1 .. SPDX-License-Identifier: GPL-2.0
7 ===================== ======================================= ================
8 /proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999
9 Bodo Bauer <bb@ricochet.net>
10 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
11 move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009
12 fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009
13 ===================== ======================================= ================
20 0.1 Introduction/Credits
23 1 Collecting System Information
24 1.1 Process-Specific Subdirectories
26 1.3 IDE devices in /proc/ide
27 1.4 Networking info in /proc/net
29 1.6 Parallel port info in /proc/parport
30 1.7 TTY info in /proc/tty
31 1.8 Miscellaneous kernel statistics in /proc/stat
32 1.9 Ext4 file system parameters
34 2 Modifying System Parameters
36 3 Per-Process Parameters
37 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer
39 3.2 /proc/<pid>/oom_score - Display current oom-killer score
40 3.3 /proc/<pid>/io - Display the IO accounting fields
41 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
42 3.5 /proc/<pid>/mountinfo - Information about mounts
43 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
44 3.7 /proc/<pid>/task/<tid>/children - Information about task children
45 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
46 3.9 /proc/<pid>/map_files - Information about memory mapped files
47 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
48 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
49 3.12 /proc/<pid>/arch_status - Task architecture specific information
57 0.1 Introduction/Credits
58 ------------------------
60 This documentation is part of a soon (or so we hope) to be released book on
61 the SuSE Linux distribution. As there is no complete documentation for the
62 /proc file system and we've used many freely available sources to write these
63 chapters, it seems only fair to give the work back to the Linux community.
64 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
65 afraid it's still far from complete, but we hope it will be useful. As far as
66 we know, it is the first 'all-in-one' document about the /proc file system. It
67 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
68 SPARC, AXP, etc., features, you probably won't find what you are looking for.
69 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
70 additions and patches are welcome and will be added to this document if you
73 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
74 other people for help compiling this documentation. We'd also like to extend a
75 special thank you to Andi Kleen for documentation, which we relied on heavily
76 to create this document, as well as the additional information he provided.
77 Thanks to everybody else who contributed source or docs to the Linux kernel
78 and helped create a great piece of software... :)
80 If you have any comments, corrections or additions, please don't hesitate to
81 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
84 The latest version of this document is available online at
85 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
87 If the above direction does not works for you, you could try the kernel
88 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
89 comandante@zaralinux.com.
94 We don't guarantee the correctness of this document, and if you come to us
95 complaining about how you screwed up your system because of incorrect
96 documentation, we won't feel responsible...
98 Chapter 1: Collecting System Information
99 ========================================
103 * Investigating the properties of the pseudo file system /proc and its
104 ability to provide information on the running Linux system
105 * Examining /proc's structure
106 * Uncovering various information about the kernel and the processes running
109 ------------------------------------------------------------------------------
111 The proc file system acts as an interface to internal data structures in the
112 kernel. It can be used to obtain information about the system and to change
113 certain kernel parameters at runtime (sysctl).
115 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
116 show you how you can use /proc/sys to change settings.
118 1.1 Process-Specific Subdirectories
119 -----------------------------------
121 The directory /proc contains (among other things) one subdirectory for each
122 process running on the system, which is named after the process ID (PID).
124 The link self points to the process reading the file system. Each process
125 subdirectory has the entries listed in Table 1-1.
127 Note that an open a file descriptor to /proc/<pid> or to any of its
128 contained files or subdirectories does not prevent <pid> being reused
129 for some other process in the event that <pid> exits. Operations on
130 open /proc/<pid> file descriptors corresponding to dead processes
131 never act on any new process that the kernel may, through chance, have
132 also assigned the process ID <pid>. Instead, operations on these FDs
133 usually fail with ESRCH.
135 .. table:: Table 1-1: Process specific entries in /proc
137 ============= ===============================================================
139 ============= ===============================================================
140 clear_refs Clears page referenced bits shown in smaps output
141 cmdline Command line arguments
142 cpu Current and last cpu in which it was executed (2.4)(smp)
143 cwd Link to the current working directory
144 environ Values of environment variables
145 exe Link to the executable of this process
146 fd Directory, which contains all file descriptors
147 maps Memory maps to executables and library files (2.4)
148 mem Memory held by this process
149 root Link to the root directory of this process
151 statm Process memory status information
152 status Process status in human readable form
153 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function
154 symbol the task is blocked in - or "0" if not blocked.
156 stack Report full stack trace, enable via CONFIG_STACKTRACE
157 smaps An extension based on maps, showing the memory consumption of
158 each mapping and flags associated with it
159 smaps_rollup Accumulated smaps stats for all mappings of the process. This
160 can be derived from smaps, but is faster and more convenient
161 numa_maps An extension based on maps, showing the memory locality and
162 binding policy as well as mem usage (in pages) of each mapping.
163 ============= ===============================================================
165 For example, to get the status information of a process, all you have to do is
166 read the file /proc/PID/status::
168 >cat /proc/self/status
198 SigPnd: 0000000000000000
199 ShdPnd: 0000000000000000
200 SigBlk: 0000000000000000
201 SigIgn: 0000000000000000
202 SigCgt: 0000000000000000
203 CapInh: 00000000fffffeff
204 CapPrm: 0000000000000000
205 CapEff: 0000000000000000
206 CapBnd: ffffffffffffffff
207 CapAmb: 0000000000000000
210 Speculation_Store_Bypass: thread vulnerable
211 voluntary_ctxt_switches: 0
212 nonvoluntary_ctxt_switches: 1
214 This shows you nearly the same information you would get if you viewed it with
215 the ps command. In fact, ps uses the proc file system to obtain its
216 information. But you get a more detailed view of the process by reading the
217 file /proc/PID/status. It fields are described in table 1-2.
219 The statm file contains more detailed information about the process
220 memory usage. Its seven fields are explained in Table 1-3. The stat file
221 contains details information about the process itself. Its fields are
222 explained in Table 1-4.
224 (for SMP CONFIG users)
226 For making accounting scalable, RSS related information are handled in an
227 asynchronous manner and the value may not be very precise. To see a precise
228 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
229 It's slow but very precise.
231 .. table:: Table 1-2: Contents of the status files (as of 4.19)
233 ========================== ===================================================
235 ========================== ===================================================
236 Name filename of the executable
237 Umask file mode creation mask
238 State state (R is running, S is sleeping, D is sleeping
239 in an uninterruptible wait, Z is zombie,
240 T is traced or stopped)
242 Ngid NUMA group ID (0 if none)
244 PPid process id of the parent process
245 TracerPid PID of process tracing this process (0 if not)
246 Uid Real, effective, saved set, and file system UIDs
247 Gid Real, effective, saved set, and file system GIDs
248 FDSize number of file descriptor slots currently allocated
249 Groups supplementary group list
250 NStgid descendant namespace thread group ID hierarchy
251 NSpid descendant namespace process ID hierarchy
252 NSpgid descendant namespace process group ID hierarchy
253 NSsid descendant namespace session ID hierarchy
254 VmPeak peak virtual memory size
255 VmSize total program size
256 VmLck locked memory size
257 VmPin pinned memory size
258 VmHWM peak resident set size ("high water mark")
259 VmRSS size of memory portions. It contains the three
261 (VmRSS = RssAnon + RssFile + RssShmem)
262 RssAnon size of resident anonymous memory
263 RssFile size of resident file mappings
264 RssShmem size of resident shmem memory (includes SysV shm,
265 mapping of tmpfs and shared anonymous mappings)
266 VmData size of private data segments
267 VmStk size of stack segments
268 VmExe size of text segment
269 VmLib size of shared library code
270 VmPTE size of page table entries
271 VmSwap amount of swap used by anonymous private data
272 (shmem swap usage is not included)
273 HugetlbPages size of hugetlb memory portions
274 CoreDumping process's memory is currently being dumped
275 (killing the process may lead to a corrupted core)
276 THP_enabled process is allowed to use THP (returns 0 when
277 PR_SET_THP_DISABLE is set on the process
278 Threads number of threads
279 SigQ number of signals queued/max. number for queue
280 SigPnd bitmap of pending signals for the thread
281 ShdPnd bitmap of shared pending signals for the process
282 SigBlk bitmap of blocked signals
283 SigIgn bitmap of ignored signals
284 SigCgt bitmap of caught signals
285 CapInh bitmap of inheritable capabilities
286 CapPrm bitmap of permitted capabilities
287 CapEff bitmap of effective capabilities
288 CapBnd bitmap of capabilities bounding set
289 CapAmb bitmap of ambient capabilities
290 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
291 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
292 Speculation_Store_Bypass speculative store bypass mitigation status
293 Cpus_allowed mask of CPUs on which this process may run
294 Cpus_allowed_list Same as previous, but in "list format"
295 Mems_allowed mask of memory nodes allowed to this process
296 Mems_allowed_list Same as previous, but in "list format"
297 voluntary_ctxt_switches number of voluntary context switches
298 nonvoluntary_ctxt_switches number of non voluntary context switches
299 ========================== ===================================================
302 .. table:: Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
304 ======== =============================== ==============================
306 ======== =============================== ==============================
307 size total program size (pages) (same as VmSize in status)
308 resident size of memory portions (pages) (same as VmRSS in status)
309 shared number of pages that are shared (i.e. backed by a file, same
310 as RssFile+RssShmem in status)
311 trs number of pages that are 'code' (not including libs; broken,
312 includes data segment)
313 lrs number of pages of library (always 0 on 2.6)
314 drs number of pages of data/stack (including libs; broken,
315 includes library text)
316 dt number of dirty pages (always 0 on 2.6)
317 ======== =============================== ==============================
320 .. table:: Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
322 ============= ===============================================================
324 ============= ===============================================================
326 tcomm filename of the executable
327 state state (R is running, S is sleeping, D is sleeping in an
328 uninterruptible wait, Z is zombie, T is traced or stopped)
329 ppid process id of the parent process
330 pgrp pgrp of the process
332 tty_nr tty the process uses
333 tty_pgrp pgrp of the tty
335 min_flt number of minor faults
336 cmin_flt number of minor faults with child's
337 maj_flt number of major faults
338 cmaj_flt number of major faults with child's
339 utime user mode jiffies
340 stime kernel mode jiffies
341 cutime user mode jiffies with child's
342 cstime kernel mode jiffies with child's
343 priority priority level
345 num_threads number of threads
346 it_real_value (obsolete, always 0)
347 start_time time the process started after system boot
348 vsize virtual memory size
349 rss resident set memory size
350 rsslim current limit in bytes on the rss
351 start_code address above which program text can run
352 end_code address below which program text can run
353 start_stack address of the start of the main process stack
354 esp current value of ESP
355 eip current value of EIP
356 pending bitmap of pending signals
357 blocked bitmap of blocked signals
358 sigign bitmap of ignored signals
359 sigcatch bitmap of caught signals
360 0 (place holder, used to be the wchan address,
361 use /proc/PID/wchan instead)
364 exit_signal signal to send to parent thread on exit
365 task_cpu which CPU the task is scheduled on
366 rt_priority realtime priority
367 policy scheduling policy (man sched_setscheduler)
368 blkio_ticks time spent waiting for block IO
369 gtime guest time of the task in jiffies
370 cgtime guest time of the task children in jiffies
371 start_data address above which program data+bss is placed
372 end_data address below which program data+bss is placed
373 start_brk address above which program heap can be expanded with brk()
374 arg_start address above which program command line is placed
375 arg_end address below which program command line is placed
376 env_start address above which program environment is placed
377 env_end address below which program environment is placed
378 exit_code the thread's exit_code in the form reported by the waitpid
380 ============= ===============================================================
382 The /proc/PID/maps file contains the currently mapped memory regions and
383 their access permissions.
387 address perms offset dev inode pathname
389 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
390 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
391 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
392 a7cb1000-a7cb2000 ---p 00000000 00:00 0
393 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
394 a7eb2000-a7eb3000 ---p 00000000 00:00 0
395 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
396 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
397 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
398 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
399 a800b000-a800e000 rw-p 00000000 00:00 0
400 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
401 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
402 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
403 a8024000-a8027000 rw-p 00000000 00:00 0
404 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
405 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
406 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
407 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
408 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
410 where "address" is the address space in the process that it occupies, "perms"
411 is a set of permissions::
417 p = private (copy on write)
419 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
420 "inode" is the inode on that device. 0 indicates that no inode is associated
421 with the memory region, as the case would be with BSS (uninitialized data).
422 The "pathname" shows the name associated file for this mapping. If the mapping
423 is not associated with a file:
425 ======= ====================================
426 [heap] the heap of the program
427 [stack] the stack of the main process
428 [vdso] the "virtual dynamic shared object",
429 the kernel system call handler
430 ======= ====================================
432 or if empty, the mapping is anonymous.
434 The /proc/PID/smaps is an extension based on maps, showing the memory
435 consumption for each of the process's mappings. For each mapping (aka Virtual
436 Memory Area, or VMA) there is a series of lines such as the following::
438 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
455 Private_Hugetlb: 0 kB
462 VmFlags: rd ex mr mw me dw
464 The first of these lines shows the same information as is displayed for the
465 mapping in /proc/PID/maps. Following lines show the size of the mapping
466 (size); the size of each page allocated when backing a VMA (KernelPageSize),
467 which is usually the same as the size in the page table entries; the page size
468 used by the MMU when backing a VMA (in most cases, the same as KernelPageSize);
469 the amount of the mapping that is currently resident in RAM (RSS); the
470 process' proportional share of this mapping (PSS); and the number of clean and
471 dirty shared and private pages in the mapping.
473 The "proportional set size" (PSS) of a process is the count of pages it has
474 in memory, where each page is divided by the number of processes sharing it.
475 So if a process has 1000 pages all to itself, and 1000 shared with one other
476 process, its PSS will be 1500.
478 Note that even a page which is part of a MAP_SHARED mapping, but has only
479 a single pte mapped, i.e. is currently used by only one process, is accounted
480 as private and not as shared.
482 "Referenced" indicates the amount of memory currently marked as referenced or
485 "Anonymous" shows the amount of memory that does not belong to any file. Even
486 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
487 and a page is modified, the file page is replaced by a private anonymous copy.
489 "LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
490 The memory isn't freed immediately with madvise(). It's freed in memory
491 pressure if the memory is clean. Please note that the printed value might
492 be lower than the real value due to optimizations used in the current
493 implementation. If this is not desirable please file a bug report.
495 "AnonHugePages" shows the ammount of memory backed by transparent hugepage.
497 "ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by
500 "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
501 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
502 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
504 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
506 For shmem mappings, "Swap" includes also the size of the mapped (and not
507 replaced by copy-on-write) part of the underlying shmem object out on swap.
508 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
509 does not take into account swapped out page of underlying shmem objects.
510 "Locked" indicates whether the mapping is locked in memory or not.
511 "THPeligible" indicates whether the mapping is eligible for allocating THP
512 pages - 1 if true, 0 otherwise. It just shows the current status.
514 "VmFlags" field deserves a separate description. This member represents the
515 kernel flags associated with the particular virtual memory area in two letter
516 encoded manner. The codes are the following:
518 == =======================================
527 gd stack segment growns down
529 dw disabled write to the mapped file
530 lo pages are locked in memory
531 io memory mapped I/O area
532 sr sequential read advise provided
533 rr random read advise provided
534 dc do not copy area on fork
535 de do not expand area on remapping
536 ac area is accountable
537 nr swap space is not reserved for the area
538 ht area uses huge tlb pages
539 ar architecture specific flag
540 dd do not include area into core dump
543 hg huge page advise flag
544 nh no huge page advise flag
545 mg mergable advise flag
546 bt - arm64 BTI guarded page
547 == =======================================
549 Note that there is no guarantee that every flag and associated mnemonic will
550 be present in all further kernel releases. Things get changed, the flags may
551 be vanished or the reverse -- new added. Interpretation of their meaning
552 might change in future as well. So each consumer of these flags has to
553 follow each specific kernel version for the exact semantic.
555 This file is only present if the CONFIG_MMU kernel configuration option is
558 Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
559 output can be achieved only in the single read call).
561 This typically manifests when doing partial reads of these files while the
562 memory map is being modified. Despite the races, we do provide the following
565 1) The mapped addresses never go backwards, which implies no two
566 regions will ever overlap.
567 2) If there is something at a given vaddr during the entirety of the
568 life of the smaps/maps walk, there will be some output for it.
570 The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps,
571 but their values are the sums of the corresponding values for all mappings of
572 the process. Additionally, it contains these fields:
578 They represent the proportional shares of anonymous, file, and shmem pages, as
579 described for smaps above. These fields are omitted in smaps since each
580 mapping identifies the type (anon, file, or shmem) of all pages it contains.
581 Thus all information in smaps_rollup can be derived from smaps, but at a
582 significantly higher cost.
584 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
585 bits on both physical and virtual pages associated with a process, and the
586 soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst
588 To clear the bits for all the pages associated with the process::
590 > echo 1 > /proc/PID/clear_refs
592 To clear the bits for the anonymous pages associated with the process::
594 > echo 2 > /proc/PID/clear_refs
596 To clear the bits for the file mapped pages associated with the process::
598 > echo 3 > /proc/PID/clear_refs
600 To clear the soft-dirty bit::
602 > echo 4 > /proc/PID/clear_refs
604 To reset the peak resident set size ("high water mark") to the process's
607 > echo 5 > /proc/PID/clear_refs
609 Any other value written to /proc/PID/clear_refs will have no effect.
611 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
612 using /proc/kpageflags and number of times a page is mapped using
613 /proc/kpagecount. For detailed explanation, see
614 Documentation/admin-guide/mm/pagemap.rst.
616 The /proc/pid/numa_maps is an extension based on maps, showing the memory
617 locality and binding policy, as well as the memory usage (in pages) of
618 each mapping. The output follows a general format where mapping details get
619 summarized separated by blank spaces, one mapping per each file line::
621 address policy mapping details
623 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
624 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
625 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
626 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
627 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
628 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
629 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
630 320698b000 default file=/lib64/libc-2.12.so
631 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
632 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
633 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
634 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
635 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
636 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
637 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
638 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
642 "address" is the starting address for the mapping;
644 "policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst);
646 "mapping details" summarizes mapping data such as mapping type, page usage counters,
647 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
648 size, in KB, that is backing the mapping up.
653 Similar to the process entries, the kernel data files give information about
654 the running kernel. The files used to obtain this information are contained in
655 /proc and are listed in Table 1-5. Not all of these will be present in your
656 system. It depends on the kernel configuration and the loaded modules, which
657 files are there, and which are missing.
659 .. table:: Table 1-5: Kernel info in /proc
661 ============ ===============================================================
663 ============ ===============================================================
664 apm Advanced power management info
665 buddyinfo Kernel memory allocator information (see text) (2.5)
666 bus Directory containing bus specific information
667 cmdline Kernel command line
668 cpuinfo Info about the CPU
669 devices Available devices (block and character)
670 dma Used DMS channels
671 filesystems Supported filesystems
672 driver Various drivers grouped here, currently rtc (2.4)
673 execdomains Execdomains, related to security (2.4)
674 fb Frame Buffer devices (2.4)
675 fs File system parameters, currently nfs/exports (2.4)
676 ide Directory containing info about the IDE subsystem
677 interrupts Interrupt usage
678 iomem Memory map (2.4)
679 ioports I/O port usage
680 irq Masks for irq to cpu affinity (2.4)(smp?)
681 isapnp ISA PnP (Plug&Play) Info (2.4)
682 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
684 ksyms Kernel symbol table
685 loadavg Load average of last 1, 5 & 15 minutes
689 modules List of loaded modules
690 mounts Mounted filesystems
691 net Networking info (see text)
692 pagetypeinfo Additional page allocator information (see text) (2.5)
693 partitions Table of partitions known to the system
694 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
695 decoupled by lspci (2.4)
697 scsi SCSI info (see text)
698 slabinfo Slab pool info
699 softirqs softirq usage
700 stat Overall statistics
701 swaps Swap space utilization
703 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
704 tty Info of tty drivers
705 uptime Wall clock since boot, combined idle time of all cpus
706 version Kernel version
707 video bttv info of video resources (2.4)
708 vmallocinfo Show vmalloced areas
709 ============ ===============================================================
711 You can, for example, check which interrupts are currently in use and what
712 they are used for by looking in the file /proc/interrupts::
714 > cat /proc/interrupts
716 0: 8728810 XT-PIC timer
717 1: 895 XT-PIC keyboard
719 3: 531695 XT-PIC aha152x
720 4: 2014133 XT-PIC serial
721 5: 44401 XT-PIC pcnet_cs
724 12: 182918 XT-PIC PS/2 Mouse
726 14: 1232265 XT-PIC ide0
730 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
731 output of a SMP machine)::
733 > cat /proc/interrupts
736 0: 1243498 1214548 IO-APIC-edge timer
737 1: 8949 8958 IO-APIC-edge keyboard
738 2: 0 0 XT-PIC cascade
739 5: 11286 10161 IO-APIC-edge soundblaster
740 8: 1 0 IO-APIC-edge rtc
741 9: 27422 27407 IO-APIC-edge 3c503
742 12: 113645 113873 IO-APIC-edge PS/2 Mouse
744 14: 22491 24012 IO-APIC-edge ide0
745 15: 2183 2415 IO-APIC-edge ide1
746 17: 30564 30414 IO-APIC-level eth0
747 18: 177 164 IO-APIC-level bttv
752 NMI is incremented in this case because every timer interrupt generates a NMI
753 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
755 LOC is the local interrupt counter of the internal APIC of every CPU.
757 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
758 connects the CPUs in a SMP system. This means that an error has been detected,
759 the IO-APIC automatically retry the transmission, so it should not be a big
760 problem, but you should read the SMP-FAQ.
762 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
763 /proc/interrupts to display every IRQ vector in use by the system, not
764 just those considered 'most important'. The new vectors are:
767 interrupt raised when a machine check threshold counter
768 (typically counting ECC corrected errors of memory or cache) exceeds
769 a configurable threshold. Only available on some systems.
772 a thermal event interrupt occurs when a temperature threshold
773 has been exceeded for the CPU. This interrupt may also be generated
774 when the temperature drops back to normal.
777 a spurious interrupt is some interrupt that was raised then lowered
778 by some IO device before it could be fully processed by the APIC. Hence
779 the APIC sees the interrupt but does not know what device it came from.
780 For this case the APIC will generate the interrupt with a IRQ vector
781 of 0xff. This might also be generated by chipset bugs.
784 rescheduling, call and TLB flush interrupts are
785 sent from one CPU to another per the needs of the OS. Typically,
786 their statistics are used by kernel developers and interested users to
787 determine the occurrence of interrupts of the given type.
789 The above IRQ vectors are displayed only when relevant. For example,
790 the threshold vector does not exist on x86_64 platforms. Others are
791 suppressed when the system is a uniprocessor. As of this writing, only
792 i386 and x86_64 platforms support the new IRQ vector displays.
794 Of some interest is the introduction of the /proc/irq directory to 2.4.
795 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
796 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
797 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
803 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
804 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
808 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
809 IRQ, you can set it by doing::
811 > echo 1 > /proc/irq/10/smp_affinity
813 This means that only the first CPU will handle the IRQ, but you can also echo
814 5 which means that only the first and third CPU can handle the IRQ.
816 The contents of each smp_affinity file is the same by default::
818 > cat /proc/irq/0/smp_affinity
821 There is an alternate interface, smp_affinity_list which allows specifying
822 a cpu range instead of a bitmask::
824 > cat /proc/irq/0/smp_affinity_list
827 The default_smp_affinity mask applies to all non-active IRQs, which are the
828 IRQs which have not yet been allocated/activated, and hence which lack a
829 /proc/irq/[0-9]* directory.
831 The node file on an SMP system shows the node to which the device using the IRQ
832 reports itself as being attached. This hardware locality information does not
833 include information about any possible driver locality preference.
835 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
836 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
838 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
839 between all the CPUs which are allowed to handle it. As usual the kernel has
840 more info than you and does a better job than you, so the defaults are the
841 best choice for almost everyone. [Note this applies only to those IO-APIC's
842 that support "Round Robin" interrupt distribution.]
844 There are three more important subdirectories in /proc: net, scsi, and sys.
845 The general rule is that the contents, or even the existence of these
846 directories, depend on your kernel configuration. If SCSI is not enabled, the
847 directory scsi may not exist. The same is true with the net, which is there
848 only when networking support is present in the running kernel.
850 The slabinfo file gives information about memory usage at the slab level.
851 Linux uses slab pools for memory management above page level in version 2.2.
852 Commonly used objects have their own slab pool (such as network buffers,
853 directory cache, and so on).
857 > cat /proc/buddyinfo
859 Node 0, zone DMA 0 4 5 4 4 3 ...
860 Node 0, zone Normal 1 0 0 1 101 8 ...
861 Node 0, zone HighMem 2 0 0 1 1 0 ...
863 External fragmentation is a problem under some workloads, and buddyinfo is a
864 useful tool for helping diagnose these problems. Buddyinfo will give you a
865 clue as to how big an area you can safely allocate, or why a previous
868 Each column represents the number of pages of a certain order which are
869 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
870 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
871 available in ZONE_NORMAL, etc...
873 More information relevant to external fragmentation can be found in
876 > cat /proc/pagetypeinfo
880 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
881 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
882 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
883 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
884 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
885 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
886 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
887 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
888 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
889 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
890 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
892 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
893 Node 0, zone DMA 2 0 5 1 0
894 Node 0, zone DMA32 41 6 967 2 0
896 Fragmentation avoidance in the kernel works by grouping pages of different
897 migrate types into the same contiguous regions of memory called page blocks.
898 A page block is typically the size of the default hugepage size e.g. 2MB on
899 X86-64. By keeping pages grouped based on their ability to move, the kernel
900 can reclaim pages within a page block to satisfy a high-order allocation.
902 The pagetypinfo begins with information on the size of a page block. It
903 then gives the same type of information as buddyinfo except broken down
904 by migrate-type and finishes with details on how many page blocks of each
907 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
908 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
909 make an estimate of the likely number of huge pages that can be allocated
910 at a given point in time. All the "Movable" blocks should be allocatable
911 unless memory has been mlock()'d. Some of the Reclaimable blocks should
912 also be allocatable although a lot of filesystem metadata may have to be
913 reclaimed to achieve this.
919 Provides information about distribution and utilization of memory. This
920 varies by architecture and compile options. The following is from a
921 16GB PIII, which has highmem enabled. You may not have all of these fields.
927 MemTotal: 16344972 kB
929 MemAvailable: 14836172 kB
935 HighTotal: 15597528 kB
936 HighFree: 13629632 kB
946 KReclaimable: 168048 kB
948 SReclaimable: 159856 kB
949 SUnreclaim: 124508 kB
954 CommitLimit: 7669796 kB
955 Committed_AS: 100056 kB
956 VmallocTotal: 112216 kB
958 VmallocChunk: 111088 kB
960 HardwareCorrupted: 0 kB
961 AnonHugePages: 49152 kB
966 Total usable ram (i.e. physical ram minus a few reserved
967 bits and the kernel binary code)
969 The sum of LowFree+HighFree
971 An estimate of how much memory is available for starting new
972 applications, without swapping. Calculated from MemFree,
973 SReclaimable, the size of the file LRU lists, and the low
974 watermarks in each zone.
975 The estimate takes into account that the system needs some
976 page cache to function well, and that not all reclaimable
977 slab will be reclaimable, due to items being in use. The
978 impact of those factors will vary from system to system.
980 Relatively temporary storage for raw disk blocks
981 shouldn't get tremendously large (20MB or so)
983 in-memory cache for files read from the disk (the
984 pagecache). Doesn't include SwapCached
986 Memory that once was swapped out, is swapped back in but
987 still also is in the swapfile (if memory is needed it
988 doesn't need to be swapped out AGAIN because it is already
989 in the swapfile. This saves I/O)
991 Memory that has been used more recently and usually not
992 reclaimed unless absolutely necessary.
994 Memory which has been less recently used. It is more
995 eligible to be reclaimed for other purposes
997 Highmem is all memory above ~860MB of physical memory
998 Highmem areas are for use by userspace programs, or
999 for the pagecache. The kernel must use tricks to access
1000 this memory, making it slower to access than lowmem.
1002 Lowmem is memory which can be used for everything that
1003 highmem can be used for, but it is also available for the
1004 kernel's use for its own data structures. Among many
1005 other things, it is where everything from the Slab is
1006 allocated. Bad things happen when you're out of lowmem.
1008 total amount of swap space available
1010 Memory which has been evicted from RAM, and is temporarily
1013 Memory which is waiting to get written back to the disk
1015 Memory which is actively being written back to the disk
1017 Non-file backed pages mapped into userspace page tables
1019 The amount of RAM/memory in KB, the kernel identifies as
1022 Non-file backed huge pages mapped into userspace page tables
1024 files which have been mmaped, such as libraries
1026 Total memory used by shared memory (shmem) and tmpfs
1028 Memory used by shared memory (shmem) and tmpfs allocated
1031 Shared memory mapped into userspace with huge pages
1033 Kernel allocations that the kernel will attempt to reclaim
1034 under memory pressure. Includes SReclaimable (below), and other
1035 direct allocations with a shrinker.
1037 in-kernel data structures cache
1039 Part of Slab, that might be reclaimed, such as caches
1041 Part of Slab, that cannot be reclaimed on memory pressure
1043 amount of memory dedicated to the lowest level of page
1046 Always zero. Previous counted pages which had been written to
1047 the server, but has not been committed to stable storage.
1049 Memory used for block device "bounce buffers"
1051 Memory used by FUSE for temporary writeback buffers
1053 Based on the overcommit ratio ('vm.overcommit_ratio'),
1054 this is the total amount of memory currently available to
1055 be allocated on the system. This limit is only adhered to
1056 if strict overcommit accounting is enabled (mode 2 in
1057 'vm.overcommit_memory').
1059 The CommitLimit is calculated with the following formula::
1061 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
1062 overcommit_ratio / 100 + [total swap pages]
1064 For example, on a system with 1G of physical RAM and 7G
1065 of swap with a `vm.overcommit_ratio` of 30 it would
1066 yield a CommitLimit of 7.3G.
1068 For more details, see the memory overcommit documentation
1069 in vm/overcommit-accounting.
1071 The amount of memory presently allocated on the system.
1072 The committed memory is a sum of all of the memory which
1073 has been allocated by processes, even if it has not been
1074 "used" by them as of yet. A process which malloc()'s 1G
1075 of memory, but only touches 300M of it will show up as
1076 using 1G. This 1G is memory which has been "committed" to
1077 by the VM and can be used at any time by the allocating
1078 application. With strict overcommit enabled on the system
1079 (mode 2 in 'vm.overcommit_memory'),allocations which would
1080 exceed the CommitLimit (detailed above) will not be permitted.
1081 This is useful if one needs to guarantee that processes will
1082 not fail due to lack of memory once that memory has been
1083 successfully allocated.
1085 total size of vmalloc memory area
1087 amount of vmalloc area which is used
1089 largest contiguous block of vmalloc area which is free
1091 Memory allocated to the percpu allocator used to back percpu
1092 allocations. This stat excludes the cost of metadata.
1097 Provides information about vmalloced/vmaped areas. One line per area,
1098 containing the virtual address range of the area, size in bytes,
1099 caller information of the creator, and optional information depending
1100 on the kind of area :
1102 ========== ===================================================
1103 pages=nr number of pages
1104 phys=addr if a physical address was specified
1105 ioremap I/O mapping (ioremap() and friends)
1106 vmalloc vmalloc() area
1108 user VM_USERMAP area
1109 vpages buffer for pages pointers was vmalloced (huge area)
1110 N<node>=nr (Only on NUMA kernels)
1111 Number of pages allocated on memory node <node>
1112 ========== ===================================================
1116 > cat /proc/vmallocinfo
1117 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
1118 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
1119 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
1120 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
1121 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
1122 phys=7fee8000 ioremap
1123 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
1124 phys=7fee7000 ioremap
1125 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
1126 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
1127 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
1128 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
1129 pages=2 vmalloc N1=2
1130 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
1131 /0x130 [x_tables] pages=4 vmalloc N0=4
1132 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
1133 pages=14 vmalloc N2=14
1134 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
1135 pages=4 vmalloc N1=4
1136 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
1137 pages=2 vmalloc N1=2
1138 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
1139 pages=10 vmalloc N0=10
1145 Provides counts of softirq handlers serviced since boot time, for each cpu.
1149 > cat /proc/softirqs
1152 TIMER: 27166 27120 27097 27034
1157 SCHED: 27035 26983 26971 26746
1159 RCU: 1678 1769 2178 2250
1162 1.3 IDE devices in /proc/ide
1163 ----------------------------
1165 The subdirectory /proc/ide contains information about all IDE devices of which
1166 the kernel is aware. There is one subdirectory for each IDE controller, the
1167 file drivers and a link for each IDE device, pointing to the device directory
1168 in the controller specific subtree.
1170 The file drivers contains general information about the drivers used for the
1173 > cat /proc/ide/drivers
1174 ide-cdrom version 4.53
1175 ide-disk version 1.08
1177 More detailed information can be found in the controller specific
1178 subdirectories. These are named ide0, ide1 and so on. Each of these
1179 directories contains the files shown in table 1-6.
1182 .. table:: Table 1-6: IDE controller info in /proc/ide/ide?
1184 ======= =======================================
1186 ======= =======================================
1187 channel IDE channel (0 or 1)
1188 config Configuration (only for PCI/IDE bridge)
1190 model Type/Chipset of IDE controller
1191 ======= =======================================
1193 Each device connected to a controller has a separate subdirectory in the
1194 controllers directory. The files listed in table 1-7 are contained in these
1198 .. table:: Table 1-7: IDE device information
1200 ================ ==========================================
1202 ================ ==========================================
1204 capacity Capacity of the medium (in 512Byte blocks)
1205 driver driver and version
1206 geometry physical and logical geometry
1207 identify device identify block
1209 model device identifier
1210 settings device setup
1211 smart_thresholds IDE disk management thresholds
1212 smart_values IDE disk management values
1213 ================ ==========================================
1215 The most interesting file is ``settings``. This file contains a nice
1216 overview of the drive parameters::
1218 # cat /proc/ide/ide0/hda/settings
1219 name value min max mode
1220 ---- ----- --- --- ----
1221 bios_cyl 526 0 65535 rw
1222 bios_head 255 0 255 rw
1223 bios_sect 63 0 63 rw
1224 breada_readahead 4 0 127 rw
1226 file_readahead 72 0 2097151 rw
1228 keepsettings 0 0 1 rw
1229 max_kb_per_request 122 1 127 rw
1233 pio_mode write-only 0 255 w
1239 1.4 Networking info in /proc/net
1240 --------------------------------
1242 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1243 additional values you get for IP version 6 if you configure the kernel to
1244 support this. Table 1-9 lists the files and their meaning.
1247 .. table:: Table 1-8: IPv6 info in /proc/net
1249 ========== =====================================================
1251 ========== =====================================================
1252 udp6 UDP sockets (IPv6)
1253 tcp6 TCP sockets (IPv6)
1254 raw6 Raw device statistics (IPv6)
1255 igmp6 IP multicast addresses, which this host joined (IPv6)
1256 if_inet6 List of IPv6 interface addresses
1257 ipv6_route Kernel routing table for IPv6
1258 rt6_stats Global IPv6 routing tables statistics
1259 sockstat6 Socket statistics (IPv6)
1260 snmp6 Snmp data (IPv6)
1261 ========== =====================================================
1263 .. table:: Table 1-9: Network info in /proc/net
1265 ============= ================================================================
1267 ============= ================================================================
1268 arp Kernel ARP table
1269 dev network devices with statistics
1270 dev_mcast the Layer2 multicast groups a device is listening too
1271 (interface index, label, number of references, number of bound
1273 dev_stat network device status
1274 ip_fwchains Firewall chain linkage
1275 ip_fwnames Firewall chain names
1276 ip_masq Directory containing the masquerading tables
1277 ip_masquerade Major masquerading table
1278 netstat Network statistics
1279 raw raw device statistics
1280 route Kernel routing table
1281 rpc Directory containing rpc info
1282 rt_cache Routing cache
1284 sockstat Socket statistics
1287 unix UNIX domain sockets
1288 wireless Wireless interface data (Wavelan etc)
1289 igmp IP multicast addresses, which this host joined
1290 psched Global packet scheduler parameters.
1291 netlink List of PF_NETLINK sockets
1292 ip_mr_vifs List of multicast virtual interfaces
1293 ip_mr_cache List of multicast routing cache
1294 ============= ================================================================
1296 You can use this information to see which network devices are available in
1297 your system and how much traffic was routed over those devices::
1300 Inter-|Receive |[...
1301 face |bytes packets errs drop fifo frame compressed multicast|[...
1302 lo: 908188 5596 0 0 0 0 0 0 [...
1303 ppp0:15475140 20721 410 0 0 410 0 0 [...
1304 eth0: 614530 7085 0 0 0 0 0 1 [...
1307 ...] bytes packets errs drop fifo colls carrier compressed
1308 ...] 908188 5596 0 0 0 0 0 0
1309 ...] 1375103 17405 0 0 0 0 0 0
1310 ...] 1703981 5535 0 0 0 3 0 0
1312 In addition, each Channel Bond interface has its own directory. For
1313 example, the bond0 device will have a directory called /proc/net/bond0/.
1314 It will contain information that is specific to that bond, such as the
1315 current slaves of the bond, the link status of the slaves, and how
1316 many times the slaves link has failed.
1321 If you have a SCSI host adapter in your system, you'll find a subdirectory
1322 named after the driver for this adapter in /proc/scsi. You'll also see a list
1323 of all recognized SCSI devices in /proc/scsi::
1325 >cat /proc/scsi/scsi
1327 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1328 Vendor: IBM Model: DGHS09U Rev: 03E0
1329 Type: Direct-Access ANSI SCSI revision: 03
1330 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1331 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1332 Type: CD-ROM ANSI SCSI revision: 02
1335 The directory named after the driver has one file for each adapter found in
1336 the system. These files contain information about the controller, including
1337 the used IRQ and the IO address range. The amount of information shown is
1338 dependent on the adapter you use. The example shows the output for an Adaptec
1339 AHA-2940 SCSI adapter::
1341 > cat /proc/scsi/aic7xxx/0
1343 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1345 TCQ Enabled By Default : Disabled
1346 AIC7XXX_PROC_STATS : Disabled
1347 AIC7XXX_RESET_DELAY : 5
1348 Adapter Configuration:
1349 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1350 Ultra Wide Controller
1351 PCI MMAPed I/O Base: 0xeb001000
1352 Adapter SEEPROM Config: SEEPROM found and used.
1353 Adaptec SCSI BIOS: Enabled
1355 SCBs: Active 0, Max Active 2,
1356 Allocated 15, HW 16, Page 255
1358 BIOS Control Word: 0x18b6
1359 Adapter Control Word: 0x005b
1360 Extended Translation: Enabled
1361 Disconnect Enable Flags: 0xffff
1362 Ultra Enable Flags: 0x0001
1363 Tag Queue Enable Flags: 0x0000
1364 Ordered Queue Tag Flags: 0x0000
1365 Default Tag Queue Depth: 8
1366 Tagged Queue By Device array for aic7xxx host instance 0:
1367 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1368 Actual queue depth per device for aic7xxx host instance 0:
1369 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1372 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1373 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1374 Total transfers 160151 (74577 reads and 85574 writes)
1376 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1377 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1378 Total transfers 0 (0 reads and 0 writes)
1381 1.6 Parallel port info in /proc/parport
1382 ---------------------------------------
1384 The directory /proc/parport contains information about the parallel ports of
1385 your system. It has one subdirectory for each port, named after the port
1388 These directories contain the four files shown in Table 1-10.
1391 .. table:: Table 1-10: Files in /proc/parport
1393 ========= ====================================================================
1395 ========= ====================================================================
1396 autoprobe Any IEEE-1284 device ID information that has been acquired.
1397 devices list of the device drivers using that port. A + will appear by the
1398 name of the device currently using the port (it might not appear
1400 hardware Parallel port's base address, IRQ line and DMA channel.
1401 irq IRQ that parport is using for that port. This is in a separate
1402 file to allow you to alter it by writing a new value in (IRQ
1404 ========= ====================================================================
1406 1.7 TTY info in /proc/tty
1407 -------------------------
1409 Information about the available and actually used tty's can be found in the
1410 directory /proc/tty.You'll find entries for drivers and line disciplines in
1411 this directory, as shown in Table 1-11.
1414 .. table:: Table 1-11: Files in /proc/tty
1416 ============= ==============================================
1418 ============= ==============================================
1419 drivers list of drivers and their usage
1420 ldiscs registered line disciplines
1421 driver/serial usage statistic and status of single tty lines
1422 ============= ==============================================
1424 To see which tty's are currently in use, you can simply look into the file
1427 > cat /proc/tty/drivers
1428 pty_slave /dev/pts 136 0-255 pty:slave
1429 pty_master /dev/ptm 128 0-255 pty:master
1430 pty_slave /dev/ttyp 3 0-255 pty:slave
1431 pty_master /dev/pty 2 0-255 pty:master
1432 serial /dev/cua 5 64-67 serial:callout
1433 serial /dev/ttyS 4 64-67 serial
1434 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1435 /dev/ptmx /dev/ptmx 5 2 system
1436 /dev/console /dev/console 5 1 system:console
1437 /dev/tty /dev/tty 5 0 system:/dev/tty
1438 unknown /dev/tty 4 1-63 console
1441 1.8 Miscellaneous kernel statistics in /proc/stat
1442 -------------------------------------------------
1444 Various pieces of information about kernel activity are available in the
1445 /proc/stat file. All of the numbers reported in this file are aggregates
1446 since the system first booted. For a quick look, simply cat the file::
1449 cpu 2255 34 2290 22625563 6290 127 456 0 0 0
1450 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0
1451 cpu1 1123 0 849 11313845 2614 0 18 0 0 0
1452 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1458 softirq 183433 0 21755 12 39 1137 231 21459 2263
1460 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1461 lines. These numbers identify the amount of time the CPU has spent performing
1462 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1463 second). The meanings of the columns are as follows, from left to right:
1465 - user: normal processes executing in user mode
1466 - nice: niced processes executing in user mode
1467 - system: processes executing in kernel mode
1468 - idle: twiddling thumbs
1469 - iowait: In a word, iowait stands for waiting for I/O to complete. But there
1470 are several problems:
1472 1. Cpu will not wait for I/O to complete, iowait is the time that a task is
1473 waiting for I/O to complete. When cpu goes into idle state for
1474 outstanding task io, another task will be scheduled on this CPU.
1475 2. In a multi-core CPU, the task waiting for I/O to complete is not running
1476 on any CPU, so the iowait of each CPU is difficult to calculate.
1477 3. The value of iowait field in /proc/stat will decrease in certain
1480 So, the iowait is not reliable by reading from /proc/stat.
1481 - irq: servicing interrupts
1482 - softirq: servicing softirqs
1483 - steal: involuntary wait
1484 - guest: running a normal guest
1485 - guest_nice: running a niced guest
1487 The "intr" line gives counts of interrupts serviced since boot time, for each
1488 of the possible system interrupts. The first column is the total of all
1489 interrupts serviced including unnumbered architecture specific interrupts;
1490 each subsequent column is the total for that particular numbered interrupt.
1491 Unnumbered interrupts are not shown, only summed into the total.
1493 The "ctxt" line gives the total number of context switches across all CPUs.
1495 The "btime" line gives the time at which the system booted, in seconds since
1498 The "processes" line gives the number of processes and threads created, which
1499 includes (but is not limited to) those created by calls to the fork() and
1500 clone() system calls.
1502 The "procs_running" line gives the total number of threads that are
1503 running or ready to run (i.e., the total number of runnable threads).
1505 The "procs_blocked" line gives the number of processes currently blocked,
1506 waiting for I/O to complete.
1508 The "softirq" line gives counts of softirqs serviced since boot time, for each
1509 of the possible system softirqs. The first column is the total of all
1510 softirqs serviced; each subsequent column is the total for that particular
1514 1.9 Ext4 file system parameters
1515 -------------------------------
1517 Information about mounted ext4 file systems can be found in
1518 /proc/fs/ext4. Each mounted filesystem will have a directory in
1519 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1520 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1521 in Table 1-12, below.
1523 .. table:: Table 1-12: Files in /proc/fs/ext4/<devname>
1525 ============== ==========================================================
1527 mb_groups details of multiblock allocator buddy cache of free blocks
1528 ============== ==========================================================
1532 Shows registered system console lines.
1534 To see which character device lines are currently used for the system console
1535 /dev/console, you may simply look into the file /proc/consoles::
1537 > cat /proc/consoles
1543 +--------------------+-------------------------------------------------------+
1544 | device | name of the device |
1545 +====================+=======================================================+
1546 | operations | * R = can do read operations |
1547 | | * W = can do write operations |
1548 | | * U = can do unblank |
1549 +--------------------+-------------------------------------------------------+
1550 | flags | * E = it is enabled |
1551 | | * C = it is preferred console |
1552 | | * B = it is primary boot console |
1553 | | * p = it is used for printk buffer |
1554 | | * b = it is not a TTY but a Braille device |
1555 | | * a = it is safe to use when cpu is offline |
1556 +--------------------+-------------------------------------------------------+
1557 | major:minor | major and minor number of the device separated by a |
1559 +--------------------+-------------------------------------------------------+
1564 The /proc file system serves information about the running system. It not only
1565 allows access to process data but also allows you to request the kernel status
1566 by reading files in the hierarchy.
1568 The directory structure of /proc reflects the types of information and makes
1569 it easy, if not obvious, where to look for specific data.
1571 Chapter 2: Modifying System Parameters
1572 ======================================
1577 * Modifying kernel parameters by writing into files found in /proc/sys
1578 * Exploring the files which modify certain parameters
1579 * Review of the /proc/sys file tree
1581 ------------------------------------------------------------------------------
1583 A very interesting part of /proc is the directory /proc/sys. This is not only
1584 a source of information, it also allows you to change parameters within the
1585 kernel. Be very careful when attempting this. You can optimize your system,
1586 but you can also cause it to crash. Never alter kernel parameters on a
1587 production system. Set up a development machine and test to make sure that
1588 everything works the way you want it to. You may have no alternative but to
1589 reboot the machine once an error has been made.
1591 To change a value, simply echo the new value into the file. An example is
1592 given below in the section on the file system data. You need to be root to do
1593 this. You can create your own boot script to perform this every time your
1596 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1597 general things in the operation of the Linux kernel. Since some of the files
1598 can inadvertently disrupt your system, it is advisable to read both
1599 documentation and source before actually making adjustments. In any case, be
1600 very careful when writing to any of these files. The entries in /proc may
1601 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1602 review the kernel documentation in the directory /usr/src/linux/Documentation.
1603 This chapter is heavily based on the documentation included in the pre 2.2
1604 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1606 Please see: Documentation/admin-guide/sysctl/ directory for descriptions of these
1612 Certain aspects of kernel behavior can be modified at runtime, without the
1613 need to recompile the kernel, or even to reboot the system. The files in the
1614 /proc/sys tree can not only be read, but also modified. You can use the echo
1615 command to write value into these files, thereby changing the default settings
1619 Chapter 3: Per-process Parameters
1620 =================================
1622 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1623 --------------------------------------------------------------------------------
1625 These file can be used to adjust the badness heuristic used to select which
1626 process gets killed in out of memory conditions.
1628 The badness heuristic assigns a value to each candidate task ranging from 0
1629 (never kill) to 1000 (always kill) to determine which process is targeted. The
1630 units are roughly a proportion along that range of allowed memory the process
1631 may allocate from based on an estimation of its current memory and swap use.
1632 For example, if a task is using all allowed memory, its badness score will be
1633 1000. If it is using half of its allowed memory, its score will be 500.
1635 There is an additional factor included in the badness score: the current memory
1636 and swap usage is discounted by 3% for root processes.
1638 The amount of "allowed" memory depends on the context in which the oom killer
1639 was called. If it is due to the memory assigned to the allocating task's cpuset
1640 being exhausted, the allowed memory represents the set of mems assigned to that
1641 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1642 memory represents the set of mempolicy nodes. If it is due to a memory
1643 limit (or swap limit) being reached, the allowed memory is that configured
1644 limit. Finally, if it is due to the entire system being out of memory, the
1645 allowed memory represents all allocatable resources.
1647 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1648 is used to determine which task to kill. Acceptable values range from -1000
1649 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1650 polarize the preference for oom killing either by always preferring a certain
1651 task or completely disabling it. The lowest possible value, -1000, is
1652 equivalent to disabling oom killing entirely for that task since it will always
1653 report a badness score of 0.
1655 Consequently, it is very simple for userspace to define the amount of memory to
1656 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1657 example, is roughly equivalent to allowing the remainder of tasks sharing the
1658 same system, cpuset, mempolicy, or memory controller resources to use at least
1659 50% more memory. A value of -500, on the other hand, would be roughly
1660 equivalent to discounting 50% of the task's allowed memory from being considered
1661 as scoring against the task.
1663 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1664 be used to tune the badness score. Its acceptable values range from -16
1665 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1666 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1667 scaled linearly with /proc/<pid>/oom_score_adj.
1669 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1670 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1671 requires CAP_SYS_RESOURCE.
1673 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1674 generation children with separate address spaces instead, if possible. This
1675 avoids servers and important system daemons from being killed and loses the
1676 minimal amount of work.
1679 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1680 -------------------------------------------------------------
1682 This file can be used to check the current score used by the oom-killer is for
1683 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1684 process should be killed in an out-of-memory situation.
1687 3.3 /proc/<pid>/io - Display the IO accounting fields
1688 -------------------------------------------------------
1690 This file contains IO statistics for each running process
1697 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1700 test:/tmp # cat /proc/3828/io
1706 write_bytes: 323932160
1707 cancelled_write_bytes: 0
1716 I/O counter: chars read
1717 The number of bytes which this task has caused to be read from storage. This
1718 is simply the sum of bytes which this process passed to read() and pread().
1719 It includes things like tty IO and it is unaffected by whether or not actual
1720 physical disk IO was required (the read might have been satisfied from
1727 I/O counter: chars written
1728 The number of bytes which this task has caused, or shall cause to be written
1729 to disk. Similar caveats apply here as with rchar.
1735 I/O counter: read syscalls
1736 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1743 I/O counter: write syscalls
1744 Attempt to count the number of write I/O operations, i.e. syscalls like
1745 write() and pwrite().
1751 I/O counter: bytes read
1752 Attempt to count the number of bytes which this process really did cause to
1753 be fetched from the storage layer. Done at the submit_bio() level, so it is
1754 accurate for block-backed filesystems. <please add status regarding NFS and
1755 CIFS at a later time>
1761 I/O counter: bytes written
1762 Attempt to count the number of bytes which this process caused to be sent to
1763 the storage layer. This is done at page-dirtying time.
1766 cancelled_write_bytes
1767 ^^^^^^^^^^^^^^^^^^^^^
1769 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1770 then deletes the file, it will in fact perform no writeout. But it will have
1771 been accounted as having caused 1MB of write.
1772 In other words: The number of bytes which this process caused to not happen,
1773 by truncating pagecache. A task can cause "negative" IO too. If this task
1774 truncates some dirty pagecache, some IO which another task has been accounted
1775 for (in its write_bytes) will not be happening. We _could_ just subtract that
1776 from the truncating task's write_bytes, but there is information loss in doing
1782 At its current implementation state, this is a bit racy on 32-bit machines:
1783 if process A reads process B's /proc/pid/io while process B is updating one
1784 of those 64-bit counters, process A could see an intermediate result.
1787 More information about this can be found within the taskstats documentation in
1788 Documentation/accounting.
1790 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1791 ---------------------------------------------------------------
1792 When a process is dumped, all anonymous memory is written to a core file as
1793 long as the size of the core file isn't limited. But sometimes we don't want
1794 to dump some memory segments, for example, huge shared memory or DAX.
1795 Conversely, sometimes we want to save file-backed memory segments into a core
1796 file, not only the individual files.
1798 /proc/<pid>/coredump_filter allows you to customize which memory segments
1799 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1800 of memory types. If a bit of the bitmask is set, memory segments of the
1801 corresponding memory type are dumped, otherwise they are not dumped.
1803 The following 9 memory types are supported:
1805 - (bit 0) anonymous private memory
1806 - (bit 1) anonymous shared memory
1807 - (bit 2) file-backed private memory
1808 - (bit 3) file-backed shared memory
1809 - (bit 4) ELF header pages in file-backed private memory areas (it is
1810 effective only if the bit 2 is cleared)
1811 - (bit 5) hugetlb private memory
1812 - (bit 6) hugetlb shared memory
1813 - (bit 7) DAX private memory
1814 - (bit 8) DAX shared memory
1816 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1817 are always dumped regardless of the bitmask status.
1819 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1820 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1822 The default value of coredump_filter is 0x33; this means all anonymous memory
1823 segments, ELF header pages and hugetlb private memory are dumped.
1825 If you don't want to dump all shared memory segments attached to pid 1234,
1826 write 0x31 to the process's proc file::
1828 $ echo 0x31 > /proc/1234/coredump_filter
1830 When a new process is created, the process inherits the bitmask status from its
1831 parent. It is useful to set up coredump_filter before the program runs.
1834 $ echo 0x7 > /proc/self/coredump_filter
1837 3.5 /proc/<pid>/mountinfo - Information about mounts
1838 --------------------------------------------------------
1840 This file contains lines of the form::
1842 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1843 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1845 (1) mount ID: unique identifier of the mount (may be reused after umount)
1846 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1847 (3) major:minor: value of st_dev for files on filesystem
1848 (4) root: root of the mount within the filesystem
1849 (5) mount point: mount point relative to the process's root
1850 (6) mount options: per mount options
1851 (7) optional fields: zero or more fields of the form "tag[:value]"
1852 (8) separator: marks the end of the optional fields
1853 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1854 (10) mount source: filesystem specific information or "none"
1855 (11) super options: per super block options
1857 Parsers should ignore all unrecognised optional fields. Currently the
1858 possible optional fields are:
1860 ================ ==============================================================
1861 shared:X mount is shared in peer group X
1862 master:X mount is slave to peer group X
1863 propagate_from:X mount is slave and receives propagation from peer group X [#]_
1864 unbindable mount is unbindable
1865 ================ ==============================================================
1867 .. [#] X is the closest dominant peer group under the process's root. If
1868 X is the immediate master of the mount, or if there's no dominant peer
1869 group under the same root, then only the "master:X" field is present
1870 and not the "propagate_from:X" field.
1872 For more information on mount propagation see:
1874 Documentation/filesystems/sharedsubtree.rst
1877 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1878 --------------------------------------------------------
1879 These files provide a method to access a tasks comm value. It also allows for
1880 a task to set its own or one of its thread siblings comm value. The comm value
1881 is limited in size compared to the cmdline value, so writing anything longer
1882 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1886 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1887 -------------------------------------------------------------------------
1888 This file provides a fast way to retrieve first level children pids
1889 of a task pointed by <pid>/<tid> pair. The format is a space separated
1892 Note the "first level" here -- if a child has own children they will
1893 not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1894 to obtain the descendants.
1896 Since this interface is intended to be fast and cheap it doesn't
1897 guarantee to provide precise results and some children might be
1898 skipped, especially if they've exited right after we printed their
1899 pids, so one need to either stop or freeze processes being inspected
1900 if precise results are needed.
1903 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1904 ---------------------------------------------------------------
1905 This file provides information associated with an opened file. The regular
1906 files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos'
1907 represents the current offset of the opened file in decimal form [see lseek(2)
1908 for details], 'flags' denotes the octal O_xxx mask the file has been
1909 created with [see open(2) for details] and 'mnt_id' represents mount ID of
1910 the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo
1913 A typical output is::
1919 All locks associated with a file descriptor are shown in its fdinfo too::
1921 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1923 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1924 pair provide additional information particular to the objects they represent.
1936 where 'eventfd-count' is hex value of a counter.
1946 sigmask: 0000000000000200
1948 where 'sigmask' is hex value of the signal mask associated
1959 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7
1961 where 'tfd' is a target file descriptor number in decimal form,
1962 'events' is events mask being watched and the 'data' is data
1963 associated with a target [see epoll(7) for more details].
1965 The 'pos' is current offset of the target file in decimal form
1966 [see lseek(2)], 'ino' and 'sdev' are inode and device numbers
1967 where target file resides, all in hex format.
1971 For inotify files the format is the following::
1975 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1977 where 'wd' is a watch descriptor in decimal form, ie a target file
1978 descriptor number, 'ino' and 'sdev' are inode and device where the
1979 target file resides and the 'mask' is the mask of events, all in hex
1980 form [see inotify(7) for more details].
1982 If the kernel was built with exportfs support, the path to the target
1983 file is encoded as a file handle. The file handle is provided by three
1984 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1987 If the kernel is built without exportfs support the file handle won't be
1990 If there is no inotify mark attached yet the 'inotify' line will be omitted.
1992 For fanotify files the format is::
1997 fanotify flags:10 event-flags:0
1998 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1999 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
2001 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
2002 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
2003 flags associated with mark which are tracked separately from events
2004 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
2005 mask and 'ignored_mask' is the mask of events which are to be ignored.
2006 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
2007 does provide information about flags and mask used in fanotify_mark
2008 call [see fsnotify manpage for details].
2010 While the first three lines are mandatory and always printed, the rest is
2011 optional and may be omitted if no marks created yet.
2024 it_value: (0, 49406829)
2027 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
2028 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
2029 flags in octal form been used to setup the timer [see timerfd_settime(2) for
2030 details]. 'it_value' is remaining time until the timer exiration.
2031 'it_interval' is the interval for the timer. Note the timer might be set up
2032 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
2033 still exhibits timer's remaining time.
2035 3.9 /proc/<pid>/map_files - Information about memory mapped files
2036 ---------------------------------------------------------------------
2037 This directory contains symbolic links which represent memory mapped files
2038 the process is maintaining. Example output::
2040 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
2041 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
2042 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
2044 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
2045 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
2047 The name of a link represents the virtual memory bounds of a mapping, i.e.
2048 vm_area_struct::vm_start-vm_area_struct::vm_end.
2050 The main purpose of the map_files is to retrieve a set of memory mapped
2051 files in a fast way instead of parsing /proc/<pid>/maps or
2052 /proc/<pid>/smaps, both of which contain many more records. At the same
2053 time one can open(2) mappings from the listings of two processes and
2054 comparing their inode numbers to figure out which anonymous memory areas
2055 are actually shared.
2057 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
2058 ---------------------------------------------------------
2059 This file provides the value of the task's timerslack value in nanoseconds.
2060 This value specifies a amount of time that normal timers may be deferred
2061 in order to coalesce timers and avoid unnecessary wakeups.
2063 This allows a task's interactivity vs power consumption trade off to be
2066 Writing 0 to the file will set the tasks timerslack to the default value.
2068 Valid values are from 0 - ULLONG_MAX
2070 An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
2071 permissions on the task specified to change its timerslack_ns value.
2073 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
2074 -----------------------------------------------------------------
2075 When CONFIG_LIVEPATCH is enabled, this file displays the value of the
2076 patch state for the task.
2078 A value of '-1' indicates that no patch is in transition.
2080 A value of '0' indicates that a patch is in transition and the task is
2081 unpatched. If the patch is being enabled, then the task hasn't been
2082 patched yet. If the patch is being disabled, then the task has already
2085 A value of '1' indicates that a patch is in transition and the task is
2086 patched. If the patch is being enabled, then the task has already been
2087 patched. If the patch is being disabled, then the task hasn't been
2090 3.12 /proc/<pid>/arch_status - task architecture specific status
2091 -------------------------------------------------------------------
2092 When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the
2093 architecture specific status of the task.
2100 $ cat /proc/6753/arch_status
2101 AVX512_elapsed_ms: 8
2106 x86 specific entries:
2107 ~~~~~~~~~~~~~~~~~~~~~
2112 If AVX512 is supported on the machine, this entry shows the milliseconds
2113 elapsed since the last time AVX512 usage was recorded. The recording
2114 happens on a best effort basis when a task is scheduled out. This means
2115 that the value depends on two factors:
2117 1) The time which the task spent on the CPU without being scheduled
2118 out. With CPU isolation and a single runnable task this can take
2121 2) The time since the task was scheduled out last. Depending on the
2122 reason for being scheduled out (time slice exhausted, syscall ...)
2123 this can be arbitrary long time.
2125 As a consequence the value cannot be considered precise and authoritative
2126 information. The application which uses this information has to be aware
2127 of the overall scenario on the system in order to determine whether a
2128 task is a real AVX512 user or not. Precise information can be obtained
2129 with performance counters.
2131 A special value of '-1' indicates that no AVX512 usage was recorded, thus
2132 the task is unlikely an AVX512 user, but depends on the workload and the
2133 scheduling scenario, it also could be a false negative mentioned above.
2139 ---------------------
2141 The following mount options are supported:
2143 ========= ========================================================
2144 hidepid= Set /proc/<pid>/ access mode.
2145 gid= Set the group authorized to learn processes information.
2146 ========= ========================================================
2148 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
2151 hidepid=1 means users may not access any /proc/<pid>/ directories but their
2152 own. Sensitive files like cmdline, sched*, status are now protected against
2153 other users. This makes it impossible to learn whether any user runs
2154 specific program (given the program doesn't reveal itself by its behaviour).
2155 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
2156 poorly written programs passing sensitive information via program arguments are
2157 now protected against local eavesdroppers.
2159 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
2160 users. It doesn't mean that it hides a fact whether a process with a specific
2161 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
2162 but it hides process' uid and gid, which may be learned by stat()'ing
2163 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
2164 information about running processes, whether some daemon runs with elevated
2165 privileges, whether other user runs some sensitive program, whether other users
2166 run any program at all, etc.
2168 gid= defines a group authorized to learn processes information otherwise
2169 prohibited by hidepid=. If you use some daemon like identd which needs to learn
2170 information about processes information, just add identd to this group.