ int msg; /* issue warning message if 1 */
+{
+ if (access(path, R_OK) < 0) {
+ if (!Fwarn && msg == 1)
+ (void) fprintf(stderr, ACCESSERRFMT, Pn, path, strerror(errno));
+ return(0);
+ }
+ return(1);
+}
+
+
+/*
+ * lstatsafely() - lstat path safely (i. e., with timeout)
+ */
+
+int
+lstatsafely(path, buf)
+ char *path; /* file path */
+ struct stat *buf; /* stat buffer address */
+{
+ if (Fblock) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: avoiding stat(%s): -b was specified.\n",
+ Pn, path);
+ errno = EWOULDBLOCK;
+ return(1);
+ }
+ return(doinchild(dolstat, path, (char *)buf, sizeof(struct stat)));
+}
+
+
+/*
+ * Readlink() - read and interpret file system symbolic links
+ */
+
+char *
+Readlink(arg)
+ char *arg; /* argument to be interpreted */
+{
+ char abuf[MAXPATHLEN+1];
+ int alen;
+ char *ap;
+ char *argp1, *argp2;
+ int i, len, llen, slen;
+ char lbuf[MAXPATHLEN+1];
+ static char *op = (char *)NULL;
+ static int ss = 0;
+ char *s1;
+ static char **stk = (char **)NULL;
+ static int sx = 0;
+ char tbuf[MAXPATHLEN+1];
+/*
+ * See if avoiding kernel blocks.
+ */
+ if (Fblock) {
+ if (!Fwarn) {
+ (void) fprintf(stderr, "%s: avoiding readlink(", Pn);
+ safestrprt(arg, stderr, 0);
+ (void) fprintf(stderr, "): -b was specified.\n");
+ }
+ op = (char *)NULL;
+ return(arg);
+ }
+/*
+ * Save the original path.
+ */
+ if (!op)
+ op = arg;
+/*
+ * Evaluate each component of the argument for a symbolic link.
+ */
+ for (alen = 0, ap = abuf, argp1 = argp2 = arg; *argp2; argp1 = argp2 ) {
+ for (argp2 = argp1 + 1; *argp2 && *argp2 != '/'; argp2++)
+ ;
+ if ((len = argp2 - arg) >= (int)sizeof(tbuf)) {
+
+path_too_long:
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: readlink() path too long: ", Pn);
+ safestrprt(op ? op : arg, stderr, 1);
+ }
+ op = (char *)NULL;
+ return((char *)NULL);
+ }
+ (void) strncpy(tbuf, arg, len);
+ tbuf[len] = '\0';
+ /*
+ * Dereference a symbolic link.
+ */
+ if ((llen=doinchild(doreadlink,tbuf,lbuf,sizeof(lbuf) - 1)) >= 0) {
+
+ /*
+ * If the link is a new absolute path, replace
+ * the previous assembly with it.
+ */
+ if (lbuf[0] == '/') {
+ (void) strncpy(abuf, lbuf, llen);
+ ap = &abuf[llen];
+ *ap = '\0';
+ alen = llen;
+ continue;
+ }
+ lbuf[llen] = '\0';
+ s1 = lbuf;
+ } else {
+ llen = argp2 - argp1;
+ s1 = argp1;
+ }
+ /*
+ * Make sure two components are separated by a `/'.
+ *
+ * If the first component is not a link, don't force
+ * a leading '/'.
+ *
+ * If the first component is a link and the source of
+ * the link has a leading '/', force a leading '/'.
+ */
+ if (*s1 == '/')
+ slen = 1;
+ else {
+ if (alen > 0) {
+
+ /*
+ * This is not the first component.
+ */
+ if (abuf[alen - 1] == '/')
+ slen = 1;
+ else
+ slen = 2;
+ } else {
+
+ /*
+ * This is the first component.
+ */
+ if (s1 == lbuf && tbuf[0] == '/')
+ slen = 2;
+ else
+ slen = 1;
+ }
+ }
+ /*
+ * Add to the path assembly.
+ */
+ if ((alen + llen + slen) >= (int)sizeof(abuf))
+ goto path_too_long;
+ if (slen == 2)
+ *ap++ = '/';
+ (void) strncpy(ap, s1, llen);
+ ap += llen;
+ *ap = '\0';
+ alen += (llen + slen - 1);
+ }
+/*
+ * If the assembled path and argument are the same, free all but the
+ * last string in the stack, and return the argument.
+ */
+ if (strcmp(arg, abuf) == 0) {
+ for (i = 0; i < sx; i++) {
+ if (i < (sx - 1))
+ (void) free((FREE_P *)stk[i]);
+ stk[i] = (char *)NULL;
+ }
+ sx = 0;
+ op = (char *)NULL;
+ return(arg);
+ }
+/*
+ * If the assembled path and argument are different, add it to the
+ * string stack, then Readlink() it.
+ */
+ if (!(s1 = mkstrcpy(abuf, (MALLOC_S *)NULL))) {
+
+no_readlink_space:
+
+ (void) fprintf(stderr, "%s: no Readlink string space for ", Pn);
+ safestrprt(abuf, stderr, 1);
+ Exit(1);
+ }
+ if (sx >= MAXSYMLINKS) {
+
+ /*
+ * If there are too many symbolic links, report an error, clear
+ * the stack, and return no path.
+ */
+ if (!Fwarn) {
+ (void) fprintf(stderr,
+ "%s: too many (> %d) symbolic links in readlink() path: ",
+ Pn, MAXSYMLINKS);
+ safestrprt(op ? op : arg, stderr, 1);
+ }
+ for (i = 0; i < sx; i++) {
+ (void) free((FREE_P *)stk[i]);
+ stk[i] = (char *)NULL;
+ }
+ (void) free((FREE_P *)stk);
+ stk = (char **)NULL;
+ ss = sx = 0;
+ op = (char *)NULL;
+ return((char *)NULL);
+ }
+ if (++sx > ss) {
+ if (!stk)
+ stk = (char **)malloc((MALLOC_S)(sizeof(char *) * sx));
+ else
+ stk = (char **)realloc((MALLOC_P *)stk,
+ (MALLOC_S)(sizeof(char *) * sx));
+ if (!stk)
+ goto no_readlink_space;
+ ss = sx;
+ }
+ stk[sx - 1] = s1;
+ return(Readlink(s1));
+}
+
+
+#if defined(HASSTREAMS)
+/*
+ * readstdata() - read stream's stdata structure
+ */
+
+int
+readstdata(addr, buf)
+ KA_T addr; /* stdata address in kernel*/
+ struct stdata *buf; /* buffer addess */
+{
+ if (!addr
+ || kread(addr, (char *)buf, sizeof(struct stdata))) {
+ (void) snpf(Namech, Namechl, "no stream data in %s",
+ print_kptr(addr, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+
+
+/*
+ * readsthead() - read stream head
+ */
+
+int
+readsthead(addr, buf)
+ KA_T addr; /* starting queue pointer in kernel */
+ struct queue *buf; /* buffer for queue head */
+{
+ KA_T qp;
+
+ if (!addr) {
+ (void) snpf(Namech, Namechl, "no stream queue head");
+ return(1);
+ }
+ for (qp = addr; qp; qp = (KA_T)buf->q_next) {
+ if (kread(qp, (char *)buf, sizeof(struct queue))) {
+ (void) snpf(Namech, Namechl, "bad stream queue link at %s",
+ print_kptr(qp, (char *)NULL, 0));
+ return(1);
+ }
+ }
+ return(0);
+}
+
+
+/*
+ * readstidnm() - read stream module ID name
+ */
+
+int
+readstidnm(addr, buf, len)
+ KA_T addr; /* module ID name address in kernel */
+ char *buf; /* receiving buffer address */
+ READLEN_T len; /* buffer length */
+{
+ if (!addr || kread(addr, buf, len)) {
+ (void) snpf(Namech, Namechl, "can't read module ID name from %s",
+ print_kptr(addr, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+
+
+/*
+ * readstmin() - read stream's module info
+ */
+
+int
+readstmin(addr, buf)
+ KA_T addr; /* module info address in kernel */
+ struct module_info *buf; /* receiving buffer address */
+{
+ if (!addr || kread(addr, (char *)buf, sizeof(struct module_info))) {
+ (void) snpf(Namech, Namechl, "can't read module info from %s",
+ print_kptr(addr, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+
+
+/*
+ * readstqinit() - read stream's queue information structure
+ */
+
+int
+readstqinit(addr, buf)
+ KA_T addr; /* queue info address in kernel */
+ struct qinit *buf; /* receiving buffer address */
+{
+ if (!addr || kread(addr, (char *)buf, sizeof(struct qinit))) {
+ (void) snpf(Namech, Namechl, "can't read queue info from %s",
+ print_kptr(addr, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* HASSTREAMS */
+
+
+/*
+ * safepup() - safely print an unprintable character -- i.e., print it in a
+ * printable form
+ *
+ * return: char * to printable equivalent
+ * cl = strlen(printable equivalent)
+ */
+
+static char *
+safepup(c, cl)
+ unsigned int c; /* unprintable (i.e., !isprint())
+ * character */
+ int *cl; /* returned printable strlen -- NULL if
+ * no return needed */
+{
+ int len;
+ char *rp;
+ static char up[8];
+
+ if (c < 0x20) {
+ switch (c) {
+ case '\b':
+ rp = "\\b";
+ break;
+ case '\f':
+ rp = "\\f";
+ break;
+ case '\n':
+ rp = "\\n";
+ break;
+ case '\r':
+ rp = "\\r";
+ break;
+ case '\t':
+ rp = "\\t";
+ break;
+ default:
+ (void) snpf(up, sizeof(up), "^%c", c + 0x40);
+ rp = up;
+ }
+ len = 2;
+ } else if (c == 0xff) {
+ rp = "^?";
+ len = 2;
+ } else {
+ (void) snpf(up, sizeof(up), "\\x%02x", (int)(c & 0xff));
+ rp = up;
+ len = 4;
+ }
+ if (cl)
+ *cl = len;
+ return(rp);
+}
+
+
+/*
+ * safestrlen() - calculate a "safe" string length -- i.e., compute space for
+ * non-printable characters when printed in a printable form
+ */
+
+int
+safestrlen(sp, flags)
+ char *sp; /* string pointer */
+ int flags; /* flags:
+ * bit 0: 0 (0) = no NL
+ * 1 (1) = add trailing NL
+ * 1: 0 (0) = ' ' printable
+ * 1 (2) = ' ' not printable
+ */
+{
+ char c;
+ int len = 0;
+
+ c = (flags & 2) ? ' ' : '\0';
+ if (sp) {
+ for (; *sp; sp++) {
+ if (!isprint((unsigned char)*sp) || *sp == c) {
+ if (*sp < 0x20 || (unsigned char)*sp == 0xff)
+ len += 2; /* length of \. or ^. form */
+ else
+ len += 4; /* length of "\x%02x" printf */
+ } else
+ len++;
+ }
+ }
+ return(len);
+}
+
+
+/*
+ * safestrprt() - print a string "safely" to the indicated stream -- i.e.,
+ * print unprintable characters in a printable form
+ */
+
+void
+safestrprt(sp, fs, flags)
+ char *sp; /* string to print pointer pointer */
+ FILE *fs; /* destination stream -- e.g., stderr
+ * or stdout */
+ int flags; /* flags:
+ * bit 0: 0 (0) = no NL
+ * 1 (1) = add trailing NL
+ * 1: 0 (0) = ' ' printable
+ * 1 (2) = ' ' not printable
+ * 2: 0 (0) = print string as is
+ * 1 (4) = surround string
+ * with '"'
+ * 4: 0 (0) = print ending '\n'
+ * 1 (8) = don't print ending
+ * '\n'
+ */
+{
+ char c;
+ int lnc, lnt, sl;
+
+#if defined(HASWIDECHAR)
+ wchar_t w;
+ int wcmx = MB_CUR_MAX;
+#else /* !defined(HASWIDECHAR) */
+ static int wcmx = 1;
+#endif /* defined(HASWIDECHAR) */
+
+ c = (flags & 2) ? ' ' : '\0';
+ if (flags & 4)
+ putc('"', fs);
+ if (sp) {
+ for (sl = strlen(sp); *sp; sl -= lnc, sp += lnc) {
+
+#if defined(HASWIDECHAR)
+ if (wcmx > 1) {
+ lnc = mblen(sp, sl);
+ if (lnc > 1) {
+ if ((mbtowc(&w, sp, sl) == lnc) && iswprint(w)) {
+ for (lnt = 0; lnt < lnc; lnt++) {
+ putc((int)*(sp + lnt), fs);
+ }
+ } else {
+ for (lnt = 0; lnt < lnc; lnt++) {
+ fputs(safepup((unsigned int)*(sp + lnt),
+ (int *)NULL), fs);
+ }
+ }
+ continue;
+ } else
+ lnc = 1;
+ } else
+ lnc = 1;
+#else /* !defined(HASWIDECHAR) */
+ lnc = 1;
+#endif /* defined(HASWIDECHAR) */
+
+ if (isprint((unsigned char)*sp) && *sp != c)
+ putc((int)(*sp & 0xff), fs);
+ else {
+ if ((flags & 8) && (*sp == '\n') && !*(sp + 1))
+ break;
+ fputs(safepup((unsigned int)*sp, (int *)NULL), fs);
+ }
+ }
+ }
+ if (flags & 4)
+ putc('"', fs);
+ if (flags & 1)
+ putc('\n', fs);
+}
+
+
+/*
+ * safestrprtn() - print a specified number of characters from a string
+ * "safely" to the indicated stream
+ */
+
+void
+safestrprtn(sp, len, fs, flags)
+ char *sp; /* string to print pointer pointer */
+ int len; /* safe number of characters to
+ * print */
+ FILE *fs; /* destination stream -- e.g., stderr
+ * or stdout */
+ int flags; /* flags:
+ * bit 0: 0 (0) = no NL
+ * 1 (1) = add trailing NL
+ * 1: 0 (0) = ' ' printable
+ * 1 (2) = ' ' not printable
+ * 2: 0 (0) = print string as is
+ * 1 (4) = surround string
+ * with '"'
+ * 4: 0 (0) = print ending '\n'
+ * 1 (8) = don't print ending
+ * '\n'
+ */
+{
+ char c, *up;
+ int cl, i;
+
+ if (flags & 4)
+ putc('"', fs);
+ if (sp) {
+ c = (flags & 2) ? ' ' : '\0';
+ for (i = 0; i < len && *sp; sp++) {
+ if (isprint((unsigned char)*sp) && *sp != c) {
+ putc((int)(*sp & 0xff), fs);
+ i++;
+ } else {
+ if ((flags & 8) && (*sp == '\n') && !*(sp + 1))
+ break;
+ up = safepup((unsigned int)*sp, &cl);
+ if ((i + cl) > len)
+ break;
+ fputs(up, fs);
+ i += cl;
+ }
+ }
+ } else
+ i = 0;
+ for (; i < len; i++)
+ putc(' ', fs);
+ if (flags & 4)
+ putc('"', fs);
+ if (flags & 1)
+ putc('\n', fs);
+}
+
+
+/*
+ * statsafely() - stat path safely (i. e., with timeout)
+ */
+
+int
+statsafely(path, buf)
+ char *path; /* file path */
+ struct stat *buf; /* stat buffer address */
+{
+ if (Fblock) {
+ if (!Fwarn)
+ (void) fprintf(stderr,
+ "%s: avoiding stat(%s): -b was specified.\n",
+ Pn, path);
+ errno = EWOULDBLOCK;
+ return(1);
+ }
+ return(doinchild(dostat, path, (char *)buf, sizeof(struct stat)));
+}
+
+
+/*
+ * stkdir() - stack directory name
+ */
+
+void
+stkdir(p)
+ char *p; /* directory path */
+{
+ MALLOC_S len;
+/*
+ * Provide adequate space for directory stack pointers.
+ */
+ if (Dstkx >= Dstkn) {
+ Dstkn += 128;
+ len = (MALLOC_S)(Dstkn * sizeof(char *));
+ if (!Dstk)
+ Dstk = (char **)malloc(len);
+ else
+ Dstk = (char **)realloc((MALLOC_P *)Dstk, len);
+ if (!Dstk) {
+ (void) fprintf(stderr,
+ "%s: no space for directory stack at: ", Pn);
+ safestrprt(p, stderr, 1);
+ Exit(1);
+ }
+ }
+/*
+ * Allocate space for the name, copy it there and put its pointer on the stack.
+ */
+ if (!(Dstk[Dstkx] = mkstrcpy(p, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: no space for: ", Pn);
+ safestrprt(p, stderr, 1);
+ Exit(1);
+ }
+ Dstkx++;
+}
+
+
+/*
+ * x2dev() - convert hexadecimal ASCII string to device number
+ */
+
+char *
+x2dev(s, d)
+ char *s; /* ASCII string */
+ dev_t *d; /* device receptacle */
+{
+ char *cp, *cp1;
+ int n;
+ dev_t r;
+
+/*
+ * Skip an optional leading 0x. Count the number of hex digits up to the end
+ * of the string, or to a space, or to a comma. Return an error if an unknown
+ * character is encountered. If the count is larger than (2 * sizeof(dev_t))
+ * -- e.g., because of sign extension -- ignore excess leading hex 0xf digits,
+ * but return an error if an excess leading digit isn't 0xf.
+ */
+ if (strncasecmp(s, "0x", 2) == 0)
+ s += 2;
+ for (cp = s, n = 0; *cp; cp++, n++) {
+ if (isdigit((unsigned char)*cp))
+ continue;
+ if ((unsigned char)*cp >= 'a' && (unsigned char)*cp <= 'f')
+ continue;
+ if ((unsigned char)*cp >= 'A' && (unsigned char)*cp <= 'F')
+ continue;
+ if (*cp == ' ' || *cp == ',')
+ break;
+ return((char *)NULL);
+ }
+ if (!n)
+ return((char *)NULL);
+ if (n > (2 * (int)sizeof(dev_t))) {
+ cp1 = s;
+ s += (n - (2 * sizeof(dev_t)));
+ while (cp1 < s) {
+ if (*cp1 != 'f' && *cp1 != 'F')
+ return((char *)NULL);
+ cp1++;
+ }
+ }
+/*
+ * Assemble the validated hex digits of the device number, starting at a point
+ * in the string relevant to sizeof(dev_t).
+ */
+ for (r = 0; s < cp; s++) {
+ r = r << 4;
+ if (isdigit((unsigned char)*s))
+ r |= (unsigned char)(*s - '0') & 0xf;
+ else {
+ if (isupper((unsigned char)*s))
+ r |= ((unsigned char)(*s - 'A') + 10) & 0xf;
+ else
+ r |= ((unsigned char)(*s - 'a') + 10) & 0xf;
+ }
+ }
+ *d = r;
+ return(s);
+}
--- /dev/null
+/*
+ * node.c - common node reading functions for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1994 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: node.c,v 1.5 2000/08/01 17:08:05 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * print_kptr() - print kernel pointer
+ */
+
+char *
+print_kptr(kp, buf, bufl)
+ KA_T kp; /* kernel pointer address */
+ char *buf; /* optional destination buffer */
+ size_t bufl; /* size of buf[] */
+{
+ static char dbuf[32];
+
+ (void) snpf(buf ? buf : dbuf,
+ buf ? bufl : sizeof(dbuf),
+ KA_T_FMT_X, kp);
+ return(buf ? buf : dbuf);
+}
+
+
+#if defined(HASCDRNODE)
+/*
+ * readcdrnode() - read CD-ROM node
+ */
+
+int
+readcdrnode(ca, c)
+ KA_T ca; /* cdrnode kernel address */
+ struct cdrnode *c; /* cdrnode buffer */
+{
+ if (kread((KA_T)ca, (char *)c, sizeof(struct cdrnode))) {
+ (void) snpf(Namech, Namechl, "can't read cdrnode at %s",
+ print_kptr(ca, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASCDRNODE) */
+
+
+#if defined(HASFIFONODE)
+/*
+ * readfifonode() - read fifonode
+ */
+
+int
+readfifonode(fa, f)
+ KA_T fa; /* fifonode kernel address */
+ struct fifonode *f; /* fifonode buffer */
+{
+ if (kread((KA_T)fa, (char *)f, sizeof(struct fifonode))) {
+ (void) snpf(Namech, Namechl, "can't read fifonode at %s",
+ print_kptr(fa, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASFIFONODE) */
+
+
+#if defined(HASGNODE)
+/*
+ * readgnode() - read gnode
+ */
+
+int
+readgnode(ga, g)
+ KA_T ga; /* gnode kernel address */
+ struct gnode *g; /* gnode buffer */
+{
+ if (kread((KA_T)ga, (char *)g, sizeof(struct gnode))) {
+ (void) snpf(Namech, Namechl, "can't read gnode at %s",
+ print_kptr(ga, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASGNODE) */
+
+
+#if defined(HASHSNODE)
+/*
+ * readhsnode() - read High Sierra file system node
+ */
+
+int
+readhsnode(ha, h)
+ KA_T ha; /* hsnode kernel address */
+ struct hsnode *h; /* hsnode buffer */
+{
+ if (kread((KA_T)ha, (char *)h, sizeof(struct hsnode))) {
+ (void) snpf(Namech, Namechl, "can't read hsnode at %s",
+ print_kptr(ha, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASHSNODE) */
+
+
+#if defined(HASINODE)
+/*
+ * readinode() - read inode
+ */
+
+int
+readinode(ia, i)
+ KA_T ia; /* inode kernel address */
+ struct inode *i; /* inode buffer */
+{
+ if (kread((KA_T)ia, (char *)i, sizeof(struct inode))) {
+ (void) snpf(Namech, Namechl, "can't read inode at %s",
+ print_kptr(ia, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASINODE) */
+
+
+#if defined(HASPIPENODE)
+/*
+ * readpipenode() - read pipe node
+ */
+
+int
+readpipenode(pa, p)
+ KA_T pa; /* pipe node kernel address */
+ struct pipenode *p; /* pipe node buffer */
+{
+ if (kread((KA_T)pa, (char *)p, sizeof(struct pipenode))) {
+ (void) snpf(Namech, Namechl, "can't read pipenode at %s",
+ print_kptr(pa, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASPIPENODE) */
+
+
+#if defined(HASRNODE)
+/*
+ * readrnode() - read rnode
+ */
+
+int
+readrnode(ra, r)
+ KA_T ra; /* rnode kernel space address */
+ struct rnode *r; /* rnode buffer pointer */
+{
+ if (kread((KA_T)ra, (char *)r, sizeof(struct rnode))) {
+ (void) snpf(Namech, Namechl, "can't read rnode at %s",
+ print_kptr(ra, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASRNODE) */
+
+
+#if defined(HASSNODE)
+/*
+ * readsnode() - read snode
+ */
+
+int
+readsnode(sa, s)
+ KA_T sa; /* snode kernel space address */
+ struct snode *s; /* snode buffer pointer */
+{
+ if (kread((KA_T)sa, (char *)s, sizeof(struct snode))) {
+ (void) snpf(Namech, Namechl, "can't read snode at %s",
+ print_kptr(sa, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASSNODE) */
+
+
+#if defined(HASTMPNODE)
+/*
+ * readtnode() - read tmpnode
+ */
+
+int
+readtnode(ta, t)
+ KA_T ta; /* tmpnode kernel space address */
+ struct tmpnode *t; /* tmpnode buffer pointer */
+{
+ if (kread((KA_T)ta, (char *)t, sizeof(struct tmpnode))) {
+ (void) snpf(Namech, Namechl, "can't read tmpnode at %s",
+ print_kptr(ta, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASTMPNODE) */
+
+
+#if defined(HASVNODE)
+/*
+ * readvnode() - read vnode
+ */
+
+int
+readvnode(va, v)
+ KA_T va; /* vnode kernel space address */
+ struct vnode *v; /* vnode buffer pointer */
+{
+ if (kread((KA_T)va, (char *)v, sizeof(struct vnode))) {
+ (void) snpf(Namech, Namechl, "can't read vnode at %s",
+ print_kptr(va, (char *)NULL, 0));
+ return(1);
+ }
+ return(0);
+}
+#endif /* defined(HASVNODE) */
--- /dev/null
+* Wed Jun 08 2011 Anas Nashif <anas.nashif@intel.com> - 4.82
+- Repackage docs
+
+* Mon Jan 11 2010 Passion Zhao <passion.zhao@intel.com> - 4.82
+- Update to 4.82
+
+* Sat Feb 07 2009 Anas Nashif <anas.nashif@intel.com> 4.81
+- Update to 4.81
+
+* Tue Sep 09 2008 Anas Nashif <anas.nashif@intel.com> 4.78
+- initial checkin into moblin
+- remove selinux support
+
--- /dev/null
+#
+# Please submit bugfixes or comments via http://bugs.meego.com/
+#
+
+Name: lsof
+Version: 4.82
+Release: 1
+License: zlib
+Summary: A utility which lists open files on a Linux/UNIX system
+Group: Development/Debuggers
+
+Url: ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof
+# lsof contains licensed code that we cannot ship. Therefore we use
+# upstream2downstream.sh script to remove the code before shipping it.
+#
+# The script you can found in CVS or download from:
+# http://cvs.fedoraproject.org/viewcvs/rpms/lsof/devel/upstream2downstream.sh
+#
+%define lsofrh lsof_4.82-rh
+Source0: %{lsofrh}.tar.bz2
+
+# 184338 - allow lsof access nptl threads
+Patch1: lsof_4.81-threads.patch
+
+%description
+Lsof stands for LiSt Open Files, and it does just that: it lists
+information about files that are open by the processes running on a
+UNIX system.
+
+%prep
+%setup -q -n %{lsofrh}
+%patch1 -p1
+
+%build
+LSOF_VSTR=2.6.16 LINUX_BASE=/proc ./Configure -n linux
+
+make DEBUG="%{optflags}" %{?_smp_mflags}
+
+%install
+mkdir -p %{buildroot}%{_sbindir}
+install -p -m 0755 lsof %{buildroot}%{_prefix}/sbin
+mkdir -p %{buildroot}%{_mandir}/man8
+install -p lsof.8 %{buildroot}%{_mandir}/man8/
+
+%clean
+rm -rf %{buildroot}
+
+%docs_package
+
+%files
+%defattr(644,root,root,755)
+%attr(0755,root,root) %{_sbindir}/lsof
+
--- /dev/null
+diff -up lsof_4.81-rh/dialects/linux/dproc.c.kzak lsof_4.81-rh/dialects/linux/dproc.c
+--- lsof_4.81-rh/dialects/linux/dproc.c.kzak 2008-10-21 18:17:25.000000000 +0200
++++ lsof_4.81-rh/dialects/linux/dproc.c 2008-12-02 10:54:54.000000000 +0100
+@@ -89,7 +89,8 @@ _PROTOTYPE(static void process_proc_map,
+ _PROTOTYPE(static int process_id,(char *idp, int idpl, char *cmd, UID_ARG uid,
+ int pid, int ppid, int pgid));
+ _PROTOTYPE(static int statEx,(char *p, struct stat *s, int *ss));
+-
++_PROTOTYPE(static int get_other_thread,(int pid, char **tid));
++
+
+ #if defined(HASSELINUX)
+ _PROTOTYPE(static int cmp_cntx_eq,(char *pcntx, char *ucntx));
+@@ -159,6 +160,7 @@ gather_proc_info()
+ struct dirent *dp;
+ struct stat sb;
+ int lwp, n, nl, pgid, pid, ppid, rv, tx;
++ char *tid = NULL;
+ static char *lwppath = (char *)NULL;
+ static int lwppathl = 0;
+ static char *path = (char *)NULL;
+@@ -252,6 +254,13 @@ gather_proc_info()
+ while ((dp = readdir(ps))) {
+ if (nm2id(dp->d_name, &pid, &n))
+ continue;
++
++ tid = NULL;
++ if (get_other_thread(pid, &tid) < 0)
++ continue;
++ if (tid)
++ n += sizeof("task/") + strlen(tid);
++
+ /*
+ * Build path to PID's directory.
+ */
+@@ -265,7 +274,14 @@ gather_proc_info()
+ Exit(1);
+ }
+ }
+- (void) snpf(pidpath + pidx, pidpathl - pidx, "%s/", dp->d_name);
++ if (tid) {
++ /* /proc/<pid> is useless (zombie), we have to use /proc/<pid>/task/<tid>
++ * where is still running thread
++ */
++ (void) snpf(pidpath + pidx, pidpathl - pidx, "%s/task/%s/", dp->d_name, tid);
++ free(tid);
++ } else
++ (void) snpf(pidpath + pidx, pidpathl - pidx, "%s/", dp->d_name);
+ n += (pidx + 1);
+ /*
+ * Process the PID's stat info.
+@@ -1007,6 +1023,64 @@ process_id(idp, idpl, cmd, uid, pid, ppi
+ return(0);
+ }
+
++/* fill tid if the initial thread is zombie,
++ * but other thread still alive
++ *
++ * returns -1=error, 0=nothing, 1=ok
++ */
++static int
++get_other_thread(pid, tid)
++ int pid;
++ char **tid;
++{
++ char path[MAXPATHLEN];
++ DIR *tdp;
++ struct dirent *td;
++ char pstate;
++ FILE *f;
++ int _pid;
++ int re = 0, x;
++
++ snpf(path, sizeof(path), "%s/%d/stat", PROCFS, pid);
++ if (!(f = fopen(path, "r")))
++ return -1;
++ x = fscanf(f, "%d %*s %c", &_pid, &pstate);
++ fclose(f);
++ if (x!=2)
++ return -1;
++ if (_pid != pid)
++ return -1; /* corrupted /proc? */
++ if (pstate!='Z')
++ return 0; /* ignore normal proceses */
++
++ snpf(path, sizeof(path), "%s/%d/task", PROCFS, pid);
++
++ /* open /proc/<pid>/task */
++ if (!(tdp = opendir(path)))
++ return 0; /* kernel < 2.6.x */
++
++ /* look for first alive thread */
++ while ((td = readdir(tdp))) {
++ if (strcmp(td->d_name, ".")==0 || strcmp(td->d_name, "..")==0)
++ continue;
++
++ /* /proc/<pid>/task/<tid>/stat */
++ snpf(path, sizeof(path), "%s/%d/task/%s/stat", PROCFS, pid, td->d_name);
++ if (!(f = fopen(path, "r")))
++ continue;
++ x = fscanf(f, "%*d %*s %c", &pstate);
++ fclose(f);
++ if (x!=1)
++ continue;
++ if (pstate!='Z') {
++ re = 1;
++ *tid = strdup(td->d_name);
++ break;
++ }
++ }
++ closedir(tdp);
++ return re;
++}
+
+ /*
+ * process_proc_map() - process the memory map of a process
+@@ -1250,12 +1324,6 @@ read_id_stat(ty, p, id, cmd, ppid, pgid)
+ return(1);
+ }
+ /*
+- * Convert the first field to an integer; its conversion must match the
+- * ID argument.
+- */
+- if (!fp[0] || (atoi(fp[0]) != id))
+- return(1);
+-/*
+ * Get the command name from the second field. Strip a starting '(' and
+ * an ending ')'. Allocate space to hold the result and return the space
+ * pointer.
--- /dev/null
+/*
+ * print.c - common print support functions for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1994 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: print.c,v 1.50 2008/10/21 16:21:41 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Local definitions, structures and function prototypes
+ */
+
+#define HCINC 64 /* host cache size increase chunk */
+#define PORTHASHBUCKETS 128 /* port hash bucket count
+ * !!MUST BE A POWER OF 2!! */
+#define PORTTABTHRESH 10 /* threshold at which we will switch
+ * from using getservbyport() to
+ * getservent() -- see lkup_port()
+ * and fill_porttab() */
+
+struct hostcache {
+ unsigned char a[MAX_AF_ADDR]; /* numeric address */
+ int af; /* address family -- e.g., AF_INET
+ * or AF_INET6 */
+ char *name; /* name */
+};
+
+struct porttab {
+ int port;
+ MALLOC_S nl; /* name length (excluding '\0') */
+ int ss; /* service name status, 0 = lookup not
+ * yet performed */
+ char *name;
+ struct porttab *next;
+};
+
+
+static struct porttab **Pth[4] = { NULL, NULL, NULL, NULL };
+ /* port hash buckets:
+ * Pth[0] for TCP service names
+ * Pth[1] for UDP service names
+ * Pth[2] for TCP portmap info
+ * Pth[3] for UDP portmap info
+ */
+#define HASHPORT(p) (((((int)(p)) * 31415) >> 3) & (PORTHASHBUCKETS - 1))
+
+
+_PROTOTYPE(static void fill_portmap,(void));
+_PROTOTYPE(static void fill_porttab,(void));
+_PROTOTYPE(static char *lkup_port,(int p, int pr, int src));
+_PROTOTYPE(static char *lkup_svcnam,(int h, int p, int pr, int ss));
+_PROTOTYPE(static int printinaddr,(void));
+_PROTOTYPE(static void update_portmap,(struct porttab *pt, char *pn));
+
+
+/*
+ * endnm() - locate end of Namech
+ */
+
+char *
+endnm(sz)
+ size_t *sz; /* returned remaining size */
+{
+ register char *s;
+ register size_t tsz;
+
+ for (s = Namech, tsz = Namechl; *s; s++, tsz--)
+ ;
+ *sz = tsz;
+ return(s);
+}
+
+
+/*
+ * fill_portmap() -- fill the RPC portmap program name table via a conversation
+ * with the portmapper
+ *
+ * The following copyright notice acknowledges that this function was adapted
+ * from getrpcportnam() of the source code of the OpenBSD netstat program.
+ */
+
+/*
+* Copyright (c) 1983, 1988, 1993
+* The Regents of the University of California. All rights reserved.
+*
+* Redistribution and use in source and binary forms, with or without
+* modification, are permitted provided that the following conditions
+* are met:
+* 1. Redistributions of source code must retain the above copyright
+* notice, this list of conditions and the following disclaimer.
+* 2. Redistributions in binary form must reproduce the above copyright
+* notice, this list of conditions and the following disclaimer in the
+* documentation and/or other materials provided with the distribution.
+* 3. All advertising materials mentioning features or use of this software
+* must display the following acknowledgement:
+* This product includes software developed by the University of
+* California, Berkeley and its contributors.
+* 4. Neither the name of the University nor the names of its contributors
+* may be used to endorse or promote products derived from this software
+* without specific prior written permission.
+*
+* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+* SUCH DAMAGE.
+*/
+
+static void
+fill_portmap()
+{
+ char buf[128], *cp, *nm;
+ CLIENT *c;
+ int h, port, pr;
+ MALLOC_S nl;
+ struct pmaplist *p = (struct pmaplist *)NULL;
+ struct porttab *pt;
+ struct rpcent *r;
+ struct TIMEVAL_LSOF tm;
+
+#if !defined(CAN_USE_CLNT_CREATE)
+ struct hostent *he;
+ struct sockaddr_in ia;
+ int s = RPC_ANYSOCK;
+#endif /* !defined(CAN_USE_CLNT_CREATE) */
+
+/*
+ * Construct structures for communicating with the portmapper.
+ */
+
+#if !defined(CAN_USE_CLNT_CREATE)
+ zeromem(&ia, sizeof(ia));
+ ia.sin_family = AF_INET;
+ if ((he = gethostbyname("localhost")))
+ MEMMOVE((caddr_t)&ia.sin_addr, he->h_addr, he->h_length);
+ ia.sin_port = htons(PMAPPORT);
+#endif /* !defined(CAN_USE_CLNT_CREATE) */
+
+ tm.tv_sec = 60;
+ tm.tv_usec = 0;
+/*
+ * Get an RPC client handle. Then ask for a dump of the port map.
+ */
+
+#if defined(CAN_USE_CLNT_CREATE)
+ if (!(c = clnt_create("localhost", PMAPPROG, PMAPVERS, "tcp")))
+#else /* !defined(CAN_USE_CLNT_CREATE) */
+ if (!(c = clnttcp_create(&ia, PMAPPROG, PMAPVERS, &s, 0, 0)))
+#endif /* defined(CAN_USE_CLNT_CREATE) */
+
+ return;
+ if (clnt_call(c, PMAPPROC_DUMP, XDR_VOID, NULL, XDR_PMAPLIST,
+ (caddr_t)&p, tm)
+ != RPC_SUCCESS) {
+ clnt_destroy(c);
+ return;
+ }
+/*
+ * Loop through the port map dump, creating portmap table entries from TCP
+ * and UDP members.
+ */
+ for (; p; p = p->pml_next) {
+
+ /*
+ * Determine the port map entry's protocol; ignore all but TCP and UDP.
+ */
+ if (p->pml_map.pm_prot == IPPROTO_TCP)
+ pr = 2;
+ else if (p->pml_map.pm_prot == IPPROTO_UDP)
+ pr = 3;
+ else
+ continue;
+ /*
+ * See if there's already a portmap entry for this port. If there is,
+ * ignore this entry.
+ */
+ h = HASHPORT((port = (int)p->pml_map.pm_port));
+ for (pt = Pth[pr][h]; pt; pt = pt->next) {
+ if (pt->port == port)
+ break;
+ }
+ if (pt)
+ continue;
+ /*
+ * Save the registration name or number.
+ */
+ cp = (char *)NULL;
+ if ((r = (struct rpcent *)getrpcbynumber(p->pml_map.pm_prog))) {
+ if (r->r_name && strlen(r->r_name))
+ cp = r->r_name;
+ }
+ if (!cp) {
+ (void) snpf(buf, sizeof(buf), "%lu",
+ (unsigned long)p->pml_map.pm_prog);
+ cp = buf;
+ }
+ if (!strlen(cp))
+ continue;
+ /*
+ * Allocate space for the portmap name entry and copy it there.
+ */
+ if (!(nm = mkstrcpy(cp, &nl))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate space for portmap entry: ", Pn);
+ safestrprt(cp, stderr, 1);
+ Exit(1);
+ }
+ if (!nl) {
+ (void) free((FREE_P *)nm);
+ continue;
+ }
+ /*
+ * Allocate and fill a porttab struct entry for the portmap table.
+ * Link it to the head of its hash bucket, and make it the new head.
+ */
+ if (!(pt = (struct porttab *)malloc(sizeof(struct porttab)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate porttab entry for portmap: ", Pn);
+ safestrprt(nm, stderr, 1);
+ Exit(1);
+ }
+ pt->name = nm;
+ pt->nl = nl;
+ pt->port = port;
+ pt->next = Pth[pr][h];
+ pt->ss = 0;
+ Pth[pr][h] = pt;
+ }
+ clnt_destroy(c);
+}
+
+
+/*
+ * fill_porttab() -- fill the TCP and UDP service name port table with a
+ * getservent() scan
+ */
+
+static void
+fill_porttab()
+{
+ int h, p, pr;
+ MALLOC_S nl;
+ char *nm;
+ struct porttab *pt;
+ struct servent *se;
+
+ (void) endservent();
+/*
+ * Scan the services data base for TCP and UDP entries that have a non-null
+ * name associated with them.
+ */
+ (void) setservent(1);
+ while ((se = getservent())) {
+ if (!se->s_name || !se->s_proto)
+ continue;
+ if (strcasecmp(se->s_proto, "TCP") == 0)
+ pr = 0;
+ else if (strcasecmp(se->s_proto, "UDP") == 0)
+ pr = 1;
+ else
+ continue;
+ if (!se->s_name || !strlen(se->s_name))
+ continue;
+ p = ntohs(se->s_port);
+ /*
+ * See if a port->service entry is already cached for this port and
+ * prototcol. If it is, leave it alone.
+ */
+ h = HASHPORT(p);
+ for (pt = Pth[pr][h]; pt; pt = pt->next) {
+ if (pt->port == p)
+ break;
+ }
+ if (pt)
+ continue;
+ /*
+ * Add a new entry to the cache for this port and protocol.
+ */
+ if (!(nm = mkstrcpy(se->s_name, &nl))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for port %d name: %s\n",
+ Pn, (int)(nl + 1), p, se->s_name);
+ Exit(1);
+ }
+ if (!nl) {
+ (void) free((FREE_P *)nm);
+ continue;
+ }
+ if (!(pt = (struct porttab *)malloc(sizeof(struct porttab)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate porttab entry for port %d: %s\n",
+ Pn, p, se->s_name);
+ Exit(1);
+ }
+ pt->name = nm;
+ pt->nl = nl - 1;
+ pt->port = p;
+ pt->next = Pth[pr][h];
+ pt->ss = 0;
+ Pth[pr][h] = pt;
+ }
+ (void) endservent();
+}
+
+
+/*
+ * gethostnm() - get host name
+ */
+
+char *
+gethostnm(ia, af)
+ unsigned char *ia; /* Internet address */
+ int af; /* address family -- e.g., AF_INET
+ * or AF_INET6 */
+{
+ int al = MIN_AF_ADDR;
+ char hbuf[256];
+ static struct hostcache *hc = (struct hostcache *)NULL;
+ static int hcx = 0;
+ char *hn, *np;
+ struct hostent *he = (struct hostent *)NULL;
+ int i, j;
+ MALLOC_S len;
+ static int nhc = 0;
+/*
+ * Search cache.
+ */
+
+#if defined(HASIPv6)
+ if (af == AF_INET6)
+ al = MAX_AF_ADDR;
+#endif /* defined(HASIPv6) */
+
+ for (i = 0; i < hcx; i++) {
+ if (af != hc[i].af)
+ continue;
+ for (j = 0; j < al; j++) {
+ if (ia[j] != hc[i].a[j])
+ break;
+ }
+ if (j >= al)
+ return(hc[i].name);
+ }
+/*
+ * If -n has been specified, construct a numeric address. Otherwise, look up
+ * host name by address. If that fails, or if there is no name in the returned
+ * hostent structure, construct a numeric version of the address.
+ */
+ if (Fhost)
+ he = gethostbyaddr((char *)ia, al, af);
+ if (!he || !he->h_name) {
+
+#if defined(HASIPv6)
+ if (af == AF_INET6) {
+
+ /*
+ * Since IPv6 numeric addresses use `:' as a separator, enclose
+ * them in brackets.
+ */
+ hbuf[0] = '[';
+ if (!inet_ntop(af, ia, hbuf + 1, sizeof(hbuf) - 3)) {
+ (void) snpf(&hbuf[1], (sizeof(hbuf) - 1),
+ "can't format IPv6 address]");
+ } else {
+ len = strlen(hbuf);
+ (void) snpf(&hbuf[len], sizeof(hbuf) - len, "]");
+ }
+ } else
+#endif /* defined(HASIPv6) */
+
+ if (af == AF_INET)
+ (void) snpf(hbuf, sizeof(hbuf), "%u.%u.%u.%u", ia[0], ia[1],
+ ia[2], ia[3]);
+ else
+ (void) snpf(hbuf, sizeof(hbuf), "(unknown AF value: %d)", af);
+ hn = hbuf;
+ } else
+ hn = (char *)he->h_name;
+/*
+ * Allocate space for name and copy name to it.
+ */
+ if (!(np = mkstrcpy(hn, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: no space for host name: ", Pn);
+ safestrprt(hn, stderr, 1);
+ Exit(1);
+ }
+/*
+ * Add address/name entry to cache. Allocate cache space in HCINC chunks.
+ */
+ if (hcx >= nhc) {
+ nhc += HCINC;
+ len = (MALLOC_S)(nhc * sizeof(struct hostcache));
+ if (!hc)
+ hc = (struct hostcache *)malloc(len);
+ else
+ hc = (struct hostcache *)realloc((MALLOC_P *)hc, len);
+ if (!hc) {
+ (void) fprintf(stderr, "%s: no space for host cache\n", Pn);
+ Exit(1);
+ }
+ }
+ hc[hcx].af = af;
+ for (i = 0; i < al; i++) {
+ hc[hcx].a[i] = ia[i];
+ }
+ hc[hcx++].name = np;
+ return(np);
+}
+
+
+/*
+ * lkup_port() - look up port for protocol
+ */
+
+static char *
+lkup_port(p, pr, src)
+ int p; /* port number */
+ int pr; /* protocol index: 0 = tcp, 1 = udp */
+ int src; /* port source: 0 = local
+ * 1 = foreign */
+{
+ int h, nh;
+ MALLOC_S nl;
+ char *nm, *pn;
+ static char pb[128];
+ static int pm = 0;
+ struct porttab *pt;
+/*
+ * If the hash buckets haven't been allocated, do so.
+ */
+ if (!Pth[0]) {
+ nh = FportMap ? 4 : 2;
+ for (h = 0; h < nh; h++) {
+ if (!(Pth[h] = (struct porttab **)calloc(PORTHASHBUCKETS,
+ sizeof(struct porttab *))))
+ {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for %s %s hash buckets\n",
+ Pn,
+ (int)(2 * (PORTHASHBUCKETS * sizeof(struct porttab *))),
+ (h & 1) ? "UDP" : "TCP",
+ (h > 1) ? "portmap" : "port");
+ Exit(1);
+ }
+ }
+ }
+/*
+ * If we're looking up program names for portmapped ports, make sure the
+ * portmap table has been loaded.
+ */
+ if (FportMap && !pm) {
+ (void) fill_portmap();
+ pm++;
+ }
+/*
+ * Hash the port and see if its name has been cached. Look for a local
+ * port first in the portmap, if portmap searching is enabled.
+ */
+ h = HASHPORT(p);
+ if (!src && FportMap) {
+ for (pt = Pth[pr+2][h]; pt; pt = pt->next) {
+ if (pt->port != p)
+ continue;
+ if (!pt->ss) {
+ pn = Fport ? lkup_svcnam(h, p, pr, 0) : (char *)NULL;
+ if (!pn) {
+ (void) snpf(pb, sizeof(pb), "%d", p);
+ pn = pb;
+ }
+ (void) update_portmap(pt, pn);
+ }
+ return(pt->name);
+ }
+ }
+ for (pt = Pth[pr][h]; pt; pt = pt->next) {
+ if (pt->port == p)
+ return(pt->name);
+ }
+/*
+ * Search for a possible service name, unless the -P option has been specified.
+ *
+ * If there is no service name, return a %d conversion.
+ *
+ * Don't cache %d conversions; a zero port number is a %d conversion that
+ * is represented by "*".
+ */
+ pn = Fport ? lkup_svcnam(h, p, pr, 1) : (char *)NULL;
+ if (!pn || !strlen(pn)) {
+ if (p) {
+ (void) snpf(pb, sizeof(pb), "%d", p);
+ return(pb);
+ } else
+ return("*");
+ }
+/*
+ * Allocate a new porttab entry for the TCP or UDP service name.
+ */
+ if (!(pt = (struct porttab *)malloc(sizeof(struct porttab)))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate porttab entry for port %d\n", Pn, p);
+ Exit(1);
+ }
+/*
+ * Allocate space for the name; copy it to the porttab entry; and link the
+ * porttab entry to its hash bucket.
+ *
+ * Return a pointer to the name.
+ */
+ if (!(nm = mkstrcpy(pn, &nl))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate space for port name: ", Pn);
+ safestrprt(pn, stderr, 1);
+ Exit(1);
+ }
+ pt->name = nm;
+ pt->nl = nl;
+ pt->port = p;
+ pt->next = Pth[pr][h];
+ pt->ss = 0;
+ Pth[pr][h] = pt;
+ return(nm);
+}
+
+
+/*
+ * lkup_svcnam() - look up service name for port
+ */
+
+static char *
+lkup_svcnam(h, p, pr, ss)
+ int h; /* porttab hash index */
+ int p; /* port number */
+ int pr; /* protocol: 0 = TCP, 1 = UDP */
+ int ss; /* search status: 1 = Pth[pr][h]
+ * already searched */
+{
+ static int fl[PORTTABTHRESH];
+ static int fln = 0;
+ static int gsbp = 0;
+ int i;
+ struct porttab *pt;
+ static int ptf = 0;
+ struct servent *se;
+/*
+ * Do nothing if -P has been specified.
+ */
+ if (!Fport)
+ return((char *)NULL);
+
+ for (;;) {
+
+ /*
+ * Search service name cache, if it hasn't already been done.
+ * Return the name of a match.
+ */
+ if (!ss) {
+ for (pt = Pth[pr][h]; pt; pt = pt->next) {
+ if (pt->port == p)
+ return(pt->name);
+ }
+ }
+/*
+ * If fill_porttab() has been called, there is no service name.
+ *
+ * Do PORTTABTHRES getservbport() calls, remembering the failures, so they
+ * won't be repeated.
+ *
+ * After PORTABTHRESH getservbyport() calls, call fill_porttab() once,
+ */
+ if (ptf)
+ break;
+ if (gsbp < PORTTABTHRESH) {
+ for (i = 0; i < fln; i++) {
+ if (fl[i] == p)
+ return((char *)NULL);
+ }
+ gsbp++;
+ if ((se = getservbyport(htons(p), pr ? "udp" : "tcp")))
+ return(se->s_name);
+ if (fln < PORTTABTHRESH)
+ fl[fln++] = p;
+ return((char *)NULL);
+ }
+ (void) fill_porttab();
+ ptf++;
+ ss = 0;
+ }
+ return((char *)NULL);
+}
+
+
+/*
+ * print_file() - print file
+ */
+
+void
+print_file()
+{
+ char buf[128];
+ char *cp = (char *)NULL;
+ dev_t dev;
+ int devs, len;
+
+ if (PrPass && !Hdr) {
+
+ /*
+ * Print the header line if this is the second pass and the
+ * header hasn't already been printed.
+ */
+ (void) printf("%-*.*s %*s", CmdColW, CmdColW, CMDTTL, PidColW,
+ PIDTTL);
+
+#if defined(HASZONES)
+ if (Fzone)
+ (void) printf(" %-*s", ZoneColW, ZONETTL);
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+ if (Fcntx)
+ (void) printf(" %-*s", CntxColW, CNTXTTL);
+#endif /* defined(HASSELINUX) */
+
+#if defined(HASPPID)
+ if (Fppid)
+ (void) printf(" %*s", PpidColW, PPIDTTL);
+#endif /* defined(HASPPID) */
+
+ if (Fpgid)
+ (void) printf(" %*s", PgidColW, PGIDTTL);
+ (void) printf(" %*s %*s %*s",
+ UserColW, USERTTL,
+ FdColW - 2, FDTTL,
+ TypeColW, TYPETTL);
+
+#if defined(HASFSTRUCT)
+ if (Fsv) {
+
+# if !defined(HASNOFSADDR)
+ if (Fsv & FSV_FA)
+ (void) printf(" %*s", FsColW, FSTTL);
+# endif /* !defined(HASNOFSADDR) */
+
+# if !defined(HASNOFSCOUNT)
+ if (Fsv & FSV_CT)
+ (void) printf(" %*s", FcColW, FCTTL);
+# endif /* !defined(HASNOFSCOUNT) */
+
+# if !defined(HASNOFSFLAGS)
+ if (Fsv & FSV_FG)
+ (void) printf(" %*s", FgColW, FGTTL);
+# endif /* !defined(HASNOFSFLAGS) */
+
+# if !defined(HASNOFSNADDR)
+ if (Fsv & FSV_NI)
+ (void) printf(" %*s", NiColW, NiTtl);
+# endif /* !defined(HASNOFSNADDR) */
+
+ }
+#endif /* defined(HASFSTRUCT) */
+
+ (void) printf(" %*s", DevColW, DEVTTL);
+ if (Foffset)
+ (void) printf(" %*s", SzOffColW, OFFTTL);
+ else if (Fsize)
+ (void) printf(" %*s", SzOffColW, SZTTL);
+ else
+ (void) printf(" %*s", SzOffColW, SZOFFTTL);
+ if (Fnlink)
+ (void) printf(" %*s", NlColW, NLTTL);
+ (void) printf(" %*s %s\n", NodeColW, NODETTL, NMTTL);
+ Hdr++;
+ }
+/*
+ * Size or print the command.
+ */
+ cp = (Lp->cmd && *Lp->cmd != '\0') ? Lp->cmd : "(unknown)";
+ if (!PrPass) {
+ len = safestrlen(cp, 2);
+ if (CmdLim && (len > CmdLim))
+ len = CmdLim;
+ if (len > CmdColW)
+ CmdColW = len;
+ } else
+ safestrprtn(cp, CmdColW, stdout, 2);
+/*
+ * Size or print the process ID.
+ */
+ if (!PrPass) {
+ (void) snpf(buf, sizeof(buf), "%d", Lp->pid);
+ if ((len = strlen(buf)) > PidColW)
+ PidColW = len;
+ } else
+ (void) printf(" %*d", PidColW, Lp->pid);
+
+#if defined(HASZONES)
+/*
+ * Size or print the zone.
+ */
+ if (Fzone) {
+ if (!PrPass) {
+ if (Lp->zn) {
+ if ((len = strlen(Lp->zn)) > ZoneColW)
+ ZoneColW = len;
+ }
+ } else
+ (void) printf(" %-*s", ZoneColW, Lp->zn ? Lp->zn : "");
+ }
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+/*
+ * Size or print the context.
+ */
+ if (Fcntx) {
+ if (!PrPass) {
+ if (Lp->cntx) {
+ if ((len = strlen(Lp->cntx)) > CntxColW)
+ CntxColW = len;
+ }
+ } else
+ (void) printf(" %-*s", CntxColW, Lp->cntx ? Lp->cntx : "");
+ }
+#endif /* defined(HASSELINUX) */
+
+#if defined(HASPPID)
+ if (Fppid) {
+
+ /*
+ * Size or print the parent process ID.
+ */
+ if (!PrPass) {
+ (void) snpf(buf, sizeof(buf), "%d", Lp->ppid);
+ if ((len = strlen(buf)) > PpidColW)
+ PpidColW = len;
+ } else
+ (void) printf(" %*d", PpidColW, Lp->ppid);
+ }
+#endif /* defined(HASPPID) */
+
+ if (Fpgid) {
+
+ /*
+ * Size or print the process group ID.
+ */
+ if (!PrPass) {
+ (void) snpf(buf, sizeof(buf), "%d", Lp->pgid);
+ if ((len = strlen(buf)) > PgidColW)
+ PgidColW = len;
+ } else
+ (void) printf(" %*d", PgidColW, Lp->pgid);
+ }
+/*
+ * Size or print the user ID or login name.
+ */
+ if (!PrPass) {
+ if ((len = strlen(printuid((UID_ARG)Lp->uid, NULL))) > UserColW)
+ UserColW = len;
+ } else
+ (void) printf(" %*.*s", UserColW, UserColW,
+ printuid((UID_ARG)Lp->uid, NULL));
+/*
+ * Size or print the file descriptor, access mode and lock status.
+ */
+ if (!PrPass) {
+ (void) snpf(buf, sizeof(buf), "%s%c%c",
+ Lf->fd,
+ (Lf->lock == ' ') ? Lf->access
+ : (Lf->access == ' ') ? '-'
+ : Lf->access,
+ Lf->lock);
+ if ((len = strlen(buf)) > FdColW)
+ FdColW = len;
+ } else
+ (void) printf(" %*.*s%c%c", FdColW - 2, FdColW - 2, Lf->fd,
+ (Lf->lock == ' ') ? Lf->access
+ : (Lf->access == ' ') ? '-'
+ : Lf->access,
+ Lf->lock);
+/*
+ * Size or print the type.
+ */
+ if (!PrPass) {
+ if ((len = strlen(Lf->type)) > TypeColW)
+ TypeColW = len;
+ } else
+ (void) printf(" %*.*s", TypeColW, TypeColW, Lf->type);
+
+#if defined(HASFSTRUCT)
+/*
+ * Size or print the file structure address, file usage count, and node
+ * ID (address).
+ */
+
+ if (Fsv) {
+
+# if !defined(HASNOFSADDR)
+ if (Fsv & FSV_FA) {
+ cp = (Lf->fsv & FSV_FA) ? print_kptr(Lf->fsa, buf, sizeof(buf))
+ : "";
+ if (!PrPass) {
+ if ((len = strlen(cp)) > FsColW)
+ FsColW = len;
+ } else
+ (void) printf(" %*.*s", FsColW, FsColW, cp);
+
+ }
+# endif /* !defined(HASNOFSADDR) */
+
+# if !defined(HASNOFSCOUNT)
+ if (Fsv & FSV_CT) {
+ if (Lf->fsv & FSV_CT) {
+ (void) snpf(buf, sizeof(buf), "%ld", Lf->fct);
+ cp = buf;
+ } else
+ cp = "";
+ if (!PrPass) {
+ if ((len = strlen(cp)) > FcColW)
+ FcColW = len;
+ } else
+ (void) printf(" %*.*s", FcColW, FcColW, cp);
+ }
+# endif /* !defined(HASNOFSCOUNT) */
+
+# if !defined(HASNOFSFLAGS)
+ if (Fsv & FSV_FG) {
+ if ((Lf->fsv & FSV_FG) && (FsvFlagX || Lf->ffg || Lf->pof))
+ cp = print_fflags(Lf->ffg, Lf->pof);
+ else
+ cp = "";
+ if (!PrPass) {
+ if ((len = strlen(cp)) > FgColW)
+ FgColW = len;
+ } else
+ (void) printf(" %*.*s", FgColW, FgColW, cp);
+ }
+# endif /* !defined(HASNOFSFLAGS) */
+
+# if !defined(HASNOFSNADDR)
+ if (Fsv & FSV_NI) {
+ cp = (Lf->fsv & FSV_NI) ? print_kptr(Lf->fna, buf, sizeof(buf))
+ : "";
+ if (!PrPass) {
+ if ((len = strlen(cp)) > NiColW)
+ NiColW = len;
+ } else
+ (void) printf(" %*.*s", NiColW, NiColW, cp);
+ }
+# endif /* !defined(HASNOFSNADDR) */
+
+ }
+#endif /* defined(HASFSTRUCT) */
+
+/*
+ * Size or print the device information.
+ */
+
+ if (Lf->rdev_def) {
+ dev = Lf->rdev;
+ devs = 1;
+ } else if (Lf->dev_def) {
+ dev = Lf->dev;
+ devs = 1;
+ } else
+ devs = 0;
+ if (devs) {
+
+#if defined(HASPRINTDEV)
+ cp = HASPRINTDEV(Lf, &dev);
+#else /* !defined(HASPRINTDEV) */
+ (void) snpf(buf, sizeof(buf), "%u,%u", GET_MAJ_DEV(dev),
+ GET_MIN_DEV(dev));
+ cp = buf;
+#endif /* defined(HASPRINTDEV) */
+
+ }
+
+ if (!PrPass) {
+ if (devs)
+ len = strlen(cp);
+ else if (Lf->dev_ch)
+ len = strlen(Lf->dev_ch);
+ else
+ len = 0;
+ if (len > DevColW)
+ DevColW = len;
+ } else {
+ if (devs)
+ (void) printf(" %*.*s", DevColW, DevColW, cp);
+ else {
+ if (Lf->dev_ch)
+ (void) printf(" %*.*s", DevColW, DevColW, Lf->dev_ch);
+ else
+ (void) printf(" %*.*s", DevColW, DevColW, "");
+ }
+ }
+/*
+ * Size or print the size or offset.
+ */
+ if (!PrPass) {
+ if (Lf->sz_def) {
+
+#if defined(HASPRINTSZ)
+ cp = HASPRINTSZ(Lf);
+#else /* !defined(HASPRINTSZ) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_d, Lf->sz);
+ cp = buf;
+#endif /* defined(HASPRINTSZ) */
+
+ len = strlen(cp);
+ } else if (Lf->off_def) {
+
+#if defined(HASPRINTOFF)
+ cp = HASPRINTOFF(Lf, 0);
+#else /* !defined(HASPRINTOFF) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_0t, Lf->off);
+ cp = buf;
+#endif /* defined(HASPRINTOFF) */
+
+ len = strlen(cp);
+ if (OffDecDig && len > (OffDecDig + 2)) {
+
+#if defined(HASPRINTOFF)
+ cp = HASPRINTOFF(Lf, 1);
+#else /* !defined(HASPRINTOFF) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_x, Lf->off);
+ cp = buf;
+#endif /* defined(HASPRINTOFF) */
+
+ len = strlen(cp);
+ }
+ } else
+ len = 0;
+ if (len > SzOffColW)
+ SzOffColW = len;
+ } else {
+ putchar(' ');
+ if (Lf->sz_def)
+
+#if defined(HASPRINTSZ)
+ (void) printf("%*.*s", SzOffColW, SzOffColW, HASPRINTSZ(Lf));
+#else /* !defined(HASPRINTSZ) */
+ (void) printf(SzOffFmt_dv, SzOffColW, Lf->sz);
+#endif /* defined(HASPRINTSZ) */
+
+ else if (Lf->off_def) {
+
+#if defined(HASPRINTOFF)
+ cp = HASPRINTOFF(Lf, 0);
+#else /* !defined(HASPRINTOFF) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_0t, Lf->off);
+ cp = buf;
+#endif /* defined(HASPRINTOFF) */
+
+ if (OffDecDig && (int)strlen(cp) > (OffDecDig + 2)) {
+
+#if defined(HASPRINTOFF)
+ cp = HASPRINTOFF(Lf, 1);
+#else /* !defined(HASPRINTOFF) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_x, Lf->off);
+ cp = buf;
+#endif /* defined(HASPRINTOFF) */
+
+ }
+ (void) printf("%*.*s", SzOffColW, SzOffColW, cp);
+ } else
+ (void) printf("%*.*s", SzOffColW, SzOffColW, "");
+ }
+/*
+ * Size or print the link count.
+ */
+ if (Fnlink) {
+ if (Lf->nlink_def) {
+ (void) snpf(buf, sizeof(buf), " %ld", Lf->nlink);
+ cp = buf;
+ } else
+ cp = "";
+ if (!PrPass) {
+ if ((len = strlen(cp)) > NlColW)
+ NlColW = len;
+ } else
+ (void) printf(" %*s", NlColW, cp);
+ }
+/*
+ * Size or print the inode information.
+ */
+ switch (Lf->inp_ty) {
+ case 1:
+
+#if defined(HASPRINTINO)
+ cp = HASPRINTINO(Lf);
+#else /* !defined(HASPRINTINO) */
+ (void) snpf(buf, sizeof(buf), InodeFmt_d, Lf->inode);
+ cp = buf;
+#endif /* defined(HASPRINTINO) */
+
+ break;
+ case 2:
+ if (Lf->iproto[0])
+ cp = Lf->iproto;
+ else
+ cp = "";
+ break;
+ case 3:
+ (void) snpf(buf, sizeof(buf), InodeFmt_x, Lf->inode);
+ cp = buf;
+ break;
+ default:
+ cp = "";
+ }
+ if (!PrPass) {
+ if ((len = strlen(cp)) > NodeColW)
+ NodeColW = len;
+ } else {
+ (void) printf(" %*.*s", NodeColW, NodeColW, cp);
+ }
+/*
+ * If this is the second pass, print the name column. (It doesn't need
+ * to be sized.)
+ */
+ if (PrPass) {
+ putchar(' ');
+
+#if defined(HASPRINTNM)
+ HASPRINTNM(Lf);
+#else /* !defined(HASPRINTNM) */
+ printname(1);
+#endif /* defined(HASPRINTNM) */
+
+ }
+}
+
+
+/*
+ * printinaddr() - print Internet addresses
+ */
+
+static int
+printinaddr()
+{
+ int i, len, src;
+ char *host, *port;
+ int nl = Namechl - 1;
+ char *np = Namech;
+ char pbuf[32];
+/*
+ * Process local network address first. If there's a foreign address,
+ * separate it from the local address with "->".
+ */
+ for (i = 0, *np = '\0'; i < 2; i++) {
+ if (!Lf->li[i].af)
+ continue;
+ host = port = (char *)NULL;
+ if (i) {
+
+ /*
+ * If this is the foreign address, insert the separator.
+ */
+ if (nl < 2)
+
+addr_too_long:
+
+ {
+ (void) snpf(Namech, Namechl,
+ "network addresses too long");
+ return(1);
+ }
+ (void) snpf(np, nl, "->");
+ np += 2;
+ nl -= 2;
+ }
+ /*
+ * Convert the address to a host name.
+ */
+
+#if defined(HASIPv6)
+ if ((Lf->li[i].af == AF_INET6
+ && IN6_IS_ADDR_UNSPECIFIED(&Lf->li[i].ia.a6))
+ || (Lf->li[i].af == AF_INET
+ && Lf->li[i].ia.a4.s_addr == INADDR_ANY))
+ host ="*";
+ else
+ host = gethostnm((unsigned char *)&Lf->li[i].ia, Lf->li[i].af);
+#else /* !defined(HASIPv6) */
+ if (Lf->li[i].ia.a4.s_addr == INADDR_ANY)
+ host ="*";
+ else
+ host = gethostnm((unsigned char *)&Lf->li[i].ia, Lf->li[i].af);
+#endif /* defined(HASIPv6) */
+
+ /*
+ * Process the port number.
+ */
+ if (Lf->li[i].p > 0) {
+ if (Fport || FportMap) {
+
+ /*
+ * If converting port numbers to service names, or looking
+ * up portmap program names and numbers, do so by protocol.
+ *
+ * Identify the port source as local if: 1) it comes from the
+ * local entry (0) of the file's Internet address array; or
+ * 2) it comes from the foreign entry (1), and the foreign
+ * Internet address matches the local one; or 3) it is the
+ * loopback address 127.0.0.1. (Test 2 may not always work
+ * -- e.g., on hosts with multiple interfaces.)
+ */
+ if ((src = i) && FportMap) {
+
+#if defined(HASIPv6)
+ if (Lf->li[0].af == AF_INET6) {
+ if (IN6_IS_ADDR_LOOPBACK(&Lf->li[i].ia.a6)
+ || IN6_ARE_ADDR_EQUAL(&Lf->li[0].ia.a6,
+ &Lf->li[1].ia.a6)
+ )
+ src = 0;
+ } else
+#endif /* defined(HASIPv6) */
+
+ if (Lf->li[0].af == AF_INET) {
+ if (Lf->li[i].ia.a4.s_addr == htonl(INADDR_LOOPBACK)
+ || Lf->li[0].ia.a4.s_addr == Lf->li[1].ia.a4.s_addr
+ )
+ src = 0;
+ }
+ }
+ if (strcasecmp(Lf->iproto, "TCP") == 0)
+ port = lkup_port(Lf->li[i].p, 0, src);
+ else if (strcasecmp(Lf->iproto, "UDP") == 0)
+ port = lkup_port(Lf->li[i].p, 1, src);
+ }
+ if (!port) {
+ (void) snpf(pbuf, sizeof(pbuf), "%d", Lf->li[i].p);
+ port = pbuf;
+ }
+ } else if (Lf->li[i].p == 0)
+ port = "*";
+ /*
+ * Enter the host name.
+ */
+ if (host) {
+ if ((len = strlen(host)) > nl)
+ goto addr_too_long;
+ if (len) {
+ (void) snpf(np, nl, "%s", host);
+ np += len;
+ nl -= len;
+ }
+ }
+ /*
+ * Enter the port number, preceded by a colon.
+ */
+ if (port) {
+ if (((len = strlen(port)) + 1) >= nl)
+ goto addr_too_long;
+ (void) snpf(np, nl, ":%s", port);
+ np += len + 1;
+ nl -= len - 1;
+ }
+ }
+ if (Namech[0]) {
+ safestrprt(Namech, stdout, 0);
+ return(1);
+ }
+ return(0);
+}
+
+
+/*
+ * print_init() - initialize for printing
+ */
+
+void
+print_init()
+{
+ PrPass = (Ffield || Fterse) ? 1 : 0;
+ CmdColW = strlen(CMDTTL);
+ DevColW = strlen(DEVTTL);
+ FdColW = strlen(FDTTL);
+ if (Fnlink)
+ NlColW = strlen(NLTTL);
+ NmColW = strlen(NMTTL);
+ NodeColW = strlen(NODETTL);
+ PgidColW = strlen(PGIDTTL);
+ PidColW = strlen(PIDTTL);
+ PpidColW = strlen(PPIDTTL);
+ if (Fsize)
+ SzOffColW = strlen(SZTTL);
+ else if (Foffset)
+ SzOffColW = strlen(OFFTTL);
+ else
+ SzOffColW = strlen(SZOFFTTL);
+ TypeColW = strlen(TYPETTL);
+ UserColW = strlen(USERTTL);
+
+#if defined(HASFSTRUCT)
+
+# if !defined(HASNOFSADDR)
+ FsColW = strlen(FSTTL);
+# endif /* !defined(HASNOFSADDR) */
+
+# if !defined(HASNOFSCOUNT)
+ FcColW = strlen(FCTTL);
+# endif /* !defined(HASNOFSCOUNT) */
+
+# if !defined(HASNOFSFLAGS)
+ FgColW = strlen(FGTTL);
+# endif /* !defined(HASNOFSFLAGS) */
+
+# if !defined(HASNOFSNADDR)
+ NiColW = strlen(NiTtl);
+# endif /* !defined(HASNOFSNADDR) */
+#endif /* defined(HASFSTRUCT) */
+
+#if defined(HASSELINUX)
+ if (Fcntx)
+ CntxColW = strlen(CNTXTTL);
+#endif /* defined(HASSELINUX) */
+
+#if defined(HASZONES)
+ if (Fzone)
+ ZoneColW = strlen(ZONETTL);
+#endif /* defined(HASZONES) */
+
+}
+
+
+#if !defined(HASPRIVPRIPP)
+/*
+ * printiproto() - print Internet protocol name
+ */
+
+void
+printiproto(p)
+ int p; /* protocol number */
+{
+ int i;
+ static int m = -1;
+ char *s;
+
+ switch (p) {
+
+#if defined(IPPROTO_TCP)
+ case IPPROTO_TCP:
+ s = "TCP";
+ break;
+#endif /* defined(IPPROTO_TCP) */
+
+#if defined(IPPROTO_UDP)
+ case IPPROTO_UDP:
+ s = "UDP";
+ break;
+#endif /* defined(IPPROTO_UDP) */
+
+#if defined(IPPROTO_IP)
+# if !defined(IPPROTO_HOPOPTS) || IPPROTO_IP!=IPPROTO_HOPOPTS
+ case IPPROTO_IP:
+ s = "IP";
+ break;
+# endif /* !defined(IPPROTO_HOPOPTS) || IPPROTO_IP!=IPPROTO_HOPOPTS */
+#endif /* defined(IPPROTO_IP) */
+
+#if defined(IPPROTO_ICMP)
+ case IPPROTO_ICMP:
+ s = "ICMP";
+ break;
+#endif /* defined(IPPROTO_ICMP) */
+
+#if defined(IPPROTO_ICMPV6)
+ case IPPROTO_ICMPV6:
+ s = "ICMPV6";
+ break;
+#endif /* defined(IPPROTO_ICMPV6) */
+
+#if defined(IPPROTO_IGMP)
+ case IPPROTO_IGMP:
+ s = "IGMP";
+ break;
+#endif /* defined(IPPROTO_IGMP) */
+
+#if defined(IPPROTO_GGP)
+ case IPPROTO_GGP:
+ s = "GGP";
+ break;
+#endif /* defined(IPPROTO_GGP) */
+
+#if defined(IPPROTO_EGP)
+ case IPPROTO_EGP:
+ s = "EGP";
+ break;
+#endif /* defined(IPPROTO_EGP) */
+
+#if defined(IPPROTO_PUP)
+ case IPPROTO_PUP:
+ s = "PUP";
+ break;
+#endif /* defined(IPPROTO_PUP) */
+
+#if defined(IPPROTO_IDP)
+ case IPPROTO_IDP:
+ s = "IDP";
+ break;
+#endif /* defined(IPPROTO_IDP) */
+
+#if defined(IPPROTO_ND)
+ case IPPROTO_ND:
+ s = "ND";
+ break;
+#endif /* defined(IPPROTO_ND) */
+
+#if defined(IPPROTO_RAW)
+ case IPPROTO_RAW:
+ s = "RAW";
+ break;
+#endif /* defined(IPPROTO_RAW) */
+
+#if defined(IPPROTO_HELLO)
+ case IPPROTO_HELLO:
+ s = "HELLO";
+ break;
+#endif /* defined(IPPROTO_HELLO) */
+
+#if defined(IPPROTO_PXP)
+ case IPPROTO_PXP:
+ s = "PXP";
+ break;
+#endif /* defined(IPPROTO_PXP) */
+
+#if defined(IPPROTO_RAWIP)
+ case IPPROTO_RAWIP:
+ s = "RAWIP";
+ break;
+#endif /* defined(IPPROTO_RAWIP) */
+
+#if defined(IPPROTO_RAWIF)
+ case IPPROTO_RAWIF:
+ s = "RAWIF";
+ break;
+#endif /* defined(IPPROTO_RAWIF) */
+
+#if defined(IPPROTO_HOPOPTS)
+ case IPPROTO_HOPOPTS:
+ s = "HOPOPTS";
+ break;
+#endif /* defined(IPPROTO_HOPOPTS) */
+
+#if defined(IPPROTO_IPIP)
+ case IPPROTO_IPIP:
+ s = "IPIP";
+ break;
+#endif /* defined(IPPROTO_IPIP) */
+
+#if defined(IPPROTO_ST)
+ case IPPROTO_ST:
+ s = "ST";
+ break;
+#endif /* defined(IPPROTO_ST) */
+
+#if defined(IPPROTO_PIGP)
+ case IPPROTO_PIGP:
+ s = "PIGP";
+ break;
+#endif /* defined(IPPROTO_PIGP) */
+
+#if defined(IPPROTO_RCCMON)
+ case IPPROTO_RCCMON:
+ s = "RCCMON";
+ break;
+#endif /* defined(IPPROTO_RCCMON) */
+
+#if defined(IPPROTO_NVPII)
+ case IPPROTO_NVPII:
+ s = "NVPII";
+ break;
+#endif /* defined(IPPROTO_NVPII) */
+
+#if defined(IPPROTO_ARGUS)
+ case IPPROTO_ARGUS:
+ s = "ARGUS";
+ break;
+#endif /* defined(IPPROTO_ARGUS) */
+
+#if defined(IPPROTO_EMCON)
+ case IPPROTO_EMCON:
+ s = "EMCON";
+ break;
+#endif /* defined(IPPROTO_EMCON) */
+
+#if defined(IPPROTO_XNET)
+ case IPPROTO_XNET:
+ s = "XNET";
+ break;
+#endif /* defined(IPPROTO_XNET) */
+
+#if defined(IPPROTO_CHAOS)
+ case IPPROTO_CHAOS:
+ s = "CHAOS";
+ break;
+#endif /* defined(IPPROTO_CHAOS) */
+
+#if defined(IPPROTO_MUX)
+ case IPPROTO_MUX:
+ s = "MUX";
+ break;
+#endif /* defined(IPPROTO_MUX) */
+
+#if defined(IPPROTO_MEAS)
+ case IPPROTO_MEAS:
+ s = "MEAS";
+ break;
+#endif /* defined(IPPROTO_MEAS) */
+
+#if defined(IPPROTO_HMP)
+ case IPPROTO_HMP:
+ s = "HMP";
+ break;
+#endif /* defined(IPPROTO_HMP) */
+
+#if defined(IPPROTO_PRM)
+ case IPPROTO_PRM:
+ s = "PRM";
+ break;
+#endif /* defined(IPPROTO_PRM) */
+
+#if defined(IPPROTO_TRUNK1)
+ case IPPROTO_TRUNK1:
+ s = "TRUNK1";
+ break;
+#endif /* defined(IPPROTO_TRUNK1) */
+
+#if defined(IPPROTO_TRUNK2)
+ case IPPROTO_TRUNK2:
+ s = "TRUNK2";
+ break;
+#endif /* defined(IPPROTO_TRUNK2) */
+
+#if defined(IPPROTO_LEAF1)
+ case IPPROTO_LEAF1:
+ s = "LEAF1";
+ break;
+#endif /* defined(IPPROTO_LEAF1) */
+
+#if defined(IPPROTO_LEAF2)
+ case IPPROTO_LEAF2:
+ s = "LEAF2";
+ break;
+#endif /* defined(IPPROTO_LEAF2) */
+
+#if defined(IPPROTO_RDP)
+ case IPPROTO_RDP:
+ s = "RDP";
+ break;
+#endif /* defined(IPPROTO_RDP) */
+
+#if defined(IPPROTO_IRTP)
+ case IPPROTO_IRTP:
+ s = "IRTP";
+ break;
+#endif /* defined(IPPROTO_IRTP) */
+
+#if defined(IPPROTO_TP)
+ case IPPROTO_TP:
+ s = "TP";
+ break;
+#endif /* defined(IPPROTO_TP) */
+
+#if defined(IPPROTO_BLT)
+ case IPPROTO_BLT:
+ s = "BLT";
+ break;
+#endif /* defined(IPPROTO_BLT) */
+
+#if defined(IPPROTO_NSP)
+ case IPPROTO_NSP:
+ s = "NSP";
+ break;
+#endif /* defined(IPPROTO_NSP) */
+
+#if defined(IPPROTO_INP)
+ case IPPROTO_INP:
+ s = "INP";
+ break;
+#endif /* defined(IPPROTO_INP) */
+
+#if defined(IPPROTO_SEP)
+ case IPPROTO_SEP:
+ s = "SEP";
+ break;
+#endif /* defined(IPPROTO_SEP) */
+
+#if defined(IPPROTO_3PC)
+ case IPPROTO_3PC:
+ s = "3PC";
+ break;
+#endif /* defined(IPPROTO_3PC) */
+
+#if defined(IPPROTO_IDPR)
+ case IPPROTO_IDPR:
+ s = "IDPR";
+ break;
+#endif /* defined(IPPROTO_IDPR) */
+
+#if defined(IPPROTO_XTP)
+ case IPPROTO_XTP:
+ s = "XTP";
+ break;
+#endif /* defined(IPPROTO_XTP) */
+
+#if defined(IPPROTO_DDP)
+ case IPPROTO_DDP:
+ s = "DDP";
+ break;
+#endif /* defined(IPPROTO_DDP) */
+
+#if defined(IPPROTO_CMTP)
+ case IPPROTO_CMTP:
+ s = "CMTP";
+ break;
+#endif /* defined(IPPROTO_CMTP) */
+
+#if defined(IPPROTO_TPXX)
+ case IPPROTO_TPXX:
+ s = "TPXX";
+ break;
+#endif /* defined(IPPROTO_TPXX) */
+
+#if defined(IPPROTO_IL)
+ case IPPROTO_IL:
+ s = "IL";
+ break;
+#endif /* defined(IPPROTO_IL) */
+
+#if defined(IPPROTO_IPV6)
+ case IPPROTO_IPV6:
+ s = "IPV6";
+ break;
+#endif /* defined(IPPROTO_IPV6) */
+
+#if defined(IPPROTO_SDRP)
+ case IPPROTO_SDRP:
+ s = "SDRP";
+ break;
+#endif /* defined(IPPROTO_SDRP) */
+
+#if defined(IPPROTO_ROUTING)
+ case IPPROTO_ROUTING:
+ s = "ROUTING";
+ break;
+#endif /* defined(IPPROTO_ROUTING) */
+
+#if defined(IPPROTO_FRAGMENT)
+ case IPPROTO_FRAGMENT:
+ s = "FRAGMNT";
+ break;
+#endif /* defined(IPPROTO_FRAGMENT) */
+
+#if defined(IPPROTO_IDRP)
+ case IPPROTO_IDRP:
+ s = "IDRP";
+ break;
+#endif /* defined(IPPROTO_IDRP) */
+
+#if defined(IPPROTO_RSVP)
+ case IPPROTO_RSVP:
+ s = "RSVP";
+ break;
+#endif /* defined(IPPROTO_RSVP) */
+
+#if defined(IPPROTO_GRE)
+ case IPPROTO_GRE:
+ s = "GRE";
+ break;
+#endif /* defined(IPPROTO_GRE) */
+
+#if defined(IPPROTO_MHRP)
+ case IPPROTO_MHRP:
+ s = "MHRP";
+ break;
+#endif /* defined(IPPROTO_MHRP) */
+
+#if defined(IPPROTO_BHA)
+ case IPPROTO_BHA:
+ s = "BHA";
+ break;
+#endif /* defined(IPPROTO_BHA) */
+
+#if defined(IPPROTO_ESP)
+ case IPPROTO_ESP:
+ s = "ESP";
+ break;
+#endif /* defined(IPPROTO_ESP) */
+
+#if defined(IPPROTO_AH)
+ case IPPROTO_AH:
+ s = "AH";
+ break;
+#endif /* defined(IPPROTO_AH) */
+
+#if defined(IPPROTO_INLSP)
+ case IPPROTO_INLSP:
+ s = "INLSP";
+ break;
+#endif /* defined(IPPROTO_INLSP) */
+
+#if defined(IPPROTO_SWIPE)
+ case IPPROTO_SWIPE:
+ s = "SWIPE";
+ break;
+#endif /* defined(IPPROTO_SWIPE) */
+
+#if defined(IPPROTO_NHRP)
+ case IPPROTO_NHRP:
+ s = "NHRP";
+ break;
+#endif /* defined(IPPROTO_NHRP) */
+
+#if defined(IPPROTO_NONE)
+ case IPPROTO_NONE:
+ s = "NONE";
+ break;
+#endif /* defined(IPPROTO_NONE) */
+
+#if defined(IPPROTO_DSTOPTS)
+ case IPPROTO_DSTOPTS:
+ s = "DSTOPTS";
+ break;
+#endif /* defined(IPPROTO_DSTOPTS) */
+
+#if defined(IPPROTO_AHIP)
+ case IPPROTO_AHIP:
+ s = "AHIP";
+ break;
+#endif /* defined(IPPROTO_AHIP) */
+
+#if defined(IPPROTO_CFTP)
+ case IPPROTO_CFTP:
+ s = "CFTP";
+ break;
+#endif /* defined(IPPROTO_CFTP) */
+
+#if defined(IPPROTO_SATEXPAK)
+ case IPPROTO_SATEXPAK:
+ s = "SATEXPK";
+ break;
+#endif /* defined(IPPROTO_SATEXPAK) */
+
+#if defined(IPPROTO_KRYPTOLAN)
+ case IPPROTO_KRYPTOLAN:
+ s = "KRYPTOL";
+ break;
+#endif /* defined(IPPROTO_KRYPTOLAN) */
+
+#if defined(IPPROTO_RVD)
+ case IPPROTO_RVD:
+ s = "RVD";
+ break;
+#endif /* defined(IPPROTO_RVD) */
+
+#if defined(IPPROTO_IPPC)
+ case IPPROTO_IPPC:
+ s = "IPPC";
+ break;
+#endif /* defined(IPPROTO_IPPC) */
+
+#if defined(IPPROTO_ADFS)
+ case IPPROTO_ADFS:
+ s = "ADFS";
+ break;
+#endif /* defined(IPPROTO_ADFS) */
+
+#if defined(IPPROTO_SATMON)
+ case IPPROTO_SATMON:
+ s = "SATMON";
+ break;
+#endif /* defined(IPPROTO_SATMON) */
+
+#if defined(IPPROTO_VISA)
+ case IPPROTO_VISA:
+ s = "VISA";
+ break;
+#endif /* defined(IPPROTO_VISA) */
+
+#if defined(IPPROTO_IPCV)
+ case IPPROTO_IPCV:
+ s = "IPCV";
+ break;
+#endif /* defined(IPPROTO_IPCV) */
+
+#if defined(IPPROTO_CPNX)
+ case IPPROTO_CPNX:
+ s = "CPNX";
+ break;
+#endif /* defined(IPPROTO_CPNX) */
+
+#if defined(IPPROTO_CPHB)
+ case IPPROTO_CPHB:
+ s = "CPHB";
+ break;
+#endif /* defined(IPPROTO_CPHB) */
+
+#if defined(IPPROTO_WSN)
+ case IPPROTO_WSN:
+ s = "WSN";
+ break;
+#endif /* defined(IPPROTO_WSN) */
+
+#if defined(IPPROTO_PVP)
+ case IPPROTO_PVP:
+ s = "PVP";
+ break;
+#endif /* defined(IPPROTO_PVP) */
+
+#if defined(IPPROTO_BRSATMON)
+ case IPPROTO_BRSATMON:
+ s = "BRSATMN";
+ break;
+#endif /* defined(IPPROTO_BRSATMON) */
+
+#if defined(IPPROTO_WBMON)
+ case IPPROTO_WBMON:
+ s = "WBMON";
+ break;
+#endif /* defined(IPPROTO_WBMON) */
+
+#if defined(IPPROTO_WBEXPAK)
+ case IPPROTO_WBEXPAK:
+ s = "WBEXPAK";
+ break;
+#endif /* defined(IPPROTO_WBEXPAK) */
+
+#if defined(IPPROTO_EON)
+ case IPPROTO_EON:
+ s = "EON";
+ break;
+#endif /* defined(IPPROTO_EON) */
+
+#if defined(IPPROTO_VMTP)
+ case IPPROTO_VMTP:
+ s = "VMTP";
+ break;
+#endif /* defined(IPPROTO_VMTP) */
+
+#if defined(IPPROTO_SVMTP)
+ case IPPROTO_SVMTP:
+ s = "SVMTP";
+ break;
+#endif /* defined(IPPROTO_SVMTP) */
+
+#if defined(IPPROTO_VINES)
+ case IPPROTO_VINES:
+ s = "VINES";
+ break;
+#endif /* defined(IPPROTO_VINES) */
+
+#if defined(IPPROTO_TTP)
+ case IPPROTO_TTP:
+ s = "TTP";
+ break;
+#endif /* defined(IPPROTO_TTP) */
+
+#if defined(IPPROTO_IGP)
+ case IPPROTO_IGP:
+ s = "IGP";
+ break;
+#endif /* defined(IPPROTO_IGP) */
+
+#if defined(IPPROTO_DGP)
+ case IPPROTO_DGP:
+ s = "DGP";
+ break;
+#endif /* defined(IPPROTO_DGP) */
+
+#if defined(IPPROTO_TCF)
+ case IPPROTO_TCF:
+ s = "TCF";
+ break;
+#endif /* defined(IPPROTO_TCF) */
+
+#if defined(IPPROTO_IGRP)
+ case IPPROTO_IGRP:
+ s = "IGRP";
+ break;
+#endif /* defined(IPPROTO_IGRP) */
+
+#if defined(IPPROTO_OSPFIGP)
+ case IPPROTO_OSPFIGP:
+ s = "OSPFIGP";
+ break;
+#endif /* defined(IPPROTO_OSPFIGP) */
+
+#if defined(IPPROTO_SRPC)
+ case IPPROTO_SRPC:
+ s = "SRPC";
+ break;
+#endif /* defined(IPPROTO_SRPC) */
+
+#if defined(IPPROTO_LARP)
+ case IPPROTO_LARP:
+ s = "LARP";
+ break;
+#endif /* defined(IPPROTO_LARP) */
+
+#if defined(IPPROTO_MTP)
+ case IPPROTO_MTP:
+ s = "MTP";
+ break;
+#endif /* defined(IPPROTO_MTP) */
+
+#if defined(IPPROTO_AX25)
+ case IPPROTO_AX25:
+ s = "AX25";
+ break;
+#endif /* defined(IPPROTO_AX25) */
+
+#if defined(IPPROTO_IPEIP)
+ case IPPROTO_IPEIP:
+ s = "IPEIP";
+ break;
+#endif /* defined(IPPROTO_IPEIP) */
+
+#if defined(IPPROTO_MICP)
+ case IPPROTO_MICP:
+ s = "MICP";
+ break;
+#endif /* defined(IPPROTO_MICP) */
+
+#if defined(IPPROTO_SCCSP)
+ case IPPROTO_SCCSP:
+ s = "SCCSP";
+ break;
+#endif /* defined(IPPROTO_SCCSP) */
+
+#if defined(IPPROTO_ETHERIP)
+ case IPPROTO_ETHERIP:
+ s = "ETHERIP";
+ break;
+#endif /* defined(IPPROTO_ETHERIP) */
+
+#if defined(IPPROTO_ENCAP)
+# if !defined(IPPROTO_IPIP) || IPPROTO_IPIP!=IPPROTO_ENCAP
+ case IPPROTO_ENCAP:
+ s = "ENCAP";
+ break;
+# endif /* !defined(IPPROTO_IPIP) || IPPROTO_IPIP!=IPPROTO_ENCAP */
+#endif /* defined(IPPROTO_ENCAP) */
+
+#if defined(IPPROTO_APES)
+ case IPPROTO_APES:
+ s = "APES";
+ break;
+#endif /* defined(IPPROTO_APES) */
+
+#if defined(IPPROTO_GMTP)
+ case IPPROTO_GMTP:
+ s = "GMTP";
+ break;
+#endif /* defined(IPPROTO_GMTP) */
+
+#if defined(IPPROTO_DIVERT)
+ case IPPROTO_DIVERT:
+ s = "DIVERT";
+ break;
+#endif /* defined(IPPROTO_DIVERT) */
+
+ default:
+ s = (char *)NULL;
+ }
+ if (s)
+ (void) snpf(Lf->iproto, sizeof(Lf->iproto), "%.*s", IPROTOL-1, s);
+ else {
+ if (m < 0) {
+ for (i = 0, m = 1; i < IPROTOL-2; i++)
+ m *= 10;
+ }
+ if (m > p)
+ (void) snpf(Lf->iproto, sizeof(Lf->iproto), "%d?", p);
+ else
+ (void) snpf(Lf->iproto, sizeof(Lf->iproto), "*%d?", p % (m/10));
+ }
+}
+#endif /* !defined(HASPRIVPRIPP) */
+
+
+/*
+ * printname() - print output name field
+ */
+
+void
+printname(nl)
+ int nl; /* NL status */
+{
+
+#if defined(HASNCACHE)
+ char buf[MAXPATHLEN];
+ char *cp;
+ int fp;
+#endif /* defined(HASNCACHE) */
+
+ int ps = 0;
+
+ if (Lf->nm && Lf->nm[0]) {
+
+ /*
+ * Print the name characters, if there are some.
+ */
+ safestrprt(Lf->nm, stdout, 0);
+ ps++;
+ if (!Lf->li[0].af && !Lf->li[1].af)
+ goto print_nma;
+ }
+ if (Lf->li[0].af || Lf->li[1].af) {
+ if (ps)
+ putchar(' ');
+ /*
+ * If the file has Internet addresses, print them.
+ */
+ if (printinaddr())
+ ps++;
+ goto print_nma;
+ }
+ if (((Lf->ntype == N_BLK) || (Lf->ntype == N_CHR))
+ && Lf->dev_def && Lf->rdev_def
+ && printdevname(&Lf->dev, &Lf->rdev, 0, Lf->ntype))
+ {
+
+ /*
+ * If this is a block or character device and it has a name, print it.
+ */
+ ps++;
+ goto print_nma;
+ }
+ if (Lf->is_com) {
+
+ /*
+ * If this is a common node, print that fact.
+ */
+ (void) fputs("COMMON: ", stdout);
+ ps++;
+ goto print_nma;
+ }
+
+#if defined(HASPRIVNMCACHE)
+ if (HASPRIVNMCACHE(Lf)) {
+ ps++;
+ goto print_nma;
+ }
+#endif /* defined(HASPRIVNMCACHE) */
+
+ if (Lf->lmi_srch) {
+ struct mounts *mp;
+ /*
+ * Do a deferred local mount info table search for the file system
+ * (mounted) directory name and inode number, and mounted device name.
+ */
+ for (mp = readmnt(); mp; mp = mp->next) {
+ if (Lf->dev == mp->dev) {
+ Lf->fsdir = mp->dir;
+ Lf->fsdev = mp->fsname;
+
+#if defined(HASFSINO)
+ Lf->fs_ino = mp->inode;
+#endif /* defined(HASFSINO) */
+
+ break;
+ }
+ }
+ Lf->lmi_srch = 0;
+ }
+ if (Lf->fsdir || Lf->fsdev) {
+
+ /*
+ * Print the file system directory name, device name, and
+ * possible path name components.
+ */
+
+#if !defined(HASNCACHE) || HASNCACHE<2
+ if (Lf->fsdir) {
+ safestrprt(Lf->fsdir, stdout, 0);
+ ps++;
+ }
+#endif /* !defined(HASNCACHE) || HASNCACHE<2 */
+
+#if defined(HASNCACHE)
+
+# if HASNCACHE<2
+ if (Lf->na) {
+ if (NcacheReload) {
+
+# if defined(NCACHELDPFX)
+ NCACHELDPFX
+# endif /* defined(NCACHELDPFX) */
+
+ (void) ncache_load();
+
+# if defined(NCACHELDSFX)
+ NCACHELDSFX
+# endif /* defined(NCACHELDSFX) */
+
+ NcacheReload = 0;
+ }
+ if ((cp = ncache_lookup(buf, sizeof(buf), &fp))) {
+ char *cp1;
+
+ if (*cp == '\0')
+ goto print_nma;
+ if (fp && Lf->fsdir) {
+ if (*cp != '/') {
+ cp1 = strrchr(Lf->fsdir, '/');
+ if (cp1 == (char *)NULL || *(cp1 + 1) != '\0')
+ putchar('/');
+ }
+ } else
+ (void) fputs(" -- ", stdout);
+ safestrprt(cp, stdout, 0);
+ ps++;
+ goto print_nma;
+ }
+ }
+# else /* HASNCACHE>1 */
+ if (NcacheReload) {
+
+# if defined(NCACHELDPFX)
+ NCACHELDPFX
+# endif /* defined(NCACHELDPFX) */
+
+ (void) ncache_load();
+
+# if defined(NCACHELDSFX)
+ NCACHELDSFX
+# endif /* defined(NCACHELDSFX) */
+
+ NcacheReload = 0;
+ }
+ if ((cp = ncache_lookup(buf, sizeof(buf), &fp))) {
+ if (fp) {
+ safestrprt(cp, stdout, 0);
+ ps++;
+ } else {
+ if (Lf->fsdir) {
+ safestrprt(Lf->fsdir, stdout, 0);
+ ps++;
+ }
+ if (*cp) {
+ (void) fputs(" -- ", stdout);
+ safestrprt(cp, stdout, 0);
+ ps++;
+ }
+ }
+ goto print_nma;
+ }
+ if (Lf->fsdir) {
+ safestrprt(Lf->fsdir, stdout, 0);
+ ps++;
+ }
+# endif /* HASNCACHE<2 */
+#endif /* defined(HASNCACHE) */
+
+ if (Lf->fsdev) {
+ if (Lf->fsdir)
+ (void) fputs(" (", stdout);
+ else
+ (void) putchar('(');
+ safestrprt(Lf->fsdev, stdout, 0);
+ (void) putchar(')');
+ ps++;
+ }
+ }
+/*
+ * Print the NAME column addition, if there is one. If there isn't
+ * make sure a NL is printed, as requested.
+ */
+
+print_nma:
+
+ if (Lf->nma) {
+ if (ps)
+ putchar(' ');
+ safestrprt(Lf->nma, stdout, 0);
+ ps++;
+ }
+/*
+ * If this file has TCP/IP state information, print it.
+ */
+ if (!Ffield && Ftcptpi
+ && (Lf->lts.type >= 0
+
+#if defined(HASTCPTPIQ)
+ || ((Ftcptpi & TCPTPI_QUEUES) && (Lf->lts.rqs || Lf->lts.sqs))
+#endif /* defined(HASTCPTPIQ) */
+
+#if defined(HASTCPTPIW)
+ || ((Ftcptpi & TCPTPI_WINDOWS) && (Lf->lts.rws || Lf->lts.wws))
+#endif /* defined(HASTCPTPIW) */
+
+ )) {
+ if (ps)
+ putchar(' ');
+ (void) print_tcptpi(1);
+ return;
+ }
+ if (nl)
+ putchar('\n');
+}
+
+
+/*
+ * printrawaddr() - print raw socket address
+ */
+
+void
+printrawaddr(sa)
+ struct sockaddr *sa; /* socket address */
+{
+ char *ep;
+ size_t sz;
+
+ ep = endnm(&sz);
+ (void) snpf(ep, sz, "%u/%u,%u,%u,%u,%u,%u,%u,%u,%u,%u,%u,%u,%u,%u",
+ sa->sa_family,
+ (unsigned char)sa->sa_data[0],
+ (unsigned char)sa->sa_data[1],
+ (unsigned char)sa->sa_data[2],
+ (unsigned char)sa->sa_data[3],
+ (unsigned char)sa->sa_data[4],
+ (unsigned char)sa->sa_data[5],
+ (unsigned char)sa->sa_data[6],
+ (unsigned char)sa->sa_data[7],
+ (unsigned char)sa->sa_data[8],
+ (unsigned char)sa->sa_data[9],
+ (unsigned char)sa->sa_data[10],
+ (unsigned char)sa->sa_data[11],
+ (unsigned char)sa->sa_data[12],
+ (unsigned char)sa->sa_data[13]);
+}
+
+
+/*
+ * printsockty() - print socket type
+ */
+
+char *
+printsockty(ty)
+ int ty; /* socket type -- e.g., from so_type */
+{
+ static char buf[64];
+ char *cp;
+
+ switch (ty) {
+
+#if defined(SOCK_STREAM)
+ case SOCK_STREAM:
+ cp = "STREAM";
+ break;
+#endif /* defined(SOCK_STREAM) */
+
+#if defined(SOCK_STREAM)
+ case SOCK_DGRAM:
+ cp = "DGRAM";
+ break;
+#endif /* defined(SOCK_DGRAM) */
+
+#if defined(SOCK_RAW)
+ case SOCK_RAW:
+ cp = "RAW";
+ break;
+#endif /* defined(SOCK_RAW) */
+
+#if defined(SOCK_RDM)
+ case SOCK_RDM:
+ cp = "RDM";
+ break;
+#endif /* defined(SOCK_RDM) */
+
+#if defined(SOCK_SEQPACKET)
+ case SOCK_SEQPACKET:
+ cp = "SEQPACKET";
+ break;
+#endif /* defined(SOCK_SEQPACKET) */
+
+ default:
+ (void) snpf(buf, sizeof(buf), "SOCK_%#x", ty);
+ return(buf);
+ }
+ (void) snpf(buf, sizeof(buf), "SOCK_%s", cp);
+ return(buf);
+}
+
+
+/*
+ * printuid() - print User ID or login name
+ */
+
+char *
+printuid(uid, ty)
+ UID_ARG uid; /* User IDentification number */
+ int *ty; /* returned UID type pointer (NULL
+ * (if none wanted). If non-NULL
+ * then: *ty = 0 = login name
+ * = 1 = UID number */
+{
+ int i;
+ struct passwd *pw;
+ struct stat sb;
+ static struct stat sbs;
+ static struct uidcache {
+ uid_t uid;
+ char nm[LOGINML+1];
+ struct uidcache *next;
+ } **uc = (struct uidcache **)NULL;
+ struct uidcache *up, *upn;
+ static char user[USERPRTL+1];
+
+ if (Futol) {
+ if (CkPasswd) {
+
+ /*
+ * Get the mtime and ctime of /etc/passwd, as required.
+ */
+ if (stat("/etc/passwd", &sb) != 0) {
+ (void) fprintf(stderr, "%s: can't stat(/etc/passwd): %s\n",
+ Pn, strerror(errno));
+ Exit(1);
+ }
+ }
+ /*
+ * Define the UID cache, if necessary.
+ */
+ if (!uc) {
+ if (!(uc = (struct uidcache **)calloc(UIDCACHEL,
+ sizeof(struct uidcache *))))
+ {
+ (void) fprintf(stderr,
+ "%s: no space for %d byte UID cache hash buckets\n",
+ Pn, (int)(UIDCACHEL * (sizeof(struct uidcache *))));
+ Exit(1);
+ }
+ if (CkPasswd) {
+ sbs = sb;
+ CkPasswd = 0;
+ }
+ }
+ /*
+ * If it's time to check /etc/passwd and if its the mtime/ctime has
+ * changed, destroy the existing UID cache.
+ */
+ if (CkPasswd) {
+ if (sbs.st_mtime != sb.st_mtime || sbs.st_ctime != sb.st_ctime)
+ {
+ for (i = 0; i < UIDCACHEL; i++) {
+ if ((up = uc[i])) {
+ do {
+ upn = up->next;
+ (void) free((FREE_P *)up);
+ } while ((up = upn) != (struct uidcache *)NULL);
+ uc[i] = (struct uidcache *)NULL;
+ }
+ }
+ sbs = sb;
+ }
+ CkPasswd = 0;
+ }
+ /*
+ * Search the UID cache.
+ */
+ i = (int)((((unsigned long)uid * 31415L) >> 7) & (UIDCACHEL - 1));
+ for (up = uc[i]; up; up = up->next) {
+ if (up->uid == (uid_t)uid) {
+ if (ty)
+ *ty = 0;
+ return(up->nm);
+ }
+ }
+ /*
+ * The UID is not in the cache.
+ *
+ * Look up the login name from the UID for a new cache entry.
+ */
+ if (!(pw = getpwuid((uid_t)uid))) {
+ if (!Fwarn) {
+ (void) fprintf(stderr, "%s: no pwd entry for UID %lu\n",
+ Pn, (unsigned long)uid);
+ }
+ } else {
+
+ /*
+ * Allocate and fill a new cache entry. Link it to its hash bucket.
+ */
+ if (!(upn = (struct uidcache *)malloc(sizeof(struct uidcache))))
+ {
+ (void) fprintf(stderr,
+ "%s: no space for UID cache entry for: %lu, %s)\n",
+ Pn, (unsigned long)uid, pw->pw_name);
+ Exit(1);
+ }
+ (void) strncpy(upn->nm, pw->pw_name, LOGINML);
+ upn->nm[LOGINML] = '\0';
+ upn->uid = (uid_t)uid;
+ upn->next = uc[i];
+ uc[i] = upn;
+ if (ty)
+ *ty = 0;
+ return(upn->nm);
+ }
+ }
+/*
+ * Produce a numeric conversion of the UID.
+ */
+ (void) snpf(user, sizeof(user), "%*lu", USERPRTL, (unsigned long)uid);
+ if (ty)
+ *ty = 1;
+ return(user);
+}
+
+
+/*
+ * printunkaf() - print unknown address family
+ */
+
+void
+printunkaf(fam, ty)
+ int fam; /* unknown address family */
+ int ty; /* output type: 0 = terse; 1 = full */
+{
+ char *p, *s;
+
+ p = "";
+ switch (fam) {
+
+#if defined(AF_UNSPEC)
+ case AF_UNSPEC:
+ s = "UNSPEC";
+ break;
+#endif /* defined(AF_UNSPEC) */
+
+#if defined(AF_UNIX)
+ case AF_UNIX:
+ s = "UNIX";
+ break;
+#endif /* defined(AF_UNIX) */
+
+#if defined(AF_INET)
+ case AF_INET:
+ s = "INET";
+ break;
+#endif /* defined(AF_INET) */
+
+#if defined(AF_INET6)
+ case AF_INET6:
+ s = "INET6";
+ break;
+#endif /* defined(AF_INET6) */
+
+#if defined(AF_IMPLINK)
+ case AF_IMPLINK:
+ s = "IMPLINK";
+ break;
+#endif /* defined(AF_IMPLINK) */
+
+#if defined(AF_PUP)
+ case AF_PUP:
+ s = "PUP";
+ break;
+#endif /* defined(AF_PUP) */
+
+#if defined(AF_CHAOS)
+ case AF_CHAOS:
+ s = "CHAOS";
+ break;
+#endif /* defined(AF_CHAOS) */
+
+#if defined(AF_NS)
+ case AF_NS:
+ s = "NS";
+ break;
+#endif /* defined(AF_NS) */
+
+#if defined(AF_ISO)
+ case AF_ISO:
+ s = "ISO";
+ break;
+#endif /* defined(AF_ISO) */
+
+#if defined(AF_NBS)
+# if !defined(AF_ISO) || AF_NBS!=AF_ISO
+ case AF_NBS:
+ s = "NBS";
+ break;
+# endif /* !defined(AF_ISO) || AF_NBS!=AF_ISO */
+#endif /* defined(AF_NBS) */
+
+#if defined(AF_ECMA)
+ case AF_ECMA:
+ s = "ECMA";
+ break;
+#endif /* defined(AF_ECMA) */
+
+#if defined(AF_DATAKIT)
+ case AF_DATAKIT:
+ s = "DATAKIT";
+ break;
+#endif /* defined(AF_DATAKIT) */
+
+#if defined(AF_CCITT)
+ case AF_CCITT:
+ s = "CCITT";
+ break;
+#endif /* defined(AF_CCITT) */
+
+#if defined(AF_SNA)
+ case AF_SNA:
+ s = "SNA";
+ break;
+#endif /* defined(AF_SNA) */
+
+#if defined(AF_DECnet)
+ case AF_DECnet:
+ s = "DECnet";
+ break;
+#endif /* defined(AF_DECnet) */
+
+#if defined(AF_DLI)
+ case AF_DLI:
+ s = "DLI";
+ break;
+#endif /* defined(AF_DLI) */
+
+#if defined(AF_LAT)
+ case AF_LAT:
+ s = "LAT";
+ break;
+#endif /* defined(AF_LAT) */
+
+#if defined(AF_HYLINK)
+ case AF_HYLINK:
+ s = "HYLINK";
+ break;
+#endif /* defined(AF_HYLINK) */
+
+#if defined(AF_APPLETALK)
+ case AF_APPLETALK:
+ s = "APPLETALK";
+ break;
+#endif /* defined(AF_APPLETALK) */
+
+#if defined(AF_BSC)
+ case AF_BSC:
+ s = "BSC";
+ break;
+#endif /* defined(AF_BSC) */
+
+#if defined(AF_DSS)
+ case AF_DSS:
+ s = "DSS";
+ break;
+#endif /* defined(AF_DSS) */
+
+#if defined(AF_ROUTE)
+ case AF_ROUTE:
+ s = "ROUTE";
+ break;
+#endif /* defined(AF_ROUTE) */
+
+#if defined(AF_RAW)
+ case AF_RAW:
+ s = "RAW";
+ break;
+#endif /* defined(AF_RAW) */
+
+#if defined(AF_LINK)
+ case AF_LINK:
+ s = "LINK";
+ break;
+#endif /* defined(AF_LINK) */
+
+#if defined(pseudo_AF_XTP)
+ case pseudo_AF_XTP:
+ p = "pseudo_";
+ s = "XTP";
+ break;
+#endif /* defined(pseudo_AF_XTP) */
+
+#if defined(AF_RMP)
+ case AF_RMP:
+ s = "RMP";
+ break;
+#endif /* defined(AF_RMP) */
+
+#if defined(AF_COIP)
+ case AF_COIP:
+ s = "COIP";
+ break;
+#endif /* defined(AF_COIP) */
+
+#if defined(AF_CNT)
+ case AF_CNT:
+ s = "CNT";
+ break;
+#endif /* defined(AF_CNT) */
+
+#if defined(pseudo_AF_RTIP)
+ case pseudo_AF_RTIP:
+ p = "pseudo_";
+ s = "RTIP";
+ break;
+#endif /* defined(pseudo_AF_RTIP) */
+
+#if defined(AF_NETMAN)
+ case AF_NETMAN:
+ s = "NETMAN";
+ break;
+#endif /* defined(AF_NETMAN) */
+
+#if defined(AF_INTF)
+ case AF_INTF:
+ s = "INTF";
+ break;
+#endif /* defined(AF_INTF) */
+
+#if defined(AF_NETWARE)
+ case AF_NETWARE:
+ s = "NETWARE";
+ break;
+#endif /* defined(AF_NETWARE) */
+
+#if defined(AF_NDD)
+ case AF_NDD:
+ s = "NDD";
+ break;
+#endif /* defined(AF_NDD) */
+
+#if defined(AF_NIT)
+# if !defined(AF_ROUTE) || AF_ROUTE!=AF_NIT
+ case AF_NIT:
+ s = "NIT";
+ break;
+# endif /* !defined(AF_ROUTE) || AF_ROUTE!=AF_NIT */
+#endif /* defined(AF_NIT) */
+
+#if defined(AF_802)
+# if !defined(AF_RAW) || AF_RAW!=AF_802
+ case AF_802:
+ s = "802";
+ break;
+# endif /* !defined(AF_RAW) || AF_RAW!=AF_802 */
+#endif /* defined(AF_802) */
+
+#if defined(AF_X25)
+ case AF_X25:
+ s = "X25";
+ break;
+#endif /* defined(AF_X25) */
+
+#if defined(AF_CTF)
+ case AF_CTF:
+ s = "CTF";
+ break;
+#endif /* defined(AF_CTF) */
+
+#if defined(AF_WAN)
+ case AF_WAN:
+ s = "WAN";
+ break;
+#endif /* defined(AF_WAN) */
+
+#if defined(AF_OSINET)
+# if defined(AF_INET) && AF_INET!=AF_OSINET
+ case AF_OSINET:
+ s = "OSINET";
+ break;
+# endif /* defined(AF_INET) && AF_INET!=AF_OSINET */
+#endif /* defined(AF_OSINET) */
+
+#if defined(AF_GOSIP)
+ case AF_GOSIP:
+ s = "GOSIP";
+ break;
+#endif /* defined(AF_GOSIP) */
+
+#if defined(AF_SDL)
+ case AF_SDL:
+ s = "SDL";
+ break;
+#endif /* defined(AF_SDL) */
+
+#if defined(AF_IPX)
+ case AF_IPX:
+ s = "IPX";
+ break;
+#endif /* defined(AF_IPX) */
+
+#if defined(AF_SIP)
+ case AF_SIP:
+ s = "SIP";
+ break;
+#endif /* defined(AF_SIP) */
+
+#if defined(psuedo_AF_PIP)
+ case psuedo_AF_PIP:
+ p = "pseudo_";
+ s = "PIP";
+ break;
+#endif /* defined(psuedo_AF_PIP) */
+
+#if defined(AF_OTS)
+ case AF_OTS:
+ s = "OTS";
+ break;
+#endif /* defined(AF_OTS) */
+
+#if defined(pseudo_AF_BLUE)
+ case pseudo_AF_BLUE: /* packets for Blue box */
+ p = "pseudo_";
+ s = "BLUE";
+ break;
+#endif /* defined(pseudo_AF_BLUE) */
+
+#if defined(AF_NDRV) /* network driver raw access */
+ case AF_NDRV:
+ s = "NDRV";
+ break;
+#endif /* defined(AF_NDRV) */
+
+#if defined(AF_SYSTEM) /* kernel event messages */
+ case AF_SYSTEM:
+ s = "SYSTEM";
+ break;
+#endif /* defined(AF_SYSTEM) */
+
+#if defined(AF_USER)
+ case AF_USER:
+ s = "USER";
+ break;
+#endif /* defined(AF_USER) */
+
+#if defined(pseudo_AF_KEY)
+ case pseudo_AF_KEY:
+ p = "pseudo_";
+ s = "KEY";
+ break;
+#endif /* defined(pseudo_AF_KEY) */
+
+#if defined(AF_KEY) /* Security Association DB socket */
+ case AF_KEY:
+ s = "KEY";
+ break;
+#endif /* defined(AF_KEY) */
+
+#if defined(AF_NCA) /* NCA socket */
+ case AF_NCA:
+ s = "NCA";
+ break;
+#endif /* defined(AF_NCA) */
+
+#if defined(AF_POLICY) /* Security Policy DB socket */
+ case AF_POLICY:
+ s = "POLICY";
+ break;
+#endif /* defined(AF_POLICY) */
+
+#if defined(AF_PPP) /* PPP socket */
+ case AF_PPP:
+ s = "PPP";
+ break;
+#endif /* defined(AF_PPP) */
+
+ default:
+ if (!ty)
+ (void) snpf(Namech, Namechl, "%#x", fam);
+ else
+ (void) snpf(Namech, Namechl,
+ "no further information on family %#x", fam);
+ return;
+ }
+ if (!ty)
+ (void) snpf(Namech, Namechl, "%sAF_%s", p, s);
+ else
+ (void) snpf(Namech, Namechl, "no further information on %sAF_%s",
+ p, s);
+ return;
+}
+
+
+/*
+ * update_portmap() - update a portmap entry with its port number or service
+ * name
+ */
+
+static void
+update_portmap(pt, pn)
+ struct porttab *pt; /* porttab entry */
+ char *pn; /* port name */
+{
+ MALLOC_S al, nl;
+ char *cp;
+
+ if (pt->ss)
+ return;
+ if (!(al = strlen(pn))) {
+ pt->ss = 1;
+ return;
+ }
+ nl = al + pt->nl + 2;
+ if (!(cp = (char *)malloc(nl + 1))) {
+ (void) fprintf(stderr,
+ "%s: can't allocate %d bytes for portmap name: %s[%s]\n",
+ Pn, (int)(nl + 1), pn, pt->name);
+ Exit(1);
+ }
+ (void) snpf(cp, nl + 1, "%s[%s]", pn, pt->name);
+ (void) free((FREE_P *)pt->name);
+ pt->name = cp;
+ pt->nl = nl;
+ pt->ss = 1;
+}
--- /dev/null
+/*
+ * proc.c - common process and file structure functions for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1994 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: proc.c,v 1.45 2009/03/25 19:20:44 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static int is_file_sel,(struct lproc *lp, struct lfile *lf));
+
+
+/*
+ * add_nma() - add to NAME column addition
+ */
+
+void
+add_nma(cp, len)
+ char *cp; /* string to add */
+ int len; /* string length */
+{
+ int nl;
+
+ if (!cp || !len)
+ return;
+ if (Lf->nma) {
+ nl = (int)strlen(Lf->nma);
+ Lf->nma = (char *)realloc((MALLOC_P *)Lf->nma,
+ (MALLOC_S)(len + nl + 2));
+ } else {
+ nl = 0;
+ Lf->nma = (char *)malloc((MALLOC_S)(len + 1));
+ }
+ if (!Lf->nma) {
+ (void) fprintf(stderr, "%s: no name addition space: PID %ld, FD %s",
+ Pn, (long)Lp->pid, Lf->fd);
+ Exit(1);
+ }
+ if (nl) {
+ Lf->nma[nl] = ' ';
+ (void) strncpy(&Lf->nma[nl + 1], cp, len);
+ Lf->nma[nl + 1 + len] = '\0';
+ } else {
+ (void) strncpy(Lf->nma, cp, len);
+ Lf->nma[len] = '\0';
+ }
+}
+
+
+#if defined(HASFSTRUCT)
+_PROTOTYPE(static char *alloc_fflbuf,(char **bp, int *al, int lr));
+
+
+/*
+ * alloc_fflbuf() - allocate file flags print buffer
+ */
+
+static char *
+alloc_fflbuf(bp, al, lr)
+ char **bp; /* current buffer pointer */
+ int *al; /* current allocated length */
+ int lr; /* length required */
+{
+ int sz;
+
+ sz = (int)(lr + 1); /* allocate '\0' space */
+ if (*bp && (sz <= *al))
+ return(*bp);
+ if (*bp)
+ *bp = (char *)realloc((MALLOC_P *)*bp, (MALLOC_S)sz);
+ else
+ *bp = (char *)malloc((MALLOC_S)sz);
+ if (!*bp) {
+ (void) fprintf(stderr, "%s: no space (%d) for print flags\n",
+ Pn, sz);
+ Exit(1);
+ }
+ *al = sz;
+ return(*bp);
+}
+#endif /* defined(HASFSTRUCT) */
+
+
+/*
+ * alloc_lfile() - allocate local file structure space
+ */
+
+void
+alloc_lfile(nm, num)
+ char *nm; /* file descriptor name (may be NULL) */
+ int num; /* file descriptor number -- -1 if
+ * none */
+{
+ int fds;
+
+ if (Lf) {
+/*
+ * If reusing a previously allocated structure, release any allocated
+ * space it was using.
+ */
+ if (Lf->dev_ch)
+ (void) free((FREE_P *)Lf->dev_ch);
+ if (Lf->nm)
+ (void) free((FREE_P *)Lf->nm);
+ if (Lf->nma)
+ (void) free((FREE_P *)Lf->nma);
+
+#if defined(HASLFILEADD) && defined(CLRLFILEADD)
+ CLRLFILEADD(Lf)
+#endif /* defined(HASLFILEADD) && defined(CLRLFILEADD) */
+
+/*
+ * Othwerise, allocate a new structure.
+ */
+ } else if (!(Lf = (struct lfile *)malloc(sizeof(struct lfile)))) {
+ (void) fprintf(stderr, "%s: no local file space at PID %d\n",
+ Pn, Lp->pid);
+ Exit(1);
+ }
+/*
+ * Initialize the structure.
+ */
+ Lf->access = Lf->lock = ' ';
+ Lf->dev_def = Lf->inp_ty = Lf->is_com = Lf->is_nfs = Lf->is_stream
+ = Lf->lmi_srch = Lf->nlink_def = Lf->off_def = Lf->sz_def
+ = Lf->rdev_def
+ = (unsigned char)0;
+ Lf->li[0].af = Lf->li[1].af = 0;
+ Lf->lts.type = -1;
+ Lf->nlink = 0l;
+
+#if defined(HASMNTSTAT)
+ Lf->mnt_stat = (unsigned char)0;
+#endif /* defined(HASMNTSTAT) */
+
+#if defined(HASSOOPT)
+ Lf->lts.kai = Lf->lts.ltm = 0;
+ Lf->lts.opt = Lf->lts.qlen = Lf->lts.qlim = Lf->lts.pqlen
+ = (unsigned int)0;
+ Lf->lts.rbsz = Lf->lts.sbsz = (unsigned long)0;
+ Lf->lts.qlens = Lf->lts.qlims = Lf->lts.pqlens = Lf->lts.rbszs
+ = Lf->lts.sbszs = (unsigned char)0;
+#endif /* defined(HASSOOPT) */
+
+#if defined(HASSOSTATE)
+ Lf->lts.ss = 0;
+#endif /* defined(HASSOSTATE) */
+
+#if defined(HASTCPOPT)
+ Lf->lts.mss = (unsigned long)0;
+ Lf->lts.msss = (unsigned char)0;
+ Lf->lts.topt = (unsigned int)0;
+#endif /* defined(HASTCPOPT) */
+
+#if defined(HASTCPTPIQ)
+ Lf->lts.rqs = Lf->lts.sqs = (unsigned char)0;
+#endif /* defined(HASTCPTPIQ) */
+
+#if defined(HASTCPTPIW)
+ Lf->lts.rws = Lf->lts.wws = (unsigned char)0;
+#endif /* defined(HASTCPTPIW) */
+
+#if defined(HASFSINO)
+ Lf->fs_ino = 0;
+#endif /* defined(HASFSINO) */
+
+#if defined(HASVXFS) && defined(HASVXFSDNLC)
+ Lf->is_vxfs = 0;
+#endif /* defined(HASVXFS) && defined(HASVXFSDNLC) */
+
+ Lf->inode = (INODETYPE)0;
+ Lf->off = (SZOFFTYPE)0;
+ if (Lp->pss & PS_PRI)
+ Lf->sf = Lp->sf;
+ else
+ Lf->sf = 0;
+ Lf->iproto[0] = Lf->type[0] = '\0';
+ if (nm) {
+ (void) strncpy(Lf->fd, nm, FDLEN - 1);
+ Lf->fd[FDLEN - 1] = '\0';
+ } else if (num >= 0) {
+ if (num < 10000)
+ (void) snpf(Lf->fd, sizeof(Lf->fd), "%4d", num);
+ else
+ (void) snpf(Lf->fd, sizeof(Lf->fd), "*%03d", num % 1000);
+ } else
+ Lf->fd[0] = '\0';
+ Lf->dev_ch = Lf->fsdir = Lf->fsdev = Lf->nm = Lf->nma = (char *)NULL;
+ Lf->ch = -1;
+
+#if defined(HASNCACHE) && HASNCACHE<2
+ Lf->na = (KA_T)NULL;
+#endif /* defined(HASNCACHE) && HASNCACHE<2 */
+
+ Lf->next = (struct lfile *)NULL;
+ Lf->ntype = Ntype = N_REGLR;
+ Namech[0] = '\0';
+
+#if defined(HASFSTRUCT)
+ Lf->fct = Lf->ffg = Lf->pof = (long)0;
+ Lf->fna = (KA_T)NULL;
+ Lf->fsv = (unsigned char)0;
+#endif /* defined(HASFSTRUCT) */
+
+#if defined(HASLFILEADD) && defined(SETLFILEADD)
+/*
+ * Do local initializations.
+ */
+ SETLFILEADD
+#endif /* defined(HASLFILEADD) && defined(SETLFILEADD) */
+
+/*
+ * See if the file descriptor has been selected.
+ */
+ if (!Fdl || (!nm && num < 0))
+ return;
+ fds = ck_fd_status(nm, num);
+ switch (FdlTy) {
+ case 0: /* inclusion list */
+ if (fds == 2)
+ Lf->sf |= SELFD;
+ break;
+ case 1: /* exclusion list */
+ if (fds != 1)
+ Lf->sf |= SELFD;
+ }
+}
+
+
+/*
+ * alloc_lproc() - allocate local proc structure space
+ */
+
+void
+alloc_lproc(pid, pgid, ppid, uid, cmd, pss, sf)
+ int pid; /* Process ID */
+ int pgid; /* process group ID */
+ int ppid; /* parent process ID */
+ UID_ARG uid; /* User ID */
+ char *cmd; /* command */
+ int pss; /* process select state */
+ int sf; /* process select flags */
+{
+ static int sz = 0;
+
+ if (!Lproc) {
+ if (!(Lproc = (struct lproc *)malloc(
+ (MALLOC_S)(LPROCINCR * sizeof(struct lproc)))))
+ {
+ (void) fprintf(stderr,
+ "%s: no malloc space for %d local proc structures\n",
+ Pn, LPROCINCR);
+ Exit(1);
+ }
+ sz = LPROCINCR;
+ } else if ((Nlproc + 1) > sz) {
+ sz += LPROCINCR;
+ if (!(Lproc = (struct lproc *)realloc((MALLOC_P *)Lproc,
+ (MALLOC_S)(sz * sizeof(struct lproc)))))
+ {
+ (void) fprintf(stderr,
+ "%s: no realloc space for %d local proc structures\n",
+ Pn, sz);
+ Exit(1);
+ }
+ }
+ Lp = &Lproc[Nlproc++];
+ Lp->pid = pid;
+ Lp->pgid = pgid;
+ Lp->ppid = ppid;
+ Lp->file = (struct lfile *)NULL;
+ Lp->sf = (short)sf;
+ Lp->pss = (short)pss;
+ Lp->uid = (uid_t)uid;
+/*
+ * Allocate space for the full command name and copy it there.
+ */
+ if (!(Lp->cmd = mkstrcpy(cmd, (MALLOC_S *)NULL))) {
+ (void) fprintf(stderr, "%s: PID %d, no space for command name: ",
+ Pn, pid);
+ safestrprt(cmd, stderr, 1);
+ Exit(1);
+ }
+
+#if defined(HASZONES)
+/*
+ * Clear the zone name pointer. The dialect's own code will set it.
+ */
+ Lp->zn = (char *)NULL;
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+/*
+ * Clear the security context pointer. The dialect's own code will
+ * set it.
+ */
+ Lp->cntx = (char *)NULL;
+#endif /* defined(HASSELINUX) */
+
+}
+
+
+/*
+ * ck_fd_status() - check FD status
+ *
+ * return: 0 == FD is neither included nor excluded
+ * 1 == FD is excluded
+ * 2 == FD is included
+ */
+
+extern int
+ck_fd_status(nm, num)
+ char *nm; /* file descriptor name (may be NULL) */
+ int num; /* file descriptor number -- -1 if
+ * none */
+{
+ char *cp;
+ struct fd_lst *fp;
+
+ if (!(fp = Fdl) || (!nm && num < 0))
+ return(0);
+ if ((cp = nm)) {
+ while (*cp && *cp == ' ')
+ cp++;
+ }
+/*
+ * Check for an exclusion match.
+ */
+ if (FdlTy == 1) {
+ for (; fp; fp = fp->next) {
+ if (cp) {
+ if (fp->nm && strcmp(fp->nm, cp) == 0)
+ return(1);
+ continue;
+ }
+ if (num >= fp->lo && num <= fp->hi)
+ return(1);
+ }
+ return(0);
+ }
+/*
+ * If Fdl isn't an exclusion list, check for an inclusion match.
+ */
+ for (; fp; fp = fp->next) {
+ if (cp) {
+ if (fp->nm && strcmp(fp->nm, cp) == 0)
+ return(2);
+ continue;
+ }
+ if (num >= fp->lo && num <= fp->hi)
+ return(2);
+ }
+ return(0);
+}
+
+
+/*
+ * comppid() - compare PIDs
+ */
+
+int
+comppid(a1, a2)
+ COMP_P *a1, *a2;
+{
+ struct lproc **p1 = (struct lproc **)a1;
+ struct lproc **p2 = (struct lproc **)a2;
+
+ if ((*p1)->pid < (*p2)->pid)
+ return(-1);
+ if ((*p1)->pid > (*p2)->pid)
+ return(1);
+ return(0);
+}
+
+
+/*
+ * ent_inaddr() - enter Internet addresses
+ */
+
+void
+ent_inaddr(la, lp, fa, fp, af)
+ unsigned char *la; /* local Internet address */
+ int lp; /* local port */
+ unsigned char *fa; /* foreign Internet address -- may
+ * be NULL to indicate no foreign
+ * address is known */
+ int fp; /* foreign port */
+ int af; /* address family -- e.g, AF_INET,
+ * AF_INET */
+{
+ int m;
+
+ if (la) {
+ Lf->li[0].af = af;
+
+#if defined(HASIPv6)
+ if (af == AF_INET6)
+ Lf->li[0].ia.a6 = *(struct in6_addr *)la;
+ else
+#endif /* defined(HASIPv6) */
+
+ Lf->li[0].ia.a4 = *(struct in_addr *)la;
+ Lf->li[0].p = lp;
+ } else
+ Lf->li[0].af = 0;
+ if (fa) {
+ Lf->li[1].af = af;
+
+#if defined(HASIPv6)
+ if (af == AF_INET6)
+ Lf->li[1].ia.a6 = *(struct in6_addr *)fa;
+ else
+#endif /* defined(HASIPv6) */
+
+ Lf->li[1].ia.a4 = *(struct in_addr *)fa;
+ Lf->li[1].p = fp;
+ } else
+ Lf->li[1].af = 0;
+/*
+ * If network address matching has been selected, check both addresses.
+ */
+ if ((Selflags & SELNA) && Nwad) {
+ m = (fa && is_nw_addr(fa, fp, af)) ? 1 : 0;
+ m |= (la && is_nw_addr(la, lp, af)) ? 1 : 0;
+ if (m)
+ Lf->sf |= SELNA;
+ }
+}
+
+
+/*
+ * examine_lproc() - examine local process
+ *
+ * return: 1 = last process
+ */
+
+int
+examine_lproc()
+{
+ int sbp = 0;
+
+ if (RptTm)
+ return(0);
+/*
+ * List the process if the process is selected and:
+ *
+ * o listing is limited to a single PID selection -- this one;
+ *
+ * o listing is selected by an ANDed option set (not all options)
+ * that includes a single PID selection -- this one.
+ */
+ if ((Lp->sf & SELPID) && !Selall) {
+ if ((Selflags == SELPID)
+ || (Fand && (Selflags & SELPID))) {
+ sbp = 1;
+ Npuns--;
+ }
+ }
+ if (Lp->pss && Npid == 1 && sbp) {
+ print_init();
+ (void) print_proc();
+ PrPass++;
+ if (PrPass < 2)
+ (void) print_proc();
+ Lp->pss = 0;
+ }
+/*
+ * Deprecate an unselected (or listed) process.
+ */
+ if ( ! Lp->pss) {
+ (void) free_lproc(Lp);
+ Nlproc--;
+ }
+/*
+ * Indicate last-process if listing is limited to PID selections,
+ * and all selected processes have been listed.
+ */
+ return((sbp && Npuns == 0) ? 1 : 0);
+}
+
+
+/*
+ * free_lproc() - free lproc entry and its associated malloc'd space
+ */
+
+void
+free_lproc(lp)
+ struct lproc *lp;
+{
+ struct lfile *lf, *nf;
+
+ for (lf = lp->file; lf; lf = nf) {
+ if (lf->dev_ch) {
+ (void) free((FREE_P *)lf->dev_ch);
+ lf->dev_ch = (char *)NULL;
+ }
+ if (lf->nm) {
+ (void) free((FREE_P *)lf->nm);
+ lf->nm = (char *)NULL;
+ }
+ if (lf->nma) {
+ (void) free((FREE_P *)lf->nma);
+ lf->nma = (char *)NULL;
+ }
+
+#if defined(HASLFILEADD) && defined(CLRLFILEADD)
+ CLRLFILEADD(lf)
+#endif /* defined(HASLFILEADD) && defined(CLRLFILEADD) */
+
+ nf = lf->next;
+ (void) free((FREE_P *)lf);
+ }
+ lp->file = (struct lfile *)NULL;
+ if (lp->cmd) {
+ (void) free((FREE_P *)lp->cmd);
+ lp->cmd = (char *)NULL;
+ }
+}
+
+
+/*
+ * is_cmd_excl() - is command excluded?
+ */
+
+int
+is_cmd_excl(cmd, pss, sf)
+ char *cmd; /* command name */
+ short *pss; /* process state */
+ short *sf; /* process select flags */
+{
+ int i;
+ struct str_lst *sp;
+/*
+ * See if the command is excluded by a "-c^<command>" option.
+ */
+ if (Cmdl && Cmdnx) {
+ for (sp = Cmdl; sp; sp = sp->next) {
+ if (sp->x && !strncmp(sp->str, cmd, sp->len))
+ return(1);
+ }
+ }
+/*
+ * The command is not excluded if no command selection was requested,
+ * or if its name matches any -c <command> specification.
+ *
+ */
+ if ((Selflags & SELCMD) == 0)
+ return(0);
+ for (sp = Cmdl; sp; sp = sp->next) {
+ if (!sp->x && !strncmp(sp->str, cmd, sp->len)) {
+ sp->f = 1;
+ *pss |= PS_PRI;
+ *sf |= SELCMD;
+ return(0);
+ }
+ }
+/*
+ * The command name doesn't match any -c <command> specification. See if it
+ * matches a -c /RE/[bix] specification.
+ */
+ for (i = 0; i < NCmdRxU; i++) {
+ if (!regexec(&CmdRx[i].cx, cmd, 0, NULL, 0)) {
+ CmdRx[i].mc = 1;
+ *pss |= PS_PRI;
+ *sf |= SELCMD;
+ return(0);
+ }
+ }
+/*
+ * The command name matches no -c specification.
+ *
+ * It's excluded if the only selection condition is command name,
+ * or if command name selection is part of an ANDed set.
+ */
+ if (Selflags == SELCMD)
+ return(1);
+ return (Fand ? 1 : 0);
+}
+
+
+/*
+ * is_file_sel() - is file selected?
+ */
+
+static int
+is_file_sel(lp, lf)
+ struct lproc *lp; /* lproc structure pointer */
+ struct lfile *lf; /* lfile structure pointer */
+{
+ if (!lf || !lf->sf)
+ return(0);
+ if (Lf->sf & SELEXCLF)
+ return(0);
+
+#if defined(HASSECURITY) && defined(HASNOSOCKSECURITY)
+ if (Myuid && (Myuid != lp->uid)) {
+ if (!(lf->sf & (SELNA | SELNET)))
+ return(0);
+ }
+#endif /* defined(HASSECURITY) && defined(HASNOSOCKSECURITY) */
+
+ if (Selall)
+ return(1);
+ if (Fand && ((lf->sf & Selflags) != Selflags))
+ return(0);
+ return(1);
+}
+
+
+/*
+ * is_proc_excl() - is process excluded?
+ */
+
+int
+is_proc_excl(pid, pgid, uid, pss, sf)
+ int pid; /* Process ID */
+ int pgid; /* process group ID */
+ UID_ARG uid; /* User ID */
+ short *pss; /* process select state for lproc */
+ short *sf; /* select flags for lproc */
+{
+ int i, j;
+
+ *pss = *sf = 0;
+
+#if defined(HASSECURITY)
+/*
+ * The process is excluded by virtue of the security option if it
+ * isn't owned by the owner of this lsof process, unless the
+ * HASNOSOCKSECURITY option is also specified. In that case the
+ * selected socket files of any process may be listed.
+ */
+# if !defined(HASNOSOCKSECURITY)
+ if (Myuid && Myuid != (uid_t)uid)
+ return(1);
+# endif /* !defined(HASNOSOCKSECURITY) */
+#endif /* defined(HASSECURITY) */
+
+/*
+ * If the excluding of process listing by UID has been specified, see if the
+ * owner of this process is excluded.
+ */
+ if (Nuidexcl) {
+ for (i = j = 0; (i < Nuid) && (j < Nuidexcl); i++) {
+ if (!Suid[i].excl)
+ continue;
+ if (Suid[i].uid == (uid_t)uid)
+ return(1);
+ j++;
+ }
+ }
+/*
+ * If the excluding of process listing by PGID has been specified, see if this
+ * PGID is excluded.
+ */
+ if (Npgidx) {
+ for (i = j = 0; (i < Npgid) && (j < Npgidx); i++) {
+ if (!Spgid[i].x)
+ continue;
+ if (Spgid[i].i == pgid)
+ return(1);
+ j++;
+ }
+ }
+/*
+ * If the excluding of process listing by PID has been specified, see if this
+ * PID is excluded.
+ */
+ if (Npidx) {
+ for (i = j = 0; (i < Npid) && (j < Npidx); i++) {
+ if (!Spid[i].x)
+ continue;
+ if (Spid[i].i == pid)
+ return(1);
+ j++;
+ }
+ }
+/*
+ * If the listing of all processes is selected, then this one is not excluded.
+ *
+ * However, if HASSECURITY and HASNOSOCKSECURITY are both specified, exclude
+ * network selections from the file flags, so that the tests in is_file_sel()
+ * work as expected.
+ */
+ if (Selall) {
+ *pss = PS_PRI;
+
+#if defined(HASSECURITY) && defined(HASNOSOCKSECURITY)
+ *sf = SELALL & ~(SELNA | SELNET);
+#else /* !defined(HASSECURITY) || !defined(HASNOSOCKSECURITY) */
+ *sf = SELALL;
+#endif /* defined(HASSECURITY) && defined(HASNOSOCKSECURITY) */
+
+ return(0);
+ }
+/*
+ * If the listing of processes has been specified by process group ID, see
+ * if this one is included or excluded.
+ */
+ if (Npgidi && (Selflags & SELPGID)) {
+ for (i = j = 0; (i < Npgid) && (j < Npgidi); i++) {
+ if (Spgid[i].x)
+ continue;
+ if (Spgid[i].i == pgid) {
+ Spgid[i].f = 1;
+ *pss = PS_PRI;
+ *sf = SELPGID;
+ if (Selflags == SELPGID)
+ return(0);
+ break;
+ }
+ j++;
+ }
+ if ((Selflags == SELPGID) && !*sf)
+ return(1);
+ }
+/*
+ * If the listing of processes has been specified by PID, see if this one is
+ * included or excluded.
+ */
+ if (Npidi && (Selflags & SELPID)) {
+ for (i = j = 0; (i < Npid) && (j < Npidi); i++) {
+ if (Spid[i].x)
+ continue;
+ if (Spid[i].i == pid) {
+ Spid[i].f = 1;
+ *pss = PS_PRI;
+ *sf |= SELPID;
+ if (Selflags == SELPID)
+ return(0);
+ break;
+ }
+ j++;
+ }
+ if ((Selflags == SELPID) && !*sf)
+ return(1);
+ }
+/*
+ * If the listing of processes has been specified by UID, see if the owner of
+* this process has been included.
+ */
+ if (Nuidincl && (Selflags & SELUID)) {
+ for (i = j = 0; (i < Nuid) && (j < Nuidincl); i++) {
+ if (Suid[i].excl)
+ continue;
+ if (Suid[i].uid == (uid_t)uid) {
+ Suid[i].f = 1;
+ *pss = PS_PRI;
+ *sf |= SELUID;
+ if (Selflags == SELUID)
+ return(0);
+ break;
+ }
+ j++;
+ }
+ if (Selflags == SELUID && (*sf & SELUID) == 0)
+ return(1);
+ }
+/*
+ * When neither the process group ID, nor the PID, nor the UID is selected:
+ *
+ * If list option ANDing of process group IDs, PIDs or UIDs is specified,
+ * the process is excluded;
+ *
+ * Otherwise, it's not excluded by the tests of this function.
+ */
+ if ( ! *sf)
+ return((Fand && (Selflags & (SELPGID|SELPID|SELUID))) ? 1 : 0);
+/*
+ * When the process group ID, PID, or UID is selected and the process group
+ * ID, PID, or UID list option has been specified:
+ *
+ * If list option ANDing has been specified, and the correct
+ * combination of process group ID, PID, and UID is selected, reply that
+ * the process is not excluded;
+ * or
+ * If list option ANDing has not been specified, reply that the
+ * process is not excluded by the tests of this function.
+ */
+ if (Selflags & (SELPGID|SELPID|SELUID)) {
+ if (Fand)
+ return(((Selflags & (SELPGID|SELPID|SELUID)) != *sf) ? 1 : 0);
+ return(0);
+ }
+/*
+ * Finally, when neither the process group ID, nor the PID, nor the UID is
+ * selected, and no process group ID, PID or UID list option has been
+ * specified:
+ *
+ * If list option ANDing has been specified, this process is
+ * excluded;
+ *
+ * Otherwise, it isn't excluded by the tests of this function.
+ */
+ return(Fand ? 1 : 0);
+}
+
+
+/*
+ * link_lfile() - link local file structures
+ */
+
+void
+link_lfile()
+{
+ if (Lf->sf & SELEXCLF)
+ return;
+ Lp->pss |= PS_SEC;
+ if (Plf)
+ Plf->next = Lf;
+ else
+ Lp->file = Lf;
+ Plf = Lf;
+ if (Fnet && (Lf->sf & SELNET))
+ Fnet = 2;
+ if (Fnfs && (Lf->sf & SELNFS))
+ Fnfs = 2;
+ Lf = (struct lfile *)NULL;
+}
+
+
+#if defined(HASFSTRUCT)
+/*
+ * print_fflags() - print interpreted f_flag[s]
+ */
+
+char *
+print_fflags(ffg, pof)
+ long ffg; /* file structure's flags value */
+ long pof; /* process open files flags value */
+{
+ int al, ct, fx;
+ static int bl = 0;
+ static char *bp = (char *)NULL;
+ char *sep;
+ int sepl;
+ struct pff_tab *tp;
+ long wf;
+ char xbuf[64];
+/*
+ * Reduce the supplied flags according to the definitions in Pff_tab[] and
+ * Pof_tab[].
+ */
+ for (ct = fx = 0; fx < 2; fx++) {
+ if (fx == 0) {
+ sep = "";
+ sepl = 0;
+ tp = Pff_tab;
+ wf = ffg;
+ } else {
+ sep = ";";
+ sepl = 1;
+ tp = Pof_tab;
+ wf = pof;
+ }
+ for (; wf && !FsvFlagX; ct += al ) {
+ while (tp->nm) {
+ if (wf & tp->val)
+ break;
+ tp++;
+ }
+ if (!tp->nm)
+ break;
+ al = (int)strlen(tp->nm) + sepl;
+ bp = alloc_fflbuf(&bp, &bl, al + ct);
+ (void) snpf(bp + ct, al + 1, "%s%s", sep, tp->nm);
+ sep = ",";
+ sepl = 1;
+ wf &= ~(tp->val);
+ }
+ /*
+ * If flag bits remain, print them in hex. If hex output was
+ * specified with +fG, print all flag values, including zero,
+ * in hex.
+ */
+ if (wf || FsvFlagX) {
+ (void) snpf(xbuf, sizeof(xbuf), "0x%lx", wf);
+ al = (int)strlen(xbuf) + sepl;
+ bp = alloc_fflbuf(&bp, &bl, al + ct);
+ (void) snpf(bp + ct, al + 1, "%s%s", sep, xbuf);
+ ct += al;
+ }
+ }
+/*
+ * Make sure there is at least a NUL terminated reply.
+ */
+ if (!bp) {
+ bp = alloc_fflbuf(&bp, &bl, 0);
+ *bp = '\0';
+ }
+ return(bp);
+}
+#endif /* defined(HASFSTRUCT) */
+
+
+/*
+ * print_proc() - print process
+ */
+
+int
+print_proc()
+{
+ char buf[128], *cp;
+ int lc, len, st, ty;
+ int rv = 0;
+ unsigned long ul;
+/*
+ * If nothing in the process has been selected, skip it.
+ */
+ if (!Lp->pss)
+ return(0);
+ if (Fterse) {
+
+ /*
+ * The mode is terse and something in the process appears to have
+ * been selected. Make sure of that by looking for a selected file,
+ * so that the HASSECURITY and HASNOSOCKSECURITY option combination
+ * won't produce a false positive result.
+ */
+ for (Lf = Lp->file; Lf; Lf = Lf->next) {
+ if (is_file_sel(Lp, Lf)) {
+ (void) printf("%d\n", Lp->pid);
+ return(1);
+ }
+ }
+ return(0);
+ }
+/*
+ * If fields have been selected, output the process-only ones, provided
+ * that some file has also been selected.
+ */
+ if (Ffield) {
+ for (Lf = Lp->file; Lf; Lf = Lf->next) {
+ if (is_file_sel(Lp, Lf))
+ break;
+ }
+ if (!Lf)
+ return(rv);
+ rv = 1;
+ (void) printf("%c%d%c", LSOF_FID_PID, Lp->pid, Terminator);
+
+#if defined(HASZONES)
+ if (FieldSel[LSOF_FIX_ZONE].st && Fzone && Lp->zn)
+ (void) printf("%c%s%c", LSOF_FID_ZONE, Lp->zn, Terminator);
+#endif /* defined(HASZONES) */
+
+#if defined(HASSELINUX)
+ if (FieldSel[LSOF_FIX_CNTX].st && Fcntx && Lp->cntx && CntxStatus)
+ (void) printf("%c%s%c", LSOF_FID_CNTX, Lp->cntx, Terminator);
+#endif /* defined(HASSELINUX) */
+
+ if (FieldSel[LSOF_FIX_PGID].st && Fpgid)
+ (void) printf("%c%d%c", LSOF_FID_PGID, Lp->pgid, Terminator);
+
+#if defined(HASPPID)
+ if (FieldSel[LSOF_FIX_PPID].st && Fppid)
+ (void) printf("%c%d%c", LSOF_FID_PPID, Lp->ppid, Terminator);
+#endif /* defined(HASPPID) */
+
+ if (FieldSel[LSOF_FIX_CMD].st) {
+ putchar(LSOF_FID_CMD);
+ safestrprt(Lp->cmd ? Lp->cmd : "(unknown)", stdout, 0);
+ putchar(Terminator);
+ }
+ if (FieldSel[LSOF_FIX_UID].st)
+ (void) printf("%c%d%c", LSOF_FID_UID, (int)Lp->uid, Terminator);
+ if (FieldSel[LSOF_FIX_LOGIN].st) {
+ cp = printuid((UID_ARG)Lp->uid, &ty);
+ if (ty == 0)
+ (void) printf("%c%s%c", LSOF_FID_LOGIN, cp, Terminator);
+ }
+ if (Terminator == '\0')
+ putchar('\n');
+ }
+/*
+ * Print files.
+ */
+ for (Lf = Lp->file; Lf; Lf = Lf->next) {
+ if (!is_file_sel(Lp, Lf))
+ continue;
+ rv = 1;
+ /*
+ * If no field output selected, print dialects-specific formatted
+ * output.
+ */
+ if (!Ffield) {
+ print_file();
+ continue;
+ }
+ /*
+ * Print selected fields.
+ */
+ lc = st = 0;
+ if (FieldSel[LSOF_FIX_FD].st) {
+ for (cp = Lf->fd; *cp == ' '; cp++)
+ ;
+ if (*cp) {
+ (void) printf("%c%s%c", LSOF_FID_FD, cp, Terminator);
+ lc++;
+ }
+ }
+ if (FieldSel[LSOF_FIX_ACCESS].st) {
+ (void) printf("%c%c%c",
+ LSOF_FID_ACCESS, Lf->access, Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_LOCK].st) {
+ (void) printf("%c%c%c", LSOF_FID_LOCK, Lf->lock, Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_TYPE].st) {
+ for (cp = Lf->type; *cp == ' '; cp++)
+ ;
+ if (*cp) {
+ (void) printf("%c%s%c", LSOF_FID_TYPE, cp, Terminator);
+ lc++;
+ }
+ }
+
+#if defined(HASFSTRUCT)
+ if (FieldSel[LSOF_FIX_FA].st && (Fsv & FSV_FA)
+ && (Lf->fsv & FSV_FA)) {
+ (void) printf("%c%s%c", LSOF_FID_FA,
+ print_kptr(Lf->fsa, (char *)NULL, 0), Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_CT].st && (Fsv & FSV_CT)
+ && (Lf->fsv & FSV_CT)) {
+ (void) printf("%c%ld%c", LSOF_FID_CT, Lf->fct, Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_FG].st && (Fsv & FSV_FG)
+ && (Lf->fsv & FSV_FG) && (FsvFlagX || Lf->ffg || Lf->pof)) {
+ (void) printf("%c%s%c", LSOF_FID_FG,
+ print_fflags(Lf->ffg, Lf->pof), Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_NI].st && (Fsv & FSV_NI)
+ && (Lf->fsv & FSV_NI)) {
+ (void) printf("%c%s%c", LSOF_FID_NI,
+ print_kptr(Lf->fna, (char *)NULL, 0), Terminator);
+ lc++;
+ }
+#endif /* defined(HASFSTRUCT) */
+
+ if (FieldSel[LSOF_FIX_DEVCH].st && Lf->dev_ch && Lf->dev_ch[0]) {
+ for (cp = Lf->dev_ch; *cp == ' '; cp++)
+ ;
+ if (*cp) {
+ (void) printf("%c%s%c", LSOF_FID_DEVCH, cp, Terminator);
+ lc++;
+ }
+ }
+ if (FieldSel[LSOF_FIX_DEVN].st && Lf->dev_def) {
+ if (sizeof(unsigned long) > sizeof(dev_t))
+ ul = (unsigned long)((unsigned int)Lf->dev);
+ else
+ ul = (unsigned long)Lf->dev;
+ (void) printf("%c0x%lx%c", LSOF_FID_DEVN, ul, Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_RDEV].st && Lf->rdev_def) {
+ if (sizeof(unsigned long) > sizeof(dev_t))
+ ul = (unsigned long)((unsigned int)Lf->rdev);
+ else
+ ul = (unsigned long)Lf->rdev;
+ (void) printf("%c0x%lx%c", LSOF_FID_RDEV, ul, Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_SIZE].st && Lf->sz_def) {
+ putchar(LSOF_FID_SIZE);
+
+#if defined(HASPRINTSZ)
+ cp = HASPRINTSZ(Lf);
+#else /* !defined(HASPRINTSZ) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_d, Lf->sz);
+ cp = buf;
+#endif /* defined(HASPRINTSZ) */
+
+ (void) printf("%s", cp);
+ putchar(Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_OFFSET].st && Lf->off_def) {
+ putchar(LSOF_FID_OFFSET);
+
+#if defined(HASPRINTOFF)
+ cp = HASPRINTOFF(Lf, 0);
+#else /* !defined(HASPRINTOFF) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_0t, Lf->off);
+ cp = buf;
+#endif /* defined(HASPRINTOFF) */
+
+ len = strlen(cp);
+ if (OffDecDig && len > (OffDecDig + 2)) {
+
+#if defined(HASPRINTOFF)
+ cp = HASPRINTOFF(Lf, 1);
+#else /* !defined(HASPRINTOFF) */
+ (void) snpf(buf, sizeof(buf), SzOffFmt_x, Lf->off);
+ cp = buf;
+#endif /* defined(HASPRINTOFF) */
+
+ }
+ (void) printf("%s", cp);
+ putchar(Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_INODE].st && Lf->inp_ty == 1) {
+ putchar(LSOF_FID_INODE);
+ (void) printf(InodeFmt_d, Lf->inode);
+ putchar(Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_NLINK].st && Lf->nlink_def) {
+ (void) printf("%c%ld%c", LSOF_FID_NLINK, Lf->nlink, Terminator);
+ lc++;
+ }
+ if (FieldSel[LSOF_FIX_PROTO].st && Lf->inp_ty == 2) {
+ for (cp = Lf->iproto; *cp == ' '; cp++)
+ ;
+ if (*cp) {
+ (void) printf("%c%s%c", LSOF_FID_PROTO, cp, Terminator);
+ lc++;
+ }
+ }
+ if (FieldSel[LSOF_FIX_STREAM].st && Lf->nm && Lf->is_stream) {
+ if (strncmp(Lf->nm, "STR:", 4) == 0
+ || strcmp(Lf->iproto, "STR") == 0) {
+ putchar(LSOF_FID_STREAM);
+ printname(0);
+ putchar(Terminator);
+ lc++;
+ st++;
+ }
+ }
+ if (st == 0 && FieldSel[LSOF_FIX_NAME].st) {
+ putchar(LSOF_FID_NAME);
+ printname(0);
+ putchar(Terminator);
+ lc++;
+ }
+ if (Lf->lts.type >= 0 && FieldSel[LSOF_FIX_TCPTPI].st) {
+ print_tcptpi(0);
+ lc++;
+ }
+ if (Terminator == '\0' && lc)
+ putchar('\n');
+ }
+ return(rv);
+}
--- /dev/null
+/*
+ * proto.h - common function prototypes for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * $Id: proto.h,v 1.34 2008/10/21 16:21:41 abe Exp $
+ */
+
+
+#if !defined(PROTO_H)
+#define PROTO_H 1
+
+
+/*
+ * The _PROTOTYPE macro provides strict ANSI C prototypes if __STDC__
+ * is defined, and old-style K&R prototypes otherwise.
+ *
+ * (With thanks to Andy Tanenbaum)
+ */
+
+# if defined(__STDC__)
+#define _PROTOTYPE(function, params) function params
+# else /* !defined(__STDC__) */
+#define _PROTOTYPE(function, params) function()
+# endif /* defined(__STDC__) */
+
+
+/*
+ * The following define keeps gcc>=2.7 from complaining about the failure
+ * of the Exit() function to return.
+ *
+ * Paul Eggert supplied it.
+ */
+
+# if defined(__GNUC__) && !(__GNUC__<2 || (__GNUC__==2 && __GNUC_MINOR__<7))
+#define exiting __attribute__((__noreturn__))
+# else /* !gcc || gcc<2.7 */
+#define exiting
+# endif /* gcc && gcc>=2.7 */
+
+
+_PROTOTYPE(extern void add_nma,(char *cp, int len));
+_PROTOTYPE(extern void alloc_lfile,(char *nm, int num));
+_PROTOTYPE(extern void alloc_lproc,(int pid, int pgid, int ppid, UID_ARG uid, char *cmd, int pss, int sf));
+_PROTOTYPE(extern void build_IPstates,(void));
+_PROTOTYPE(extern void childx,(void));
+_PROTOTYPE(extern int ck_fd_status,(char *nm, int num));
+_PROTOTYPE(extern int ck_file_arg,(int i, int ac, char *av[], int fv, int rs, struct stat *sbp));
+_PROTOTYPE(extern void ckkv,(char *d, char *er, char *ev, char *ea));
+_PROTOTYPE(extern void clr_devtab,(void));
+_PROTOTYPE(extern int compdev,(COMP_P *a1, COMP_P *a2));
+_PROTOTYPE(extern int comppid,(COMP_P *a1, COMP_P *a2));
+
+# if defined(WILLDROPGID)
+_PROTOTYPE(extern void dropgid,(void));
+# endif /* defined(WILLDROPGID) */
+
+_PROTOTYPE(extern char *endnm,(size_t *sz));
+_PROTOTYPE(extern int enter_cmd_rx,(char *x));
+_PROTOTYPE(extern void enter_dev_ch,(char *m));
+_PROTOTYPE(extern int enter_dir,(char *d, int descend));
+_PROTOTYPE(extern int enter_fd,(char *f));
+_PROTOTYPE(extern int enter_network_address,(char *na));
+_PROTOTYPE(extern int enter_id,(enum IDType ty, char *p));
+_PROTOTYPE(extern void enter_IPstate,(char *ty, char *nm, int nr));
+_PROTOTYPE(extern void enter_nm,(char *m));
+
+# if defined(HASTCPUDPSTATE)
+_PROTOTYPE(extern int enter_state_spec,(char *ss));
+# endif /* defined(HASTCPUDPSTATE) */
+
+_PROTOTYPE(extern int enter_str_lst,(char *opt, char *s, struct str_lst **lp,
+ int *incl, int *excl));
+_PROTOTYPE(extern int enter_uid,(char *us));
+_PROTOTYPE(extern void ent_inaddr,(unsigned char *la, int lp, unsigned char *fa, int fp, int af));
+_PROTOTYPE(extern int examine_lproc,(void));
+_PROTOTYPE(extern void Exit,(int xv)) exiting;
+_PROTOTYPE(extern void find_ch_ino,(void));
+_PROTOTYPE(extern void free_lproc,(struct lproc *lp));
+_PROTOTYPE(extern void gather_proc_info,(void));
+_PROTOTYPE(extern char *gethostnm,(unsigned char *ia, int af));
+
+# if !defined(GET_MAX_FD)
+/*
+ * This is not strictly a prototype, but GET_MAX_FD is the name of the
+ * function that, in lieu of getdtablesize(), returns the maximum file
+ * descriptor plus one (or file descriptor count). GET_MAX_FD may be
+ * defined in the dialect's machine.h. If it is not, the following
+ * selects getdtablesize().
+ */
+
+#define GET_MAX_FD getdtablesize
+# endif /* !defined(GET_MAX_FD) */
+
+_PROTOTYPE(extern int hashbyname,(char *nm, int mod));
+_PROTOTYPE(extern void hashSfile,(void));
+_PROTOTYPE(extern void initialize,(void));
+_PROTOTYPE(extern int is_cmd_excl,(char *cmd, short *pss, short *sf));
+_PROTOTYPE(extern int is_nw_addr,(unsigned char *ia, int p, int af));
+_PROTOTYPE(extern int is_proc_excl,(int pid, int pgid, UID_ARG uid, short *pss, short *sf));
+_PROTOTYPE(extern int is_readable,(char *path, int msg));
+_PROTOTYPE(extern int kread,(KA_T addr, char *buf, READLEN_T len));
+_PROTOTYPE(extern void link_lfile,(void));
+_PROTOTYPE(extern struct l_dev *lkupdev,(dev_t *dev,dev_t *rdev,int i,int r));
+_PROTOTYPE(extern int main,(int argc, char *argv[]));
+_PROTOTYPE(extern int lstatsafely,(char *path, struct stat *buf));
+_PROTOTYPE(extern char *mkstrcpy,(char *src, MALLOC_S *rlp));
+_PROTOTYPE(extern char *mkstrcat,(char *s1, int l1, char *s2, int l2, char *s3, int l3, MALLOC_S *clp));
+_PROTOTYPE(extern int printdevname,(dev_t *dev, dev_t *rdev, int f, int nty));
+_PROTOTYPE(extern void print_file,(void));
+_PROTOTYPE(extern void print_init,(void));
+_PROTOTYPE(extern void printname,(int nl));
+_PROTOTYPE(extern char *print_kptr,(KA_T kp, char *buf, size_t bufl));
+_PROTOTYPE(extern int print_proc,(void));
+_PROTOTYPE(extern void printrawaddr,(struct sockaddr *sa));
+_PROTOTYPE(extern void print_tcptpi,(int nl));
+_PROTOTYPE(extern char *printuid,(UID_ARG uid, int *ty));
+_PROTOTYPE(extern void printunkaf,(int fam, int ty));
+_PROTOTYPE(extern char *printsockty,(int ty));
+_PROTOTYPE(extern void process_file,(KA_T fp));
+_PROTOTYPE(extern void process_node,(KA_T f));
+_PROTOTYPE(extern char *Readlink,(char *arg));
+_PROTOTYPE(extern void readdev,(int skip));
+_PROTOTYPE(extern struct mounts *readmnt,(void));
+_PROTOTYPE(extern void rereaddev,(void));
+_PROTOTYPE(extern int safestrlen,(char *sp, int flags));
+_PROTOTYPE(extern void safestrprtn,(char *sp, int len, FILE *fs, int flags));
+_PROTOTYPE(extern void safestrprt,(char *sp, FILE *fs, int flags));
+_PROTOTYPE(extern int statsafely,(char *path, struct stat *buf));
+_PROTOTYPE(extern void stkdir,(char *p));
+_PROTOTYPE(extern void usage,(int xv, int fh, int version));
+_PROTOTYPE(extern int util_strftime,(char *fmtr, int fmtl, char *fmt));
+_PROTOTYPE(extern int vfy_dev,(struct l_dev *dp));
+_PROTOTYPE(extern char *x2dev,(char *s, dev_t *d));
+
+# if defined(HASBLKDEV)
+_PROTOTYPE(extern void find_bl_ino,(void));
+_PROTOTYPE(extern struct l_dev *lkupbdev,(dev_t *dev,dev_t *rdev,int i,int r));
+_PROTOTYPE(extern int printbdevname,(dev_t *dev, dev_t *rdev, int f));
+# endif /* defined(HASBLKDEV) */
+
+# if defined(HASCDRNODE)
+_PROTOTYPE(extern int readcdrnode,(KA_T ca, struct cdrnode *c));
+# endif /* defined(HASCDRNODE) */
+
+# if defined(HASDCACHE)
+_PROTOTYPE(extern void alloc_dcache,(void));
+_PROTOTYPE(extern void crc,(char *b, int l, unsigned *s));
+_PROTOTYPE(extern void crdbld,(void));
+_PROTOTYPE(extern int ctrl_dcache,(char *p));
+_PROTOTYPE(extern int dcpath,(int rw, int npw));
+_PROTOTYPE(extern int open_dcache,(int m, int r, struct stat *sb));
+_PROTOTYPE(extern int read_dcache,(void));
+_PROTOTYPE(extern int wr2DCfd,(char *b, unsigned *c));
+_PROTOTYPE(extern void write_dcache,(void));
+# endif /* defined(HASDCACHE) */
+
+# if defined(HASFIFONODE)
+_PROTOTYPE(extern int readfifonode,(KA_T fa, struct fifonode *f));
+# endif /* defined(HASFIFONODE) */
+
+# if defined(HASFSTRUCT)
+_PROTOTYPE(extern char *print_fflags,(long ffg, long pof));
+# endif /* defined(HASFSTRUCT) */
+
+# if defined(HASGNODE)
+_PROTOTYPE(extern int readgnode,(KA_T ga, struct gnode *g));
+# endif /* defined(HASGNODE) */
+
+# if defined(HASKQUEUE)
+_PROTOTYPE(extern void process_kqueue,(KA_T ka));
+# endif /* defined(HASKQUEUE) */
+
+# if defined(HASHSNODE)
+_PROTOTYPE(extern int readhsnode,(KA_T ha, struct hsnode *h));
+# endif /* defined(HASHSNODE) */
+
+# if defined(HASINODE)
+_PROTOTYPE(extern int readinode,(KA_T ia, struct inode *i));
+# endif /* defined(HASINODE) */
+
+# if defined(HASNCACHE)
+_PROTOTYPE(extern void ncache_load,(void));
+_PROTOTYPE(extern char *ncache_lookup,(char *buf, int blen, int *fp));
+# endif /* defined(HASNCACHE) */
+
+# if defined(HASNLIST)
+_PROTOTYPE(extern void build_Nl,(struct drive_Nl *d));
+_PROTOTYPE(extern int get_Nl_value,(char *nn, struct drive_Nl *d, KA_T *v));
+# endif /* defined(HASNLIST) */
+
+# if defined(HASPIPENODE)
+_PROTOTYPE(extern int readpipenode,(KA_T pa, struct pipenode *p));
+# endif /* defined(HASPIPENODE) */
+
+# if defined(HASPRINTDEV)
+_PROTOTYPE(extern char *HASPRINTDEV,(struct lfile *lf, dev_t *dev));
+# endif /* defined(HASPRINTDEV) */
+
+# if defined(HASPRINTINO)
+_PROTOTYPE(extern char *HASPRINTINO,(struct lfile *lf));
+# endif /* defined(HASPRINTINO) */
+
+# if defined(HASPRINTNM)
+_PROTOTYPE(extern void HASPRINTNM,(struct lfile *lf));
+# endif /* defined(HASPRINTNM) */
+
+# if defined(HASPRINTOFF)
+_PROTOTYPE(extern char *HASPRINTOFF,(struct lfile *lf, int ty));
+# endif /* defined(HASPRINTOFF) */
+
+# if defined(HASPRINTSZ)
+_PROTOTYPE(extern char *HASPRINTSZ,(struct lfile *lf));
+# endif /* defined(HASPRINTSZ) */
+
+# if defined(HASPRIVNMCACHE)
+_PROTOTYPE(extern int HASPRIVNMCACHE,(struct lfile *lf));
+# endif /* defined(HASPRIVNMCACHE) */
+
+# if !defined(HASPRIVPRIPP)
+_PROTOTYPE(extern void printiproto,(int p));
+# endif /* !defined(HASPRIVPRIPP) */
+
+# if defined(HASRNODE)
+_PROTOTYPE(extern int readrnode,(KA_T ra, struct rnode *r));
+# endif /* defined(HASRNODE) */
+
+# if defined(HASSPECDEVD)
+_PROTOTYPE(extern void HASSPECDEVD,(char *p, struct stat *s));
+# endif /* defined(HASSPECDEVD) */
+
+# if defined(HASSNODE)
+_PROTOTYPE(extern int readsnode,(KA_T sa, struct snode *s));
+# endif /* defined(HASSNODE) */
+
+# if defined(HASSTREAMS)
+_PROTOTYPE(extern int readstdata,(KA_T addr, struct stdata *buf));
+_PROTOTYPE(extern int readsthead,(KA_T addr, struct queue *buf));
+_PROTOTYPE(extern int readstidnm,(KA_T addr, char *buf, READLEN_T len));
+_PROTOTYPE(extern int readstmin,(KA_T addr, struct module_info *buf));
+_PROTOTYPE(extern int readstqinit,(KA_T addr, struct qinit *buf));
+# endif /* defined(HASSTREAMS) */
+
+# if defined(HASTMPNODE)
+_PROTOTYPE(extern int readtnode,(KA_T ta, struct tmpnode *t));
+# endif /* defined(HASTMPNODE) */
+
+# if defined(HASVNODE)
+_PROTOTYPE(extern int readvnode,(KA_T va, struct vnode *v));
+# endif /* defined(HASVNODE) */
+
+# if defined(USE_LIB_SNPF)
+_PROTOTYPE(extern int snpf,(char *str, int len, char *fmt, ...));
+# endif /* defined(USE_LIB_SNPF) */
+
+# endif /* !defined(PROTO_H) */
--- /dev/null
+/*
+ * regex.h -- regular expression definitions for lsof
+ *
+ * This header file is used only when the dialect has no POSIX-conformant
+ * regular expression function set. When that is the case, the dialect's
+ * machine.h will define USE_LIB_REGEX.
+ *
+ * When the dialect has a POSIX-conformant regular expression function set,
+ * USE_LIB_REGEX is not defined and this header file #include's <regex.h>.
+ *
+ * V. Abell <abe@purdue.edu>
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2000 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * This software has been adapted from snprintf.c in sendmail 8.9.3. It
+ * is subject to the sendmail copyright statements listed below, and the
+ * sendmail licensing terms stated in the sendmail LICENSE file comment
+ * section of this file.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#ifdef USE_LIB_REGEX
+/*
+ * This section comes from GLIBC 2.2. It is used only when the dialect
+ * has no POSIX-conformant regular expression function set. When that is
+ * the case, the dialect's machine.h will define USE_LIB_REGEX.
+ */
+
+/* Definitions for data structures and routines for the regular
+ expression library, version 0.12.
+ Copyright (C) 1985,1989-1993,1995-1998, 2000 Free Software Foundation, Inc.
+
+ This file is part of the GNU C Library. Its master source is NOT part of
+ the C library, however. The master source lives in /gd/gnu/lib.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Library General Public License as
+ published by the Free Software Foundation; either version 2 of the
+ License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Library General Public License for more details.
+
+ You should have received a copy of the GNU Library General Public
+ License along with the GNU C Library; see the file COPYING.LIB. If not,
+ write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ Boston, MA 02111-1307, USA. */
+
+#ifndef _REGEX_H
+#define _REGEX_H 1
+
+/* Allow the use in C++ code. */
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* POSIX says that <sys/types.h> must be included (by the caller) before
+ <regex.h>. */
+
+#if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
+/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
+ should be there. */
+# include <stddef.h>
+#endif
+
+/* The following two types have to be signed and unsigned integer type
+ wide enough to hold a value of a pointer. For most ANSI compilers
+ ptrdiff_t and size_t should be likely OK. Still size of these two
+ types is 2 for Microsoft C. Ugh... */
+typedef long int s_reg_t;
+typedef unsigned long int active_reg_t;
+
+/* The following bits are used to determine the regexp syntax we
+ recognize. The set/not-set meanings are chosen so that Emacs syntax
+ remains the value 0. The bits are given in alphabetical order, and
+ the definitions shifted by one from the previous bit; thus, when we
+ add or remove a bit, only one other definition need change. */
+typedef unsigned long int reg_syntax_t;
+
+/* If this bit is not set, then \ inside a bracket expression is literal.
+ If set, then such a \ quotes the following character. */
+#define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
+
+/* If this bit is not set, then + and ? are operators, and \+ and \? are
+ literals.
+ If set, then \+ and \? are operators and + and ? are literals. */
+#define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
+
+/* If this bit is set, then character classes are supported. They are:
+ [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
+ [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
+ If not set, then character classes are not supported. */
+#define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
+
+/* If this bit is set, then ^ and $ are always anchors (outside bracket
+ expressions, of course).
+ If this bit is not set, then it depends:
+ ^ is an anchor if it is at the beginning of a regular
+ expression or after an open-group or an alternation operator;
+ $ is an anchor if it is at the end of a regular expression, or
+ before a close-group or an alternation operator.
+
+ This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
+ POSIX draft 11.2 says that * etc. in leading positions is undefined.
+ We already implemented a previous draft which made those constructs
+ invalid, though, so we haven't changed the code back. */
+#define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
+
+/* If this bit is set, then special characters are always special
+ regardless of where they are in the pattern.
+ If this bit is not set, then special characters are special only in
+ some contexts; otherwise they are ordinary. Specifically,
+ * + ? and intervals are only special when not after the beginning,
+ open-group, or alternation operator. */
+#define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
+
+/* If this bit is set, then *, +, ?, and { cannot be first in an re or
+ immediately after an alternation or begin-group operator. */
+#define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
+
+/* If this bit is set, then . matches newline.
+ If not set, then it doesn't. */
+#define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
+
+/* If this bit is set, then . doesn't match NUL.
+ If not set, then it does. */
+#define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
+
+/* If this bit is set, nonmatching lists [^...] do not match newline.
+ If not set, they do. */
+#define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
+
+/* If this bit is set, either \{...\} or {...} defines an
+ interval, depending on RE_NO_BK_BRACES.
+ If not set, \{, \}, {, and } are literals. */
+#define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
+
+/* If this bit is set, +, ? and | aren't recognized as operators.
+ If not set, they are. */
+#define RE_LIMITED_OPS (RE_INTERVALS << 1)
+
+/* If this bit is set, newline is an alternation operator.
+ If not set, newline is literal. */
+#define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
+
+/* If this bit is set, then `{...}' defines an interval, and \{ and \}
+ are literals.
+ If not set, then `\{...\}' defines an interval. */
+#define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
+
+/* If this bit is set, (...) defines a group, and \( and \) are literals.
+ If not set, \(...\) defines a group, and ( and ) are literals. */
+#define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
+
+/* If this bit is set, then \<digit> matches <digit>.
+ If not set, then \<digit> is a back-reference. */
+#define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
+
+/* If this bit is set, then | is an alternation operator, and \| is literal.
+ If not set, then \| is an alternation operator, and | is literal. */
+#define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
+
+/* If this bit is set, then an ending range point collating higher
+ than the starting range point, as in [z-a], is invalid.
+ If not set, then when ending range point collates higher than the
+ starting range point, the range is ignored. */
+#define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
+
+/* If this bit is set, then an unmatched ) is ordinary.
+ If not set, then an unmatched ) is invalid. */
+#define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
+
+/* If this bit is set, succeed as soon as we match the whole pattern,
+ without further backtracking. */
+#define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
+
+/* If this bit is set, do not process the GNU regex operators.
+ If not set, then the GNU regex operators are recognized. */
+#define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
+
+/* If this bit is set, turn on internal regex debugging.
+ If not set, and debugging was on, turn it off.
+ This only works if regex.c is compiled -DDEBUG.
+ We define this bit always, so that all that's needed to turn on
+ debugging is to recompile regex.c; the calling code can always have
+ this bit set, and it won't affect anything in the normal case. */
+#define RE_DEBUG (RE_NO_GNU_OPS << 1)
+
+/* This global variable defines the particular regexp syntax to use (for
+ some interfaces). When a regexp is compiled, the syntax used is
+ stored in the pattern buffer, so changing this does not affect
+ already-compiled regexps. */
+extern reg_syntax_t re_syntax_options;
+\f
+/* Define combinations of the above bits for the standard possibilities.
+ (The [[[ comments delimit what gets put into the Texinfo file, so
+ don't delete them!) */
+/* [[[begin syntaxes]]] */
+#define RE_SYNTAX_EMACS 0
+
+#define RE_SYNTAX_AWK \
+ (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
+ | RE_NO_BK_PARENS | RE_NO_BK_REFS \
+ | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
+ | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
+
+#define RE_SYNTAX_GNU_AWK \
+ ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
+ & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
+
+#define RE_SYNTAX_POSIX_AWK \
+ (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
+ | RE_INTERVALS | RE_NO_GNU_OPS)
+
+#define RE_SYNTAX_GREP \
+ (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
+ | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
+ | RE_NEWLINE_ALT)
+
+#define RE_SYNTAX_EGREP \
+ (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
+ | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
+ | RE_NO_BK_VBAR)
+
+#define RE_SYNTAX_POSIX_EGREP \
+ (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
+
+/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
+#define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
+
+#define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
+
+/* Syntax bits common to both basic and extended POSIX regex syntax. */
+#define _RE_SYNTAX_POSIX_COMMON \
+ (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
+ | RE_INTERVALS | RE_NO_EMPTY_RANGES)
+
+#define RE_SYNTAX_POSIX_BASIC \
+ (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
+
+/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
+ RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
+ isn't minimal, since other operators, such as \`, aren't disabled. */
+#define RE_SYNTAX_POSIX_MINIMAL_BASIC \
+ (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
+
+#define RE_SYNTAX_POSIX_EXTENDED \
+ (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
+ | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
+ | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
+
+/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
+ removed and RE_NO_BK_REFS is added. */
+#define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
+ (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
+ | RE_NO_BK_PARENS | RE_NO_BK_REFS \
+ | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
+/* [[[end syntaxes]]] */
+\f
+/* Maximum number of duplicates an interval can allow. Some systems
+ (erroneously) define this in other header files, but we want our
+ value, so remove any previous define. */
+#ifdef RE_DUP_MAX
+# undef RE_DUP_MAX
+#endif
+/* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
+#define RE_DUP_MAX (0x7fff)
+
+
+/* POSIX `cflags' bits (i.e., information for `regcomp'). */
+
+/* If this bit is set, then use extended regular expression syntax.
+ If not set, then use basic regular expression syntax. */
+#define REG_EXTENDED 1
+
+/* If this bit is set, then ignore case when matching.
+ If not set, then case is significant. */
+#define REG_ICASE (REG_EXTENDED << 1)
+
+/* If this bit is set, then anchors do not match at newline
+ characters in the string.
+ If not set, then anchors do match at newlines. */
+#define REG_NEWLINE (REG_ICASE << 1)
+
+/* If this bit is set, then report only success or fail in regexec.
+ If not set, then returns differ between not matching and errors. */
+#define REG_NOSUB (REG_NEWLINE << 1)
+
+
+/* POSIX `eflags' bits (i.e., information for regexec). */
+
+/* If this bit is set, then the beginning-of-line operator doesn't match
+ the beginning of the string (presumably because it's not the
+ beginning of a line).
+ If not set, then the beginning-of-line operator does match the
+ beginning of the string. */
+#define REG_NOTBOL 1
+
+/* Like REG_NOTBOL, except for the end-of-line. */
+#define REG_NOTEOL (1 << 1)
+
+
+/* If any error codes are removed, changed, or added, update the
+ `re_error_msg' table in regex.c. */
+typedef enum
+{
+#ifdef _XOPEN_SOURCE
+ REG_ENOSYS = -1, /* This will never happen for this implementation. */
+#endif
+
+ REG_NOERROR = 0, /* Success. */
+ REG_NOMATCH, /* Didn't find a match (for regexec). */
+
+ /* POSIX regcomp return error codes. (In the order listed in the
+ standard.) */
+ REG_BADPAT, /* Invalid pattern. */
+ REG_ECOLLATE, /* Not implemented. */
+ REG_ECTYPE, /* Invalid character class name. */
+ REG_EESCAPE, /* Trailing backslash. */
+ REG_ESUBREG, /* Invalid back reference. */
+ REG_EBRACK, /* Unmatched left bracket. */
+ REG_EPAREN, /* Parenthesis imbalance. */
+ REG_EBRACE, /* Unmatched \{. */
+ REG_BADBR, /* Invalid contents of \{\}. */
+ REG_ERANGE, /* Invalid range end. */
+ REG_ESPACE, /* Ran out of memory. */
+ REG_BADRPT, /* No preceding re for repetition op. */
+
+ /* Error codes we've added. */
+ REG_EEND, /* Premature end. */
+ REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
+ REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
+} reg_errcode_t;
+\f
+/* This data structure represents a compiled pattern. Before calling
+ the pattern compiler, the fields `buffer', `allocated', `fastmap',
+ `translate', and `no_sub' can be set. After the pattern has been
+ compiled, the `re_nsub' field is available. All other fields are
+ private to the regex routines. */
+
+#ifndef RE_TRANSLATE_TYPE
+# define RE_TRANSLATE_TYPE char *
+#endif
+
+struct re_pattern_buffer
+{
+/* [[[begin pattern_buffer]]] */
+ /* Space that holds the compiled pattern. It is declared as
+ `unsigned char *' because its elements are
+ sometimes used as array indexes. */
+ unsigned char *buffer;
+
+ /* Number of bytes to which `buffer' points. */
+ unsigned long int allocated;
+
+ /* Number of bytes actually used in `buffer'. */
+ unsigned long int used;
+
+ /* Syntax setting with which the pattern was compiled. */
+ reg_syntax_t syntax;
+
+ /* Pointer to a fastmap, if any, otherwise zero. re_search uses
+ the fastmap, if there is one, to skip over impossible
+ starting points for matches. */
+ char *fastmap;
+
+ /* Either a translate table to apply to all characters before
+ comparing them, or zero for no translation. The translation
+ is applied to a pattern when it is compiled and to a string
+ when it is matched. */
+ RE_TRANSLATE_TYPE translate;
+
+ /* Number of subexpressions found by the compiler. */
+ size_t re_nsub;
+
+ /* Zero if this pattern cannot match the empty string, one else.
+ Well, in truth it's used only in `re_search_2', to see
+ whether or not we should use the fastmap, so we don't set
+ this absolutely perfectly; see `re_compile_fastmap' (the
+ `duplicate' case). */
+ unsigned can_be_null : 1;
+
+ /* If REGS_UNALLOCATED, allocate space in the `regs' structure
+ for `max (RE_NREGS, re_nsub + 1)' groups.
+ If REGS_REALLOCATE, reallocate space if necessary.
+ If REGS_FIXED, use what's there. */
+#define REGS_UNALLOCATED 0
+#define REGS_REALLOCATE 1
+#define REGS_FIXED 2
+ unsigned regs_allocated : 2;
+
+ /* Set to zero when `regex_compile' compiles a pattern; set to one
+ by `re_compile_fastmap' if it updates the fastmap. */
+ unsigned fastmap_accurate : 1;
+
+ /* If set, `re_match_2' does not return information about
+ subexpressions. */
+ unsigned no_sub : 1;
+
+ /* If set, a beginning-of-line anchor doesn't match at the
+ beginning of the string. */
+ unsigned not_bol : 1;
+
+ /* Similarly for an end-of-line anchor. */
+ unsigned not_eol : 1;
+
+ /* If true, an anchor at a newline matches. */
+ unsigned newline_anchor : 1;
+
+/* [[[end pattern_buffer]]] */
+};
+
+typedef struct re_pattern_buffer regex_t;
+\f
+/* Type for byte offsets within the string. POSIX mandates this. */
+typedef int regoff_t;
+
+
+/* This is the structure we store register match data in. See
+ regex.texinfo for a full description of what registers match. */
+struct re_registers
+{
+ unsigned num_regs;
+ regoff_t *start;
+ regoff_t *end;
+};
+
+
+/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
+ `re_match_2' returns information about at least this many registers
+ the first time a `regs' structure is passed. */
+#ifndef RE_NREGS
+# define RE_NREGS 30
+#endif
+
+
+/* POSIX specification for registers. Aside from the different names than
+ `re_registers', POSIX uses an array of structures, instead of a
+ structure of arrays. */
+typedef struct
+{
+ regoff_t rm_so; /* Byte offset from string's start to substring's start. */
+ regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
+} regmatch_t;
+\f
+/* Declarations for routines. */
+
+/* To avoid duplicating every routine declaration -- once with a
+ prototype (if we are ANSI), and once without (if we aren't) -- we
+ use the following macro to declare argument types. This
+ unfortunately clutters up the declarations a bit, but I think it's
+ worth it. */
+
+#if __STDC__
+
+# define _RE_ARGS(args) args
+
+#else /* not __STDC__ */
+
+# define _RE_ARGS(args) ()
+
+#endif /* not __STDC__ */
+
+/* Sets the current default syntax to SYNTAX, and return the old syntax.
+ You can also simply assign to the `re_syntax_options' variable. */
+extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
+
+/* Compile the regular expression PATTERN, with length LENGTH
+ and syntax given by the global `re_syntax_options', into the buffer
+ BUFFER. Return NULL if successful, and an error string if not. */
+extern const char *re_compile_pattern
+ _RE_ARGS ((const char *pattern, size_t length,
+ struct re_pattern_buffer *buffer));
+
+
+/* Compile a fastmap for the compiled pattern in BUFFER; used to
+ accelerate searches. Return 0 if successful and -2 if was an
+ internal error. */
+extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
+
+
+/* Search in the string STRING (with length LENGTH) for the pattern
+ compiled into BUFFER. Start searching at position START, for RANGE
+ characters. Return the starting position of the match, -1 for no
+ match, or -2 for an internal error. Also return register
+ information in REGS (if REGS and BUFFER->no_sub are nonzero). */
+extern int re_search
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
+ int length, int start, int range, struct re_registers *regs));
+
+
+/* Like `re_search', but search in the concatenation of STRING1 and
+ STRING2. Also, stop searching at index START + STOP. */
+extern int re_search_2
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
+ int length1, const char *string2, int length2,
+ int start, int range, struct re_registers *regs, int stop));
+
+
+/* Like `re_search', but return how many characters in STRING the regexp
+ in BUFFER matched, starting at position START. */
+extern int re_match
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
+ int length, int start, struct re_registers *regs));
+
+
+/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
+extern int re_match_2
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
+ int length1, const char *string2, int length2,
+ int start, struct re_registers *regs, int stop));
+
+
+/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
+ ENDS. Subsequent matches using BUFFER and REGS will use this memory
+ for recording register information. STARTS and ENDS must be
+ allocated with malloc, and must each be at least `NUM_REGS * sizeof
+ (regoff_t)' bytes long.
+
+ If NUM_REGS == 0, then subsequent matches should allocate their own
+ register data.
+
+ Unless this function is called, the first search or match using
+ PATTERN_BUFFER will allocate its own register data, without
+ freeing the old data. */
+extern void re_set_registers
+ _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
+ unsigned num_regs, regoff_t *starts, regoff_t *ends));
+
+#if defined _REGEX_RE_COMP || defined _LIBC
+# ifndef _CRAY
+/* 4.2 bsd compatibility. */
+extern char *re_comp _RE_ARGS ((const char *));
+extern int re_exec _RE_ARGS ((const char *));
+# endif
+#endif
+
+/* GCC 2.95 and later have "__restrict"; C99 compilers have
+ "restrict", and "configure" may have defined "restrict". */
+#ifndef __restrict
+# if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
+# if defined restrict || 199901L <= __STDC_VERSION__
+# define __restrict restrict
+# else
+# define __restrict
+# endif
+# endif
+#endif
+/* For now unconditionally define __restrict_arr to expand to nothing.
+ Ideally we would have a test for the compiler which allows defining
+ it to restrict. */
+#define __restrict_arr
+
+/* POSIX compatibility. */
+extern int regcomp _RE_ARGS ((regex_t *__restrict __preg,
+ const char *__restrict __pattern,
+ int __cflags));
+
+extern int regexec _RE_ARGS ((const regex_t *__restrict __preg,
+ const char *__restrict __string, size_t __nmatch,
+ regmatch_t __pmatch[__restrict_arr],
+ int __eflags));
+
+extern size_t regerror _RE_ARGS ((int __errcode, const regex_t *__preg,
+ char *__errbuf, size_t __errbuf_size));
+
+extern void regfree _RE_ARGS ((regex_t *__preg));
+
+#ifdef __cplusplus
+}
+#endif /* C++ */
+
+#endif /* regex.h */
+\f
+/*
+Local variables:
+make-backup-files: t
+version-control: t
+trim-versions-without-asking: nil
+End:
+*/
+
+#else /* !defined(USE_LIB_REGEX) */
+#include <regex.h>
+#endif /* defined(USE_LIB_REGEX) */
--- /dev/null
+The scripts in this subdirectory give examples of using lsof's
+field output.
+
+big_brother.perl5 Perl 5 script, contributed by Lionel Cons
+ <Lionel.Cons@cern.ch>, that watches for new
+ network connections.
+
+count_pf.perl Perl 4 or 5 script that runs lsof in repeat
+ mode, gathering process, file, TCP, and UDP
+ counts
+
+count_pf.perl5 Perl 5 script that runs lsof in repeat mode,
+ gathering process, file, TCP, and UDP counts
+
+ This script uses NUL terminated lsof field
+ output.
+
+identd.perl5 Perl 5 script, contributed by Kapil Chowksey
+ <kchowksey@hss.hns.com> that implements an
+ identd server. (Thanks, Kapil!)
+
+idrlogin.perl Perl 4 script that identifies the shell and
+ network source address of users who have logged
+ on from remote locations via rlogin, ssh, or
+ telnet
+
+idrlogin.perl5 Perl 5 script that identifies the shell and
+ network source address of users who have logged
+ on from remote locations via rlogin, ssh, or
+ telnet
+
+list_NULf.perl5 Perl 5 script that prints lsof's NUL terminated
+ field output
+
+list_fields.awk AWK script that prints lsof's field output
+
+list_fields.perl Perl 4 or 5 script that prints lsof's field
+ output
+
+shared.perl5 Perl 5 script that uses +ffn output to produce
+ a list of file descriptors or files shared by
+ processes.
+
+sort_res.perl5 Perl 5 script, contributed by Fabian Frederick
+ <fabian.frederick@gmx.fr>, to display top resource
+ usage.
+
+watch_a_file.perl Perl 4 or 5 script that watches the use of a
+ named file
+
+xusers.awk an AWK (actually NAWK) script, written by
+ Dan A. Mercer <damercer@mmm.com> that, "Prints
+ list of users and applications signed on X
+ workstations." This script was developed
+ and is used with lsof on HP-UX systems.
+
+Vic Abell
+December 28, 1998
--- /dev/null
+
+ Notes on Using the Scripts in This Subdirectory
+
+The scripts in this subdirectory are examples of post-processing
+lsof field output. Some are contributed by lsof users and are
+reproduced substantially as written by those users. Since the
+scripts are examples, they are not guaranteed to work on all UNIX
+dialects. Use them to learn about processing field output, don't
+expect them to be ready for production, and expect to be required
+to modify them to make them work.
+
+If you want to do field output post-processing in a C program, take
+a look at the test suite C library in ../tests/LTlib.c. You may
+be able to adapt it to your needs.
+
+The scripts are written in AWK, Perl 4 (4.036), and Perl 5 (5.001e
+through 5.006). AWK scripts have a suffix of ``.awk''; Perl 4
+(which will work under Perl 5) scripts have a ``.perl4'' suffix;
+and Perl 5 scripts, ``.perl''.
+
+Supply AWK scripts to your AWK interpreter with its -f option. Supply
+lsof field output via a pipe -- e.g.,
+
+ lsof -F | awk -f list_fields.awk
+
+The Perl scripts use the Unix command interpreter line feature to
+specify the location of Perl -- i.e., the first line begins with
+``#!'' and the path to the Perl interpreter follows. If your system
+supports the command interpreter feature, but your Perl interpreters
+have different paths to them, just change the interpreter lines in
+the scripts. These scripts assume:
+
+ Path to: Is:
+ ======= ==
+
+ Perl 4 /usr/local/bin/perl4
+
+ Perl 5 /usr/local/bin/perl
+
+If your system doesn't support the command interpreter feature,
+you'll have to supply the scripts to your Perl interpreter on its
+command line -- e.g.,
+
+ lsof -F | /<path_to_your_perl_4> list_fields.perl
+
+The Perl scripts attempt to establish a path to lsof, putting their
+result in the $LSOF variable. Assuming you'll run them from the
+scripts subdirectory, they look there first, then in the directories
+of the PATH environment variable. If that proves unsuitable, modify
+the &isexec() subroutine calls in the scripts to suit your lsof
+location.
+
+
+Vic Abell
+April 4, 2002
--- /dev/null
+#!/usr/local/bin/perl -w
+#+##############################################################################
+# #
+# File: big_brother.perl #
+# #
+# Description: check the network sockets with lsof to detect new connections #
+# #
+# Contributed by Lionel Cons <Lionel.Cons@cern.ch> #
+# #
+#-##############################################################################
+
+# @(#)big_brother 1.12 08/14/96 Written by Lionel.Cons@cern.ch
+
+# no waranty! use this at your own risks!
+
+#
+# init & setup
+#
+$verbose = 1;
+$lsof_opt = "-itcp -iudp -Di -FcLPn -r 5";
+$SIG{'HUP'} = \&hangup;
+chop($hostname = `/bin/hostname`);
+$fq_hostname = (gethostbyname($hostname))[0];
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+#
+# spy forever...
+#
+$| = 1;
+die "$LSOF is not executable\n" unless -x $LSOF;
+while (1) {
+ $lsof_pid = open(PIPE, "$LSOF $lsof_opt 2>&1 |")
+ || die "can't start $LSOF: $!\n";
+ print "# ", ×tamp, " $LSOF $lsof_opt, pid=$lsof_pid\n"
+ if $verbose;
+ print "#COMMAND PID USER P NAME\n";
+ $printed = $hanguped = $pid = $proto = 0;
+ while (<PIPE>) {
+ if (/^lsof: PID \d+, /) {
+ # fatal error message?
+ print "*** $_";
+ last;
+ } elsif (/^lsof: /) {
+ # warning
+ warn "* $_";
+ } elsif (/^p(\d+)$/) {
+ &flush;
+ $pid = $1;
+ $proto = 0;
+ } elsif (/^c(.*)$/) {
+ $command = $1;
+ } elsif (/^L(.*)$/) {
+ $user = $1;
+ } elsif (/^P(.*)$/) {
+ &flush;
+ $proto = $1;
+ } elsif (/^n(.*)$/) {
+ $name = $1;
+ # replace local hostname by 'localhost'
+ $name =~ s/\Q$fq_hostname\E/localhost/g;
+ $name =~ s/[0-9hms]+ ago//g;
+ } elsif (/^m$/) {
+ &flush;
+ &clean;
+ } else {
+ warn "* bad output ignored: $_";
+ }
+ }
+ kill('INT', $lsof_pid);
+ kill('KILL', $lsof_pid);
+ close(PIPE);
+}
+
+sub hangup {
+ $hanguped = 1;
+ $SIG{'HUP'} = \&hangup;
+}
+
+sub flush {
+ return unless $pid && $proto;
+ return if &skip;
+ $tag = sprintf("%-9s %5d %8s %1s %s", $command, $pid, $user,
+ substr($proto, 0, 1), $name);
+ unless (defined($seen{$tag})) {
+ print "+$tag\n";
+ $printed++;
+ }
+ $seen{$tag} = 1;
+}
+
+sub clean {
+ my(@to_delete, $tag);
+
+ if ($hanguped) {
+ $hanguped = 0;
+ @to_delete = keys(%seen);
+ print "# ", ×tamp, " hangup received, rescanning all connections\n"
+ if $verbose;
+ } else {
+ @to_delete = ();
+ foreach $tag (keys(%seen)) {
+ if ($seen{$tag} == 0) {
+ # not seen this time: delete it
+ push(@to_delete, $tag);
+ print "-$tag\n";
+ $printed++;
+ } else {
+ # seen this time: reset the flag
+ $seen{$tag} = 0;
+ }
+ }
+ }
+ grep(delete($seen{$_}), @to_delete);
+ if ($printed > 10) {
+ print "# ", ×tamp, "\n" if $verbose;
+ $printed = 0;
+ }
+}
+
+sub skip {
+ #
+ # put stuff here to ignore some connections, for instance:
+ #
+
+ # what we get when the socket gets created...
+ return(1) if $name eq '*:0';
+ return(1) if $name =~ /^localhost:(\d+)$/ && $1 > 1000;
+#
+# UDP & TCP stuff
+#
+ #
+ # ignore common daemons
+ #
+ if ($name =~ /^\*:/ && $user eq 'root' && $pid < 300) {
+ return(1) if $command =~ /^inetd(\.afs)?$/;
+ return(1) if $command =~ /^rpc\.(stat|lock)d$/;
+ return(1) if $command eq 'syslogd' && $name eq '*:syslog';
+ }
+ #
+ # forking beasts: portmap, ypbind, inetd
+ #
+ if ($command eq 'portmap' && $user eq 'daemon') {
+ return(1) if $name =~ /^\*:/;
+ } elsif ($command eq 'ypbind') {
+ return(1) if $name =~ /^\*:\d+$/;
+ }
+#
+# TCP-only stuff
+#
+ return(0) unless $proto eq 'TCP';
+ #
+ # outgoing commands: ftp, telnet, r*
+ #
+ if ($command eq 'ftp') {
+ return(1) if $name =~ /:ftp(-data)?$/;
+ } elsif ($command eq 'telnet') {
+ return(1) if $name =~ /:telnet$/;
+ } elsif ($command eq 'remsh') {
+ if ($name =~ /:(\d?\d\d\d)->.+:(\d?\d\d\d)$/) {
+ return(1) if $1 < 1024 && $1 > 990 && $2 < 1024 && $2 > 990;
+ } elsif ($name =~ /:(\d?\d\d\d)->.+:(shell|ta-rauth)$/) {
+ return(1) if $1 < 1024 && $1 > 990;
+ } elsif ($name =~ /^\*:(\d?\d\d\d)$/) {
+ return(1) if $1 < 1024 && $1 > 990;
+ }
+ }
+ return(0);
+}
+
+sub timestamp {
+ my($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst);
+
+ ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime(time);
+ sprintf("%d/%02d/%02d-%02d:%02d:%02d", $year + 1900, $mon+1, $mday,
+ $hour, $min, $sec);
+}
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/local/bin/perl
+#
+# count_pf.perl-- run lsof in repeat mode and count processes and
+# files
+
+sub interrupt { print "\n"; exit 0; }
+
+$RPT = 15; # lsof repeat time
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+# Read lsof -nPF output repeatedly from a pipe.
+
+$| = 1; # unbuffer output
+$SIG{'INT'} = 'interrupt'; # catch interrupt
+$proc = $files = $proto{'TCP'} = $proto{'UDP'} = 0;
+$progress="/"; # used to show "progress"
+open(P, "$LSOF -nPF -r $RPT|") || die "can't open pipe to $LSOF\n";
+
+while (<P>) {
+ chop;
+ if (/^m/) {
+
+ # A marker line signals the end of an lsof repetition.
+
+ printf "%s Processes: %5d, Files: %6d, TCP: %6d, UDP: %6d\r",
+ $progress, $proc, $files, $proto{'TCP'}, $proto{'UDP'};
+ $proc = $files = $proto{'TCP'} = $proto{'UDP'} = 0;
+ if ($progress eq "/") { $progress = "\\"; } else { $progress = "/"; }
+ next;
+ }
+ if (/^p/) { $proc++; next; } # Count processes.
+ if (/^f/) { $files++; next; } # Count files.
+ if (/^P(.*)/) { $proto{$1}++; next; } # Count protocols.
+}
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/local/bin/perl
+#
+# count_pf.perl5 -- run lsof in repeat mode and count processes and
+# files
+
+sub interrupt { print "\n"; exit 0; }
+
+$RPT = 15; # lsof repeat time
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+# Read lsof -nPF0 output repeatedly from a pipe.
+
+$| = 1; # unbuffer output
+$SIG{'INT'} = 'interrupt'; # catch interrupt
+$proc = $files = $tcp = $udp = 0;
+$progress="/";
+open(P, "$LSOF -nPF0 -r $RPT|") || die "can't open pipe to $LSOF\n";
+
+LSOF_LINE:
+
+while (<P>) {
+ chop;
+ if (/^m/) {
+
+ # A marker line signals the end of an lsof repetition.
+
+ printf "%s Processes: %5d, Files: %6d, TCP: %6d, UDP: %6d\r",
+ $progress, $proc, $files, $tcp, $udp;
+ $proc = $files = $tcp = $udp = 0;
+ if ($progress eq "/") { $progress = "\\"; } else { $progress = "/"; }
+ next LSOF_LINE;
+ }
+ if (/^p/) {
+
+ # Count process.
+
+ $proc++;
+ next LSOF_LINE;
+ }
+ if (/^f/) {
+
+ # Count files.
+
+ $files++;
+ @F = split("\0", $_, 999);
+ foreach $i (0 .. ($#F - 1)) {
+
+ # Search for protocol field.
+
+ if ($F[$i] =~ /^P(.*)/) {
+
+ # Count instances of TCP and UDP protocols.
+
+ if ($1 eq "TCP") { $tcp++; }
+ elsif ($1 eq "UDP") { $udp++; }
+ next LSOF_LINE;
+ }
+ }
+ }
+}
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/local/bin/perl
+###################################################################
+# identd.perl5 : An implementation of RFC 1413 Ident Server
+# using Vic Abell's lsof.
+#
+# - Started from inetd with 'nowait' option. This entry in
+# /etc/inetd.conf will suffice :
+#
+# ident stream tcp nowait root /usr/local/bin/identd.perl5 -t200
+#
+# - Multiple instances of the server are not a performance penalty
+# since they shall use lsof's cacheing mechanism. (compare with
+# Peter Eriksson's pidentd)
+# - assumes 'lsof' binary in /usr/local/sbin
+# - Command line arguments :
+# -t TIMEOUT Number of seconds to wait for a query before aborting.
+# Default is 120.
+#
+# Kapil Chowksey <kchowksey@hss.hns.com>
+###################################################################
+
+use Socket;
+require 'getopts.pl';
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+# redirect lsof's warnings/errors to /dev/null
+close(STDERR);
+open(STDERR, ">/dev/null");
+
+$Timeout = "120";
+
+&Getopts('t:');
+if ($opt_t) {
+ $Timeout = $opt_t;
+}
+
+($port, $iaddr) = sockaddr_in(getpeername(STDIN));
+$peer_addr = inet_ntoa($iaddr);
+
+# read ident-query from socket (STDIN) with a timeout.
+$timeout = int($Timeout);
+eval {
+ local $SIG{ALRM} = sub { die "alarm\n" };
+ alarm $timeout;
+ $query = <STDIN>;
+ alarm 0;
+};
+die if $@ && $@ ne "alarm\n";
+if ($@) {
+ # timed out
+ exit;
+}
+
+# remove all white-spaces from query
+$query =~ s/\s//g;
+
+$serv_port = "";
+$cli_port = "";
+($serv_port,$cli_port) = split(/,/,$query);
+
+if ($serv_port =~ /^[0-9]+$/) {
+ if (int($serv_port) < 1 || int($serv_port) > 65535) {
+ print $query." : ERROR : INVALID-PORT"."\n";
+ exit;
+ }
+} else {
+ print $query." : ERROR : INVALID-PORT"."\n";
+ exit;
+}
+
+if ($cli_port =~ /^[0-9]+$/) {
+ if (int($cli_port) < 1 || int($cli_port) > 65535) {
+ print $query." : ERROR : INVALID-PORT"."\n";
+ exit;
+ }
+} else {
+ print $query." : ERROR : INVALID-PORT"."\n";
+ exit;
+}
+
+open(LSOFP,"$LSOF -nPDi -T -FLn -iTCP@".$peer_addr.":".$cli_port."|");
+
+$user = "UNKNOWN";
+while ($a_line = <LSOFP>) {
+ # extract user name.
+ if ($a_line =~ /^L.*/) {
+ ($user) = ($a_line =~ /^L(.*)/);
+ }
+
+ # make sure local port matches.
+ if ($a_line =~ /^n.*:\Q$serv_port->/) {
+ print $serv_port.", ".$cli_port." : USERID : UNIX :".$user."\n";
+ exit;
+ }
+}
+
+print $serv_port.", ".$cli_port." : ERROR : NO-USER"."\n";
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/local/bin/perl
+#
+# $Id: idrlogin.perl,v 1.5 2001/11/18 12:20:46 abe Exp $
+#
+# idrlogin.perl -- sample Perl script to identify the network source of a
+# network (remote) login via rlogind, sshd, or telnetd
+
+
+# IMPORTANT DEFINITIONS
+# =====================
+#
+# 1. Set the interpreter line of this script to the local path of the
+# Perl executable.
+
+
+#
+# Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+# 47907. All rights reserved.
+#
+# Written by Victor A. Abell
+#
+# This software is not subject to any license of the American Telephone
+# and Telegraph Company or the Regents of the University of California.
+#
+# Permission is granted to anyone to use this software for any purpose on
+# any computer system, and to alter it and redistribute it freely, subject
+# to the following restrictions:
+#
+# 1. Neither the authors nor Purdue University are responsible for any
+# consequences of the use of this software.
+#
+# 2. The origin of this software must not be misrepresented, either by
+# explicit claim or by omission. Credit to the authors and Purdue
+# University must appear in documentation and sources.
+#
+# 3. Altered versions must be plainly marked as such, and must not be
+# misrepresented as being the original software.
+#
+# 4. This notice may not be removed or altered.
+
+# Initialize variables.
+
+$dev = $name = $proto = ""; # fd variables
+$fdst = 0; # fd state
+$pidst = 0; # process state
+$cmd = $login = $pid = $ppid = ""; # process var.
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+# Open a pipe from lsof.
+
+open(P, "$LSOF -R -FcDfLpPRn|") || die "Can't pipe from $LSOF\n";
+
+# Process the ``lsof -FcDfLpPRn'' output a line at a time
+
+while (<P>) {
+ chop;
+ if (/^p(.*)/) {
+
+# A process set begins with a PID field whose ID character is `p'.
+
+ $tpid = $1;
+ if ($pidst && $fdst) { &save_proc }
+ $pidst = 1;
+ $pid = $tpid;
+ $cmd = $login = $ppid = "";
+ $fdst = 0;
+ $dev = $name = $proto = "";
+ next;
+ }
+
+# Save process-related values.
+
+ if (/^c(.*)/) { $cmd = $1; next; }
+ if (/^L(.*)/) { $login = $1; next; }
+ if (/^R(.*)/) { $ppid = $1; next; }
+
+# A file set begins with a file descriptor field.
+
+ if (/^f/) {
+ if ($pidst && $fdst) { &save_proc }
+ $fdst = 0;
+ $dev = $name = $proto = "";
+ next;
+ }
+
+# Accumulate file information.
+
+ if (/^D(.*)/) { $dev = $1; next; }
+ if (/^P(.*)/) { $proto = $1; next; }
+ if (/^n(.*)/) { $name = $1; $fdst = 1; next; }
+}
+
+# Flush any stored file or process output.
+
+if ($pidst && $fdst) { &save_proc }
+
+# List the shell processes that have rlogind/sshd//telnetd parents.
+
+$hdr = 0;
+foreach $pid (sort keys(%shcmd)) {
+ $p = $pid;
+ if (!defined($raddr{$pid})) {
+ for ($ff = 0; !$ff && defined($Ppid{$p}); ) {
+ $p = $Ppid{$p};
+ if ($p < 2 || defined($raddr{$p})) { $ff = 1; }
+ }
+ } else { $ff = 2; }
+ if ($ff && defined($raddr{$p})) {
+ if (!$hdr) {
+ printf "%-8.8s %-8.8s %6s %-10.10s %6s %-10.10s %s\n",
+ "Login", "Shell", "PID", "Via", "PID", "TTY", "From";
+ $hdr = 1;
+ }
+ printf "%-8.8s %-8.8s %6d %-10.10s %6s %-10.10s %s\n",
+ $shlogin{$pid}, $shcmd{$pid}, $pid,
+ ($ff == 2) ? "(direct)" : $rcmd{$p},
+ ($ff == 2) ? "" : $p,
+ ($shtty{$pid} eq "") ? "(unknown)" : $shtty{$pid},
+ $raddr{$p};
+ }
+}
+exit(0);
+
+
+# save_proc -- save process information
+# Values are stored inelegantly in global variables.
+
+sub save_proc {
+ if ($cmd eq ""
+ || $login eq ""
+ || $ppid eq ""
+ || $pid eq ""
+ || $name eq ""
+ ) { return; }
+ if (!defined($Ppid{$pid})) { $Ppid{$pid} = $ppid; }
+ if ($proto eq "TCP"
+ && (($cmd =~ /rlogind/) || ($cmd =~ /sshd/) || ($cmd =~ /telnetd/))) {
+ if (defined($raddr{$pid})) { return; }
+ if (($name =~ /[^:]*:[^-]*->([^:]*):.*/)) {
+ $raddr{$pid} = $1;
+ $rcmd{$pid} = $cmd;
+ return;
+ }
+ }
+ if (($cmd =~ /.*sh$/)) {
+ if (defined($shcmd{$pid})) { return; }
+ if ($proto eq "TCP") {
+ if (defined($raddr{$pid})) { return; }
+ if (($name =~ /[^:]*:[^-]*->([^:]*):.*/)) {
+ $raddr{$pid} = $1;
+ $shcmd{$pid} = $cmd;
+ $shlogin{$pid} = $login;
+ }
+ }
+ if (($name =~ m#/dev.*ty.*#)) {
+ ($tty) = ($name =~ m#/dev.*/(.*)#);
+ } elsif (($name =~ m#/dev/(pts/\d+)#)) {
+ $tty = $1;
+ } elsif (($name =~ m#/dev.*pts.*#)) {
+ $d = oct($dev);
+ $tty = sprintf("pts/%d", $d & 0xffff);
+ } else { return; }
+ } else { return; }
+ $shcmd{$pid} = $cmd;
+ $shtty{$pid} = $tty;
+ $shlogin{$pid} = $login;
+}
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/local/bin/perl
+#
+# $Id: idrlogin.perl5,v 1.5 2001/11/18 12:20:46 abe Exp $
+#
+# idrlogin.perl5 -- sample Perl 5 script to identify the network source of a
+# network (remote) login via rlogind, sshd, or telnetd
+
+
+# IMPORTANT DEFINITIONS
+# =====================
+#
+# 1. Set the interpreter line of this script to the local path of the
+# Perl 5 executable.
+
+
+# Copyright 1997 Purdue Research Foundation, West Lafayette, Indiana
+# 47907. All rights reserved.
+#
+# Written by Victor A. Abell
+#
+# This software is not subject to any license of the American Telephone
+# and Telegraph Company or the Regents of the University of California.
+#
+# Permission is granted to anyone to use this software for any purpose on
+# any computer system, and to alter it and redistribute it freely, subject
+# to the following restrictions:
+#
+# 1. Neither the authors nor Purdue University are responsible for any
+# consequences of the use of this software.
+#
+# 2. The origin of this software must not be misrepresented, either by
+# explicit claim or by omission. Credit to the authors and Purdue
+# University must appear in documentation and sources.
+#
+# 3. Altered versions must be plainly marked as such, and must not be
+# misrepresented as being the original software.
+#
+# 4. This notice may not be removed or altered.
+
+# Initialize variables.
+
+$dev = $faddr = $tty = ""; # fd variables
+$pidst = 0; # process state
+$cmd = $login = $pgrp = $pid = $ppid = ""; # process var.
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+# Open a pipe from lsof
+
+if (! -x "$LSOF") { die "Can't execute $LSOF\n"; }
+open (P, "$LSOF -R -FcDfLpPRn0|") || die "Can't pipe from $LSOF\n";
+
+# Process the lsof output a line at a time
+
+while (<P>) {
+ chop;
+ @F = split('\0', $_, 999);
+ if ($F[0] =~ /^p/) {
+
+# A process set begins with a PID field whose ID character is `p'.
+
+ if ($pidst) { &save_proc }
+ foreach $i (0 .. ($#F - 1)) {
+
+ PROC: {
+ if ($F[$i] =~ /^c(.*)/) { $cmd = $1; last PROC }
+ if ($F[$i] =~ /^p(.*)/) { $pid = $1; last PROC }
+ if ($F[$i] =~ /^R(.*)/) { $ppid = $1; last PROC }
+ if ($F[$i] =~ /^L(.*)/) { $login = $1; last PROC }
+ }
+ }
+ $pidst = 1;
+ next;
+ }
+
+# A file descriptor set begins with a file descriptor field whose ID
+# character is `f'.
+
+ if ($F[0] =~ /^f/) {
+ if ($faddr ne "") { next; }
+ $proto = $name = "";
+ foreach $i (0 .. ($#F - 1)) {
+
+ FD: {
+ if ($F[$i] =~ /^P(.*)/) { $proto = $1; last FD; }
+ if ($F[$i] =~ /^n(.*)/) { $name = $1; last FD; }
+ if ($F[$i] =~ /^D(.*)/) { $dev = $1; last FD; }
+ }
+ }
+ if ($proto eq "TCP"
+ && $faddr eq ""
+ && (($cmd =~ /rlogind/) || ($cmd =~ /sshd/) || ($cmd =~ /telnetd/))) {
+ if (($name =~ /[^:]*:[^-]*->([^:]*):.*/)) {
+ $faddr = $1;
+ }
+ } elsif ($tty eq "" && ($cmd =~ /.*sh$/)) {
+ if (($name =~ m#/dev.*ty.*#)) {
+ ($tty) = ($name =~ m#/dev.*/(.*)#);
+ } elsif (($name =~ m#/dev/(pts/\d+)#)) {
+ $tty = $1;
+ } elsif (($name =~ m#/dev.*pts.*#)) {
+ $d = oct($dev);
+ $tty = sprintf("pts/%d", $d & 0xffff);
+ }
+ }
+ next;
+ }
+}
+
+# Flush any stored file or process output.
+
+if ($pidst) { &save_proc }
+
+# List the shell processes that have rlogind/sshd/telnetd parents.
+
+$hdr = 0;
+foreach $pid (sort keys(%shcmd)) {
+ $p = $pid;
+ if (!defined($raddr{$pid})) {
+ for ($ff = 0; !$ff && defined($Ppid{$p}); ) {
+ $p = $Ppid{$p};
+ if ($p < 2 || defined($raddr{$p})) { $ff = 1; }
+ }
+ } else { $ff = 2; }
+ if ($ff && defined($raddr{$p})) {
+ if (!$hdr) {
+ printf "%-8.8s %-8.8s %6s %-10.10s %6s %-10.10s %s\n",
+ "Login", "Shell", "PID", "Via", "PID", "TTY", "From";
+ $hdr = 1;
+ }
+ printf "%-8.8s %-8.8s %6d %-10.10s %6s %-10.10s %s\n",
+ $shlogin{$pid}, $shcmd{$pid}, $pid,
+ ($ff == 2) ? "(direct)" : $rcmd{$p},
+ ($ff == 2) ? "" : $p,
+ ($shtty{$pid} eq "") ? "(unknown)" : $shtty{$pid},
+ $raddr{$p};
+ }
+}
+exit(0);
+
+
+# save_proc -- save process information
+# Values are stored inelegantly in global variables.
+
+sub save_proc {
+ if (!defined($Ppid{$pid})) { $Ppid{$pid} = $ppid; }
+ if ($faddr ne "") {
+ $raddr{$pid} = $faddr;
+ if (($cmd =~ /.*sh$/)) {
+ $shcmd{$pid} = $cmd;
+ $shlogin{$pid} = $login;
+ } else { $rcmd{$pid} = $cmd; }
+ }
+ if ($tty ne "") {
+ $shcmd{$pid} = $cmd;
+ $shtty{$pid} = $tty;
+ $shlogin{$pid} = $login;
+ }
+
+# Clear variables.
+
+ $cmd = $dev = $faddr = $pgrp = $pid = $ppid = $tty = "";
+ $pidst = 0;
+}
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/local/bin/perl
+#
+# $Id: list_NULf.perl5,v 1.5 2000/07/14 17:03:37 abe Exp $
+#
+# list_NULf.perl5 -- sample Perl 5 script to list lsof NUL-terminated
+# full field output (i.e., -F0 output)
+#
+# This script has been tested under perl version 5.001e.
+#
+# Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+# 47907. All rights reserved.
+#
+# Written by Victor A. Abell
+#
+# This software is not subject to any license of the American Telephone
+# and Telegraph Company or the Regents of the University of California.
+#
+# Permission is granted to anyone to use this software for any purpose on
+# any computer system, and to alter it and redistribute it freely, subject
+# to the following restrictions:
+#
+# 1. Neither the authors nor Purdue University are responsible for any
+# consequences of the use of this software.
+#
+# 2. The origin of this software must not be misrepresented, either by
+# explicit claim or by omission. Credit to the authors and Purdue
+# University must appear in documentation and sources.
+#
+# 3. Altered versions must be plainly marked as such, and must not be
+# misrepresented as being the original software.
+#
+# 4. This notice may not be removed or altered.
+
+# Initialize variables.
+
+$fhdr = 0; # fd hdr. flag
+$fdst = 0; # fd state
+$access = $devch = $devn = $fd = $inode = $lock = $name = ""; # | file descr.
+$offset = $proto = $size = $state = $stream = $type = ""; # | variables
+$pidst = 0; # process state
+$cmd = $login = $pgrp = $pid = $ppid = $uid = ""; # process var.
+
+# Process the ``lsof -F'' output a line at a time, gathering
+# the variables for a process together before printing them;
+# then gathering the variables for each file descriptor
+# together before printing them.
+
+while (<>) {
+ chop;
+ @F = split('\0', $_, 999);
+ if ($F[0] =~ /^p/) {
+
+# A process set begins with a PID field whose ID character is `p'.
+
+ if ($pidst) { &list_proc }
+ if ($fdst) { &list_fd; $fdst = 0; }
+ foreach $i (0 .. ($#F - 1)) {
+
+ PROC: {
+ if ($F[$i] =~ /^c(.*)/) { $cmd = $1; last PROC }
+ if ($F[$i] =~ /^g(.*)/) { $pgrp = $1; last PROC }
+ if ($F[$i] =~ /^p(.*)/) { $pid = $1; last PROC }
+ if ($F[$i] =~ /^u(.*)/) { $uid = $1; last PROC }
+ if ($F[$i] =~ /^L(.*)/) { $login = $1; last PROC }
+ if ($F[$i] =~ /^R(.*)/) { $ppid = $1; last PROC }
+ print "ERROR: unrecognized process field: \"$F[$i]\"\n";
+ }
+ }
+ $pidst = 1;
+ next;
+ }
+
+# A file descriptor set begins with a file descriptor field whose ID
+# character is `f'.
+
+ if ($F[0] =~ /^f/) {
+ if ($pidst) { &list_proc }
+ if ($fdst) { &list_fd }
+ foreach $i (0 .. ($#F - 1)) {
+
+ FD: {
+ if ($F[$i] =~ /^a(.*)/) { $access = $1; last FD; }
+ if ($F[$i] =~ /^C(.*)/) { last FD; }
+ if ($F[$i] =~ /^f(.*)/) { $fd = $1; last FD; }
+ if ($F[$i] =~ /^F(.*)/) { last FD; }
+ if ($F[$i] =~ /^d(.*)/) { $devch = $1; last FD; }
+ if ($F[$i] =~ /^D(.*)/) { $devn = $1; last FD; }
+ if ($F[$i] =~ /^G(.*)/) { last FD; }
+ if ($F[$i] =~ /^i(.*)/) { $inode = $1; last FD; }
+ if ($F[$i] =~ /^k(.*)/) { last FD; }
+ if ($F[$i] =~ /^l(.*)/) { $lock = $1; last FD; }
+ if ($F[$i] =~ /^N(.*)/) { last FD; }
+ if ($F[$i] =~ /^o(.*)/) { $offset = $1; last FD; }
+ if ($F[$i] =~ /^P(.*)/) { $proto = $1; last FD; }
+ if ($F[$i] =~ /^s(.*)/) { $size = $1; last FD; }
+ if ($F[$i] =~ /^S(.*)/) { $stream = $1; last FD; }
+ if ($F[$i] =~ /^t(.*)/) { $type = $1; last FD; }
+ if ($F[$i] =~ /^T(.*)/) {
+ if ($state eq "") { $state = "(" . $1; }
+ else { $state = $state . " " . $1; }
+ last FD;
+ }
+ if ($F[$i] =~ /^n(.*)/) { $name = $1; last FD; }
+ print "ERROR: unrecognized file set field: \"$F[$i]\"\n";
+ }
+ }
+ $fdst = 1;
+ next;
+ }
+ print "ERROR: unrecognized: \"$_\"\n";
+}
+
+# Flush any stored file or process output.
+
+if ($fdst) { &list_fd }
+if ($pidst) { &list_proc }
+exit(0);
+
+
+## list_fd -- list file descriptor information
+# Values are stored inelegantly in global variables.
+
+sub list_fd {
+ if ( ! $fhdr) {
+
+ # Print header once.
+
+ print " FD TYPE DEVICE SIZE/OFF INODE NAME\n";
+ $fhdr = 1;
+ }
+ printf " %4s%1.1s%1.1s %4.4s", $fd, $access, $lock, $type;
+ $tmp = $devn; if ($devch ne "") { $tmp = $devch }
+ printf " %10.10s", $tmp;
+ $tmp = $size; if ($offset ne "") { $tmp = $offset }
+ printf " %10.10s", $tmp;
+ $tmp = $inode; if ($proto ne "") { $tmp = $proto }
+ printf " %10.10s", $tmp;
+ $tmp = $stream; if ($name ne "") { $tmp = $name }
+ print " ", $tmp;
+ if ($state ne "") { printf " %s)\n", $state; } else { print "\n"; }
+
+# Clear variables.
+
+ $access = $devch = $devn = $fd = $inode = $lock = "";
+ $name = $offset = $proto = $size = $state = $stream = $type = "";
+}
+
+
+# list_proc -- list process information
+# Values are stored inelegantly in global variables.
+
+sub list_proc {
+ print "COMMAND PID PGRP PPID USER\n";
+ $tmp = $uid; if ($login ne "") {$tmp = $login }
+ printf "%-9.9s %6d %6d %6d %s\n", $cmd, $pid, $pgrp, $ppid, $tmp;
+
+# Clear variables.
+
+ $cmd = $login = $pgrp = $pid = $uid = "";
+ $fhdr = $pidst = 0;
+}
--- /dev/null
+# $Id: list_fields.awk,v 1.3 97/09/23 09:32:38 abe Exp $
+#
+# list_fields.awk -- sample awk script to list lsof full field output
+# (i.e., -F output without -0)
+#
+# NB: this is not particularly elegant awk; several sections were
+# replicated, perhaps unnecessarily, to produce a sample quickly
+# and simply.
+#
+#
+# Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+# 47907. All rights reserved.
+#
+# Written by Victor A. Abell
+#
+# This software is not subject to any license of the American Telephone
+# and Telegraph Company or the Regents of the University of California.
+#
+# Permission is granted to anyone to use this software for any purpose on
+# any computer system, and to alter it and redistribute it freely, subject
+# to the following restrictions:
+#
+# 1. Neither the authors nor Purdue University are responsible for any
+# consequences of the use of this software.
+#
+# 2. The origin of this software must not be misrepresented, either by
+# explicit claim or by omission. Credit to the authors and Purdue
+# University must appear in documentation and sources.
+#
+# 3. Altered versions must be plainly marked as such, and must not be
+# misrepresented as being the original software.
+#
+# 4. This notice may not be removed or altered.
+
+# Clear file and process status.
+
+BEGIN {
+ fhdr = fdst = pidst = 0;
+ access = dev = devch = fd = inode = lock = name = offset = "";
+ proto = size = state = stream = type = "";
+ cmd = login = pgrp = pid = ppid = uid = "";
+}
+
+# Start a new process.
+
+/^p/ {
+ val = substr($0, 2);
+ if (pidst) {
+
+ # Print a previously accumulated process set.
+
+ printf "COMMAND PID PGRP PPID USER\n";
+ printf "%-9.9s %6d %6d %6d", cmd, pid, pgrp, ppid;
+ if (login != "") { printf " %s\n", login }
+ else { printf " %s\n", uid }
+ pidst = 0;
+ cmd = login = pgrp = pid = uid = "";
+ }
+ if (fdst) {
+
+ # Print a previously accumulated file set.
+
+ if (fhdr == 0) {
+ printf " FD TYPE DEVICE SIZE/OFF INODE NAME\n";
+ }
+ printf " %4.4s%1.1s%1.1s %4.4s", fd, access, lock, type;
+ t = dev; if (devch != "") { t = devch }
+ printf(" %10.10s", t);
+ t = size; if (offset != "") { t = offset }
+ printf " %10.10s", t;
+ t = inode; if (proto != "") { t = proto }
+ printf " %10.10s", t;
+ t = stream; if (name != "") {t = name }
+ printf " %s", t;
+ if (state != "") { printf " %s)\n", state } else { printf "\n" }
+ access = dev = devch = fd = inode = lock = name = offset = "";
+ proto = size = state = stream = type = "";
+ fdst = fhdr = 0
+ }
+
+# Record a new process.
+
+ pidst = 1;
+ pid = val;
+}
+
+/^g|^c|^u|^L|^R/ {
+
+# Save process set information.
+
+ id = substr($0, 1, 1);
+ val = substr($0, 2);
+ if (id == "g") { pgrp = val; next } # PGRP
+ if (id == "c") { cmd = val; next } # command
+ if (id == "u") { uid = val; next } # UID
+ if (id == "L") { login = val; next } # login name
+ if (id == "R") { ppid = val; next } # PPID
+}
+
+/^f|^a|^l|^t|^d|^D|^s|^o|^i|^P|^S|^T|^n/ {
+
+# Save file set information.
+
+ id = substr($0, 1, 1);
+ val = substr($0, 2);
+ if (id == "f") {
+ if (pidst) {
+
+ # Print a previously accumulated process set.
+
+ printf "COMMAND PID PGRP PPID USER\n";
+ printf "%-9.9s %6d %6d %6d", cmd, pid, pgrp, ppid;
+ if (login != "") { printf " %s\n", login }
+ else { printf " %s\n", uid }
+ pidst = 0;
+ cmd = login = pgrp = pid = uid = "";
+ }
+ if (fdst) {
+
+ # Print a previously accumulated file set.
+
+ if (fhdr == 0) {
+ printf " FD TYPE DEVICE SIZE/OFF INODE NAME\n";
+ }
+ fhdr = 1;
+ printf " %4.4s%1.1s%1.1s %4.4s", fd, access, lock, type;
+ t = dev; if (devch != "") { t = devch }
+ printf(" %10.10s", t);
+ t = size; if (offset != "") { t = offset }
+ printf " %10.10s", t;
+ t = inode; if (proto != "") { t = proto }
+ printf " %10.10s", t;
+ t = stream; if (name != "") {t = name }
+ printf " %s", t;
+ if (state != "") { printf " %s)\n", state } else { printf "\n" }
+ access = dev = devch = fd = inode = lock = name = offset = "";
+ proto = size = state = stream = type = "";
+ }
+
+ # Start an new file set.
+
+ fd = val;
+ fdst = 1;
+ next;
+ }
+
+# Save file set information.
+
+ if (id == "a") { access = val; next } # access
+ if (id == "l") { lock = val; next } # lock
+ if (id == "t") { type = val; next } # type
+ if (id == "d") { devch = val; next } # device characters
+ if (id == "D") { dev = val; next } # device major/minor numbers
+ if (id == "s") { size = val; next } # size
+ if (id == "o") { offset = val; next } # offset
+ if (id == "i") { inode = val; next } # inode number
+ if (id == "P") { proto = val; next } # protocol
+ if (id == "S") { stream = val; next } # stream name
+ if (id == "T") { # TCP/TPI state
+ if (state == "") {
+ state = sprintf("(%s", val);
+ } else {
+ state = sprintf("%s %s", state, val);
+ }
+ next
+ }
+ if (id == "n") { name = val; next } # name, comment, etc.
+}
+
+END {
+ if (pidst) {
+
+ # Print last process set.
+
+ printf "COMMAND PID PGRP PPID USER\n";
+ printf "%-9.9s %6d %6d %6d", cmd, pid, pgrp, ppid;
+ if (login != "") { printf " %s\n", login }
+ else { printf " %s\n", uid }
+ }
+ if (fdst) {
+
+ # Print last file set.
+
+ if (fhdr == 0) {
+ printf " FD TYPE DEVICE SIZE/OFF INODE NAME\n";
+ }
+ printf " %4.4s%1.1s%1.1s %4.4s", fd, access, lock, type;
+ t = dev; if (devch != "") { t = devch }
+ printf(" %10.10s", t);
+ t = size; if (offset != "") { t = offset }
+ printf " %10.10s", t;
+ t = inode; if (proto != "") { t = proto }
+ printf " %10.10s", t;
+ t = stream; if (name != "") {t = name }
+ printf " %s", t;
+ if (state != "") { printf " %s)\n", state; } else { printf "\n"; }
+ }
+}
--- /dev/null
+#!/usr/local/bin/perl4
+#
+# $Id: list_fields.perl,v 1.5 2000/07/14 17:03:37 abe Exp $
+#
+# list_fields.perl -- sample Perl script to list lsof full field output
+# (i.e., -F output without -0)
+#
+# This script has been tested under perl versions 4.036 and 5.001e.
+#
+# Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+# 47907. All rights reserved.
+#
+# Written by Victor A. Abell
+#
+# This software is not subject to any license of the American Telephone
+# and Telegraph Company or the Regents of the University of California.
+#
+# Permission is granted to anyone to use this software for any purpose on
+# any computer system, and to alter it and redistribute it freely, subject
+# to the following restrictions:
+#
+# 1. Neither the authors nor Purdue University are responsible for any
+# consequences of the use of this software.
+#
+# 2. The origin of this software must not be misrepresented, either by
+# explicit claim or by omission. Credit to the authors and Purdue
+# University must appear in documentation and sources.
+#
+# 3. Altered versions must be plainly marked as such, and must not be
+# misrepresented as being the original software.
+#
+# 4. This notice may not be removed or altered.
+
+# Initialize variables.
+
+$fhdr = 0; # fd hdr. flag
+$fdst = 0; # fd state
+$access = $devch = $devn = $fd = $inode = $lock = $name = ""; # | file descr.
+$offset = $proto = $size = $state = $stream = $type = ""; # | variables
+$pidst = 0; # process state
+$cmd = $login = $pgrp = $pid = $ppid = $uid = ""; # process var.
+
+# Process the ``lsof -F'' output a line at a time, gathering
+# the variables for a process together before printing them;
+# then gathering the variables for each file descriptor
+# together before printing them.
+
+while (<>) {
+ chop;
+ if (/^p(.*)/) {
+
+# A process set begins with a PID field whose ID character is `p'.
+
+ $tpid = $1;
+ if ($pidst) { &list_proc }
+ $pidst = 1;
+ $pid = $tpid;
+ if ($fdst) { &list_fd; $fdst = 0; }
+ next;
+ }
+
+# Save process-related values.
+
+ if (/^g(.*)/) { $pgrp = $1; next; }
+ if (/^c(.*)/) { $cmd = $1; next; }
+ if (/^u(.*)/) { $uid = $1; next; }
+ if (/^L(.*)/) { $login = $1; next; }
+ if (/^R(.*)/) { $ppid = $1; next; }
+
+# A file descriptor set begins with a file descriptor field whose ID
+# character is `f'.
+
+ if (/^f(.*)/) {
+ $tfd = $1;
+ if ($pidst) { &list_proc }
+ if ($fdst) { &list_fd }
+ $fd = $tfd;
+ $fdst = 1;
+ next;
+ }
+
+# Save file set information.
+
+ if (/^a(.*)/) { $access = $1; next; }
+ if (/^C(.*)/) { next; }
+ if (/^d(.*)/) { $devch = $1; next; }
+ if (/^D(.*)/) { $devn = $1; next; }
+ if (/^F(.*)/) { next; }
+ if (/^G(.*)/) { next; }
+ if (/^i(.*)/) { $inode = $1; next; }
+ if (/^k(.*)/) { next; }
+ if (/^l(.*)/) { $lock = $1; next; }
+ if (/^N(.*)/) { next; }
+ if (/^o(.*)/) { $offset = $1; next; }
+ if (/^P(.*)/) { $proto = $1; next; }
+ if (/^s(.*)/) { $size = $1; next; }
+ if (/^S(.*)/) { $stream = $1; next; }
+ if (/^t(.*)/) { $type = $1; next; }
+ if (/^T(.*)/) {
+ if ($state eq "") { $state = "(" . $1; }
+ else { $state = $state . " " . $1; }
+ next;
+ }
+ if (/^n(.*)/) { $name = $1; next; }
+ print "ERROR: unrecognized: \"$_\"\n";
+}
+
+# Flush any stored file or process output.
+
+if ($fdst) { &list_fd }
+if ($pidst) { &list_proc }
+exit(0);
+
+
+## list_fd -- list file descriptor information
+# Values are stored inelegantly in global variables.
+
+sub list_fd {
+ if ( ! $fhdr) {
+
+ # Print header once.
+
+ print " FD TYPE DEVICE SIZE/OFF INODE NAME\n";
+ $fhdr = 1;
+ }
+ printf " %4s%1.1s%1.1s %4.4s", $fd, $access, $lock, $type;
+ $tmp = $devn; if ($devch ne "") { $tmp = $devch }
+ printf " %10.10s", $tmp;
+ $tmp = $size; if ($offset ne "") { $tmp = $offset }
+ printf " %10.10s", $tmp;
+ $tmp = $inode; if ($proto ne "") { $tmp = $proto }
+ printf " %10.10s", $tmp;
+ $tmp = $stream; if ($name ne "") { $tmp = $name }
+ print " ", $tmp;
+ if ($state ne "") { printf " %s)\n", $state; } else { print "\n"; }
+
+# Clear variables.
+
+ $access = $devch = $devn = $fd = $inode = $lock = $name = "";
+ $offset = $proto = $size = $state = $stream = $type = "";
+}
+
+
+# list_proc -- list process information
+# Values are stored inelegantly in global variables.
+
+sub list_proc {
+ print "COMMAND PID PGRP PPID USER\n";
+ $tmp = $uid; if ($login ne "") {$tmp = $login }
+ printf "%-9.9s %6d %6d %6d %s\n", $cmd, $pid, $pgrp, $ppid, $tmp;
+
+# Clear variables.
+
+ $cmd = $login = $pgrp = $pid = $uid = "";
+ $fhdr = $pidst = 0;
+}
--- /dev/null
+#!/usr/local/bin/perl
+#
+# $Id: shared.perl5,v 1.4 2001/11/18 12:20:46 abe Exp $
+#
+# shared.perl5 -- sample Perl 5 script to list processes that share
+# file descriptors or files, using `lsof +ffn -F..."
+# output
+#
+# Usage: shared [fd|file]
+#
+# where: fd to list file descriptors (default)
+#
+# file to list files
+#
+# This script has been tested under perl version 5.001e.
+
+
+# IMPORTANT DEFINITIONS
+# =====================
+#
+# 1. Set the interpreter line of this script to the local path of the
+# Perl5 executable.
+
+
+# Copyright 1998 Purdue Research Foundation, West Lafayette, Indiana
+# 47907. All rights reserved.
+#
+# Written by Victor A. Abell <abe@purdue.edu>
+#
+# This software is not subject to any license of the American Telephone
+# and Telegraph Company or the Regents of the University of California.
+#
+# Permission is granted to anyone to use this software for any purpose on
+# any computer system, and to alter it and redistribute it freely, subject
+# to the following restrictions:
+#
+# 1. Neither the authors nor Purdue University are responsible for any
+# consequences of the use of this software.
+#
+# 2. The origin of this software must not be misrepresented, either by
+# explicit claim or by omission. Credit to the authors and Purdue
+# University must appear in documentation and sources.
+#
+# 3. Altered versions must be plainly marked as such, and must not be
+# misrepresented as being the original software.
+#
+# 4. This notice may not be removed or altered.
+
+# Initialize variables.
+
+$Access = $Devch = $Devn = $Fd = $Fsa = $Inode = $Lock = # file
+ $Na = $Name = ""; # | descriptor
+$Cmd = $Login = $Pgrp = $Pid = $Ppid = $Uid = ""; # process var.
+$Fdst = 0; # fd state
+$Hdr = 0; # header state
+$Offset = $Proto = $Size = $State = $Stream = $Type = ""; # | variables
+$Pidst = 0; # process state
+$Pn = "shared";
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+# Define print field constants.
+
+$CmdTtl = "CMD";
+$CmdW = length($CmdTtl);
+$DevTtl = "DEVICE";
+$DevW = length($DevTtl);
+$FdTtl = "FD";
+$FdW = length($FdTtl);
+$InoTtl = "NODE";
+$InoW = length($InoTtl);
+$KeyTtl = "FILEADDR";
+$KeyW = length($KeyTtl);
+$PidTtl = "PID";
+$PidW = length($PidTtl);
+$PpidTtl = "PPID";
+$PpidW = length(PpidTtl);
+
+# Process one (optional) argument.
+
+if ($#ARGV >= 0) {
+ $err = 0;
+ if ($#ARGV > 1) { $err = 1; }
+ elsif ($ARGV[0] eq "fd") {
+ $KeyTtl = "FILEADDR";
+ $Shfd = 1;
+ $Shfile = 0;
+ } elsif ($ARGV[0] eq "file") {
+ $KeyTtl = "NODEID";
+ $Shfd = 0;
+ $Shfile = 1;
+ } else { $err = 1; }
+ if ($err) { die "$Pn: usage [fd|file]\n"; }
+ shift;
+} else { $Shfd = 1; $Shfile = 0; }
+$KeyW = length($KeyTtl);
+
+# Open a pipe from lsof.
+
+if (!open(LSOF_PIPE, "$LSOF -R +ffn -F0pcRDfFinN |")) {
+ die "$Pn: can't open pipe to: $LSOF\n";
+}
+
+# Process the lsof output a line at a time, gathering the variables for
+# processes and files.
+
+while (<LSOF_PIPE>) {
+ chop;
+ @F = split('\0', $_, 999);
+ if ($F[0] =~ /^p/) {
+
+# A process set begins with a PID field whose ID character is `p'.
+
+ if ($Fdst) { &End_fd }
+ if ($Pidst) { &End_proc }
+ foreach $i (0 .. ($#F - 1)) {
+
+ PROC: {
+ if ($F[$i] =~ /^c(.*)/) { $Cmd = $1; last PROC }
+ if ($F[$i] =~ /^g(.*)/) { $Pgrp = $1; last PROC }
+ if ($F[$i] =~ /^p(.*)/) { $Pid = $1; last PROC }
+ if ($F[$i] =~ /^u(.*)/) { $Uid = $1; last PROC }
+ if ($F[$i] =~ /^L(.*)/) { $Login = $1; last PROC }
+ if ($F[$i] =~ /^R(.*)/) { $Ppid = $1; last PROC }
+ print "ERROR: unrecognized process field: \"$F[$i]\"\n";
+ }
+ }
+ $Pidst = 1;
+ next;
+ }
+
+# A file descriptor set begins with a file descriptor field whose ID
+# character is `f'.
+
+ if ($F[0] =~ /^f/) {
+ if ($Fdst) { &End_fd }
+ foreach $i (0 .. ($#F - 1)) {
+
+ FD: {
+ if ($F[$i] =~ /^a(.*)/) { $Access = $1; last FD; }
+ if ($F[$i] =~ /^f(.*)/) { $Fd = $1; last FD; }
+ if ($F[$i] =~ /^F(.*)/) { $Fsa = $1; last FD; }
+ if ($F[$i] =~ /^l(.*)/) { $Lock = $1; last FD; }
+ if ($F[$i] =~ /^t(.*)/) { $Type = $1; last FD; }
+ if ($F[$i] =~ /^d(.*)/) { $Devch = $1; last FD; }
+ if ($F[$i] =~ /^D(.*)/) { $Devn = $1; last FD; }
+ if ($F[$i] =~ /^s(.*)/) { $Size = $1; last FD; }
+ if ($F[$i] =~ /^o(.*)/) { $Offset = $1; last FD; }
+ if ($F[$i] =~ /^i(.*)/) { $Inode = $1; last FD; }
+ if ($F[$i] =~ /^P(.*)/) { $Proto = $1; last FD; }
+ if ($F[$i] =~ /^S(.*)/) { $Stream = $1; last FD; }
+ if ($F[$i] =~ /^T(.*)/) {
+ if ($State eq "") { $State = "(" . $1; }
+ else { $State = $State . " " . $1; }
+ last FD;
+ }
+ if ($F[$i] =~ /^n(.*)/) { $Name = $1; last FD; }
+ if ($F[$i] =~ /^N(.*)/) { $Na = $1; last FD; }
+ print "ERROR: unrecognized file set field: \"$F[$i]\"\n";
+ }
+ }
+ $Fdst = 1;
+ next;
+ }
+ print "ERROR: unrecognized: \"$_\"\n";
+}
+close(LSOF_PIPE);
+if ($Fdst) { &End_fd }
+if ($Pidst) { &End_proc }
+
+# List matching files or file descriptors.
+
+for ($pass = 0; $pass < 2; $pass++) {
+ foreach $key (sort keys(%Fds)) {
+ @Praw = split(' ', $Fds{$key}, 999);
+ if ($#Praw < 1) { next; }
+ if ($Shfd) { @P = sort Sort_by_FD_and_PID @Praw; }
+ else { @P = sort Sort_by_PID_and_FD @Praw; }
+
+ # Accumulate and print blocks of (key, PID, FD) triplets.
+
+ for ($i = 0; $i < $#P; $i++) {
+ if ($Shfile) {
+ for ($n = 0; $n <= $#P; $n++) {
+ ($pid, $fd) = split(",", $P[$n], 999);
+ $PrtPid[$n] = $pid;
+ $PrtFd[$n] = $fd;
+ }
+ $i = $n;
+ } else {
+ ($pid, $fd) = split(",", $P[$i], 999);
+ $PrtFd[0] = $fd;
+ $PrtPid[0] = $pid;
+ for ($n = 1; $i < $#P; $i++, $n++) {
+ ($nxtpid, $nxtfd) = split(",", $P[$i + 1], 999);
+ if ($fd ne $nxtfd) { last; }
+ $PrtFd[$n] = $nxtfd;
+ $PrtPid[$n] = $nxtpid;
+ }
+ }
+ if ($n > 1) { &Print_block($key, $n, $pass); }
+ }
+ }
+}
+exit(0);
+
+
+## End_fd() -- process end of file descriptor
+
+sub End_fd {
+
+ local ($key);
+
+ if ($Fdst && $Pidst && $Pid ne "") {
+ if ($Cmd ne "") { $Cmds{$Pid} = $Cmd; }
+ if ($Ppid ne "") { $Ppids{$Pid} = $Ppid; }
+ $key = $Shfd ? $Fsa : $Na;
+ if ($key ne "") {
+ if (!defined($Fds{$key})) { $Fds{$key} = "$Pid,$Fd"; }
+ else { $Fds{$key} .= " $Pid,$Fd"; }
+ if ($Name ne "" && !defined($Name{$key})) { $Name{$key} = $Name }
+ if ($Inode ne "" && !defined($Inodes{$key})) {
+ $Inodes{$key} = $Inode;
+ }
+ if ($Devn ne "" && !defined($Devns{$key})) {
+ $Devns{$key} = $Devn;
+ }
+ }
+ }
+
+# Clear variables.
+
+ $Access = $Devch = $Devn = $Fd = $Fsa = $Inode = $Lock = "";
+ $Na = $Name = $Offset = $Proto = $Size = $State = $Stream = $Type = "";
+ $Fdst = 0;
+}
+
+
+## End_proc() -- process end of process
+
+sub End_proc {
+
+# Clear variables.
+
+ $Cmd = $Login = $Pgrp = $Pid = $Ppid = $Uid = "";
+ $Fdst = $Pidst = 0;
+}
+
+
+## Print_block() -- print a block of entries
+#
+# entry:
+#
+# @_[0] = block's key
+# @_[1] = number of entries in the block
+# @_[2] = print pass status (1 == print)
+
+sub Print_block {
+
+ my ($key, $n, $pass) = @_;
+
+ local ($fd, $i, $pid, $t, $tW);
+
+ if ($pass) {
+ if (!$Hdr) {
+ printf "%${KeyW}.${KeyW}s", $KeyTtl;
+ printf " %${PidW}.${PidW}s", $PidTtl;
+ printf " %${PpidW}.${PpidW}s", $PpidTtl;
+ printf " %-${CmdW}.${CmdW}s", $CmdTtl;
+ printf " %${FdW}.${FdW}s", $FdTtl;
+ printf " %${DevW}.${DevW}s", $DevTtl;
+ printf " %${InoW}.${InoW}s", $InoTtl;
+ printf " NAME\n";
+ $Hdr = 1;
+ } else { print "\n"; }
+ }
+
+# Loop through block. During a non-print pass, caclulate maximum field widths.
+
+ for ($i = 0; $i < $n; $i++) {
+ $fd = $PrtFd[$i];
+ $pid = $PrtPid[$i];
+
+ # Process key.
+
+ if (!$pass) {
+ $tW = length(sprintf("%s", $key));
+ if ($tW > $KeyW) { $KeyW = $tW; }
+ } else { printf "%s", $key; }
+
+ # Process PID.
+
+ if (!$pass) {
+ $tW = length(sprintf(" %s", $pid));
+ if ($tW > $PidW) { $PidW = $tW; }
+ } else { printf " %${PidW}.${PidW}s", $pid; }
+
+ # Process parent PID.
+
+ $t = defined($Ppids{$pid}) ? $Ppids{$pid} : "";
+ if (!$pass) {
+ $tW = length(sprintf(" %s", $t));
+ if ($tW > $PpidW) { $PpidW = $tW; }
+ } else { printf " %${PpidW}.${PpidW}s", $t; }
+
+ # Process command name.
+
+ $t = defined($Cmds{$pid}) ? $Cmds{$pid} : "";
+ if (!$pass) {
+ $tW = length(sprintf(" %s", $t));
+ if ($tW > $CmdW) { $CmdW = $tW; }
+ } else { printf " %-${CmdW}.${CmdW}s", $t; }
+
+ # Process file descriptor.
+
+ if (!$pass) {
+ $tW = length(sprintf(" %s", $fd));
+ if ($tW > $FdW) { $FdW = $tW; }
+ } else { printf " %${FdW}.${FdW}s", $fd; }
+
+ # Process device number.
+
+ $t = defined($Devns{$key}) ? $Devns{$key} : "";
+ if (!$pass) {
+ $tW = length(sprintf(" %s", $t));
+ if ($tW > $DevW) { $DevW = $tW; }
+ } else { printf " %${DevW}.${DevW}s", $t; }
+
+ # Process node number.
+
+ $t = defined($Inodes{$key}) ? $Inodes{$key} : $t;
+ if (!$pass) {
+ $tW = length(sprintf (" %s", $t));
+ if ($tW > $InoW) { $InoW = $tW; }
+ } else { printf " %${InoW}.${InoW}s", $t; }
+
+ # Print name and line terminater, if this is a print pass.
+
+ if ($pass) {
+ if (defined($Name{$key})) { print " $Name{$key}\n"; }
+ else { print "\n"; }
+ }
+ }
+}
+
+
+## Sort_by_FD_and_PID() -- sort (PID,FD) doublets by FD first, then PID
+
+sub Sort_by_FD_and_PID {
+
+ local ($pida, $pidb, $fda, $fdj, $rv);
+
+ ($pida, $fda) = split(",", $a);
+ ($pidb, $fdb) = split(",", $b);
+ if ($fda < $fdb) { return(-1); }
+ if ($fda > $fdb) { return(1); }
+ if ($pida < $pidb) { return(-1); }
+ if ($pida > $pidb) { return(1); }
+ return(0);
+}
+
+
+## Sort_by_PID_and_FD() -- sort (PID,FD) doublets by PID first, then FD
+
+sub Sort_by_PID_and_FD {
+
+ local ($pida, $pidb, $fda, $fdj, $rv);
+
+ ($pida, $fda) = split(",", $a);
+ ($pidb, $fdb) = split(",", $b);
+ if ($pida < $pidb) { return(-1); }
+ if ($pida > $pidb) { return(1); }
+ if ($fda < $fdb) { return(-1); }
+ return(0);
+ if ($fda > $fdb) { return(1); }
+}
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/bin/perl
+# sort_res.perl5 - Script to group & sort lsof output by resource
+#
+# Copyright (c) 2004, 2005 - Fabian Frederick <fabian.frederick@gmx.fr>
+#
+# This program/include file is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as published
+# by the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program/include file is distributed in the hope that it will be
+# useful, but WITHOUT ANY WARRANTY; without even the implied warranty
+# of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program (in the main directory of the Linux-NTFS
+# distribution in the file COPYING); if not, write to the Free Software
+# Foundation,Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+#
+# Note :
+# -This script uses lsof released by Victor A. Abell
+# -lsof path recovery comes from standard perl scripts in there.
+#
+# Usage :
+# perl sort_res.perl5 -> display used resources + size
+# or perl sort_res.perl5 <program name>
+#
+# 12/2005 (FabF)
+# -size reset in loop (script was broken in 4.76)
+# -isexec looking in .. (like other scripts)
+# -display for one or all processes
+# -removing unuseful line number arg.
+# -display global size
+
+require 'getopts.pl';
+my @args = @_;
+
+# Set path to lsof.
+if (($LSOF = &isexec("../lsof")) eq "") { # Some distros use lsof
+ # out of $PATH
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ if (($LSOF = &isexec("../lsof")) eq "") { # Then try ..
+ print "can't execute $LSOF\n"; exit 1
+ }
+ }
+}
+
+if ($ARGV[0] ne ""){
+ $cmd="$LSOF -nPl -Fcns -c".$ARGV[0]."|";
+}else{
+ $cmd="$LSOF -nPl -Fcns|";
+}
+
+#Parse lsof output to gather command, resource name, pid and size
+#Some extradata stand to keep script genericity
+$i=0;
+if (open(FILE, $cmd)){
+ while (defined ($line=<FILE>)){
+ $cline=$line;
+ $cline =~ s"^(.)"";
+ $cline =~ s/^\s+|\s+$//g;
+ if($line=~m/^p/){
+ $pid=$cline;
+ }else{
+ if($line=~/^s/){
+ $size = $cline;
+ }else{
+ if($line=~/^c/){
+ $command = $cline;
+ }else{
+ if($line=~/^n/){
+ $name = $cline;
+ $data{$i} = { command => $command, name => $name,
+ pid => $pid , size => $size};
+ $size=0;
+ $i = $i+1;
+ }
+ }
+ }
+ }
+ }
+}
+
+#Resource name sorting
+sub byresname { $data{$a}{name} cmp $data{$b}{name}}
+@ks=sort byresname (keys %data);
+
+#Resource grouping
+$i=0;
+$cname="a";
+foreach $k (@ks){
+ if ($data{$k}{name} ne $cname){
+ $dgroup{$i} = { name => $data{$k}{name}, size => $data{$k}{size}};
+ $cname = $data{$k}{name};
+ $i++;
+ }
+}
+
+#Size sort on resource hash
+sub bysize { $dgroup{$a}{size} <=> $dgroup{$b}{size} }
+@ks=sort bysize (keys %dgroup);
+$gsize=0;
+printf(" -- KB -- -- Resource --\n", );
+foreach $k (@ks){
+ printf("%10d %s\n", $dgroup{$k}{size}/1024, $dgroup{$k}{name});
+ $gsize+=$dgroup{$k}{size};
+}
+
+printf("Total KB : %10d\n", $gsize/1024);
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/local/bin/perl
+#
+# watch_a_file.perl -- use lsof -F output to watch a specific file
+# (or file system)
+#
+# usage: watch_a_file.perl file_name
+
+## Interrupt handler
+
+sub interrupt { wait; print "\n"; exit 0; }
+
+
+## Start main program
+
+$Pn = "watch_a_file";
+# Check file argument.
+
+if ($#ARGV != 0) { print "$#ARGV\n"; die "$Pn usage: file_name\n"; }
+$fnm = $ARGV[0];
+if (! -r $fnm) { die "$Pn: can't read $fnm\n"; }
+
+# Do setup.
+
+$RPT = 15; # lsof repeat time
+$| = 1; # unbuffer output
+$SIG{'INT'} = 'interrupt'; # catch interrupt
+
+# Set path to lsof.
+
+if (($LSOF = &isexec("../lsof")) eq "") { # Try .. first
+ if (($LSOF = &isexec("lsof")) eq "") { # Then try . and $PATH
+ print "can't execute $LSOF\n"; exit 1
+ }
+}
+
+# Read lsof -nPF output from a pipe and gather the PIDs of the processes
+# and file descriptors to watch.
+
+open(P, "$LSOF -nPFpf $fnm|") || die "$Pn: can't pipe to $LSOF\n";
+
+$curpid = -1;
+$pids = "";
+while (<P>) {
+ chop;
+ if (/^p(.*)/) { $curpid = $1; next; } # Identify process.
+ if (/^f/) {
+ if ($curpid > 0) {
+ if ($pids eq "") { $pids = $curpid; }
+ else { $pids = $pids . "," . $curpid; }
+ $curpid = -1;
+ }
+ }
+}
+close(P);
+wait;
+if ($pids eq "") { die "$Pn: no processes using $fnm located.\n"; }
+print "watch_file: $fnm being used by processes:\n\t$pids\n\n";
+
+# Read repeated lsof output from a pipe and display.
+
+$pipe = "$LSOF -ap $pids -r $RPT $fnm";
+open(P, "$pipe|") || die "$Pn: can't pipe: $pipe\n";
+
+while (<P>) { print $_; }
+close(P);
+print "$Pn: unexpected EOF from \"$pipe\"\n";
+exit 1;
+
+
+## isexec($path) -- is $path executable
+#
+# $path = absolute or relative path to file to test for executabiity.
+# Paths that begin with neither '/' nor '.' that arent't found as
+# simple references are also tested with the path prefixes of the
+# PATH environment variable.
+
+sub
+isexec {
+ my ($path) = @_;
+ my ($i, @P, $PATH);
+
+ $path =~ s/^\s+|\s+$//g;
+ if ($path eq "") { return(""); }
+ if (($path =~ m#^[\/\.]#)) {
+ if (-x $path) { return($path); }
+ return("");
+ }
+ $PATH = $ENV{PATH};
+ @P = split(":", $PATH);
+ for ($i = 0; $i <= $#P; $i++) {
+ if (-x "$P[$i]/$path") { return("$P[$i]/$path"); }
+ }
+ return("");
+}
--- /dev/null
+#!/usr/bin/awk -f
+################################################################
+#
+# Program Name : xusers
+# Date Created : 02-27-97
+# Author : Dan A. Mercer
+# Email : damercer@mmm.com
+# :
+# Description : Print list of users and applications signed on
+# : X workstations
+################################################################
+# standard help message
+function help(hlpmsg) {
+basename = ARGV[0]
+sub(/.*\//,"",basename)
+printf "Format: %s [o=[hi]] [s=cdlp] [pattern]\n", basename
+print "Print list of users and applications signed on X workstations"
+print "NOTE: applicationname is truncated to 9 chars"
+print "Arguments:"
+print " o=[h|i] - Options"
+print " h - help - print this message"
+print " i - case insensitive pattern search"
+print " s=[c|d|l|p] - Sort Options"
+print " c - sort by command"
+print " d - sort by display name"
+print " l - sort by login name"
+print " p - sort by pid"
+print " pattern - regex pattern to search commands against"
+
+if (length(hlpmsg)) print hlpmsg
+exit
+}
+BEGIN {
+# process command line
+for (i=1;i<ARGC;i++) {
+ if (ARGV[i] ~ /^o=/) {
+ if (options)
+ help("duplicate option string")
+ options = ARGV[i]
+ sub(/^o=/,"",options)
+ if (options !~ /^[hi]$/)
+ help("Invalid options " options)
+ if ("h" == options)
+ help("")
+ else
+ igncase = 1
+ }
+ else if (ARGV[i] ~ /^s=/) {
+ if (sortorder)
+ help("duplicate sort order string")
+ sortorder = ARGV[i]
+ sub(/^s=/,"",sortorder)
+ if (sortorder !~ /^[cdlp]$/)
+ help("Invalid sort order: '" sortorder "'")
+ if ("p" == sortorder) {
+ sort = "sort -kn2"
+ }
+ else if ("c" == sortorder) {
+ # the 'b' option means ignore leading blanks
+ sort = "sort -kb3"
+ }
+ else if ("l" == sortorder) {
+ sort = "sort -kb1"
+ }
+ else {
+ sort = "sort -kb4"
+ }
+ }
+ else {
+ if (pattern)
+ help("duplicate pattern string")
+ pattern = ARGV[i]
+ }
+ }
+
+# default is to sort by pid
+sort = (sort) ? sort : "sort -kn2"
+
+# check for igncase
+if (pattern && igncase)
+ pattern = tolower(pattern)
+
+# set default pattern
+pattern = (pattern) ? pattern : ".*"
+
+cmd = "lsof -FpLcn -awP -iTCP:6000"
+# ||||| |||| |
+# ||||| |||| X servers use port 6000
+# ||||| |||don't list port names
+# ||||| ||suppress warning messages
+# ||||| |and all conditions
+# ||||| |options
+# |||||
+# ||||Internet addresses
+# |||command name
+# ||login name
+# |process id
+# Format string
+# Output consists of one record per pid, followed by newline
+# delimited fields for command, Login name, and network address
+# The pid is preceded by a 'p', command by a 'c',
+# Login name by an L, and network connection by an 'n'. There may
+# be multiple 'n' entries (for instance for vuewm)
+
+while ((cmd | getline field) > 0) {
+ type = substr(field,1,1)
+ sub("^.","",field)
+ if ("p" == type) {
+ # always output first
+ pid = field
+ PID[pid] = ++ct
+ }
+ else if ("c" == type) {
+ # always output second
+ XAPPL[pid] = field
+ }
+ else if ("L" == type) {
+ # always output fourth
+ USER[pid] = field
+ }
+ else if ("n" == type) {
+ # may be multiple instances - we just use the last
+ gsub(".*->|:6000","",field)
+ DPY[pid] = field
+ }
+ }
+close(cmd)
+
+printf "%8s %5s %-9s %s\n","USER","PID","COMMAND","DISPLAY"
+for (pid in PID) {
+ if (((igncase) ? tolower(XAPPL[pid]) : XAPPL[pid]) ~ pattern)
+ printf "%8s %5d %-9s %s\n", USER[pid],pid,XAPPL[pid],DPY[pid] | sort
+ }
+
+close(sort)
+exit
+}
--- /dev/null
+/*
+ * store.c - common global storage for lsof
+ */
+
+
+/*
+ * Copyright 1994 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1994 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: store.c,v 1.38 2008/10/21 16:21:41 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+
+
+/*
+ * Global storage definitions
+ */
+
+#if defined(HASBLKDEV)
+struct l_dev *BDevtp = (struct l_dev *)NULL;
+ /* block device table pointer */
+int BNdev = 0; /* number of entries in BDevtp[] */
+struct l_dev **BSdev = (struct l_dev **)NULL;
+ /* pointer to BDevtp[] pointers, sorted
+ * by device */
+#endif /* defined(HASBLKDEV) */
+
+int CkPasswd = 0; /* time to check /etc/passwd for change */
+
+#if defined(HAS_STD_CLONE)
+struct clone *Clone = (struct clone *)NULL;
+ /* clone device list */
+#endif /* defined(HAS_STD_CLONE) */
+
+int CmdColW; /* COMMAND column width */
+struct str_lst *Cmdl = (struct str_lst *)NULL;
+ /* command names selected with -c */
+int CmdLim = CMDL; /* COMMAND column width limit */
+int Cmdni = 0; /* command name inclusions selected with -c */
+int Cmdnx = 0; /* command name exclusions selected with -c */
+lsof_rx_t *CmdRx = (lsof_rx_t *)NULL;
+ /* command regular expression table */
+
+#if defined(HASSELINUX)
+cntxlist_t *CntxArg = (cntxlist_t *)NULL;
+ /* security context arguments supplied with
+ * -Z */
+int CntxColW; /* security context column width */
+int CntxStatus = 0; /* security context status: 0 == disabled,
+ * 1 == enabled */
+#endif /* defined(HASSELINUX) */
+
+#if defined(HASDCACHE)
+unsigned DCcksum; /* device cache file checksum */
+int DCfd = -1; /* device cache file descriptor */
+FILE *DCfs = (FILE *)NULL; /* stream pointer for DCfd */
+char *DCpathArg = (char *)NULL; /* device cache path from -D[b|r|u]<path> */
+char *DCpath[] = { /* device cache paths, indexed by DCpathX
+ *when it's >= 0 */
+ (char *)NULL, (char *)NULL, (char *)NULL, (char *)NULL
+};
+int DCpathX = -1; /* device cache path index:
+ * -1 = path not defined
+ * 0 = defined via -D
+ * 1 = defined via HASENVDC
+ * 2 = defined via HASSYSDC
+ * 3 = defined via HASPERSDC and
+ * HASPERSDCPATH */
+int DCrebuilt = 0; /* an unsafe device cache file has been
+ * rebuilt */
+int DCstate = 3; /* device cache state:
+ * 0 = ignore (-Di)
+ * 1 = build (-Db[path])
+ * 2 = read; don't rebuild (-Dr[path])
+ * 3 = update; read and rebuild if
+ * necessary (-Du[path])
+ */
+int DCunsafe = 0; /* device cache file is potentially unsafe,
+ * (The [cm]time check failed.) */
+#endif /* defined(HASDCACHE) */
+
+int DChelp = 0; /* -D? status */
+
+int DevColW; /* DEVICE column width */
+dev_t DevDev; /* device number of /dev or its equivalent */
+struct l_dev *Devtp = (struct l_dev *)NULL;
+ /* device table pointer */
+
+
+/*
+ * Externals for a stkdir(), dumbed-down for older AIX compilers.
+ */
+
+char **Dstk = (char **)NULL; /* the directory stack */
+int Dstkx = 0; /* Dstk[] index */
+int Dstkn = 0; /* Dstk[] entries allocated */
+
+int ErrStat = 0; /* path stat() error count */
+uid_t Euid; /* effective UID of this lsof process */
+int Fand = 0; /* -a option status */
+int Fblock = 0; /* -b option status */
+int Fcntx = 0; /* -Z option status */
+int FdColW; /* FD column width */
+int Ffilesys = 0; /* -f option status:
+ * 0 = paths may be file systems
+ * 1 = paths are just files
+ * 2 = paths must be file systems */
+
+#if defined(HASNCACHE)
+int Fncache = 1; /* -C option status */
+int NcacheReload = 1; /* 1 == call ncache_load() */
+#endif /* defined(HASNCACHE) */
+
+int Ffield = 0; /* -f and -F status */
+int Fhelp = 0; /* -h option status */
+int Fhost = 1; /* -H option status */
+int Fnet = 0; /* -i option status: 0==none
+ * 1==find all
+ * 2==some found*/
+int FnetTy = 0; /* Fnet type request: 0==all
+ * 4==IPv4
+ * 6==IPv6 */
+int Fnfs = 0; /* -N option status: 0==none, 1==find all,
+ * 2==some found*/
+int Fnlink = 0; /* -L option status */
+int Foffset = 0; /* -o option status */
+int Fovhd = 0; /* -O option status */
+int Fport = 1; /* -P option status */
+
+#if defined(HASPMAPENABLED)
+int FportMap = 1; /* +|-M option status */
+#else /* !defined(HASPMAPENABLED) */
+int FportMap = 0; /* +|-M option status */
+#endif /* defined(HASPMAPENABLED) */
+
+int Fpgid = 0; /* -g option status */
+int Fppid = 0; /* -R option status */
+int Fsize = 0; /* -s option status */
+int FcColW; /* FCT column width */
+int FgColW; /* FILE-FLAG column width */
+int FsColW; /* FSTR-ADDR column width */
+int Fsv = FSV_DEFAULT; /* file struct value selections */
+int FsvByf = 0; /* Fsv was set by +f */
+int FsvFlagX = 0; /* hex format status for FSV_FG */
+int NiColW; /* NODE-ID column width */
+char *NiTtl = NITTL; /* NODE-ID column title */
+int Ftcptpi = TCPTPI_STATE; /* -T option status */
+int Fterse = 0; /* -t option status */
+int Funix = 0; /* -U option status */
+int Futol = 1; /* -l option status */
+int Fverbose = 0; /* -V option status */
+
+#if defined(WARNINGSTATE)
+int Fwarn = 1; /* +|-w option status */
+#else /* !defined(WARNINGSTATE) */
+int Fwarn = 0; /* +|-w option status */
+#endif /* defined(WARNINGSTATE) */
+
+#if defined(HASXOPT_VALUE)
+int Fxopt = HASXOPT_VALUE; /* -X option status */
+#endif /* defined(HASXOPT_VALUE) */
+
+int Fxover = 0; /* -x option value */
+int Fzone = 0; /* -z option status */
+
+struct fd_lst *Fdl = (struct fd_lst *)NULL;
+ /* file descriptors selected with -d */
+int FdlTy = -1; /* Fdl[] type: -1 == none
+ * 0 == include
+ * 1 == exclude */
+
+struct fieldsel FieldSel[] = {
+ { LSOF_FID_ACCESS, 0, LSOF_FNM_ACCESS, NULL, 0 }, /* 0 */
+ { LSOF_FID_CMD, 0, LSOF_FNM_CMD, NULL, 0 }, /* 1 */
+ { LSOF_FID_CT, 0, LSOF_FNM_CT, &Fsv, FSV_CT }, /* 2 */
+ { LSOF_FID_DEVCH, 0, LSOF_FNM_DEVCH, NULL, 0 }, /* 3 */
+ { LSOF_FID_DEVN, 0, LSOF_FNM_DEVN, NULL, 0 }, /* 4 */
+ { LSOF_FID_FD, 0, LSOF_FNM_FD, NULL, 0 }, /* 5 */
+ { LSOF_FID_FA, 0, LSOF_FNM_FA, &Fsv, FSV_FA }, /* 6 */
+ { LSOF_FID_FG, 0, LSOF_FNM_FG, &Fsv, FSV_FG }, /* 7 */
+ { LSOF_FID_INODE, 0, LSOF_FNM_INODE, NULL, 0 }, /* 8 */
+ { LSOF_FID_NLINK, 0, LSOF_FNM_NLINK, &Fnlink, 1 }, /* 9 */
+ { LSOF_FID_LOCK, 0, LSOF_FNM_LOCK, NULL, 0 }, /* 10 */
+ { LSOF_FID_LOGIN, 0, LSOF_FNM_LOGIN, NULL, 0 }, /* 11 */
+ { LSOF_FID_MARK, 1, LSOF_FNM_MARK, NULL, 0 }, /* 12 */
+ { LSOF_FID_NAME, 0, LSOF_FNM_NAME, NULL, 0 }, /* 13 */
+ { LSOF_FID_NI, 0, LSOF_FNM_NI, &Fsv, FSV_NI }, /* 14 */
+ { LSOF_FID_OFFSET, 0, LSOF_FNM_OFFSET, NULL, 0 }, /* 15 */
+ { LSOF_FID_PID, 1, LSOF_FNM_PID, NULL, 0 }, /* 16 */
+ { LSOF_FID_PGID, 0, LSOF_FNM_PGID, &Fpgid, 1 }, /* 17 */
+ { LSOF_FID_PROTO, 0, LSOF_FNM_PROTO, NULL, 0 }, /* 18 */
+ { LSOF_FID_RDEV, 0, LSOF_FNM_RDEV, NULL, 0 }, /* 19 */
+ { LSOF_FID_PPID, 0, LSOF_FNM_PPID, &Fppid, 1 }, /* 20 */
+ { LSOF_FID_SIZE, 0, LSOF_FNM_SIZE, NULL, 0 }, /* 21 */
+ { LSOF_FID_STREAM, 0, LSOF_FNM_STREAM, NULL, 0 }, /* 22 */
+ { LSOF_FID_TYPE, 0, LSOF_FNM_TYPE, NULL, 0 }, /* 23 */
+ { LSOF_FID_TCPTPI, 0, LSOF_FNM_TCPTPI, &Ftcptpi, TCPTPI_ALL }, /* 24 */
+ { LSOF_FID_UID, 0, LSOF_FNM_UID, NULL, 0 }, /* 25 */
+ { LSOF_FID_ZONE, 0, LSOF_FNM_ZONE, &Fzone, 1 }, /* 26 */
+ { LSOF_FID_CNTX, 0, LSOF_FNM_CNTX, &Fcntx, 1 }, /* 27 */
+ { LSOF_FID_TERM, 0, LSOF_FNM_TERM, NULL, 0 }, /* 28 */
+
+#if defined(HASFIELDAP1)
+ { '1', 0, HASFIELDAP1, NULL, 0 }, /* TERM+1 */
+#endif /* defined(HASFIELDAP1) */
+
+#if defined(HASFIELDAP2)
+ { '2', 0, HASFIELDAP2, NULL, 0 }, /* TERM+2 */
+#endif /* defined(HASFIELDAP2) */
+
+#if defined(HASFIELDAP3)
+ { '3', 0, HASFIELDAP3, NULL, 0 }, /* TERM+3 */
+#endif /* defined(HASFIELDAP3) */
+
+#if defined(HASFIELDAP4)
+ { '4', 0, HASFIELDAP4, NULL, 0 }, /* TERM+4 */
+#endif /* defined(HASFIELDAP4) */
+
+#if defined(HASFIELDAP5)
+ { '5', 0, HASFIELDAP5, NULL, 0 }, /* TERM+5 */
+#endif /* defined(HASFIELDAP5) */
+
+#if defined(HASFIELDAP6)
+ { '6', 0, HASFIELDAP6, NULL, 0 }, /* TERM+6 */
+#endif /* defined(HASFIELDAP6) */
+
+#if defined(HASFIELDAP7)
+ { '7', 0, HASFIELDAP7, NULL, 0 }, /* TERM+7 */
+#endif /* defined(HASFIELDAP7) */
+
+#if defined(HASFIELDAP8)
+ { '8', 0, HASFIELDAP8, NULL, 0 }, /* TERM+8 */
+#endif /* defined(HASFIELDAP8) */
+
+#if defined(HASFIELDAP9)
+ { '9', 0, HASFIELDAP9, NULL, 0 }, /* TERM+9 */
+#endif /* defined(HASFIELDAP9) */
+
+ { ' ', 0, NULL, NULL, 0 }
+};
+
+int Hdr = 0; /* header print status */
+char *InodeFmt_d = (char *) NULL;
+ /* INODETYPE decimal printf specification */
+char *InodeFmt_x = (char *) NULL;
+ /* INODETYPE hexadecimal printf specification */
+struct lfile *Lf = (struct lfile *)NULL;
+ /* current local file structure */
+struct lproc *Lp = (struct lproc *)NULL;
+ /* current local process table entry */
+struct lproc *Lproc = (struct lproc *)NULL;
+ /* local process table */
+char *Memory = (char *)NULL; /* core file path */
+int MntSup = 0; /* mount supplement state: 0 == none
+ * 1 == create
+ * 2 == read */
+char *MntSupP = (char *)NULL; /* mount supplement path -- if MntSup == 2 */
+
+#if defined(HASPROCFS)
+struct mounts *Mtprocfs = (struct mounts *)NULL;
+ /* /proc mount entry */
+#endif /* defined(HASPROCFS) */
+
+int Mxpgid = 0; /* maximum process group ID table entries */
+int Mxpid = 0; /* maximum PID table entries */
+int Mxuid = 0; /* maximum UID table entries */
+gid_t Mygid; /* real GID of this lsof process */
+int Mypid; /* lsof's process ID */
+uid_t Myuid; /* real UID of this lsof process */
+char *Namech = (char *)NULL; /* name characters for printing */
+size_t Namechl = (size_t)0; /* sizeof(Namech) */
+int NCmdRxU = 0; /* number of CmdRx[] entries */
+int Ndev = 0; /* number of entries in Devtp[] */
+
+#if defined(HASNLIST)
+struct NLIST_TYPE *Nl = (struct NLIST_TYPE *)NULL;
+ /* kernel name list */
+int Nll = 0; /* Nl calloc'd length */
+#endif /* defined(HASNLIST) */
+
+long Nlink = 0l; /* report nlink values below this number
+ * (0 = report all nlink values) */
+int Nlproc = 0; /* number of entries in Lproc[] */
+int NlColW; /* NLINK column width */
+int NmColW; /* NAME column width */
+char *Nmlst = (char *)NULL; /* namelist file path */
+int NodeColW; /* NODE column width */
+int Npgid = 0; /* -g option count */
+int Npgidi = 0; /* -g option inclusion count */
+int Npgidx = 0; /* -g option exclusion count */
+int Npid = 0; /* -p option count */
+int Npidi = 0; /* -p option inclusion count */
+int Npidx = 0; /* -p option exclusion count */
+int Npuns; /* number of unselected PIDs (starts at Npid) */
+int Ntype; /* node type (see N_* symbols) */
+int Nuid = 0; /* -u option count */
+int Nuidexcl = 0; /* -u option count of UIDs excluded */
+int Nuidincl = 0; /* -u option count of UIDs included */
+struct nwad *Nwad = (struct nwad *)NULL;
+ /* list of network addresses */
+int OffDecDig = OFFDECDIG; /* offset decimal form (0t...) digit limit */
+int OffColW; /* OFFSET column width */
+int PgidColW; /* PGID column width */
+int PidColW; /* PID column width */
+struct lfile *Plf = (struct lfile *)NULL;
+ /* previous local file structure */
+char *Pn; /* program name */
+int PpidColW; /* PPID column width */
+
+#if defined(HASPROCFS)
+int Procfind = 0; /* 1 when searching for an proc file system
+ * file and one was found */
+struct procfsid *Procfsid = (struct procfsid *)NULL;
+ /* proc file system PID search table */
+int Procsrch = 0; /* 1 if searching for any proc file system
+ * file */
+#endif /* defined(HASPROCFS) */
+
+int PrPass = 0; /* print pass: 0 = compute column widths
+ * 1 = print */
+int RptTm = 0; /* repeat time -- set by -r */
+struct l_dev **Sdev = (struct l_dev **)NULL;
+ /* pointer to Devtp[] pointers, sorted
+ * by device */
+int Selall = 1; /* all processes are selected (default) */
+int Selflags = 0; /* selection flags -- see SEL* in lsof.h */
+int Setgid = 0; /* setgid state */
+int Selinet = 0; /* select only Internet socket files */
+int Setuidroot = 0; /* setuid-root state */
+struct sfile *Sfile = (struct sfile *)NULL;
+ /* chain of files to search for */
+struct int_lst *Spgid = (struct int_lst *)NULL;
+ /* process group IDs to search for */
+struct int_lst *Spid = (struct int_lst *)NULL;
+ /* Process IDs to search for */
+struct seluid *Suid = (struct seluid *)NULL;
+ /* User IDs to include or exclude */
+int SzColW; /* SIZE column width */
+int SzOffColW; /* SIZE/OFF column width */
+char *SzOffFmt_0t = (char *)NULL;
+ /* SZOFFTYPE 0t%u printf specification */
+char *SzOffFmt_d = (char *)NULL;
+ /* SZOFFTYPE %d printf specification */
+char *SzOffFmt_dv = (char *)NULL;
+ /* SZOFFTYPE %*d printf specification */
+char *SzOffFmt_x = (char *)NULL;
+ /* SZOFFTYPE %#x printf specification */
+int TcpStAlloc = 0; /* allocated (possibly unused) entries in TCP
+ * state tables */
+unsigned char *TcpStI = (unsigned char *)NULL;
+ /* included TCP states */
+int TcpStIn = 0; /* number of entries in TcpStI[] */
+int TcpStOff = 0; /* offset for TCP state number to adjust
+ * negative numbers to an index into TcpSt[],
+ * TcpStI[] and TcpStX[] */
+unsigned char *TcpStX = (unsigned char *)NULL;
+ /* excluded TCP states */
+int TcpStXn = 0; /* number of entries in TcpStX[] */
+int TcpNstates = 0; /* number of TCP states -- either in
+ * tcpstates[] or TcpSt[] */
+char **TcpSt = (char **)NULL; /* local TCP state names, indexed by system
+ * state value */
+char Terminator = '\n'; /* output field terminator */
+int TmLimit = TMLIMIT; /* Readlink() and stat() timeout (seconds) */
+int TypeColW; /* TYPE column width */
+int UdpStAlloc = 0; /* allocated (possibly unused) entries in UDP
+ * state tables */
+unsigned char *UdpStI = (unsigned char *)NULL;
+ /* included UDP states */
+int UdpStIn = 0; /* number of entries in UdpStI[] */
+int UdpStOff = 0; /* offset for UDP state number to adjust
+ * negative numbers to an index into UdpSt[],
+ * UdpStI[] and UdpStX[] */
+unsigned char *UdpStX = (unsigned char *)NULL;
+ /* excluded UDP states */
+int UdpStXn = 0; /* number of entries in UdpStX[] */
+int UdpNstates = 0; /* number of UDP states in UdpSt[] */
+char **UdpSt = (char **)NULL; /* local UDP state names, indexed by system
+ * state number */
+int UserColW; /* USER column width */
+
+#if defined(HASZONES)
+znhash_t **ZoneArg = (znhash_t **)NULL;
+ /* zone arguments supplied with -z */
+#endif /* defined(HASZONES) */
+
+int ZoneColW; /* ZONE column width */
--- /dev/null
+
+ .../lsof_<version>/tests
+
+This sub-directory contains support for lsof's test suite. Find
+more information about the test suite in the 00TESTS file of the
+lsof distribution, which should be in in the parent of this
+subdirectory.
+
+These tests can be activated from .. with:
+
+ $ make test
+
+They can be activated from this directory with:
+
+ $ make
+ $ make test
+ $ make all
+
+These tests are all written in C, so individual tests may be
+activated by executing them directly -- e.g.,
+
+ $ ./LTlock
+
+It may sometimes be necessary to use execution-time options
+alter test behavior. (Some tests will suggest that when they
+encounter certain kinds of errors.) See the 00FAQ and 00TEST files
+in .. for more information.
+
+These tests check lsof field output, not lsof text output. There
+are no tests for lsof text output.
+
+Here is a brief description of the files in this subdirectory:
+
+ 00README this file
+
+ Add2TestDB a script to add the identity of the current
+ test to TestDB
+
+ CkTestDB a script to check the identity of this
+ dialect against the TestDB file
+
+ config.cc a file prepared by ../Configure that contains
+ the name (and possibly the path) to the C
+ compiler for the programs of this sub-directory
+
+ config.cflags a file prepared by ../Configure that contains
+ C compiler flags for the programs of this
+ sub-directory
+
+ config.libs a file prepared by ../Configure that contains
+ library load specifications -- i.e, make(1)
+ LDFLAGS
+
+ config.xobj a file prepared by ../Configure that contains
+ paths to any extra object files (*.o) needed
+ by the C programs in this directory
+
+ LsofTest.h lsof test definitions for C programs
+
+ LTbasic.c C source to basic lsof tests
+
+ LTbigf.c C source to a program that tests large file
+ sizes and offsets on dialects that support
+ file sizes > 32 bits
+
+ LTdnlc.c C source to a program that tests the
+ effectiveness of assembling path names from
+ the kernel's Dynamic Name Lookup Cache
+ (DNLC)
+
+ LTlib.c a support library in C
+
+ LTlock.c C source to a program that tests lock reporting
+
+ LTnfs C source to a program that tests for open NFS
+ files
+
+ LTnlink.c C source to a program that tests lsof's
+ reporting of open file link counts
+
+ LTsock.c C source to program that tests the finding
+ of IPv4 sockets
+
+ LTszoff.c C source to a program that tests file sizes
+ and offsets -- see LTbigf.c for a large
+ file (size > 32 bits) test
+
+ LTunix.c C source to a program that tests the finding
+ of UNIX domain sockets
+
+ Makefile the make(1) control file
+
+ The Makefile clean rule will not remove
+ config.* files, but the spotless rule will.
+ One the spotless rule has been used,
+ ../Configure must be re-run.
+
+ TestDB a data base of dialects where the test
+ suite has been validated
+
+Vic Abell
+April 11, 2002
--- /dev/null
+#!/bin/sh
+#
+# Add2TestDB -- add the current test to the lsof test suite DB
+#
+# This script saves the current TestDB file in TestDB.old and adds
+# the words in config.cflags to it. "-D" prefixes on the words are
+# removed, the words are sorted, and they are joint in a single
+# line that is catenated to TestDB if it isn't already there.
+#
+# $Id: Add2TestDB,v 1.2 2002/04/19 11:53:37 abe Exp $
+
+# Check for config.flags.
+
+if test ! -r config.cflags
+then
+ echo "$0: no ./config.cflags file"
+ exit 1
+fi
+
+# Check for a current data base file.
+
+if test ! -r TestDB
+then
+ echo "$0: no ./TestDB file"
+ exit 1
+fi
+
+# Form a new data base line.
+
+new=""
+for i in `sort < config.cflags`
+do
+ w=`echo $i | sed 's/^-D//'`
+ if test "X$new" = "X"
+ then
+ new=$w
+ else
+ new="$new $w"
+ fi
+done
+
+# See if the new line is already in the data base.
+
+grep "$new" TestDB > /dev/null 2>&1
+if test $? -eq 0
+then
+ echo "\"$new\" is already in TestDB."
+ exit 1
+fi
+
+# Build a new data base file.
+
+if test ! -w TestDB
+then
+ echo "$0: can't write the following to the end of TestDB:"
+ echo " \"$new\""
+ exit 1
+fi
+rm -f TestDB.new
+cp TestDB TestDB.new
+chmod 644 TestDB.new
+echo "$new" >> TestDB.new
+
+# Archive the current data base file, if possible.
+
+if test -d OLD
+then
+ dt=`date`
+ dtm="========== $dt =========="
+ if test -r OLD/TestDB
+ then
+ echo "$dtm" >> OLD/TestDB
+ else
+ echo "$dtm" > OLD/TestDB
+ fi
+ cat TestDB >> OLD/TestDB
+fi
+
+# Put the new data base file in place.
+
+mv TestDB.new TestDB
+echo "\"$new\" added to TestDB."
+exit 0
--- /dev/null
+#!/bin/sh
+#
+# CkTestDB -- see if this dialect is has been tested
+#
+# This script builds a line from config.flags in the form of lines in
+# ./TestDB, (See Add2TestDB.)
+#
+# It then compares the line to TestDB. If the line is found, the script
+# exits. if the line is not found, the script issues a warning and requests
+# a go-ahead confirmation.
+#
+# The script will exit 0 if the test line is in the DB or the go-ahead
+# confirmation is positive.
+#
+# $Id: CkTestDB,v 1.2 2002/04/19 11:54:00 abe Exp $
+
+# Check for config.flags.
+
+if test ! -r config.cflags
+then
+ echo "$0: no ./config.cflags file"
+ exit 1
+fi
+
+# Check for a current data base file.
+
+if test ! -r TestDB
+then
+ echo "$0: no ./TestDB file"
+ exit 1
+fi
+
+# Form a data base line.
+
+new=""
+for i in `sort < config.cflags`
+do
+ w=`echo $i | sed 's/^-D//'`
+ if test "X$new" = "X"
+ then
+ new=$w
+ else
+ new="$new $w"
+ fi
+done
+
+# See if the line is already in the data base. Exit with success (0), if it is.
+
+grep "^$new\$" TestDB > /dev/null 2>&1
+if test $? -eq 0
+then
+ exit 0
+fi
+
+# This dialect may never have been validated with the test suite.
+
+# If the standard input is not a TTY, quit, because no interaction
+# is possible.
+
+tty -s > /dev/null 2>&1
+if test $? -ne 0
+then
+ echo ""
+ echo "This suite has not been validated on:"
+ echo ""
+ echo " $new"
+ echo ""
+ exit 1
+fi
+
+# Establish trap and stty handling.
+
+ISIG=":"
+trap '$ISIG; exit 1' 1 2 3 15
+stty -a 2>&1 | grep isig > /dev/null
+if test $? -eq 0
+then
+ stty -a 2>&1 | egrep -e -isig > /dev/null
+ if test $? -eq 0
+ then
+ ISIG="stty -isig"
+ stty isig
+ fi
+fi
+
+# Establish echo type -- Berkeley or SYSV.
+
+j=`echo -n ""`
+if test "X$j" = "X-n "
+then
+ EC="\c"
+ EO=""
+else
+ EC=""
+ EO="-n"
+fi
+
+# Display a validation warning.
+
+cat << .CAT_MARK > /dev/tty
+
+==================================================================
+
+!!!WARNING!!!
+
+This dialect or its particular version may not have been validated
+with the lsof test suite. Consequently some tests may fail or may
+not even compile.
+
+This is the computed identity of this dialect, not found in the
+test data base file, ./TestDB:
+
+.CAT_MARK
+echo " $new" > /dev/tty
+END=0
+while test $END = 0
+do
+ echo "" > /dev/tty
+ echo $EO "Do you want to continue (y|n) [n]? $EC" > /dev/tty
+ read ANS EXCESS
+ if test "X$ANS" = "Xn" -o "X$ANS" = "XN"
+ then
+ exit 1
+ fi
+ if test "X$ANS" = "Xy" -o "X$ANS" = "XY"
+ then
+ exit 0
+ else
+ echo "Please answer y or n." > /dev/tty
+ fi
+done
+
+# Should never get here!
+
+echo "$0: unexpected failure!"
+exit 2
--- /dev/null
+/*
+ * LTbasic.c -- Lsof Test basic tests
+ *
+ * The basic tests measure the finding by lsof of its own open CWD, open
+ * executable (when possible), and open /dev/kmem files.
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+
+/*
+ * Local definitions
+ */
+
+
+/*
+ * Globals
+ */
+
+char *Pn = (char *)NULL; /* program name */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static char *tstlsof,(char **texec, char **tkmem, char **tproc));
+
+
+/*
+ * Main program for dialects that support locking tests.
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ char *em; /* error message pointer */
+ char *texec = (char *)NULL; /* lsof executable test result */
+ char *tkmem = (char *)NULL; /* /dev/kmem test result */
+ char *tproc = (char *)NULL; /* lsof process test result */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "h", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h]", Pn);
+ PrtMsgX (" -h print help (this panel)", Pn, cleanup,
+ xv);
+ }
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * Test lsof.
+ */
+ if ((em = tstlsof(&texec, &tkmem, &tproc)))
+ PrtMsg(em, Pn);
+ if (texec)
+ PrtMsg(texec, Pn);
+ if (tkmem)
+ PrtMsg(tkmem, Pn);
+ if (tproc)
+ PrtMsg(tproc, Pn);
+/*
+ * Compute exit value and exit.
+ */
+ if (em || texec || tkmem || tproc) {
+ if (strcmp(LT_DEF_LSOF_PATH, LsofPath)) {
+ PrtMsg (" ", Pn);
+ PrtMsg ("Hint: you used the LT_LSOF_PATH environment variable to",
+ Pn);
+ PrtMsg (" specify this path to the lsof executable:\n", Pn);
+ (void) snprintf(buf, sizeof(buf) - 1, " %s\n", LsofPath);
+ buf[sizeof(buf) - 1] = '\0';
+ PrtMsg (buf, Pn);
+ PrtMsgX(" Make sure its revision is 4.63 or higher.",
+ Pn, cleanup, 1);
+ } else
+ PrtMsgX("", Pn, cleanup, 1);
+ }
+ (void) PrtMsgX("OK", Pn, cleanup, 0);
+ return(0);
+}
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+}
+
+
+/*
+ * tstlsof() -- test for the lsof process
+ */
+
+static char *
+tstlsof(texec, tkmem, tproc)
+ char **texec; /* result of the executable test */
+ char **tkmem; /* result of the /dev/kmem test */
+ char **tproc; /* result of the lsof process test */
+{
+ char buf[2048]; /* temporary buffer */
+ char *cem; /* current error message pointer */
+ LTfldo_t *cmdp; /* command pointer */
+ LTdev_t cwddc; /* CWD device components */
+ struct stat cwdsb; /* CWD stat(2) buffer */
+ LTfldo_t *devp; /* device pointer */
+ int execs = 0; /* executable status */
+ int fdn; /* FD is a number */
+ LTfldo_t *fdp; /* file descriptor pointer */
+ LTfldo_t *fop; /* field output pointer */
+ char ibuf[64]; /* inode string buffer */
+ LTfldo_t *inop; /* inode number pointer */
+ LTdev_t kmemdc; /* /dev/kmem device components */
+ int kmems = 0; /* kmem status */
+ struct stat kmemsb; /* /dev/kmem stat(2) buffer */
+ LTdev_t lsofdc; /* lsof device components */
+ struct stat lsofsb; /* lsof stat(2) buffer */
+ int nf; /* number of fields */
+ char *opv[4]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ int procs = 0; /* process status */
+ LTfldo_t *rdevp; /* raw device pointer */
+ char *tcp; /* temporary character pointer */
+ int ti; /* temporary integer */
+ LTdev_t tmpdc; /* temporary device components */
+ LTfldo_t *typ; /* file type pointer */
+ int xwhile; /* exit while() flag */
+
+/*
+ * Get lsof executable's stat(2) information.
+ */
+ if (stat(LsofPath, &lsofsb)) {
+ (void) snprintf(buf, sizeof(buf) - 1, "ERROR!!! stat(%s): %s",
+ LsofPath, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ execs = 1;
+ } else if ((cem = ConvStatDev(&lsofsb.st_dev, &lsofdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ execs = 1;
+ }
+
+#if defined(LT_KMEM)
+/*
+ * Get /dev/kmem's stat(2) information.
+ */
+ if (stat("/dev/kmem", &kmemsb)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't stat(2) /dev/kmem: %s", strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ kmems = 1;
+ } else if ((cem = ConvStatDev(&kmemsb.st_rdev, &kmemdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ kmems = 1;
+ }
+#else /* !defined(LT_KMEM) */
+ kmems = 1;
+#endif /* defined(LT_KMEM) */
+
+/*
+ * Get CWD's stat(2) information.
+ */
+ if (stat(".", &cwdsb)) {
+ (void) snprintf(buf, sizeof(buf) - 1, "ERROR!!! stat(.): %s",
+ strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ procs = 1;
+ } else if ((cem = ConvStatDev(&cwdsb.st_dev, &cwddc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ procs = 1;
+ }
+
+/*
+ * Complete the option vector and start lsof execution.
+ */
+ ti = 0;
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+#if defined(USE_LSOF_X_OPT)
+ opv[ti++] = "-X";
+#endif /* defined(USE_LSOF_X_OPT) */
+
+ opv[ti++] = "-clsof";
+ opv[ti] = (char *)NULL;
+ if ((cem = ExecLsof(opv))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+/*
+ * Read lsof output.
+ */
+ xwhile = execs + kmems + procs;
+ while ((xwhile < 3) && (fop = RdFrLsof(&nf, &cem))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != LsofPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line. Scan its fields.
+ */
+ if (!pids)
+ break;
+ devp = inop = rdevp = typ = (LTfldo_t *)NULL;
+ fdp = fop;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch(fop->ft) {
+ case LSOF_FID_DEVN:
+ devp = fop;
+ break;
+ case LSOF_FID_INODE:
+ inop = fop;
+ break;
+ case LSOF_FID_RDEV:
+ rdevp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * A file descriptor line has been processes.
+ *
+ * Set the descriptor's numeric status.
+ *
+ * Check descriptor by FD type.
+ */
+
+ for (fdn = 0, tcp = fdp->v; *tcp; tcp++) {
+ if (!isdigit((unsigned char)*tcp)) {
+ fdn = -1;
+ break;
+ }
+ fdn = (fdn * 10) + (int)(*tcp - '0');
+ }
+ if (!procs
+ && (fdn == -1)
+ && !strcasecmp(fdp->v, "cwd")
+ && typ
+ && (!strcasecmp(typ->v, "DIR") || !strcasecmp(typ->v, "VDIR"))
+ ) {
+
+ /*
+ * This is the CWD for the process. Make sure its information
+ * matches what stat(2) said about the CWD.
+ */
+ if (!devp || !inop)
+ break;
+ if ((cem = ConvLsofDev(devp->v, &tmpdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ (void) snprintf(ibuf, sizeof(ibuf) - 1, "%u",
+ (unsigned int)cwdsb.st_ino);
+ ibuf[sizeof(ibuf) - 1] = '\0';
+ if ((tmpdc.maj == cwddc.maj)
+ && (tmpdc.min == cwddc.min)
+ && (tmpdc.unit == cwddc.unit)
+ && !strcmp(inop->v, ibuf)
+ ) {
+ procs = 1;
+ xwhile++;
+ }
+ break;
+ }
+ if (!kmems
+ && (fdn >= 0)
+ && typ
+ && (!strcasecmp(typ->v, "CHR") || !strcasecmp(typ->v, "VCHR"))
+ ) {
+
+ /*
+ * /dev/kmem hasn't been found and this is an open character device
+ * file with a numeric descriptor.
+ *
+ * See if it is /dev/kmem.
+ */
+ if (!inop || !rdevp)
+ break;
+ if ((cem = ConvLsofDev(rdevp->v, &tmpdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ (void) snprintf(ibuf, sizeof(ibuf) - 1, "%u",
+ (unsigned int)kmemsb.st_ino);
+ ibuf[sizeof(ibuf) - 1] = '\0';
+ if ((tmpdc.maj == kmemdc.maj)
+ && (tmpdc.min == kmemdc.min)
+ && (tmpdc.unit == kmemdc.unit)
+ && !strcmp(inop->v, ibuf)
+ ) {
+ kmems = 1;
+ xwhile++;
+ }
+ break;
+ }
+ if (!execs
+ && (fdn == -1)
+ && typ
+ && (!strcasecmp(typ->v, "REG") || !strcasecmp(typ->v, "VREG"))
+ ) {
+
+ /*
+ * If this is a regular file with a non-numeric FD, it may be the
+ * executable.
+ */
+ if (!devp || !inop)
+ break;
+ if ((cem = ConvLsofDev(devp->v, &lsofdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ (void) snprintf(ibuf, sizeof(ibuf) - 1, "%u",
+ (unsigned int)lsofsb.st_ino);
+ ibuf[sizeof(ibuf) - 1] = '\0';
+ if ((tmpdc.maj == lsofdc.maj)
+ && (tmpdc.min == lsofdc.min)
+ && (tmpdc.unit == lsofdc.unit)
+ && !strcmp(inop->v, ibuf)
+ ) {
+ execs = 1;
+ xwhile++;
+ }
+ }
+ }
+ }
+ (void) StopLsof();
+ if (!execs)
+ *texec = "ERROR!!! open lsof executable wasn't found.";
+ if (!kmems)
+ *tkmem = "ERROR!!! open lsof /dev/kmem usage wasn't found.";
+ if (!procs)
+ *tproc = "ERROR!!! lsof process wasn't found.";
+ return(pem);
+}
--- /dev/null
+/*
+ * LTbigf.c -- Lsof Test big file size and offset tests
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+
+#if !defined(LT_BIGF)
+
+/*
+ * Here begins the version of this program for dialects that don't support
+ * large files.
+ */
+
+
+/*
+ * Main program for dialects that don't support large files
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char *pn; /* program name */
+/*
+ * Get program name and issue start and exit message.
+ */
+ if ((pn = (char *)strrchr(argv[0], '/')))
+ pn++;
+ else
+ pn = argv[0];
+
+ (void) printf("%s ... %s\n", pn, LT_DONT_DO_TEST);
+ return(0);
+}
+#else /* defined(LT_BIGF) */
+
+/*
+ * Here begins the version of this program for dialects that support
+ * large files.
+ */
+
+#include "lsof_fields.h"
+
+
+/*
+ * Pre-definitions that may be changed by specific dialects
+ */
+
+#define OFFTST_STAT 1 /* offset tests status */
+
+
+#if defined(LT_DIAL_aix)
+/*
+ * AIX-specific definitions
+ */
+
+#define OFFSET_T off64_t /* define offset type */
+#endif /* defined(LT_DIAL_aix) */
+
+
+#if defined(LT_DIAL_bsdi)
+/*
+ * BSDI-specific definitions
+ */
+
+#define OFFSET_T off_t /* define offset type */
+#define OPENF open /* define open function */
+#define SEEKF lseek /* define seek function */
+#define STATF stat /* define stat function */
+#define STATS struct stat /* define stat structure */
+#endif /* defined(LT_DIAL_bsdi) */
+
+
+#if defined(LT_DIAL_darwin)
+/*
+ * Darwin-specific definitions
+ */
+
+# if LT_VERS>=900
+#define OFFSET_T off_t /* define offset type */
+#define OPENF open /* define open function */
+#define SEEKF lseek /* define seek function */
+#define STATF stat /* define stat function */
+#define STATS struct stat /* define stat structure */
+# endif /* LT_VERS>=900 */
+#endif /* defined(LT_DIAL_darwin) */
+
+
+#if defined(LT_DIAL_du)
+/*
+ * DEC_OSF/1|Digital_UNIX|Tru64_UNIX-specific items
+ */
+
+#define OFFSET_T off_t /* define offset type */
+#define OPENF open /* define open function */
+#define SEEKF lseek /* define seek function */
+#define STATF stat /* define stat function */
+#define STATS struct stat /* define stat structure */
+#endif /* defined(LT_DIAL_du) */
+
+
+#if defined(LT_DIAL_freebsd)
+/*
+ * FreeBSD-specific definitions
+ */
+
+#define OFFSET_T off_t /* define offset type */
+#define OPENF open /* define open function */
+#define SEEKF lseek /* define seek function */
+#define STATF stat /* define stat function */
+#define STATS struct stat /* define stat structure */
+#endif /* defined(LT_DIAL_freebsd) */
+
+
+#if defined(LT_DIAL_linux)
+/*
+ * Linux-specific definitions
+ */
+
+#undef OFFTST_STAT
+#define OFFTST_STAT 0 /* Linux lsof may not be able to report
+ * offsets -- see the function
+ * ck_Linux_offset_support() */
+#define OFFSET_T off_t /* define offset type */
+#define OPENF open /* define open function */
+#define SEEKF lseek /* define seek function */
+#define STATF stat /* define stat function */
+#define STATS struct stat /* define stat structure */
+
+_PROTOTYPE(static int ck_Linux_offset_support,(void));
+#endif /* defined(LT_DIAL_linux) */
+
+
+#if defined(LT_DIAL_hpux)
+/*
+ * HP-UX-specific definitions
+ */
+
+#define OFFSET_T off64_t /* define offset type */
+#endif /* defined(LT_DIAL_hpux) */
+
+
+#if defined(LT_DIAL_netbsd)
+/*
+ * NetBSD-specific definitions
+ */
+
+#define OFFSET_T off_t /* define offset type */
+#define OPENF open /* define open function */
+#define SEEKF lseek /* define seek function */
+#define STATF stat /* define stat function */
+#define STATS struct stat /* define stat structure */
+#endif /* defined(LT_DIAL_netbsd) */
+
+
+#if defined(LT_DIAL_openbsd)
+/*
+ * OpenBSD-specific definitions
+ */
+
+#define OFFSET_T off_t /* define offset type */
+#define OPENF open /* define open function */
+#define SEEKF lseek /* define seek function */
+#define STATF stat /* define stat function */
+#define STATS struct stat /* define stat structure */
+#endif /* defined(LT_DIAL_openbsd) */
+
+
+#if defined(LT_DIAL_ou)
+/*
+ * OpenUNIX-specific items
+ */
+
+#include <signal.h>
+
+#define IGNORE_SIGXFSZ
+#define OFFSET_T off64_t /* define offset type */
+#endif /* defined(LT_DIAL_ou) */
+
+
+#if defined(LT_DIAL_solaris)
+/*
+ * Solaris-specific definitions
+ */
+
+#define OFFSET_T off64_t /* define offset type */
+#endif /* defined(LT_DIAL_solaris) */
+
+
+#if defined(LT_DIAL_uw)
+/*
+ * UnixWare-specific items
+ */
+
+#include <signal.h>
+
+#define IGNORE_SIGXFSZ
+#define OFFSET_T off64_t /* define offset type */
+#endif /* defined(LT_DIAL_uw) */
+
+
+/*
+ * Local definitions
+ */
+
+#if !defined(OPENF)
+#define OPENF open64 /* open() function */
+#endif /* !defined(OPENF) */
+
+#if !defined(OFFSET_T)
+#define OFFSET_T unsigned long long /* offset type */
+#endif /* !defined(OFFSET_T) */
+
+#if !defined(SEEKF)
+#define SEEKF lseek64 /* seek() function */
+# endif /* !defined(SEEKF) */
+
+#if !defined(STATF)
+#define STATF stat64 /* stat(2) structure */
+#endif /* !defined(STATF) */
+
+#if !defined(STATS)
+#define STATS struct stat64 /* stat(2) structure */
+#endif /* !defined(STATS) */
+
+#define TST_OFFT 0 /* test offset in 0t decimal*/
+#define TST_OFFX 1 /* test offset in hex */
+#define TST_SZ 2 /* test size */
+
+
+/*
+ * Globals
+ */
+
+int Fd = -1; /* test file descriptor; open if >= 0 */
+pid_t MyPid = (pid_t)0; /* PID of this process */
+char *Path = (char *)NULL; /* test file path; none if NULL */
+char *Pn = (char *)NULL; /* program name */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static int tstwlsof,(int tt, char *opt, OFFSET_T sz));
+
+
+/*
+ * Main program for dialects that support large files
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ int do_offt = OFFTST_STAT; /* do offset tests if == 1 */
+ char *em; /* error message pointer */
+ int i; /* temporary integer */
+ int len; /* string length */
+ OFFSET_T sz = 0x140000000ll; /* test file size */
+ char szbuf[64]; /* size buffer */
+ char *tcp; /* temporary character pointer */
+ int tofft = 0; /* 0t offset test result */
+ int toffx = 0; /* 0x offset test result */
+ int tsz = 0; /* size test result */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "hp:", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h] [-p path]", Pn);
+ PrtMsg (" -h print help (this panel)", Pn);
+ PrtMsgX (" -p path define test file path", Pn, cleanup, xv);
+ }
+
+#if defined(LT_DIAL_linux)
+/*
+ * If this is Linux, see if lsof can report file offsets.
+ */
+ do_offt = ck_Linux_offset_support();
+#endif /* defined(LT_DIAL_linux) */
+
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * Construct the path. If LT_BIGSZOFF_PATH is defined in the environment,
+ * use it. otherwise construct a path in the CWD.
+ */
+ if (!(Path = LTopt_p)) {
+ (void) snprintf(buf, sizeof(buf), "./config.LTbigf%ld",
+ (long)MyPid);
+ buf[sizeof(buf) - 1] = '\0';
+ Path = MkStrCpy(buf, &len);
+ }
+/*
+ * Fill buffer for writing to the test file.
+ */
+ for (i = 0; i < sizeof(buf); i++) {
+ buf[i] = (char)(i & 0xff);
+ }
+
+#if defined(IGNORE_SIGXFSZ)
+/*
+ * Ignore SIGXFSZ, if directed by a dialect-specific option.
+ */
+ (void) signal(SIGXFSZ, SIG_IGN);
+#endif /* defined(IGNORE_SIGXFSZ) */
+
+/*
+ * Open a new test file at the specified path.
+ */
+ (void) unlink(Path);
+ if ((Fd = OPENF(Path, O_RDWR|O_CREAT, 0600)) < 0) {
+ (void) fprintf(stderr, "ERROR!!! can't open %s\n", Path);
+
+print_hint:
+
+ /*
+ * Print a hint about the LT_BIGSZOFF_PATH environment variable.
+ */
+
+ MsgStat = 1;
+ (void) snprintf(buf, sizeof(buf) - 1, " Errno %d: %s",
+ errno, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsg(buf, Pn);
+ (void) PrtMsg("Hint: try using \"-p path\" to supply a path in a", Pn);
+ (void) PrtMsg("file system that has large file support enabled.\n", Pn);
+ (void) PrtMsg("Hint: try raising the process ulimit file block", Pn);
+ (void) PrtMsg("size to a value that will permit this test to", Pn);
+ (void) snprintf(szbuf, sizeof(szbuf) - 1, "%lld", (long long)sz);
+ szbuf[sizeof(szbuf) - 1] = '\0';
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "write a file whose size appears to be %s", szbuf);
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsg(buf, Pn);
+ (void) PrtMsg("bytes. (The file really isn't that big -- it", Pn);
+ (void) PrtMsg("just has a large \"hole\" in its mid-section.)\n", Pn);
+ (void) PrtMsgX("See 00FAQ and 00TEST for more information.", Pn,
+ cleanup, 1);
+ }
+/*
+ * Write a buffer load at the beginning of the file.
+ */
+ if (SEEKF(Fd, (OFFSET_T)0, SEEK_SET) < 0) {
+ (void) fprintf(stderr,
+ "ERROR!!! can't seek to the beginning of %s\n", Path);
+ goto print_hint;
+ }
+ if (write(Fd, buf, sizeof(buf)) != sizeof(buf)) {
+ (void) fprintf(stderr,
+ "ERROR!!! can't write %d bytes to the beginning of %s\n",
+ (int)sizeof(buf), Path);
+ goto print_hint;
+ }
+/*
+ * Write a buffer load near the end of the file to bring it to the
+ * specified length. Leave the file open so lsof can find it.
+ */
+ if (SEEKF(Fd, (OFFSET_T)(sz - sizeof(buf)), SEEK_SET) < 0) {
+ (void) snprintf(szbuf, sizeof(szbuf) - 1, "%lld",
+ (unsigned long long)(sz - sizeof(buf)));
+ (void) fprintf(stderr, "ERROR!!! can't seek to %s in %s\n", szbuf,
+ Path);
+ goto print_hint;
+ }
+ if (write(Fd, buf, sizeof(buf)) != sizeof(buf)) {
+ (void) fprintf(stderr,
+ "ERROR!!! can't write %d bytes near the end of %s\n",
+ (int)sizeof(buf), Path);
+ goto print_hint;
+ }
+/*
+ * Fsync() the file.
+ */
+ if (fsync(Fd)) {
+ (void) fprintf(stderr, "ERROR!!! can't fsync %s\n", Path);
+ goto print_hint;
+ }
+
+/*
+ * If this dialect can't report offsets, disable the offset tests.
+ */
+ if (!do_offt) {
+ tofft = toffx = 1;
+ PrtMsg("WARNING!!! lsof can't return file offsets for this dialect,",
+ Pn);
+ PrtMsg(" so offset tests have been disabled.", Pn);
+ }
+/*
+ * Do file size test.
+ */
+ tsz = tstwlsof(TST_SZ, "-s", sz);
+/*
+ * If enabled, do offset tests.
+ */
+ if (!tofft)
+ tofft = tstwlsof(TST_OFFT, "-oo20", sz);
+ if (!toffx)
+ toffx = tstwlsof(TST_OFFX, "-oo2", sz);
+/*
+ * Compute exit value and exit.
+ */
+ if ((tsz != 1) || (tofft != 1) || (toffx != 1)) {
+ tcp = (char *)NULL;
+ xv = 1;
+ } else {
+ tcp = "OK";
+ xv = 0;
+ }
+ (void) PrtMsgX(tcp, Pn, cleanup, xv);
+ return(0);
+}
+
+
+#if defined(LT_DIAL_linux)
+/*
+ * ck_Linux_offset_support() -- see if lsof can report offsets for this
+ * Linux implementation
+ */
+
+static int
+ck_Linux_offset_support()
+{
+ char buf[1024]; /* lsof output line buffer */
+ int bufl = sizeof(buf); /* size of buf[] */
+ char *opv[5]; /* option vector for lsof */
+ int rv = 1; /* return value:
+ * 0 == no lsof offset support
+ * 1 == lsof offset support */
+/*
+ * Ask lsof to report the test's FD zero offset.
+ */
+ if (IsLsofExec())
+ return(0);
+ opv[0] = "-o";
+ snprintf(buf, bufl - 1, "-p%d", (int)getpid());
+ opv[1] = buf;
+ opv[2] = "-ad0";
+ opv[3] = "+w";
+ opv[4] = (char *)NULL;
+ if (ExecLsof(opv))
+ return(0);
+/*
+ * Read the lsof output. Look for a line with "WARNING: can't report offset"
+ * in it. If it is found, then this Linux lsof can't report offsets.
+ */
+ while(fgets(buf, bufl - 1, LsofFs)) {
+ if (strstr(buf, "WARNING: can't report offset")) {
+ rv = 0;
+ break;
+ }
+ }
+ (void) StopLsof();
+ return(rv);
+}
+#endif /* defined(LT_DIAL_linux) */
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+ if (Fd >= 0) {
+/*
+ * Close the test file.
+ *
+ * But first unlink it to discourage some kernel file system implementations
+ * (e.g., HFS on Apple Darwin, aka Mac OS X) from trying to fill the file's
+ * large holes. (Filling can take a long time.)
+ */
+ if (Path) {
+ (void) unlink(Path);
+ Path = (char *)NULL;
+ }
+ (void) close(Fd);
+ Fd = -1;
+ }
+}
+
+
+/*
+ * tstwlsof() -- test the open file with lsof
+ */
+
+static int
+tstwlsof(tt, opt, sz)
+ int tt; /* test type -- i.e., TST_* */
+ char *opt; /* additional lsof options */
+ OFFSET_T sz; /* expected size (and offset) */
+{
+ char buf[2048], buf1[2048]; /* temporary buffers */
+ LTfldo_t *cmdp; /* command pointer */
+ LTfldo_t *devp; /* device pointer */
+ char *em; /* error message pointer */
+ int ff = 0; /* file found status */
+ LTfldo_t *fop; /* field output pointer */
+ LTfldo_t *inop; /* inode number pointer */
+ LTdev_t lsofdc; /* lsof device components */
+ int nf; /* number of fields */
+ LTfldo_t *nmp; /* file name pointer */
+ LTfldo_t *offp; /* file offset pointer */
+ char *opv[4]; /* option vector for ExecLsof() */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ STATS sb; /* stat(2) buffer */
+ LTdev_t stdc; /* stat(2) device components */
+ LTfldo_t *szp; /* file size pointer */
+ LTfldo_t *tfop; /* temporary field output pointer */
+ int ti; /* temporary index */
+ LTfldo_t *typ; /* file type pointer */
+ int xv = 0; /* exit value */
+/*
+ * Check the test type.
+ */
+ switch (tt) {
+ case TST_OFFT:
+ case TST_OFFX:
+ case TST_SZ:
+ break;
+ default:
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! unknown test type: %d", tt);
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Get test file's information.
+ */
+ if (STATF(Path, &sb)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't stat(2) %s: %s", Path, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Extract components from test file's device number.
+ */
+ if ((em = ConvStatDev(&sb.st_dev, &stdc))) {
+ (void) PrtMsg(em, Pn);
+ return(0);
+ }
+/*
+ * Complete the option vector and start lsof execution.
+ */
+ ti = 0;
+ if (opt && *opt)
+ opv[ti++] = opt;
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#else /* !defined(USE_LSOF_C_OPT) */
+ opv[ti++] = "--";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+ opv[ti++] = Path;
+ opv[ti] = (char *)NULL;
+ if ((em = ExecLsof(opv))) {
+ (void) PrtMsg(em, Pn);
+ return(0);
+ }
+/*
+ * Read lsof output.
+ */
+ while (!ff && (fop = RdFrLsof(&nf, &em))) {
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != MyPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line.
+ *
+ * Scan for device number, inode number, name, offset, size, and type
+ * fields.
+ */
+ if (!pids)
+ break;
+ devp = inop = nmp = offp = szp = typ = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch(fop->ft) {
+ case LSOF_FID_DEVN:
+ devp = fop;
+ break;
+ case LSOF_FID_INODE:
+ inop = fop;
+ break;
+ case LSOF_FID_NAME:
+ nmp = fop;
+ break;
+ case LSOF_FID_OFFSET:
+ offp = fop;
+ break;
+ case LSOF_FID_SIZE:
+ szp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the results of the file descriptor field scan.
+ *
+ * (Don't compare path names because of symbolic link interference.)
+ */
+ if (!devp || !inop || !nmp || !typ)
+ break;
+ if (strcasecmp(typ->v, "reg") && strcasecmp(typ->v, "vreg"))
+ break;
+ if (ConvLsofDev(devp->v, &lsofdc))
+ break;
+ if ((stdc.maj != lsofdc.maj)
+ || (stdc.min != lsofdc.min)
+ || (stdc.unit != lsofdc.unit))
+ break;
+ (void) snprintf(buf, sizeof(buf) - 1, "%llu",
+ (unsigned long long)sb.st_ino);
+ buf[sizeof(buf) - 1] = '\0';
+ if (strcmp(inop->v, buf))
+ break;
+ /*
+ * The specifed file has been located. Check its size or offset,
+ * according to the tt argument.
+ */
+ ff = 1;
+ switch (tt) {
+ case TST_OFFT:
+ case TST_SZ:
+
+ /*
+ * Test the size as an offset in decimal with a leading "0t", or
+ * test the size as a size in decimal.
+ */
+ (void) snprintf(buf, sizeof(buf) - 1,
+ (tt == TST_SZ) ? "%llu" : "0t%llu",
+ (unsigned long long)sz);
+ buf[sizeof(buf) - 1] = '\0';
+ tfop = (tt == TST_SZ) ? szp : offp;
+ if (!tfop || strcmp(tfop->v, buf)) {
+ (void) snprintf(buf1, sizeof(buf1) - 1,
+ "%s mismatch: expected %s, got %s",
+ (tt == TST_SZ) ? "size" : "0t offset",
+ buf,
+ tfop ? tfop->v : "nothing");
+ buf1[sizeof(buf1) - 1] = '\0';
+ (void) PrtMsg(buf1, Pn);
+ xv = 0;
+ } else
+ xv = 1;
+ break;
+ case TST_OFFX:
+
+ /*
+ * Test the size as an offset in hex.
+ */
+ (void) snprintf(buf, sizeof(buf) - 1, "0x%llx",
+ (unsigned long long)sz);
+ buf[sizeof(buf) - 1] = '\0';
+ if (!offp || strcmp(offp->v, buf)) {
+ (void) snprintf(buf1, sizeof(buf1) - 1,
+ "0x offset mismatch: expected %s, got %s",
+ buf,
+ offp ? offp->v : "nothing");
+ buf1[sizeof(buf1) - 1] = '\0';
+ (void) PrtMsg(buf1, Pn);
+ xv = 0;
+ } else
+ xv = 1;
+ }
+ break;
+ }
+ }
+ (void) StopLsof();
+ if (em) {
+
+ /*
+ * RdFrLsof() encountered an error.
+ */
+ (void) PrtMsg(em, Pn);
+ xv = 0;
+ }
+ if (!ff) {
+ (void) snprintf(buf, sizeof(buf) - 1, "%s not found by lsof", Path);
+ buf[sizeof(buf) - 1] = '\0';
+ PrtMsg(buf, Pn);
+ xv = 0;
+ }
+ return(xv);
+}
+#endif /* defined(LT_BIG) */
--- /dev/null
+/*
+ * LTdnlc.c -- Lsof Test Dynamic Name Lookup Cache test
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+
+/*
+ * Pre-definitions that may be revoked by specific dialects
+ */
+
+#define DO_TEST /* do the test */
+
+
+/*
+ * Dialect-specific items
+ */
+
+
+#if defined(LT_DIAL_aix)
+/*
+ * AIX-specific items
+ */
+
+#undef DO_TEST
+#endif /* defined(LT_DIAL_aix) */
+
+
+#if defined(LT_DIAL_darwin)
+/*
+ * Darwin-specific items
+ */
+
+# if LT_VERS<800
+#undef DO_TEST
+# endif /* LT_VERS<800 */
+#endif /* defined(LT_DIAL_darwin) */
+
+
+/*
+ * Local definitions
+ */
+
+#define ATTEMPT_CT 5 /* number of lsof CWD lookup attempts */
+#define LSPATH "/bin/ls" /* path to ls(1) */
+#define SUCCESS_THRESH 50.0 /* success threshold */
+
+
+/*
+ * Globals
+ */
+
+pid_t MyPid = (pid_t)0; /* PID of this process */
+char *Pn = (char *)NULL; /* program name */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static char *FindLsofCwd,(int *ff, LTdev_t *cwddc, char *ibuf));
+
+
+/*
+ * Main program
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ char cwd[MAXPATHLEN + 1]; /* CWD */
+ LTdev_t cwddc; /* CWD device components */
+ char *em; /* error message pointer */
+ int ff; /* FindFile() file-found flag */
+ int fpathct; /* full path found count */
+ char ibuf[32]; /* inode buffer */
+ char lsbuf[2048 + MAXPATHLEN + 1]; /* ls(1) system() command */
+ double pct; /* performance percentage */
+ struct stat sb; /* CWD stat(2) results */
+ int ti; /* temporary index */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "h", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h] [-p path]", Pn);
+ PrtMsgX(" -h print help (this panel)", Pn, cleanup, xv);
+ }
+
+#if !defined(DO_TEST)
+/*
+ * If the dialect has disabled the test, echo that result and exit with
+ * a successful return code.
+ */
+ (void) PrtMsgX(LT_DONT_DO_TEST, Pn, cleanup, 0);
+#endif /* !defined(DO_TEST) */
+
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * Get the CWD and form the ls(1) system() command.
+ */
+
+#if defined(USE_GETCWD)
+ em = "getcwd";
+ if (!getcwd(cwd, sizeof(cwd)))
+#else /* ! defined(USE_GETCWD) */
+ em = "getwd";
+ if (!getwd(cwd))
+#endif /* defined(USE_GETCWD) */
+
+ {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! %s() error: %s", em, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+ (void) snprintf(lsbuf, sizeof(lsbuf) - 1, "%s %s > /dev/null 2>&1",
+ LSPATH, cwd);
+/*
+ * Get the CWD stat(2) results.
+ */
+ if (stat(cwd, &sb)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! stat(%s) error: %s", cwd, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+ if ((em = ConvStatDev(&sb.st_dev, &cwddc)))
+ PrtMsgX(em, Pn, cleanup, 1);
+ (void) snprintf(ibuf, sizeof(ibuf) - 1, "%u", (unsigned int)sb.st_ino);
+ ibuf[sizeof(ibuf) - 1] = '\0';
+/*
+ * Loop ATTEMPT_CT times.
+ */
+ for (fpathct = ti = 0; ti < ATTEMPT_CT; ti++) {
+
+ /*
+ * Call ls(1) to list the CWD to /dev/null.
+ */
+ (void) system(lsbuf);
+ /*
+ * Call lsof to look up its own CWD -- i.e., this one.
+ */
+ if ((em = FindLsofCwd(&ff, &cwddc, ibuf))) {
+
+ /*
+ * FindLsofCwd() returned a message. Decode it via ff.
+ */
+ if (ff == -1)
+ PrtMsgX(em, Pn, cleanup, 1);
+ else if (ff == 1) {
+
+ /*
+ * This shouldn't happen. If FindLsof() found lsof's CWD, it
+ * should set ff to one and return NULL.
+ */
+ PrtMsgX("ERROR!!! inconsistent FindLsofCwd() return", Pn,
+ cleanup, 1);
+ }
+ } else if (ff == 1) {
+ fpathct++;
+ }
+ }
+/*
+ * Compute, display, and measure the success percentage.
+ */
+ pct = ((double)fpathct * (double)100.0) / (double)ATTEMPT_CT;
+ PrtMsg((char *)NULL, Pn);
+ (void) printf("%s found: %.2f%%\n", cwd, pct); /* NeXT snpf.c has no
+ * %f support */
+ MsgStat = 1;
+ if (pct < (double)SUCCESS_THRESH) {
+ PrtMsg("ERROR!!! the find rate was too low.", Pn);
+ if (!fpathct) {
+ (void) PrtMsg(
+ "Hint: since the find rate is zero, it may be that this file",
+ Pn);
+ (void) PrtMsg(
+ "system does not fully participate in kernel DNLC processing",
+ Pn);
+ (void) PrtMsg(
+ "-- e.g., NFS file systems often do not, /tmp file systems",
+ Pn);
+ (void) PrtMsg(
+ "sometimes do not, Solaris loopback file systems do not.\n",
+ Pn);
+ (void) PrtMsg(
+ "As a work-around rebuild and test lsof on a file system that",
+ Pn);
+ (void) PrtMsg(
+ "fully participates in kernel DNLC processing.\n",
+ Pn);
+ (void) PrtMsg("See 00FAQ and 00TEST for more information.", Pn);
+ }
+ exit(1);
+ }
+/*
+ * Exit successfully.
+ */
+ (void) PrtMsgX("OK", Pn, cleanup, 0);
+ return(0);
+}
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+}
+
+
+/*
+ * FindLsofCwd() -- find the lsof CWD
+ */
+
+static char *
+FindLsofCwd(ff, cwddc, ibuf)
+ int *ff; /* file-found response receptor */
+ LTdev_t *cwddc; /* CWD device components */
+ char *ibuf; /* CWD inode number in ASCII */
+{
+ char *cp; /* temporary character pointer */
+ char *cem; /* current error message pointer */
+ LTfldo_t *cmdp; /* command pointer */
+ LTdev_t devdc; /* devp->v device components */
+ LTfldo_t *devp; /* device pointer */
+ LTfldo_t *fop; /* field output pointer */
+ LTfldo_t *inop; /* inode number pointer */
+ int nf; /* number of fields */
+ LTfldo_t *nmp; /* name pointer */
+ char *opv[3]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message pointer */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ int ti; /* temporary integer */
+ LTfldo_t *typ; /* file type pointer */
+/*
+ * Check the argument pointers.
+ *
+ * Set the file-found response false.
+ */
+ if (!ff || !cwddc || !ibuf)
+ (void) PrtMsgX("ERROR!!! missing argument to FindFile()",
+ Pn, cleanup, 1);
+ *ff = 0;
+/*
+ * Complete the option vector and start lsof execution.
+ */
+ opv[0] = "-clsof";
+ opv[1] = "-adcwd";
+ opv[2] = (char *)NULL;
+ if ((cem = ExecLsof(opv))) {
+ *ff = -1;
+ return(cem);
+ }
+/*
+ * Read lsof output.
+ */
+ while (!*ff && (fop = RdFrLsof(&nf, &cem))) {
+ if (cem) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ *ff = -1;
+ return(cem);
+ }
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != LsofPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line. Make sure it's for the expected
+ * PID and its type is "cwd".
+ */
+ if (!pids)
+ break;
+ if (strcasecmp(fop->v, "cwd"))
+ break;
+ /*
+ * Scan for device, inode, name, and type fields.
+ */
+ devp = inop = nmp = typ = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_DEVN:
+ devp = fop;
+ break;
+ case LSOF_FID_INODE:
+ inop = fop;
+ break;
+ case LSOF_FID_NAME:
+ nmp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the device, inode, and type of the file.
+ */
+ if (!devp || !inop || !nmp || !typ)
+ break;
+ if (strcasecmp(typ->v, "dir") && strcasecmp(typ->v, "vdir"))
+ break;
+ if ((cem = ConvLsofDev(devp->v, &devdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if ((cwddc->maj != devdc.maj)
+ || (cwddc->min != devdc.min)
+ || (cwddc->unit != devdc.unit)
+ || strcmp(inop->v, ibuf)
+ ) {
+ break;
+ }
+ /*
+ * Check the name for spaces. If it has none, set a file-found
+ * response.
+ */
+ if (!(cp = strchr(nmp->v, ' ')))
+ *ff = 1;
+ else {
+
+ /*
+ * If a parenthesized file system name follows the space in the
+ * file's name, it probably is an NFS file system name and can
+ * be ignored. Accordingly set a file-found response.
+ */
+ if ((*(cp + 1) == '(') && *(cp + 2) && !strchr(cp + 2, ' ')) {
+ if ((cp = strchr(cp + 2, ')')) && !*(cp + 1))
+ *ff = 1;
+ }
+ }
+ }
+ }
+/*
+ * Clean up and return.
+ */
+ (void) StopLsof();
+ if (pem) {
+ *ff = -1;
+ return(pem);
+ }
+ return((char *)NULL);
+}
--- /dev/null
+/*
+ * LTlib.c -- the lsof test library
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+
+
+/*
+ * Pre-defintions that may be changed by a specific dialect
+ */
+
+#define X2DEV_T unsigned int /* cast for result of x2dev() */
+#define XDINDEV 8 /* number of hex digits in an lsof
+ * device field -- should be
+ * 2 X sizeof(X2DEV_T) */
+
+
+#if defined(LT_DIAL_aix)
+/*
+ * AIX-specific items
+ */
+
+#include <sys/sysmacros.h>
+
+# if defined(LT_AIXA) && LT_AIXA>=1
+
+/*
+ * Note: the DEVNO64 and ISDEVNO54 #define's come from <sys/sysmacros.h>, but
+ * only when _KERNEL is #define'd.
+ */
+
+#undef DEVNO64
+#define DEVNO64 0x8000000000000000LL
+#undef ISDEVNO64
+#define ISDEVNO64(d) (((ulong)(d) & DEVNO64) ? 1 : 0)
+
+/*
+ * Define major and minor extraction macros that work on 64 bit AIX
+ * architectures.
+ */
+
+#define major_S(d) (ISDEVNO64(d) ? major64(d) : minor(d & ~SDEV_REMOTE))
+#define minor_S(d) (ISDEVNO64(d) ? (minor64(d) & ~SDEV_REMOTE) : minor(d))
+#undef X2DEV_T
+#define X2DEV_T unsigned long long
+#undef XDINDEV
+#define XDINDEV 16
+#define major_X(dp, em) major_S(x2dev(dp, em))
+#define minor_X(dp, em) minor_S(x2dev(dp, em))
+# endif /* defined(LT_AIXA) && LT_AIXA>=1 */
+
+#endif /* defined(LT_DIAL_aix) */
+
+
+#if defined(LT_DIAL_bsdi)
+/*
+ * BSDI-specific items
+ */
+
+#define minor_S(dev) dv_subunit(dev)
+#define unit_S(dev) dv_unit(dev)
+#define minor_X(dp, em) dv_subunit(x2dev(dp, em))
+#define unit_X(dp, em) dv_unit(x2dev(dp, em))
+#endif /* defined(LT_DIAL_bsdi) */
+
+
+#if defined(LT_DIAL_osr)
+/*
+ * OpenUNIX-specific items
+ */
+
+#include <sys/sysmacros.h>
+#endif /* defined(LT_DIAL_osr) */
+
+
+#if defined(LT_DIAL_ou)
+/*
+ * OpenUNIX-specific items
+ */
+
+#include <sys/mkdev.h>
+#endif /* defined(LT_DIAL_ou) */
+
+
+#if defined(LT_DIAL_solaris)
+/*
+ * Solaris-specific items
+ */
+
+#include <sys/sysmacros.h>
+
+
+/*
+ * Define maximum major device number in a stat(2) dev_t
+ */
+
+# if LT_VERS>=20501
+#define LT_MJX L_MAXMAJ /* Get maximum major device number from
+ * <sys/sysmacros.h>. */
+# else /* LT_VERS<20501 */
+#define LT_MJX 0x3fff /* Avoid <sys/sysmacros.h> when
+ * Solaris < 2.5.1. */
+# endif /* LT_VERS>=20501 */
+
+#define major_S(dev) ((int)((dev >> L_BITSMINOR) & LT_MJX))
+#define minor_S(dev) ((int)(dev & L_MAXMIN))
+
+# if defined(LT_K64)
+
+/*
+ * Solaris 64 bit kernel
+ */
+
+#undef X2DEV_T
+#define X2DEV_T unsigned long long
+#undef XDINDEV
+#define XDINDEV 16
+
+#define major_X(dp, em) ((int)((x2dev(dp, em) >> 32) & 0xffffffff))
+#define minor_X(dp, em) ((int)(x2dev(dp, em) & 0xffffffff))
+# else /* !defined(LT_K64) */
+
+/*
+ * Solaris 32 bit kernel
+ */
+
+#define major_X(dp, em) ((int)((x2dev(dp, em) >> L_BITSMINOR) & LT_MJX))
+#define minor_X(dp, em) ((int)(x2dev(dp, em) & L_MAXMIN))
+# endif /* LT_K64 */
+#endif /* defined(LT_DIAL_solaris) */
+
+
+#if defined(LT_DIAL_uw)
+/*
+ * UnixWare-specific items
+ */
+
+#include <sys/mkdev.h>
+#endif /* defined(LT_DIAL_uw) */
+
+
+/*
+ * Global variables
+ */
+
+int LsofFd = -1; /* lsof pipe FD */
+FILE *LsofFs = (FILE *)NULL; /* stream for lsof pipe FD */
+char *LsofPath = (char *)NULL; /* path to lsof executable */
+pid_t LsofPid = (pid_t)0; /* PID of lsof child process */
+int LTopt_h = 0; /* "-h" option's switch value */
+char *LTopt_p = (char *)NULL; /* "-p path" option's path value */
+int MsgStat = 0; /* message status: 1 means prefix needs
+ * to be issued */
+
+
+/*
+ * Local static variables
+ */
+
+static int Afo = 0; /* Fo[] structures allocated */
+static char *GOv = (char *)NULL; /* option `:' value pointer */
+static int GOx1 = 1; /* first opt[][] index */
+static int GOx2 = 0; /* second opt[][] index */
+static LTfldo_t *Fo = (LTfldo_t *)NULL; /* allocated LTfldo_t structures */
+static int Ufo = 0; /* Fo[] structures used */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void closepipe,(void));
+_PROTOTYPE(static void getlsofpath,(void));
+_PROTOTYPE(static int GetOpt,(int ct, char *opt[], char *rules, char **em,
+ char *pn));
+_PROTOTYPE(static X2DEV_T x2dev,(char *x, char **em));
+
+
+/*
+ * Default major, minor, and unit macros.
+ */
+
+#if !defined(major_S)
+#define major_S major
+#endif /* defined(major_S) */
+
+#if !defined(minor_S)
+#define minor_S minor
+#endif /* defined(minor_S) */
+
+#if !defined(unit_S)
+#define unit_S(x) 0
+#endif /* defined(unit_S) */
+
+#if !defined(major_X)
+#define major_X(dp, em) major(x2dev(dp, em))
+#endif /* defined(major_X) */
+
+#if !defined(minor_X)
+#define minor_X(dp, em) minor(x2dev(dp, em))
+#endif /* defined(minor_X) */
+
+#if !defined(unit_X)
+#define unit_X(dp, em) 0
+#endif /* defined(unit_X) */
+
+
+/*
+ * CanRdKmem() -- can lsof read kernel memory devices?
+ */
+
+char *
+CanRdKmem()
+{
+
+#if defined(LT_KMEM)
+ char buf[2048]; /* temporary buffer */
+ char *dn; /* memory device name */
+ char *em; /* error message pointer */
+ int fd; /* temporary file descriptor */
+ struct stat sb; /* memory device stat(2) buffer */
+ int ti; /* temporary integer */
+/*
+ * Get the lsof path. If it is not the default, check no further.
+ */
+ (void) getlsofpath();
+ if (!strcmp(LsofPath, LT_DEF_LSOF_PATH))
+ return((char *)NULL);
+/*
+ * Check /dev/kmem access.
+ */
+ dn = "/dev/kmem";
+ if (stat(dn, &sb)) {
+ em = "stat";
+
+kmem_error:
+
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't %s(%s): %s\n", em, dn, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ }
+ if ((fd = open(dn, O_RDONLY, 0)) < 0) {
+ em = "open";
+ goto kmem_error;
+ }
+ (void) close(fd);
+/*
+ * Check /dev/mem access.
+ */
+ dn = "/dev/mem";
+ if (stat(dn, &sb)) {
+
+ /*
+ * If /dev/mem can't be found, ignore the error.
+ */
+ return((char *)NULL);
+ }
+ if ((fd = open(dn, O_RDONLY, 0)) < 0) {
+ em = "open";
+ goto kmem_error;
+ }
+ (void) close(fd);
+#endif /* defined(LT_KMEM) */
+
+ return((char *)NULL);
+}
+
+
+/*
+ * closepipe() -- close pipe from lsof
+ */
+
+static void
+closepipe()
+{
+ if (LsofFd >= 0) {
+
+ /*
+ * A pipe from lsof is open. Close it and the associated stream.
+ */
+ if (LsofFs) {
+ (void) fclose(LsofFs);
+ LsofFs = (FILE *)NULL;
+ }
+ (void) close(LsofFd);
+ LsofFd = -1;
+ }
+}
+
+
+/*
+ * ConvLsofDev() -- convert lsof device string
+ *
+ * Note: this function is dialect-specific.
+ */
+
+char *
+ConvLsofDev(dev, ldev)
+ char *dev; /* lsof device string -- the value to the
+ * LSOF_FID_DEVN field of a LSOF_FID_FD block
+ * (see lsof_fields.h) */
+ LTdev_t *ldev; /* results are returned to this structure */
+{
+ char *dp; /* device pointer */
+ char *em; /* error message pointer */
+ int tlen; /* temporary length */
+/*
+ * Check function arguments.
+ *
+ * Establish values for decoding the device string.
+ */
+ if (!dev)
+ return("ERROR!!! no ConvLsofDev() device");
+ if (!ldev)
+ return("ERROR!!! no ConvLsofDev() result pointer");
+ if (strncmp(dev, "0x", 2))
+ return("ERROR!!! no leading 0x in ConvLsofDev() device");
+ dp = dev + 2;
+ if (((tlen = (int)strlen(dp)) < 1) || (tlen > XDINDEV))
+ return("ERROR!!! bad ConvLsofDev() device length");
+/*
+ * Use the pre-defined *_X() macros to do the decomposition.
+ */
+ ldev->maj = (unsigned int)major_X(dp, &em);
+ if (em)
+ return(em);
+ ldev->min = (unsigned int)minor_X(dp, &em);
+ if (em)
+ return(em);
+ ldev->unit = (unsigned int)unit_X(dp, &em);
+ return(em);
+}
+
+
+/*
+ * ConvStatDev() -- convert stat(2) device number
+ *
+ * Note: this function is dialect-specific.
+ */
+
+char *
+ConvStatDev(dev, ldev)
+ dev_t *dev; /* device number to be converted */
+ LTdev_t *ldev; /* results are returned to this structure */
+{
+
+/*
+ * Check function arguments.
+ */
+ if (!dev)
+ return("ERROR!!! no ConvStatDev() device");
+ if (!ldev)
+ return("ERROR!!! no ConvStatDev() result pointer");
+/*
+ * Use the pre-defined *_S() macros to do the decomposition.
+ */
+ ldev->maj = (unsigned int)major_S(*dev);
+ ldev->min = (unsigned int)minor_S(*dev);
+ ldev->unit = (unsigned int)unit_S(*dev);
+ return((char *)NULL);
+}
+
+
+/*
+ * ExecLsof() -- execute lsof with full field output and a NUL field terminator
+ * in a child process
+ */
+
+char *
+ExecLsof(opt)
+ char **opt; /* lsof options -- a pointer to an
+ * array of character pointers,
+ * terminated by a NULL pointer */
+{
+ static char **av = (char **)NULL; /* lsof argument vector, dynamically
+ * allocated */
+ static int ava = 0; /* **av entries allocated */
+ char buf[2048]; /* temporary buffer */
+ char *em; /* error message pointer */
+ int fd; /* temporary file descriptor */
+ int optc; /* option count */
+ int nf; /* number of files */
+ int p[2]; /* pipe FDs */
+ char **tcpp; /* temporary character pointer
+ * pointer */
+ int ti; /* temporary integer */
+ int tlen; /* temporary length */
+ pid_t tpid; /* temporary PID holder */
+/*
+ * It's an error if lsof is already in execution or if no lsof options
+ * were supplied.
+ */
+ (void) getlsofpath();
+ if (LsofPid)
+ return("ERROR!!! ExecLsof() says lsof is already in execution");
+ if (!opt)
+ return("ERROR!!! no ExecLsof() option list");
+ for (optc = 0, tcpp = opt; *tcpp; optc++, tcpp++)
+ ;
+/*
+ * Make sure lsof is executable.
+ */
+ if ((em = IsLsofExec()))
+ return(em);
+/*
+ * Open a pipe through which lsof can return output.
+ */
+ if (pipe(p)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't open pipe: %s", strerror(errno));
+ return(MkStrCpy(buf, &ti));
+ }
+/*
+ * Allocate and build an argument vector. The first entry will be set
+ * to "lsof", the second to "-wFr", and the third to "-F0". Additional
+ * entries will be set as supplied by the caller.
+ */
+ if ((optc + 4) > ava) {
+ tlen = (int)(sizeof(char *) * (optc + 4));
+ if (!av)
+ av = (char **)malloc(tlen);
+ else
+ av = (char **)realloc((void *)av, tlen);
+ if (!av) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "LTlib: ExecLsof() can't allocat pointers for %d arguments",
+ optc + 4);
+ return(MkStrCpy(buf, &ti));
+ }
+ ava = optc + 4;
+ }
+ for (ti = 0, tcpp = opt; ti < (optc + 3); ti++) {
+ switch(ti) {
+ case 0:
+ av[ti] = "lsof";
+ break;
+ case 1:
+ av[ti] = "-wFr";
+ break;
+ case 2:
+ av[ti] = "-F0";
+ break;
+ default:
+ av[ti] = *tcpp;
+ tcpp++;
+ }
+ }
+ av[ti] = (char *)NULL;
+/*
+ * Fork a child process to run lsof.
+ */
+ switch((tpid = fork())) {
+ case (pid_t)0:
+
+ /*
+ * This is the child process.
+ *
+ * First close all file descriptors except the output side of the pipe.
+ *
+ * Make the output side of the pipe STDOUT and STDERR.
+ */
+ for (fd = 0, nf = getdtablesize(); fd < nf; fd++) {
+ if (fd == p[1])
+ continue;
+ (void) close(fd);
+ }
+ if (p[1] != 1)
+ (void) dup2(p[1], 1);
+ if (p[1] != 2)
+ (void) dup2(p[1], 2);
+ if ((p[1] != 1) && (p[1] != 2))
+ (void) close(p[1]);
+ /*
+ * Execute lsof.
+ */
+ (void) execv(LsofPath, av);
+ _exit(0); /* (Shouldn't get here.) */
+ case (pid_t)-1:
+
+ /*
+ * A fork error occurred. Form and return a message.
+ */
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! ExecLsof() can't fork: %s", strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ default:
+
+ /*
+ * This is the parent.
+ *
+ * Save the lsof child PID.
+ *
+ * Close the output side of the pipe.
+ *
+ * Save the input side of the pipe as LsofFd; open a stream for it.
+ */
+ LsofPid = tpid;
+ (void) close(p[1]);
+ LsofFd = p[0];
+ if (!(LsofFs = fdopen(LsofFd, "r")))
+ return("ERROR!!! ExecLsof() can't open stream to lsof output FD");
+ }
+/*
+ * Wait a bit for lsof to start and put something in its pipe, then return
+ * an "All is well." response.
+ */
+ sleep(1);
+ return((char *)NULL);
+}
+
+
+/*
+ * getlsofpath() -- get lsof path, either from LT_LSOF_PATH in the environment
+ * or from LT_DEF_LSOF_PATH
+ */
+
+static void
+getlsofpath()
+{
+ char *tcp; /* temporary character pointer */
+ int ti; /* temporary integer */
+
+ if (LsofPath)
+ return;
+ if ((tcp = getenv("LT_LSOF_PATH")))
+ LsofPath = MkStrCpy(tcp, &ti);
+ else
+ LsofPath = LT_DEF_LSOF_PATH;
+}
+
+
+/*
+ * GetOpt() -- Local get option
+ *
+ * Borrowed from lsof's main.c source file.
+ *
+ * Liberally adapted from the public domain AT&T getopt() source,
+ * distributed at the 1985 UNIFORM conference in Dallas
+ *
+ * The modifications allow `?' to be an option character and allow
+ * the caller to decide that an option that may be followed by a
+ * value doesn't have one -- e.g., has a default instead.
+ */
+
+static int
+GetOpt(ct, opt, rules, em, pn)
+ int ct; /* option count */
+ char *opt[]; /* options */
+ char *rules; /* option rules */
+ char **em; /* error message return */
+ char *pn;
+{
+ register int c; /* character value */
+ register char *cp = (char *)NULL; /* character pointer */
+ char embf[2048]; /* error message buffer */
+ int tlen; /* temporary message length from
+ * MkStrCpy() */
+
+ *em = (char *)NULL;
+ if (GOx2 == 0) {
+
+ /*
+ * Move to a new entry of the option array.
+ *
+ * EOF if:
+ *
+ * Option list has been exhausted;
+ * Next option doesn't start with `-' or `+';
+ * Next option has nothing but `-' or `+';
+ * Next option is ``--'' or ``++''.
+ */
+ if (GOx1 >= ct
+ || (opt[GOx1][0] != '-' && opt[GOx1][0] != '+')
+ || !opt[GOx1][1])
+ return(EOF);
+ if (strcmp(opt[GOx1], "--") == 0 || strcmp(opt[GOx1], "++") == 0) {
+ GOx1++;
+ return(EOF);
+ }
+ GOx2 = 1;
+ }
+/*
+ * Flag `:' option character as an error.
+ *
+ * Check for a rule on this option character.
+ */
+ if ((c = opt[GOx1][GOx2]) == ':') {
+ (void) snprintf(embf, sizeof(embf) - 1,
+ "ERROR!!! colon is an illegal option character.");
+ embf[sizeof(embf) - 1] = '\0';
+ *em = MkStrCpy(embf, &tlen);
+ } else if (!(cp = strchr(rules, c))) {
+ (void) snprintf(embf, sizeof(embf) - 1,
+ "ERROR!!! illegal option character: %c", c);
+ embf[sizeof(embf) - 1] = '\0';
+ *em = MkStrCpy(embf, &tlen);
+ }
+ if (*em) {
+
+ /*
+ * An error was detected.
+ *
+ * Advance to the next option character.
+ *
+ * Return the character causing the error.
+ */
+ if (opt[GOx1][++GOx2] == '\0') {
+ GOx1++;
+ GOx2 = 0;
+ }
+ return(c);
+ }
+ if (*(cp + 1) == ':') {
+
+ /*
+ * The option may have a following value. The caller decides if it does.
+ *
+ * Don't indicate that an option of ``--'' is a possible value.
+ *
+ * Finally, on the assumption that the caller will decide that the possible
+ * value belongs to the option, position to the option following the
+ * possible value, so that the next call to GetOpt() will find it.
+ */
+ if(opt[GOx1][GOx2 + 1] != '\0') {
+ GOv = &opt[GOx1++][GOx2];
+ } else if (++GOx1 >= ct)
+ GOv = (char *)NULL;
+ else {
+ GOv = opt[GOx1];
+ if (strcmp(GOv, "--") == 0)
+ GOv = (char *)NULL;
+ else
+ GOx1++;
+ }
+ GOx2 = 0;
+ } else {
+
+ /*
+ * The option character stands alone with no following value.
+ *
+ * Advance to the next option character.
+ */
+ if (opt[GOx1][++GOx2] == '\0') {
+ GOx2 = 0;
+ GOx1++;
+ }
+ GOv = (char *)NULL;
+ }
+/*
+ * Return the option character.
+ */
+ return(c);
+}
+
+
+/*
+ * IsLsofExec() -- see if lsof is executable
+ */
+
+char *
+IsLsofExec()
+{
+ char buf[2048]; /* temporary buffer */
+ int len; /* temporary length */
+
+ (void) getlsofpath();
+ if (access(LsofPath, X_OK) < 0) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't execute %s: %s", LsofPath, strerror(errno));
+ return(MkStrCpy(buf, &len));
+ }
+ return((char *)NULL);
+}
+
+
+/*
+ * LTlibClean() -- clean up LTlib resource accesses
+ */
+
+void
+LTlibClean()
+{
+ (void) StopLsof();
+}
+
+
+/*
+ * MkStrCpy() -- make string copy
+ */
+
+char *
+MkStrCpy(src, len)
+ char *src; /* string source to copy */
+ int *len; /* returned length allocation */
+{
+ char *rp; /* return pointer */
+ int srclen; /* source string length */
+
+ if (!src) {
+ (void) fprintf(stderr, "ERROR!!! no string supplied to MkStrCpy()\n");
+ exit(1);
+ }
+ srclen = (int)strlen(src);
+ *len = srclen++;
+ if (!(rp = (char *)malloc(srclen))) {
+ (void) fprintf(stderr, "ERROR!!! MkStrCpy() -- no malloc() space");
+ exit(1);
+ }
+ (void) strcpy(rp, src);
+ return(rp);
+}
+
+
+/*
+ * PrtMsg() -- print message
+ */
+
+void
+PrtMsg(mp, pn)
+ char *mp; /* message pointer -- may be NULL to
+ * trigger space prefix initialization
+ */
+ char *pn; /* program name */
+{
+ static int pfxlen = -1; /* prefix length, based on program */
+ /* name -- computed on first call
+ * when pfxlen == -1 */
+ static char *pfx = (char *)NULL; /* prefix (spaces) */
+ int ti; /* temporary index */
+
+ if (pfxlen == -1) {
+
+ /*
+ * This is the first call. Compute the prefix length and build the
+ * prefix.
+ */
+ if (!pn)
+ pfxlen = 0;
+ else
+ pfxlen = (int)(strlen(pn));
+ pfxlen += (int)strlen(" ... ");
+ if (!(pfx = (char *)malloc(pfxlen + 1))) {
+ (void) printf( "ERROR!!! not enough space for %d space prefix\n",
+ pfxlen);
+ exit(1);
+ }
+ for (ti = 0; ti < pfxlen; ti++) {
+ pfx[ti] = ' ';
+ }
+ pfx[pfxlen] = '\0';
+ MsgStat = 0;
+ }
+/*
+ * Process the message.
+ */
+ if (MsgStat)
+ (void) printf("%s", pfx);
+ if (mp && *mp) {
+ (void) printf("%s\n", mp);
+ MsgStat = 1;
+ }
+}
+
+
+/*
+ * PrtMsgX() -- print message and exit
+ */
+
+void
+PrtMsgX(mp, pn, f, xv)
+ char *mp; /* message pointer */
+ char *pn; /* program name */
+ void (*f)(); /* clean-up function pointer */
+ int xv; /* exit value */
+{
+ if (mp)
+ PrtMsg(mp, pn);
+ if (f)
+ (void) (*f)();
+ (void) LTlibClean();
+ exit(xv);
+}
+
+
+/*
+ * RdFrLsof() -- read from lsof
+ */
+
+LTfldo_t *
+RdFrLsof(nf, em)
+ int *nf; /* number of fields receiver */
+ char **em; /* error message pointer receiver */
+{
+ char buf[2048]; /* temporary buffer */
+ int bufl = (int)sizeof(buf); /* size of buf[] */
+ char *blim = &buf[bufl - 1]; /* buf[] limit (last character
+ * address) */
+ char *fsp; /* field start pointer */
+ char *tcp; /* temporary character pointer */
+ LTfldo_t *tfop; /* temporary field output pointer */
+ int ti; /* temporary index */
+ int tlen; /* remporary length */
+ char *vp; /* value character pointer */
+/*
+ * Check for errors.
+ */
+ if (!em)
+ return((LTfldo_t *)NULL);
+ if (!nf) {
+ *em = "ERROR!!! RdFrLsof() not given a count return pointer";
+ return((LTfldo_t *)NULL);
+ }
+ *em = (char *)NULL;
+ *nf = 0;
+/*
+ * If fields are in use, release their resources.
+ */
+ for (ti = 0, tfop = Fo; (ti < Ufo); ti++, tfop++) {
+ if (tfop->v)
+ (void) free((void *)tfop->v);
+ }
+ Ufo = 0;
+/*
+ * Read a line from lsof.
+ */
+ if (!fgets(buf, bufl - 2, LsofFs)) {
+
+ /*
+ * An lsof pipe EOF has been reached. Indicate that with a NULL
+ * pointer return, coupled with a NULL error message return pointer
+ * (set above), and a field count of zero (set above).
+ */
+ return((LTfldo_t *)NULL);
+ }
+/*
+ * Parse the lsof line, allocating field output structures as appropriate.
+ *
+ * It is expected that fields will end in a NUL ('\0') or a NL ('\0') and
+ * that a NL ends all fields in the lsof line.
+ */
+ for (tcp = buf, Ufo = 0; (*tcp != '\n') && (tcp < blim); tcp++) {
+
+ /*
+ * Start a new field. The first character is the LSOF_FID_*.
+ *
+ * First allocate an LTfldo_t structure.
+ */
+ if (Ufo >= Afo) {
+
+ /*
+ * More LTfldo_t space is required.
+ */
+ Afo += LT_FLDO_ALLOC;
+ tlen = (int)(Afo * sizeof(LTfldo_t));
+ if (Fo)
+ Fo = (LTfldo_t *)realloc(Fo, tlen);
+ else
+ Fo = (LTfldo_t *)malloc(tlen);
+ if (!Fo) {
+
+ /*
+ * A serious error has occurred; no LTfldo_t space is available.
+ */
+ (void) snprintf(buf, bufl,
+ "ERROR!!! RdFrLsof() can't allocate %d pointer bytes",
+ tlen);
+ *em = MkStrCpy(buf, &ti);
+ *nf = -1;
+ return((LTfldo_t *)NULL);
+ }
+ }
+ tfop = Fo + Ufo;
+ tfop->v = (char *)NULL;
+ Ufo++;
+ /*
+ * Save the LSOF_FID_* character. Then compute the field value length,
+ * and make a copy of it.
+ */
+ tfop->ft = *tcp++;
+ fsp = tcp;
+ tlen = 0;
+ while (*tcp && (*tcp != '\n') && (tcp < blim)) {
+ tcp++;
+ tlen++;
+ }
+ if (!(vp = (char *)malloc(tlen + 1))) {
+
+ /*
+ * A serious error has occurred; there's no space for the field value.
+ */
+ (void) snprintf(buf, bufl,
+ "ERROR!!! RdFrLsof() can't allocate %d field bytes", tlen + 1);
+ *em = MkStrCpy(buf, &ti);
+ *nf = -1;
+ return((LTfldo_t *)NULL);
+ }
+ (void) memcpy((void *)vp, (void *)fsp, tlen);
+ vp[tlen] = '\0';
+ tfop->v = vp;
+ if (*tcp == '\n')
+ break;
+ if (tcp >= blim) {
+
+ /*
+ * The lsof line has no NL terminator; that's an error.
+ */
+ *em = "ERROR!!! RdFrLsof() didn't find a NL";
+ *nf = -1;
+ return((LTfldo_t *)NULL);
+ }
+ }
+/*
+ * The end of the lsof line has been reached. If no fields were assembled,
+ * return an error indicate. Otherwise return the fields and their count.
+ */
+ if (!Ufo) {
+ *em = "ERROR!!! RdFrLsof() read an empty lsof line";
+ *nf = -1;
+ return((LTfldo_t *)NULL);
+ }
+ *nf = Ufo;
+ *em = (char *)NULL;
+ return(Fo);
+}
+
+
+/*
+ * ScanArg() -- scan arguments
+ */
+
+int
+ScanArg(ac, av, opt, pn)
+ int ac; /* argument count */
+ char *av[]; /* argument pointers */
+ char *opt; /* option string */
+ char *pn; /* program name */
+{
+ char *em; /* pointer to error message returned by
+ * GetOpt() */
+ char embf[2048]; /* error message buffer */
+ int rv = 0; /* return value */
+ int tc; /* temporary character value */
+/*
+ * Preset possible argument values.
+ */
+ LTopt_h = 0;
+ if (LTopt_p) {
+ (void) free((void *)LTopt_p);
+ LTopt_p = (char *)NULL;
+ }
+/*
+ * Process the options according to the supplied option string.
+ */
+ while ((tc = GetOpt(ac, av, opt, &em, pn)) != EOF) {
+ if (em) {
+ rv = 1;
+ PrtMsg(em, pn);
+ continue;
+ }
+ switch (tc) {
+ case 'h':
+ LTopt_h = 1;
+ break;
+ case 'p':
+ if (!GOv || *GOv == '-' || *GOv == '+') {
+ rv = 1;
+ (void) PrtMsg("ERROR!!! -p not followed by a path", pn);
+ } else
+ LTopt_p = GOv;
+ break;
+ default:
+ rv = 1;
+ (void) snprintf(embf, sizeof(embf) - 1,
+ "ERROR!!! unknown option: %c", tc);
+ PrtMsg(embf, pn);
+ }
+ }
+ for (; GOx1 < ac; GOx1++) {
+
+ /*
+ * Report extraneous arguments.
+ */
+ rv = 1;
+ (void) snprintf(embf, sizeof(embf) - 1,
+ "ERROR!!! extraneous option: \"%s\"", av[GOx1]);
+ PrtMsg(embf, pn);
+ }
+ return(rv);
+}
+
+
+/*
+ * StopLsof() -- stop a running lsof process and close the pipe from it
+ */
+
+void
+StopLsof()
+{
+ pid_t pid;
+
+ if (LsofPid) {
+
+ /*
+ * An lsof child process may be active. Wait for (or kill) it.
+ */
+ pid = wait3(NULL, WNOHANG, NULL);
+ if (pid != LsofPid) {
+ (void) kill(LsofPid, SIGKILL);
+ sleep(2);
+ pid = wait3(NULL, WNOHANG, NULL);
+ }
+ LsofPid = (pid_t)0;
+ }
+ (void) closepipe();
+}
+
+
+/*
+ * x2dev() -- convert hex string to device number
+ */
+
+static X2DEV_T
+x2dev(x, em)
+ char *x; /* hex string */
+ char **em; /* error message receiver */
+{
+ char buf[2048]; /* temporary message buffer */
+ int c; /* character holder */
+ X2DEV_T dev; /* device number result */
+ char *wx; /* working hex string pointer */
+ int xl; /* hex string length */
+
+ if (!x || !*x) {
+ *em = "ERROR!!! no hex string supplied to x2dev()";
+ return(0);
+ }
+ wx = strncasecmp(x, "0x", 2) ? x : (x + 2);
+ if (((xl = (int)strlen(wx)) < 1) || (xl > XDINDEV)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! x2dev(\"%s\") bad length: %d", x, xl + 2);
+ buf[sizeof(buf) - 1] = '\0';
+ *em = MkStrCpy(buf, &c);
+ return(0);
+ }
+/*
+ * Assemble the device number result from the hex string.
+ */
+ for (dev = (X2DEV_T)0; *wx; wx++) {
+ if (isdigit((unsigned char)*wx)) {
+ dev = (dev << 4) | (unsigned int)(((int)*wx - (int)'0') & 0xf);
+ continue;
+ }
+ c = (int) tolower((unsigned char)*wx);
+ if ((c >= (int)'a') && (c <= (int)'f')) {
+ dev = (dev << 4) | (unsigned int)((c - 'a' + 10) & 0xf);
+ continue;
+ }
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! x2dev(\"%s\") non-hex character: %c", x, c);
+ *em = MkStrCpy(buf, &c);
+ }
+/*
+ * Return result and no error indication.
+ */
+ *em = (char *)NULL;
+ return(dev);
+}
--- /dev/null
+/*
+ * LTlock.c -- Lsof Test locking tests
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+
+#if defined(LT_DIAL_aix)
+/*
+ * AIX-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_aix) */
+
+
+#if defined(LT_DIAL_bsdi)
+/*
+ * BSDI-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_bsdi) */
+
+
+#if defined(LT_DIAL_darwin)
+/*
+ * Darwin-specific items
+ */
+
+/*
+ * There is no Darwin USE_* definition, because lock support in lsof for
+ * Darwin is inadequate for this test.
+ */
+#endif /* defined(LT_DIAL_darwin) */
+
+
+#if defined(LT_DIAL_du)
+/*
+ * DEC_OSF/1|Digital_UNIX|Tru64_UNIX-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_du) */
+
+
+#if defined(LT_DIAL_freebsd)
+/*
+ * FreeBSD-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_freebsd) */
+
+
+#if defined(LT_DIAL_linux)
+/*
+ * Linux-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_linux) */
+
+
+#if defined(LT_DIAL_netbsd)
+/*
+ * NetBSD-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_netbsd) */
+
+
+#if defined(LT_DIAL_openbsd)
+/*
+ * OpenBSD-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_openbsd) */
+
+
+#if defined(LT_DIAL_hpux)
+/*
+ * HP-UX-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_hpux) */
+
+
+#if defined(LT_DIAL_ns)
+/*
+ * NEXTSTEP-specific items
+ */
+
+#define USE_FLOCK
+#endif /* defined(LT_DIAL_ns) */
+
+
+#if defined(LT_DIAL_osr)
+/*
+ * OSR-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_osr) */
+
+
+#if defined(LT_DIAL_ou)
+/*
+ * OpenUNIX-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_ou) */
+
+
+#if defined(LT_DIAL_openbsd)
+/*
+ * OpenBSD-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_openbsd) */
+
+
+#if defined(LT_DIAL_solaris)
+/*
+ * Solaris-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(solaris) */
+
+
+#if defined(LT_DIAL_uw)
+/*
+ * UnixWare-specific items
+ */
+
+#define USE_FCNTL
+#endif /* defined(LT_DIAL_uw) */
+
+
+#if !defined(USE_FLOCK) && !defined(USE_FCNTL)
+/*
+ * Here begins the version of this program for dialects that don't support
+ * flock() or fcntl() locking.
+ */
+
+
+/*
+ * Main program for dialects that don't support flock() of fcntl() locking.
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char *pn; /* program name */
+/*
+ * Get program name and issue error message.
+ */
+ if ((pn = (char *)strrchr(argv[0], '/')))
+ pn++;
+ else
+ pn = argv[0];
+ (void) printf("%s ... %s\n", pn, LT_DONT_DO_TEST);
+ return(0);
+}
+#else /* defined(USE_FLOCK) || defined(USE_FCNTL) */
+
+
+/*
+ * Local definitions
+ */
+
+#define FULL_EX_LOCK 0 /* get a full file exclusive lock */
+#define FULL_SH_LOCK 1 /* get a full file shared lock */
+#define PART_EX_LOCK 2 /* get a partial file exclusive lock */
+#define PART_SH_LOCK 3 /* get a partial file shared lock */
+
+
+/*
+ * Globals
+ */
+
+int Fd = -1; /* test file descriptor; open if >= 0 */
+pid_t MyPid = (pid_t)0; /* PID of this process */
+char *Path = (char *)NULL; /* test file path; none if NULL */
+char *Pn = (char *)NULL; /* program name */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static char *lkfile,(int ty));
+_PROTOTYPE(static char *tstwlsof,(char *opt, char *xlk));
+_PROTOTYPE(static char *unlkfile,(int ty));
+
+
+/*
+ * Main program for dialects that support locking tests.
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ char *em; /* error message pointer */
+ int ti; /* temporary index */
+ char *tcp; /* temporary character pointer */
+ int tlen; /* temporary length -- e.g., as
+ * returned by MkStrCpy() */
+ char *tstR = (char *)NULL; /* "R" lock test result */
+ char *tstr = (char *)NULL; /* "r" lock test result */
+ char *tstW = (char *)NULL; /* "W" lock test result */
+ char *tstw = (char *)NULL; /* "w" lock test result */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ (void) PrtMsg((char *)NULL, Pn);
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "hp:", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg ("usage: [-h] [-p path]", Pn);
+ (void) PrtMsg (" -h print help (this panel)", Pn);
+ (void) PrtMsgX(" -p path define test file path", Pn, cleanup,
+ xv);
+ }
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * If a path was supplied in an "-p path" option, use it. Otherwise construct
+ * a path in the CWD.
+ */
+ if (!(Path = LTopt_p)) {
+ (void) snprintf(buf, sizeof(buf), "./config.LTlock%ld",
+ (long)MyPid);
+ buf[sizeof(buf) - 1] = '\0';
+ Path = MkStrCpy(buf, &tlen);
+ }
+/*
+ * Fill buffer for writing to the test file.
+ */
+ for (ti = 0; ti < sizeof(buf); ti++) {
+ buf[ti] = (char)(ti & 0xff);
+ }
+/*
+ * Open a new test file at the specified path.
+ */
+ (void) unlink(Path);
+ if ((Fd = open(Path, O_RDWR|O_CREAT, 0600)) < 0) {
+ (void) fprintf(stderr, "ERROR!!! can't open %s\n", Path);
+
+print_file_error:
+
+ MsgStat = 1;
+ (void) snprintf(buf, sizeof(buf) - 1, " Errno %d: %s",
+ errno, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Write a buffer load at the beginning of the file.
+ */
+ if (write(Fd, buf, sizeof(buf)) != sizeof(buf)) {
+ (void) fprintf(stderr,
+ "ERROR!!! can't write %d bytes to the beginning of %s\n",
+ (int)sizeof(buf), Path);
+ goto print_file_error;
+ }
+/*
+ * Fsync() the file.
+ */
+ if (fsync(Fd)) {
+ (void) fprintf(stderr, "ERROR!!! can't fsync %s\n", Path);
+ goto print_file_error;
+ }
+/*
+ * Quit (with a hint) if the test file is on an NFS file system.
+ */
+ if (!tstwlsof("-wNa", " ")) {
+ (void) printf("ERROR!!! %s is NFS-mounted.\n", Path);
+ MsgStat = 1;
+ (void) PrtMsg ("Lsof can't report lock information on files that", Pn);
+ (void) PrtMsg ("are located on file systems mounted from a remote", Pn);
+ (void) PrtMsg ("NFS server.\n", Pn);
+ (void) PrtMsg ("Hint: try using \"-p path\" to supply a path in a", Pn);
+ (void) PrtMsg ("non-NFS file system.\n", Pn);
+ (void) PrtMsgX("See 00FAQ and 00TEST for more information.", Pn,
+ cleanup, 1);
+ }
+/*
+ * Get an exclusive lock on the entire file and test it with lsof.
+ */
+ if ((em = lkfile(FULL_EX_LOCK)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((tstW = tstwlsof("-w", "W")))
+ (void) PrtMsg(tstW, Pn);
+/*
+ * Get a shared lock on the entire file and test it with lsof.
+ */
+ if ((em = unlkfile(FULL_EX_LOCK)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = lkfile(FULL_SH_LOCK)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((tstR = tstwlsof("-w", "R")))
+ (void) PrtMsg(tstR, Pn);
+
+# if defined(USE_FLOCK)
+/*
+ * If using flock(), skip the byte lock tests.
+ */
+ tstr = tstw = (char *)NULL;
+# endif /* defined(USE_FLOCK) */
+
+# if defined(USE_FCNTL)
+/*
+ * If using fcntl(), do exclusive and shared byte lock tests,
+ */
+ if ((em = unlkfile(FULL_SH_LOCK)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = lkfile(PART_EX_LOCK)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((tstw = tstwlsof("-w", "w")))
+ (void) PrtMsg(tstw, Pn);
+ if ((em = unlkfile(PART_EX_LOCK)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = lkfile(PART_SH_LOCK)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((tstr = tstwlsof("-w", "r")))
+ (void) PrtMsg(tstr, Pn);
+# endif /* defined(USE_FCNTL) */
+
+/*
+ * Compute exit value and exit.
+ */
+ if (tstr || tstR || tstw || tstW) {
+ tcp = (char *)NULL;
+ xv = 1;
+ } else {
+ tcp = "OK";
+ xv = 0;
+ }
+ (void) PrtMsgX(tcp, Pn, cleanup, xv);
+ return(0);
+}
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+ if (Fd >= 0) {
+ (void) close(Fd);
+ Fd = -1;
+ if (Path) {
+ (void) unlink(Path);
+ Path = (char *)NULL;
+ }
+ }
+}
+
+
+/*
+ * lkfile() -- lock the test file
+ */
+
+static char *
+lkfile(ty)
+ int ty; /* a *_*_LOCK requested */
+{
+ char buf[2048]; /* temporary buffer */
+ int ti; /* temporary integer */
+
+# if defined(USE_FLOCK)
+ int flf; /* flock() function */
+# endif /* defined(USE_FLOCK) */
+
+# if defined(USE_FCNTL)
+ struct flock fl; /* flock control structure */
+/*
+ * Check fcntl() lock request.
+ */
+ (void) memset((void *)&fl, 0, sizeof(fl));
+ switch(ty) {
+ case FULL_EX_LOCK:
+ fl.l_type = F_WRLCK;
+ break;
+ case FULL_SH_LOCK:
+ fl.l_type = F_RDLCK;
+ break;
+ case PART_EX_LOCK:
+ fl.l_type = F_WRLCK;
+ fl.l_len = (off_t)1;
+ break;
+ case PART_SH_LOCK:
+ fl.l_type = F_RDLCK;
+ fl.l_len = (off_t)1;
+ break;
+ default:
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! unknown lock type: %d", ty);
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ }
+/*
+ * Lock test file with fcntl().
+ */
+ if (fcntl(Fd, F_SETLK, &fl) != -1)
+ return((char *)NULL);
+ (void) snprintf(buf, sizeof(buf) - 1, "ERROR!!! fcntl() lock error: %s",
+ strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+# endif /* defined(USE_FCNTL) */
+
+# if defined(USE_FLOCK)
+/*
+ * Check flock() lock request.
+ */
+ switch(ty) {
+ case FULL_EX_LOCK:
+ flf = LOCK_EX;
+ break;
+ case FULL_SH_LOCK:
+ flf = LOCK_SH;
+ break;
+ case PART_EX_LOCK:
+ case PART_SH_LOCK:
+ return("ERROR!!! flock() doesn't support partial locks");
+ break;
+ default:
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! unknown flock() type: %d", ty);
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ }
+/*
+ * Acquire lock.
+ */
+ if (!flock(Fd, flf))
+ return((char *)NULL);
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! flock() %s lock failed: %s",
+ (flf == LOCK_EX) ? "exclusive" : "shared",
+ strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+# endif /* defined(USE_FLOCK) */
+
+}
+
+
+/*
+ * tstwlsof() -- test the open file with lsof
+ */
+
+static char *
+tstwlsof(opt, xlk)
+ char *opt; /* extra lsof options */
+ char *xlk; /* expected lock value */
+{
+ char buf[2048]; /* temporary buffer */
+ LTfldo_t *cmdp; /* command pointer */
+ LTfldo_t *devp; /* device pointer */
+ char *cem; /* current error message pointer */
+ int ff = 0; /* file found status */
+ LTfldo_t *fop; /* field output pointer */
+ LTfldo_t *inop; /* inode number pointer */
+ LTfldo_t *lkp; /* lock pointer */
+ LTdev_t lsofdc; /* lsof device components */
+ int nf; /* number of fields */
+ LTfldo_t *nmp; /* file name pointer */
+ char *opv[4]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message pointer */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ struct stat sb; /* stat(2) buffer */
+ LTdev_t stdc; /* stat(2) device components */
+ char *tcp; /* temporary character pointer */
+ int ti; /* temporary integer */
+ LTfldo_t *typ; /* file type pointer */
+/*
+ * Make sure there is an expected lock value.
+ */
+ if (!xlk || !*xlk)
+ (void) PrtMsgX("ERROR!!! no expected lock value", Pn, cleanup, 1);
+/*
+ * Get test file's information.
+ */
+ if (stat(Path, &sb)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't stat(2) %s: %s", Path, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Extract components from test file's device number.
+ */
+ if ((cem = ConvStatDev(&sb.st_dev, &stdc)))
+ (void) PrtMsgX(cem, Pn, cleanup, 1);
+/*
+ * Complete the option vector and start lsof execution.
+ */
+ ti = 0;
+ if (opt && *opt)
+ opv[ti++] = opt;
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+ opv[ti++] = Path;
+ opv[ti] = (char *)NULL;
+ if ((cem = ExecLsof(opv)))
+ return(cem);
+/*
+ * Read lsof output.
+ */
+ while (!ff && (fop = RdFrLsof(&nf, &cem))) {
+ if (cem) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != MyPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line. Make sure its number matches the
+ * test file's descriptor number.
+ *
+ * Scan for lock and name fields.
+ */
+ if (!pids)
+ break;
+ for (ti = 0, tcp = fop->v; *tcp; tcp++) {
+
+ /*
+ * Convert file descriptor to a number.
+ */
+ if (*tcp == ' ')
+ continue;
+ if (((int)*tcp < (int)'0') || ((int)*tcp > (int)'9')) {
+ ti = -1;
+ break;
+ }
+ ti = (ti * 10) + (int)*tcp - (int)'0';
+ }
+ if (Fd != ti)
+ break;
+ devp = inop = lkp = nmp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch(fop->ft) {
+ case LSOF_FID_DEVN:
+ devp = fop;
+ break;
+ case LSOF_FID_INODE:
+ inop = fop;
+ break;
+ case LSOF_FID_LOCK:
+ lkp = fop;
+ break;
+ case LSOF_FID_NAME:
+ nmp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the results of the file descriptor field scan.
+ *
+ * (Don't compare path names because of symbolic link interference.)
+ */
+ if (!devp || !inop || !nmp || !typ)
+ break;
+ if (strcasecmp(typ->v, "reg") && strcasecmp(typ->v, "vreg"))
+ break;
+ if (ConvLsofDev(devp->v, &lsofdc))
+ break;
+ if ((stdc.maj != lsofdc.maj)
+ || (stdc.min != lsofdc.min)
+ || (stdc.unit != lsofdc.unit))
+ break;
+ (void) snprintf(buf, sizeof(buf) - 1, "%u",
+ (unsigned int)sb.st_ino);
+ buf[sizeof(buf) - 1] = '\0';
+ if (strcmp(inop->v, buf))
+ break;
+ /*
+ * The specified file has been located. Check its lock status.
+ */
+ ff = 1;
+ if (!lkp || strcmp(lkp->v, xlk)) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "lock mismatch: expected %s, got \"%s\"", xlk,
+ lkp ? lkp->v : "(none)");
+ pem = MkStrCpy(buf, &ti);
+ }
+ break;
+ }
+ }
+ (void) StopLsof();
+ if (!ff) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "lock test file %s not found by lsof", Path);
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ }
+ return(pem);
+}
+
+
+/*
+ * unlkfile() -- unlock the test file
+ */
+
+static char *
+unlkfile(ty)
+ int ty; /* current *_*_LOCK lock typ */
+{
+ char buf[2048]; /* temporary buffer */
+ int ti; /* temporary integer */
+
+# if defined(USE_FCNTL)
+ struct flock fl; /* flock control structure */
+/*
+ * Check current fcntl() lock type.
+ */
+ (void) memset((void *)&fl, 0, sizeof(fl));
+ switch(ty) {
+ case FULL_EX_LOCK:
+ case FULL_SH_LOCK:
+ break;
+ case PART_EX_LOCK:
+ case PART_SH_LOCK:
+ fl.l_len = (off_t)1;
+ break;
+ default:
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! unknown unlock type: %d", ty);
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ }
+/*
+ * Unlock test file with fcntl().
+ */
+ fl.l_type = F_UNLCK;
+ if (fcntl(Fd, F_SETLK, &fl) != -1)
+ return((char *)NULL);
+ (void) snprintf(buf, sizeof(buf) - 1, "ERROR!!! fcntl() unlock error: %s",
+ strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+# endif /* defined(USE_FCNTL) */
+
+# if defined(USE_FLOCK)
+/*
+ * Check current flock() lock type.
+ */
+ switch(ty) {
+ case FULL_EX_LOCK:
+ case FULL_SH_LOCK:
+ break;
+ default:
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! unknown unlock type: %s", ty);
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ }
+/*
+ * Unlock file with flock().
+ */
+ if (!flock(Fd, LOCK_UN))
+ return((char *)NULL);
+ (void) snprintf(buf, sizeof(buf) - 1, "ERROR!!! flock() unlock error: %s",
+ strerror(errno));
+ return(MkStrCpy(buf, &ti));
+# endif /* defined(USE_FLOCK) */
+
+}
+#endif /* !defined(USE_FLOCK) && !defined(USE_FCNTL) */
--- /dev/null
+/*
+ * LTnfs.c -- Lsof Test NFS tests
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+
+/*
+ * Pre-definitions that may be revoked by specific dialects
+ */
+
+#define DO_TEST /* do the test */
+
+
+#if defined(LT_DIAL_darwin)
+/*
+ * Darwin-specific items
+ */
+
+# if LT_VERS<800
+#undef DO_TEST
+# endif /* LT_VERS<800 */
+#endif /* defined(LT_DIAL_darwin) */
+
+
+/*
+ * Globals
+ */
+
+int Fd = -1; /* test file descriptor; open if >= 0 */
+pid_t MyPid = (pid_t)0; /* PID of this process */
+int NFstat = 0; /* NFS file status: 0 == closed
+ * 1 == not created by this
+ * these and must not be
+ * unlinked
+ * 2 == created by this test
+ * and must be unlinked
+ */
+char *Path = (char *)NULL; /* test file path; none if NULL */
+char *Pn = (char *)NULL; /* program name */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static char *FindNFSfile,(int *ff, char *szbuf));
+
+
+/*
+ * Main program
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ char *em; /* error message pointer */
+ int ff; /* FindNFSfile() file-found flag */
+ int sz; /* file size (if created) */
+ char szbuf[32]; /* created test file size in ASCII */
+ int ti; /* temporary index */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+
+#if !defined(DO_TEST)
+/*
+ * If the dialect has disabled the test, echo that result and exit with
+ * a successful return code.
+ */
+ (void) PrtMsgX(LT_DONT_DO_TEST, Pn, cleanup, 0);
+#endif /* !defined(DO_TEST) */
+
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "hp:", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h] [-p path]", Pn);
+ PrtMsg (" -h print help (this panel)", Pn);
+ PrtMsgX (" -p path define test file path", Pn, cleanup, xv);
+ }
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * Process the file path and open it.
+ */
+ if ((Path = LTopt_p)) {
+
+ /*
+ * The file path was supplied. Open the file read-only.
+ */
+ if ((Fd = open(Path, O_RDONLY, 0400)) < 0) {
+ (void) fprintf(stderr, "ERROR!!! can't read-only open %s\n",
+ Path);
+ goto print_file_error;
+ }
+ /*
+ * Record that an existing file is being used. Clear its ASCII size.
+ */
+ NFstat = 1;
+ szbuf[0] = '\0';
+ } else {
+
+ /*
+ * The file path wasn't supplied with -p, so generate one.
+ */
+ (void) snprintf(buf, sizeof(buf) - 1, "./config.LTnfs%ld",
+ (long)MyPid);
+ buf[sizeof(buf) - 1] = '\0';
+ Path = MkStrCpy(buf, &ti);
+ /*
+ * Open a new test file at the specified path.
+ */
+ (void) unlink(Path);
+ if ((Fd = open(Path, O_RDWR|O_CREAT, 0600)) < 0) {
+ (void) fprintf(stderr, "ERROR!!! can't create %s\n", Path);
+
+print_file_error:
+
+ MsgStat = 1;
+ (void) snprintf(buf, sizeof(buf) - 1, " Errno %d: %s",
+ errno, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+ NFstat = 2;
+ /*
+ * Write the test file to its expected size.
+ */
+ sz = sizeof(buf);
+ for (ti = 0; ti < sz; ti++) {
+ buf[ti] = (char)(ti & 0xff);
+ }
+ if (write(Fd, buf, sz) != sz) {
+ (void) fprintf(stderr, "ERROR!!! can't write %d bytes to %s\n",
+ sz, Path);
+ goto print_file_error;
+ }
+ /*
+ * Fsync() the file.
+ */
+ if (fsync(Fd)) {
+ (void) fprintf(stderr, "ERROR!!! can't fsync %s\n", Path);
+ goto print_file_error;
+ }
+ /*
+ * Convert the file size to ASCII.
+ */
+ (void) snprintf(szbuf, sizeof(szbuf) - 1, "%d", sz);
+ szbuf[sizeof(szbuf) - 1] = '\0';
+ }
+/*
+ * Make sure the test file can be found on an NFS file system.
+ */
+ if ((em = FindNFSfile(&ff, szbuf))) {
+
+ /*
+ * Print the error message returned by FindNFSfile().
+ */
+ (void) PrtMsg(em, Pn);
+ if (!ff) {
+
+ /*
+ * If the file couldn't be found, print hints.
+ */
+ if (NFstat == 1) {
+ (void) PrtMsg(
+ "Hint: this test must be able to open for read access",
+ Pn);
+ (void) PrtMsg(
+ "the file at the path supplied with the -p option and",
+ Pn);
+ (void) PrtMsg(
+ "that file must be a regular file (not a directory) on",
+ Pn);
+ (void) PrtMsg(
+ "an NFS file system.\n",
+ Pn);
+ (void) PrtMsgX(
+ "See 00FAQ and 00TEST for more information.",
+ Pn, cleanup, 1);
+ } else if (NFstat == 2) {
+ (void) PrtMsg(
+ "Hint: the temporary path generated by this test might",
+ Pn);
+ (void) PrtMsg(
+ "not be on an NFS file system, or this test might be",
+ Pn);
+ (void) PrtMsg(
+ "unable to create a file on the NFS file system.\n",
+ Pn);
+ (void) PrtMsg(
+ "As a work-around use the -p option to specify a path to",
+ Pn);
+ (void) PrtMsg(
+ "a regular file (not a directory) on an NFS file system",
+ Pn);
+ (void) PrtMsg(
+ "to which this test will have read access.\n",
+ Pn);
+ (void) PrtMsgX(
+ "See 00FAQ and 00TEST for more information.",
+ Pn, cleanup, 1);
+ }
+ }
+ }
+/*
+ * Exit successfully.
+ */
+ (void) PrtMsgX("OK", Pn, cleanup, 0);
+ return(0);
+}
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+ if (Fd >= 0) {
+ (void) close(Fd);
+ Fd = -1;
+ if (Path) {
+ if (NFstat == 2)
+ (void) unlink(Path);
+ Path = (char *)NULL;
+ }
+ }
+}
+
+
+/*
+ * FindNFSfile() -- find the NFS file with lsof
+ */
+
+static char *
+FindNFSfile(ff, szbuf)
+ int *ff; /* file-found response receptor */
+ char *szbuf; /* expected file size in ASCII (if
+ * the file was created by this test */
+{
+ char buf[2048]; /* temporary buffer */
+ char *cem; /* current error message pointer */
+ LTfldo_t *cmdp; /* command pointer */
+ LTfldo_t *devp; /* device pointer */
+ LTfldo_t *fop; /* field output pointer */
+ char ibuf[64]; /* inode number buffer */
+ LTfldo_t *inop; /* inode number pointer */
+ LTdev_t lsofdc; /* lsof device components */
+ int nf; /* number of fields */
+ char nlkbuf[32]; /* link count buffer */
+ LTfldo_t *nlkp; /* nlink pointer */
+ char *opv[5]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message pointer */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ struct stat sb; /* stat(2) buffer */
+ LTdev_t stdc; /* stat(2) device components */
+ LTfldo_t *szp; /* size pointer */
+ char *tcp; /* temporary character pointer */
+ int ti; /* temporary integer */
+ LTfldo_t *typ; /* file type pointer */
+/*
+ * Check the argument pointers.
+ *
+ * Set the file-found response false.
+ */
+ if (!ff || !szbuf)
+ (void) PrtMsgX("ERROR!!! missing argument to FindNFSfile()",
+ Pn, cleanup, 1);
+ *ff = 0;
+/*
+ * Get test file's information.
+ */
+ if (stat(Path, &sb)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't stat(2) %s: %s", Path, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Extract components from test file's stat buffer.
+ */
+ if ((cem = ConvStatDev(&sb.st_dev, &stdc)))
+ PrtMsgX(cem, Pn, cleanup, 1);
+ (void) snprintf(ibuf, sizeof(ibuf) - 1, "%u", (unsigned int)sb.st_ino);
+ ibuf[sizeof(ibuf) - 1] = '\0';
+ (void) snprintf(nlkbuf, sizeof(nlkbuf) - 1, "%d", (int)sb.st_nlink);
+ nlkbuf[sizeof(nlkbuf) - 1] = '\0';
+/*
+ * Complete the option vector and start lsof execution.
+ */
+ ti = 0;
+ opv[ti++] = "-s";
+ opv[ti++] = "-Na";
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+ opv[ti++] = Path;
+ opv[ti] = (char *)NULL;
+ if ((cem = ExecLsof(opv)))
+ return(cem);
+/*
+ * Read lsof output.
+ */
+ while (!*ff && (fop = RdFrLsof(&nf, &cem))) {
+ if (cem) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != MyPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line. Make sure its number matches the
+ * test file's descriptor number.
+ */
+ if (!pids)
+ break;
+ for (ti = 0, tcp = fop->v; *tcp; tcp++) {
+
+ /*
+ * Convert file descriptor to a number.
+ */
+ if (*tcp == ' ')
+ continue;
+ if (((int)*tcp < (int)'0') || ((int)*tcp > (int)'9')) {
+ ti = -1;
+ break;
+ }
+ ti = (ti * 10) + (int)*tcp - (int)'0';
+ }
+ if (Fd != ti)
+ break;
+ /*
+ * Scan for device, inode, nlink, offset, size and type fields.
+ */
+ devp = inop = nlkp, szp = typ = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_DEVN:
+ devp = fop;
+ break;
+ case LSOF_FID_INODE:
+ inop = fop;
+ break;
+ case LSOF_FID_NLINK:
+ nlkp = fop;
+ break;
+ case LSOF_FID_OFFSET:
+ break;
+ case LSOF_FID_SIZE:
+ szp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the device, inode, and type of the file.
+ */
+ if (!devp || !inop || !typ)
+ break;
+ if (strcasecmp(typ->v, "reg") && strcasecmp(typ->v, "vreg"))
+ break;
+ if ((cem = ConvLsofDev(devp->v, &lsofdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if ((stdc.maj != lsofdc.maj)
+ || (stdc.min != lsofdc.min)
+ || (stdc.unit != lsofdc.unit)
+ || strcmp(inop->v, ibuf)
+ ) {
+ break;
+ }
+ /*
+ * Indicate the file was found.
+ */
+ *ff = 1;
+ /*
+ * Check the link count.
+ */
+ if (!nlkp) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! lsof didn't report a link count for %s", Path);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if (strcmp(nlkp->v, nlkbuf)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! wrong link count: expected %s, got %s",
+ nlkbuf, nlkp->v);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ /*
+ * If the file was created by this test, check its size.
+ */
+ if (NFstat == 2) {
+ if (!szp) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! lsof didn't report a size for %s", Path);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if (strcmp(szp->v, szbuf)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! wrong file size: expected %s, got %s",
+ szbuf, szp->v);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ }
+ /*
+ * The requested file was located. Return the previous error message
+ * pointer. (It will be NULL if no error was detected.)
+ */
+ (void) StopLsof();
+ return(pem);
+ }
+ }
+/*
+ * The test file wasn't found.
+ */
+ (void) StopLsof();
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! test file %s not found by lsof", Path);
+ buf[sizeof(buf) - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+}
--- /dev/null
+/*
+ * LTnlink.c -- Lsof Test nlink tests
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+
+/*
+ * Pre-definitions that may be changed by specific dialects
+ */
+
+#define DO_TEST /* do the test */
+
+
+/*
+ * Dialect-specific items
+ */
+
+
+#if defined(LT_DIAL_darwin)
+/*
+ * Darwin-specific items
+ */
+
+# if defined(LT_KMEM)
+#undef DO_TEST
+# endif /* defined(LT_KMEM) */
+
+#endif /* defined(LT_DIAL_darwin) */
+
+/*
+ * Globals
+ */
+
+int Fd = -1; /* test file descriptor; open if >= 0 */
+pid_t MyPid = (pid_t)0; /* PID of this process */
+char *Path = (char *)NULL; /* test file path; none if NULL */
+char *Pn = (char *)NULL; /* program name */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static char *FindFile,(char *opt, int *ff, int ie, LTdev_t *tfdc,
+ char *ibuf, char *xlnk, char *szbuf));
+
+
+/*
+ * Main program
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ int do_unlink = 1; /* do the unlink test section */
+ char *em; /* error message pointer */
+ int ff; /* FindFile() file-found flag */
+ char ibuf[32]; /* inode number in ASCII */
+ char *opt; /* lsof option */
+ int sz; /* file size */
+ char szbuf[32]; /* file size in ASCII */
+ LTdev_t tfdc; /* device components */
+ struct stat tfsb; /* test file stat(2) buffer */
+ int ti, tj; /* temporary indexes */
+ char xlnk[32]; /* expected link count in ASCII */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+
+#if !defined(DO_TEST)
+/*
+ * Quit if lsof for this dialect doesn't support adequate nlink reporting.
+ */
+ (void) PrtMsgX(LT_DONT_DO_TEST, Pn, cleanup, 0);
+#endif /* !defined(DO_TEST) */
+
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "hp:", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h] [-p path]", Pn);
+ PrtMsg (" -h print help (this panel)", Pn);
+ PrtMsgX (" -p path define test file path", Pn, cleanup, xv);
+ }
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * Process the file path.
+ */
+ if (!(Path = LTopt_p)) {
+
+ /*
+ * The file path was not supplied, so make one.
+ */
+ (void) snprintf(buf, sizeof(buf) - 1, "./config.LTnlink%ld",
+ (long)MyPid);
+ buf[sizeof(buf) - 1] = '\0';
+ Path = MkStrCpy(buf, &ti);
+ }
+/*
+ * Create the test file.
+ */
+ (void) unlink(Path);
+ if ((Fd = open(Path, O_RDWR|O_CREAT, 0600)) < 0) {
+ (void) fprintf(stderr, "ERROR!!! can't create %s\n", Path);
+
+print_file_error:
+
+ MsgStat = 1;
+ (void) snprintf(buf, sizeof(buf) - 1, " Errno %d: %s",
+ errno, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Write the test file to its expected size.
+ */
+ sz = sizeof(buf);
+ for (ti = 0; ti < sz; ti++) {
+ buf[ti] = (char)(ti & 0xff);
+ }
+ if (write(Fd, buf, sz) != sz) {
+ (void) fprintf(stderr, "ERROR!!! can't write %d bytes to %s\n",
+ sz, Path);
+ goto print_file_error;
+ }
+/*
+ * Fsync() the file.
+ */
+ if (fsync(Fd)) {
+ (void) fprintf(stderr, "ERROR!!! can't fsync %s\n", Path);
+ goto print_file_error;
+ }
+/*
+ * Stat(2) the test file.
+ */
+ if (stat(Path, &tfsb)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't stat(2) %s: %s", Path, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Set the test file status to open and linked.
+ *
+ * Get the test file's parameters:
+ *
+ * * device paramters in LTdev_t form;
+ * * inode number in ASCII;
+ * * link count in ASCII;
+ * * file size in ASCII.
+ */
+ if ((em = ConvStatDev(&tfsb.st_dev, &tfdc)))
+ PrtMsgX(em, Pn, cleanup, 1);
+ (void) snprintf(ibuf, sizeof(ibuf) - 1, "%u", (unsigned int)tfsb.st_ino);
+ ibuf[sizeof(szbuf) - 1] = '\0';
+ (void) snprintf(xlnk, sizeof(xlnk) - 1, "%d", tfsb.st_nlink);
+ ibuf[sizeof(szbuf) - 1] = '\0';
+ (void) snprintf(szbuf, sizeof(szbuf) - 1, "%d", sz);
+ szbuf[sizeof(szbuf) - 1] = '\0';
+/*
+ * See if the file is on an NFS file system.
+ */
+ (void) FindFile("-Na", &ff, 1, &tfdc, ibuf, xlnk, szbuf);
+ if (ff) {
+
+ /*
+ * The file was found on an NFS file system.
+ */
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "WARNING!!! Test file %s is NFS mounted.", Path);
+ (void) PrtMsg(buf, Pn);
+ (void) PrtMsg(
+ " As a result this test probably won't be able to unlink it and",
+ Pn);
+ (void) PrtMsg(
+ " find its open and unlinked instance with lsof's +L option.",
+ Pn);
+ (void) PrtMsg(
+ " Therefore, that section of this test has been disabled.\n",
+ Pn);
+ (void) PrtMsg(
+ " Hint: supply a path with the -p option to a file in a non-NFS",
+ Pn);
+ (void) PrtMsg(
+ " file system that this test can write and unlink.\n",
+ Pn);
+ (void) PrtMsg(
+ " See 00FAQ and 00TEST for more information.",
+ Pn);
+ do_unlink = 0;
+ }
+/*
+ * Find the test file.
+ */
+ if ((em = FindFile("+L", &ff, 0, &tfdc, ibuf, xlnk, szbuf)))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * If the unlink test is enabled, do it.
+ */
+ if (do_unlink) {
+ (void) unlink(Path);
+ for (opt = "+L1", ti = 0, tj = 30; ti < tj; ti++) {
+
+ /*
+ * Wait a while for the link count to be updated before concluding
+ * lsof can't find the unlinked file. Use "+L1" for only the first
+ * third of the tries, then switch to "+L".
+ */
+ if ((ti + ti + ti) >= tj)
+ opt = "+L";
+ if (!(em = FindFile(opt, &ff, 0, &tfdc, ibuf, "0", szbuf)))
+ break;
+ if (ti)
+ (void) printf(".");
+ else
+ (void) printf("waiting for link count update: .");
+ (void) fflush(stdout);
+ (void) sleep(2);
+ }
+ if (ti) {
+
+ /*
+ * End the delay message.
+ */
+ printf("\n");
+ (void) fflush(stdout);
+ MsgStat = 1;
+ }
+ if (em)
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ }
+/*
+ * Exit successfully.
+ */
+ (void) PrtMsgX("OK", Pn, cleanup, 0);
+ return(0);
+}
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+ if (Fd >= 0) {
+ (void) close(Fd);
+ Fd = -1;
+ }
+ if (Path)
+ (void) unlink(Path);
+}
+
+
+/*
+ * FindFile() -- find a file with lsof
+ */
+
+static char *
+FindFile(opt, ff, ie, tfdc, ibuf, xlnk, szbuf)
+ char *opt; /* additional lsof options */
+ int *ff; /* file-found response receptor */
+ int ie; /* ignore errors if == 1 */
+ LTdev_t *tfdc; /* test file device components */
+ char *ibuf; /* inode number in ASCII */
+ char *xlnk; /* expected link count */
+ char *szbuf; /* file size in ASCII */
+{
+ char buf[2048]; /* temporary buffer */
+ char *cem; /* current error message pointer */
+ LTfldo_t *cmdp; /* command pointer */
+ LTfldo_t *devp; /* device pointer */
+ LTfldo_t *fop; /* field output pointer */
+ LTfldo_t *inop; /* inode number pointer */
+ LTdev_t lsofdc; /* lsof device components */
+ int nf; /* number of fields */
+ LTfldo_t *nlkp; /* nlink pointer */
+ char *opv[4]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message pointer */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ LTfldo_t *szp; /* size pointer */
+ char *tcp; /* temporary character pointer */
+ int ti; /* temporary integer */
+ LTfldo_t *typ; /* file type pointer */
+/*
+ * Check the argument pointers.
+ *
+ * Set the file-found response false.
+ */
+ if (!ff || !ibuf || !szbuf || !tfdc || !xlnk)
+ (void) PrtMsgX("ERROR!!! missing argument to FindFile()",
+ Pn, cleanup, 1);
+ *ff = 0;
+/*
+ * Complete the option vector and start lsof execution.
+ */
+ ti = 0;
+ if (opt && *opt)
+ opv[ti++] = opt;
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+ if (strcmp(xlnk, "0"))
+ opv[ti++] = Path;
+ opv[ti] = (char *)NULL;
+ if ((cem = ExecLsof(opv))) {
+ if (ie)
+ return((char *)NULL);
+ return(cem);
+ }
+/*
+ * Read lsof output.
+ */
+ while (!*ff && (fop = RdFrLsof(&nf, &cem))) {
+ if (cem) {
+ if (ie)
+ return((char *)NULL);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != MyPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line. Make sure its number matches the
+ * test file's descriptor number.
+ */
+ if (!pids)
+ break;
+ for (ti = 0, tcp = fop->v; *tcp; tcp++) {
+
+ /*
+ * Convert file descriptor to a number.
+ */
+ if (*tcp == ' ')
+ continue;
+ if (((int)*tcp < (int)'0') || ((int)*tcp > (int)'9')) {
+ ti = -1;
+ break;
+ }
+ ti = (ti * 10) + (int)*tcp - (int)'0';
+ }
+ if (Fd != ti)
+ break;
+ /*
+ * Scan for device, inode, nlink, size and type fields.
+ */
+ devp = inop = nlkp = szp = typ = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_DEVN:
+ devp = fop;
+ break;
+ case LSOF_FID_INODE:
+ inop = fop;
+ break;
+ case LSOF_FID_NLINK:
+ nlkp = fop;
+ break;
+ case LSOF_FID_SIZE:
+ szp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the device, inode, and type of the file.
+ */
+ if (!devp || !inop || !szp || !typ)
+ break;
+ if (strcasecmp(typ->v, "reg") && strcasecmp(typ->v, "vreg"))
+ break;
+ if ((cem = ConvLsofDev(devp->v, &lsofdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if ((tfdc->maj != lsofdc.maj)
+ || (tfdc->min != lsofdc.min)
+ || (tfdc->unit != lsofdc.unit)
+ || strcmp(inop->v, ibuf)
+ ) {
+ break;
+ }
+ /*
+ * Indicate the file was found.
+ */
+ *ff = 1;
+ /*
+ * Check the size and link count.
+ */
+ if (!szp) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! lsof didn't report a file size for %s", Path);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if (strcmp(szp->v, szbuf)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! wrong file size: expected %s, got %s",
+ szbuf, szp->v);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if (!nlkp) {
+ if (strcmp(xlnk, "0")) {
+
+ /*
+ * If lsof returned no link count and the expected return is
+ * not "0", it's an error. Otherwise, interpret no link count
+ * as equivalent to a "0" link count.
+ */
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! lsof didn't report a link count for %s",
+ Path);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ } else {
+ if (strcmp(nlkp->v, xlnk)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! wrong link count: expected %s, got %s",
+ xlnk, nlkp->v);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ }
+ /*
+ * The requested file was located. Return the previous error message
+ * pointer unless errors are being ignored. (The previous error
+ * message pointer will be NULL if no error was detected.)
+ */
+ (void) StopLsof();
+ if (ie)
+ return((char *)NULL);
+ return(pem);
+ }
+ }
+/*
+ * Clean up and return.
+ */
+ (void) StopLsof();
+ if (!*ff && !ie) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! %s test file %s not found by lsof",
+ strcmp(xlnk, "0") ? "linked" : "unlinked",
+ Path);
+ buf[sizeof(buf) - 1] = '\0';
+ pem = MkStrCpy(buf, &ti);
+ }
+ if (ie)
+ return((char *)NULL);
+ return(pem);
+}
--- /dev/null
+/*
+ * LTsock.c -- Lsof Test IPv4 sockets
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+#include <netdb.h>
+#include <signal.h>
+#include <sys/socket.h>
+#include <netinet/in.h>
+#include <arpa/inet.h>
+
+
+/*
+ * Pre-definitions that make be changed or revoked by dialects
+ */
+
+#define SIGHANDLER_T void /* signal handler function type */
+#define LT_SOCKLEN_T int /* socket length type */
+
+
+#if defined(LT_DIAL_aix)
+/*
+ * AIX-specific items
+ */
+
+#undef LT_SOCKLEN_T
+#define LT_SOCKLEN_T size_t
+#endif /* defined(LT_DIAL_aix) */
+
+
+#if defined(LT_DIAL_darwin)
+/*
+ * Darwin-specific items
+ */
+
+# if LT_VERS>=800
+#undef LT_SOCKLEN_T
+#define LT_SOCKLEN_T socklen_t
+# endif /* LT_VERS>=800 */
+#endif /* defined(LT_DIAL_darwin) */
+
+
+#if defined(LT_DIAL_hpux)
+/*
+ * HP-UX-specific items
+ */
+
+# if LT_VERS>=1123 && defined(__GNUC__)
+#undef LT_SOCKLEN_T
+#define LT_SOCKLEN_T size_t
+# endif /* LT_VERS>=1123 && defined(__GNUC__) */
+#endif /* defined(LT_DIAL_hpux) */
+
+
+#if defined(LT_DIAL_ou)
+/*
+ * OpenUNIX-specific items
+ */
+
+#undef LT_SOCKLEN_T
+#define LT_SOCKLEN_T size_t
+#endif /* defined(LT_DIAL_ou) */
+
+
+#if defined(LT_DIAL_uw)
+/*
+ * UnixWare-specific items
+ */
+
+#undef LT_SOCKLEN_T
+#define LT_SOCKLEN_T size_t
+#endif /* defined(LT_DIAL_uw) */
+
+
+/*
+ * Local definitions
+ */
+
+#define ALARMTM 30 /* alarm timer */
+
+#define LT_CLNT 0 /* child process index */
+#define LT_SRVR 1 /* parent process index */
+
+#define LT_FNF 0 /* file not found */
+#define LT_FBYIP 1 /* file found by IP address */
+#define LT_FBYHN 2 /* file found by host name */
+#define LT_FBYPORT 4 /* file found by port */
+
+#if !defined(MAXHOSTNAMELEN)
+#define MAXHOSTNAMELEN 256 /* maximum host name length */
+#endif /* !defined(MAXHOSTNAMELEN) */
+
+#if !defined(MAXPATHLEN)
+#define MAXPATHLEN 1024 /* maximum path length */
+#endif /* !defined(MAXPATHLEN) */
+
+
+/*
+ * Local structure definitions.
+ */
+
+
+typedef struct fdpara { /* file descriptor parameters */
+ int fd; /* FD */
+ char *fds; /* FD in ASCII */
+ int ff; /* file found flags (see LT_F*) */
+ char *host; /* host name */
+ int hlen; /* strlen(host) */
+ char *ipaddr; /* dotted IP address */
+ int ilen; /* strlen(ipaddr) */
+ pid_t pid; /* PID of process */
+ char *port; /* port in ASCII */
+ int plen; /* strlen(port) */
+ struct sockaddr_in sa; /* socket's address */
+} fdpara_t;
+
+
+/*
+ * Globals
+ */
+
+pid_t CPid = (pid_t)0; /* client PID */
+fdpara_t FdPara[2]; /* file descriptor parameters */
+#define NFDPARA (sizeof(FdPara) /sizeof(fdpara_t))
+struct sockaddr_in Myad; /* my (server) socket address */
+pid_t MyPid = (pid_t)0; /* PID of this process */
+char *Pn = (char *)NULL; /* program name */
+char *PtNm[] = { "client", "server" };
+ /* program type name */
+int Ssock = -1; /* server socket */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void CleanupClnt,(void));
+_PROTOTYPE(static void CleanupSrvr,(void));
+_PROTOTYPE(static SIGHANDLER_T HandleClntAlarm,(int sig));
+_PROTOTYPE(static SIGHANDLER_T HandleSrvrAlarm,(int sig));
+_PROTOTYPE(static char *FindSock,(int fn));
+_PROTOTYPE(static void StartClnt,(struct sockaddr_in *cad));
+
+
+/*
+ * Main program
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ struct sockaddr_in aa; /* accept address */
+ struct sockaddr_in ba; /* bind address */
+ char buf[2048]; /* temporary buffer */
+ int bufl = sizeof(buf); /* size of buf[] */
+ struct sockaddr_in ca; /* connect address */
+ char *cem; /* current error message pointer */
+ char *ep; /* error message parameter */
+ char hnm[MAXHOSTNAMELEN + 1]; /* this host's name */
+ char *host; /* host name */
+ struct hostent *hp; /* this host's hostent structure */
+ char *ipaddr; /* IP address */
+ char *pem = (char *)NULL; /* previous error message */
+ char *port; /* port */
+ LT_SOCKLEN_T sal; /* socket address length */
+ char *tcp; /* temporary character size */
+ int ti, tj, tk; /* temporary indexes */
+ int tsfd; /* temporary socket FD */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+/*
+ * Initalize the FdPara[] array before any CleanupClnt() call.
+ */
+ for (ti = 0; ti < NFDPARA; ti++) {
+ (void) memset((void *)&FdPara[ti], 0, sizeof(fdpara_t));
+ FdPara[ti].fd = -1;
+ FdPara[ti].ff = LT_FNF;
+ }
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "h", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h]", Pn);
+ PrtMsgX(" -h print help (this panel)", Pn, CleanupSrvr,
+ xv);
+ }
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((cem = IsLsofExec()))
+ (void) PrtMsgX(cem, Pn, CleanupSrvr, 1);
+ if ((cem = CanRdKmem()))
+ (void) PrtMsgX(cem, Pn, CleanupSrvr, 1);
+/*
+ * Get the host name and its IP address. Convert the IP address to dotted
+ * ASCII form.
+ */
+ if (gethostname(hnm, sizeof(hnm) - 1)) {
+ cem = "ERROR!!! can't get this host's name";
+ goto print_errno;
+ }
+ hnm[sizeof(hnm) - 1] = '\0';
+ if (!(hp = gethostbyname(hnm))) {
+ (void) snprintf(buf, bufl - 1, "ERROR!!! can't get IP address for %s",
+ hnm);
+ buf[bufl - 1] = '\0';
+ cem = buf;
+ goto print_errno;
+ }
+ (void) memset((void *)&Myad, 0, sizeof(Myad));
+ if ((ti = hp->h_length) > sizeof(Myad.sin_addr))
+ ti = sizeof(Myad.sin_addr);
+ (void) memcpy((void *)&Myad.sin_addr, (void *)hp->h_addr, ti);
+ Myad.sin_family = hp->h_addrtype;
+/*
+ * Get INET domain socket FDs.
+ */
+ for (ti = 0; ti < NFDPARA; ti++) {
+ if ((tsfd = socket(AF_INET, SOCK_STREAM, 0)) < 0) {
+ ep = "socket";
+
+print_errno_by_ti:
+
+ /*
+ * Report socket function error.
+ *
+ * Entry: ep = function name
+ * hnm = this host's name
+ * Myad = this host's IP address
+ * ti = FdPara[] index
+ */
+ (void) snprintf(buf, bufl - 1, "ERROR!!! %s %s() failure",
+ PtNm[ti], ep);
+ buf[bufl - 1] = '\0';
+ PrtMsg(buf, Pn);
+ (void) snprintf(buf, bufl - 1, " host: %s",
+ FdPara[ti].host ? FdPara[ti].host : hnm);
+ buf[bufl - 1] = '\0';
+ PrtMsg(buf, Pn);
+ (void) snprintf(buf, bufl - 1, " IP: %s",
+ FdPara[ti].ipaddr ? FdPara[ti].ipaddr
+ : inet_ntoa(Myad.sin_addr));
+ buf[bufl - 1] = '\0';
+ cem = buf;
+
+print_errno:
+
+ /*
+ * Report errno.
+ *
+ * Entry: errno = error number
+ */
+ PrtMsg(cem, Pn);
+ (void) snprintf(buf, bufl - 1, " Errno %d: %s", errno,
+ strerror(errno));
+ buf[bufl - 1] = '\0';
+ PrtMsgX(buf, Pn, CleanupSrvr, 1);
+ }
+ /*
+ * Put the FD just acquired in FdPara[ti].fd.
+ *
+ * Set the file-not-found to LT_FNF.
+ *
+ * Save the server socket if this FdPara[] is for it.
+ */
+ FdPara[ti].fd = tsfd;
+ (void) snprintf(buf, bufl - 1, "%d", tsfd);
+ buf[bufl - 1] = '\0';
+ FdPara[ti].fds = MkStrCpy(buf, &tj);
+ if (ti == LT_SRVR)
+ Ssock = tsfd;
+ }
+/*
+ * Bind the host name to the server socket.
+ *
+ * Get and save the server's socket address.
+ *
+ * Initiate a listen with an address list of one.
+ */
+ (void) memcpy((void *)&ba, (void *)&Myad, sizeof(ba));
+ ti = LT_SRVR;
+ FdPara[ti].pid = MyPid;
+ if (bind(Ssock, (struct sockaddr *)&ba, sizeof(ba)) < 0) {
+ ep = "bind";
+ goto print_errno_by_ti;
+ }
+ sal = (LT_SOCKLEN_T)sizeof(ca);
+ if (getsockname(Ssock, (struct sockaddr *)&ca, &sal)) {
+ ep = "getsockname";
+ goto print_errno_by_ti;
+ }
+ (void) memcpy((void *)&FdPara[ti].sa, (void *)&ca, sizeof(FdPara[ti].sa));
+ if (listen(Ssock, 1) < 0) {
+ ep = "listen";
+ goto print_errno_by_ti;
+ }
+/*
+ * Fork a child process to run as the client.
+ */
+ switch ((CPid = (pid_t)fork())) {
+ case (pid_t)0:
+
+ /*
+ * This is the child. Start the client.
+ */
+ StartClnt(&ca);
+ (void) PrtMsgX("ERROR!!! unexpected client return", Pn, CleanupSrvr,
+ 1);
+ case (pid_t)-1:
+
+ /*
+ * This is a fork error.
+ */
+ cem = "ERROR!!! fork() error";
+ goto print_errno;
+ default:
+
+ /*
+ * This is the parent.
+ *
+ * Save the client's PID.
+ *
+ * Close the client's socket.
+ */
+ FdPara[LT_CLNT].pid = CPid;
+ if (FdPara[LT_CLNT].fd >= 0) {
+ (void) close(FdPara[LT_CLNT].fd);
+ FdPara[LT_CLNT].fd = -1;
+ }
+ }
+/*
+ * Set a SIGALRM, then accept() the connection from the client.
+ *
+ * Save the client's socket address.
+ *
+ * Replace the server's FD with the accepted one and close the original.
+ */
+ sal = (LT_SOCKLEN_T)sizeof(aa);
+ (void) alarm(0);
+ (void) signal(SIGALRM, HandleSrvrAlarm);
+ (void) alarm(ALARMTM);
+ tsfd = FdPara[LT_SRVR].fd = accept(Ssock, (struct sockaddr *)&aa, &sal);
+ (void) alarm(0);
+ (void) signal(SIGALRM, SIG_DFL);
+ if (tsfd < 0) {
+ ep = "accept";
+ goto print_errno_by_ti;
+ }
+ (void) snprintf(buf, bufl - 1, "%d", tsfd);
+ buf[bufl - 1] = '\0';
+ if (FdPara[LT_SRVR].fds)
+ (void) free((void *)FdPara[LT_SRVR].fds);
+ FdPara[LT_SRVR].fds = MkStrCpy(buf, &tj);
+ ti = LT_CLNT;
+ (void) memcpy((void *)&FdPara[ti].sa, (void *)&aa, sizeof(FdPara[ti].sa));
+ (void) close(Ssock);
+ Ssock = -1;
+/*
+ * Convert the client and server IP address to ASCII form.
+ *
+ * Look up the client and server host names for their IP addresses.
+ *
+ * Convert the port from the socket address to host form.
+ */
+ for (ti = 0; ti < NFDPARA; ti++) {
+ tcp = inet_ntoa(FdPara[ti].sa.sin_addr);
+ FdPara[ti].ipaddr = MkStrCpy(tcp, &FdPara[ti].ilen);
+ (void) snprintf(buf, bufl - 1, "%d",
+ (int)ntohs(FdPara[ti].sa.sin_port));
+ buf[bufl - 1] = '\0';
+ FdPara[ti].port = MkStrCpy(buf, &FdPara[ti].plen);
+ if (!(hp = gethostbyaddr((char *)&FdPara[ti].sa.sin_addr,
+ sizeof(FdPara[ti].sa.sin_addr),
+ FdPara[ti].sa.sin_family))
+ ) {
+ ep = "gethostbyaddr";
+ goto print_errno_by_ti;
+ }
+ if (hp->h_name)
+ FdPara[ti].host = MkStrCpy(hp->h_name, &FdPara[ti].hlen);
+ else {
+
+ /*
+ * The connected client's socket address can't be mapped to a host
+ * name.
+ */
+
+ (void) snprintf(buf, bufl - 1,
+ "ERROR!!! can't map %s (client) to a host name",
+ FdPara[ti].ipaddr);
+ buf[bufl - 1] = '\0';
+ PrtMsgX(buf, Pn, CleanupSrvr, 1);
+ }
+ }
+/*
+ * Call lsof three times to find the two sockets: 1) by host name and port;
+ * 2) by IP address and port; and 3) by port.
+ */
+ if ((cem = FindSock(LT_FBYHN)))
+ PrtMsgX(cem, Pn, CleanupSrvr, 1);
+ if ((cem = FindSock(LT_FBYIP)))
+ PrtMsgX(cem, Pn, CleanupSrvr, 1);
+ if ((cem = FindSock(LT_FBYPORT)))
+ PrtMsgX(cem, Pn, CleanupSrvr, 1);
+/*
+ * Check the FindSock() results.
+ */
+ for (pem = (char *)NULL, ti = 0; ti < NFDPARA; ti++) {
+ if ((tj = FdPara[ti].ff) != (LT_FBYHN | LT_FBYIP | LT_FBYPORT)) {
+ host = FdPara[ti].host;
+ ipaddr = FdPara[ti].ipaddr;
+ port = FdPara[ti].port;
+
+ /*
+ * This FD wasn't found by some search method.
+ */
+ if (!(tj & LT_FBYHN)) {
+
+ /*
+ * The search by host name and port failed.
+ */
+ (void) snprintf(buf, bufl - 1,
+ "ERROR!!! no %s socket by host and port: %s@%s",
+ PtNm[ti], host, port);
+ buf[bufl - 1] = '\0';
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = MkStrCpy(buf, &tk);
+ }
+ if (!(tj & LT_FBYIP)) {
+
+ /*
+ * The search by IP address and port failed.
+ */
+ (void) snprintf(buf, bufl - 1,
+ "ERROR!!! no %s socket by IP and port: %s@%s",
+ PtNm[ti], ipaddr, port);
+ buf[bufl - 1] = '\0';
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = MkStrCpy(buf, &tk);
+ }
+ if (!(tj & LT_FBYPORT)) {
+
+ /*
+ * The search by port number failed.
+ */
+ (void) snprintf(buf, bufl - 1,
+ "ERROR!!! no %s socket by port: %s",
+ PtNm[ti], port);
+ buf[bufl - 1] = '\0';
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = MkStrCpy(buf, &tk);
+ }
+ }
+ }
+ if (pem)
+ (void) PrtMsgX(pem, Pn, CleanupSrvr, 1);
+/*
+ * Exit successfully.
+ */
+ (void) PrtMsgX("OK", Pn, CleanupSrvr, 0);
+ return(0);
+}
+
+
+/*
+ * ClntCleanup() -- release client resources
+ */
+
+static void
+CleanupClnt()
+{
+ int tfd; /* temporary file descriptor */
+
+ if ((tfd = FdPara[LT_CLNT].fd) >= 0) {
+ (void) shutdown(tfd, 2);
+ (void) close(tfd);
+ FdPara[LT_CLNT].fd = -1;
+ }
+}
+
+
+/*
+ * CleanupSrvr() -- release server resources
+ */
+
+static void
+CleanupSrvr()
+{
+ int tfd; /* temporary file descriptor */
+ int ti; /* temporary index */
+ pid_t wpid; /* wait() PID */
+
+ if ((Ssock >= 0) && (Ssock != FdPara[LT_SRVR].fd)) {
+ (void) shutdown(Ssock, 2);
+ (void) close(Ssock);
+ Ssock = -1;
+ }
+ for (ti = 0; ti < NFDPARA; ti++) {
+ if ((tfd = FdPara[ti].fd) >= 0) {
+ (void) shutdown(tfd, 2);
+ (void) close(tfd);
+ FdPara[ti].fd = -1;
+ }
+ }
+ if (CPid > 0) {
+ wpid = wait3(NULL, WNOHANG, NULL);
+ if (wpid != CPid) {
+ kill(CPid, SIGKILL);
+ (void) wait3(NULL, WNOHANG, NULL);
+ }
+ CPid = (pid_t)0;
+ }
+}
+
+
+/*
+ * FindSock() -- find sockets with lsof
+ */
+
+static char *
+FindSock(fn)
+ int fn; /* function -- an LT_FBY* value */
+{
+ char buf[2048]; /* temporary buffer */
+ int bufl = sizeof(buf); /* size of buf[] */
+ char *cem; /* current error message pointer */
+ LTfldo_t *cmdp; /* command pointer */
+ LTfldo_t *fop; /* field output pointer */
+ int nf; /* number of fields */
+ int nl; /* name length */
+ LTfldo_t *nmp; /* name pointer */
+ char *opv[5]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message pointer */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ int pl; /* port length */
+ int px; /* process index -- LT_CLNT or
+ * LT_SRVR */
+ char *tcp, *tcp1; /* temporary character pointers */
+ int ti, tj; /* temporary integers */
+ LTfldo_t *typ; /* file type pointer */
+/*
+ * Check the function and determine the first lsof option from it.
+ */
+ ti = 0;
+ switch (fn) {
+ case LT_FBYHN:
+ opv[ti++] = "-P";
+ for (tj = 0; tj < NFDPARA; tj++) {
+ (void) snprintf(buf, bufl - 1, "-i@%s:%s", FdPara[tj].host,
+ FdPara[tj].port);
+ buf[bufl - 1] = '\0';
+ opv[ti++] = MkStrCpy(buf, &pl);
+ }
+ break;
+ case LT_FBYIP:
+ opv[ti++] = "-Pn";
+ for (tj = 0; tj < NFDPARA; tj++) {
+ (void) snprintf(buf, bufl - 1, "-i@%s:%s", FdPara[tj].ipaddr,
+ FdPara[tj].port);
+ buf[bufl - 1] = '\0';
+ opv[ti++] = MkStrCpy(buf, &pl);
+ }
+ break;
+ case LT_FBYPORT:
+ opv[ti++] = "-P";
+ for (tj = 0; tj < NFDPARA; tj++) {
+ (void) snprintf(buf, bufl - 1, "-i:%s", FdPara[tj].port);
+ buf[bufl - 1] = '\0';
+ opv[ti++] = MkStrCpy(buf, &pl);
+ }
+ break;
+ default:
+ (void) snprintf(buf, bufl - 1,
+ "ERROR!!! illegal FindSock() function: %d", fn);
+ buf[bufl - 1] = '\0';
+ return(MkStrCpy(buf, &ti));
+ }
+/*
+ * Complete the option vector and start lsof execution.
+ */
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+ opv[ti] = (char *)NULL;
+ if ((cem = ExecLsof(opv)))
+ return(cem);
+/*
+ * Read lsof output.
+ */
+ while ((((FdPara[LT_CLNT].ff & fn) == 0)
+ || ((FdPara[LT_SRVR].ff & fn) == 0))
+ && (fop = RdFrLsof(&nf, &cem))
+ ) {
+ if (cem) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || ((pid != CPid) && (pid != MyPid)))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line.
+ *
+ * Identify the process -- client or server.
+ */
+ if (!pids)
+ break;
+ if (pid == CPid)
+ px = LT_CLNT;
+ else if (pid == MyPid)
+ px = LT_SRVR;
+ else
+ break;
+ /*
+ * Make sure the FD matches the identified process.
+ */
+ if (strcmp(fop->v, FdPara[px].fds))
+ break;
+ /*
+ * Scan for name and type.
+ */
+ nmp = typ = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_NAME:
+ nmp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the type of the file.
+ */
+ if (!typ
+ || (strcasecmp(typ->v, "inet") && strcasecmp(typ->v, "ipv4"))
+ ) {
+ break;
+ }
+ /*
+ * Check the addess in the name, based on the calling function.
+ */
+ if (!nmp)
+ break;
+ tcp = nmp->v;
+ switch (fn) {
+ case LT_FBYHN:
+ if (((nl = FdPara[px].hlen) <= 0)
+ || !(tcp1 = FdPara[px].host)
+ || strncasecmp(tcp, tcp1, nl)
+ ) {
+ break;
+ }
+ tcp += nl;
+ if ((*tcp++ != ':')
+ || !(tcp1 = FdPara[px].port)
+ || ((pl = FdPara[px].plen) <= 0)
+ || strncmp(tcp, tcp1, pl)
+ ) {
+ break;
+ }
+ tcp += pl;
+ if ((*tcp == '-') || (*tcp == ' ') || !*tcp) {
+ FdPara[px].ff |= LT_FBYHN;
+ }
+ break;
+ case LT_FBYIP:
+ if (((nl = FdPara[px].ilen) <= 0)
+ || !(tcp1 = FdPara[px].ipaddr)
+ || strncasecmp(tcp, tcp1, nl)
+ ) {
+ break;
+ }
+ tcp += nl;
+ if ((*tcp++ != ':')
+ || !(tcp1 = FdPara[px].port)
+ || ((pl = FdPara[px].plen) <= 0)
+ || strncmp(tcp, tcp1, pl)
+ ) {
+ break;
+ }
+ tcp += pl;
+ if ((*tcp == '-') || (*tcp == ' ') || !*tcp) {
+ FdPara[px].ff |= LT_FBYIP;
+ }
+ break;
+ case LT_FBYPORT:
+ if (!(tcp = strchr(tcp, ':')))
+ break;
+ tcp++;
+ if (!(tcp1 = FdPara[px].port)
+ || ((pl = FdPara[px].plen) <= 0)
+ || strncmp(tcp, tcp1, pl)
+ ) {
+ break;
+ }
+ tcp += pl;
+ if ((*tcp == '-') || (*tcp == ' ') || !*tcp) {
+ FdPara[px].ff |= LT_FBYPORT;
+ }
+ break;
+ }
+ }
+ }
+/*
+ * Clean up and return.
+ */
+ (void) StopLsof();
+ return(pem);
+}
+
+
+/*
+ * HandleClntAlarm() -- handle client alarm
+ */
+
+static SIGHANDLER_T
+HandleClntAlarm(sig)
+ int sig; /* the signal (SIGALRM) */
+{
+ (void) PrtMsgX("ERROR!!! client caught an alarm signal", Pn,
+ CleanupClnt, 1);
+}
+
+
+/*
+ * Handle SrvrAlarm() -- handle server alarm
+ */
+
+static SIGHANDLER_T
+HandleSrvrAlarm(sig)
+ int sig; /* the signal (SIGALRM) */
+{
+ (void) PrtMsgX("ERROR!!! server caught an alarm signal.", Pn,
+ CleanupSrvr, 1);
+}
+
+
+/*
+ * StartClnt() -- start network client
+ */
+
+static void
+StartClnt(cad)
+ struct sockaddr_in *cad; /* connection address */
+{
+ struct sockaddr_in ba; /* bind address */
+ int br; /* bytes read */
+ char buf[2048]; /* temporary buffer */
+ int bufl = sizeof(buf); /* size of buf[] */
+ int cr; /* connect() reply */
+ char *em; /* error message pointer */
+ int fd = FdPara[LT_CLNT].fd; /* client's socket FD */
+/*
+ * Close the server's sockets.
+ */
+ if ((Ssock >= 0) && (Ssock != FdPara[LT_SRVR].fd)) {
+ (void) close(Ssock);
+ Ssock = -1;
+ }
+ if (FdPara[LT_SRVR].fd >= 0) {
+ (void) close(FdPara[LT_SRVR].fd);
+ FdPara[LT_SRVR].fd = -1;
+ }
+/*
+ * Bind to the local address.
+ */
+ (void) memcpy((void *)&ba, (void *)&Myad, sizeof(ba));
+ if (bind(fd, (struct sockaddr *)&ba, sizeof(ba)) < 0) {
+ em = "bind";
+
+client_errno:
+
+ (void) snprintf(buf, bufl - 1,
+ "ERROR!!! client %s error: %s", em, strerror(errno));
+ buf[bufl - 1] = '\0';
+ (void) PrtMsgX(em, Pn, CleanupClnt, 1);
+ }
+/*
+ * Set an alarm timeout and connect to the server.
+ */
+ (void) signal(SIGALRM, HandleClntAlarm);
+ (void) alarm(ALARMTM);
+ cr = connect(fd, (struct sockaddr *)cad, sizeof(struct sockaddr_in));
+ (void) alarm(0);
+ (void) signal(SIGALRM, SIG_DFL);
+ if (cr) {
+ em = "connect";
+ goto client_errno;
+ }
+/*
+ * Sleep until the socket closes or the parent kills the process.
+ */
+ for (br = 0; br >= 0;) {
+ sleep(1);
+ br = read(fd, buf, bufl);
+ }
+ (void) CleanupClnt();
+ exit(0);
+}
--- /dev/null
+/*
+ * LTszoff.c -- Lsof Test small file (< 32 bits) size and offset tests
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+
+/*
+ * Pre-definitions that might be undefined by dialects
+ */
+
+#define OFFTST_STAT 1 /* offset tests status */
+
+
+#if defined(LT_DIAL_linux)
+/*
+ * Linux-specific items
+ */
+
+#undef OFFTST_STAT
+#define OFFTST_STAT 0 /* Linux lsof may not be able to report
+ * offsets -- see the function
+ * ck_Linux_offset_support() */
+
+_PROTOTYPE(static int ck_Linux_offset_support,(void));
+#endif /* defined(LT_DIAL_linux) */
+
+
+/*
+ * Local definitions
+ */
+
+#define TYTST_SZ 0 /* size test type */
+#define TYTST_0to 1 /* 0t offset test type */
+#define TYTST_0xo 2 /* 0x offset test type */
+#define TSTFSZ 32768 /* test file size */
+
+
+/*
+ * Globals
+ */
+
+int Fd = -1; /* test file descriptor; open if >= 0 */
+pid_t MyPid = (pid_t)0; /* PID of this process */
+char *Path = (char *)NULL; /* test file path; none if NULL */
+char *Pn = (char *)NULL; /* program name */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static char *testlsof,(int tt, char *opt, char *xval));
+
+
+/*
+ * Main program
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ int do_offt = OFFTST_STAT; /* do offset tests if == 1 */
+ char *em; /* error message pointer */
+ int ti; /* temporary index */
+ char *tcp; /* temporary character pointer */
+ char *tstsz = (char *)NULL; /* size test status */
+ char *tst0to = (char *)NULL; /* offset 0t form test */
+ char *tst0xo = (char *)NULL; /* offset 0x form test */
+ int xv = 0; /* exit value */
+ char xbuf[64]; /* expected value buffer */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "hp:", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h] [-p path]", Pn);
+ PrtMsg (" -h print help (this panel)", Pn);
+ PrtMsgX (" -p path define test file path", Pn, cleanup, xv);
+ }
+
+#if defined(LT_DIAL_linux)
+/*
+ * If this is Linux, see if lsof can report file offsets.
+ */
+ do_offt = ck_Linux_offset_support();
+#endif /* defined(LT_DIAL_linux) */
+
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * If a path was supplied in an "-p path" option, use it. Otherwise construct
+ * a path in the CWD.
+ */
+ if (!(Path = LTopt_p)) {
+ (void) snprintf(buf, sizeof(buf) - 1, "./config.LTszoff%ld",
+ (long)MyPid);
+ buf[sizeof(buf) - 1] = '\0';
+ Path = MkStrCpy(buf, &ti);
+ }
+/*
+ * Open a new test file at the specified path.
+ */
+ (void) unlink(Path);
+ if ((Fd = open(Path, O_RDWR|O_CREAT, 0600)) < 0) {
+ (void) fprintf(stderr, "ERROR!!! can't open %s\n", Path);
+
+print_file_error:
+
+ MsgStat = 1;
+ (void) snprintf(buf, sizeof(buf) - 1, " Errno %d: %s",
+ errno, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Write the test file to its expected size.
+ */
+ for (ti = 0; ti < sizeof(buf); ti++) {
+ buf[ti] = (char)(ti & 0xff);
+ }
+ for (ti = 0; ti < TSTFSZ; ti += sizeof(buf)) {
+ if (write(Fd, buf, sizeof(buf)) != sizeof(buf)) {
+ (void) fprintf(stderr, "ERROR!!! can't write %d bytes to %s\n",
+ (int)sizeof(buf), Path);
+ goto print_file_error;
+ }
+ }
+/*
+ * Fsync() the file.
+ */
+ if (fsync(Fd)) {
+ (void) fprintf(stderr, "ERROR!!! can't fsync %s\n", Path);
+ goto print_file_error;
+ }
+/*
+ * Do the tests. Skip offset tests as indicated.
+ */
+ (void) snprintf(xbuf, sizeof(xbuf) - 1, "%d", TSTFSZ);
+ xbuf[sizeof(xbuf) - 1] = '\0';
+ if ((tstsz = testlsof(TYTST_SZ, "-s", xbuf)))
+ (void) PrtMsg(tstsz, Pn);
+ if (do_offt) {
+ (void) snprintf(xbuf, sizeof(xbuf) - 1, "0t%d", TSTFSZ);
+ xbuf[sizeof(xbuf) - 1] = '\0';
+ if ((tst0to = testlsof(TYTST_0to, "-o", xbuf)))
+ (void) PrtMsg(tst0to, Pn);
+ (void) snprintf(xbuf, sizeof(xbuf) - 1, "0x%x", TSTFSZ);
+ xbuf[sizeof(xbuf) - 1] = '\0';
+ if ((tst0xo = testlsof(TYTST_0xo, "-oo2", xbuf)))
+ (void) PrtMsg(tst0to, Pn);
+ } else {
+ PrtMsg("WARNING!!! lsof can't return file offsets for this dialect,",
+ Pn);
+ PrtMsg(" so offset tests have been disabled.", Pn);
+ }
+/*
+ * Compute exit value and exit.
+ */
+ if (tstsz || tst0to || tst0xo) {
+ tcp = (char *)NULL;
+ xv = 1;
+ } else {
+ tcp = "OK";
+ xv = 0;
+ }
+ (void) PrtMsgX(tcp, Pn, cleanup, xv);
+ return(0);
+}
+
+
+#if defined(LT_DIAL_linux)
+/*
+ * ck_Linux_offset_support() -- see if lsof can report offsets for this
+ * Linux implementation
+ */
+
+static int
+ck_Linux_offset_support()
+{
+ char buf[1024]; /* lsof output line buffer */
+ int bufl = sizeof(buf); /* size of buf[] */
+ char *opv[5]; /* option vector for lsof */
+ int rv = 1; /* return value:
+ * 0 == no lsof offset support
+ * 1 == lsof offset support */
+/*
+ * Ask lsof to report the test's FD zero offset.
+ */
+ if (IsLsofExec())
+ return(0);
+ opv[0] = "-o";
+ snprintf(buf, bufl - 1, "-p%d", (int)getpid());
+ opv[1] = buf;
+ opv[2] = "-ad0";
+ opv[3] = "+w";
+ opv[4] = (char *)NULL;
+ if (ExecLsof(opv))
+ return(0);
+/*
+ * Read the lsof output. Look for a line with "WARNING: can't report offset"
+ * in it. If it is found, then this Linux lsof can't report offsets.
+ */
+ while(fgets(buf, bufl - 1, LsofFs)) {
+ if (strstr(buf, "WARNING: can't report offset")) {
+ rv = 0;
+ break;
+ }
+ }
+ (void) StopLsof();
+ return(rv);
+}
+#endif /* defined(LT_DIAL_linux) */
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+ if (Fd >= 0) {
+ (void) close(Fd);
+ Fd = -1;
+ if (Path) {
+ (void) unlink(Path);
+ Path = (char *)NULL;
+ }
+ }
+}
+
+
+/*
+ * testlsof() -- test the open file with lsof
+ */
+
+static char *
+testlsof(tt, opt, xval)
+ int tt; /* test type -- TYTST_* symbol */
+ char *opt; /* extra lsof options */
+ char *xval; /* expected value */
+{
+ char buf[2048]; /* temporary buffer */
+ char *cem; /* current error message pointer */
+ LTfldo_t *cmdp; /* command pointer */
+ LTfldo_t *devp; /* device pointer */
+ int ff = 0; /* file found status */
+ LTfldo_t *fop; /* field output pointer */
+ char ibuf[64]; /* inode number buffer */
+ LTfldo_t *inop; /* inode number pointer */
+ LTdev_t lsofdc; /* lsof device components */
+ int nf; /* number of fields */
+ LTfldo_t *offp; /* offset pointer */
+ char *opv[4]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message pointer */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ struct stat sb; /* stat(2) buffer */
+ LTdev_t stdc; /* stat(2) device components */
+ LTfldo_t *szp; /* size pointer */
+ char *tcp; /* temporary character pointer */
+ int ti; /* temporary integer */
+ char *tnm1, *tnm2; /* test names */
+ int ts = 0; /* test status flag */
+ LTfldo_t *typ; /* file type pointer */
+/*
+ * Check the test type.
+ */
+ switch (tt) {
+ case TYTST_SZ:
+ tnm1 = "";
+ tnm2 = " size";
+ break;
+ case TYTST_0to:
+ tnm1 = " 0t";
+ tnm2 = " offset";
+ break;
+ case TYTST_0xo:
+ tnm1 = " 0x";
+ tnm2 = " offset";
+ break;
+ default:
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! illegal test type: %d", tt);
+ buf[sizeof(buf) - 1] = '\0';
+ (void) PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Get test file's information.
+ */
+ if (stat(Path, &sb)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! can't stat(2) %s: %s", Path, strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ PrtMsgX(buf, Pn, cleanup, 1);
+ }
+/*
+ * Extract components from test file's stat buffer.
+ */
+ if ((cem = ConvStatDev(&sb.st_dev, &stdc)))
+ PrtMsgX(buf, Pn, cleanup, 1);
+ (void) snprintf(ibuf, sizeof(ibuf) - 1, "%u", (unsigned int)sb.st_ino);
+ ibuf[sizeof(ibuf) - 1] = '\0';
+/*
+ * Complete the option vector and start lsof execution.
+ */
+ ti = 0;
+ if (opt && *opt)
+ opv[ti++] = opt;
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#else /* !defined(USE_LSOF_C_OPT) */
+ opv[ti++] = "--";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+ opv[ti++] = Path;
+ opv[ti] = (char *)NULL;
+ if ((cem = ExecLsof(opv)))
+ return(cem);
+/*
+ * Read lsof output.
+ */
+ while (!ff && !cem && (fop = RdFrLsof(&nf, &cem))) {
+ if (cem) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != MyPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line. Make sure its number matches the
+ * test file's descriptor number.
+ */
+ if (!pids)
+ break;
+ for (ti = 0, tcp = fop->v; *tcp; tcp++) {
+
+ /*
+ * Convert file descriptor to a number.
+ */
+ if (*tcp == ' ')
+ continue;
+ if (((int)*tcp < (int)'0') || ((int)*tcp > (int)'9')) {
+ ti = -1;
+ break;
+ }
+ ti = (ti * 10) + (int)*tcp - (int)'0';
+ }
+ if (Fd != ti)
+ break;
+ /*
+ * Scan for device, inode, offset, size and type fields.
+ */
+ devp = inop = offp = szp = typ = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_DEVN:
+ devp = fop;
+ break;
+ case LSOF_FID_INODE:
+ inop = fop;
+ break;
+ case LSOF_FID_OFFSET:
+ offp = fop;
+ break;
+ case LSOF_FID_SIZE:
+ szp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the results of the file descriptor field scan.
+ */
+ if (!devp || !inop || !typ)
+ break;
+ if (strcasecmp(typ->v, "reg") && strcasecmp(typ->v, "vreg"))
+ break;
+ if ((cem = ConvLsofDev(devp->v, &lsofdc))) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ break;
+ }
+ if ((stdc.maj != lsofdc.maj)
+ || (stdc.min != lsofdc.min)
+ || (stdc.unit != lsofdc.unit)
+ || strcmp(inop->v, ibuf)
+ ) {
+ break;
+ }
+ /*
+ * The specified file has been located. Do the specified test.
+ */
+ ff = 1;
+ fop = (tt == TYTST_SZ) ? szp : offp;
+ if (!fop) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! %s%s test, but no lsof%s", tnm1, tnm2, tnm2);
+ ts = 1;
+ } else if (strcmp(fop->v, xval)) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! %s%s mismatch: expected %s, got %s",
+ tnm1, tnm2, xval, fop->v);
+ ts = 1;
+ }
+ if (ts) {
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = cem;
+ }
+ break;
+ }
+ }
+ (void) StopLsof();
+ if (!ff) {
+ (void) snprintf(buf, sizeof(buf) - 1,
+ "ERROR!!! test file %s not found by lsof", Path);
+ buf[sizeof(buf) - 1] = '\0';
+ cem = MkStrCpy(buf, &ti);
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+ return(pem);
+}
--- /dev/null
+/*
+ * LTunix.c -- Lsof Test UNIX domain socket test
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by V. Abell.
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2002 Purdue Research Foundation.\nAll rights reserved.\n";
+#endif
+
+#include "LsofTest.h"
+#include "lsof_fields.h"
+
+#include <sys/socket.h>
+#include <sys/un.h>
+
+
+/*
+ * Local definitions
+ */
+
+#if !defined(MAXPATHLEN)
+#define MAXPATHLEN 1024 /* maximum path length */
+#endif /* !defined(MAXPATHLEN) */
+
+
+/*
+ * Globals
+ */
+
+pid_t MyPid = (pid_t)0; /* PID of this process */
+char *Pn = (char *)NULL; /* program name */
+int SpFd[2] = {-1,-1}; /* socket pair FDs */
+char *Path[2] = {(char *)NULL, (char *)NULL};
+ /* socket pair paths */
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static void cleanup,(void));
+_PROTOTYPE(static char *FindUsocks,(void));
+
+
+/*
+ * Main program
+ */
+
+int
+main(argc, argv)
+ int argc; /* argument count */
+ char *argv[]; /* arguments */
+{
+ char buf[2048]; /* temporary buffer */
+ char cwd[MAXPATHLEN + 1]; /* CWD buffer */
+ char *em; /* error message pointer */
+ int ti, tj; /* temporary indexes */
+ struct sockaddr_un ua; /* UNIX socket address */
+ int xv = 0; /* exit value */
+/*
+ * Get program name and PID, issue start message, and build space prefix.
+ */
+ if ((Pn = strrchr(argv[0], '/')))
+ Pn++;
+ else
+ Pn = argv[0];
+ MyPid = getpid();
+ (void) printf("%s ... ", Pn);
+ (void) fflush(stdout);
+ PrtMsg((char *)NULL, Pn);
+/*
+ * Process arguments.
+ */
+ if (ScanArg(argc, argv, "h", Pn))
+ xv = 1;
+ if (xv || LTopt_h) {
+ (void) PrtMsg("usage: [-h]", Pn);
+ PrtMsgX(" -h print help (this panel)", Pn, cleanup, xv);
+ }
+/*
+ * See if lsof can be executed and can access kernel memory.
+ */
+ if ((em = IsLsofExec()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+ if ((em = CanRdKmem()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * Construct the socket paths.
+ */
+
+#if defined(USE_GETCWD)
+ if (!getcwd(cwd, sizeof(cwd)))
+#else /* ! defined(USE_GETCWD) */
+ if (!getwd(cwd))
+#endif /* defined(USE_GETCWD) */
+
+ {
+ em = "ERROR!!! can't get CWD";
+ goto print_errno;
+ }
+ cwd[sizeof(cwd) - 1] = '\0';
+ if ((strlen(cwd) + strlen("/config.LT#U9223372036854775807") + 1)
+ > sizeof(ua.sun_path))
+ {
+ strncpy(cwd, "/tmp", sizeof(cwd) - 1);
+ }
+ for (ti = 0; ti < 2; ti++) {
+ (void) snprintf(buf, sizeof(buf) - 1, "%s/config.LT%dU%ld", cwd, ti,
+ (long)MyPid);
+ buf[sizeof(buf) - 1] = '\0';
+ Path[ti] = MkStrCpy(buf, &tj);
+ (void) unlink(Path[ti]);
+ }
+/*
+ * Get two UNIX domain socket FDs.
+ */
+ for (ti = 0; ti < 2; ti++) {
+ if ((SpFd[ti] = socket(AF_UNIX, SOCK_STREAM, PF_UNSPEC)) < 0) {
+ em = "socket";
+
+print_errno_by_ti:
+
+ (void) snprintf(buf, sizeof(buf) - 1, "ERROR!!! %s(%s) failure",
+ em, Path[ti]);
+ buf[sizeof(buf) - 1] = '\0';
+ em = buf;
+
+print_errno:
+
+ PrtMsg(em, Pn);
+ (void) snprintf(buf, sizeof(buf) - 1, " Errno %d: %s", errno,
+ strerror(errno));
+ buf[sizeof(buf) - 1] = '\0';
+ PrtMsgX(buf, Pn, cleanup, 1);
+ }
+ }
+/*
+ * Bind file system names to the sockets.
+ */
+ for (ti = 0; ti < 2; ti++) {
+ (void) memset((void *)&ua, 0, sizeof(ua));
+ ua.sun_family = AF_UNIX;
+ (void) strncpy(ua.sun_path, Path[ti], sizeof(ua.sun_path));
+ ua.sun_path[sizeof(ua.sun_path) - 1] = '\0';
+ if (bind(SpFd[ti], (struct sockaddr *)&ua, sizeof(ua)) < 0) {
+ em = "bind";
+ goto print_errno_by_ti;
+ }
+ }
+/*
+ * Look for the open UNIX domain socket files with lsof.
+ */
+ if ((em = FindUsocks()))
+ (void) PrtMsgX(em, Pn, cleanup, 1);
+/*
+ * Exit successfully.
+ */
+ (void) PrtMsgX("OK", Pn, cleanup, 0);
+ return(0);
+}
+
+
+/*
+ * cleanup() -- release resources
+ */
+
+static void
+cleanup()
+{
+ int ti;
+
+ for (ti = 0; ti < 2; ti++) {
+ if (SpFd[ti] >= 0) {
+ (void) close(SpFd[ti]);
+ SpFd[ti] = -1;
+ }
+ if (Path[ti]) {
+ (void) unlink(Path[ti]);
+ (void) free((void *)Path[ti]);
+ Path[ti] = (char *)NULL;
+ }
+ }
+}
+
+
+/*
+ * FindUsocks() -- find UNIX sockets with lsof
+ */
+
+static char *
+FindUsocks()
+{
+ char buf[2048]; /* temporary buffer */
+ char *cem; /* current error message pointer */
+ LTfldo_t *cmdp; /* command pointer */
+ int ff[2]; /* file-found flags */
+ LTfldo_t *fop; /* field output pointer */
+ int nf; /* number of fields */
+ int nl; /* name length */
+ LTfldo_t *nmp; /* name pointer */
+ char *opv[5]; /* option vector for ExecLsof() */
+ char *pem = (char *)NULL; /* previous error message pointer */
+ pid_t pid; /* PID */
+ int pids = 0; /* PID found status */
+ char *tcp; /* temporary character pointer */
+ int ti, tj; /* temporary integers */
+ LTfldo_t *typ; /* file type pointer */
+/*
+ * Build the option vector and start lsof execution.
+ */
+ ff[0] = ff[1] = ti = 0;
+ opv[ti++] = "-aU";
+ opv[ti++] = "-p";
+ (void) snprintf(buf, sizeof(buf) - 1, "%ld", (long)MyPid);
+ buf[sizeof(buf) - 1] = '\0';
+ opv[ti++] = MkStrCpy(buf, &tj);
+
+#if defined(USE_LSOF_C_OPT)
+ opv[ti++] = "-C";
+#endif /* defined(USE_LSOF_C_OPT) */
+
+ opv[ti] = (char *)NULL;
+ if ((cem = ExecLsof(opv)))
+ return(cem);
+/*
+ * Read lsof output.
+ */
+ while (((ff[0] + ff[1]) < 2) && (fop = RdFrLsof(&nf, &cem))) {
+ if (cem) {
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ return(cem);
+ }
+ switch (fop->ft) {
+ case LSOF_FID_PID:
+
+ /*
+ * This is a process information line.
+ */
+ pid = (pid_t)atoi(fop->v);
+ pids = 1;
+ cmdp = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_CMD:
+ cmdp = fop;
+ break;
+ }
+ }
+ if (!cmdp || (pid != MyPid))
+ pids = 0;
+ break;
+ case LSOF_FID_FD:
+
+ /*
+ * This is a file descriptor line. Make sure its number matches a
+ * test file descriptor number.
+ */
+ if (!pids)
+ break;
+ for (ti = 0, tcp = fop->v; *tcp; tcp++) {
+
+ /*
+ * Convert file descriptor to a number.
+ */
+ if (*tcp == ' ')
+ continue;
+ if (((int)*tcp < (int)'0') || ((int)*tcp > (int)'9')) {
+ ti = -1;
+ break;
+ }
+ ti = (ti * 10) + (int)*tcp - (int)'0';
+ }
+ for (tj = 0; tj < 2; tj++) {
+ if (ff[tj])
+ continue;
+ if (SpFd[tj] == ti)
+ break;
+ }
+ if (tj >= 2)
+ break;
+ /*
+ * Scan for name and type.
+ */
+ nmp = typ = (LTfldo_t *)NULL;
+ for (fop++, ti = 1; ti < nf; fop++, ti++) {
+ switch (fop->ft) {
+ case LSOF_FID_NAME:
+ nmp = fop;
+ break;
+ case LSOF_FID_TYPE:
+ typ = fop;
+ break;
+ }
+ }
+ /*
+ * Check the type of the file.
+ */
+ if (!typ || strcasecmp(typ->v, "unix"))
+ break;
+ /*
+ * Look for the name.
+ */
+ if (!nmp)
+ break;
+ nl = strlen(Path[tj]);
+ for (tcp = nmp->v; tcp; tcp = strchr(tcp + 1, '/')) {
+ if (!strncmp(tcp, Path[tj], nl)) {
+
+ /*
+ * Mark a file as found.
+ */
+ ff[tj] = 1;
+ break;
+ }
+ }
+ }
+ }
+/*
+ * Clean up and return.
+ */
+ (void) StopLsof();
+ for (ti = 0; ti < 2; ti++) {
+ if (ff[tj])
+ continue;
+ (void) snprintf(buf, sizeof(buf) - 1, "ERROR!!! not found: %s",
+ Path[ti]);
+ buf[sizeof(buf) - 1] = '\0';
+ if (pem)
+ (void) PrtMsg(pem, Pn);
+ pem = MkStrCpy(buf, &tj);
+ }
+ return(pem);
+}
--- /dev/null
+/*
+ * LsofTest.h -- header file for lsof tests
+ */
+
+
+/*
+ * Copyright 2002 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+/*
+ * $Id: LsofTest.h,v 1.12 2008/07/05 16:21:07 abe Exp $
+ */
+
+
+#if !defined(LSOF_TEST_H)
+#define LSOF_TEST_H 1
+
+
+/*
+ * The _PROTOTYPE macro provides strict ANSI C prototypes if __STDC__
+ * is defined, and old-style K&R prototypes otherwise.
+ *
+ * (With thanks to Andy Tanenbaum)
+ */
+
+# if defined(__STDC__)
+#define _PROTOTYPE(function, params) function params
+# else /* !defined(__STDC__) */
+#define _PROTOTYPE(function, params) function()
+# endif /* defined(__STDC__) */
+
+
+/*
+ * The following define keeps gcc>=2.7 from complaining about the failure
+ * of the Exit() function to return.
+ *
+ * Paul Eggert <eggert@twinsun.com> supplied it.
+ */
+
+# if defined(__GNUC__) && !(__GNUC__<2 || (__GNUC__==2 && __GNUC_MINOR__<7))
+#define exiting __attribute__((__noreturn__))
+# else /* !gcc || gcc<2.7 */
+#define exiting
+# endif /* gcc && gcc>=2.7 */
+
+
+/*
+ * Necessary header files.
+ */
+
+#include <stdio.h>
+#include <ctype.h>
+#include <errno.h>
+#include <signal.h>
+
+#include <sys/types.h>
+#include <sys/param.h>
+#include <sys/stat.h>
+
+
+/*
+ * Definitions that may be revoked by a particular dialect.
+ */
+
+#define USE_GETCWD /* use the POSIX getcwd() function in
+ * place of getwd() */
+#define USE_LSOF_C_OPT /* use lsof's -C option */
+#undef USE_LSOF_X_OPT /* don't use lsof's -X option */
+
+
+# if defined(LT_DIAL_aix)
+/*
+ * AIX-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/access.h>
+#undef USE_LSOF_C_OPT
+#define USE_LSOF_X_OPT
+# endif /* defined(LT_DIAL_aix) */
+
+
+# if defined(LT_DIAL_bsdi)
+/*
+ * OpenBSD-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/wait.h>
+# endif /* defined(LT_DIAL_bsdi) */
+
+
+# if defined(LT_DIAL_darwin)
+/*
+ * Darwin-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/wait.h>
+#undef USE_LSOF_C_OPT
+# endif /* defined(LT_DIAL_darwin) */
+
+
+# if defined(LT_DIAL_du)
+/*
+ * DEC_OSF/1|Digital_UNIX|Tru64_UNIX-specific items
+ */
+
+#include <fcntl.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/wait.h>
+
+# if LT_VERS<50000
+#define snprintf snpf /* use lsof's snpf() */
+# endif /* LT_VERS<50000 */
+# endif /* defined(LT_DIAL_du) */
+
+
+# if defined(LT_DIAL_freebsd)
+/*
+ * FreeBSD-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/wait.h>
+# endif /* defined(LT_DIAL_freebsd) */
+
+
+# if defined(LT_DIAL_linux)
+/*
+ * Linux-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/wait.h>
+#undef USE_LSOF_C_OPT
+# endif /* defined(LT_DIAL_linux) */
+
+
+# if defined(LT_DIAL_hpux)
+/*
+ * HP-UX-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <strings.h>
+#include <unistd.h>
+# endif /* defined(LT_DIAL_hpux) */
+
+
+# if defined(LT_DIAL_netbsd)
+/*
+ * NetBSD-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/wait.h>
+# endif /* defined(LT_DIAL_netbsd) */
+
+
+# if defined(LT_DIAL_openbsd)
+/*
+ * OpenBSD-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/wait.h>
+# endif /* defined(LT_DIAL_openbsd) */
+
+
+# if defined(LT_DIAL_ou)
+/*
+ * OpenUNIX-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+# endif /* defined(LT_DIAL_ou) */
+
+
+# if defined(LT_DIAL_osr)
+/*
+ * OSR-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+# endif /* defined(LT_DIAL_osr) */
+
+
+# if defined(LT_DIAL_ns)
+/*
+ * NEXTSTEP-specific items
+ */
+
+#include <stdlib.h>
+#include <string.h>
+#include <sys/file.h>
+#include <sys/wait.h>
+
+typedef int pid_t;
+#define snprintf snpf
+
+#undef USE_GETCWD
+# endif /* defined(LT_DIAL_ns) */
+
+
+# if defined(LT_DIAL_solaris)
+/*
+ * Solaris-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <strings.h>
+#include <unistd.h>
+#include <sys/wait.h>
+
+# if defined(LT_VPATH)
+#undef USE_LSOF_C_OPT
+#endif /* defined(LT_VPATH) */
+# endif /* defined(LT_DIAL_solaris) */
+
+
+# if defined(LT_DIAL_uw)
+/*
+ * UnixWare-specific items
+ */
+
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+# endif /* defined(LT_DIAL_uw) */
+
+
+/*
+ * Local definitions, including ones may have been left undefined by
+ * dialect-specific header files
+ */
+
+#define LT_DONT_DO_TEST "this test does not run on this dialect."
+#define LT_DEF_LSOF_PATH "../lsof"
+
+# if !defined(MAXPATHLEN)
+#define MAXPATHLEN 1024
+# endif /* !defined(MAXPATHLEN) */
+
+
+/*
+ * Local structure definitions
+ */
+
+typedef struct LTdev { /* local device parameters */
+ unsigned int maj; /* major device number */
+ unsigned int min; /* minor device number */
+ unsigned int unit; /* unit number (where applicable) */
+} LTdev_t;
+
+typedef struct LTfldo { /* lsof field output information */
+ char ft; /* field identifier (see the LSOF_FID_*
+ * definitions in ../lsof_fields.h) */
+ char *v; /* field value character string */
+} LTfldo_t;
+#define LT_FLDO_ALLOC 16 /* LTfldo_t allocation increment */
+
+
+/*
+ * Lsof test library global variable external declarations:
+ *
+ * these global variables may be found in LTlib.c.
+ */
+
+extern int LsofFd; /* lsof pipe FD */
+extern FILE *LsofFs; /* stream for lsof pipe FD */
+extern char *LsofPath; /* path to lsof executable */
+extern pid_t LsofPid; /* PID of lsof child process */
+extern int LTopt_h; /* "-h" option's switch value */
+extern char *LTopt_p; /* "-p path" option's path value */
+extern int MsgStat; /* message status */
+
+
+/*
+ * External declarations
+ */
+
+extern int errno; /* error number */
+
+
+/*
+ * Lsof test library function prototypes:
+ *
+ * these functions may be found in LTlib.c.
+ */
+
+_PROTOTYPE(extern char *CanRdKmem,(void));
+_PROTOTYPE(extern char *ConvStatDev,(dev_t *dev, LTdev_t *ldev));
+_PROTOTYPE(extern char *ConvLsofDev,(char *dev, LTdev_t *ldev));
+_PROTOTYPE(extern char *ExecLsof,(char **opt));
+_PROTOTYPE(extern char *IsLsofExec,(void));
+_PROTOTYPE(extern void LTlibClean,(void));
+_PROTOTYPE(extern char *MkStrCpy,(char *src, int *len));
+_PROTOTYPE(extern LTfldo_t *RdFrLsof,(int *nf, char **em));
+_PROTOTYPE(extern void PrtMsg,(char *mp, char *pn));
+_PROTOTYPE(extern void PrtMsgX,(char *mp, char *pn, void (*f)(), int xv));
+_PROTOTYPE(extern int ScanArg,(int ac, char *av[], char *opt, char *pn));
+_PROTOTYPE(extern void StopLsof,(void));
+
+#endif /* LSOF_TEST_H */
--- /dev/null
+# Makefile for testing lsof
+#
+# V. Abell
+# Purdue University
+#
+# $Id: Makefile,v 1.17 2005/05/17 00:40:53 abe Exp abe $
+
+DEBUG=
+CFLAGS= ${DEBUG} -I. -I..
+
+HDR= LsofTest.h
+
+CKTSTDB= CkTestDB
+CONFCFL= ./config.cflags
+CONFIG= ./config.cc ${CONFCFL} ./config.xobj
+LTOBJ= LTlib.o
+LTSRC= LTlib.c
+LIBOBJ= ${LTOBJ}
+
+BASTST= LTbasic
+STDTST= LTnlink LTsock LTszoff LTunix
+OPTTST= LTbigf LTdnlc LTlock LTnfs
+
+all: ${CKTSTDB} ${BASTST} ${STDTST} FRC
+ @./${CKTSTDB}; xv=$$?; \
+ if [ $$xv -ne 0 ]; then \
+ exit 1 ;\
+ fi
+ @rm -f config.LT*
+ -@err=0; \
+ echo ""; \
+ echo "Basic test:"; \
+ ./${BASTST}; \
+ if [ $$? -ne 0 ]; then \
+ exit 1; \
+ fi; \
+ echo ""; \
+ echo "Standard tests:"; \
+ for i in ${STDTST}; do \
+ ./$$i; \
+ if [ $$? -ne 0 ]; then \
+ err=`expr $$err + 1`; \
+ fi; \
+ done; \
+ if [ $$err -ne 0 ]; then \
+ echo "Failed tests: $$err"; \
+ echo ""; \
+ echo "See 00FAQ and 00TEST for more information."; \
+ else \
+ echo "All standard tests succeeded."; \
+ echo ""; \
+ grep LT_DIAL_darwin ${CONFCFL} > /dev/null 2>&1; \
+ if [ $$? -ne 0 ]; then \
+ echo "Suggestion: try the optional tests: \"make opt\""; \
+ echo ""; \
+ fi; \
+ fi;
+ @rm -f config.LT*
+
+auto: ckDB silent FRC
+
+ckDB: ${CKTSTDB} FRC
+ @echo "" | ./${CKTSTDB}; xv=$$?; \
+ if [ $$xv -ne 0 ]; then \
+ exit 1 ;\
+ fi
+
+clean: FRC
+ rm -f ${BASTST} ${STDTST} ${OPTTST} *.o *.err *.out config.LT*
+
+FRC:
+
+LTbasic: LTbasic.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTbasic.c \
+ ${LIBOBJ} `cat config.xobj` -o LTbasic
+
+LTbigf: LTbigf.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTbigf.c \
+ ${LIBOBJ} `cat config.xobj` -o LTbigf
+
+LTdnlc: LTdnlc.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTdnlc.c \
+ ${LIBOBJ} `cat config.xobj` -o LTdnlc
+
+LTlock: LTlock.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTlock.c \
+ ${LIBOBJ} `cat config.xobj` -o LTlock
+
+${LTOBJ}: ${HDR} ${LTSRC} config.cflags config.cc
+ `cat config.cc` ${CFLAGS} `cat config.cflags` -c ${LTSRC} \
+ -o ${LTOBJ}
+
+LTnfs: LTnfs.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTnfs.c \
+ ${LIBOBJ} `cat config.xobj` -o LTnfs
+
+LTnlink: LTnlink.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTnlink.c \
+ ${LIBOBJ} `cat config.xobj` -o LTnlink
+
+LTsock: LTsock.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTsock.c \
+ ${LIBOBJ} `cat config.xobj` -o LTsock `cat config.ldflags`
+
+LTszoff: LTszoff.c ${CONFIG} ${LIBOBJ} ${HDR}
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTszoff.c \
+ ${LIBOBJ} `cat config.xobj` -o LTszoff
+
+LTunix: LTunix.c ${CONFIG} ${LIBOBJ} ${HDR} config.ldflags
+ `cat config.cc` ${CFLAGS} `cat config.cflags` LTunix.c \
+ ${LIBOBJ} `cat config.xobj` -o LTunix `cat config.ldflags`
+
+opt: ${CKTSTDB} ${OPTTST} FRC
+ @rm -f config.LT*
+ -@err=0; \
+ echo ""; \
+ echo "Optional tests:"; \
+ for i in ${OPTTST}; do \
+ ./$$i; \
+ if [ $$? -ne 0 ]; then \
+ err=`expr $$err + 1`; \
+ fi; \
+ done; \
+ if [ $$err -ne 0 ]; then \
+ echo "Failed tests: $$err"; \
+ else \
+ echo "All optional tests succeeded."; \
+ fi; \
+ echo "";
+ @rm -f config.LT*
+
+optional: opt
+
+silent: ${BASTST} ${STDTST} FRC
+ @rm -f config.LT*
+ @err=0; \
+ ./${BASTST} > /dev/null 2>&1; \
+ if [ $$? -ne 0 ]; then \
+ exit 1; \
+ fi; \
+ for i in ${STDTST}; do \
+ ./$$i > /dev/null 2>&1; \
+ if [ $$? -ne 0 ]; then \
+ err=`expr $$err + 1`; \
+ fi; \
+ done; \
+ rm -f config.LT*; \
+ if [ $$err -ne 0 ]; then \
+ exit 1; \
+ fi
+
+spotless: clean
+ rm -f config.*
+
+standard: all
+
+std: all
+
+test: all
--- /dev/null
+# TestDB -- lsof test suite data base
+#
+# This file contains the sorted words from config.cflags, less any leading "-D"
+# strings, joined on one line.
+#
+# See Add2TestDB for a script that will build a line for this file.
+#
+# $Id: TestDB,v 1.33 2009/03/25 19:15:29 abe Exp $
+
+LT_AIXA=0 LT_BIGF LT_CC LT_DIAL_aix LT_KMEM LT_VERS=4320
+LT_AIXA=0 LT_BIGF LT_CC LT_DIAL_aix LT_KMEM LT_VERS=4330
+LT_AIXA=1 LT_BIGF LT_CC LT_DIAL_aix LT_K64 LT_KMEM LT_VERS=5000
+LT_AIXA=0 LT_BIGF LT_CC LT_DIAL_aix LT_KMEM LT_VERS=5100
+LT_AIXA=0 LT_BIGF LT_DIAL_aix LT_GCC LT_KMEM LT_VERS=5100
+LT_AIXA=1 LT_BIGF LT_CC LT_DIAL_aix LT_K64 LT_KMEM LT_VERS=5100
+LT_AIXA=0 LT_BIGF LT_CC LT_DIAL_aix LT_KMEM LT_VERS=5200
+LT_AIXA=0 LT_BIGF LT_DIAL_aix LT_GCC LT_KMEM LT_VERS=5200
+LT_AIXA=1 LT_BIGF LT_CC LT_DIAL_aix LT_K64 LT_KMEM LT_VERS=5200
+LT_AIXA=1 LT_BIGF LT_DIAL_aix LT_GCC LT_K64 LT_KMEM LT_VERS=5200
+LT_AIXA=1 LT_BIGF LT_CC LT_DIAL_aix LT_K64 LT_KMEM LT_VERS=5300
+LT_BIGF LT_DIAL_bsdi LT_GCC LT_KMEM LT_VERS=40100
+LT_BIGF LT_DIAL_bsdi LT_GCC LT_KMEM LT_VERS=40300
+LT_CC LT_DIAL_darwin LT_KMEM LT_VERS=140
+LT_CC LT_DIAL_darwin LT_KMEM LT_VERS=530
+LT_CC LT_DIAL_darwin LT_KMEM LT_VERS=600
+LT_CC LT_DIAL_darwin LT_KMEM LT_VERS=700
+LT_CC LT_DIAL_darwin LT_KMEM LT_VERS=800
+LT_CC LT_DIAL_darwin LT_VERS=800
+LT_BIGF LT_CC LT_DIAL_darwin LT_KMEM LT_VERS=800
+LT_CC LT_DIAL_darwin LT_VERS=900
+LT_BIGF LT_CC LT_DIAL_darwin LT_VERS=900
+LT_BIGF LT_CC LT_DIAL_du LT_K64 LT_KMEM LT_VERS=40000
+LT_BIGF LT_CC LT_DIAL_du LT_K64 LT_KMEM LT_VERS=50000
+LT_BIGF LT_CC LT_DIAL_du LT_K64 LT_KMEM LT_VERS=50100
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=4050
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=4060
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=4070
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=4080
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=4090
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=4100
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=4110
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=5000
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=5010
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=5020
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=5030
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=5040
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=5050
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=6000
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=6010
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=6020
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=7000
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=7010
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=7020
+LT_BIGF LT_CC LT_DIAL_freebsd LT_KMEM LT_VERS=8000
+LT_BIGF LT_CC LT_DIAL_hpux LT_KMEM LT_VERS=1020 _LARGEFILE64_SOURCE
+LT_BIGF LT_DIAL_hpux LT_GCC LT_KMEM LT_VERS=1020 _LARGEFILE64_SOURCE
+LT_BIGF LT_CC LT_DIAL_hpux LT_KMEM LT_VERS=1100 _LARGEFILE64_SOURCE
+LT_BIGF LT_CC LT_DIAL_hpux LT_K64 LT_KMEM LT_VERS=1100 _LARGEFILE64_SOURCE
+LT_BIGF LT_CC LT_DIAL_hpux LT_K64 LT_VERS=1111 _LARGEFILE64_SOURCE
+LT_BIGF LT_CC LT_DIAL_hpux LT_K64 LT_VERS=1123 _LARGEFILE64_SOURCE
+LT_BIGF LT_DIAL_hpux LT_GCC LT_K64 LT_VERS=1123 _LARGEFILE64_SOURCE
+LT_BIGF LT_CC LT_DIAL_hpux LT_K64 LT_VERS=1131 _LARGEFILE64_SOURCE
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24012 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24018 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24021 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24023 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24024 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24025 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24026 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24027 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24028 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24029 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=24030 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=26000 _FILE_OFFSET_BITS=64
+LT_BIGF LT_CC LT_DIAL_linux LT_VERS=26018 _FILE_OFFSET_BITS=64
+_FILE_OFFSET_BITS=64 LT_BIGF LT_CC LT_DIAL_linux LT_VERS=26022
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=1005000
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=1006000
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=2000000
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=2099009
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=2099010
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=2099011
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=2099012
+LT_BIGF LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=3099000
+LT_CC LT_DIAL_netbsd LT_KMEM LT_VERS=1040
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3000
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3010
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3020
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3030
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3040
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3050
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3060
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3070
+LT_BIGF LT_CC LT_DIAL_openbsd LT_KMEM LT_VERS=3090
+LT_CC LT_DIAL_osr LT_KMEM LT_VERS=504
+LT_CC LT_DIAL_osr LT_KMEM LT_VERS=506
+LT_DIAL_ns LT_GCC LT_KMEM LT_VERS=31
+LT_CC LT_DIAL_ns LT_KMEM LT_VERS=42
+LT_CC LT_DIAL_solaris LT_KMEM LT_VERS=20600
+LT_DIAL_solaris LT_GCC LT_KMEM LT_VERS=20600
+LT_BIGF LT_CC LT_DIAL_solaris LT_KMEM LT_VERS=70000
+LT_BIGF LT_CC LT_DIAL_solaris LT_K64 LT_KMEM LT_VERS=70000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_KMEM LT_VERS=70000
+LT_BIGF LT_CC LT_DIAL_solaris LT_KMEM LT_VERS=80000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_KMEM LT_VERS=80000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_K64 LT_KMEM LT_VERS=80000
+LT_BIGF LT_CC LT_DIAL_solaris LT_K64 LT_KMEM LT_VERS=80000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_KMEM LT_VERS=90000
+LT_BIGF LT_CC LT_DIAL_solaris LT_K64 LT_KMEM LT_VERS=90000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_K64 LT_KMEM LT_VERS=90000
+LT_BIGF LT_CC LT_DIAL_solaris LT_KMEM LT_VERS=100000
+LT_BIGF LT_CC LT_DIAL_solaris LT_K64 LT_KMEM LT_VERS=100000 LT_VPATH
+LT_BIGF LT_DIAL_solaris LT_GCC LT_KMEM LT_VERS=100000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_KMEM LT_VERS=100000 LT_VPATH
+LT_BIGF LT_CC LT_DIAL_solaris LT_K64 LT_KMEM LT_VERS=100000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_K64 LT_KMEM LT_VERS=100000
+LT_BIGF LT_DIAL_solaris LT_GCC LT_K64 LT_KMEM LT_VERS=100000 LT_VPATH
+LT_BIGF LT_CC LT_DIAL_uw LT_KMEM LT_VERS=70101
+LT_BIGF LT_CC LT_DIAL_uw LT_KMEM LT_VERS=70103
+LT_BIGF LT_CC LT_DIAL_uw LT_KMEM LT_VERS=70104
--- /dev/null
+/*
+ * usage.c - usage functions for lsof
+ */
+
+
+/*
+ * Copyright 1998 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 1998 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: usage.c,v 1.28 2008/10/21 16:21:41 abe Exp $";
+#endif
+
+
+#include "lsof.h"
+#include "version.h"
+
+
+/*
+ * Local function prototypes
+ */
+
+_PROTOTYPE(static char *isnullstr,(char *s));
+_PROTOTYPE(static void report_HASDCACHE,(int type, char *ttl, char *det));
+_PROTOTYPE(static void report_HASKERNIDCK,(char *pfx, char *verb));
+_PROTOTYPE(static void report_SECURITY,(char *pfx, char *punct));
+_PROTOTYPE(static void report_WARNDEVACCESS,(char *pfx, char *verb,
+ char *punct));
+
+
+/*
+ * isnullstr() - is it a null string?
+ */
+
+static char *
+isnullstr(s)
+ char *s; /* string pointer */
+{
+ if (!s)
+ return((char *)NULL);
+ while (*s) {
+ if (*s != ' ')
+ return(s);
+ s++;
+ }
+ return((char *)NULL);
+}
+
+
+/*
+ * report_HASDCACHE() -- report device cache file state
+ */
+
+static void
+report_HASDCACHE(type, ttl, det)
+ int type; /* type: 0 == read path report
+ * 1 == full report */
+ char *ttl; /* title lines prefix
+ * (NULL if none) */
+ char *det; /* detail lines prefix
+ * (NULL if none) */
+{
+
+#if defined(HASDCACHE)
+ char *cp;
+ int dx;
+
+# if defined(WILLDROPGID)
+ int saved_Setgid = Setgid;
+
+ Setgid = 0;
+# endif /* defined(WILLDROPGID) */
+
+ if (type) {
+
+ /*
+ * Report full device cache information.
+ */
+ (void) fprintf(stderr, "%sDevice cache file read-only paths:\n",
+ ttl ? ttl : "");
+ if ((dx = dcpath(1, 0)) < 0)
+ (void) fprintf(stderr, "%snone\n", det ? det : "");
+ else {
+ (void) fprintf(stderr, "%sNamed via -D: %s\n",
+ det ? det : "",
+ DCpath[0] ? DCpath[0] : "none");
+
+# if defined(HASENVDC)
+ (void) fprintf(stderr,
+ "%sNamed in environment variable %s: %s\n",
+ det ? det : "",
+ HASENVDC, DCpath[1] ? DCpath[1] : "none");
+# endif /* defined(HASENVDC) */
+
+# if defined(HASSYSDC)
+ if (DCpath[2])
+ (void) fprintf(stderr,
+ "%sSystem-wide device cache: %s\n",
+ det ? det : "",
+ DCpath[2]);
+# endif /* defined(HASSYSDC) */
+
+# if defined(HASPERSDC)
+ (void) fprintf(stderr,
+ "%sPersonal path format (HASPERSDC): \"%s\"\n",
+ det ? det : "",
+ HASPERSDC);
+# if defined(HASPERSDCPATH)
+ (void) fprintf(stderr,
+ "%sModified personal path environment variable: %s\n",
+ det ? det : "",
+ HASPERSDCPATH);
+ cp = getenv(HASPERSDCPATH);
+ (void) fprintf(stderr, "%s%s value: %s\n",
+ det ? det : "",
+ HASPERSDCPATH, cp ? cp : "none");
+# endif /* defined(HASPERSDCPATH) */
+ (void) fprintf(stderr, "%sPersonal path: %s\n",
+ det ? det : "",
+ DCpath[3] ? DCpath[3] : "none");
+# endif /* defined(HASPERSDC) */
+ }
+ (void) fprintf(stderr, "%sDevice cache file write paths:\n",
+ ttl ? ttl : "");
+ if ((dx = dcpath(2, 0)) < 0)
+ (void) fprintf(stderr, "%snone\n", det ? det : "");
+ else {
+ (void) fprintf(stderr, "%sNamed via -D: %s\n",
+ det ? det : "",
+ DCstate == 2 ? "none"
+ : DCpath[0] ? DCpath[0] : "none");
+
+# if defined(HASENVDC)
+ (void) fprintf(stderr,
+ "%sNamed in environment variable %s: %s\n",
+ det ? det : "",
+ HASENVDC, DCpath[1] ? DCpath[1] : "none");
+# endif /* defined(HASENVDC) */
+
+# if defined(HASPERSDC)
+ (void) fprintf(stderr,
+ "%sPersonal path format (HASPERSDC): \"%s\"\n",
+ det ? det : "",
+ HASPERSDC);
+# if defined(HASPERSDCPATH)
+ (void) fprintf(stderr,
+ "%sModified personal path environment variable: %s\n",
+ det ? det : "",
+ HASPERSDCPATH);
+ cp = getenv(HASPERSDCPATH);
+ (void) fprintf(stderr, "%s%s value: %s\n",
+ det ? det : "",
+ HASPERSDCPATH, cp ? cp : "none");
+# endif /* defined(HASPERSDCPATH) */
+ (void) fprintf(stderr, "%sPersonal path: %s\n",
+ det ? det : "",
+ DCpath[3] ? DCpath[3] : "none");
+# endif /* defined(HASPERSDC) */
+ }
+ } else {
+
+ /*
+ * Report device cache read file path.
+ */
+
+# if defined(HASENVDC) || defined(HASPERSDC) || defined(HASSYSDC)
+ cp = NULL;
+# if defined(HASENVDC)
+ if ((dx = dcpath(1, 0)) >= 0)
+ cp = DCpath[1];
+# endif /* defined(HASENVDC) */
+# if defined(HASSYSDC)
+ if (!cp)
+ cp = HASSYSDC;
+# endif /* defined(HASSYSDC) */
+# if defined(HASPERSDC)
+ if (!cp && dx != -1 && (dx = dcpath(1, 0)) >= 0)
+ cp = DCpath[3];
+# endif /* defined(HASPERSDC) */
+ if (cp)
+ (void) fprintf(stderr,
+ "%s%s is the default device cache file read path.\n",
+ ttl ? ttl : "",
+ cp
+ );
+# endif /* defined(HASENVDC) || defined(HASPERSDC) || defined(HASSYSDC) */
+ }
+
+# if defined(WILLDROPGID)
+ Setgid = saved_Setgid;
+# endif /* defined(WILLDROPGID) */
+
+#endif /* defined(HASDCACHE) */
+
+}
+
+
+/*
+ * report_HASKERNIDCK() -- report HASKERNIDCK state
+ */
+
+static void
+report_HASKERNIDCK(pfx, verb)
+ char *pfx; /* prefix (NULL if none) */
+ char *verb; /* verb (NULL if none) */
+{
+ (void) fprintf(stderr, "%sernel ID check %s%s%s.\n",
+ pfx ? pfx : "",
+ verb ? verb : "",
+ verb ? " " : "",
+
+#if defined(HASKERNIDCK)
+ "enabled"
+#else /* !defined(HASKERNIDCK) */
+ "disabled"
+#endif /* defined(HASKERNIDCK) */
+
+ );
+}
+
+
+/*
+ * report_SECURITY() -- report *SECURITY states
+ */
+
+static void
+report_SECURITY(pfx, punct)
+ char *pfx; /* prefix (NULL if none) */
+ char *punct; /* short foem punctuation
+ * (NULL if none) */
+{
+ fprintf(stderr, "%s%s can list all files%s",
+ pfx ? pfx : "",
+
+#if defined(HASSECURITY)
+ "Only root",
+# if defined(HASNOSOCKSECURITY)
+ ", but anyone can list socket files.\n"
+# else /* !defined(HASNOSOCKSECURITY) */
+ punct ? punct : ""
+# endif /* defined(HASNOSOCKSECURITY) */
+#else /* !defined(HASSECURITY) */
+ "Anyone",
+ punct ? punct : ""
+#endif /* defined(HASSECURITY) */
+
+ );
+}
+
+
+/*
+ * report_WARNDEVACCESS() -- report WEARNDEVACCESS state
+ */
+
+static void
+report_WARNDEVACCESS(pfx, verb, punct)
+ char *pfx; /* prefix (NULL if none) */
+ char *verb; /* verb (NULL if none) */
+ char *punct; /* punctuation */
+{
+ (void) fprintf(stderr, "%s/dev warnings %s%s%s%s",
+ pfx ? pfx : "",
+ verb ? verb : "",
+ verb ? " " : "",
+
+#if defined(WARNDEVACCESS)
+ "enabled",
+#else /* !defined(WARNDEVACCESS) */
+ "disabled",
+#endif /* defined(WARNDEVACCESS) */
+
+ punct);
+}
+
+
+/*
+ * usage() - display usage and exit
+ */
+
+void
+usage(xv, fh, version)
+ int xv; /* exit value */
+ int fh; /* ``-F ?'' status */
+ int version; /* ``-v'' status */
+{
+ char buf[MAXPATHLEN+1], *cp, *cp1, *cp2;
+ int i;
+
+ if (Fhelp || xv) {
+ (void) fprintf(stderr, "%s %s\n latest revision: %s\n",
+ Pn, LSOF_VERSION, LSOF_URL);
+ (void) fprintf(stderr, " latest FAQ: %sFAQ\n", LSOF_URL);
+ (void) fprintf(stderr, " latest man page: %slsof_man\n", LSOF_URL);
+ (void) fprintf(stderr,
+ " usage: [-?ab%shlnNoOP%s%stUvV%s]",
+
+#if defined(HASNCACHE)
+ "C",
+#else /* !defined(HASNCACHE) */
+ "",
+#endif /* defined(HASNCACHE) */
+
+#if defined(HASPPID)
+ "R",
+#else /* !defined(HASPPID) */
+ "",
+#endif /* defined(HASPPID) */
+
+#if defined(HASTCPUDPSTATE)
+ "",
+#else /* !defined(HASTCPUDPSTATE) */
+ "s",
+#endif /* defined(HASTCPUDPSTATE) */
+
+#if defined(HASXOPT)
+# if defined(HASXOPT_ROOT)
+ (Myuid == 0) ? "X" : ""
+# else /* !defined(HASXOPT_ROOT) */
+ "X"
+# endif /* defined(HASXOPT_ROOT) */
+#else /* !defined(HASXOPT) */
+ ""
+#endif /* defined(HASXOPT) */
+
+ );
+
+#if defined(HAS_AFS) && defined(HASAOPT)
+ (void) fprintf(stderr, " [-A A]");
+#endif /* defined(HAS_AFS) && defined(HASAOPT) */
+
+ (void) fprintf(stderr, " [+|-c c] [+|-d s] [+%sD D]",
+
+#if defined(HASDCACHE)
+ "|-"
+#else /* !defined(HASDCACHE) */
+ ""
+#endif /* defined(HASDCACHE) */
+
+ );
+
+ (void) fprintf(stderr,
+ " [+|-f%s%s%s%s%s%s]\n [-F [f]] [-g [s]] [-i [i]]",
+
+#if defined(HASFSTRUCT)
+ "[",
+
+# if defined(HASNOFSCOUNT)
+ "",
+# else /* !defined(HASNOFSCOUNT) */
+ "c",
+# endif /* defined(HASNOFSCOUNT) */
+
+# if defined(HASNOFSADDR)
+ "",
+# else /* !defined(HASNOFSADDR) */
+ "f",
+# endif /* defined(HASNOFSADDR) */
+
+# if defined(HASNOFSFLAGS)
+ "",
+# else /* !defined(HASNOFSFLAGS) */
+ "gG",
+# endif /* defined(HASNOFSFLAGS) */
+
+# if defined(HASNOFSNADDR)
+ "",
+# else /* !defined(HASNOFSNADDR) */
+ "n",
+# endif /* defined(HASNOFSNADDR) */
+
+ "]"
+#else /* !defined(HASFSTRUCT) */
+ "", "", "", "", "", ""
+#endif /* defined(HASFSTRUCT) */
+
+ );
+
+#if defined(HASKOPT)
+ (void) fprintf(stderr, " [-k k]");
+#endif /* defined(HASKOPT) */
+
+ (void) fprintf(stderr, " [+|-L [l]]");
+
+#if defined(HASMOPT) || defined(HASMNTSUP)
+ (void) fprintf(stderr,
+# if defined(HASMOPT)
+# if defined(HASMNTSUP)
+ " [+|-m [m]]"
+# else /* !defined(HASMNTSUP) */
+ " [-m m]"
+# endif /* defined(HASMNTSUP) */
+# else /* !defined(HASMOPT) */
+ " [+m [m]]"
+# endif /* defined(HASMOPT) */
+ );
+#endif /* defined(HASMOPT) || defined(HASMNTSUP) */
+
+ (void) fprintf(stderr,
+ " [+|-M] [-o [o]] [-p s]\n[+|-r [t]]%s [-S [t]] [-T [t]]",
+
+#if defined(HASTCPUDPSTATE)
+ " [-s [p:s]]"
+#else /* !defined(HASTCPUDPSTATE) */
+ ""
+#endif /* defined(HASTCPUDPSTATE) */
+
+ );
+ (void) fprintf(stderr, " [-u s] [+|-w] [-x [fl]]");
+
+#if defined(HASZONES)
+ (void) fprintf(stderr, " [-z [z]]");
+#else /* !defined(HASZONES) */
+# if defined(HASSELINUX)
+ if (CntxStatus)
+ (void) fprintf(stderr, " [-Z [Z]]");
+# endif /* defined(HASSELINUX) */
+#endif /* defined(HASZONES) */
+
+ (void) fprintf(stderr, " [--] [names]\n");
+ }
+ if (xv && !Fhelp) {
+ (void) fprintf(stderr,
+ "Use the ``-h'' option to get more help information.\n");
+ if (!fh)
+ Exit(xv);
+ }
+ if (Fhelp) {
+ (void) fprintf(stderr,
+ "Defaults in parentheses; comma-separated set (s) items;");
+ (void) fprintf(stderr, " dash-separated ranges.\n");
+ (void) fprintf(stderr, " %-23.23s", "-?|-h list help");
+ (void) fprintf(stderr, " %-25.25s", "-a AND selections (OR)");
+ (void) fprintf(stderr, " %s\n", "-b avoid kernel blocks");
+ (void) fprintf(stderr, " %-23.23s", "-c c cmd c ^c /c/[bix]");
+ (void) snpf(buf, sizeof(buf), "+c w COMMAND width (%d)", CMDL);
+ (void) fprintf(stderr, " %-25.25s", buf);
+
+ (void) fprintf(stderr, " %s\n",
+
+#if defined(HASNCACHE)
+ "-C no kernel name cache");
+#else /* !defined(HASNCACHE) */
+ " ");
+#endif /* defined(HASNCACHE) */
+
+ (void) fprintf(stderr, " %-23.23s", "+d s dir s files");
+ (void) fprintf(stderr, " %-25.25s", "-d s select by FD set");
+ (void) fprintf(stderr, " %s\n", "+D D dir D tree *SLOW?*");
+
+#if defined(HASDCACHE)
+ if (Setuidroot)
+ cp = "?|i|r";
+
+# if !defined(WILLDROPGID)
+ else if (Myuid)
+ cp = "?|i|r<path>";
+# endif /* !defined(WILLDROPGID) */
+
+ else
+ cp = "?|i|b|r|u[path]";
+ (void) snpf(buf, sizeof(buf), "-D D %s", cp);
+#else /* !defined(HASDCACHE) */
+ (void) snpf(buf, sizeof(buf), " ");
+#endif /* defined(HASDCACHE) */
+
+ (void) fprintf(stderr, " %-23.23s", buf);
+ (void) snpf(buf, sizeof(buf), "-i select IPv%s files",
+
+#if defined(HASIPv6)
+ "[46]"
+#else /* !defined(HASIPv6) */
+ "4"
+#endif /* defined(HASIPv6) */
+
+ );
+ (void) fprintf(stderr, " %-25.25s", buf);
+ (void) fprintf(stderr, " %s\n", "-l list UID numbers");
+ (void) fprintf(stderr, " %-23.23s", "-n no host names");
+ (void) fprintf(stderr, " %-25.25s", "-N select NFS files");
+ (void) fprintf(stderr, " %s\n", "-o list file offset");
+ (void) fprintf(stderr, " %-23.23s", "-O avoid overhead *RISKY*");
+ (void) fprintf(stderr, " %-25.25s", "-P no port names");
+ (void) fprintf(stderr, " %s\n",
+
+#if defined(HASPPID)
+ "-R list paRent PID"
+#else /* !defined(HASPPID) */
+ ""
+#endif /* defined(HASPPID) */
+
+ );
+ (void) fprintf(stderr, " %-23.23s", "-s list file size");
+ (void) fprintf(stderr, " %-25.25s", "-t terse listing");
+ (void) fprintf(stderr, " %s\n", "-T disable TCP/TPI info");
+ (void) fprintf(stderr, " %-23.23s", "-U select Unix socket");
+ (void) fprintf(stderr, " %-25.25s", "-v list version info");
+ (void) fprintf(stderr, " %s\n", "-V verbose search");
+ (void) snpf(buf, sizeof(buf), "+|-w Warnings (%s)",
+
+#if defined(WARNINGSTATE)
+ "-"
+#else /* !defined(WARNINGSTATE) */
+ "+"
+#endif /* defined(WARNINGSTATE) */
+
+ );
+ (void) fprintf(stderr, " %-23.23s", buf);
+
+#if defined(HASXOPT)
+# if defined(HASXOPT_ROOT)
+ if (Myuid == 0)
+ (void) snpf(buf, sizeof(buf), "-X %s", HASXOPT);
+ else
+ buf[0] = '\0';
+# else /* !defined(HASXOPT_ROOT) */
+ (void) snpf(buf, sizeof(buf), "-X %s", HASXOPT);
+# endif /* defined(HASXOPT_ROOT) */
+# else /* !defined(HASXOPT) */
+ buf[0] = '\0';
+#endif /* defined(HASXOPT) */
+
+ if (buf[0])
+ (void) fprintf(stderr, " %-25.25s", buf);
+
+#if defined(HASZONES)
+ (void) fprintf(stderr,
+ (buf[0]) ? " %s\n" : " %-25.25s", "-z z zone [z]");
+#else /* !defined(HASZONES) */
+# if defined(HASSELINUX)
+ (void) fprintf(stderr,
+ (buf[0]) ? " %s\n" : " %-25.25s", "-Z Z context [Z]");
+# endif /* defined(HASSELINUX) */
+#endif /* defined(HASZONES) */
+
+ (void) fprintf(stderr, " %s\n", "-- end option scan");
+ (void) fprintf(stderr, " %-36.36s",
+ "+f|-f +filesystem or -file names");
+
+#if defined(HASFSTRUCT)
+ (void) fprintf(stderr,
+ " +|-f[%s%s%s%s]%s%s%s%s %s%s%s%s%s%s%s\n",
+
+# if defined(HASNOFSCOUNT)
+ "",
+# else /* !defined(HASNOFSCOUNT) */
+ "c",
+# endif /* defined(HASNOFSCOUNT) */
+
+# if defined(HASNOFSADDR)
+ "",
+# else /* !defined(HASNOFSADDR) */
+ "f",
+# endif /* defined(HASNOFSADDR) */
+
+# if defined(HASNOFSFLAGS)
+ "",
+# else /* !defined(HASNOFSFLAGS) */
+ "gG",
+# endif /* defined(HASNOFSFLAGS) */
+
+# if defined(HASNOFSNADDR)
+ "",
+# else /* !defined(HASNOFSNADDR) */
+ "n",
+# endif /* defined(HASNOFSNADDR) */
+
+# if defined(HASNOFSCOUNT)
+ "",
+# else /* !defined(HASNOFSCOUNT) */
+ " Ct",
+# endif /* defined(HASNOFSCOUNT) */
+
+# if defined(HASNOFSADDR)
+ "",
+# else /* !defined(HASNOFSADDR) */
+ " Fstr",
+# endif /* defined(HASNOFSADDR) */
+
+# if defined(HASNOFSFLAGS)
+ "",
+# else /* !defined(HASNOFSFLAGS) */
+ " flaGs",
+# endif /* defined(HASNOFSFLAGS) */
+
+# if defined(HASNOFSNADDR)
+ "",
+# else /* !defined(HASNOFSNADDR) */
+ " Node",
+# endif /* defined(HASNOFSNADDR) */
+
+ Fsv ? "(" : "",
+ (Fsv & FSV_CT) ? "C" : "",
+ (Fsv & FSV_FA) ? "F" : "",
+ ((Fsv & FSV_FG) && FsvFlagX) ? "g" : "",
+ ((Fsv & FSV_FG) && !FsvFlagX) ? "G" : "",
+ (Fsv & FSV_NI) ? "N" : "",
+ Fsv ? ")" : "");
+#else /* !defined(HASFSTRUCT) */
+ putc('\n', stderr);
+#endif /* defined(HASFSTRUCT) */
+
+ (void) fprintf(stderr, " %-36.36s",
+ "-F [f] select fields; -F? for help");
+
+#if defined(HASKOPT)
+ (void) fprintf(stderr,
+ " -k k kernel symbols (%s)\n",
+ Nmlst ? Nmlst
+# if defined(N_UNIX)
+ : N_UNIX
+# else /* !defined(N_UNIX) */
+ : (Nmlst = get_nlist_path(1)) ? Nmlst
+ : "none found"
+# endif /* defined(N_UNIX) */
+
+ );
+#else /* !defined(HASKOPT) */
+ putc('\n', stderr);
+#endif /* defined(HASKOPT) */
+
+ (void) fprintf(stderr,
+ " +|-L [l] list (+) suppress (-) link counts < l (0 = all; default = 0)\n");
+
+#if defined(HASMOPT) || defined(HASMNTSUP)
+# if defined(HASMOPT)
+ (void) snpf(buf, sizeof(buf), "-m m kernel memory (%s)", KMEM);
+# else /* !defined(HASMOPT) */
+ buf[0] = '\0';
+# endif /* defined(HASMOPT) */
+
+ (void) fprintf(stderr, " %-36.36s", buf);
+
+# if defined(HASMNTSUP)
+ (void) fprintf(stderr, " +m [m] use|create mount supplement\n");
+# else /* !defined(HASMNTSUP) */
+ (void) fprintf(stderr, "\n");
+# endif /* defined(HASMNTSUP) */
+#endif /* defined(HASMOPT) || defined(HASMNTSUP) */
+
+ (void) snpf(buf, sizeof(buf), "+|-M portMap registration (%s)",
+
+#if defined(HASPMAPENABLED)
+ "+"
+#else /* !defined(HASPMAPENABLED) */
+ "-"
+#endif /* defined(HASPMAPENABLED) */
+
+ );
+ (void) fprintf(stderr, " %-36.36s", buf);
+ (void) snpf(buf, sizeof(buf), "-o o o 0t offset digits (%d)",
+ OFFDECDIG);
+ (void) fprintf(stderr, " %s\n", buf);
+ (void) fprintf(stderr, " %-36.36s",
+ "-p s exclude(^)|select PIDs");
+ (void) fprintf(stderr, " -S [t] t second stat timeout (%d)\n",
+ TMLIMIT);
+ (void) snpf(buf, sizeof(buf),
+ "-T %s%ss%s TCP/TPI %s%sSt%s (s) info",
+
+#if defined(HASSOOPT) || defined(HASSOSTATE) || defined(HASTCPOPT)
+ "f",
+#else /* !defined(HASSOOPT) && !defined(HASSOSTATE) && !defined(HASTCPOPT)*/
+ "",
+#endif /* defined(HASSOOPT) || defined(HASSOSTATE) || defined(HASTCPOPT)*/
+
+#if defined(HASTCPTPIQ)
+ "q",
+#else /* !defined(HASTCPTPIQ) */
+ " ",
+#endif /* defined(HASTCPTPIQ) */
+
+#if defined(HASTCPTPIW)
+ "w",
+#else /* !defined(HASTCPTPIW) */
+ "",
+#endif /* defined(HASTCPTPIW) */
+
+#if defined(HASSOOPT) || defined(HASSOSTATE) || defined(HASTCPOPT)
+ "Fl,",
+#else /* !defined(HASSOOPT) && !defined(HASSOSTATE) && !defined(HASTCPOPT)*/
+ "",
+#endif /* defined(HASSOOPT) || defined(HASSOSTATE) || defined(HASTCPOPT)*/
+
+#if defined(HASTCPTPIQ)
+ "Q,",
+#else /* !defined(HASTCPTPIQ) */
+ "",
+#endif /* defined(HASTCPTPIQ) */
+
+#if defined(HASTCPTPIW)
+ ",Win"
+#else /* !defined(HASTCPTPIW) */
+ ""
+#endif /* defined(HASTCPTPIW) */
+
+ );
+ (void) fprintf(stderr, " %s\n", buf);
+
+#if defined(HAS_AFS) && defined(HASAOPT)
+ (void) fprintf(stderr,
+ " -A A AFS name list file (%s)\n", AFSAPATHDEF);
+#endif /* defined(HAS_AFS) && defined(HASAOPT) */
+
+ (void) fprintf(stderr,
+ " -g [s] exclude(^)|select and print process group IDs\n");
+ (void) fprintf(stderr, " -i i select by IPv%s address:",
+
+#if defined(HASIPv6)
+ "[46]"
+#else /* !defined(HASIPv6) */
+ "4"
+#endif /* defined(HASIPv6) */
+
+ );
+ (void) fprintf(stderr,
+ " [%s][proto][@host|addr][:svc_list|port_list]\n",
+
+#if defined(HASIPv6)
+ "46"
+#else /* !defined(HASIPv6) */
+ "4"
+#endif /* defined(HASIPv6) */
+
+ );
+
+ (void) fprintf(stderr,
+ " +|-r [%s] repeat every t seconds (%d); %s",
+
+#if defined(HAS_STRFTIME)
+ "t[m<fmt>]",
+#else /* !defined(has_STRFTIME) */
+ "t",
+#endif /* defined(HAS_STRFTIME) */
+
+ RPTTM,
+ " + until no files, - forever.\n");
+
+#if defined(HAS_STRFTIME)
+ (void) fprintf(stderr,
+ " An optional suffix to t is m<fmt>; m must separate %s",
+ "t from <fmt> and\n");
+ (void) fprintf(stderr, " <fmt> is an strftime(3) format %s",
+ "for the marker line.\n");
+#endif /* defined(HAS_STRFTIME) */
+
+#if defined(HASTCPUDPSTATE)
+ (void) fprintf(stderr,
+ " -s p:s exclude(^)|select protocol (p = TCP|UDP) states");
+ (void) fprintf(stderr, " by name(s).\n");
+#endif /* defined(HASTCPUDPSTATE) */
+
+ (void) fprintf(stderr,
+ " -u s exclude(^)|select login|UID set s\n");
+ (void) fprintf(stderr,
+ " -x [fl] cross over +d|+D File systems or symbolic Links\n");
+ (void) fprintf(stderr,
+ " names select named files or files on named file systems\n");
+ (void) report_SECURITY(NULL, "; ");
+ (void) report_WARNDEVACCESS(NULL, NULL, ";");
+ (void) report_HASKERNIDCK(" k", NULL);
+ (void) report_HASDCACHE(0, NULL, NULL);
+
+#if defined(DIALECT_WARNING)
+ (void) fprintf(stderr, "WARNING: %s\n", DIALECT_WARNING);
+#endif /* defined(DIALECT_WARNING) */
+
+ }
+ if (fh) {
+ (void) fprintf(stderr, "%s:\tID field description\n", Pn);
+ for (i = 0; FieldSel[i].nm; i++) {
+
+#if !defined(HASPPID)
+ if (FieldSel[i].id == LSOF_FID_PPID)
+ continue;
+#endif /* !defined(HASPPID) */
+
+#if !defined(HASFSTRUCT)
+ if (FieldSel[i].id == LSOF_FID_FA
+ || FieldSel[i].id == LSOF_FID_CT
+ || FieldSel[i].id == LSOF_FID_FG
+ || FieldSel[i].id == LSOF_FID_NI)
+ continue;
+#else /* defined(HASFSTRUCT) */
+# if defined(HASNOFSADDR)
+ if (FieldSel[i].id == LSOF_FID_FA)
+ continue;
+# endif /* defined(HASNOFSADDR) */
+
+# if defined(HASNOFSCOUNT)
+ if (FieldSel[i].id == LSOF_FID_CT)
+ continue;
+# endif /* !defined(HASNOFSCOUNT) */
+
+# if defined(HASNOFSFLAGS)
+ if (FieldSel[i].id == LSOF_FID_FG)
+ continue;
+# endif /* defined(HASNOFSFLAGS) */
+
+# if defined(HASNOFSNADDR)
+ if (FieldSel[i].id == LSOF_FID_NI)
+ continue;
+# endif /* defined(HASNOFSNADDR) */
+#endif /* !defined(HASFSTRUCT) */
+
+#if !defined(HASZONES)
+ if (FieldSel[i].id == LSOF_FID_ZONE)
+ continue;
+#endif /* !defined(HASZONES) */
+
+#if defined(HASSELINUX)
+ if ((FieldSel[i].id == LSOF_FID_CNTX) && !CntxStatus)
+ continue;
+#else /* !defined(HASSELINUX) */
+ if (FieldSel[i].id == LSOF_FID_CNTX)
+ continue;
+#endif /* !defined(HASSELINUX) */
+
+ (void) fprintf(stderr, "\t %c %s\n",
+ FieldSel[i].id, FieldSel[i].nm);
+ }
+ }
+
+#if defined(HASDCACHE)
+ if (DChelp)
+ report_HASDCACHE(1, NULL, " ");
+#endif /* defined(HASDCACHE) */
+
+ if (version) {
+
+ /*
+ * Display version information in reponse to ``-v''.
+ */
+ (void) fprintf(stderr, "%s version information:\n", Pn);
+ (void) fprintf(stderr, " revision: %s\n", LSOF_VERSION);
+ (void) fprintf(stderr, " latest revision: %s\n", LSOF_URL);
+ (void) fprintf(stderr, " latest FAQ: %sFAQ\n",
+ LSOF_URL);
+ (void) fprintf(stderr, " latest man page: %slsof_man\n",
+ LSOF_URL);
+
+#if defined(LSOF_CINFO)
+ if ((cp = isnullstr(LSOF_CINFO)))
+ (void) fprintf(stderr, " configuration info: %s\n", cp);
+#endif /* defined(LSOF_CINFO) */
+
+ if ((cp = isnullstr(LSOF_CCDATE)))
+ (void) fprintf(stderr, " constructed: %s\n", cp);
+ cp = isnullstr(LSOF_HOST);
+ if (!(cp1 = isnullstr(LSOF_LOGNAME)))
+ cp1 = isnullstr(LSOF_USER);
+ if (cp || cp1) {
+ if (cp && cp1)
+ cp2 = "by and on";
+ else if (cp)
+ cp2 = "on";
+ else
+ cp2 = "by";
+ (void) fprintf(stderr, " constructed %s: %s%s%s\n",
+ cp2,
+ cp1 ? cp1 : "",
+ (cp && cp1) ? "@" : "",
+ cp ? cp : ""
+ );
+ }
+
+#if defined(LSOF_BLDCMT)
+ if ((cp = isnullstr(LSOF_BLDCMT)))
+ (void) fprintf(stderr, " builder's comment: %s\n", cp);
+#endif /* defined(LSOF_BLDCMT) */
+
+ if ((cp = isnullstr(LSOF_CC)))
+ (void) fprintf(stderr, " compiler: %s\n", cp);
+ if ((cp = isnullstr(LSOF_CCV)))
+ (void) fprintf(stderr, " compiler version: %s\n", cp);
+ if ((cp = isnullstr(LSOF_CCFLAGS)))
+ (void) fprintf(stderr, " compiler flags: %s\n", cp);
+ if ((cp = isnullstr(LSOF_LDFLAGS)))
+ (void) fprintf(stderr, " loader flags: %s\n", cp);
+ if ((cp = isnullstr(LSOF_SYSINFO)))
+ (void) fprintf(stderr, " system info: %s\n", cp);
+ (void) report_SECURITY(" ", ".\n");
+ (void) report_WARNDEVACCESS(" ", "are", ".\n");
+ (void) report_HASKERNIDCK(" K", "is");
+
+#if defined(DIALECT_WARNING)
+ (void) fprintf(stderr, " WARNING: %s\n", DIALECT_WARNING);
+#endif /* defined(DIALECT_WARNING) */
+
+ (void) report_HASDCACHE(1, " ", "\t");
+ }
+ Exit(xv);
+}
--- /dev/null
+/*
+ * dutil.c - AIX utility functions whose compilation conflicts with the
+ * general header file tree defined by lsof.h and dlsof.h -- e.g.,
+ * the conflict between <time.h> and <sys/time.h> for the time(2)
+ * and localtime(3) functions
+ *
+ * V. Abell
+ * Purdue University
+ */
+
+
+/*
+ * Copyright 2008 Purdue Research Foundation, West Lafayette, Indiana
+ * 47907. All rights reserved.
+ *
+ * Written by Victor A. Abell
+ *
+ * This software is not subject to any license of the American Telephone
+ * and Telegraph Company or the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on
+ * any computer system, and to alter it and redistribute it freely, subject
+ * to the following restrictions:
+ *
+ * 1. Neither the authors nor Purdue University are responsible for any
+ * consequences of the use of this software.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Credit to the authors and Purdue
+ * University must appear in documentation and sources.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ *
+ * 4. This notice may not be removed or altered.
+ */
+
+
+#ifndef lint
+static char copyright[] =
+"@(#) Copyright 2008 Purdue Research Foundation.\nAll rights reserved.\n";
+static char *rcsid = "$Id: util.c,v 1.1 2008/04/01 11:56:53 abe Exp $";
+#endif
+
+#if defined(HAS_STRFTIME)
+#include <time.h>
+#endif /* defined(HAS_STRFTIME) */
+
+
+/*
+ * util_strftime() -- utility function to call strftime(3) without header
+ * file distractions
+ */
+
+int
+util_strftime(fmtr, fmtl, fmt)
+ char *fmtr; /* format output receiver */
+ int fmtl; /* sizeof(*fmtr) */
+ char *fmt; /* format */
+{
+
+#if defined(HAS_STRFTIME)
+ struct tm *lt;
+ time_t tm;
+
+ tm = time((time_t *)NULL);
+ lt = localtime(&tm);
+ return(strftime(fmtr, fmtl, fmt, lt));
+#else /* !defined(HAS_STRFTIME) */
+ return(0);
+#endif /* defined(HAS_STRFTIME) */
+
+}
--- /dev/null
+.ds VN 4.82
projects / external / lsof.git / commitdiff