| blob | fde9fd06fa988b2eac60f10880973351e783fe37 |
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
47 4 Configuring procfs
48 4.1 Mount options
50 ------------------------------------------------------------------------------
51 Preface
52 ------------------------------------------------------------------------------
54 0.1 Introduction/Credits
55 ------------------------
57 This documentation is part of a soon (or so we hope) to be released book on
58 the SuSE Linux distribution. As there is no complete documentation for the
59 /proc file system and we've used many freely available sources to write these
60 chapters, it seems only fair to give the work back to the Linux community.
61 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
62 afraid it's still far from complete, but we hope it will be useful. As far as
63 we know, it is the first 'all-in-one' document about the /proc file system. It
64 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
65 SPARC, AXP, etc., features, you probably won't find what you are looking for.
66 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
67 additions and patches are welcome and will be added to this document if you
68 mail them to Bodo.
70 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
71 other people for help compiling this documentation. We'd also like to extend a
72 special thank you to Andi Kleen for documentation, which we relied on heavily
73 to create this document, as well as the additional information he provided.
74 Thanks to everybody else who contributed source or docs to the Linux kernel
75 and helped create a great piece of software... :)
77 If you have any comments, corrections or additions, please don't hesitate to
78 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
79 document.
81 The latest version of this document is available online at
82 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
84 If the above direction does not works for you, you could try the kernel
85 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
86 comandante@zaralinux.com.
88 0.2 Legal Stuff
89 ---------------
91 We don't guarantee the correctness of this document, and if you come to us
92 complaining about how you screwed up your system because of incorrect
93 documentation, we won't feel responsible...
95 ------------------------------------------------------------------------------
96 CHAPTER 1: COLLECTING SYSTEM INFORMATION
97 ------------------------------------------------------------------------------
99 ------------------------------------------------------------------------------
100 In This Chapter
101 ------------------------------------------------------------------------------
102 * Investigating the properties of the pseudo file system /proc and its
103 ability to provide information on the running Linux system
104 * Examining /proc's structure
105 * Uncovering various information about the kernel and the processes running
106 on the system
107 ------------------------------------------------------------------------------
110 The proc file system acts as an interface to internal data structures in the
111 kernel. It can be used to obtain information about the system and to change
112 certain kernel parameters at runtime (sysctl).
114 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
115 show you how you can use /proc/sys to change settings.
117 1.1 Process-Specific Subdirectories
118 -----------------------------------
120 The directory /proc contains (among other things) one subdirectory for each
121 process running on the system, which is named after the process ID (PID).
123 The link self points to the process reading the file system. Each process
124 subdirectory has the entries listed in Table 1-1.
127 Table 1-1: Process specific entries in /proc
128 ..............................................................................
129 File Content
130 clear_refs Clears page referenced bits shown in smaps output
131 cmdline Command line arguments
132 cpu Current and last cpu in which it was executed (2.4)(smp)
133 cwd Link to the current working directory
134 environ Values of environment variables
135 exe Link to the executable of this process
136 fd Directory, which contains all file descriptors
137 maps Memory maps to executables and library files (2.4)
138 mem Memory held by this process
139 root Link to the root directory of this process
140 stat Process status
141 statm Process memory status information
142 status Process status in human readable form
143 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function
144 symbol the task is blocked in - or "0" if not blocked.
145 pagemap Page table
146 stack Report full stack trace, enable via CONFIG_STACKTRACE
147 smaps a extension based on maps, showing the memory consumption of
148 each mapping and flags associated with it
149 numa_maps an extension based on maps, showing the memory locality and
150 binding policy as well as mem usage (in pages) of each mapping.
151 ..............................................................................
153 For example, to get the status information of a process, all you have to do is
154 read the file /proc/PID/status:
156 >cat /proc/self/status
157 Name: cat
158 State: R (running)
159 Tgid: 5452
160 Pid: 5452
161 PPid: 743
162 TracerPid: 0 (2.4)
163 Uid: 501 501 501 501
164 Gid: 100 100 100 100
165 FDSize: 256
166 Groups: 100 14 16
167 VmPeak: 5004 kB
168 VmSize: 5004 kB
169 VmLck: 0 kB
170 VmHWM: 476 kB
171 VmRSS: 476 kB
172 RssAnon: 352 kB
173 RssFile: 120 kB
174 RssShmem: 4 kB
175 VmData: 156 kB
176 VmStk: 88 kB
177 VmExe: 68 kB
178 VmLib: 1412 kB
179 VmPTE: 20 kb
180 VmSwap: 0 kB
181 HugetlbPages: 0 kB
182 Threads: 1
183 SigQ: 0/28578
184 SigPnd: 0000000000000000
185 ShdPnd: 0000000000000000
186 SigBlk: 0000000000000000
187 SigIgn: 0000000000000000
188 SigCgt: 0000000000000000
189 CapInh: 00000000fffffeff
190 CapPrm: 0000000000000000
191 CapEff: 0000000000000000
192 CapBnd: ffffffffffffffff
193 Seccomp: 0
194 voluntary_ctxt_switches: 0
195 nonvoluntary_ctxt_switches: 1
197 This shows you nearly the same information you would get if you viewed it with
198 the ps command. In fact, ps uses the proc file system to obtain its
199 information. But you get a more detailed view of the process by reading the
200 file /proc/PID/status. It fields are described in table 1-2.
202 The statm file contains more detailed information about the process
203 memory usage. Its seven fields are explained in Table 1-3. The stat file
204 contains details information about the process itself. Its fields are
205 explained in Table 1-4.
207 (for SMP CONFIG users)
208 For making accounting scalable, RSS related information are handled in an
209 asynchronous manner and the value may not be very precise. To see a precise
210 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
211 It's slow but very precise.
213 Table 1-2: Contents of the status files (as of 4.1)
214 ..............................................................................
215 Field Content
216 Name filename of the executable
217 State state (R is running, S is sleeping, D is sleeping
218 in an uninterruptible wait, Z is zombie,
219 T is traced or stopped)
220 Tgid thread group ID
221 Ngid NUMA group ID (0 if none)
222 Pid process id
223 PPid process id of the parent process
224 TracerPid PID of process tracing this process (0 if not)
225 Uid Real, effective, saved set, and file system UIDs
226 Gid Real, effective, saved set, and file system GIDs
227 FDSize number of file descriptor slots currently allocated
228 Groups supplementary group list
229 NStgid descendant namespace thread group ID hierarchy
230 NSpid descendant namespace process ID hierarchy
231 NSpgid descendant namespace process group ID hierarchy
232 NSsid descendant namespace session ID hierarchy
233 VmPeak peak virtual memory size
234 VmSize total program size
235 VmLck locked memory size
236 VmHWM peak resident set size ("high water mark")
237 VmRSS size of memory portions. It contains the three
238 following parts (VmRSS = RssAnon + RssFile + RssShmem)
239 RssAnon size of resident anonymous memory
240 RssFile size of resident file mappings
241 RssShmem size of resident shmem memory (includes SysV shm,
242 mapping of tmpfs and shared anonymous mappings)
243 VmData size of data, stack, and text segments
244 VmStk size of data, stack, and text segments
245 VmExe size of text segment
246 VmLib size of shared library code
247 VmPTE size of page table entries
248 VmPMD size of second level page tables
249 VmSwap amount of swap used by anonymous private data
250 (shmem swap usage is not included)
251 HugetlbPages size of hugetlb memory portions
252 Threads number of threads
253 SigQ number of signals queued/max. number for queue
254 SigPnd bitmap of pending signals for the thread
255 ShdPnd bitmap of shared pending signals for the process
256 SigBlk bitmap of blocked signals
257 SigIgn bitmap of ignored signals
258 SigCgt bitmap of caught signals
259 CapInh bitmap of inheritable capabilities
260 CapPrm bitmap of permitted capabilities
261 CapEff bitmap of effective capabilities
262 CapBnd bitmap of capabilities bounding set
263 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
264 Cpus_allowed mask of CPUs on which this process may run
265 Cpus_allowed_list Same as previous, but in "list format"
266 Mems_allowed mask of memory nodes allowed to this process
267 Mems_allowed_list Same as previous, but in "list format"
268 voluntary_ctxt_switches number of voluntary context switches
269 nonvoluntary_ctxt_switches number of non voluntary context switches
270 ..............................................................................
272 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
273 ..............................................................................
274 Field Content
275 size total program size (pages) (same as VmSize in status)
276 resident size of memory portions (pages) (same as VmRSS in status)
277 shared number of pages that are shared (i.e. backed by a file, same
278 as RssFile+RssShmem in status)
279 trs number of pages that are 'code' (not including libs; broken,
280 includes data segment)
281 lrs number of pages of library (always 0 on 2.6)
282 drs number of pages of data/stack (including libs; broken,
283 includes library text)
284 dt number of dirty pages (always 0 on 2.6)
285 ..............................................................................
288 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
289 ..............................................................................
290 Field Content
291 pid process id
292 tcomm filename of the executable
293 state state (R is running, S is sleeping, D is sleeping in an
294 uninterruptible wait, Z is zombie, T is traced or stopped)
295 ppid process id of the parent process
296 pgrp pgrp of the process
297 sid session id
298 tty_nr tty the process uses
299 tty_pgrp pgrp of the tty
300 flags task flags
301 min_flt number of minor faults
302 cmin_flt number of minor faults with child's
303 maj_flt number of major faults
304 cmaj_flt number of major faults with child's
305 utime user mode jiffies
306 stime kernel mode jiffies
307 cutime user mode jiffies with child's
308 cstime kernel mode jiffies with child's
309 priority priority level
310 nice nice level
311 num_threads number of threads
312 it_real_value (obsolete, always 0)
313 start_time time the process started after system boot
314 vsize virtual memory size
315 rss resident set memory size
316 rsslim current limit in bytes on the rss
317 start_code address above which program text can run
318 end_code address below which program text can run
319 start_stack address of the start of the main process stack
320 esp current value of ESP
321 eip current value of EIP
322 pending bitmap of pending signals
323 blocked bitmap of blocked signals
324 sigign bitmap of ignored signals
325 sigcatch bitmap of caught signals
326 0 (place holder, used to be the wchan address, use /proc/PID/wchan instead)
327 0 (place holder)
328 0 (place holder)
329 exit_signal signal to send to parent thread on exit
330 task_cpu which CPU the task is scheduled on
331 rt_priority realtime priority
332 policy scheduling policy (man sched_setscheduler)
333 blkio_ticks time spent waiting for block IO
334 gtime guest time of the task in jiffies
335 cgtime guest time of the task children in jiffies
336 start_data address above which program data+bss is placed
337 end_data address below which program data+bss is placed
338 start_brk address above which program heap can be expanded with brk()
339 arg_start address above which program command line is placed
340 arg_end address below which program command line is placed
341 env_start address above which program environment is placed
342 env_end address below which program environment is placed
343 exit_code the thread's exit_code in the form reported by the waitpid system call
344 ..............................................................................
346 The /proc/PID/maps file containing the currently mapped memory regions and
347 their access permissions.
349 The format is:
351 address perms offset dev inode pathname
353 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
354 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
355 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
356 a7cb1000-a7cb2000 ---p 00000000 00:00 0
357 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
358 a7eb2000-a7eb3000 ---p 00000000 00:00 0
359 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack:1001]
360 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
361 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
362 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
363 a800b000-a800e000 rw-p 00000000 00:00 0
364 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
365 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
366 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
367 a8024000-a8027000 rw-p 00000000 00:00 0
368 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
369 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
370 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
371 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
372 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
374 where "address" is the address space in the process that it occupies, "perms"
375 is a set of permissions:
377 r = read
378 w = write
379 x = execute
380 s = shared
381 p = private (copy on write)
383 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
384 "inode" is the inode on that device. 0 indicates that no inode is associated
385 with the memory region, as the case would be with BSS (uninitialized data).
386 The "pathname" shows the name associated file for this mapping. If the mapping
387 is not associated with a file:
389 [heap] = the heap of the program
390 [stack] = the stack of the main process
391 [stack:1001] = the stack of the thread with tid 1001
392 [vdso] = the "virtual dynamic shared object",
393 the kernel system call handler
395 or if empty, the mapping is anonymous.
397 The /proc/PID/task/TID/maps is a view of the virtual memory from the viewpoint
398 of the individual tasks of a process. In this file you will see a mapping marked
399 as [stack] if that task sees it as a stack. This is a key difference from the
400 content of /proc/PID/maps, where you will see all mappings that are being used
401 as stack by all of those tasks. Hence, for the example above, the task-level
402 map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this:
404 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
405 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
406 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
407 a7cb1000-a7cb2000 ---p 00000000 00:00 0
408 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
409 a7eb2000-a7eb3000 ---p 00000000 00:00 0
410 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack]
411 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
412 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
413 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
414 a800b000-a800e000 rw-p 00000000 00:00 0
415 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
416 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
417 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
418 a8024000-a8027000 rw-p 00000000 00:00 0
419 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
420 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
421 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
422 aff35000-aff4a000 rw-p 00000000 00:00 0
423 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
425 The /proc/PID/smaps is an extension based on maps, showing the memory
426 consumption for each of the process's mappings. For each of mappings there
427 is a series of lines such as the following:
429 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
430 Size: 1084 kB
431 Rss: 892 kB
432 Pss: 374 kB
433 Shared_Clean: 892 kB
434 Shared_Dirty: 0 kB
435 Private_Clean: 0 kB
436 Private_Dirty: 0 kB
437 Referenced: 892 kB
438 Anonymous: 0 kB
439 AnonHugePages: 0 kB
440 Shared_Hugetlb: 0 kB
441 Private_Hugetlb: 0 kB
442 Swap: 0 kB
443 SwapPss: 0 kB
444 KernelPageSize: 4 kB
445 MMUPageSize: 4 kB
446 Locked: 0 kB
447 VmFlags: rd ex mr mw me dw
449 the first of these lines shows the same information as is displayed for the
450 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
451 (size), the amount of the mapping that is currently resident in RAM (RSS), the
452 process' proportional share of this mapping (PSS), the number of clean and
453 dirty private pages in the mapping.
455 The "proportional set size" (PSS) of a process is the count of pages it has
456 in memory, where each page is divided by the number of processes sharing it.
457 So if a process has 1000 pages all to itself, and 1000 shared with one other
458 process, its PSS will be 1500.
459 Note that even a page which is part of a MAP_SHARED mapping, but has only
460 a single pte mapped, i.e. is currently used by only one process, is accounted
461 as private and not as shared.
462 "Referenced" indicates the amount of memory currently marked as referenced or
463 accessed.
464 "Anonymous" shows the amount of memory that does not belong to any file. Even
465 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
466 and a page is modified, the file page is replaced by a private anonymous copy.
467 "AnonHugePages" shows the ammount of memory backed by transparent hugepage.
468 "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
469 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
470 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
471 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
472 For shmem mappings, "Swap" includes also the size of the mapped (and not
473 replaced by copy-on-write) part of the underlying shmem object out on swap.
474 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
475 does not take into account swapped out page of underlying shmem objects.
476 "Locked" indicates whether the mapping is locked in memory or not.
478 "VmFlags" field deserves a separate description. This member represents the kernel
479 flags associated with the particular virtual memory area in two letter encoded
480 manner. The codes are the following:
481 rd - readable
482 wr - writeable
483 ex - executable
484 sh - shared
485 mr - may read
486 mw - may write
487 me - may execute
488 ms - may share
489 gd - stack segment growns down
490 pf - pure PFN range
491 dw - disabled write to the mapped file
492 lo - pages are locked in memory
493 io - memory mapped I/O area
494 sr - sequential read advise provided
495 rr - random read advise provided
496 dc - do not copy area on fork
497 de - do not expand area on remapping
498 ac - area is accountable
499 nr - swap space is not reserved for the area
500 ht - area uses huge tlb pages
501 ar - architecture specific flag
502 dd - do not include area into core dump
503 sd - soft-dirty flag
504 mm - mixed map area
505 hg - huge page advise flag
506 nh - no-huge page advise flag
507 mg - mergable advise flag
509 Note that there is no guarantee that every flag and associated mnemonic will
510 be present in all further kernel releases. Things get changed, the flags may
511 be vanished or the reverse -- new added.
513 This file is only present if the CONFIG_MMU kernel configuration option is
514 enabled.
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/vm/soft-dirty.txt for details).
519 To clear the bits for all the pages associated with the process
520 > echo 1 > /proc/PID/clear_refs
522 To clear the bits for the anonymous pages associated with the process
523 > echo 2 > /proc/PID/clear_refs
525 To clear the bits for the file mapped pages associated with the process
526 > echo 3 > /proc/PID/clear_refs
528 To clear the soft-dirty bit
529 > echo 4 > /proc/PID/clear_refs
531 To reset the peak resident set size ("high water mark") to the process's
532 current value:
533 > echo 5 > /proc/PID/clear_refs
535 Any other value written to /proc/PID/clear_refs will have no effect.
537 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
538 using /proc/kpageflags and number of times a page is mapped using
539 /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
541 The /proc/pid/numa_maps is an extension based on maps, showing the memory
542 locality and binding policy, as well as the memory usage (in pages) of
543 each mapping. The output follows a general format where mapping details get
544 summarized separated by blank spaces, one mapping per each file line:
546 address policy mapping details
548 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
549 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
550 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
551 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
552 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
553 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
554 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
555 320698b000 default file=/lib64/libc-2.12.so
556 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
557 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
558 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
559 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
560 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
561 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
562 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
563 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
565 Where:
566 "address" is the starting address for the mapping;
567 "policy" reports the NUMA memory policy set for the mapping (see vm/numa_memory_policy.txt);
568 "mapping details" summarizes mapping data such as mapping type, page usage counters,
569 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
570 size, in KB, that is backing the mapping up.
572 1.2 Kernel data
573 ---------------
575 Similar to the process entries, the kernel data files give information about
576 the running kernel. The files used to obtain this information are contained in
577 /proc and are listed in Table 1-5. Not all of these will be present in your
578 system. It depends on the kernel configuration and the loaded modules, which
579 files are there, and which are missing.
581 Table 1-5: Kernel info in /proc
582 ..............................................................................
583 File Content
584 apm Advanced power management info
585 buddyinfo Kernel memory allocator information (see text) (2.5)
586 bus Directory containing bus specific information
587 cmdline Kernel command line
588 cpuinfo Info about the CPU
589 devices Available devices (block and character)
590 dma Used DMS channels
591 filesystems Supported filesystems
592 driver Various drivers grouped here, currently rtc (2.4)
593 execdomains Execdomains, related to security (2.4)
594 fb Frame Buffer devices (2.4)
595 fs File system parameters, currently nfs/exports (2.4)
596 ide Directory containing info about the IDE subsystem
597 interrupts Interrupt usage
598 iomem Memory map (2.4)
599 ioports I/O port usage
600 irq Masks for irq to cpu affinity (2.4)(smp?)
601 isapnp ISA PnP (Plug&Play) Info (2.4)
602 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
603 kmsg Kernel messages
604 ksyms Kernel symbol table
605 loadavg Load average of last 1, 5 & 15 minutes
606 locks Kernel locks
607 meminfo Memory info
608 misc Miscellaneous
609 modules List of loaded modules
610 mounts Mounted filesystems
611 net Networking info (see text)
612 pagetypeinfo Additional page allocator information (see text) (2.5)
613 partitions Table of partitions known to the system
614 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
615 decoupled by lspci (2.4)
616 rtc Real time clock
617 scsi SCSI info (see text)
618 slabinfo Slab pool info
619 softirqs softirq usage
620 stat Overall statistics
621 swaps Swap space utilization
622 sys See chapter 2
623 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
624 tty Info of tty drivers
625 uptime Wall clock since boot, combined idle time of all cpus
626 version Kernel version
627 video bttv info of video resources (2.4)
628 vmallocinfo Show vmalloced areas
629 ..............................................................................
631 You can, for example, check which interrupts are currently in use and what
632 they are used for by looking in the file /proc/interrupts:
634 > cat /proc/interrupts
635 CPU0
636 0: 8728810 XT-PIC timer
637 1: 895 XT-PIC keyboard
638 2: 0 XT-PIC cascade
639 3: 531695 XT-PIC aha152x
640 4: 2014133 XT-PIC serial
641 5: 44401 XT-PIC pcnet_cs
642 8: 2 XT-PIC rtc
643 11: 8 XT-PIC i82365
644 12: 182918 XT-PIC PS/2 Mouse
645 13: 1 XT-PIC fpu
646 14: 1232265 XT-PIC ide0
647 15: 7 XT-PIC ide1
648 NMI: 0
650 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
651 output of a SMP machine):
653 > cat /proc/interrupts
655 CPU0 CPU1
656 0: 1243498 1214548 IO-APIC-edge timer
657 1: 8949 8958 IO-APIC-edge keyboard
658 2: 0 0 XT-PIC cascade
659 5: 11286 10161 IO-APIC-edge soundblaster
660 8: 1 0 IO-APIC-edge rtc
661 9: 27422 27407 IO-APIC-edge 3c503
662 12: 113645 113873 IO-APIC-edge PS/2 Mouse
663 13: 0 0 XT-PIC fpu
664 14: 22491 24012 IO-APIC-edge ide0
665 15: 2183 2415 IO-APIC-edge ide1
666 17: 30564 30414 IO-APIC-level eth0
667 18: 177 164 IO-APIC-level bttv
668 NMI: 2457961 2457959
669 LOC: 2457882 2457881
670 ERR: 2155
672 NMI is incremented in this case because every timer interrupt generates a NMI
673 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
675 LOC is the local interrupt counter of the internal APIC of every CPU.
677 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
678 connects the CPUs in a SMP system. This means that an error has been detected,
679 the IO-APIC automatically retry the transmission, so it should not be a big
680 problem, but you should read the SMP-FAQ.
682 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
683 /proc/interrupts to display every IRQ vector in use by the system, not
684 just those considered 'most important'. The new vectors are:
686 THR -- interrupt raised when a machine check threshold counter
687 (typically counting ECC corrected errors of memory or cache) exceeds
688 a configurable threshold. Only available on some systems.
690 TRM -- a thermal event interrupt occurs when a temperature threshold
691 has been exceeded for the CPU. This interrupt may also be generated
692 when the temperature drops back to normal.
694 SPU -- a spurious interrupt is some interrupt that was raised then lowered
695 by some IO device before it could be fully processed by the APIC. Hence
696 the APIC sees the interrupt but does not know what device it came from.
697 For this case the APIC will generate the interrupt with a IRQ vector
698 of 0xff. This might also be generated by chipset bugs.
700 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
701 sent from one CPU to another per the needs of the OS. Typically,
702 their statistics are used by kernel developers and interested users to
703 determine the occurrence of interrupts of the given type.
705 The above IRQ vectors are displayed only when relevant. For example,
706 the threshold vector does not exist on x86_64 platforms. Others are
707 suppressed when the system is a uniprocessor. As of this writing, only
708 i386 and x86_64 platforms support the new IRQ vector displays.
710 Of some interest is the introduction of the /proc/irq directory to 2.4.
711 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
712 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
713 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
714 prof_cpu_mask.
716 For example
717 > ls /proc/irq/
718 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
719 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
720 > ls /proc/irq/0/
721 smp_affinity
723 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
724 IRQ, you can set it by doing:
726 > echo 1 > /proc/irq/10/smp_affinity
728 This means that only the first CPU will handle the IRQ, but you can also echo
729 5 which means that only the first and fourth CPU can handle the IRQ.
731 The contents of each smp_affinity file is the same by default:
733 > cat /proc/irq/0/smp_affinity
734 ffffffff
736 There is an alternate interface, smp_affinity_list which allows specifying
737 a cpu range instead of a bitmask:
739 > cat /proc/irq/0/smp_affinity_list
740 1024-1031
742 The default_smp_affinity mask applies to all non-active IRQs, which are the
743 IRQs which have not yet been allocated/activated, and hence which lack a
744 /proc/irq/[0-9]* directory.
746 The node file on an SMP system shows the node to which the device using the IRQ
747 reports itself as being attached. This hardware locality information does not
748 include information about any possible driver locality preference.
750 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
751 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
753 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
754 between all the CPUs which are allowed to handle it. As usual the kernel has
755 more info than you and does a better job than you, so the defaults are the
756 best choice for almost everyone. [Note this applies only to those IO-APIC's
757 that support "Round Robin" interrupt distribution.]
759 There are three more important subdirectories in /proc: net, scsi, and sys.
760 The general rule is that the contents, or even the existence of these
761 directories, depend on your kernel configuration. If SCSI is not enabled, the
762 directory scsi may not exist. The same is true with the net, which is there
763 only when networking support is present in the running kernel.
765 The slabinfo file gives information about memory usage at the slab level.
766 Linux uses slab pools for memory management above page level in version 2.2.
767 Commonly used objects have their own slab pool (such as network buffers,
768 directory cache, and so on).
770 ..............................................................................
772 > cat /proc/buddyinfo
774 Node 0, zone DMA 0 4 5 4 4 3 ...
775 Node 0, zone Normal 1 0 0 1 101 8 ...
776 Node 0, zone HighMem 2 0 0 1 1 0 ...
778 External fragmentation is a problem under some workloads, and buddyinfo is a
779 useful tool for helping diagnose these problems. Buddyinfo will give you a
780 clue as to how big an area you can safely allocate, or why a previous
781 allocation failed.
783 Each column represents the number of pages of a certain order which are
784 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
785 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
786 available in ZONE_NORMAL, etc...
788 More information relevant to external fragmentation can be found in
789 pagetypeinfo.
791 > cat /proc/pagetypeinfo
792 Page block order: 9
793 Pages per block: 512
795 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
796 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
797 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
798 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
799 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
800 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
801 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
802 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
803 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
804 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
805 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
807 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
808 Node 0, zone DMA 2 0 5 1 0
809 Node 0, zone DMA32 41 6 967 2 0
811 Fragmentation avoidance in the kernel works by grouping pages of different
812 migrate types into the same contiguous regions of memory called page blocks.
813 A page block is typically the size of the default hugepage size e.g. 2MB on
814 X86-64. By keeping pages grouped based on their ability to move, the kernel
815 can reclaim pages within a page block to satisfy a high-order allocation.
817 The pagetypinfo begins with information on the size of a page block. It
818 then gives the same type of information as buddyinfo except broken down
819 by migrate-type and finishes with details on how many page blocks of each
820 type exist.
822 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
823 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
824 make an estimate of the likely number of huge pages that can be allocated
825 at a given point in time. All the "Movable" blocks should be allocatable
826 unless memory has been mlock()'d. Some of the Reclaimable blocks should
827 also be allocatable although a lot of filesystem metadata may have to be
828 reclaimed to achieve this.
830 ..............................................................................
832 meminfo:
834 Provides information about distribution and utilization of memory. This
835 varies by architecture and compile options. The following is from a
836 16GB PIII, which has highmem enabled. You may not have all of these fields.
838 > cat /proc/meminfo
840 MemTotal: 16344972 kB
841 MemFree: 13634064 kB
842 MemAvailable: 14836172 kB
843 Buffers: 3656 kB
844 Cached: 1195708 kB
845 SwapCached: 0 kB
846 Active: 891636 kB
847 Inactive: 1077224 kB
848 HighTotal: 15597528 kB
849 HighFree: 13629632 kB
850 LowTotal: 747444 kB
851 LowFree: 4432 kB
852 SwapTotal: 0 kB
853 SwapFree: 0 kB
854 Dirty: 968 kB
855 Writeback: 0 kB
856 AnonPages: 861800 kB
857 Mapped: 280372 kB
858 Shmem: 644 kB
859 Slab: 284364 kB
860 SReclaimable: 159856 kB
861 SUnreclaim: 124508 kB
862 PageTables: 24448 kB
863 NFS_Unstable: 0 kB
864 Bounce: 0 kB
865 WritebackTmp: 0 kB
866 CommitLimit: 7669796 kB
867 Committed_AS: 100056 kB
868 VmallocTotal: 112216 kB
869 VmallocUsed: 428 kB
870 VmallocChunk: 111088 kB
871 AnonHugePages: 49152 kB
873 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
874 bits and the kernel binary code)
875 MemFree: The sum of LowFree+HighFree
876 MemAvailable: An estimate of how much memory is available for starting new
877 applications, without swapping. Calculated from MemFree,
878 SReclaimable, the size of the file LRU lists, and the low
879 watermarks in each zone.
880 The estimate takes into account that the system needs some
881 page cache to function well, and that not all reclaimable
882 slab will be reclaimable, due to items being in use. The
883 impact of those factors will vary from system to system.
884 Buffers: Relatively temporary storage for raw disk blocks
885 shouldn't get tremendously large (20MB or so)
886 Cached: in-memory cache for files read from the disk (the
887 pagecache). Doesn't include SwapCached
888 SwapCached: Memory that once was swapped out, is swapped back in but
889 still also is in the swapfile (if memory is needed it
890 doesn't need to be swapped out AGAIN because it is already
891 in the swapfile. This saves I/O)
892 Active: Memory that has been used more recently and usually not
893 reclaimed unless absolutely necessary.
894 Inactive: Memory which has been less recently used. It is more
895 eligible to be reclaimed for other purposes
896 HighTotal:
897 HighFree: Highmem is all memory above ~860MB of physical memory
898 Highmem areas are for use by userspace programs, or
899 for the pagecache. The kernel must use tricks to access
900 this memory, making it slower to access than lowmem.
901 LowTotal:
902 LowFree: Lowmem is memory which can be used for everything that
903 highmem can be used for, but it is also available for the
904 kernel's use for its own data structures. Among many
905 other things, it is where everything from the Slab is
906 allocated. Bad things happen when you're out of lowmem.
907 SwapTotal: total amount of swap space available
908 SwapFree: Memory which has been evicted from RAM, and is temporarily
909 on the disk
910 Dirty: Memory which is waiting to get written back to the disk
911 Writeback: Memory which is actively being written back to the disk
912 AnonPages: Non-file backed pages mapped into userspace page tables
913 AnonHugePages: Non-file backed huge pages mapped into userspace page tables
914 Mapped: files which have been mmaped, such as libraries
915 Shmem: Total memory used by shared memory (shmem) and tmpfs
916 Slab: in-kernel data structures cache
917 SReclaimable: Part of Slab, that might be reclaimed, such as caches
918 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
919 PageTables: amount of memory dedicated to the lowest level of page
920 tables.
921 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
922 storage
923 Bounce: Memory used for block device "bounce buffers"
924 WritebackTmp: Memory used by FUSE for temporary writeback buffers
925 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
926 this is the total amount of memory currently available to
927 be allocated on the system. This limit is only adhered to
928 if strict overcommit accounting is enabled (mode 2 in
929 'vm.overcommit_memory').
930 The CommitLimit is calculated with the following formula:
931 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
932 overcommit_ratio / 100 + [total swap pages]
933 For example, on a system with 1G of physical RAM and 7G
934 of swap with a `vm.overcommit_ratio` of 30 it would
935 yield a CommitLimit of 7.3G.
936 For more details, see the memory overcommit documentation
937 in vm/overcommit-accounting.
938 Committed_AS: The amount of memory presently allocated on the system.
939 The committed memory is a sum of all of the memory which
940 has been allocated by processes, even if it has not been
941 "used" by them as of yet. A process which malloc()'s 1G
942 of memory, but only touches 300M of it will show up as
943 using 1G. This 1G is memory which has been "committed" to
944 by the VM and can be used at any time by the allocating
945 application. With strict overcommit enabled on the system
946 (mode 2 in 'vm.overcommit_memory'),allocations which would
947 exceed the CommitLimit (detailed above) will not be permitted.
948 This is useful if one needs to guarantee that processes will
949 not fail due to lack of memory once that memory has been
950 successfully allocated.
951 VmallocTotal: total size of vmalloc memory area
952 VmallocUsed: amount of vmalloc area which is used
953 VmallocChunk: largest contiguous block of vmalloc area which is free
955 ..............................................................................
957 vmallocinfo:
959 Provides information about vmalloced/vmaped areas. One line per area,
960 containing the virtual address range of the area, size in bytes,
961 caller information of the creator, and optional information depending
962 on the kind of area :
964 pages=nr number of pages
965 phys=addr if a physical address was specified
966 ioremap I/O mapping (ioremap() and friends)
967 vmalloc vmalloc() area
968 vmap vmap()ed pages
969 user VM_USERMAP area
970 vpages buffer for pages pointers was vmalloced (huge area)
971 N<node>=nr (Only on NUMA kernels)
972 Number of pages allocated on memory node <node>
974 > cat /proc/vmallocinfo
975 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
976 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
977 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
978 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
979 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
980 phys=7fee8000 ioremap
981 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
982 phys=7fee7000 ioremap
983 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
984 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
985 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
986 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
987 pages=2 vmalloc N1=2
988 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
989 /0x130 [x_tables] pages=4 vmalloc N0=4
990 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
991 pages=14 vmalloc N2=14
992 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
993 pages=4 vmalloc N1=4
994 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
995 pages=2 vmalloc N1=2
996 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
997 pages=10 vmalloc N0=10
999 ..............................................................................
1001 softirqs:
1003 Provides counts of softirq handlers serviced since boot time, for each cpu.
1005 > cat /proc/softirqs
1006 CPU0 CPU1 CPU2 CPU3
1007 HI: 0 0 0 0
1008 TIMER: 27166 27120 27097 27034
1009 NET_TX: 0 0 0 17
1010 NET_RX: 42 0 0 39
1011 BLOCK: 0 0 107 1121
1012 TASKLET: 0 0 0 290
1013 SCHED: 27035 26983 26971 26746
1014 HRTIMER: 0 0 0 0
1015 RCU: 1678 1769 2178 2250
1018 1.3 IDE devices in /proc/ide
1019 ----------------------------
1021 The subdirectory /proc/ide contains information about all IDE devices of which
1022 the kernel is aware. There is one subdirectory for each IDE controller, the
1023 file drivers and a link for each IDE device, pointing to the device directory
1024 in the controller specific subtree.
1026 The file drivers contains general information about the drivers used for the
1027 IDE devices:
1029 > cat /proc/ide/drivers
1030 ide-cdrom version 4.53
1031 ide-disk version 1.08
1033 More detailed information can be found in the controller specific
1034 subdirectories. These are named ide0, ide1 and so on. Each of these
1035 directories contains the files shown in table 1-6.
1038 Table 1-6: IDE controller info in /proc/ide/ide?
1039 ..............................................................................
1040 File Content
1041 channel IDE channel (0 or 1)
1042 config Configuration (only for PCI/IDE bridge)
1043 mate Mate name
1044 model Type/Chipset of IDE controller
1045 ..............................................................................
1047 Each device connected to a controller has a separate subdirectory in the
1048 controllers directory. The files listed in table 1-7 are contained in these
1049 directories.
1052 Table 1-7: IDE device information
1053 ..............................................................................
1054 File Content
1055 cache The cache
1056 capacity Capacity of the medium (in 512Byte blocks)
1057 driver driver and version
1058 geometry physical and logical geometry
1059 identify device identify block
1060 media media type
1061 model device identifier
1062 settings device setup
1063 smart_thresholds IDE disk management thresholds
1064 smart_values IDE disk management values
1065 ..............................................................................
1067 The most interesting file is settings. This file contains a nice overview of
1068 the drive parameters:
1070 # cat /proc/ide/ide0/hda/settings
1071 name value min max mode
1072 ---- ----- --- --- ----
1073 bios_cyl 526 0 65535 rw
1074 bios_head 255 0 255 rw
1075 bios_sect 63 0 63 rw
1076 breada_readahead 4 0 127 rw
1077 bswap 0 0 1 r
1078 file_readahead 72 0 2097151 rw
1079 io_32bit 0 0 3 rw
1080 keepsettings 0 0 1 rw
1081 max_kb_per_request 122 1 127 rw
1082 multcount 0 0 8 rw
1083 nice1 1 0 1 rw
1084 nowerr 0 0 1 rw
1085 pio_mode write-only 0 255 w
1086 slow 0 0 1 rw
1087 unmaskirq 0 0 1 rw
1088 using_dma 0 0 1 rw
1091 1.4 Networking info in /proc/net
1092 --------------------------------
1094 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1095 additional values you get for IP version 6 if you configure the kernel to
1096 support this. Table 1-9 lists the files and their meaning.
1099 Table 1-8: IPv6 info in /proc/net
1100 ..............................................................................
1101 File Content
1102 udp6 UDP sockets (IPv6)
1103 tcp6 TCP sockets (IPv6)
1104 raw6 Raw device statistics (IPv6)
1105 igmp6 IP multicast addresses, which this host joined (IPv6)
1106 if_inet6 List of IPv6 interface addresses
1107 ipv6_route Kernel routing table for IPv6
1108 rt6_stats Global IPv6 routing tables statistics
1109 sockstat6 Socket statistics (IPv6)
1110 snmp6 Snmp data (IPv6)
1111 ..............................................................................
1114 Table 1-9: Network info in /proc/net
1115 ..............................................................................
1116 File Content
1117 arp Kernel ARP table
1118 dev network devices with statistics
1119 dev_mcast the Layer2 multicast groups a device is listening too
1120 (interface index, label, number of references, number of bound
1121 addresses).
1122 dev_stat network device status
1123 ip_fwchains Firewall chain linkage
1124 ip_fwnames Firewall chain names
1125 ip_masq Directory containing the masquerading tables
1126 ip_masquerade Major masquerading table
1127 netstat Network statistics
1128 raw raw device statistics
1129 route Kernel routing table
1130 rpc Directory containing rpc info
1131 rt_cache Routing cache
1132 snmp SNMP data
1133 sockstat Socket statistics
1134 tcp TCP sockets
1135 udp UDP sockets
1136 unix UNIX domain sockets
1137 wireless Wireless interface data (Wavelan etc)
1138 igmp IP multicast addresses, which this host joined
1139 psched Global packet scheduler parameters.
1140 netlink List of PF_NETLINK sockets
1141 ip_mr_vifs List of multicast virtual interfaces
1142 ip_mr_cache List of multicast routing cache
1143 ..............................................................................
1145 You can use this information to see which network devices are available in
1146 your system and how much traffic was routed over those devices:
1148 > cat /proc/net/dev
1149 Inter-|Receive |[...
1150 face |bytes packets errs drop fifo frame compressed multicast|[...
1151 lo: 908188 5596 0 0 0 0 0 0 [...
1152 ppp0:15475140 20721 410 0 0 410 0 0 [...
1153 eth0: 614530 7085 0 0 0 0 0 1 [...
1155 ...] Transmit
1156 ...] bytes packets errs drop fifo colls carrier compressed
1157 ...] 908188 5596 0 0 0 0 0 0
1158 ...] 1375103 17405 0 0 0 0 0 0
1159 ...] 1703981 5535 0 0 0 3 0 0
1161 In addition, each Channel Bond interface has its own directory. For
1162 example, the bond0 device will have a directory called /proc/net/bond0/.
1163 It will contain information that is specific to that bond, such as the
1164 current slaves of the bond, the link status of the slaves, and how
1165 many times the slaves link has failed.
1167 1.5 SCSI info
1168 -------------
1170 If you have a SCSI host adapter in your system, you'll find a subdirectory
1171 named after the driver for this adapter in /proc/scsi. You'll also see a list
1172 of all recognized SCSI devices in /proc/scsi:
1174 >cat /proc/scsi/scsi
1175 Attached devices:
1176 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1177 Vendor: IBM Model: DGHS09U Rev: 03E0
1178 Type: Direct-Access ANSI SCSI revision: 03
1179 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1180 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1181 Type: CD-ROM ANSI SCSI revision: 02
1184 The directory named after the driver has one file for each adapter found in
1185 the system. These files contain information about the controller, including
1186 the used IRQ and the IO address range. The amount of information shown is
1187 dependent on the adapter you use. The example shows the output for an Adaptec
1188 AHA-2940 SCSI adapter:
1190 > cat /proc/scsi/aic7xxx/0
1192 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1193 Compile Options:
1194 TCQ Enabled By Default : Disabled
1195 AIC7XXX_PROC_STATS : Disabled
1196 AIC7XXX_RESET_DELAY : 5
1197 Adapter Configuration:
1198 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1199 Ultra Wide Controller
1200 PCI MMAPed I/O Base: 0xeb001000
1201 Adapter SEEPROM Config: SEEPROM found and used.
1202 Adaptec SCSI BIOS: Enabled
1203 IRQ: 10
1204 SCBs: Active 0, Max Active 2,
1205 Allocated 15, HW 16, Page 255
1206 Interrupts: 160328
1207 BIOS Control Word: 0x18b6
1208 Adapter Control Word: 0x005b
1209 Extended Translation: Enabled
1210 Disconnect Enable Flags: 0xffff
1211 Ultra Enable Flags: 0x0001
1212 Tag Queue Enable Flags: 0x0000
1213 Ordered Queue Tag Flags: 0x0000
1214 Default Tag Queue Depth: 8
1215 Tagged Queue By Device array for aic7xxx host instance 0:
1216 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1217 Actual queue depth per device for aic7xxx host instance 0:
1218 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1219 Statistics:
1220 (scsi0:0:0:0)
1221 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1222 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1223 Total transfers 160151 (74577 reads and 85574 writes)
1224 (scsi0:0:6:0)
1225 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1226 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1227 Total transfers 0 (0 reads and 0 writes)
1230 1.6 Parallel port info in /proc/parport
1231 ---------------------------------------
1233 The directory /proc/parport contains information about the parallel ports of
1234 your system. It has one subdirectory for each port, named after the port
1235 number (0,1,2,...).
1237 These directories contain the four files shown in Table 1-10.
1240 Table 1-10: Files in /proc/parport
1241 ..............................................................................
1242 File Content
1243 autoprobe Any IEEE-1284 device ID information that has been acquired.
1244 devices list of the device drivers using that port. A + will appear by the
1245 name of the device currently using the port (it might not appear
1246 against any).
1247 hardware Parallel port's base address, IRQ line and DMA channel.
1248 irq IRQ that parport is using for that port. This is in a separate
1249 file to allow you to alter it by writing a new value in (IRQ
1250 number or none).
1251 ..............................................................................
1253 1.7 TTY info in /proc/tty
1254 -------------------------
1256 Information about the available and actually used tty's can be found in the
1257 directory /proc/tty.You'll find entries for drivers and line disciplines in
1258 this directory, as shown in Table 1-11.
1261 Table 1-11: Files in /proc/tty
1262 ..............................................................................
1263 File Content
1264 drivers list of drivers and their usage
1265 ldiscs registered line disciplines
1266 driver/serial usage statistic and status of single tty lines
1267 ..............................................................................
1269 To see which tty's are currently in use, you can simply look into the file
1270 /proc/tty/drivers:
1272 > cat /proc/tty/drivers
1273 pty_slave /dev/pts 136 0-255 pty:slave
1274 pty_master /dev/ptm 128 0-255 pty:master
1275 pty_slave /dev/ttyp 3 0-255 pty:slave
1276 pty_master /dev/pty 2 0-255 pty:master
1277 serial /dev/cua 5 64-67 serial:callout
1278 serial /dev/ttyS 4 64-67 serial
1279 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1280 /dev/ptmx /dev/ptmx 5 2 system
1281 /dev/console /dev/console 5 1 system:console
1282 /dev/tty /dev/tty 5 0 system:/dev/tty
1283 unknown /dev/tty 4 1-63 console
1286 1.8 Miscellaneous kernel statistics in /proc/stat
1287 -------------------------------------------------
1289 Various pieces of information about kernel activity are available in the
1290 /proc/stat file. All of the numbers reported in this file are aggregates
1291 since the system first booted. For a quick look, simply cat the file:
1293 > cat /proc/stat
1294 cpu 2255 34 2290 22625563 6290 127 456 0 0 0
1295 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0
1296 cpu1 1123 0 849 11313845 2614 0 18 0 0 0
1297 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1298 ctxt 1990473
1299 btime 1062191376
1300 processes 2915
1301 procs_running 1
1302 procs_blocked 0
1303 softirq 183433 0 21755 12 39 1137 231 21459 2263
1305 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1306 lines. These numbers identify the amount of time the CPU has spent performing
1307 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1308 second). The meanings of the columns are as follows, from left to right:
1310 - user: normal processes executing in user mode
1311 - nice: niced processes executing in user mode
1312 - system: processes executing in kernel mode
1313 - idle: twiddling thumbs
1314 - iowait: waiting for I/O to complete
1315 - irq: servicing interrupts
1316 - softirq: servicing softirqs
1317 - steal: involuntary wait
1318 - guest: running a normal guest
1319 - guest_nice: running a niced guest
1321 The "intr" line gives counts of interrupts serviced since boot time, for each
1322 of the possible system interrupts. The first column is the total of all
1323 interrupts serviced including unnumbered architecture specific interrupts;
1324 each subsequent column is the total for that particular numbered interrupt.
1325 Unnumbered interrupts are not shown, only summed into the total.
1327 The "ctxt" line gives the total number of context switches across all CPUs.
1329 The "btime" line gives the time at which the system booted, in seconds since
1330 the Unix epoch.
1332 The "processes" line gives the number of processes and threads created, which
1333 includes (but is not limited to) those created by calls to the fork() and
1334 clone() system calls.
1336 The "procs_running" line gives the total number of threads that are
1337 running or ready to run (i.e., the total number of runnable threads).
1339 The "procs_blocked" line gives the number of processes currently blocked,
1340 waiting for I/O to complete.
1342 The "softirq" line gives counts of softirqs serviced since boot time, for each
1343 of the possible system softirqs. The first column is the total of all
1344 softirqs serviced; each subsequent column is the total for that particular
1345 softirq.
1348 1.9 Ext4 file system parameters
1349 -------------------------------
1351 Information about mounted ext4 file systems can be found in
1352 /proc/fs/ext4. Each mounted filesystem will have a directory in
1353 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1354 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1355 in Table 1-12, below.
1357 Table 1-12: Files in /proc/fs/ext4/<devname>
1358 ..............................................................................
1359 File Content
1360 mb_groups details of multiblock allocator buddy cache of free blocks
1361 ..............................................................................
1363 2.0 /proc/consoles
1364 ------------------
1365 Shows registered system console lines.
1367 To see which character device lines are currently used for the system console
1368 /dev/console, you may simply look into the file /proc/consoles:
1370 > cat /proc/consoles
1371 tty0 -WU (ECp) 4:7
1372 ttyS0 -W- (Ep) 4:64
1374 The columns are:
1376 device name of the device
1377 operations R = can do read operations
1378 W = can do write operations
1379 U = can do unblank
1380 flags E = it is enabled
1381 C = it is preferred console
1382 B = it is primary boot console
1383 p = it is used for printk buffer
1384 b = it is not a TTY but a Braille device
1385 a = it is safe to use when cpu is offline
1386 major:minor major and minor number of the device separated by a colon
1388 ------------------------------------------------------------------------------
1389 Summary
1390 ------------------------------------------------------------------------------
1391 The /proc file system serves information about the running system. It not only
1392 allows access to process data but also allows you to request the kernel status
1393 by reading files in the hierarchy.
1395 The directory structure of /proc reflects the types of information and makes
1396 it easy, if not obvious, where to look for specific data.
1397 ------------------------------------------------------------------------------
1399 ------------------------------------------------------------------------------
1400 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1401 ------------------------------------------------------------------------------
1403 ------------------------------------------------------------------------------
1404 In This Chapter
1405 ------------------------------------------------------------------------------
1406 * Modifying kernel parameters by writing into files found in /proc/sys
1407 * Exploring the files which modify certain parameters
1408 * Review of the /proc/sys file tree
1409 ------------------------------------------------------------------------------
1412 A very interesting part of /proc is the directory /proc/sys. This is not only
1413 a source of information, it also allows you to change parameters within the
1414 kernel. Be very careful when attempting this. You can optimize your system,
1415 but you can also cause it to crash. Never alter kernel parameters on a
1416 production system. Set up a development machine and test to make sure that
1417 everything works the way you want it to. You may have no alternative but to
1418 reboot the machine once an error has been made.
1420 To change a value, simply echo the new value into the file. An example is
1421 given below in the section on the file system data. You need to be root to do
1422 this. You can create your own boot script to perform this every time your
1423 system boots.
1425 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1426 general things in the operation of the Linux kernel. Since some of the files
1427 can inadvertently disrupt your system, it is advisable to read both
1428 documentation and source before actually making adjustments. In any case, be
1429 very careful when writing to any of these files. The entries in /proc may
1430 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1431 review the kernel documentation in the directory /usr/src/linux/Documentation.
1432 This chapter is heavily based on the documentation included in the pre 2.2
1433 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1435 Please see: Documentation/sysctl/ directory for descriptions of these
1436 entries.
1438 ------------------------------------------------------------------------------
1439 Summary
1440 ------------------------------------------------------------------------------
1441 Certain aspects of kernel behavior can be modified at runtime, without the
1442 need to recompile the kernel, or even to reboot the system. The files in the
1443 /proc/sys tree can not only be read, but also modified. You can use the echo
1444 command to write value into these files, thereby changing the default settings
1445 of the kernel.
1446 ------------------------------------------------------------------------------
1448 ------------------------------------------------------------------------------
1449 CHAPTER 3: PER-PROCESS PARAMETERS
1450 ------------------------------------------------------------------------------
1452 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1453 --------------------------------------------------------------------------------
1455 These file can be used to adjust the badness heuristic used to select which
1456 process gets killed in out of memory conditions.
1458 The badness heuristic assigns a value to each candidate task ranging from 0
1459 (never kill) to 1000 (always kill) to determine which process is targeted. The
1460 units are roughly a proportion along that range of allowed memory the process
1461 may allocate from based on an estimation of its current memory and swap use.
1462 For example, if a task is using all allowed memory, its badness score will be
1463 1000. If it is using half of its allowed memory, its score will be 500.
1465 There is an additional factor included in the badness score: the current memory
1466 and swap usage is discounted by 3% for root processes.
1468 The amount of "allowed" memory depends on the context in which the oom killer
1469 was called. If it is due to the memory assigned to the allocating task's cpuset
1470 being exhausted, the allowed memory represents the set of mems assigned to that
1471 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1472 memory represents the set of mempolicy nodes. If it is due to a memory
1473 limit (or swap limit) being reached, the allowed memory is that configured
1474 limit. Finally, if it is due to the entire system being out of memory, the
1475 allowed memory represents all allocatable resources.
1477 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1478 is used to determine which task to kill. Acceptable values range from -1000
1479 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1480 polarize the preference for oom killing either by always preferring a certain
1481 task or completely disabling it. The lowest possible value, -1000, is
1482 equivalent to disabling oom killing entirely for that task since it will always
1483 report a badness score of 0.
1485 Consequently, it is very simple for userspace to define the amount of memory to
1486 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1487 example, is roughly equivalent to allowing the remainder of tasks sharing the
1488 same system, cpuset, mempolicy, or memory controller resources to use at least
1489 50% more memory. A value of -500, on the other hand, would be roughly
1490 equivalent to discounting 50% of the task's allowed memory from being considered
1491 as scoring against the task.
1493 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1494 be used to tune the badness score. Its acceptable values range from -16
1495 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1496 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1497 scaled linearly with /proc/<pid>/oom_score_adj.
1499 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1500 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1501 requires CAP_SYS_RESOURCE.
1503 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1504 generation children with separate address spaces instead, if possible. This
1505 avoids servers and important system daemons from being killed and loses the
1506 minimal amount of work.
1509 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1510 -------------------------------------------------------------
1512 This file can be used to check the current score used by the oom-killer is for
1513 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1514 process should be killed in an out-of-memory situation.
1517 3.3 /proc/<pid>/io - Display the IO accounting fields
1518 -------------------------------------------------------
1520 This file contains IO statistics for each running process
1522 Example
1523 -------
1525 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1526 [1] 3828
1528 test:/tmp # cat /proc/3828/io
1529 rchar: 323934931
1530 wchar: 323929600
1531 syscr: 632687
1532 syscw: 632675
1533 read_bytes: 0
1534 write_bytes: 323932160
1535 cancelled_write_bytes: 0
1538 Description
1539 -----------
1541 rchar
1542 -----
1544 I/O counter: chars read
1545 The number of bytes which this task has caused to be read from storage. This
1546 is simply the sum of bytes which this process passed to read() and pread().
1547 It includes things like tty IO and it is unaffected by whether or not actual
1548 physical disk IO was required (the read might have been satisfied from
1549 pagecache)
1552 wchar
1553 -----
1555 I/O counter: chars written
1556 The number of bytes which this task has caused, or shall cause to be written
1557 to disk. Similar caveats apply here as with rchar.
1560 syscr
1561 -----
1563 I/O counter: read syscalls
1564 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1565 and pread().
1568 syscw
1569 -----
1571 I/O counter: write syscalls
1572 Attempt to count the number of write I/O operations, i.e. syscalls like
1573 write() and pwrite().
1576 read_bytes
1577 ----------
1579 I/O counter: bytes read
1580 Attempt to count the number of bytes which this process really did cause to
1581 be fetched from the storage layer. Done at the submit_bio() level, so it is
1582 accurate for block-backed filesystems. <please add status regarding NFS and
1583 CIFS at a later time>
1586 write_bytes
1587 -----------
1589 I/O counter: bytes written
1590 Attempt to count the number of bytes which this process caused to be sent to
1591 the storage layer. This is done at page-dirtying time.
1594 cancelled_write_bytes
1595 ---------------------
1597 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1598 then deletes the file, it will in fact perform no writeout. But it will have
1599 been accounted as having caused 1MB of write.
1600 In other words: The number of bytes which this process caused to not happen,
1601 by truncating pagecache. A task can cause "negative" IO too. If this task
1602 truncates some dirty pagecache, some IO which another task has been accounted
1603 for (in its write_bytes) will not be happening. We _could_ just subtract that
1604 from the truncating task's write_bytes, but there is information loss in doing
1605 that.
1608 Note
1609 ----
1611 At its current implementation state, this is a bit racy on 32-bit machines: if
1612 process A reads process B's /proc/pid/io while process B is updating one of
1613 those 64-bit counters, process A could see an intermediate result.
1616 More information about this can be found within the taskstats documentation in
1617 Documentation/accounting.
1619 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1620 ---------------------------------------------------------------
1621 When a process is dumped, all anonymous memory is written to a core file as
1622 long as the size of the core file isn't limited. But sometimes we don't want
1623 to dump some memory segments, for example, huge shared memory or DAX.
1624 Conversely, sometimes we want to save file-backed memory segments into a core
1625 file, not only the individual files.
1627 /proc/<pid>/coredump_filter allows you to customize which memory segments
1628 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1629 of memory types. If a bit of the bitmask is set, memory segments of the
1630 corresponding memory type are dumped, otherwise they are not dumped.
1632 The following 9 memory types are supported:
1633 - (bit 0) anonymous private memory
1634 - (bit 1) anonymous shared memory
1635 - (bit 2) file-backed private memory
1636 - (bit 3) file-backed shared memory
1637 - (bit 4) ELF header pages in file-backed private memory areas (it is
1638 effective only if the bit 2 is cleared)
1639 - (bit 5) hugetlb private memory
1640 - (bit 6) hugetlb shared memory
1641 - (bit 7) DAX private memory
1642 - (bit 8) DAX shared memory
1644 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1645 are always dumped regardless of the bitmask status.
1647 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1648 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1650 The default value of coredump_filter is 0x33; this means all anonymous memory
1651 segments, ELF header pages and hugetlb private memory are dumped.
1653 If you don't want to dump all shared memory segments attached to pid 1234,
1654 write 0x31 to the process's proc file.
1656 $ echo 0x31 > /proc/1234/coredump_filter
1658 When a new process is created, the process inherits the bitmask status from its
1659 parent. It is useful to set up coredump_filter before the program runs.
1660 For example:
1662 $ echo 0x7 > /proc/self/coredump_filter
1663 $ ./some_program
1665 3.5 /proc/<pid>/mountinfo - Information about mounts
1666 --------------------------------------------------------
1668 This file contains lines of the form:
1670 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1671 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1673 (1) mount ID: unique identifier of the mount (may be reused after umount)
1674 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1675 (3) major:minor: value of st_dev for files on filesystem
1676 (4) root: root of the mount within the filesystem
1677 (5) mount point: mount point relative to the process's root
1678 (6) mount options: per mount options
1679 (7) optional fields: zero or more fields of the form "tag[:value]"
1680 (8) separator: marks the end of the optional fields
1681 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1682 (10) mount source: filesystem specific information or "none"
1683 (11) super options: per super block options
1685 Parsers should ignore all unrecognised optional fields. Currently the
1686 possible optional fields are:
1688 shared:X mount is shared in peer group X
1689 master:X mount is slave to peer group X
1690 propagate_from:X mount is slave and receives propagation from peer group X (*)
1691 unbindable mount is unbindable
1693 (*) X is the closest dominant peer group under the process's root. If
1694 X is the immediate master of the mount, or if there's no dominant peer
1695 group under the same root, then only the "master:X" field is present
1696 and not the "propagate_from:X" field.
1698 For more information on mount propagation see:
1700 Documentation/filesystems/sharedsubtree.txt
1703 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1704 --------------------------------------------------------
1705 These files provide a method to access a tasks comm value. It also allows for
1706 a task to set its own or one of its thread siblings comm value. The comm value
1707 is limited in size compared to the cmdline value, so writing anything longer
1708 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1709 comm value.
1712 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1713 -------------------------------------------------------------------------
1714 This file provides a fast way to retrieve first level children pids
1715 of a task pointed by <pid>/<tid> pair. The format is a space separated
1716 stream of pids.
1718 Note the "first level" here -- if a child has own children they will
1719 not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1720 to obtain the descendants.
1722 Since this interface is intended to be fast and cheap it doesn't
1723 guarantee to provide precise results and some children might be
1724 skipped, especially if they've exited right after we printed their
1725 pids, so one need to either stop or freeze processes being inspected
1726 if precise results are needed.
1729 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1730 ---------------------------------------------------------------
1731 This file provides information associated with an opened file. The regular
1732 files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos'
1733 represents the current offset of the opened file in decimal form [see lseek(2)
1734 for details], 'flags' denotes the octal O_xxx mask the file has been
1735 created with [see open(2) for details] and 'mnt_id' represents mount ID of
1736 the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo
1737 for details].
1739 A typical output is
1741 pos: 0
1742 flags: 0100002
1743 mnt_id: 19
1745 All locks associated with a file descriptor are shown in its fdinfo too.
1747 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1749 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1750 pair provide additional information particular to the objects they represent.
1752 Eventfd files
1753 ~~~~~~~~~~~~~
1754 pos: 0
1755 flags: 04002
1756 mnt_id: 9
1757 eventfd-count: 5a
1759 where 'eventfd-count' is hex value of a counter.
1761 Signalfd files
1762 ~~~~~~~~~~~~~~
1763 pos: 0
1764 flags: 04002
1765 mnt_id: 9
1766 sigmask: 0000000000000200
1768 where 'sigmask' is hex value of the signal mask associated
1769 with a file.
1771 Epoll files
1772 ~~~~~~~~~~~
1773 pos: 0
1774 flags: 02
1775 mnt_id: 9
1776 tfd: 5 events: 1d data: ffffffffffffffff
1778 where 'tfd' is a target file descriptor number in decimal form,
1779 'events' is events mask being watched and the 'data' is data
1780 associated with a target [see epoll(7) for more details].
1782 Fsnotify files
1783 ~~~~~~~~~~~~~~
1784 For inotify files the format is the following
1786 pos: 0
1787 flags: 02000000
1788 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1790 where 'wd' is a watch descriptor in decimal form, ie a target file
1791 descriptor number, 'ino' and 'sdev' are inode and device where the
1792 target file resides and the 'mask' is the mask of events, all in hex
1793 form [see inotify(7) for more details].
1795 If the kernel was built with exportfs support, the path to the target
1796 file is encoded as a file handle. The file handle is provided by three
1797 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1798 format.
1800 If the kernel is built without exportfs support the file handle won't be
1801 printed out.
1803 If there is no inotify mark attached yet the 'inotify' line will be omitted.
1805 For fanotify files the format is
1807 pos: 0
1808 flags: 02
1809 mnt_id: 9
1810 fanotify flags:10 event-flags:0
1811 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1812 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
1814 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
1815 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
1816 flags associated with mark which are tracked separately from events
1817 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
1818 mask and 'ignored_mask' is the mask of events which are to be ignored.
1819 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
1820 does provide information about flags and mask used in fanotify_mark
1821 call [see fsnotify manpage for details].
1823 While the first three lines are mandatory and always printed, the rest is
1824 optional and may be omitted if no marks created yet.
1826 Timerfd files
1827 ~~~~~~~~~~~~~
1829 pos: 0
1830 flags: 02
1831 mnt_id: 9
1832 clockid: 0
1833 ticks: 0
1834 settime flags: 01
1835 it_value: (0, 49406829)
1836 it_interval: (1, 0)
1838 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
1839 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
1840 flags in octal form been used to setup the timer [see timerfd_settime(2) for
1841 details]. 'it_value' is remaining time until the timer exiration.
1842 'it_interval' is the interval for the timer. Note the timer might be set up
1843 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
1844 still exhibits timer's remaining time.
1846 3.9 /proc/<pid>/map_files - Information about memory mapped files
1847 ---------------------------------------------------------------------
1848 This directory contains symbolic links which represent memory mapped files
1849 the process is maintaining. Example output:
1851 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
1852 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
1853 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
1854 | ...
1855 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
1856 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
1858 The name of a link represents the virtual memory bounds of a mapping, i.e.
1859 vm_area_struct::vm_start-vm_area_struct::vm_end.
1861 The main purpose of the map_files is to retrieve a set of memory mapped
1862 files in a fast way instead of parsing /proc/<pid>/maps or
1863 /proc/<pid>/smaps, both of which contain many more records. At the same
1864 time one can open(2) mappings from the listings of two processes and
1865 comparing their inode numbers to figure out which anonymous memory areas
1866 are actually shared.
1868 ------------------------------------------------------------------------------
1869 Configuring procfs
1870 ------------------------------------------------------------------------------
1872 4.1 Mount options
1873 ---------------------
1875 The following mount options are supported:
1877 hidepid= Set /proc/<pid>/ access mode.
1878 gid= Set the group authorized to learn processes information.
1880 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1881 (default).
1883 hidepid=1 means users may not access any /proc/<pid>/ directories but their
1884 own. Sensitive files like cmdline, sched*, status are now protected against
1885 other users. This makes it impossible to learn whether any user runs
1886 specific program (given the program doesn't reveal itself by its behaviour).
1887 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1888 poorly written programs passing sensitive information via program arguments are
1889 now protected against local eavesdroppers.
1891 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1892 users. It doesn't mean that it hides a fact whether a process with a specific
1893 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1894 but it hides process' uid and gid, which may be learned by stat()'ing
1895 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
1896 information about running processes, whether some daemon runs with elevated
1897 privileges, whether other user runs some sensitive program, whether other users
1898 run any program at all, etc.
1900 gid= defines a group authorized to learn processes information otherwise
1901 prohibited by hidepid=. If you use some daemon like identd which needs to learn
1902 information about processes information, just add identd to this group.