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 - Adjust the oom-killer score
37 3.2 /proc/<pid>/oom_score - Display current oom-killer score
38 3.3 /proc/<pid>/io - Display the IO accounting fields
39 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
40 3.5 /proc/<pid>/mountinfo - Information about mounts
41 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
44 ------------------------------------------------------------------------------
46 ------------------------------------------------------------------------------
48 0.1 Introduction/Credits
49 ------------------------
51 This documentation is part of a soon (or so we hope) to be released book on
52 the SuSE Linux distribution. As there is no complete documentation for the
53 /proc file system and we've used many freely available sources to write these
54 chapters, it seems only fair to give the work back to the Linux community.
55 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
56 afraid it's still far from complete, but we hope it will be useful. As far as
57 we know, it is the first 'all-in-one' document about the /proc file system. It
58 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
59 SPARC, AXP, etc., features, you probably won't find what you are looking for.
60 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
61 additions and patches are welcome and will be added to this document if you
64 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
65 other people for help compiling this documentation. We'd also like to extend a
66 special thank you to Andi Kleen for documentation, which we relied on heavily
67 to create this document, as well as the additional information he provided.
68 Thanks to everybody else who contributed source or docs to the Linux kernel
69 and helped create a great piece of software... :)
71 If you have any comments, corrections or additions, please don't hesitate to
72 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
75 The latest version of this document is available online at
76 http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
78 If the above direction does not works for you, ypu could try the kernel
79 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
80 comandante@zaralinux.com.
85 We don't guarantee the correctness of this document, and if you come to us
86 complaining about how you screwed up your system because of incorrect
87 documentation, we won't feel responsible...
89 ------------------------------------------------------------------------------
90 CHAPTER 1: COLLECTING SYSTEM INFORMATION
91 ------------------------------------------------------------------------------
93 ------------------------------------------------------------------------------
95 ------------------------------------------------------------------------------
96 * Investigating the properties of the pseudo file system /proc and its
97 ability to provide information on the running Linux system
98 * Examining /proc's structure
99 * Uncovering various information about the kernel and the processes running
101 ------------------------------------------------------------------------------
104 The proc file system acts as an interface to internal data structures in the
105 kernel. It can be used to obtain information about the system and to change
106 certain kernel parameters at runtime (sysctl).
108 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
109 show you how you can use /proc/sys to change settings.
111 1.1 Process-Specific Subdirectories
112 -----------------------------------
114 The directory /proc contains (among other things) one subdirectory for each
115 process running on the system, which is named after the process ID (PID).
117 The link self points to the process reading the file system. Each process
118 subdirectory has the entries listed in Table 1-1.
121 Table 1-1: Process specific entries in /proc
122 ..............................................................................
124 clear_refs Clears page referenced bits shown in smaps output
125 cmdline Command line arguments
126 cpu Current and last cpu in which it was executed (2.4)(smp)
127 cwd Link to the current working directory
128 environ Values of environment variables
129 exe Link to the executable of this process
130 fd Directory, which contains all file descriptors
131 maps Memory maps to executables and library files (2.4)
132 mem Memory held by this process
133 root Link to the root directory of this process
135 statm Process memory status information
136 status Process status in human readable form
137 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
138 stack Report full stack trace, enable via CONFIG_STACKTRACE
139 smaps a extension based on maps, showing the memory consumption of
141 ..............................................................................
143 For example, to get the status information of a process, all you have to do is
144 read the file /proc/PID/status:
146 >cat /proc/self/status
169 SigPnd: 0000000000000000
170 ShdPnd: 0000000000000000
171 SigBlk: 0000000000000000
172 SigIgn: 0000000000000000
173 SigCgt: 0000000000000000
174 CapInh: 00000000fffffeff
175 CapPrm: 0000000000000000
176 CapEff: 0000000000000000
177 CapBnd: ffffffffffffffff
178 voluntary_ctxt_switches: 0
179 nonvoluntary_ctxt_switches: 1
181 This shows you nearly the same information you would get if you viewed it with
182 the ps command. In fact, ps uses the proc file system to obtain its
183 information. But you get a more detailed view of the process by reading the
184 file /proc/PID/status. It fields are described in table 1-2.
186 The statm file contains more detailed information about the process
187 memory usage. Its seven fields are explained in Table 1-3. The stat file
188 contains details information about the process itself. Its fields are
189 explained in Table 1-4.
191 (for SMP CONFIG users)
192 For making accounting scalable, RSS related information are handled in
193 asynchronous manner and the vaule may not be very precise. To see a precise
194 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
195 It's slow but very precise.
197 Table 1-2: Contents of the statm files (as of 2.6.30-rc7)
198 ..............................................................................
200 Name filename of the executable
201 State state (R is running, S is sleeping, D is sleeping
202 in an uninterruptible wait, Z is zombie,
203 T is traced or stopped)
206 PPid process id of the parent process
207 TracerPid PID of process tracing this process (0 if not)
208 Uid Real, effective, saved set, and file system UIDs
209 Gid Real, effective, saved set, and file system GIDs
210 FDSize number of file descriptor slots currently allocated
211 Groups supplementary group list
212 VmPeak peak virtual memory size
213 VmSize total program size
214 VmLck locked memory size
215 VmHWM peak resident set size ("high water mark")
216 VmRSS size of memory portions
217 VmData size of data, stack, and text segments
218 VmStk size of data, stack, and text segments
219 VmExe size of text segment
220 VmLib size of shared library code
221 VmPTE size of page table entries
222 Threads number of threads
223 SigQ number of signals queued/max. number for queue
224 SigPnd bitmap of pending signals for the thread
225 ShdPnd bitmap of shared pending signals for the process
226 SigBlk bitmap of blocked signals
227 SigIgn bitmap of ignored signals
228 SigCgt bitmap of catched signals
229 CapInh bitmap of inheritable capabilities
230 CapPrm bitmap of permitted capabilities
231 CapEff bitmap of effective capabilities
232 CapBnd bitmap of capabilities bounding set
233 Cpus_allowed mask of CPUs on which this process may run
234 Cpus_allowed_list Same as previous, but in "list format"
235 Mems_allowed mask of memory nodes allowed to this process
236 Mems_allowed_list Same as previous, but in "list format"
237 voluntary_ctxt_switches number of voluntary context switches
238 nonvoluntary_ctxt_switches number of non voluntary context switches
239 ..............................................................................
241 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
242 ..............................................................................
244 size total program size (pages) (same as VmSize in status)
245 resident size of memory portions (pages) (same as VmRSS in status)
246 shared number of pages that are shared (i.e. backed by a file)
247 trs number of pages that are 'code' (not including libs; broken,
248 includes data segment)
249 lrs number of pages of library (always 0 on 2.6)
250 drs number of pages of data/stack (including libs; broken,
251 includes library text)
252 dt number of dirty pages (always 0 on 2.6)
253 ..............................................................................
256 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
257 ..............................................................................
260 tcomm filename of the executable
261 state state (R is running, S is sleeping, D is sleeping in an
262 uninterruptible wait, Z is zombie, T is traced or stopped)
263 ppid process id of the parent process
264 pgrp pgrp of the process
266 tty_nr tty the process uses
267 tty_pgrp pgrp of the tty
269 min_flt number of minor faults
270 cmin_flt number of minor faults with child's
271 maj_flt number of major faults
272 cmaj_flt number of major faults with child's
273 utime user mode jiffies
274 stime kernel mode jiffies
275 cutime user mode jiffies with child's
276 cstime kernel mode jiffies with child's
277 priority priority level
279 num_threads number of threads
280 it_real_value (obsolete, always 0)
281 start_time time the process started after system boot
282 vsize virtual memory size
283 rss resident set memory size
284 rsslim current limit in bytes on the rss
285 start_code address above which program text can run
286 end_code address below which program text can run
287 start_stack address of the start of the stack
288 esp current value of ESP
289 eip current value of EIP
290 pending bitmap of pending signals
291 blocked bitmap of blocked signals
292 sigign bitmap of ignored signals
293 sigcatch bitmap of catched signals
294 wchan address where process went to sleep
297 exit_signal signal to send to parent thread on exit
298 task_cpu which CPU the task is scheduled on
299 rt_priority realtime priority
300 policy scheduling policy (man sched_setscheduler)
301 blkio_ticks time spent waiting for block IO
302 gtime guest time of the task in jiffies
303 cgtime guest time of the task children in jiffies
304 ..............................................................................
306 The /proc/PID/map file containing the currently mapped memory regions and
307 their access permissions.
311 address perms offset dev inode pathname
313 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
314 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
315 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
316 a7cb1000-a7cb2000 ---p 00000000 00:00 0
317 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 [threadstack:001ff4b4]
318 a7eb2000-a7eb3000 ---p 00000000 00:00 0
319 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
320 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
321 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
322 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
323 a800b000-a800e000 rw-p 00000000 00:00 0
324 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
325 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
326 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
327 a8024000-a8027000 rw-p 00000000 00:00 0
328 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
329 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
330 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
331 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
332 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
334 where "address" is the address space in the process that it occupies, "perms"
335 is a set of permissions:
341 p = private (copy on write)
343 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
344 "inode" is the inode on that device. 0 indicates that no inode is associated
345 with the memory region, as the case would be with BSS (uninitialized data).
346 The "pathname" shows the name associated file for this mapping. If the mapping
347 is not associated with a file:
349 [heap] = the heap of the program
350 [stack] = the stack of the main process
351 [vdso] = the "virtual dynamic shared object",
352 the kernel system call handler
353 [threadstack:xxxxxxxx] = the stack of the thread, xxxxxxxx is the stack size
355 or if empty, the mapping is anonymous.
358 The /proc/PID/smaps is an extension based on maps, showing the memory
359 consumption for each of the process's mappings. For each of mappings there
360 is a series of lines such as the following:
362 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
375 The first of these lines shows the same information as is displayed for the
376 mapping in /proc/PID/maps. The remaining lines show the size of the mapping,
377 the amount of the mapping that is currently resident in RAM, the "proportional
378 set size” (divide each shared page by the number of processes sharing it), the
379 number of clean and dirty shared pages in the mapping, and the number of clean
380 and dirty private pages in the mapping. The "Referenced" indicates the amount
381 of memory currently marked as referenced or accessed.
383 This file is only present if the CONFIG_MMU kernel configuration option is
386 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
387 bits on both physical and virtual pages associated with a process.
388 To clear the bits for all the pages associated with the process
389 > echo 1 > /proc/PID/clear_refs
391 To clear the bits for the anonymous pages associated with the process
392 > echo 2 > /proc/PID/clear_refs
394 To clear the bits for the file mapped pages associated with the process
395 > echo 3 > /proc/PID/clear_refs
396 Any other value written to /proc/PID/clear_refs will have no effect.
402 Similar to the process entries, the kernel data files give information about
403 the running kernel. The files used to obtain this information are contained in
404 /proc and are listed in Table 1-5. Not all of these will be present in your
405 system. It depends on the kernel configuration and the loaded modules, which
406 files are there, and which are missing.
408 Table 1-5: Kernel info in /proc
409 ..............................................................................
411 apm Advanced power management info
412 buddyinfo Kernel memory allocator information (see text) (2.5)
413 bus Directory containing bus specific information
414 cmdline Kernel command line
415 cpuinfo Info about the CPU
416 devices Available devices (block and character)
417 dma Used DMS channels
418 filesystems Supported filesystems
419 driver Various drivers grouped here, currently rtc (2.4)
420 execdomains Execdomains, related to security (2.4)
421 fb Frame Buffer devices (2.4)
422 fs File system parameters, currently nfs/exports (2.4)
423 ide Directory containing info about the IDE subsystem
424 interrupts Interrupt usage
425 iomem Memory map (2.4)
426 ioports I/O port usage
427 irq Masks for irq to cpu affinity (2.4)(smp?)
428 isapnp ISA PnP (Plug&Play) Info (2.4)
429 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
431 ksyms Kernel symbol table
432 loadavg Load average of last 1, 5 & 15 minutes
436 modules List of loaded modules
437 mounts Mounted filesystems
438 net Networking info (see text)
439 partitions Table of partitions known to the system
440 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
441 decoupled by lspci (2.4)
443 scsi SCSI info (see text)
444 slabinfo Slab pool info
445 softirqs softirq usage
446 stat Overall statistics
447 swaps Swap space utilization
449 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
450 tty Info of tty drivers
452 version Kernel version
453 video bttv info of video resources (2.4)
454 vmallocinfo Show vmalloced areas
455 ..............................................................................
457 You can, for example, check which interrupts are currently in use and what
458 they are used for by looking in the file /proc/interrupts:
460 > cat /proc/interrupts
462 0: 8728810 XT-PIC timer
463 1: 895 XT-PIC keyboard
465 3: 531695 XT-PIC aha152x
466 4: 2014133 XT-PIC serial
467 5: 44401 XT-PIC pcnet_cs
470 12: 182918 XT-PIC PS/2 Mouse
472 14: 1232265 XT-PIC ide0
476 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
477 output of a SMP machine):
479 > cat /proc/interrupts
482 0: 1243498 1214548 IO-APIC-edge timer
483 1: 8949 8958 IO-APIC-edge keyboard
484 2: 0 0 XT-PIC cascade
485 5: 11286 10161 IO-APIC-edge soundblaster
486 8: 1 0 IO-APIC-edge rtc
487 9: 27422 27407 IO-APIC-edge 3c503
488 12: 113645 113873 IO-APIC-edge PS/2 Mouse
490 14: 22491 24012 IO-APIC-edge ide0
491 15: 2183 2415 IO-APIC-edge ide1
492 17: 30564 30414 IO-APIC-level eth0
493 18: 177 164 IO-APIC-level bttv
498 NMI is incremented in this case because every timer interrupt generates a NMI
499 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
501 LOC is the local interrupt counter of the internal APIC of every CPU.
503 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
504 connects the CPUs in a SMP system. This means that an error has been detected,
505 the IO-APIC automatically retry the transmission, so it should not be a big
506 problem, but you should read the SMP-FAQ.
508 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
509 /proc/interrupts to display every IRQ vector in use by the system, not
510 just those considered 'most important'. The new vectors are:
512 THR -- interrupt raised when a machine check threshold counter
513 (typically counting ECC corrected errors of memory or cache) exceeds
514 a configurable threshold. Only available on some systems.
516 TRM -- a thermal event interrupt occurs when a temperature threshold
517 has been exceeded for the CPU. This interrupt may also be generated
518 when the temperature drops back to normal.
520 SPU -- a spurious interrupt is some interrupt that was raised then lowered
521 by some IO device before it could be fully processed by the APIC. Hence
522 the APIC sees the interrupt but does not know what device it came from.
523 For this case the APIC will generate the interrupt with a IRQ vector
524 of 0xff. This might also be generated by chipset bugs.
526 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
527 sent from one CPU to another per the needs of the OS. Typically,
528 their statistics are used by kernel developers and interested users to
529 determine the occurrence of interrupts of the given type.
531 The above IRQ vectors are displayed only when relevent. For example,
532 the threshold vector does not exist on x86_64 platforms. Others are
533 suppressed when the system is a uniprocessor. As of this writing, only
534 i386 and x86_64 platforms support the new IRQ vector displays.
536 Of some interest is the introduction of the /proc/irq directory to 2.4.
537 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
538 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
539 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
544 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
545 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
549 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
550 IRQ, you can set it by doing:
552 > echo 1 > /proc/irq/10/smp_affinity
554 This means that only the first CPU will handle the IRQ, but you can also echo
555 5 which means that only the first and fourth CPU can handle the IRQ.
557 The contents of each smp_affinity file is the same by default:
559 > cat /proc/irq/0/smp_affinity
562 The default_smp_affinity mask applies to all non-active IRQs, which are the
563 IRQs which have not yet been allocated/activated, and hence which lack a
564 /proc/irq/[0-9]* directory.
566 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
567 profiler. Default value is ffffffff (all cpus).
569 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
570 between all the CPUs which are allowed to handle it. As usual the kernel has
571 more info than you and does a better job than you, so the defaults are the
572 best choice for almost everyone.
574 There are three more important subdirectories in /proc: net, scsi, and sys.
575 The general rule is that the contents, or even the existence of these
576 directories, depend on your kernel configuration. If SCSI is not enabled, the
577 directory scsi may not exist. The same is true with the net, which is there
578 only when networking support is present in the running kernel.
580 The slabinfo file gives information about memory usage at the slab level.
581 Linux uses slab pools for memory management above page level in version 2.2.
582 Commonly used objects have their own slab pool (such as network buffers,
583 directory cache, and so on).
585 ..............................................................................
587 > cat /proc/buddyinfo
589 Node 0, zone DMA 0 4 5 4 4 3 ...
590 Node 0, zone Normal 1 0 0 1 101 8 ...
591 Node 0, zone HighMem 2 0 0 1 1 0 ...
593 Memory fragmentation is a problem under some workloads, and buddyinfo is a
594 useful tool for helping diagnose these problems. Buddyinfo will give you a
595 clue as to how big an area you can safely allocate, or why a previous
598 Each column represents the number of pages of a certain order which are
599 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
600 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
601 available in ZONE_NORMAL, etc...
603 ..............................................................................
607 Provides information about distribution and utilization of memory. This
608 varies by architecture and compile options. The following is from a
609 16GB PIII, which has highmem enabled. You may not have all of these fields.
614 MemTotal: 16344972 kB
621 HighTotal: 15597528 kB
622 HighFree: 13629632 kB
632 SReclaimable: 159856 kB
633 SUnreclaim: 124508 kB
638 CommitLimit: 7669796 kB
639 Committed_AS: 100056 kB
640 VmallocTotal: 112216 kB
642 VmallocChunk: 111088 kB
644 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
645 bits and the kernel binary code)
646 MemFree: The sum of LowFree+HighFree
647 Buffers: Relatively temporary storage for raw disk blocks
648 shouldn't get tremendously large (20MB or so)
649 Cached: in-memory cache for files read from the disk (the
650 pagecache). Doesn't include SwapCached
651 SwapCached: Memory that once was swapped out, is swapped back in but
652 still also is in the swapfile (if memory is needed it
653 doesn't need to be swapped out AGAIN because it is already
654 in the swapfile. This saves I/O)
655 Active: Memory that has been used more recently and usually not
656 reclaimed unless absolutely necessary.
657 Inactive: Memory which has been less recently used. It is more
658 eligible to be reclaimed for other purposes
660 HighFree: Highmem is all memory above ~860MB of physical memory
661 Highmem areas are for use by userspace programs, or
662 for the pagecache. The kernel must use tricks to access
663 this memory, making it slower to access than lowmem.
665 LowFree: Lowmem is memory which can be used for everything that
666 highmem can be used for, but it is also available for the
667 kernel's use for its own data structures. Among many
668 other things, it is where everything from the Slab is
669 allocated. Bad things happen when you're out of lowmem.
670 SwapTotal: total amount of swap space available
671 SwapFree: Memory which has been evicted from RAM, and is temporarily
673 Dirty: Memory which is waiting to get written back to the disk
674 Writeback: Memory which is actively being written back to the disk
675 AnonPages: Non-file backed pages mapped into userspace page tables
676 Mapped: files which have been mmaped, such as libraries
677 Slab: in-kernel data structures cache
678 SReclaimable: Part of Slab, that might be reclaimed, such as caches
679 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
680 PageTables: amount of memory dedicated to the lowest level of page
682 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
684 Bounce: Memory used for block device "bounce buffers"
685 WritebackTmp: Memory used by FUSE for temporary writeback buffers
686 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
687 this is the total amount of memory currently available to
688 be allocated on the system. This limit is only adhered to
689 if strict overcommit accounting is enabled (mode 2 in
690 'vm.overcommit_memory').
691 The CommitLimit is calculated with the following formula:
692 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
693 For example, on a system with 1G of physical RAM and 7G
694 of swap with a `vm.overcommit_ratio` of 30 it would
695 yield a CommitLimit of 7.3G.
696 For more details, see the memory overcommit documentation
697 in vm/overcommit-accounting.
698 Committed_AS: The amount of memory presently allocated on the system.
699 The committed memory is a sum of all of the memory which
700 has been allocated by processes, even if it has not been
701 "used" by them as of yet. A process which malloc()'s 1G
702 of memory, but only touches 300M of it will only show up
703 as using 300M of memory even if it has the address space
704 allocated for the entire 1G. This 1G is memory which has
705 been "committed" to by the VM and can be used at any time
706 by the allocating application. With strict overcommit
707 enabled on the system (mode 2 in 'vm.overcommit_memory'),
708 allocations which would exceed the CommitLimit (detailed
709 above) will not be permitted. This is useful if one needs
710 to guarantee that processes will not fail due to lack of
711 memory once that memory has been successfully allocated.
712 VmallocTotal: total size of vmalloc memory area
713 VmallocUsed: amount of vmalloc area which is used
714 VmallocChunk: largest contiguous block of vmalloc area which is free
716 ..............................................................................
720 Provides information about vmalloced/vmaped areas. One line per area,
721 containing the virtual address range of the area, size in bytes,
722 caller information of the creator, and optional information depending
723 on the kind of area :
725 pages=nr number of pages
726 phys=addr if a physical address was specified
727 ioremap I/O mapping (ioremap() and friends)
728 vmalloc vmalloc() area
731 vpages buffer for pages pointers was vmalloced (huge area)
732 N<node>=nr (Only on NUMA kernels)
733 Number of pages allocated on memory node <node>
735 > cat /proc/vmallocinfo
736 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
737 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
738 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
739 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
740 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
741 phys=7fee8000 ioremap
742 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
743 phys=7fee7000 ioremap
744 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
745 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
746 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
747 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
749 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
750 /0x130 [x_tables] pages=4 vmalloc N0=4
751 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
752 pages=14 vmalloc N2=14
753 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
755 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
757 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
758 pages=10 vmalloc N0=10
760 ..............................................................................
764 Provides counts of softirq handlers serviced since boot time, for each cpu.
769 TIMER: 27166 27120 27097 27034
774 SCHED: 27035 26983 26971 26746
776 RCU: 1678 1769 2178 2250
779 1.3 IDE devices in /proc/ide
780 ----------------------------
782 The subdirectory /proc/ide contains information about all IDE devices of which
783 the kernel is aware. There is one subdirectory for each IDE controller, the
784 file drivers and a link for each IDE device, pointing to the device directory
785 in the controller specific subtree.
787 The file drivers contains general information about the drivers used for the
790 > cat /proc/ide/drivers
791 ide-cdrom version 4.53
792 ide-disk version 1.08
794 More detailed information can be found in the controller specific
795 subdirectories. These are named ide0, ide1 and so on. Each of these
796 directories contains the files shown in table 1-6.
799 Table 1-6: IDE controller info in /proc/ide/ide?
800 ..............................................................................
802 channel IDE channel (0 or 1)
803 config Configuration (only for PCI/IDE bridge)
805 model Type/Chipset of IDE controller
806 ..............................................................................
808 Each device connected to a controller has a separate subdirectory in the
809 controllers directory. The files listed in table 1-7 are contained in these
813 Table 1-7: IDE device information
814 ..............................................................................
817 capacity Capacity of the medium (in 512Byte blocks)
818 driver driver and version
819 geometry physical and logical geometry
820 identify device identify block
822 model device identifier
823 settings device setup
824 smart_thresholds IDE disk management thresholds
825 smart_values IDE disk management values
826 ..............................................................................
828 The most interesting file is settings. This file contains a nice overview of
829 the drive parameters:
831 # cat /proc/ide/ide0/hda/settings
832 name value min max mode
833 ---- ----- --- --- ----
834 bios_cyl 526 0 65535 rw
835 bios_head 255 0 255 rw
837 breada_readahead 4 0 127 rw
839 file_readahead 72 0 2097151 rw
841 keepsettings 0 0 1 rw
842 max_kb_per_request 122 1 127 rw
846 pio_mode write-only 0 255 w
852 1.4 Networking info in /proc/net
853 --------------------------------
855 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
856 additional values you get for IP version 6 if you configure the kernel to
857 support this. Table 1-9 lists the files and their meaning.
860 Table 1-8: IPv6 info in /proc/net
861 ..............................................................................
863 udp6 UDP sockets (IPv6)
864 tcp6 TCP sockets (IPv6)
865 raw6 Raw device statistics (IPv6)
866 igmp6 IP multicast addresses, which this host joined (IPv6)
867 if_inet6 List of IPv6 interface addresses
868 ipv6_route Kernel routing table for IPv6
869 rt6_stats Global IPv6 routing tables statistics
870 sockstat6 Socket statistics (IPv6)
871 snmp6 Snmp data (IPv6)
872 ..............................................................................
875 Table 1-9: Network info in /proc/net
876 ..............................................................................
879 dev network devices with statistics
880 dev_mcast the Layer2 multicast groups a device is listening too
881 (interface index, label, number of references, number of bound
883 dev_stat network device status
884 ip_fwchains Firewall chain linkage
885 ip_fwnames Firewall chain names
886 ip_masq Directory containing the masquerading tables
887 ip_masquerade Major masquerading table
888 netstat Network statistics
889 raw raw device statistics
890 route Kernel routing table
891 rpc Directory containing rpc info
892 rt_cache Routing cache
894 sockstat Socket statistics
896 tr_rif Token ring RIF routing table
898 unix UNIX domain sockets
899 wireless Wireless interface data (Wavelan etc)
900 igmp IP multicast addresses, which this host joined
901 psched Global packet scheduler parameters.
902 netlink List of PF_NETLINK sockets
903 ip_mr_vifs List of multicast virtual interfaces
904 ip_mr_cache List of multicast routing cache
905 ..............................................................................
907 You can use this information to see which network devices are available in
908 your system and how much traffic was routed over those devices:
912 face |bytes packets errs drop fifo frame compressed multicast|[...
913 lo: 908188 5596 0 0 0 0 0 0 [...
914 ppp0:15475140 20721 410 0 0 410 0 0 [...
915 eth0: 614530 7085 0 0 0 0 0 1 [...
918 ...] bytes packets errs drop fifo colls carrier compressed
919 ...] 908188 5596 0 0 0 0 0 0
920 ...] 1375103 17405 0 0 0 0 0 0
921 ...] 1703981 5535 0 0 0 3 0 0
923 In addition, each Channel Bond interface has it's own directory. For
924 example, the bond0 device will have a directory called /proc/net/bond0/.
925 It will contain information that is specific to that bond, such as the
926 current slaves of the bond, the link status of the slaves, and how
927 many times the slaves link has failed.
932 If you have a SCSI host adapter in your system, you'll find a subdirectory
933 named after the driver for this adapter in /proc/scsi. You'll also see a list
934 of all recognized SCSI devices in /proc/scsi:
938 Host: scsi0 Channel: 00 Id: 00 Lun: 00
939 Vendor: IBM Model: DGHS09U Rev: 03E0
940 Type: Direct-Access ANSI SCSI revision: 03
941 Host: scsi0 Channel: 00 Id: 06 Lun: 00
942 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
943 Type: CD-ROM ANSI SCSI revision: 02
946 The directory named after the driver has one file for each adapter found in
947 the system. These files contain information about the controller, including
948 the used IRQ and the IO address range. The amount of information shown is
949 dependent on the adapter you use. The example shows the output for an Adaptec
950 AHA-2940 SCSI adapter:
952 > cat /proc/scsi/aic7xxx/0
954 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
956 TCQ Enabled By Default : Disabled
957 AIC7XXX_PROC_STATS : Disabled
958 AIC7XXX_RESET_DELAY : 5
959 Adapter Configuration:
960 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
961 Ultra Wide Controller
962 PCI MMAPed I/O Base: 0xeb001000
963 Adapter SEEPROM Config: SEEPROM found and used.
964 Adaptec SCSI BIOS: Enabled
966 SCBs: Active 0, Max Active 2,
967 Allocated 15, HW 16, Page 255
969 BIOS Control Word: 0x18b6
970 Adapter Control Word: 0x005b
971 Extended Translation: Enabled
972 Disconnect Enable Flags: 0xffff
973 Ultra Enable Flags: 0x0001
974 Tag Queue Enable Flags: 0x0000
975 Ordered Queue Tag Flags: 0x0000
976 Default Tag Queue Depth: 8
977 Tagged Queue By Device array for aic7xxx host instance 0:
978 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
979 Actual queue depth per device for aic7xxx host instance 0:
980 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
983 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
984 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
985 Total transfers 160151 (74577 reads and 85574 writes)
987 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
988 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
989 Total transfers 0 (0 reads and 0 writes)
992 1.6 Parallel port info in /proc/parport
993 ---------------------------------------
995 The directory /proc/parport contains information about the parallel ports of
996 your system. It has one subdirectory for each port, named after the port
999 These directories contain the four files shown in Table 1-10.
1002 Table 1-10: Files in /proc/parport
1003 ..............................................................................
1005 autoprobe Any IEEE-1284 device ID information that has been acquired.
1006 devices list of the device drivers using that port. A + will appear by the
1007 name of the device currently using the port (it might not appear
1009 hardware Parallel port's base address, IRQ line and DMA channel.
1010 irq IRQ that parport is using for that port. This is in a separate
1011 file to allow you to alter it by writing a new value in (IRQ
1013 ..............................................................................
1015 1.7 TTY info in /proc/tty
1016 -------------------------
1018 Information about the available and actually used tty's can be found in the
1019 directory /proc/tty.You'll find entries for drivers and line disciplines in
1020 this directory, as shown in Table 1-11.
1023 Table 1-11: Files in /proc/tty
1024 ..............................................................................
1026 drivers list of drivers and their usage
1027 ldiscs registered line disciplines
1028 driver/serial usage statistic and status of single tty lines
1029 ..............................................................................
1031 To see which tty's are currently in use, you can simply look into the file
1034 > cat /proc/tty/drivers
1035 pty_slave /dev/pts 136 0-255 pty:slave
1036 pty_master /dev/ptm 128 0-255 pty:master
1037 pty_slave /dev/ttyp 3 0-255 pty:slave
1038 pty_master /dev/pty 2 0-255 pty:master
1039 serial /dev/cua 5 64-67 serial:callout
1040 serial /dev/ttyS 4 64-67 serial
1041 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1042 /dev/ptmx /dev/ptmx 5 2 system
1043 /dev/console /dev/console 5 1 system:console
1044 /dev/tty /dev/tty 5 0 system:/dev/tty
1045 unknown /dev/tty 4 1-63 console
1048 1.8 Miscellaneous kernel statistics in /proc/stat
1049 -------------------------------------------------
1051 Various pieces of information about kernel activity are available in the
1052 /proc/stat file. All of the numbers reported in this file are aggregates
1053 since the system first booted. For a quick look, simply cat the file:
1056 cpu 2255 34 2290 22625563 6290 127 456 0 0
1057 cpu0 1132 34 1441 11311718 3675 127 438 0 0
1058 cpu1 1123 0 849 11313845 2614 0 18 0 0
1059 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1065 softirq 183433 0 21755 12 39 1137 231 21459 2263
1067 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1068 lines. These numbers identify the amount of time the CPU has spent performing
1069 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1070 second). The meanings of the columns are as follows, from left to right:
1072 - user: normal processes executing in user mode
1073 - nice: niced processes executing in user mode
1074 - system: processes executing in kernel mode
1075 - idle: twiddling thumbs
1076 - iowait: waiting for I/O to complete
1077 - irq: servicing interrupts
1078 - softirq: servicing softirqs
1079 - steal: involuntary wait
1080 - guest: running a normal guest
1081 - guest_nice: running a niced guest
1083 The "intr" line gives counts of interrupts serviced since boot time, for each
1084 of the possible system interrupts. The first column is the total of all
1085 interrupts serviced; each subsequent column is the total for that particular
1088 The "ctxt" line gives the total number of context switches across all CPUs.
1090 The "btime" line gives the time at which the system booted, in seconds since
1093 The "processes" line gives the number of processes and threads created, which
1094 includes (but is not limited to) those created by calls to the fork() and
1095 clone() system calls.
1097 The "procs_running" line gives the total number of threads that are
1098 running or ready to run (i.e., the total number of runnable threads).
1100 The "procs_blocked" line gives the number of processes currently blocked,
1101 waiting for I/O to complete.
1103 The "softirq" line gives counts of softirqs serviced since boot time, for each
1104 of the possible system softirqs. The first column is the total of all
1105 softirqs serviced; each subsequent column is the total for that particular
1109 1.9 Ext4 file system parameters
1110 ------------------------------
1112 Information about mounted ext4 file systems can be found in
1113 /proc/fs/ext4. Each mounted filesystem will have a directory in
1114 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1115 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1116 in Table 1-12, below.
1118 Table 1-12: Files in /proc/fs/ext4/<devname>
1119 ..............................................................................
1121 mb_groups details of multiblock allocator buddy cache of free blocks
1122 ..............................................................................
1125 ------------------------------------------------------------------------------
1127 ------------------------------------------------------------------------------
1128 The /proc file system serves information about the running system. It not only
1129 allows access to process data but also allows you to request the kernel status
1130 by reading files in the hierarchy.
1132 The directory structure of /proc reflects the types of information and makes
1133 it easy, if not obvious, where to look for specific data.
1134 ------------------------------------------------------------------------------
1136 ------------------------------------------------------------------------------
1137 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1138 ------------------------------------------------------------------------------
1140 ------------------------------------------------------------------------------
1142 ------------------------------------------------------------------------------
1143 * Modifying kernel parameters by writing into files found in /proc/sys
1144 * Exploring the files which modify certain parameters
1145 * Review of the /proc/sys file tree
1146 ------------------------------------------------------------------------------
1149 A very interesting part of /proc is the directory /proc/sys. This is not only
1150 a source of information, it also allows you to change parameters within the
1151 kernel. Be very careful when attempting this. You can optimize your system,
1152 but you can also cause it to crash. Never alter kernel parameters on a
1153 production system. Set up a development machine and test to make sure that
1154 everything works the way you want it to. You may have no alternative but to
1155 reboot the machine once an error has been made.
1157 To change a value, simply echo the new value into the file. An example is
1158 given below in the section on the file system data. You need to be root to do
1159 this. You can create your own boot script to perform this every time your
1162 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1163 general things in the operation of the Linux kernel. Since some of the files
1164 can inadvertently disrupt your system, it is advisable to read both
1165 documentation and source before actually making adjustments. In any case, be
1166 very careful when writing to any of these files. The entries in /proc may
1167 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1168 review the kernel documentation in the directory /usr/src/linux/Documentation.
1169 This chapter is heavily based on the documentation included in the pre 2.2
1170 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1172 Please see: Documentation/sysctls/ directory for descriptions of these
1175 ------------------------------------------------------------------------------
1177 ------------------------------------------------------------------------------
1178 Certain aspects of kernel behavior can be modified at runtime, without the
1179 need to recompile the kernel, or even to reboot the system. The files in the
1180 /proc/sys tree can not only be read, but also modified. You can use the echo
1181 command to write value into these files, thereby changing the default settings
1183 ------------------------------------------------------------------------------
1185 ------------------------------------------------------------------------------
1186 CHAPTER 3: PER-PROCESS PARAMETERS
1187 ------------------------------------------------------------------------------
1189 3.1 /proc/<pid>/oom_adj - Adjust the oom-killer score
1190 ------------------------------------------------------
1192 This file can be used to adjust the score used to select which processes
1193 should be killed in an out-of-memory situation. Giving it a high score will
1194 increase the likelihood of this process being killed by the oom-killer. Valid
1195 values are in the range -16 to +15, plus the special value -17, which disables
1196 oom-killing altogether for this process.
1198 The process to be killed in an out-of-memory situation is selected among all others
1199 based on its badness score. This value equals the original memory size of the process
1200 and is then updated according to its CPU time (utime + stime) and the
1201 run time (uptime - start time). The longer it runs the smaller is the score.
1202 Badness score is divided by the square root of the CPU time and then by
1203 the double square root of the run time.
1205 Swapped out tasks are killed first. Half of each child's memory size is added to
1206 the parent's score if they do not share the same memory. Thus forking servers
1207 are the prime candidates to be killed. Having only one 'hungry' child will make
1208 parent less preferable than the child.
1210 /proc/<pid>/oom_score shows process' current badness score.
1212 The following heuristics are then applied:
1213 * if the task was reniced, its score doubles
1214 * superuser or direct hardware access tasks (CAP_SYS_ADMIN, CAP_SYS_RESOURCE
1215 or CAP_SYS_RAWIO) have their score divided by 4
1216 * if oom condition happened in one cpuset and checked process does not belong
1217 to it, its score is divided by 8
1218 * the resulting score is multiplied by two to the power of oom_adj, i.e.
1219 points <<= oom_adj when it is positive and
1220 points >>= -(oom_adj) otherwise
1222 The task with the highest badness score is then selected and its children
1223 are killed, process itself will be killed in an OOM situation when it does
1224 not have children or some of them disabled oom like described above.
1226 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1227 -------------------------------------------------------------
1229 This file can be used to check the current score used by the oom-killer is for
1230 any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
1231 process should be killed in an out-of-memory situation.
1234 3.3 /proc/<pid>/io - Display the IO accounting fields
1235 -------------------------------------------------------
1237 This file contains IO statistics for each running process
1242 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1245 test:/tmp # cat /proc/3828/io
1251 write_bytes: 323932160
1252 cancelled_write_bytes: 0
1261 I/O counter: chars read
1262 The number of bytes which this task has caused to be read from storage. This
1263 is simply the sum of bytes which this process passed to read() and pread().
1264 It includes things like tty IO and it is unaffected by whether or not actual
1265 physical disk IO was required (the read might have been satisfied from
1272 I/O counter: chars written
1273 The number of bytes which this task has caused, or shall cause to be written
1274 to disk. Similar caveats apply here as with rchar.
1280 I/O counter: read syscalls
1281 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1288 I/O counter: write syscalls
1289 Attempt to count the number of write I/O operations, i.e. syscalls like
1290 write() and pwrite().
1296 I/O counter: bytes read
1297 Attempt to count the number of bytes which this process really did cause to
1298 be fetched from the storage layer. Done at the submit_bio() level, so it is
1299 accurate for block-backed filesystems. <please add status regarding NFS and
1300 CIFS at a later time>
1306 I/O counter: bytes written
1307 Attempt to count the number of bytes which this process caused to be sent to
1308 the storage layer. This is done at page-dirtying time.
1311 cancelled_write_bytes
1312 ---------------------
1314 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1315 then deletes the file, it will in fact perform no writeout. But it will have
1316 been accounted as having caused 1MB of write.
1317 In other words: The number of bytes which this process caused to not happen,
1318 by truncating pagecache. A task can cause "negative" IO too. If this task
1319 truncates some dirty pagecache, some IO which another task has been accounted
1320 for (in it's write_bytes) will not be happening. We _could_ just subtract that
1321 from the truncating task's write_bytes, but there is information loss in doing
1328 At its current implementation state, this is a bit racy on 32-bit machines: if
1329 process A reads process B's /proc/pid/io while process B is updating one of
1330 those 64-bit counters, process A could see an intermediate result.
1333 More information about this can be found within the taskstats documentation in
1334 Documentation/accounting.
1336 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1337 ---------------------------------------------------------------
1338 When a process is dumped, all anonymous memory is written to a core file as
1339 long as the size of the core file isn't limited. But sometimes we don't want
1340 to dump some memory segments, for example, huge shared memory. Conversely,
1341 sometimes we want to save file-backed memory segments into a core file, not
1342 only the individual files.
1344 /proc/<pid>/coredump_filter allows you to customize which memory segments
1345 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1346 of memory types. If a bit of the bitmask is set, memory segments of the
1347 corresponding memory type are dumped, otherwise they are not dumped.
1349 The following 7 memory types are supported:
1350 - (bit 0) anonymous private memory
1351 - (bit 1) anonymous shared memory
1352 - (bit 2) file-backed private memory
1353 - (bit 3) file-backed shared memory
1354 - (bit 4) ELF header pages in file-backed private memory areas (it is
1355 effective only if the bit 2 is cleared)
1356 - (bit 5) hugetlb private memory
1357 - (bit 6) hugetlb shared memory
1359 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1360 are always dumped regardless of the bitmask status.
1362 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
1363 effected by bit 5-6.
1365 Default value of coredump_filter is 0x23; this means all anonymous memory
1366 segments and hugetlb private memory are dumped.
1368 If you don't want to dump all shared memory segments attached to pid 1234,
1369 write 0x21 to the process's proc file.
1371 $ echo 0x21 > /proc/1234/coredump_filter
1373 When a new process is created, the process inherits the bitmask status from its
1374 parent. It is useful to set up coredump_filter before the program runs.
1377 $ echo 0x7 > /proc/self/coredump_filter
1380 3.5 /proc/<pid>/mountinfo - Information about mounts
1381 --------------------------------------------------------
1383 This file contains lines of the form:
1385 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1386 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1388 (1) mount ID: unique identifier of the mount (may be reused after umount)
1389 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1390 (3) major:minor: value of st_dev for files on filesystem
1391 (4) root: root of the mount within the filesystem
1392 (5) mount point: mount point relative to the process's root
1393 (6) mount options: per mount options
1394 (7) optional fields: zero or more fields of the form "tag[:value]"
1395 (8) separator: marks the end of the optional fields
1396 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1397 (10) mount source: filesystem specific information or "none"
1398 (11) super options: per super block options
1400 Parsers should ignore all unrecognised optional fields. Currently the
1401 possible optional fields are:
1403 shared:X mount is shared in peer group X
1404 master:X mount is slave to peer group X
1405 propagate_from:X mount is slave and receives propagation from peer group X (*)
1406 unbindable mount is unbindable
1408 (*) X is the closest dominant peer group under the process's root. If
1409 X is the immediate master of the mount, or if there's no dominant peer
1410 group under the same root, then only the "master:X" field is present
1411 and not the "propagate_from:X" field.
1413 For more information on mount propagation see:
1415 Documentation/filesystems/sharedsubtree.txt
1418 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1419 --------------------------------------------------------
1420 These files provide a method to access a tasks comm value. It also allows for
1421 a task to set its own or one of its thread siblings comm value. The comm value
1422 is limited in size compared to the cmdline value, so writing anything longer
1423 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
git://bbs.cooldavid.org / net-next-2.6.git / blob