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 72.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 8move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 9------------------------------------------------------------------------------ 10Version 1.3 Kernel version 2.2.12 11 Kernel version 2.4.0-test11-pre4 12------------------------------------------------------------------------------ 13fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 14 15Table of Contents 16----------------- 17 18 0 Preface 19 0.1 Introduction/Credits 20 0.2 Legal Stuff 21 22 1 Collecting System Information 23 1.1 Process-Specific Subdirectories 24 1.2 Kernel data 25 1.3 IDE devices in /proc/ide 26 1.4 Networking info in /proc/net 27 1.5 SCSI info 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 32 33 2 Modifying System Parameters 34 35 3 Per-Process Parameters 36 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 37 score 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 47 4 Configuring procfs 48 4.1 Mount options 49 50------------------------------------------------------------------------------ 51Preface 52------------------------------------------------------------------------------ 53 540.1 Introduction/Credits 55------------------------ 56 57This documentation is part of a soon (or so we hope) to be released book on 58the SuSE Linux distribution. As there is no complete documentation for the 59/proc file system and we've used many freely available sources to write these 60chapters, it seems only fair to give the work back to the Linux community. 61This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 62afraid it's still far from complete, but we hope it will be useful. As far as 63we know, it is the first 'all-in-one' document about the /proc file system. It 64is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 65SPARC, AXP, etc., features, you probably won't find what you are looking for. 66It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 67additions and patches are welcome and will be added to this document if you 68mail them to Bodo. 69 70We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 71other people for help compiling this documentation. We'd also like to extend a 72special thank you to Andi Kleen for documentation, which we relied on heavily 73to create this document, as well as the additional information he provided. 74Thanks to everybody else who contributed source or docs to the Linux kernel 75and helped create a great piece of software... :) 76 77If you have any comments, corrections or additions, please don't hesitate to 78contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 79document. 80 81The latest version of this document is available online at 82http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html 83 84If the above direction does not works for you, you could try the kernel 85mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 86comandante@zaralinux.com. 87 880.2 Legal Stuff 89--------------- 90 91We don't guarantee the correctness of this document, and if you come to us 92complaining about how you screwed up your system because of incorrect 93documentation, we won't feel responsible... 94 95------------------------------------------------------------------------------ 96CHAPTER 1: COLLECTING SYSTEM INFORMATION 97------------------------------------------------------------------------------ 98 99------------------------------------------------------------------------------ 100In This Chapter 101------------------------------------------------------------------------------ 102* Investigating the properties of the pseudo file system /proc and its 103 ability to provide information on the running Linux system 104* Examining /proc's structure 105* Uncovering various information about the kernel and the processes running 106 on the system 107------------------------------------------------------------------------------ 108 109 110The proc file system acts as an interface to internal data structures in the 111kernel. It can be used to obtain information about the system and to change 112certain kernel parameters at runtime (sysctl). 113 114First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 115show you how you can use /proc/sys to change settings. 116 1171.1 Process-Specific Subdirectories 118----------------------------------- 119 120The directory /proc contains (among other things) one subdirectory for each 121process running on the system, which is named after the process ID (PID). 122 123The link self points to the process reading the file system. Each process 124subdirectory has the entries listed in Table 1-1. 125 126 127Table 1-1: Process specific entries in /proc 128.............................................................................. 129 File Content 130 clear_refs Clears page referenced bits shown in smaps output 131 cmdline Command line arguments 132 cpu Current and last cpu in which it was executed (2.4)(smp) 133 cwd Link to the current working directory 134 environ Values of environment variables 135 exe Link to the executable of this process 136 fd Directory, which contains all file descriptors 137 maps Memory maps to executables and library files (2.4) 138 mem Memory held by this process 139 root Link to the root directory of this process 140 stat Process status 141 statm Process memory status information 142 status Process status in human readable form 143 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 144 symbol the task is blocked in - or "0" if not blocked. 145 pagemap Page table 146 stack Report full stack trace, enable via CONFIG_STACKTRACE 147 smaps a extension based on maps, showing the memory consumption of 148 each mapping and flags associated with it 149 numa_maps an extension based on maps, showing the memory locality and 150 binding policy as well as mem usage (in pages) of each mapping. 151.............................................................................. 152 153For example, to get the status information of a process, all you have to do is 154read the file /proc/PID/status: 155 156 >cat /proc/self/status 157 Name: cat 158 State: R (running) 159 Tgid: 5452 160 Pid: 5452 161 PPid: 743 162 TracerPid: 0 (2.4) 163 Uid: 501 501 501 501 164 Gid: 100 100 100 100 165 FDSize: 256 166 Groups: 100 14 16 167 VmPeak: 5004 kB 168 VmSize: 5004 kB 169 VmLck: 0 kB 170 VmHWM: 476 kB 171 VmRSS: 476 kB 172 VmData: 156 kB 173 VmStk: 88 kB 174 VmExe: 68 kB 175 VmLib: 1412 kB 176 VmPTE: 20 kb 177 VmSwap: 0 kB 178 HugetlbPages: 0 kB 179 Threads: 1 180 SigQ: 0/28578 181 SigPnd: 0000000000000000 182 ShdPnd: 0000000000000000 183 SigBlk: 0000000000000000 184 SigIgn: 0000000000000000 185 SigCgt: 0000000000000000 186 CapInh: 00000000fffffeff 187 CapPrm: 0000000000000000 188 CapEff: 0000000000000000 189 CapBnd: ffffffffffffffff 190 Seccomp: 0 191 voluntary_ctxt_switches: 0 192 nonvoluntary_ctxt_switches: 1 193 194This shows you nearly the same information you would get if you viewed it with 195the ps command. In fact, ps uses the proc file system to obtain its 196information. But you get a more detailed view of the process by reading the 197file /proc/PID/status. It fields are described in table 1-2. 198 199The statm file contains more detailed information about the process 200memory usage. Its seven fields are explained in Table 1-3. The stat file 201contains details information about the process itself. Its fields are 202explained in Table 1-4. 203 204(for SMP CONFIG users) 205For making accounting scalable, RSS related information are handled in an 206asynchronous manner and the value may not be very precise. To see a precise 207snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 208It's slow but very precise. 209 210Table 1-2: Contents of the status files (as of 4.1) 211.............................................................................. 212 Field Content 213 Name filename of the executable 214 State state (R is running, S is sleeping, D is sleeping 215 in an uninterruptible wait, Z is zombie, 216 T is traced or stopped) 217 Tgid thread group ID 218 Ngid NUMA group ID (0 if none) 219 Pid process id 220 PPid process id of the parent process 221 TracerPid PID of process tracing this process (0 if not) 222 Uid Real, effective, saved set, and file system UIDs 223 Gid Real, effective, saved set, and file system GIDs 224 FDSize number of file descriptor slots currently allocated 225 Groups supplementary group list 226 NStgid descendant namespace thread group ID hierarchy 227 NSpid descendant namespace process ID hierarchy 228 NSpgid descendant namespace process group ID hierarchy 229 NSsid descendant namespace session ID hierarchy 230 VmPeak peak virtual memory size 231 VmSize total program size 232 VmLck locked memory size 233 VmHWM peak resident set size ("high water mark") 234 VmRSS size of memory portions 235 VmData size of data, stack, and text segments 236 VmStk size of data, stack, and text segments 237 VmExe size of text segment 238 VmLib size of shared library code 239 VmPTE size of page table entries 240 VmPMD size of second level page tables 241 VmSwap size of swap usage (the number of referred swapents) 242 HugetlbPages size of hugetlb memory portions 243 Threads number of threads 244 SigQ number of signals queued/max. number for queue 245 SigPnd bitmap of pending signals for the thread 246 ShdPnd bitmap of shared pending signals for the process 247 SigBlk bitmap of blocked signals 248 SigIgn bitmap of ignored signals 249 SigCgt bitmap of caught signals 250 CapInh bitmap of inheritable capabilities 251 CapPrm bitmap of permitted capabilities 252 CapEff bitmap of effective capabilities 253 CapBnd bitmap of capabilities bounding set 254 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 255 Cpus_allowed mask of CPUs on which this process may run 256 Cpus_allowed_list Same as previous, but in "list format" 257 Mems_allowed mask of memory nodes allowed to this process 258 Mems_allowed_list Same as previous, but in "list format" 259 voluntary_ctxt_switches number of voluntary context switches 260 nonvoluntary_ctxt_switches number of non voluntary context switches 261.............................................................................. 262 263Table 1-3: Contents of the statm files (as of 2.6.8-rc3) 264.............................................................................. 265 Field Content 266 size total program size (pages) (same as VmSize in status) 267 resident size of memory portions (pages) (same as VmRSS in status) 268 shared number of pages that are shared (i.e. backed by a file) 269 trs number of pages that are 'code' (not including libs; broken, 270 includes data segment) 271 lrs number of pages of library (always 0 on 2.6) 272 drs number of pages of data/stack (including libs; broken, 273 includes library text) 274 dt number of dirty pages (always 0 on 2.6) 275.............................................................................. 276 277 278Table 1-4: Contents of the stat files (as of 2.6.30-rc7) 279.............................................................................. 280 Field Content 281 pid process id 282 tcomm filename of the executable 283 state state (R is running, S is sleeping, D is sleeping in an 284 uninterruptible wait, Z is zombie, T is traced or stopped) 285 ppid process id of the parent process 286 pgrp pgrp of the process 287 sid session id 288 tty_nr tty the process uses 289 tty_pgrp pgrp of the tty 290 flags task flags 291 min_flt number of minor faults 292 cmin_flt number of minor faults with child's 293 maj_flt number of major faults 294 cmaj_flt number of major faults with child's 295 utime user mode jiffies 296 stime kernel mode jiffies 297 cutime user mode jiffies with child's 298 cstime kernel mode jiffies with child's 299 priority priority level 300 nice nice level 301 num_threads number of threads 302 it_real_value (obsolete, always 0) 303 start_time time the process started after system boot 304 vsize virtual memory size 305 rss resident set memory size 306 rsslim current limit in bytes on the rss 307 start_code address above which program text can run 308 end_code address below which program text can run 309 start_stack address of the start of the main process stack 310 esp current value of ESP 311 eip current value of EIP 312 pending bitmap of pending signals 313 blocked bitmap of blocked signals 314 sigign bitmap of ignored signals 315 sigcatch bitmap of caught signals 316 0 (place holder, used to be the wchan address, use /proc/PID/wchan instead) 317 0 (place holder) 318 0 (place holder) 319 exit_signal signal to send to parent thread on exit 320 task_cpu which CPU the task is scheduled on 321 rt_priority realtime priority 322 policy scheduling policy (man sched_setscheduler) 323 blkio_ticks time spent waiting for block IO 324 gtime guest time of the task in jiffies 325 cgtime guest time of the task children in jiffies 326 start_data address above which program data+bss is placed 327 end_data address below which program data+bss is placed 328 start_brk address above which program heap can be expanded with brk() 329 arg_start address above which program command line is placed 330 arg_end address below which program command line is placed 331 env_start address above which program environment is placed 332 env_end address below which program environment is placed 333 exit_code the thread's exit_code in the form reported by the waitpid system call 334.............................................................................. 335 336The /proc/PID/maps file containing the currently mapped memory regions and 337their access permissions. 338 339The format is: 340 341address perms offset dev inode pathname 342 34308048000-08049000 r-xp 00000000 03:00 8312 /opt/test 34408049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 3450804a000-0806b000 rw-p 00000000 00:00 0 [heap] 346a7cb1000-a7cb2000 ---p 00000000 00:00 0 347a7cb2000-a7eb2000 rw-p 00000000 00:00 0 348a7eb2000-a7eb3000 ---p 00000000 00:00 0 349a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack:1001] 350a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 351a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 352a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 353a800b000-a800e000 rw-p 00000000 00:00 0 354a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 355a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 356a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 357a8024000-a8027000 rw-p 00000000 00:00 0 358a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 359a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 360a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 361aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 362ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 363 364where "address" is the address space in the process that it occupies, "perms" 365is a set of permissions: 366 367 r = read 368 w = write 369 x = execute 370 s = shared 371 p = private (copy on write) 372 373"offset" is the offset into the mapping, "dev" is the device (major:minor), and 374"inode" is the inode on that device. 0 indicates that no inode is associated 375with the memory region, as the case would be with BSS (uninitialized data). 376The "pathname" shows the name associated file for this mapping. If the mapping 377is not associated with a file: 378 379 [heap] = the heap of the program 380 [stack] = the stack of the main process 381 [stack:1001] = the stack of the thread with tid 1001 382 [vdso] = the "virtual dynamic shared object", 383 the kernel system call handler 384 385 or if empty, the mapping is anonymous. 386 387The /proc/PID/task/TID/maps is a view of the virtual memory from the viewpoint 388of the individual tasks of a process. In this file you will see a mapping marked 389as [stack] if that task sees it as a stack. This is a key difference from the 390content of /proc/PID/maps, where you will see all mappings that are being used 391as stack by all of those tasks. Hence, for the example above, the task-level 392map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this: 393 39408048000-08049000 r-xp 00000000 03:00 8312 /opt/test 39508049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 3960804a000-0806b000 rw-p 00000000 00:00 0 [heap] 397a7cb1000-a7cb2000 ---p 00000000 00:00 0 398a7cb2000-a7eb2000 rw-p 00000000 00:00 0 399a7eb2000-a7eb3000 ---p 00000000 00:00 0 400a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack] 401a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 402a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 403a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 404a800b000-a800e000 rw-p 00000000 00:00 0 405a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 406a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 407a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 408a8024000-a8027000 rw-p 00000000 00:00 0 409a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 410a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 411a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 412aff35000-aff4a000 rw-p 00000000 00:00 0 413ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 414 415The /proc/PID/smaps is an extension based on maps, showing the memory 416consumption for each of the process's mappings. For each of mappings there 417is a series of lines such as the following: 418 41908048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 420Size: 1084 kB 421Rss: 892 kB 422Pss: 374 kB 423Shared_Clean: 892 kB 424Shared_Dirty: 0 kB 425Private_Clean: 0 kB 426Private_Dirty: 0 kB 427Referenced: 892 kB 428Anonymous: 0 kB 429AnonHugePages: 0 kB 430Shared_Hugetlb: 0 kB 431Private_Hugetlb: 0 kB 432Swap: 0 kB 433SwapPss: 0 kB 434KernelPageSize: 4 kB 435MMUPageSize: 4 kB 436Locked: 0 kB 437VmFlags: rd ex mr mw me dw 438 439the first of these lines shows the same information as is displayed for the 440mapping in /proc/PID/maps. The remaining lines show the size of the mapping 441(size), the amount of the mapping that is currently resident in RAM (RSS), the 442process' proportional share of this mapping (PSS), the number of clean and 443dirty private pages in the mapping. 444 445The "proportional set size" (PSS) of a process is the count of pages it has 446in memory, where each page is divided by the number of processes sharing it. 447So if a process has 1000 pages all to itself, and 1000 shared with one other 448process, its PSS will be 1500. 449Note that even a page which is part of a MAP_SHARED mapping, but has only 450a single pte mapped, i.e. is currently used by only one process, is accounted 451as private and not as shared. 452"Referenced" indicates the amount of memory currently marked as referenced or 453accessed. 454"Anonymous" shows the amount of memory that does not belong to any file. Even 455a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 456and a page is modified, the file page is replaced by a private anonymous copy. 457"AnonHugePages" shows the ammount of memory backed by transparent hugepage. 458"Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by 459hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 460reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 461"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 462"SwapPss" shows proportional swap share of this mapping. 463"Locked" indicates whether the mapping is locked in memory or not. 464 465"VmFlags" field deserves a separate description. This member represents the kernel 466flags associated with the particular virtual memory area in two letter encoded 467manner. The codes are the following: 468 rd - readable 469 wr - writeable 470 ex - executable 471 sh - shared 472 mr - may read 473 mw - may write 474 me - may execute 475 ms - may share 476 gd - stack segment growns down 477 pf - pure PFN range 478 dw - disabled write to the mapped file 479 lo - pages are locked in memory 480 io - memory mapped I/O area 481 sr - sequential read advise provided 482 rr - random read advise provided 483 dc - do not copy area on fork 484 de - do not expand area on remapping 485 ac - area is accountable 486 nr - swap space is not reserved for the area 487 ht - area uses huge tlb pages 488 ar - architecture specific flag 489 dd - do not include area into core dump 490 sd - soft-dirty flag 491 mm - mixed map area 492 hg - huge page advise flag 493 nh - no-huge page advise flag 494 mg - mergable advise flag 495 496Note that there is no guarantee that every flag and associated mnemonic will 497be present in all further kernel releases. Things get changed, the flags may 498be vanished or the reverse -- new added. 499 500This file is only present if the CONFIG_MMU kernel configuration option is 501enabled. 502 503The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 504bits on both physical and virtual pages associated with a process, and the 505soft-dirty bit on pte (see Documentation/vm/soft-dirty.txt for details). 506To clear the bits for all the pages associated with the process 507 > echo 1 > /proc/PID/clear_refs 508 509To clear the bits for the anonymous pages associated with the process 510 > echo 2 > /proc/PID/clear_refs 511 512To clear the bits for the file mapped pages associated with the process 513 > echo 3 > /proc/PID/clear_refs 514 515To clear the soft-dirty bit 516 > echo 4 > /proc/PID/clear_refs 517 518To reset the peak resident set size ("high water mark") to the process's 519current value: 520 > echo 5 > /proc/PID/clear_refs 521 522Any other value written to /proc/PID/clear_refs will have no effect. 523 524The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 525using /proc/kpageflags and number of times a page is mapped using 526/proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt. 527 528The /proc/pid/numa_maps is an extension based on maps, showing the memory 529locality and binding policy, as well as the memory usage (in pages) of 530each mapping. The output follows a general format where mapping details get 531summarized separated by blank spaces, one mapping per each file line: 532 533address policy mapping details 534 53500400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 53600600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 5373206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 538320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 5393206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 5403206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 5413206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 542320698b000 default file=/lib64/libc-2.12.so 5433206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 5443206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 5453206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 5467f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 5477f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 5487f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 5497fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 5507fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 551 552Where: 553"address" is the starting address for the mapping; 554"policy" reports the NUMA memory policy set for the mapping (see vm/numa_memory_policy.txt); 555"mapping details" summarizes mapping data such as mapping type, page usage counters, 556node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 557size, in KB, that is backing the mapping up. 558 5591.2 Kernel data 560--------------- 561 562Similar to the process entries, the kernel data files give information about 563the running kernel. The files used to obtain this information are contained in 564/proc and are listed in Table 1-5. Not all of these will be present in your 565system. It depends on the kernel configuration and the loaded modules, which 566files are there, and which are missing. 567 568Table 1-5: Kernel info in /proc 569.............................................................................. 570 File Content 571 apm Advanced power management info 572 buddyinfo Kernel memory allocator information (see text) (2.5) 573 bus Directory containing bus specific information 574 cmdline Kernel command line 575 cpuinfo Info about the CPU 576 devices Available devices (block and character) 577 dma Used DMS channels 578 filesystems Supported filesystems 579 driver Various drivers grouped here, currently rtc (2.4) 580 execdomains Execdomains, related to security (2.4) 581 fb Frame Buffer devices (2.4) 582 fs File system parameters, currently nfs/exports (2.4) 583 ide Directory containing info about the IDE subsystem 584 interrupts Interrupt usage 585 iomem Memory map (2.4) 586 ioports I/O port usage 587 irq Masks for irq to cpu affinity (2.4)(smp?) 588 isapnp ISA PnP (Plug&Play) Info (2.4) 589 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 590 kmsg Kernel messages 591 ksyms Kernel symbol table 592 loadavg Load average of last 1, 5 & 15 minutes 593 locks Kernel locks 594 meminfo Memory info 595 misc Miscellaneous 596 modules List of loaded modules 597 mounts Mounted filesystems 598 net Networking info (see text) 599 pagetypeinfo Additional page allocator information (see text) (2.5) 600 partitions Table of partitions known to the system 601 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 602 decoupled by lspci (2.4) 603 rtc Real time clock 604 scsi SCSI info (see text) 605 slabinfo Slab pool info 606 softirqs softirq usage 607 stat Overall statistics 608 swaps Swap space utilization 609 sys See chapter 2 610 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 611 tty Info of tty drivers 612 uptime Wall clock since boot, combined idle time of all cpus 613 version Kernel version 614 video bttv info of video resources (2.4) 615 vmallocinfo Show vmalloced areas 616.............................................................................. 617 618You can, for example, check which interrupts are currently in use and what 619they are used for by looking in the file /proc/interrupts: 620 621 > cat /proc/interrupts 622 CPU0 623 0: 8728810 XT-PIC timer 624 1: 895 XT-PIC keyboard 625 2: 0 XT-PIC cascade 626 3: 531695 XT-PIC aha152x 627 4: 2014133 XT-PIC serial 628 5: 44401 XT-PIC pcnet_cs 629 8: 2 XT-PIC rtc 630 11: 8 XT-PIC i82365 631 12: 182918 XT-PIC PS/2 Mouse 632 13: 1 XT-PIC fpu 633 14: 1232265 XT-PIC ide0 634 15: 7 XT-PIC ide1 635 NMI: 0 636 637In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 638output of a SMP machine): 639 640 > cat /proc/interrupts 641 642 CPU0 CPU1 643 0: 1243498 1214548 IO-APIC-edge timer 644 1: 8949 8958 IO-APIC-edge keyboard 645 2: 0 0 XT-PIC cascade 646 5: 11286 10161 IO-APIC-edge soundblaster 647 8: 1 0 IO-APIC-edge rtc 648 9: 27422 27407 IO-APIC-edge 3c503 649 12: 113645 113873 IO-APIC-edge PS/2 Mouse 650 13: 0 0 XT-PIC fpu 651 14: 22491 24012 IO-APIC-edge ide0 652 15: 2183 2415 IO-APIC-edge ide1 653 17: 30564 30414 IO-APIC-level eth0 654 18: 177 164 IO-APIC-level bttv 655 NMI: 2457961 2457959 656 LOC: 2457882 2457881 657 ERR: 2155 658 659NMI is incremented in this case because every timer interrupt generates a NMI 660(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 661 662LOC is the local interrupt counter of the internal APIC of every CPU. 663 664ERR is incremented in the case of errors in the IO-APIC bus (the bus that 665connects the CPUs in a SMP system. This means that an error has been detected, 666the IO-APIC automatically retry the transmission, so it should not be a big 667problem, but you should read the SMP-FAQ. 668 669In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 670/proc/interrupts to display every IRQ vector in use by the system, not 671just those considered 'most important'. The new vectors are: 672 673 THR -- interrupt raised when a machine check threshold counter 674 (typically counting ECC corrected errors of memory or cache) exceeds 675 a configurable threshold. Only available on some systems. 676 677 TRM -- a thermal event interrupt occurs when a temperature threshold 678 has been exceeded for the CPU. This interrupt may also be generated 679 when the temperature drops back to normal. 680 681 SPU -- a spurious interrupt is some interrupt that was raised then lowered 682 by some IO device before it could be fully processed by the APIC. Hence 683 the APIC sees the interrupt but does not know what device it came from. 684 For this case the APIC will generate the interrupt with a IRQ vector 685 of 0xff. This might also be generated by chipset bugs. 686 687 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are 688 sent from one CPU to another per the needs of the OS. Typically, 689 their statistics are used by kernel developers and interested users to 690 determine the occurrence of interrupts of the given type. 691 692The above IRQ vectors are displayed only when relevant. For example, 693the threshold vector does not exist on x86_64 platforms. Others are 694suppressed when the system is a uniprocessor. As of this writing, only 695i386 and x86_64 platforms support the new IRQ vector displays. 696 697Of some interest is the introduction of the /proc/irq directory to 2.4. 698It could be used to set IRQ to CPU affinity, this means that you can "hook" an 699IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 700irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 701prof_cpu_mask. 702 703For example 704 > ls /proc/irq/ 705 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 706 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 707 > ls /proc/irq/0/ 708 smp_affinity 709 710smp_affinity is a bitmask, in which you can specify which CPUs can handle the 711IRQ, you can set it by doing: 712 713 > echo 1 > /proc/irq/10/smp_affinity 714 715This means that only the first CPU will handle the IRQ, but you can also echo 7165 which means that only the first and fourth CPU can handle the IRQ. 717 718The contents of each smp_affinity file is the same by default: 719 720 > cat /proc/irq/0/smp_affinity 721 ffffffff 722 723There is an alternate interface, smp_affinity_list which allows specifying 724a cpu range instead of a bitmask: 725 726 > cat /proc/irq/0/smp_affinity_list 727 1024-1031 728 729The default_smp_affinity mask applies to all non-active IRQs, which are the 730IRQs which have not yet been allocated/activated, and hence which lack a 731/proc/irq/[0-9]* directory. 732 733The node file on an SMP system shows the node to which the device using the IRQ 734reports itself as being attached. This hardware locality information does not 735include information about any possible driver locality preference. 736 737prof_cpu_mask specifies which CPUs are to be profiled by the system wide 738profiler. Default value is ffffffff (all cpus if there are only 32 of them). 739 740The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 741between all the CPUs which are allowed to handle it. As usual the kernel has 742more info than you and does a better job than you, so the defaults are the 743best choice for almost everyone. [Note this applies only to those IO-APIC's 744that support "Round Robin" interrupt distribution.] 745 746There are three more important subdirectories in /proc: net, scsi, and sys. 747The general rule is that the contents, or even the existence of these 748directories, depend on your kernel configuration. If SCSI is not enabled, the 749directory scsi may not exist. The same is true with the net, which is there 750only when networking support is present in the running kernel. 751 752The slabinfo file gives information about memory usage at the slab level. 753Linux uses slab pools for memory management above page level in version 2.2. 754Commonly used objects have their own slab pool (such as network buffers, 755directory cache, and so on). 756 757.............................................................................. 758 759> cat /proc/buddyinfo 760 761Node 0, zone DMA 0 4 5 4 4 3 ... 762Node 0, zone Normal 1 0 0 1 101 8 ... 763Node 0, zone HighMem 2 0 0 1 1 0 ... 764 765External fragmentation is a problem under some workloads, and buddyinfo is a 766useful tool for helping diagnose these problems. Buddyinfo will give you a 767clue as to how big an area you can safely allocate, or why a previous 768allocation failed. 769 770Each column represents the number of pages of a certain order which are 771available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 772ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 773available in ZONE_NORMAL, etc... 774 775More information relevant to external fragmentation can be found in 776pagetypeinfo. 777 778> cat /proc/pagetypeinfo 779Page block order: 9 780Pages per block: 512 781 782Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 783Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 784Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 785Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 786Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 787Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 788Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 789Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 790Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 791Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 792Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 793 794Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 795Node 0, zone DMA 2 0 5 1 0 796Node 0, zone DMA32 41 6 967 2 0 797 798Fragmentation avoidance in the kernel works by grouping pages of different 799migrate types into the same contiguous regions of memory called page blocks. 800A page block is typically the size of the default hugepage size e.g. 2MB on 801X86-64. By keeping pages grouped based on their ability to move, the kernel 802can reclaim pages within a page block to satisfy a high-order allocation. 803 804The pagetypinfo begins with information on the size of a page block. It 805then gives the same type of information as buddyinfo except broken down 806by migrate-type and finishes with details on how many page blocks of each 807type exist. 808 809If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 810from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can 811make an estimate of the likely number of huge pages that can be allocated 812at a given point in time. All the "Movable" blocks should be allocatable 813unless memory has been mlock()'d. Some of the Reclaimable blocks should 814also be allocatable although a lot of filesystem metadata may have to be 815reclaimed to achieve this. 816 817.............................................................................. 818 819meminfo: 820 821Provides information about distribution and utilization of memory. This 822varies by architecture and compile options. The following is from a 82316GB PIII, which has highmem enabled. You may not have all of these fields. 824 825> cat /proc/meminfo 826 827MemTotal: 16344972 kB 828MemFree: 13634064 kB 829MemAvailable: 14836172 kB 830Buffers: 3656 kB 831Cached: 1195708 kB 832SwapCached: 0 kB 833Active: 891636 kB 834Inactive: 1077224 kB 835HighTotal: 15597528 kB 836HighFree: 13629632 kB 837LowTotal: 747444 kB 838LowFree: 4432 kB 839SwapTotal: 0 kB 840SwapFree: 0 kB 841Dirty: 968 kB 842Writeback: 0 kB 843AnonPages: 861800 kB 844Mapped: 280372 kB 845Slab: 284364 kB 846SReclaimable: 159856 kB 847SUnreclaim: 124508 kB 848PageTables: 24448 kB 849NFS_Unstable: 0 kB 850Bounce: 0 kB 851WritebackTmp: 0 kB 852CommitLimit: 7669796 kB 853Committed_AS: 100056 kB 854VmallocTotal: 112216 kB 855VmallocUsed: 428 kB 856VmallocChunk: 111088 kB 857AnonHugePages: 49152 kB 858 859 MemTotal: Total usable ram (i.e. physical ram minus a few reserved 860 bits and the kernel binary code) 861 MemFree: The sum of LowFree+HighFree 862MemAvailable: An estimate of how much memory is available for starting new 863 applications, without swapping. Calculated from MemFree, 864 SReclaimable, the size of the file LRU lists, and the low 865 watermarks in each zone. 866 The estimate takes into account that the system needs some 867 page cache to function well, and that not all reclaimable 868 slab will be reclaimable, due to items being in use. The 869 impact of those factors will vary from system to system. 870 Buffers: Relatively temporary storage for raw disk blocks 871 shouldn't get tremendously large (20MB or so) 872 Cached: in-memory cache for files read from the disk (the 873 pagecache). Doesn't include SwapCached 874 SwapCached: Memory that once was swapped out, is swapped back in but 875 still also is in the swapfile (if memory is needed it 876 doesn't need to be swapped out AGAIN because it is already 877 in the swapfile. This saves I/O) 878 Active: Memory that has been used more recently and usually not 879 reclaimed unless absolutely necessary. 880 Inactive: Memory which has been less recently used. It is more 881 eligible to be reclaimed for other purposes 882 HighTotal: 883 HighFree: Highmem is all memory above ~860MB of physical memory 884 Highmem areas are for use by userspace programs, or 885 for the pagecache. The kernel must use tricks to access 886 this memory, making it slower to access than lowmem. 887 LowTotal: 888 LowFree: Lowmem is memory which can be used for everything that 889 highmem can be used for, but it is also available for the 890 kernel's use for its own data structures. Among many 891 other things, it is where everything from the Slab is 892 allocated. Bad things happen when you're out of lowmem. 893 SwapTotal: total amount of swap space available 894 SwapFree: Memory which has been evicted from RAM, and is temporarily 895 on the disk 896 Dirty: Memory which is waiting to get written back to the disk 897 Writeback: Memory which is actively being written back to the disk 898 AnonPages: Non-file backed pages mapped into userspace page tables 899AnonHugePages: Non-file backed huge pages mapped into userspace page tables 900 Mapped: files which have been mmaped, such as libraries 901 Slab: in-kernel data structures cache 902SReclaimable: Part of Slab, that might be reclaimed, such as caches 903 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure 904 PageTables: amount of memory dedicated to the lowest level of page 905 tables. 906NFS_Unstable: NFS pages sent to the server, but not yet committed to stable 907 storage 908 Bounce: Memory used for block device "bounce buffers" 909WritebackTmp: Memory used by FUSE for temporary writeback buffers 910 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'), 911 this is the total amount of memory currently available to 912 be allocated on the system. This limit is only adhered to 913 if strict overcommit accounting is enabled (mode 2 in 914 'vm.overcommit_memory'). 915 The CommitLimit is calculated with the following formula: 916 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 917 overcommit_ratio / 100 + [total swap pages] 918 For example, on a system with 1G of physical RAM and 7G 919 of swap with a `vm.overcommit_ratio` of 30 it would 920 yield a CommitLimit of 7.3G. 921 For more details, see the memory overcommit documentation 922 in vm/overcommit-accounting. 923Committed_AS: The amount of memory presently allocated on the system. 924 The committed memory is a sum of all of the memory which 925 has been allocated by processes, even if it has not been 926 "used" by them as of yet. A process which malloc()'s 1G 927 of memory, but only touches 300M of it will show up as 928 using 1G. This 1G is memory which has been "committed" to 929 by the VM and can be used at any time by the allocating 930 application. With strict overcommit enabled on the system 931 (mode 2 in 'vm.overcommit_memory'),allocations which would 932 exceed the CommitLimit (detailed above) will not be permitted. 933 This is useful if one needs to guarantee that processes will 934 not fail due to lack of memory once that memory has been 935 successfully allocated. 936VmallocTotal: total size of vmalloc memory area 937 VmallocUsed: amount of vmalloc area which is used 938VmallocChunk: largest contiguous block of vmalloc area which is free 939 940.............................................................................. 941 942vmallocinfo: 943 944Provides information about vmalloced/vmaped areas. One line per area, 945containing the virtual address range of the area, size in bytes, 946caller information of the creator, and optional information depending 947on the kind of area : 948 949 pages=nr number of pages 950 phys=addr if a physical address was specified 951 ioremap I/O mapping (ioremap() and friends) 952 vmalloc vmalloc() area 953 vmap vmap()ed pages 954 user VM_USERMAP area 955 vpages buffer for pages pointers was vmalloced (huge area) 956 N<node>=nr (Only on NUMA kernels) 957 Number of pages allocated on memory node <node> 958 959> cat /proc/vmallocinfo 9600xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 961 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 9620xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 963 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 9640xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 965 phys=7fee8000 ioremap 9660xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 967 phys=7fee7000 ioremap 9680xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 9690xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 970 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 9710xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 972 pages=2 vmalloc N1=2 9730xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 974 /0x130 [x_tables] pages=4 vmalloc N0=4 9750xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 976 pages=14 vmalloc N2=14 9770xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 978 pages=4 vmalloc N1=4 9790xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 980 pages=2 vmalloc N1=2 9810xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 982 pages=10 vmalloc N0=10 983 984.............................................................................. 985 986softirqs: 987 988Provides counts of softirq handlers serviced since boot time, for each cpu. 989 990> cat /proc/softirqs 991 CPU0 CPU1 CPU2 CPU3 992 HI: 0 0 0 0 993 TIMER: 27166 27120 27097 27034 994 NET_TX: 0 0 0 17 995 NET_RX: 42 0 0 39 996 BLOCK: 0 0 107 1121 997 TASKLET: 0 0 0 290 998 SCHED: 27035 26983 26971 26746 999 HRTIMER: 0 0 0 0 1000 RCU: 1678 1769 2178 2250 1001 1002 10031.3 IDE devices in /proc/ide 1004---------------------------- 1005 1006The subdirectory /proc/ide contains information about all IDE devices of which 1007the kernel is aware. There is one subdirectory for each IDE controller, the 1008file drivers and a link for each IDE device, pointing to the device directory 1009in the controller specific subtree. 1010 1011The file drivers contains general information about the drivers used for the 1012IDE devices: 1013 1014 > cat /proc/ide/drivers 1015 ide-cdrom version 4.53 1016 ide-disk version 1.08 1017 1018More detailed information can be found in the controller specific 1019subdirectories. These are named ide0, ide1 and so on. Each of these 1020directories contains the files shown in table 1-6. 1021 1022 1023Table 1-6: IDE controller info in /proc/ide/ide? 1024.............................................................................. 1025 File Content 1026 channel IDE channel (0 or 1) 1027 config Configuration (only for PCI/IDE bridge) 1028 mate Mate name 1029 model Type/Chipset of IDE controller 1030.............................................................................. 1031 1032Each device connected to a controller has a separate subdirectory in the 1033controllers directory. The files listed in table 1-7 are contained in these 1034directories. 1035 1036 1037Table 1-7: IDE device information 1038.............................................................................. 1039 File Content 1040 cache The cache 1041 capacity Capacity of the medium (in 512Byte blocks) 1042 driver driver and version 1043 geometry physical and logical geometry 1044 identify device identify block 1045 media media type 1046 model device identifier 1047 settings device setup 1048 smart_thresholds IDE disk management thresholds 1049 smart_values IDE disk management values 1050.............................................................................. 1051 1052The most interesting file is settings. This file contains a nice overview of 1053the drive parameters: 1054 1055 # cat /proc/ide/ide0/hda/settings 1056 name value min max mode 1057 ---- ----- --- --- ---- 1058 bios_cyl 526 0 65535 rw 1059 bios_head 255 0 255 rw 1060 bios_sect 63 0 63 rw 1061 breada_readahead 4 0 127 rw 1062 bswap 0 0 1 r 1063 file_readahead 72 0 2097151 rw 1064 io_32bit 0 0 3 rw 1065 keepsettings 0 0 1 rw 1066 max_kb_per_request 122 1 127 rw 1067 multcount 0 0 8 rw 1068 nice1 1 0 1 rw 1069 nowerr 0 0 1 rw 1070 pio_mode write-only 0 255 w 1071 slow 0 0 1 rw 1072 unmaskirq 0 0 1 rw 1073 using_dma 0 0 1 rw 1074 1075 10761.4 Networking info in /proc/net 1077-------------------------------- 1078 1079The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1080additional values you get for IP version 6 if you configure the kernel to 1081support this. Table 1-9 lists the files and their meaning. 1082 1083 1084Table 1-8: IPv6 info in /proc/net 1085.............................................................................. 1086 File Content 1087 udp6 UDP sockets (IPv6) 1088 tcp6 TCP sockets (IPv6) 1089 raw6 Raw device statistics (IPv6) 1090 igmp6 IP multicast addresses, which this host joined (IPv6) 1091 if_inet6 List of IPv6 interface addresses 1092 ipv6_route Kernel routing table for IPv6 1093 rt6_stats Global IPv6 routing tables statistics 1094 sockstat6 Socket statistics (IPv6) 1095 snmp6 Snmp data (IPv6) 1096.............................................................................. 1097 1098 1099Table 1-9: Network info in /proc/net 1100.............................................................................. 1101 File Content 1102 arp Kernel ARP table 1103 dev network devices with statistics 1104 dev_mcast the Layer2 multicast groups a device is listening too 1105 (interface index, label, number of references, number of bound 1106 addresses). 1107 dev_stat network device status 1108 ip_fwchains Firewall chain linkage 1109 ip_fwnames Firewall chain names 1110 ip_masq Directory containing the masquerading tables 1111 ip_masquerade Major masquerading table 1112 netstat Network statistics 1113 raw raw device statistics 1114 route Kernel routing table 1115 rpc Directory containing rpc info 1116 rt_cache Routing cache 1117 snmp SNMP data 1118 sockstat Socket statistics 1119 tcp TCP sockets 1120 udp UDP sockets 1121 unix UNIX domain sockets 1122 wireless Wireless interface data (Wavelan etc) 1123 igmp IP multicast addresses, which this host joined 1124 psched Global packet scheduler parameters. 1125 netlink List of PF_NETLINK sockets 1126 ip_mr_vifs List of multicast virtual interfaces 1127 ip_mr_cache List of multicast routing cache 1128.............................................................................. 1129 1130You can use this information to see which network devices are available in 1131your system and how much traffic was routed over those devices: 1132 1133 > cat /proc/net/dev 1134 Inter-|Receive |[... 1135 face |bytes packets errs drop fifo frame compressed multicast|[... 1136 lo: 908188 5596 0 0 0 0 0 0 [... 1137 ppp0:15475140 20721 410 0 0 410 0 0 [... 1138 eth0: 614530 7085 0 0 0 0 0 1 [... 1139 1140 ...] Transmit 1141 ...] bytes packets errs drop fifo colls carrier compressed 1142 ...] 908188 5596 0 0 0 0 0 0 1143 ...] 1375103 17405 0 0 0 0 0 0 1144 ...] 1703981 5535 0 0 0 3 0 0 1145 1146In addition, each Channel Bond interface has its own directory. For 1147example, the bond0 device will have a directory called /proc/net/bond0/. 1148It will contain information that is specific to that bond, such as the 1149current slaves of the bond, the link status of the slaves, and how 1150many times the slaves link has failed. 1151 11521.5 SCSI info 1153------------- 1154 1155If you have a SCSI host adapter in your system, you'll find a subdirectory 1156named after the driver for this adapter in /proc/scsi. You'll also see a list 1157of all recognized SCSI devices in /proc/scsi: 1158 1159 >cat /proc/scsi/scsi 1160 Attached devices: 1161 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1162 Vendor: IBM Model: DGHS09U Rev: 03E0 1163 Type: Direct-Access ANSI SCSI revision: 03 1164 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1165 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1166 Type: CD-ROM ANSI SCSI revision: 02 1167 1168 1169The directory named after the driver has one file for each adapter found in 1170the system. These files contain information about the controller, including 1171the used IRQ and the IO address range. The amount of information shown is 1172dependent on the adapter you use. The example shows the output for an Adaptec 1173AHA-2940 SCSI adapter: 1174 1175 > cat /proc/scsi/aic7xxx/0 1176 1177 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1178 Compile Options: 1179 TCQ Enabled By Default : Disabled 1180 AIC7XXX_PROC_STATS : Disabled 1181 AIC7XXX_RESET_DELAY : 5 1182 Adapter Configuration: 1183 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1184 Ultra Wide Controller 1185 PCI MMAPed I/O Base: 0xeb001000 1186 Adapter SEEPROM Config: SEEPROM found and used. 1187 Adaptec SCSI BIOS: Enabled 1188 IRQ: 10 1189 SCBs: Active 0, Max Active 2, 1190 Allocated 15, HW 16, Page 255 1191 Interrupts: 160328 1192 BIOS Control Word: 0x18b6 1193 Adapter Control Word: 0x005b 1194 Extended Translation: Enabled 1195 Disconnect Enable Flags: 0xffff 1196 Ultra Enable Flags: 0x0001 1197 Tag Queue Enable Flags: 0x0000 1198 Ordered Queue Tag Flags: 0x0000 1199 Default Tag Queue Depth: 8 1200 Tagged Queue By Device array for aic7xxx host instance 0: 1201 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1202 Actual queue depth per device for aic7xxx host instance 0: 1203 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1204 Statistics: 1205 (scsi0:0:0:0) 1206 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1207 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1208 Total transfers 160151 (74577 reads and 85574 writes) 1209 (scsi0:0:6:0) 1210 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1211 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1212 Total transfers 0 (0 reads and 0 writes) 1213 1214 12151.6 Parallel port info in /proc/parport 1216--------------------------------------- 1217 1218The directory /proc/parport contains information about the parallel ports of 1219your system. It has one subdirectory for each port, named after the port 1220number (0,1,2,...). 1221 1222These directories contain the four files shown in Table 1-10. 1223 1224 1225Table 1-10: Files in /proc/parport 1226.............................................................................. 1227 File Content 1228 autoprobe Any IEEE-1284 device ID information that has been acquired. 1229 devices list of the device drivers using that port. A + will appear by the 1230 name of the device currently using the port (it might not appear 1231 against any). 1232 hardware Parallel port's base address, IRQ line and DMA channel. 1233 irq IRQ that parport is using for that port. This is in a separate 1234 file to allow you to alter it by writing a new value in (IRQ 1235 number or none). 1236.............................................................................. 1237 12381.7 TTY info in /proc/tty 1239------------------------- 1240 1241Information about the available and actually used tty's can be found in the 1242directory /proc/tty.You'll find entries for drivers and line disciplines in 1243this directory, as shown in Table 1-11. 1244 1245 1246Table 1-11: Files in /proc/tty 1247.............................................................................. 1248 File Content 1249 drivers list of drivers and their usage 1250 ldiscs registered line disciplines 1251 driver/serial usage statistic and status of single tty lines 1252.............................................................................. 1253 1254To see which tty's are currently in use, you can simply look into the file 1255/proc/tty/drivers: 1256 1257 > cat /proc/tty/drivers 1258 pty_slave /dev/pts 136 0-255 pty:slave 1259 pty_master /dev/ptm 128 0-255 pty:master 1260 pty_slave /dev/ttyp 3 0-255 pty:slave 1261 pty_master /dev/pty 2 0-255 pty:master 1262 serial /dev/cua 5 64-67 serial:callout 1263 serial /dev/ttyS 4 64-67 serial 1264 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1265 /dev/ptmx /dev/ptmx 5 2 system 1266 /dev/console /dev/console 5 1 system:console 1267 /dev/tty /dev/tty 5 0 system:/dev/tty 1268 unknown /dev/tty 4 1-63 console 1269 1270 12711.8 Miscellaneous kernel statistics in /proc/stat 1272------------------------------------------------- 1273 1274Various pieces of information about kernel activity are available in the 1275/proc/stat file. All of the numbers reported in this file are aggregates 1276since the system first booted. For a quick look, simply cat the file: 1277 1278 > cat /proc/stat 1279 cpu 2255 34 2290 22625563 6290 127 456 0 0 0 1280 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0 1281 cpu1 1123 0 849 11313845 2614 0 18 0 0 0 1282 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...] 1283 ctxt 1990473 1284 btime 1062191376 1285 processes 2915 1286 procs_running 1 1287 procs_blocked 0 1288 softirq 183433 0 21755 12 39 1137 231 21459 2263 1289 1290The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1291lines. These numbers identify the amount of time the CPU has spent performing 1292different kinds of work. Time units are in USER_HZ (typically hundredths of a 1293second). The meanings of the columns are as follows, from left to right: 1294 1295- user: normal processes executing in user mode 1296- nice: niced processes executing in user mode 1297- system: processes executing in kernel mode 1298- idle: twiddling thumbs 1299- iowait: waiting for I/O to complete 1300- irq: servicing interrupts 1301- softirq: servicing softirqs 1302- steal: involuntary wait 1303- guest: running a normal guest 1304- guest_nice: running a niced guest 1305 1306The "intr" line gives counts of interrupts serviced since boot time, for each 1307of the possible system interrupts. The first column is the total of all 1308interrupts serviced including unnumbered architecture specific interrupts; 1309each subsequent column is the total for that particular numbered interrupt. 1310Unnumbered interrupts are not shown, only summed into the total. 1311 1312The "ctxt" line gives the total number of context switches across all CPUs. 1313 1314The "btime" line gives the time at which the system booted, in seconds since 1315the Unix epoch. 1316 1317The "processes" line gives the number of processes and threads created, which 1318includes (but is not limited to) those created by calls to the fork() and 1319clone() system calls. 1320 1321The "procs_running" line gives the total number of threads that are 1322running or ready to run (i.e., the total number of runnable threads). 1323 1324The "procs_blocked" line gives the number of processes currently blocked, 1325waiting for I/O to complete. 1326 1327The "softirq" line gives counts of softirqs serviced since boot time, for each 1328of the possible system softirqs. The first column is the total of all 1329softirqs serviced; each subsequent column is the total for that particular 1330softirq. 1331 1332 13331.9 Ext4 file system parameters 1334------------------------------- 1335 1336Information about mounted ext4 file systems can be found in 1337/proc/fs/ext4. Each mounted filesystem will have a directory in 1338/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1339/proc/fs/ext4/dm-0). The files in each per-device directory are shown 1340in Table 1-12, below. 1341 1342Table 1-12: Files in /proc/fs/ext4/<devname> 1343.............................................................................. 1344 File Content 1345 mb_groups details of multiblock allocator buddy cache of free blocks 1346.............................................................................. 1347 13482.0 /proc/consoles 1349------------------ 1350Shows registered system console lines. 1351 1352To see which character device lines are currently used for the system console 1353/dev/console, you may simply look into the file /proc/consoles: 1354 1355 > cat /proc/consoles 1356 tty0 -WU (ECp) 4:7 1357 ttyS0 -W- (Ep) 4:64 1358 1359The columns are: 1360 1361 device name of the device 1362 operations R = can do read operations 1363 W = can do write operations 1364 U = can do unblank 1365 flags E = it is enabled 1366 C = it is preferred console 1367 B = it is primary boot console 1368 p = it is used for printk buffer 1369 b = it is not a TTY but a Braille device 1370 a = it is safe to use when cpu is offline 1371 major:minor major and minor number of the device separated by a colon 1372 1373------------------------------------------------------------------------------ 1374Summary 1375------------------------------------------------------------------------------ 1376The /proc file system serves information about the running system. It not only 1377allows access to process data but also allows you to request the kernel status 1378by reading files in the hierarchy. 1379 1380The directory structure of /proc reflects the types of information and makes 1381it easy, if not obvious, where to look for specific data. 1382------------------------------------------------------------------------------ 1383 1384------------------------------------------------------------------------------ 1385CHAPTER 2: MODIFYING SYSTEM PARAMETERS 1386------------------------------------------------------------------------------ 1387 1388------------------------------------------------------------------------------ 1389In This Chapter 1390------------------------------------------------------------------------------ 1391* Modifying kernel parameters by writing into files found in /proc/sys 1392* Exploring the files which modify certain parameters 1393* Review of the /proc/sys file tree 1394------------------------------------------------------------------------------ 1395 1396 1397A very interesting part of /proc is the directory /proc/sys. This is not only 1398a source of information, it also allows you to change parameters within the 1399kernel. Be very careful when attempting this. You can optimize your system, 1400but you can also cause it to crash. Never alter kernel parameters on a 1401production system. Set up a development machine and test to make sure that 1402everything works the way you want it to. You may have no alternative but to 1403reboot the machine once an error has been made. 1404 1405To change a value, simply echo the new value into the file. An example is 1406given below in the section on the file system data. You need to be root to do 1407this. You can create your own boot script to perform this every time your 1408system boots. 1409 1410The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1411general things in the operation of the Linux kernel. Since some of the files 1412can inadvertently disrupt your system, it is advisable to read both 1413documentation and source before actually making adjustments. In any case, be 1414very careful when writing to any of these files. The entries in /proc may 1415change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1416review the kernel documentation in the directory /usr/src/linux/Documentation. 1417This chapter is heavily based on the documentation included in the pre 2.2 1418kernels, and became part of it in version 2.2.1 of the Linux kernel. 1419 1420Please see: Documentation/sysctl/ directory for descriptions of these 1421entries. 1422 1423------------------------------------------------------------------------------ 1424Summary 1425------------------------------------------------------------------------------ 1426Certain aspects of kernel behavior can be modified at runtime, without the 1427need to recompile the kernel, or even to reboot the system. The files in the 1428/proc/sys tree can not only be read, but also modified. You can use the echo 1429command to write value into these files, thereby changing the default settings 1430of the kernel. 1431------------------------------------------------------------------------------ 1432 1433------------------------------------------------------------------------------ 1434CHAPTER 3: PER-PROCESS PARAMETERS 1435------------------------------------------------------------------------------ 1436 14373.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1438-------------------------------------------------------------------------------- 1439 1440These file can be used to adjust the badness heuristic used to select which 1441process gets killed in out of memory conditions. 1442 1443The badness heuristic assigns a value to each candidate task ranging from 0 1444(never kill) to 1000 (always kill) to determine which process is targeted. The 1445units are roughly a proportion along that range of allowed memory the process 1446may allocate from based on an estimation of its current memory and swap use. 1447For example, if a task is using all allowed memory, its badness score will be 14481000. If it is using half of its allowed memory, its score will be 500. 1449 1450There is an additional factor included in the badness score: the current memory 1451and swap usage is discounted by 3% for root processes. 1452 1453The amount of "allowed" memory depends on the context in which the oom killer 1454was called. If it is due to the memory assigned to the allocating task's cpuset 1455being exhausted, the allowed memory represents the set of mems assigned to that 1456cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1457memory represents the set of mempolicy nodes. If it is due to a memory 1458limit (or swap limit) being reached, the allowed memory is that configured 1459limit. Finally, if it is due to the entire system being out of memory, the 1460allowed memory represents all allocatable resources. 1461 1462The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1463is used to determine which task to kill. Acceptable values range from -1000 1464(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1465polarize the preference for oom killing either by always preferring a certain 1466task or completely disabling it. The lowest possible value, -1000, is 1467equivalent to disabling oom killing entirely for that task since it will always 1468report a badness score of 0. 1469 1470Consequently, it is very simple for userspace to define the amount of memory to 1471consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1472example, is roughly equivalent to allowing the remainder of tasks sharing the 1473same system, cpuset, mempolicy, or memory controller resources to use at least 147450% more memory. A value of -500, on the other hand, would be roughly 1475equivalent to discounting 50% of the task's allowed memory from being considered 1476as scoring against the task. 1477 1478For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1479be used to tune the badness score. Its acceptable values range from -16 1480(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1481(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1482scaled linearly with /proc/<pid>/oom_score_adj. 1483 1484The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1485value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1486requires CAP_SYS_RESOURCE. 1487 1488Caveat: when a parent task is selected, the oom killer will sacrifice any first 1489generation children with separate address spaces instead, if possible. This 1490avoids servers and important system daemons from being killed and loses the 1491minimal amount of work. 1492 1493 14943.2 /proc/<pid>/oom_score - Display current oom-killer score 1495------------------------------------------------------------- 1496 1497This file can be used to check the current score used by the oom-killer is for 1498any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1499process should be killed in an out-of-memory situation. 1500 1501 15023.3 /proc/<pid>/io - Display the IO accounting fields 1503------------------------------------------------------- 1504 1505This file contains IO statistics for each running process 1506 1507Example 1508------- 1509 1510test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1511[1] 3828 1512 1513test:/tmp # cat /proc/3828/io 1514rchar: 323934931 1515wchar: 323929600 1516syscr: 632687 1517syscw: 632675 1518read_bytes: 0 1519write_bytes: 323932160 1520cancelled_write_bytes: 0 1521 1522 1523Description 1524----------- 1525 1526rchar 1527----- 1528 1529I/O counter: chars read 1530The number of bytes which this task has caused to be read from storage. This 1531is simply the sum of bytes which this process passed to read() and pread(). 1532It includes things like tty IO and it is unaffected by whether or not actual 1533physical disk IO was required (the read might have been satisfied from 1534pagecache) 1535 1536 1537wchar 1538----- 1539 1540I/O counter: chars written 1541The number of bytes which this task has caused, or shall cause to be written 1542to disk. Similar caveats apply here as with rchar. 1543 1544 1545syscr 1546----- 1547 1548I/O counter: read syscalls 1549Attempt to count the number of read I/O operations, i.e. syscalls like read() 1550and pread(). 1551 1552 1553syscw 1554----- 1555 1556I/O counter: write syscalls 1557Attempt to count the number of write I/O operations, i.e. syscalls like 1558write() and pwrite(). 1559 1560 1561read_bytes 1562---------- 1563 1564I/O counter: bytes read 1565Attempt to count the number of bytes which this process really did cause to 1566be fetched from the storage layer. Done at the submit_bio() level, so it is 1567accurate for block-backed filesystems. <please add status regarding NFS and 1568CIFS at a later time> 1569 1570 1571write_bytes 1572----------- 1573 1574I/O counter: bytes written 1575Attempt to count the number of bytes which this process caused to be sent to 1576the storage layer. This is done at page-dirtying time. 1577 1578 1579cancelled_write_bytes 1580--------------------- 1581 1582The big inaccuracy here is truncate. If a process writes 1MB to a file and 1583then deletes the file, it will in fact perform no writeout. But it will have 1584been accounted as having caused 1MB of write. 1585In other words: The number of bytes which this process caused to not happen, 1586by truncating pagecache. A task can cause "negative" IO too. If this task 1587truncates some dirty pagecache, some IO which another task has been accounted 1588for (in its write_bytes) will not be happening. We _could_ just subtract that 1589from the truncating task's write_bytes, but there is information loss in doing 1590that. 1591 1592 1593Note 1594---- 1595 1596At its current implementation state, this is a bit racy on 32-bit machines: if 1597process A reads process B's /proc/pid/io while process B is updating one of 1598those 64-bit counters, process A could see an intermediate result. 1599 1600 1601More information about this can be found within the taskstats documentation in 1602Documentation/accounting. 1603 16043.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1605--------------------------------------------------------------- 1606When a process is dumped, all anonymous memory is written to a core file as 1607long as the size of the core file isn't limited. But sometimes we don't want 1608to dump some memory segments, for example, huge shared memory or DAX. 1609Conversely, sometimes we want to save file-backed memory segments into a core 1610file, not only the individual files. 1611 1612/proc/<pid>/coredump_filter allows you to customize which memory segments 1613will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1614of memory types. If a bit of the bitmask is set, memory segments of the 1615corresponding memory type are dumped, otherwise they are not dumped. 1616 1617The following 9 memory types are supported: 1618 - (bit 0) anonymous private memory 1619 - (bit 1) anonymous shared memory 1620 - (bit 2) file-backed private memory 1621 - (bit 3) file-backed shared memory 1622 - (bit 4) ELF header pages in file-backed private memory areas (it is 1623 effective only if the bit 2 is cleared) 1624 - (bit 5) hugetlb private memory 1625 - (bit 6) hugetlb shared memory 1626 - (bit 7) DAX private memory 1627 - (bit 8) DAX shared memory 1628 1629 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1630 are always dumped regardless of the bitmask status. 1631 1632 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1633 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1634 1635The default value of coredump_filter is 0x33; this means all anonymous memory 1636segments, ELF header pages and hugetlb private memory are dumped. 1637 1638If you don't want to dump all shared memory segments attached to pid 1234, 1639write 0x31 to the process's proc file. 1640 1641 $ echo 0x31 > /proc/1234/coredump_filter 1642 1643When a new process is created, the process inherits the bitmask status from its 1644parent. It is useful to set up coredump_filter before the program runs. 1645For example: 1646 1647 $ echo 0x7 > /proc/self/coredump_filter 1648 $ ./some_program 1649 16503.5 /proc/<pid>/mountinfo - Information about mounts 1651-------------------------------------------------------- 1652 1653This file contains lines of the form: 1654 165536 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1656(1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11) 1657 1658(1) mount ID: unique identifier of the mount (may be reused after umount) 1659(2) parent ID: ID of parent (or of self for the top of the mount tree) 1660(3) major:minor: value of st_dev for files on filesystem 1661(4) root: root of the mount within the filesystem 1662(5) mount point: mount point relative to the process's root 1663(6) mount options: per mount options 1664(7) optional fields: zero or more fields of the form "tag[:value]" 1665(8) separator: marks the end of the optional fields 1666(9) filesystem type: name of filesystem of the form "type[.subtype]" 1667(10) mount source: filesystem specific information or "none" 1668(11) super options: per super block options 1669 1670Parsers should ignore all unrecognised optional fields. Currently the 1671possible optional fields are: 1672 1673shared:X mount is shared in peer group X 1674master:X mount is slave to peer group X 1675propagate_from:X mount is slave and receives propagation from peer group X (*) 1676unbindable mount is unbindable 1677 1678(*) X is the closest dominant peer group under the process's root. If 1679X is the immediate master of the mount, or if there's no dominant peer 1680group under the same root, then only the "master:X" field is present 1681and not the "propagate_from:X" field. 1682 1683For more information on mount propagation see: 1684 1685 Documentation/filesystems/sharedsubtree.txt 1686 1687 16883.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1689-------------------------------------------------------- 1690These files provide a method to access a tasks comm value. It also allows for 1691a task to set its own or one of its thread siblings comm value. The comm value 1692is limited in size compared to the cmdline value, so writing anything longer 1693then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated 1694comm value. 1695 1696 16973.7 /proc/<pid>/task/<tid>/children - Information about task children 1698------------------------------------------------------------------------- 1699This file provides a fast way to retrieve first level children pids 1700of a task pointed by <pid>/<tid> pair. The format is a space separated 1701stream of pids. 1702 1703Note the "first level" here -- if a child has own children they will 1704not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children 1705to obtain the descendants. 1706 1707Since this interface is intended to be fast and cheap it doesn't 1708guarantee to provide precise results and some children might be 1709skipped, especially if they've exited right after we printed their 1710pids, so one need to either stop or freeze processes being inspected 1711if precise results are needed. 1712 1713 17143.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1715--------------------------------------------------------------- 1716This file provides information associated with an opened file. The regular 1717files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos' 1718represents the current offset of the opened file in decimal form [see lseek(2) 1719for details], 'flags' denotes the octal O_xxx mask the file has been 1720created with [see open(2) for details] and 'mnt_id' represents mount ID of 1721the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo 1722for details]. 1723 1724A typical output is 1725 1726 pos: 0 1727 flags: 0100002 1728 mnt_id: 19 1729 1730All locks associated with a file descriptor are shown in its fdinfo too. 1731 1732lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1733 1734The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1735pair provide additional information particular to the objects they represent. 1736 1737 Eventfd files 1738 ~~~~~~~~~~~~~ 1739 pos: 0 1740 flags: 04002 1741 mnt_id: 9 1742 eventfd-count: 5a 1743 1744 where 'eventfd-count' is hex value of a counter. 1745 1746 Signalfd files 1747 ~~~~~~~~~~~~~~ 1748 pos: 0 1749 flags: 04002 1750 mnt_id: 9 1751 sigmask: 0000000000000200 1752 1753 where 'sigmask' is hex value of the signal mask associated 1754 with a file. 1755 1756 Epoll files 1757 ~~~~~~~~~~~ 1758 pos: 0 1759 flags: 02 1760 mnt_id: 9 1761 tfd: 5 events: 1d data: ffffffffffffffff 1762 1763 where 'tfd' is a target file descriptor number in decimal form, 1764 'events' is events mask being watched and the 'data' is data 1765 associated with a target [see epoll(7) for more details]. 1766 1767 Fsnotify files 1768 ~~~~~~~~~~~~~~ 1769 For inotify files the format is the following 1770 1771 pos: 0 1772 flags: 02000000 1773 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 1774 1775 where 'wd' is a watch descriptor in decimal form, ie a target file 1776 descriptor number, 'ino' and 'sdev' are inode and device where the 1777 target file resides and the 'mask' is the mask of events, all in hex 1778 form [see inotify(7) for more details]. 1779 1780 If the kernel was built with exportfs support, the path to the target 1781 file is encoded as a file handle. The file handle is provided by three 1782 fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 1783 format. 1784 1785 If the kernel is built without exportfs support the file handle won't be 1786 printed out. 1787 1788 If there is no inotify mark attached yet the 'inotify' line will be omitted. 1789 1790 For fanotify files the format is 1791 1792 pos: 0 1793 flags: 02 1794 mnt_id: 9 1795 fanotify flags:10 event-flags:0 1796 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 1797 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 1798 1799 where fanotify 'flags' and 'event-flags' are values used in fanotify_init 1800 call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 1801 flags associated with mark which are tracked separately from events 1802 mask. 'ino', 'sdev' are target inode and device, 'mask' is the events 1803 mask and 'ignored_mask' is the mask of events which are to be ignored. 1804 All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 1805 does provide information about flags and mask used in fanotify_mark 1806 call [see fsnotify manpage for details]. 1807 1808 While the first three lines are mandatory and always printed, the rest is 1809 optional and may be omitted if no marks created yet. 1810 1811 Timerfd files 1812 ~~~~~~~~~~~~~ 1813 1814 pos: 0 1815 flags: 02 1816 mnt_id: 9 1817 clockid: 0 1818 ticks: 0 1819 settime flags: 01 1820 it_value: (0, 49406829) 1821 it_interval: (1, 0) 1822 1823 where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 1824 that have occurred [see timerfd_create(2) for details]. 'settime flags' are 1825 flags in octal form been used to setup the timer [see timerfd_settime(2) for 1826 details]. 'it_value' is remaining time until the timer exiration. 1827 'it_interval' is the interval for the timer. Note the timer might be set up 1828 with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 1829 still exhibits timer's remaining time. 1830 18313.9 /proc/<pid>/map_files - Information about memory mapped files 1832--------------------------------------------------------------------- 1833This directory contains symbolic links which represent memory mapped files 1834the process is maintaining. Example output: 1835 1836 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 1837 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 1838 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 1839 | ... 1840 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 1841 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 1842 1843The name of a link represents the virtual memory bounds of a mapping, i.e. 1844vm_area_struct::vm_start-vm_area_struct::vm_end. 1845 1846The main purpose of the map_files is to retrieve a set of memory mapped 1847files in a fast way instead of parsing /proc/<pid>/maps or 1848/proc/<pid>/smaps, both of which contain many more records. At the same 1849time one can open(2) mappings from the listings of two processes and 1850comparing their inode numbers to figure out which anonymous memory areas 1851are actually shared. 1852 1853------------------------------------------------------------------------------ 1854Configuring procfs 1855------------------------------------------------------------------------------ 1856 18574.1 Mount options 1858--------------------- 1859 1860The following mount options are supported: 1861 1862 hidepid= Set /proc/<pid>/ access mode. 1863 gid= Set the group authorized to learn processes information. 1864 1865hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories 1866(default). 1867 1868hidepid=1 means users may not access any /proc/<pid>/ directories but their 1869own. Sensitive files like cmdline, sched*, status are now protected against 1870other users. This makes it impossible to learn whether any user runs 1871specific program (given the program doesn't reveal itself by its behaviour). 1872As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users, 1873poorly written programs passing sensitive information via program arguments are 1874now protected against local eavesdroppers. 1875 1876hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other 1877users. It doesn't mean that it hides a fact whether a process with a specific 1878pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"), 1879but it hides process' uid and gid, which may be learned by stat()'ing 1880/proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering 1881information about running processes, whether some daemon runs with elevated 1882privileges, whether other user runs some sensitive program, whether other users 1883run any program at all, etc. 1884 1885gid= defines a group authorized to learn processes information otherwise 1886prohibited by hidepid=. If you use some daemon like identd which needs to learn 1887information about processes information, just add identd to this group. 1888