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
45 ------------------------------------------------------------------------------
47 ------------------------------------------------------------------------------
49 0.1 Introduction/Credits
50 ------------------------
52 This documentation is part of a soon (or so we hope) to be released book on
53 the SuSE Linux distribution. As there is no complete documentation for the
54 /proc file system and we've used many freely available sources to write these
55 chapters, it seems only fair to give the work back to the Linux community.
56 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
57 afraid it's still far from complete, but we hope it will be useful. As far as
58 we know, it is the first 'all-in-one' document about the /proc file system. It
59 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
60 SPARC, AXP, etc., features, you probably won't find what you are looking for.
61 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
62 additions and patches are welcome and will be added to this document if you
65 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
66 other people for help compiling this documentation. We'd also like to extend a
67 special thank you to Andi Kleen for documentation, which we relied on heavily
68 to create this document, as well as the additional information he provided.
69 Thanks to everybody else who contributed source or docs to the Linux kernel
70 and helped create a great piece of software... :)
72 If you have any comments, corrections or additions, please don't hesitate to
73 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
76 The latest version of this document is available online at
77 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
79 If the above direction does not works for you, you could try the kernel
80 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
81 comandante@zaralinux.com.
86 We don't guarantee the correctness of this document, and if you come to us
87 complaining about how you screwed up your system because of incorrect
88 documentation, we won't feel responsible...
90 ------------------------------------------------------------------------------
91 CHAPTER 1: COLLECTING SYSTEM INFORMATION
92 ------------------------------------------------------------------------------
94 ------------------------------------------------------------------------------
96 ------------------------------------------------------------------------------
97 * Investigating the properties of the pseudo file system /proc and its
98 ability to provide information on the running Linux system
99 * Examining /proc's structure
100 * Uncovering various information about the kernel and the processes running
102 ------------------------------------------------------------------------------
105 The proc file system acts as an interface to internal data structures in the
106 kernel. It can be used to obtain information about the system and to change
107 certain kernel parameters at runtime (sysctl).
109 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
110 show you how you can use /proc/sys to change settings.
112 1.1 Process-Specific Subdirectories
113 -----------------------------------
115 The directory /proc contains (among other things) one subdirectory for each
116 process running on the system, which is named after the process ID (PID).
118 The link self points to the process reading the file system. Each process
119 subdirectory has the entries listed in Table 1-1.
122 Table 1-1: Process specific entries in /proc
123 ..............................................................................
125 clear_refs Clears page referenced bits shown in smaps output
126 cmdline Command line arguments
127 cpu Current and last cpu in which it was executed (2.4)(smp)
128 cwd Link to the current working directory
129 environ Values of environment variables
130 exe Link to the executable of this process
131 fd Directory, which contains all file descriptors
132 maps Memory maps to executables and library files (2.4)
133 mem Memory held by this process
134 root Link to the root directory of this process
136 statm Process memory status information
137 status Process status in human readable form
138 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
139 stack Report full stack trace, enable via CONFIG_STACKTRACE
140 smaps a extension based on maps, showing the memory consumption of
142 ..............................................................................
144 For example, to get the status information of a process, all you have to do is
145 read the file /proc/PID/status:
147 >cat /proc/self/status
171 SigPnd: 0000000000000000
172 ShdPnd: 0000000000000000
173 SigBlk: 0000000000000000
174 SigIgn: 0000000000000000
175 SigCgt: 0000000000000000
176 CapInh: 00000000fffffeff
177 CapPrm: 0000000000000000
178 CapEff: 0000000000000000
179 CapBnd: ffffffffffffffff
180 voluntary_ctxt_switches: 0
181 nonvoluntary_ctxt_switches: 1
183 This shows you nearly the same information you would get if you viewed it with
184 the ps command. In fact, ps uses the proc file system to obtain its
185 information. But you get a more detailed view of the process by reading the
186 file /proc/PID/status. It fields are described in table 1-2.
188 The statm file contains more detailed information about the process
189 memory usage. Its seven fields are explained in Table 1-3. The stat file
190 contains details information about the process itself. Its fields are
191 explained in Table 1-4.
193 (for SMP CONFIG users)
194 For making accounting scalable, RSS related information are handled in
195 asynchronous manner and the vaule may not be very precise. To see a precise
196 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
197 It's slow but very precise.
199 Table 1-2: Contents of the status files (as of 2.6.30-rc7)
200 ..............................................................................
202 Name filename of the executable
203 State state (R is running, S is sleeping, D is sleeping
204 in an uninterruptible wait, Z is zombie,
205 T is traced or stopped)
208 PPid process id of the parent process
209 TracerPid PID of process tracing this process (0 if not)
210 Uid Real, effective, saved set, and file system UIDs
211 Gid Real, effective, saved set, and file system GIDs
212 FDSize number of file descriptor slots currently allocated
213 Groups supplementary group list
214 VmPeak peak virtual memory size
215 VmSize total program size
216 VmLck locked memory size
217 VmHWM peak resident set size ("high water mark")
218 VmRSS size of memory portions
219 VmData size of data, stack, and text segments
220 VmStk size of data, stack, and text segments
221 VmExe size of text segment
222 VmLib size of shared library code
223 VmPTE size of page table entries
224 VmSwap size of swap usage (the number of referred swapents)
225 Threads number of threads
226 SigQ number of signals queued/max. number for queue
227 SigPnd bitmap of pending signals for the thread
228 ShdPnd bitmap of shared pending signals for the process
229 SigBlk bitmap of blocked signals
230 SigIgn bitmap of ignored signals
231 SigCgt bitmap of catched signals
232 CapInh bitmap of inheritable capabilities
233 CapPrm bitmap of permitted capabilities
234 CapEff bitmap of effective capabilities
235 CapBnd bitmap of capabilities bounding set
236 Cpus_allowed mask of CPUs on which this process may run
237 Cpus_allowed_list Same as previous, but in "list format"
238 Mems_allowed mask of memory nodes allowed to this process
239 Mems_allowed_list Same as previous, but in "list format"
240 voluntary_ctxt_switches number of voluntary context switches
241 nonvoluntary_ctxt_switches number of non voluntary context switches
242 ..............................................................................
244 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
245 ..............................................................................
247 size total program size (pages) (same as VmSize in status)
248 resident size of memory portions (pages) (same as VmRSS in status)
249 shared number of pages that are shared (i.e. backed by a file)
250 trs number of pages that are 'code' (not including libs; broken,
251 includes data segment)
252 lrs number of pages of library (always 0 on 2.6)
253 drs number of pages of data/stack (including libs; broken,
254 includes library text)
255 dt number of dirty pages (always 0 on 2.6)
256 ..............................................................................
259 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
260 ..............................................................................
263 tcomm filename of the executable
264 state state (R is running, S is sleeping, D is sleeping in an
265 uninterruptible wait, Z is zombie, T is traced or stopped)
266 ppid process id of the parent process
267 pgrp pgrp of the process
269 tty_nr tty the process uses
270 tty_pgrp pgrp of the tty
272 min_flt number of minor faults
273 cmin_flt number of minor faults with child's
274 maj_flt number of major faults
275 cmaj_flt number of major faults with child's
276 utime user mode jiffies
277 stime kernel mode jiffies
278 cutime user mode jiffies with child's
279 cstime kernel mode jiffies with child's
280 priority priority level
282 num_threads number of threads
283 it_real_value (obsolete, always 0)
284 start_time time the process started after system boot
285 vsize virtual memory size
286 rss resident set memory size
287 rsslim current limit in bytes on the rss
288 start_code address above which program text can run
289 end_code address below which program text can run
290 start_stack address of the start of the stack
291 esp current value of ESP
292 eip current value of EIP
293 pending bitmap of pending signals
294 blocked bitmap of blocked signals
295 sigign bitmap of ignored signals
296 sigcatch bitmap of catched signals
297 wchan address where process went to sleep
300 exit_signal signal to send to parent thread on exit
301 task_cpu which CPU the task is scheduled on
302 rt_priority realtime priority
303 policy scheduling policy (man sched_setscheduler)
304 blkio_ticks time spent waiting for block IO
305 gtime guest time of the task in jiffies
306 cgtime guest time of the task children in jiffies
307 ..............................................................................
309 The /proc/PID/maps file containing the currently mapped memory regions and
310 their access permissions.
314 address perms offset dev inode pathname
316 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
317 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
318 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
319 a7cb1000-a7cb2000 ---p 00000000 00:00 0
320 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
321 a7eb2000-a7eb3000 ---p 00000000 00:00 0
322 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
323 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
324 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
325 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
326 a800b000-a800e000 rw-p 00000000 00:00 0
327 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
328 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
329 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
330 a8024000-a8027000 rw-p 00000000 00:00 0
331 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
332 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
333 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
334 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
335 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
337 where "address" is the address space in the process that it occupies, "perms"
338 is a set of permissions:
344 p = private (copy on write)
346 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
347 "inode" is the inode on that device. 0 indicates that no inode is associated
348 with the memory region, as the case would be with BSS (uninitialized data).
349 The "pathname" shows the name associated file for this mapping. If the mapping
350 is not associated with a file:
352 [heap] = the heap of the program
353 [stack] = the stack of the main process
354 [vdso] = the "virtual dynamic shared object",
355 the kernel system call handler
357 or if empty, the mapping is anonymous.
360 The /proc/PID/smaps is an extension based on maps, showing the memory
361 consumption for each of the process's mappings. For each of mappings there
362 is a series of lines such as the following:
364 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
378 The first of these lines shows the same information as is displayed for the
379 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
380 (size), the amount of the mapping that is currently resident in RAM (RSS), the
381 process' proportional share of this mapping (PSS), the number of clean and
382 dirty private pages in the mapping. Note that even a page which is part of a
383 MAP_SHARED mapping, but has only a single pte mapped, i.e. is currently used
384 by only one process, is accounted as private and not as shared. "Referenced"
385 indicates the amount of memory currently marked as referenced or accessed.
386 "Anonymous" shows the amount of memory that does not belong to any file. Even
387 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
388 and a page is modified, the file page is replaced by a private anonymous copy.
389 "Swap" shows how much would-be-anonymous memory is also used, but out on
392 This file is only present if the CONFIG_MMU kernel configuration option is
395 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
396 bits on both physical and virtual pages associated with a process.
397 To clear the bits for all the pages associated with the process
398 > echo 1 > /proc/PID/clear_refs
400 To clear the bits for the anonymous pages associated with the process
401 > echo 2 > /proc/PID/clear_refs
403 To clear the bits for the file mapped pages associated with the process
404 > echo 3 > /proc/PID/clear_refs
405 Any other value written to /proc/PID/clear_refs will have no effect.
411 Similar to the process entries, the kernel data files give information about
412 the running kernel. The files used to obtain this information are contained in
413 /proc and are listed in Table 1-5. Not all of these will be present in your
414 system. It depends on the kernel configuration and the loaded modules, which
415 files are there, and which are missing.
417 Table 1-5: Kernel info in /proc
418 ..............................................................................
420 apm Advanced power management info
421 buddyinfo Kernel memory allocator information (see text) (2.5)
422 bus Directory containing bus specific information
423 cmdline Kernel command line
424 cpuinfo Info about the CPU
425 devices Available devices (block and character)
426 dma Used DMS channels
427 filesystems Supported filesystems
428 driver Various drivers grouped here, currently rtc (2.4)
429 execdomains Execdomains, related to security (2.4)
430 fb Frame Buffer devices (2.4)
431 fs File system parameters, currently nfs/exports (2.4)
432 ide Directory containing info about the IDE subsystem
433 interrupts Interrupt usage
434 iomem Memory map (2.4)
435 ioports I/O port usage
436 irq Masks for irq to cpu affinity (2.4)(smp?)
437 isapnp ISA PnP (Plug&Play) Info (2.4)
438 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
440 ksyms Kernel symbol table
441 loadavg Load average of last 1, 5 & 15 minutes
445 modules List of loaded modules
446 mounts Mounted filesystems
447 net Networking info (see text)
448 pagetypeinfo Additional page allocator information (see text) (2.5)
449 partitions Table of partitions known to the system
450 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
451 decoupled by lspci (2.4)
453 scsi SCSI info (see text)
454 slabinfo Slab pool info
455 softirqs softirq usage
456 stat Overall statistics
457 swaps Swap space utilization
459 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
460 tty Info of tty drivers
462 version Kernel version
463 video bttv info of video resources (2.4)
464 vmallocinfo Show vmalloced areas
465 ..............................................................................
467 You can, for example, check which interrupts are currently in use and what
468 they are used for by looking in the file /proc/interrupts:
470 > cat /proc/interrupts
472 0: 8728810 XT-PIC timer
473 1: 895 XT-PIC keyboard
475 3: 531695 XT-PIC aha152x
476 4: 2014133 XT-PIC serial
477 5: 44401 XT-PIC pcnet_cs
480 12: 182918 XT-PIC PS/2 Mouse
482 14: 1232265 XT-PIC ide0
486 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
487 output of a SMP machine):
489 > cat /proc/interrupts
492 0: 1243498 1214548 IO-APIC-edge timer
493 1: 8949 8958 IO-APIC-edge keyboard
494 2: 0 0 XT-PIC cascade
495 5: 11286 10161 IO-APIC-edge soundblaster
496 8: 1 0 IO-APIC-edge rtc
497 9: 27422 27407 IO-APIC-edge 3c503
498 12: 113645 113873 IO-APIC-edge PS/2 Mouse
500 14: 22491 24012 IO-APIC-edge ide0
501 15: 2183 2415 IO-APIC-edge ide1
502 17: 30564 30414 IO-APIC-level eth0
503 18: 177 164 IO-APIC-level bttv
508 NMI is incremented in this case because every timer interrupt generates a NMI
509 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
511 LOC is the local interrupt counter of the internal APIC of every CPU.
513 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
514 connects the CPUs in a SMP system. This means that an error has been detected,
515 the IO-APIC automatically retry the transmission, so it should not be a big
516 problem, but you should read the SMP-FAQ.
518 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
519 /proc/interrupts to display every IRQ vector in use by the system, not
520 just those considered 'most important'. The new vectors are:
522 THR -- interrupt raised when a machine check threshold counter
523 (typically counting ECC corrected errors of memory or cache) exceeds
524 a configurable threshold. Only available on some systems.
526 TRM -- a thermal event interrupt occurs when a temperature threshold
527 has been exceeded for the CPU. This interrupt may also be generated
528 when the temperature drops back to normal.
530 SPU -- a spurious interrupt is some interrupt that was raised then lowered
531 by some IO device before it could be fully processed by the APIC. Hence
532 the APIC sees the interrupt but does not know what device it came from.
533 For this case the APIC will generate the interrupt with a IRQ vector
534 of 0xff. This might also be generated by chipset bugs.
536 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
537 sent from one CPU to another per the needs of the OS. Typically,
538 their statistics are used by kernel developers and interested users to
539 determine the occurrence of interrupts of the given type.
541 The above IRQ vectors are displayed only when relevent. For example,
542 the threshold vector does not exist on x86_64 platforms. Others are
543 suppressed when the system is a uniprocessor. As of this writing, only
544 i386 and x86_64 platforms support the new IRQ vector displays.
546 Of some interest is the introduction of the /proc/irq directory to 2.4.
547 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
548 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
549 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
554 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
555 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
559 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
560 IRQ, you can set it by doing:
562 > echo 1 > /proc/irq/10/smp_affinity
564 This means that only the first CPU will handle the IRQ, but you can also echo
565 5 which means that only the first and fourth CPU can handle the IRQ.
567 The contents of each smp_affinity file is the same by default:
569 > cat /proc/irq/0/smp_affinity
572 The default_smp_affinity mask applies to all non-active IRQs, which are the
573 IRQs which have not yet been allocated/activated, and hence which lack a
574 /proc/irq/[0-9]* directory.
576 The node file on an SMP system shows the node to which the device using the IRQ
577 reports itself as being attached. This hardware locality information does not
578 include information about any possible driver locality preference.
580 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
581 profiler. Default value is ffffffff (all cpus).
583 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
584 between all the CPUs which are allowed to handle it. As usual the kernel has
585 more info than you and does a better job than you, so the defaults are the
586 best choice for almost everyone.
588 There are three more important subdirectories in /proc: net, scsi, and sys.
589 The general rule is that the contents, or even the existence of these
590 directories, depend on your kernel configuration. If SCSI is not enabled, the
591 directory scsi may not exist. The same is true with the net, which is there
592 only when networking support is present in the running kernel.
594 The slabinfo file gives information about memory usage at the slab level.
595 Linux uses slab pools for memory management above page level in version 2.2.
596 Commonly used objects have their own slab pool (such as network buffers,
597 directory cache, and so on).
599 ..............................................................................
601 > cat /proc/buddyinfo
603 Node 0, zone DMA 0 4 5 4 4 3 ...
604 Node 0, zone Normal 1 0 0 1 101 8 ...
605 Node 0, zone HighMem 2 0 0 1 1 0 ...
607 External fragmentation is a problem under some workloads, and buddyinfo is a
608 useful tool for helping diagnose these problems. Buddyinfo will give you a
609 clue as to how big an area you can safely allocate, or why a previous
612 Each column represents the number of pages of a certain order which are
613 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
614 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
615 available in ZONE_NORMAL, etc...
617 More information relevant to external fragmentation can be found in
620 > cat /proc/pagetypeinfo
624 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
625 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
626 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
627 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
628 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
629 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
630 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
631 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
632 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
633 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
634 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
636 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
637 Node 0, zone DMA 2 0 5 1 0
638 Node 0, zone DMA32 41 6 967 2 0
640 Fragmentation avoidance in the kernel works by grouping pages of different
641 migrate types into the same contiguous regions of memory called page blocks.
642 A page block is typically the size of the default hugepage size e.g. 2MB on
643 X86-64. By keeping pages grouped based on their ability to move, the kernel
644 can reclaim pages within a page block to satisfy a high-order allocation.
646 The pagetypinfo begins with information on the size of a page block. It
647 then gives the same type of information as buddyinfo except broken down
648 by migrate-type and finishes with details on how many page blocks of each
651 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
652 from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can
653 make an estimate of the likely number of huge pages that can be allocated
654 at a given point in time. All the "Movable" blocks should be allocatable
655 unless memory has been mlock()'d. Some of the Reclaimable blocks should
656 also be allocatable although a lot of filesystem metadata may have to be
657 reclaimed to achieve this.
659 ..............................................................................
663 Provides information about distribution and utilization of memory. This
664 varies by architecture and compile options. The following is from a
665 16GB PIII, which has highmem enabled. You may not have all of these fields.
670 MemTotal: 16344972 kB
677 HighTotal: 15597528 kB
678 HighFree: 13629632 kB
688 SReclaimable: 159856 kB
689 SUnreclaim: 124508 kB
694 CommitLimit: 7669796 kB
695 Committed_AS: 100056 kB
696 VmallocTotal: 112216 kB
698 VmallocChunk: 111088 kB
700 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
701 bits and the kernel binary code)
702 MemFree: The sum of LowFree+HighFree
703 Buffers: Relatively temporary storage for raw disk blocks
704 shouldn't get tremendously large (20MB or so)
705 Cached: in-memory cache for files read from the disk (the
706 pagecache). Doesn't include SwapCached
707 SwapCached: Memory that once was swapped out, is swapped back in but
708 still also is in the swapfile (if memory is needed it
709 doesn't need to be swapped out AGAIN because it is already
710 in the swapfile. This saves I/O)
711 Active: Memory that has been used more recently and usually not
712 reclaimed unless absolutely necessary.
713 Inactive: Memory which has been less recently used. It is more
714 eligible to be reclaimed for other purposes
716 HighFree: Highmem is all memory above ~860MB of physical memory
717 Highmem areas are for use by userspace programs, or
718 for the pagecache. The kernel must use tricks to access
719 this memory, making it slower to access than lowmem.
721 LowFree: Lowmem is memory which can be used for everything that
722 highmem can be used for, but it is also available for the
723 kernel's use for its own data structures. Among many
724 other things, it is where everything from the Slab is
725 allocated. Bad things happen when you're out of lowmem.
726 SwapTotal: total amount of swap space available
727 SwapFree: Memory which has been evicted from RAM, and is temporarily
729 Dirty: Memory which is waiting to get written back to the disk
730 Writeback: Memory which is actively being written back to the disk
731 AnonPages: Non-file backed pages mapped into userspace page tables
732 Mapped: files which have been mmaped, such as libraries
733 Slab: in-kernel data structures cache
734 SReclaimable: Part of Slab, that might be reclaimed, such as caches
735 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
736 PageTables: amount of memory dedicated to the lowest level of page
738 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
740 Bounce: Memory used for block device "bounce buffers"
741 WritebackTmp: Memory used by FUSE for temporary writeback buffers
742 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
743 this is the total amount of memory currently available to
744 be allocated on the system. This limit is only adhered to
745 if strict overcommit accounting is enabled (mode 2 in
746 'vm.overcommit_memory').
747 The CommitLimit is calculated with the following formula:
748 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
749 For example, on a system with 1G of physical RAM and 7G
750 of swap with a `vm.overcommit_ratio` of 30 it would
751 yield a CommitLimit of 7.3G.
752 For more details, see the memory overcommit documentation
753 in vm/overcommit-accounting.
754 Committed_AS: The amount of memory presently allocated on the system.
755 The committed memory is a sum of all of the memory which
756 has been allocated by processes, even if it has not been
757 "used" by them as of yet. A process which malloc()'s 1G
758 of memory, but only touches 300M of it will only show up
759 as using 300M of memory even if it has the address space
760 allocated for the entire 1G. This 1G is memory which has
761 been "committed" to by the VM and can be used at any time
762 by the allocating application. With strict overcommit
763 enabled on the system (mode 2 in 'vm.overcommit_memory'),
764 allocations which would exceed the CommitLimit (detailed
765 above) will not be permitted. This is useful if one needs
766 to guarantee that processes will not fail due to lack of
767 memory once that memory has been successfully allocated.
768 VmallocTotal: total size of vmalloc memory area
769 VmallocUsed: amount of vmalloc area which is used
770 VmallocChunk: largest contiguous block of vmalloc area which is free
772 ..............................................................................
776 Provides information about vmalloced/vmaped areas. One line per area,
777 containing the virtual address range of the area, size in bytes,
778 caller information of the creator, and optional information depending
779 on the kind of area :
781 pages=nr number of pages
782 phys=addr if a physical address was specified
783 ioremap I/O mapping (ioremap() and friends)
784 vmalloc vmalloc() area
787 vpages buffer for pages pointers was vmalloced (huge area)
788 N<node>=nr (Only on NUMA kernels)
789 Number of pages allocated on memory node <node>
791 > cat /proc/vmallocinfo
792 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
793 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
794 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
795 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
796 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
797 phys=7fee8000 ioremap
798 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
799 phys=7fee7000 ioremap
800 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
801 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
802 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
803 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
805 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
806 /0x130 [x_tables] pages=4 vmalloc N0=4
807 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
808 pages=14 vmalloc N2=14
809 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
811 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
813 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
814 pages=10 vmalloc N0=10
816 ..............................................................................
820 Provides counts of softirq handlers serviced since boot time, for each cpu.
825 TIMER: 27166 27120 27097 27034
830 SCHED: 27035 26983 26971 26746
832 RCU: 1678 1769 2178 2250
835 1.3 IDE devices in /proc/ide
836 ----------------------------
838 The subdirectory /proc/ide contains information about all IDE devices of which
839 the kernel is aware. There is one subdirectory for each IDE controller, the
840 file drivers and a link for each IDE device, pointing to the device directory
841 in the controller specific subtree.
843 The file drivers contains general information about the drivers used for the
846 > cat /proc/ide/drivers
847 ide-cdrom version 4.53
848 ide-disk version 1.08
850 More detailed information can be found in the controller specific
851 subdirectories. These are named ide0, ide1 and so on. Each of these
852 directories contains the files shown in table 1-6.
855 Table 1-6: IDE controller info in /proc/ide/ide?
856 ..............................................................................
858 channel IDE channel (0 or 1)
859 config Configuration (only for PCI/IDE bridge)
861 model Type/Chipset of IDE controller
862 ..............................................................................
864 Each device connected to a controller has a separate subdirectory in the
865 controllers directory. The files listed in table 1-7 are contained in these
869 Table 1-7: IDE device information
870 ..............................................................................
873 capacity Capacity of the medium (in 512Byte blocks)
874 driver driver and version
875 geometry physical and logical geometry
876 identify device identify block
878 model device identifier
879 settings device setup
880 smart_thresholds IDE disk management thresholds
881 smart_values IDE disk management values
882 ..............................................................................
884 The most interesting file is settings. This file contains a nice overview of
885 the drive parameters:
887 # cat /proc/ide/ide0/hda/settings
888 name value min max mode
889 ---- ----- --- --- ----
890 bios_cyl 526 0 65535 rw
891 bios_head 255 0 255 rw
893 breada_readahead 4 0 127 rw
895 file_readahead 72 0 2097151 rw
897 keepsettings 0 0 1 rw
898 max_kb_per_request 122 1 127 rw
902 pio_mode write-only 0 255 w
908 1.4 Networking info in /proc/net
909 --------------------------------
911 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
912 additional values you get for IP version 6 if you configure the kernel to
913 support this. Table 1-9 lists the files and their meaning.
916 Table 1-8: IPv6 info in /proc/net
917 ..............................................................................
919 udp6 UDP sockets (IPv6)
920 tcp6 TCP sockets (IPv6)
921 raw6 Raw device statistics (IPv6)
922 igmp6 IP multicast addresses, which this host joined (IPv6)
923 if_inet6 List of IPv6 interface addresses
924 ipv6_route Kernel routing table for IPv6
925 rt6_stats Global IPv6 routing tables statistics
926 sockstat6 Socket statistics (IPv6)
927 snmp6 Snmp data (IPv6)
928 ..............................................................................
931 Table 1-9: Network info in /proc/net
932 ..............................................................................
935 dev network devices with statistics
936 dev_mcast the Layer2 multicast groups a device is listening too
937 (interface index, label, number of references, number of bound
939 dev_stat network device status
940 ip_fwchains Firewall chain linkage
941 ip_fwnames Firewall chain names
942 ip_masq Directory containing the masquerading tables
943 ip_masquerade Major masquerading table
944 netstat Network statistics
945 raw raw device statistics
946 route Kernel routing table
947 rpc Directory containing rpc info
948 rt_cache Routing cache
950 sockstat Socket statistics
952 tr_rif Token ring RIF routing table
954 unix UNIX domain sockets
955 wireless Wireless interface data (Wavelan etc)
956 igmp IP multicast addresses, which this host joined
957 psched Global packet scheduler parameters.
958 netlink List of PF_NETLINK sockets
959 ip_mr_vifs List of multicast virtual interfaces
960 ip_mr_cache List of multicast routing cache
961 ..............................................................................
963 You can use this information to see which network devices are available in
964 your system and how much traffic was routed over those devices:
968 face |bytes packets errs drop fifo frame compressed multicast|[...
969 lo: 908188 5596 0 0 0 0 0 0 [...
970 ppp0:15475140 20721 410 0 0 410 0 0 [...
971 eth0: 614530 7085 0 0 0 0 0 1 [...
974 ...] bytes packets errs drop fifo colls carrier compressed
975 ...] 908188 5596 0 0 0 0 0 0
976 ...] 1375103 17405 0 0 0 0 0 0
977 ...] 1703981 5535 0 0 0 3 0 0
979 In addition, each Channel Bond interface has its own directory. For
980 example, the bond0 device will have a directory called /proc/net/bond0/.
981 It will contain information that is specific to that bond, such as the
982 current slaves of the bond, the link status of the slaves, and how
983 many times the slaves link has failed.
988 If you have a SCSI host adapter in your system, you'll find a subdirectory
989 named after the driver for this adapter in /proc/scsi. You'll also see a list
990 of all recognized SCSI devices in /proc/scsi:
994 Host: scsi0 Channel: 00 Id: 00 Lun: 00
995 Vendor: IBM Model: DGHS09U Rev: 03E0
996 Type: Direct-Access ANSI SCSI revision: 03
997 Host: scsi0 Channel: 00 Id: 06 Lun: 00
998 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
999 Type: CD-ROM ANSI SCSI revision: 02
1002 The directory named after the driver has one file for each adapter found in
1003 the system. These files contain information about the controller, including
1004 the used IRQ and the IO address range. The amount of information shown is
1005 dependent on the adapter you use. The example shows the output for an Adaptec
1006 AHA-2940 SCSI adapter:
1008 > cat /proc/scsi/aic7xxx/0
1010 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1012 TCQ Enabled By Default : Disabled
1013 AIC7XXX_PROC_STATS : Disabled
1014 AIC7XXX_RESET_DELAY : 5
1015 Adapter Configuration:
1016 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1017 Ultra Wide Controller
1018 PCI MMAPed I/O Base: 0xeb001000
1019 Adapter SEEPROM Config: SEEPROM found and used.
1020 Adaptec SCSI BIOS: Enabled
1022 SCBs: Active 0, Max Active 2,
1023 Allocated 15, HW 16, Page 255
1025 BIOS Control Word: 0x18b6
1026 Adapter Control Word: 0x005b
1027 Extended Translation: Enabled
1028 Disconnect Enable Flags: 0xffff
1029 Ultra Enable Flags: 0x0001
1030 Tag Queue Enable Flags: 0x0000
1031 Ordered Queue Tag Flags: 0x0000
1032 Default Tag Queue Depth: 8
1033 Tagged Queue By Device array for aic7xxx host instance 0:
1034 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1035 Actual queue depth per device for aic7xxx host instance 0:
1036 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1039 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1040 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1041 Total transfers 160151 (74577 reads and 85574 writes)
1043 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1044 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1045 Total transfers 0 (0 reads and 0 writes)
1048 1.6 Parallel port info in /proc/parport
1049 ---------------------------------------
1051 The directory /proc/parport contains information about the parallel ports of
1052 your system. It has one subdirectory for each port, named after the port
1055 These directories contain the four files shown in Table 1-10.
1058 Table 1-10: Files in /proc/parport
1059 ..............................................................................
1061 autoprobe Any IEEE-1284 device ID information that has been acquired.
1062 devices list of the device drivers using that port. A + will appear by the
1063 name of the device currently using the port (it might not appear
1065 hardware Parallel port's base address, IRQ line and DMA channel.
1066 irq IRQ that parport is using for that port. This is in a separate
1067 file to allow you to alter it by writing a new value in (IRQ
1069 ..............................................................................
1071 1.7 TTY info in /proc/tty
1072 -------------------------
1074 Information about the available and actually used tty's can be found in the
1075 directory /proc/tty.You'll find entries for drivers and line disciplines in
1076 this directory, as shown in Table 1-11.
1079 Table 1-11: Files in /proc/tty
1080 ..............................................................................
1082 drivers list of drivers and their usage
1083 ldiscs registered line disciplines
1084 driver/serial usage statistic and status of single tty lines
1085 ..............................................................................
1087 To see which tty's are currently in use, you can simply look into the file
1090 > cat /proc/tty/drivers
1091 pty_slave /dev/pts 136 0-255 pty:slave
1092 pty_master /dev/ptm 128 0-255 pty:master
1093 pty_slave /dev/ttyp 3 0-255 pty:slave
1094 pty_master /dev/pty 2 0-255 pty:master
1095 serial /dev/cua 5 64-67 serial:callout
1096 serial /dev/ttyS 4 64-67 serial
1097 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1098 /dev/ptmx /dev/ptmx 5 2 system
1099 /dev/console /dev/console 5 1 system:console
1100 /dev/tty /dev/tty 5 0 system:/dev/tty
1101 unknown /dev/tty 4 1-63 console
1104 1.8 Miscellaneous kernel statistics in /proc/stat
1105 -------------------------------------------------
1107 Various pieces of information about kernel activity are available in the
1108 /proc/stat file. All of the numbers reported in this file are aggregates
1109 since the system first booted. For a quick look, simply cat the file:
1112 cpu 2255 34 2290 22625563 6290 127 456 0 0
1113 cpu0 1132 34 1441 11311718 3675 127 438 0 0
1114 cpu1 1123 0 849 11313845 2614 0 18 0 0
1115 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1121 softirq 183433 0 21755 12 39 1137 231 21459 2263
1123 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1124 lines. These numbers identify the amount of time the CPU has spent performing
1125 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1126 second). The meanings of the columns are as follows, from left to right:
1128 - user: normal processes executing in user mode
1129 - nice: niced processes executing in user mode
1130 - system: processes executing in kernel mode
1131 - idle: twiddling thumbs
1132 - iowait: waiting for I/O to complete
1133 - irq: servicing interrupts
1134 - softirq: servicing softirqs
1135 - steal: involuntary wait
1136 - guest: running a normal guest
1137 - guest_nice: running a niced guest
1139 The "intr" line gives counts of interrupts serviced since boot time, for each
1140 of the possible system interrupts. The first column is the total of all
1141 interrupts serviced; each subsequent column is the total for that particular
1144 The "ctxt" line gives the total number of context switches across all CPUs.
1146 The "btime" line gives the time at which the system booted, in seconds since
1149 The "processes" line gives the number of processes and threads created, which
1150 includes (but is not limited to) those created by calls to the fork() and
1151 clone() system calls.
1153 The "procs_running" line gives the total number of threads that are
1154 running or ready to run (i.e., the total number of runnable threads).
1156 The "procs_blocked" line gives the number of processes currently blocked,
1157 waiting for I/O to complete.
1159 The "softirq" line gives counts of softirqs serviced since boot time, for each
1160 of the possible system softirqs. The first column is the total of all
1161 softirqs serviced; each subsequent column is the total for that particular
1165 1.9 Ext4 file system parameters
1166 ------------------------------
1168 Information about mounted ext4 file systems can be found in
1169 /proc/fs/ext4. Each mounted filesystem will have a directory in
1170 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1171 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1172 in Table 1-12, below.
1174 Table 1-12: Files in /proc/fs/ext4/<devname>
1175 ..............................................................................
1177 mb_groups details of multiblock allocator buddy cache of free blocks
1178 ..............................................................................
1181 ------------------------------------------------------------------------------
1183 ------------------------------------------------------------------------------
1184 The /proc file system serves information about the running system. It not only
1185 allows access to process data but also allows you to request the kernel status
1186 by reading files in the hierarchy.
1188 The directory structure of /proc reflects the types of information and makes
1189 it easy, if not obvious, where to look for specific data.
1190 ------------------------------------------------------------------------------
1192 ------------------------------------------------------------------------------
1193 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1194 ------------------------------------------------------------------------------
1196 ------------------------------------------------------------------------------
1198 ------------------------------------------------------------------------------
1199 * Modifying kernel parameters by writing into files found in /proc/sys
1200 * Exploring the files which modify certain parameters
1201 * Review of the /proc/sys file tree
1202 ------------------------------------------------------------------------------
1205 A very interesting part of /proc is the directory /proc/sys. This is not only
1206 a source of information, it also allows you to change parameters within the
1207 kernel. Be very careful when attempting this. You can optimize your system,
1208 but you can also cause it to crash. Never alter kernel parameters on a
1209 production system. Set up a development machine and test to make sure that
1210 everything works the way you want it to. You may have no alternative but to
1211 reboot the machine once an error has been made.
1213 To change a value, simply echo the new value into the file. An example is
1214 given below in the section on the file system data. You need to be root to do
1215 this. You can create your own boot script to perform this every time your
1218 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1219 general things in the operation of the Linux kernel. Since some of the files
1220 can inadvertently disrupt your system, it is advisable to read both
1221 documentation and source before actually making adjustments. In any case, be
1222 very careful when writing to any of these files. The entries in /proc may
1223 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1224 review the kernel documentation in the directory /usr/src/linux/Documentation.
1225 This chapter is heavily based on the documentation included in the pre 2.2
1226 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1228 Please see: Documentation/sysctls/ directory for descriptions of these
1231 ------------------------------------------------------------------------------
1233 ------------------------------------------------------------------------------
1234 Certain aspects of kernel behavior can be modified at runtime, without the
1235 need to recompile the kernel, or even to reboot the system. The files in the
1236 /proc/sys tree can not only be read, but also modified. You can use the echo
1237 command to write value into these files, thereby changing the default settings
1239 ------------------------------------------------------------------------------
1241 ------------------------------------------------------------------------------
1242 CHAPTER 3: PER-PROCESS PARAMETERS
1243 ------------------------------------------------------------------------------
1245 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1246 --------------------------------------------------------------------------------
1248 These file can be used to adjust the badness heuristic used to select which
1249 process gets killed in out of memory conditions.
1251 The badness heuristic assigns a value to each candidate task ranging from 0
1252 (never kill) to 1000 (always kill) to determine which process is targeted. The
1253 units are roughly a proportion along that range of allowed memory the process
1254 may allocate from based on an estimation of its current memory and swap use.
1255 For example, if a task is using all allowed memory, its badness score will be
1256 1000. If it is using half of its allowed memory, its score will be 500.
1258 There is an additional factor included in the badness score: root
1259 processes are given 3% extra memory over other tasks.
1261 The amount of "allowed" memory depends on the context in which the oom killer
1262 was called. If it is due to the memory assigned to the allocating task's cpuset
1263 being exhausted, the allowed memory represents the set of mems assigned to that
1264 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1265 memory represents the set of mempolicy nodes. If it is due to a memory
1266 limit (or swap limit) being reached, the allowed memory is that configured
1267 limit. Finally, if it is due to the entire system being out of memory, the
1268 allowed memory represents all allocatable resources.
1270 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1271 is used to determine which task to kill. Acceptable values range from -1000
1272 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1273 polarize the preference for oom killing either by always preferring a certain
1274 task or completely disabling it. The lowest possible value, -1000, is
1275 equivalent to disabling oom killing entirely for that task since it will always
1276 report a badness score of 0.
1278 Consequently, it is very simple for userspace to define the amount of memory to
1279 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1280 example, is roughly equivalent to allowing the remainder of tasks sharing the
1281 same system, cpuset, mempolicy, or memory controller resources to use at least
1282 50% more memory. A value of -500, on the other hand, would be roughly
1283 equivalent to discounting 50% of the task's allowed memory from being considered
1284 as scoring against the task.
1286 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1287 be used to tune the badness score. Its acceptable values range from -16
1288 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1289 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1290 scaled linearly with /proc/<pid>/oom_score_adj.
1292 Writing to /proc/<pid>/oom_score_adj or /proc/<pid>/oom_adj will change the
1293 other with its scaled value.
1295 NOTICE: /proc/<pid>/oom_adj is deprecated and will be removed, please see
1296 Documentation/feature-removal-schedule.txt.
1298 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1299 generation children with seperate address spaces instead, if possible. This
1300 avoids servers and important system daemons from being killed and loses the
1301 minimal amount of work.
1304 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1305 -------------------------------------------------------------
1307 This file can be used to check the current score used by the oom-killer is for
1308 any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
1309 process should be killed in an out-of-memory situation.
1312 3.3 /proc/<pid>/io - Display the IO accounting fields
1313 -------------------------------------------------------
1315 This file contains IO statistics for each running process
1320 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1323 test:/tmp # cat /proc/3828/io
1329 write_bytes: 323932160
1330 cancelled_write_bytes: 0
1339 I/O counter: chars read
1340 The number of bytes which this task has caused to be read from storage. This
1341 is simply the sum of bytes which this process passed to read() and pread().
1342 It includes things like tty IO and it is unaffected by whether or not actual
1343 physical disk IO was required (the read might have been satisfied from
1350 I/O counter: chars written
1351 The number of bytes which this task has caused, or shall cause to be written
1352 to disk. Similar caveats apply here as with rchar.
1358 I/O counter: read syscalls
1359 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1366 I/O counter: write syscalls
1367 Attempt to count the number of write I/O operations, i.e. syscalls like
1368 write() and pwrite().
1374 I/O counter: bytes read
1375 Attempt to count the number of bytes which this process really did cause to
1376 be fetched from the storage layer. Done at the submit_bio() level, so it is
1377 accurate for block-backed filesystems. <please add status regarding NFS and
1378 CIFS at a later time>
1384 I/O counter: bytes written
1385 Attempt to count the number of bytes which this process caused to be sent to
1386 the storage layer. This is done at page-dirtying time.
1389 cancelled_write_bytes
1390 ---------------------
1392 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1393 then deletes the file, it will in fact perform no writeout. But it will have
1394 been accounted as having caused 1MB of write.
1395 In other words: The number of bytes which this process caused to not happen,
1396 by truncating pagecache. A task can cause "negative" IO too. If this task
1397 truncates some dirty pagecache, some IO which another task has been accounted
1398 for (in its write_bytes) will not be happening. We _could_ just subtract that
1399 from the truncating task's write_bytes, but there is information loss in doing
1406 At its current implementation state, this is a bit racy on 32-bit machines: if
1407 process A reads process B's /proc/pid/io while process B is updating one of
1408 those 64-bit counters, process A could see an intermediate result.
1411 More information about this can be found within the taskstats documentation in
1412 Documentation/accounting.
1414 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1415 ---------------------------------------------------------------
1416 When a process is dumped, all anonymous memory is written to a core file as
1417 long as the size of the core file isn't limited. But sometimes we don't want
1418 to dump some memory segments, for example, huge shared memory. Conversely,
1419 sometimes we want to save file-backed memory segments into a core file, not
1420 only the individual files.
1422 /proc/<pid>/coredump_filter allows you to customize which memory segments
1423 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1424 of memory types. If a bit of the bitmask is set, memory segments of the
1425 corresponding memory type are dumped, otherwise they are not dumped.
1427 The following 7 memory types are supported:
1428 - (bit 0) anonymous private memory
1429 - (bit 1) anonymous shared memory
1430 - (bit 2) file-backed private memory
1431 - (bit 3) file-backed shared memory
1432 - (bit 4) ELF header pages in file-backed private memory areas (it is
1433 effective only if the bit 2 is cleared)
1434 - (bit 5) hugetlb private memory
1435 - (bit 6) hugetlb shared memory
1437 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1438 are always dumped regardless of the bitmask status.
1440 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
1441 effected by bit 5-6.
1443 Default value of coredump_filter is 0x23; this means all anonymous memory
1444 segments and hugetlb private memory are dumped.
1446 If you don't want to dump all shared memory segments attached to pid 1234,
1447 write 0x21 to the process's proc file.
1449 $ echo 0x21 > /proc/1234/coredump_filter
1451 When a new process is created, the process inherits the bitmask status from its
1452 parent. It is useful to set up coredump_filter before the program runs.
1455 $ echo 0x7 > /proc/self/coredump_filter
1458 3.5 /proc/<pid>/mountinfo - Information about mounts
1459 --------------------------------------------------------
1461 This file contains lines of the form:
1463 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1464 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1466 (1) mount ID: unique identifier of the mount (may be reused after umount)
1467 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1468 (3) major:minor: value of st_dev for files on filesystem
1469 (4) root: root of the mount within the filesystem
1470 (5) mount point: mount point relative to the process's root
1471 (6) mount options: per mount options
1472 (7) optional fields: zero or more fields of the form "tag[:value]"
1473 (8) separator: marks the end of the optional fields
1474 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1475 (10) mount source: filesystem specific information or "none"
1476 (11) super options: per super block options
1478 Parsers should ignore all unrecognised optional fields. Currently the
1479 possible optional fields are:
1481 shared:X mount is shared in peer group X
1482 master:X mount is slave to peer group X
1483 propagate_from:X mount is slave and receives propagation from peer group X (*)
1484 unbindable mount is unbindable
1486 (*) X is the closest dominant peer group under the process's root. If
1487 X is the immediate master of the mount, or if there's no dominant peer
1488 group under the same root, then only the "master:X" field is present
1489 and not the "propagate_from:X" field.
1491 For more information on mount propagation see:
1493 Documentation/filesystems/sharedsubtree.txt
1496 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1497 --------------------------------------------------------
1498 These files provide a method to access a tasks comm value. It also allows for
1499 a task to set its own or one of its thread siblings comm value. The comm value
1500 is limited in size compared to the cmdline value, so writing anything longer
1501 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated