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
202 SigPnd: 0000000000000000
203 ShdPnd: 0000000000000000
204 SigBlk: 0000000000000000
205 SigIgn: 0000000000000000
206 SigCgt: 0000000000000000
207 CapInh: 00000000fffffeff
208 CapPrm: 0000000000000000
209 CapEff: 0000000000000000
210 CapBnd: ffffffffffffffff
211 CapAmb: 0000000000000000
214 Speculation_Store_Bypass: thread vulnerable
215 SpeculationIndirectBranch: conditional enabled
216 voluntary_ctxt_switches: 0
217 nonvoluntary_ctxt_switches: 1
219 This shows you nearly the same information you would get if you viewed it with
220 the ps command. In fact, ps uses the proc file system to obtain its
221 information. But you get a more detailed view of the process by reading the
222 file /proc/PID/status. It fields are described in table 1-2.
224 The statm file contains more detailed information about the process
225 memory usage. Its seven fields are explained in Table 1-3. The stat file
226 contains detailed information about the process itself. Its fields are
227 explained in Table 1-4.
229 (for SMP CONFIG users)
231 For making accounting scalable, RSS related information are handled in an
232 asynchronous manner and the value may not be very precise. To see a precise
233 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
234 It's slow but very precise.
236 .. table:: Table 1-2: Contents of the status fields (as of 4.19)
238 ========================== ===================================================
240 ========================== ===================================================
241 Name filename of the executable
242 Umask file mode creation mask
243 State state (R is running, S is sleeping, D is sleeping
244 in an uninterruptible wait, Z is zombie,
245 T is traced or stopped)
247 Ngid NUMA group ID (0 if none)
249 PPid process id of the parent process
250 TracerPid PID of process tracing this process (0 if not, or
251 the tracer is outside of the current pid namespace)
252 Uid Real, effective, saved set, and file system UIDs
253 Gid Real, effective, saved set, and file system GIDs
254 FDSize number of file descriptor slots currently allocated
255 Groups supplementary group list
256 NStgid descendant namespace thread group ID hierarchy
257 NSpid descendant namespace process ID hierarchy
258 NSpgid descendant namespace process group ID hierarchy
259 NSsid descendant namespace session ID hierarchy
260 Kthread kernel thread flag, 1 is yes, 0 is no
261 VmPeak peak virtual memory size
262 VmSize total program size
263 VmLck locked memory size
264 VmPin pinned memory size
265 VmHWM peak resident set size ("high water mark")
266 VmRSS size of memory portions. It contains the three
268 (VmRSS = RssAnon + RssFile + RssShmem)
269 RssAnon size of resident anonymous memory
270 RssFile size of resident file mappings
271 RssShmem size of resident shmem memory (includes SysV shm,
272 mapping of tmpfs and shared anonymous mappings)
273 VmData size of private data segments
274 VmStk size of stack segments
275 VmExe size of text segment
276 VmLib size of shared library code
277 VmPTE size of page table entries
278 VmSwap amount of swap used by anonymous private data
279 (shmem swap usage is not included)
280 HugetlbPages size of hugetlb memory portions
281 CoreDumping process's memory is currently being dumped
282 (killing the process may lead to a corrupted core)
283 THP_enabled process is allowed to use THP (returns 0 when
284 PR_SET_THP_DISABLE is set on the process
285 Threads number of threads
286 SigQ number of signals queued/max. number for queue
287 SigPnd bitmap of pending signals for the thread
288 ShdPnd bitmap of shared pending signals for the process
289 SigBlk bitmap of blocked signals
290 SigIgn bitmap of ignored signals
291 SigCgt bitmap of caught signals
292 CapInh bitmap of inheritable capabilities
293 CapPrm bitmap of permitted capabilities
294 CapEff bitmap of effective capabilities
295 CapBnd bitmap of capabilities bounding set
296 CapAmb bitmap of ambient capabilities
297 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
298 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
299 Speculation_Store_Bypass speculative store bypass mitigation status
300 SpeculationIndirectBranch indirect branch speculation mode
301 Cpus_allowed mask of CPUs on which this process may run
302 Cpus_allowed_list Same as previous, but in "list format"
303 Mems_allowed mask of memory nodes allowed to this process
304 Mems_allowed_list Same as previous, but in "list format"
305 voluntary_ctxt_switches number of voluntary context switches
306 nonvoluntary_ctxt_switches number of non voluntary context switches
307 ========================== ===================================================
310 .. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3)
312 ======== =============================== ==============================
314 ======== =============================== ==============================
315 size total program size (pages) (same as VmSize in status)
316 resident size of memory portions (pages) (same as VmRSS in status)
317 shared number of pages that are shared (i.e. backed by a file, same
318 as RssFile+RssShmem in status)
319 trs number of pages that are 'code' (not including libs; broken,
320 includes data segment)
321 lrs number of pages of library (always 0 on 2.6)
322 drs number of pages of data/stack (including libs; broken,
323 includes library text)
324 dt number of dirty pages (always 0 on 2.6)
325 ======== =============================== ==============================
328 .. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7)
330 ============= ===============================================================
332 ============= ===============================================================
334 tcomm filename of the executable
335 state state (R is running, S is sleeping, D is sleeping in an
336 uninterruptible wait, Z is zombie, T is traced or stopped)
337 ppid process id of the parent process
338 pgrp pgrp of the process
340 tty_nr tty the process uses
341 tty_pgrp pgrp of the tty
343 min_flt number of minor faults
344 cmin_flt number of minor faults with child's
345 maj_flt number of major faults
346 cmaj_flt number of major faults with child's
347 utime user mode jiffies
348 stime kernel mode jiffies
349 cutime user mode jiffies with child's
350 cstime kernel mode jiffies with child's
351 priority priority level
353 num_threads number of threads
354 it_real_value (obsolete, always 0)
355 start_time time the process started after system boot
356 vsize virtual memory size
357 rss resident set memory size
358 rsslim current limit in bytes on the rss
359 start_code address above which program text can run
360 end_code address below which program text can run
361 start_stack address of the start of the main process stack
362 esp current value of ESP
363 eip current value of EIP
364 pending bitmap of pending signals
365 blocked bitmap of blocked signals
366 sigign bitmap of ignored signals
367 sigcatch bitmap of caught signals
368 0 (place holder, used to be the wchan address,
369 use /proc/PID/wchan instead)
372 exit_signal signal to send to parent thread on exit
373 task_cpu which CPU the task is scheduled on
374 rt_priority realtime priority
375 policy scheduling policy (man sched_setscheduler)
376 blkio_ticks time spent waiting for block IO
377 gtime guest time of the task in jiffies
378 cgtime guest time of the task children in jiffies
379 start_data address above which program data+bss is placed
380 end_data address below which program data+bss is placed
381 start_brk address above which program heap can be expanded with brk()
382 arg_start address above which program command line is placed
383 arg_end address below which program command line is placed
384 env_start address above which program environment is placed
385 env_end address below which program environment is placed
386 exit_code the thread's exit_code in the form reported by the waitpid
388 ============= ===============================================================
390 The /proc/PID/maps file contains the currently mapped memory regions and
391 their access permissions.
395 address perms offset dev inode pathname
397 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
398 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
399 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
400 a7cb1000-a7cb2000 ---p 00000000 00:00 0
401 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
402 a7eb2000-a7eb3000 ---p 00000000 00:00 0
403 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
404 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
405 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
406 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
407 a800b000-a800e000 rw-p 00000000 00:00 0
408 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
409 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
410 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
411 a8024000-a8027000 rw-p 00000000 00:00 0
412 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
413 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
414 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
415 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
416 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
418 where "address" is the address space in the process that it occupies, "perms"
419 is a set of permissions::
425 p = private (copy on write)
427 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
428 "inode" is the inode on that device. 0 indicates that no inode is associated
429 with the memory region, as the case would be with BSS (uninitialized data).
430 The "pathname" shows the name associated file for this mapping. If the mapping
431 is not associated with a file:
433 =================== ===========================================
434 [heap] the heap of the program
435 [stack] the stack of the main process
436 [vdso] the "virtual dynamic shared object",
437 the kernel system call handler
438 [anon:<name>] a private anonymous mapping that has been
440 [anon_shmem:<name>] an anonymous shared memory mapping that has
441 been named by userspace
442 =================== ===========================================
444 or if empty, the mapping is anonymous.
446 The /proc/PID/smaps is an extension based on maps, showing the memory
447 consumption for each of the process's mappings. For each mapping (aka Virtual
448 Memory Area, or VMA) there is a series of lines such as the following::
450 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
469 Private_Hugetlb: 0 kB
476 VmFlags: rd ex mr mw me dw
478 The first of these lines shows the same information as is displayed for the
479 mapping in /proc/PID/maps. Following lines show the size of the mapping
480 (size); the size of each page allocated when backing a VMA (KernelPageSize),
481 which is usually the same as the size in the page table entries; the page size
482 used by the MMU when backing a VMA (in most cases, the same as KernelPageSize);
483 the amount of the mapping that is currently resident in RAM (RSS); the
484 process' proportional share of this mapping (PSS); and the number of clean and
485 dirty shared and private pages in the mapping.
487 The "proportional set size" (PSS) of a process is the count of pages it has
488 in memory, where each page is divided by the number of processes sharing it.
489 So if a process has 1000 pages all to itself, and 1000 shared with one other
490 process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which
491 consists of dirty pages. ("Pss_Clean" is not included, but it can be
492 calculated by subtracting "Pss_Dirty" from "Pss".)
494 Note that even a page which is part of a MAP_SHARED mapping, but has only
495 a single pte mapped, i.e. is currently used by only one process, is accounted
496 as private and not as shared.
498 "Referenced" indicates the amount of memory currently marked as referenced or
501 "Anonymous" shows the amount of memory that does not belong to any file. Even
502 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
503 and a page is modified, the file page is replaced by a private anonymous copy.
505 "KSM" reports how many of the pages are KSM pages. Note that KSM-placed zeropages
506 are not included, only actual KSM pages.
508 "LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
509 The memory isn't freed immediately with madvise(). It's freed in memory
510 pressure if the memory is clean. Please note that the printed value might
511 be lower than the real value due to optimizations used in the current
512 implementation. If this is not desirable please file a bug report.
514 "AnonHugePages" shows the amount of memory backed by transparent hugepage.
516 "ShmemPmdMapped" shows the amount of shared (shmem/tmpfs) memory backed by
519 "Shared_Hugetlb" and "Private_Hugetlb" show the amounts of memory backed by
520 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
521 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
523 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
525 For shmem mappings, "Swap" includes also the size of the mapped (and not
526 replaced by copy-on-write) part of the underlying shmem object out on swap.
527 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
528 does not take into account swapped out page of underlying shmem objects.
529 "Locked" indicates whether the mapping is locked in memory or not.
531 "THPeligible" indicates whether the mapping is eligible for allocating THP
532 pages as well as the THP is PMD mappable or not - 1 if true, 0 otherwise.
533 It just shows the current status.
535 "VmFlags" field deserves a separate description. This member represents the
536 kernel flags associated with the particular virtual memory area in two letter
537 encoded manner. The codes are the following:
539 == =======================================
548 gd stack segment growns down
550 dw disabled write to the mapped file
551 lo pages are locked in memory
552 io memory mapped I/O area
553 sr sequential read advise provided
554 rr random read advise provided
555 dc do not copy area on fork
556 de do not expand area on remapping
557 ac area is accountable
558 nr swap space is not reserved for the area
559 ht area uses huge tlb pages
560 sf synchronous page fault
561 ar architecture specific flag
563 dd do not include area into core dump
566 hg huge page advise flag
567 nh no huge page advise flag
568 mg mergeable advise flag
569 bt arm64 BTI guarded page
570 mt arm64 MTE allocation tags are enabled
571 um userfaultfd missing tracking
572 uw userfaultfd wr-protect tracking
574 == =======================================
576 Note that there is no guarantee that every flag and associated mnemonic will
577 be present in all further kernel releases. Things get changed, the flags may
578 be vanished or the reverse -- new added. Interpretation of their meaning
579 might change in future as well. So each consumer of these flags has to
580 follow each specific kernel version for the exact semantic.
582 This file is only present if the CONFIG_MMU kernel configuration option is
585 Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
586 output can be achieved only in the single read call).
588 This typically manifests when doing partial reads of these files while the
589 memory map is being modified. Despite the races, we do provide the following
592 1) The mapped addresses never go backwards, which implies no two
593 regions will ever overlap.
594 2) If there is something at a given vaddr during the entirety of the
595 life of the smaps/maps walk, there will be some output for it.
597 The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps,
598 but their values are the sums of the corresponding values for all mappings of
599 the process. Additionally, it contains these fields:
605 They represent the proportional shares of anonymous, file, and shmem pages, as
606 described for smaps above. These fields are omitted in smaps since each
607 mapping identifies the type (anon, file, or shmem) of all pages it contains.
608 Thus all information in smaps_rollup can be derived from smaps, but at a
609 significantly higher cost.
611 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
612 bits on both physical and virtual pages associated with a process, and the
613 soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst
615 To clear the bits for all the pages associated with the process::
617 > echo 1 > /proc/PID/clear_refs
619 To clear the bits for the anonymous pages associated with the process::
621 > echo 2 > /proc/PID/clear_refs
623 To clear the bits for the file mapped pages associated with the process::
625 > echo 3 > /proc/PID/clear_refs
627 To clear the soft-dirty bit::
629 > echo 4 > /proc/PID/clear_refs
631 To reset the peak resident set size ("high water mark") to the process's
634 > echo 5 > /proc/PID/clear_refs
636 Any other value written to /proc/PID/clear_refs will have no effect.
638 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
639 using /proc/kpageflags and number of times a page is mapped using
640 /proc/kpagecount. For detailed explanation, see
641 Documentation/admin-guide/mm/pagemap.rst.
643 The /proc/pid/numa_maps is an extension based on maps, showing the memory
644 locality and binding policy, as well as the memory usage (in pages) of
645 each mapping. The output follows a general format where mapping details get
646 summarized separated by blank spaces, one mapping per each file line::
648 address policy mapping details
650 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
651 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
652 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
653 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
654 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
655 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
656 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
657 320698b000 default file=/lib64/libc-2.12.so
658 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
659 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
660 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
661 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
662 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
663 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
664 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
665 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
669 "address" is the starting address for the mapping;
671 "policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst);
673 "mapping details" summarizes mapping data such as mapping type, page usage counters,
674 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
675 size, in KB, that is backing the mapping up.
680 Similar to the process entries, the kernel data files give information about
681 the running kernel. The files used to obtain this information are contained in
682 /proc and are listed in Table 1-5. Not all of these will be present in your
683 system. It depends on the kernel configuration and the loaded modules, which
684 files are there, and which are missing.
686 .. table:: Table 1-5: Kernel info in /proc
688 ============ ===============================================================
690 ============ ===============================================================
691 apm Advanced power management info
692 bootconfig Kernel command line obtained from boot config,
693 and, if there were kernel parameters from the
694 boot loader, a "# Parameters from bootloader:"
695 line followed by a line containing those
696 parameters prefixed by "# ". (5.5)
697 buddyinfo Kernel memory allocator information (see text) (2.5)
698 bus Directory containing bus specific information
699 cmdline Kernel command line, both from bootloader and embedded
701 cpuinfo Info about the CPU
702 devices Available devices (block and character)
703 dma Used DMS channels
704 filesystems Supported filesystems
705 driver Various drivers grouped here, currently rtc (2.4)
706 execdomains Execdomains, related to security (2.4)
707 fb Frame Buffer devices (2.4)
708 fs File system parameters, currently nfs/exports (2.4)
709 ide Directory containing info about the IDE subsystem
710 interrupts Interrupt usage
711 iomem Memory map (2.4)
712 ioports I/O port usage
713 irq Masks for irq to cpu affinity (2.4)(smp?)
714 isapnp ISA PnP (Plug&Play) Info (2.4)
715 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
717 ksyms Kernel symbol table
718 loadavg Load average of last 1, 5 & 15 minutes;
719 number of processes currently runnable (running or on ready queue);
720 total number of processes in system;
722 All fields are separated by one space except "number of
723 processes currently runnable" and "total number of processes
724 in system", which are separated by a slash ('/'). Example:
725 0.61 0.61 0.55 3/828 22084
729 modules List of loaded modules
730 mounts Mounted filesystems
731 net Networking info (see text)
732 pagetypeinfo Additional page allocator information (see text) (2.5)
733 partitions Table of partitions known to the system
734 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
735 decoupled by lspci (2.4)
737 scsi SCSI info (see text)
738 slabinfo Slab pool info
739 softirqs softirq usage
740 stat Overall statistics
741 swaps Swap space utilization
743 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
744 tty Info of tty drivers
745 uptime Wall clock since boot, combined idle time of all cpus
746 version Kernel version
747 video bttv info of video resources (2.4)
748 vmallocinfo Show vmalloced areas
749 ============ ===============================================================
751 You can, for example, check which interrupts are currently in use and what
752 they are used for by looking in the file /proc/interrupts::
754 > cat /proc/interrupts
756 0: 8728810 XT-PIC timer
757 1: 895 XT-PIC keyboard
759 3: 531695 XT-PIC aha152x
760 4: 2014133 XT-PIC serial
761 5: 44401 XT-PIC pcnet_cs
764 12: 182918 XT-PIC PS/2 Mouse
766 14: 1232265 XT-PIC ide0
770 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
771 output of a SMP machine)::
773 > cat /proc/interrupts
776 0: 1243498 1214548 IO-APIC-edge timer
777 1: 8949 8958 IO-APIC-edge keyboard
778 2: 0 0 XT-PIC cascade
779 5: 11286 10161 IO-APIC-edge soundblaster
780 8: 1 0 IO-APIC-edge rtc
781 9: 27422 27407 IO-APIC-edge 3c503
782 12: 113645 113873 IO-APIC-edge PS/2 Mouse
784 14: 22491 24012 IO-APIC-edge ide0
785 15: 2183 2415 IO-APIC-edge ide1
786 17: 30564 30414 IO-APIC-level eth0
787 18: 177 164 IO-APIC-level bttv
792 NMI is incremented in this case because every timer interrupt generates a NMI
793 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
795 LOC is the local interrupt counter of the internal APIC of every CPU.
797 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
798 connects the CPUs in a SMP system. This means that an error has been detected,
799 the IO-APIC automatically retry the transmission, so it should not be a big
800 problem, but you should read the SMP-FAQ.
802 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
803 /proc/interrupts to display every IRQ vector in use by the system, not
804 just those considered 'most important'. The new vectors are:
807 interrupt raised when a machine check threshold counter
808 (typically counting ECC corrected errors of memory or cache) exceeds
809 a configurable threshold. Only available on some systems.
812 a thermal event interrupt occurs when a temperature threshold
813 has been exceeded for the CPU. This interrupt may also be generated
814 when the temperature drops back to normal.
817 a spurious interrupt is some interrupt that was raised then lowered
818 by some IO device before it could be fully processed by the APIC. Hence
819 the APIC sees the interrupt but does not know what device it came from.
820 For this case the APIC will generate the interrupt with a IRQ vector
821 of 0xff. This might also be generated by chipset bugs.
824 rescheduling, call and TLB flush interrupts are
825 sent from one CPU to another per the needs of the OS. Typically,
826 their statistics are used by kernel developers and interested users to
827 determine the occurrence of interrupts of the given type.
829 The above IRQ vectors are displayed only when relevant. For example,
830 the threshold vector does not exist on x86_64 platforms. Others are
831 suppressed when the system is a uniprocessor. As of this writing, only
832 i386 and x86_64 platforms support the new IRQ vector displays.
834 Of some interest is the introduction of the /proc/irq directory to 2.4.
835 It could be used to set IRQ to CPU affinity. This means that you can "hook" an
836 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
837 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
843 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
844 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
848 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
849 IRQ. You can set it by doing::
851 > echo 1 > /proc/irq/10/smp_affinity
853 This means that only the first CPU will handle the IRQ, but you can also echo
854 5 which means that only the first and third CPU can handle the IRQ.
856 The contents of each smp_affinity file is the same by default::
858 > cat /proc/irq/0/smp_affinity
861 There is an alternate interface, smp_affinity_list which allows specifying
862 a CPU range instead of a bitmask::
864 > cat /proc/irq/0/smp_affinity_list
867 The default_smp_affinity mask applies to all non-active IRQs, which are the
868 IRQs which have not yet been allocated/activated, and hence which lack a
869 /proc/irq/[0-9]* directory.
871 The node file on an SMP system shows the node to which the device using the IRQ
872 reports itself as being attached. This hardware locality information does not
873 include information about any possible driver locality preference.
875 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
876 profiler. Default value is ffffffff (all CPUs if there are only 32 of them).
878 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
879 between all the CPUs which are allowed to handle it. As usual the kernel has
880 more info than you and does a better job than you, so the defaults are the
881 best choice for almost everyone. [Note this applies only to those IO-APIC's
882 that support "Round Robin" interrupt distribution.]
884 There are three more important subdirectories in /proc: net, scsi, and sys.
885 The general rule is that the contents, or even the existence of these
886 directories, depend on your kernel configuration. If SCSI is not enabled, the
887 directory scsi may not exist. The same is true with the net, which is there
888 only when networking support is present in the running kernel.
890 The slabinfo file gives information about memory usage at the slab level.
891 Linux uses slab pools for memory management above page level in version 2.2.
892 Commonly used objects have their own slab pool (such as network buffers,
893 directory cache, and so on).
897 > cat /proc/buddyinfo
899 Node 0, zone DMA 0 4 5 4 4 3 ...
900 Node 0, zone Normal 1 0 0 1 101 8 ...
901 Node 0, zone HighMem 2 0 0 1 1 0 ...
903 External fragmentation is a problem under some workloads, and buddyinfo is a
904 useful tool for helping diagnose these problems. Buddyinfo will give you a
905 clue as to how big an area you can safely allocate, or why a previous
908 Each column represents the number of pages of a certain order which are
909 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
910 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
911 available in ZONE_NORMAL, etc...
913 More information relevant to external fragmentation can be found in
916 > cat /proc/pagetypeinfo
920 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
921 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
922 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
923 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
924 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
925 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
926 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
927 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
928 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
929 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
930 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
932 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
933 Node 0, zone DMA 2 0 5 1 0
934 Node 0, zone DMA32 41 6 967 2 0
936 Fragmentation avoidance in the kernel works by grouping pages of different
937 migrate types into the same contiguous regions of memory called page blocks.
938 A page block is typically the size of the default hugepage size, e.g. 2MB on
939 X86-64. By keeping pages grouped based on their ability to move, the kernel
940 can reclaim pages within a page block to satisfy a high-order allocation.
942 The pagetypinfo begins with information on the size of a page block. It
943 then gives the same type of information as buddyinfo except broken down
944 by migrate-type and finishes with details on how many page blocks of each
947 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
948 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
949 make an estimate of the likely number of huge pages that can be allocated
950 at a given point in time. All the "Movable" blocks should be allocatable
951 unless memory has been mlock()'d. Some of the Reclaimable blocks should
952 also be allocatable although a lot of filesystem metadata may have to be
953 reclaimed to achieve this.
959 Provides information about distribution and utilization of memory. This
960 varies by architecture and compile options. Some of the counters reported
961 here overlap. The memory reported by the non overlapping counters may not
962 add up to the overall memory usage and the difference for some workloads
963 can be substantial. In many cases there are other means to find out
964 additional memory using subsystem specific interfaces, for instance
965 /proc/net/sockstat for TCP memory allocations.
967 Example output. You may not have all of these fields.
973 MemTotal: 32858820 kB
975 MemAvailable: 27214312 kB
981 Active(anon): 94064 kB
982 Inactive(anon): 4570616 kB
983 Active(file): 3143088 kB
984 Inactive(file): 3015640 kB
993 AnonPages: 4654780 kB
996 KReclaimable: 517708 kB
998 SReclaimable: 517708 kB
999 SUnreclaim: 142336 kB
1000 KernelStack: 11168 kB
1001 PageTables: 20540 kB
1006 CommitLimit: 16429408 kB
1007 Committed_AS: 7715148 kB
1008 VmallocTotal: 34359738367 kB
1009 VmallocUsed: 40444 kB
1012 EarlyMemtestBad: 0 kB
1013 HardwareCorrupted: 0 kB
1014 AnonHugePages: 4149248 kB
1015 ShmemHugePages: 0 kB
1016 ShmemPmdMapped: 0 kB
1025 Hugepagesize: 2048 kB
1027 DirectMap4k: 401152 kB
1028 DirectMap2M: 10008576 kB
1029 DirectMap1G: 24117248 kB
1032 Total usable RAM (i.e. physical RAM minus a few reserved
1033 bits and the kernel binary code)
1035 Total free RAM. On highmem systems, the sum of LowFree+HighFree
1037 An estimate of how much memory is available for starting new
1038 applications, without swapping. Calculated from MemFree,
1039 SReclaimable, the size of the file LRU lists, and the low
1040 watermarks in each zone.
1041 The estimate takes into account that the system needs some
1042 page cache to function well, and that not all reclaimable
1043 slab will be reclaimable, due to items being in use. The
1044 impact of those factors will vary from system to system.
1046 Relatively temporary storage for raw disk blocks
1047 shouldn't get tremendously large (20MB or so)
1049 In-memory cache for files read from the disk (the
1050 pagecache) as well as tmpfs & shmem.
1051 Doesn't include SwapCached.
1053 Memory that once was swapped out, is swapped back in but
1054 still also is in the swapfile (if memory is needed it
1055 doesn't need to be swapped out AGAIN because it is already
1056 in the swapfile. This saves I/O)
1058 Memory that has been used more recently and usually not
1059 reclaimed unless absolutely necessary.
1061 Memory which has been less recently used. It is more
1062 eligible to be reclaimed for other purposes
1064 Memory allocated for userspace which cannot be reclaimed, such
1065 as mlocked pages, ramfs backing pages, secret memfd pages etc.
1067 Memory locked with mlock().
1069 Highmem is all memory above ~860MB of physical memory.
1070 Highmem areas are for use by userspace programs, or
1071 for the pagecache. The kernel must use tricks to access
1072 this memory, making it slower to access than lowmem.
1074 Lowmem is memory which can be used for everything that
1075 highmem can be used for, but it is also available for the
1076 kernel's use for its own data structures. Among many
1077 other things, it is where everything from the Slab is
1078 allocated. Bad things happen when you're out of lowmem.
1080 total amount of swap space available
1082 Memory which has been evicted from RAM, and is temporarily
1085 Memory consumed by the zswap backend (compressed size)
1087 Amount of anonymous memory stored in zswap (original size)
1089 Memory which is waiting to get written back to the disk
1091 Memory which is actively being written back to the disk
1093 Non-file backed pages mapped into userspace page tables
1095 files which have been mmapped, such as libraries
1097 Total memory used by shared memory (shmem) and tmpfs
1099 Kernel allocations that the kernel will attempt to reclaim
1100 under memory pressure. Includes SReclaimable (below), and other
1101 direct allocations with a shrinker.
1103 in-kernel data structures cache
1105 Part of Slab, that might be reclaimed, such as caches
1107 Part of Slab, that cannot be reclaimed on memory pressure
1109 Memory consumed by the kernel stacks of all tasks
1111 Memory consumed by userspace page tables
1113 Memory consumed by secondary page tables, this currently
1114 currently includes KVM mmu allocations on x86 and arm64.
1116 Always zero. Previous counted pages which had been written to
1117 the server, but has not been committed to stable storage.
1119 Memory used for block device "bounce buffers"
1121 Memory used by FUSE for temporary writeback buffers
1123 Based on the overcommit ratio ('vm.overcommit_ratio'),
1124 this is the total amount of memory currently available to
1125 be allocated on the system. This limit is only adhered to
1126 if strict overcommit accounting is enabled (mode 2 in
1127 'vm.overcommit_memory').
1129 The CommitLimit is calculated with the following formula::
1131 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
1132 overcommit_ratio / 100 + [total swap pages]
1134 For example, on a system with 1G of physical RAM and 7G
1135 of swap with a `vm.overcommit_ratio` of 30 it would
1136 yield a CommitLimit of 7.3G.
1138 For more details, see the memory overcommit documentation
1139 in mm/overcommit-accounting.
1141 The amount of memory presently allocated on the system.
1142 The committed memory is a sum of all of the memory which
1143 has been allocated by processes, even if it has not been
1144 "used" by them as of yet. A process which malloc()'s 1G
1145 of memory, but only touches 300M of it will show up as
1146 using 1G. This 1G is memory which has been "committed" to
1147 by the VM and can be used at any time by the allocating
1148 application. With strict overcommit enabled on the system
1149 (mode 2 in 'vm.overcommit_memory'), allocations which would
1150 exceed the CommitLimit (detailed above) will not be permitted.
1151 This is useful if one needs to guarantee that processes will
1152 not fail due to lack of memory once that memory has been
1153 successfully allocated.
1155 total size of vmalloc virtual address space
1157 amount of vmalloc area which is used
1159 largest contiguous block of vmalloc area which is free
1161 Memory allocated to the percpu allocator used to back percpu
1162 allocations. This stat excludes the cost of metadata.
1164 The amount of RAM/memory in kB, that was identified as corrupted
1165 by early memtest. If memtest was not run, this field will not
1166 be displayed at all. Size is never rounded down to 0 kB.
1167 That means if 0 kB is reported, you can safely assume
1168 there was at least one pass of memtest and none of the passes
1169 found a single faulty byte of RAM.
1171 The amount of RAM/memory in KB, the kernel identifies as
1174 Non-file backed huge pages mapped into userspace page tables
1176 Memory used by shared memory (shmem) and tmpfs allocated
1179 Shared memory mapped into userspace with huge pages
1181 Memory used for filesystem data (page cache) allocated
1184 Page cache mapped into userspace with huge pages
1186 Memory reserved for the Contiguous Memory Allocator (CMA)
1188 Free remaining memory in the CMA reserves
1189 HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb
1190 See Documentation/admin-guide/mm/hugetlbpage.rst.
1191 DirectMap4k, DirectMap2M, DirectMap1G
1192 Breakdown of page table sizes used in the kernel's
1193 identity mapping of RAM
1198 Provides information about vmalloced/vmaped areas. One line per area,
1199 containing the virtual address range of the area, size in bytes,
1200 caller information of the creator, and optional information depending
1201 on the kind of area:
1203 ========== ===================================================
1204 pages=nr number of pages
1205 phys=addr if a physical address was specified
1206 ioremap I/O mapping (ioremap() and friends)
1207 vmalloc vmalloc() area
1209 user VM_USERMAP area
1210 vpages buffer for pages pointers was vmalloced (huge area)
1211 N<node>=nr (Only on NUMA kernels)
1212 Number of pages allocated on memory node <node>
1213 ========== ===================================================
1217 > cat /proc/vmallocinfo
1218 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
1219 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
1220 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
1221 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
1222 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
1223 phys=7fee8000 ioremap
1224 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
1225 phys=7fee7000 ioremap
1226 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
1227 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
1228 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
1229 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
1230 pages=2 vmalloc N1=2
1231 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
1232 /0x130 [x_tables] pages=4 vmalloc N0=4
1233 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
1234 pages=14 vmalloc N2=14
1235 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
1236 pages=4 vmalloc N1=4
1237 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
1238 pages=2 vmalloc N1=2
1239 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
1240 pages=10 vmalloc N0=10
1246 Provides counts of softirq handlers serviced since boot time, for each CPU.
1250 > cat /proc/softirqs
1253 TIMER: 27166 27120 27097 27034
1258 SCHED: 27035 26983 26971 26746
1260 RCU: 1678 1769 2178 2250
1262 1.3 Networking info in /proc/net
1263 --------------------------------
1265 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1266 additional values you get for IP version 6 if you configure the kernel to
1267 support this. Table 1-9 lists the files and their meaning.
1270 .. table:: Table 1-8: IPv6 info in /proc/net
1272 ========== =====================================================
1274 ========== =====================================================
1275 udp6 UDP sockets (IPv6)
1276 tcp6 TCP sockets (IPv6)
1277 raw6 Raw device statistics (IPv6)
1278 igmp6 IP multicast addresses, which this host joined (IPv6)
1279 if_inet6 List of IPv6 interface addresses
1280 ipv6_route Kernel routing table for IPv6
1281 rt6_stats Global IPv6 routing tables statistics
1282 sockstat6 Socket statistics (IPv6)
1283 snmp6 Snmp data (IPv6)
1284 ========== =====================================================
1286 .. table:: Table 1-9: Network info in /proc/net
1288 ============= ================================================================
1290 ============= ================================================================
1291 arp Kernel ARP table
1292 dev network devices with statistics
1293 dev_mcast the Layer2 multicast groups a device is listening too
1294 (interface index, label, number of references, number of bound
1296 dev_stat network device status
1297 ip_fwchains Firewall chain linkage
1298 ip_fwnames Firewall chain names
1299 ip_masq Directory containing the masquerading tables
1300 ip_masquerade Major masquerading table
1301 netstat Network statistics
1302 raw raw device statistics
1303 route Kernel routing table
1304 rpc Directory containing rpc info
1305 rt_cache Routing cache
1307 sockstat Socket statistics
1308 softnet_stat Per-CPU incoming packets queues statistics of online CPUs
1311 unix UNIX domain sockets
1312 wireless Wireless interface data (Wavelan etc)
1313 igmp IP multicast addresses, which this host joined
1314 psched Global packet scheduler parameters.
1315 netlink List of PF_NETLINK sockets
1316 ip_mr_vifs List of multicast virtual interfaces
1317 ip_mr_cache List of multicast routing cache
1318 ============= ================================================================
1320 You can use this information to see which network devices are available in
1321 your system and how much traffic was routed over those devices::
1324 Inter-|Receive |[...
1325 face |bytes packets errs drop fifo frame compressed multicast|[...
1326 lo: 908188 5596 0 0 0 0 0 0 [...
1327 ppp0:15475140 20721 410 0 0 410 0 0 [...
1328 eth0: 614530 7085 0 0 0 0 0 1 [...
1331 ...] bytes packets errs drop fifo colls carrier compressed
1332 ...] 908188 5596 0 0 0 0 0 0
1333 ...] 1375103 17405 0 0 0 0 0 0
1334 ...] 1703981 5535 0 0 0 3 0 0
1336 In addition, each Channel Bond interface has its own directory. For
1337 example, the bond0 device will have a directory called /proc/net/bond0/.
1338 It will contain information that is specific to that bond, such as the
1339 current slaves of the bond, the link status of the slaves, and how
1340 many times the slaves link has failed.
1345 If you have a SCSI or ATA host adapter in your system, you'll find a
1346 subdirectory named after the driver for this adapter in /proc/scsi.
1347 You'll also see a list of all recognized SCSI devices in /proc/scsi::
1349 >cat /proc/scsi/scsi
1351 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1352 Vendor: IBM Model: DGHS09U Rev: 03E0
1353 Type: Direct-Access ANSI SCSI revision: 03
1354 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1355 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1356 Type: CD-ROM ANSI SCSI revision: 02
1359 The directory named after the driver has one file for each adapter found in
1360 the system. These files contain information about the controller, including
1361 the used IRQ and the IO address range. The amount of information shown is
1362 dependent on the adapter you use. The example shows the output for an Adaptec
1363 AHA-2940 SCSI adapter::
1365 > cat /proc/scsi/aic7xxx/0
1367 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1369 TCQ Enabled By Default : Disabled
1370 AIC7XXX_PROC_STATS : Disabled
1371 AIC7XXX_RESET_DELAY : 5
1372 Adapter Configuration:
1373 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1374 Ultra Wide Controller
1375 PCI MMAPed I/O Base: 0xeb001000
1376 Adapter SEEPROM Config: SEEPROM found and used.
1377 Adaptec SCSI BIOS: Enabled
1379 SCBs: Active 0, Max Active 2,
1380 Allocated 15, HW 16, Page 255
1382 BIOS Control Word: 0x18b6
1383 Adapter Control Word: 0x005b
1384 Extended Translation: Enabled
1385 Disconnect Enable Flags: 0xffff
1386 Ultra Enable Flags: 0x0001
1387 Tag Queue Enable Flags: 0x0000
1388 Ordered Queue Tag Flags: 0x0000
1389 Default Tag Queue Depth: 8
1390 Tagged Queue By Device array for aic7xxx host instance 0:
1391 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1392 Actual queue depth per device for aic7xxx host instance 0:
1393 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1396 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1397 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1398 Total transfers 160151 (74577 reads and 85574 writes)
1400 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1401 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1402 Total transfers 0 (0 reads and 0 writes)
1405 1.5 Parallel port info in /proc/parport
1406 ---------------------------------------
1408 The directory /proc/parport contains information about the parallel ports of
1409 your system. It has one subdirectory for each port, named after the port
1412 These directories contain the four files shown in Table 1-10.
1415 .. table:: Table 1-10: Files in /proc/parport
1417 ========= ====================================================================
1419 ========= ====================================================================
1420 autoprobe Any IEEE-1284 device ID information that has been acquired.
1421 devices list of the device drivers using that port. A + will appear by the
1422 name of the device currently using the port (it might not appear
1424 hardware Parallel port's base address, IRQ line and DMA channel.
1425 irq IRQ that parport is using for that port. This is in a separate
1426 file to allow you to alter it by writing a new value in (IRQ
1428 ========= ====================================================================
1430 1.6 TTY info in /proc/tty
1431 -------------------------
1433 Information about the available and actually used tty's can be found in the
1434 directory /proc/tty. You'll find entries for drivers and line disciplines in
1435 this directory, as shown in Table 1-11.
1438 .. table:: Table 1-11: Files in /proc/tty
1440 ============= ==============================================
1442 ============= ==============================================
1443 drivers list of drivers and their usage
1444 ldiscs registered line disciplines
1445 driver/serial usage statistic and status of single tty lines
1446 ============= ==============================================
1448 To see which tty's are currently in use, you can simply look into the file
1451 > cat /proc/tty/drivers
1452 pty_slave /dev/pts 136 0-255 pty:slave
1453 pty_master /dev/ptm 128 0-255 pty:master
1454 pty_slave /dev/ttyp 3 0-255 pty:slave
1455 pty_master /dev/pty 2 0-255 pty:master
1456 serial /dev/cua 5 64-67 serial:callout
1457 serial /dev/ttyS 4 64-67 serial
1458 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1459 /dev/ptmx /dev/ptmx 5 2 system
1460 /dev/console /dev/console 5 1 system:console
1461 /dev/tty /dev/tty 5 0 system:/dev/tty
1462 unknown /dev/tty 4 1-63 console
1465 1.7 Miscellaneous kernel statistics in /proc/stat
1466 -------------------------------------------------
1468 Various pieces of information about kernel activity are available in the
1469 /proc/stat file. All of the numbers reported in this file are aggregates
1470 since the system first booted. For a quick look, simply cat the file::
1473 cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0
1474 cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0
1475 cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0
1476 cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0
1477 cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0
1478 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>
1484 softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354
1486 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1487 lines. These numbers identify the amount of time the CPU has spent performing
1488 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1489 second). The meanings of the columns are as follows, from left to right:
1491 - user: normal processes executing in user mode
1492 - nice: niced processes executing in user mode
1493 - system: processes executing in kernel mode
1494 - idle: twiddling thumbs
1495 - iowait: In a word, iowait stands for waiting for I/O to complete. But there
1496 are several problems:
1498 1. CPU will not wait for I/O to complete, iowait is the time that a task is
1499 waiting for I/O to complete. When CPU goes into idle state for
1500 outstanding task I/O, another task will be scheduled on this CPU.
1501 2. In a multi-core CPU, the task waiting for I/O to complete is not running
1502 on any CPU, so the iowait of each CPU is difficult to calculate.
1503 3. The value of iowait field in /proc/stat will decrease in certain
1506 So, the iowait is not reliable by reading from /proc/stat.
1507 - irq: servicing interrupts
1508 - softirq: servicing softirqs
1509 - steal: involuntary wait
1510 - guest: running a normal guest
1511 - guest_nice: running a niced guest
1513 The "intr" line gives counts of interrupts serviced since boot time, for each
1514 of the possible system interrupts. The first column is the total of all
1515 interrupts serviced including unnumbered architecture specific interrupts;
1516 each subsequent column is the total for that particular numbered interrupt.
1517 Unnumbered interrupts are not shown, only summed into the total.
1519 The "ctxt" line gives the total number of context switches across all CPUs.
1521 The "btime" line gives the time at which the system booted, in seconds since
1524 The "processes" line gives the number of processes and threads created, which
1525 includes (but is not limited to) those created by calls to the fork() and
1526 clone() system calls.
1528 The "procs_running" line gives the total number of threads that are
1529 running or ready to run (i.e., the total number of runnable threads).
1531 The "procs_blocked" line gives the number of processes currently blocked,
1532 waiting for I/O to complete.
1534 The "softirq" line gives counts of softirqs serviced since boot time, for each
1535 of the possible system softirqs. The first column is the total of all
1536 softirqs serviced; each subsequent column is the total for that particular
1540 1.8 Ext4 file system parameters
1541 -------------------------------
1543 Information about mounted ext4 file systems can be found in
1544 /proc/fs/ext4. Each mounted filesystem will have a directory in
1545 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1546 /proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device
1547 directory are shown in Table 1-12, below.
1549 .. table:: Table 1-12: Files in /proc/fs/ext4/<devname>
1551 ============== ==========================================================
1553 mb_groups details of multiblock allocator buddy cache of free blocks
1554 ============== ==========================================================
1558 Shows registered system console lines.
1560 To see which character device lines are currently used for the system console
1561 /dev/console, you may simply look into the file /proc/consoles::
1563 > cat /proc/consoles
1569 +--------------------+-------------------------------------------------------+
1570 | device | name of the device |
1571 +====================+=======================================================+
1572 | operations | * R = can do read operations |
1573 | | * W = can do write operations |
1574 | | * U = can do unblank |
1575 +--------------------+-------------------------------------------------------+
1576 | flags | * E = it is enabled |
1577 | | * C = it is preferred console |
1578 | | * B = it is primary boot console |
1579 | | * p = it is used for printk buffer |
1580 | | * b = it is not a TTY but a Braille device |
1581 | | * a = it is safe to use when cpu is offline |
1582 +--------------------+-------------------------------------------------------+
1583 | major:minor | major and minor number of the device separated by a |
1585 +--------------------+-------------------------------------------------------+
1590 The /proc file system serves information about the running system. It not only
1591 allows access to process data but also allows you to request the kernel status
1592 by reading files in the hierarchy.
1594 The directory structure of /proc reflects the types of information and makes
1595 it easy, if not obvious, where to look for specific data.
1597 Chapter 2: Modifying System Parameters
1598 ======================================
1603 * Modifying kernel parameters by writing into files found in /proc/sys
1604 * Exploring the files which modify certain parameters
1605 * Review of the /proc/sys file tree
1607 ------------------------------------------------------------------------------
1609 A very interesting part of /proc is the directory /proc/sys. This is not only
1610 a source of information, it also allows you to change parameters within the
1611 kernel. Be very careful when attempting this. You can optimize your system,
1612 but you can also cause it to crash. Never alter kernel parameters on a
1613 production system. Set up a development machine and test to make sure that
1614 everything works the way you want it to. You may have no alternative but to
1615 reboot the machine once an error has been made.
1617 To change a value, simply echo the new value into the file.
1618 You need to be root to do this. You can create your own boot script
1619 to perform this every time your system boots.
1621 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1622 general things in the operation of the Linux kernel. Since some of the files
1623 can inadvertently disrupt your system, it is advisable to read both
1624 documentation and source before actually making adjustments. In any case, be
1625 very careful when writing to any of these files. The entries in /proc may
1626 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1627 review the kernel documentation in the directory linux/Documentation.
1628 This chapter is heavily based on the documentation included in the pre 2.2
1629 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1631 Please see: Documentation/admin-guide/sysctl/ directory for descriptions of
1637 Certain aspects of kernel behavior can be modified at runtime, without the
1638 need to recompile the kernel, or even to reboot the system. The files in the
1639 /proc/sys tree can not only be read, but also modified. You can use the echo
1640 command to write value into these files, thereby changing the default settings
1644 Chapter 3: Per-process Parameters
1645 =================================
1647 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1648 --------------------------------------------------------------------------------
1650 These files can be used to adjust the badness heuristic used to select which
1651 process gets killed in out of memory (oom) conditions.
1653 The badness heuristic assigns a value to each candidate task ranging from 0
1654 (never kill) to 1000 (always kill) to determine which process is targeted. The
1655 units are roughly a proportion along that range of allowed memory the process
1656 may allocate from based on an estimation of its current memory and swap use.
1657 For example, if a task is using all allowed memory, its badness score will be
1658 1000. If it is using half of its allowed memory, its score will be 500.
1660 The amount of "allowed" memory depends on the context in which the oom killer
1661 was called. If it is due to the memory assigned to the allocating task's cpuset
1662 being exhausted, the allowed memory represents the set of mems assigned to that
1663 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1664 memory represents the set of mempolicy nodes. If it is due to a memory
1665 limit (or swap limit) being reached, the allowed memory is that configured
1666 limit. Finally, if it is due to the entire system being out of memory, the
1667 allowed memory represents all allocatable resources.
1669 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1670 is used to determine which task to kill. Acceptable values range from -1000
1671 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1672 polarize the preference for oom killing either by always preferring a certain
1673 task or completely disabling it. The lowest possible value, -1000, is
1674 equivalent to disabling oom killing entirely for that task since it will always
1675 report a badness score of 0.
1677 Consequently, it is very simple for userspace to define the amount of memory to
1678 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1679 example, is roughly equivalent to allowing the remainder of tasks sharing the
1680 same system, cpuset, mempolicy, or memory controller resources to use at least
1681 50% more memory. A value of -500, on the other hand, would be roughly
1682 equivalent to discounting 50% of the task's allowed memory from being considered
1683 as scoring against the task.
1685 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1686 be used to tune the badness score. Its acceptable values range from -16
1687 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1688 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1689 scaled linearly with /proc/<pid>/oom_score_adj.
1691 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1692 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1693 requires CAP_SYS_RESOURCE.
1696 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1697 -------------------------------------------------------------
1699 This file can be used to check the current score used by the oom-killer for
1700 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1701 process should be killed in an out-of-memory situation.
1703 Please note that the exported value includes oom_score_adj so it is
1704 effectively in range [0,2000].
1707 3.3 /proc/<pid>/io - Display the IO accounting fields
1708 -------------------------------------------------------
1710 This file contains IO statistics for each running process.
1717 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1720 test:/tmp # cat /proc/3828/io
1726 write_bytes: 323932160
1727 cancelled_write_bytes: 0
1736 I/O counter: chars read
1737 The number of bytes which this task has caused to be read from storage. This
1738 is simply the sum of bytes which this process passed to read() and pread().
1739 It includes things like tty IO and it is unaffected by whether or not actual
1740 physical disk IO was required (the read might have been satisfied from
1747 I/O counter: chars written
1748 The number of bytes which this task has caused, or shall cause to be written
1749 to disk. Similar caveats apply here as with rchar.
1755 I/O counter: read syscalls
1756 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1763 I/O counter: write syscalls
1764 Attempt to count the number of write I/O operations, i.e. syscalls like
1765 write() and pwrite().
1771 I/O counter: bytes read
1772 Attempt to count the number of bytes which this process really did cause to
1773 be fetched from the storage layer. Done at the submit_bio() level, so it is
1774 accurate for block-backed filesystems. <please add status regarding NFS and
1775 CIFS at a later time>
1781 I/O counter: bytes written
1782 Attempt to count the number of bytes which this process caused to be sent to
1783 the storage layer. This is done at page-dirtying time.
1786 cancelled_write_bytes
1787 ^^^^^^^^^^^^^^^^^^^^^
1789 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1790 then deletes the file, it will in fact perform no writeout. But it will have
1791 been accounted as having caused 1MB of write.
1792 In other words: The number of bytes which this process caused to not happen,
1793 by truncating pagecache. A task can cause "negative" IO too. If this task
1794 truncates some dirty pagecache, some IO which another task has been accounted
1795 for (in its write_bytes) will not be happening. We _could_ just subtract that
1796 from the truncating task's write_bytes, but there is information loss in doing
1802 At its current implementation state, this is a bit racy on 32-bit machines:
1803 if process A reads process B's /proc/pid/io while process B is updating one
1804 of those 64-bit counters, process A could see an intermediate result.
1807 More information about this can be found within the taskstats documentation in
1808 Documentation/accounting.
1810 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1811 ---------------------------------------------------------------
1812 When a process is dumped, all anonymous memory is written to a core file as
1813 long as the size of the core file isn't limited. But sometimes we don't want
1814 to dump some memory segments, for example, huge shared memory or DAX.
1815 Conversely, sometimes we want to save file-backed memory segments into a core
1816 file, not only the individual files.
1818 /proc/<pid>/coredump_filter allows you to customize which memory segments
1819 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1820 of memory types. If a bit of the bitmask is set, memory segments of the
1821 corresponding memory type are dumped, otherwise they are not dumped.
1823 The following 9 memory types are supported:
1825 - (bit 0) anonymous private memory
1826 - (bit 1) anonymous shared memory
1827 - (bit 2) file-backed private memory
1828 - (bit 3) file-backed shared memory
1829 - (bit 4) ELF header pages in file-backed private memory areas (it is
1830 effective only if the bit 2 is cleared)
1831 - (bit 5) hugetlb private memory
1832 - (bit 6) hugetlb shared memory
1833 - (bit 7) DAX private memory
1834 - (bit 8) DAX shared memory
1836 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1837 are always dumped regardless of the bitmask status.
1839 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1840 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1842 The default value of coredump_filter is 0x33; this means all anonymous memory
1843 segments, ELF header pages and hugetlb private memory are dumped.
1845 If you don't want to dump all shared memory segments attached to pid 1234,
1846 write 0x31 to the process's proc file::
1848 $ echo 0x31 > /proc/1234/coredump_filter
1850 When a new process is created, the process inherits the bitmask status from its
1851 parent. It is useful to set up coredump_filter before the program runs.
1854 $ echo 0x7 > /proc/self/coredump_filter
1857 3.5 /proc/<pid>/mountinfo - Information about mounts
1858 --------------------------------------------------------
1860 This file contains lines of the form::
1862 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1863 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4)
1865 (1) mount ID: unique identifier of the mount (may be reused after umount)
1866 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1867 (3) major:minor: value of st_dev for files on filesystem
1868 (4) root: root of the mount within the filesystem
1869 (5) mount point: mount point relative to the process's root
1870 (6) mount options: per mount options
1871 (n…m) optional fields: zero or more fields of the form "tag[:value]"
1872 (m+1) separator: marks the end of the optional fields
1873 (m+2) filesystem type: name of filesystem of the form "type[.subtype]"
1874 (m+3) mount source: filesystem specific information or "none"
1875 (m+4) super options: per super block options
1877 Parsers should ignore all unrecognised optional fields. Currently the
1878 possible optional fields are:
1880 ================ ==============================================================
1881 shared:X mount is shared in peer group X
1882 master:X mount is slave to peer group X
1883 propagate_from:X mount is slave and receives propagation from peer group X [#]_
1884 unbindable mount is unbindable
1885 ================ ==============================================================
1887 .. [#] X is the closest dominant peer group under the process's root. If
1888 X is the immediate master of the mount, or if there's no dominant peer
1889 group under the same root, then only the "master:X" field is present
1890 and not the "propagate_from:X" field.
1892 For more information on mount propagation see:
1894 Documentation/filesystems/sharedsubtree.rst
1897 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1898 --------------------------------------------------------
1899 These files provide a method to access a task's comm value. It also allows for
1900 a task to set its own or one of its thread siblings comm value. The comm value
1901 is limited in size compared to the cmdline value, so writing anything longer
1902 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1906 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1907 -------------------------------------------------------------------------
1908 This file provides a fast way to retrieve first level children pids
1909 of a task pointed by <pid>/<tid> pair. The format is a space separated
1912 Note the "first level" here -- if a child has its own children they will
1913 not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children
1914 to obtain the descendants.
1916 Since this interface is intended to be fast and cheap it doesn't
1917 guarantee to provide precise results and some children might be
1918 skipped, especially if they've exited right after we printed their
1919 pids, so one needs to either stop or freeze processes being inspected
1920 if precise results are needed.
1923 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1924 ---------------------------------------------------------------
1925 This file provides information associated with an opened file. The regular
1926 files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'.
1927 The 'pos' represents the current offset of the opened file in decimal
1928 form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the
1929 file has been created with [see open(2) for details] and 'mnt_id' represents
1930 mount ID of the file system containing the opened file [see 3.5
1931 /proc/<pid>/mountinfo for details]. 'ino' represents the inode number of
1934 A typical output is::
1941 All locks associated with a file descriptor are shown in its fdinfo too::
1943 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1945 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1946 pair provide additional information particular to the objects they represent.
1959 where 'eventfd-count' is hex value of a counter.
1970 sigmask: 0000000000000200
1972 where 'sigmask' is hex value of the signal mask associated
1984 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7
1986 where 'tfd' is a target file descriptor number in decimal form,
1987 'events' is events mask being watched and the 'data' is data
1988 associated with a target [see epoll(7) for more details].
1990 The 'pos' is current offset of the target file in decimal form
1991 [see lseek(2)], 'ino' and 'sdev' are inode and device numbers
1992 where target file resides, all in hex format.
1996 For inotify files the format is the following::
2002 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
2004 where 'wd' is a watch descriptor in decimal form, i.e. a target file
2005 descriptor number, 'ino' and 'sdev' are inode and device where the
2006 target file resides and the 'mask' is the mask of events, all in hex
2007 form [see inotify(7) for more details].
2009 If the kernel was built with exportfs support, the path to the target
2010 file is encoded as a file handle. The file handle is provided by three
2011 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
2014 If the kernel is built without exportfs support the file handle won't be
2017 If there is no inotify mark attached yet the 'inotify' line will be omitted.
2019 For fanotify files the format is::
2025 fanotify flags:10 event-flags:0
2026 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
2027 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
2029 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
2030 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
2031 flags associated with mark which are tracked separately from events
2032 mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events
2033 mask and 'ignored_mask' is the mask of events which are to be ignored.
2034 All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
2035 provide information about flags and mask used in fanotify_mark
2036 call [see fsnotify manpage for details].
2038 While the first three lines are mandatory and always printed, the rest is
2039 optional and may be omitted if no marks created yet.
2053 it_value: (0, 49406829)
2056 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
2057 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
2058 flags in octal form been used to setup the timer [see timerfd_settime(2) for
2059 details]. 'it_value' is remaining time until the timer expiration.
2060 'it_interval' is the interval for the timer. Note the timer might be set up
2061 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
2062 still exhibits timer's remaining time.
2075 exp_name: system-heap
2077 where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of
2078 the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter.
2080 3.9 /proc/<pid>/map_files - Information about memory mapped files
2081 ---------------------------------------------------------------------
2082 This directory contains symbolic links which represent memory mapped files
2083 the process is maintaining. Example output::
2085 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
2086 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
2087 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
2089 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
2090 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
2092 The name of a link represents the virtual memory bounds of a mapping, i.e.
2093 vm_area_struct::vm_start-vm_area_struct::vm_end.
2095 The main purpose of the map_files is to retrieve a set of memory mapped
2096 files in a fast way instead of parsing /proc/<pid>/maps or
2097 /proc/<pid>/smaps, both of which contain many more records. At the same
2098 time one can open(2) mappings from the listings of two processes and
2099 comparing their inode numbers to figure out which anonymous memory areas
2100 are actually shared.
2102 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
2103 ---------------------------------------------------------
2104 This file provides the value of the task's timerslack value in nanoseconds.
2105 This value specifies an amount of time that normal timers may be deferred
2106 in order to coalesce timers and avoid unnecessary wakeups.
2108 This allows a task's interactivity vs power consumption tradeoff to be
2111 Writing 0 to the file will set the task's timerslack to the default value.
2113 Valid values are from 0 - ULLONG_MAX
2115 An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
2116 permissions on the task specified to change its timerslack_ns value.
2118 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
2119 -----------------------------------------------------------------
2120 When CONFIG_LIVEPATCH is enabled, this file displays the value of the
2121 patch state for the task.
2123 A value of '-1' indicates that no patch is in transition.
2125 A value of '0' indicates that a patch is in transition and the task is
2126 unpatched. If the patch is being enabled, then the task hasn't been
2127 patched yet. If the patch is being disabled, then the task has already
2130 A value of '1' indicates that a patch is in transition and the task is
2131 patched. If the patch is being enabled, then the task has already been
2132 patched. If the patch is being disabled, then the task hasn't been
2135 3.12 /proc/<pid>/arch_status - task architecture specific status
2136 -------------------------------------------------------------------
2137 When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the
2138 architecture specific status of the task.
2145 $ cat /proc/6753/arch_status
2146 AVX512_elapsed_ms: 8
2151 x86 specific entries
2152 ~~~~~~~~~~~~~~~~~~~~~
2157 If AVX512 is supported on the machine, this entry shows the milliseconds
2158 elapsed since the last time AVX512 usage was recorded. The recording
2159 happens on a best effort basis when a task is scheduled out. This means
2160 that the value depends on two factors:
2162 1) The time which the task spent on the CPU without being scheduled
2163 out. With CPU isolation and a single runnable task this can take
2166 2) The time since the task was scheduled out last. Depending on the
2167 reason for being scheduled out (time slice exhausted, syscall ...)
2168 this can be arbitrary long time.
2170 As a consequence the value cannot be considered precise and authoritative
2171 information. The application which uses this information has to be aware
2172 of the overall scenario on the system in order to determine whether a
2173 task is a real AVX512 user or not. Precise information can be obtained
2174 with performance counters.
2176 A special value of '-1' indicates that no AVX512 usage was recorded, thus
2177 the task is unlikely an AVX512 user, but depends on the workload and the
2178 scheduling scenario, it also could be a false negative mentioned above.
2180 3.13 /proc/<pid>/fd - List of symlinks to open files
2181 -------------------------------------------------------
2182 This directory contains symbolic links which represent open files
2183 the process is maintaining. Example output::
2185 lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null
2186 l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null
2187 lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]'
2188 lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]'
2189 lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]'
2191 The number of open files for the process is stored in 'size' member
2192 of stat() output for /proc/<pid>/fd for fast access.
2193 -------------------------------------------------------
2196 Chapter 4: Configuring procfs
2197 =============================
2200 ---------------------
2202 The following mount options are supported:
2204 ========= ========================================================
2205 hidepid= Set /proc/<pid>/ access mode.
2206 gid= Set the group authorized to learn processes information.
2207 subset= Show only the specified subset of procfs.
2208 ========= ========================================================
2210 hidepid=off or hidepid=0 means classic mode - everybody may access all
2211 /proc/<pid>/ directories (default).
2213 hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/
2214 directories but their own. Sensitive files like cmdline, sched*, status are now
2215 protected against other users. This makes it impossible to learn whether any
2216 user runs specific program (given the program doesn't reveal itself by its
2217 behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for
2218 other users, poorly written programs passing sensitive information via program
2219 arguments are now protected against local eavesdroppers.
2221 hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be
2222 fully invisible to other users. It doesn't mean that it hides a fact whether a
2223 process with a specific pid value exists (it can be learned by other means, e.g.
2224 by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by
2225 stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of
2226 gathering information about running processes, whether some daemon runs with
2227 elevated privileges, whether other user runs some sensitive program, whether
2228 other users run any program at all, etc.
2230 hidepid=ptraceable or hidepid=4 means that procfs should only contain
2231 /proc/<pid>/ directories that the caller can ptrace.
2233 gid= defines a group authorized to learn processes information otherwise
2234 prohibited by hidepid=. If you use some daemon like identd which needs to learn
2235 information about processes information, just add identd to this group.
2237 subset=pid hides all top level files and directories in the procfs that
2238 are not related to tasks.
2240 Chapter 5: Filesystem behavior
2241 ==============================
2243 Originally, before the advent of pid namespace, procfs was a global file
2244 system. It means that there was only one procfs instance in the system.
2246 When pid namespace was added, a separate procfs instance was mounted in
2247 each pid namespace. So, procfs mount options are global among all
2248 mountpoints within the same namespace::
2250 # grep ^proc /proc/mounts
2251 proc /proc proc rw,relatime,hidepid=2 0 0
2253 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc
2254 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0
2255 +++ exited with 0 +++
2257 # grep ^proc /proc/mounts
2258 proc /proc proc rw,relatime,hidepid=2 0 0
2259 proc /tmp/proc proc rw,relatime,hidepid=2 0 0
2261 and only after remounting procfs mount options will change at all
2264 # mount -o remount,hidepid=1 -t proc proc /tmp/proc
2266 # grep ^proc /proc/mounts
2267 proc /proc proc rw,relatime,hidepid=1 0 0
2268 proc /tmp/proc proc rw,relatime,hidepid=1 0 0
2270 This behavior is different from the behavior of other filesystems.
2272 The new procfs behavior is more like other filesystems. Each procfs mount
2273 creates a new procfs instance. Mount options affect own procfs instance.
2274 It means that it became possible to have several procfs instances
2275 displaying tasks with different filtering options in one pid namespace::
2277 # mount -o hidepid=invisible -t proc proc /proc
2278 # mount -o hidepid=noaccess -t proc proc /tmp/proc
2279 # grep ^proc /proc/mounts
2280 proc /proc proc rw,relatime,hidepid=invisible 0 0
2281 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0