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
59 0.1 Introduction/Credits
60 ------------------------
62 This documentation is part of a soon (or so we hope) to be released book on
63 the SuSE Linux distribution. As there is no complete documentation for the
64 /proc file system and we've used many freely available sources to write these
65 chapters, it seems only fair to give the work back to the Linux community.
66 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
67 afraid it's still far from complete, but we hope it will be useful. As far as
68 we know, it is the first 'all-in-one' document about the /proc file system. It
69 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
70 SPARC, AXP, etc., features, you probably won't find what you are looking for.
71 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
72 additions and patches are welcome and will be added to this document if you
75 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
76 other people for help compiling this documentation. We'd also like to extend a
77 special thank you to Andi Kleen for documentation, which we relied on heavily
78 to create this document, as well as the additional information he provided.
79 Thanks to everybody else who contributed source or docs to the Linux kernel
80 and helped create a great piece of software... :)
82 If you have any comments, corrections or additions, please don't hesitate to
83 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
86 The latest version of this document is available online at
87 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
89 If the above direction does not works for you, you could try the kernel
90 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
91 comandante@zaralinux.com.
96 We don't guarantee the correctness of this document, and if you come to us
97 complaining about how you screwed up your system because of incorrect
98 documentation, we won't feel responsible...
100 Chapter 1: Collecting System Information
101 ========================================
105 * Investigating the properties of the pseudo file system /proc and its
106 ability to provide information on the running Linux system
107 * Examining /proc's structure
108 * Uncovering various information about the kernel and the processes running
111 ------------------------------------------------------------------------------
113 The proc file system acts as an interface to internal data structures in the
114 kernel. It can be used to obtain information about the system and to change
115 certain kernel parameters at runtime (sysctl).
117 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
118 show you how you can use /proc/sys to change settings.
120 1.1 Process-Specific Subdirectories
121 -----------------------------------
123 The directory /proc contains (among other things) one subdirectory for each
124 process running on the system, which is named after the process ID (PID).
126 The link 'self' points to the process reading the file system. Each process
127 subdirectory has the entries listed in Table 1-1.
129 Note that an open file descriptor to /proc/<pid> or to any of its
130 contained files or subdirectories does not prevent <pid> being reused
131 for some other process in the event that <pid> exits. Operations on
132 open /proc/<pid> file descriptors corresponding to dead processes
133 never act on any new process that the kernel may, through chance, have
134 also assigned the process ID <pid>. Instead, operations on these FDs
135 usually fail with ESRCH.
137 .. table:: Table 1-1: Process specific entries in /proc
139 ============= ===============================================================
141 ============= ===============================================================
142 clear_refs Clears page referenced bits shown in smaps output
143 cmdline Command line arguments
144 cpu Current and last cpu in which it was executed (2.4)(smp)
145 cwd Link to the current working directory
146 environ Values of environment variables
147 exe Link to the executable of this process
148 fd Directory, which contains all file descriptors
149 maps Memory maps to executables and library files (2.4)
150 mem Memory held by this process
151 root Link to the root directory of this process
153 statm Process memory status information
154 status Process status in human readable form
155 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function
156 symbol the task is blocked in - or "0" if not blocked.
158 stack Report full stack trace, enable via CONFIG_STACKTRACE
159 smaps An extension based on maps, showing the memory consumption of
160 each mapping and flags associated with it
161 smaps_rollup Accumulated smaps stats for all mappings of the process. This
162 can be derived from smaps, but is faster and more convenient
163 numa_maps An extension based on maps, showing the memory locality and
164 binding policy as well as mem usage (in pages) of each mapping.
165 ============= ===============================================================
167 For example, to get the status information of a process, all you have to do is
168 read the file /proc/PID/status::
170 >cat /proc/self/status
200 SigPnd: 0000000000000000
201 ShdPnd: 0000000000000000
202 SigBlk: 0000000000000000
203 SigIgn: 0000000000000000
204 SigCgt: 0000000000000000
205 CapInh: 00000000fffffeff
206 CapPrm: 0000000000000000
207 CapEff: 0000000000000000
208 CapBnd: ffffffffffffffff
209 CapAmb: 0000000000000000
212 Speculation_Store_Bypass: thread vulnerable
213 SpeculationIndirectBranch: conditional enabled
214 voluntary_ctxt_switches: 0
215 nonvoluntary_ctxt_switches: 1
217 This shows you nearly the same information you would get if you viewed it with
218 the ps command. In fact, ps uses the proc file system to obtain its
219 information. But you get a more detailed view of the process by reading the
220 file /proc/PID/status. It fields are described in table 1-2.
222 The statm file contains more detailed information about the process
223 memory usage. Its seven fields are explained in Table 1-3. The stat file
224 contains detailed information about the process itself. Its fields are
225 explained in Table 1-4.
227 (for SMP CONFIG users)
229 For making accounting scalable, RSS related information are handled in an
230 asynchronous manner and the value may not be very precise. To see a precise
231 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
232 It's slow but very precise.
234 .. table:: Table 1-2: Contents of the status files (as of 4.19)
236 ========================== ===================================================
238 ========================== ===================================================
239 Name filename of the executable
240 Umask file mode creation mask
241 State state (R is running, S is sleeping, D is sleeping
242 in an uninterruptible wait, Z is zombie,
243 T is traced or stopped)
245 Ngid NUMA group ID (0 if none)
247 PPid process id of the parent process
248 TracerPid PID of process tracing this process (0 if not)
249 Uid Real, effective, saved set, and file system UIDs
250 Gid Real, effective, saved set, and file system GIDs
251 FDSize number of file descriptor slots currently allocated
252 Groups supplementary group list
253 NStgid descendant namespace thread group ID hierarchy
254 NSpid descendant namespace process ID hierarchy
255 NSpgid descendant namespace process group ID hierarchy
256 NSsid descendant namespace session ID hierarchy
257 VmPeak peak virtual memory size
258 VmSize total program size
259 VmLck locked memory size
260 VmPin pinned memory size
261 VmHWM peak resident set size ("high water mark")
262 VmRSS size of memory portions. It contains the three
264 (VmRSS = RssAnon + RssFile + RssShmem)
265 RssAnon size of resident anonymous memory
266 RssFile size of resident file mappings
267 RssShmem size of resident shmem memory (includes SysV shm,
268 mapping of tmpfs and shared anonymous mappings)
269 VmData size of private data segments
270 VmStk size of stack segments
271 VmExe size of text segment
272 VmLib size of shared library code
273 VmPTE size of page table entries
274 VmSwap amount of swap used by anonymous private data
275 (shmem swap usage is not included)
276 HugetlbPages size of hugetlb memory portions
277 CoreDumping process's memory is currently being dumped
278 (killing the process may lead to a corrupted core)
279 THP_enabled process is allowed to use THP (returns 0 when
280 PR_SET_THP_DISABLE is set on the process
281 Threads number of threads
282 SigQ number of signals queued/max. number for queue
283 SigPnd bitmap of pending signals for the thread
284 ShdPnd bitmap of shared pending signals for the process
285 SigBlk bitmap of blocked signals
286 SigIgn bitmap of ignored signals
287 SigCgt bitmap of caught signals
288 CapInh bitmap of inheritable capabilities
289 CapPrm bitmap of permitted capabilities
290 CapEff bitmap of effective capabilities
291 CapBnd bitmap of capabilities bounding set
292 CapAmb bitmap of ambient capabilities
293 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
294 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
295 Speculation_Store_Bypass speculative store bypass mitigation status
296 SpeculationIndirectBranch indirect branch speculation mode
297 Cpus_allowed mask of CPUs on which this process may run
298 Cpus_allowed_list Same as previous, but in "list format"
299 Mems_allowed mask of memory nodes allowed to this process
300 Mems_allowed_list Same as previous, but in "list format"
301 voluntary_ctxt_switches number of voluntary context switches
302 nonvoluntary_ctxt_switches number of non voluntary context switches
303 ========================== ===================================================
306 .. table:: Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
308 ======== =============================== ==============================
310 ======== =============================== ==============================
311 size total program size (pages) (same as VmSize in status)
312 resident size of memory portions (pages) (same as VmRSS in status)
313 shared number of pages that are shared (i.e. backed by a file, same
314 as RssFile+RssShmem in status)
315 trs number of pages that are 'code' (not including libs; broken,
316 includes data segment)
317 lrs number of pages of library (always 0 on 2.6)
318 drs number of pages of data/stack (including libs; broken,
319 includes library text)
320 dt number of dirty pages (always 0 on 2.6)
321 ======== =============================== ==============================
324 .. table:: Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
326 ============= ===============================================================
328 ============= ===============================================================
330 tcomm filename of the executable
331 state state (R is running, S is sleeping, D is sleeping in an
332 uninterruptible wait, Z is zombie, T is traced or stopped)
333 ppid process id of the parent process
334 pgrp pgrp of the process
336 tty_nr tty the process uses
337 tty_pgrp pgrp of the tty
339 min_flt number of minor faults
340 cmin_flt number of minor faults with child's
341 maj_flt number of major faults
342 cmaj_flt number of major faults with child's
343 utime user mode jiffies
344 stime kernel mode jiffies
345 cutime user mode jiffies with child's
346 cstime kernel mode jiffies with child's
347 priority priority level
349 num_threads number of threads
350 it_real_value (obsolete, always 0)
351 start_time time the process started after system boot
352 vsize virtual memory size
353 rss resident set memory size
354 rsslim current limit in bytes on the rss
355 start_code address above which program text can run
356 end_code address below which program text can run
357 start_stack address of the start of the main process stack
358 esp current value of ESP
359 eip current value of EIP
360 pending bitmap of pending signals
361 blocked bitmap of blocked signals
362 sigign bitmap of ignored signals
363 sigcatch bitmap of caught signals
364 0 (place holder, used to be the wchan address,
365 use /proc/PID/wchan instead)
368 exit_signal signal to send to parent thread on exit
369 task_cpu which CPU the task is scheduled on
370 rt_priority realtime priority
371 policy scheduling policy (man sched_setscheduler)
372 blkio_ticks time spent waiting for block IO
373 gtime guest time of the task in jiffies
374 cgtime guest time of the task children in jiffies
375 start_data address above which program data+bss is placed
376 end_data address below which program data+bss is placed
377 start_brk address above which program heap can be expanded with brk()
378 arg_start address above which program command line is placed
379 arg_end address below which program command line is placed
380 env_start address above which program environment is placed
381 env_end address below which program environment is placed
382 exit_code the thread's exit_code in the form reported by the waitpid
384 ============= ===============================================================
386 The /proc/PID/maps file contains the currently mapped memory regions and
387 their access permissions.
391 address perms offset dev inode pathname
393 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
394 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
395 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
396 a7cb1000-a7cb2000 ---p 00000000 00:00 0
397 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
398 a7eb2000-a7eb3000 ---p 00000000 00:00 0
399 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
400 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
401 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
402 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
403 a800b000-a800e000 rw-p 00000000 00:00 0
404 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
405 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
406 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
407 a8024000-a8027000 rw-p 00000000 00:00 0
408 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
409 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
410 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
411 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
412 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
414 where "address" is the address space in the process that it occupies, "perms"
415 is a set of permissions::
421 p = private (copy on write)
423 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
424 "inode" is the inode on that device. 0 indicates that no inode is associated
425 with the memory region, as the case would be with BSS (uninitialized data).
426 The "pathname" shows the name associated file for this mapping. If the mapping
427 is not associated with a file:
429 ============= ====================================
430 [heap] the heap of the program
431 [stack] the stack of the main process
432 [vdso] the "virtual dynamic shared object",
433 the kernel system call handler
434 [anon:<name>] an anonymous mapping that has been
436 ============= ====================================
438 or if empty, the mapping is anonymous.
440 The /proc/PID/smaps is an extension based on maps, showing the memory
441 consumption for each of the process's mappings. For each mapping (aka Virtual
442 Memory Area, or VMA) there is a series of lines such as the following::
444 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
461 Private_Hugetlb: 0 kB
468 VmFlags: rd ex mr mw me dw
470 The first of these lines shows the same information as is displayed for the
471 mapping in /proc/PID/maps. Following lines show the size of the mapping
472 (size); the size of each page allocated when backing a VMA (KernelPageSize),
473 which is usually the same as the size in the page table entries; the page size
474 used by the MMU when backing a VMA (in most cases, the same as KernelPageSize);
475 the amount of the mapping that is currently resident in RAM (RSS); the
476 process' proportional share of this mapping (PSS); and the number of clean and
477 dirty shared and private pages in the mapping.
479 The "proportional set size" (PSS) of a process is the count of pages it has
480 in memory, where each page is divided by the number of processes sharing it.
481 So if a process has 1000 pages all to itself, and 1000 shared with one other
482 process, its PSS will be 1500.
484 Note that even a page which is part of a MAP_SHARED mapping, but has only
485 a single pte mapped, i.e. is currently used by only one process, is accounted
486 as private and not as shared.
488 "Referenced" indicates the amount of memory currently marked as referenced or
491 "Anonymous" shows the amount of memory that does not belong to any file. Even
492 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
493 and a page is modified, the file page is replaced by a private anonymous copy.
495 "LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
496 The memory isn't freed immediately with madvise(). It's freed in memory
497 pressure if the memory is clean. Please note that the printed value might
498 be lower than the real value due to optimizations used in the current
499 implementation. If this is not desirable please file a bug report.
501 "AnonHugePages" shows the ammount of memory backed by transparent hugepage.
503 "ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by
506 "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
507 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
508 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
510 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
512 For shmem mappings, "Swap" includes also the size of the mapped (and not
513 replaced by copy-on-write) part of the underlying shmem object out on swap.
514 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
515 does not take into account swapped out page of underlying shmem objects.
516 "Locked" indicates whether the mapping is locked in memory or not.
517 "THPeligible" indicates whether the mapping is eligible for allocating THP
518 pages - 1 if true, 0 otherwise. It just shows the current status.
520 "VmFlags" field deserves a separate description. This member represents the
521 kernel flags associated with the particular virtual memory area in two letter
522 encoded manner. The codes are the following:
524 == =======================================
533 gd stack segment growns down
535 dw disabled write to the mapped file
536 lo pages are locked in memory
537 io memory mapped I/O area
538 sr sequential read advise provided
539 rr random read advise provided
540 dc do not copy area on fork
541 de do not expand area on remapping
542 ac area is accountable
543 nr swap space is not reserved for the area
544 ht area uses huge tlb pages
545 sf synchronous page fault
546 ar architecture specific flag
548 dd do not include area into core dump
551 hg huge page advise flag
552 nh no huge page advise flag
553 mg mergable advise flag
554 bt arm64 BTI guarded page
555 mt arm64 MTE allocation tags are enabled
556 um userfaultfd missing tracking
557 uw userfaultfd wr-protect tracking
558 == =======================================
560 Note that there is no guarantee that every flag and associated mnemonic will
561 be present in all further kernel releases. Things get changed, the flags may
562 be vanished or the reverse -- new added. Interpretation of their meaning
563 might change in future as well. So each consumer of these flags has to
564 follow each specific kernel version for the exact semantic.
566 This file is only present if the CONFIG_MMU kernel configuration option is
569 Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
570 output can be achieved only in the single read call).
572 This typically manifests when doing partial reads of these files while the
573 memory map is being modified. Despite the races, we do provide the following
576 1) The mapped addresses never go backwards, which implies no two
577 regions will ever overlap.
578 2) If there is something at a given vaddr during the entirety of the
579 life of the smaps/maps walk, there will be some output for it.
581 The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps,
582 but their values are the sums of the corresponding values for all mappings of
583 the process. Additionally, it contains these fields:
589 They represent the proportional shares of anonymous, file, and shmem pages, as
590 described for smaps above. These fields are omitted in smaps since each
591 mapping identifies the type (anon, file, or shmem) of all pages it contains.
592 Thus all information in smaps_rollup can be derived from smaps, but at a
593 significantly higher cost.
595 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
596 bits on both physical and virtual pages associated with a process, and the
597 soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst
599 To clear the bits for all the pages associated with the process::
601 > echo 1 > /proc/PID/clear_refs
603 To clear the bits for the anonymous pages associated with the process::
605 > echo 2 > /proc/PID/clear_refs
607 To clear the bits for the file mapped pages associated with the process::
609 > echo 3 > /proc/PID/clear_refs
611 To clear the soft-dirty bit::
613 > echo 4 > /proc/PID/clear_refs
615 To reset the peak resident set size ("high water mark") to the process's
618 > echo 5 > /proc/PID/clear_refs
620 Any other value written to /proc/PID/clear_refs will have no effect.
622 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
623 using /proc/kpageflags and number of times a page is mapped using
624 /proc/kpagecount. For detailed explanation, see
625 Documentation/admin-guide/mm/pagemap.rst.
627 The /proc/pid/numa_maps is an extension based on maps, showing the memory
628 locality and binding policy, as well as the memory usage (in pages) of
629 each mapping. The output follows a general format where mapping details get
630 summarized separated by blank spaces, one mapping per each file line::
632 address policy mapping details
634 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
635 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
636 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
637 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
638 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
639 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
640 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
641 320698b000 default file=/lib64/libc-2.12.so
642 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
643 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
644 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
645 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
646 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
647 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
648 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
649 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
653 "address" is the starting address for the mapping;
655 "policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst);
657 "mapping details" summarizes mapping data such as mapping type, page usage counters,
658 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
659 size, in KB, that is backing the mapping up.
664 Similar to the process entries, the kernel data files give information about
665 the running kernel. The files used to obtain this information are contained in
666 /proc and are listed in Table 1-5. Not all of these will be present in your
667 system. It depends on the kernel configuration and the loaded modules, which
668 files are there, and which are missing.
670 .. table:: Table 1-5: Kernel info in /proc
672 ============ ===============================================================
674 ============ ===============================================================
675 apm Advanced power management info
676 buddyinfo Kernel memory allocator information (see text) (2.5)
677 bus Directory containing bus specific information
678 cmdline Kernel command line
679 cpuinfo Info about the CPU
680 devices Available devices (block and character)
681 dma Used DMS channels
682 filesystems Supported filesystems
683 driver Various drivers grouped here, currently rtc (2.4)
684 execdomains Execdomains, related to security (2.4)
685 fb Frame Buffer devices (2.4)
686 fs File system parameters, currently nfs/exports (2.4)
687 ide Directory containing info about the IDE subsystem
688 interrupts Interrupt usage
689 iomem Memory map (2.4)
690 ioports I/O port usage
691 irq Masks for irq to cpu affinity (2.4)(smp?)
692 isapnp ISA PnP (Plug&Play) Info (2.4)
693 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
695 ksyms Kernel symbol table
696 loadavg Load average of last 1, 5 & 15 minutes;
697 number of processes currently runnable (running or on ready queue);
698 total number of processes in system;
700 All fields are separated by one space except "number of
701 processes currently runnable" and "total number of processes
702 in system", which are separated by a slash ('/'). Example:
703 0.61 0.61 0.55 3/828 22084
707 modules List of loaded modules
708 mounts Mounted filesystems
709 net Networking info (see text)
710 pagetypeinfo Additional page allocator information (see text) (2.5)
711 partitions Table of partitions known to the system
712 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
713 decoupled by lspci (2.4)
715 scsi SCSI info (see text)
716 slabinfo Slab pool info
717 softirqs softirq usage
718 stat Overall statistics
719 swaps Swap space utilization
721 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
722 tty Info of tty drivers
723 uptime Wall clock since boot, combined idle time of all cpus
724 version Kernel version
725 video bttv info of video resources (2.4)
726 vmallocinfo Show vmalloced areas
727 ============ ===============================================================
729 You can, for example, check which interrupts are currently in use and what
730 they are used for by looking in the file /proc/interrupts::
732 > cat /proc/interrupts
734 0: 8728810 XT-PIC timer
735 1: 895 XT-PIC keyboard
737 3: 531695 XT-PIC aha152x
738 4: 2014133 XT-PIC serial
739 5: 44401 XT-PIC pcnet_cs
742 12: 182918 XT-PIC PS/2 Mouse
744 14: 1232265 XT-PIC ide0
748 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
749 output of a SMP machine)::
751 > cat /proc/interrupts
754 0: 1243498 1214548 IO-APIC-edge timer
755 1: 8949 8958 IO-APIC-edge keyboard
756 2: 0 0 XT-PIC cascade
757 5: 11286 10161 IO-APIC-edge soundblaster
758 8: 1 0 IO-APIC-edge rtc
759 9: 27422 27407 IO-APIC-edge 3c503
760 12: 113645 113873 IO-APIC-edge PS/2 Mouse
762 14: 22491 24012 IO-APIC-edge ide0
763 15: 2183 2415 IO-APIC-edge ide1
764 17: 30564 30414 IO-APIC-level eth0
765 18: 177 164 IO-APIC-level bttv
770 NMI is incremented in this case because every timer interrupt generates a NMI
771 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
773 LOC is the local interrupt counter of the internal APIC of every CPU.
775 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
776 connects the CPUs in a SMP system. This means that an error has been detected,
777 the IO-APIC automatically retry the transmission, so it should not be a big
778 problem, but you should read the SMP-FAQ.
780 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
781 /proc/interrupts to display every IRQ vector in use by the system, not
782 just those considered 'most important'. The new vectors are:
785 interrupt raised when a machine check threshold counter
786 (typically counting ECC corrected errors of memory or cache) exceeds
787 a configurable threshold. Only available on some systems.
790 a thermal event interrupt occurs when a temperature threshold
791 has been exceeded for the CPU. This interrupt may also be generated
792 when the temperature drops back to normal.
795 a spurious interrupt is some interrupt that was raised then lowered
796 by some IO device before it could be fully processed by the APIC. Hence
797 the APIC sees the interrupt but does not know what device it came from.
798 For this case the APIC will generate the interrupt with a IRQ vector
799 of 0xff. This might also be generated by chipset bugs.
802 rescheduling, call and TLB flush interrupts are
803 sent from one CPU to another per the needs of the OS. Typically,
804 their statistics are used by kernel developers and interested users to
805 determine the occurrence of interrupts of the given type.
807 The above IRQ vectors are displayed only when relevant. For example,
808 the threshold vector does not exist on x86_64 platforms. Others are
809 suppressed when the system is a uniprocessor. As of this writing, only
810 i386 and x86_64 platforms support the new IRQ vector displays.
812 Of some interest is the introduction of the /proc/irq directory to 2.4.
813 It could be used to set IRQ to CPU affinity. This means that you can "hook" an
814 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
815 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
821 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
822 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
826 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
827 IRQ. You can set it by doing::
829 > echo 1 > /proc/irq/10/smp_affinity
831 This means that only the first CPU will handle the IRQ, but you can also echo
832 5 which means that only the first and third CPU can handle the IRQ.
834 The contents of each smp_affinity file is the same by default::
836 > cat /proc/irq/0/smp_affinity
839 There is an alternate interface, smp_affinity_list which allows specifying
840 a CPU range instead of a bitmask::
842 > cat /proc/irq/0/smp_affinity_list
845 The default_smp_affinity mask applies to all non-active IRQs, which are the
846 IRQs which have not yet been allocated/activated, and hence which lack a
847 /proc/irq/[0-9]* directory.
849 The node file on an SMP system shows the node to which the device using the IRQ
850 reports itself as being attached. This hardware locality information does not
851 include information about any possible driver locality preference.
853 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
854 profiler. Default value is ffffffff (all CPUs if there are only 32 of them).
856 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
857 between all the CPUs which are allowed to handle it. As usual the kernel has
858 more info than you and does a better job than you, so the defaults are the
859 best choice for almost everyone. [Note this applies only to those IO-APIC's
860 that support "Round Robin" interrupt distribution.]
862 There are three more important subdirectories in /proc: net, scsi, and sys.
863 The general rule is that the contents, or even the existence of these
864 directories, depend on your kernel configuration. If SCSI is not enabled, the
865 directory scsi may not exist. The same is true with the net, which is there
866 only when networking support is present in the running kernel.
868 The slabinfo file gives information about memory usage at the slab level.
869 Linux uses slab pools for memory management above page level in version 2.2.
870 Commonly used objects have their own slab pool (such as network buffers,
871 directory cache, and so on).
875 > cat /proc/buddyinfo
877 Node 0, zone DMA 0 4 5 4 4 3 ...
878 Node 0, zone Normal 1 0 0 1 101 8 ...
879 Node 0, zone HighMem 2 0 0 1 1 0 ...
881 External fragmentation is a problem under some workloads, and buddyinfo is a
882 useful tool for helping diagnose these problems. Buddyinfo will give you a
883 clue as to how big an area you can safely allocate, or why a previous
886 Each column represents the number of pages of a certain order which are
887 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
888 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
889 available in ZONE_NORMAL, etc...
891 More information relevant to external fragmentation can be found in
894 > cat /proc/pagetypeinfo
898 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
899 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
900 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
901 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
902 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
903 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
904 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
905 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
906 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
907 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
908 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
910 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
911 Node 0, zone DMA 2 0 5 1 0
912 Node 0, zone DMA32 41 6 967 2 0
914 Fragmentation avoidance in the kernel works by grouping pages of different
915 migrate types into the same contiguous regions of memory called page blocks.
916 A page block is typically the size of the default hugepage size, e.g. 2MB on
917 X86-64. By keeping pages grouped based on their ability to move, the kernel
918 can reclaim pages within a page block to satisfy a high-order allocation.
920 The pagetypinfo begins with information on the size of a page block. It
921 then gives the same type of information as buddyinfo except broken down
922 by migrate-type and finishes with details on how many page blocks of each
925 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
926 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
927 make an estimate of the likely number of huge pages that can be allocated
928 at a given point in time. All the "Movable" blocks should be allocatable
929 unless memory has been mlock()'d. Some of the Reclaimable blocks should
930 also be allocatable although a lot of filesystem metadata may have to be
931 reclaimed to achieve this.
937 Provides information about distribution and utilization of memory. This
938 varies by architecture and compile options. Some of the counters reported
939 here overlap. The memory reported by the non overlapping counters may not
940 add up to the overall memory usage and the difference for some workloads
941 can be substantial. In many cases there are other means to find out
942 additional memory using subsystem specific interfaces, for instance
943 /proc/net/sockstat for TCP memory allocations.
945 The following is from a 16GB PIII, which has highmem enabled.
946 You may not have all of these fields.
952 MemTotal: 16344972 kB
954 MemAvailable: 14836172 kB
960 HighTotal: 15597528 kB
961 HighFree: 13629632 kB
971 KReclaimable: 168048 kB
973 SReclaimable: 159856 kB
974 SUnreclaim: 124508 kB
979 CommitLimit: 7669796 kB
980 Committed_AS: 100056 kB
981 VmallocTotal: 112216 kB
983 VmallocChunk: 111088 kB
985 HardwareCorrupted: 0 kB
986 AnonHugePages: 49152 kB
991 Total usable RAM (i.e. physical RAM minus a few reserved
992 bits and the kernel binary code)
994 The sum of LowFree+HighFree
996 An estimate of how much memory is available for starting new
997 applications, without swapping. Calculated from MemFree,
998 SReclaimable, the size of the file LRU lists, and the low
999 watermarks in each zone.
1000 The estimate takes into account that the system needs some
1001 page cache to function well, and that not all reclaimable
1002 slab will be reclaimable, due to items being in use. The
1003 impact of those factors will vary from system to system.
1005 Relatively temporary storage for raw disk blocks
1006 shouldn't get tremendously large (20MB or so)
1008 in-memory cache for files read from the disk (the
1009 pagecache). Doesn't include SwapCached
1011 Memory that once was swapped out, is swapped back in but
1012 still also is in the swapfile (if memory is needed it
1013 doesn't need to be swapped out AGAIN because it is already
1014 in the swapfile. This saves I/O)
1016 Memory that has been used more recently and usually not
1017 reclaimed unless absolutely necessary.
1019 Memory which has been less recently used. It is more
1020 eligible to be reclaimed for other purposes
1022 Highmem is all memory above ~860MB of physical memory.
1023 Highmem areas are for use by userspace programs, or
1024 for the pagecache. The kernel must use tricks to access
1025 this memory, making it slower to access than lowmem.
1027 Lowmem is memory which can be used for everything that
1028 highmem can be used for, but it is also available for the
1029 kernel's use for its own data structures. Among many
1030 other things, it is where everything from the Slab is
1031 allocated. Bad things happen when you're out of lowmem.
1033 total amount of swap space available
1035 Memory which has been evicted from RAM, and is temporarily
1038 Memory which is waiting to get written back to the disk
1040 Memory which is actively being written back to the disk
1042 Non-file backed pages mapped into userspace page tables
1044 The amount of RAM/memory in KB, the kernel identifies as
1047 Non-file backed huge pages mapped into userspace page tables
1049 files which have been mmaped, such as libraries
1051 Total memory used by shared memory (shmem) and tmpfs
1053 Memory used by shared memory (shmem) and tmpfs allocated
1056 Shared memory mapped into userspace with huge pages
1058 Kernel allocations that the kernel will attempt to reclaim
1059 under memory pressure. Includes SReclaimable (below), and other
1060 direct allocations with a shrinker.
1062 in-kernel data structures cache
1064 Part of Slab, that might be reclaimed, such as caches
1066 Part of Slab, that cannot be reclaimed on memory pressure
1068 amount of memory dedicated to the lowest level of page
1071 Always zero. Previous counted pages which had been written to
1072 the server, but has not been committed to stable storage.
1074 Memory used for block device "bounce buffers"
1076 Memory used by FUSE for temporary writeback buffers
1078 Based on the overcommit ratio ('vm.overcommit_ratio'),
1079 this is the total amount of memory currently available to
1080 be allocated on the system. This limit is only adhered to
1081 if strict overcommit accounting is enabled (mode 2 in
1082 'vm.overcommit_memory').
1084 The CommitLimit is calculated with the following formula::
1086 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
1087 overcommit_ratio / 100 + [total swap pages]
1089 For example, on a system with 1G of physical RAM and 7G
1090 of swap with a `vm.overcommit_ratio` of 30 it would
1091 yield a CommitLimit of 7.3G.
1093 For more details, see the memory overcommit documentation
1094 in vm/overcommit-accounting.
1096 The amount of memory presently allocated on the system.
1097 The committed memory is a sum of all of the memory which
1098 has been allocated by processes, even if it has not been
1099 "used" by them as of yet. A process which malloc()'s 1G
1100 of memory, but only touches 300M of it will show up as
1101 using 1G. This 1G is memory which has been "committed" to
1102 by the VM and can be used at any time by the allocating
1103 application. With strict overcommit enabled on the system
1104 (mode 2 in 'vm.overcommit_memory'), allocations which would
1105 exceed the CommitLimit (detailed above) will not be permitted.
1106 This is useful if one needs to guarantee that processes will
1107 not fail due to lack of memory once that memory has been
1108 successfully allocated.
1110 total size of vmalloc memory area
1112 amount of vmalloc area which is used
1114 largest contiguous block of vmalloc area which is free
1116 Memory allocated to the percpu allocator used to back percpu
1117 allocations. This stat excludes the cost of metadata.
1122 Provides information about vmalloced/vmaped areas. One line per area,
1123 containing the virtual address range of the area, size in bytes,
1124 caller information of the creator, and optional information depending
1125 on the kind of area:
1127 ========== ===================================================
1128 pages=nr number of pages
1129 phys=addr if a physical address was specified
1130 ioremap I/O mapping (ioremap() and friends)
1131 vmalloc vmalloc() area
1133 user VM_USERMAP area
1134 vpages buffer for pages pointers was vmalloced (huge area)
1135 N<node>=nr (Only on NUMA kernels)
1136 Number of pages allocated on memory node <node>
1137 ========== ===================================================
1141 > cat /proc/vmallocinfo
1142 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
1143 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
1144 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
1145 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
1146 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
1147 phys=7fee8000 ioremap
1148 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
1149 phys=7fee7000 ioremap
1150 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
1151 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
1152 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
1153 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
1154 pages=2 vmalloc N1=2
1155 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
1156 /0x130 [x_tables] pages=4 vmalloc N0=4
1157 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
1158 pages=14 vmalloc N2=14
1159 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
1160 pages=4 vmalloc N1=4
1161 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
1162 pages=2 vmalloc N1=2
1163 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
1164 pages=10 vmalloc N0=10
1170 Provides counts of softirq handlers serviced since boot time, for each CPU.
1174 > cat /proc/softirqs
1177 TIMER: 27166 27120 27097 27034
1182 SCHED: 27035 26983 26971 26746
1184 RCU: 1678 1769 2178 2250
1187 1.3 IDE devices in /proc/ide
1188 ----------------------------
1190 The subdirectory /proc/ide contains information about all IDE devices of which
1191 the kernel is aware. There is one subdirectory for each IDE controller, the
1192 file drivers and a link for each IDE device, pointing to the device directory
1193 in the controller specific subtree.
1195 The file 'drivers' contains general information about the drivers used for the
1198 > cat /proc/ide/drivers
1199 ide-cdrom version 4.53
1200 ide-disk version 1.08
1202 More detailed information can be found in the controller specific
1203 subdirectories. These are named ide0, ide1 and so on. Each of these
1204 directories contains the files shown in table 1-6.
1207 .. table:: Table 1-6: IDE controller info in /proc/ide/ide?
1209 ======= =======================================
1211 ======= =======================================
1212 channel IDE channel (0 or 1)
1213 config Configuration (only for PCI/IDE bridge)
1215 model Type/Chipset of IDE controller
1216 ======= =======================================
1218 Each device connected to a controller has a separate subdirectory in the
1219 controllers directory. The files listed in table 1-7 are contained in these
1223 .. table:: Table 1-7: IDE device information
1225 ================ ==========================================
1227 ================ ==========================================
1229 capacity Capacity of the medium (in 512Byte blocks)
1230 driver driver and version
1231 geometry physical and logical geometry
1232 identify device identify block
1234 model device identifier
1235 settings device setup
1236 smart_thresholds IDE disk management thresholds
1237 smart_values IDE disk management values
1238 ================ ==========================================
1240 The most interesting file is ``settings``. This file contains a nice
1241 overview of the drive parameters::
1243 # cat /proc/ide/ide0/hda/settings
1244 name value min max mode
1245 ---- ----- --- --- ----
1246 bios_cyl 526 0 65535 rw
1247 bios_head 255 0 255 rw
1248 bios_sect 63 0 63 rw
1249 breada_readahead 4 0 127 rw
1251 file_readahead 72 0 2097151 rw
1253 keepsettings 0 0 1 rw
1254 max_kb_per_request 122 1 127 rw
1258 pio_mode write-only 0 255 w
1264 1.4 Networking info in /proc/net
1265 --------------------------------
1267 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1268 additional values you get for IP version 6 if you configure the kernel to
1269 support this. Table 1-9 lists the files and their meaning.
1272 .. table:: Table 1-8: IPv6 info in /proc/net
1274 ========== =====================================================
1276 ========== =====================================================
1277 udp6 UDP sockets (IPv6)
1278 tcp6 TCP sockets (IPv6)
1279 raw6 Raw device statistics (IPv6)
1280 igmp6 IP multicast addresses, which this host joined (IPv6)
1281 if_inet6 List of IPv6 interface addresses
1282 ipv6_route Kernel routing table for IPv6
1283 rt6_stats Global IPv6 routing tables statistics
1284 sockstat6 Socket statistics (IPv6)
1285 snmp6 Snmp data (IPv6)
1286 ========== =====================================================
1288 .. table:: Table 1-9: Network info in /proc/net
1290 ============= ================================================================
1292 ============= ================================================================
1293 arp Kernel ARP table
1294 dev network devices with statistics
1295 dev_mcast the Layer2 multicast groups a device is listening too
1296 (interface index, label, number of references, number of bound
1298 dev_stat network device status
1299 ip_fwchains Firewall chain linkage
1300 ip_fwnames Firewall chain names
1301 ip_masq Directory containing the masquerading tables
1302 ip_masquerade Major masquerading table
1303 netstat Network statistics
1304 raw raw device statistics
1305 route Kernel routing table
1306 rpc Directory containing rpc info
1307 rt_cache Routing cache
1309 sockstat Socket statistics
1312 unix UNIX domain sockets
1313 wireless Wireless interface data (Wavelan etc)
1314 igmp IP multicast addresses, which this host joined
1315 psched Global packet scheduler parameters.
1316 netlink List of PF_NETLINK sockets
1317 ip_mr_vifs List of multicast virtual interfaces
1318 ip_mr_cache List of multicast routing cache
1319 ============= ================================================================
1321 You can use this information to see which network devices are available in
1322 your system and how much traffic was routed over those devices::
1325 Inter-|Receive |[...
1326 face |bytes packets errs drop fifo frame compressed multicast|[...
1327 lo: 908188 5596 0 0 0 0 0 0 [...
1328 ppp0:15475140 20721 410 0 0 410 0 0 [...
1329 eth0: 614530 7085 0 0 0 0 0 1 [...
1332 ...] bytes packets errs drop fifo colls carrier compressed
1333 ...] 908188 5596 0 0 0 0 0 0
1334 ...] 1375103 17405 0 0 0 0 0 0
1335 ...] 1703981 5535 0 0 0 3 0 0
1337 In addition, each Channel Bond interface has its own directory. For
1338 example, the bond0 device will have a directory called /proc/net/bond0/.
1339 It will contain information that is specific to that bond, such as the
1340 current slaves of the bond, the link status of the slaves, and how
1341 many times the slaves link has failed.
1346 If you have a SCSI host adapter in your system, you'll find a subdirectory
1347 named after the driver for this adapter in /proc/scsi. You'll also see a list
1348 of all recognized SCSI devices in /proc/scsi::
1350 >cat /proc/scsi/scsi
1352 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1353 Vendor: IBM Model: DGHS09U Rev: 03E0
1354 Type: Direct-Access ANSI SCSI revision: 03
1355 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1356 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1357 Type: CD-ROM ANSI SCSI revision: 02
1360 The directory named after the driver has one file for each adapter found in
1361 the system. These files contain information about the controller, including
1362 the used IRQ and the IO address range. The amount of information shown is
1363 dependent on the adapter you use. The example shows the output for an Adaptec
1364 AHA-2940 SCSI adapter::
1366 > cat /proc/scsi/aic7xxx/0
1368 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1370 TCQ Enabled By Default : Disabled
1371 AIC7XXX_PROC_STATS : Disabled
1372 AIC7XXX_RESET_DELAY : 5
1373 Adapter Configuration:
1374 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1375 Ultra Wide Controller
1376 PCI MMAPed I/O Base: 0xeb001000
1377 Adapter SEEPROM Config: SEEPROM found and used.
1378 Adaptec SCSI BIOS: Enabled
1380 SCBs: Active 0, Max Active 2,
1381 Allocated 15, HW 16, Page 255
1383 BIOS Control Word: 0x18b6
1384 Adapter Control Word: 0x005b
1385 Extended Translation: Enabled
1386 Disconnect Enable Flags: 0xffff
1387 Ultra Enable Flags: 0x0001
1388 Tag Queue Enable Flags: 0x0000
1389 Ordered Queue Tag Flags: 0x0000
1390 Default Tag Queue Depth: 8
1391 Tagged Queue By Device array for aic7xxx host instance 0:
1392 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1393 Actual queue depth per device for aic7xxx host instance 0:
1394 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1397 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1398 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1399 Total transfers 160151 (74577 reads and 85574 writes)
1401 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1402 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1403 Total transfers 0 (0 reads and 0 writes)
1406 1.6 Parallel port info in /proc/parport
1407 ---------------------------------------
1409 The directory /proc/parport contains information about the parallel ports of
1410 your system. It has one subdirectory for each port, named after the port
1413 These directories contain the four files shown in Table 1-10.
1416 .. table:: Table 1-10: Files in /proc/parport
1418 ========= ====================================================================
1420 ========= ====================================================================
1421 autoprobe Any IEEE-1284 device ID information that has been acquired.
1422 devices list of the device drivers using that port. A + will appear by the
1423 name of the device currently using the port (it might not appear
1425 hardware Parallel port's base address, IRQ line and DMA channel.
1426 irq IRQ that parport is using for that port. This is in a separate
1427 file to allow you to alter it by writing a new value in (IRQ
1429 ========= ====================================================================
1431 1.7 TTY info in /proc/tty
1432 -------------------------
1434 Information about the available and actually used tty's can be found in the
1435 directory /proc/tty. You'll find entries for drivers and line disciplines in
1436 this directory, as shown in Table 1-11.
1439 .. table:: Table 1-11: Files in /proc/tty
1441 ============= ==============================================
1443 ============= ==============================================
1444 drivers list of drivers and their usage
1445 ldiscs registered line disciplines
1446 driver/serial usage statistic and status of single tty lines
1447 ============= ==============================================
1449 To see which tty's are currently in use, you can simply look into the file
1452 > cat /proc/tty/drivers
1453 pty_slave /dev/pts 136 0-255 pty:slave
1454 pty_master /dev/ptm 128 0-255 pty:master
1455 pty_slave /dev/ttyp 3 0-255 pty:slave
1456 pty_master /dev/pty 2 0-255 pty:master
1457 serial /dev/cua 5 64-67 serial:callout
1458 serial /dev/ttyS 4 64-67 serial
1459 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1460 /dev/ptmx /dev/ptmx 5 2 system
1461 /dev/console /dev/console 5 1 system:console
1462 /dev/tty /dev/tty 5 0 system:/dev/tty
1463 unknown /dev/tty 4 1-63 console
1466 1.8 Miscellaneous kernel statistics in /proc/stat
1467 -------------------------------------------------
1469 Various pieces of information about kernel activity are available in the
1470 /proc/stat file. All of the numbers reported in this file are aggregates
1471 since the system first booted. For a quick look, simply cat the file::
1474 cpu 2255 34 2290 22625563 6290 127 456 0 0 0
1475 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0
1476 cpu1 1123 0 849 11313845 2614 0 18 0 0 0
1477 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1483 softirq 183433 0 21755 12 39 1137 231 21459 2263
1485 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1486 lines. These numbers identify the amount of time the CPU has spent performing
1487 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1488 second). The meanings of the columns are as follows, from left to right:
1490 - user: normal processes executing in user mode
1491 - nice: niced processes executing in user mode
1492 - system: processes executing in kernel mode
1493 - idle: twiddling thumbs
1494 - iowait: In a word, iowait stands for waiting for I/O to complete. But there
1495 are several problems:
1497 1. CPU will not wait for I/O to complete, iowait is the time that a task is
1498 waiting for I/O to complete. When CPU goes into idle state for
1499 outstanding task I/O, another task will be scheduled on this CPU.
1500 2. In a multi-core CPU, the task waiting for I/O to complete is not running
1501 on any CPU, so the iowait of each CPU is difficult to calculate.
1502 3. The value of iowait field in /proc/stat will decrease in certain
1505 So, the iowait is not reliable by reading from /proc/stat.
1506 - irq: servicing interrupts
1507 - softirq: servicing softirqs
1508 - steal: involuntary wait
1509 - guest: running a normal guest
1510 - guest_nice: running a niced guest
1512 The "intr" line gives counts of interrupts serviced since boot time, for each
1513 of the possible system interrupts. The first column is the total of all
1514 interrupts serviced including unnumbered architecture specific interrupts;
1515 each subsequent column is the total for that particular numbered interrupt.
1516 Unnumbered interrupts are not shown, only summed into the total.
1518 The "ctxt" line gives the total number of context switches across all CPUs.
1520 The "btime" line gives the time at which the system booted, in seconds since
1523 The "processes" line gives the number of processes and threads created, which
1524 includes (but is not limited to) those created by calls to the fork() and
1525 clone() system calls.
1527 The "procs_running" line gives the total number of threads that are
1528 running or ready to run (i.e., the total number of runnable threads).
1530 The "procs_blocked" line gives the number of processes currently blocked,
1531 waiting for I/O to complete.
1533 The "softirq" line gives counts of softirqs serviced since boot time, for each
1534 of the possible system softirqs. The first column is the total of all
1535 softirqs serviced; each subsequent column is the total for that particular
1539 1.9 Ext4 file system parameters
1540 -------------------------------
1542 Information about mounted ext4 file systems can be found in
1543 /proc/fs/ext4. Each mounted filesystem will have a directory in
1544 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1545 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1546 in Table 1-12, below.
1548 .. table:: Table 1-12: Files in /proc/fs/ext4/<devname>
1550 ============== ==========================================================
1552 mb_groups details of multiblock allocator buddy cache of free blocks
1553 ============== ==========================================================
1557 Shows registered system console lines.
1559 To see which character device lines are currently used for the system console
1560 /dev/console, you may simply look into the file /proc/consoles::
1562 > cat /proc/consoles
1568 +--------------------+-------------------------------------------------------+
1569 | device | name of the device |
1570 +====================+=======================================================+
1571 | operations | * R = can do read operations |
1572 | | * W = can do write operations |
1573 | | * U = can do unblank |
1574 +--------------------+-------------------------------------------------------+
1575 | flags | * E = it is enabled |
1576 | | * C = it is preferred console |
1577 | | * B = it is primary boot console |
1578 | | * p = it is used for printk buffer |
1579 | | * b = it is not a TTY but a Braille device |
1580 | | * a = it is safe to use when cpu is offline |
1581 +--------------------+-------------------------------------------------------+
1582 | major:minor | major and minor number of the device separated by a |
1584 +--------------------+-------------------------------------------------------+
1589 The /proc file system serves information about the running system. It not only
1590 allows access to process data but also allows you to request the kernel status
1591 by reading files in the hierarchy.
1593 The directory structure of /proc reflects the types of information and makes
1594 it easy, if not obvious, where to look for specific data.
1596 Chapter 2: Modifying System Parameters
1597 ======================================
1602 * Modifying kernel parameters by writing into files found in /proc/sys
1603 * Exploring the files which modify certain parameters
1604 * Review of the /proc/sys file tree
1606 ------------------------------------------------------------------------------
1608 A very interesting part of /proc is the directory /proc/sys. This is not only
1609 a source of information, it also allows you to change parameters within the
1610 kernel. Be very careful when attempting this. You can optimize your system,
1611 but you can also cause it to crash. Never alter kernel parameters on a
1612 production system. Set up a development machine and test to make sure that
1613 everything works the way you want it to. You may have no alternative but to
1614 reboot the machine once an error has been made.
1616 To change a value, simply echo the new value into the file.
1617 You need to be root to do this. You can create your own boot script
1618 to perform this every time your system boots.
1620 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1621 general things in the operation of the Linux kernel. Since some of the files
1622 can inadvertently disrupt your system, it is advisable to read both
1623 documentation and source before actually making adjustments. In any case, be
1624 very careful when writing to any of these files. The entries in /proc may
1625 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1626 review the kernel documentation in the directory /usr/src/linux/Documentation.
1627 This chapter is heavily based on the documentation included in the pre 2.2
1628 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1630 Please see: Documentation/admin-guide/sysctl/ directory for descriptions of these
1636 Certain aspects of kernel behavior can be modified at runtime, without the
1637 need to recompile the kernel, or even to reboot the system. The files in the
1638 /proc/sys tree can not only be read, but also modified. You can use the echo
1639 command to write value into these files, thereby changing the default settings
1643 Chapter 3: Per-process Parameters
1644 =================================
1646 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1647 --------------------------------------------------------------------------------
1649 These files can be used to adjust the badness heuristic used to select which
1650 process gets killed in out of memory (oom) conditions.
1652 The badness heuristic assigns a value to each candidate task ranging from 0
1653 (never kill) to 1000 (always kill) to determine which process is targeted. The
1654 units are roughly a proportion along that range of allowed memory the process
1655 may allocate from based on an estimation of its current memory and swap use.
1656 For example, if a task is using all allowed memory, its badness score will be
1657 1000. If it is using half of its allowed memory, its score will be 500.
1659 The amount of "allowed" memory depends on the context in which the oom killer
1660 was called. If it is due to the memory assigned to the allocating task's cpuset
1661 being exhausted, the allowed memory represents the set of mems assigned to that
1662 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1663 memory represents the set of mempolicy nodes. If it is due to a memory
1664 limit (or swap limit) being reached, the allowed memory is that configured
1665 limit. Finally, if it is due to the entire system being out of memory, the
1666 allowed memory represents all allocatable resources.
1668 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1669 is used to determine which task to kill. Acceptable values range from -1000
1670 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1671 polarize the preference for oom killing either by always preferring a certain
1672 task or completely disabling it. The lowest possible value, -1000, is
1673 equivalent to disabling oom killing entirely for that task since it will always
1674 report a badness score of 0.
1676 Consequently, it is very simple for userspace to define the amount of memory to
1677 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1678 example, is roughly equivalent to allowing the remainder of tasks sharing the
1679 same system, cpuset, mempolicy, or memory controller resources to use at least
1680 50% more memory. A value of -500, on the other hand, would be roughly
1681 equivalent to discounting 50% of the task's allowed memory from being considered
1682 as scoring against the task.
1684 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1685 be used to tune the badness score. Its acceptable values range from -16
1686 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1687 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1688 scaled linearly with /proc/<pid>/oom_score_adj.
1690 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1691 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1692 requires CAP_SYS_RESOURCE.
1695 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1696 -------------------------------------------------------------
1698 This file can be used to check the current score used by the oom-killer for
1699 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1700 process should be killed in an out-of-memory situation.
1702 Please note that the exported value includes oom_score_adj so it is
1703 effectively in range [0,2000].
1706 3.3 /proc/<pid>/io - Display the IO accounting fields
1707 -------------------------------------------------------
1709 This file contains IO statistics for each running process.
1716 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1719 test:/tmp # cat /proc/3828/io
1725 write_bytes: 323932160
1726 cancelled_write_bytes: 0
1735 I/O counter: chars read
1736 The number of bytes which this task has caused to be read from storage. This
1737 is simply the sum of bytes which this process passed to read() and pread().
1738 It includes things like tty IO and it is unaffected by whether or not actual
1739 physical disk IO was required (the read might have been satisfied from
1746 I/O counter: chars written
1747 The number of bytes which this task has caused, or shall cause to be written
1748 to disk. Similar caveats apply here as with rchar.
1754 I/O counter: read syscalls
1755 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1762 I/O counter: write syscalls
1763 Attempt to count the number of write I/O operations, i.e. syscalls like
1764 write() and pwrite().
1770 I/O counter: bytes read
1771 Attempt to count the number of bytes which this process really did cause to
1772 be fetched from the storage layer. Done at the submit_bio() level, so it is
1773 accurate for block-backed filesystems. <please add status regarding NFS and
1774 CIFS at a later time>
1780 I/O counter: bytes written
1781 Attempt to count the number of bytes which this process caused to be sent to
1782 the storage layer. This is done at page-dirtying time.
1785 cancelled_write_bytes
1786 ^^^^^^^^^^^^^^^^^^^^^
1788 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1789 then deletes the file, it will in fact perform no writeout. But it will have
1790 been accounted as having caused 1MB of write.
1791 In other words: The number of bytes which this process caused to not happen,
1792 by truncating pagecache. A task can cause "negative" IO too. If this task
1793 truncates some dirty pagecache, some IO which another task has been accounted
1794 for (in its write_bytes) will not be happening. We _could_ just subtract that
1795 from the truncating task's write_bytes, but there is information loss in doing
1801 At its current implementation state, this is a bit racy on 32-bit machines:
1802 if process A reads process B's /proc/pid/io while process B is updating one
1803 of those 64-bit counters, process A could see an intermediate result.
1806 More information about this can be found within the taskstats documentation in
1807 Documentation/accounting.
1809 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1810 ---------------------------------------------------------------
1811 When a process is dumped, all anonymous memory is written to a core file as
1812 long as the size of the core file isn't limited. But sometimes we don't want
1813 to dump some memory segments, for example, huge shared memory or DAX.
1814 Conversely, sometimes we want to save file-backed memory segments into a core
1815 file, not only the individual files.
1817 /proc/<pid>/coredump_filter allows you to customize which memory segments
1818 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1819 of memory types. If a bit of the bitmask is set, memory segments of the
1820 corresponding memory type are dumped, otherwise they are not dumped.
1822 The following 9 memory types are supported:
1824 - (bit 0) anonymous private memory
1825 - (bit 1) anonymous shared memory
1826 - (bit 2) file-backed private memory
1827 - (bit 3) file-backed shared memory
1828 - (bit 4) ELF header pages in file-backed private memory areas (it is
1829 effective only if the bit 2 is cleared)
1830 - (bit 5) hugetlb private memory
1831 - (bit 6) hugetlb shared memory
1832 - (bit 7) DAX private memory
1833 - (bit 8) DAX shared memory
1835 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1836 are always dumped regardless of the bitmask status.
1838 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1839 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1841 The default value of coredump_filter is 0x33; this means all anonymous memory
1842 segments, ELF header pages and hugetlb private memory are dumped.
1844 If you don't want to dump all shared memory segments attached to pid 1234,
1845 write 0x31 to the process's proc file::
1847 $ echo 0x31 > /proc/1234/coredump_filter
1849 When a new process is created, the process inherits the bitmask status from its
1850 parent. It is useful to set up coredump_filter before the program runs.
1853 $ echo 0x7 > /proc/self/coredump_filter
1856 3.5 /proc/<pid>/mountinfo - Information about mounts
1857 --------------------------------------------------------
1859 This file contains lines of the form::
1861 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1862 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4)
1864 (1) mount ID: unique identifier of the mount (may be reused after umount)
1865 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1866 (3) major:minor: value of st_dev for files on filesystem
1867 (4) root: root of the mount within the filesystem
1868 (5) mount point: mount point relative to the process's root
1869 (6) mount options: per mount options
1870 (n…m) optional fields: zero or more fields of the form "tag[:value]"
1871 (m+1) separator: marks the end of the optional fields
1872 (m+2) filesystem type: name of filesystem of the form "type[.subtype]"
1873 (m+3) mount source: filesystem specific information or "none"
1874 (m+4) super options: per super block options
1876 Parsers should ignore all unrecognised optional fields. Currently the
1877 possible optional fields are:
1879 ================ ==============================================================
1880 shared:X mount is shared in peer group X
1881 master:X mount is slave to peer group X
1882 propagate_from:X mount is slave and receives propagation from peer group X [#]_
1883 unbindable mount is unbindable
1884 ================ ==============================================================
1886 .. [#] X is the closest dominant peer group under the process's root. If
1887 X is the immediate master of the mount, or if there's no dominant peer
1888 group under the same root, then only the "master:X" field is present
1889 and not the "propagate_from:X" field.
1891 For more information on mount propagation see:
1893 Documentation/filesystems/sharedsubtree.rst
1896 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1897 --------------------------------------------------------
1898 These files provide a method to access a task's comm value. It also allows for
1899 a task to set its own or one of its thread siblings comm value. The comm value
1900 is limited in size compared to the cmdline value, so writing anything longer
1901 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1905 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1906 -------------------------------------------------------------------------
1907 This file provides a fast way to retrieve first level children pids
1908 of a task pointed by <pid>/<tid> pair. The format is a space separated
1911 Note the "first level" here -- if a child has its own children they will
1912 not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children
1913 to obtain the descendants.
1915 Since this interface is intended to be fast and cheap it doesn't
1916 guarantee to provide precise results and some children might be
1917 skipped, especially if they've exited right after we printed their
1918 pids, so one needs to either stop or freeze processes being inspected
1919 if precise results are needed.
1922 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1923 ---------------------------------------------------------------
1924 This file provides information associated with an opened file. The regular
1925 files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'.
1926 The 'pos' represents the current offset of the opened file in decimal
1927 form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the
1928 file has been created with [see open(2) for details] and 'mnt_id' represents
1929 mount ID of the file system containing the opened file [see 3.5
1930 /proc/<pid>/mountinfo for details]. 'ino' represents the inode number of
1933 A typical output is::
1940 All locks associated with a file descriptor are shown in its fdinfo too::
1942 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1944 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1945 pair provide additional information particular to the objects they represent.
1958 where 'eventfd-count' is hex value of a counter.
1969 sigmask: 0000000000000200
1971 where 'sigmask' is hex value of the signal mask associated
1983 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7
1985 where 'tfd' is a target file descriptor number in decimal form,
1986 'events' is events mask being watched and the 'data' is data
1987 associated with a target [see epoll(7) for more details].
1989 The 'pos' is current offset of the target file in decimal form
1990 [see lseek(2)], 'ino' and 'sdev' are inode and device numbers
1991 where target file resides, all in hex format.
1995 For inotify files the format is the following::
2001 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
2003 where 'wd' is a watch descriptor in decimal form, i.e. a target file
2004 descriptor number, 'ino' and 'sdev' are inode and device where the
2005 target file resides and the 'mask' is the mask of events, all in hex
2006 form [see inotify(7) for more details].
2008 If the kernel was built with exportfs support, the path to the target
2009 file is encoded as a file handle. The file handle is provided by three
2010 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
2013 If the kernel is built without exportfs support the file handle won't be
2016 If there is no inotify mark attached yet the 'inotify' line will be omitted.
2018 For fanotify files the format is::
2024 fanotify flags:10 event-flags:0
2025 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
2026 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
2028 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
2029 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
2030 flags associated with mark which are tracked separately from events
2031 mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events
2032 mask and 'ignored_mask' is the mask of events which are to be ignored.
2033 All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
2034 provide information about flags and mask used in fanotify_mark
2035 call [see fsnotify manpage for details].
2037 While the first three lines are mandatory and always printed, the rest is
2038 optional and may be omitted if no marks created yet.
2052 it_value: (0, 49406829)
2055 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
2056 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
2057 flags in octal form been used to setup the timer [see timerfd_settime(2) for
2058 details]. 'it_value' is remaining time until the timer expiration.
2059 'it_interval' is the interval for the timer. Note the timer might be set up
2060 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
2061 still exhibits timer's remaining time.
2074 exp_name: system-heap
2076 where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of
2077 the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter.
2079 3.9 /proc/<pid>/map_files - Information about memory mapped files
2080 ---------------------------------------------------------------------
2081 This directory contains symbolic links which represent memory mapped files
2082 the process is maintaining. Example output::
2084 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
2085 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
2086 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
2088 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
2089 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
2091 The name of a link represents the virtual memory bounds of a mapping, i.e.
2092 vm_area_struct::vm_start-vm_area_struct::vm_end.
2094 The main purpose of the map_files is to retrieve a set of memory mapped
2095 files in a fast way instead of parsing /proc/<pid>/maps or
2096 /proc/<pid>/smaps, both of which contain many more records. At the same
2097 time one can open(2) mappings from the listings of two processes and
2098 comparing their inode numbers to figure out which anonymous memory areas
2099 are actually shared.
2101 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
2102 ---------------------------------------------------------
2103 This file provides the value of the task's timerslack value in nanoseconds.
2104 This value specifies an amount of time that normal timers may be deferred
2105 in order to coalesce timers and avoid unnecessary wakeups.
2107 This allows a task's interactivity vs power consumption tradeoff to be
2110 Writing 0 to the file will set the task's timerslack to the default value.
2112 Valid values are from 0 - ULLONG_MAX
2114 An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
2115 permissions on the task specified to change its timerslack_ns value.
2117 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
2118 -----------------------------------------------------------------
2119 When CONFIG_LIVEPATCH is enabled, this file displays the value of the
2120 patch state for the task.
2122 A value of '-1' indicates that no patch is in transition.
2124 A value of '0' indicates that a patch is in transition and the task is
2125 unpatched. If the patch is being enabled, then the task hasn't been
2126 patched yet. If the patch is being disabled, then the task has already
2129 A value of '1' indicates that a patch is in transition and the task is
2130 patched. If the patch is being enabled, then the task has already been
2131 patched. If the patch is being disabled, then the task hasn't been
2134 3.12 /proc/<pid>/arch_status - task architecture specific status
2135 -------------------------------------------------------------------
2136 When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the
2137 architecture specific status of the task.
2144 $ cat /proc/6753/arch_status
2145 AVX512_elapsed_ms: 8
2150 x86 specific entries
2151 ~~~~~~~~~~~~~~~~~~~~~
2156 If AVX512 is supported on the machine, this entry shows the milliseconds
2157 elapsed since the last time AVX512 usage was recorded. The recording
2158 happens on a best effort basis when a task is scheduled out. This means
2159 that the value depends on two factors:
2161 1) The time which the task spent on the CPU without being scheduled
2162 out. With CPU isolation and a single runnable task this can take
2165 2) The time since the task was scheduled out last. Depending on the
2166 reason for being scheduled out (time slice exhausted, syscall ...)
2167 this can be arbitrary long time.
2169 As a consequence the value cannot be considered precise and authoritative
2170 information. The application which uses this information has to be aware
2171 of the overall scenario on the system in order to determine whether a
2172 task is a real AVX512 user or not. Precise information can be obtained
2173 with performance counters.
2175 A special value of '-1' indicates that no AVX512 usage was recorded, thus
2176 the task is unlikely an AVX512 user, but depends on the workload and the
2177 scheduling scenario, it also could be a false negative mentioned above.
2179 Chapter 4: Configuring procfs
2180 =============================
2183 ---------------------
2185 The following mount options are supported:
2187 ========= ========================================================
2188 hidepid= Set /proc/<pid>/ access mode.
2189 gid= Set the group authorized to learn processes information.
2190 subset= Show only the specified subset of procfs.
2191 ========= ========================================================
2193 hidepid=off or hidepid=0 means classic mode - everybody may access all
2194 /proc/<pid>/ directories (default).
2196 hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/
2197 directories but their own. Sensitive files like cmdline, sched*, status are now
2198 protected against other users. This makes it impossible to learn whether any
2199 user runs specific program (given the program doesn't reveal itself by its
2200 behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for
2201 other users, poorly written programs passing sensitive information via program
2202 arguments are now protected against local eavesdroppers.
2204 hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be
2205 fully invisible to other users. It doesn't mean that it hides a fact whether a
2206 process with a specific pid value exists (it can be learned by other means, e.g.
2207 by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by
2208 stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of
2209 gathering information about running processes, whether some daemon runs with
2210 elevated privileges, whether other user runs some sensitive program, whether
2211 other users run any program at all, etc.
2213 hidepid=ptraceable or hidepid=4 means that procfs should only contain
2214 /proc/<pid>/ directories that the caller can ptrace.
2216 gid= defines a group authorized to learn processes information otherwise
2217 prohibited by hidepid=. If you use some daemon like identd which needs to learn
2218 information about processes information, just add identd to this group.
2220 subset=pid hides all top level files and directories in the procfs that
2221 are not related to tasks.
2223 Chapter 5: Filesystem behavior
2224 ==============================
2226 Originally, before the advent of pid namepsace, procfs was a global file
2227 system. It means that there was only one procfs instance in the system.
2229 When pid namespace was added, a separate procfs instance was mounted in
2230 each pid namespace. So, procfs mount options are global among all
2231 mountpoints within the same namespace::
2233 # grep ^proc /proc/mounts
2234 proc /proc proc rw,relatime,hidepid=2 0 0
2236 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc
2237 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0
2238 +++ exited with 0 +++
2240 # grep ^proc /proc/mounts
2241 proc /proc proc rw,relatime,hidepid=2 0 0
2242 proc /tmp/proc proc rw,relatime,hidepid=2 0 0
2244 and only after remounting procfs mount options will change at all
2247 # mount -o remount,hidepid=1 -t proc proc /tmp/proc
2249 # grep ^proc /proc/mounts
2250 proc /proc proc rw,relatime,hidepid=1 0 0
2251 proc /tmp/proc proc rw,relatime,hidepid=1 0 0
2253 This behavior is different from the behavior of other filesystems.
2255 The new procfs behavior is more like other filesystems. Each procfs mount
2256 creates a new procfs instance. Mount options affect own procfs instance.
2257 It means that it became possible to have several procfs instances
2258 displaying tasks with different filtering options in one pid namespace::
2260 # mount -o hidepid=invisible -t proc proc /proc
2261 # mount -o hidepid=noaccess -t proc proc /tmp/proc
2262 # grep ^proc /proc/mounts
2263 proc /proc proc rw,relatime,hidepid=invisible 0 0
2264 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0