blob: 0d07513a67a661440fc1599121bb1dc38560bc33 [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001------------------------------------------------------------------------------
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>
6
72.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
Stefani Seibold349888e2009-06-17 16:26:01 -07008move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009
Linus Torvalds1da177e2005-04-16 15:20:36 -07009------------------------------------------------------------------------------
10Version 1.3 Kernel version 2.2.12
11 Kernel version 2.4.0-test11-pre4
12------------------------------------------------------------------------------
Stefani Seibold349888e2009-06-17 16:26:01 -070013fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009
Linus Torvalds1da177e2005-04-16 15:20:36 -070014
15Table of Contents
16-----------------
17
18 0 Preface
19 0.1 Introduction/Credits
20 0.2 Legal Stuff
21
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
Shen Feng760df932009-04-02 16:57:20 -070031 1.9 Ext4 file system parameters
Linus Torvalds1da177e2005-04-16 15:20:36 -070032
33 2 Modifying System Parameters
Shen Feng760df932009-04-02 16:57:20 -070034
35 3 Per-Process Parameters
36 3.1 /proc/<pid>/oom_adj - Adjust the oom-killer score
37 3.2 /proc/<pid>/oom_score - Display current oom-killer score
38 3.3 /proc/<pid>/io - Display the IO accounting fields
39 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
40 3.5 /proc/<pid>/mountinfo - Information about mounts
john stultz4614a696b2009-12-14 18:00:05 -080041 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
Shen Feng760df932009-04-02 16:57:20 -070042
Linus Torvalds1da177e2005-04-16 15:20:36 -070043
44------------------------------------------------------------------------------
45Preface
46------------------------------------------------------------------------------
47
480.1 Introduction/Credits
49------------------------
50
51This documentation is part of a soon (or so we hope) to be released book on
52the SuSE Linux distribution. As there is no complete documentation for the
53/proc file system and we've used many freely available sources to write these
54chapters, it seems only fair to give the work back to the Linux community.
55This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
56afraid it's still far from complete, but we hope it will be useful. As far as
57we know, it is the first 'all-in-one' document about the /proc file system. It
58is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
59SPARC, AXP, etc., features, you probably won't find what you are looking for.
60It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
61additions and patches are welcome and will be added to this document if you
62mail them to Bodo.
63
64We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
65other people for help compiling this documentation. We'd also like to extend a
66special thank you to Andi Kleen for documentation, which we relied on heavily
67to create this document, as well as the additional information he provided.
68Thanks to everybody else who contributed source or docs to the Linux kernel
69and helped create a great piece of software... :)
70
71If you have any comments, corrections or additions, please don't hesitate to
72contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
73document.
74
75The latest version of this document is available online at
76http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
77
78If the above direction does not works for you, ypu could try the kernel
79mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
80comandante@zaralinux.com.
81
820.2 Legal Stuff
83---------------
84
85We don't guarantee the correctness of this document, and if you come to us
86complaining about how you screwed up your system because of incorrect
87documentation, we won't feel responsible...
88
89------------------------------------------------------------------------------
90CHAPTER 1: COLLECTING SYSTEM INFORMATION
91------------------------------------------------------------------------------
92
93------------------------------------------------------------------------------
94In This Chapter
95------------------------------------------------------------------------------
96* Investigating the properties of the pseudo file system /proc and its
97 ability to provide information on the running Linux system
98* Examining /proc's structure
99* Uncovering various information about the kernel and the processes running
100 on the system
101------------------------------------------------------------------------------
102
103
104The proc file system acts as an interface to internal data structures in the
105kernel. It can be used to obtain information about the system and to change
106certain kernel parameters at runtime (sysctl).
107
108First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
109show you how you can use /proc/sys to change settings.
110
1111.1 Process-Specific Subdirectories
112-----------------------------------
113
114The directory /proc contains (among other things) one subdirectory for each
115process running on the system, which is named after the process ID (PID).
116
117The link self points to the process reading the file system. Each process
118subdirectory has the entries listed in Table 1-1.
119
120
Stefani Seibold349888e2009-06-17 16:26:01 -0700121Table 1-1: Process specific entries in /proc
Linus Torvalds1da177e2005-04-16 15:20:36 -0700122..............................................................................
David Rientjesb813e932007-05-06 14:49:24 -0700123 File Content
124 clear_refs Clears page referenced bits shown in smaps output
125 cmdline Command line arguments
126 cpu Current and last cpu in which it was executed (2.4)(smp)
127 cwd Link to the current working directory
128 environ Values of environment variables
129 exe Link to the executable of this process
130 fd Directory, which contains all file descriptors
131 maps Memory maps to executables and library files (2.4)
132 mem Memory held by this process
133 root Link to the root directory of this process
134 stat Process status
135 statm Process memory status information
136 status Process status in human readable form
137 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
Ken Chen2ec220e2008-11-10 11:26:08 +0300138 stack Report full stack trace, enable via CONFIG_STACKTRACE
Stefani Seibold349888e2009-06-17 16:26:01 -0700139 smaps a extension based on maps, showing the memory consumption of
140 each mapping
Linus Torvalds1da177e2005-04-16 15:20:36 -0700141..............................................................................
142
143For example, to get the status information of a process, all you have to do is
144read the file /proc/PID/status:
145
Stefani Seibold349888e2009-06-17 16:26:01 -0700146 >cat /proc/self/status
147 Name: cat
148 State: R (running)
149 Tgid: 5452
150 Pid: 5452
151 PPid: 743
Linus Torvalds1da177e2005-04-16 15:20:36 -0700152 TracerPid: 0 (2.4)
Stefani Seibold349888e2009-06-17 16:26:01 -0700153 Uid: 501 501 501 501
154 Gid: 100 100 100 100
155 FDSize: 256
156 Groups: 100 14 16
157 VmPeak: 5004 kB
158 VmSize: 5004 kB
159 VmLck: 0 kB
160 VmHWM: 476 kB
161 VmRSS: 476 kB
162 VmData: 156 kB
163 VmStk: 88 kB
164 VmExe: 68 kB
165 VmLib: 1412 kB
166 VmPTE: 20 kb
167 Threads: 1
168 SigQ: 0/28578
169 SigPnd: 0000000000000000
170 ShdPnd: 0000000000000000
171 SigBlk: 0000000000000000
172 SigIgn: 0000000000000000
173 SigCgt: 0000000000000000
174 CapInh: 00000000fffffeff
175 CapPrm: 0000000000000000
176 CapEff: 0000000000000000
177 CapBnd: ffffffffffffffff
178 voluntary_ctxt_switches: 0
179 nonvoluntary_ctxt_switches: 1
Linus Torvalds1da177e2005-04-16 15:20:36 -0700180
181This shows you nearly the same information you would get if you viewed it with
182the ps command. In fact, ps uses the proc file system to obtain its
Stefani Seibold349888e2009-06-17 16:26:01 -0700183information. But you get a more detailed view of the process by reading the
184file /proc/PID/status. It fields are described in table 1-2.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700185
Stefani Seibold349888e2009-06-17 16:26:01 -0700186The statm file contains more detailed information about the process
187memory usage. Its seven fields are explained in Table 1-3. The stat file
188contains details information about the process itself. Its fields are
189explained in Table 1-4.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700190
Stefani Seibold349888e2009-06-17 16:26:01 -0700191Table 1-2: Contents of the statm files (as of 2.6.30-rc7)
192..............................................................................
193 Field Content
194 Name filename of the executable
195 State state (R is running, S is sleeping, D is sleeping
196 in an uninterruptible wait, Z is zombie,
197 T is traced or stopped)
198 Tgid thread group ID
199 Pid process id
200 PPid process id of the parent process
201 TracerPid PID of process tracing this process (0 if not)
202 Uid Real, effective, saved set, and file system UIDs
203 Gid Real, effective, saved set, and file system GIDs
204 FDSize number of file descriptor slots currently allocated
205 Groups supplementary group list
206 VmPeak peak virtual memory size
207 VmSize total program size
208 VmLck locked memory size
209 VmHWM peak resident set size ("high water mark")
210 VmRSS size of memory portions
211 VmData size of data, stack, and text segments
212 VmStk size of data, stack, and text segments
213 VmExe size of text segment
214 VmLib size of shared library code
215 VmPTE size of page table entries
216 Threads number of threads
217 SigQ number of signals queued/max. number for queue
218 SigPnd bitmap of pending signals for the thread
219 ShdPnd bitmap of shared pending signals for the process
220 SigBlk bitmap of blocked signals
221 SigIgn bitmap of ignored signals
222 SigCgt bitmap of catched signals
223 CapInh bitmap of inheritable capabilities
224 CapPrm bitmap of permitted capabilities
225 CapEff bitmap of effective capabilities
226 CapBnd bitmap of capabilities bounding set
227 Cpus_allowed mask of CPUs on which this process may run
228 Cpus_allowed_list Same as previous, but in "list format"
229 Mems_allowed mask of memory nodes allowed to this process
230 Mems_allowed_list Same as previous, but in "list format"
231 voluntary_ctxt_switches number of voluntary context switches
232 nonvoluntary_ctxt_switches number of non voluntary context switches
233..............................................................................
234
235Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700236..............................................................................
237 Field Content
238 size total program size (pages) (same as VmSize in status)
239 resident size of memory portions (pages) (same as VmRSS in status)
240 shared number of pages that are shared (i.e. backed by a file)
241 trs number of pages that are 'code' (not including libs; broken,
242 includes data segment)
243 lrs number of pages of library (always 0 on 2.6)
244 drs number of pages of data/stack (including libs; broken,
245 includes library text)
246 dt number of dirty pages (always 0 on 2.6)
247..............................................................................
248
Kees Cook18d96772007-07-15 23:40:38 -0700249
Stefani Seibold349888e2009-06-17 16:26:01 -0700250Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
Kees Cook18d96772007-07-15 23:40:38 -0700251..............................................................................
252 Field Content
253 pid process id
254 tcomm filename of the executable
255 state state (R is running, S is sleeping, D is sleeping in an
256 uninterruptible wait, Z is zombie, T is traced or stopped)
257 ppid process id of the parent process
258 pgrp pgrp of the process
259 sid session id
260 tty_nr tty the process uses
261 tty_pgrp pgrp of the tty
262 flags task flags
263 min_flt number of minor faults
264 cmin_flt number of minor faults with child's
265 maj_flt number of major faults
266 cmaj_flt number of major faults with child's
267 utime user mode jiffies
268 stime kernel mode jiffies
269 cutime user mode jiffies with child's
270 cstime kernel mode jiffies with child's
271 priority priority level
272 nice nice level
273 num_threads number of threads
Leonardo Chiquitto2e01e002008-02-03 16:17:16 +0200274 it_real_value (obsolete, always 0)
Kees Cook18d96772007-07-15 23:40:38 -0700275 start_time time the process started after system boot
276 vsize virtual memory size
277 rss resident set memory size
278 rsslim current limit in bytes on the rss
279 start_code address above which program text can run
280 end_code address below which program text can run
281 start_stack address of the start of the stack
282 esp current value of ESP
283 eip current value of EIP
Stefani Seibold349888e2009-06-17 16:26:01 -0700284 pending bitmap of pending signals
285 blocked bitmap of blocked signals
286 sigign bitmap of ignored signals
287 sigcatch bitmap of catched signals
Kees Cook18d96772007-07-15 23:40:38 -0700288 wchan address where process went to sleep
289 0 (place holder)
290 0 (place holder)
291 exit_signal signal to send to parent thread on exit
292 task_cpu which CPU the task is scheduled on
293 rt_priority realtime priority
294 policy scheduling policy (man sched_setscheduler)
295 blkio_ticks time spent waiting for block IO
Stefani Seibold349888e2009-06-17 16:26:01 -0700296 gtime guest time of the task in jiffies
297 cgtime guest time of the task children in jiffies
Kees Cook18d96772007-07-15 23:40:38 -0700298..............................................................................
299
Stefani Seibold349888e2009-06-17 16:26:01 -0700300The /proc/PID/map file containing the currently mapped memory regions and
301their access permissions.
302
303The format is:
304
305address perms offset dev inode pathname
306
30708048000-08049000 r-xp 00000000 03:00 8312 /opt/test
30808049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
3090804a000-0806b000 rw-p 00000000 00:00 0 [heap]
310a7cb1000-a7cb2000 ---p 00000000 00:00 0
Stefani Seiboldd899bf72009-09-22 16:45:40 -0700311a7cb2000-a7eb2000 rw-p 00000000 00:00 0 [threadstack:001ff4b4]
Stefani Seibold349888e2009-06-17 16:26:01 -0700312a7eb2000-a7eb3000 ---p 00000000 00:00 0
313a7eb3000-a7ed5000 rw-p 00000000 00:00 0
314a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
315a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
316a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
317a800b000-a800e000 rw-p 00000000 00:00 0
318a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
319a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
320a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
321a8024000-a8027000 rw-p 00000000 00:00 0
322a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
323a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
324a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
325aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
326ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
327
328where "address" is the address space in the process that it occupies, "perms"
329is a set of permissions:
330
331 r = read
332 w = write
333 x = execute
334 s = shared
335 p = private (copy on write)
336
337"offset" is the offset into the mapping, "dev" is the device (major:minor), and
338"inode" is the inode on that device. 0 indicates that no inode is associated
339with the memory region, as the case would be with BSS (uninitialized data).
340The "pathname" shows the name associated file for this mapping. If the mapping
341is not associated with a file:
342
343 [heap] = the heap of the program
344 [stack] = the stack of the main process
345 [vdso] = the "virtual dynamic shared object",
346 the kernel system call handler
Stefani Seiboldd899bf72009-09-22 16:45:40 -0700347 [threadstack:xxxxxxxx] = the stack of the thread, xxxxxxxx is the stack size
Stefani Seibold349888e2009-06-17 16:26:01 -0700348
349 or if empty, the mapping is anonymous.
350
351
352The /proc/PID/smaps is an extension based on maps, showing the memory
353consumption for each of the process's mappings. For each of mappings there
354is a series of lines such as the following:
355
35608048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
357Size: 1084 kB
358Rss: 892 kB
359Pss: 374 kB
360Shared_Clean: 892 kB
361Shared_Dirty: 0 kB
362Private_Clean: 0 kB
363Private_Dirty: 0 kB
364Referenced: 892 kB
365Swap: 0 kB
366KernelPageSize: 4 kB
367MMUPageSize: 4 kB
368
369The first of these lines shows the same information as is displayed for the
370mapping in /proc/PID/maps. The remaining lines show the size of the mapping,
371the amount of the mapping that is currently resident in RAM, the "proportional
372set size” (divide each shared page by the number of processes sharing it), the
373number of clean and dirty shared pages in the mapping, and the number of clean
374and dirty private pages in the mapping. The "Referenced" indicates the amount
375of memory currently marked as referenced or accessed.
376
377This file is only present if the CONFIG_MMU kernel configuration option is
378enabled.
Kees Cook18d96772007-07-15 23:40:38 -0700379
Moussa A. Ba398499d2009-09-21 17:02:29 -0700380The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
381bits on both physical and virtual pages associated with a process.
382To clear the bits for all the pages associated with the process
383 > echo 1 > /proc/PID/clear_refs
384
385To clear the bits for the anonymous pages associated with the process
386 > echo 2 > /proc/PID/clear_refs
387
388To clear the bits for the file mapped pages associated with the process
389 > echo 3 > /proc/PID/clear_refs
390Any other value written to /proc/PID/clear_refs will have no effect.
391
392
Linus Torvalds1da177e2005-04-16 15:20:36 -07003931.2 Kernel data
394---------------
395
396Similar to the process entries, the kernel data files give information about
397the running kernel. The files used to obtain this information are contained in
Stefani Seibold349888e2009-06-17 16:26:01 -0700398/proc and are listed in Table 1-5. Not all of these will be present in your
Linus Torvalds1da177e2005-04-16 15:20:36 -0700399system. It depends on the kernel configuration and the loaded modules, which
400files are there, and which are missing.
401
Stefani Seibold349888e2009-06-17 16:26:01 -0700402Table 1-5: Kernel info in /proc
Linus Torvalds1da177e2005-04-16 15:20:36 -0700403..............................................................................
404 File Content
405 apm Advanced power management info
406 buddyinfo Kernel memory allocator information (see text) (2.5)
407 bus Directory containing bus specific information
408 cmdline Kernel command line
409 cpuinfo Info about the CPU
410 devices Available devices (block and character)
411 dma Used DMS channels
412 filesystems Supported filesystems
413 driver Various drivers grouped here, currently rtc (2.4)
414 execdomains Execdomains, related to security (2.4)
415 fb Frame Buffer devices (2.4)
416 fs File system parameters, currently nfs/exports (2.4)
417 ide Directory containing info about the IDE subsystem
418 interrupts Interrupt usage
419 iomem Memory map (2.4)
420 ioports I/O port usage
421 irq Masks for irq to cpu affinity (2.4)(smp?)
422 isapnp ISA PnP (Plug&Play) Info (2.4)
423 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
424 kmsg Kernel messages
425 ksyms Kernel symbol table
426 loadavg Load average of last 1, 5 & 15 minutes
427 locks Kernel locks
428 meminfo Memory info
429 misc Miscellaneous
430 modules List of loaded modules
431 mounts Mounted filesystems
432 net Networking info (see text)
433 partitions Table of partitions known to the system
Randy Dunlap8b607562007-05-09 07:19:14 +0200434 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
Linus Torvalds1da177e2005-04-16 15:20:36 -0700435 decoupled by lspci (2.4)
436 rtc Real time clock
437 scsi SCSI info (see text)
438 slabinfo Slab pool info
Keika Kobayashid3d64df2009-06-17 16:25:55 -0700439 softirqs softirq usage
Linus Torvalds1da177e2005-04-16 15:20:36 -0700440 stat Overall statistics
441 swaps Swap space utilization
442 sys See chapter 2
443 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
444 tty Info of tty drivers
445 uptime System uptime
446 version Kernel version
447 video bttv info of video resources (2.4)
Eric Dumazeta47a1262008-07-23 21:27:38 -0700448 vmallocinfo Show vmalloced areas
Linus Torvalds1da177e2005-04-16 15:20:36 -0700449..............................................................................
450
451You can, for example, check which interrupts are currently in use and what
452they are used for by looking in the file /proc/interrupts:
453
454 > cat /proc/interrupts
455 CPU0
456 0: 8728810 XT-PIC timer
457 1: 895 XT-PIC keyboard
458 2: 0 XT-PIC cascade
459 3: 531695 XT-PIC aha152x
460 4: 2014133 XT-PIC serial
461 5: 44401 XT-PIC pcnet_cs
462 8: 2 XT-PIC rtc
463 11: 8 XT-PIC i82365
464 12: 182918 XT-PIC PS/2 Mouse
465 13: 1 XT-PIC fpu
466 14: 1232265 XT-PIC ide0
467 15: 7 XT-PIC ide1
468 NMI: 0
469
470In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
471output of a SMP machine):
472
473 > cat /proc/interrupts
474
475 CPU0 CPU1
476 0: 1243498 1214548 IO-APIC-edge timer
477 1: 8949 8958 IO-APIC-edge keyboard
478 2: 0 0 XT-PIC cascade
479 5: 11286 10161 IO-APIC-edge soundblaster
480 8: 1 0 IO-APIC-edge rtc
481 9: 27422 27407 IO-APIC-edge 3c503
482 12: 113645 113873 IO-APIC-edge PS/2 Mouse
483 13: 0 0 XT-PIC fpu
484 14: 22491 24012 IO-APIC-edge ide0
485 15: 2183 2415 IO-APIC-edge ide1
486 17: 30564 30414 IO-APIC-level eth0
487 18: 177 164 IO-APIC-level bttv
488 NMI: 2457961 2457959
489 LOC: 2457882 2457881
490 ERR: 2155
491
492NMI is incremented in this case because every timer interrupt generates a NMI
493(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
494
495LOC is the local interrupt counter of the internal APIC of every CPU.
496
497ERR is incremented in the case of errors in the IO-APIC bus (the bus that
498connects the CPUs in a SMP system. This means that an error has been detected,
499the IO-APIC automatically retry the transmission, so it should not be a big
500problem, but you should read the SMP-FAQ.
501
Joe Korty38e760a2007-10-17 18:04:40 +0200502In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
503/proc/interrupts to display every IRQ vector in use by the system, not
504just those considered 'most important'. The new vectors are:
505
506 THR -- interrupt raised when a machine check threshold counter
507 (typically counting ECC corrected errors of memory or cache) exceeds
508 a configurable threshold. Only available on some systems.
509
510 TRM -- a thermal event interrupt occurs when a temperature threshold
511 has been exceeded for the CPU. This interrupt may also be generated
512 when the temperature drops back to normal.
513
514 SPU -- a spurious interrupt is some interrupt that was raised then lowered
515 by some IO device before it could be fully processed by the APIC. Hence
516 the APIC sees the interrupt but does not know what device it came from.
517 For this case the APIC will generate the interrupt with a IRQ vector
518 of 0xff. This might also be generated by chipset bugs.
519
520 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
521 sent from one CPU to another per the needs of the OS. Typically,
522 their statistics are used by kernel developers and interested users to
Matt LaPlante19f59462009-04-27 15:06:31 +0200523 determine the occurrence of interrupts of the given type.
Joe Korty38e760a2007-10-17 18:04:40 +0200524
525The above IRQ vectors are displayed only when relevent. For example,
526the threshold vector does not exist on x86_64 platforms. Others are
527suppressed when the system is a uniprocessor. As of this writing, only
528i386 and x86_64 platforms support the new IRQ vector displays.
529
530Of some interest is the introduction of the /proc/irq directory to 2.4.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700531It could be used to set IRQ to CPU affinity, this means that you can "hook" an
532IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
Max Krasnyansky18404752008-05-29 11:02:52 -0700533irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
534prof_cpu_mask.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700535
536For example
537 > ls /proc/irq/
538 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
Max Krasnyansky18404752008-05-29 11:02:52 -0700539 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
Linus Torvalds1da177e2005-04-16 15:20:36 -0700540 > ls /proc/irq/0/
541 smp_affinity
542
Max Krasnyansky18404752008-05-29 11:02:52 -0700543smp_affinity is a bitmask, in which you can specify which CPUs can handle the
544IRQ, you can set it by doing:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700545
Max Krasnyansky18404752008-05-29 11:02:52 -0700546 > echo 1 > /proc/irq/10/smp_affinity
547
548This means that only the first CPU will handle the IRQ, but you can also echo
5495 which means that only the first and fourth CPU can handle the IRQ.
550
551The contents of each smp_affinity file is the same by default:
552
553 > cat /proc/irq/0/smp_affinity
Linus Torvalds1da177e2005-04-16 15:20:36 -0700554 ffffffff
555
Max Krasnyansky18404752008-05-29 11:02:52 -0700556The default_smp_affinity mask applies to all non-active IRQs, which are the
557IRQs which have not yet been allocated/activated, and hence which lack a
558/proc/irq/[0-9]* directory.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700559
Max Krasnyansky18404752008-05-29 11:02:52 -0700560prof_cpu_mask specifies which CPUs are to be profiled by the system wide
561profiler. Default value is ffffffff (all cpus).
Linus Torvalds1da177e2005-04-16 15:20:36 -0700562
563The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
564between all the CPUs which are allowed to handle it. As usual the kernel has
565more info than you and does a better job than you, so the defaults are the
566best choice for almost everyone.
567
568There are three more important subdirectories in /proc: net, scsi, and sys.
569The general rule is that the contents, or even the existence of these
570directories, depend on your kernel configuration. If SCSI is not enabled, the
571directory scsi may not exist. The same is true with the net, which is there
572only when networking support is present in the running kernel.
573
574The slabinfo file gives information about memory usage at the slab level.
575Linux uses slab pools for memory management above page level in version 2.2.
576Commonly used objects have their own slab pool (such as network buffers,
577directory cache, and so on).
578
579..............................................................................
580
581> cat /proc/buddyinfo
582
583Node 0, zone DMA 0 4 5 4 4 3 ...
584Node 0, zone Normal 1 0 0 1 101 8 ...
585Node 0, zone HighMem 2 0 0 1 1 0 ...
586
587Memory fragmentation is a problem under some workloads, and buddyinfo is a
588useful tool for helping diagnose these problems. Buddyinfo will give you a
589clue as to how big an area you can safely allocate, or why a previous
590allocation failed.
591
592Each column represents the number of pages of a certain order which are
593available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
594ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
595available in ZONE_NORMAL, etc...
596
597..............................................................................
598
599meminfo:
600
601Provides information about distribution and utilization of memory. This
602varies by architecture and compile options. The following is from a
60316GB PIII, which has highmem enabled. You may not have all of these fields.
604
605> cat /proc/meminfo
606
607
608MemTotal: 16344972 kB
609MemFree: 13634064 kB
610Buffers: 3656 kB
611Cached: 1195708 kB
612SwapCached: 0 kB
613Active: 891636 kB
614Inactive: 1077224 kB
615HighTotal: 15597528 kB
616HighFree: 13629632 kB
617LowTotal: 747444 kB
618LowFree: 4432 kB
619SwapTotal: 0 kB
620SwapFree: 0 kB
621Dirty: 968 kB
622Writeback: 0 kB
Miklos Szeredib88473f2008-04-30 00:54:39 -0700623AnonPages: 861800 kB
Linus Torvalds1da177e2005-04-16 15:20:36 -0700624Mapped: 280372 kB
Miklos Szeredib88473f2008-04-30 00:54:39 -0700625Slab: 284364 kB
626SReclaimable: 159856 kB
627SUnreclaim: 124508 kB
628PageTables: 24448 kB
629NFS_Unstable: 0 kB
630Bounce: 0 kB
631WritebackTmp: 0 kB
Linus Torvalds1da177e2005-04-16 15:20:36 -0700632CommitLimit: 7669796 kB
633Committed_AS: 100056 kB
Linus Torvalds1da177e2005-04-16 15:20:36 -0700634VmallocTotal: 112216 kB
635VmallocUsed: 428 kB
636VmallocChunk: 111088 kB
637
638 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
639 bits and the kernel binary code)
640 MemFree: The sum of LowFree+HighFree
641 Buffers: Relatively temporary storage for raw disk blocks
642 shouldn't get tremendously large (20MB or so)
643 Cached: in-memory cache for files read from the disk (the
644 pagecache). Doesn't include SwapCached
645 SwapCached: Memory that once was swapped out, is swapped back in but
646 still also is in the swapfile (if memory is needed it
647 doesn't need to be swapped out AGAIN because it is already
648 in the swapfile. This saves I/O)
649 Active: Memory that has been used more recently and usually not
650 reclaimed unless absolutely necessary.
651 Inactive: Memory which has been less recently used. It is more
652 eligible to be reclaimed for other purposes
653 HighTotal:
654 HighFree: Highmem is all memory above ~860MB of physical memory
655 Highmem areas are for use by userspace programs, or
656 for the pagecache. The kernel must use tricks to access
657 this memory, making it slower to access than lowmem.
658 LowTotal:
659 LowFree: Lowmem is memory which can be used for everything that
Matt LaPlante3f6dee92006-10-03 22:45:33 +0200660 highmem can be used for, but it is also available for the
Linus Torvalds1da177e2005-04-16 15:20:36 -0700661 kernel's use for its own data structures. Among many
662 other things, it is where everything from the Slab is
663 allocated. Bad things happen when you're out of lowmem.
664 SwapTotal: total amount of swap space available
665 SwapFree: Memory which has been evicted from RAM, and is temporarily
666 on the disk
667 Dirty: Memory which is waiting to get written back to the disk
668 Writeback: Memory which is actively being written back to the disk
Miklos Szeredib88473f2008-04-30 00:54:39 -0700669 AnonPages: Non-file backed pages mapped into userspace page tables
Linus Torvalds1da177e2005-04-16 15:20:36 -0700670 Mapped: files which have been mmaped, such as libraries
Adrian Bunke82443c2006-01-10 00:20:30 +0100671 Slab: in-kernel data structures cache
Miklos Szeredib88473f2008-04-30 00:54:39 -0700672SReclaimable: Part of Slab, that might be reclaimed, such as caches
673 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
674 PageTables: amount of memory dedicated to the lowest level of page
675 tables.
676NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
677 storage
678 Bounce: Memory used for block device "bounce buffers"
679WritebackTmp: Memory used by FUSE for temporary writeback buffers
Linus Torvalds1da177e2005-04-16 15:20:36 -0700680 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
681 this is the total amount of memory currently available to
682 be allocated on the system. This limit is only adhered to
683 if strict overcommit accounting is enabled (mode 2 in
684 'vm.overcommit_memory').
685 The CommitLimit is calculated with the following formula:
686 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
687 For example, on a system with 1G of physical RAM and 7G
688 of swap with a `vm.overcommit_ratio` of 30 it would
689 yield a CommitLimit of 7.3G.
690 For more details, see the memory overcommit documentation
691 in vm/overcommit-accounting.
692Committed_AS: The amount of memory presently allocated on the system.
693 The committed memory is a sum of all of the memory which
694 has been allocated by processes, even if it has not been
695 "used" by them as of yet. A process which malloc()'s 1G
696 of memory, but only touches 300M of it will only show up
697 as using 300M of memory even if it has the address space
698 allocated for the entire 1G. This 1G is memory which has
699 been "committed" to by the VM and can be used at any time
700 by the allocating application. With strict overcommit
701 enabled on the system (mode 2 in 'vm.overcommit_memory'),
702 allocations which would exceed the CommitLimit (detailed
703 above) will not be permitted. This is useful if one needs
704 to guarantee that processes will not fail due to lack of
705 memory once that memory has been successfully allocated.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700706VmallocTotal: total size of vmalloc memory area
707 VmallocUsed: amount of vmalloc area which is used
Matt LaPlante19f59462009-04-27 15:06:31 +0200708VmallocChunk: largest contiguous block of vmalloc area which is free
Linus Torvalds1da177e2005-04-16 15:20:36 -0700709
Eric Dumazeta47a1262008-07-23 21:27:38 -0700710..............................................................................
711
712vmallocinfo:
713
714Provides information about vmalloced/vmaped areas. One line per area,
715containing the virtual address range of the area, size in bytes,
716caller information of the creator, and optional information depending
717on the kind of area :
718
719 pages=nr number of pages
720 phys=addr if a physical address was specified
721 ioremap I/O mapping (ioremap() and friends)
722 vmalloc vmalloc() area
723 vmap vmap()ed pages
724 user VM_USERMAP area
725 vpages buffer for pages pointers was vmalloced (huge area)
726 N<node>=nr (Only on NUMA kernels)
727 Number of pages allocated on memory node <node>
728
729> cat /proc/vmallocinfo
7300xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
731 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
7320xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
733 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
7340xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
735 phys=7fee8000 ioremap
7360xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
737 phys=7fee7000 ioremap
7380xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
7390xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
740 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
7410xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
742 pages=2 vmalloc N1=2
7430xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
744 /0x130 [x_tables] pages=4 vmalloc N0=4
7450xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
746 pages=14 vmalloc N2=14
7470xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
748 pages=4 vmalloc N1=4
7490xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
750 pages=2 vmalloc N1=2
7510xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
752 pages=10 vmalloc N0=10
Linus Torvalds1da177e2005-04-16 15:20:36 -0700753
Keika Kobayashid3d64df2009-06-17 16:25:55 -0700754..............................................................................
755
756softirqs:
757
758Provides counts of softirq handlers serviced since boot time, for each cpu.
759
760> cat /proc/softirqs
761 CPU0 CPU1 CPU2 CPU3
762 HI: 0 0 0 0
763 TIMER: 27166 27120 27097 27034
764 NET_TX: 0 0 0 17
765 NET_RX: 42 0 0 39
766 BLOCK: 0 0 107 1121
767 TASKLET: 0 0 0 290
768 SCHED: 27035 26983 26971 26746
769 HRTIMER: 0 0 0 0
770 RCU: 1678 1769 2178 2250
771
772
Linus Torvalds1da177e2005-04-16 15:20:36 -07007731.3 IDE devices in /proc/ide
774----------------------------
775
776The subdirectory /proc/ide contains information about all IDE devices of which
777the kernel is aware. There is one subdirectory for each IDE controller, the
778file drivers and a link for each IDE device, pointing to the device directory
779in the controller specific subtree.
780
781The file drivers contains general information about the drivers used for the
782IDE devices:
783
784 > cat /proc/ide/drivers
785 ide-cdrom version 4.53
786 ide-disk version 1.08
787
788More detailed information can be found in the controller specific
789subdirectories. These are named ide0, ide1 and so on. Each of these
Stefani Seibold349888e2009-06-17 16:26:01 -0700790directories contains the files shown in table 1-6.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700791
792
Stefani Seibold349888e2009-06-17 16:26:01 -0700793Table 1-6: IDE controller info in /proc/ide/ide?
Linus Torvalds1da177e2005-04-16 15:20:36 -0700794..............................................................................
795 File Content
796 channel IDE channel (0 or 1)
797 config Configuration (only for PCI/IDE bridge)
798 mate Mate name
799 model Type/Chipset of IDE controller
800..............................................................................
801
802Each device connected to a controller has a separate subdirectory in the
Stefani Seibold349888e2009-06-17 16:26:01 -0700803controllers directory. The files listed in table 1-7 are contained in these
Linus Torvalds1da177e2005-04-16 15:20:36 -0700804directories.
805
806
Stefani Seibold349888e2009-06-17 16:26:01 -0700807Table 1-7: IDE device information
Linus Torvalds1da177e2005-04-16 15:20:36 -0700808..............................................................................
809 File Content
810 cache The cache
811 capacity Capacity of the medium (in 512Byte blocks)
812 driver driver and version
813 geometry physical and logical geometry
814 identify device identify block
815 media media type
816 model device identifier
817 settings device setup
818 smart_thresholds IDE disk management thresholds
819 smart_values IDE disk management values
820..............................................................................
821
822The most interesting file is settings. This file contains a nice overview of
823the drive parameters:
824
825 # cat /proc/ide/ide0/hda/settings
826 name value min max mode
827 ---- ----- --- --- ----
828 bios_cyl 526 0 65535 rw
829 bios_head 255 0 255 rw
830 bios_sect 63 0 63 rw
831 breada_readahead 4 0 127 rw
832 bswap 0 0 1 r
833 file_readahead 72 0 2097151 rw
834 io_32bit 0 0 3 rw
835 keepsettings 0 0 1 rw
836 max_kb_per_request 122 1 127 rw
837 multcount 0 0 8 rw
838 nice1 1 0 1 rw
839 nowerr 0 0 1 rw
840 pio_mode write-only 0 255 w
841 slow 0 0 1 rw
842 unmaskirq 0 0 1 rw
843 using_dma 0 0 1 rw
844
845
8461.4 Networking info in /proc/net
847--------------------------------
848
Stefani Seibold349888e2009-06-17 16:26:01 -0700849The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
Linus Torvalds1da177e2005-04-16 15:20:36 -0700850additional values you get for IP version 6 if you configure the kernel to
Stefani Seibold349888e2009-06-17 16:26:01 -0700851support this. Table 1-9 lists the files and their meaning.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700852
853
Stefani Seibold349888e2009-06-17 16:26:01 -0700854Table 1-8: IPv6 info in /proc/net
Linus Torvalds1da177e2005-04-16 15:20:36 -0700855..............................................................................
856 File Content
857 udp6 UDP sockets (IPv6)
858 tcp6 TCP sockets (IPv6)
859 raw6 Raw device statistics (IPv6)
860 igmp6 IP multicast addresses, which this host joined (IPv6)
861 if_inet6 List of IPv6 interface addresses
862 ipv6_route Kernel routing table for IPv6
863 rt6_stats Global IPv6 routing tables statistics
864 sockstat6 Socket statistics (IPv6)
865 snmp6 Snmp data (IPv6)
866..............................................................................
867
868
Stefani Seibold349888e2009-06-17 16:26:01 -0700869Table 1-9: Network info in /proc/net
Linus Torvalds1da177e2005-04-16 15:20:36 -0700870..............................................................................
871 File Content
872 arp Kernel ARP table
873 dev network devices with statistics
874 dev_mcast the Layer2 multicast groups a device is listening too
875 (interface index, label, number of references, number of bound
876 addresses).
877 dev_stat network device status
878 ip_fwchains Firewall chain linkage
879 ip_fwnames Firewall chain names
880 ip_masq Directory containing the masquerading tables
881 ip_masquerade Major masquerading table
882 netstat Network statistics
883 raw raw device statistics
884 route Kernel routing table
885 rpc Directory containing rpc info
886 rt_cache Routing cache
887 snmp SNMP data
888 sockstat Socket statistics
889 tcp TCP sockets
890 tr_rif Token ring RIF routing table
891 udp UDP sockets
892 unix UNIX domain sockets
893 wireless Wireless interface data (Wavelan etc)
894 igmp IP multicast addresses, which this host joined
895 psched Global packet scheduler parameters.
896 netlink List of PF_NETLINK sockets
897 ip_mr_vifs List of multicast virtual interfaces
898 ip_mr_cache List of multicast routing cache
899..............................................................................
900
901You can use this information to see which network devices are available in
902your system and how much traffic was routed over those devices:
903
904 > cat /proc/net/dev
905 Inter-|Receive |[...
906 face |bytes packets errs drop fifo frame compressed multicast|[...
907 lo: 908188 5596 0 0 0 0 0 0 [...
908 ppp0:15475140 20721 410 0 0 410 0 0 [...
909 eth0: 614530 7085 0 0 0 0 0 1 [...
910
911 ...] Transmit
912 ...] bytes packets errs drop fifo colls carrier compressed
913 ...] 908188 5596 0 0 0 0 0 0
914 ...] 1375103 17405 0 0 0 0 0 0
915 ...] 1703981 5535 0 0 0 3 0 0
916
917In addition, each Channel Bond interface has it's own directory. For
918example, the bond0 device will have a directory called /proc/net/bond0/.
919It will contain information that is specific to that bond, such as the
920current slaves of the bond, the link status of the slaves, and how
921many times the slaves link has failed.
922
9231.5 SCSI info
924-------------
925
926If you have a SCSI host adapter in your system, you'll find a subdirectory
927named after the driver for this adapter in /proc/scsi. You'll also see a list
928of all recognized SCSI devices in /proc/scsi:
929
930 >cat /proc/scsi/scsi
931 Attached devices:
932 Host: scsi0 Channel: 00 Id: 00 Lun: 00
933 Vendor: IBM Model: DGHS09U Rev: 03E0
934 Type: Direct-Access ANSI SCSI revision: 03
935 Host: scsi0 Channel: 00 Id: 06 Lun: 00
936 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
937 Type: CD-ROM ANSI SCSI revision: 02
938
939
940The directory named after the driver has one file for each adapter found in
941the system. These files contain information about the controller, including
942the used IRQ and the IO address range. The amount of information shown is
943dependent on the adapter you use. The example shows the output for an Adaptec
944AHA-2940 SCSI adapter:
945
946 > cat /proc/scsi/aic7xxx/0
947
948 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
949 Compile Options:
950 TCQ Enabled By Default : Disabled
951 AIC7XXX_PROC_STATS : Disabled
952 AIC7XXX_RESET_DELAY : 5
953 Adapter Configuration:
954 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
955 Ultra Wide Controller
956 PCI MMAPed I/O Base: 0xeb001000
957 Adapter SEEPROM Config: SEEPROM found and used.
958 Adaptec SCSI BIOS: Enabled
959 IRQ: 10
960 SCBs: Active 0, Max Active 2,
961 Allocated 15, HW 16, Page 255
962 Interrupts: 160328
963 BIOS Control Word: 0x18b6
964 Adapter Control Word: 0x005b
965 Extended Translation: Enabled
966 Disconnect Enable Flags: 0xffff
967 Ultra Enable Flags: 0x0001
968 Tag Queue Enable Flags: 0x0000
969 Ordered Queue Tag Flags: 0x0000
970 Default Tag Queue Depth: 8
971 Tagged Queue By Device array for aic7xxx host instance 0:
972 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
973 Actual queue depth per device for aic7xxx host instance 0:
974 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
975 Statistics:
976 (scsi0:0:0:0)
977 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
978 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
979 Total transfers 160151 (74577 reads and 85574 writes)
980 (scsi0:0:6:0)
981 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
982 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
983 Total transfers 0 (0 reads and 0 writes)
984
985
9861.6 Parallel port info in /proc/parport
987---------------------------------------
988
989The directory /proc/parport contains information about the parallel ports of
990your system. It has one subdirectory for each port, named after the port
991number (0,1,2,...).
992
Stefani Seibold349888e2009-06-17 16:26:01 -0700993These directories contain the four files shown in Table 1-10.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700994
995
Stefani Seibold349888e2009-06-17 16:26:01 -0700996Table 1-10: Files in /proc/parport
Linus Torvalds1da177e2005-04-16 15:20:36 -0700997..............................................................................
998 File Content
999 autoprobe Any IEEE-1284 device ID information that has been acquired.
1000 devices list of the device drivers using that port. A + will appear by the
1001 name of the device currently using the port (it might not appear
1002 against any).
1003 hardware Parallel port's base address, IRQ line and DMA channel.
1004 irq IRQ that parport is using for that port. This is in a separate
1005 file to allow you to alter it by writing a new value in (IRQ
1006 number or none).
1007..............................................................................
1008
10091.7 TTY info in /proc/tty
1010-------------------------
1011
1012Information about the available and actually used tty's can be found in the
1013directory /proc/tty.You'll find entries for drivers and line disciplines in
Stefani Seibold349888e2009-06-17 16:26:01 -07001014this directory, as shown in Table 1-11.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001015
1016
Stefani Seibold349888e2009-06-17 16:26:01 -07001017Table 1-11: Files in /proc/tty
Linus Torvalds1da177e2005-04-16 15:20:36 -07001018..............................................................................
1019 File Content
1020 drivers list of drivers and their usage
1021 ldiscs registered line disciplines
1022 driver/serial usage statistic and status of single tty lines
1023..............................................................................
1024
1025To see which tty's are currently in use, you can simply look into the file
1026/proc/tty/drivers:
1027
1028 > cat /proc/tty/drivers
1029 pty_slave /dev/pts 136 0-255 pty:slave
1030 pty_master /dev/ptm 128 0-255 pty:master
1031 pty_slave /dev/ttyp 3 0-255 pty:slave
1032 pty_master /dev/pty 2 0-255 pty:master
1033 serial /dev/cua 5 64-67 serial:callout
1034 serial /dev/ttyS 4 64-67 serial
1035 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1036 /dev/ptmx /dev/ptmx 5 2 system
1037 /dev/console /dev/console 5 1 system:console
1038 /dev/tty /dev/tty 5 0 system:/dev/tty
1039 unknown /dev/tty 4 1-63 console
1040
1041
10421.8 Miscellaneous kernel statistics in /proc/stat
1043-------------------------------------------------
1044
1045Various pieces of information about kernel activity are available in the
1046/proc/stat file. All of the numbers reported in this file are aggregates
1047since the system first booted. For a quick look, simply cat the file:
1048
1049 > cat /proc/stat
Eric Dumazetc5743582009-09-21 17:01:06 -07001050 cpu 2255 34 2290 22625563 6290 127 456 0 0
1051 cpu0 1132 34 1441 11311718 3675 127 438 0 0
1052 cpu1 1123 0 849 11313845 2614 0 18 0 0
Linus Torvalds1da177e2005-04-16 15:20:36 -07001053 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1054 ctxt 1990473
1055 btime 1062191376
1056 processes 2915
1057 procs_running 1
1058 procs_blocked 0
Keika Kobayashid3d64df2009-06-17 16:25:55 -07001059 softirq 183433 0 21755 12 39 1137 231 21459 2263
Linus Torvalds1da177e2005-04-16 15:20:36 -07001060
1061The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1062lines. These numbers identify the amount of time the CPU has spent performing
1063different kinds of work. Time units are in USER_HZ (typically hundredths of a
1064second). The meanings of the columns are as follows, from left to right:
1065
1066- user: normal processes executing in user mode
1067- nice: niced processes executing in user mode
1068- system: processes executing in kernel mode
1069- idle: twiddling thumbs
1070- iowait: waiting for I/O to complete
1071- irq: servicing interrupts
1072- softirq: servicing softirqs
Leonardo Chiquittob68f2c3a2007-10-20 03:03:38 +02001073- steal: involuntary wait
Ryota Ozakice0e7b22009-10-24 01:20:10 +09001074- guest: running a normal guest
1075- guest_nice: running a niced guest
Linus Torvalds1da177e2005-04-16 15:20:36 -07001076
1077The "intr" line gives counts of interrupts serviced since boot time, for each
1078of the possible system interrupts. The first column is the total of all
1079interrupts serviced; each subsequent column is the total for that particular
1080interrupt.
1081
1082The "ctxt" line gives the total number of context switches across all CPUs.
1083
1084The "btime" line gives the time at which the system booted, in seconds since
1085the Unix epoch.
1086
1087The "processes" line gives the number of processes and threads created, which
1088includes (but is not limited to) those created by calls to the fork() and
1089clone() system calls.
1090
Luis Garces-Ericee3cc2222009-12-06 18:30:44 -08001091The "procs_running" line gives the total number of threads that are
1092running or ready to run (i.e., the total number of runnable threads).
Linus Torvalds1da177e2005-04-16 15:20:36 -07001093
1094The "procs_blocked" line gives the number of processes currently blocked,
1095waiting for I/O to complete.
1096
Keika Kobayashid3d64df2009-06-17 16:25:55 -07001097The "softirq" line gives counts of softirqs serviced since boot time, for each
1098of the possible system softirqs. The first column is the total of all
1099softirqs serviced; each subsequent column is the total for that particular
1100softirq.
1101
Theodore Ts'o37515fa2008-10-09 23:21:54 -04001102
Alex Tomasc9de5602008-01-29 00:19:52 -050011031.9 Ext4 file system parameters
1104------------------------------
Alex Tomasc9de5602008-01-29 00:19:52 -05001105
Theodore Ts'o37515fa2008-10-09 23:21:54 -04001106Information about mounted ext4 file systems can be found in
1107/proc/fs/ext4. Each mounted filesystem will have a directory in
1108/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1109/proc/fs/ext4/dm-0). The files in each per-device directory are shown
Stefani Seibold349888e2009-06-17 16:26:01 -07001110in Table 1-12, below.
Alex Tomasc9de5602008-01-29 00:19:52 -05001111
Stefani Seibold349888e2009-06-17 16:26:01 -07001112Table 1-12: Files in /proc/fs/ext4/<devname>
Theodore Ts'o37515fa2008-10-09 23:21:54 -04001113..............................................................................
1114 File Content
1115 mb_groups details of multiblock allocator buddy cache of free blocks
Theodore Ts'o37515fa2008-10-09 23:21:54 -04001116..............................................................................
Alex Tomasc9de5602008-01-29 00:19:52 -05001117
Linus Torvalds1da177e2005-04-16 15:20:36 -07001118
1119------------------------------------------------------------------------------
1120Summary
1121------------------------------------------------------------------------------
1122The /proc file system serves information about the running system. It not only
1123allows access to process data but also allows you to request the kernel status
1124by reading files in the hierarchy.
1125
1126The directory structure of /proc reflects the types of information and makes
1127it easy, if not obvious, where to look for specific data.
1128------------------------------------------------------------------------------
1129
1130------------------------------------------------------------------------------
1131CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1132------------------------------------------------------------------------------
1133
1134------------------------------------------------------------------------------
1135In This Chapter
1136------------------------------------------------------------------------------
1137* Modifying kernel parameters by writing into files found in /proc/sys
1138* Exploring the files which modify certain parameters
1139* Review of the /proc/sys file tree
1140------------------------------------------------------------------------------
1141
1142
1143A very interesting part of /proc is the directory /proc/sys. This is not only
1144a source of information, it also allows you to change parameters within the
1145kernel. Be very careful when attempting this. You can optimize your system,
1146but you can also cause it to crash. Never alter kernel parameters on a
1147production system. Set up a development machine and test to make sure that
1148everything works the way you want it to. You may have no alternative but to
1149reboot the machine once an error has been made.
1150
1151To change a value, simply echo the new value into the file. An example is
1152given below in the section on the file system data. You need to be root to do
1153this. You can create your own boot script to perform this every time your
1154system boots.
1155
1156The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1157general things in the operation of the Linux kernel. Since some of the files
1158can inadvertently disrupt your system, it is advisable to read both
1159documentation and source before actually making adjustments. In any case, be
1160very careful when writing to any of these files. The entries in /proc may
1161change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1162review the kernel documentation in the directory /usr/src/linux/Documentation.
1163This chapter is heavily based on the documentation included in the pre 2.2
1164kernels, and became part of it in version 2.2.1 of the Linux kernel.
1165
Shen Feng760df932009-04-02 16:57:20 -07001166Please see: Documentation/sysctls/ directory for descriptions of these
Peter W Morrealedb0fb182009-01-15 13:50:42 -08001167entries.
Andrew Morton9d0243b2006-01-08 01:00:39 -08001168
Shen Feng760df932009-04-02 16:57:20 -07001169------------------------------------------------------------------------------
1170Summary
1171------------------------------------------------------------------------------
1172Certain aspects of kernel behavior can be modified at runtime, without the
1173need to recompile the kernel, or even to reboot the system. The files in the
1174/proc/sys tree can not only be read, but also modified. You can use the echo
1175command to write value into these files, thereby changing the default settings
1176of the kernel.
1177------------------------------------------------------------------------------
Andrew Morton9d0243b2006-01-08 01:00:39 -08001178
Shen Feng760df932009-04-02 16:57:20 -07001179------------------------------------------------------------------------------
1180CHAPTER 3: PER-PROCESS PARAMETERS
1181------------------------------------------------------------------------------
Linus Torvalds1da177e2005-04-16 15:20:36 -07001182
Shen Feng760df932009-04-02 16:57:20 -070011833.1 /proc/<pid>/oom_adj - Adjust the oom-killer score
Jan-Frode Myklebustd7ff0db2006-09-29 01:59:45 -07001184------------------------------------------------------
1185
KOSAKI Motohiro0753ba02009-08-18 14:11:10 -07001186This file can be used to adjust the score used to select which processes
1187should be killed in an out-of-memory situation. Giving it a high score will
1188increase the likelihood of this process being killed by the oom-killer. Valid
1189values are in the range -16 to +15, plus the special value -17, which disables
1190oom-killing altogether for this process.
Jan-Frode Myklebustd7ff0db2006-09-29 01:59:45 -07001191
Evgeniy Polyakov9e9e3cb2009-01-29 14:25:09 -08001192The process to be killed in an out-of-memory situation is selected among all others
1193based on its badness score. This value equals the original memory size of the process
1194and is then updated according to its CPU time (utime + stime) and the
1195run time (uptime - start time). The longer it runs the smaller is the score.
1196Badness score is divided by the square root of the CPU time and then by
1197the double square root of the run time.
1198
1199Swapped out tasks are killed first. Half of each child's memory size is added to
1200the parent's score if they do not share the same memory. Thus forking servers
1201are the prime candidates to be killed. Having only one 'hungry' child will make
1202parent less preferable than the child.
1203
1204/proc/<pid>/oom_score shows process' current badness score.
1205
1206The following heuristics are then applied:
1207 * if the task was reniced, its score doubles
1208 * superuser or direct hardware access tasks (CAP_SYS_ADMIN, CAP_SYS_RESOURCE
1209 or CAP_SYS_RAWIO) have their score divided by 4
KOSAKI Motohiro495789a2009-09-21 17:03:14 -07001210 * if oom condition happened in one cpuset and checked process does not belong
Evgeniy Polyakov9e9e3cb2009-01-29 14:25:09 -08001211 to it, its score is divided by 8
1212 * the resulting score is multiplied by two to the power of oom_adj, i.e.
1213 points <<= oom_adj when it is positive and
1214 points >>= -(oom_adj) otherwise
1215
1216The task with the highest badness score is then selected and its children
1217are killed, process itself will be killed in an OOM situation when it does
1218not have children or some of them disabled oom like described above.
1219
Shen Feng760df932009-04-02 16:57:20 -070012203.2 /proc/<pid>/oom_score - Display current oom-killer score
Jan-Frode Myklebustd7ff0db2006-09-29 01:59:45 -07001221-------------------------------------------------------------
1222
Jan-Frode Myklebustd7ff0db2006-09-29 01:59:45 -07001223This file can be used to check the current score used by the oom-killer is for
1224any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
1225process should be killed in an out-of-memory situation.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001226
Roland Kletzingf9c99462007-03-05 00:30:54 -08001227
Shen Feng760df932009-04-02 16:57:20 -070012283.3 /proc/<pid>/io - Display the IO accounting fields
Roland Kletzingf9c99462007-03-05 00:30:54 -08001229-------------------------------------------------------
1230
1231This file contains IO statistics for each running process
1232
1233Example
1234-------
1235
1236test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1237[1] 3828
1238
1239test:/tmp # cat /proc/3828/io
1240rchar: 323934931
1241wchar: 323929600
1242syscr: 632687
1243syscw: 632675
1244read_bytes: 0
1245write_bytes: 323932160
1246cancelled_write_bytes: 0
1247
1248
1249Description
1250-----------
1251
1252rchar
1253-----
1254
1255I/O counter: chars read
1256The number of bytes which this task has caused to be read from storage. This
1257is simply the sum of bytes which this process passed to read() and pread().
1258It includes things like tty IO and it is unaffected by whether or not actual
1259physical disk IO was required (the read might have been satisfied from
1260pagecache)
1261
1262
1263wchar
1264-----
1265
1266I/O counter: chars written
1267The number of bytes which this task has caused, or shall cause to be written
1268to disk. Similar caveats apply here as with rchar.
1269
1270
1271syscr
1272-----
1273
1274I/O counter: read syscalls
1275Attempt to count the number of read I/O operations, i.e. syscalls like read()
1276and pread().
1277
1278
1279syscw
1280-----
1281
1282I/O counter: write syscalls
1283Attempt to count the number of write I/O operations, i.e. syscalls like
1284write() and pwrite().
1285
1286
1287read_bytes
1288----------
1289
1290I/O counter: bytes read
1291Attempt to count the number of bytes which this process really did cause to
1292be fetched from the storage layer. Done at the submit_bio() level, so it is
1293accurate for block-backed filesystems. <please add status regarding NFS and
1294CIFS at a later time>
1295
1296
1297write_bytes
1298-----------
1299
1300I/O counter: bytes written
1301Attempt to count the number of bytes which this process caused to be sent to
1302the storage layer. This is done at page-dirtying time.
1303
1304
1305cancelled_write_bytes
1306---------------------
1307
1308The big inaccuracy here is truncate. If a process writes 1MB to a file and
1309then deletes the file, it will in fact perform no writeout. But it will have
1310been accounted as having caused 1MB of write.
1311In other words: The number of bytes which this process caused to not happen,
1312by truncating pagecache. A task can cause "negative" IO too. If this task
1313truncates some dirty pagecache, some IO which another task has been accounted
1314for (in it's write_bytes) will not be happening. We _could_ just subtract that
1315from the truncating task's write_bytes, but there is information loss in doing
1316that.
1317
1318
1319Note
1320----
1321
1322At its current implementation state, this is a bit racy on 32-bit machines: if
1323process A reads process B's /proc/pid/io while process B is updating one of
1324those 64-bit counters, process A could see an intermediate result.
1325
1326
1327More information about this can be found within the taskstats documentation in
1328Documentation/accounting.
1329
Shen Feng760df932009-04-02 16:57:20 -070013303.4 /proc/<pid>/coredump_filter - Core dump filtering settings
Kawai, Hidehirobb901102007-07-19 01:48:31 -07001331---------------------------------------------------------------
1332When a process is dumped, all anonymous memory is written to a core file as
1333long as the size of the core file isn't limited. But sometimes we don't want
1334to dump some memory segments, for example, huge shared memory. Conversely,
1335sometimes we want to save file-backed memory segments into a core file, not
1336only the individual files.
1337
1338/proc/<pid>/coredump_filter allows you to customize which memory segments
1339will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1340of memory types. If a bit of the bitmask is set, memory segments of the
1341corresponding memory type are dumped, otherwise they are not dumped.
1342
KOSAKI Motohiroe575f112008-10-18 20:27:08 -07001343The following 7 memory types are supported:
Kawai, Hidehirobb901102007-07-19 01:48:31 -07001344 - (bit 0) anonymous private memory
1345 - (bit 1) anonymous shared memory
1346 - (bit 2) file-backed private memory
1347 - (bit 3) file-backed shared memory
Hidehiro Kawaib261dfe2008-09-13 02:33:10 -07001348 - (bit 4) ELF header pages in file-backed private memory areas (it is
1349 effective only if the bit 2 is cleared)
KOSAKI Motohiroe575f112008-10-18 20:27:08 -07001350 - (bit 5) hugetlb private memory
1351 - (bit 6) hugetlb shared memory
Kawai, Hidehirobb901102007-07-19 01:48:31 -07001352
1353 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1354 are always dumped regardless of the bitmask status.
1355
KOSAKI Motohiroe575f112008-10-18 20:27:08 -07001356 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
1357 effected by bit 5-6.
1358
1359Default value of coredump_filter is 0x23; this means all anonymous memory
1360segments and hugetlb private memory are dumped.
Kawai, Hidehirobb901102007-07-19 01:48:31 -07001361
1362If you don't want to dump all shared memory segments attached to pid 1234,
KOSAKI Motohiroe575f112008-10-18 20:27:08 -07001363write 0x21 to the process's proc file.
Kawai, Hidehirobb901102007-07-19 01:48:31 -07001364
KOSAKI Motohiroe575f112008-10-18 20:27:08 -07001365 $ echo 0x21 > /proc/1234/coredump_filter
Kawai, Hidehirobb901102007-07-19 01:48:31 -07001366
1367When a new process is created, the process inherits the bitmask status from its
1368parent. It is useful to set up coredump_filter before the program runs.
1369For example:
1370
1371 $ echo 0x7 > /proc/self/coredump_filter
1372 $ ./some_program
1373
Shen Feng760df932009-04-02 16:57:20 -070013743.5 /proc/<pid>/mountinfo - Information about mounts
Ram Pai2d4d4862008-03-27 13:06:25 +01001375--------------------------------------------------------
1376
1377This file contains lines of the form:
1378
137936 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1380(1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1381
1382(1) mount ID: unique identifier of the mount (may be reused after umount)
1383(2) parent ID: ID of parent (or of self for the top of the mount tree)
1384(3) major:minor: value of st_dev for files on filesystem
1385(4) root: root of the mount within the filesystem
1386(5) mount point: mount point relative to the process's root
1387(6) mount options: per mount options
1388(7) optional fields: zero or more fields of the form "tag[:value]"
1389(8) separator: marks the end of the optional fields
1390(9) filesystem type: name of filesystem of the form "type[.subtype]"
1391(10) mount source: filesystem specific information or "none"
1392(11) super options: per super block options
1393
1394Parsers should ignore all unrecognised optional fields. Currently the
1395possible optional fields are:
1396
1397shared:X mount is shared in peer group X
1398master:X mount is slave to peer group X
Miklos Szeredi97e7e0f2008-03-27 13:06:26 +01001399propagate_from:X mount is slave and receives propagation from peer group X (*)
Ram Pai2d4d4862008-03-27 13:06:25 +01001400unbindable mount is unbindable
1401
Miklos Szeredi97e7e0f2008-03-27 13:06:26 +01001402(*) X is the closest dominant peer group under the process's root. If
1403X is the immediate master of the mount, or if there's no dominant peer
1404group under the same root, then only the "master:X" field is present
1405and not the "propagate_from:X" field.
1406
Ram Pai2d4d4862008-03-27 13:06:25 +01001407For more information on mount propagation see:
1408
1409 Documentation/filesystems/sharedsubtree.txt
1410
john stultz4614a696b2009-12-14 18:00:05 -08001411
14123.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1413--------------------------------------------------------
1414These files provide a method to access a tasks comm value. It also allows for
1415a task to set its own or one of its thread siblings comm value. The comm value
1416is limited in size compared to the cmdline value, so writing anything longer
1417then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1418comm value.