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
49 ------------------------------------------------------------------------------
51 ------------------------------------------------------------------------------
53 0.1 Introduction/Credits
54 ------------------------
56 This documentation is part of a soon (or so we hope) to be released book on
57 the SuSE Linux distribution. As there is no complete documentation for the
58 /proc file system and we've used many freely available sources to write these
59 chapters, it seems only fair to give the work back to the Linux community.
60 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
61 afraid it's still far from complete, but we hope it will be useful. As far as
62 we know, it is the first 'all-in-one' document about the /proc file system. It
63 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
64 SPARC, AXP, etc., features, you probably won't find what you are looking for.
65 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
66 additions and patches are welcome and will be added to this document if you
69 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
70 other people for help compiling this documentation. We'd also like to extend a
71 special thank you to Andi Kleen for documentation, which we relied on heavily
72 to create this document, as well as the additional information he provided.
73 Thanks to everybody else who contributed source or docs to the Linux kernel
74 and helped create a great piece of software... :)
76 If you have any comments, corrections or additions, please don't hesitate to
77 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
80 The latest version of this document is available online at
81 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
83 If the above direction does not works for you, you could try the kernel
84 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
85 comandante@zaralinux.com.
90 We don't guarantee the correctness of this document, and if you come to us
91 complaining about how you screwed up your system because of incorrect
92 documentation, we won't feel responsible...
94 ------------------------------------------------------------------------------
95 CHAPTER 1: COLLECTING SYSTEM INFORMATION
96 ------------------------------------------------------------------------------
98 ------------------------------------------------------------------------------
100 ------------------------------------------------------------------------------
101 * Investigating the properties of the pseudo file system /proc and its
102 ability to provide information on the running Linux system
103 * Examining /proc's structure
104 * Uncovering various information about the kernel and the processes running
106 ------------------------------------------------------------------------------
109 The proc file system acts as an interface to internal data structures in the
110 kernel. It can be used to obtain information about the system and to change
111 certain kernel parameters at runtime (sysctl).
113 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
114 show you how you can use /proc/sys to change settings.
116 1.1 Process-Specific Subdirectories
117 -----------------------------------
119 The directory /proc contains (among other things) one subdirectory for each
120 process running on the system, which is named after the process ID (PID).
122 The link self points to the process reading the file system. Each process
123 subdirectory has the entries listed in Table 1-1.
126 Table 1-1: Process specific entries in /proc
127 ..............................................................................
129 clear_refs Clears page referenced bits shown in smaps output
130 cmdline Command line arguments
131 cpu Current and last cpu in which it was executed (2.4)(smp)
132 cwd Link to the current working directory
133 environ Values of environment variables
134 exe Link to the executable of this process
135 fd Directory, which contains all file descriptors
136 maps Memory maps to executables and library files (2.4)
137 mem Memory held by this process
138 root Link to the root directory of this process
140 statm Process memory status information
141 status Process status in human readable form
142 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
144 stack Report full stack trace, enable via CONFIG_STACKTRACE
145 smaps a extension based on maps, showing the memory consumption of
146 each mapping and flags associated with it
147 ..............................................................................
149 For example, to get the status information of a process, all you have to do is
150 read the file /proc/PID/status:
152 >cat /proc/self/status
176 SigPnd: 0000000000000000
177 ShdPnd: 0000000000000000
178 SigBlk: 0000000000000000
179 SigIgn: 0000000000000000
180 SigCgt: 0000000000000000
181 CapInh: 00000000fffffeff
182 CapPrm: 0000000000000000
183 CapEff: 0000000000000000
184 CapBnd: ffffffffffffffff
186 voluntary_ctxt_switches: 0
187 nonvoluntary_ctxt_switches: 1
189 This shows you nearly the same information you would get if you viewed it with
190 the ps command. In fact, ps uses the proc file system to obtain its
191 information. But you get a more detailed view of the process by reading the
192 file /proc/PID/status. It fields are described in table 1-2.
194 The statm file contains more detailed information about the process
195 memory usage. Its seven fields are explained in Table 1-3. The stat file
196 contains details information about the process itself. Its fields are
197 explained in Table 1-4.
199 (for SMP CONFIG users)
200 For making accounting scalable, RSS related information are handled in
201 asynchronous manner and the vaule may not be very precise. To see a precise
202 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
203 It's slow but very precise.
205 Table 1-2: Contents of the status files (as of 2.6.30-rc7)
206 ..............................................................................
208 Name filename of the executable
209 State state (R is running, S is sleeping, D is sleeping
210 in an uninterruptible wait, Z is zombie,
211 T is traced or stopped)
214 PPid process id of the parent process
215 TracerPid PID of process tracing this process (0 if not)
216 Uid Real, effective, saved set, and file system UIDs
217 Gid Real, effective, saved set, and file system GIDs
218 FDSize number of file descriptor slots currently allocated
219 Groups supplementary group list
220 VmPeak peak virtual memory size
221 VmSize total program size
222 VmLck locked memory size
223 VmHWM peak resident set size ("high water mark")
224 VmRSS size of memory portions
225 VmData size of data, stack, and text segments
226 VmStk size of data, stack, and text segments
227 VmExe size of text segment
228 VmLib size of shared library code
229 VmPTE size of page table entries
230 VmSwap size of swap usage (the number of referred swapents)
231 Threads number of threads
232 SigQ number of signals queued/max. number for queue
233 SigPnd bitmap of pending signals for the thread
234 ShdPnd bitmap of shared pending signals for the process
235 SigBlk bitmap of blocked signals
236 SigIgn bitmap of ignored signals
237 SigCgt bitmap of catched signals
238 CapInh bitmap of inheritable capabilities
239 CapPrm bitmap of permitted capabilities
240 CapEff bitmap of effective capabilities
241 CapBnd bitmap of capabilities bounding set
242 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
243 Cpus_allowed mask of CPUs on which this process may run
244 Cpus_allowed_list Same as previous, but in "list format"
245 Mems_allowed mask of memory nodes allowed to this process
246 Mems_allowed_list Same as previous, but in "list format"
247 voluntary_ctxt_switches number of voluntary context switches
248 nonvoluntary_ctxt_switches number of non voluntary context switches
249 ..............................................................................
251 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
252 ..............................................................................
254 size total program size (pages) (same as VmSize in status)
255 resident size of memory portions (pages) (same as VmRSS in status)
256 shared number of pages that are shared (i.e. backed by a file)
257 trs number of pages that are 'code' (not including libs; broken,
258 includes data segment)
259 lrs number of pages of library (always 0 on 2.6)
260 drs number of pages of data/stack (including libs; broken,
261 includes library text)
262 dt number of dirty pages (always 0 on 2.6)
263 ..............................................................................
266 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
267 ..............................................................................
270 tcomm filename of the executable
271 state state (R is running, S is sleeping, D is sleeping in an
272 uninterruptible wait, Z is zombie, T is traced or stopped)
273 ppid process id of the parent process
274 pgrp pgrp of the process
276 tty_nr tty the process uses
277 tty_pgrp pgrp of the tty
279 min_flt number of minor faults
280 cmin_flt number of minor faults with child's
281 maj_flt number of major faults
282 cmaj_flt number of major faults with child's
283 utime user mode jiffies
284 stime kernel mode jiffies
285 cutime user mode jiffies with child's
286 cstime kernel mode jiffies with child's
287 priority priority level
289 num_threads number of threads
290 it_real_value (obsolete, always 0)
291 start_time time the process started after system boot
292 vsize virtual memory size
293 rss resident set memory size
294 rsslim current limit in bytes on the rss
295 start_code address above which program text can run
296 end_code address below which program text can run
297 start_stack address of the start of the main process stack
298 esp current value of ESP
299 eip current value of EIP
300 pending bitmap of pending signals
301 blocked bitmap of blocked signals
302 sigign bitmap of ignored signals
303 sigcatch bitmap of catched signals
304 wchan address where process went to sleep
307 exit_signal signal to send to parent thread on exit
308 task_cpu which CPU the task is scheduled on
309 rt_priority realtime priority
310 policy scheduling policy (man sched_setscheduler)
311 blkio_ticks time spent waiting for block IO
312 gtime guest time of the task in jiffies
313 cgtime guest time of the task children in jiffies
314 start_data address above which program data+bss is placed
315 end_data address below which program data+bss is placed
316 start_brk address above which program heap can be expanded with brk()
317 arg_start address above which program command line is placed
318 arg_end address below which program command line is placed
319 env_start address above which program environment is placed
320 env_end address below which program environment is placed
321 exit_code the thread's exit_code in the form reported by the waitpid system call
322 ..............................................................................
324 The /proc/PID/maps file containing the currently mapped memory regions and
325 their access permissions.
329 address perms offset dev inode pathname
331 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
332 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
333 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
334 a7cb1000-a7cb2000 ---p 00000000 00:00 0
335 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
336 a7eb2000-a7eb3000 ---p 00000000 00:00 0
337 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack:1001]
338 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
339 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
340 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
341 a800b000-a800e000 rw-p 00000000 00:00 0
342 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
343 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
344 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
345 a8024000-a8027000 rw-p 00000000 00:00 0
346 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
347 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
348 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
349 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
350 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
352 where "address" is the address space in the process that it occupies, "perms"
353 is a set of permissions:
359 p = private (copy on write)
361 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
362 "inode" is the inode on that device. 0 indicates that no inode is associated
363 with the memory region, as the case would be with BSS (uninitialized data).
364 The "pathname" shows the name associated file for this mapping. If the mapping
365 is not associated with a file:
367 [heap] = the heap of the program
368 [stack] = the stack of the main process
369 [stack:1001] = the stack of the thread with tid 1001
370 [vdso] = the "virtual dynamic shared object",
371 the kernel system call handler
373 or if empty, the mapping is anonymous.
375 The /proc/PID/task/TID/maps is a view of the virtual memory from the viewpoint
376 of the individual tasks of a process. In this file you will see a mapping marked
377 as [stack] if that task sees it as a stack. This is a key difference from the
378 content of /proc/PID/maps, where you will see all mappings that are being used
379 as stack by all of those tasks. Hence, for the example above, the task-level
380 map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this:
382 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
383 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
384 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
385 a7cb1000-a7cb2000 ---p 00000000 00:00 0
386 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
387 a7eb2000-a7eb3000 ---p 00000000 00:00 0
388 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack]
389 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
390 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
391 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
392 a800b000-a800e000 rw-p 00000000 00:00 0
393 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
394 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
395 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
396 a8024000-a8027000 rw-p 00000000 00:00 0
397 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
398 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
399 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
400 aff35000-aff4a000 rw-p 00000000 00:00 0
401 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
403 The /proc/PID/smaps is an extension based on maps, showing the memory
404 consumption for each of the process's mappings. For each of mappings there
405 is a series of lines such as the following:
407 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
421 VmFlags: rd ex mr mw me de
423 the first of these lines shows the same information as is displayed for the
424 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
425 (size), the amount of the mapping that is currently resident in RAM (RSS), the
426 process' proportional share of this mapping (PSS), the number of clean and
427 dirty private pages in the mapping. Note that even a page which is part of a
428 MAP_SHARED mapping, but has only a single pte mapped, i.e. is currently used
429 by only one process, is accounted as private and not as shared. "Referenced"
430 indicates the amount of memory currently marked as referenced or accessed.
431 "Anonymous" shows the amount of memory that does not belong to any file. Even
432 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
433 and a page is modified, the file page is replaced by a private anonymous copy.
434 "Swap" shows how much would-be-anonymous memory is also used, but out on
437 "VmFlags" field deserves a separate description. This member represents the kernel
438 flags associated with the particular virtual memory area in two letter encoded
439 manner. The codes are the following:
448 gd - stack segment growns down
450 dw - disabled write to the mapped file
451 lo - pages are locked in memory
452 io - memory mapped I/O area
453 sr - sequential read advise provided
454 rr - random read advise provided
455 dc - do not copy area on fork
456 de - do not expand area on remapping
457 ac - area is accountable
458 nr - swap space is not reserved for the area
459 ht - area uses huge tlb pages
460 nl - non-linear mapping
461 ar - architecture specific flag
462 dd - do not include area into core dump
464 hg - huge page advise flag
465 nh - no-huge page advise flag
466 mg - mergable advise flag
468 Note that there is no guarantee that every flag and associated mnemonic will
469 be present in all further kernel releases. Things get changed, the flags may
470 be vanished or the reverse -- new added.
472 This file is only present if the CONFIG_MMU kernel configuration option is
475 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
476 bits on both physical and virtual pages associated with a process.
477 To clear the bits for all the pages associated with the process
478 > echo 1 > /proc/PID/clear_refs
480 To clear the bits for the anonymous pages associated with the process
481 > echo 2 > /proc/PID/clear_refs
483 To clear the bits for the file mapped pages associated with the process
484 > echo 3 > /proc/PID/clear_refs
485 Any other value written to /proc/PID/clear_refs will have no effect.
487 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
488 using /proc/kpageflags and number of times a page is mapped using
489 /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
494 Similar to the process entries, the kernel data files give information about
495 the running kernel. The files used to obtain this information are contained in
496 /proc and are listed in Table 1-5. Not all of these will be present in your
497 system. It depends on the kernel configuration and the loaded modules, which
498 files are there, and which are missing.
500 Table 1-5: Kernel info in /proc
501 ..............................................................................
503 apm Advanced power management info
504 buddyinfo Kernel memory allocator information (see text) (2.5)
505 bus Directory containing bus specific information
506 cmdline Kernel command line
507 cpuinfo Info about the CPU
508 devices Available devices (block and character)
509 dma Used DMS channels
510 filesystems Supported filesystems
511 driver Various drivers grouped here, currently rtc (2.4)
512 execdomains Execdomains, related to security (2.4)
513 fb Frame Buffer devices (2.4)
514 fs File system parameters, currently nfs/exports (2.4)
515 ide Directory containing info about the IDE subsystem
516 interrupts Interrupt usage
517 iomem Memory map (2.4)
518 ioports I/O port usage
519 irq Masks for irq to cpu affinity (2.4)(smp?)
520 isapnp ISA PnP (Plug&Play) Info (2.4)
521 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
523 ksyms Kernel symbol table
524 loadavg Load average of last 1, 5 & 15 minutes
528 modules List of loaded modules
529 mounts Mounted filesystems
530 net Networking info (see text)
531 pagetypeinfo Additional page allocator information (see text) (2.5)
532 partitions Table of partitions known to the system
533 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
534 decoupled by lspci (2.4)
536 scsi SCSI info (see text)
537 slabinfo Slab pool info
538 softirqs softirq usage
539 stat Overall statistics
540 swaps Swap space utilization
542 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
543 tty Info of tty drivers
545 version Kernel version
546 video bttv info of video resources (2.4)
547 vmallocinfo Show vmalloced areas
548 ..............................................................................
550 You can, for example, check which interrupts are currently in use and what
551 they are used for by looking in the file /proc/interrupts:
553 > cat /proc/interrupts
555 0: 8728810 XT-PIC timer
556 1: 895 XT-PIC keyboard
558 3: 531695 XT-PIC aha152x
559 4: 2014133 XT-PIC serial
560 5: 44401 XT-PIC pcnet_cs
563 12: 182918 XT-PIC PS/2 Mouse
565 14: 1232265 XT-PIC ide0
569 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
570 output of a SMP machine):
572 > cat /proc/interrupts
575 0: 1243498 1214548 IO-APIC-edge timer
576 1: 8949 8958 IO-APIC-edge keyboard
577 2: 0 0 XT-PIC cascade
578 5: 11286 10161 IO-APIC-edge soundblaster
579 8: 1 0 IO-APIC-edge rtc
580 9: 27422 27407 IO-APIC-edge 3c503
581 12: 113645 113873 IO-APIC-edge PS/2 Mouse
583 14: 22491 24012 IO-APIC-edge ide0
584 15: 2183 2415 IO-APIC-edge ide1
585 17: 30564 30414 IO-APIC-level eth0
586 18: 177 164 IO-APIC-level bttv
591 NMI is incremented in this case because every timer interrupt generates a NMI
592 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
594 LOC is the local interrupt counter of the internal APIC of every CPU.
596 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
597 connects the CPUs in a SMP system. This means that an error has been detected,
598 the IO-APIC automatically retry the transmission, so it should not be a big
599 problem, but you should read the SMP-FAQ.
601 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
602 /proc/interrupts to display every IRQ vector in use by the system, not
603 just those considered 'most important'. The new vectors are:
605 THR -- interrupt raised when a machine check threshold counter
606 (typically counting ECC corrected errors of memory or cache) exceeds
607 a configurable threshold. Only available on some systems.
609 TRM -- a thermal event interrupt occurs when a temperature threshold
610 has been exceeded for the CPU. This interrupt may also be generated
611 when the temperature drops back to normal.
613 SPU -- a spurious interrupt is some interrupt that was raised then lowered
614 by some IO device before it could be fully processed by the APIC. Hence
615 the APIC sees the interrupt but does not know what device it came from.
616 For this case the APIC will generate the interrupt with a IRQ vector
617 of 0xff. This might also be generated by chipset bugs.
619 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
620 sent from one CPU to another per the needs of the OS. Typically,
621 their statistics are used by kernel developers and interested users to
622 determine the occurrence of interrupts of the given type.
624 The above IRQ vectors are displayed only when relevant. For example,
625 the threshold vector does not exist on x86_64 platforms. Others are
626 suppressed when the system is a uniprocessor. As of this writing, only
627 i386 and x86_64 platforms support the new IRQ vector displays.
629 Of some interest is the introduction of the /proc/irq directory to 2.4.
630 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
631 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
632 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
637 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
638 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
642 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
643 IRQ, you can set it by doing:
645 > echo 1 > /proc/irq/10/smp_affinity
647 This means that only the first CPU will handle the IRQ, but you can also echo
648 5 which means that only the first and fourth CPU can handle the IRQ.
650 The contents of each smp_affinity file is the same by default:
652 > cat /proc/irq/0/smp_affinity
655 There is an alternate interface, smp_affinity_list which allows specifying
656 a cpu range instead of a bitmask:
658 > cat /proc/irq/0/smp_affinity_list
661 The default_smp_affinity mask applies to all non-active IRQs, which are the
662 IRQs which have not yet been allocated/activated, and hence which lack a
663 /proc/irq/[0-9]* directory.
665 The node file on an SMP system shows the node to which the device using the IRQ
666 reports itself as being attached. This hardware locality information does not
667 include information about any possible driver locality preference.
669 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
670 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
672 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
673 between all the CPUs which are allowed to handle it. As usual the kernel has
674 more info than you and does a better job than you, so the defaults are the
675 best choice for almost everyone. [Note this applies only to those IO-APIC's
676 that support "Round Robin" interrupt distribution.]
678 There are three more important subdirectories in /proc: net, scsi, and sys.
679 The general rule is that the contents, or even the existence of these
680 directories, depend on your kernel configuration. If SCSI is not enabled, the
681 directory scsi may not exist. The same is true with the net, which is there
682 only when networking support is present in the running kernel.
684 The slabinfo file gives information about memory usage at the slab level.
685 Linux uses slab pools for memory management above page level in version 2.2.
686 Commonly used objects have their own slab pool (such as network buffers,
687 directory cache, and so on).
689 ..............................................................................
691 > cat /proc/buddyinfo
693 Node 0, zone DMA 0 4 5 4 4 3 ...
694 Node 0, zone Normal 1 0 0 1 101 8 ...
695 Node 0, zone HighMem 2 0 0 1 1 0 ...
697 External fragmentation is a problem under some workloads, and buddyinfo is a
698 useful tool for helping diagnose these problems. Buddyinfo will give you a
699 clue as to how big an area you can safely allocate, or why a previous
702 Each column represents the number of pages of a certain order which are
703 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
704 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
705 available in ZONE_NORMAL, etc...
707 More information relevant to external fragmentation can be found in
710 > cat /proc/pagetypeinfo
714 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
715 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
716 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
717 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
718 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
719 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
720 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
721 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
722 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
723 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
724 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
726 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
727 Node 0, zone DMA 2 0 5 1 0
728 Node 0, zone DMA32 41 6 967 2 0
730 Fragmentation avoidance in the kernel works by grouping pages of different
731 migrate types into the same contiguous regions of memory called page blocks.
732 A page block is typically the size of the default hugepage size e.g. 2MB on
733 X86-64. By keeping pages grouped based on their ability to move, the kernel
734 can reclaim pages within a page block to satisfy a high-order allocation.
736 The pagetypinfo begins with information on the size of a page block. It
737 then gives the same type of information as buddyinfo except broken down
738 by migrate-type and finishes with details on how many page blocks of each
741 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
742 from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can
743 make an estimate of the likely number of huge pages that can be allocated
744 at a given point in time. All the "Movable" blocks should be allocatable
745 unless memory has been mlock()'d. Some of the Reclaimable blocks should
746 also be allocatable although a lot of filesystem metadata may have to be
747 reclaimed to achieve this.
749 ..............................................................................
753 Provides information about distribution and utilization of memory. This
754 varies by architecture and compile options. The following is from a
755 16GB PIII, which has highmem enabled. You may not have all of these fields.
759 The "Locked" indicates whether the mapping is locked in memory or not.
762 MemTotal: 16344972 kB
769 HighTotal: 15597528 kB
770 HighFree: 13629632 kB
780 SReclaimable: 159856 kB
781 SUnreclaim: 124508 kB
786 CommitLimit: 7669796 kB
787 Committed_AS: 100056 kB
788 VmallocTotal: 112216 kB
790 VmallocChunk: 111088 kB
791 AnonHugePages: 49152 kB
793 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
794 bits and the kernel binary code)
795 MemFree: The sum of LowFree+HighFree
796 Buffers: Relatively temporary storage for raw disk blocks
797 shouldn't get tremendously large (20MB or so)
798 Cached: in-memory cache for files read from the disk (the
799 pagecache). Doesn't include SwapCached
800 SwapCached: Memory that once was swapped out, is swapped back in but
801 still also is in the swapfile (if memory is needed it
802 doesn't need to be swapped out AGAIN because it is already
803 in the swapfile. This saves I/O)
804 Active: Memory that has been used more recently and usually not
805 reclaimed unless absolutely necessary.
806 Inactive: Memory which has been less recently used. It is more
807 eligible to be reclaimed for other purposes
809 HighFree: Highmem is all memory above ~860MB of physical memory
810 Highmem areas are for use by userspace programs, or
811 for the pagecache. The kernel must use tricks to access
812 this memory, making it slower to access than lowmem.
814 LowFree: Lowmem is memory which can be used for everything that
815 highmem can be used for, but it is also available for the
816 kernel's use for its own data structures. Among many
817 other things, it is where everything from the Slab is
818 allocated. Bad things happen when you're out of lowmem.
819 SwapTotal: total amount of swap space available
820 SwapFree: Memory which has been evicted from RAM, and is temporarily
822 Dirty: Memory which is waiting to get written back to the disk
823 Writeback: Memory which is actively being written back to the disk
824 AnonPages: Non-file backed pages mapped into userspace page tables
825 AnonHugePages: Non-file backed huge pages mapped into userspace page tables
826 Mapped: files which have been mmaped, such as libraries
827 Slab: in-kernel data structures cache
828 SReclaimable: Part of Slab, that might be reclaimed, such as caches
829 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
830 PageTables: amount of memory dedicated to the lowest level of page
832 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
834 Bounce: Memory used for block device "bounce buffers"
835 WritebackTmp: Memory used by FUSE for temporary writeback buffers
836 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
837 this is the total amount of memory currently available to
838 be allocated on the system. This limit is only adhered to
839 if strict overcommit accounting is enabled (mode 2 in
840 'vm.overcommit_memory').
841 The CommitLimit is calculated with the following formula:
842 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
843 For example, on a system with 1G of physical RAM and 7G
844 of swap with a `vm.overcommit_ratio` of 30 it would
845 yield a CommitLimit of 7.3G.
846 For more details, see the memory overcommit documentation
847 in vm/overcommit-accounting.
848 Committed_AS: The amount of memory presently allocated on the system.
849 The committed memory is a sum of all of the memory which
850 has been allocated by processes, even if it has not been
851 "used" by them as of yet. A process which malloc()'s 1G
852 of memory, but only touches 300M of it will only show up
853 as using 300M of memory even if it has the address space
854 allocated for the entire 1G. This 1G is memory which has
855 been "committed" to by the VM and can be used at any time
856 by the allocating application. With strict overcommit
857 enabled on the system (mode 2 in 'vm.overcommit_memory'),
858 allocations which would exceed the CommitLimit (detailed
859 above) will not be permitted. This is useful if one needs
860 to guarantee that processes will not fail due to lack of
861 memory once that memory has been successfully allocated.
862 VmallocTotal: total size of vmalloc memory area
863 VmallocUsed: amount of vmalloc area which is used
864 VmallocChunk: largest contiguous block of vmalloc area which is free
866 ..............................................................................
870 Provides information about vmalloced/vmaped areas. One line per area,
871 containing the virtual address range of the area, size in bytes,
872 caller information of the creator, and optional information depending
873 on the kind of area :
875 pages=nr number of pages
876 phys=addr if a physical address was specified
877 ioremap I/O mapping (ioremap() and friends)
878 vmalloc vmalloc() area
881 vpages buffer for pages pointers was vmalloced (huge area)
882 N<node>=nr (Only on NUMA kernels)
883 Number of pages allocated on memory node <node>
885 > cat /proc/vmallocinfo
886 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
887 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
888 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
889 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
890 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
891 phys=7fee8000 ioremap
892 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
893 phys=7fee7000 ioremap
894 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
895 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
896 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
897 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
899 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
900 /0x130 [x_tables] pages=4 vmalloc N0=4
901 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
902 pages=14 vmalloc N2=14
903 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
905 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
907 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
908 pages=10 vmalloc N0=10
910 ..............................................................................
914 Provides counts of softirq handlers serviced since boot time, for each cpu.
919 TIMER: 27166 27120 27097 27034
924 SCHED: 27035 26983 26971 26746
926 RCU: 1678 1769 2178 2250
929 1.3 IDE devices in /proc/ide
930 ----------------------------
932 The subdirectory /proc/ide contains information about all IDE devices of which
933 the kernel is aware. There is one subdirectory for each IDE controller, the
934 file drivers and a link for each IDE device, pointing to the device directory
935 in the controller specific subtree.
937 The file drivers contains general information about the drivers used for the
940 > cat /proc/ide/drivers
941 ide-cdrom version 4.53
942 ide-disk version 1.08
944 More detailed information can be found in the controller specific
945 subdirectories. These are named ide0, ide1 and so on. Each of these
946 directories contains the files shown in table 1-6.
949 Table 1-6: IDE controller info in /proc/ide/ide?
950 ..............................................................................
952 channel IDE channel (0 or 1)
953 config Configuration (only for PCI/IDE bridge)
955 model Type/Chipset of IDE controller
956 ..............................................................................
958 Each device connected to a controller has a separate subdirectory in the
959 controllers directory. The files listed in table 1-7 are contained in these
963 Table 1-7: IDE device information
964 ..............................................................................
967 capacity Capacity of the medium (in 512Byte blocks)
968 driver driver and version
969 geometry physical and logical geometry
970 identify device identify block
972 model device identifier
973 settings device setup
974 smart_thresholds IDE disk management thresholds
975 smart_values IDE disk management values
976 ..............................................................................
978 The most interesting file is settings. This file contains a nice overview of
979 the drive parameters:
981 # cat /proc/ide/ide0/hda/settings
982 name value min max mode
983 ---- ----- --- --- ----
984 bios_cyl 526 0 65535 rw
985 bios_head 255 0 255 rw
987 breada_readahead 4 0 127 rw
989 file_readahead 72 0 2097151 rw
991 keepsettings 0 0 1 rw
992 max_kb_per_request 122 1 127 rw
996 pio_mode write-only 0 255 w
1002 1.4 Networking info in /proc/net
1003 --------------------------------
1005 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1006 additional values you get for IP version 6 if you configure the kernel to
1007 support this. Table 1-9 lists the files and their meaning.
1010 Table 1-8: IPv6 info in /proc/net
1011 ..............................................................................
1013 udp6 UDP sockets (IPv6)
1014 tcp6 TCP sockets (IPv6)
1015 raw6 Raw device statistics (IPv6)
1016 igmp6 IP multicast addresses, which this host joined (IPv6)
1017 if_inet6 List of IPv6 interface addresses
1018 ipv6_route Kernel routing table for IPv6
1019 rt6_stats Global IPv6 routing tables statistics
1020 sockstat6 Socket statistics (IPv6)
1021 snmp6 Snmp data (IPv6)
1022 ..............................................................................
1025 Table 1-9: Network info in /proc/net
1026 ..............................................................................
1028 arp Kernel ARP table
1029 dev network devices with statistics
1030 dev_mcast the Layer2 multicast groups a device is listening too
1031 (interface index, label, number of references, number of bound
1033 dev_stat network device status
1034 ip_fwchains Firewall chain linkage
1035 ip_fwnames Firewall chain names
1036 ip_masq Directory containing the masquerading tables
1037 ip_masquerade Major masquerading table
1038 netstat Network statistics
1039 raw raw device statistics
1040 route Kernel routing table
1041 rpc Directory containing rpc info
1042 rt_cache Routing cache
1044 sockstat Socket statistics
1047 unix UNIX domain sockets
1048 wireless Wireless interface data (Wavelan etc)
1049 igmp IP multicast addresses, which this host joined
1050 psched Global packet scheduler parameters.
1051 netlink List of PF_NETLINK sockets
1052 ip_mr_vifs List of multicast virtual interfaces
1053 ip_mr_cache List of multicast routing cache
1054 ..............................................................................
1056 You can use this information to see which network devices are available in
1057 your system and how much traffic was routed over those devices:
1060 Inter-|Receive |[...
1061 face |bytes packets errs drop fifo frame compressed multicast|[...
1062 lo: 908188 5596 0 0 0 0 0 0 [...
1063 ppp0:15475140 20721 410 0 0 410 0 0 [...
1064 eth0: 614530 7085 0 0 0 0 0 1 [...
1067 ...] bytes packets errs drop fifo colls carrier compressed
1068 ...] 908188 5596 0 0 0 0 0 0
1069 ...] 1375103 17405 0 0 0 0 0 0
1070 ...] 1703981 5535 0 0 0 3 0 0
1072 In addition, each Channel Bond interface has its own directory. For
1073 example, the bond0 device will have a directory called /proc/net/bond0/.
1074 It will contain information that is specific to that bond, such as the
1075 current slaves of the bond, the link status of the slaves, and how
1076 many times the slaves link has failed.
1081 If you have a SCSI host adapter in your system, you'll find a subdirectory
1082 named after the driver for this adapter in /proc/scsi. You'll also see a list
1083 of all recognized SCSI devices in /proc/scsi:
1085 >cat /proc/scsi/scsi
1087 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1088 Vendor: IBM Model: DGHS09U Rev: 03E0
1089 Type: Direct-Access ANSI SCSI revision: 03
1090 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1091 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1092 Type: CD-ROM ANSI SCSI revision: 02
1095 The directory named after the driver has one file for each adapter found in
1096 the system. These files contain information about the controller, including
1097 the used IRQ and the IO address range. The amount of information shown is
1098 dependent on the adapter you use. The example shows the output for an Adaptec
1099 AHA-2940 SCSI adapter:
1101 > cat /proc/scsi/aic7xxx/0
1103 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1105 TCQ Enabled By Default : Disabled
1106 AIC7XXX_PROC_STATS : Disabled
1107 AIC7XXX_RESET_DELAY : 5
1108 Adapter Configuration:
1109 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1110 Ultra Wide Controller
1111 PCI MMAPed I/O Base: 0xeb001000
1112 Adapter SEEPROM Config: SEEPROM found and used.
1113 Adaptec SCSI BIOS: Enabled
1115 SCBs: Active 0, Max Active 2,
1116 Allocated 15, HW 16, Page 255
1118 BIOS Control Word: 0x18b6
1119 Adapter Control Word: 0x005b
1120 Extended Translation: Enabled
1121 Disconnect Enable Flags: 0xffff
1122 Ultra Enable Flags: 0x0001
1123 Tag Queue Enable Flags: 0x0000
1124 Ordered Queue Tag Flags: 0x0000
1125 Default Tag Queue Depth: 8
1126 Tagged Queue By Device array for aic7xxx host instance 0:
1127 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1128 Actual queue depth per device for aic7xxx host instance 0:
1129 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1132 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1133 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1134 Total transfers 160151 (74577 reads and 85574 writes)
1136 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1137 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1138 Total transfers 0 (0 reads and 0 writes)
1141 1.6 Parallel port info in /proc/parport
1142 ---------------------------------------
1144 The directory /proc/parport contains information about the parallel ports of
1145 your system. It has one subdirectory for each port, named after the port
1148 These directories contain the four files shown in Table 1-10.
1151 Table 1-10: Files in /proc/parport
1152 ..............................................................................
1154 autoprobe Any IEEE-1284 device ID information that has been acquired.
1155 devices list of the device drivers using that port. A + will appear by the
1156 name of the device currently using the port (it might not appear
1158 hardware Parallel port's base address, IRQ line and DMA channel.
1159 irq IRQ that parport is using for that port. This is in a separate
1160 file to allow you to alter it by writing a new value in (IRQ
1162 ..............................................................................
1164 1.7 TTY info in /proc/tty
1165 -------------------------
1167 Information about the available and actually used tty's can be found in the
1168 directory /proc/tty.You'll find entries for drivers and line disciplines in
1169 this directory, as shown in Table 1-11.
1172 Table 1-11: Files in /proc/tty
1173 ..............................................................................
1175 drivers list of drivers and their usage
1176 ldiscs registered line disciplines
1177 driver/serial usage statistic and status of single tty lines
1178 ..............................................................................
1180 To see which tty's are currently in use, you can simply look into the file
1183 > cat /proc/tty/drivers
1184 pty_slave /dev/pts 136 0-255 pty:slave
1185 pty_master /dev/ptm 128 0-255 pty:master
1186 pty_slave /dev/ttyp 3 0-255 pty:slave
1187 pty_master /dev/pty 2 0-255 pty:master
1188 serial /dev/cua 5 64-67 serial:callout
1189 serial /dev/ttyS 4 64-67 serial
1190 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1191 /dev/ptmx /dev/ptmx 5 2 system
1192 /dev/console /dev/console 5 1 system:console
1193 /dev/tty /dev/tty 5 0 system:/dev/tty
1194 unknown /dev/tty 4 1-63 console
1197 1.8 Miscellaneous kernel statistics in /proc/stat
1198 -------------------------------------------------
1200 Various pieces of information about kernel activity are available in the
1201 /proc/stat file. All of the numbers reported in this file are aggregates
1202 since the system first booted. For a quick look, simply cat the file:
1205 cpu 2255 34 2290 22625563 6290 127 456 0 0
1206 cpu0 1132 34 1441 11311718 3675 127 438 0 0
1207 cpu1 1123 0 849 11313845 2614 0 18 0 0
1208 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1214 softirq 183433 0 21755 12 39 1137 231 21459 2263
1216 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1217 lines. These numbers identify the amount of time the CPU has spent performing
1218 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1219 second). The meanings of the columns are as follows, from left to right:
1221 - user: normal processes executing in user mode
1222 - nice: niced processes executing in user mode
1223 - system: processes executing in kernel mode
1224 - idle: twiddling thumbs
1225 - iowait: waiting for I/O to complete
1226 - irq: servicing interrupts
1227 - softirq: servicing softirqs
1228 - steal: involuntary wait
1229 - guest: running a normal guest
1230 - guest_nice: running a niced guest
1232 The "intr" line gives counts of interrupts serviced since boot time, for each
1233 of the possible system interrupts. The first column is the total of all
1234 interrupts serviced; each subsequent column is the total for that particular
1237 The "ctxt" line gives the total number of context switches across all CPUs.
1239 The "btime" line gives the time at which the system booted, in seconds since
1242 The "processes" line gives the number of processes and threads created, which
1243 includes (but is not limited to) those created by calls to the fork() and
1244 clone() system calls.
1246 The "procs_running" line gives the total number of threads that are
1247 running or ready to run (i.e., the total number of runnable threads).
1249 The "procs_blocked" line gives the number of processes currently blocked,
1250 waiting for I/O to complete.
1252 The "softirq" line gives counts of softirqs serviced since boot time, for each
1253 of the possible system softirqs. The first column is the total of all
1254 softirqs serviced; each subsequent column is the total for that particular
1258 1.9 Ext4 file system parameters
1259 ------------------------------
1261 Information about mounted ext4 file systems can be found in
1262 /proc/fs/ext4. Each mounted filesystem will have a directory in
1263 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1264 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1265 in Table 1-12, below.
1267 Table 1-12: Files in /proc/fs/ext4/<devname>
1268 ..............................................................................
1270 mb_groups details of multiblock allocator buddy cache of free blocks
1271 ..............................................................................
1275 Shows registered system console lines.
1277 To see which character device lines are currently used for the system console
1278 /dev/console, you may simply look into the file /proc/consoles:
1280 > cat /proc/consoles
1286 device name of the device
1287 operations R = can do read operations
1288 W = can do write operations
1290 flags E = it is enabled
1291 C = it is preferred console
1292 B = it is primary boot console
1293 p = it is used for printk buffer
1294 b = it is not a TTY but a Braille device
1295 a = it is safe to use when cpu is offline
1296 major:minor major and minor number of the device separated by a colon
1298 ------------------------------------------------------------------------------
1300 ------------------------------------------------------------------------------
1301 The /proc file system serves information about the running system. It not only
1302 allows access to process data but also allows you to request the kernel status
1303 by reading files in the hierarchy.
1305 The directory structure of /proc reflects the types of information and makes
1306 it easy, if not obvious, where to look for specific data.
1307 ------------------------------------------------------------------------------
1309 ------------------------------------------------------------------------------
1310 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1311 ------------------------------------------------------------------------------
1313 ------------------------------------------------------------------------------
1315 ------------------------------------------------------------------------------
1316 * Modifying kernel parameters by writing into files found in /proc/sys
1317 * Exploring the files which modify certain parameters
1318 * Review of the /proc/sys file tree
1319 ------------------------------------------------------------------------------
1322 A very interesting part of /proc is the directory /proc/sys. This is not only
1323 a source of information, it also allows you to change parameters within the
1324 kernel. Be very careful when attempting this. You can optimize your system,
1325 but you can also cause it to crash. Never alter kernel parameters on a
1326 production system. Set up a development machine and test to make sure that
1327 everything works the way you want it to. You may have no alternative but to
1328 reboot the machine once an error has been made.
1330 To change a value, simply echo the new value into the file. An example is
1331 given below in the section on the file system data. You need to be root to do
1332 this. You can create your own boot script to perform this every time your
1335 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1336 general things in the operation of the Linux kernel. Since some of the files
1337 can inadvertently disrupt your system, it is advisable to read both
1338 documentation and source before actually making adjustments. In any case, be
1339 very careful when writing to any of these files. The entries in /proc may
1340 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1341 review the kernel documentation in the directory /usr/src/linux/Documentation.
1342 This chapter is heavily based on the documentation included in the pre 2.2
1343 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1345 Please see: Documentation/sysctl/ directory for descriptions of these
1348 ------------------------------------------------------------------------------
1350 ------------------------------------------------------------------------------
1351 Certain aspects of kernel behavior can be modified at runtime, without the
1352 need to recompile the kernel, or even to reboot the system. The files in the
1353 /proc/sys tree can not only be read, but also modified. You can use the echo
1354 command to write value into these files, thereby changing the default settings
1356 ------------------------------------------------------------------------------
1358 ------------------------------------------------------------------------------
1359 CHAPTER 3: PER-PROCESS PARAMETERS
1360 ------------------------------------------------------------------------------
1362 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1363 --------------------------------------------------------------------------------
1365 These file can be used to adjust the badness heuristic used to select which
1366 process gets killed in out of memory conditions.
1368 The badness heuristic assigns a value to each candidate task ranging from 0
1369 (never kill) to 1000 (always kill) to determine which process is targeted. The
1370 units are roughly a proportion along that range of allowed memory the process
1371 may allocate from based on an estimation of its current memory and swap use.
1372 For example, if a task is using all allowed memory, its badness score will be
1373 1000. If it is using half of its allowed memory, its score will be 500.
1375 There is an additional factor included in the badness score: the current memory
1376 and swap usage is discounted by 3% for root processes.
1378 The amount of "allowed" memory depends on the context in which the oom killer
1379 was called. If it is due to the memory assigned to the allocating task's cpuset
1380 being exhausted, the allowed memory represents the set of mems assigned to that
1381 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1382 memory represents the set of mempolicy nodes. If it is due to a memory
1383 limit (or swap limit) being reached, the allowed memory is that configured
1384 limit. Finally, if it is due to the entire system being out of memory, the
1385 allowed memory represents all allocatable resources.
1387 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1388 is used to determine which task to kill. Acceptable values range from -1000
1389 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1390 polarize the preference for oom killing either by always preferring a certain
1391 task or completely disabling it. The lowest possible value, -1000, is
1392 equivalent to disabling oom killing entirely for that task since it will always
1393 report a badness score of 0.
1395 Consequently, it is very simple for userspace to define the amount of memory to
1396 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1397 example, is roughly equivalent to allowing the remainder of tasks sharing the
1398 same system, cpuset, mempolicy, or memory controller resources to use at least
1399 50% more memory. A value of -500, on the other hand, would be roughly
1400 equivalent to discounting 50% of the task's allowed memory from being considered
1401 as scoring against the task.
1403 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1404 be used to tune the badness score. Its acceptable values range from -16
1405 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1406 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1407 scaled linearly with /proc/<pid>/oom_score_adj.
1409 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1410 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1411 requires CAP_SYS_RESOURCE.
1413 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1414 generation children with separate address spaces instead, if possible. This
1415 avoids servers and important system daemons from being killed and loses the
1416 minimal amount of work.
1419 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1420 -------------------------------------------------------------
1422 This file can be used to check the current score used by the oom-killer is for
1423 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1424 process should be killed in an out-of-memory situation.
1427 3.3 /proc/<pid>/io - Display the IO accounting fields
1428 -------------------------------------------------------
1430 This file contains IO statistics for each running process
1435 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1438 test:/tmp # cat /proc/3828/io
1444 write_bytes: 323932160
1445 cancelled_write_bytes: 0
1454 I/O counter: chars read
1455 The number of bytes which this task has caused to be read from storage. This
1456 is simply the sum of bytes which this process passed to read() and pread().
1457 It includes things like tty IO and it is unaffected by whether or not actual
1458 physical disk IO was required (the read might have been satisfied from
1465 I/O counter: chars written
1466 The number of bytes which this task has caused, or shall cause to be written
1467 to disk. Similar caveats apply here as with rchar.
1473 I/O counter: read syscalls
1474 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1481 I/O counter: write syscalls
1482 Attempt to count the number of write I/O operations, i.e. syscalls like
1483 write() and pwrite().
1489 I/O counter: bytes read
1490 Attempt to count the number of bytes which this process really did cause to
1491 be fetched from the storage layer. Done at the submit_bio() level, so it is
1492 accurate for block-backed filesystems. <please add status regarding NFS and
1493 CIFS at a later time>
1499 I/O counter: bytes written
1500 Attempt to count the number of bytes which this process caused to be sent to
1501 the storage layer. This is done at page-dirtying time.
1504 cancelled_write_bytes
1505 ---------------------
1507 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1508 then deletes the file, it will in fact perform no writeout. But it will have
1509 been accounted as having caused 1MB of write.
1510 In other words: The number of bytes which this process caused to not happen,
1511 by truncating pagecache. A task can cause "negative" IO too. If this task
1512 truncates some dirty pagecache, some IO which another task has been accounted
1513 for (in its write_bytes) will not be happening. We _could_ just subtract that
1514 from the truncating task's write_bytes, but there is information loss in doing
1521 At its current implementation state, this is a bit racy on 32-bit machines: if
1522 process A reads process B's /proc/pid/io while process B is updating one of
1523 those 64-bit counters, process A could see an intermediate result.
1526 More information about this can be found within the taskstats documentation in
1527 Documentation/accounting.
1529 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1530 ---------------------------------------------------------------
1531 When a process is dumped, all anonymous memory is written to a core file as
1532 long as the size of the core file isn't limited. But sometimes we don't want
1533 to dump some memory segments, for example, huge shared memory. Conversely,
1534 sometimes we want to save file-backed memory segments into a core file, not
1535 only the individual files.
1537 /proc/<pid>/coredump_filter allows you to customize which memory segments
1538 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1539 of memory types. If a bit of the bitmask is set, memory segments of the
1540 corresponding memory type are dumped, otherwise they are not dumped.
1542 The following 7 memory types are supported:
1543 - (bit 0) anonymous private memory
1544 - (bit 1) anonymous shared memory
1545 - (bit 2) file-backed private memory
1546 - (bit 3) file-backed shared memory
1547 - (bit 4) ELF header pages in file-backed private memory areas (it is
1548 effective only if the bit 2 is cleared)
1549 - (bit 5) hugetlb private memory
1550 - (bit 6) hugetlb shared memory
1552 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1553 are always dumped regardless of the bitmask status.
1555 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
1556 effected by bit 5-6.
1558 Default value of coredump_filter is 0x23; this means all anonymous memory
1559 segments and hugetlb private memory are dumped.
1561 If you don't want to dump all shared memory segments attached to pid 1234,
1562 write 0x21 to the process's proc file.
1564 $ echo 0x21 > /proc/1234/coredump_filter
1566 When a new process is created, the process inherits the bitmask status from its
1567 parent. It is useful to set up coredump_filter before the program runs.
1570 $ echo 0x7 > /proc/self/coredump_filter
1573 3.5 /proc/<pid>/mountinfo - Information about mounts
1574 --------------------------------------------------------
1576 This file contains lines of the form:
1578 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1579 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1581 (1) mount ID: unique identifier of the mount (may be reused after umount)
1582 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1583 (3) major:minor: value of st_dev for files on filesystem
1584 (4) root: root of the mount within the filesystem
1585 (5) mount point: mount point relative to the process's root
1586 (6) mount options: per mount options
1587 (7) optional fields: zero or more fields of the form "tag[:value]"
1588 (8) separator: marks the end of the optional fields
1589 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1590 (10) mount source: filesystem specific information or "none"
1591 (11) super options: per super block options
1593 Parsers should ignore all unrecognised optional fields. Currently the
1594 possible optional fields are:
1596 shared:X mount is shared in peer group X
1597 master:X mount is slave to peer group X
1598 propagate_from:X mount is slave and receives propagation from peer group X (*)
1599 unbindable mount is unbindable
1601 (*) X is the closest dominant peer group under the process's root. If
1602 X is the immediate master of the mount, or if there's no dominant peer
1603 group under the same root, then only the "master:X" field is present
1604 and not the "propagate_from:X" field.
1606 For more information on mount propagation see:
1608 Documentation/filesystems/sharedsubtree.txt
1611 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1612 --------------------------------------------------------
1613 These files provide a method to access a tasks comm value. It also allows for
1614 a task to set its own or one of its thread siblings comm value. The comm value
1615 is limited in size compared to the cmdline value, so writing anything longer
1616 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1620 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1621 -------------------------------------------------------------------------
1622 This file provides a fast way to retrieve first level children pids
1623 of a task pointed by <pid>/<tid> pair. The format is a space separated
1626 Note the "first level" here -- if a child has own children they will
1627 not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1628 to obtain the descendants.
1630 Since this interface is intended to be fast and cheap it doesn't
1631 guarantee to provide precise results and some children might be
1632 skipped, especially if they've exited right after we printed their
1633 pids, so one need to either stop or freeze processes being inspected
1634 if precise results are needed.
1637 3.7 /proc/<pid>/fdinfo/<fd> - Information about opened file
1638 ---------------------------------------------------------------
1639 This file provides information associated with an opened file. The regular
1640 files have at least two fields -- 'pos' and 'flags'. The 'pos' represents
1641 the current offset of the opened file in decimal form [see lseek(2) for
1642 details] and 'flags' denotes the octal O_xxx mask the file has been
1643 created with [see open(2) for details].
1650 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1651 pair provide additional information particular to the objects they represent.
1659 where 'eventfd-count' is hex value of a counter.
1665 sigmask: 0000000000000200
1667 where 'sigmask' is hex value of the signal mask associated
1674 tfd: 5 events: 1d data: ffffffffffffffff
1676 where 'tfd' is a target file descriptor number in decimal form,
1677 'events' is events mask being watched and the 'data' is data
1678 associated with a target [see epoll(7) for more details].
1682 For inotify files the format is the following
1686 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1688 where 'wd' is a watch descriptor in decimal form, ie a target file
1689 descriptor number, 'ino' and 'sdev' are inode and device where the
1690 target file resides and the 'mask' is the mask of events, all in hex
1691 form [see inotify(7) for more details].
1693 If the kernel was built with exportfs support, the path to the target
1694 file is encoded as a file handle. The file handle is provided by three
1695 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1698 If the kernel is built without exportfs support the file handle won't be
1701 If there is no inotify mark attached yet the 'inotify' line will be omitted.
1703 For fanotify files the format is
1707 fanotify flags:10 event-flags:0
1708 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1709 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
1711 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
1712 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
1713 flags associated with mark which are tracked separately from events
1714 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
1715 mask and 'ignored_mask' is the mask of events which are to be ignored.
1716 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
1717 does provide information about flags and mask used in fanotify_mark
1718 call [see fsnotify manpage for details].
1720 While the first three lines are mandatory and always printed, the rest is
1721 optional and may be omitted if no marks created yet.
1724 ------------------------------------------------------------------------------
1726 ------------------------------------------------------------------------------
1729 ---------------------
1731 The following mount options are supported:
1733 hidepid= Set /proc/<pid>/ access mode.
1734 gid= Set the group authorized to learn processes information.
1736 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1739 hidepid=1 means users may not access any /proc/<pid>/ directories but their
1740 own. Sensitive files like cmdline, sched*, status are now protected against
1741 other users. This makes it impossible to learn whether any user runs
1742 specific program (given the program doesn't reveal itself by its behaviour).
1743 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1744 poorly written programs passing sensitive information via program arguments are
1745 now protected against local eavesdroppers.
1747 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1748 users. It doesn't mean that it hides a fact whether a process with a specific
1749 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1750 but it hides process' uid and gid, which may be learned by stat()'ing
1751 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
1752 information about running processes, whether some daemon runs with elevated
1753 privileges, whether other user runs some sensitive program, whether other users
1754 run any program at all, etc.
1756 gid= defines a group authorized to learn processes information otherwise
1757 prohibited by hidepid=. If you use some daemon like identd which needs to learn
1758 information about processes information, just add identd to this group.
projects / firefly-linux-kernel-4.4.55.git / blob