| blob | 520f6a84cf50ffefacea03a47685e5dcffa4ba76 |
1 ------------------------------------------------------------------------------
2 T H E /proc F I L E S Y S T E M
3 ------------------------------------------------------------------------------
4 /proc/sys Terrehon Bowden <terrehon@pacbell.net> October 7 1999
5 Bodo Bauer <bb@ricochet.net>
7 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8 move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009
9 ------------------------------------------------------------------------------
10 Version 1.3 Kernel version 2.2.12
11 Kernel version 2.4.0-test11-pre4
12 ------------------------------------------------------------------------------
13 fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009
15 Table of Contents
16 -----------------
18 0 Preface
19 0.1 Introduction/Credits
20 0.2 Legal Stuff
22 1 Collecting System Information
23 1.1 Process-Specific Subdirectories
24 1.2 Kernel data
25 1.3 IDE devices in /proc/ide
26 1.4 Networking info in /proc/net
27 1.5 SCSI info
28 1.6 Parallel port info in /proc/parport
29 1.7 TTY info in /proc/tty
30 1.8 Miscellaneous kernel statistics in /proc/stat
31 1.9 Ext4 file system parameters
33 2 Modifying System Parameters
35 3 Per-Process Parameters
36 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer
37 score
38 3.2 /proc/<pid>/oom_score - Display current oom-killer score
39 3.3 /proc/<pid>/io - Display the IO accounting fields
40 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
41 3.5 /proc/<pid>/mountinfo - Information about mounts
42 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
43 3.7 /proc/<pid>/task/<tid>/children - Information about task children
44 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
45 3.9 /proc/<pid>/map_files - Information about memory mapped files
46 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
47 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
49 4 Configuring procfs
50 4.1 Mount options
52 ------------------------------------------------------------------------------
53 Preface
54 ------------------------------------------------------------------------------
56 0.1 Introduction/Credits
57 ------------------------
59 This documentation is part of a soon (or so we hope) to be released book on
60 the SuSE Linux distribution. As there is no complete documentation for the
61 /proc file system and we've used many freely available sources to write these
62 chapters, it seems only fair to give the work back to the Linux community.
63 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
64 afraid it's still far from complete, but we hope it will be useful. As far as
65 we know, it is the first 'all-in-one' document about the /proc file system. It
66 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
67 SPARC, AXP, etc., features, you probably won't find what you are looking for.
68 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
69 additions and patches are welcome and will be added to this document if you
70 mail them to Bodo.
72 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
73 other people for help compiling this documentation. We'd also like to extend a
74 special thank you to Andi Kleen for documentation, which we relied on heavily
75 to create this document, as well as the additional information he provided.
76 Thanks to everybody else who contributed source or docs to the Linux kernel
77 and helped create a great piece of software... :)
79 If you have any comments, corrections or additions, please don't hesitate to
80 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
81 document.
83 The latest version of this document is available online at
84 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
86 If the above direction does not works for you, you could try the kernel
87 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
88 comandante@zaralinux.com.
90 0.2 Legal Stuff
91 ---------------
93 We don't guarantee the correctness of this document, and if you come to us
94 complaining about how you screwed up your system because of incorrect
95 documentation, we won't feel responsible...
97 ------------------------------------------------------------------------------
98 CHAPTER 1: COLLECTING SYSTEM INFORMATION
99 ------------------------------------------------------------------------------
101 ------------------------------------------------------------------------------
102 In This Chapter
103 ------------------------------------------------------------------------------
104 * Investigating the properties of the pseudo file system /proc and its
105 ability to provide information on the running Linux system
106 * Examining /proc's structure
107 * Uncovering various information about the kernel and the processes running
108 on the system
109 ------------------------------------------------------------------------------
112 The proc file system acts as an interface to internal data structures in the
113 kernel. It can be used to obtain information about the system and to change
114 certain kernel parameters at runtime (sysctl).
116 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
117 show you how you can use /proc/sys to change settings.
119 1.1 Process-Specific Subdirectories
120 -----------------------------------
122 The directory /proc contains (among other things) one subdirectory for each
123 process running on the system, which is named after the process ID (PID).
125 The link self points to the process reading the file system. Each process
126 subdirectory has the entries listed in Table 1-1.
129 Table 1-1: Process specific entries in /proc
130 ..............................................................................
131 File Content
132 clear_refs Clears page referenced bits shown in smaps output
133 cmdline Command line arguments
134 cpu Current and last cpu in which it was executed (2.4)(smp)
135 cwd Link to the current working directory
136 environ Values of environment variables
137 exe Link to the executable of this process
138 fd Directory, which contains all file descriptors
139 maps Memory maps to executables and library files (2.4)
140 mem Memory held by this process
141 root Link to the root directory of this process
142 stat Process status
143 statm Process memory status information
144 status Process status in human readable form
145 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function
146 symbol the task is blocked in - or "0" if not blocked.
147 pagemap Page table
148 stack Report full stack trace, enable via CONFIG_STACKTRACE
149 smaps an extension based on maps, showing the memory consumption of
150 each mapping and flags associated with it
151 numa_maps an extension based on maps, showing the memory locality and
152 binding policy as well as mem usage (in pages) of each mapping.
153 ..............................................................................
155 For example, to get the status information of a process, all you have to do is
156 read the file /proc/PID/status:
158 >cat /proc/self/status
159 Name: cat
160 State: R (running)
161 Tgid: 5452
162 Pid: 5452
163 PPid: 743
164 TracerPid: 0 (2.4)
165 Uid: 501 501 501 501
166 Gid: 100 100 100 100
167 FDSize: 256
168 Groups: 100 14 16
169 VmPeak: 5004 kB
170 VmSize: 5004 kB
171 VmLck: 0 kB
172 VmHWM: 476 kB
173 VmRSS: 476 kB
174 RssAnon: 352 kB
175 RssFile: 120 kB
176 RssShmem: 4 kB
177 VmData: 156 kB
178 VmStk: 88 kB
179 VmExe: 68 kB
180 VmLib: 1412 kB
181 VmPTE: 20 kb
182 VmSwap: 0 kB
183 HugetlbPages: 0 kB
184 CoreDumping: 0
185 Threads: 1
186 SigQ: 0/28578
187 SigPnd: 0000000000000000
188 ShdPnd: 0000000000000000
189 SigBlk: 0000000000000000
190 SigIgn: 0000000000000000
191 SigCgt: 0000000000000000
192 CapInh: 00000000fffffeff
193 CapPrm: 0000000000000000
194 CapEff: 0000000000000000
195 CapBnd: ffffffffffffffff
196 NoNewPrivs: 0
197 Seccomp: 0
198 voluntary_ctxt_switches: 0
199 nonvoluntary_ctxt_switches: 1
201 This shows you nearly the same information you would get if you viewed it with
202 the ps command. In fact, ps uses the proc file system to obtain its
203 information. But you get a more detailed view of the process by reading the
204 file /proc/PID/status. It fields are described in table 1-2.
206 The statm file contains more detailed information about the process
207 memory usage. Its seven fields are explained in Table 1-3. The stat file
208 contains details information about the process itself. Its fields are
209 explained in Table 1-4.
211 (for SMP CONFIG users)
212 For making accounting scalable, RSS related information are handled in an
213 asynchronous manner and the value may not be very precise. To see a precise
214 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
215 It's slow but very precise.
217 Table 1-2: Contents of the status files (as of 4.8)
218 ..............................................................................
219 Field Content
220 Name filename of the executable
221 Umask file mode creation mask
222 State state (R is running, S is sleeping, D is sleeping
223 in an uninterruptible wait, Z is zombie,
224 T is traced or stopped)
225 Tgid thread group ID
226 Ngid NUMA group ID (0 if none)
227 Pid process id
228 PPid process id of the parent process
229 TracerPid PID of process tracing this process (0 if not)
230 Uid Real, effective, saved set, and file system UIDs
231 Gid Real, effective, saved set, and file system GIDs
232 FDSize number of file descriptor slots currently allocated
233 Groups supplementary group list
234 NStgid descendant namespace thread group ID hierarchy
235 NSpid descendant namespace process ID hierarchy
236 NSpgid descendant namespace process group ID hierarchy
237 NSsid descendant namespace session ID hierarchy
238 VmPeak peak virtual memory size
239 VmSize total program size
240 VmLck locked memory size
241 VmPin pinned memory size
242 VmHWM peak resident set size ("high water mark")
243 VmRSS size of memory portions. It contains the three
244 following parts (VmRSS = RssAnon + RssFile + RssShmem)
245 RssAnon size of resident anonymous memory
246 RssFile size of resident file mappings
247 RssShmem size of resident shmem memory (includes SysV shm,
248 mapping of tmpfs and shared anonymous mappings)
249 VmData size of private data segments
250 VmStk size of stack segments
251 VmExe size of text segment
252 VmLib size of shared library code
253 VmPTE size of page table entries
254 VmSwap amount of swap used by anonymous private data
255 (shmem swap usage is not included)
256 HugetlbPages size of hugetlb memory portions
257 CoreDumping process's memory is currently being dumped
258 (killing the process may lead to a corrupted core)
259 Threads number of threads
260 SigQ number of signals queued/max. number for queue
261 SigPnd bitmap of pending signals for the thread
262 ShdPnd bitmap of shared pending signals for the process
263 SigBlk bitmap of blocked signals
264 SigIgn bitmap of ignored signals
265 SigCgt bitmap of caught signals
266 CapInh bitmap of inheritable capabilities
267 CapPrm bitmap of permitted capabilities
268 CapEff bitmap of effective capabilities
269 CapBnd bitmap of capabilities bounding set
270 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
271 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
272 Cpus_allowed mask of CPUs on which this process may run
273 Cpus_allowed_list Same as previous, but in "list format"
274 Mems_allowed mask of memory nodes allowed to this process
275 Mems_allowed_list Same as previous, but in "list format"
276 voluntary_ctxt_switches number of voluntary context switches
277 nonvoluntary_ctxt_switches number of non voluntary context switches
278 ..............................................................................
280 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
281 ..............................................................................
282 Field Content
283 size total program size (pages) (same as VmSize in status)
284 resident size of memory portions (pages) (same as VmRSS in status)
285 shared number of pages that are shared (i.e. backed by a file, same
286 as RssFile+RssShmem in status)
287 trs number of pages that are 'code' (not including libs; broken,
288 includes data segment)
289 lrs number of pages of library (always 0 on 2.6)
290 drs number of pages of data/stack (including libs; broken,
291 includes library text)
292 dt number of dirty pages (always 0 on 2.6)
293 ..............................................................................
296 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
297 ..............................................................................
298 Field Content
299 pid process id
300 tcomm filename of the executable
301 state state (R is running, S is sleeping, D is sleeping in an
302 uninterruptible wait, Z is zombie, T is traced or stopped)
303 ppid process id of the parent process
304 pgrp pgrp of the process
305 sid session id
306 tty_nr tty the process uses
307 tty_pgrp pgrp of the tty
308 flags task flags
309 min_flt number of minor faults
310 cmin_flt number of minor faults with child's
311 maj_flt number of major faults
312 cmaj_flt number of major faults with child's
313 utime user mode jiffies
314 stime kernel mode jiffies
315 cutime user mode jiffies with child's
316 cstime kernel mode jiffies with child's
317 priority priority level
318 nice nice level
319 num_threads number of threads
320 it_real_value (obsolete, always 0)
321 start_time time the process started after system boot
322 vsize virtual memory size
323 rss resident set memory size
324 rsslim current limit in bytes on the rss
325 start_code address above which program text can run
326 end_code address below which program text can run
327 start_stack address of the start of the main process stack
328 esp current value of ESP
329 eip current value of EIP
330 pending bitmap of pending signals
331 blocked bitmap of blocked signals
332 sigign bitmap of ignored signals
333 sigcatch bitmap of caught signals
334 0 (place holder, used to be the wchan address, use /proc/PID/wchan instead)
335 0 (place holder)
336 0 (place holder)
337 exit_signal signal to send to parent thread on exit
338 task_cpu which CPU the task is scheduled on
339 rt_priority realtime priority
340 policy scheduling policy (man sched_setscheduler)
341 blkio_ticks time spent waiting for block IO
342 gtime guest time of the task in jiffies
343 cgtime guest time of the task children in jiffies
344 start_data address above which program data+bss is placed
345 end_data address below which program data+bss is placed
346 start_brk address above which program heap can be expanded with brk()
347 arg_start address above which program command line is placed
348 arg_end address below which program command line is placed
349 env_start address above which program environment is placed
350 env_end address below which program environment is placed
351 exit_code the thread's exit_code in the form reported by the waitpid system call
352 ..............................................................................
354 The /proc/PID/maps file containing the currently mapped memory regions and
355 their access permissions.
357 The format is:
359 address perms offset dev inode pathname
361 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
362 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
363 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
364 a7cb1000-a7cb2000 ---p 00000000 00:00 0
365 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
366 a7eb2000-a7eb3000 ---p 00000000 00:00 0
367 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
368 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
369 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
370 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
371 a800b000-a800e000 rw-p 00000000 00:00 0
372 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
373 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
374 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
375 a8024000-a8027000 rw-p 00000000 00:00 0
376 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
377 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
378 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
379 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
380 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
382 where "address" is the address space in the process that it occupies, "perms"
383 is a set of permissions:
385 r = read
386 w = write
387 x = execute
388 s = shared
389 p = private (copy on write)
391 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
392 "inode" is the inode on that device. 0 indicates that no inode is associated
393 with the memory region, as the case would be with BSS (uninitialized data).
394 The "pathname" shows the name associated file for this mapping. If the mapping
395 is not associated with a file:
397 [heap] = the heap of the program
398 [stack] = the stack of the main process
399 [vdso] = the "virtual dynamic shared object",
400 the kernel system call handler
402 or if empty, the mapping is anonymous.
404 The /proc/PID/smaps is an extension based on maps, showing the memory
405 consumption for each of the process's mappings. For each of mappings there
406 is a series of lines such as the following:
408 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
409 Size: 1084 kB
410 Rss: 892 kB
411 Pss: 374 kB
412 Shared_Clean: 892 kB
413 Shared_Dirty: 0 kB
414 Private_Clean: 0 kB
415 Private_Dirty: 0 kB
416 Referenced: 892 kB
417 Anonymous: 0 kB
418 LazyFree: 0 kB
419 AnonHugePages: 0 kB
420 ShmemPmdMapped: 0 kB
421 Shared_Hugetlb: 0 kB
422 Private_Hugetlb: 0 kB
423 Swap: 0 kB
424 SwapPss: 0 kB
425 KernelPageSize: 4 kB
426 MMUPageSize: 4 kB
427 Locked: 0 kB
428 VmFlags: rd ex mr mw me dw
430 the first of these lines shows the same information as is displayed for the
431 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
432 (size), the amount of the mapping that is currently resident in RAM (RSS), the
433 process' proportional share of this mapping (PSS), the number of clean and
434 dirty private pages in the mapping.
436 The "proportional set size" (PSS) of a process is the count of pages it has
437 in memory, where each page is divided by the number of processes sharing it.
438 So if a process has 1000 pages all to itself, and 1000 shared with one other
439 process, its PSS will be 1500.
440 Note that even a page which is part of a MAP_SHARED mapping, but has only
441 a single pte mapped, i.e. is currently used by only one process, is accounted
442 as private and not as shared.
443 "Referenced" indicates the amount of memory currently marked as referenced or
444 accessed.
445 "Anonymous" shows the amount of memory that does not belong to any file. Even
446 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
447 and a page is modified, the file page is replaced by a private anonymous copy.
448 "LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
449 The memory isn't freed immediately with madvise(). It's freed in memory
450 pressure if the memory is clean. Please note that the printed value might
451 be lower than the real value due to optimizations used in the current
452 implementation. If this is not desirable please file a bug report.
453 "AnonHugePages" shows the ammount of memory backed by transparent hugepage.
454 "ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by
455 huge pages.
456 "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
457 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
458 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
459 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
460 For shmem mappings, "Swap" includes also the size of the mapped (and not
461 replaced by copy-on-write) part of the underlying shmem object out on swap.
462 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
463 does not take into account swapped out page of underlying shmem objects.
464 "Locked" indicates whether the mapping is locked in memory or not.
466 "VmFlags" field deserves a separate description. This member represents the kernel
467 flags associated with the particular virtual memory area in two letter encoded
468 manner. The codes are the following:
469 rd - readable
470 wr - writeable
471 ex - executable
472 sh - shared
473 mr - may read
474 mw - may write
475 me - may execute
476 ms - may share
477 gd - stack segment growns down
478 pf - pure PFN range
479 dw - disabled write to the mapped file
480 lo - pages are locked in memory
481 io - memory mapped I/O area
482 sr - sequential read advise provided
483 rr - random read advise provided
484 dc - do not copy area on fork
485 de - do not expand area on remapping
486 ac - area is accountable
487 nr - swap space is not reserved for the area
488 ht - area uses huge tlb pages
489 ar - architecture specific flag
490 dd - do not include area into core dump
491 sd - soft-dirty flag
492 mm - mixed map area
493 hg - huge page advise flag
494 nh - no-huge page advise flag
495 mg - mergable advise flag
497 Note that there is no guarantee that every flag and associated mnemonic will
498 be present in all further kernel releases. Things get changed, the flags may
499 be vanished or the reverse -- new added.
501 This file is only present if the CONFIG_MMU kernel configuration option is
502 enabled.
504 Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
505 output can be achieved only in the single read call).
506 This typically manifests when doing partial reads of these files while the
507 memory map is being modified. Despite the races, we do provide the following
508 guarantees:
510 1) The mapped addresses never go backwards, which implies no two
511 regions will ever overlap.
512 2) If there is something at a given vaddr during the entirety of the
513 life of the smaps/maps walk, there will be some output for it.
516 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
517 bits on both physical and virtual pages associated with a process, and the
518 soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst
519 for details).
520 To clear the bits for all the pages associated with the process
521 > echo 1 > /proc/PID/clear_refs
523 To clear the bits for the anonymous pages associated with the process
524 > echo 2 > /proc/PID/clear_refs
526 To clear the bits for the file mapped pages associated with the process
527 > echo 3 > /proc/PID/clear_refs
529 To clear the soft-dirty bit
530 > echo 4 > /proc/PID/clear_refs
532 To reset the peak resident set size ("high water mark") to the process's
533 current value:
534 > echo 5 > /proc/PID/clear_refs
536 Any other value written to /proc/PID/clear_refs will have no effect.
538 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
539 using /proc/kpageflags and number of times a page is mapped using
540 /proc/kpagecount. For detailed explanation, see
541 Documentation/admin-guide/mm/pagemap.rst.
543 The /proc/pid/numa_maps is an extension based on maps, showing the memory
544 locality and binding policy, as well as the memory usage (in pages) of
545 each mapping. The output follows a general format where mapping details get
546 summarized separated by blank spaces, one mapping per each file line:
548 address policy mapping details
550 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
551 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
552 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
553 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
554 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
555 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
556 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
557 320698b000 default file=/lib64/libc-2.12.so
558 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
559 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
560 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
561 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
562 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
563 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
564 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
565 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
567 Where:
568 "address" is the starting address for the mapping;
569 "policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst);
570 "mapping details" summarizes mapping data such as mapping type, page usage counters,
571 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
572 size, in KB, that is backing the mapping up.
574 1.2 Kernel data
575 ---------------
577 Similar to the process entries, the kernel data files give information about
578 the running kernel. The files used to obtain this information are contained in
579 /proc and are listed in Table 1-5. Not all of these will be present in your
580 system. It depends on the kernel configuration and the loaded modules, which
581 files are there, and which are missing.
583 Table 1-5: Kernel info in /proc
584 ..............................................................................
585 File Content
586 apm Advanced power management info
587 buddyinfo Kernel memory allocator information (see text) (2.5)
588 bus Directory containing bus specific information
589 cmdline Kernel command line
590 cpuinfo Info about the CPU
591 devices Available devices (block and character)
592 dma Used DMS channels
593 filesystems Supported filesystems
594 driver Various drivers grouped here, currently rtc (2.4)
595 execdomains Execdomains, related to security (2.4)
596 fb Frame Buffer devices (2.4)
597 fs File system parameters, currently nfs/exports (2.4)
598 ide Directory containing info about the IDE subsystem
599 interrupts Interrupt usage
600 iomem Memory map (2.4)
601 ioports I/O port usage
602 irq Masks for irq to cpu affinity (2.4)(smp?)
603 isapnp ISA PnP (Plug&Play) Info (2.4)
604 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
605 kmsg Kernel messages
606 ksyms Kernel symbol table
607 loadavg Load average of last 1, 5 & 15 minutes
608 locks Kernel locks
609 meminfo Memory info
610 misc Miscellaneous
611 modules List of loaded modules
612 mounts Mounted filesystems
613 net Networking info (see text)
614 pagetypeinfo Additional page allocator information (see text) (2.5)
615 partitions Table of partitions known to the system
616 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
617 decoupled by lspci (2.4)
618 rtc Real time clock
619 scsi SCSI info (see text)
620 slabinfo Slab pool info
621 softirqs softirq usage
622 stat Overall statistics
623 swaps Swap space utilization
624 sys See chapter 2
625 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
626 tty Info of tty drivers
627 uptime Wall clock since boot, combined idle time of all cpus
628 version Kernel version
629 video bttv info of video resources (2.4)
630 vmallocinfo Show vmalloced areas
631 ..............................................................................
633 You can, for example, check which interrupts are currently in use and what
634 they are used for by looking in the file /proc/interrupts:
636 > cat /proc/interrupts
637 CPU0
638 0: 8728810 XT-PIC timer
639 1: 895 XT-PIC keyboard
640 2: 0 XT-PIC cascade
641 3: 531695 XT-PIC aha152x
642 4: 2014133 XT-PIC serial
643 5: 44401 XT-PIC pcnet_cs
644 8: 2 XT-PIC rtc
645 11: 8 XT-PIC i82365
646 12: 182918 XT-PIC PS/2 Mouse
647 13: 1 XT-PIC fpu
648 14: 1232265 XT-PIC ide0
649 15: 7 XT-PIC ide1
650 NMI: 0
652 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
653 output of a SMP machine):
655 > cat /proc/interrupts
657 CPU0 CPU1
658 0: 1243498 1214548 IO-APIC-edge timer
659 1: 8949 8958 IO-APIC-edge keyboard
660 2: 0 0 XT-PIC cascade
661 5: 11286 10161 IO-APIC-edge soundblaster
662 8: 1 0 IO-APIC-edge rtc
663 9: 27422 27407 IO-APIC-edge 3c503
664 12: 113645 113873 IO-APIC-edge PS/2 Mouse
665 13: 0 0 XT-PIC fpu
666 14: 22491 24012 IO-APIC-edge ide0
667 15: 2183 2415 IO-APIC-edge ide1
668 17: 30564 30414 IO-APIC-level eth0
669 18: 177 164 IO-APIC-level bttv
670 NMI: 2457961 2457959
671 LOC: 2457882 2457881
672 ERR: 2155
674 NMI is incremented in this case because every timer interrupt generates a NMI
675 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
677 LOC is the local interrupt counter of the internal APIC of every CPU.
679 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
680 connects the CPUs in a SMP system. This means that an error has been detected,
681 the IO-APIC automatically retry the transmission, so it should not be a big
682 problem, but you should read the SMP-FAQ.
684 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
685 /proc/interrupts to display every IRQ vector in use by the system, not
686 just those considered 'most important'. The new vectors are:
688 THR -- interrupt raised when a machine check threshold counter
689 (typically counting ECC corrected errors of memory or cache) exceeds
690 a configurable threshold. Only available on some systems.
692 TRM -- a thermal event interrupt occurs when a temperature threshold
693 has been exceeded for the CPU. This interrupt may also be generated
694 when the temperature drops back to normal.
696 SPU -- a spurious interrupt is some interrupt that was raised then lowered
697 by some IO device before it could be fully processed by the APIC. Hence
698 the APIC sees the interrupt but does not know what device it came from.
699 For this case the APIC will generate the interrupt with a IRQ vector
700 of 0xff. This might also be generated by chipset bugs.
702 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
703 sent from one CPU to another per the needs of the OS. Typically,
704 their statistics are used by kernel developers and interested users to
705 determine the occurrence of interrupts of the given type.
707 The above IRQ vectors are displayed only when relevant. For example,
708 the threshold vector does not exist on x86_64 platforms. Others are
709 suppressed when the system is a uniprocessor. As of this writing, only
710 i386 and x86_64 platforms support the new IRQ vector displays.
712 Of some interest is the introduction of the /proc/irq directory to 2.4.
713 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
714 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
715 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
716 prof_cpu_mask.
718 For example
719 > ls /proc/irq/
720 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
721 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
722 > ls /proc/irq/0/
723 smp_affinity
725 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
726 IRQ, you can set it by doing:
728 > echo 1 > /proc/irq/10/smp_affinity
730 This means that only the first CPU will handle the IRQ, but you can also echo
731 5 which means that only the first and third CPU can handle the IRQ.
733 The contents of each smp_affinity file is the same by default:
735 > cat /proc/irq/0/smp_affinity
736 ffffffff
738 There is an alternate interface, smp_affinity_list which allows specifying
739 a cpu range instead of a bitmask:
741 > cat /proc/irq/0/smp_affinity_list
742 1024-1031
744 The default_smp_affinity mask applies to all non-active IRQs, which are the
745 IRQs which have not yet been allocated/activated, and hence which lack a
746 /proc/irq/[0-9]* directory.
748 The node file on an SMP system shows the node to which the device using the IRQ
749 reports itself as being attached. This hardware locality information does not
750 include information about any possible driver locality preference.
752 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
753 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
755 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
756 between all the CPUs which are allowed to handle it. As usual the kernel has
757 more info than you and does a better job than you, so the defaults are the
758 best choice for almost everyone. [Note this applies only to those IO-APIC's
759 that support "Round Robin" interrupt distribution.]
761 There are three more important subdirectories in /proc: net, scsi, and sys.
762 The general rule is that the contents, or even the existence of these
763 directories, depend on your kernel configuration. If SCSI is not enabled, the
764 directory scsi may not exist. The same is true with the net, which is there
765 only when networking support is present in the running kernel.
767 The slabinfo file gives information about memory usage at the slab level.
768 Linux uses slab pools for memory management above page level in version 2.2.
769 Commonly used objects have their own slab pool (such as network buffers,
770 directory cache, and so on).
772 ..............................................................................
774 > cat /proc/buddyinfo
776 Node 0, zone DMA 0 4 5 4 4 3 ...
777 Node 0, zone Normal 1 0 0 1 101 8 ...
778 Node 0, zone HighMem 2 0 0 1 1 0 ...
780 External fragmentation is a problem under some workloads, and buddyinfo is a
781 useful tool for helping diagnose these problems. Buddyinfo will give you a
782 clue as to how big an area you can safely allocate, or why a previous
783 allocation failed.
785 Each column represents the number of pages of a certain order which are
786 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
787 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
788 available in ZONE_NORMAL, etc...
790 More information relevant to external fragmentation can be found in
791 pagetypeinfo.
793 > cat /proc/pagetypeinfo
794 Page block order: 9
795 Pages per block: 512
797 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
798 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
799 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
800 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
801 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
802 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
803 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
804 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
805 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
806 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
807 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
809 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
810 Node 0, zone DMA 2 0 5 1 0
811 Node 0, zone DMA32 41 6 967 2 0
813 Fragmentation avoidance in the kernel works by grouping pages of different
814 migrate types into the same contiguous regions of memory called page blocks.
815 A page block is typically the size of the default hugepage size e.g. 2MB on
816 X86-64. By keeping pages grouped based on their ability to move, the kernel
817 can reclaim pages within a page block to satisfy a high-order allocation.
819 The pagetypinfo begins with information on the size of a page block. It
820 then gives the same type of information as buddyinfo except broken down
821 by migrate-type and finishes with details on how many page blocks of each
822 type exist.
824 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
825 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
826 make an estimate of the likely number of huge pages that can be allocated
827 at a given point in time. All the "Movable" blocks should be allocatable
828 unless memory has been mlock()'d. Some of the Reclaimable blocks should
829 also be allocatable although a lot of filesystem metadata may have to be
830 reclaimed to achieve this.
832 ..............................................................................
834 meminfo:
836 Provides information about distribution and utilization of memory. This
837 varies by architecture and compile options. The following is from a
838 16GB PIII, which has highmem enabled. You may not have all of these fields.
840 > cat /proc/meminfo
842 MemTotal: 16344972 kB
843 MemFree: 13634064 kB
844 MemAvailable: 14836172 kB
845 Buffers: 3656 kB
846 Cached: 1195708 kB
847 SwapCached: 0 kB
848 Active: 891636 kB
849 Inactive: 1077224 kB
850 HighTotal: 15597528 kB
851 HighFree: 13629632 kB
852 LowTotal: 747444 kB
853 LowFree: 4432 kB
854 SwapTotal: 0 kB
855 SwapFree: 0 kB
856 Dirty: 968 kB
857 Writeback: 0 kB
858 AnonPages: 861800 kB
859 Mapped: 280372 kB
860 Shmem: 644 kB
861 Slab: 284364 kB
862 SReclaimable: 159856 kB
863 SUnreclaim: 124508 kB
864 PageTables: 24448 kB
865 NFS_Unstable: 0 kB
866 Bounce: 0 kB
867 WritebackTmp: 0 kB
868 CommitLimit: 7669796 kB
869 Committed_AS: 100056 kB
870 VmallocTotal: 112216 kB
871 VmallocUsed: 428 kB
872 VmallocChunk: 111088 kB
873 AnonHugePages: 49152 kB
874 ShmemHugePages: 0 kB
875 ShmemPmdMapped: 0 kB
878 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
879 bits and the kernel binary code)
880 MemFree: The sum of LowFree+HighFree
881 MemAvailable: An estimate of how much memory is available for starting new
882 applications, without swapping. Calculated from MemFree,
883 SReclaimable, the size of the file LRU lists, and the low
884 watermarks in each zone.
885 The estimate takes into account that the system needs some
886 page cache to function well, and that not all reclaimable
887 slab will be reclaimable, due to items being in use. The
888 impact of those factors will vary from system to system.
889 Buffers: Relatively temporary storage for raw disk blocks
890 shouldn't get tremendously large (20MB or so)
891 Cached: in-memory cache for files read from the disk (the
892 pagecache). Doesn't include SwapCached
893 SwapCached: Memory that once was swapped out, is swapped back in but
894 still also is in the swapfile (if memory is needed it
895 doesn't need to be swapped out AGAIN because it is already
896 in the swapfile. This saves I/O)
897 Active: Memory that has been used more recently and usually not
898 reclaimed unless absolutely necessary.
899 Inactive: Memory which has been less recently used. It is more
900 eligible to be reclaimed for other purposes
901 HighTotal:
902 HighFree: Highmem is all memory above ~860MB of physical memory
903 Highmem areas are for use by userspace programs, or
904 for the pagecache. The kernel must use tricks to access
905 this memory, making it slower to access than lowmem.
906 LowTotal:
907 LowFree: Lowmem is memory which can be used for everything that
908 highmem can be used for, but it is also available for the
909 kernel's use for its own data structures. Among many
910 other things, it is where everything from the Slab is
911 allocated. Bad things happen when you're out of lowmem.
912 SwapTotal: total amount of swap space available
913 SwapFree: Memory which has been evicted from RAM, and is temporarily
914 on the disk
915 Dirty: Memory which is waiting to get written back to the disk
916 Writeback: Memory which is actively being written back to the disk
917 AnonPages: Non-file backed pages mapped into userspace page tables
918 AnonHugePages: Non-file backed huge pages mapped into userspace page tables
919 Mapped: files which have been mmaped, such as libraries
920 Shmem: Total memory used by shared memory (shmem) and tmpfs
921 ShmemHugePages: Memory used by shared memory (shmem) and tmpfs allocated
922 with huge pages
923 ShmemPmdMapped: Shared memory mapped into userspace with huge pages
924 Slab: in-kernel data structures cache
925 SReclaimable: Part of Slab, that might be reclaimed, such as caches
926 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
927 PageTables: amount of memory dedicated to the lowest level of page
928 tables.
929 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
930 storage
931 Bounce: Memory used for block device "bounce buffers"
932 WritebackTmp: Memory used by FUSE for temporary writeback buffers
933 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
934 this is the total amount of memory currently available to
935 be allocated on the system. This limit is only adhered to
936 if strict overcommit accounting is enabled (mode 2 in
937 'vm.overcommit_memory').
938 The CommitLimit is calculated with the following formula:
939 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
940 overcommit_ratio / 100 + [total swap pages]
941 For example, on a system with 1G of physical RAM and 7G
942 of swap with a `vm.overcommit_ratio` of 30 it would
943 yield a CommitLimit of 7.3G.
944 For more details, see the memory overcommit documentation
945 in vm/overcommit-accounting.
946 Committed_AS: The amount of memory presently allocated on the system.
947 The committed memory is a sum of all of the memory which
948 has been allocated by processes, even if it has not been
949 "used" by them as of yet. A process which malloc()'s 1G
950 of memory, but only touches 300M of it will show up as
951 using 1G. This 1G is memory which has been "committed" to
952 by the VM and can be used at any time by the allocating
953 application. With strict overcommit enabled on the system
954 (mode 2 in 'vm.overcommit_memory'),allocations which would
955 exceed the CommitLimit (detailed above) will not be permitted.
956 This is useful if one needs to guarantee that processes will
957 not fail due to lack of memory once that memory has been
958 successfully allocated.
959 VmallocTotal: total size of vmalloc memory area
960 VmallocUsed: amount of vmalloc area which is used
961 VmallocChunk: largest contiguous block of vmalloc area which is free
963 ..............................................................................
965 vmallocinfo:
967 Provides information about vmalloced/vmaped areas. One line per area,
968 containing the virtual address range of the area, size in bytes,
969 caller information of the creator, and optional information depending
970 on the kind of area :
972 pages=nr number of pages
973 phys=addr if a physical address was specified
974 ioremap I/O mapping (ioremap() and friends)
975 vmalloc vmalloc() area
976 vmap vmap()ed pages
977 user VM_USERMAP area
978 vpages buffer for pages pointers was vmalloced (huge area)
979 N<node>=nr (Only on NUMA kernels)
980 Number of pages allocated on memory node <node>
982 > cat /proc/vmallocinfo
983 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
984 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
985 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
986 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
987 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
988 phys=7fee8000 ioremap
989 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
990 phys=7fee7000 ioremap
991 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
992 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
993 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
994 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
995 pages=2 vmalloc N1=2
996 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
997 /0x130 [x_tables] pages=4 vmalloc N0=4
998 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
999 pages=14 vmalloc N2=14
1000 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
1001 pages=4 vmalloc N1=4
1002 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
1003 pages=2 vmalloc N1=2
1004 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
1005 pages=10 vmalloc N0=10
1007 ..............................................................................
1009 softirqs:
1011 Provides counts of softirq handlers serviced since boot time, for each cpu.
1013 > cat /proc/softirqs
1014 CPU0 CPU1 CPU2 CPU3
1015 HI: 0 0 0 0
1016 TIMER: 27166 27120 27097 27034
1017 NET_TX: 0 0 0 17
1018 NET_RX: 42 0 0 39
1019 BLOCK: 0 0 107 1121
1020 TASKLET: 0 0 0 290
1021 SCHED: 27035 26983 26971 26746
1022 HRTIMER: 0 0 0 0
1023 RCU: 1678 1769 2178 2250
1026 1.3 IDE devices in /proc/ide
1027 ----------------------------
1029 The subdirectory /proc/ide contains information about all IDE devices of which
1030 the kernel is aware. There is one subdirectory for each IDE controller, the
1031 file drivers and a link for each IDE device, pointing to the device directory
1032 in the controller specific subtree.
1034 The file drivers contains general information about the drivers used for the
1035 IDE devices:
1037 > cat /proc/ide/drivers
1038 ide-cdrom version 4.53
1039 ide-disk version 1.08
1041 More detailed information can be found in the controller specific
1042 subdirectories. These are named ide0, ide1 and so on. Each of these
1043 directories contains the files shown in table 1-6.
1046 Table 1-6: IDE controller info in /proc/ide/ide?
1047 ..............................................................................
1048 File Content
1049 channel IDE channel (0 or 1)
1050 config Configuration (only for PCI/IDE bridge)
1051 mate Mate name
1052 model Type/Chipset of IDE controller
1053 ..............................................................................
1055 Each device connected to a controller has a separate subdirectory in the
1056 controllers directory. The files listed in table 1-7 are contained in these
1057 directories.
1060 Table 1-7: IDE device information
1061 ..............................................................................
1062 File Content
1063 cache The cache
1064 capacity Capacity of the medium (in 512Byte blocks)
1065 driver driver and version
1066 geometry physical and logical geometry
1067 identify device identify block
1068 media media type
1069 model device identifier
1070 settings device setup
1071 smart_thresholds IDE disk management thresholds
1072 smart_values IDE disk management values
1073 ..............................................................................
1075 The most interesting file is settings. This file contains a nice overview of
1076 the drive parameters:
1078 # cat /proc/ide/ide0/hda/settings
1079 name value min max mode
1080 ---- ----- --- --- ----
1081 bios_cyl 526 0 65535 rw
1082 bios_head 255 0 255 rw
1083 bios_sect 63 0 63 rw
1084 breada_readahead 4 0 127 rw
1085 bswap 0 0 1 r
1086 file_readahead 72 0 2097151 rw
1087 io_32bit 0 0 3 rw
1088 keepsettings 0 0 1 rw
1089 max_kb_per_request 122 1 127 rw
1090 multcount 0 0 8 rw
1091 nice1 1 0 1 rw
1092 nowerr 0 0 1 rw
1093 pio_mode write-only 0 255 w
1094 slow 0 0 1 rw
1095 unmaskirq 0 0 1 rw
1096 using_dma 0 0 1 rw
1099 1.4 Networking info in /proc/net
1100 --------------------------------
1102 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1103 additional values you get for IP version 6 if you configure the kernel to
1104 support this. Table 1-9 lists the files and their meaning.
1107 Table 1-8: IPv6 info in /proc/net
1108 ..............................................................................
1109 File Content
1110 udp6 UDP sockets (IPv6)
1111 tcp6 TCP sockets (IPv6)
1112 raw6 Raw device statistics (IPv6)
1113 igmp6 IP multicast addresses, which this host joined (IPv6)
1114 if_inet6 List of IPv6 interface addresses
1115 ipv6_route Kernel routing table for IPv6
1116 rt6_stats Global IPv6 routing tables statistics
1117 sockstat6 Socket statistics (IPv6)
1118 snmp6 Snmp data (IPv6)
1119 ..............................................................................
1122 Table 1-9: Network info in /proc/net
1123 ..............................................................................
1124 File Content
1125 arp Kernel ARP table
1126 dev network devices with statistics
1127 dev_mcast the Layer2 multicast groups a device is listening too
1128 (interface index, label, number of references, number of bound
1129 addresses).
1130 dev_stat network device status
1131 ip_fwchains Firewall chain linkage
1132 ip_fwnames Firewall chain names
1133 ip_masq Directory containing the masquerading tables
1134 ip_masquerade Major masquerading table
1135 netstat Network statistics
1136 raw raw device statistics
1137 route Kernel routing table
1138 rpc Directory containing rpc info
1139 rt_cache Routing cache
1140 snmp SNMP data
1141 sockstat Socket statistics
1142 tcp TCP sockets
1143 udp UDP sockets
1144 unix UNIX domain sockets
1145 wireless Wireless interface data (Wavelan etc)
1146 igmp IP multicast addresses, which this host joined
1147 psched Global packet scheduler parameters.
1148 netlink List of PF_NETLINK sockets
1149 ip_mr_vifs List of multicast virtual interfaces
1150 ip_mr_cache List of multicast routing cache
1151 ..............................................................................
1153 You can use this information to see which network devices are available in
1154 your system and how much traffic was routed over those devices:
1156 > cat /proc/net/dev
1157 Inter-|Receive |[...
1158 face |bytes packets errs drop fifo frame compressed multicast|[...
1159 lo: 908188 5596 0 0 0 0 0 0 [...
1160 ppp0:15475140 20721 410 0 0 410 0 0 [...
1161 eth0: 614530 7085 0 0 0 0 0 1 [...
1163 ...] Transmit
1164 ...] bytes packets errs drop fifo colls carrier compressed
1165 ...] 908188 5596 0 0 0 0 0 0
1166 ...] 1375103 17405 0 0 0 0 0 0
1167 ...] 1703981 5535 0 0 0 3 0 0
1169 In addition, each Channel Bond interface has its own directory. For
1170 example, the bond0 device will have a directory called /proc/net/bond0/.
1171 It will contain information that is specific to that bond, such as the
1172 current slaves of the bond, the link status of the slaves, and how
1173 many times the slaves link has failed.
1175 1.5 SCSI info
1176 -------------
1178 If you have a SCSI host adapter in your system, you'll find a subdirectory
1179 named after the driver for this adapter in /proc/scsi. You'll also see a list
1180 of all recognized SCSI devices in /proc/scsi:
1182 >cat /proc/scsi/scsi
1183 Attached devices:
1184 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1185 Vendor: IBM Model: DGHS09U Rev: 03E0
1186 Type: Direct-Access ANSI SCSI revision: 03
1187 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1188 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1189 Type: CD-ROM ANSI SCSI revision: 02
1192 The directory named after the driver has one file for each adapter found in
1193 the system. These files contain information about the controller, including
1194 the used IRQ and the IO address range. The amount of information shown is
1195 dependent on the adapter you use. The example shows the output for an Adaptec
1196 AHA-2940 SCSI adapter:
1198 > cat /proc/scsi/aic7xxx/0
1200 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1201 Compile Options:
1202 TCQ Enabled By Default : Disabled
1203 AIC7XXX_PROC_STATS : Disabled
1204 AIC7XXX_RESET_DELAY : 5
1205 Adapter Configuration:
1206 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1207 Ultra Wide Controller
1208 PCI MMAPed I/O Base: 0xeb001000
1209 Adapter SEEPROM Config: SEEPROM found and used.
1210 Adaptec SCSI BIOS: Enabled
1211 IRQ: 10
1212 SCBs: Active 0, Max Active 2,
1213 Allocated 15, HW 16, Page 255
1214 Interrupts: 160328
1215 BIOS Control Word: 0x18b6
1216 Adapter Control Word: 0x005b
1217 Extended Translation: Enabled
1218 Disconnect Enable Flags: 0xffff
1219 Ultra Enable Flags: 0x0001
1220 Tag Queue Enable Flags: 0x0000
1221 Ordered Queue Tag Flags: 0x0000
1222 Default Tag Queue Depth: 8
1223 Tagged Queue By Device array for aic7xxx host instance 0:
1224 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1225 Actual queue depth per device for aic7xxx host instance 0:
1226 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1227 Statistics:
1228 (scsi0:0:0:0)
1229 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1230 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1231 Total transfers 160151 (74577 reads and 85574 writes)
1232 (scsi0:0:6:0)
1233 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1234 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1235 Total transfers 0 (0 reads and 0 writes)
1238 1.6 Parallel port info in /proc/parport
1239 ---------------------------------------
1241 The directory /proc/parport contains information about the parallel ports of
1242 your system. It has one subdirectory for each port, named after the port
1243 number (0,1,2,...).
1245 These directories contain the four files shown in Table 1-10.
1248 Table 1-10: Files in /proc/parport
1249 ..............................................................................
1250 File Content
1251 autoprobe Any IEEE-1284 device ID information that has been acquired.
1252 devices list of the device drivers using that port. A + will appear by the
1253 name of the device currently using the port (it might not appear
1254 against any).
1255 hardware Parallel port's base address, IRQ line and DMA channel.
1256 irq IRQ that parport is using for that port. This is in a separate
1257 file to allow you to alter it by writing a new value in (IRQ
1258 number or none).
1259 ..............................................................................
1261 1.7 TTY info in /proc/tty
1262 -------------------------
1264 Information about the available and actually used tty's can be found in the
1265 directory /proc/tty.You'll find entries for drivers and line disciplines in
1266 this directory, as shown in Table 1-11.
1269 Table 1-11: Files in /proc/tty
1270 ..............................................................................
1271 File Content
1272 drivers list of drivers and their usage
1273 ldiscs registered line disciplines
1274 driver/serial usage statistic and status of single tty lines
1275 ..............................................................................
1277 To see which tty's are currently in use, you can simply look into the file
1278 /proc/tty/drivers:
1280 > cat /proc/tty/drivers
1281 pty_slave /dev/pts 136 0-255 pty:slave
1282 pty_master /dev/ptm 128 0-255 pty:master
1283 pty_slave /dev/ttyp 3 0-255 pty:slave
1284 pty_master /dev/pty 2 0-255 pty:master
1285 serial /dev/cua 5 64-67 serial:callout
1286 serial /dev/ttyS 4 64-67 serial
1287 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1288 /dev/ptmx /dev/ptmx 5 2 system
1289 /dev/console /dev/console 5 1 system:console
1290 /dev/tty /dev/tty 5 0 system:/dev/tty
1291 unknown /dev/tty 4 1-63 console
1294 1.8 Miscellaneous kernel statistics in /proc/stat
1295 -------------------------------------------------
1297 Various pieces of information about kernel activity are available in the
1298 /proc/stat file. All of the numbers reported in this file are aggregates
1299 since the system first booted. For a quick look, simply cat the file:
1301 > cat /proc/stat
1302 cpu 2255 34 2290 22625563 6290 127 456 0 0 0
1303 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0
1304 cpu1 1123 0 849 11313845 2614 0 18 0 0 0
1305 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1306 ctxt 1990473
1307 btime 1062191376
1308 processes 2915
1309 procs_running 1
1310 procs_blocked 0
1311 softirq 183433 0 21755 12 39 1137 231 21459 2263
1313 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1314 lines. These numbers identify the amount of time the CPU has spent performing
1315 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1316 second). The meanings of the columns are as follows, from left to right:
1318 - user: normal processes executing in user mode
1319 - nice: niced processes executing in user mode
1320 - system: processes executing in kernel mode
1321 - idle: twiddling thumbs
1322 - iowait: In a word, iowait stands for waiting for I/O to complete. But there
1323 are several problems:
1324 1. Cpu will not wait for I/O to complete, iowait is the time that a task is
1325 waiting for I/O to complete. When cpu goes into idle state for
1326 outstanding task io, another task will be scheduled on this CPU.
1327 2. In a multi-core CPU, the task waiting for I/O to complete is not running
1328 on any CPU, so the iowait of each CPU is difficult to calculate.
1329 3. The value of iowait field in /proc/stat will decrease in certain
1330 conditions.
1331 So, the iowait is not reliable by reading from /proc/stat.
1332 - irq: servicing interrupts
1333 - softirq: servicing softirqs
1334 - steal: involuntary wait
1335 - guest: running a normal guest
1336 - guest_nice: running a niced guest
1338 The "intr" line gives counts of interrupts serviced since boot time, for each
1339 of the possible system interrupts. The first column is the total of all
1340 interrupts serviced including unnumbered architecture specific interrupts;
1341 each subsequent column is the total for that particular numbered interrupt.
1342 Unnumbered interrupts are not shown, only summed into the total.
1344 The "ctxt" line gives the total number of context switches across all CPUs.
1346 The "btime" line gives the time at which the system booted, in seconds since
1347 the Unix epoch.
1349 The "processes" line gives the number of processes and threads created, which
1350 includes (but is not limited to) those created by calls to the fork() and
1351 clone() system calls.
1353 The "procs_running" line gives the total number of threads that are
1354 running or ready to run (i.e., the total number of runnable threads).
1356 The "procs_blocked" line gives the number of processes currently blocked,
1357 waiting for I/O to complete.
1359 The "softirq" line gives counts of softirqs serviced since boot time, for each
1360 of the possible system softirqs. The first column is the total of all
1361 softirqs serviced; each subsequent column is the total for that particular
1362 softirq.
1365 1.9 Ext4 file system parameters
1366 -------------------------------
1368 Information about mounted ext4 file systems can be found in
1369 /proc/fs/ext4. Each mounted filesystem will have a directory in
1370 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1371 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1372 in Table 1-12, below.
1374 Table 1-12: Files in /proc/fs/ext4/<devname>
1375 ..............................................................................
1376 File Content
1377 mb_groups details of multiblock allocator buddy cache of free blocks
1378 ..............................................................................
1380 2.0 /proc/consoles
1381 ------------------
1382 Shows registered system console lines.
1384 To see which character device lines are currently used for the system console
1385 /dev/console, you may simply look into the file /proc/consoles:
1387 > cat /proc/consoles
1388 tty0 -WU (ECp) 4:7
1389 ttyS0 -W- (Ep) 4:64
1391 The columns are:
1393 device name of the device
1394 operations R = can do read operations
1395 W = can do write operations
1396 U = can do unblank
1397 flags E = it is enabled
1398 C = it is preferred console
1399 B = it is primary boot console
1400 p = it is used for printk buffer
1401 b = it is not a TTY but a Braille device
1402 a = it is safe to use when cpu is offline
1403 major:minor major and minor number of the device separated by a colon
1405 ------------------------------------------------------------------------------
1406 Summary
1407 ------------------------------------------------------------------------------
1408 The /proc file system serves information about the running system. It not only
1409 allows access to process data but also allows you to request the kernel status
1410 by reading files in the hierarchy.
1412 The directory structure of /proc reflects the types of information and makes
1413 it easy, if not obvious, where to look for specific data.
1414 ------------------------------------------------------------------------------
1416 ------------------------------------------------------------------------------
1417 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1418 ------------------------------------------------------------------------------
1420 ------------------------------------------------------------------------------
1421 In This Chapter
1422 ------------------------------------------------------------------------------
1423 * Modifying kernel parameters by writing into files found in /proc/sys
1424 * Exploring the files which modify certain parameters
1425 * Review of the /proc/sys file tree
1426 ------------------------------------------------------------------------------
1429 A very interesting part of /proc is the directory /proc/sys. This is not only
1430 a source of information, it also allows you to change parameters within the
1431 kernel. Be very careful when attempting this. You can optimize your system,
1432 but you can also cause it to crash. Never alter kernel parameters on a
1433 production system. Set up a development machine and test to make sure that
1434 everything works the way you want it to. You may have no alternative but to
1435 reboot the machine once an error has been made.
1437 To change a value, simply echo the new value into the file. An example is
1438 given below in the section on the file system data. You need to be root to do
1439 this. You can create your own boot script to perform this every time your
1440 system boots.
1442 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1443 general things in the operation of the Linux kernel. Since some of the files
1444 can inadvertently disrupt your system, it is advisable to read both
1445 documentation and source before actually making adjustments. In any case, be
1446 very careful when writing to any of these files. The entries in /proc may
1447 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1448 review the kernel documentation in the directory /usr/src/linux/Documentation.
1449 This chapter is heavily based on the documentation included in the pre 2.2
1450 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1452 Please see: Documentation/sysctl/ directory for descriptions of these
1453 entries.
1455 ------------------------------------------------------------------------------
1456 Summary
1457 ------------------------------------------------------------------------------
1458 Certain aspects of kernel behavior can be modified at runtime, without the
1459 need to recompile the kernel, or even to reboot the system. The files in the
1460 /proc/sys tree can not only be read, but also modified. You can use the echo
1461 command to write value into these files, thereby changing the default settings
1462 of the kernel.
1463 ------------------------------------------------------------------------------
1465 ------------------------------------------------------------------------------
1466 CHAPTER 3: PER-PROCESS PARAMETERS
1467 ------------------------------------------------------------------------------
1469 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1470 --------------------------------------------------------------------------------
1472 These file can be used to adjust the badness heuristic used to select which
1473 process gets killed in out of memory conditions.
1475 The badness heuristic assigns a value to each candidate task ranging from 0
1476 (never kill) to 1000 (always kill) to determine which process is targeted. The
1477 units are roughly a proportion along that range of allowed memory the process
1478 may allocate from based on an estimation of its current memory and swap use.
1479 For example, if a task is using all allowed memory, its badness score will be
1480 1000. If it is using half of its allowed memory, its score will be 500.
1482 There is an additional factor included in the badness score: the current memory
1483 and swap usage is discounted by 3% for root processes.
1485 The amount of "allowed" memory depends on the context in which the oom killer
1486 was called. If it is due to the memory assigned to the allocating task's cpuset
1487 being exhausted, the allowed memory represents the set of mems assigned to that
1488 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1489 memory represents the set of mempolicy nodes. If it is due to a memory
1490 limit (or swap limit) being reached, the allowed memory is that configured
1491 limit. Finally, if it is due to the entire system being out of memory, the
1492 allowed memory represents all allocatable resources.
1494 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1495 is used to determine which task to kill. Acceptable values range from -1000
1496 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1497 polarize the preference for oom killing either by always preferring a certain
1498 task or completely disabling it. The lowest possible value, -1000, is
1499 equivalent to disabling oom killing entirely for that task since it will always
1500 report a badness score of 0.
1502 Consequently, it is very simple for userspace to define the amount of memory to
1503 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1504 example, is roughly equivalent to allowing the remainder of tasks sharing the
1505 same system, cpuset, mempolicy, or memory controller resources to use at least
1506 50% more memory. A value of -500, on the other hand, would be roughly
1507 equivalent to discounting 50% of the task's allowed memory from being considered
1508 as scoring against the task.
1510 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1511 be used to tune the badness score. Its acceptable values range from -16
1512 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1513 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1514 scaled linearly with /proc/<pid>/oom_score_adj.
1516 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1517 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1518 requires CAP_SYS_RESOURCE.
1520 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1521 generation children with separate address spaces instead, if possible. This
1522 avoids servers and important system daemons from being killed and loses the
1523 minimal amount of work.
1526 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1527 -------------------------------------------------------------
1529 This file can be used to check the current score used by the oom-killer is for
1530 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1531 process should be killed in an out-of-memory situation.
1534 3.3 /proc/<pid>/io - Display the IO accounting fields
1535 -------------------------------------------------------
1537 This file contains IO statistics for each running process
1539 Example
1540 -------
1542 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1543 [1] 3828
1545 test:/tmp # cat /proc/3828/io
1546 rchar: 323934931
1547 wchar: 323929600
1548 syscr: 632687
1549 syscw: 632675
1550 read_bytes: 0
1551 write_bytes: 323932160
1552 cancelled_write_bytes: 0
1555 Description
1556 -----------
1558 rchar
1559 -----
1561 I/O counter: chars read
1562 The number of bytes which this task has caused to be read from storage. This
1563 is simply the sum of bytes which this process passed to read() and pread().
1564 It includes things like tty IO and it is unaffected by whether or not actual
1565 physical disk IO was required (the read might have been satisfied from
1566 pagecache)
1569 wchar
1570 -----
1572 I/O counter: chars written
1573 The number of bytes which this task has caused, or shall cause to be written
1574 to disk. Similar caveats apply here as with rchar.
1577 syscr
1578 -----
1580 I/O counter: read syscalls
1581 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1582 and pread().
1585 syscw
1586 -----
1588 I/O counter: write syscalls
1589 Attempt to count the number of write I/O operations, i.e. syscalls like
1590 write() and pwrite().
1593 read_bytes
1594 ----------
1596 I/O counter: bytes read
1597 Attempt to count the number of bytes which this process really did cause to
1598 be fetched from the storage layer. Done at the submit_bio() level, so it is
1599 accurate for block-backed filesystems. <please add status regarding NFS and
1600 CIFS at a later time>
1603 write_bytes
1604 -----------
1606 I/O counter: bytes written
1607 Attempt to count the number of bytes which this process caused to be sent to
1608 the storage layer. This is done at page-dirtying time.
1611 cancelled_write_bytes
1612 ---------------------
1614 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1615 then deletes the file, it will in fact perform no writeout. But it will have
1616 been accounted as having caused 1MB of write.
1617 In other words: The number of bytes which this process caused to not happen,
1618 by truncating pagecache. A task can cause "negative" IO too. If this task
1619 truncates some dirty pagecache, some IO which another task has been accounted
1620 for (in its write_bytes) will not be happening. We _could_ just subtract that
1621 from the truncating task's write_bytes, but there is information loss in doing
1622 that.
1625 Note
1626 ----
1628 At its current implementation state, this is a bit racy on 32-bit machines: if
1629 process A reads process B's /proc/pid/io while process B is updating one of
1630 those 64-bit counters, process A could see an intermediate result.
1633 More information about this can be found within the taskstats documentation in
1634 Documentation/accounting.
1636 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1637 ---------------------------------------------------------------
1638 When a process is dumped, all anonymous memory is written to a core file as
1639 long as the size of the core file isn't limited. But sometimes we don't want
1640 to dump some memory segments, for example, huge shared memory or DAX.
1641 Conversely, sometimes we want to save file-backed memory segments into a core
1642 file, not only the individual files.
1644 /proc/<pid>/coredump_filter allows you to customize which memory segments
1645 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1646 of memory types. If a bit of the bitmask is set, memory segments of the
1647 corresponding memory type are dumped, otherwise they are not dumped.
1649 The following 9 memory types are supported:
1650 - (bit 0) anonymous private memory
1651 - (bit 1) anonymous shared memory
1652 - (bit 2) file-backed private memory
1653 - (bit 3) file-backed shared memory
1654 - (bit 4) ELF header pages in file-backed private memory areas (it is
1655 effective only if the bit 2 is cleared)
1656 - (bit 5) hugetlb private memory
1657 - (bit 6) hugetlb shared memory
1658 - (bit 7) DAX private memory
1659 - (bit 8) DAX shared memory
1661 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1662 are always dumped regardless of the bitmask status.
1664 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1665 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1667 The default value of coredump_filter is 0x33; this means all anonymous memory
1668 segments, ELF header pages and hugetlb private memory are dumped.
1670 If you don't want to dump all shared memory segments attached to pid 1234,
1671 write 0x31 to the process's proc file.
1673 $ echo 0x31 > /proc/1234/coredump_filter
1675 When a new process is created, the process inherits the bitmask status from its
1676 parent. It is useful to set up coredump_filter before the program runs.
1677 For example:
1679 $ echo 0x7 > /proc/self/coredump_filter
1680 $ ./some_program
1682 3.5 /proc/<pid>/mountinfo - Information about mounts
1683 --------------------------------------------------------
1685 This file contains lines of the form:
1687 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1688 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1690 (1) mount ID: unique identifier of the mount (may be reused after umount)
1691 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1692 (3) major:minor: value of st_dev for files on filesystem
1693 (4) root: root of the mount within the filesystem
1694 (5) mount point: mount point relative to the process's root
1695 (6) mount options: per mount options
1696 (7) optional fields: zero or more fields of the form "tag[:value]"
1697 (8) separator: marks the end of the optional fields
1698 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1699 (10) mount source: filesystem specific information or "none"
1700 (11) super options: per super block options
1702 Parsers should ignore all unrecognised optional fields. Currently the
1703 possible optional fields are:
1705 shared:X mount is shared in peer group X
1706 master:X mount is slave to peer group X
1707 propagate_from:X mount is slave and receives propagation from peer group X (*)
1708 unbindable mount is unbindable
1710 (*) X is the closest dominant peer group under the process's root. If
1711 X is the immediate master of the mount, or if there's no dominant peer
1712 group under the same root, then only the "master:X" field is present
1713 and not the "propagate_from:X" field.
1715 For more information on mount propagation see:
1717 Documentation/filesystems/sharedsubtree.txt
1720 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1721 --------------------------------------------------------
1722 These files provide a method to access a tasks comm value. It also allows for
1723 a task to set its own or one of its thread siblings comm value. The comm value
1724 is limited in size compared to the cmdline value, so writing anything longer
1725 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1726 comm value.
1729 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1730 -------------------------------------------------------------------------
1731 This file provides a fast way to retrieve first level children pids
1732 of a task pointed by <pid>/<tid> pair. The format is a space separated
1733 stream of pids.
1735 Note the "first level" here -- if a child has own children they will
1736 not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1737 to obtain the descendants.
1739 Since this interface is intended to be fast and cheap it doesn't
1740 guarantee to provide precise results and some children might be
1741 skipped, especially if they've exited right after we printed their
1742 pids, so one need to either stop or freeze processes being inspected
1743 if precise results are needed.
1746 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1747 ---------------------------------------------------------------
1748 This file provides information associated with an opened file. The regular
1749 files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos'
1750 represents the current offset of the opened file in decimal form [see lseek(2)
1751 for details], 'flags' denotes the octal O_xxx mask the file has been
1752 created with [see open(2) for details] and 'mnt_id' represents mount ID of
1753 the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo
1754 for details].
1756 A typical output is
1758 pos: 0
1759 flags: 0100002
1760 mnt_id: 19
1762 All locks associated with a file descriptor are shown in its fdinfo too.
1764 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1766 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1767 pair provide additional information particular to the objects they represent.
1769 Eventfd files
1770 ~~~~~~~~~~~~~
1771 pos: 0
1772 flags: 04002
1773 mnt_id: 9
1774 eventfd-count: 5a
1776 where 'eventfd-count' is hex value of a counter.
1778 Signalfd files
1779 ~~~~~~~~~~~~~~
1780 pos: 0
1781 flags: 04002
1782 mnt_id: 9
1783 sigmask: 0000000000000200
1785 where 'sigmask' is hex value of the signal mask associated
1786 with a file.
1788 Epoll files
1789 ~~~~~~~~~~~
1790 pos: 0
1791 flags: 02
1792 mnt_id: 9
1793 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7
1795 where 'tfd' is a target file descriptor number in decimal form,
1796 'events' is events mask being watched and the 'data' is data
1797 associated with a target [see epoll(7) for more details].
1799 The 'pos' is current offset of the target file in decimal form
1800 [see lseek(2)], 'ino' and 'sdev' are inode and device numbers
1801 where target file resides, all in hex format.
1803 Fsnotify files
1804 ~~~~~~~~~~~~~~
1805 For inotify files the format is the following
1807 pos: 0
1808 flags: 02000000
1809 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1811 where 'wd' is a watch descriptor in decimal form, ie a target file
1812 descriptor number, 'ino' and 'sdev' are inode and device where the
1813 target file resides and the 'mask' is the mask of events, all in hex
1814 form [see inotify(7) for more details].
1816 If the kernel was built with exportfs support, the path to the target
1817 file is encoded as a file handle. The file handle is provided by three
1818 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1819 format.
1821 If the kernel is built without exportfs support the file handle won't be
1822 printed out.
1824 If there is no inotify mark attached yet the 'inotify' line will be omitted.
1826 For fanotify files the format is
1828 pos: 0
1829 flags: 02
1830 mnt_id: 9
1831 fanotify flags:10 event-flags:0
1832 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1833 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
1835 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
1836 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
1837 flags associated with mark which are tracked separately from events
1838 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
1839 mask and 'ignored_mask' is the mask of events which are to be ignored.
1840 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
1841 does provide information about flags and mask used in fanotify_mark
1842 call [see fsnotify manpage for details].
1844 While the first three lines are mandatory and always printed, the rest is
1845 optional and may be omitted if no marks created yet.
1847 Timerfd files
1848 ~~~~~~~~~~~~~
1850 pos: 0
1851 flags: 02
1852 mnt_id: 9
1853 clockid: 0
1854 ticks: 0
1855 settime flags: 01
1856 it_value: (0, 49406829)
1857 it_interval: (1, 0)
1859 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
1860 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
1861 flags in octal form been used to setup the timer [see timerfd_settime(2) for
1862 details]. 'it_value' is remaining time until the timer exiration.
1863 'it_interval' is the interval for the timer. Note the timer might be set up
1864 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
1865 still exhibits timer's remaining time.
1867 3.9 /proc/<pid>/map_files - Information about memory mapped files
1868 ---------------------------------------------------------------------
1869 This directory contains symbolic links which represent memory mapped files
1870 the process is maintaining. Example output:
1872 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
1873 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
1874 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
1875 | ...
1876 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
1877 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
1879 The name of a link represents the virtual memory bounds of a mapping, i.e.
1880 vm_area_struct::vm_start-vm_area_struct::vm_end.
1882 The main purpose of the map_files is to retrieve a set of memory mapped
1883 files in a fast way instead of parsing /proc/<pid>/maps or
1884 /proc/<pid>/smaps, both of which contain many more records. At the same
1885 time one can open(2) mappings from the listings of two processes and
1886 comparing their inode numbers to figure out which anonymous memory areas
1887 are actually shared.
1889 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
1890 ---------------------------------------------------------
1891 This file provides the value of the task's timerslack value in nanoseconds.
1892 This value specifies a amount of time that normal timers may be deferred
1893 in order to coalesce timers and avoid unnecessary wakeups.
1895 This allows a task's interactivity vs power consumption trade off to be
1896 adjusted.
1898 Writing 0 to the file will set the tasks timerslack to the default value.
1900 Valid values are from 0 - ULLONG_MAX
1902 An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
1903 permissions on the task specified to change its timerslack_ns value.
1905 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
1906 -----------------------------------------------------------------
1907 When CONFIG_LIVEPATCH is enabled, this file displays the value of the
1908 patch state for the task.
1910 A value of '-1' indicates that no patch is in transition.
1912 A value of '0' indicates that a patch is in transition and the task is
1913 unpatched. If the patch is being enabled, then the task hasn't been
1914 patched yet. If the patch is being disabled, then the task has already
1915 been unpatched.
1917 A value of '1' indicates that a patch is in transition and the task is
1918 patched. If the patch is being enabled, then the task has already been
1919 patched. If the patch is being disabled, then the task hasn't been
1920 unpatched yet.
1923 ------------------------------------------------------------------------------
1924 Configuring procfs
1925 ------------------------------------------------------------------------------
1927 4.1 Mount options
1928 ---------------------
1930 The following mount options are supported:
1932 hidepid= Set /proc/<pid>/ access mode.
1933 gid= Set the group authorized to learn processes information.
1935 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1936 (default).
1938 hidepid=1 means users may not access any /proc/<pid>/ directories but their
1939 own. Sensitive files like cmdline, sched*, status are now protected against
1940 other users. This makes it impossible to learn whether any user runs
1941 specific program (given the program doesn't reveal itself by its behaviour).
1942 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1943 poorly written programs passing sensitive information via program arguments are
1944 now protected against local eavesdroppers.
1946 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1947 users. It doesn't mean that it hides a fact whether a process with a specific
1948 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1949 but it hides process' uid and gid, which may be learned by stat()'ing
1950 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
1951 information about running processes, whether some daemon runs with elevated
1952 privileges, whether other user runs some sensitive program, whether other users
1953 run any program at all, etc.
1955 gid= defines a group authorized to learn processes information otherwise
1956 prohibited by hidepid=. If you use some daemon like identd which needs to learn
1957 information about processes information, just add identd to this group.