proc.txt 74 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394959697989910010110210310410510610710810911011111211311411511611711811912012112212312412512612712812913013113213313413513613713813914014114214314414514614714814915015115215315415515615715815916016116216316416516616716816917017117217317417517617717817918018118218318418518618718818919019119219319419519619719819920020120220320420520620720820921021121221321421521621721821922022122222322422522622722822923023123223323423523623723823924024124224324424524624724824925025125225325425525625725825926026126226326426526626726826927027127227327427527627727827928028128228328428528628728828929029129229329429529629729829930030130230330430530630730830931031131231331431531631731831932032132232332432532632732832933033133233333433533633733833934034134234334434534634734834935035135235335435535635735835936036136236336436536636736836937037137237337437537637737837938038138238338438538638738838939039139239339439539639739839940040140240340440540640740840941041141241341441541641741841942042142242342442542642742842943043143243343443543643743843944044144244344444544644744844945045145245345445545645745845946046146246346446546646746846947047147247347447547647747847948048148248348448548648748848949049149249349449549649749849950050150250350450550650750850951051151251351451551651751851952052152252352452552652752852953053153253353453553653753853954054154254354454554654754854955055155255355455555655755855956056156256356456556656756856957057157257357457557657757857958058158258358458558658758858959059159259359459559659759859960060160260360460560660760860961061161261361461561661761861962062162262362462562662762862963063163263363463563663763863964064164264364464564664764864965065165265365465565665765865966066166266366466566666766866967067167267367467567667767867968068168268368468568668768868969069169269369469569669769869970070170270370470570670770870971071171271371471571671771871972072172272372472572672772872973073173273373473573673773873974074174274374474574674774874975075175275375475575675775875976076176276376476576676776876977077177277377477577677777877978078178278378478578678778878979079179279379479579679779879980080180280380480580680780880981081181281381481581681781881982082182282382482582682782882983083183283383483583683783883984084184284384484584684784884985085185285385485585685785885986086186286386486586686786886987087187287387487587687787887988088188288388488588688788888989089189289389489589689789889990090190290390490590690790890991091191291391491591691791891992092192292392492592692792892993093193293393493593693793893994094194294394494594694794894995095195295395495595695795895996096196296396496596696796896997097197297397497597697797897998098198298398498598698798898999099199299399499599699799899910001001100210031004100510061007100810091010101110121013101410151016101710181019102010211022102310241025102610271028102910301031103210331034103510361037103810391040104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094109510961097109810991100110111021103110411051106110711081109111011111112111311141115111611171118111911201121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178117911801181118211831184118511861187118811891190119111921193119411951196119711981199120012011202120312041205120612071208120912101211121212131214121512161217121812191220122112221223122412251226122712281229123012311232123312341235123612371238123912401241124212431244124512461247124812491250125112521253125412551256125712581259126012611262126312641265126612671268126912701271127212731274127512761277127812791280128112821283128412851286128712881289129012911292129312941295129612971298129913001301130213031304130513061307130813091310131113121313131413151316131713181319132013211322132313241325132613271328132913301331133213331334133513361337133813391340134113421343134413451346134713481349135013511352135313541355135613571358135913601361136213631364136513661367136813691370137113721373137413751376137713781379138013811382138313841385138613871388138913901391139213931394139513961397139813991400140114021403140414051406140714081409141014111412141314141415141614171418141914201421142214231424142514261427142814291430143114321433143414351436143714381439144014411442144314441445144614471448144914501451145214531454145514561457145814591460146114621463146414651466146714681469147014711472147314741475147614771478147914801481148214831484148514861487148814891490149114921493149414951496149714981499150015011502150315041505150615071508150915101511151215131514151515161517151815191520152115221523152415251526152715281529153015311532153315341535153615371538153915401541154215431544154515461547154815491550155115521553155415551556155715581559156015611562156315641565156615671568156915701571157215731574157515761577157815791580158115821583158415851586158715881589159015911592159315941595159615971598159916001601160216031604160516061607160816091610161116121613161416151616161716181619162016211622162316241625162616271628162916301631163216331634163516361637
  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>
  6. 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
  7. move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009
  8. ------------------------------------------------------------------------------
  9. Version 1.3 Kernel version 2.2.12
  10. Kernel version 2.4.0-test11-pre4
  11. ------------------------------------------------------------------------------
  12. fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009
  13. Table of Contents
  14. -----------------
  15. 0 Preface
  16. 0.1 Introduction/Credits
  17. 0.2 Legal Stuff
  18. 1 Collecting System Information
  19. 1.1 Process-Specific Subdirectories
  20. 1.2 Kernel data
  21. 1.3 IDE devices in /proc/ide
  22. 1.4 Networking info in /proc/net
  23. 1.5 SCSI info
  24. 1.6 Parallel port info in /proc/parport
  25. 1.7 TTY info in /proc/tty
  26. 1.8 Miscellaneous kernel statistics in /proc/stat
  27. 1.9 Ext4 file system parameters
  28. 2 Modifying System Parameters
  29. 3 Per-Process Parameters
  30. 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer
  31. score
  32. 3.2 /proc/<pid>/oom_score - Display current oom-killer score
  33. 3.3 /proc/<pid>/io - Display the IO accounting fields
  34. 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
  35. 3.5 /proc/<pid>/mountinfo - Information about mounts
  36. 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
  37. 4 Configuring procfs
  38. 4.1 Mount options
  39. ------------------------------------------------------------------------------
  40. Preface
  41. ------------------------------------------------------------------------------
  42. 0.1 Introduction/Credits
  43. ------------------------
  44. This documentation is part of a soon (or so we hope) to be released book on
  45. the SuSE Linux distribution. As there is no complete documentation for the
  46. /proc file system and we've used many freely available sources to write these
  47. chapters, it seems only fair to give the work back to the Linux community.
  48. This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
  49. afraid it's still far from complete, but we hope it will be useful. As far as
  50. we know, it is the first 'all-in-one' document about the /proc file system. It
  51. is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
  52. SPARC, AXP, etc., features, you probably won't find what you are looking for.
  53. It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
  54. additions and patches are welcome and will be added to this document if you
  55. mail them to Bodo.
  56. We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
  57. other people for help compiling this documentation. We'd also like to extend a
  58. special thank you to Andi Kleen for documentation, which we relied on heavily
  59. to create this document, as well as the additional information he provided.
  60. Thanks to everybody else who contributed source or docs to the Linux kernel
  61. and helped create a great piece of software... :)
  62. If you have any comments, corrections or additions, please don't hesitate to
  63. contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
  64. document.
  65. The latest version of this document is available online at
  66. http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
  67. If the above direction does not works for you, you could try the kernel
  68. mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
  69. comandante@zaralinux.com.
  70. 0.2 Legal Stuff
  71. ---------------
  72. We don't guarantee the correctness of this document, and if you come to us
  73. complaining about how you screwed up your system because of incorrect
  74. documentation, we won't feel responsible...
  75. ------------------------------------------------------------------------------
  76. CHAPTER 1: COLLECTING SYSTEM INFORMATION
  77. ------------------------------------------------------------------------------
  78. ------------------------------------------------------------------------------
  79. In This Chapter
  80. ------------------------------------------------------------------------------
  81. * Investigating the properties of the pseudo file system /proc and its
  82. ability to provide information on the running Linux system
  83. * Examining /proc's structure
  84. * Uncovering various information about the kernel and the processes running
  85. on the system
  86. ------------------------------------------------------------------------------
  87. The proc file system acts as an interface to internal data structures in the
  88. kernel. It can be used to obtain information about the system and to change
  89. certain kernel parameters at runtime (sysctl).
  90. First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
  91. show you how you can use /proc/sys to change settings.
  92. 1.1 Process-Specific Subdirectories
  93. -----------------------------------
  94. The directory /proc contains (among other things) one subdirectory for each
  95. process running on the system, which is named after the process ID (PID).
  96. The link self points to the process reading the file system. Each process
  97. subdirectory has the entries listed in Table 1-1.
  98. Table 1-1: Process specific entries in /proc
  99. ..............................................................................
  100. File Content
  101. clear_refs Clears page referenced bits shown in smaps output
  102. cmdline Command line arguments
  103. cpu Current and last cpu in which it was executed (2.4)(smp)
  104. cwd Link to the current working directory
  105. environ Values of environment variables
  106. exe Link to the executable of this process
  107. fd Directory, which contains all file descriptors
  108. maps Memory maps to executables and library files (2.4)
  109. mem Memory held by this process
  110. root Link to the root directory of this process
  111. stat Process status
  112. statm Process memory status information
  113. status Process status in human readable form
  114. wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
  115. pagemap Page table
  116. stack Report full stack trace, enable via CONFIG_STACKTRACE
  117. smaps a extension based on maps, showing the memory consumption of
  118. each mapping
  119. ..............................................................................
  120. For example, to get the status information of a process, all you have to do is
  121. read the file /proc/PID/status:
  122. >cat /proc/self/status
  123. Name: cat
  124. State: R (running)
  125. Tgid: 5452
  126. Pid: 5452
  127. PPid: 743
  128. TracerPid: 0 (2.4)
  129. Uid: 501 501 501 501
  130. Gid: 100 100 100 100
  131. FDSize: 256
  132. Groups: 100 14 16
  133. VmPeak: 5004 kB
  134. VmSize: 5004 kB
  135. VmLck: 0 kB
  136. VmHWM: 476 kB
  137. VmRSS: 476 kB
  138. VmData: 156 kB
  139. VmStk: 88 kB
  140. VmExe: 68 kB
  141. VmLib: 1412 kB
  142. VmPTE: 20 kb
  143. VmSwap: 0 kB
  144. Threads: 1
  145. SigQ: 0/28578
  146. SigPnd: 0000000000000000
  147. ShdPnd: 0000000000000000
  148. SigBlk: 0000000000000000
  149. SigIgn: 0000000000000000
  150. SigCgt: 0000000000000000
  151. CapInh: 00000000fffffeff
  152. CapPrm: 0000000000000000
  153. CapEff: 0000000000000000
  154. CapBnd: ffffffffffffffff
  155. Seccomp: 0
  156. voluntary_ctxt_switches: 0
  157. nonvoluntary_ctxt_switches: 1
  158. This shows you nearly the same information you would get if you viewed it with
  159. the ps command. In fact, ps uses the proc file system to obtain its
  160. information. But you get a more detailed view of the process by reading the
  161. file /proc/PID/status. It fields are described in table 1-2.
  162. The statm file contains more detailed information about the process
  163. memory usage. Its seven fields are explained in Table 1-3. The stat file
  164. contains details information about the process itself. Its fields are
  165. explained in Table 1-4.
  166. (for SMP CONFIG users)
  167. For making accounting scalable, RSS related information are handled in
  168. asynchronous manner and the vaule may not be very precise. To see a precise
  169. snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
  170. It's slow but very precise.
  171. Table 1-2: Contents of the status files (as of 2.6.30-rc7)
  172. ..............................................................................
  173. Field Content
  174. Name filename of the executable
  175. State state (R is running, S is sleeping, D is sleeping
  176. in an uninterruptible wait, Z is zombie,
  177. T is traced or stopped)
  178. Tgid thread group ID
  179. Pid process id
  180. PPid process id of the parent process
  181. TracerPid PID of process tracing this process (0 if not)
  182. Uid Real, effective, saved set, and file system UIDs
  183. Gid Real, effective, saved set, and file system GIDs
  184. FDSize number of file descriptor slots currently allocated
  185. Groups supplementary group list
  186. VmPeak peak virtual memory size
  187. VmSize total program size
  188. VmLck locked memory size
  189. VmHWM peak resident set size ("high water mark")
  190. VmRSS size of memory portions
  191. VmData size of data, stack, and text segments
  192. VmStk size of data, stack, and text segments
  193. VmExe size of text segment
  194. VmLib size of shared library code
  195. VmPTE size of page table entries
  196. VmSwap size of swap usage (the number of referred swapents)
  197. Threads number of threads
  198. SigQ number of signals queued/max. number for queue
  199. SigPnd bitmap of pending signals for the thread
  200. ShdPnd bitmap of shared pending signals for the process
  201. SigBlk bitmap of blocked signals
  202. SigIgn bitmap of ignored signals
  203. SigCgt bitmap of catched signals
  204. CapInh bitmap of inheritable capabilities
  205. CapPrm bitmap of permitted capabilities
  206. CapEff bitmap of effective capabilities
  207. CapBnd bitmap of capabilities bounding set
  208. Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...)
  209. Cpus_allowed mask of CPUs on which this process may run
  210. Cpus_allowed_list Same as previous, but in "list format"
  211. Mems_allowed mask of memory nodes allowed to this process
  212. Mems_allowed_list Same as previous, but in "list format"
  213. voluntary_ctxt_switches number of voluntary context switches
  214. nonvoluntary_ctxt_switches number of non voluntary context switches
  215. ..............................................................................
  216. Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
  217. ..............................................................................
  218. Field Content
  219. size total program size (pages) (same as VmSize in status)
  220. resident size of memory portions (pages) (same as VmRSS in status)
  221. shared number of pages that are shared (i.e. backed by a file)
  222. trs number of pages that are 'code' (not including libs; broken,
  223. includes data segment)
  224. lrs number of pages of library (always 0 on 2.6)
  225. drs number of pages of data/stack (including libs; broken,
  226. includes library text)
  227. dt number of dirty pages (always 0 on 2.6)
  228. ..............................................................................
  229. Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
  230. ..............................................................................
  231. Field Content
  232. pid process id
  233. tcomm filename of the executable
  234. state state (R is running, S is sleeping, D is sleeping in an
  235. uninterruptible wait, Z is zombie, T is traced or stopped)
  236. ppid process id of the parent process
  237. pgrp pgrp of the process
  238. sid session id
  239. tty_nr tty the process uses
  240. tty_pgrp pgrp of the tty
  241. flags task flags
  242. min_flt number of minor faults
  243. cmin_flt number of minor faults with child's
  244. maj_flt number of major faults
  245. cmaj_flt number of major faults with child's
  246. utime user mode jiffies
  247. stime kernel mode jiffies
  248. cutime user mode jiffies with child's
  249. cstime kernel mode jiffies with child's
  250. priority priority level
  251. nice nice level
  252. num_threads number of threads
  253. it_real_value (obsolete, always 0)
  254. start_time time the process started after system boot
  255. vsize virtual memory size
  256. rss resident set memory size
  257. rsslim current limit in bytes on the rss
  258. start_code address above which program text can run
  259. end_code address below which program text can run
  260. start_stack address of the start of the main process stack
  261. esp current value of ESP
  262. eip current value of EIP
  263. pending bitmap of pending signals
  264. blocked bitmap of blocked signals
  265. sigign bitmap of ignored signals
  266. sigcatch bitmap of catched signals
  267. wchan address where process went to sleep
  268. 0 (place holder)
  269. 0 (place holder)
  270. exit_signal signal to send to parent thread on exit
  271. task_cpu which CPU the task is scheduled on
  272. rt_priority realtime priority
  273. policy scheduling policy (man sched_setscheduler)
  274. blkio_ticks time spent waiting for block IO
  275. gtime guest time of the task in jiffies
  276. cgtime guest time of the task children in jiffies
  277. start_data address above which program data+bss is placed
  278. end_data address below which program data+bss is placed
  279. start_brk address above which program heap can be expanded with brk()
  280. ..............................................................................
  281. The /proc/PID/maps file containing the currently mapped memory regions and
  282. their access permissions.
  283. The format is:
  284. address perms offset dev inode pathname
  285. 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
  286. 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
  287. 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
  288. a7cb1000-a7cb2000 ---p 00000000 00:00 0
  289. a7cb2000-a7eb2000 rw-p 00000000 00:00 0
  290. a7eb2000-a7eb3000 ---p 00000000 00:00 0
  291. a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack:1001]
  292. a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
  293. a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
  294. a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
  295. a800b000-a800e000 rw-p 00000000 00:00 0
  296. a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
  297. a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
  298. a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
  299. a8024000-a8027000 rw-p 00000000 00:00 0
  300. a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
  301. a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
  302. a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
  303. aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
  304. ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
  305. where "address" is the address space in the process that it occupies, "perms"
  306. is a set of permissions:
  307. r = read
  308. w = write
  309. x = execute
  310. s = shared
  311. p = private (copy on write)
  312. "offset" is the offset into the mapping, "dev" is the device (major:minor), and
  313. "inode" is the inode on that device. 0 indicates that no inode is associated
  314. with the memory region, as the case would be with BSS (uninitialized data).
  315. The "pathname" shows the name associated file for this mapping. If the mapping
  316. is not associated with a file:
  317. [heap] = the heap of the program
  318. [stack] = the stack of the main process
  319. [stack:1001] = the stack of the thread with tid 1001
  320. [vdso] = the "virtual dynamic shared object",
  321. the kernel system call handler
  322. [anon:<name>] = an anonymous mapping that has been
  323. named by userspace
  324. or if empty, the mapping is anonymous.
  325. The /proc/PID/task/TID/maps is a view of the virtual memory from the viewpoint
  326. of the individual tasks of a process. In this file you will see a mapping marked
  327. as [stack] if that task sees it as a stack. This is a key difference from the
  328. content of /proc/PID/maps, where you will see all mappings that are being used
  329. as stack by all of those tasks. Hence, for the example above, the task-level
  330. map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this:
  331. 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
  332. 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
  333. 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
  334. a7cb1000-a7cb2000 ---p 00000000 00:00 0
  335. a7cb2000-a7eb2000 rw-p 00000000 00:00 0
  336. a7eb2000-a7eb3000 ---p 00000000 00:00 0
  337. a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack]
  338. a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
  339. a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
  340. a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
  341. a800b000-a800e000 rw-p 00000000 00:00 0
  342. a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
  343. a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
  344. a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
  345. a8024000-a8027000 rw-p 00000000 00:00 0
  346. a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
  347. a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
  348. a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
  349. aff35000-aff4a000 rw-p 00000000 00:00 0
  350. ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
  351. The /proc/PID/smaps is an extension based on maps, showing the memory
  352. consumption for each of the process's mappings. For each of mappings there
  353. is a series of lines such as the following:
  354. 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
  355. Size: 1084 kB
  356. Rss: 892 kB
  357. Pss: 374 kB
  358. Shared_Clean: 892 kB
  359. Shared_Dirty: 0 kB
  360. Private_Clean: 0 kB
  361. Private_Dirty: 0 kB
  362. Referenced: 892 kB
  363. Anonymous: 0 kB
  364. Swap: 0 kB
  365. SwapPss: 0 kB
  366. KernelPageSize: 4 kB
  367. MMUPageSize: 4 kB
  368. Locked: 374 kB
  369. Name: name from userspace
  370. The first of these lines shows the same information as is displayed for the
  371. mapping in /proc/PID/maps. The remaining lines show the size of the mapping
  372. (size), the amount of the mapping that is currently resident in RAM (RSS), the
  373. process' proportional share of this mapping (PSS), the number of clean and
  374. dirty private pages in the mapping.
  375. The "proportional set size" (PSS) of a process is the count of pages it has
  376. in memory, where each page is divided by the number of processes sharing it.
  377. So if a process has 1000 pages all to itself, and 1000 shared with one other
  378. process, its PSS will be 1500.
  379. Note that even a page which is part of a MAP_SHARED mapping, but has only
  380. a single pte mapped, i.e. is currently used by only one process, is accounted
  381. as private and not as shared.
  382. "Referenced" indicates the amount of memory currently marked as referenced or
  383. accessed.
  384. "Anonymous" shows the amount of memory that does not belong to any file. Even
  385. a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
  386. and a page is modified, the file page is replaced by a private anonymous copy.
  387. "Swap" shows how much would-be-anonymous memory is also used, but out on
  388. swap.
  389. "SwapPss" shows proportional swap share of this mapping.
  390. The "Name" field will only be present on a mapping that has been named by
  391. userspace, and will show the name passed in by userspace.
  392. This file is only present if the CONFIG_MMU kernel configuration option is
  393. enabled.
  394. The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
  395. bits on both physical and virtual pages associated with a process.
  396. To clear the bits for all the pages associated with the process
  397. > echo 1 > /proc/PID/clear_refs
  398. To clear the bits for the anonymous pages associated with the process
  399. > echo 2 > /proc/PID/clear_refs
  400. To clear the bits for the file mapped pages associated with the process
  401. > echo 3 > /proc/PID/clear_refs
  402. Any other value written to /proc/PID/clear_refs will have no effect.
  403. To reset the peak resident set size ("high water mark") to the process's
  404. current value:
  405. > echo 5 > /proc/PID/clear_refs
  406. The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
  407. using /proc/kpageflags and number of times a page is mapped using
  408. /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
  409. 1.2 Kernel data
  410. ---------------
  411. Similar to the process entries, the kernel data files give information about
  412. the running kernel. The files used to obtain this information are contained in
  413. /proc and are listed in Table 1-5. Not all of these will be present in your
  414. system. It depends on the kernel configuration and the loaded modules, which
  415. files are there, and which are missing.
  416. Table 1-5: Kernel info in /proc
  417. ..............................................................................
  418. File Content
  419. apm Advanced power management info
  420. buddyinfo Kernel memory allocator information (see text) (2.5)
  421. bus Directory containing bus specific information
  422. cmdline Kernel command line
  423. cpuinfo Info about the CPU
  424. devices Available devices (block and character)
  425. dma Used DMS channels
  426. filesystems Supported filesystems
  427. driver Various drivers grouped here, currently rtc (2.4)
  428. execdomains Execdomains, related to security (2.4)
  429. fb Frame Buffer devices (2.4)
  430. fs File system parameters, currently nfs/exports (2.4)
  431. ide Directory containing info about the IDE subsystem
  432. interrupts Interrupt usage
  433. iomem Memory map (2.4)
  434. ioports I/O port usage
  435. irq Masks for irq to cpu affinity (2.4)(smp?)
  436. isapnp ISA PnP (Plug&Play) Info (2.4)
  437. kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
  438. kmsg Kernel messages
  439. ksyms Kernel symbol table
  440. loadavg Load average of last 1, 5 & 15 minutes
  441. locks Kernel locks
  442. meminfo Memory info
  443. misc Miscellaneous
  444. modules List of loaded modules
  445. mounts Mounted filesystems
  446. net Networking info (see text)
  447. pagetypeinfo Additional page allocator information (see text) (2.5)
  448. partitions Table of partitions known to the system
  449. pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
  450. decoupled by lspci (2.4)
  451. rtc Real time clock
  452. scsi SCSI info (see text)
  453. slabinfo Slab pool info
  454. softirqs softirq usage
  455. stat Overall statistics
  456. swaps Swap space utilization
  457. sys See chapter 2
  458. sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
  459. tty Info of tty drivers
  460. uptime System uptime
  461. version Kernel version
  462. video bttv info of video resources (2.4)
  463. vmallocinfo Show vmalloced areas
  464. ..............................................................................
  465. You can, for example, check which interrupts are currently in use and what
  466. they are used for by looking in the file /proc/interrupts:
  467. > cat /proc/interrupts
  468. CPU0
  469. 0: 8728810 XT-PIC timer
  470. 1: 895 XT-PIC keyboard
  471. 2: 0 XT-PIC cascade
  472. 3: 531695 XT-PIC aha152x
  473. 4: 2014133 XT-PIC serial
  474. 5: 44401 XT-PIC pcnet_cs
  475. 8: 2 XT-PIC rtc
  476. 11: 8 XT-PIC i82365
  477. 12: 182918 XT-PIC PS/2 Mouse
  478. 13: 1 XT-PIC fpu
  479. 14: 1232265 XT-PIC ide0
  480. 15: 7 XT-PIC ide1
  481. NMI: 0
  482. In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
  483. output of a SMP machine):
  484. > cat /proc/interrupts
  485. CPU0 CPU1
  486. 0: 1243498 1214548 IO-APIC-edge timer
  487. 1: 8949 8958 IO-APIC-edge keyboard
  488. 2: 0 0 XT-PIC cascade
  489. 5: 11286 10161 IO-APIC-edge soundblaster
  490. 8: 1 0 IO-APIC-edge rtc
  491. 9: 27422 27407 IO-APIC-edge 3c503
  492. 12: 113645 113873 IO-APIC-edge PS/2 Mouse
  493. 13: 0 0 XT-PIC fpu
  494. 14: 22491 24012 IO-APIC-edge ide0
  495. 15: 2183 2415 IO-APIC-edge ide1
  496. 17: 30564 30414 IO-APIC-level eth0
  497. 18: 177 164 IO-APIC-level bttv
  498. NMI: 2457961 2457959
  499. LOC: 2457882 2457881
  500. ERR: 2155
  501. NMI is incremented in this case because every timer interrupt generates a NMI
  502. (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
  503. LOC is the local interrupt counter of the internal APIC of every CPU.
  504. ERR is incremented in the case of errors in the IO-APIC bus (the bus that
  505. connects the CPUs in a SMP system. This means that an error has been detected,
  506. the IO-APIC automatically retry the transmission, so it should not be a big
  507. 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:
  511. THR -- interrupt raised when a machine check threshold counter
  512. (typically counting ECC corrected errors of memory or cache) exceeds
  513. a configurable threshold. Only available on some systems.
  514. TRM -- a thermal event interrupt occurs when a temperature threshold
  515. has been exceeded for the CPU. This interrupt may also be generated
  516. when the temperature drops back to normal.
  517. SPU -- a spurious interrupt is some interrupt that was raised then lowered
  518. by some IO device before it could be fully processed by the APIC. Hence
  519. the APIC sees the interrupt but does not know what device it came from.
  520. For this case the APIC will generate the interrupt with a IRQ vector
  521. of 0xff. This might also be generated by chipset bugs.
  522. RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
  523. sent from one CPU to another per the needs of the OS. Typically,
  524. their statistics are used by kernel developers and interested users to
  525. determine the occurrence of interrupts of the given type.
  526. The above IRQ vectors are displayed only when relevant. For example,
  527. the threshold vector does not exist on x86_64 platforms. Others are
  528. suppressed when the system is a uniprocessor. As of this writing, only
  529. i386 and x86_64 platforms support the new IRQ vector displays.
  530. Of some interest is the introduction of the /proc/irq directory to 2.4.
  531. It could be used to set IRQ to CPU affinity, this means that you can "hook" an
  532. IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
  533. irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
  534. prof_cpu_mask.
  535. For example
  536. > ls /proc/irq/
  537. 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
  538. 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
  539. > ls /proc/irq/0/
  540. smp_affinity
  541. smp_affinity is a bitmask, in which you can specify which CPUs can handle the
  542. IRQ, you can set it by doing:
  543. > echo 1 > /proc/irq/10/smp_affinity
  544. This means that only the first CPU will handle the IRQ, but you can also echo
  545. 5 which means that only the first and fourth CPU can handle the IRQ.
  546. The contents of each smp_affinity file is the same by default:
  547. > cat /proc/irq/0/smp_affinity
  548. ffffffff
  549. There is an alternate interface, smp_affinity_list which allows specifying
  550. a cpu range instead of a bitmask:
  551. > cat /proc/irq/0/smp_affinity_list
  552. 1024-1031
  553. The default_smp_affinity mask applies to all non-active IRQs, which are the
  554. IRQs which have not yet been allocated/activated, and hence which lack a
  555. /proc/irq/[0-9]* directory.
  556. The node file on an SMP system shows the node to which the device using the IRQ
  557. reports itself as being attached. This hardware locality information does not
  558. include information about any possible driver locality preference.
  559. prof_cpu_mask specifies which CPUs are to be profiled by the system wide
  560. profiler. Default value is ffffffff (all cpus if there are only 32 of them).
  561. The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
  562. between all the CPUs which are allowed to handle it. As usual the kernel has
  563. more info than you and does a better job than you, so the defaults are the
  564. best choice for almost everyone. [Note this applies only to those IO-APIC's
  565. that support "Round Robin" interrupt distribution.]
  566. There are three more important subdirectories in /proc: net, scsi, and sys.
  567. The general rule is that the contents, or even the existence of these
  568. directories, depend on your kernel configuration. If SCSI is not enabled, the
  569. directory scsi may not exist. The same is true with the net, which is there
  570. only when networking support is present in the running kernel.
  571. The slabinfo file gives information about memory usage at the slab level.
  572. Linux uses slab pools for memory management above page level in version 2.2.
  573. Commonly used objects have their own slab pool (such as network buffers,
  574. directory cache, and so on).
  575. ..............................................................................
  576. > cat /proc/buddyinfo
  577. Node 0, zone DMA 0 4 5 4 4 3 ...
  578. Node 0, zone Normal 1 0 0 1 101 8 ...
  579. Node 0, zone HighMem 2 0 0 1 1 0 ...
  580. External fragmentation is a problem under some workloads, and buddyinfo is a
  581. useful tool for helping diagnose these problems. Buddyinfo will give you a
  582. clue as to how big an area you can safely allocate, or why a previous
  583. allocation failed.
  584. Each column represents the number of pages of a certain order which are
  585. available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
  586. ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
  587. available in ZONE_NORMAL, etc...
  588. More information relevant to external fragmentation can be found in
  589. pagetypeinfo.
  590. > cat /proc/pagetypeinfo
  591. Page block order: 9
  592. Pages per block: 512
  593. Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
  594. Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
  595. Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
  596. Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
  597. Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
  598. Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
  599. Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
  600. Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
  601. Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
  602. Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
  603. Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
  604. Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
  605. Node 0, zone DMA 2 0 5 1 0
  606. Node 0, zone DMA32 41 6 967 2 0
  607. Fragmentation avoidance in the kernel works by grouping pages of different
  608. migrate types into the same contiguous regions of memory called page blocks.
  609. A page block is typically the size of the default hugepage size e.g. 2MB on
  610. X86-64. By keeping pages grouped based on their ability to move, the kernel
  611. can reclaim pages within a page block to satisfy a high-order allocation.
  612. The pagetypinfo begins with information on the size of a page block. It
  613. then gives the same type of information as buddyinfo except broken down
  614. by migrate-type and finishes with details on how many page blocks of each
  615. type exist.
  616. If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
  617. from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can
  618. make an estimate of the likely number of huge pages that can be allocated
  619. at a given point in time. All the "Movable" blocks should be allocatable
  620. unless memory has been mlock()'d. Some of the Reclaimable blocks should
  621. also be allocatable although a lot of filesystem metadata may have to be
  622. reclaimed to achieve this.
  623. ..............................................................................
  624. meminfo:
  625. Provides information about distribution and utilization of memory. This
  626. varies by architecture and compile options. The following is from a
  627. 16GB PIII, which has highmem enabled. You may not have all of these fields.
  628. > cat /proc/meminfo
  629. The "Locked" indicates whether the mapping is locked in memory or not.
  630. MemTotal: 16344972 kB
  631. MemFree: 13634064 kB
  632. Buffers: 3656 kB
  633. Cached: 1195708 kB
  634. SwapCached: 0 kB
  635. Active: 891636 kB
  636. Inactive: 1077224 kB
  637. HighTotal: 15597528 kB
  638. HighFree: 13629632 kB
  639. LowTotal: 747444 kB
  640. LowFree: 4432 kB
  641. SwapTotal: 0 kB
  642. SwapFree: 0 kB
  643. Dirty: 968 kB
  644. Writeback: 0 kB
  645. AnonPages: 861800 kB
  646. Mapped: 280372 kB
  647. Slab: 284364 kB
  648. SReclaimable: 159856 kB
  649. SUnreclaim: 124508 kB
  650. PageTables: 24448 kB
  651. NFS_Unstable: 0 kB
  652. Bounce: 0 kB
  653. WritebackTmp: 0 kB
  654. CommitLimit: 7669796 kB
  655. Committed_AS: 100056 kB
  656. VmallocTotal: 112216 kB
  657. VmallocUsed: 428 kB
  658. VmallocChunk: 111088 kB
  659. MemTotal: Total usable ram (i.e. physical ram minus a few reserved
  660. bits and the kernel binary code)
  661. MemFree: The sum of LowFree+HighFree
  662. Buffers: Relatively temporary storage for raw disk blocks
  663. shouldn't get tremendously large (20MB or so)
  664. Cached: in-memory cache for files read from the disk (the
  665. pagecache). Doesn't include SwapCached
  666. SwapCached: Memory that once was swapped out, is swapped back in but
  667. still also is in the swapfile (if memory is needed it
  668. doesn't need to be swapped out AGAIN because it is already
  669. in the swapfile. This saves I/O)
  670. Active: Memory that has been used more recently and usually not
  671. reclaimed unless absolutely necessary.
  672. Inactive: Memory which has been less recently used. It is more
  673. eligible to be reclaimed for other purposes
  674. HighTotal:
  675. HighFree: Highmem is all memory above ~860MB of physical memory
  676. Highmem areas are for use by userspace programs, or
  677. for the pagecache. The kernel must use tricks to access
  678. this memory, making it slower to access than lowmem.
  679. LowTotal:
  680. LowFree: Lowmem is memory which can be used for everything that
  681. highmem can be used for, but it is also available for the
  682. kernel's use for its own data structures. Among many
  683. other things, it is where everything from the Slab is
  684. allocated. Bad things happen when you're out of lowmem.
  685. SwapTotal: total amount of swap space available
  686. SwapFree: Memory which has been evicted from RAM, and is temporarily
  687. on the disk
  688. Dirty: Memory which is waiting to get written back to the disk
  689. Writeback: Memory which is actively being written back to the disk
  690. AnonPages: Non-file backed pages mapped into userspace page tables
  691. Mapped: files which have been mmaped, such as libraries
  692. Slab: in-kernel data structures cache
  693. SReclaimable: Part of Slab, that might be reclaimed, such as caches
  694. SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
  695. PageTables: amount of memory dedicated to the lowest level of page
  696. tables.
  697. NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
  698. storage
  699. Bounce: Memory used for block device "bounce buffers"
  700. WritebackTmp: Memory used by FUSE for temporary writeback buffers
  701. CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
  702. this is the total amount of memory currently available to
  703. be allocated on the system. This limit is only adhered to
  704. if strict overcommit accounting is enabled (mode 2 in
  705. 'vm.overcommit_memory').
  706. The CommitLimit is calculated with the following formula:
  707. CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
  708. For example, on a system with 1G of physical RAM and 7G
  709. of swap with a `vm.overcommit_ratio` of 30 it would
  710. yield a CommitLimit of 7.3G.
  711. For more details, see the memory overcommit documentation
  712. in vm/overcommit-accounting.
  713. Committed_AS: The amount of memory presently allocated on the system.
  714. The committed memory is a sum of all of the memory which
  715. has been allocated by processes, even if it has not been
  716. "used" by them as of yet. A process which malloc()'s 1G
  717. of memory, but only touches 300M of it will only show up
  718. as using 300M of memory even if it has the address space
  719. allocated for the entire 1G. This 1G is memory which has
  720. been "committed" to by the VM and can be used at any time
  721. by the allocating application. With strict overcommit
  722. enabled on the system (mode 2 in 'vm.overcommit_memory'),
  723. allocations which would exceed the CommitLimit (detailed
  724. above) will not be permitted. This is useful if one needs
  725. to guarantee that processes will not fail due to lack of
  726. memory once that memory has been successfully allocated.
  727. VmallocTotal: total size of vmalloc memory area
  728. VmallocUsed: amount of vmalloc area which is used
  729. VmallocChunk: largest contiguous block of vmalloc area which is free
  730. ..............................................................................
  731. vmallocinfo:
  732. Provides information about vmalloced/vmaped areas. One line per area,
  733. containing the virtual address range of the area, size in bytes,
  734. caller information of the creator, and optional information depending
  735. on the kind of area :
  736. pages=nr number of pages
  737. phys=addr if a physical address was specified
  738. ioremap I/O mapping (ioremap() and friends)
  739. vmalloc vmalloc() area
  740. vmap vmap()ed pages
  741. user VM_USERMAP area
  742. vpages buffer for pages pointers was vmalloced (huge area)
  743. N<node>=nr (Only on NUMA kernels)
  744. Number of pages allocated on memory node <node>
  745. > cat /proc/vmallocinfo
  746. 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
  747. /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
  748. 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
  749. /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
  750. 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
  751. phys=7fee8000 ioremap
  752. 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
  753. phys=7fee7000 ioremap
  754. 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
  755. 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
  756. /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
  757. 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
  758. pages=2 vmalloc N1=2
  759. 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
  760. /0x130 [x_tables] pages=4 vmalloc N0=4
  761. 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
  762. pages=14 vmalloc N2=14
  763. 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
  764. pages=4 vmalloc N1=4
  765. 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
  766. pages=2 vmalloc N1=2
  767. 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
  768. pages=10 vmalloc N0=10
  769. ..............................................................................
  770. softirqs:
  771. Provides counts of softirq handlers serviced since boot time, for each cpu.
  772. > cat /proc/softirqs
  773. CPU0 CPU1 CPU2 CPU3
  774. HI: 0 0 0 0
  775. TIMER: 27166 27120 27097 27034
  776. NET_TX: 0 0 0 17
  777. NET_RX: 42 0 0 39
  778. BLOCK: 0 0 107 1121
  779. TASKLET: 0 0 0 290
  780. SCHED: 27035 26983 26971 26746
  781. HRTIMER: 0 0 0 0
  782. RCU: 1678 1769 2178 2250
  783. 1.3 IDE devices in /proc/ide
  784. ----------------------------
  785. The subdirectory /proc/ide contains information about all IDE devices of which
  786. the kernel is aware. There is one subdirectory for each IDE controller, the
  787. file drivers and a link for each IDE device, pointing to the device directory
  788. in the controller specific subtree.
  789. The file drivers contains general information about the drivers used for the
  790. IDE devices:
  791. > cat /proc/ide/drivers
  792. ide-cdrom version 4.53
  793. 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.
  797. Table 1-6: IDE controller info in /proc/ide/ide?
  798. ..............................................................................
  799. File Content
  800. channel IDE channel (0 or 1)
  801. config Configuration (only for PCI/IDE bridge)
  802. mate Mate name
  803. model Type/Chipset of IDE controller
  804. ..............................................................................
  805. Each device connected to a controller has a separate subdirectory in the
  806. controllers directory. The files listed in table 1-7 are contained in these
  807. directories.
  808. Table 1-7: IDE device information
  809. ..............................................................................
  810. File Content
  811. cache The cache
  812. capacity Capacity of the medium (in 512Byte blocks)
  813. driver driver and version
  814. geometry physical and logical geometry
  815. identify device identify block
  816. media media type
  817. model device identifier
  818. settings device setup
  819. smart_thresholds IDE disk management thresholds
  820. smart_values IDE disk management values
  821. ..............................................................................
  822. The most interesting file is settings. This file contains a nice overview of
  823. the drive parameters:
  824. # cat /proc/ide/ide0/hda/settings
  825. name value min max mode
  826. ---- ----- --- --- ----
  827. bios_cyl 526 0 65535 rw
  828. bios_head 255 0 255 rw
  829. bios_sect 63 0 63 rw
  830. breada_readahead 4 0 127 rw
  831. bswap 0 0 1 r
  832. file_readahead 72 0 2097151 rw
  833. io_32bit 0 0 3 rw
  834. keepsettings 0 0 1 rw
  835. max_kb_per_request 122 1 127 rw
  836. multcount 0 0 8 rw
  837. nice1 1 0 1 rw
  838. nowerr 0 0 1 rw
  839. pio_mode write-only 0 255 w
  840. slow 0 0 1 rw
  841. unmaskirq 0 0 1 rw
  842. using_dma 0 0 1 rw
  843. 1.4 Networking info in /proc/net
  844. --------------------------------
  845. The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
  846. additional values you get for IP version 6 if you configure the kernel to
  847. support this. Table 1-9 lists the files and their meaning.
  848. Table 1-8: IPv6 info in /proc/net
  849. ..............................................................................
  850. File Content
  851. udp6 UDP sockets (IPv6)
  852. tcp6 TCP sockets (IPv6)
  853. raw6 Raw device statistics (IPv6)
  854. igmp6 IP multicast addresses, which this host joined (IPv6)
  855. if_inet6 List of IPv6 interface addresses
  856. ipv6_route Kernel routing table for IPv6
  857. rt6_stats Global IPv6 routing tables statistics
  858. sockstat6 Socket statistics (IPv6)
  859. snmp6 Snmp data (IPv6)
  860. ..............................................................................
  861. Table 1-9: Network info in /proc/net
  862. ..............................................................................
  863. File Content
  864. arp Kernel ARP table
  865. dev network devices with statistics
  866. dev_mcast the Layer2 multicast groups a device is listening too
  867. (interface index, label, number of references, number of bound
  868. addresses).
  869. dev_stat network device status
  870. ip_fwchains Firewall chain linkage
  871. ip_fwnames Firewall chain names
  872. ip_masq Directory containing the masquerading tables
  873. ip_masquerade Major masquerading table
  874. netstat Network statistics
  875. raw raw device statistics
  876. route Kernel routing table
  877. rpc Directory containing rpc info
  878. rt_cache Routing cache
  879. snmp SNMP data
  880. sockstat Socket statistics
  881. tcp TCP sockets
  882. tr_rif Token ring RIF routing table
  883. udp UDP sockets
  884. unix UNIX domain sockets
  885. wireless Wireless interface data (Wavelan etc)
  886. igmp IP multicast addresses, which this host joined
  887. psched Global packet scheduler parameters.
  888. netlink List of PF_NETLINK sockets
  889. ip_mr_vifs List of multicast virtual interfaces
  890. ip_mr_cache List of multicast routing cache
  891. ..............................................................................
  892. You can use this information to see which network devices are available in
  893. your system and how much traffic was routed over those devices:
  894. > cat /proc/net/dev
  895. Inter-|Receive |[...
  896. face |bytes packets errs drop fifo frame compressed multicast|[...
  897. lo: 908188 5596 0 0 0 0 0 0 [...
  898. ppp0:15475140 20721 410 0 0 410 0 0 [...
  899. eth0: 614530 7085 0 0 0 0 0 1 [...
  900. ...] Transmit
  901. ...] bytes packets errs drop fifo colls carrier compressed
  902. ...] 908188 5596 0 0 0 0 0 0
  903. ...] 1375103 17405 0 0 0 0 0 0
  904. ...] 1703981 5535 0 0 0 3 0 0
  905. In addition, each Channel Bond interface has its own directory. For
  906. example, the bond0 device will have a directory called /proc/net/bond0/.
  907. It will contain information that is specific to that bond, such as the
  908. current slaves of the bond, the link status of the slaves, and how
  909. many times the slaves link has failed.
  910. 1.5 SCSI info
  911. -------------
  912. If you have a SCSI host adapter in your system, you'll find a subdirectory
  913. named after the driver for this adapter in /proc/scsi. You'll also see a list
  914. of all recognized SCSI devices in /proc/scsi:
  915. >cat /proc/scsi/scsi
  916. Attached devices:
  917. Host: scsi0 Channel: 00 Id: 00 Lun: 00
  918. Vendor: IBM Model: DGHS09U Rev: 03E0
  919. Type: Direct-Access ANSI SCSI revision: 03
  920. Host: scsi0 Channel: 00 Id: 06 Lun: 00
  921. Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
  922. Type: CD-ROM ANSI SCSI revision: 02
  923. The directory named after the driver has one file for each adapter found in
  924. the system. These files contain information about the controller, including
  925. the used IRQ and the IO address range. The amount of information shown is
  926. dependent on the adapter you use. The example shows the output for an Adaptec
  927. AHA-2940 SCSI adapter:
  928. > cat /proc/scsi/aic7xxx/0
  929. Adaptec AIC7xxx driver version: 5.1.19/3.2.4
  930. Compile Options:
  931. TCQ Enabled By Default : Disabled
  932. AIC7XXX_PROC_STATS : Disabled
  933. AIC7XXX_RESET_DELAY : 5
  934. Adapter Configuration:
  935. SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
  936. Ultra Wide Controller
  937. PCI MMAPed I/O Base: 0xeb001000
  938. Adapter SEEPROM Config: SEEPROM found and used.
  939. Adaptec SCSI BIOS: Enabled
  940. IRQ: 10
  941. SCBs: Active 0, Max Active 2,
  942. Allocated 15, HW 16, Page 255
  943. Interrupts: 160328
  944. BIOS Control Word: 0x18b6
  945. Adapter Control Word: 0x005b
  946. Extended Translation: Enabled
  947. Disconnect Enable Flags: 0xffff
  948. Ultra Enable Flags: 0x0001
  949. Tag Queue Enable Flags: 0x0000
  950. Ordered Queue Tag Flags: 0x0000
  951. Default Tag Queue Depth: 8
  952. Tagged Queue By Device array for aic7xxx host instance 0:
  953. {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
  954. Actual queue depth per device for aic7xxx host instance 0:
  955. {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
  956. Statistics:
  957. (scsi0:0:0:0)
  958. Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
  959. Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
  960. Total transfers 160151 (74577 reads and 85574 writes)
  961. (scsi0:0:6:0)
  962. Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
  963. Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
  964. Total transfers 0 (0 reads and 0 writes)
  965. 1.6 Parallel port info in /proc/parport
  966. ---------------------------------------
  967. The directory /proc/parport contains information about the parallel ports of
  968. your system. It has one subdirectory for each port, named after the port
  969. number (0,1,2,...).
  970. These directories contain the four files shown in Table 1-10.
  971. Table 1-10: Files in /proc/parport
  972. ..............................................................................
  973. File Content
  974. autoprobe Any IEEE-1284 device ID information that has been acquired.
  975. devices list of the device drivers using that port. A + will appear by the
  976. name of the device currently using the port (it might not appear
  977. against any).
  978. hardware Parallel port's base address, IRQ line and DMA channel.
  979. irq IRQ that parport is using for that port. This is in a separate
  980. file to allow you to alter it by writing a new value in (IRQ
  981. number or none).
  982. ..............................................................................
  983. 1.7 TTY info in /proc/tty
  984. -------------------------
  985. Information about the available and actually used tty's can be found in the
  986. directory /proc/tty.You'll find entries for drivers and line disciplines in
  987. this directory, as shown in Table 1-11.
  988. Table 1-11: Files in /proc/tty
  989. ..............................................................................
  990. File Content
  991. drivers list of drivers and their usage
  992. ldiscs registered line disciplines
  993. driver/serial usage statistic and status of single tty lines
  994. ..............................................................................
  995. To see which tty's are currently in use, you can simply look into the file
  996. /proc/tty/drivers:
  997. > cat /proc/tty/drivers
  998. pty_slave /dev/pts 136 0-255 pty:slave
  999. pty_master /dev/ptm 128 0-255 pty:master
  1000. pty_slave /dev/ttyp 3 0-255 pty:slave
  1001. pty_master /dev/pty 2 0-255 pty:master
  1002. serial /dev/cua 5 64-67 serial:callout
  1003. serial /dev/ttyS 4 64-67 serial
  1004. /dev/tty0 /dev/tty0 4 0 system:vtmaster
  1005. /dev/ptmx /dev/ptmx 5 2 system
  1006. /dev/console /dev/console 5 1 system:console
  1007. /dev/tty /dev/tty 5 0 system:/dev/tty
  1008. unknown /dev/tty 4 1-63 console
  1009. 1.8 Miscellaneous kernel statistics in /proc/stat
  1010. -------------------------------------------------
  1011. Various pieces of information about kernel activity are available in the
  1012. /proc/stat file. All of the numbers reported in this file are aggregates
  1013. since the system first booted. For a quick look, simply cat the file:
  1014. > cat /proc/stat
  1015. cpu 2255 34 2290 22625563 6290 127 456 0 0
  1016. cpu0 1132 34 1441 11311718 3675 127 438 0 0
  1017. cpu1 1123 0 849 11313845 2614 0 18 0 0
  1018. intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
  1019. ctxt 1990473
  1020. btime 1062191376
  1021. processes 2915
  1022. procs_running 1
  1023. procs_blocked 0
  1024. softirq 183433 0 21755 12 39 1137 231 21459 2263
  1025. The very first "cpu" line aggregates the numbers in all of the other "cpuN"
  1026. lines. These numbers identify the amount of time the CPU has spent performing
  1027. different kinds of work. Time units are in USER_HZ (typically hundredths of a
  1028. second). The meanings of the columns are as follows, from left to right:
  1029. - user: normal processes executing in user mode
  1030. - nice: niced processes executing in user mode
  1031. - system: processes executing in kernel mode
  1032. - idle: twiddling thumbs
  1033. - iowait: waiting for I/O to complete
  1034. - irq: servicing interrupts
  1035. - softirq: servicing softirqs
  1036. - steal: involuntary wait
  1037. - guest: running a normal guest
  1038. - guest_nice: running a niced guest
  1039. The "intr" line gives counts of interrupts serviced since boot time, for each
  1040. of the possible system interrupts. The first column is the total of all
  1041. interrupts serviced; each subsequent column is the total for that particular
  1042. interrupt.
  1043. The "ctxt" line gives the total number of context switches across all CPUs.
  1044. The "btime" line gives the time at which the system booted, in seconds since
  1045. the Unix epoch.
  1046. The "processes" line gives the number of processes and threads created, which
  1047. includes (but is not limited to) those created by calls to the fork() and
  1048. clone() system calls.
  1049. The "procs_running" line gives the total number of threads that are
  1050. running or ready to run (i.e., the total number of runnable threads).
  1051. The "procs_blocked" line gives the number of processes currently blocked,
  1052. waiting for I/O to complete.
  1053. The "softirq" line gives counts of softirqs serviced since boot time, for each
  1054. of the possible system softirqs. The first column is the total of all
  1055. softirqs serviced; each subsequent column is the total for that particular
  1056. softirq.
  1057. 1.9 Ext4 file system parameters
  1058. ------------------------------
  1059. Information about mounted ext4 file systems can be found in
  1060. /proc/fs/ext4. Each mounted filesystem will have a directory in
  1061. /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
  1062. /proc/fs/ext4/dm-0). The files in each per-device directory are shown
  1063. in Table 1-12, below.
  1064. Table 1-12: Files in /proc/fs/ext4/<devname>
  1065. ..............................................................................
  1066. File Content
  1067. mb_groups details of multiblock allocator buddy cache of free blocks
  1068. ..............................................................................
  1069. 2.0 /proc/consoles
  1070. ------------------
  1071. Shows registered system console lines.
  1072. To see which character device lines are currently used for the system console
  1073. /dev/console, you may simply look into the file /proc/consoles:
  1074. > cat /proc/consoles
  1075. tty0 -WU (ECp) 4:7
  1076. ttyS0 -W- (Ep) 4:64
  1077. The columns are:
  1078. device name of the device
  1079. operations R = can do read operations
  1080. W = can do write operations
  1081. U = can do unblank
  1082. flags E = it is enabled
  1083. C = it is preferred console
  1084. B = it is primary boot console
  1085. p = it is used for printk buffer
  1086. b = it is not a TTY but a Braille device
  1087. a = it is safe to use when cpu is offline
  1088. major:minor major and minor number of the device separated by a colon
  1089. ------------------------------------------------------------------------------
  1090. Summary
  1091. ------------------------------------------------------------------------------
  1092. The /proc file system serves information about the running system. It not only
  1093. allows access to process data but also allows you to request the kernel status
  1094. by reading files in the hierarchy.
  1095. The directory structure of /proc reflects the types of information and makes
  1096. it easy, if not obvious, where to look for specific data.
  1097. ------------------------------------------------------------------------------
  1098. ------------------------------------------------------------------------------
  1099. CHAPTER 2: MODIFYING SYSTEM PARAMETERS
  1100. ------------------------------------------------------------------------------
  1101. ------------------------------------------------------------------------------
  1102. In This Chapter
  1103. ------------------------------------------------------------------------------
  1104. * Modifying kernel parameters by writing into files found in /proc/sys
  1105. * Exploring the files which modify certain parameters
  1106. * Review of the /proc/sys file tree
  1107. ------------------------------------------------------------------------------
  1108. A very interesting part of /proc is the directory /proc/sys. This is not only
  1109. a source of information, it also allows you to change parameters within the
  1110. kernel. Be very careful when attempting this. You can optimize your system,
  1111. but you can also cause it to crash. Never alter kernel parameters on a
  1112. production system. Set up a development machine and test to make sure that
  1113. everything works the way you want it to. You may have no alternative but to
  1114. reboot the machine once an error has been made.
  1115. To change a value, simply echo the new value into the file. An example is
  1116. given below in the section on the file system data. You need to be root to do
  1117. this. You can create your own boot script to perform this every time your
  1118. system boots.
  1119. The files in /proc/sys can be used to fine tune and monitor miscellaneous and
  1120. general things in the operation of the Linux kernel. Since some of the files
  1121. can inadvertently disrupt your system, it is advisable to read both
  1122. documentation and source before actually making adjustments. In any case, be
  1123. very careful when writing to any of these files. The entries in /proc may
  1124. change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
  1125. review the kernel documentation in the directory /usr/src/linux/Documentation.
  1126. This chapter is heavily based on the documentation included in the pre 2.2
  1127. kernels, and became part of it in version 2.2.1 of the Linux kernel.
  1128. Please see: Documentation/sysctl/ directory for descriptions of these
  1129. entries.
  1130. ------------------------------------------------------------------------------
  1131. Summary
  1132. ------------------------------------------------------------------------------
  1133. Certain aspects of kernel behavior can be modified at runtime, without the
  1134. need to recompile the kernel, or even to reboot the system. The files in the
  1135. /proc/sys tree can not only be read, but also modified. You can use the echo
  1136. command to write value into these files, thereby changing the default settings
  1137. of the kernel.
  1138. ------------------------------------------------------------------------------
  1139. ------------------------------------------------------------------------------
  1140. CHAPTER 3: PER-PROCESS PARAMETERS
  1141. ------------------------------------------------------------------------------
  1142. 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
  1143. --------------------------------------------------------------------------------
  1144. These file can be used to adjust the badness heuristic used to select which
  1145. process gets killed in out of memory conditions.
  1146. The badness heuristic assigns a value to each candidate task ranging from 0
  1147. (never kill) to 1000 (always kill) to determine which process is targeted. The
  1148. units are roughly a proportion along that range of allowed memory the process
  1149. may allocate from based on an estimation of its current memory and swap use.
  1150. For example, if a task is using all allowed memory, its badness score will be
  1151. 1000. If it is using half of its allowed memory, its score will be 500.
  1152. There is an additional factor included in the badness score: root
  1153. processes are given 3% extra memory over other tasks.
  1154. The amount of "allowed" memory depends on the context in which the oom killer
  1155. was called. If it is due to the memory assigned to the allocating task's cpuset
  1156. being exhausted, the allowed memory represents the set of mems assigned to that
  1157. cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
  1158. memory represents the set of mempolicy nodes. If it is due to a memory
  1159. limit (or swap limit) being reached, the allowed memory is that configured
  1160. limit. Finally, if it is due to the entire system being out of memory, the
  1161. allowed memory represents all allocatable resources.
  1162. The value of /proc/<pid>/oom_score_adj is added to the badness score before it
  1163. is used to determine which task to kill. Acceptable values range from -1000
  1164. (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
  1165. polarize the preference for oom killing either by always preferring a certain
  1166. task or completely disabling it. The lowest possible value, -1000, is
  1167. equivalent to disabling oom killing entirely for that task since it will always
  1168. report a badness score of 0.
  1169. Consequently, it is very simple for userspace to define the amount of memory to
  1170. consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
  1171. example, is roughly equivalent to allowing the remainder of tasks sharing the
  1172. same system, cpuset, mempolicy, or memory controller resources to use at least
  1173. 50% more memory. A value of -500, on the other hand, would be roughly
  1174. equivalent to discounting 50% of the task's allowed memory from being considered
  1175. as scoring against the task.
  1176. For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
  1177. be used to tune the badness score. Its acceptable values range from -16
  1178. (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
  1179. (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
  1180. scaled linearly with /proc/<pid>/oom_score_adj.
  1181. Writing to /proc/<pid>/oom_score_adj or /proc/<pid>/oom_adj will change the
  1182. other with its scaled value.
  1183. The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
  1184. value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
  1185. requires CAP_SYS_RESOURCE.
  1186. NOTICE: /proc/<pid>/oom_adj is deprecated and will be removed, please see
  1187. Documentation/feature-removal-schedule.txt.
  1188. Caveat: when a parent task is selected, the oom killer will sacrifice any first
  1189. generation children with separate address spaces instead, if possible. This
  1190. avoids servers and important system daemons from being killed and loses the
  1191. minimal amount of work.
  1192. 3.2 /proc/<pid>/oom_score - Display current oom-killer score
  1193. -------------------------------------------------------------
  1194. This file can be used to check the current score used by the oom-killer is for
  1195. any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
  1196. process should be killed in an out-of-memory situation.
  1197. 3.3 /proc/<pid>/io - Display the IO accounting fields
  1198. -------------------------------------------------------
  1199. This file contains IO statistics for each running process
  1200. Example
  1201. -------
  1202. test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
  1203. [1] 3828
  1204. test:/tmp # cat /proc/3828/io
  1205. rchar: 323934931
  1206. wchar: 323929600
  1207. syscr: 632687
  1208. syscw: 632675
  1209. read_bytes: 0
  1210. write_bytes: 323932160
  1211. cancelled_write_bytes: 0
  1212. Description
  1213. -----------
  1214. rchar
  1215. -----
  1216. I/O counter: chars read
  1217. The number of bytes which this task has caused to be read from storage. This
  1218. is simply the sum of bytes which this process passed to read() and pread().
  1219. It includes things like tty IO and it is unaffected by whether or not actual
  1220. physical disk IO was required (the read might have been satisfied from
  1221. pagecache)
  1222. wchar
  1223. -----
  1224. I/O counter: chars written
  1225. The number of bytes which this task has caused, or shall cause to be written
  1226. to disk. Similar caveats apply here as with rchar.
  1227. syscr
  1228. -----
  1229. I/O counter: read syscalls
  1230. Attempt to count the number of read I/O operations, i.e. syscalls like read()
  1231. and pread().
  1232. syscw
  1233. -----
  1234. I/O counter: write syscalls
  1235. Attempt to count the number of write I/O operations, i.e. syscalls like
  1236. write() and pwrite().
  1237. read_bytes
  1238. ----------
  1239. I/O counter: bytes read
  1240. Attempt to count the number of bytes which this process really did cause to
  1241. be fetched from the storage layer. Done at the submit_bio() level, so it is
  1242. accurate for block-backed filesystems. <please add status regarding NFS and
  1243. CIFS at a later time>
  1244. write_bytes
  1245. -----------
  1246. I/O counter: bytes written
  1247. Attempt to count the number of bytes which this process caused to be sent to
  1248. the storage layer. This is done at page-dirtying time.
  1249. cancelled_write_bytes
  1250. ---------------------
  1251. The big inaccuracy here is truncate. If a process writes 1MB to a file and
  1252. then deletes the file, it will in fact perform no writeout. But it will have
  1253. been accounted as having caused 1MB of write.
  1254. In other words: The number of bytes which this process caused to not happen,
  1255. by truncating pagecache. A task can cause "negative" IO too. If this task
  1256. truncates some dirty pagecache, some IO which another task has been accounted
  1257. for (in its write_bytes) will not be happening. We _could_ just subtract that
  1258. from the truncating task's write_bytes, but there is information loss in doing
  1259. that.
  1260. Note
  1261. ----
  1262. At its current implementation state, this is a bit racy on 32-bit machines: if
  1263. process A reads process B's /proc/pid/io while process B is updating one of
  1264. those 64-bit counters, process A could see an intermediate result.
  1265. More information about this can be found within the taskstats documentation in
  1266. Documentation/accounting.
  1267. 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
  1268. ---------------------------------------------------------------
  1269. When a process is dumped, all anonymous memory is written to a core file as
  1270. long as the size of the core file isn't limited. But sometimes we don't want
  1271. to dump some memory segments, for example, huge shared memory. Conversely,
  1272. sometimes we want to save file-backed memory segments into a core file, not
  1273. only the individual files.
  1274. /proc/<pid>/coredump_filter allows you to customize which memory segments
  1275. will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
  1276. of memory types. If a bit of the bitmask is set, memory segments of the
  1277. corresponding memory type are dumped, otherwise they are not dumped.
  1278. The following 7 memory types are supported:
  1279. - (bit 0) anonymous private memory
  1280. - (bit 1) anonymous shared memory
  1281. - (bit 2) file-backed private memory
  1282. - (bit 3) file-backed shared memory
  1283. - (bit 4) ELF header pages in file-backed private memory areas (it is
  1284. effective only if the bit 2 is cleared)
  1285. - (bit 5) hugetlb private memory
  1286. - (bit 6) hugetlb shared memory
  1287. Note that MMIO pages such as frame buffer are never dumped and vDSO pages
  1288. are always dumped regardless of the bitmask status.
  1289. Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
  1290. effected by bit 5-6.
  1291. Default value of coredump_filter is 0x23; this means all anonymous memory
  1292. segments and hugetlb private memory are dumped.
  1293. If you don't want to dump all shared memory segments attached to pid 1234,
  1294. write 0x21 to the process's proc file.
  1295. $ echo 0x21 > /proc/1234/coredump_filter
  1296. When a new process is created, the process inherits the bitmask status from its
  1297. parent. It is useful to set up coredump_filter before the program runs.
  1298. For example:
  1299. $ echo 0x7 > /proc/self/coredump_filter
  1300. $ ./some_program
  1301. 3.5 /proc/<pid>/mountinfo - Information about mounts
  1302. --------------------------------------------------------
  1303. This file contains lines of the form:
  1304. 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
  1305. (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
  1306. (1) mount ID: unique identifier of the mount (may be reused after umount)
  1307. (2) parent ID: ID of parent (or of self for the top of the mount tree)
  1308. (3) major:minor: value of st_dev for files on filesystem
  1309. (4) root: root of the mount within the filesystem
  1310. (5) mount point: mount point relative to the process's root
  1311. (6) mount options: per mount options
  1312. (7) optional fields: zero or more fields of the form "tag[:value]"
  1313. (8) separator: marks the end of the optional fields
  1314. (9) filesystem type: name of filesystem of the form "type[.subtype]"
  1315. (10) mount source: filesystem specific information or "none"
  1316. (11) super options: per super block options
  1317. Parsers should ignore all unrecognised optional fields. Currently the
  1318. possible optional fields are:
  1319. shared:X mount is shared in peer group X
  1320. master:X mount is slave to peer group X
  1321. propagate_from:X mount is slave and receives propagation from peer group X (*)
  1322. unbindable mount is unbindable
  1323. (*) X is the closest dominant peer group under the process's root. If
  1324. X is the immediate master of the mount, or if there's no dominant peer
  1325. group under the same root, then only the "master:X" field is present
  1326. and not the "propagate_from:X" field.
  1327. For more information on mount propagation see:
  1328. Documentation/filesystems/sharedsubtree.txt
  1329. 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
  1330. --------------------------------------------------------
  1331. These files provide a method to access a tasks comm value. It also allows for
  1332. a task to set its own or one of its thread siblings comm value. The comm value
  1333. is limited in size compared to the cmdline value, so writing anything longer
  1334. then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
  1335. comm value.
  1336. ------------------------------------------------------------------------------
  1337. Configuring procfs
  1338. ------------------------------------------------------------------------------
  1339. 4.1 Mount options
  1340. ---------------------
  1341. The following mount options are supported:
  1342. hidepid= Set /proc/<pid>/ access mode.
  1343. gid= Set the group authorized to learn processes information.
  1344. hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
  1345. (default).
  1346. hidepid=1 means users may not access any /proc/<pid>/ directories but their
  1347. own. Sensitive files like cmdline, sched*, status are now protected against
  1348. other users. This makes it impossible to learn whether any user runs
  1349. specific program (given the program doesn't reveal itself by its behaviour).
  1350. As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
  1351. poorly written programs passing sensitive information via program arguments are
  1352. now protected against local eavesdroppers.
  1353. hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
  1354. users. It doesn't mean that it hides a fact whether a process with a specific
  1355. pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
  1356. but it hides process' uid and gid, which may be learned by stat()'ing
  1357. /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
  1358. information about running processes, whether some daemon runs with elevated
  1359. privileges, whether other user runs some sensitive program, whether other users
  1360. run any program at all, etc.
  1361. gid= defines a group authorized to learn processes information otherwise
  1362. prohibited by hidepid=. If you use some daemon like identd which needs to learn
  1363. information about processes information, just add identd to this group.