Lucas De Marchi [Wed, 25 Jan 2012 01:31:46 +0000 (23:31 -0200)]
testsuite: export environment with flags and LD_PRELOAD
A certain config can add flags and each flag may be associated with a
lib to LD_PRELOAD. It's now done for uname(2), which requires uname.so
in order to trap the calls.
Other trap will be added in later commits.
Lucas De Marchi [Wed, 25 Jan 2012 01:28:39 +0000 (23:28 -0200)]
testsuite: move oneshot to inside the test struct
Lucas De Marchi [Wed, 25 Jan 2012 00:04:46 +0000 (22:04 -0200)]
testsuite: trap calls to uname
Lucas De Marchi [Tue, 24 Jan 2012 23:53:27 +0000 (21:53 -0200)]
testsuite: test libkmod initialization
Lucas De Marchi [Tue, 24 Jan 2012 22:59:54 +0000 (20:59 -0200)]
testsuite: add skeleton
Lucas De Marchi [Thu, 26 Jan 2012 18:01:34 +0000 (16:01 -0200)]
modprobe: fix build 'cause of excessive number of arguments
Dave Reisner [Thu, 26 Jan 2012 16:36:35 +0000 (11:36 -0500)]
modprobe: remove code referring to -t, --type
This was only useful with the --list function, which isn't implemented.
Lucas De Marchi [Mon, 23 Jan 2012 14:29:38 +0000 (12:29 -0200)]
README: add link to packages place
A lot of people trying to get latest package from cgit is not good.
They should get the package from the released versions instead.
Lucas De Marchi [Sat, 21 Jan 2012 17:38:25 +0000 (15:38 -0200)]
build-sys: do not create symlinks by default
Distro packagers should create them instead. It's too much trouble to
create them in the build system and every distro wants a different path
for them.
Lucas De Marchi [Sat, 21 Jan 2012 04:45:06 +0000 (02:45 -0200)]
modprobe: kill operations depending on path
It was not on module-init-tools and it doesn't make much sense. It will
deal with dependencies, but looking at modules in the index. This might
not be the module we want if we are loading another from outside of the
tree.
Dealing with paths causes this bug (supposing there's a module names
squashfs):
# cd /
# touch squashfs
# modprobe squashfs
That is because it detects that squashfs exists as a file and it will
try to load it instead of the alias "squashfs".
If you need to load a module from a path, use insmod.
Thanks to Silvan Calarco <silvan.calarco@mambasoft.it> who reported the
bug and helped debugging it.
Lucas De Marchi [Fri, 20 Jan 2012 15:04:43 +0000 (13:04 -0200)]
TODO: system() should not be used inside a library
Lucas De Marchi [Wed, 18 Jan 2012 18:37:59 +0000 (16:37 -0200)]
build-sys: forcefully create links
If link already existed we would fail. Really not important for
packagers (the link should not exist), but for those who manually install it.
Lucas De Marchi [Wed, 18 Jan 2012 18:02:45 +0000 (16:02 -0200)]
build-sys: move modprobe to sbindir
Lucas De Marchi [Wed, 18 Jan 2012 18:01:58 +0000 (16:01 -0200)]
Add program to calculate the shortest relative path for symlinks
Lucas De Marchi [Tue, 17 Jan 2012 23:16:23 +0000 (21:16 -0200)]
libkmod-list: allow to append an empty list
Lucas De Marchi [Tue, 17 Jan 2012 21:33:32 +0000 (19:33 -0200)]
build-sys: create symlinks if we are installing tools
Thierry Vignaud [Tue, 17 Jan 2012 19:32:48 +0000 (17:32 -0200)]
WEXITSTATUS is defined in sys/wait.h
Fix compiling with dietlibc:
../libkmod/libkmod-module.c:858:2: warning: implicit declaration of function 'WEXITSTATUS' [-Wimplicit-function-declaration]
../libkmod/libkmod-module.c:858:2: warning: nested extern declaration of 'WEXITSTATUS' [-Wnested-externs]
(...)
kmod-modprobe.c:(.text.command_do+0x157): undefined reference to `WEXITSTATUS'
Lucas De Marchi [Tue, 17 Jan 2012 14:10:42 +0000 (12:10 -0200)]
Check if struct stat has mtim member
Not all libc's have a mtim member in struct stat (dietlibc doesn't).
Change ts_usec() to receive a struct stat as parameter and implement it
accordingly for both cases.
Lucas De Marchi [Tue, 17 Jan 2012 12:05:02 +0000 (10:05 -0200)]
modprobe: flush stdout before dumping indexes
Index dump doesn't use stdio.h function and instead call write()
directly on STDOUT_FILENO file descriptor. Therefore we need to flush
stdio buffers before calling it, to be sure the configuration dump will
appear before index's.
Lucas De Marchi [Tue, 17 Jan 2012 01:03:33 +0000 (23:03 -0200)]
Fix common misspelling with codespell
-------8<-------
SUMMARY:
ommitted 1
Miklos Vajna [Tue, 17 Jan 2012 00:48:11 +0000 (01:48 +0100)]
man: spelling fixes
Lucas De Marchi [Mon, 16 Jan 2012 18:53:04 +0000 (16:53 -0200)]
kmod 4
Lucas De Marchi [Mon, 16 Jan 2012 02:42:28 +0000 (00:42 -0200)]
build-sys: add release helpers
Lucas De Marchi [Mon, 16 Jan 2012 18:43:47 +0000 (16:43 -0200)]
Add doc to kmod_dump_index
Lucas De Marchi [Mon, 16 Jan 2012 18:00:35 +0000 (16:00 -0200)]
modprobe: dump indexes as well
Lucas De Marchi [Mon, 16 Jan 2012 17:56:17 +0000 (15:56 -0200)]
libkmod: dump index files
Provide a function to dump the index files to a certain fd. It could be
more optimized (particularly the functions to dump the index that were
copied and pasted from m-i-t), but it seems like the only user of it is
'modprobe -c', used for debugging purposes. So, keep it as is.
Lucas De Marchi [Mon, 16 Jan 2012 17:46:01 +0000 (15:46 -0200)]
libkmod-util: add helper function to write to fd
Lucas De Marchi [Mon, 16 Jan 2012 17:45:12 +0000 (15:45 -0200)]
libkmod-util: adhere to coding style
Lucas De Marchi [Mon, 16 Jan 2012 14:01:48 +0000 (12:01 -0200)]
libkmod: export enum kmod_index and rename members
Lucas De Marchi [Mon, 16 Jan 2012 12:43:34 +0000 (10:43 -0200)]
libkmod: store prefix of each index
Prefix is useful when dumping the index (to be added later).
Lucas De Marchi [Sat, 14 Jan 2012 14:09:34 +0000 (12:09 -0200)]
Update documentation with recent changes
Lucas De Marchi [Sat, 14 Jan 2012 04:46:58 +0000 (02:46 -0200)]
depmod: deal with relative root dir
module paths coming from libkmod are absolute. Make sure we store the
absolute dirname to compare later.
Lucas De Marchi [Sat, 14 Jan 2012 04:31:51 +0000 (02:31 -0200)]
kmod_new(): deal with relative paths for dirname
Lucas De Marchi [Sat, 14 Jan 2012 04:04:56 +0000 (02:04 -0200)]
depmod: fix idx calculation after postponed array creation
We need to set up each module's idx when the array is created. So,
postpone its initialization as well.
Lucas De Marchi [Sat, 14 Jan 2012 04:04:31 +0000 (02:04 -0200)]
Reduce scope of counter variable
Lucas De Marchi [Sat, 14 Jan 2012 04:03:21 +0000 (02:03 -0200)]
depmod: adhere to coding style
Lucas De Marchi [Fri, 13 Jan 2012 04:35:34 +0000 (02:35 -0200)]
modprobe: dump configuration
Lucas De Marchi [Fri, 13 Jan 2012 12:49:31 +0000 (10:49 -0200)]
config: let softdeps dump their data
Lucas De Marchi [Fri, 13 Jan 2012 04:24:18 +0000 (02:24 -0200)]
config: add exported iterator functions
Config iterators are useful to get each configuration list, remember its
type and how to get their key/value pair.
softdeps don't have the value yet, because they are stored as string
vectors.
Lucas De Marchi [Fri, 13 Jan 2012 03:14:46 +0000 (01:14 -0200)]
libkmod-private: allow to get aliases from config
Lucas De Marchi [Thu, 12 Jan 2012 20:31:10 +0000 (18:31 -0200)]
build-sys: add script to bootstrap and configure
Lucas De Marchi [Thu, 12 Jan 2012 20:23:32 +0000 (18:23 -0200)]
build-sys: rename autogen.sh to bootstrap and keep a symlink
Lucas De Marchi [Thu, 12 Jan 2012 20:14:32 +0000 (18:14 -0200)]
TODO: add code unification to list of tasks
Lucas De Marchi [Thu, 12 Jan 2012 19:14:30 +0000 (17:14 -0200)]
modprobe: abort on dependency loop that cannot be broken
Lucas De Marchi [Thu, 12 Jan 2012 17:36:54 +0000 (15:36 -0200)]
modprobe: fix leak on error path
Lucas De Marchi [Thu, 12 Jan 2012 17:28:19 +0000 (15:28 -0200)]
TODO: update tasks
Lucas De Marchi [Thu, 12 Jan 2012 17:23:51 +0000 (15:23 -0200)]
modprobe: rework module removal without tree traversing
Just like the module insertion, module removal is remade.
The dependencies line that comes from modules.dep already contains all
the dependencies necessary to remove that module. Therefore modprobe
doesn't have to do the recursion between the modules in order to remove
it. All we have to do is to remove in order:
For the module being removed:
----------------------------
1. softdeps (in reverse order)
2. deps (in reverse order)
3. module
4. postdeps (in reverse order)
For any of the dependencies:
----------------------------
1. softdeps (in reverse order)
2. module
3. softdeps (in reverse order)
Lucas De Marchi [Thu, 12 Jan 2012 16:45:11 +0000 (14:45 -0200)]
TODO: update differences with m-i-t
Lucas De Marchi [Thu, 12 Jan 2012 04:21:26 +0000 (02:21 -0200)]
modprobe: rework module insertion without tree traversing
The dependencies line that comes from modules.dep already contains all
the dependencies necessary to insert that module. Therefore modprobe
doesn't have to do the recursion between the modules in order to load a
module. All we have to do is to load in order:
For the module being loaded:
----------------------------
1. softdeps
2. deps
3. module
4. postdeps
For any of the dependencies:
----------------------------
1. softdeps
2. module
3. softdeps
Lucas De Marchi [Thu, 12 Jan 2012 16:20:10 +0000 (14:20 -0200)]
man: build modinfo man page
Lucas De Marchi [Thu, 12 Jan 2012 16:12:27 +0000 (14:12 -0200)]
man: build modprobe man page
Lucas De Marchi [Thu, 12 Jan 2012 16:02:27 +0000 (14:02 -0200)]
man: build rmmod man page
Lucas De Marchi [Thu, 12 Jan 2012 15:57:45 +0000 (13:57 -0200)]
man: build lsmod man page
Lucas De Marchi [Thu, 12 Jan 2012 15:53:55 +0000 (13:53 -0200)]
man: build insmod man page
Lucas De Marchi [Thu, 12 Jan 2012 15:47:21 +0000 (13:47 -0200)]
man: build depmod man page
Lucas De Marchi [Thu, 12 Jan 2012 15:34:09 +0000 (13:34 -0200)]
man: build modules.dep{,.bin} man pages
Lucas De Marchi [Thu, 12 Jan 2012 15:18:08 +0000 (13:18 -0200)]
man: build modprobe.d man page
Lucas De Marchi [Thu, 12 Jan 2012 13:48:21 +0000 (11:48 -0200)]
build-sys: build man pages
Lucas De Marchi [Thu, 12 Jan 2012 13:24:38 +0000 (11:24 -0200)]
man: convert depmod.d.sgml to xml and update it
Lucas De Marchi [Thu, 12 Jan 2012 13:52:58 +0000 (11:52 -0200)]
build-sys: fix clean rule removing tracked files
Lucas De Marchi [Thu, 12 Jan 2012 12:05:25 +0000 (10:05 -0200)]
rename doc dir to man
Robby Workman [Thu, 12 Jan 2012 06:00:27 +0000 (00:00 -0600)]
doc/modprobe.sgml: Various updates wrt inclusion with kmod
* TODO: document "--remove-dependencies" flag
Robby Workman [Thu, 12 Jan 2012 05:47:11 +0000 (23:47 -0600)]
doc/depmod.sgml: Remove references to legacy map files
Robby Workman [Thu, 12 Jan 2012 05:40:24 +0000 (23:40 -0600)]
doc/depmod.sgml: Remove refs to depmod.conf
Robby Workman [Thu, 12 Jan 2012 05:37:03 +0000 (23:37 -0600)]
doc/depmod.d.sgml: Various touchups; mostly s/depmod.conf/depmod.d/g
Robby Workman [Thu, 12 Jan 2012 06:06:27 +0000 (00:06 -0600)]
doc/modprobe.d.sgml: Various touchups
* change references to "kmod" instead of "module-init-tools"
* remove references to modprobe.conf
Robby Workman [Thu, 12 Jan 2012 06:04:57 +0000 (00:04 -0600)]
doc: Import sgml manpages from module-init-tools repo
Lucas De Marchi [Wed, 11 Jan 2012 23:48:08 +0000 (21:48 -0200)]
modprobe: show if module is in kernel
Now with './tools/modprobe --show-depends ahci' (ahci is builtin) we have the following
output:
$ ./tools/modprobe --show-depends ahci
builtin ahci
Just like modprobe from m-i-t. Previously we had:
$ ./tools/modprobe --show-depends ahci
FATAL: Module ahci not found.
Lucas De Marchi [Wed, 11 Jan 2012 23:22:21 +0000 (21:22 -0200)]
test: add check of module's state
Jan Alexander Steffens (heftig) [Wed, 11 Jan 2012 22:17:38 +0000 (23:17 +0100)]
depmod: Postpone creation of module array
Deleting modules (we have found replacements) invalidates the indices
because the array collapses removed elements, hitting the assertion.
Since we don't make use of the array until the sorting step, build it from
the modules_by_name hash instead.
Lucas De Marchi [Wed, 11 Jan 2012 20:29:55 +0000 (18:29 -0200)]
modprobe: break dependency loop by checking if module is loaded
modprobe doesn't have support for handling dependency loop. That happens
with poorly written softdeps that can introduce a loop. We must deal
with them like it's being done in libkmod.
However, we can break a dependency loop when the dependency was already
inserted. This commit fixes this issue, that happens in the following
scenario:
dependencies:
-------------
modA:
modB: modA
modC: modA
config:
softdep modA post: modB modC
This creates the following loop:
modA
inserted ok
handle post-soft-deps of modA -> modB modC
modB
handle dependencies of modB -> modA
modA is already inserted
handle post-soft-deps of modA -> modB modC
And so on and so forth.
Now we break the loop by checking if module is already inserted, before
handling it. Thus this gives us:
modA
inserted ok
handle post-soft-deps of modA -> modB modC
modB
handle dependencies of modB -> modA
modA is already inserted
inserted ok
modC
handle dependencies of modC -> modA
modA is already inserted
inserted ok
Lucas De Marchi [Wed, 11 Jan 2012 19:54:05 +0000 (17:54 -0200)]
autogen.sh: configure with 3-clicks + paste
Lucas De Marchi [Wed, 11 Jan 2012 17:38:50 +0000 (15:38 -0200)]
modprobe: adhere do coding style
Dave Reisner [Wed, 11 Jan 2012 00:28:30 +0000 (19:28 -0500)]
rmmod: behavior more like m-i-t's rmmod
In line with m-i-t's behavior, we should check to see if each module is:
- loaded
- has any holders
- has a 0 refcnt
Detecting any of these lets us provide a more useful message than the
kernel's EPERM response to delete_module(2).
Additionally, alter the main loop behavior to avoid exiting early on the
first error.
Dave Reisner [Tue, 10 Jan 2012 18:36:27 +0000 (13:36 -0500)]
modprobe: check for EPERM on insertion
Throw an appropriate error when an unprivileged user attempts to load a
module.
Lucas De Marchi [Wed, 11 Jan 2012 02:28:12 +0000 (00:28 -0200)]
config: deprecate 'include' and 'config' commands
Lucas De Marchi [Wed, 11 Jan 2012 02:08:14 +0000 (00:08 -0200)]
Fix error code returned on module removal
Lucas De Marchi [Tue, 10 Jan 2012 18:22:59 +0000 (16:22 -0200)]
build-sys: fix build with zlib/xz after private lib
Lucas De Marchi [Tue, 10 Jan 2012 17:45:14 +0000 (15:45 -0200)]
TODO: add task to modprobe
Lucas De Marchi [Tue, 10 Jan 2012 16:06:24 +0000 (14:06 -0200)]
build-sys: workaround libtool issue with argv[0]
Symlinking tools to kmod doesn't work because argv[0] is not the name of
the symlink, but rather 'kmod' (since libtool's wrapper script calls the
tools/.libs/kmod directly)
Now we create another binary kmod-nolib that is statically linked to
libkmod so we can call the binary directly and do not worry about
LD_LIBRARY_PATH.
Lucas De Marchi [Tue, 10 Jan 2012 13:34:12 +0000 (11:34 -0200)]
build-sys: build tests as part of check phase
Do not unconditionally compile tests, make them as target of 'make
check'. Only those listed in TESTS var are run.
Lucas De Marchi [Tue, 10 Jan 2012 13:15:19 +0000 (11:15 -0200)]
build-sys: create symlinks instead of building separate tools
Lucas De Marchi [Tue, 10 Jan 2012 14:55:13 +0000 (12:55 -0200)]
build-sys: allow building without having gtk-doc installed
Lucas De Marchi [Tue, 10 Jan 2012 17:00:07 +0000 (15:00 -0200)]
config: be reliable againt syntax errors in configs
Lucas De Marchi [Tue, 10 Jan 2012 12:19:01 +0000 (10:19 -0200)]
TODO: add tasks to modprobe
Lucas De Marchi [Mon, 9 Jan 2012 15:32:31 +0000 (13:32 -0200)]
doc: fix out of tree build
Lucas De Marchi [Mon, 9 Jan 2012 15:26:06 +0000 (13:26 -0200)]
build-sys: give recommended configure options for hacking
Lucas De Marchi [Mon, 9 Jan 2012 15:15:46 +0000 (13:15 -0200)]
README: add more information
Lucas De Marchi [Mon, 9 Jan 2012 15:00:28 +0000 (13:00 -0200)]
build-sys: do not run configure in autogen.sh
Lucas De Marchi [Mon, 9 Jan 2012 06:47:40 +0000 (04:47 -0200)]
doc: organize sections
Lucas De Marchi [Mon, 9 Jan 2012 06:20:55 +0000 (04:20 -0200)]
doc: add gtk-doc to generate documentation
Current limitation is horrible no support to sections: we have to to
have separate header files or to maintain the libkmod-sections.txt file.
We are doing the latter.
Lucas De Marchi [Mon, 9 Jan 2012 05:45:19 +0000 (03:45 -0200)]
doc: fix broken links to other functions
Lucas De Marchi [Mon, 9 Jan 2012 05:09:49 +0000 (03:09 -0200)]
doc: fix order of param descriptions
Lucas De Marchi [Mon, 9 Jan 2012 05:08:26 +0000 (03:08 -0200)]
Match param names on header and source code
This is needed by gtk-doc (to be added later) to be able to properly
document the function.
Lucas De Marchi [Mon, 9 Jan 2012 04:35:41 +0000 (02:35 -0200)]
doc: fix comments format
Lucas De Marchi [Mon, 9 Jan 2012 02:40:50 +0000 (00:40 -0200)]
Update copyright
Lucas De Marchi [Sun, 8 Jan 2012 20:08:05 +0000 (18:08 -0200)]
modprobe: fix error path when loading dependencies
demarchi> scenario is the following:
demarchi> modA depends on modB and modC
demarchi> if there's a race when trying to insert a dependency of a module, say
modB, it will stop loading all the modules
demarchi> it should check by "module already loaded error"
demarchi> like it does for modA
Gustavo Sverzut Barbieri [Sun, 8 Jan 2012 15:58:28 +0000 (13:58 -0200)]
modprobe: show is independent from verbose level.
"modprobe --quiet --show-depends" now works.
Lucas De Marchi [Sun, 8 Jan 2012 03:02:29 +0000 (01:02 -0200)]
Replace NAME_MAX with PATH_MAX for module aliases
Module aliases can be bigger than NAME_MAX. So, replace with PATH_MAX
that is bigger enough to hold them.
Technically in some places NAME_MAX would be sufficient (those using
module names only), but they use functions that can be called with
alias. So increase the buffers in these cases to PATH_MAX too.
Lucas De Marchi [Fri, 6 Jan 2012 21:22:41 +0000 (19:22 -0200)]
config: check if opening /proc/cmdline succeeded