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
50 ------------------------------------------------------------------------------
52 ------------------------------------------------------------------------------
54 0.1 Introduction/Credits
55 ------------------------
57 This documentation is part of a soon (or so we hope) to be released book on
58 the SuSE Linux distribution. As there is no complete documentation for the
59 /proc file system and we've used many freely available sources to write these
60 chapters, it seems only fair to give the work back to the Linux community.
61 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
62 afraid it's still far from complete, but we hope it will be useful. As far as
63 we know, it is the first 'all-in-one' document about the /proc file system. It
64 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
65 SPARC, AXP, etc., features, you probably won't find what you are looking for.
66 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
67 additions and patches are welcome and will be added to this document if you
70 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
71 other people for help compiling this documentation. We'd also like to extend a
72 special thank you to Andi Kleen for documentation, which we relied on heavily
73 to create this document, as well as the additional information he provided.
74 Thanks to everybody else who contributed source or docs to the Linux kernel
75 and helped create a great piece of software... :)
77 If you have any comments, corrections or additions, please don't hesitate to
78 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
81 The latest version of this document is available online at
82 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
84 If the above direction does not works for you, you could try the kernel
85 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
86 comandante@zaralinux.com.
91 We don't guarantee the correctness of this document, and if you come to us
92 complaining about how you screwed up your system because of incorrect
93 documentation, we won't feel responsible...
95 ------------------------------------------------------------------------------
96 CHAPTER 1: COLLECTING SYSTEM INFORMATION
97 ------------------------------------------------------------------------------
99 ------------------------------------------------------------------------------
101 ------------------------------------------------------------------------------
102 * Investigating the properties of the pseudo file system /proc and its
103 ability to provide information on the running Linux system
104 * Examining /proc's structure
105 * Uncovering various information about the kernel and the processes running
107 ------------------------------------------------------------------------------
110 The proc file system acts as an interface to internal data structures in the
111 kernel. It can be used to obtain information about the system and to change
112 certain kernel parameters at runtime (sysctl).
114 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
115 show you how you can use /proc/sys to change settings.
117 1.1 Process-Specific Subdirectories
118 -----------------------------------
120 The directory /proc contains (among other things) one subdirectory for each
121 process running on the system, which is named after the process ID (PID).
123 The link self points to the process reading the file system. Each process
124 subdirectory has the entries listed in Table 1-1.
127 Table 1-1: Process specific entries in /proc
128 ..............................................................................
130 clear_refs Clears page referenced bits shown in smaps output
131 cmdline Command line arguments
132 cpu Current and last cpu in which it was executed (2.4)(smp)
133 cwd Link to the current working directory
134 environ Values of environment variables
135 exe Link to the executable of this process
136 fd Directory, which contains all file descriptors
137 maps Memory maps to executables and library files (2.4)
138 mem Memory held by this process
139 root Link to the root directory of this process
141 statm Process memory status information
142 status Process status in human readable form
143 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
145 stack Report full stack trace, enable via CONFIG_STACKTRACE
146 smaps a extension based on maps, showing the memory consumption of
147 each mapping and flags associated with it
148 numa_maps an extension based on maps, showing the memory locality and
149 binding policy as well as mem usage (in pages) of each mapping.
150 ..............................................................................
152 For example, to get the status information of a process, all you have to do is
153 read the file /proc/PID/status:
155 >cat /proc/self/status
179 SigPnd: 0000000000000000
180 ShdPnd: 0000000000000000
181 SigBlk: 0000000000000000
182 SigIgn: 0000000000000000
183 SigCgt: 0000000000000000
184 CapInh: 00000000fffffeff
185 CapPrm: 0000000000000000
186 CapEff: 0000000000000000
187 CapBnd: ffffffffffffffff
189 voluntary_ctxt_switches: 0
190 nonvoluntary_ctxt_switches: 1
192 This shows you nearly the same information you would get if you viewed it with
193 the ps command. In fact, ps uses the proc file system to obtain its
194 information. But you get a more detailed view of the process by reading the
195 file /proc/PID/status. It fields are described in table 1-2.
197 The statm file contains more detailed information about the process
198 memory usage. Its seven fields are explained in Table 1-3. The stat file
199 contains details information about the process itself. Its fields are
200 explained in Table 1-4.
202 (for SMP CONFIG users)
203 For making accounting scalable, RSS related information are handled in
204 asynchronous manner and the vaule may not be very precise. To see a precise
205 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
206 It's slow but very precise.
208 Table 1-2: Contents of the status files (as of 2.6.30-rc7)
209 ..............................................................................
211 Name filename of the executable
212 State state (R is running, S is sleeping, D is sleeping
213 in an uninterruptible wait, Z is zombie,
214 T is traced or stopped)
217 PPid process id of the parent process
218 TracerPid PID of process tracing this process (0 if not)
219 Uid Real, effective, saved set, and file system UIDs
220 Gid Real, effective, saved set, and file system GIDs
221 FDSize number of file descriptor slots currently allocated
222 Groups supplementary group list
223 VmPeak peak virtual memory size
224 VmSize total program size
225 VmLck locked memory size
226 VmHWM peak resident set size ("high water mark")
227 VmRSS size of memory portions
228 VmData size of data, stack, and text segments
229 VmStk size of data, stack, and text segments
230 VmExe size of text segment
231 VmLib size of shared library code
232 VmPTE size of page table entries
233 VmSwap size of swap usage (the number of referred swapents)
234 Threads number of threads
235 SigQ number of signals queued/max. number for queue
236 SigPnd bitmap of pending signals for the thread
237 ShdPnd bitmap of shared pending signals for the process
238 SigBlk bitmap of blocked signals
239 SigIgn bitmap of ignored signals
240 SigCgt bitmap of caught signals
241 CapInh bitmap of inheritable capabilities
242 CapPrm bitmap of permitted capabilities
243 CapEff bitmap of effective capabilities
244 CapBnd bitmap of capabilities bounding set
245 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
246 Cpus_allowed mask of CPUs on which this process may run
247 Cpus_allowed_list Same as previous, but in "list format"
248 Mems_allowed mask of memory nodes allowed to this process
249 Mems_allowed_list Same as previous, but in "list format"
250 voluntary_ctxt_switches number of voluntary context switches
251 nonvoluntary_ctxt_switches number of non voluntary context switches
252 ..............................................................................
254 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
255 ..............................................................................
257 size total program size (pages) (same as VmSize in status)
258 resident size of memory portions (pages) (same as VmRSS in status)
259 shared number of pages that are shared (i.e. backed by a file)
260 trs number of pages that are 'code' (not including libs; broken,
261 includes data segment)
262 lrs number of pages of library (always 0 on 2.6)
263 drs number of pages of data/stack (including libs; broken,
264 includes library text)
265 dt number of dirty pages (always 0 on 2.6)
266 ..............................................................................
269 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
270 ..............................................................................
273 tcomm filename of the executable
274 state state (R is running, S is sleeping, D is sleeping in an
275 uninterruptible wait, Z is zombie, T is traced or stopped)
276 ppid process id of the parent process
277 pgrp pgrp of the process
279 tty_nr tty the process uses
280 tty_pgrp pgrp of the tty
282 min_flt number of minor faults
283 cmin_flt number of minor faults with child's
284 maj_flt number of major faults
285 cmaj_flt number of major faults with child's
286 utime user mode jiffies
287 stime kernel mode jiffies
288 cutime user mode jiffies with child's
289 cstime kernel mode jiffies with child's
290 priority priority level
292 num_threads number of threads
293 it_real_value (obsolete, always 0)
294 start_time time the process started after system boot
295 vsize virtual memory size
296 rss resident set memory size
297 rsslim current limit in bytes on the rss
298 start_code address above which program text can run
299 end_code address below which program text can run
300 start_stack address of the start of the main process stack
301 esp current value of ESP
302 eip current value of EIP
303 pending bitmap of pending signals
304 blocked bitmap of blocked signals
305 sigign bitmap of ignored signals
306 sigcatch bitmap of caught signals
307 wchan address where process went to sleep
310 exit_signal signal to send to parent thread on exit
311 task_cpu which CPU the task is scheduled on
312 rt_priority realtime priority
313 policy scheduling policy (man sched_setscheduler)
314 blkio_ticks time spent waiting for block IO
315 gtime guest time of the task in jiffies
316 cgtime guest time of the task children in jiffies
317 start_data address above which program data+bss is placed
318 end_data address below which program data+bss is placed
319 start_brk address above which program heap can be expanded with brk()
320 arg_start address above which program command line is placed
321 arg_end address below which program command line is placed
322 env_start address above which program environment is placed
323 env_end address below which program environment is placed
324 exit_code the thread's exit_code in the form reported by the waitpid system call
325 ..............................................................................
327 The /proc/PID/maps file containing the currently mapped memory regions and
328 their access permissions.
332 address perms offset dev inode pathname
334 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
335 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
336 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
337 a7cb1000-a7cb2000 ---p 00000000 00:00 0
338 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
339 a7eb2000-a7eb3000 ---p 00000000 00:00 0
340 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack:1001]
341 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
342 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
343 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
344 a800b000-a800e000 rw-p 00000000 00:00 0
345 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
346 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
347 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
348 a8024000-a8027000 rw-p 00000000 00:00 0
349 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
350 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
351 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
352 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
353 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
355 where "address" is the address space in the process that it occupies, "perms"
356 is a set of permissions:
362 p = private (copy on write)
364 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
365 "inode" is the inode on that device. 0 indicates that no inode is associated
366 with the memory region, as the case would be with BSS (uninitialized data).
367 The "pathname" shows the name associated file for this mapping. If the mapping
368 is not associated with a file:
370 [heap] = the heap of the program
371 [stack] = the stack of the main process
372 [stack:1001] = the stack of the thread with tid 1001
373 [vdso] = the "virtual dynamic shared object",
374 the kernel system call handler
376 or if empty, the mapping is anonymous.
378 The /proc/PID/task/TID/maps is a view of the virtual memory from the viewpoint
379 of the individual tasks of a process. In this file you will see a mapping marked
380 as [stack] if that task sees it as a stack. This is a key difference from the
381 content of /proc/PID/maps, where you will see all mappings that are being used
382 as stack by all of those tasks. Hence, for the example above, the task-level
383 map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this:
385 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
386 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
387 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
388 a7cb1000-a7cb2000 ---p 00000000 00:00 0
389 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
390 a7eb2000-a7eb3000 ---p 00000000 00:00 0
391 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack]
392 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
393 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
394 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
395 a800b000-a800e000 rw-p 00000000 00:00 0
396 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
397 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
398 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
399 a8024000-a8027000 rw-p 00000000 00:00 0
400 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
401 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
402 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
403 aff35000-aff4a000 rw-p 00000000 00:00 0
404 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
406 The /proc/PID/smaps is an extension based on maps, showing the memory
407 consumption for each of the process's mappings. For each of mappings there
408 is a series of lines such as the following:
410 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
424 VmFlags: rd ex mr mw me de
426 the first of these lines shows the same information as is displayed for the
427 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
428 (size), the amount of the mapping that is currently resident in RAM (RSS), the
429 process' proportional share of this mapping (PSS), the number of clean and
430 dirty private pages in the mapping. Note that even a page which is part of a
431 MAP_SHARED mapping, but has only a single pte mapped, i.e. is currently used
432 by only one process, is accounted as private and not as shared. "Referenced"
433 indicates the amount of memory currently marked as referenced or accessed.
434 "Anonymous" shows the amount of memory that does not belong to any file. Even
435 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
436 and a page is modified, the file page is replaced by a private anonymous copy.
437 "Swap" shows how much would-be-anonymous memory is also used, but out on
440 "VmFlags" field deserves a separate description. This member represents the kernel
441 flags associated with the particular virtual memory area in two letter encoded
442 manner. The codes are the following:
451 gd - stack segment growns down
453 dw - disabled write to the mapped file
454 lo - pages are locked in memory
455 io - memory mapped I/O area
456 sr - sequential read advise provided
457 rr - random read advise provided
458 dc - do not copy area on fork
459 de - do not expand area on remapping
460 ac - area is accountable
461 nr - swap space is not reserved for the area
462 ht - area uses huge tlb pages
463 nl - non-linear mapping
464 ar - architecture specific flag
465 dd - do not include area into core dump
468 hg - huge page advise flag
469 nh - no-huge page advise flag
470 mg - mergable advise flag
472 Note that there is no guarantee that every flag and associated mnemonic will
473 be present in all further kernel releases. Things get changed, the flags may
474 be vanished or the reverse -- new added.
476 This file is only present if the CONFIG_MMU kernel configuration option is
479 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
480 bits on both physical and virtual pages associated with a process, and the
481 soft-dirty bit on pte (see Documentation/vm/soft-dirty.txt for details).
482 To clear the bits for all the pages associated with the process
483 > echo 1 > /proc/PID/clear_refs
485 To clear the bits for the anonymous pages associated with the process
486 > echo 2 > /proc/PID/clear_refs
488 To clear the bits for the file mapped pages associated with the process
489 > echo 3 > /proc/PID/clear_refs
491 To clear the soft-dirty bit
492 > echo 4 > /proc/PID/clear_refs
494 To reset the peak resident set size ("high water mark") to the process's
496 > echo 5 > /proc/PID/clear_refs
498 Any other value written to /proc/PID/clear_refs will have no effect.
500 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
501 using /proc/kpageflags and number of times a page is mapped using
502 /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
504 The /proc/pid/numa_maps is an extension based on maps, showing the memory
505 locality and binding policy, as well as the memory usage (in pages) of
506 each mapping. The output follows a general format where mapping details get
507 summarized separated by blank spaces, one mapping per each file line:
509 address policy mapping details
511 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
512 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
513 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
514 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
515 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
516 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
517 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
518 320698b000 default file=/lib64/libc-2.12.so
519 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
520 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
521 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
522 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
523 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
524 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
525 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
526 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
529 "address" is the starting address for the mapping;
530 "policy" reports the NUMA memory policy set for the mapping (see vm/numa_memory_policy.txt);
531 "mapping details" summarizes mapping data such as mapping type, page usage counters,
532 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
533 size, in KB, that is backing the mapping up.
538 Similar to the process entries, the kernel data files give information about
539 the running kernel. The files used to obtain this information are contained in
540 /proc and are listed in Table 1-5. Not all of these will be present in your
541 system. It depends on the kernel configuration and the loaded modules, which
542 files are there, and which are missing.
544 Table 1-5: Kernel info in /proc
545 ..............................................................................
547 apm Advanced power management info
548 buddyinfo Kernel memory allocator information (see text) (2.5)
549 bus Directory containing bus specific information
550 cmdline Kernel command line
551 cpuinfo Info about the CPU
552 devices Available devices (block and character)
553 dma Used DMS channels
554 filesystems Supported filesystems
555 driver Various drivers grouped here, currently rtc (2.4)
556 execdomains Execdomains, related to security (2.4)
557 fb Frame Buffer devices (2.4)
558 fs File system parameters, currently nfs/exports (2.4)
559 ide Directory containing info about the IDE subsystem
560 interrupts Interrupt usage
561 iomem Memory map (2.4)
562 ioports I/O port usage
563 irq Masks for irq to cpu affinity (2.4)(smp?)
564 isapnp ISA PnP (Plug&Play) Info (2.4)
565 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
567 ksyms Kernel symbol table
568 loadavg Load average of last 1, 5 & 15 minutes
572 modules List of loaded modules
573 mounts Mounted filesystems
574 net Networking info (see text)
575 pagetypeinfo Additional page allocator information (see text) (2.5)
576 partitions Table of partitions known to the system
577 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
578 decoupled by lspci (2.4)
580 scsi SCSI info (see text)
581 slabinfo Slab pool info
582 softirqs softirq usage
583 stat Overall statistics
584 swaps Swap space utilization
586 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
587 tty Info of tty drivers
588 uptime Wall clock since boot, combined idle time of all cpus
589 version Kernel version
590 video bttv info of video resources (2.4)
591 vmallocinfo Show vmalloced areas
592 ..............................................................................
594 You can, for example, check which interrupts are currently in use and what
595 they are used for by looking in the file /proc/interrupts:
597 > cat /proc/interrupts
599 0: 8728810 XT-PIC timer
600 1: 895 XT-PIC keyboard
602 3: 531695 XT-PIC aha152x
603 4: 2014133 XT-PIC serial
604 5: 44401 XT-PIC pcnet_cs
607 12: 182918 XT-PIC PS/2 Mouse
609 14: 1232265 XT-PIC ide0
613 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
614 output of a SMP machine):
616 > cat /proc/interrupts
619 0: 1243498 1214548 IO-APIC-edge timer
620 1: 8949 8958 IO-APIC-edge keyboard
621 2: 0 0 XT-PIC cascade
622 5: 11286 10161 IO-APIC-edge soundblaster
623 8: 1 0 IO-APIC-edge rtc
624 9: 27422 27407 IO-APIC-edge 3c503
625 12: 113645 113873 IO-APIC-edge PS/2 Mouse
627 14: 22491 24012 IO-APIC-edge ide0
628 15: 2183 2415 IO-APIC-edge ide1
629 17: 30564 30414 IO-APIC-level eth0
630 18: 177 164 IO-APIC-level bttv
635 NMI is incremented in this case because every timer interrupt generates a NMI
636 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
638 LOC is the local interrupt counter of the internal APIC of every CPU.
640 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
641 connects the CPUs in a SMP system. This means that an error has been detected,
642 the IO-APIC automatically retry the transmission, so it should not be a big
643 problem, but you should read the SMP-FAQ.
645 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
646 /proc/interrupts to display every IRQ vector in use by the system, not
647 just those considered 'most important'. The new vectors are:
649 THR -- interrupt raised when a machine check threshold counter
650 (typically counting ECC corrected errors of memory or cache) exceeds
651 a configurable threshold. Only available on some systems.
653 TRM -- a thermal event interrupt occurs when a temperature threshold
654 has been exceeded for the CPU. This interrupt may also be generated
655 when the temperature drops back to normal.
657 SPU -- a spurious interrupt is some interrupt that was raised then lowered
658 by some IO device before it could be fully processed by the APIC. Hence
659 the APIC sees the interrupt but does not know what device it came from.
660 For this case the APIC will generate the interrupt with a IRQ vector
661 of 0xff. This might also be generated by chipset bugs.
663 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
664 sent from one CPU to another per the needs of the OS. Typically,
665 their statistics are used by kernel developers and interested users to
666 determine the occurrence of interrupts of the given type.
668 The above IRQ vectors are displayed only when relevant. For example,
669 the threshold vector does not exist on x86_64 platforms. Others are
670 suppressed when the system is a uniprocessor. As of this writing, only
671 i386 and x86_64 platforms support the new IRQ vector displays.
673 Of some interest is the introduction of the /proc/irq directory to 2.4.
674 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
675 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
676 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
681 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
682 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
686 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
687 IRQ, you can set it by doing:
689 > echo 1 > /proc/irq/10/smp_affinity
691 This means that only the first CPU will handle the IRQ, but you can also echo
692 5 which means that only the first and fourth CPU can handle the IRQ.
694 The contents of each smp_affinity file is the same by default:
696 > cat /proc/irq/0/smp_affinity
699 There is an alternate interface, smp_affinity_list which allows specifying
700 a cpu range instead of a bitmask:
702 > cat /proc/irq/0/smp_affinity_list
705 The default_smp_affinity mask applies to all non-active IRQs, which are the
706 IRQs which have not yet been allocated/activated, and hence which lack a
707 /proc/irq/[0-9]* directory.
709 The node file on an SMP system shows the node to which the device using the IRQ
710 reports itself as being attached. This hardware locality information does not
711 include information about any possible driver locality preference.
713 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
714 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
716 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
717 between all the CPUs which are allowed to handle it. As usual the kernel has
718 more info than you and does a better job than you, so the defaults are the
719 best choice for almost everyone. [Note this applies only to those IO-APIC's
720 that support "Round Robin" interrupt distribution.]
722 There are three more important subdirectories in /proc: net, scsi, and sys.
723 The general rule is that the contents, or even the existence of these
724 directories, depend on your kernel configuration. If SCSI is not enabled, the
725 directory scsi may not exist. The same is true with the net, which is there
726 only when networking support is present in the running kernel.
728 The slabinfo file gives information about memory usage at the slab level.
729 Linux uses slab pools for memory management above page level in version 2.2.
730 Commonly used objects have their own slab pool (such as network buffers,
731 directory cache, and so on).
733 ..............................................................................
735 > cat /proc/buddyinfo
737 Node 0, zone DMA 0 4 5 4 4 3 ...
738 Node 0, zone Normal 1 0 0 1 101 8 ...
739 Node 0, zone HighMem 2 0 0 1 1 0 ...
741 External fragmentation is a problem under some workloads, and buddyinfo is a
742 useful tool for helping diagnose these problems. Buddyinfo will give you a
743 clue as to how big an area you can safely allocate, or why a previous
746 Each column represents the number of pages of a certain order which are
747 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
748 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
749 available in ZONE_NORMAL, etc...
751 More information relevant to external fragmentation can be found in
754 > cat /proc/pagetypeinfo
758 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
759 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
760 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
761 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
762 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
763 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
764 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
765 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
766 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
767 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
768 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
770 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
771 Node 0, zone DMA 2 0 5 1 0
772 Node 0, zone DMA32 41 6 967 2 0
774 Fragmentation avoidance in the kernel works by grouping pages of different
775 migrate types into the same contiguous regions of memory called page blocks.
776 A page block is typically the size of the default hugepage size e.g. 2MB on
777 X86-64. By keeping pages grouped based on their ability to move, the kernel
778 can reclaim pages within a page block to satisfy a high-order allocation.
780 The pagetypinfo begins with information on the size of a page block. It
781 then gives the same type of information as buddyinfo except broken down
782 by migrate-type and finishes with details on how many page blocks of each
785 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
786 from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can
787 make an estimate of the likely number of huge pages that can be allocated
788 at a given point in time. All the "Movable" blocks should be allocatable
789 unless memory has been mlock()'d. Some of the Reclaimable blocks should
790 also be allocatable although a lot of filesystem metadata may have to be
791 reclaimed to achieve this.
793 ..............................................................................
797 Provides information about distribution and utilization of memory. This
798 varies by architecture and compile options. The following is from a
799 16GB PIII, which has highmem enabled. You may not have all of these fields.
803 The "Locked" indicates whether the mapping is locked in memory or not.
806 MemTotal: 16344972 kB
808 MemAvailable: 14836172 kB
814 HighTotal: 15597528 kB
815 HighFree: 13629632 kB
825 SReclaimable: 159856 kB
826 SUnreclaim: 124508 kB
831 CommitLimit: 7669796 kB
832 Committed_AS: 100056 kB
833 VmallocTotal: 112216 kB
835 VmallocChunk: 111088 kB
836 AnonHugePages: 49152 kB
838 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
839 bits and the kernel binary code)
840 MemFree: The sum of LowFree+HighFree
841 MemAvailable: An estimate of how much memory is available for starting new
842 applications, without swapping. Calculated from MemFree,
843 SReclaimable, the size of the file LRU lists, and the low
844 watermarks in each zone.
845 The estimate takes into account that the system needs some
846 page cache to function well, and that not all reclaimable
847 slab will be reclaimable, due to items being in use. The
848 impact of those factors will vary from system to system.
849 Buffers: Relatively temporary storage for raw disk blocks
850 shouldn't get tremendously large (20MB or so)
851 Cached: in-memory cache for files read from the disk (the
852 pagecache). Doesn't include SwapCached
853 SwapCached: Memory that once was swapped out, is swapped back in but
854 still also is in the swapfile (if memory is needed it
855 doesn't need to be swapped out AGAIN because it is already
856 in the swapfile. This saves I/O)
857 Active: Memory that has been used more recently and usually not
858 reclaimed unless absolutely necessary.
859 Inactive: Memory which has been less recently used. It is more
860 eligible to be reclaimed for other purposes
862 HighFree: Highmem is all memory above ~860MB of physical memory
863 Highmem areas are for use by userspace programs, or
864 for the pagecache. The kernel must use tricks to access
865 this memory, making it slower to access than lowmem.
867 LowFree: Lowmem is memory which can be used for everything that
868 highmem can be used for, but it is also available for the
869 kernel's use for its own data structures. Among many
870 other things, it is where everything from the Slab is
871 allocated. Bad things happen when you're out of lowmem.
872 SwapTotal: total amount of swap space available
873 SwapFree: Memory which has been evicted from RAM, and is temporarily
875 Dirty: Memory which is waiting to get written back to the disk
876 Writeback: Memory which is actively being written back to the disk
877 AnonPages: Non-file backed pages mapped into userspace page tables
878 AnonHugePages: Non-file backed huge pages mapped into userspace page tables
879 Mapped: files which have been mmaped, such as libraries
880 Slab: in-kernel data structures cache
881 SReclaimable: Part of Slab, that might be reclaimed, such as caches
882 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
883 PageTables: amount of memory dedicated to the lowest level of page
885 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
887 Bounce: Memory used for block device "bounce buffers"
888 WritebackTmp: Memory used by FUSE for temporary writeback buffers
889 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
890 this is the total amount of memory currently available to
891 be allocated on the system. This limit is only adhered to
892 if strict overcommit accounting is enabled (mode 2 in
893 'vm.overcommit_memory').
894 The CommitLimit is calculated with the following formula:
895 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
896 overcommit_ratio / 100 + [total swap pages]
897 For example, on a system with 1G of physical RAM and 7G
898 of swap with a `vm.overcommit_ratio` of 30 it would
899 yield a CommitLimit of 7.3G.
900 For more details, see the memory overcommit documentation
901 in vm/overcommit-accounting.
902 Committed_AS: The amount of memory presently allocated on the system.
903 The committed memory is a sum of all of the memory which
904 has been allocated by processes, even if it has not been
905 "used" by them as of yet. A process which malloc()'s 1G
906 of memory, but only touches 300M of it will show up as
907 using 1G. This 1G is memory which has been "committed" to
908 by the VM and can be used at any time by the allocating
909 application. With strict overcommit enabled on the system
910 (mode 2 in 'vm.overcommit_memory'),allocations which would
911 exceed the CommitLimit (detailed above) will not be permitted.
912 This is useful if one needs to guarantee that processes will
913 not fail due to lack of memory once that memory has been
914 successfully allocated.
915 VmallocTotal: total size of vmalloc memory area
916 VmallocUsed: amount of vmalloc area which is used
917 VmallocChunk: largest contiguous block of vmalloc area which is free
919 ..............................................................................
923 Provides information about vmalloced/vmaped areas. One line per area,
924 containing the virtual address range of the area, size in bytes,
925 caller information of the creator, and optional information depending
926 on the kind of area :
928 pages=nr number of pages
929 phys=addr if a physical address was specified
930 ioremap I/O mapping (ioremap() and friends)
931 vmalloc vmalloc() area
934 vpages buffer for pages pointers was vmalloced (huge area)
935 N<node>=nr (Only on NUMA kernels)
936 Number of pages allocated on memory node <node>
938 > cat /proc/vmallocinfo
939 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
940 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
941 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
942 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
943 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
944 phys=7fee8000 ioremap
945 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
946 phys=7fee7000 ioremap
947 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
948 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
949 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
950 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
952 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
953 /0x130 [x_tables] pages=4 vmalloc N0=4
954 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
955 pages=14 vmalloc N2=14
956 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
958 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
960 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
961 pages=10 vmalloc N0=10
963 ..............................................................................
967 Provides counts of softirq handlers serviced since boot time, for each cpu.
972 TIMER: 27166 27120 27097 27034
977 SCHED: 27035 26983 26971 26746
979 RCU: 1678 1769 2178 2250
982 1.3 IDE devices in /proc/ide
983 ----------------------------
985 The subdirectory /proc/ide contains information about all IDE devices of which
986 the kernel is aware. There is one subdirectory for each IDE controller, the
987 file drivers and a link for each IDE device, pointing to the device directory
988 in the controller specific subtree.
990 The file drivers contains general information about the drivers used for the
993 > cat /proc/ide/drivers
994 ide-cdrom version 4.53
995 ide-disk version 1.08
997 More detailed information can be found in the controller specific
998 subdirectories. These are named ide0, ide1 and so on. Each of these
999 directories contains the files shown in table 1-6.
1002 Table 1-6: IDE controller info in /proc/ide/ide?
1003 ..............................................................................
1005 channel IDE channel (0 or 1)
1006 config Configuration (only for PCI/IDE bridge)
1008 model Type/Chipset of IDE controller
1009 ..............................................................................
1011 Each device connected to a controller has a separate subdirectory in the
1012 controllers directory. The files listed in table 1-7 are contained in these
1016 Table 1-7: IDE device information
1017 ..............................................................................
1020 capacity Capacity of the medium (in 512Byte blocks)
1021 driver driver and version
1022 geometry physical and logical geometry
1023 identify device identify block
1025 model device identifier
1026 settings device setup
1027 smart_thresholds IDE disk management thresholds
1028 smart_values IDE disk management values
1029 ..............................................................................
1031 The most interesting file is settings. This file contains a nice overview of
1032 the drive parameters:
1034 # cat /proc/ide/ide0/hda/settings
1035 name value min max mode
1036 ---- ----- --- --- ----
1037 bios_cyl 526 0 65535 rw
1038 bios_head 255 0 255 rw
1039 bios_sect 63 0 63 rw
1040 breada_readahead 4 0 127 rw
1042 file_readahead 72 0 2097151 rw
1044 keepsettings 0 0 1 rw
1045 max_kb_per_request 122 1 127 rw
1049 pio_mode write-only 0 255 w
1055 1.4 Networking info in /proc/net
1056 --------------------------------
1058 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1059 additional values you get for IP version 6 if you configure the kernel to
1060 support this. Table 1-9 lists the files and their meaning.
1063 Table 1-8: IPv6 info in /proc/net
1064 ..............................................................................
1066 udp6 UDP sockets (IPv6)
1067 tcp6 TCP sockets (IPv6)
1068 raw6 Raw device statistics (IPv6)
1069 igmp6 IP multicast addresses, which this host joined (IPv6)
1070 if_inet6 List of IPv6 interface addresses
1071 ipv6_route Kernel routing table for IPv6
1072 rt6_stats Global IPv6 routing tables statistics
1073 sockstat6 Socket statistics (IPv6)
1074 snmp6 Snmp data (IPv6)
1075 ..............................................................................
1078 Table 1-9: Network info in /proc/net
1079 ..............................................................................
1081 arp Kernel ARP table
1082 dev network devices with statistics
1083 dev_mcast the Layer2 multicast groups a device is listening too
1084 (interface index, label, number of references, number of bound
1086 dev_stat network device status
1087 ip_fwchains Firewall chain linkage
1088 ip_fwnames Firewall chain names
1089 ip_masq Directory containing the masquerading tables
1090 ip_masquerade Major masquerading table
1091 netstat Network statistics
1092 raw raw device statistics
1093 route Kernel routing table
1094 rpc Directory containing rpc info
1095 rt_cache Routing cache
1097 sockstat Socket statistics
1100 unix UNIX domain sockets
1101 wireless Wireless interface data (Wavelan etc)
1102 igmp IP multicast addresses, which this host joined
1103 psched Global packet scheduler parameters.
1104 netlink List of PF_NETLINK sockets
1105 ip_mr_vifs List of multicast virtual interfaces
1106 ip_mr_cache List of multicast routing cache
1107 ..............................................................................
1109 You can use this information to see which network devices are available in
1110 your system and how much traffic was routed over those devices:
1113 Inter-|Receive |[...
1114 face |bytes packets errs drop fifo frame compressed multicast|[...
1115 lo: 908188 5596 0 0 0 0 0 0 [...
1116 ppp0:15475140 20721 410 0 0 410 0 0 [...
1117 eth0: 614530 7085 0 0 0 0 0 1 [...
1120 ...] bytes packets errs drop fifo colls carrier compressed
1121 ...] 908188 5596 0 0 0 0 0 0
1122 ...] 1375103 17405 0 0 0 0 0 0
1123 ...] 1703981 5535 0 0 0 3 0 0
1125 In addition, each Channel Bond interface has its own directory. For
1126 example, the bond0 device will have a directory called /proc/net/bond0/.
1127 It will contain information that is specific to that bond, such as the
1128 current slaves of the bond, the link status of the slaves, and how
1129 many times the slaves link has failed.
1134 If you have a SCSI host adapter in your system, you'll find a subdirectory
1135 named after the driver for this adapter in /proc/scsi. You'll also see a list
1136 of all recognized SCSI devices in /proc/scsi:
1138 >cat /proc/scsi/scsi
1140 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1141 Vendor: IBM Model: DGHS09U Rev: 03E0
1142 Type: Direct-Access ANSI SCSI revision: 03
1143 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1144 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1145 Type: CD-ROM ANSI SCSI revision: 02
1148 The directory named after the driver has one file for each adapter found in
1149 the system. These files contain information about the controller, including
1150 the used IRQ and the IO address range. The amount of information shown is
1151 dependent on the adapter you use. The example shows the output for an Adaptec
1152 AHA-2940 SCSI adapter:
1154 > cat /proc/scsi/aic7xxx/0
1156 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1158 TCQ Enabled By Default : Disabled
1159 AIC7XXX_PROC_STATS : Disabled
1160 AIC7XXX_RESET_DELAY : 5
1161 Adapter Configuration:
1162 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1163 Ultra Wide Controller
1164 PCI MMAPed I/O Base: 0xeb001000
1165 Adapter SEEPROM Config: SEEPROM found and used.
1166 Adaptec SCSI BIOS: Enabled
1168 SCBs: Active 0, Max Active 2,
1169 Allocated 15, HW 16, Page 255
1171 BIOS Control Word: 0x18b6
1172 Adapter Control Word: 0x005b
1173 Extended Translation: Enabled
1174 Disconnect Enable Flags: 0xffff
1175 Ultra Enable Flags: 0x0001
1176 Tag Queue Enable Flags: 0x0000
1177 Ordered Queue Tag Flags: 0x0000
1178 Default Tag Queue Depth: 8
1179 Tagged Queue By Device array for aic7xxx host instance 0:
1180 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1181 Actual queue depth per device for aic7xxx host instance 0:
1182 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1185 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1186 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1187 Total transfers 160151 (74577 reads and 85574 writes)
1189 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1190 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1191 Total transfers 0 (0 reads and 0 writes)
1194 1.6 Parallel port info in /proc/parport
1195 ---------------------------------------
1197 The directory /proc/parport contains information about the parallel ports of
1198 your system. It has one subdirectory for each port, named after the port
1201 These directories contain the four files shown in Table 1-10.
1204 Table 1-10: Files in /proc/parport
1205 ..............................................................................
1207 autoprobe Any IEEE-1284 device ID information that has been acquired.
1208 devices list of the device drivers using that port. A + will appear by the
1209 name of the device currently using the port (it might not appear
1211 hardware Parallel port's base address, IRQ line and DMA channel.
1212 irq IRQ that parport is using for that port. This is in a separate
1213 file to allow you to alter it by writing a new value in (IRQ
1215 ..............................................................................
1217 1.7 TTY info in /proc/tty
1218 -------------------------
1220 Information about the available and actually used tty's can be found in the
1221 directory /proc/tty.You'll find entries for drivers and line disciplines in
1222 this directory, as shown in Table 1-11.
1225 Table 1-11: Files in /proc/tty
1226 ..............................................................................
1228 drivers list of drivers and their usage
1229 ldiscs registered line disciplines
1230 driver/serial usage statistic and status of single tty lines
1231 ..............................................................................
1233 To see which tty's are currently in use, you can simply look into the file
1236 > cat /proc/tty/drivers
1237 pty_slave /dev/pts 136 0-255 pty:slave
1238 pty_master /dev/ptm 128 0-255 pty:master
1239 pty_slave /dev/ttyp 3 0-255 pty:slave
1240 pty_master /dev/pty 2 0-255 pty:master
1241 serial /dev/cua 5 64-67 serial:callout
1242 serial /dev/ttyS 4 64-67 serial
1243 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1244 /dev/ptmx /dev/ptmx 5 2 system
1245 /dev/console /dev/console 5 1 system:console
1246 /dev/tty /dev/tty 5 0 system:/dev/tty
1247 unknown /dev/tty 4 1-63 console
1250 1.8 Miscellaneous kernel statistics in /proc/stat
1251 -------------------------------------------------
1253 Various pieces of information about kernel activity are available in the
1254 /proc/stat file. All of the numbers reported in this file are aggregates
1255 since the system first booted. For a quick look, simply cat the file:
1258 cpu 2255 34 2290 22625563 6290 127 456 0 0
1259 cpu0 1132 34 1441 11311718 3675 127 438 0 0
1260 cpu1 1123 0 849 11313845 2614 0 18 0 0
1261 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1267 softirq 183433 0 21755 12 39 1137 231 21459 2263
1269 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1270 lines. These numbers identify the amount of time the CPU has spent performing
1271 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1272 second). The meanings of the columns are as follows, from left to right:
1274 - user: normal processes executing in user mode
1275 - nice: niced processes executing in user mode
1276 - system: processes executing in kernel mode
1277 - idle: twiddling thumbs
1278 - iowait: waiting for I/O to complete
1279 - irq: servicing interrupts
1280 - softirq: servicing softirqs
1281 - steal: involuntary wait
1282 - guest: running a normal guest
1283 - guest_nice: running a niced guest
1285 The "intr" line gives counts of interrupts serviced since boot time, for each
1286 of the possible system interrupts. The first column is the total of all
1287 interrupts serviced including unnumbered architecture specific interrupts;
1288 each subsequent column is the total for that particular numbered interrupt.
1289 Unnumbered interrupts are not shown, only summed into the total.
1291 The "ctxt" line gives the total number of context switches across all CPUs.
1293 The "btime" line gives the time at which the system booted, in seconds since
1296 The "processes" line gives the number of processes and threads created, which
1297 includes (but is not limited to) those created by calls to the fork() and
1298 clone() system calls.
1300 The "procs_running" line gives the total number of threads that are
1301 running or ready to run (i.e., the total number of runnable threads).
1303 The "procs_blocked" line gives the number of processes currently blocked,
1304 waiting for I/O to complete.
1306 The "softirq" line gives counts of softirqs serviced since boot time, for each
1307 of the possible system softirqs. The first column is the total of all
1308 softirqs serviced; each subsequent column is the total for that particular
1312 1.9 Ext4 file system parameters
1313 -------------------------------
1315 Information about mounted ext4 file systems can be found in
1316 /proc/fs/ext4. Each mounted filesystem will have a directory in
1317 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1318 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1319 in Table 1-12, below.
1321 Table 1-12: Files in /proc/fs/ext4/<devname>
1322 ..............................................................................
1324 mb_groups details of multiblock allocator buddy cache of free blocks
1325 ..............................................................................
1329 Shows registered system console lines.
1331 To see which character device lines are currently used for the system console
1332 /dev/console, you may simply look into the file /proc/consoles:
1334 > cat /proc/consoles
1340 device name of the device
1341 operations R = can do read operations
1342 W = can do write operations
1344 flags E = it is enabled
1345 C = it is preferred console
1346 B = it is primary boot console
1347 p = it is used for printk buffer
1348 b = it is not a TTY but a Braille device
1349 a = it is safe to use when cpu is offline
1350 major:minor major and minor number of the device separated by a colon
1352 ------------------------------------------------------------------------------
1354 ------------------------------------------------------------------------------
1355 The /proc file system serves information about the running system. It not only
1356 allows access to process data but also allows you to request the kernel status
1357 by reading files in the hierarchy.
1359 The directory structure of /proc reflects the types of information and makes
1360 it easy, if not obvious, where to look for specific data.
1361 ------------------------------------------------------------------------------
1363 ------------------------------------------------------------------------------
1364 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1365 ------------------------------------------------------------------------------
1367 ------------------------------------------------------------------------------
1369 ------------------------------------------------------------------------------
1370 * Modifying kernel parameters by writing into files found in /proc/sys
1371 * Exploring the files which modify certain parameters
1372 * Review of the /proc/sys file tree
1373 ------------------------------------------------------------------------------
1376 A very interesting part of /proc is the directory /proc/sys. This is not only
1377 a source of information, it also allows you to change parameters within the
1378 kernel. Be very careful when attempting this. You can optimize your system,
1379 but you can also cause it to crash. Never alter kernel parameters on a
1380 production system. Set up a development machine and test to make sure that
1381 everything works the way you want it to. You may have no alternative but to
1382 reboot the machine once an error has been made.
1384 To change a value, simply echo the new value into the file. An example is
1385 given below in the section on the file system data. You need to be root to do
1386 this. You can create your own boot script to perform this every time your
1389 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1390 general things in the operation of the Linux kernel. Since some of the files
1391 can inadvertently disrupt your system, it is advisable to read both
1392 documentation and source before actually making adjustments. In any case, be
1393 very careful when writing to any of these files. The entries in /proc may
1394 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1395 review the kernel documentation in the directory /usr/src/linux/Documentation.
1396 This chapter is heavily based on the documentation included in the pre 2.2
1397 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1399 Please see: Documentation/sysctl/ directory for descriptions of these
1402 ------------------------------------------------------------------------------
1404 ------------------------------------------------------------------------------
1405 Certain aspects of kernel behavior can be modified at runtime, without the
1406 need to recompile the kernel, or even to reboot the system. The files in the
1407 /proc/sys tree can not only be read, but also modified. You can use the echo
1408 command to write value into these files, thereby changing the default settings
1410 ------------------------------------------------------------------------------
1412 ------------------------------------------------------------------------------
1413 CHAPTER 3: PER-PROCESS PARAMETERS
1414 ------------------------------------------------------------------------------
1416 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1417 --------------------------------------------------------------------------------
1419 These file can be used to adjust the badness heuristic used to select which
1420 process gets killed in out of memory conditions.
1422 The badness heuristic assigns a value to each candidate task ranging from 0
1423 (never kill) to 1000 (always kill) to determine which process is targeted. The
1424 units are roughly a proportion along that range of allowed memory the process
1425 may allocate from based on an estimation of its current memory and swap use.
1426 For example, if a task is using all allowed memory, its badness score will be
1427 1000. If it is using half of its allowed memory, its score will be 500.
1429 There is an additional factor included in the badness score: the current memory
1430 and swap usage is discounted by 3% for root processes.
1432 The amount of "allowed" memory depends on the context in which the oom killer
1433 was called. If it is due to the memory assigned to the allocating task's cpuset
1434 being exhausted, the allowed memory represents the set of mems assigned to that
1435 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1436 memory represents the set of mempolicy nodes. If it is due to a memory
1437 limit (or swap limit) being reached, the allowed memory is that configured
1438 limit. Finally, if it is due to the entire system being out of memory, the
1439 allowed memory represents all allocatable resources.
1441 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1442 is used to determine which task to kill. Acceptable values range from -1000
1443 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1444 polarize the preference for oom killing either by always preferring a certain
1445 task or completely disabling it. The lowest possible value, -1000, is
1446 equivalent to disabling oom killing entirely for that task since it will always
1447 report a badness score of 0.
1449 Consequently, it is very simple for userspace to define the amount of memory to
1450 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1451 example, is roughly equivalent to allowing the remainder of tasks sharing the
1452 same system, cpuset, mempolicy, or memory controller resources to use at least
1453 50% more memory. A value of -500, on the other hand, would be roughly
1454 equivalent to discounting 50% of the task's allowed memory from being considered
1455 as scoring against the task.
1457 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1458 be used to tune the badness score. Its acceptable values range from -16
1459 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1460 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1461 scaled linearly with /proc/<pid>/oom_score_adj.
1463 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1464 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1465 requires CAP_SYS_RESOURCE.
1467 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1468 generation children with separate address spaces instead, if possible. This
1469 avoids servers and important system daemons from being killed and loses the
1470 minimal amount of work.
1473 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1474 -------------------------------------------------------------
1476 This file can be used to check the current score used by the oom-killer is for
1477 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1478 process should be killed in an out-of-memory situation.
1481 3.3 /proc/<pid>/io - Display the IO accounting fields
1482 -------------------------------------------------------
1484 This file contains IO statistics for each running process
1489 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1492 test:/tmp # cat /proc/3828/io
1498 write_bytes: 323932160
1499 cancelled_write_bytes: 0
1508 I/O counter: chars read
1509 The number of bytes which this task has caused to be read from storage. This
1510 is simply the sum of bytes which this process passed to read() and pread().
1511 It includes things like tty IO and it is unaffected by whether or not actual
1512 physical disk IO was required (the read might have been satisfied from
1519 I/O counter: chars written
1520 The number of bytes which this task has caused, or shall cause to be written
1521 to disk. Similar caveats apply here as with rchar.
1527 I/O counter: read syscalls
1528 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1535 I/O counter: write syscalls
1536 Attempt to count the number of write I/O operations, i.e. syscalls like
1537 write() and pwrite().
1543 I/O counter: bytes read
1544 Attempt to count the number of bytes which this process really did cause to
1545 be fetched from the storage layer. Done at the submit_bio() level, so it is
1546 accurate for block-backed filesystems. <please add status regarding NFS and
1547 CIFS at a later time>
1553 I/O counter: bytes written
1554 Attempt to count the number of bytes which this process caused to be sent to
1555 the storage layer. This is done at page-dirtying time.
1558 cancelled_write_bytes
1559 ---------------------
1561 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1562 then deletes the file, it will in fact perform no writeout. But it will have
1563 been accounted as having caused 1MB of write.
1564 In other words: The number of bytes which this process caused to not happen,
1565 by truncating pagecache. A task can cause "negative" IO too. If this task
1566 truncates some dirty pagecache, some IO which another task has been accounted
1567 for (in its write_bytes) will not be happening. We _could_ just subtract that
1568 from the truncating task's write_bytes, but there is information loss in doing
1575 At its current implementation state, this is a bit racy on 32-bit machines: if
1576 process A reads process B's /proc/pid/io while process B is updating one of
1577 those 64-bit counters, process A could see an intermediate result.
1580 More information about this can be found within the taskstats documentation in
1581 Documentation/accounting.
1583 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1584 ---------------------------------------------------------------
1585 When a process is dumped, all anonymous memory is written to a core file as
1586 long as the size of the core file isn't limited. But sometimes we don't want
1587 to dump some memory segments, for example, huge shared memory. Conversely,
1588 sometimes we want to save file-backed memory segments into a core file, not
1589 only the individual files.
1591 /proc/<pid>/coredump_filter allows you to customize which memory segments
1592 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1593 of memory types. If a bit of the bitmask is set, memory segments of the
1594 corresponding memory type are dumped, otherwise they are not dumped.
1596 The following 7 memory types are supported:
1597 - (bit 0) anonymous private memory
1598 - (bit 1) anonymous shared memory
1599 - (bit 2) file-backed private memory
1600 - (bit 3) file-backed shared memory
1601 - (bit 4) ELF header pages in file-backed private memory areas (it is
1602 effective only if the bit 2 is cleared)
1603 - (bit 5) hugetlb private memory
1604 - (bit 6) hugetlb shared memory
1606 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1607 are always dumped regardless of the bitmask status.
1609 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
1610 effected by bit 5-6.
1612 Default value of coredump_filter is 0x23; this means all anonymous memory
1613 segments and hugetlb private memory are dumped.
1615 If you don't want to dump all shared memory segments attached to pid 1234,
1616 write 0x21 to the process's proc file.
1618 $ echo 0x21 > /proc/1234/coredump_filter
1620 When a new process is created, the process inherits the bitmask status from its
1621 parent. It is useful to set up coredump_filter before the program runs.
1624 $ echo 0x7 > /proc/self/coredump_filter
1627 3.5 /proc/<pid>/mountinfo - Information about mounts
1628 --------------------------------------------------------
1630 This file contains lines of the form:
1632 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1633 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1635 (1) mount ID: unique identifier of the mount (may be reused after umount)
1636 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1637 (3) major:minor: value of st_dev for files on filesystem
1638 (4) root: root of the mount within the filesystem
1639 (5) mount point: mount point relative to the process's root
1640 (6) mount options: per mount options
1641 (7) optional fields: zero or more fields of the form "tag[:value]"
1642 (8) separator: marks the end of the optional fields
1643 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1644 (10) mount source: filesystem specific information or "none"
1645 (11) super options: per super block options
1647 Parsers should ignore all unrecognised optional fields. Currently the
1648 possible optional fields are:
1650 shared:X mount is shared in peer group X
1651 master:X mount is slave to peer group X
1652 propagate_from:X mount is slave and receives propagation from peer group X (*)
1653 unbindable mount is unbindable
1655 (*) X is the closest dominant peer group under the process's root. If
1656 X is the immediate master of the mount, or if there's no dominant peer
1657 group under the same root, then only the "master:X" field is present
1658 and not the "propagate_from:X" field.
1660 For more information on mount propagation see:
1662 Documentation/filesystems/sharedsubtree.txt
1665 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1666 --------------------------------------------------------
1667 These files provide a method to access a tasks comm value. It also allows for
1668 a task to set its own or one of its thread siblings comm value. The comm value
1669 is limited in size compared to the cmdline value, so writing anything longer
1670 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1674 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1675 -------------------------------------------------------------------------
1676 This file provides a fast way to retrieve first level children pids
1677 of a task pointed by <pid>/<tid> pair. The format is a space separated
1680 Note the "first level" here -- if a child has own children they will
1681 not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1682 to obtain the descendants.
1684 Since this interface is intended to be fast and cheap it doesn't
1685 guarantee to provide precise results and some children might be
1686 skipped, especially if they've exited right after we printed their
1687 pids, so one need to either stop or freeze processes being inspected
1688 if precise results are needed.
1691 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1692 ---------------------------------------------------------------
1693 This file provides information associated with an opened file. The regular
1694 files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos'
1695 represents the current offset of the opened file in decimal form [see lseek(2)
1696 for details], 'flags' denotes the octal O_xxx mask the file has been
1697 created with [see open(2) for details] and 'mnt_id' represents mount ID of
1698 the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo
1707 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1708 pair provide additional information particular to the objects they represent.
1717 where 'eventfd-count' is hex value of a counter.
1724 sigmask: 0000000000000200
1726 where 'sigmask' is hex value of the signal mask associated
1734 tfd: 5 events: 1d data: ffffffffffffffff
1736 where 'tfd' is a target file descriptor number in decimal form,
1737 'events' is events mask being watched and the 'data' is data
1738 associated with a target [see epoll(7) for more details].
1742 For inotify files the format is the following
1746 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1748 where 'wd' is a watch descriptor in decimal form, ie a target file
1749 descriptor number, 'ino' and 'sdev' are inode and device where the
1750 target file resides and the 'mask' is the mask of events, all in hex
1751 form [see inotify(7) for more details].
1753 If the kernel was built with exportfs support, the path to the target
1754 file is encoded as a file handle. The file handle is provided by three
1755 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1758 If the kernel is built without exportfs support the file handle won't be
1761 If there is no inotify mark attached yet the 'inotify' line will be omitted.
1763 For fanotify files the format is
1768 fanotify flags:10 event-flags:0
1769 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1770 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
1772 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
1773 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
1774 flags associated with mark which are tracked separately from events
1775 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
1776 mask and 'ignored_mask' is the mask of events which are to be ignored.
1777 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
1778 does provide information about flags and mask used in fanotify_mark
1779 call [see fsnotify manpage for details].
1781 While the first three lines are mandatory and always printed, the rest is
1782 optional and may be omitted if no marks created yet.
1793 it_value: (0, 49406829)
1796 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
1797 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
1798 flags in octal form been used to setup the timer [see timerfd_settime(2) for
1799 details]. 'it_value' is remaining time until the timer exiration.
1800 'it_interval' is the interval for the timer. Note the timer might be set up
1801 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
1802 still exhibits timer's remaining time.
1804 3.9 /proc/<pid>/map_files - Information about memory mapped files
1805 ---------------------------------------------------------------------
1806 This directory contains symbolic links which represent memory mapped files
1807 the process is maintaining. Example output:
1809 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
1810 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
1811 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
1813 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
1814 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
1816 The name of a link represents the virtual memory bounds of a mapping, i.e.
1817 vm_area_struct::vm_start-vm_area_struct::vm_end.
1819 The main purpose of the map_files is to retrieve a set of memory mapped
1820 files in a fast way instead of parsing /proc/<pid>/maps or
1821 /proc/<pid>/smaps, both of which contain many more records. At the same
1822 time one can open(2) mappings from the listings of two processes and
1823 comparing their inode numbers to figure out which anonymous memory areas
1824 are actually shared.
1826 ------------------------------------------------------------------------------
1828 ------------------------------------------------------------------------------
1831 ---------------------
1833 The following mount options are supported:
1835 hidepid= Set /proc/<pid>/ access mode.
1836 gid= Set the group authorized to learn processes information.
1838 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1841 hidepid=1 means users may not access any /proc/<pid>/ directories but their
1842 own. Sensitive files like cmdline, sched*, status are now protected against
1843 other users. This makes it impossible to learn whether any user runs
1844 specific program (given the program doesn't reveal itself by its behaviour).
1845 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1846 poorly written programs passing sensitive information via program arguments are
1847 now protected against local eavesdroppers.
1849 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1850 users. It doesn't mean that it hides a fact whether a process with a specific
1851 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1852 but it hides process' uid and gid, which may be learned by stat()'ing
1853 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
1854 information about running processes, whether some daemon runs with elevated
1855 privileges, whether other user runs some sensitive program, whether other users
1856 run any program at all, etc.
1858 gid= defines a group authorized to learn processes information otherwise
1859 prohibited by hidepid=. If you use some daemon like identd which needs to learn
1860 information about processes information, just add identd to this group.
projects / mirror_ubuntu-zesty-kernel.git / blob