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
19 0.1 Introduction/Credits
22 1 Collecting System Information
23 1.1 Process-Specific Subdirectories
25 1.3 IDE devices in /proc/ide
26 1.4 Networking info in /proc/net
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
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
51 ------------------------------------------------------------------------------
53 ------------------------------------------------------------------------------
55 0.1 Introduction/Credits
56 ------------------------
58 This documentation is part of a soon (or so we hope) to be released book on
59 the SuSE Linux distribution. As there is no complete documentation for the
60 /proc file system and we've used many freely available sources to write these
61 chapters, it seems only fair to give the work back to the Linux community.
62 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
63 afraid it's still far from complete, but we hope it will be useful. As far as
64 we know, it is the first 'all-in-one' document about the /proc file system. It
65 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
66 SPARC, AXP, etc., features, you probably won't find what you are looking for.
67 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
68 additions and patches are welcome and will be added to this document if you
71 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
72 other people for help compiling this documentation. We'd also like to extend a
73 special thank you to Andi Kleen for documentation, which we relied on heavily
74 to create this document, as well as the additional information he provided.
75 Thanks to everybody else who contributed source or docs to the Linux kernel
76 and helped create a great piece of software... :)
78 If you have any comments, corrections or additions, please don't hesitate to
79 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
82 The latest version of this document is available online at
83 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
85 If the above direction does not works for you, you could try the kernel
86 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
87 comandante@zaralinux.com.
92 We don't guarantee the correctness of this document, and if you come to us
93 complaining about how you screwed up your system because of incorrect
94 documentation, we won't feel responsible...
96 ------------------------------------------------------------------------------
97 CHAPTER 1: COLLECTING SYSTEM INFORMATION
98 ------------------------------------------------------------------------------
100 ------------------------------------------------------------------------------
102 ------------------------------------------------------------------------------
103 * Investigating the properties of the pseudo file system /proc and its
104 ability to provide information on the running Linux system
105 * Examining /proc's structure
106 * Uncovering various information about the kernel and the processes running
108 ------------------------------------------------------------------------------
111 The proc file system acts as an interface to internal data structures in the
112 kernel. It can be used to obtain information about the system and to change
113 certain kernel parameters at runtime (sysctl).
115 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
116 show you how you can use /proc/sys to change settings.
118 1.1 Process-Specific Subdirectories
119 -----------------------------------
121 The directory /proc contains (among other things) one subdirectory for each
122 process running on the system, which is named after the process ID (PID).
124 The link self points to the process reading the file system. Each process
125 subdirectory has the entries listed in Table 1-1.
128 Table 1-1: Process specific entries in /proc
129 ..............................................................................
131 clear_refs Clears page referenced bits shown in smaps output
132 cmdline Command line arguments
133 cpu Current and last cpu in which it was executed (2.4)(smp)
134 cwd Link to the current working directory
135 environ Values of environment variables
136 exe Link to the executable of this process
137 fd Directory, which contains all file descriptors
138 maps Memory maps to executables and library files (2.4)
139 mem Memory held by this process
140 root Link to the root directory of this process
142 statm Process memory status information
143 status Process status in human readable form
144 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function
145 symbol the task is blocked in - or "0" if not blocked.
147 stack Report full stack trace, enable via CONFIG_STACKTRACE
148 smaps a extension based on maps, showing the memory consumption of
149 each mapping and flags associated with it
150 numa_maps an extension based on maps, showing the memory locality and
151 binding policy as well as mem usage (in pages) of each mapping.
152 ..............................................................................
154 For example, to get the status information of a process, all you have to do is
155 read the file /proc/PID/status:
157 >cat /proc/self/status
185 SigPnd: 0000000000000000
186 ShdPnd: 0000000000000000
187 SigBlk: 0000000000000000
188 SigIgn: 0000000000000000
189 SigCgt: 0000000000000000
190 CapInh: 00000000fffffeff
191 CapPrm: 0000000000000000
192 CapEff: 0000000000000000
193 CapBnd: ffffffffffffffff
195 voluntary_ctxt_switches: 0
196 nonvoluntary_ctxt_switches: 1
198 This shows you nearly the same information you would get if you viewed it with
199 the ps command. In fact, ps uses the proc file system to obtain its
200 information. But you get a more detailed view of the process by reading the
201 file /proc/PID/status. It fields are described in table 1-2.
203 The statm file contains more detailed information about the process
204 memory usage. Its seven fields are explained in Table 1-3. The stat file
205 contains details information about the process itself. Its fields are
206 explained in Table 1-4.
208 (for SMP CONFIG users)
209 For making accounting scalable, RSS related information are handled in an
210 asynchronous manner and the value may not be very precise. To see a precise
211 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
212 It's slow but very precise.
214 Table 1-2: Contents of the status files (as of 4.1)
215 ..............................................................................
217 Name filename of the executable
218 State state (R is running, S is sleeping, D is sleeping
219 in an uninterruptible wait, Z is zombie,
220 T is traced or stopped)
222 Ngid NUMA group ID (0 if none)
224 PPid process id of the parent process
225 TracerPid PID of process tracing this process (0 if not)
226 Uid Real, effective, saved set, and file system UIDs
227 Gid Real, effective, saved set, and file system GIDs
228 Umask file mode creation mask
229 FDSize number of file descriptor slots currently allocated
230 Groups supplementary group list
231 NStgid descendant namespace thread group ID hierarchy
232 NSpid descendant namespace process ID hierarchy
233 NSpgid descendant namespace process group ID hierarchy
234 NSsid descendant namespace session ID hierarchy
235 VmPeak peak virtual memory size
236 VmSize total program size
237 VmLck locked memory size
238 VmHWM peak resident set size ("high water mark")
239 VmRSS size of memory portions. It contains the three
240 following parts (VmRSS = RssAnon + RssFile + RssShmem)
241 RssAnon size of resident anonymous memory
242 RssFile size of resident file mappings
243 RssShmem size of resident shmem memory (includes SysV shm,
244 mapping of tmpfs and shared anonymous mappings)
245 VmData size of private data segments
246 VmStk size of stack segments
247 VmExe size of text segment
248 VmLib size of shared library code
249 VmPTE size of page table entries
250 VmPMD size of second level page tables
251 VmSwap amount of swap used by anonymous private data
252 (shmem swap usage is not included)
253 HugetlbPages size of hugetlb memory portions
254 Threads number of threads
255 SigQ number of signals queued/max. number for queue
256 SigPnd bitmap of pending signals for the thread
257 ShdPnd bitmap of shared pending signals for the process
258 SigBlk bitmap of blocked signals
259 SigIgn bitmap of ignored signals
260 SigCgt bitmap of caught signals
261 CapInh bitmap of inheritable capabilities
262 CapPrm bitmap of permitted capabilities
263 CapEff bitmap of effective capabilities
264 CapBnd bitmap of capabilities bounding set
265 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
266 Cpus_allowed mask of CPUs on which this process may run
267 Cpus_allowed_list Same as previous, but in "list format"
268 Mems_allowed mask of memory nodes allowed to this process
269 Mems_allowed_list Same as previous, but in "list format"
270 voluntary_ctxt_switches number of voluntary context switches
271 nonvoluntary_ctxt_switches number of non voluntary context switches
272 ..............................................................................
274 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
275 ..............................................................................
277 size total program size (pages) (same as VmSize in status)
278 resident size of memory portions (pages) (same as VmRSS in status)
279 shared number of pages that are shared (i.e. backed by a file, same
280 as RssFile+RssShmem in status)
281 trs number of pages that are 'code' (not including libs; broken,
282 includes data segment)
283 lrs number of pages of library (always 0 on 2.6)
284 drs number of pages of data/stack (including libs; broken,
285 includes library text)
286 dt number of dirty pages (always 0 on 2.6)
287 ..............................................................................
290 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
291 ..............................................................................
294 tcomm filename of the executable
295 state state (R is running, S is sleeping, D is sleeping in an
296 uninterruptible wait, Z is zombie, T is traced or stopped)
297 ppid process id of the parent process
298 pgrp pgrp of the process
300 tty_nr tty the process uses
301 tty_pgrp pgrp of the tty
303 min_flt number of minor faults
304 cmin_flt number of minor faults with child's
305 maj_flt number of major faults
306 cmaj_flt number of major faults with child's
307 utime user mode jiffies
308 stime kernel mode jiffies
309 cutime user mode jiffies with child's
310 cstime kernel mode jiffies with child's
311 priority priority level
313 num_threads number of threads
314 it_real_value (obsolete, always 0)
315 start_time time the process started after system boot
316 vsize virtual memory size
317 rss resident set memory size
318 rsslim current limit in bytes on the rss
319 start_code address above which program text can run
320 end_code address below which program text can run
321 start_stack address of the start of the main process stack
322 esp current value of ESP
323 eip current value of EIP
324 pending bitmap of pending signals
325 blocked bitmap of blocked signals
326 sigign bitmap of ignored signals
327 sigcatch bitmap of caught signals
328 0 (place holder, used to be the wchan address, use /proc/PID/wchan instead)
331 exit_signal signal to send to parent thread on exit
332 task_cpu which CPU the task is scheduled on
333 rt_priority realtime priority
334 policy scheduling policy (man sched_setscheduler)
335 blkio_ticks time spent waiting for block IO
336 gtime guest time of the task in jiffies
337 cgtime guest time of the task children in jiffies
338 start_data address above which program data+bss is placed
339 end_data address below which program data+bss is placed
340 start_brk address above which program heap can be expanded with brk()
341 arg_start address above which program command line is placed
342 arg_end address below which program command line is placed
343 env_start address above which program environment is placed
344 env_end address below which program environment is placed
345 exit_code the thread's exit_code in the form reported by the waitpid system call
346 ..............................................................................
348 The /proc/PID/maps file containing the currently mapped memory regions and
349 their access permissions.
353 address perms offset dev inode pathname
355 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
356 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
357 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
358 a7cb1000-a7cb2000 ---p 00000000 00:00 0
359 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
360 a7eb2000-a7eb3000 ---p 00000000 00:00 0
361 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
362 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
363 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
364 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
365 a800b000-a800e000 rw-p 00000000 00:00 0
366 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
367 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
368 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
369 a8024000-a8027000 rw-p 00000000 00:00 0
370 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
371 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
372 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
373 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
374 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
376 where "address" is the address space in the process that it occupies, "perms"
377 is a set of permissions:
383 p = private (copy on write)
385 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
386 "inode" is the inode on that device. 0 indicates that no inode is associated
387 with the memory region, as the case would be with BSS (uninitialized data).
388 The "pathname" shows the name associated file for this mapping. If the mapping
389 is not associated with a file:
391 [heap] = the heap of the program
392 [stack] = the stack of the main process
393 [vdso] = the "virtual dynamic shared object",
394 the kernel system call handler
396 or if empty, the mapping is anonymous.
398 The /proc/PID/task/TID/maps is a view of the virtual memory from the viewpoint
399 of the individual tasks of a process. In this file you will see a mapping marked
400 as [stack] if that task sees it as a stack. Hence, for the example above, the
401 task-level map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this:
403 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
404 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
405 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
406 a7cb1000-a7cb2000 ---p 00000000 00:00 0
407 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
408 a7eb2000-a7eb3000 ---p 00000000 00:00 0
409 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack]
410 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
411 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
412 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
413 a800b000-a800e000 rw-p 00000000 00:00 0
414 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
415 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
416 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
417 a8024000-a8027000 rw-p 00000000 00:00 0
418 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
419 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
420 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
421 aff35000-aff4a000 rw-p 00000000 00:00 0
422 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
424 The /proc/PID/smaps is an extension based on maps, showing the memory
425 consumption for each of the process's mappings. For each of mappings there
426 is a series of lines such as the following:
428 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
440 Private_Hugetlb: 0 kB
446 VmFlags: rd ex mr mw me dw
448 the first of these lines shows the same information as is displayed for the
449 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
450 (size), the amount of the mapping that is currently resident in RAM (RSS), the
451 process' proportional share of this mapping (PSS), the number of clean and
452 dirty private pages in the mapping.
454 The "proportional set size" (PSS) of a process is the count of pages it has
455 in memory, where each page is divided by the number of processes sharing it.
456 So if a process has 1000 pages all to itself, and 1000 shared with one other
457 process, its PSS will be 1500.
458 Note that even a page which is part of a MAP_SHARED mapping, but has only
459 a single pte mapped, i.e. is currently used by only one process, is accounted
460 as private and not as shared.
461 "Referenced" indicates the amount of memory currently marked as referenced or
463 "Anonymous" shows the amount of memory that does not belong to any file. Even
464 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
465 and a page is modified, the file page is replaced by a private anonymous copy.
466 "AnonHugePages" shows the ammount of memory backed by transparent hugepage.
467 "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
468 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
469 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
470 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
471 For shmem mappings, "Swap" includes also the size of the mapped (and not
472 replaced by copy-on-write) part of the underlying shmem object out on swap.
473 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
474 does not take into account swapped out page of underlying shmem objects.
475 "Locked" indicates whether the mapping is locked in memory or not.
477 "VmFlags" field deserves a separate description. This member represents the kernel
478 flags associated with the particular virtual memory area in two letter encoded
479 manner. The codes are the following:
488 gd - stack segment growns down
490 dw - disabled write to the mapped file
491 lo - pages are locked in memory
492 io - memory mapped I/O area
493 sr - sequential read advise provided
494 rr - random read advise provided
495 dc - do not copy area on fork
496 de - do not expand area on remapping
497 ac - area is accountable
498 nr - swap space is not reserved for the area
499 ht - area uses huge tlb pages
500 ar - architecture specific flag
501 dd - do not include area into core dump
504 hg - huge page advise flag
505 nh - no-huge page advise flag
506 mg - mergable advise flag
508 Note that there is no guarantee that every flag and associated mnemonic will
509 be present in all further kernel releases. Things get changed, the flags may
510 be vanished or the reverse -- new added.
512 This file is only present if the CONFIG_MMU kernel configuration option is
515 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
516 bits on both physical and virtual pages associated with a process, and the
517 soft-dirty bit on pte (see Documentation/vm/soft-dirty.txt for details).
518 To clear the bits for all the pages associated with the process
519 > echo 1 > /proc/PID/clear_refs
521 To clear the bits for the anonymous pages associated with the process
522 > echo 2 > /proc/PID/clear_refs
524 To clear the bits for the file mapped pages associated with the process
525 > echo 3 > /proc/PID/clear_refs
527 To clear the soft-dirty bit
528 > echo 4 > /proc/PID/clear_refs
530 To reset the peak resident set size ("high water mark") to the process's
532 > echo 5 > /proc/PID/clear_refs
534 Any other value written to /proc/PID/clear_refs will have no effect.
536 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
537 using /proc/kpageflags and number of times a page is mapped using
538 /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
540 The /proc/pid/numa_maps is an extension based on maps, showing the memory
541 locality and binding policy, as well as the memory usage (in pages) of
542 each mapping. The output follows a general format where mapping details get
543 summarized separated by blank spaces, one mapping per each file line:
545 address policy mapping details
547 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
548 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
549 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
550 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
551 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
552 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
553 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
554 320698b000 default file=/lib64/libc-2.12.so
555 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
556 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
557 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
558 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
559 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
560 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
561 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
562 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
565 "address" is the starting address for the mapping;
566 "policy" reports the NUMA memory policy set for the mapping (see vm/numa_memory_policy.txt);
567 "mapping details" summarizes mapping data such as mapping type, page usage counters,
568 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
569 size, in KB, that is backing the mapping up.
574 Similar to the process entries, the kernel data files give information about
575 the running kernel. The files used to obtain this information are contained in
576 /proc and are listed in Table 1-5. Not all of these will be present in your
577 system. It depends on the kernel configuration and the loaded modules, which
578 files are there, and which are missing.
580 Table 1-5: Kernel info in /proc
581 ..............................................................................
583 apm Advanced power management info
584 buddyinfo Kernel memory allocator information (see text) (2.5)
585 bus Directory containing bus specific information
586 cmdline Kernel command line
587 cpuinfo Info about the CPU
588 devices Available devices (block and character)
589 dma Used DMS channels
590 filesystems Supported filesystems
591 driver Various drivers grouped here, currently rtc (2.4)
592 execdomains Execdomains, related to security (2.4)
593 fb Frame Buffer devices (2.4)
594 fs File system parameters, currently nfs/exports (2.4)
595 ide Directory containing info about the IDE subsystem
596 interrupts Interrupt usage
597 iomem Memory map (2.4)
598 ioports I/O port usage
599 irq Masks for irq to cpu affinity (2.4)(smp?)
600 isapnp ISA PnP (Plug&Play) Info (2.4)
601 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
603 ksyms Kernel symbol table
604 loadavg Load average of last 1, 5 & 15 minutes
608 modules List of loaded modules
609 mounts Mounted filesystems
610 net Networking info (see text)
611 pagetypeinfo Additional page allocator information (see text) (2.5)
612 partitions Table of partitions known to the system
613 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
614 decoupled by lspci (2.4)
616 scsi SCSI info (see text)
617 slabinfo Slab pool info
618 softirqs softirq usage
619 stat Overall statistics
620 swaps Swap space utilization
622 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
623 tty Info of tty drivers
624 uptime Wall clock since boot, combined idle time of all cpus
625 version Kernel version
626 video bttv info of video resources (2.4)
627 vmallocinfo Show vmalloced areas
628 ..............................................................................
630 You can, for example, check which interrupts are currently in use and what
631 they are used for by looking in the file /proc/interrupts:
633 > cat /proc/interrupts
635 0: 8728810 XT-PIC timer
636 1: 895 XT-PIC keyboard
638 3: 531695 XT-PIC aha152x
639 4: 2014133 XT-PIC serial
640 5: 44401 XT-PIC pcnet_cs
643 12: 182918 XT-PIC PS/2 Mouse
645 14: 1232265 XT-PIC ide0
649 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
650 output of a SMP machine):
652 > cat /proc/interrupts
655 0: 1243498 1214548 IO-APIC-edge timer
656 1: 8949 8958 IO-APIC-edge keyboard
657 2: 0 0 XT-PIC cascade
658 5: 11286 10161 IO-APIC-edge soundblaster
659 8: 1 0 IO-APIC-edge rtc
660 9: 27422 27407 IO-APIC-edge 3c503
661 12: 113645 113873 IO-APIC-edge PS/2 Mouse
663 14: 22491 24012 IO-APIC-edge ide0
664 15: 2183 2415 IO-APIC-edge ide1
665 17: 30564 30414 IO-APIC-level eth0
666 18: 177 164 IO-APIC-level bttv
671 NMI is incremented in this case because every timer interrupt generates a NMI
672 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
674 LOC is the local interrupt counter of the internal APIC of every CPU.
676 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
677 connects the CPUs in a SMP system. This means that an error has been detected,
678 the IO-APIC automatically retry the transmission, so it should not be a big
679 problem, but you should read the SMP-FAQ.
681 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
682 /proc/interrupts to display every IRQ vector in use by the system, not
683 just those considered 'most important'. The new vectors are:
685 THR -- interrupt raised when a machine check threshold counter
686 (typically counting ECC corrected errors of memory or cache) exceeds
687 a configurable threshold. Only available on some systems.
689 TRM -- a thermal event interrupt occurs when a temperature threshold
690 has been exceeded for the CPU. This interrupt may also be generated
691 when the temperature drops back to normal.
693 SPU -- a spurious interrupt is some interrupt that was raised then lowered
694 by some IO device before it could be fully processed by the APIC. Hence
695 the APIC sees the interrupt but does not know what device it came from.
696 For this case the APIC will generate the interrupt with a IRQ vector
697 of 0xff. This might also be generated by chipset bugs.
699 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
700 sent from one CPU to another per the needs of the OS. Typically,
701 their statistics are used by kernel developers and interested users to
702 determine the occurrence of interrupts of the given type.
704 The above IRQ vectors are displayed only when relevant. For example,
705 the threshold vector does not exist on x86_64 platforms. Others are
706 suppressed when the system is a uniprocessor. As of this writing, only
707 i386 and x86_64 platforms support the new IRQ vector displays.
709 Of some interest is the introduction of the /proc/irq directory to 2.4.
710 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
711 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
712 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
717 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
718 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
722 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
723 IRQ, you can set it by doing:
725 > echo 1 > /proc/irq/10/smp_affinity
727 This means that only the first CPU will handle the IRQ, but you can also echo
728 5 which means that only the first and fourth CPU can handle the IRQ.
730 The contents of each smp_affinity file is the same by default:
732 > cat /proc/irq/0/smp_affinity
735 There is an alternate interface, smp_affinity_list which allows specifying
736 a cpu range instead of a bitmask:
738 > cat /proc/irq/0/smp_affinity_list
741 The default_smp_affinity mask applies to all non-active IRQs, which are the
742 IRQs which have not yet been allocated/activated, and hence which lack a
743 /proc/irq/[0-9]* directory.
745 The node file on an SMP system shows the node to which the device using the IRQ
746 reports itself as being attached. This hardware locality information does not
747 include information about any possible driver locality preference.
749 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
750 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
752 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
753 between all the CPUs which are allowed to handle it. As usual the kernel has
754 more info than you and does a better job than you, so the defaults are the
755 best choice for almost everyone. [Note this applies only to those IO-APIC's
756 that support "Round Robin" interrupt distribution.]
758 There are three more important subdirectories in /proc: net, scsi, and sys.
759 The general rule is that the contents, or even the existence of these
760 directories, depend on your kernel configuration. If SCSI is not enabled, the
761 directory scsi may not exist. The same is true with the net, which is there
762 only when networking support is present in the running kernel.
764 The slabinfo file gives information about memory usage at the slab level.
765 Linux uses slab pools for memory management above page level in version 2.2.
766 Commonly used objects have their own slab pool (such as network buffers,
767 directory cache, and so on).
769 ..............................................................................
771 > cat /proc/buddyinfo
773 Node 0, zone DMA 0 4 5 4 4 3 ...
774 Node 0, zone Normal 1 0 0 1 101 8 ...
775 Node 0, zone HighMem 2 0 0 1 1 0 ...
777 External fragmentation is a problem under some workloads, and buddyinfo is a
778 useful tool for helping diagnose these problems. Buddyinfo will give you a
779 clue as to how big an area you can safely allocate, or why a previous
782 Each column represents the number of pages of a certain order which are
783 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
784 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
785 available in ZONE_NORMAL, etc...
787 More information relevant to external fragmentation can be found in
790 > cat /proc/pagetypeinfo
794 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
795 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
796 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
797 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
798 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
799 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
800 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
801 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
802 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
803 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
804 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
806 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
807 Node 0, zone DMA 2 0 5 1 0
808 Node 0, zone DMA32 41 6 967 2 0
810 Fragmentation avoidance in the kernel works by grouping pages of different
811 migrate types into the same contiguous regions of memory called page blocks.
812 A page block is typically the size of the default hugepage size e.g. 2MB on
813 X86-64. By keeping pages grouped based on their ability to move, the kernel
814 can reclaim pages within a page block to satisfy a high-order allocation.
816 The pagetypinfo begins with information on the size of a page block. It
817 then gives the same type of information as buddyinfo except broken down
818 by migrate-type and finishes with details on how many page blocks of each
821 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
822 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
823 make an estimate of the likely number of huge pages that can be allocated
824 at a given point in time. All the "Movable" blocks should be allocatable
825 unless memory has been mlock()'d. Some of the Reclaimable blocks should
826 also be allocatable although a lot of filesystem metadata may have to be
827 reclaimed to achieve this.
829 ..............................................................................
833 Provides information about distribution and utilization of memory. This
834 varies by architecture and compile options. The following is from a
835 16GB PIII, which has highmem enabled. You may not have all of these fields.
839 MemTotal: 16344972 kB
841 MemAvailable: 14836172 kB
847 HighTotal: 15597528 kB
848 HighFree: 13629632 kB
859 SReclaimable: 159856 kB
860 SUnreclaim: 124508 kB
865 CommitLimit: 7669796 kB
866 Committed_AS: 100056 kB
867 VmallocTotal: 112216 kB
869 VmallocChunk: 111088 kB
870 AnonHugePages: 49152 kB
872 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
873 bits and the kernel binary code)
874 MemFree: The sum of LowFree+HighFree
875 MemAvailable: An estimate of how much memory is available for starting new
876 applications, without swapping. Calculated from MemFree,
877 SReclaimable, the size of the file LRU lists, and the low
878 watermarks in each zone.
879 The estimate takes into account that the system needs some
880 page cache to function well, and that not all reclaimable
881 slab will be reclaimable, due to items being in use. The
882 impact of those factors will vary from system to system.
883 Buffers: Relatively temporary storage for raw disk blocks
884 shouldn't get tremendously large (20MB or so)
885 Cached: in-memory cache for files read from the disk (the
886 pagecache). Doesn't include SwapCached
887 SwapCached: Memory that once was swapped out, is swapped back in but
888 still also is in the swapfile (if memory is needed it
889 doesn't need to be swapped out AGAIN because it is already
890 in the swapfile. This saves I/O)
891 Active: Memory that has been used more recently and usually not
892 reclaimed unless absolutely necessary.
893 Inactive: Memory which has been less recently used. It is more
894 eligible to be reclaimed for other purposes
896 HighFree: Highmem is all memory above ~860MB of physical memory
897 Highmem areas are for use by userspace programs, or
898 for the pagecache. The kernel must use tricks to access
899 this memory, making it slower to access than lowmem.
901 LowFree: Lowmem is memory which can be used for everything that
902 highmem can be used for, but it is also available for the
903 kernel's use for its own data structures. Among many
904 other things, it is where everything from the Slab is
905 allocated. Bad things happen when you're out of lowmem.
906 SwapTotal: total amount of swap space available
907 SwapFree: Memory which has been evicted from RAM, and is temporarily
909 Dirty: Memory which is waiting to get written back to the disk
910 Writeback: Memory which is actively being written back to the disk
911 AnonPages: Non-file backed pages mapped into userspace page tables
912 AnonHugePages: Non-file backed huge pages mapped into userspace page tables
913 Mapped: files which have been mmaped, such as libraries
914 Shmem: Total memory used by shared memory (shmem) and tmpfs
915 Slab: in-kernel data structures cache
916 SReclaimable: Part of Slab, that might be reclaimed, such as caches
917 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
918 PageTables: amount of memory dedicated to the lowest level of page
920 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
922 Bounce: Memory used for block device "bounce buffers"
923 WritebackTmp: Memory used by FUSE for temporary writeback buffers
924 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
925 this is the total amount of memory currently available to
926 be allocated on the system. This limit is only adhered to
927 if strict overcommit accounting is enabled (mode 2 in
928 'vm.overcommit_memory').
929 The CommitLimit is calculated with the following formula:
930 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
931 overcommit_ratio / 100 + [total swap pages]
932 For example, on a system with 1G of physical RAM and 7G
933 of swap with a `vm.overcommit_ratio` of 30 it would
934 yield a CommitLimit of 7.3G.
935 For more details, see the memory overcommit documentation
936 in vm/overcommit-accounting.
937 Committed_AS: The amount of memory presently allocated on the system.
938 The committed memory is a sum of all of the memory which
939 has been allocated by processes, even if it has not been
940 "used" by them as of yet. A process which malloc()'s 1G
941 of memory, but only touches 300M of it will show up as
942 using 1G. This 1G is memory which has been "committed" to
943 by the VM and can be used at any time by the allocating
944 application. With strict overcommit enabled on the system
945 (mode 2 in 'vm.overcommit_memory'),allocations which would
946 exceed the CommitLimit (detailed above) will not be permitted.
947 This is useful if one needs to guarantee that processes will
948 not fail due to lack of memory once that memory has been
949 successfully allocated.
950 VmallocTotal: total size of vmalloc memory area
951 VmallocUsed: amount of vmalloc area which is used
952 VmallocChunk: largest contiguous block of vmalloc area which is free
954 ..............................................................................
958 Provides information about vmalloced/vmaped areas. One line per area,
959 containing the virtual address range of the area, size in bytes,
960 caller information of the creator, and optional information depending
961 on the kind of area :
963 pages=nr number of pages
964 phys=addr if a physical address was specified
965 ioremap I/O mapping (ioremap() and friends)
966 vmalloc vmalloc() area
969 vpages buffer for pages pointers was vmalloced (huge area)
970 N<node>=nr (Only on NUMA kernels)
971 Number of pages allocated on memory node <node>
973 > cat /proc/vmallocinfo
974 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
975 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
976 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
977 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
978 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
979 phys=7fee8000 ioremap
980 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
981 phys=7fee7000 ioremap
982 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
983 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
984 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
985 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
987 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
988 /0x130 [x_tables] pages=4 vmalloc N0=4
989 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
990 pages=14 vmalloc N2=14
991 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
993 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
995 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
996 pages=10 vmalloc N0=10
998 ..............................................................................
1002 Provides counts of softirq handlers serviced since boot time, for each cpu.
1004 > cat /proc/softirqs
1007 TIMER: 27166 27120 27097 27034
1012 SCHED: 27035 26983 26971 26746
1014 RCU: 1678 1769 2178 2250
1017 1.3 IDE devices in /proc/ide
1018 ----------------------------
1020 The subdirectory /proc/ide contains information about all IDE devices of which
1021 the kernel is aware. There is one subdirectory for each IDE controller, the
1022 file drivers and a link for each IDE device, pointing to the device directory
1023 in the controller specific subtree.
1025 The file drivers contains general information about the drivers used for the
1028 > cat /proc/ide/drivers
1029 ide-cdrom version 4.53
1030 ide-disk version 1.08
1032 More detailed information can be found in the controller specific
1033 subdirectories. These are named ide0, ide1 and so on. Each of these
1034 directories contains the files shown in table 1-6.
1037 Table 1-6: IDE controller info in /proc/ide/ide?
1038 ..............................................................................
1040 channel IDE channel (0 or 1)
1041 config Configuration (only for PCI/IDE bridge)
1043 model Type/Chipset of IDE controller
1044 ..............................................................................
1046 Each device connected to a controller has a separate subdirectory in the
1047 controllers directory. The files listed in table 1-7 are contained in these
1051 Table 1-7: IDE device information
1052 ..............................................................................
1055 capacity Capacity of the medium (in 512Byte blocks)
1056 driver driver and version
1057 geometry physical and logical geometry
1058 identify device identify block
1060 model device identifier
1061 settings device setup
1062 smart_thresholds IDE disk management thresholds
1063 smart_values IDE disk management values
1064 ..............................................................................
1066 The most interesting file is settings. This file contains a nice overview of
1067 the drive parameters:
1069 # cat /proc/ide/ide0/hda/settings
1070 name value min max mode
1071 ---- ----- --- --- ----
1072 bios_cyl 526 0 65535 rw
1073 bios_head 255 0 255 rw
1074 bios_sect 63 0 63 rw
1075 breada_readahead 4 0 127 rw
1077 file_readahead 72 0 2097151 rw
1079 keepsettings 0 0 1 rw
1080 max_kb_per_request 122 1 127 rw
1084 pio_mode write-only 0 255 w
1090 1.4 Networking info in /proc/net
1091 --------------------------------
1093 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1094 additional values you get for IP version 6 if you configure the kernel to
1095 support this. Table 1-9 lists the files and their meaning.
1098 Table 1-8: IPv6 info in /proc/net
1099 ..............................................................................
1101 udp6 UDP sockets (IPv6)
1102 tcp6 TCP sockets (IPv6)
1103 raw6 Raw device statistics (IPv6)
1104 igmp6 IP multicast addresses, which this host joined (IPv6)
1105 if_inet6 List of IPv6 interface addresses
1106 ipv6_route Kernel routing table for IPv6
1107 rt6_stats Global IPv6 routing tables statistics
1108 sockstat6 Socket statistics (IPv6)
1109 snmp6 Snmp data (IPv6)
1110 ..............................................................................
1113 Table 1-9: Network info in /proc/net
1114 ..............................................................................
1116 arp Kernel ARP table
1117 dev network devices with statistics
1118 dev_mcast the Layer2 multicast groups a device is listening too
1119 (interface index, label, number of references, number of bound
1121 dev_stat network device status
1122 ip_fwchains Firewall chain linkage
1123 ip_fwnames Firewall chain names
1124 ip_masq Directory containing the masquerading tables
1125 ip_masquerade Major masquerading table
1126 netstat Network statistics
1127 raw raw device statistics
1128 route Kernel routing table
1129 rpc Directory containing rpc info
1130 rt_cache Routing cache
1132 sockstat Socket statistics
1135 unix UNIX domain sockets
1136 wireless Wireless interface data (Wavelan etc)
1137 igmp IP multicast addresses, which this host joined
1138 psched Global packet scheduler parameters.
1139 netlink List of PF_NETLINK sockets
1140 ip_mr_vifs List of multicast virtual interfaces
1141 ip_mr_cache List of multicast routing cache
1142 ..............................................................................
1144 You can use this information to see which network devices are available in
1145 your system and how much traffic was routed over those devices:
1148 Inter-|Receive |[...
1149 face |bytes packets errs drop fifo frame compressed multicast|[...
1150 lo: 908188 5596 0 0 0 0 0 0 [...
1151 ppp0:15475140 20721 410 0 0 410 0 0 [...
1152 eth0: 614530 7085 0 0 0 0 0 1 [...
1155 ...] bytes packets errs drop fifo colls carrier compressed
1156 ...] 908188 5596 0 0 0 0 0 0
1157 ...] 1375103 17405 0 0 0 0 0 0
1158 ...] 1703981 5535 0 0 0 3 0 0
1160 In addition, each Channel Bond interface has its own directory. For
1161 example, the bond0 device will have a directory called /proc/net/bond0/.
1162 It will contain information that is specific to that bond, such as the
1163 current slaves of the bond, the link status of the slaves, and how
1164 many times the slaves link has failed.
1169 If you have a SCSI host adapter in your system, you'll find a subdirectory
1170 named after the driver for this adapter in /proc/scsi. You'll also see a list
1171 of all recognized SCSI devices in /proc/scsi:
1173 >cat /proc/scsi/scsi
1175 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1176 Vendor: IBM Model: DGHS09U Rev: 03E0
1177 Type: Direct-Access ANSI SCSI revision: 03
1178 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1179 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1180 Type: CD-ROM ANSI SCSI revision: 02
1183 The directory named after the driver has one file for each adapter found in
1184 the system. These files contain information about the controller, including
1185 the used IRQ and the IO address range. The amount of information shown is
1186 dependent on the adapter you use. The example shows the output for an Adaptec
1187 AHA-2940 SCSI adapter:
1189 > cat /proc/scsi/aic7xxx/0
1191 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1193 TCQ Enabled By Default : Disabled
1194 AIC7XXX_PROC_STATS : Disabled
1195 AIC7XXX_RESET_DELAY : 5
1196 Adapter Configuration:
1197 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1198 Ultra Wide Controller
1199 PCI MMAPed I/O Base: 0xeb001000
1200 Adapter SEEPROM Config: SEEPROM found and used.
1201 Adaptec SCSI BIOS: Enabled
1203 SCBs: Active 0, Max Active 2,
1204 Allocated 15, HW 16, Page 255
1206 BIOS Control Word: 0x18b6
1207 Adapter Control Word: 0x005b
1208 Extended Translation: Enabled
1209 Disconnect Enable Flags: 0xffff
1210 Ultra Enable Flags: 0x0001
1211 Tag Queue Enable Flags: 0x0000
1212 Ordered Queue Tag Flags: 0x0000
1213 Default Tag Queue Depth: 8
1214 Tagged Queue By Device array for aic7xxx host instance 0:
1215 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1216 Actual queue depth per device for aic7xxx host instance 0:
1217 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1220 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1221 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1222 Total transfers 160151 (74577 reads and 85574 writes)
1224 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1225 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1226 Total transfers 0 (0 reads and 0 writes)
1229 1.6 Parallel port info in /proc/parport
1230 ---------------------------------------
1232 The directory /proc/parport contains information about the parallel ports of
1233 your system. It has one subdirectory for each port, named after the port
1236 These directories contain the four files shown in Table 1-10.
1239 Table 1-10: Files in /proc/parport
1240 ..............................................................................
1242 autoprobe Any IEEE-1284 device ID information that has been acquired.
1243 devices list of the device drivers using that port. A + will appear by the
1244 name of the device currently using the port (it might not appear
1246 hardware Parallel port's base address, IRQ line and DMA channel.
1247 irq IRQ that parport is using for that port. This is in a separate
1248 file to allow you to alter it by writing a new value in (IRQ
1250 ..............................................................................
1252 1.7 TTY info in /proc/tty
1253 -------------------------
1255 Information about the available and actually used tty's can be found in the
1256 directory /proc/tty.You'll find entries for drivers and line disciplines in
1257 this directory, as shown in Table 1-11.
1260 Table 1-11: Files in /proc/tty
1261 ..............................................................................
1263 drivers list of drivers and their usage
1264 ldiscs registered line disciplines
1265 driver/serial usage statistic and status of single tty lines
1266 ..............................................................................
1268 To see which tty's are currently in use, you can simply look into the file
1271 > cat /proc/tty/drivers
1272 pty_slave /dev/pts 136 0-255 pty:slave
1273 pty_master /dev/ptm 128 0-255 pty:master
1274 pty_slave /dev/ttyp 3 0-255 pty:slave
1275 pty_master /dev/pty 2 0-255 pty:master
1276 serial /dev/cua 5 64-67 serial:callout
1277 serial /dev/ttyS 4 64-67 serial
1278 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1279 /dev/ptmx /dev/ptmx 5 2 system
1280 /dev/console /dev/console 5 1 system:console
1281 /dev/tty /dev/tty 5 0 system:/dev/tty
1282 unknown /dev/tty 4 1-63 console
1285 1.8 Miscellaneous kernel statistics in /proc/stat
1286 -------------------------------------------------
1288 Various pieces of information about kernel activity are available in the
1289 /proc/stat file. All of the numbers reported in this file are aggregates
1290 since the system first booted. For a quick look, simply cat the file:
1293 cpu 2255 34 2290 22625563 6290 127 456 0 0 0
1294 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0
1295 cpu1 1123 0 849 11313845 2614 0 18 0 0 0
1296 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1302 softirq 183433 0 21755 12 39 1137 231 21459 2263
1304 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1305 lines. These numbers identify the amount of time the CPU has spent performing
1306 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1307 second). The meanings of the columns are as follows, from left to right:
1309 - user: normal processes executing in user mode
1310 - nice: niced processes executing in user mode
1311 - system: processes executing in kernel mode
1312 - idle: twiddling thumbs
1313 - iowait: waiting for I/O to complete
1314 - irq: servicing interrupts
1315 - softirq: servicing softirqs
1316 - steal: involuntary wait
1317 - guest: running a normal guest
1318 - guest_nice: running a niced guest
1320 The "intr" line gives counts of interrupts serviced since boot time, for each
1321 of the possible system interrupts. The first column is the total of all
1322 interrupts serviced including unnumbered architecture specific interrupts;
1323 each subsequent column is the total for that particular numbered interrupt.
1324 Unnumbered interrupts are not shown, only summed into the total.
1326 The "ctxt" line gives the total number of context switches across all CPUs.
1328 The "btime" line gives the time at which the system booted, in seconds since
1331 The "processes" line gives the number of processes and threads created, which
1332 includes (but is not limited to) those created by calls to the fork() and
1333 clone() system calls.
1335 The "procs_running" line gives the total number of threads that are
1336 running or ready to run (i.e., the total number of runnable threads).
1338 The "procs_blocked" line gives the number of processes currently blocked,
1339 waiting for I/O to complete.
1341 The "softirq" line gives counts of softirqs serviced since boot time, for each
1342 of the possible system softirqs. The first column is the total of all
1343 softirqs serviced; each subsequent column is the total for that particular
1347 1.9 Ext4 file system parameters
1348 -------------------------------
1350 Information about mounted ext4 file systems can be found in
1351 /proc/fs/ext4. Each mounted filesystem will have a directory in
1352 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1353 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1354 in Table 1-12, below.
1356 Table 1-12: Files in /proc/fs/ext4/<devname>
1357 ..............................................................................
1359 mb_groups details of multiblock allocator buddy cache of free blocks
1360 ..............................................................................
1364 Shows registered system console lines.
1366 To see which character device lines are currently used for the system console
1367 /dev/console, you may simply look into the file /proc/consoles:
1369 > cat /proc/consoles
1375 device name of the device
1376 operations R = can do read operations
1377 W = can do write operations
1379 flags E = it is enabled
1380 C = it is preferred console
1381 B = it is primary boot console
1382 p = it is used for printk buffer
1383 b = it is not a TTY but a Braille device
1384 a = it is safe to use when cpu is offline
1385 major:minor major and minor number of the device separated by a colon
1387 ------------------------------------------------------------------------------
1389 ------------------------------------------------------------------------------
1390 The /proc file system serves information about the running system. It not only
1391 allows access to process data but also allows you to request the kernel status
1392 by reading files in the hierarchy.
1394 The directory structure of /proc reflects the types of information and makes
1395 it easy, if not obvious, where to look for specific data.
1396 ------------------------------------------------------------------------------
1398 ------------------------------------------------------------------------------
1399 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1400 ------------------------------------------------------------------------------
1402 ------------------------------------------------------------------------------
1404 ------------------------------------------------------------------------------
1405 * Modifying kernel parameters by writing into files found in /proc/sys
1406 * Exploring the files which modify certain parameters
1407 * Review of the /proc/sys file tree
1408 ------------------------------------------------------------------------------
1411 A very interesting part of /proc is the directory /proc/sys. This is not only
1412 a source of information, it also allows you to change parameters within the
1413 kernel. Be very careful when attempting this. You can optimize your system,
1414 but you can also cause it to crash. Never alter kernel parameters on a
1415 production system. Set up a development machine and test to make sure that
1416 everything works the way you want it to. You may have no alternative but to
1417 reboot the machine once an error has been made.
1419 To change a value, simply echo the new value into the file. An example is
1420 given below in the section on the file system data. You need to be root to do
1421 this. You can create your own boot script to perform this every time your
1424 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1425 general things in the operation of the Linux kernel. Since some of the files
1426 can inadvertently disrupt your system, it is advisable to read both
1427 documentation and source before actually making adjustments. In any case, be
1428 very careful when writing to any of these files. The entries in /proc may
1429 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1430 review the kernel documentation in the directory /usr/src/linux/Documentation.
1431 This chapter is heavily based on the documentation included in the pre 2.2
1432 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1434 Please see: Documentation/sysctl/ directory for descriptions of these
1437 ------------------------------------------------------------------------------
1439 ------------------------------------------------------------------------------
1440 Certain aspects of kernel behavior can be modified at runtime, without the
1441 need to recompile the kernel, or even to reboot the system. The files in the
1442 /proc/sys tree can not only be read, but also modified. You can use the echo
1443 command to write value into these files, thereby changing the default settings
1445 ------------------------------------------------------------------------------
1447 ------------------------------------------------------------------------------
1448 CHAPTER 3: PER-PROCESS PARAMETERS
1449 ------------------------------------------------------------------------------
1451 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1452 --------------------------------------------------------------------------------
1454 These file can be used to adjust the badness heuristic used to select which
1455 process gets killed in out of memory conditions.
1457 The badness heuristic assigns a value to each candidate task ranging from 0
1458 (never kill) to 1000 (always kill) to determine which process is targeted. The
1459 units are roughly a proportion along that range of allowed memory the process
1460 may allocate from based on an estimation of its current memory and swap use.
1461 For example, if a task is using all allowed memory, its badness score will be
1462 1000. If it is using half of its allowed memory, its score will be 500.
1464 There is an additional factor included in the badness score: the current memory
1465 and swap usage is discounted by 3% for root processes.
1467 The amount of "allowed" memory depends on the context in which the oom killer
1468 was called. If it is due to the memory assigned to the allocating task's cpuset
1469 being exhausted, the allowed memory represents the set of mems assigned to that
1470 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1471 memory represents the set of mempolicy nodes. If it is due to a memory
1472 limit (or swap limit) being reached, the allowed memory is that configured
1473 limit. Finally, if it is due to the entire system being out of memory, the
1474 allowed memory represents all allocatable resources.
1476 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1477 is used to determine which task to kill. Acceptable values range from -1000
1478 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1479 polarize the preference for oom killing either by always preferring a certain
1480 task or completely disabling it. The lowest possible value, -1000, is
1481 equivalent to disabling oom killing entirely for that task since it will always
1482 report a badness score of 0.
1484 Consequently, it is very simple for userspace to define the amount of memory to
1485 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1486 example, is roughly equivalent to allowing the remainder of tasks sharing the
1487 same system, cpuset, mempolicy, or memory controller resources to use at least
1488 50% more memory. A value of -500, on the other hand, would be roughly
1489 equivalent to discounting 50% of the task's allowed memory from being considered
1490 as scoring against the task.
1492 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1493 be used to tune the badness score. Its acceptable values range from -16
1494 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1495 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1496 scaled linearly with /proc/<pid>/oom_score_adj.
1498 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1499 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1500 requires CAP_SYS_RESOURCE.
1502 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1503 generation children with separate address spaces instead, if possible. This
1504 avoids servers and important system daemons from being killed and loses the
1505 minimal amount of work.
1508 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1509 -------------------------------------------------------------
1511 This file can be used to check the current score used by the oom-killer is for
1512 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1513 process should be killed in an out-of-memory situation.
1516 3.3 /proc/<pid>/io - Display the IO accounting fields
1517 -------------------------------------------------------
1519 This file contains IO statistics for each running process
1524 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1527 test:/tmp # cat /proc/3828/io
1533 write_bytes: 323932160
1534 cancelled_write_bytes: 0
1543 I/O counter: chars read
1544 The number of bytes which this task has caused to be read from storage. This
1545 is simply the sum of bytes which this process passed to read() and pread().
1546 It includes things like tty IO and it is unaffected by whether or not actual
1547 physical disk IO was required (the read might have been satisfied from
1554 I/O counter: chars written
1555 The number of bytes which this task has caused, or shall cause to be written
1556 to disk. Similar caveats apply here as with rchar.
1562 I/O counter: read syscalls
1563 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1570 I/O counter: write syscalls
1571 Attempt to count the number of write I/O operations, i.e. syscalls like
1572 write() and pwrite().
1578 I/O counter: bytes read
1579 Attempt to count the number of bytes which this process really did cause to
1580 be fetched from the storage layer. Done at the submit_bio() level, so it is
1581 accurate for block-backed filesystems. <please add status regarding NFS and
1582 CIFS at a later time>
1588 I/O counter: bytes written
1589 Attempt to count the number of bytes which this process caused to be sent to
1590 the storage layer. This is done at page-dirtying time.
1593 cancelled_write_bytes
1594 ---------------------
1596 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1597 then deletes the file, it will in fact perform no writeout. But it will have
1598 been accounted as having caused 1MB of write.
1599 In other words: The number of bytes which this process caused to not happen,
1600 by truncating pagecache. A task can cause "negative" IO too. If this task
1601 truncates some dirty pagecache, some IO which another task has been accounted
1602 for (in its write_bytes) will not be happening. We _could_ just subtract that
1603 from the truncating task's write_bytes, but there is information loss in doing
1610 At its current implementation state, this is a bit racy on 32-bit machines: if
1611 process A reads process B's /proc/pid/io while process B is updating one of
1612 those 64-bit counters, process A could see an intermediate result.
1615 More information about this can be found within the taskstats documentation in
1616 Documentation/accounting.
1618 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1619 ---------------------------------------------------------------
1620 When a process is dumped, all anonymous memory is written to a core file as
1621 long as the size of the core file isn't limited. But sometimes we don't want
1622 to dump some memory segments, for example, huge shared memory or DAX.
1623 Conversely, sometimes we want to save file-backed memory segments into a core
1624 file, not only the individual files.
1626 /proc/<pid>/coredump_filter allows you to customize which memory segments
1627 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1628 of memory types. If a bit of the bitmask is set, memory segments of the
1629 corresponding memory type are dumped, otherwise they are not dumped.
1631 The following 9 memory types are supported:
1632 - (bit 0) anonymous private memory
1633 - (bit 1) anonymous shared memory
1634 - (bit 2) file-backed private memory
1635 - (bit 3) file-backed shared memory
1636 - (bit 4) ELF header pages in file-backed private memory areas (it is
1637 effective only if the bit 2 is cleared)
1638 - (bit 5) hugetlb private memory
1639 - (bit 6) hugetlb shared memory
1640 - (bit 7) DAX private memory
1641 - (bit 8) DAX shared memory
1643 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1644 are always dumped regardless of the bitmask status.
1646 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1647 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1649 The default value of coredump_filter is 0x33; this means all anonymous memory
1650 segments, ELF header pages and hugetlb private memory are dumped.
1652 If you don't want to dump all shared memory segments attached to pid 1234,
1653 write 0x31 to the process's proc file.
1655 $ echo 0x31 > /proc/1234/coredump_filter
1657 When a new process is created, the process inherits the bitmask status from its
1658 parent. It is useful to set up coredump_filter before the program runs.
1661 $ echo 0x7 > /proc/self/coredump_filter
1664 3.5 /proc/<pid>/mountinfo - Information about mounts
1665 --------------------------------------------------------
1667 This file contains lines of the form:
1669 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1670 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1672 (1) mount ID: unique identifier of the mount (may be reused after umount)
1673 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1674 (3) major:minor: value of st_dev for files on filesystem
1675 (4) root: root of the mount within the filesystem
1676 (5) mount point: mount point relative to the process's root
1677 (6) mount options: per mount options
1678 (7) optional fields: zero or more fields of the form "tag[:value]"
1679 (8) separator: marks the end of the optional fields
1680 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1681 (10) mount source: filesystem specific information or "none"
1682 (11) super options: per super block options
1684 Parsers should ignore all unrecognised optional fields. Currently the
1685 possible optional fields are:
1687 shared:X mount is shared in peer group X
1688 master:X mount is slave to peer group X
1689 propagate_from:X mount is slave and receives propagation from peer group X (*)
1690 unbindable mount is unbindable
1692 (*) X is the closest dominant peer group under the process's root. If
1693 X is the immediate master of the mount, or if there's no dominant peer
1694 group under the same root, then only the "master:X" field is present
1695 and not the "propagate_from:X" field.
1697 For more information on mount propagation see:
1699 Documentation/filesystems/sharedsubtree.txt
1702 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1703 --------------------------------------------------------
1704 These files provide a method to access a tasks comm value. It also allows for
1705 a task to set its own or one of its thread siblings comm value. The comm value
1706 is limited in size compared to the cmdline value, so writing anything longer
1707 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1711 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1712 -------------------------------------------------------------------------
1713 This file provides a fast way to retrieve first level children pids
1714 of a task pointed by <pid>/<tid> pair. The format is a space separated
1717 Note the "first level" here -- if a child has own children they will
1718 not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1719 to obtain the descendants.
1721 Since this interface is intended to be fast and cheap it doesn't
1722 guarantee to provide precise results and some children might be
1723 skipped, especially if they've exited right after we printed their
1724 pids, so one need to either stop or freeze processes being inspected
1725 if precise results are needed.
1728 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1729 ---------------------------------------------------------------
1730 This file provides information associated with an opened file. The regular
1731 files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos'
1732 represents the current offset of the opened file in decimal form [see lseek(2)
1733 for details], 'flags' denotes the octal O_xxx mask the file has been
1734 created with [see open(2) for details] and 'mnt_id' represents mount ID of
1735 the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo
1744 All locks associated with a file descriptor are shown in its fdinfo too.
1746 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1748 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1749 pair provide additional information particular to the objects they represent.
1758 where 'eventfd-count' is hex value of a counter.
1765 sigmask: 0000000000000200
1767 where 'sigmask' is hex value of the signal mask associated
1775 tfd: 5 events: 1d data: ffffffffffffffff
1777 where 'tfd' is a target file descriptor number in decimal form,
1778 'events' is events mask being watched and the 'data' is data
1779 associated with a target [see epoll(7) for more details].
1783 For inotify files the format is the following
1787 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1789 where 'wd' is a watch descriptor in decimal form, ie a target file
1790 descriptor number, 'ino' and 'sdev' are inode and device where the
1791 target file resides and the 'mask' is the mask of events, all in hex
1792 form [see inotify(7) for more details].
1794 If the kernel was built with exportfs support, the path to the target
1795 file is encoded as a file handle. The file handle is provided by three
1796 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1799 If the kernel is built without exportfs support the file handle won't be
1802 If there is no inotify mark attached yet the 'inotify' line will be omitted.
1804 For fanotify files the format is
1809 fanotify flags:10 event-flags:0
1810 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1811 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
1813 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
1814 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
1815 flags associated with mark which are tracked separately from events
1816 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
1817 mask and 'ignored_mask' is the mask of events which are to be ignored.
1818 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
1819 does provide information about flags and mask used in fanotify_mark
1820 call [see fsnotify manpage for details].
1822 While the first three lines are mandatory and always printed, the rest is
1823 optional and may be omitted if no marks created yet.
1834 it_value: (0, 49406829)
1837 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
1838 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
1839 flags in octal form been used to setup the timer [see timerfd_settime(2) for
1840 details]. 'it_value' is remaining time until the timer exiration.
1841 'it_interval' is the interval for the timer. Note the timer might be set up
1842 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
1843 still exhibits timer's remaining time.
1845 3.9 /proc/<pid>/map_files - Information about memory mapped files
1846 ---------------------------------------------------------------------
1847 This directory contains symbolic links which represent memory mapped files
1848 the process is maintaining. Example output:
1850 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
1851 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
1852 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
1854 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
1855 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
1857 The name of a link represents the virtual memory bounds of a mapping, i.e.
1858 vm_area_struct::vm_start-vm_area_struct::vm_end.
1860 The main purpose of the map_files is to retrieve a set of memory mapped
1861 files in a fast way instead of parsing /proc/<pid>/maps or
1862 /proc/<pid>/smaps, both of which contain many more records. At the same
1863 time one can open(2) mappings from the listings of two processes and
1864 comparing their inode numbers to figure out which anonymous memory areas
1865 are actually shared.
1867 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
1868 ---------------------------------------------------------
1869 This file provides the value of the task's timerslack value in nanoseconds.
1870 This value specifies a amount of time that normal timers may be deferred
1871 in order to coalesce timers and avoid unnecessary wakeups.
1873 This allows a task's interactivity vs power consumption trade off to be
1876 Writing 0 to the file will set the tasks timerslack to the default value.
1878 Valid values are from 0 - ULLONG_MAX
1880 An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
1881 permissions on the task specified to change its timerslack_ns value.
1884 ------------------------------------------------------------------------------
1886 ------------------------------------------------------------------------------
1889 ---------------------
1891 The following mount options are supported:
1893 hidepid= Set /proc/<pid>/ access mode.
1894 gid= Set the group authorized to learn processes information.
1896 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1899 hidepid=1 means users may not access any /proc/<pid>/ directories but their
1900 own. Sensitive files like cmdline, sched*, status are now protected against
1901 other users. This makes it impossible to learn whether any user runs
1902 specific program (given the program doesn't reveal itself by its behaviour).
1903 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1904 poorly written programs passing sensitive information via program arguments are
1905 now protected against local eavesdroppers.
1907 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1908 users. It doesn't mean that it hides a fact whether a process with a specific
1909 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1910 but it hides process' uid and gid, which may be learned by stat()'ing
1911 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
1912 information about running processes, whether some daemon runs with elevated
1913 privileges, whether other user runs some sensitive program, whether other users
1914 run any program at all, etc.
1916 gid= defines a group authorized to learn processes information otherwise
1917 prohibited by hidepid=. If you use some daemon like identd which needs to learn
1918 information about processes information, just add identd to this group.