1 ------------------------------------------------------------------------------
2 T H E /proc F I L E S Y S T E M
3 ------------------------------------------------------------------------------
4 /proc/sys Terrehon Bowden <terrehon@pacbell.net> October 7 1999
5 Bodo Bauer <bb@ricochet.net>
7 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8 move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009
9 ------------------------------------------------------------------------------
10 Version 1.3 Kernel version 2.2.12
11 Kernel version 2.4.0-test11-pre4
12 ------------------------------------------------------------------------------
13 fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009
19 0.1 Introduction/Credits
22 1 Collecting System Information
23 1.1 Process-Specific Subdirectories
25 1.3 IDE devices in /proc/ide
26 1.4 Networking info in /proc/net
28 1.6 Parallel port info in /proc/parport
29 1.7 TTY info in /proc/tty
30 1.8 Miscellaneous kernel statistics in /proc/stat
31 1.9 Ext4 file system parameters
33 2 Modifying System Parameters
35 3 Per-Process Parameters
36 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer
38 3.2 /proc/<pid>/oom_score - Display current oom-killer score
39 3.3 /proc/<pid>/io - Display the IO accounting fields
40 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
41 3.5 /proc/<pid>/mountinfo - Information about mounts
42 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
43 3.7 /proc/<pid>/task/<tid>/children - Information about task children
44 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
45 3.9 /proc/<pid>/map_files - Information about memory mapped files
46 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
47 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
52 ------------------------------------------------------------------------------
54 ------------------------------------------------------------------------------
56 0.1 Introduction/Credits
57 ------------------------
59 This documentation is part of a soon (or so we hope) to be released book on
60 the SuSE Linux distribution. As there is no complete documentation for the
61 /proc file system and we've used many freely available sources to write these
62 chapters, it seems only fair to give the work back to the Linux community.
63 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
64 afraid it's still far from complete, but we hope it will be useful. As far as
65 we know, it is the first 'all-in-one' document about the /proc file system. It
66 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
67 SPARC, AXP, etc., features, you probably won't find what you are looking for.
68 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
69 additions and patches are welcome and will be added to this document if you
72 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
73 other people for help compiling this documentation. We'd also like to extend a
74 special thank you to Andi Kleen for documentation, which we relied on heavily
75 to create this document, as well as the additional information he provided.
76 Thanks to everybody else who contributed source or docs to the Linux kernel
77 and helped create a great piece of software... :)
79 If you have any comments, corrections or additions, please don't hesitate to
80 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
83 The latest version of this document is available online at
84 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
86 If the above direction does not works for you, you could try the kernel
87 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
88 comandante@zaralinux.com.
93 We don't guarantee the correctness of this document, and if you come to us
94 complaining about how you screwed up your system because of incorrect
95 documentation, we won't feel responsible...
97 ------------------------------------------------------------------------------
98 CHAPTER 1: COLLECTING SYSTEM INFORMATION
99 ------------------------------------------------------------------------------
101 ------------------------------------------------------------------------------
103 ------------------------------------------------------------------------------
104 * Investigating the properties of the pseudo file system /proc and its
105 ability to provide information on the running Linux system
106 * Examining /proc's structure
107 * Uncovering various information about the kernel and the processes running
109 ------------------------------------------------------------------------------
112 The proc file system acts as an interface to internal data structures in the
113 kernel. It can be used to obtain information about the system and to change
114 certain kernel parameters at runtime (sysctl).
116 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
117 show you how you can use /proc/sys to change settings.
119 1.1 Process-Specific Subdirectories
120 -----------------------------------
122 The directory /proc contains (among other things) one subdirectory for each
123 process running on the system, which is named after the process ID (PID).
125 The link self points to the process reading the file system. Each process
126 subdirectory has the entries listed in Table 1-1.
129 Table 1-1: Process specific entries in /proc
130 ..............................................................................
132 clear_refs Clears page referenced bits shown in smaps output
133 cmdline Command line arguments
134 cpu Current and last cpu in which it was executed (2.4)(smp)
135 cwd Link to the current working directory
136 environ Values of environment variables
137 exe Link to the executable of this process
138 fd Directory, which contains all file descriptors
139 maps Memory maps to executables and library files (2.4)
140 mem Memory held by this process
141 root Link to the root directory of this process
143 statm Process memory status information
144 status Process status in human readable form
145 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function
146 symbol the task is blocked in - or "0" if not blocked.
148 stack Report full stack trace, enable via CONFIG_STACKTRACE
149 smaps an extension based on maps, showing the memory consumption of
150 each mapping and flags associated with it
151 numa_maps an extension based on maps, showing the memory locality and
152 binding policy as well as mem usage (in pages) of each mapping.
153 ..............................................................................
155 For example, to get the status information of a process, all you have to do is
156 read the file /proc/PID/status:
158 >cat /proc/self/status
188 SigPnd: 0000000000000000
189 ShdPnd: 0000000000000000
190 SigBlk: 0000000000000000
191 SigIgn: 0000000000000000
192 SigCgt: 0000000000000000
193 CapInh: 00000000fffffeff
194 CapPrm: 0000000000000000
195 CapEff: 0000000000000000
196 CapBnd: ffffffffffffffff
199 voluntary_ctxt_switches: 0
200 nonvoluntary_ctxt_switches: 1
202 This shows you nearly the same information you would get if you viewed it with
203 the ps command. In fact, ps uses the proc file system to obtain its
204 information. But you get a more detailed view of the process by reading the
205 file /proc/PID/status. It fields are described in table 1-2.
207 The statm file contains more detailed information about the process
208 memory usage. Its seven fields are explained in Table 1-3. The stat file
209 contains details information about the process itself. Its fields are
210 explained in Table 1-4.
212 (for SMP CONFIG users)
213 For making accounting scalable, RSS related information are handled in an
214 asynchronous manner and the value may not be very precise. To see a precise
215 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
216 It's slow but very precise.
218 Table 1-2: Contents of the status files (as of 4.8)
219 ..............................................................................
221 Name filename of the executable
222 Umask file mode creation mask
223 State state (R is running, S is sleeping, D is sleeping
224 in an uninterruptible wait, Z is zombie,
225 T is traced or stopped)
227 Ngid NUMA group ID (0 if none)
229 PPid process id of the parent process
230 TracerPid PID of process tracing this process (0 if not)
231 Uid Real, effective, saved set, and file system UIDs
232 Gid Real, effective, saved set, and file system GIDs
233 FDSize number of file descriptor slots currently allocated
234 Groups supplementary group list
235 NStgid descendant namespace thread group ID hierarchy
236 NSpid descendant namespace process ID hierarchy
237 NSpgid descendant namespace process group ID hierarchy
238 NSsid descendant namespace session ID hierarchy
239 VmPeak peak virtual memory size
240 VmSize total program size
241 VmLck locked memory size
242 VmPin pinned memory size
243 VmHWM peak resident set size ("high water mark")
244 VmRSS size of memory portions. It contains the three
245 following parts (VmRSS = RssAnon + RssFile + RssShmem)
246 RssAnon size of resident anonymous memory
247 RssFile size of resident file mappings
248 RssShmem size of resident shmem memory (includes SysV shm,
249 mapping of tmpfs and shared anonymous mappings)
250 VmData size of private data segments
251 VmStk size of stack segments
252 VmExe size of text segment
253 VmLib size of shared library code
254 VmPTE size of page table entries
255 VmSwap amount of swap used by anonymous private data
256 (shmem swap usage is not included)
257 HugetlbPages size of hugetlb memory portions
258 CoreDumping process's memory is currently being dumped
259 (killing the process may lead to a corrupted core)
260 THP_enabled process is allowed to use THP (returns 0 when
261 PR_SET_THP_DISABLE is set on the process
262 Threads number of threads
263 SigQ number of signals queued/max. number for queue
264 SigPnd bitmap of pending signals for the thread
265 ShdPnd bitmap of shared pending signals for the process
266 SigBlk bitmap of blocked signals
267 SigIgn bitmap of ignored signals
268 SigCgt bitmap of caught signals
269 CapInh bitmap of inheritable capabilities
270 CapPrm bitmap of permitted capabilities
271 CapEff bitmap of effective capabilities
272 CapBnd bitmap of capabilities bounding set
273 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
274 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
275 Cpus_allowed mask of CPUs on which this process may run
276 Cpus_allowed_list Same as previous, but in "list format"
277 Mems_allowed mask of memory nodes allowed to this process
278 Mems_allowed_list Same as previous, but in "list format"
279 voluntary_ctxt_switches number of voluntary context switches
280 nonvoluntary_ctxt_switches number of non voluntary context switches
281 ..............................................................................
283 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
284 ..............................................................................
286 size total program size (pages) (same as VmSize in status)
287 resident size of memory portions (pages) (same as VmRSS in status)
288 shared number of pages that are shared (i.e. backed by a file, same
289 as RssFile+RssShmem in status)
290 trs number of pages that are 'code' (not including libs; broken,
291 includes data segment)
292 lrs number of pages of library (always 0 on 2.6)
293 drs number of pages of data/stack (including libs; broken,
294 includes library text)
295 dt number of dirty pages (always 0 on 2.6)
296 ..............................................................................
299 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
300 ..............................................................................
303 tcomm filename of the executable
304 state state (R is running, S is sleeping, D is sleeping in an
305 uninterruptible wait, Z is zombie, T is traced or stopped)
306 ppid process id of the parent process
307 pgrp pgrp of the process
309 tty_nr tty the process uses
310 tty_pgrp pgrp of the tty
312 min_flt number of minor faults
313 cmin_flt number of minor faults with child's
314 maj_flt number of major faults
315 cmaj_flt number of major faults with child's
316 utime user mode jiffies
317 stime kernel mode jiffies
318 cutime user mode jiffies with child's
319 cstime kernel mode jiffies with child's
320 priority priority level
322 num_threads number of threads
323 it_real_value (obsolete, always 0)
324 start_time time the process started after system boot
325 vsize virtual memory size
326 rss resident set memory size
327 rsslim current limit in bytes on the rss
328 start_code address above which program text can run
329 end_code address below which program text can run
330 start_stack address of the start of the main process stack
331 esp current value of ESP
332 eip current value of EIP
333 pending bitmap of pending signals
334 blocked bitmap of blocked signals
335 sigign bitmap of ignored signals
336 sigcatch bitmap of caught signals
337 0 (place holder, used to be the wchan address, use /proc/PID/wchan instead)
340 exit_signal signal to send to parent thread on exit
341 task_cpu which CPU the task is scheduled on
342 rt_priority realtime priority
343 policy scheduling policy (man sched_setscheduler)
344 blkio_ticks time spent waiting for block IO
345 gtime guest time of the task in jiffies
346 cgtime guest time of the task children in jiffies
347 start_data address above which program data+bss is placed
348 end_data address below which program data+bss is placed
349 start_brk address above which program heap can be expanded with brk()
350 arg_start address above which program command line is placed
351 arg_end address below which program command line is placed
352 env_start address above which program environment is placed
353 env_end address below which program environment is placed
354 exit_code the thread's exit_code in the form reported by the waitpid system call
355 ..............................................................................
357 The /proc/PID/maps file containing the currently mapped memory regions and
358 their access permissions.
362 address perms offset dev inode pathname
364 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
365 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
366 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
367 a7cb1000-a7cb2000 ---p 00000000 00:00 0
368 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
369 a7eb2000-a7eb3000 ---p 00000000 00:00 0
370 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
371 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
372 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
373 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
374 a800b000-a800e000 rw-p 00000000 00:00 0
375 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
376 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
377 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
378 a8024000-a8027000 rw-p 00000000 00:00 0
379 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
380 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
381 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
382 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
383 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
385 where "address" is the address space in the process that it occupies, "perms"
386 is a set of permissions:
392 p = private (copy on write)
394 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
395 "inode" is the inode on that device. 0 indicates that no inode is associated
396 with the memory region, as the case would be with BSS (uninitialized data).
397 The "pathname" shows the name associated file for this mapping. If the mapping
398 is not associated with a file:
400 [heap] = the heap of the program
401 [stack] = the stack of the main process
402 [vdso] = the "virtual dynamic shared object",
403 the kernel system call handler
405 or if empty, the mapping is anonymous.
407 The /proc/PID/smaps is an extension based on maps, showing the memory
408 consumption for each of the process's mappings. For each of mappings there
409 is a series of lines such as the following:
411 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
425 Private_Hugetlb: 0 kB
432 VmFlags: rd ex mr mw me dw
434 the first of these lines shows the same information as is displayed for the
435 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
436 (size), the amount of the mapping that is currently resident in RAM (RSS), the
437 process' proportional share of this mapping (PSS), the number of clean and
438 dirty private pages in the mapping.
440 The "proportional set size" (PSS) of a process is the count of pages it has
441 in memory, where each page is divided by the number of processes sharing it.
442 So if a process has 1000 pages all to itself, and 1000 shared with one other
443 process, its PSS will be 1500.
444 Note that even a page which is part of a MAP_SHARED mapping, but has only
445 a single pte mapped, i.e. is currently used by only one process, is accounted
446 as private and not as shared.
447 "Referenced" indicates the amount of memory currently marked as referenced or
449 "Anonymous" shows the amount of memory that does not belong to any file. Even
450 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
451 and a page is modified, the file page is replaced by a private anonymous copy.
452 "LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
453 The memory isn't freed immediately with madvise(). It's freed in memory
454 pressure if the memory is clean. Please note that the printed value might
455 be lower than the real value due to optimizations used in the current
456 implementation. If this is not desirable please file a bug report.
457 "AnonHugePages" shows the ammount of memory backed by transparent hugepage.
458 "ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by
460 "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
461 hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
462 reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
463 "Swap" shows how much would-be-anonymous memory is also used, but out on swap.
464 For shmem mappings, "Swap" includes also the size of the mapped (and not
465 replaced by copy-on-write) part of the underlying shmem object out on swap.
466 "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
467 does not take into account swapped out page of underlying shmem objects.
468 "Locked" indicates whether the mapping is locked in memory or not.
469 "THPeligible" indicates whether the mapping is eligible for THP pages - 1 if
472 "VmFlags" field deserves a separate description. This member represents the kernel
473 flags associated with the particular virtual memory area in two letter encoded
474 manner. The codes are the following:
483 gd - stack segment growns down
485 dw - disabled write to the mapped file
486 lo - pages are locked in memory
487 io - memory mapped I/O area
488 sr - sequential read advise provided
489 rr - random read advise provided
490 dc - do not copy area on fork
491 de - do not expand area on remapping
492 ac - area is accountable
493 nr - swap space is not reserved for the area
494 ht - area uses huge tlb pages
495 ar - architecture specific flag
496 dd - do not include area into core dump
499 hg - huge page advise flag
500 nh - no-huge page advise flag
501 mg - mergable advise flag
503 Note that there is no guarantee that every flag and associated mnemonic will
504 be present in all further kernel releases. Things get changed, the flags may
505 be vanished or the reverse -- new added. Interpretation of their meaning
506 might change in future as well. So each consumer of these flags has to
507 follow each specific kernel version for the exact semantic.
509 This file is only present if the CONFIG_MMU kernel configuration option is
512 Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
513 output can be achieved only in the single read call).
514 This typically manifests when doing partial reads of these files while the
515 memory map is being modified. Despite the races, we do provide the following
518 1) The mapped addresses never go backwards, which implies no two
519 regions will ever overlap.
520 2) If there is something at a given vaddr during the entirety of the
521 life of the smaps/maps walk, there will be some output for it.
524 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
525 bits on both physical and virtual pages associated with a process, and the
526 soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst
528 To clear the bits for all the pages associated with the process
529 > echo 1 > /proc/PID/clear_refs
531 To clear the bits for the anonymous pages associated with the process
532 > echo 2 > /proc/PID/clear_refs
534 To clear the bits for the file mapped pages associated with the process
535 > echo 3 > /proc/PID/clear_refs
537 To clear the soft-dirty bit
538 > echo 4 > /proc/PID/clear_refs
540 To reset the peak resident set size ("high water mark") to the process's
542 > echo 5 > /proc/PID/clear_refs
544 Any other value written to /proc/PID/clear_refs will have no effect.
546 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
547 using /proc/kpageflags and number of times a page is mapped using
548 /proc/kpagecount. For detailed explanation, see
549 Documentation/admin-guide/mm/pagemap.rst.
551 The /proc/pid/numa_maps is an extension based on maps, showing the memory
552 locality and binding policy, as well as the memory usage (in pages) of
553 each mapping. The output follows a general format where mapping details get
554 summarized separated by blank spaces, one mapping per each file line:
556 address policy mapping details
558 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
559 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
560 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
561 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
562 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
563 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
564 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
565 320698b000 default file=/lib64/libc-2.12.so
566 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
567 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
568 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
569 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
570 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
571 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
572 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
573 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
576 "address" is the starting address for the mapping;
577 "policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst);
578 "mapping details" summarizes mapping data such as mapping type, page usage counters,
579 node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
580 size, in KB, that is backing the mapping up.
585 Similar to the process entries, the kernel data files give information about
586 the running kernel. The files used to obtain this information are contained in
587 /proc and are listed in Table 1-5. Not all of these will be present in your
588 system. It depends on the kernel configuration and the loaded modules, which
589 files are there, and which are missing.
591 Table 1-5: Kernel info in /proc
592 ..............................................................................
594 apm Advanced power management info
595 buddyinfo Kernel memory allocator information (see text) (2.5)
596 bus Directory containing bus specific information
597 cmdline Kernel command line
598 cpuinfo Info about the CPU
599 devices Available devices (block and character)
600 dma Used DMS channels
601 filesystems Supported filesystems
602 driver Various drivers grouped here, currently rtc (2.4)
603 execdomains Execdomains, related to security (2.4)
604 fb Frame Buffer devices (2.4)
605 fs File system parameters, currently nfs/exports (2.4)
606 ide Directory containing info about the IDE subsystem
607 interrupts Interrupt usage
608 iomem Memory map (2.4)
609 ioports I/O port usage
610 irq Masks for irq to cpu affinity (2.4)(smp?)
611 isapnp ISA PnP (Plug&Play) Info (2.4)
612 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
614 ksyms Kernel symbol table
615 loadavg Load average of last 1, 5 & 15 minutes
619 modules List of loaded modules
620 mounts Mounted filesystems
621 net Networking info (see text)
622 pagetypeinfo Additional page allocator information (see text) (2.5)
623 partitions Table of partitions known to the system
624 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
625 decoupled by lspci (2.4)
627 scsi SCSI info (see text)
628 slabinfo Slab pool info
629 softirqs softirq usage
630 stat Overall statistics
631 swaps Swap space utilization
633 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
634 tty Info of tty drivers
635 uptime Wall clock since boot, combined idle time of all cpus
636 version Kernel version
637 video bttv info of video resources (2.4)
638 vmallocinfo Show vmalloced areas
639 ..............................................................................
641 You can, for example, check which interrupts are currently in use and what
642 they are used for by looking in the file /proc/interrupts:
644 > cat /proc/interrupts
646 0: 8728810 XT-PIC timer
647 1: 895 XT-PIC keyboard
649 3: 531695 XT-PIC aha152x
650 4: 2014133 XT-PIC serial
651 5: 44401 XT-PIC pcnet_cs
654 12: 182918 XT-PIC PS/2 Mouse
656 14: 1232265 XT-PIC ide0
660 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
661 output of a SMP machine):
663 > cat /proc/interrupts
666 0: 1243498 1214548 IO-APIC-edge timer
667 1: 8949 8958 IO-APIC-edge keyboard
668 2: 0 0 XT-PIC cascade
669 5: 11286 10161 IO-APIC-edge soundblaster
670 8: 1 0 IO-APIC-edge rtc
671 9: 27422 27407 IO-APIC-edge 3c503
672 12: 113645 113873 IO-APIC-edge PS/2 Mouse
674 14: 22491 24012 IO-APIC-edge ide0
675 15: 2183 2415 IO-APIC-edge ide1
676 17: 30564 30414 IO-APIC-level eth0
677 18: 177 164 IO-APIC-level bttv
682 NMI is incremented in this case because every timer interrupt generates a NMI
683 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
685 LOC is the local interrupt counter of the internal APIC of every CPU.
687 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
688 connects the CPUs in a SMP system. This means that an error has been detected,
689 the IO-APIC automatically retry the transmission, so it should not be a big
690 problem, but you should read the SMP-FAQ.
692 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
693 /proc/interrupts to display every IRQ vector in use by the system, not
694 just those considered 'most important'. The new vectors are:
696 THR -- interrupt raised when a machine check threshold counter
697 (typically counting ECC corrected errors of memory or cache) exceeds
698 a configurable threshold. Only available on some systems.
700 TRM -- a thermal event interrupt occurs when a temperature threshold
701 has been exceeded for the CPU. This interrupt may also be generated
702 when the temperature drops back to normal.
704 SPU -- a spurious interrupt is some interrupt that was raised then lowered
705 by some IO device before it could be fully processed by the APIC. Hence
706 the APIC sees the interrupt but does not know what device it came from.
707 For this case the APIC will generate the interrupt with a IRQ vector
708 of 0xff. This might also be generated by chipset bugs.
710 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
711 sent from one CPU to another per the needs of the OS. Typically,
712 their statistics are used by kernel developers and interested users to
713 determine the occurrence of interrupts of the given type.
715 The above IRQ vectors are displayed only when relevant. For example,
716 the threshold vector does not exist on x86_64 platforms. Others are
717 suppressed when the system is a uniprocessor. As of this writing, only
718 i386 and x86_64 platforms support the new IRQ vector displays.
720 Of some interest is the introduction of the /proc/irq directory to 2.4.
721 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
722 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
723 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
728 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
729 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
733 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
734 IRQ, you can set it by doing:
736 > echo 1 > /proc/irq/10/smp_affinity
738 This means that only the first CPU will handle the IRQ, but you can also echo
739 5 which means that only the first and third CPU can handle the IRQ.
741 The contents of each smp_affinity file is the same by default:
743 > cat /proc/irq/0/smp_affinity
746 There is an alternate interface, smp_affinity_list which allows specifying
747 a cpu range instead of a bitmask:
749 > cat /proc/irq/0/smp_affinity_list
752 The default_smp_affinity mask applies to all non-active IRQs, which are the
753 IRQs which have not yet been allocated/activated, and hence which lack a
754 /proc/irq/[0-9]* directory.
756 The node file on an SMP system shows the node to which the device using the IRQ
757 reports itself as being attached. This hardware locality information does not
758 include information about any possible driver locality preference.
760 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
761 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
763 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
764 between all the CPUs which are allowed to handle it. As usual the kernel has
765 more info than you and does a better job than you, so the defaults are the
766 best choice for almost everyone. [Note this applies only to those IO-APIC's
767 that support "Round Robin" interrupt distribution.]
769 There are three more important subdirectories in /proc: net, scsi, and sys.
770 The general rule is that the contents, or even the existence of these
771 directories, depend on your kernel configuration. If SCSI is not enabled, the
772 directory scsi may not exist. The same is true with the net, which is there
773 only when networking support is present in the running kernel.
775 The slabinfo file gives information about memory usage at the slab level.
776 Linux uses slab pools for memory management above page level in version 2.2.
777 Commonly used objects have their own slab pool (such as network buffers,
778 directory cache, and so on).
780 ..............................................................................
782 > cat /proc/buddyinfo
784 Node 0, zone DMA 0 4 5 4 4 3 ...
785 Node 0, zone Normal 1 0 0 1 101 8 ...
786 Node 0, zone HighMem 2 0 0 1 1 0 ...
788 External fragmentation is a problem under some workloads, and buddyinfo is a
789 useful tool for helping diagnose these problems. Buddyinfo will give you a
790 clue as to how big an area you can safely allocate, or why a previous
793 Each column represents the number of pages of a certain order which are
794 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
795 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
796 available in ZONE_NORMAL, etc...
798 More information relevant to external fragmentation can be found in
801 > cat /proc/pagetypeinfo
805 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
806 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
807 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
808 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
809 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
810 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
811 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
812 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
813 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
814 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
815 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
817 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
818 Node 0, zone DMA 2 0 5 1 0
819 Node 0, zone DMA32 41 6 967 2 0
821 Fragmentation avoidance in the kernel works by grouping pages of different
822 migrate types into the same contiguous regions of memory called page blocks.
823 A page block is typically the size of the default hugepage size e.g. 2MB on
824 X86-64. By keeping pages grouped based on their ability to move, the kernel
825 can reclaim pages within a page block to satisfy a high-order allocation.
827 The pagetypinfo begins with information on the size of a page block. It
828 then gives the same type of information as buddyinfo except broken down
829 by migrate-type and finishes with details on how many page blocks of each
832 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
833 from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
834 make an estimate of the likely number of huge pages that can be allocated
835 at a given point in time. All the "Movable" blocks should be allocatable
836 unless memory has been mlock()'d. Some of the Reclaimable blocks should
837 also be allocatable although a lot of filesystem metadata may have to be
838 reclaimed to achieve this.
840 ..............................................................................
844 Provides information about distribution and utilization of memory. This
845 varies by architecture and compile options. The following is from a
846 16GB PIII, which has highmem enabled. You may not have all of these fields.
850 MemTotal: 16344972 kB
852 MemAvailable: 14836172 kB
858 HighTotal: 15597528 kB
859 HighFree: 13629632 kB
869 KReclaimable: 168048 kB
871 SReclaimable: 159856 kB
872 SUnreclaim: 124508 kB
877 CommitLimit: 7669796 kB
878 Committed_AS: 100056 kB
879 VmallocTotal: 112216 kB
881 VmallocChunk: 111088 kB
883 HardwareCorrupted: 0 kB
884 AnonHugePages: 49152 kB
889 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
890 bits and the kernel binary code)
891 MemFree: The sum of LowFree+HighFree
892 MemAvailable: An estimate of how much memory is available for starting new
893 applications, without swapping. Calculated from MemFree,
894 SReclaimable, the size of the file LRU lists, and the low
895 watermarks in each zone.
896 The estimate takes into account that the system needs some
897 page cache to function well, and that not all reclaimable
898 slab will be reclaimable, due to items being in use. The
899 impact of those factors will vary from system to system.
900 Buffers: Relatively temporary storage for raw disk blocks
901 shouldn't get tremendously large (20MB or so)
902 Cached: in-memory cache for files read from the disk (the
903 pagecache). Doesn't include SwapCached
904 SwapCached: Memory that once was swapped out, is swapped back in but
905 still also is in the swapfile (if memory is needed it
906 doesn't need to be swapped out AGAIN because it is already
907 in the swapfile. This saves I/O)
908 Active: Memory that has been used more recently and usually not
909 reclaimed unless absolutely necessary.
910 Inactive: Memory which has been less recently used. It is more
911 eligible to be reclaimed for other purposes
913 HighFree: Highmem is all memory above ~860MB of physical memory
914 Highmem areas are for use by userspace programs, or
915 for the pagecache. The kernel must use tricks to access
916 this memory, making it slower to access than lowmem.
918 LowFree: Lowmem is memory which can be used for everything that
919 highmem can be used for, but it is also available for the
920 kernel's use for its own data structures. Among many
921 other things, it is where everything from the Slab is
922 allocated. Bad things happen when you're out of lowmem.
923 SwapTotal: total amount of swap space available
924 SwapFree: Memory which has been evicted from RAM, and is temporarily
926 Dirty: Memory which is waiting to get written back to the disk
927 Writeback: Memory which is actively being written back to the disk
928 AnonPages: Non-file backed pages mapped into userspace page tables
929 HardwareCorrupted: The amount of RAM/memory in KB, the kernel identifies as
931 AnonHugePages: Non-file backed huge pages mapped into userspace page tables
932 Mapped: files which have been mmaped, such as libraries
933 Shmem: Total memory used by shared memory (shmem) and tmpfs
934 ShmemHugePages: Memory used by shared memory (shmem) and tmpfs allocated
936 ShmemPmdMapped: Shared memory mapped into userspace with huge pages
937 KReclaimable: Kernel allocations that the kernel will attempt to reclaim
938 under memory pressure. Includes SReclaimable (below), and other
939 direct allocations with a shrinker.
940 Slab: in-kernel data structures cache
941 SReclaimable: Part of Slab, that might be reclaimed, such as caches
942 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
943 PageTables: amount of memory dedicated to the lowest level of page
945 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
947 Bounce: Memory used for block device "bounce buffers"
948 WritebackTmp: Memory used by FUSE for temporary writeback buffers
949 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
950 this is the total amount of memory currently available to
951 be allocated on the system. This limit is only adhered to
952 if strict overcommit accounting is enabled (mode 2 in
953 'vm.overcommit_memory').
954 The CommitLimit is calculated with the following formula:
955 CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
956 overcommit_ratio / 100 + [total swap pages]
957 For example, on a system with 1G of physical RAM and 7G
958 of swap with a `vm.overcommit_ratio` of 30 it would
959 yield a CommitLimit of 7.3G.
960 For more details, see the memory overcommit documentation
961 in vm/overcommit-accounting.
962 Committed_AS: The amount of memory presently allocated on the system.
963 The committed memory is a sum of all of the memory which
964 has been allocated by processes, even if it has not been
965 "used" by them as of yet. A process which malloc()'s 1G
966 of memory, but only touches 300M of it will show up as
967 using 1G. This 1G is memory which has been "committed" to
968 by the VM and can be used at any time by the allocating
969 application. With strict overcommit enabled on the system
970 (mode 2 in 'vm.overcommit_memory'),allocations which would
971 exceed the CommitLimit (detailed above) will not be permitted.
972 This is useful if one needs to guarantee that processes will
973 not fail due to lack of memory once that memory has been
974 successfully allocated.
975 VmallocTotal: total size of vmalloc memory area
976 VmallocUsed: amount of vmalloc area which is used
977 VmallocChunk: largest contiguous block of vmalloc area which is free
978 Percpu: Memory allocated to the percpu allocator used to back percpu
979 allocations. This stat excludes the cost of metadata.
981 ..............................................................................
985 Provides information about vmalloced/vmaped areas. One line per area,
986 containing the virtual address range of the area, size in bytes,
987 caller information of the creator, and optional information depending
988 on the kind of area :
990 pages=nr number of pages
991 phys=addr if a physical address was specified
992 ioremap I/O mapping (ioremap() and friends)
993 vmalloc vmalloc() area
996 vpages buffer for pages pointers was vmalloced (huge area)
997 N<node>=nr (Only on NUMA kernels)
998 Number of pages allocated on memory node <node>
1000 > cat /proc/vmallocinfo
1001 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
1002 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
1003 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
1004 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
1005 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
1006 phys=7fee8000 ioremap
1007 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
1008 phys=7fee7000 ioremap
1009 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
1010 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
1011 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
1012 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
1013 pages=2 vmalloc N1=2
1014 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
1015 /0x130 [x_tables] pages=4 vmalloc N0=4
1016 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
1017 pages=14 vmalloc N2=14
1018 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
1019 pages=4 vmalloc N1=4
1020 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
1021 pages=2 vmalloc N1=2
1022 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
1023 pages=10 vmalloc N0=10
1025 ..............................................................................
1029 Provides counts of softirq handlers serviced since boot time, for each cpu.
1031 > cat /proc/softirqs
1034 TIMER: 27166 27120 27097 27034
1039 SCHED: 27035 26983 26971 26746
1041 RCU: 1678 1769 2178 2250
1044 1.3 IDE devices in /proc/ide
1045 ----------------------------
1047 The subdirectory /proc/ide contains information about all IDE devices of which
1048 the kernel is aware. There is one subdirectory for each IDE controller, the
1049 file drivers and a link for each IDE device, pointing to the device directory
1050 in the controller specific subtree.
1052 The file drivers contains general information about the drivers used for the
1055 > cat /proc/ide/drivers
1056 ide-cdrom version 4.53
1057 ide-disk version 1.08
1059 More detailed information can be found in the controller specific
1060 subdirectories. These are named ide0, ide1 and so on. Each of these
1061 directories contains the files shown in table 1-6.
1064 Table 1-6: IDE controller info in /proc/ide/ide?
1065 ..............................................................................
1067 channel IDE channel (0 or 1)
1068 config Configuration (only for PCI/IDE bridge)
1070 model Type/Chipset of IDE controller
1071 ..............................................................................
1073 Each device connected to a controller has a separate subdirectory in the
1074 controllers directory. The files listed in table 1-7 are contained in these
1078 Table 1-7: IDE device information
1079 ..............................................................................
1082 capacity Capacity of the medium (in 512Byte blocks)
1083 driver driver and version
1084 geometry physical and logical geometry
1085 identify device identify block
1087 model device identifier
1088 settings device setup
1089 smart_thresholds IDE disk management thresholds
1090 smart_values IDE disk management values
1091 ..............................................................................
1093 The most interesting file is settings. This file contains a nice overview of
1094 the drive parameters:
1096 # cat /proc/ide/ide0/hda/settings
1097 name value min max mode
1098 ---- ----- --- --- ----
1099 bios_cyl 526 0 65535 rw
1100 bios_head 255 0 255 rw
1101 bios_sect 63 0 63 rw
1102 breada_readahead 4 0 127 rw
1104 file_readahead 72 0 2097151 rw
1106 keepsettings 0 0 1 rw
1107 max_kb_per_request 122 1 127 rw
1111 pio_mode write-only 0 255 w
1117 1.4 Networking info in /proc/net
1118 --------------------------------
1120 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
1121 additional values you get for IP version 6 if you configure the kernel to
1122 support this. Table 1-9 lists the files and their meaning.
1125 Table 1-8: IPv6 info in /proc/net
1126 ..............................................................................
1128 udp6 UDP sockets (IPv6)
1129 tcp6 TCP sockets (IPv6)
1130 raw6 Raw device statistics (IPv6)
1131 igmp6 IP multicast addresses, which this host joined (IPv6)
1132 if_inet6 List of IPv6 interface addresses
1133 ipv6_route Kernel routing table for IPv6
1134 rt6_stats Global IPv6 routing tables statistics
1135 sockstat6 Socket statistics (IPv6)
1136 snmp6 Snmp data (IPv6)
1137 ..............................................................................
1140 Table 1-9: Network info in /proc/net
1141 ..............................................................................
1143 arp Kernel ARP table
1144 dev network devices with statistics
1145 dev_mcast the Layer2 multicast groups a device is listening too
1146 (interface index, label, number of references, number of bound
1148 dev_stat network device status
1149 ip_fwchains Firewall chain linkage
1150 ip_fwnames Firewall chain names
1151 ip_masq Directory containing the masquerading tables
1152 ip_masquerade Major masquerading table
1153 netstat Network statistics
1154 raw raw device statistics
1155 route Kernel routing table
1156 rpc Directory containing rpc info
1157 rt_cache Routing cache
1159 sockstat Socket statistics
1162 unix UNIX domain sockets
1163 wireless Wireless interface data (Wavelan etc)
1164 igmp IP multicast addresses, which this host joined
1165 psched Global packet scheduler parameters.
1166 netlink List of PF_NETLINK sockets
1167 ip_mr_vifs List of multicast virtual interfaces
1168 ip_mr_cache List of multicast routing cache
1169 ..............................................................................
1171 You can use this information to see which network devices are available in
1172 your system and how much traffic was routed over those devices:
1175 Inter-|Receive |[...
1176 face |bytes packets errs drop fifo frame compressed multicast|[...
1177 lo: 908188 5596 0 0 0 0 0 0 [...
1178 ppp0:15475140 20721 410 0 0 410 0 0 [...
1179 eth0: 614530 7085 0 0 0 0 0 1 [...
1182 ...] bytes packets errs drop fifo colls carrier compressed
1183 ...] 908188 5596 0 0 0 0 0 0
1184 ...] 1375103 17405 0 0 0 0 0 0
1185 ...] 1703981 5535 0 0 0 3 0 0
1187 In addition, each Channel Bond interface has its own directory. For
1188 example, the bond0 device will have a directory called /proc/net/bond0/.
1189 It will contain information that is specific to that bond, such as the
1190 current slaves of the bond, the link status of the slaves, and how
1191 many times the slaves link has failed.
1196 If you have a SCSI host adapter in your system, you'll find a subdirectory
1197 named after the driver for this adapter in /proc/scsi. You'll also see a list
1198 of all recognized SCSI devices in /proc/scsi:
1200 >cat /proc/scsi/scsi
1202 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1203 Vendor: IBM Model: DGHS09U Rev: 03E0
1204 Type: Direct-Access ANSI SCSI revision: 03
1205 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1206 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1207 Type: CD-ROM ANSI SCSI revision: 02
1210 The directory named after the driver has one file for each adapter found in
1211 the system. These files contain information about the controller, including
1212 the used IRQ and the IO address range. The amount of information shown is
1213 dependent on the adapter you use. The example shows the output for an Adaptec
1214 AHA-2940 SCSI adapter:
1216 > cat /proc/scsi/aic7xxx/0
1218 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1220 TCQ Enabled By Default : Disabled
1221 AIC7XXX_PROC_STATS : Disabled
1222 AIC7XXX_RESET_DELAY : 5
1223 Adapter Configuration:
1224 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1225 Ultra Wide Controller
1226 PCI MMAPed I/O Base: 0xeb001000
1227 Adapter SEEPROM Config: SEEPROM found and used.
1228 Adaptec SCSI BIOS: Enabled
1230 SCBs: Active 0, Max Active 2,
1231 Allocated 15, HW 16, Page 255
1233 BIOS Control Word: 0x18b6
1234 Adapter Control Word: 0x005b
1235 Extended Translation: Enabled
1236 Disconnect Enable Flags: 0xffff
1237 Ultra Enable Flags: 0x0001
1238 Tag Queue Enable Flags: 0x0000
1239 Ordered Queue Tag Flags: 0x0000
1240 Default Tag Queue Depth: 8
1241 Tagged Queue By Device array for aic7xxx host instance 0:
1242 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1243 Actual queue depth per device for aic7xxx host instance 0:
1244 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1247 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1248 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1249 Total transfers 160151 (74577 reads and 85574 writes)
1251 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1252 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1253 Total transfers 0 (0 reads and 0 writes)
1256 1.6 Parallel port info in /proc/parport
1257 ---------------------------------------
1259 The directory /proc/parport contains information about the parallel ports of
1260 your system. It has one subdirectory for each port, named after the port
1263 These directories contain the four files shown in Table 1-10.
1266 Table 1-10: Files in /proc/parport
1267 ..............................................................................
1269 autoprobe Any IEEE-1284 device ID information that has been acquired.
1270 devices list of the device drivers using that port. A + will appear by the
1271 name of the device currently using the port (it might not appear
1273 hardware Parallel port's base address, IRQ line and DMA channel.
1274 irq IRQ that parport is using for that port. This is in a separate
1275 file to allow you to alter it by writing a new value in (IRQ
1277 ..............................................................................
1279 1.7 TTY info in /proc/tty
1280 -------------------------
1282 Information about the available and actually used tty's can be found in the
1283 directory /proc/tty.You'll find entries for drivers and line disciplines in
1284 this directory, as shown in Table 1-11.
1287 Table 1-11: Files in /proc/tty
1288 ..............................................................................
1290 drivers list of drivers and their usage
1291 ldiscs registered line disciplines
1292 driver/serial usage statistic and status of single tty lines
1293 ..............................................................................
1295 To see which tty's are currently in use, you can simply look into the file
1298 > cat /proc/tty/drivers
1299 pty_slave /dev/pts 136 0-255 pty:slave
1300 pty_master /dev/ptm 128 0-255 pty:master
1301 pty_slave /dev/ttyp 3 0-255 pty:slave
1302 pty_master /dev/pty 2 0-255 pty:master
1303 serial /dev/cua 5 64-67 serial:callout
1304 serial /dev/ttyS 4 64-67 serial
1305 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1306 /dev/ptmx /dev/ptmx 5 2 system
1307 /dev/console /dev/console 5 1 system:console
1308 /dev/tty /dev/tty 5 0 system:/dev/tty
1309 unknown /dev/tty 4 1-63 console
1312 1.8 Miscellaneous kernel statistics in /proc/stat
1313 -------------------------------------------------
1315 Various pieces of information about kernel activity are available in the
1316 /proc/stat file. All of the numbers reported in this file are aggregates
1317 since the system first booted. For a quick look, simply cat the file:
1320 cpu 2255 34 2290 22625563 6290 127 456 0 0 0
1321 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0
1322 cpu1 1123 0 849 11313845 2614 0 18 0 0 0
1323 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1329 softirq 183433 0 21755 12 39 1137 231 21459 2263
1331 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1332 lines. These numbers identify the amount of time the CPU has spent performing
1333 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1334 second). The meanings of the columns are as follows, from left to right:
1336 - user: normal processes executing in user mode
1337 - nice: niced processes executing in user mode
1338 - system: processes executing in kernel mode
1339 - idle: twiddling thumbs
1340 - iowait: In a word, iowait stands for waiting for I/O to complete. But there
1341 are several problems:
1342 1. Cpu will not wait for I/O to complete, iowait is the time that a task is
1343 waiting for I/O to complete. When cpu goes into idle state for
1344 outstanding task io, another task will be scheduled on this CPU.
1345 2. In a multi-core CPU, the task waiting for I/O to complete is not running
1346 on any CPU, so the iowait of each CPU is difficult to calculate.
1347 3. The value of iowait field in /proc/stat will decrease in certain
1349 So, the iowait is not reliable by reading from /proc/stat.
1350 - irq: servicing interrupts
1351 - softirq: servicing softirqs
1352 - steal: involuntary wait
1353 - guest: running a normal guest
1354 - guest_nice: running a niced guest
1356 The "intr" line gives counts of interrupts serviced since boot time, for each
1357 of the possible system interrupts. The first column is the total of all
1358 interrupts serviced including unnumbered architecture specific interrupts;
1359 each subsequent column is the total for that particular numbered interrupt.
1360 Unnumbered interrupts are not shown, only summed into the total.
1362 The "ctxt" line gives the total number of context switches across all CPUs.
1364 The "btime" line gives the time at which the system booted, in seconds since
1367 The "processes" line gives the number of processes and threads created, which
1368 includes (but is not limited to) those created by calls to the fork() and
1369 clone() system calls.
1371 The "procs_running" line gives the total number of threads that are
1372 running or ready to run (i.e., the total number of runnable threads).
1374 The "procs_blocked" line gives the number of processes currently blocked,
1375 waiting for I/O to complete.
1377 The "softirq" line gives counts of softirqs serviced since boot time, for each
1378 of the possible system softirqs. The first column is the total of all
1379 softirqs serviced; each subsequent column is the total for that particular
1383 1.9 Ext4 file system parameters
1384 -------------------------------
1386 Information about mounted ext4 file systems can be found in
1387 /proc/fs/ext4. Each mounted filesystem will have a directory in
1388 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1389 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1390 in Table 1-12, below.
1392 Table 1-12: Files in /proc/fs/ext4/<devname>
1393 ..............................................................................
1395 mb_groups details of multiblock allocator buddy cache of free blocks
1396 ..............................................................................
1400 Shows registered system console lines.
1402 To see which character device lines are currently used for the system console
1403 /dev/console, you may simply look into the file /proc/consoles:
1405 > cat /proc/consoles
1411 device name of the device
1412 operations R = can do read operations
1413 W = can do write operations
1415 flags E = it is enabled
1416 C = it is preferred console
1417 B = it is primary boot console
1418 p = it is used for printk buffer
1419 b = it is not a TTY but a Braille device
1420 a = it is safe to use when cpu is offline
1421 major:minor major and minor number of the device separated by a colon
1423 ------------------------------------------------------------------------------
1425 ------------------------------------------------------------------------------
1426 The /proc file system serves information about the running system. It not only
1427 allows access to process data but also allows you to request the kernel status
1428 by reading files in the hierarchy.
1430 The directory structure of /proc reflects the types of information and makes
1431 it easy, if not obvious, where to look for specific data.
1432 ------------------------------------------------------------------------------
1434 ------------------------------------------------------------------------------
1435 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1436 ------------------------------------------------------------------------------
1438 ------------------------------------------------------------------------------
1440 ------------------------------------------------------------------------------
1441 * Modifying kernel parameters by writing into files found in /proc/sys
1442 * Exploring the files which modify certain parameters
1443 * Review of the /proc/sys file tree
1444 ------------------------------------------------------------------------------
1447 A very interesting part of /proc is the directory /proc/sys. This is not only
1448 a source of information, it also allows you to change parameters within the
1449 kernel. Be very careful when attempting this. You can optimize your system,
1450 but you can also cause it to crash. Never alter kernel parameters on a
1451 production system. Set up a development machine and test to make sure that
1452 everything works the way you want it to. You may have no alternative but to
1453 reboot the machine once an error has been made.
1455 To change a value, simply echo the new value into the file. An example is
1456 given below in the section on the file system data. You need to be root to do
1457 this. You can create your own boot script to perform this every time your
1460 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1461 general things in the operation of the Linux kernel. Since some of the files
1462 can inadvertently disrupt your system, it is advisable to read both
1463 documentation and source before actually making adjustments. In any case, be
1464 very careful when writing to any of these files. The entries in /proc may
1465 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1466 review the kernel documentation in the directory /usr/src/linux/Documentation.
1467 This chapter is heavily based on the documentation included in the pre 2.2
1468 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1470 Please see: Documentation/sysctl/ directory for descriptions of these
1473 ------------------------------------------------------------------------------
1475 ------------------------------------------------------------------------------
1476 Certain aspects of kernel behavior can be modified at runtime, without the
1477 need to recompile the kernel, or even to reboot the system. The files in the
1478 /proc/sys tree can not only be read, but also modified. You can use the echo
1479 command to write value into these files, thereby changing the default settings
1481 ------------------------------------------------------------------------------
1483 ------------------------------------------------------------------------------
1484 CHAPTER 3: PER-PROCESS PARAMETERS
1485 ------------------------------------------------------------------------------
1487 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1488 --------------------------------------------------------------------------------
1490 These file can be used to adjust the badness heuristic used to select which
1491 process gets killed in out of memory conditions.
1493 The badness heuristic assigns a value to each candidate task ranging from 0
1494 (never kill) to 1000 (always kill) to determine which process is targeted. The
1495 units are roughly a proportion along that range of allowed memory the process
1496 may allocate from based on an estimation of its current memory and swap use.
1497 For example, if a task is using all allowed memory, its badness score will be
1498 1000. If it is using half of its allowed memory, its score will be 500.
1500 There is an additional factor included in the badness score: the current memory
1501 and swap usage is discounted by 3% for root processes.
1503 The amount of "allowed" memory depends on the context in which the oom killer
1504 was called. If it is due to the memory assigned to the allocating task's cpuset
1505 being exhausted, the allowed memory represents the set of mems assigned to that
1506 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1507 memory represents the set of mempolicy nodes. If it is due to a memory
1508 limit (or swap limit) being reached, the allowed memory is that configured
1509 limit. Finally, if it is due to the entire system being out of memory, the
1510 allowed memory represents all allocatable resources.
1512 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1513 is used to determine which task to kill. Acceptable values range from -1000
1514 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1515 polarize the preference for oom killing either by always preferring a certain
1516 task or completely disabling it. The lowest possible value, -1000, is
1517 equivalent to disabling oom killing entirely for that task since it will always
1518 report a badness score of 0.
1520 Consequently, it is very simple for userspace to define the amount of memory to
1521 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1522 example, is roughly equivalent to allowing the remainder of tasks sharing the
1523 same system, cpuset, mempolicy, or memory controller resources to use at least
1524 50% more memory. A value of -500, on the other hand, would be roughly
1525 equivalent to discounting 50% of the task's allowed memory from being considered
1526 as scoring against the task.
1528 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1529 be used to tune the badness score. Its acceptable values range from -16
1530 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1531 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1532 scaled linearly with /proc/<pid>/oom_score_adj.
1534 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1535 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1536 requires CAP_SYS_RESOURCE.
1538 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1539 generation children with separate address spaces instead, if possible. This
1540 avoids servers and important system daemons from being killed and loses the
1541 minimal amount of work.
1544 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1545 -------------------------------------------------------------
1547 This file can be used to check the current score used by the oom-killer is for
1548 any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1549 process should be killed in an out-of-memory situation.
1552 3.3 /proc/<pid>/io - Display the IO accounting fields
1553 -------------------------------------------------------
1555 This file contains IO statistics for each running process
1560 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1563 test:/tmp # cat /proc/3828/io
1569 write_bytes: 323932160
1570 cancelled_write_bytes: 0
1579 I/O counter: chars read
1580 The number of bytes which this task has caused to be read from storage. This
1581 is simply the sum of bytes which this process passed to read() and pread().
1582 It includes things like tty IO and it is unaffected by whether or not actual
1583 physical disk IO was required (the read might have been satisfied from
1590 I/O counter: chars written
1591 The number of bytes which this task has caused, or shall cause to be written
1592 to disk. Similar caveats apply here as with rchar.
1598 I/O counter: read syscalls
1599 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1606 I/O counter: write syscalls
1607 Attempt to count the number of write I/O operations, i.e. syscalls like
1608 write() and pwrite().
1614 I/O counter: bytes read
1615 Attempt to count the number of bytes which this process really did cause to
1616 be fetched from the storage layer. Done at the submit_bio() level, so it is
1617 accurate for block-backed filesystems. <please add status regarding NFS and
1618 CIFS at a later time>
1624 I/O counter: bytes written
1625 Attempt to count the number of bytes which this process caused to be sent to
1626 the storage layer. This is done at page-dirtying time.
1629 cancelled_write_bytes
1630 ---------------------
1632 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1633 then deletes the file, it will in fact perform no writeout. But it will have
1634 been accounted as having caused 1MB of write.
1635 In other words: The number of bytes which this process caused to not happen,
1636 by truncating pagecache. A task can cause "negative" IO too. If this task
1637 truncates some dirty pagecache, some IO which another task has been accounted
1638 for (in its write_bytes) will not be happening. We _could_ just subtract that
1639 from the truncating task's write_bytes, but there is information loss in doing
1646 At its current implementation state, this is a bit racy on 32-bit machines: if
1647 process A reads process B's /proc/pid/io while process B is updating one of
1648 those 64-bit counters, process A could see an intermediate result.
1651 More information about this can be found within the taskstats documentation in
1652 Documentation/accounting.
1654 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1655 ---------------------------------------------------------------
1656 When a process is dumped, all anonymous memory is written to a core file as
1657 long as the size of the core file isn't limited. But sometimes we don't want
1658 to dump some memory segments, for example, huge shared memory or DAX.
1659 Conversely, sometimes we want to save file-backed memory segments into a core
1660 file, not only the individual files.
1662 /proc/<pid>/coredump_filter allows you to customize which memory segments
1663 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1664 of memory types. If a bit of the bitmask is set, memory segments of the
1665 corresponding memory type are dumped, otherwise they are not dumped.
1667 The following 9 memory types are supported:
1668 - (bit 0) anonymous private memory
1669 - (bit 1) anonymous shared memory
1670 - (bit 2) file-backed private memory
1671 - (bit 3) file-backed shared memory
1672 - (bit 4) ELF header pages in file-backed private memory areas (it is
1673 effective only if the bit 2 is cleared)
1674 - (bit 5) hugetlb private memory
1675 - (bit 6) hugetlb shared memory
1676 - (bit 7) DAX private memory
1677 - (bit 8) DAX shared memory
1679 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1680 are always dumped regardless of the bitmask status.
1682 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1683 only affected by bit 5-6, and DAX is only affected by bits 7-8.
1685 The default value of coredump_filter is 0x33; this means all anonymous memory
1686 segments, ELF header pages and hugetlb private memory are dumped.
1688 If you don't want to dump all shared memory segments attached to pid 1234,
1689 write 0x31 to the process's proc file.
1691 $ echo 0x31 > /proc/1234/coredump_filter
1693 When a new process is created, the process inherits the bitmask status from its
1694 parent. It is useful to set up coredump_filter before the program runs.
1697 $ echo 0x7 > /proc/self/coredump_filter
1700 3.5 /proc/<pid>/mountinfo - Information about mounts
1701 --------------------------------------------------------
1703 This file contains lines of the form:
1705 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1706 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1708 (1) mount ID: unique identifier of the mount (may be reused after umount)
1709 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1710 (3) major:minor: value of st_dev for files on filesystem
1711 (4) root: root of the mount within the filesystem
1712 (5) mount point: mount point relative to the process's root
1713 (6) mount options: per mount options
1714 (7) optional fields: zero or more fields of the form "tag[:value]"
1715 (8) separator: marks the end of the optional fields
1716 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1717 (10) mount source: filesystem specific information or "none"
1718 (11) super options: per super block options
1720 Parsers should ignore all unrecognised optional fields. Currently the
1721 possible optional fields are:
1723 shared:X mount is shared in peer group X
1724 master:X mount is slave to peer group X
1725 propagate_from:X mount is slave and receives propagation from peer group X (*)
1726 unbindable mount is unbindable
1728 (*) X is the closest dominant peer group under the process's root. If
1729 X is the immediate master of the mount, or if there's no dominant peer
1730 group under the same root, then only the "master:X" field is present
1731 and not the "propagate_from:X" field.
1733 For more information on mount propagation see:
1735 Documentation/filesystems/sharedsubtree.txt
1738 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1739 --------------------------------------------------------
1740 These files provide a method to access a tasks comm value. It also allows for
1741 a task to set its own or one of its thread siblings comm value. The comm value
1742 is limited in size compared to the cmdline value, so writing anything longer
1743 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1747 3.7 /proc/<pid>/task/<tid>/children - Information about task children
1748 -------------------------------------------------------------------------
1749 This file provides a fast way to retrieve first level children pids
1750 of a task pointed by <pid>/<tid> pair. The format is a space separated
1753 Note the "first level" here -- if a child has own children they will
1754 not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1755 to obtain the descendants.
1757 Since this interface is intended to be fast and cheap it doesn't
1758 guarantee to provide precise results and some children might be
1759 skipped, especially if they've exited right after we printed their
1760 pids, so one need to either stop or freeze processes being inspected
1761 if precise results are needed.
1764 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
1765 ---------------------------------------------------------------
1766 This file provides information associated with an opened file. The regular
1767 files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos'
1768 represents the current offset of the opened file in decimal form [see lseek(2)
1769 for details], 'flags' denotes the octal O_xxx mask the file has been
1770 created with [see open(2) for details] and 'mnt_id' represents mount ID of
1771 the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo
1780 All locks associated with a file descriptor are shown in its fdinfo too.
1782 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF
1784 The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1785 pair provide additional information particular to the objects they represent.
1794 where 'eventfd-count' is hex value of a counter.
1801 sigmask: 0000000000000200
1803 where 'sigmask' is hex value of the signal mask associated
1811 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7
1813 where 'tfd' is a target file descriptor number in decimal form,
1814 'events' is events mask being watched and the 'data' is data
1815 associated with a target [see epoll(7) for more details].
1817 The 'pos' is current offset of the target file in decimal form
1818 [see lseek(2)], 'ino' and 'sdev' are inode and device numbers
1819 where target file resides, all in hex format.
1823 For inotify files the format is the following
1827 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1829 where 'wd' is a watch descriptor in decimal form, ie a target file
1830 descriptor number, 'ino' and 'sdev' are inode and device where the
1831 target file resides and the 'mask' is the mask of events, all in hex
1832 form [see inotify(7) for more details].
1834 If the kernel was built with exportfs support, the path to the target
1835 file is encoded as a file handle. The file handle is provided by three
1836 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1839 If the kernel is built without exportfs support the file handle won't be
1842 If there is no inotify mark attached yet the 'inotify' line will be omitted.
1844 For fanotify files the format is
1849 fanotify flags:10 event-flags:0
1850 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1851 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
1853 where fanotify 'flags' and 'event-flags' are values used in fanotify_init
1854 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
1855 flags associated with mark which are tracked separately from events
1856 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
1857 mask and 'ignored_mask' is the mask of events which are to be ignored.
1858 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
1859 does provide information about flags and mask used in fanotify_mark
1860 call [see fsnotify manpage for details].
1862 While the first three lines are mandatory and always printed, the rest is
1863 optional and may be omitted if no marks created yet.
1874 it_value: (0, 49406829)
1877 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
1878 that have occurred [see timerfd_create(2) for details]. 'settime flags' are
1879 flags in octal form been used to setup the timer [see timerfd_settime(2) for
1880 details]. 'it_value' is remaining time until the timer exiration.
1881 'it_interval' is the interval for the timer. Note the timer might be set up
1882 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
1883 still exhibits timer's remaining time.
1885 3.9 /proc/<pid>/map_files - Information about memory mapped files
1886 ---------------------------------------------------------------------
1887 This directory contains symbolic links which represent memory mapped files
1888 the process is maintaining. Example output:
1890 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
1891 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
1892 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
1894 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
1895 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
1897 The name of a link represents the virtual memory bounds of a mapping, i.e.
1898 vm_area_struct::vm_start-vm_area_struct::vm_end.
1900 The main purpose of the map_files is to retrieve a set of memory mapped
1901 files in a fast way instead of parsing /proc/<pid>/maps or
1902 /proc/<pid>/smaps, both of which contain many more records. At the same
1903 time one can open(2) mappings from the listings of two processes and
1904 comparing their inode numbers to figure out which anonymous memory areas
1905 are actually shared.
1907 3.10 /proc/<pid>/timerslack_ns - Task timerslack value
1908 ---------------------------------------------------------
1909 This file provides the value of the task's timerslack value in nanoseconds.
1910 This value specifies a amount of time that normal timers may be deferred
1911 in order to coalesce timers and avoid unnecessary wakeups.
1913 This allows a task's interactivity vs power consumption trade off to be
1916 Writing 0 to the file will set the tasks timerslack to the default value.
1918 Valid values are from 0 - ULLONG_MAX
1920 An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
1921 permissions on the task specified to change its timerslack_ns value.
1923 3.11 /proc/<pid>/patch_state - Livepatch patch operation state
1924 -----------------------------------------------------------------
1925 When CONFIG_LIVEPATCH is enabled, this file displays the value of the
1926 patch state for the task.
1928 A value of '-1' indicates that no patch is in transition.
1930 A value of '0' indicates that a patch is in transition and the task is
1931 unpatched. If the patch is being enabled, then the task hasn't been
1932 patched yet. If the patch is being disabled, then the task has already
1935 A value of '1' indicates that a patch is in transition and the task is
1936 patched. If the patch is being enabled, then the task has already been
1937 patched. If the patch is being disabled, then the task hasn't been
1941 ------------------------------------------------------------------------------
1943 ------------------------------------------------------------------------------
1946 ---------------------
1948 The following mount options are supported:
1950 hidepid= Set /proc/<pid>/ access mode.
1951 gid= Set the group authorized to learn processes information.
1953 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1956 hidepid=1 means users may not access any /proc/<pid>/ directories but their
1957 own. Sensitive files like cmdline, sched*, status are now protected against
1958 other users. This makes it impossible to learn whether any user runs
1959 specific program (given the program doesn't reveal itself by its behaviour).
1960 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1961 poorly written programs passing sensitive information via program arguments are
1962 now protected against local eavesdroppers.
1964 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1965 users. It doesn't mean that it hides a fact whether a process with a specific
1966 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1967 but it hides process' uid and gid, which may be learned by stat()'ing
1968 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
1969 information about running processes, whether some daemon runs with elevated
1970 privileges, whether other user runs some sensitive program, whether other users
1971 run any program at all, etc.
1973 gid= defines a group authorized to learn processes information otherwise
1974 prohibited by hidepid=. If you use some daemon like identd which needs to learn
1975 information about processes information, just add identd to this group.