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
50 3.13 /proc/<pid>/fd - List of symlinks to open files
60 0.1 Introduction/Credits
61 ------------------------
63 This documentation is part of a soon (or so we hope) to be released book on
64 the SuSE Linux distribution. As there is no complete documentation for the
65 /proc file system and we've used many freely available sources to write these
66 chapters, it seems only fair to give the work back to the Linux community.
67 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
68 afraid it's still far from complete, but we hope it will be useful. As far as
69 we know, it is the first 'all-in-one' document about the /proc file system. It
70 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
71 SPARC, AXP, etc., features, you probably won't find what you are looking for.
72 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
73 additions and patches are welcome and will be added to this document if you
76 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
77 other people for help compiling this documentation. We'd also like to extend a
78 special thank you to Andi Kleen for documentation, which we relied on heavily
79 to create this document, as well as the additional information he provided.
80 Thanks to everybody else who contributed source or docs to the Linux kernel
81 and helped create a great piece of software... :)
83 If you have any comments, corrections or additions, please don't hesitate to
84 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
87 The latest version of this document is available online at
88 https://www.kernel.org/doc/html/latest/filesystems/proc.html
90 If the above direction does not works for you, you could try the kernel
91 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
92 comandante@zaralinux.com.
97 We don't guarantee the correctness of this document, and if you come to us
98 complaining about how you screwed up your system because of incorrect
99 documentation, we won't feel responsible...
101 Chapter 1: Collecting System Information
102 ========================================
106 * Investigating the properties of the pseudo file system /proc and its
107 ability to provide information on the running Linux system
108 * Examining /proc's structure
109 * Uncovering various information about the kernel and the processes running
112 ------------------------------------------------------------------------------
114 The proc file system acts as an interface to internal data structures in the
115 kernel. It can be used to obtain information about the system and to change
116 certain kernel parameters at runtime (sysctl).
118 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
119 show you how you can use /proc/sys to change settings.
121 1.1 Process-Specific Subdirectories
122 -----------------------------------
124 The directory /proc contains (among other things) one subdirectory for each
125 process running on the system, which is named after the process ID (PID).
127 The link 'self' points to the process reading the file system. Each process
128 subdirectory has the entries listed in Table 1-1.
130 Note that an open file descriptor to /proc/<pid> or to any of its
131 contained files or subdirectories does not prevent <pid> being reused
132 for some other process in the event that <pid> exits. Operations on
133 open /proc/<pid> file descriptors corresponding to dead processes
134 never act on any new process that the kernel may, through chance, have
135 also assigned the process ID <pid>. Instead, operations on these FDs
136 usually fail with ESRCH.
138 .. table:: Table 1-1: Process specific entries in /proc
140 ============= ===============================================================
142 ============= ===============================================================
143 clear_refs Clears page referenced bits shown in smaps output
144 cmdline Command line arguments
145 cpu Current and last cpu in which it was executed (2.4)(smp)
146 cwd Link to the current working directory
147 environ Values of environment variables
148 exe Link to the executable of this process
149 fd Directory, which contains all file descriptors
150 maps Memory maps to executables and library files (2.4)
151 mem Memory held by this process
152 root Link to the root directory of this process
154 statm Process memory status information
155 status Process status in human readable form
156 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function
157 symbol the task is blocked in - or "0" if not blocked.
159 stack Report full stack trace, enable via CONFIG_STACKTRACE
160 smaps An extension based on maps, showing the memory consumption of
161 each mapping and flags associated with it
162 smaps_rollup Accumulated smaps stats for all mappings of the process. This
163 can be derived from smaps, but is faster and more convenient
164 numa_maps An extension based on maps, showing the memory locality and
165 binding policy as well as mem usage (in pages) of each mapping.
166 ============= ===============================================================
168 For example, to get the status information of a process, all you have to do is
169 read the file /proc/PID/status::
171 >cat /proc/self/status
201 SigPnd: 0000000000000000
202 ShdPnd: 0000000000000000
203 SigBlk: 0000000000000000
204 SigIgn: 0000000000000000
205 SigCgt: 0000000000000000
206 CapInh: 00000000fffffeff
207 CapPrm: 0000000000000000
208 CapEff: 0000000000000000
209 CapBnd: ffffffffffffffff
210 CapAmb: 0000000000000000
213 Speculation_Store_Bypass: thread vulnerable
214 SpeculationIndirectBranch: conditional enabled
215 voluntary_ctxt_switches: 0
216 nonvoluntary_ctxt_switches: 1
218 This shows you nearly the same information you would get if you viewed it with
219 the ps command. In fact, ps uses the proc file system to obtain its
220 information. But you get a more detailed view of the process by reading the
221 file /proc/PID/status. It fields are described in table 1-2.
223 The statm file contains more detailed information about the process
224 memory usage. Its seven fields are explained in Table 1-3. The stat file
225 contains detailed information about the process itself. Its fields are
226 explained in Table 1-4.
228 (for SMP CONFIG users)
230 For making accounting scalable, RSS related information are handled in an
231 asynchronous manner and the value may not be very precise. To see a precise
232 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
233 It's slow but very precise.
235 .. table:: Table 1-2: Contents of the status fields (as of 4.19)
237 ========================== ===================================================
239 ========================== ===================================================
240 Name filename of the executable
241 Umask file mode creation mask
242 State state (R is running, S is sleeping, D is sleeping
243 in an uninterruptible wait, Z is zombie,
244 T is traced or stopped)
246 Ngid NUMA group ID (0 if none)
248 PPid process id of the parent process
249 TracerPid PID of process tracing this process (0 if not, or
250 the tracer is outside of the current pid namespace)
251 Uid Real, effective, saved set, and file system UIDs
252 Gid Real, effective, saved set, and file system GIDs
253 FDSize number of file descriptor slots currently allocated
254 Groups supplementary group list
255 NStgid descendant namespace thread group ID hierarchy
256 NSpid descendant namespace process ID hierarchy
257 NSpgid descendant namespace process group ID hierarchy
258 NSsid descendant namespace session ID hierarchy
259 VmPeak peak virtual memory size
260 VmSize total program size
261 VmLck locked memory size
262 VmPin pinned memory size
263 VmHWM peak resident set size ("high water mark")
264 VmRSS size of memory portions. It contains the three
266 (VmRSS = RssAnon + RssFile + RssShmem)
267 RssAnon size of resident anonymous memory
268 RssFile size of resident file mappings
269 RssShmem size of resident shmem memory (includes SysV shm,
270 mapping of tmpfs and shared anonymous mappings)
271 VmData size of private data segments
272 VmStk size of stack segments
273 VmExe size of text segment
274 VmLib size of shared library code
275 VmPTE size of page table entries
276 VmSwap amount of swap used by anonymous private data
277 (shmem swap usage is not included)
278 HugetlbPages size of hugetlb memory portions
279 CoreDumping process's memory is currently being dumped
280 (killing the process may lead to a corrupted core)
281 THP_enabled process is allowed to use THP (returns 0 when
282 PR_SET_THP_DISABLE is set on the process
283 Threads number of threads
284 SigQ number of signals queued/max. number for queue
285 SigPnd bitmap of pending signals for the thread
286 ShdPnd bitmap of shared pending signals for the process
287 SigBlk bitmap of blocked signals
288 SigIgn bitmap of ignored signals
289 SigCgt bitmap of caught signals
290 CapInh bitmap of inheritable capabilities
291 CapPrm bitmap of permitted capabilities
292 CapEff bitmap of effective capabilities
293 CapBnd bitmap of capabilities bounding set
294 CapAmb bitmap of ambient capabilities
295 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
296 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
297 Speculation_Store_Bypass speculative store bypass mitigation status
298 SpeculationIndirectBranch indirect branch speculation mode
299 Cpus_allowed mask of CPUs on which this process may run
300 Cpus_allowed_list Same as previous, but in "list format"
301 Mems_allowed mask of memory nodes allowed to this process
302 Mems_allowed_list Same as previous, but in "list format"
303 voluntary_ctxt_switches number of voluntary context switches
304 nonvoluntary_ctxt_switches number of non voluntary context switches
305 ========================== ===================================================
308 .. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3)
310 ======== =============================== ==============================
312 ======== =============================== ==============================
313 size total program size (pages) (same as VmSize in status)
314 resident size of memory portions (pages) (same as VmRSS in status)
315 shared number of pages that are shared (i.e. backed by a file, same
316 as RssFile+RssShmem in status)
317 trs number of pages that are 'code' (not including libs; broken,
318 includes data segment)
319 lrs number of pages of library (always 0 on 2.6)
320 drs number of pages of data/stack (including libs; broken,
321 includes library text)
322 dt number of dirty pages (always 0 on 2.6)
323 ======== =============================== ==============================
326 .. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7)
328 ============= ===============================================================
330 ============= ===============================================================
332 tcomm filename of the executable
333 state state (R is running, S is sleeping, D is sleeping in an
334 uninterruptible wait, Z is zombie, T is traced or stopped)
335 ppid process id of the parent process
336 pgrp pgrp of the process
338 tty_nr tty the process uses
339 tty_pgrp pgrp of the tty
341 min_flt number of minor faults
342 cmin_flt number of minor faults with child's
343 maj_flt number of major faults
344 cmaj_flt number of major faults with child's
345 utime user mode jiffies
346 stime kernel mode jiffies
347 cutime user mode jiffies with child's
348 cstime kernel mode jiffies with child's
349 priority priority level
351 num_threads number of threads
352 it_real_value (obsolete, always 0)
353 start_time time the process started after system boot
354 vsize virtual memory size
355 rss resident set memory size
356 rsslim current limit in bytes on the rss
357 start_code address above which program text can run
358 end_code address below which program text can run
359 start_stack address of the start of the main process stack
360 esp current value of ESP
361 eip current value of EIP
362 pending bitmap of pending signals
363 blocked bitmap of blocked signals
364 sigign bitmap of ignored signals
365 sigcatch bitmap of caught signals
366 0 (place holder, used to be the wchan address,
367 use /proc/PID/wchan instead)
370 exit_signal signal to send to parent thread on exit
371 task_cpu which CPU the task is scheduled on
372 rt_priority realtime priority
373 policy scheduling policy (man sched_setscheduler)
374 blkio_ticks time spent waiting for block IO
375 gtime guest time of the task in jiffies
376 cgtime guest time of the task children in jiffies
377 start_data address above which program data+bss is placed
378 end_data address below which program data+bss is placed
379 start_brk address above which program heap can be expanded with brk()
380 arg_start address above which program command line is placed
381 arg_end address below which program command line is placed
382 env_start address above which program environment is placed
383 env_end address below which program environment is placed
384 exit_code the thread's exit_code in the form reported by the waitpid
386 ============= ===============================================================
388 The /proc/PID/maps file contains the currently mapped memory regions and
389 their access permissions.
393 address perms offset dev inode pathname
395 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
396 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
397 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
398 a7cb1000-a7cb2000 ---p 00000000 00:00 0
399 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
400 a7eb2000-a7eb3000 ---p 00000000 00:00 0
401 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
402 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
403 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
404 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
405 a800b000-a800e000 rw-p 00000000 00:00 0
406 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
407 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
408 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
409 a8024000-a8027000 rw-p 00000000 00:00 0
410 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
411 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
412 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
413 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
414 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
416 where "address" is the address space in the process that it occupies, "perms"
417 is a set of permissions::
423 p = private (copy on write)
425 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
426 "inode" is the inode on that device. 0 indicates that no inode is associated
427 with the memory region, as the case would be with BSS (uninitialized data).
428 The "pathname" shows the name associated file for this mapping. If the mapping
429 is not associated with a file:
431 =================== ===========================================
432 [heap] the heap of the program
433 [stack] the stack of the main process
434 [vdso] the "virtual dynamic shared object",
435 the kernel system call handler
436 [anon:<name>] a private anonymous mapping that has been
438 [anon_shmem:<name>] an anonymous shared memory mapping that has
439 been named by userspace
440 =================== ===========================================
442 or if empty, the mapping is anonymous.
444 The /proc/PID/smaps is an extension based on maps, showing the memory
445 consumption for each of the process's mappings. For each mapping (aka Virtual
446 Memory Area, or VMA) there is a series of lines such as the following::
448 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
466 Private_Hugetlb: 0 kB
473 VmFlags: rd ex mr mw me dw
475 The first of these lines shows the same information as is displayed for the
476 mapping in /proc/PID/maps. Following lines show the size of the mapping
477 (size); the size of each page allocated when backing a VMA (KernelPageSize),
478 which is usually the same as the size in the page table entries; the page size
479 used by the MMU when backing a VMA (in most cases, the same as KernelPageSize);
480 the amount of the mapping that is currently resident in RAM (RSS); the
481 process' proportional share of this mapping (PSS); and the number of clean and
482 dirty shared and private pages in the mapping.
484 The "proportional set size" (PSS) of a process is the count of pages it has
485 in memory, where each page is divided by the number of processes sharing it.
486 So if a process has 1000 pages all to itself, and 1000 shared with one other
487 process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which
488 consists of dirty pages. ("Pss_Clean" is not included, but it can be
489 calculated by subtracting "Pss_Dirty" from "Pss".)
491 Note that even a page which is part of a MAP_SHARED mapping, but has only
492 a single pte mapped, i.e. is currently used by only one process, is accounted
493 as private and not as shared.
495 "Referenced" indicates the amount of memory currently marked as referenced or
498 "Anonymous" shows the amount of memory that does not belong to any file. Even
499 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
500 and a page is modified, the file page is replaced by a private anonymous copy.
502 "LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
503 The memory isn't freed immediately with madvise(). It's freed in memory
504 pressure if the memory is clean. Please note that the printed value might
505 be lower than the real value due to optimizations used in the current
506 implementation. If this is not desirable please file a bug report.
508 "AnonHugePages" shows the ammount of memory backed by transparent hugepage.
510 "ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by
513 "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
514 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
515 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
517 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
519 For shmem mappings, "Swap" includes also the size of the mapped (and not
520 replaced by copy-on-write) part of the underlying shmem object out on swap.
521 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
522 does not take into account swapped out page of underlying shmem objects.
523 "Locked" indicates whether the mapping is locked in memory or not.
525 "THPeligible" indicates whether the mapping is eligible for allocating THP
526 pages as well as the THP is PMD mappable or not - 1 if true, 0 otherwise.
527 It just shows the current status.
529 "VmFlags" field deserves a separate description. This member represents the
530 kernel flags associated with the particular virtual memory area in two letter
531 encoded manner. The codes are the following:
533 == =======================================
542 gd stack segment growns down
544 dw disabled write to the mapped file
545 lo pages are locked in memory
546 io memory mapped I/O area
547 sr sequential read advise provided
548 rr random read advise provided
549 dc do not copy area on fork
550 de do not expand area on remapping
551 ac area is accountable
552 nr swap space is not reserved for the area
553 ht area uses huge tlb pages
554 sf synchronous page fault
555 ar architecture specific flag
557 dd do not include area into core dump
560 hg huge page advise flag
561 nh no huge page advise flag
562 mg mergable advise flag
563 bt arm64 BTI guarded page
564 mt arm64 MTE allocation tags are enabled
565 um userfaultfd missing tracking
566 uw userfaultfd wr-protect tracking
567 == =======================================
569 Note that there is no guarantee that every flag and associated mnemonic will
570 be present in all further kernel releases. Things get changed, the flags may
571 be vanished or the reverse -- new added. Interpretation of their meaning
572 might change in future as well. So each consumer of these flags has to
573 follow each specific kernel version for the exact semantic.
575 This file is only present if the CONFIG_MMU kernel configuration option is
578 Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
579 output can be achieved only in the single read call).
581 This typically manifests when doing partial reads of these files while the
582 memory map is being modified. Despite the races, we do provide the following
585 1) The mapped addresses never go backwards, which implies no two
586 regions will ever overlap.
587 2) If there is something at a given vaddr during the entirety of the
588 life of the smaps/maps walk, there will be some output for it.
590 The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps,
591 but their values are the sums of the corresponding values for all mappings of
592 the process. Additionally, it contains these fields:
598 They represent the proportional shares of anonymous, file, and shmem pages, as
599 described for smaps above. These fields are omitted in smaps since each
600 mapping identifies the type (anon, file, or shmem) of all pages it contains.
601 Thus all information in smaps_rollup can be derived from smaps, but at a
602 significantly higher cost.
604 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
605 bits on both physical and virtual pages associated with a process, and the
606 soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst
608 To clear the bits for all the pages associated with the process::
610 > echo 1 > /proc/PID/clear_refs
612 To clear the bits for the anonymous pages associated with the process::
614 > echo 2 > /proc/PID/clear_refs
616 To clear the bits for the file mapped pages associated with the process::
618 > echo 3 > /proc/PID/clear_refs
620 To clear the soft-dirty bit::
622 > echo 4 > /proc/PID/clear_refs
624 To reset the peak resident set size ("high water mark") to the process's
627 > echo 5 > /proc/PID/clear_refs
629 Any other value written to /proc/PID/clear_refs will have no effect.
631 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
632 using /proc/kpageflags and number of times a page is mapped using
633 /proc/kpagecount. For detailed explanation, see
634 Documentation/admin-guide/mm/pagemap.rst.
636 The /proc/pid/numa_maps is an extension based on maps, showing the memory
637 locality and binding policy, as well as the memory usage (in pages) of
638 each mapping. The output follows a general format where mapping details get
639 summarized separated by blank spaces, one mapping per each file line::
641 address policy mapping details
643 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
644 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
645 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
646 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
647 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
648 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
649 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
650 320698b000 default file=/lib64/libc-2.12.so
651 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
652 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
653 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
654 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
655 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
656 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
657 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
658 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
662 "address" is the starting address for the mapping;
664 "policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst);
666 "mapping details" summarizes mapping data such as mapping type, page usage counters,
667 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
668 size, in KB, that is backing the mapping up.
673 Similar to the process entries, the kernel data files give information about
674 the running kernel. The files used to obtain this information are contained in
675 /proc and are listed in Table 1-5. Not all of these will be present in your
676 system. It depends on the kernel configuration and the loaded modules, which
677 files are there, and which are missing.
679 .. table:: Table 1-5: Kernel info in /proc
681 ============ ===============================================================
683 ============ ===============================================================
684 apm Advanced power management info
685 buddyinfo Kernel memory allocator information (see text) (2.5)
686 bus Directory containing bus specific information
687 cmdline Kernel command line
688 cpuinfo Info about the CPU
689 devices Available devices (block and character)
690 dma Used DMS channels
691 filesystems Supported filesystems
692 driver Various drivers grouped here, currently rtc (2.4)
693 execdomains Execdomains, related to security (2.4)
694 fb Frame Buffer devices (2.4)
695 fs File system parameters, currently nfs/exports (2.4)
696 ide Directory containing info about the IDE subsystem
697 interrupts Interrupt usage
698 iomem Memory map (2.4)
699 ioports I/O port usage
700 irq Masks for irq to cpu affinity (2.4)(smp?)
701 isapnp ISA PnP (Plug&Play) Info (2.4)
702 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
704 ksyms Kernel symbol table
705 loadavg Load average of last 1, 5 & 15 minutes;
706 number of processes currently runnable (running or on ready queue);
707 total number of processes in system;
709 All fields are separated by one space except "number of
710 processes currently runnable" and "total number of processes
711 in system", which are separated by a slash ('/'). Example:
712 0.61 0.61 0.55 3/828 22084
716 modules List of loaded modules
717 mounts Mounted filesystems
718 net Networking info (see text)
719 pagetypeinfo Additional page allocator information (see text) (2.5)
720 partitions Table of partitions known to the system
721 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
722 decoupled by lspci (2.4)
724 scsi SCSI info (see text)
725 slabinfo Slab pool info
726 softirqs softirq usage
727 stat Overall statistics
728 swaps Swap space utilization
730 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
731 tty Info of tty drivers
732 uptime Wall clock since boot, combined idle time of all cpus
733 version Kernel version
734 video bttv info of video resources (2.4)
735 vmallocinfo Show vmalloced areas
736 ============ ===============================================================
738 You can, for example, check which interrupts are currently in use and what
739 they are used for by looking in the file /proc/interrupts::
741 > cat /proc/interrupts
743 0: 8728810 XT-PIC timer
744 1: 895 XT-PIC keyboard
746 3: 531695 XT-PIC aha152x
747 4: 2014133 XT-PIC serial
748 5: 44401 XT-PIC pcnet_cs
751 12: 182918 XT-PIC PS/2 Mouse
753 14: 1232265 XT-PIC ide0
757 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
758 output of a SMP machine)::
760 > cat /proc/interrupts
763 0: 1243498 1214548 IO-APIC-edge timer
764 1: 8949 8958 IO-APIC-edge keyboard
765 2: 0 0 XT-PIC cascade
766 5: 11286 10161 IO-APIC-edge soundblaster
767 8: 1 0 IO-APIC-edge rtc
768 9: 27422 27407 IO-APIC-edge 3c503
769 12: 113645 113873 IO-APIC-edge PS/2 Mouse
771 14: 22491 24012 IO-APIC-edge ide0
772 15: 2183 2415 IO-APIC-edge ide1
773 17: 30564 30414 IO-APIC-level eth0
774 18: 177 164 IO-APIC-level bttv
779 NMI is incremented in this case because every timer interrupt generates a NMI
780 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
782 LOC is the local interrupt counter of the internal APIC of every CPU.
784 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
785 connects the CPUs in a SMP system. This means that an error has been detected,
786 the IO-APIC automatically retry the transmission, so it should not be a big
787 problem, but you should read the SMP-FAQ.
789 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
790 /proc/interrupts to display every IRQ vector in use by the system, not
791 just those considered 'most important'. The new vectors are:
794 interrupt raised when a machine check threshold counter
795 (typically counting ECC corrected errors of memory or cache) exceeds
796 a configurable threshold. Only available on some systems.
799 a thermal event interrupt occurs when a temperature threshold
800 has been exceeded for the CPU. This interrupt may also be generated
801 when the temperature drops back to normal.
804 a spurious interrupt is some interrupt that was raised then lowered
805 by some IO device before it could be fully processed by the APIC. Hence
806 the APIC sees the interrupt but does not know what device it came from.
807 For this case the APIC will generate the interrupt with a IRQ vector
808 of 0xff. This might also be generated by chipset bugs.
811 rescheduling, call and TLB flush interrupts are
812 sent from one CPU to another per the needs of the OS. Typically,
813 their statistics are used by kernel developers and interested users to
814 determine the occurrence of interrupts of the given type.
816 The above IRQ vectors are displayed only when relevant. For example,
817 the threshold vector does not exist on x86_64 platforms. Others are
818 suppressed when the system is a uniprocessor. As of this writing, only
819 i386 and x86_64 platforms support the new IRQ vector displays.
821 Of some interest is the introduction of the /proc/irq directory to 2.4.
822 It could be used to set IRQ to CPU affinity. This means that you can "hook" an
823 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
824 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
830 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
831 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
835 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
836 IRQ. You can set it by doing::
838 > echo 1 > /proc/irq/10/smp_affinity
840 This means that only the first CPU will handle the IRQ, but you can also echo
841 5 which means that only the first and third CPU can handle the IRQ.
843 The contents of each smp_affinity file is the same by default::
845 > cat /proc/irq/0/smp_affinity
848 There is an alternate interface, smp_affinity_list which allows specifying
849 a CPU range instead of a bitmask::
851 > cat /proc/irq/0/smp_affinity_list
854 The default_smp_affinity mask applies to all non-active IRQs, which are the
855 IRQs which have not yet been allocated/activated, and hence which lack a
856 /proc/irq/[0-9]* directory.
858 The node file on an SMP system shows the node to which the device using the IRQ
859 reports itself as being attached. This hardware locality information does not
860 include information about any possible driver locality preference.
862 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
863 profiler. Default value is ffffffff (all CPUs if there are only 32 of them).
865 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
866 between all the CPUs which are allowed to handle it. As usual the kernel has
867 more info than you and does a better job than you, so the defaults are the
868 best choice for almost everyone. [Note this applies only to those IO-APIC's
869 that support "Round Robin" interrupt distribution.]
871 There are three more important subdirectories in /proc: net, scsi, and sys.
872 The general rule is that the contents, or even the existence of these
873 directories, depend on your kernel configuration. If SCSI is not enabled, the
874 directory scsi may not exist. The same is true with the net, which is there
875 only when networking support is present in the running kernel.
877 The slabinfo file gives information about memory usage at the slab level.
878 Linux uses slab pools for memory management above page level in version 2.2.
879 Commonly used objects have their own slab pool (such as network buffers,
880 directory cache, and so on).
884 > cat /proc/buddyinfo
886 Node 0, zone DMA 0 4 5 4 4 3 ...
887 Node 0, zone Normal 1 0 0 1 101 8 ...
888 Node 0, zone HighMem 2 0 0 1 1 0 ...
890 External fragmentation is a problem under some workloads, and buddyinfo is a
891 useful tool for helping diagnose these problems. Buddyinfo will give you a
892 clue as to how big an area you can safely allocate, or why a previous
895 Each column represents the number of pages of a certain order which are
896 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
897 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
898 available in ZONE_NORMAL, etc...
900 More information relevant to external fragmentation can be found in
903 > cat /proc/pagetypeinfo
907 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
908 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
909 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
910 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
911 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
912 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
913 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
914 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
915 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
916 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
917 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
919 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
920 Node 0, zone DMA 2 0 5 1 0
921 Node 0, zone DMA32 41 6 967 2 0
923 Fragmentation avoidance in the kernel works by grouping pages of different
924 migrate types into the same contiguous regions of memory called page blocks.
925 A page block is typically the size of the default hugepage size, e.g. 2MB on
926 X86-64. By keeping pages grouped based on their ability to move, the kernel
927 can reclaim pages within a page block to satisfy a high-order allocation.
929 The pagetypinfo begins with information on the size of a page block. It
930 then gives the same type of information as buddyinfo except broken down
931 by migrate-type and finishes with details on how many page blocks of each
934 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
935 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
936 make an estimate of the likely number of huge pages that can be allocated
937 at a given point in time. All the "Movable" blocks should be allocatable
938 unless memory has been mlock()'d. Some of the Reclaimable blocks should
939 also be allocatable although a lot of filesystem metadata may have to be
940 reclaimed to achieve this.
946 Provides information about distribution and utilization of memory. This
947 varies by architecture and compile options. Some of the counters reported
948 here overlap. The memory reported by the non overlapping counters may not
949 add up to the overall memory usage and the difference for some workloads
950 can be substantial. In many cases there are other means to find out
951 additional memory using subsystem specific interfaces, for instance
952 /proc/net/sockstat for TCP memory allocations.
954 Example output. You may not have all of these fields.
960 MemTotal: 32858820 kB
962 MemAvailable: 27214312 kB
968 Active(anon): 94064 kB
969 Inactive(anon): 4570616 kB
970 Active(file): 3143088 kB
971 Inactive(file): 3015640 kB
980 AnonPages: 4654780 kB
983 KReclaimable: 517708 kB
985 SReclaimable: 517708 kB
986 SUnreclaim: 142336 kB
987 KernelStack: 11168 kB
993 CommitLimit: 16429408 kB
994 Committed_AS: 7715148 kB
995 VmallocTotal: 34359738367 kB
996 VmallocUsed: 40444 kB
999 EarlyMemtestBad: 0 kB
1000 HardwareCorrupted: 0 kB
1001 AnonHugePages: 4149248 kB
1002 ShmemHugePages: 0 kB
1003 ShmemPmdMapped: 0 kB
1012 Hugepagesize: 2048 kB
1014 DirectMap4k: 401152 kB
1015 DirectMap2M: 10008576 kB
1016 DirectMap1G: 24117248 kB
1019 Total usable RAM (i.e. physical RAM minus a few reserved
1020 bits and the kernel binary code)
1022 Total free RAM. On highmem systems, the sum of LowFree+HighFree
1024 An estimate of how much memory is available for starting new
1025 applications, without swapping. Calculated from MemFree,
1026 SReclaimable, the size of the file LRU lists, and the low
1027 watermarks in each zone.
1028 The estimate takes into account that the system needs some
1029 page cache to function well, and that not all reclaimable
1030 slab will be reclaimable, due to items being in use. The
1031 impact of those factors will vary from system to system.
1033 Relatively temporary storage for raw disk blocks
1034 shouldn't get tremendously large (20MB or so)
1036 In-memory cache for files read from the disk (the
1037 pagecache) as well as tmpfs & shmem.
1038 Doesn't include SwapCached.
1040 Memory that once was swapped out, is swapped back in but
1041 still also is in the swapfile (if memory is needed it
1042 doesn't need to be swapped out AGAIN because it is already
1043 in the swapfile. This saves I/O)
1045 Memory that has been used more recently and usually not
1046 reclaimed unless absolutely necessary.
1048 Memory which has been less recently used. It is more
1049 eligible to be reclaimed for other purposes
1051 Memory allocated for userspace which cannot be reclaimed, such
1052 as mlocked pages, ramfs backing pages, secret memfd pages etc.
1054 Memory locked with mlock().
1056 Highmem is all memory above ~860MB of physical memory.
1057 Highmem areas are for use by userspace programs, or
1058 for the pagecache. The kernel must use tricks to access
1059 this memory, making it slower to access than lowmem.
1061 Lowmem is memory which can be used for everything that
1062 highmem can be used for, but it is also available for the
1063 kernel's use for its own data structures. Among many
1064 other things, it is where everything from the Slab is
1065 allocated. Bad things happen when you're out of lowmem.
1067 total amount of swap space available
1069 Memory which has been evicted from RAM, and is temporarily
1072 Memory consumed by the zswap backend (compressed size)
1074 Amount of anonymous memory stored in zswap (original size)
1076 Memory which is waiting to get written back to the disk
1078 Memory which is actively being written back to the disk
1080 Non-file backed pages mapped into userspace page tables
1082 files which have been mmaped, such as libraries
1084 Total memory used by shared memory (shmem) and tmpfs
1086 Kernel allocations that the kernel will attempt to reclaim
1087 under memory pressure. Includes SReclaimable (below), and other
1088 direct allocations with a shrinker.
1090 in-kernel data structures cache
1092 Part of Slab, that might be reclaimed, such as caches
1094 Part of Slab, that cannot be reclaimed on memory pressure
1096 Memory consumed by the kernel stacks of all tasks
1098 Memory consumed by userspace page tables
1100 Memory consumed by secondary page tables, this currently
1101 currently includes KVM mmu allocations on x86 and arm64.
1103 Always zero. Previous counted pages which had been written to
1104 the server, but has not been committed to stable storage.
1106 Memory used for block device "bounce buffers"
1108 Memory used by FUSE for temporary writeback buffers
1110 Based on the overcommit ratio ('vm.overcommit_ratio'),
1111 this is the total amount of memory currently available to
1112 be allocated on the system. This limit is only adhered to
1113 if strict overcommit accounting is enabled (mode 2 in
1114 'vm.overcommit_memory').
1116 The CommitLimit is calculated with the following formula::
1118 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
1119 overcommit_ratio / 100 + [total swap pages]
1121 For example, on a system with 1G of physical RAM and 7G
1122 of swap with a `vm.overcommit_ratio` of 30 it would
1123 yield a CommitLimit of 7.3G.
1125 For more details, see the memory overcommit documentation
1126 in mm/overcommit-accounting.
1128 The amount of memory presently allocated on the system.
1129 The committed memory is a sum of all of the memory which
1130 has been allocated by processes, even if it has not been
1131 "used" by them as of yet. A process which malloc()'s 1G
1132 of memory, but only touches 300M of it will show up as
1133 using 1G. This 1G is memory which has been "committed" to
1134 by the VM and can be used at any time by the allocating
1135 application. With strict overcommit enabled on the system
1136 (mode 2 in 'vm.overcommit_memory'), allocations which would
1137 exceed the CommitLimit (detailed above) will not be permitted.
1138 This is useful if one needs to guarantee that processes will
1139 not fail due to lack of memory once that memory has been
1140 successfully allocated.
1142 total size of vmalloc virtual address space
1144 amount of vmalloc area which is used
1146 largest contiguous block of vmalloc area which is free
1148 Memory allocated to the percpu allocator used to back percpu
1149 allocations. This stat excludes the cost of metadata.
1151 The amount of RAM/memory in kB, that was identified as corrupted
1152 by early memtest. If memtest was not run, this field will not
1153 be displayed at all. Size is never rounded down to 0 kB.
1154 That means if 0 kB is reported, you can safely assume
1155 there was at least one pass of memtest and none of the passes
1156 found a single faulty byte of RAM.
1158 The amount of RAM/memory in KB, the kernel identifies as
1161 Non-file backed huge pages mapped into userspace page tables
1163 Memory used by shared memory (shmem) and tmpfs allocated
1166 Shared memory mapped into userspace with huge pages
1168 Memory used for filesystem data (page cache) allocated
1171 Page cache mapped into userspace with huge pages
1173 Memory reserved for the Contiguous Memory Allocator (CMA)
1175 Free remaining memory in the CMA reserves
1176 HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb
1177 See Documentation/admin-guide/mm/hugetlbpage.rst.
1178 DirectMap4k, DirectMap2M, DirectMap1G
1179 Breakdown of page table sizes used in the kernel's
1180 identity mapping of RAM
1185 Provides information about vmalloced/vmaped areas. One line per area,
1186 containing the virtual address range of the area, size in bytes,
1187 caller information of the creator, and optional information depending
1188 on the kind of area:
1190 ========== ===================================================
1191 pages=nr number of pages
1192 phys=addr if a physical address was specified
1193 ioremap I/O mapping (ioremap() and friends)
1194 vmalloc vmalloc() area
1196 user VM_USERMAP area
1197 vpages buffer for pages pointers was vmalloced (huge area)
1198 N<node>=nr (Only on NUMA kernels)
1199 Number of pages allocated on memory node <node>
1200 ========== ===================================================
1204 > cat /proc/vmallocinfo
1205 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
1206 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
1207 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
1208 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
1209 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
1210 phys=7fee8000 ioremap
1211 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
1212 phys=7fee7000 ioremap
1213 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
1214 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
1215 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
1216 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
1217 pages=2 vmalloc N1=2
1218 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
1219 /0x130 [x_tables] pages=4 vmalloc N0=4
1220 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
1221 pages=14 vmalloc N2=14
1222 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
1223 pages=4 vmalloc N1=4
1224 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
1225 pages=2 vmalloc N1=2
1226 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
1227 pages=10 vmalloc N0=10
1233 Provides counts of softirq handlers serviced since boot time, for each CPU.
1237 > cat /proc/softirqs
1240 TIMER: 27166 27120 27097 27034
1245 SCHED: 27035 26983 26971 26746
1247 RCU: 1678 1769 2178 2250
1249 1.3 Networking info in /proc/net
1250 --------------------------------
1252 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1253 additional values you get for IP version 6 if you configure the kernel to
1254 support this. Table 1-9 lists the files and their meaning.
1257 .. table:: Table 1-8: IPv6 info in /proc/net
1259 ========== =====================================================
1261 ========== =====================================================
1262 udp6 UDP sockets (IPv6)
1263 tcp6 TCP sockets (IPv6)
1264 raw6 Raw device statistics (IPv6)
1265 igmp6 IP multicast addresses, which this host joined (IPv6)
1266 if_inet6 List of IPv6 interface addresses
1267 ipv6_route Kernel routing table for IPv6
1268 rt6_stats Global IPv6 routing tables statistics
1269 sockstat6 Socket statistics (IPv6)
1270 snmp6 Snmp data (IPv6)
1271 ========== =====================================================
1273 .. table:: Table 1-9: Network info in /proc/net
1275 ============= ================================================================
1277 ============= ================================================================
1278 arp Kernel ARP table
1279 dev network devices with statistics
1280 dev_mcast the Layer2 multicast groups a device is listening too
1281 (interface index, label, number of references, number of bound
1283 dev_stat network device status
1284 ip_fwchains Firewall chain linkage
1285 ip_fwnames Firewall chain names
1286 ip_masq Directory containing the masquerading tables
1287 ip_masquerade Major masquerading table
1288 netstat Network statistics
1289 raw raw device statistics
1290 route Kernel routing table
1291 rpc Directory containing rpc info
1292 rt_cache Routing cache
1294 sockstat Socket statistics
1295 softnet_stat Per-CPU incoming packets queues statistics of online CPUs
1298 unix UNIX domain sockets
1299 wireless Wireless interface data (Wavelan etc)
1300 igmp IP multicast addresses, which this host joined
1301 psched Global packet scheduler parameters.
1302 netlink List of PF_NETLINK sockets
1303 ip_mr_vifs List of multicast virtual interfaces
1304 ip_mr_cache List of multicast routing cache
1305 ============= ================================================================
1307 You can use this information to see which network devices are available in
1308 your system and how much traffic was routed over those devices::
1311 Inter-|Receive |[...
1312 face |bytes packets errs drop fifo frame compressed multicast|[...
1313 lo: 908188 5596 0 0 0 0 0 0 [...
1314 ppp0:15475140 20721 410 0 0 410 0 0 [...
1315 eth0: 614530 7085 0 0 0 0 0 1 [...
1318 ...] bytes packets errs drop fifo colls carrier compressed
1319 ...] 908188 5596 0 0 0 0 0 0
1320 ...] 1375103 17405 0 0 0 0 0 0
1321 ...] 1703981 5535 0 0 0 3 0 0
1323 In addition, each Channel Bond interface has its own directory. For
1324 example, the bond0 device will have a directory called /proc/net/bond0/.
1325 It will contain information that is specific to that bond, such as the
1326 current slaves of the bond, the link status of the slaves, and how
1327 many times the slaves link has failed.
1332 If you have a SCSI or ATA host adapter in your system, you'll find a
1333 subdirectory named after the driver for this adapter in /proc/scsi.
1334 You'll also see a list of all recognized SCSI devices in /proc/scsi::
1336 >cat /proc/scsi/scsi
1338 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1339 Vendor: IBM Model: DGHS09U Rev: 03E0
1340 Type: Direct-Access ANSI SCSI revision: 03
1341 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1342 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1343 Type: CD-ROM ANSI SCSI revision: 02
1346 The directory named after the driver has one file for each adapter found in
1347 the system. These files contain information about the controller, including
1348 the used IRQ and the IO address range. The amount of information shown is
1349 dependent on the adapter you use. The example shows the output for an Adaptec
1350 AHA-2940 SCSI adapter::
1352 > cat /proc/scsi/aic7xxx/0
1354 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1356 TCQ Enabled By Default : Disabled
1357 AIC7XXX_PROC_STATS : Disabled
1358 AIC7XXX_RESET_DELAY : 5
1359 Adapter Configuration:
1360 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1361 Ultra Wide Controller
1362 PCI MMAPed I/O Base: 0xeb001000
1363 Adapter SEEPROM Config: SEEPROM found and used.
1364 Adaptec SCSI BIOS: Enabled
1366 SCBs: Active 0, Max Active 2,
1367 Allocated 15, HW 16, Page 255
1369 BIOS Control Word: 0x18b6
1370 Adapter Control Word: 0x005b
1371 Extended Translation: Enabled
1372 Disconnect Enable Flags: 0xffff
1373 Ultra Enable Flags: 0x0001
1374 Tag Queue Enable Flags: 0x0000
1375 Ordered Queue Tag Flags: 0x0000
1376 Default Tag Queue Depth: 8
1377 Tagged Queue By Device array for aic7xxx host instance 0:
1378 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1379 Actual queue depth per device for aic7xxx host instance 0:
1380 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1383 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1384 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1385 Total transfers 160151 (74577 reads and 85574 writes)
1387 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1388 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1389 Total transfers 0 (0 reads and 0 writes)
1392 1.5 Parallel port info in /proc/parport
1393 ---------------------------------------
1395 The directory /proc/parport contains information about the parallel ports of
1396 your system. It has one subdirectory for each port, named after the port
1399 These directories contain the four files shown in Table 1-10.
1402 .. table:: Table 1-10: Files in /proc/parport
1404 ========= ====================================================================
1406 ========= ====================================================================
1407 autoprobe Any IEEE-1284 device ID information that has been acquired.
1408 devices list of the device drivers using that port. A + will appear by the
1409 name of the device currently using the port (it might not appear
1411 hardware Parallel port's base address, IRQ line and DMA channel.
1412 irq IRQ that parport is using for that port. This is in a separate
1413 file to allow you to alter it by writing a new value in (IRQ
1415 ========= ====================================================================
1417 1.6 TTY info in /proc/tty
1418 -------------------------
1420 Information about the available and actually used tty's can be found in the
1421 directory /proc/tty. You'll find entries for drivers and line disciplines in
1422 this directory, as shown in Table 1-11.
1425 .. table:: Table 1-11: Files in /proc/tty
1427 ============= ==============================================
1429 ============= ==============================================
1430 drivers list of drivers and their usage
1431 ldiscs registered line disciplines
1432 driver/serial usage statistic and status of single tty lines
1433 ============= ==============================================
1435 To see which tty's are currently in use, you can simply look into the file
1438 > cat /proc/tty/drivers
1439 pty_slave /dev/pts 136 0-255 pty:slave
1440 pty_master /dev/ptm 128 0-255 pty:master
1441 pty_slave /dev/ttyp 3 0-255 pty:slave
1442 pty_master /dev/pty 2 0-255 pty:master
1443 serial /dev/cua 5 64-67 serial:callout
1444 serial /dev/ttyS 4 64-67 serial
1445 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1446 /dev/ptmx /dev/ptmx 5 2 system
1447 /dev/console /dev/console 5 1 system:console
1448 /dev/tty /dev/tty 5 0 system:/dev/tty
1449 unknown /dev/tty 4 1-63 console
1452 1.7 Miscellaneous kernel statistics in /proc/stat
1453 -------------------------------------------------
1455 Various pieces of information about kernel activity are available in the
1456 /proc/stat file. All of the numbers reported in this file are aggregates
1457 since the system first booted. For a quick look, simply cat the file::
1460 cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0
1461 cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0
1462 cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0
1463 cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0
1464 cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0
1465 intr 8688370575 8 3373 0 0 0 0 0 0 1 40791 0 0 353317 0 0 0 0 224789828 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 190974333 41958554 123983334 43 0 224593 0 0 0 <more 0's deleted>
1471 softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354
1473 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1474 lines. These numbers identify the amount of time the CPU has spent performing
1475 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1476 second). The meanings of the columns are as follows, from left to right:
1478 - user: normal processes executing in user mode
1479 - nice: niced processes executing in user mode
1480 - system: processes executing in kernel mode
1481 - idle: twiddling thumbs
1482 - iowait: In a word, iowait stands for waiting for I/O to complete. But there
1483 are several problems:
1485 1. CPU will not wait for I/O to complete, iowait is the time that a task is
1486 waiting for I/O to complete. When CPU goes into idle state for
1487 outstanding task I/O, another task will be scheduled on this CPU.
1488 2. In a multi-core CPU, the task waiting for I/O to complete is not running
1489 on any CPU, so the iowait of each CPU is difficult to calculate.
1490 3. The value of iowait field in /proc/stat will decrease in certain
1493 So, the iowait is not reliable by reading from /proc/stat.
1494 - irq: servicing interrupts
1495 - softirq: servicing softirqs
1496 - steal: involuntary wait
1497 - guest: running a normal guest
1498 - guest_nice: running a niced guest
1500 The "intr" line gives counts of interrupts serviced since boot time, for each
1501 of the possible system interrupts. The first column is the total of all
1502 interrupts serviced including unnumbered architecture specific interrupts;
1503 each subsequent column is the total for that particular numbered interrupt.
1504 Unnumbered interrupts are not shown, only summed into the total.
1506 The "ctxt" line gives the total number of context switches across all CPUs.
1508 The "btime" line gives the time at which the system booted, in seconds since
1511 The "processes" line gives the number of processes and threads created, which
1512 includes (but is not limited to) those created by calls to the fork() and
1513 clone() system calls.
1515 The "procs_running" line gives the total number of threads that are
1516 running or ready to run (i.e., the total number of runnable threads).
1518 The "procs_blocked" line gives the number of processes currently blocked,
1519 waiting for I/O to complete.
1521 The "softirq" line gives counts of softirqs serviced since boot time, for each
1522 of the possible system softirqs. The first column is the total of all
1523 softirqs serviced; each subsequent column is the total for that particular
1527 1.8 Ext4 file system parameters
1528 -------------------------------
1530 Information about mounted ext4 file systems can be found in
1531 /proc/fs/ext4. Each mounted filesystem will have a directory in
1532 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1533 /proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device
1534 directory are shown in Table 1-12, below.
1536 .. table:: Table 1-12: Files in /proc/fs/ext4/<devname>
1538 ============== ==========================================================
1540 mb_groups details of multiblock allocator buddy cache of free blocks
1541 ============== ==========================================================
1545 Shows registered system console lines.
1547 To see which character device lines are currently used for the system console
1548 /dev/console, you may simply look into the file /proc/consoles::
1550 > cat /proc/consoles
1556 +--------------------+-------------------------------------------------------+
1557 | device | name of the device |
1558 +====================+=======================================================+
1559 | operations | * R = can do read operations |
1560 | | * W = can do write operations |
1561 | | * U = can do unblank |
1562 +--------------------+-------------------------------------------------------+
1563 | flags | * E = it is enabled |
1564 | | * C = it is preferred console |
1565 | | * B = it is primary boot console |
1566 | | * p = it is used for printk buffer |
1567 | | * b = it is not a TTY but a Braille device |
1568 | | * a = it is safe to use when cpu is offline |
1569 +--------------------+-------------------------------------------------------+
1570 | major:minor | major and minor number of the device separated by a |
1572 +--------------------+-------------------------------------------------------+
1577 The /proc file system serves information about the running system. It not only
1578 allows access to process data but also allows you to request the kernel status
1579 by reading files in the hierarchy.
1581 The directory structure of /proc reflects the types of information and makes
1582 it easy, if not obvious, where to look for specific data.
1584 Chapter 2: Modifying System Parameters
1585 ======================================
1590 * Modifying kernel parameters by writing into files found in /proc/sys
1591 * Exploring the files which modify certain parameters
1592 * Review of the /proc/sys file tree
1594 ------------------------------------------------------------------------------
1596 A very interesting part of /proc is the directory /proc/sys. This is not only
1597 a source of information, it also allows you to change parameters within the
1598 kernel. Be very careful when attempting this. You can optimize your system,
1599 but you can also cause it to crash. Never alter kernel parameters on a
1600 production system. Set up a development machine and test to make sure that
1601 everything works the way you want it to. You may have no alternative but to
1602 reboot the machine once an error has been made.
1604 To change a value, simply echo the new value into the file.
1605 You need to be root to do this. You can create your own boot script
1606 to perform this every time your system boots.
1608 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1609 general things in the operation of the Linux kernel. Since some of the files
1610 can inadvertently disrupt your system, it is advisable to read both
1611 documentation and source before actually making adjustments. In any case, be
1612 very careful when writing to any of these files. The entries in /proc may
1613 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1614 review the kernel documentation in the directory linux/Documentation.
1615 This chapter is heavily based on the documentation included in the pre 2.2
1616 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1618 Please see: Documentation/admin-guide/sysctl/ directory for descriptions of
1624 Certain aspects of kernel behavior can be modified at runtime, without the
1625 need to recompile the kernel, or even to reboot the system. The files in the
1626 /proc/sys tree can not only be read, but also modified. You can use the echo
1627 command to write value into these files, thereby changing the default settings
1631 Chapter 3: Per-process Parameters
1632 =================================
1634 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1635 --------------------------------------------------------------------------------
1637 These files can be used to adjust the badness heuristic used to select which
1638 process gets killed in out of memory (oom) conditions.
1640 The badness heuristic assigns a value to each candidate task ranging from 0
1641 (never kill) to 1000 (always kill) to determine which process is targeted. The
1642 units are roughly a proportion along that range of allowed memory the process
1643 may allocate from based on an estimation of its current memory and swap use.
1644 For example, if a task is using all allowed memory, its badness score will be
1645 1000. If it is using half of its allowed memory, its score will be 500.
1647 The amount of "allowed" memory depends on the context in which the oom killer
1648 was called. If it is due to the memory assigned to the allocating task's cpuset
1649 being exhausted, the allowed memory represents the set of mems assigned to that
1650 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1651 memory represents the set of mempolicy nodes. If it is due to a memory
1652 limit (or swap limit) being reached, the allowed memory is that configured
1653 limit. Finally, if it is due to the entire system being out of memory, the
1654 allowed memory represents all allocatable resources.
1656 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1657 is used to determine which task to kill. Acceptable values range from -1000
1658 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1659 polarize the preference for oom killing either by always preferring a certain
1660 task or completely disabling it. The lowest possible value, -1000, is
1661 equivalent to disabling oom killing entirely for that task since it will always
1662 report a badness score of 0.
1664 Consequently, it is very simple for userspace to define the amount of memory to
1665 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1666 example, is roughly equivalent to allowing the remainder of tasks sharing the
1667 same system, cpuset, mempolicy, or memory controller resources to use at least
1668 50% more memory. A value of -500, on the other hand, would be roughly
1669 equivalent to discounting 50% of the task's allowed memory from being considered
1670 as scoring against the task.
1672 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1673 be used to tune the badness score. Its acceptable values range from -16
1674 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1675 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1676 scaled linearly with /proc/<pid>/oom_score_adj.
1678 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1679 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1680 requires CAP_SYS_RESOURCE.
1683 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1684 -------------------------------------------------------------
1686 This file can be used to check the current score used by the oom-killer for
1687 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1688 process should be killed in an out-of-memory situation.
1690 Please note that the exported value includes oom_score_adj so it is
1691 effectively in range [0,2000].
1694 3.3 /proc/<pid>/io - Display the IO accounting fields
1695 -------------------------------------------------------
1697 This file contains IO statistics for each running process.
1704 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1707 test:/tmp # cat /proc/3828/io
1713 write_bytes: 323932160
1714 cancelled_write_bytes: 0
1723 I/O counter: chars read
1724 The number of bytes which this task has caused to be read from storage. This
1725 is simply the sum of bytes which this process passed to read() and pread().
1726 It includes things like tty IO and it is unaffected by whether or not actual
1727 physical disk IO was required (the read might have been satisfied from
1734 I/O counter: chars written
1735 The number of bytes which this task has caused, or shall cause to be written
1736 to disk. Similar caveats apply here as with rchar.
1742 I/O counter: read syscalls
1743 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1750 I/O counter: write syscalls
1751 Attempt to count the number of write I/O operations, i.e. syscalls like
1752 write() and pwrite().
1758 I/O counter: bytes read
1759 Attempt to count the number of bytes which this process really did cause to
1760 be fetched from the storage layer. Done at the submit_bio() level, so it is
1761 accurate for block-backed filesystems. <please add status regarding NFS and
1762 CIFS at a later time>
1768 I/O counter: bytes written
1769 Attempt to count the number of bytes which this process caused to be sent to
1770 the storage layer. This is done at page-dirtying time.
1773 cancelled_write_bytes
1774 ^^^^^^^^^^^^^^^^^^^^^
1776 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1777 then deletes the file, it will in fact perform no writeout. But it will have
1778 been accounted as having caused 1MB of write.
1779 In other words: The number of bytes which this process caused to not happen,
1780 by truncating pagecache. A task can cause "negative" IO too. If this task
1781 truncates some dirty pagecache, some IO which another task has been accounted
1782 for (in its write_bytes) will not be happening. We _could_ just subtract that
1783 from the truncating task's write_bytes, but there is information loss in doing
1789 At its current implementation state, this is a bit racy on 32-bit machines:
1790 if process A reads process B's /proc/pid/io while process B is updating one
1791 of those 64-bit counters, process A could see an intermediate result.
1794 More information about this can be found within the taskstats documentation in
1795 Documentation/accounting.
1797 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1798 ---------------------------------------------------------------
1799 When a process is dumped, all anonymous memory is written to a core file as
1800 long as the size of the core file isn't limited. But sometimes we don't want
1801 to dump some memory segments, for example, huge shared memory or DAX.
1802 Conversely, sometimes we want to save file-backed memory segments into a core
1803 file, not only the individual files.
1805 /proc/<pid>/coredump_filter allows you to customize which memory segments
1806 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1807 of memory types. If a bit of the bitmask is set, memory segments of the
1808 corresponding memory type are dumped, otherwise they are not dumped.
1810 The following 9 memory types are supported:
1812 - (bit 0) anonymous private memory
1813 - (bit 1) anonymous shared memory
1814 - (bit 2) file-backed private memory
1815 - (bit 3) file-backed shared memory
1816 - (bit 4) ELF header pages in file-backed private memory areas (it is
1817 effective only if the bit 2 is cleared)
1818 - (bit 5) hugetlb private memory
1819 - (bit 6) hugetlb shared memory
1820 - (bit 7) DAX private memory
1821 - (bit 8) DAX shared memory
1823 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1824 are always dumped regardless of the bitmask status.
1826 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1827 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1829 The default value of coredump_filter is 0x33; this means all anonymous memory
1830 segments, ELF header pages and hugetlb private memory are dumped.
1832 If you don't want to dump all shared memory segments attached to pid 1234,
1833 write 0x31 to the process's proc file::
1835 $ echo 0x31 > /proc/1234/coredump_filter
1837 When a new process is created, the process inherits the bitmask status from its
1838 parent. It is useful to set up coredump_filter before the program runs.
1841 $ echo 0x7 > /proc/self/coredump_filter
1844 3.5 /proc/<pid>/mountinfo - Information about mounts
1845 --------------------------------------------------------
1847 This file contains lines of the form::
1849 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1850 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4)
1852 (1) mount ID: unique identifier of the mount (may be reused after umount)
1853 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1854 (3) major:minor: value of st_dev for files on filesystem
1855 (4) root: root of the mount within the filesystem
1856 (5) mount point: mount point relative to the process's root
1857 (6) mount options: per mount options
1858 (n…m) optional fields: zero or more fields of the form "tag[:value]"
1859 (m+1) separator: marks the end of the optional fields
1860 (m+2) filesystem type: name of filesystem of the form "type[.subtype]"
1861 (m+3) mount source: filesystem specific information or "none"
1862 (m+4) super options: per super block options
1864 Parsers should ignore all unrecognised optional fields. Currently the
1865 possible optional fields are:
1867 ================ ==============================================================
1868 shared:X mount is shared in peer group X
1869 master:X mount is slave to peer group X
1870 propagate_from:X mount is slave and receives propagation from peer group X [#]_
1871 unbindable mount is unbindable
1872 ================ ==============================================================
1874 .. [#] X is the closest dominant peer group under the process's root. If
1875 X is the immediate master of the mount, or if there's no dominant peer
1876 group under the same root, then only the "master:X" field is present
1877 and not the "propagate_from:X" field.
1879 For more information on mount propagation see:
1881 Documentation/filesystems/sharedsubtree.rst
1884 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1885 --------------------------------------------------------
1886 These files provide a method to access a task's comm value. It also allows for
1887 a task to set its own or one of its thread siblings comm value. The comm value
1888 is limited in size compared to the cmdline value, so writing anything longer
1889 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1893 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1894 -------------------------------------------------------------------------
1895 This file provides a fast way to retrieve first level children pids
1896 of a task pointed by <pid>/<tid> pair. The format is a space separated
1899 Note the "first level" here -- if a child has its own children they will
1900 not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children
1901 to obtain the descendants.
1903 Since this interface is intended to be fast and cheap it doesn't
1904 guarantee to provide precise results and some children might be
1905 skipped, especially if they've exited right after we printed their
1906 pids, so one needs to either stop or freeze processes being inspected
1907 if precise results are needed.
1910 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1911 ---------------------------------------------------------------
1912 This file provides information associated with an opened file. The regular
1913 files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'.
1914 The 'pos' represents the current offset of the opened file in decimal
1915 form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the
1916 file has been created with [see open(2) for details] and 'mnt_id' represents
1917 mount ID of the file system containing the opened file [see 3.5
1918 /proc/<pid>/mountinfo for details]. 'ino' represents the inode number of
1921 A typical output is::
1928 All locks associated with a file descriptor are shown in its fdinfo too::
1930 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1932 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1933 pair provide additional information particular to the objects they represent.
1946 where 'eventfd-count' is hex value of a counter.
1957 sigmask: 0000000000000200
1959 where 'sigmask' is hex value of the signal mask associated
1971 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7
1973 where 'tfd' is a target file descriptor number in decimal form,
1974 'events' is events mask being watched and the 'data' is data
1975 associated with a target [see epoll(7) for more details].
1977 The 'pos' is current offset of the target file in decimal form
1978 [see lseek(2)], 'ino' and 'sdev' are inode and device numbers
1979 where target file resides, all in hex format.
1983 For inotify files the format is the following::
1989 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1991 where 'wd' is a watch descriptor in decimal form, i.e. a target file
1992 descriptor number, 'ino' and 'sdev' are inode and device where the
1993 target file resides and the 'mask' is the mask of events, all in hex
1994 form [see inotify(7) for more details].
1996 If the kernel was built with exportfs support, the path to the target
1997 file is encoded as a file handle. The file handle is provided by three
1998 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
2001 If the kernel is built without exportfs support the file handle won't be
2004 If there is no inotify mark attached yet the 'inotify' line will be omitted.
2006 For fanotify files the format is::
2012 fanotify flags:10 event-flags:0
2013 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
2014 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
2016 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
2017 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
2018 flags associated with mark which are tracked separately from events
2019 mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events
2020 mask and 'ignored_mask' is the mask of events which are to be ignored.
2021 All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
2022 provide information about flags and mask used in fanotify_mark
2023 call [see fsnotify manpage for details].
2025 While the first three lines are mandatory and always printed, the rest is
2026 optional and may be omitted if no marks created yet.
2040 it_value: (0, 49406829)
2043 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
2044 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
2045 flags in octal form been used to setup the timer [see timerfd_settime(2) for
2046 details]. 'it_value' is remaining time until the timer expiration.
2047 'it_interval' is the interval for the timer. Note the timer might be set up
2048 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
2049 still exhibits timer's remaining time.
2062 exp_name: system-heap
2064 where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of
2065 the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter.
2067 3.9 /proc/<pid>/map_files - Information about memory mapped files
2068 ---------------------------------------------------------------------
2069 This directory contains symbolic links which represent memory mapped files
2070 the process is maintaining. Example output::
2072 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
2073 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
2074 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
2076 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
2077 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
2079 The name of a link represents the virtual memory bounds of a mapping, i.e.
2080 vm_area_struct::vm_start-vm_area_struct::vm_end.
2082 The main purpose of the map_files is to retrieve a set of memory mapped
2083 files in a fast way instead of parsing /proc/<pid>/maps or
2084 /proc/<pid>/smaps, both of which contain many more records. At the same
2085 time one can open(2) mappings from the listings of two processes and
2086 comparing their inode numbers to figure out which anonymous memory areas
2087 are actually shared.
2089 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
2090 ---------------------------------------------------------
2091 This file provides the value of the task's timerslack value in nanoseconds.
2092 This value specifies an amount of time that normal timers may be deferred
2093 in order to coalesce timers and avoid unnecessary wakeups.
2095 This allows a task's interactivity vs power consumption tradeoff to be
2098 Writing 0 to the file will set the task's timerslack to the default value.
2100 Valid values are from 0 - ULLONG_MAX
2102 An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
2103 permissions on the task specified to change its timerslack_ns value.
2105 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
2106 -----------------------------------------------------------------
2107 When CONFIG_LIVEPATCH is enabled, this file displays the value of the
2108 patch state for the task.
2110 A value of '-1' indicates that no patch is in transition.
2112 A value of '0' indicates that a patch is in transition and the task is
2113 unpatched. If the patch is being enabled, then the task hasn't been
2114 patched yet. If the patch is being disabled, then the task has already
2117 A value of '1' indicates that a patch is in transition and the task is
2118 patched. If the patch is being enabled, then the task has already been
2119 patched. If the patch is being disabled, then the task hasn't been
2122 3.12 /proc/<pid>/arch_status - task architecture specific status
2123 -------------------------------------------------------------------
2124 When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the
2125 architecture specific status of the task.
2132 $ cat /proc/6753/arch_status
2133 AVX512_elapsed_ms: 8
2138 x86 specific entries
2139 ~~~~~~~~~~~~~~~~~~~~~
2144 If AVX512 is supported on the machine, this entry shows the milliseconds
2145 elapsed since the last time AVX512 usage was recorded. The recording
2146 happens on a best effort basis when a task is scheduled out. This means
2147 that the value depends on two factors:
2149 1) The time which the task spent on the CPU without being scheduled
2150 out. With CPU isolation and a single runnable task this can take
2153 2) The time since the task was scheduled out last. Depending on the
2154 reason for being scheduled out (time slice exhausted, syscall ...)
2155 this can be arbitrary long time.
2157 As a consequence the value cannot be considered precise and authoritative
2158 information. The application which uses this information has to be aware
2159 of the overall scenario on the system in order to determine whether a
2160 task is a real AVX512 user or not. Precise information can be obtained
2161 with performance counters.
2163 A special value of '-1' indicates that no AVX512 usage was recorded, thus
2164 the task is unlikely an AVX512 user, but depends on the workload and the
2165 scheduling scenario, it also could be a false negative mentioned above.
2167 3.13 /proc/<pid>/fd - List of symlinks to open files
2168 -------------------------------------------------------
2169 This directory contains symbolic links which represent open files
2170 the process is maintaining. Example output::
2172 lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null
2173 l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null
2174 lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]'
2175 lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]'
2176 lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]'
2178 The number of open files for the process is stored in 'size' member
2179 of stat() output for /proc/<pid>/fd for fast access.
2180 -------------------------------------------------------
2183 Chapter 4: Configuring procfs
2184 =============================
2187 ---------------------
2189 The following mount options are supported:
2191 ========= ========================================================
2192 hidepid= Set /proc/<pid>/ access mode.
2193 gid= Set the group authorized to learn processes information.
2194 subset= Show only the specified subset of procfs.
2195 ========= ========================================================
2197 hidepid=off or hidepid=0 means classic mode - everybody may access all
2198 /proc/<pid>/ directories (default).
2200 hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/
2201 directories but their own. Sensitive files like cmdline, sched*, status are now
2202 protected against other users. This makes it impossible to learn whether any
2203 user runs specific program (given the program doesn't reveal itself by its
2204 behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for
2205 other users, poorly written programs passing sensitive information via program
2206 arguments are now protected against local eavesdroppers.
2208 hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be
2209 fully invisible to other users. It doesn't mean that it hides a fact whether a
2210 process with a specific pid value exists (it can be learned by other means, e.g.
2211 by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by
2212 stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of
2213 gathering information about running processes, whether some daemon runs with
2214 elevated privileges, whether other user runs some sensitive program, whether
2215 other users run any program at all, etc.
2217 hidepid=ptraceable or hidepid=4 means that procfs should only contain
2218 /proc/<pid>/ directories that the caller can ptrace.
2220 gid= defines a group authorized to learn processes information otherwise
2221 prohibited by hidepid=. If you use some daemon like identd which needs to learn
2222 information about processes information, just add identd to this group.
2224 subset=pid hides all top level files and directories in the procfs that
2225 are not related to tasks.
2227 Chapter 5: Filesystem behavior
2228 ==============================
2230 Originally, before the advent of pid namepsace, procfs was a global file
2231 system. It means that there was only one procfs instance in the system.
2233 When pid namespace was added, a separate procfs instance was mounted in
2234 each pid namespace. So, procfs mount options are global among all
2235 mountpoints within the same namespace::
2237 # grep ^proc /proc/mounts
2238 proc /proc proc rw,relatime,hidepid=2 0 0
2240 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc
2241 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0
2242 +++ exited with 0 +++
2244 # grep ^proc /proc/mounts
2245 proc /proc proc rw,relatime,hidepid=2 0 0
2246 proc /tmp/proc proc rw,relatime,hidepid=2 0 0
2248 and only after remounting procfs mount options will change at all
2251 # mount -o remount,hidepid=1 -t proc proc /tmp/proc
2253 # grep ^proc /proc/mounts
2254 proc /proc proc rw,relatime,hidepid=1 0 0
2255 proc /tmp/proc proc rw,relatime,hidepid=1 0 0
2257 This behavior is different from the behavior of other filesystems.
2259 The new procfs behavior is more like other filesystems. Each procfs mount
2260 creates a new procfs instance. Mount options affect own procfs instance.
2261 It means that it became possible to have several procfs instances
2262 displaying tasks with different filtering options in one pid namespace::
2264 # mount -o hidepid=invisible -t proc proc /proc
2265 # mount -o hidepid=noaccess -t proc proc /tmp/proc
2266 # grep ^proc /proc/mounts
2267 proc /proc proc rw,relatime,hidepid=invisible 0 0
2268 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0