1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 9 Bodo Bauer <bb@ricochet.net> 102.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 11move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 12fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 13===================== ======================================= ================ 14 15 16 17.. Table of Contents 18 19 0 Preface 20 0.1 Introduction/Credits 21 0.2 Legal Stuff 22 23 1 Collecting System Information 24 1.1 Process-Specific Subdirectories 25 1.2 Kernel data 26 1.3 IDE devices in /proc/ide 27 1.4 Networking info in /proc/net 28 1.5 SCSI info 29 1.6 Parallel port info in /proc/parport 30 1.7 TTY info in /proc/tty 31 1.8 Miscellaneous kernel statistics in /proc/stat 32 1.9 Ext4 file system parameters 33 34 2 Modifying System Parameters 35 36 3 Per-Process Parameters 37 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 38 score 39 3.2 /proc/<pid>/oom_score - Display current oom-killer score 40 3.3 /proc/<pid>/io - Display the IO accounting fields 41 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings 42 3.5 /proc/<pid>/mountinfo - Information about mounts 43 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 44 3.7 /proc/<pid>/task/<tid>/children - Information about task children 45 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 46 3.9 /proc/<pid>/map_files - Information about memory mapped files 47 3.10 /proc/<pid>/timerslack_ns - Task timerslack value 48 3.11 /proc/<pid>/patch_state - Livepatch patch operation state 49 3.12 /proc/<pid>/arch_status - Task architecture specific information 50 51 4 Configuring procfs 52 4.1 Mount options 53 54 5 Filesystem behavior 55 56Preface 57======= 58 590.1 Introduction/Credits 60------------------------ 61 62This documentation is part of a soon (or so we hope) to be released book on 63the SuSE Linux distribution. As there is no complete documentation for the 64/proc file system and we've used many freely available sources to write these 65chapters, it seems only fair to give the work back to the Linux community. 66This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 67afraid it's still far from complete, but we hope it will be useful. As far as 68we know, it is the first 'all-in-one' document about the /proc file system. It 69is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 70SPARC, AXP, etc., features, you probably won't find what you are looking for. 71It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 72additions and patches are welcome and will be added to this document if you 73mail them to Bodo. 74 75We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 76other people for help compiling this documentation. We'd also like to extend a 77special thank you to Andi Kleen for documentation, which we relied on heavily 78to create this document, as well as the additional information he provided. 79Thanks to everybody else who contributed source or docs to the Linux kernel 80and helped create a great piece of software... :) 81 82If you have any comments, corrections or additions, please don't hesitate to 83contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 84document. 85 86The latest version of this document is available online at 87http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html 88 89If the above direction does not works for you, you could try the kernel 90mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 91comandante@zaralinux.com. 92 930.2 Legal Stuff 94--------------- 95 96We don't guarantee the correctness of this document, and if you come to us 97complaining about how you screwed up your system because of incorrect 98documentation, we won't feel responsible... 99 100Chapter 1: Collecting System Information 101======================================== 102 103In This Chapter 104--------------- 105* Investigating the properties of the pseudo file system /proc and its 106 ability to provide information on the running Linux system 107* Examining /proc's structure 108* Uncovering various information about the kernel and the processes running 109 on the system 110 111------------------------------------------------------------------------------ 112 113The proc file system acts as an interface to internal data structures in the 114kernel. It can be used to obtain information about the system and to change 115certain kernel parameters at runtime (sysctl). 116 117First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 118show you how you can use /proc/sys to change settings. 119 1201.1 Process-Specific Subdirectories 121----------------------------------- 122 123The directory /proc contains (among other things) one subdirectory for each 124process running on the system, which is named after the process ID (PID). 125 126The link 'self' points to the process reading the file system. Each process 127subdirectory has the entries listed in Table 1-1. 128 129Note that an open file descriptor to /proc/<pid> or to any of its 130contained files or subdirectories does not prevent <pid> being reused 131for some other process in the event that <pid> exits. Operations on 132open /proc/<pid> file descriptors corresponding to dead processes 133never act on any new process that the kernel may, through chance, have 134also assigned the process ID <pid>. Instead, operations on these FDs 135usually fail with ESRCH. 136 137.. table:: Table 1-1: Process specific entries in /proc 138 139 ============= =============================================================== 140 File Content 141 ============= =============================================================== 142 clear_refs Clears page referenced bits shown in smaps output 143 cmdline Command line arguments 144 cpu Current and last cpu in which it was executed (2.4)(smp) 145 cwd Link to the current working directory 146 environ Values of environment variables 147 exe Link to the executable of this process 148 fd Directory, which contains all file descriptors 149 maps Memory maps to executables and library files (2.4) 150 mem Memory held by this process 151 root Link to the root directory of this process 152 stat Process status 153 statm Process memory status information 154 status Process status in human readable form 155 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 156 symbol the task is blocked in - or "0" if not blocked. 157 pagemap Page table 158 stack Report full stack trace, enable via CONFIG_STACKTRACE 159 smaps An extension based on maps, showing the memory consumption of 160 each mapping and flags associated with it 161 smaps_rollup Accumulated smaps stats for all mappings of the process. This 162 can be derived from smaps, but is faster and more convenient 163 numa_maps An extension based on maps, showing the memory locality and 164 binding policy as well as mem usage (in pages) of each mapping. 165 ============= =============================================================== 166 167For example, to get the status information of a process, all you have to do is 168read the file /proc/PID/status:: 169 170 >cat /proc/self/status 171 Name: cat 172 State: R (running) 173 Tgid: 5452 174 Pid: 5452 175 PPid: 743 176 TracerPid: 0 (2.4) 177 Uid: 501 501 501 501 178 Gid: 100 100 100 100 179 FDSize: 256 180 Groups: 100 14 16 181 VmPeak: 5004 kB 182 VmSize: 5004 kB 183 VmLck: 0 kB 184 VmHWM: 476 kB 185 VmRSS: 476 kB 186 RssAnon: 352 kB 187 RssFile: 120 kB 188 RssShmem: 4 kB 189 VmData: 156 kB 190 VmStk: 88 kB 191 VmExe: 68 kB 192 VmLib: 1412 kB 193 VmPTE: 20 kb 194 VmSwap: 0 kB 195 HugetlbPages: 0 kB 196 CoreDumping: 0 197 THP_enabled: 1 198 Threads: 1 199 SigQ: 0/28578 200 SigPnd: 0000000000000000 201 ShdPnd: 0000000000000000 202 SigBlk: 0000000000000000 203 SigIgn: 0000000000000000 204 SigCgt: 0000000000000000 205 CapInh: 00000000fffffeff 206 CapPrm: 0000000000000000 207 CapEff: 0000000000000000 208 CapBnd: ffffffffffffffff 209 CapAmb: 0000000000000000 210 NoNewPrivs: 0 211 Seccomp: 0 212 Speculation_Store_Bypass: thread vulnerable 213 SpeculationIndirectBranch: conditional enabled 214 voluntary_ctxt_switches: 0 215 nonvoluntary_ctxt_switches: 1 216 217This shows you nearly the same information you would get if you viewed it with 218the ps command. In fact, ps uses the proc file system to obtain its 219information. But you get a more detailed view of the process by reading the 220file /proc/PID/status. It fields are described in table 1-2. 221 222The statm file contains more detailed information about the process 223memory usage. Its seven fields are explained in Table 1-3. The stat file 224contains detailed information about the process itself. Its fields are 225explained in Table 1-4. 226 227(for SMP CONFIG users) 228 229For making accounting scalable, RSS related information are handled in an 230asynchronous manner and the value may not be very precise. To see a precise 231snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 232It's slow but very precise. 233 234.. table:: Table 1-2: Contents of the status files (as of 4.19) 235 236 ========================== =================================================== 237 Field Content 238 ========================== =================================================== 239 Name filename of the executable 240 Umask file mode creation mask 241 State state (R is running, S is sleeping, D is sleeping 242 in an uninterruptible wait, Z is zombie, 243 T is traced or stopped) 244 Tgid thread group ID 245 Ngid NUMA group ID (0 if none) 246 Pid process id 247 PPid process id of the parent process 248 TracerPid PID of process tracing this process (0 if not) 249 Uid Real, effective, saved set, and file system UIDs 250 Gid Real, effective, saved set, and file system GIDs 251 FDSize number of file descriptor slots currently allocated 252 Groups supplementary group list 253 NStgid descendant namespace thread group ID hierarchy 254 NSpid descendant namespace process ID hierarchy 255 NSpgid descendant namespace process group ID hierarchy 256 NSsid descendant namespace session ID hierarchy 257 VmPeak peak virtual memory size 258 VmSize total program size 259 VmLck locked memory size 260 VmPin pinned memory size 261 VmHWM peak resident set size ("high water mark") 262 VmRSS size of memory portions. It contains the three 263 following parts 264 (VmRSS = RssAnon + RssFile + RssShmem) 265 RssAnon size of resident anonymous memory 266 RssFile size of resident file mappings 267 RssShmem size of resident shmem memory (includes SysV shm, 268 mapping of tmpfs and shared anonymous mappings) 269 VmData size of private data segments 270 VmStk size of stack segments 271 VmExe size of text segment 272 VmLib size of shared library code 273 VmPTE size of page table entries 274 VmSwap amount of swap used by anonymous private data 275 (shmem swap usage is not included) 276 HugetlbPages size of hugetlb memory portions 277 CoreDumping process's memory is currently being dumped 278 (killing the process may lead to a corrupted core) 279 THP_enabled process is allowed to use THP (returns 0 when 280 PR_SET_THP_DISABLE is set on the process 281 Threads number of threads 282 SigQ number of signals queued/max. number for queue 283 SigPnd bitmap of pending signals for the thread 284 ShdPnd bitmap of shared pending signals for the process 285 SigBlk bitmap of blocked signals 286 SigIgn bitmap of ignored signals 287 SigCgt bitmap of caught signals 288 CapInh bitmap of inheritable capabilities 289 CapPrm bitmap of permitted capabilities 290 CapEff bitmap of effective capabilities 291 CapBnd bitmap of capabilities bounding set 292 CapAmb bitmap of ambient capabilities 293 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) 294 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 295 Speculation_Store_Bypass speculative store bypass mitigation status 296 SpeculationIndirectBranch indirect branch speculation mode 297 Cpus_allowed mask of CPUs on which this process may run 298 Cpus_allowed_list Same as previous, but in "list format" 299 Mems_allowed mask of memory nodes allowed to this process 300 Mems_allowed_list Same as previous, but in "list format" 301 voluntary_ctxt_switches number of voluntary context switches 302 nonvoluntary_ctxt_switches number of non voluntary context switches 303 ========================== =================================================== 304 305 306.. table:: Table 1-3: Contents of the statm files (as of 2.6.8-rc3) 307 308 ======== =============================== ============================== 309 Field Content 310 ======== =============================== ============================== 311 size total program size (pages) (same as VmSize in status) 312 resident size of memory portions (pages) (same as VmRSS in status) 313 shared number of pages that are shared (i.e. backed by a file, same 314 as RssFile+RssShmem in status) 315 trs number of pages that are 'code' (not including libs; broken, 316 includes data segment) 317 lrs number of pages of library (always 0 on 2.6) 318 drs number of pages of data/stack (including libs; broken, 319 includes library text) 320 dt number of dirty pages (always 0 on 2.6) 321 ======== =============================== ============================== 322 323 324.. table:: Table 1-4: Contents of the stat files (as of 2.6.30-rc7) 325 326 ============= =============================================================== 327 Field Content 328 ============= =============================================================== 329 pid process id 330 tcomm filename of the executable 331 state state (R is running, S is sleeping, D is sleeping in an 332 uninterruptible wait, Z is zombie, T is traced or stopped) 333 ppid process id of the parent process 334 pgrp pgrp of the process 335 sid session id 336 tty_nr tty the process uses 337 tty_pgrp pgrp of the tty 338 flags task flags 339 min_flt number of minor faults 340 cmin_flt number of minor faults with child's 341 maj_flt number of major faults 342 cmaj_flt number of major faults with child's 343 utime user mode jiffies 344 stime kernel mode jiffies 345 cutime user mode jiffies with child's 346 cstime kernel mode jiffies with child's 347 priority priority level 348 nice nice level 349 num_threads number of threads 350 it_real_value (obsolete, always 0) 351 start_time time the process started after system boot 352 vsize virtual memory size 353 rss resident set memory size 354 rsslim current limit in bytes on the rss 355 start_code address above which program text can run 356 end_code address below which program text can run 357 start_stack address of the start of the main process stack 358 esp current value of ESP 359 eip current value of EIP 360 pending bitmap of pending signals 361 blocked bitmap of blocked signals 362 sigign bitmap of ignored signals 363 sigcatch bitmap of caught signals 364 0 (place holder, used to be the wchan address, 365 use /proc/PID/wchan instead) 366 0 (place holder) 367 0 (place holder) 368 exit_signal signal to send to parent thread on exit 369 task_cpu which CPU the task is scheduled on 370 rt_priority realtime priority 371 policy scheduling policy (man sched_setscheduler) 372 blkio_ticks time spent waiting for block IO 373 gtime guest time of the task in jiffies 374 cgtime guest time of the task children in jiffies 375 start_data address above which program data+bss is placed 376 end_data address below which program data+bss is placed 377 start_brk address above which program heap can be expanded with brk() 378 arg_start address above which program command line is placed 379 arg_end address below which program command line is placed 380 env_start address above which program environment is placed 381 env_end address below which program environment is placed 382 exit_code the thread's exit_code in the form reported by the waitpid 383 system call 384 ============= =============================================================== 385 386The /proc/PID/maps file contains the currently mapped memory regions and 387their access permissions. 388 389The format is:: 390 391 address perms offset dev inode pathname 392 393 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 394 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 395 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] 396 a7cb1000-a7cb2000 ---p 00000000 00:00 0 397 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 398 a7eb2000-a7eb3000 ---p 00000000 00:00 0 399 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 400 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 401 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 402 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 403 a800b000-a800e000 rw-p 00000000 00:00 0 404 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 405 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 406 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 407 a8024000-a8027000 rw-p 00000000 00:00 0 408 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 409 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 410 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 411 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 412 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 413 414where "address" is the address space in the process that it occupies, "perms" 415is a set of permissions:: 416 417 r = read 418 w = write 419 x = execute 420 s = shared 421 p = private (copy on write) 422 423"offset" is the offset into the mapping, "dev" is the device (major:minor), and 424"inode" is the inode on that device. 0 indicates that no inode is associated 425with the memory region, as the case would be with BSS (uninitialized data). 426The "pathname" shows the name associated file for this mapping. If the mapping 427is not associated with a file: 428 429 ======= ==================================== 430 [heap] the heap of the program 431 [stack] the stack of the main process 432 [vdso] the "virtual dynamic shared object", 433 the kernel system call handler 434 ======= ==================================== 435 436 or if empty, the mapping is anonymous. 437 438The /proc/PID/smaps is an extension based on maps, showing the memory 439consumption for each of the process's mappings. For each mapping (aka Virtual 440Memory Area, or VMA) there is a series of lines such as the following:: 441 442 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 443 444 Size: 1084 kB 445 KernelPageSize: 4 kB 446 MMUPageSize: 4 kB 447 Rss: 892 kB 448 Pss: 374 kB 449 Shared_Clean: 892 kB 450 Shared_Dirty: 0 kB 451 Private_Clean: 0 kB 452 Private_Dirty: 0 kB 453 Referenced: 892 kB 454 Anonymous: 0 kB 455 LazyFree: 0 kB 456 AnonHugePages: 0 kB 457 ShmemPmdMapped: 0 kB 458 Shared_Hugetlb: 0 kB 459 Private_Hugetlb: 0 kB 460 Swap: 0 kB 461 SwapPss: 0 kB 462 KernelPageSize: 4 kB 463 MMUPageSize: 4 kB 464 Locked: 0 kB 465 THPeligible: 0 466 VmFlags: rd ex mr mw me dw 467 468The first of these lines shows the same information as is displayed for the 469mapping in /proc/PID/maps. Following lines show the size of the mapping 470(size); the size of each page allocated when backing a VMA (KernelPageSize), 471which is usually the same as the size in the page table entries; the page size 472used by the MMU when backing a VMA (in most cases, the same as KernelPageSize); 473the amount of the mapping that is currently resident in RAM (RSS); the 474process' proportional share of this mapping (PSS); and the number of clean and 475dirty shared and private pages in the mapping. 476 477The "proportional set size" (PSS) of a process is the count of pages it has 478in memory, where each page is divided by the number of processes sharing it. 479So if a process has 1000 pages all to itself, and 1000 shared with one other 480process, its PSS will be 1500. 481 482Note that even a page which is part of a MAP_SHARED mapping, but has only 483a single pte mapped, i.e. is currently used by only one process, is accounted 484as private and not as shared. 485 486"Referenced" indicates the amount of memory currently marked as referenced or 487accessed. 488 489"Anonymous" shows the amount of memory that does not belong to any file. Even 490a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 491and a page is modified, the file page is replaced by a private anonymous copy. 492 493"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 494The memory isn't freed immediately with madvise(). It's freed in memory 495pressure if the memory is clean. Please note that the printed value might 496be lower than the real value due to optimizations used in the current 497implementation. If this is not desirable please file a bug report. 498 499"AnonHugePages" shows the ammount of memory backed by transparent hugepage. 500 501"ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by 502huge pages. 503 504"Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by 505hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 506reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 507 508"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 509 510For shmem mappings, "Swap" includes also the size of the mapped (and not 511replaced by copy-on-write) part of the underlying shmem object out on swap. 512"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 513does not take into account swapped out page of underlying shmem objects. 514"Locked" indicates whether the mapping is locked in memory or not. 515"THPeligible" indicates whether the mapping is eligible for allocating THP 516pages - 1 if true, 0 otherwise. It just shows the current status. 517 518"VmFlags" field deserves a separate description. This member represents the 519kernel flags associated with the particular virtual memory area in two letter 520encoded manner. The codes are the following: 521 522 == ======================================= 523 rd readable 524 wr writeable 525 ex executable 526 sh shared 527 mr may read 528 mw may write 529 me may execute 530 ms may share 531 gd stack segment growns down 532 pf pure PFN range 533 dw disabled write to the mapped file 534 lo pages are locked in memory 535 io memory mapped I/O area 536 sr sequential read advise provided 537 rr random read advise provided 538 dc do not copy area on fork 539 de do not expand area on remapping 540 ac area is accountable 541 nr swap space is not reserved for the area 542 ht area uses huge tlb pages 543 sf synchronous page fault 544 ar architecture specific flag 545 wf wipe on fork 546 dd do not include area into core dump 547 sd soft dirty flag 548 mm mixed map area 549 hg huge page advise flag 550 nh no huge page advise flag 551 mg mergable advise flag 552 bt arm64 BTI guarded page 553 mt arm64 MTE allocation tags are enabled 554 um userfaultfd missing tracking 555 uw userfaultfd wr-protect tracking 556 == ======================================= 557 558Note that there is no guarantee that every flag and associated mnemonic will 559be present in all further kernel releases. Things get changed, the flags may 560be vanished or the reverse -- new added. Interpretation of their meaning 561might change in future as well. So each consumer of these flags has to 562follow each specific kernel version for the exact semantic. 563 564This file is only present if the CONFIG_MMU kernel configuration option is 565enabled. 566 567Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 568output can be achieved only in the single read call). 569 570This typically manifests when doing partial reads of these files while the 571memory map is being modified. Despite the races, we do provide the following 572guarantees: 573 5741) The mapped addresses never go backwards, which implies no two 575 regions will ever overlap. 5762) If there is something at a given vaddr during the entirety of the 577 life of the smaps/maps walk, there will be some output for it. 578 579The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 580but their values are the sums of the corresponding values for all mappings of 581the process. Additionally, it contains these fields: 582 583- Pss_Anon 584- Pss_File 585- Pss_Shmem 586 587They represent the proportional shares of anonymous, file, and shmem pages, as 588described for smaps above. These fields are omitted in smaps since each 589mapping identifies the type (anon, file, or shmem) of all pages it contains. 590Thus all information in smaps_rollup can be derived from smaps, but at a 591significantly higher cost. 592 593The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 594bits on both physical and virtual pages associated with a process, and the 595soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 596for details). 597To clear the bits for all the pages associated with the process:: 598 599 > echo 1 > /proc/PID/clear_refs 600 601To clear the bits for the anonymous pages associated with the process:: 602 603 > echo 2 > /proc/PID/clear_refs 604 605To clear the bits for the file mapped pages associated with the process:: 606 607 > echo 3 > /proc/PID/clear_refs 608 609To clear the soft-dirty bit:: 610 611 > echo 4 > /proc/PID/clear_refs 612 613To reset the peak resident set size ("high water mark") to the process's 614current value:: 615 616 > echo 5 > /proc/PID/clear_refs 617 618Any other value written to /proc/PID/clear_refs will have no effect. 619 620The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 621using /proc/kpageflags and number of times a page is mapped using 622/proc/kpagecount. For detailed explanation, see 623Documentation/admin-guide/mm/pagemap.rst. 624 625The /proc/pid/numa_maps is an extension based on maps, showing the memory 626locality and binding policy, as well as the memory usage (in pages) of 627each mapping. The output follows a general format where mapping details get 628summarized separated by blank spaces, one mapping per each file line:: 629 630 address policy mapping details 631 632 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 633 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 634 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 635 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 636 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 637 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 638 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 639 320698b000 default file=/lib64/libc-2.12.so 640 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 641 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 642 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 643 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 644 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 645 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 646 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 647 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 648 649Where: 650 651"address" is the starting address for the mapping; 652 653"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 654 655"mapping details" summarizes mapping data such as mapping type, page usage counters, 656node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 657size, in KB, that is backing the mapping up. 658 6591.2 Kernel data 660--------------- 661 662Similar to the process entries, the kernel data files give information about 663the running kernel. The files used to obtain this information are contained in 664/proc and are listed in Table 1-5. Not all of these will be present in your 665system. It depends on the kernel configuration and the loaded modules, which 666files are there, and which are missing. 667 668.. table:: Table 1-5: Kernel info in /proc 669 670 ============ =============================================================== 671 File Content 672 ============ =============================================================== 673 apm Advanced power management info 674 buddyinfo Kernel memory allocator information (see text) (2.5) 675 bus Directory containing bus specific information 676 cmdline Kernel command line 677 cpuinfo Info about the CPU 678 devices Available devices (block and character) 679 dma Used DMS channels 680 filesystems Supported filesystems 681 driver Various drivers grouped here, currently rtc (2.4) 682 execdomains Execdomains, related to security (2.4) 683 fb Frame Buffer devices (2.4) 684 fs File system parameters, currently nfs/exports (2.4) 685 ide Directory containing info about the IDE subsystem 686 interrupts Interrupt usage 687 iomem Memory map (2.4) 688 ioports I/O port usage 689 irq Masks for irq to cpu affinity (2.4)(smp?) 690 isapnp ISA PnP (Plug&Play) Info (2.4) 691 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 692 kmsg Kernel messages 693 ksyms Kernel symbol table 694 loadavg Load average of last 1, 5 & 15 minutes; 695 number of processes currently runnable (running or on ready queue); 696 total number of processes in system; 697 last pid created. 698 All fields are separated by one space except "number of 699 processes currently runnable" and "total number of processes 700 in system", which are separated by a slash ('/'). Example: 701 0.61 0.61 0.55 3/828 22084 702 locks Kernel locks 703 meminfo Memory info 704 misc Miscellaneous 705 modules List of loaded modules 706 mounts Mounted filesystems 707 net Networking info (see text) 708 pagetypeinfo Additional page allocator information (see text) (2.5) 709 partitions Table of partitions known to the system 710 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 711 decoupled by lspci (2.4) 712 rtc Real time clock 713 scsi SCSI info (see text) 714 slabinfo Slab pool info 715 softirqs softirq usage 716 stat Overall statistics 717 swaps Swap space utilization 718 sys See chapter 2 719 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 720 tty Info of tty drivers 721 uptime Wall clock since boot, combined idle time of all cpus 722 version Kernel version 723 video bttv info of video resources (2.4) 724 vmallocinfo Show vmalloced areas 725 ============ =============================================================== 726 727You can, for example, check which interrupts are currently in use and what 728they are used for by looking in the file /proc/interrupts:: 729 730 > cat /proc/interrupts 731 CPU0 732 0: 8728810 XT-PIC timer 733 1: 895 XT-PIC keyboard 734 2: 0 XT-PIC cascade 735 3: 531695 XT-PIC aha152x 736 4: 2014133 XT-PIC serial 737 5: 44401 XT-PIC pcnet_cs 738 8: 2 XT-PIC rtc 739 11: 8 XT-PIC i82365 740 12: 182918 XT-PIC PS/2 Mouse 741 13: 1 XT-PIC fpu 742 14: 1232265 XT-PIC ide0 743 15: 7 XT-PIC ide1 744 NMI: 0 745 746In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 747output of a SMP machine):: 748 749 > cat /proc/interrupts 750 751 CPU0 CPU1 752 0: 1243498 1214548 IO-APIC-edge timer 753 1: 8949 8958 IO-APIC-edge keyboard 754 2: 0 0 XT-PIC cascade 755 5: 11286 10161 IO-APIC-edge soundblaster 756 8: 1 0 IO-APIC-edge rtc 757 9: 27422 27407 IO-APIC-edge 3c503 758 12: 113645 113873 IO-APIC-edge PS/2 Mouse 759 13: 0 0 XT-PIC fpu 760 14: 22491 24012 IO-APIC-edge ide0 761 15: 2183 2415 IO-APIC-edge ide1 762 17: 30564 30414 IO-APIC-level eth0 763 18: 177 164 IO-APIC-level bttv 764 NMI: 2457961 2457959 765 LOC: 2457882 2457881 766 ERR: 2155 767 768NMI is incremented in this case because every timer interrupt generates a NMI 769(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 770 771LOC is the local interrupt counter of the internal APIC of every CPU. 772 773ERR is incremented in the case of errors in the IO-APIC bus (the bus that 774connects the CPUs in a SMP system. This means that an error has been detected, 775the IO-APIC automatically retry the transmission, so it should not be a big 776problem, but you should read the SMP-FAQ. 777 778In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 779/proc/interrupts to display every IRQ vector in use by the system, not 780just those considered 'most important'. The new vectors are: 781 782THR 783 interrupt raised when a machine check threshold counter 784 (typically counting ECC corrected errors of memory or cache) exceeds 785 a configurable threshold. Only available on some systems. 786 787TRM 788 a thermal event interrupt occurs when a temperature threshold 789 has been exceeded for the CPU. This interrupt may also be generated 790 when the temperature drops back to normal. 791 792SPU 793 a spurious interrupt is some interrupt that was raised then lowered 794 by some IO device before it could be fully processed by the APIC. Hence 795 the APIC sees the interrupt but does not know what device it came from. 796 For this case the APIC will generate the interrupt with a IRQ vector 797 of 0xff. This might also be generated by chipset bugs. 798 799RES, CAL, TLB 800 rescheduling, call and TLB flush interrupts are 801 sent from one CPU to another per the needs of the OS. Typically, 802 their statistics are used by kernel developers and interested users to 803 determine the occurrence of interrupts of the given type. 804 805The above IRQ vectors are displayed only when relevant. For example, 806the threshold vector does not exist on x86_64 platforms. Others are 807suppressed when the system is a uniprocessor. As of this writing, only 808i386 and x86_64 platforms support the new IRQ vector displays. 809 810Of some interest is the introduction of the /proc/irq directory to 2.4. 811It could be used to set IRQ to CPU affinity. This means that you can "hook" an 812IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 813irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 814prof_cpu_mask. 815 816For example:: 817 818 > ls /proc/irq/ 819 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 820 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 821 > ls /proc/irq/0/ 822 smp_affinity 823 824smp_affinity is a bitmask, in which you can specify which CPUs can handle the 825IRQ. You can set it by doing:: 826 827 > echo 1 > /proc/irq/10/smp_affinity 828 829This means that only the first CPU will handle the IRQ, but you can also echo 8305 which means that only the first and third CPU can handle the IRQ. 831 832The contents of each smp_affinity file is the same by default:: 833 834 > cat /proc/irq/0/smp_affinity 835 ffffffff 836 837There is an alternate interface, smp_affinity_list which allows specifying 838a CPU range instead of a bitmask:: 839 840 > cat /proc/irq/0/smp_affinity_list 841 1024-1031 842 843The default_smp_affinity mask applies to all non-active IRQs, which are the 844IRQs which have not yet been allocated/activated, and hence which lack a 845/proc/irq/[0-9]* directory. 846 847The node file on an SMP system shows the node to which the device using the IRQ 848reports itself as being attached. This hardware locality information does not 849include information about any possible driver locality preference. 850 851prof_cpu_mask specifies which CPUs are to be profiled by the system wide 852profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 853 854The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 855between all the CPUs which are allowed to handle it. As usual the kernel has 856more info than you and does a better job than you, so the defaults are the 857best choice for almost everyone. [Note this applies only to those IO-APIC's 858that support "Round Robin" interrupt distribution.] 859 860There are three more important subdirectories in /proc: net, scsi, and sys. 861The general rule is that the contents, or even the existence of these 862directories, depend on your kernel configuration. If SCSI is not enabled, the 863directory scsi may not exist. The same is true with the net, which is there 864only when networking support is present in the running kernel. 865 866The slabinfo file gives information about memory usage at the slab level. 867Linux uses slab pools for memory management above page level in version 2.2. 868Commonly used objects have their own slab pool (such as network buffers, 869directory cache, and so on). 870 871:: 872 873 > cat /proc/buddyinfo 874 875 Node 0, zone DMA 0 4 5 4 4 3 ... 876 Node 0, zone Normal 1 0 0 1 101 8 ... 877 Node 0, zone HighMem 2 0 0 1 1 0 ... 878 879External fragmentation is a problem under some workloads, and buddyinfo is a 880useful tool for helping diagnose these problems. Buddyinfo will give you a 881clue as to how big an area you can safely allocate, or why a previous 882allocation failed. 883 884Each column represents the number of pages of a certain order which are 885available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 886ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 887available in ZONE_NORMAL, etc... 888 889More information relevant to external fragmentation can be found in 890pagetypeinfo:: 891 892 > cat /proc/pagetypeinfo 893 Page block order: 9 894 Pages per block: 512 895 896 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 897 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 898 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 899 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 900 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 901 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 902 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 903 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 904 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 905 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 906 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 907 908 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 909 Node 0, zone DMA 2 0 5 1 0 910 Node 0, zone DMA32 41 6 967 2 0 911 912Fragmentation avoidance in the kernel works by grouping pages of different 913migrate types into the same contiguous regions of memory called page blocks. 914A page block is typically the size of the default hugepage size, e.g. 2MB on 915X86-64. By keeping pages grouped based on their ability to move, the kernel 916can reclaim pages within a page block to satisfy a high-order allocation. 917 918The pagetypinfo begins with information on the size of a page block. It 919then gives the same type of information as buddyinfo except broken down 920by migrate-type and finishes with details on how many page blocks of each 921type exist. 922 923If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 924from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 925make an estimate of the likely number of huge pages that can be allocated 926at a given point in time. All the "Movable" blocks should be allocatable 927unless memory has been mlock()'d. Some of the Reclaimable blocks should 928also be allocatable although a lot of filesystem metadata may have to be 929reclaimed to achieve this. 930 931 932meminfo 933~~~~~~~ 934 935Provides information about distribution and utilization of memory. This 936varies by architecture and compile options. Some of the counters reported 937here overlap. The memory reported by the non overlapping counters may not 938add up to the overall memory usage and the difference for some workloads 939can be substantial. In many cases there are other means to find out 940additional memory using subsystem specific interfaces, for instance 941/proc/net/sockstat for TCP memory allocations. 942 943The following is from a 16GB PIII, which has highmem enabled. 944You may not have all of these fields. 945 946:: 947 948 > cat /proc/meminfo 949 950 MemTotal: 16344972 kB 951 MemFree: 13634064 kB 952 MemAvailable: 14836172 kB 953 Buffers: 3656 kB 954 Cached: 1195708 kB 955 SwapCached: 0 kB 956 Active: 891636 kB 957 Inactive: 1077224 kB 958 HighTotal: 15597528 kB 959 HighFree: 13629632 kB 960 LowTotal: 747444 kB 961 LowFree: 4432 kB 962 SwapTotal: 0 kB 963 SwapFree: 0 kB 964 Dirty: 968 kB 965 Writeback: 0 kB 966 AnonPages: 861800 kB 967 Mapped: 280372 kB 968 Shmem: 644 kB 969 KReclaimable: 168048 kB 970 Slab: 284364 kB 971 SReclaimable: 159856 kB 972 SUnreclaim: 124508 kB 973 PageTables: 24448 kB 974 NFS_Unstable: 0 kB 975 Bounce: 0 kB 976 WritebackTmp: 0 kB 977 CommitLimit: 7669796 kB 978 Committed_AS: 100056 kB 979 VmallocTotal: 112216 kB 980 VmallocUsed: 428 kB 981 VmallocChunk: 111088 kB 982 Percpu: 62080 kB 983 HardwareCorrupted: 0 kB 984 AnonHugePages: 49152 kB 985 ShmemHugePages: 0 kB 986 ShmemPmdMapped: 0 kB 987 988MemTotal 989 Total usable RAM (i.e. physical RAM minus a few reserved 990 bits and the kernel binary code) 991MemFree 992 The sum of LowFree+HighFree 993MemAvailable 994 An estimate of how much memory is available for starting new 995 applications, without swapping. Calculated from MemFree, 996 SReclaimable, the size of the file LRU lists, and the low 997 watermarks in each zone. 998 The estimate takes into account that the system needs some 999 page cache to function well, and that not all reclaimable 1000 slab will be reclaimable, due to items being in use. The 1001 impact of those factors will vary from system to system. 1002Buffers 1003 Relatively temporary storage for raw disk blocks 1004 shouldn't get tremendously large (20MB or so) 1005Cached 1006 in-memory cache for files read from the disk (the 1007 pagecache). Doesn't include SwapCached 1008SwapCached 1009 Memory that once was swapped out, is swapped back in but 1010 still also is in the swapfile (if memory is needed it 1011 doesn't need to be swapped out AGAIN because it is already 1012 in the swapfile. This saves I/O) 1013Active 1014 Memory that has been used more recently and usually not 1015 reclaimed unless absolutely necessary. 1016Inactive 1017 Memory which has been less recently used. It is more 1018 eligible to be reclaimed for other purposes 1019HighTotal, HighFree 1020 Highmem is all memory above ~860MB of physical memory. 1021 Highmem areas are for use by userspace programs, or 1022 for the pagecache. The kernel must use tricks to access 1023 this memory, making it slower to access than lowmem. 1024LowTotal, LowFree 1025 Lowmem is memory which can be used for everything that 1026 highmem can be used for, but it is also available for the 1027 kernel's use for its own data structures. Among many 1028 other things, it is where everything from the Slab is 1029 allocated. Bad things happen when you're out of lowmem. 1030SwapTotal 1031 total amount of swap space available 1032SwapFree 1033 Memory which has been evicted from RAM, and is temporarily 1034 on the disk 1035Dirty 1036 Memory which is waiting to get written back to the disk 1037Writeback 1038 Memory which is actively being written back to the disk 1039AnonPages 1040 Non-file backed pages mapped into userspace page tables 1041HardwareCorrupted 1042 The amount of RAM/memory in KB, the kernel identifies as 1043 corrupted. 1044AnonHugePages 1045 Non-file backed huge pages mapped into userspace page tables 1046Mapped 1047 files which have been mmaped, such as libraries 1048Shmem 1049 Total memory used by shared memory (shmem) and tmpfs 1050ShmemHugePages 1051 Memory used by shared memory (shmem) and tmpfs allocated 1052 with huge pages 1053ShmemPmdMapped 1054 Shared memory mapped into userspace with huge pages 1055KReclaimable 1056 Kernel allocations that the kernel will attempt to reclaim 1057 under memory pressure. Includes SReclaimable (below), and other 1058 direct allocations with a shrinker. 1059Slab 1060 in-kernel data structures cache 1061SReclaimable 1062 Part of Slab, that might be reclaimed, such as caches 1063SUnreclaim 1064 Part of Slab, that cannot be reclaimed on memory pressure 1065PageTables 1066 amount of memory dedicated to the lowest level of page 1067 tables. 1068NFS_Unstable 1069 Always zero. Previous counted pages which had been written to 1070 the server, but has not been committed to stable storage. 1071Bounce 1072 Memory used for block device "bounce buffers" 1073WritebackTmp 1074 Memory used by FUSE for temporary writeback buffers 1075CommitLimit 1076 Based on the overcommit ratio ('vm.overcommit_ratio'), 1077 this is the total amount of memory currently available to 1078 be allocated on the system. This limit is only adhered to 1079 if strict overcommit accounting is enabled (mode 2 in 1080 'vm.overcommit_memory'). 1081 1082 The CommitLimit is calculated with the following formula:: 1083 1084 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1085 overcommit_ratio / 100 + [total swap pages] 1086 1087 For example, on a system with 1G of physical RAM and 7G 1088 of swap with a `vm.overcommit_ratio` of 30 it would 1089 yield a CommitLimit of 7.3G. 1090 1091 For more details, see the memory overcommit documentation 1092 in vm/overcommit-accounting. 1093Committed_AS 1094 The amount of memory presently allocated on the system. 1095 The committed memory is a sum of all of the memory which 1096 has been allocated by processes, even if it has not been 1097 "used" by them as of yet. A process which malloc()'s 1G 1098 of memory, but only touches 300M of it will show up as 1099 using 1G. This 1G is memory which has been "committed" to 1100 by the VM and can be used at any time by the allocating 1101 application. With strict overcommit enabled on the system 1102 (mode 2 in 'vm.overcommit_memory'), allocations which would 1103 exceed the CommitLimit (detailed above) will not be permitted. 1104 This is useful if one needs to guarantee that processes will 1105 not fail due to lack of memory once that memory has been 1106 successfully allocated. 1107VmallocTotal 1108 total size of vmalloc memory area 1109VmallocUsed 1110 amount of vmalloc area which is used 1111VmallocChunk 1112 largest contiguous block of vmalloc area which is free 1113Percpu 1114 Memory allocated to the percpu allocator used to back percpu 1115 allocations. This stat excludes the cost of metadata. 1116 1117vmallocinfo 1118~~~~~~~~~~~ 1119 1120Provides information about vmalloced/vmaped areas. One line per area, 1121containing the virtual address range of the area, size in bytes, 1122caller information of the creator, and optional information depending 1123on the kind of area: 1124 1125 ========== =================================================== 1126 pages=nr number of pages 1127 phys=addr if a physical address was specified 1128 ioremap I/O mapping (ioremap() and friends) 1129 vmalloc vmalloc() area 1130 vmap vmap()ed pages 1131 user VM_USERMAP area 1132 vpages buffer for pages pointers was vmalloced (huge area) 1133 N<node>=nr (Only on NUMA kernels) 1134 Number of pages allocated on memory node <node> 1135 ========== =================================================== 1136 1137:: 1138 1139 > cat /proc/vmallocinfo 1140 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1141 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1142 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1143 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1144 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1145 phys=7fee8000 ioremap 1146 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1147 phys=7fee7000 ioremap 1148 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1149 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1150 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1151 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1152 pages=2 vmalloc N1=2 1153 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1154 /0x130 [x_tables] pages=4 vmalloc N0=4 1155 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1156 pages=14 vmalloc N2=14 1157 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1158 pages=4 vmalloc N1=4 1159 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1160 pages=2 vmalloc N1=2 1161 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1162 pages=10 vmalloc N0=10 1163 1164 1165softirqs 1166~~~~~~~~ 1167 1168Provides counts of softirq handlers serviced since boot time, for each CPU. 1169 1170:: 1171 1172 > cat /proc/softirqs 1173 CPU0 CPU1 CPU2 CPU3 1174 HI: 0 0 0 0 1175 TIMER: 27166 27120 27097 27034 1176 NET_TX: 0 0 0 17 1177 NET_RX: 42 0 0 39 1178 BLOCK: 0 0 107 1121 1179 TASKLET: 0 0 0 290 1180 SCHED: 27035 26983 26971 26746 1181 HRTIMER: 0 0 0 0 1182 RCU: 1678 1769 2178 2250 1183 1184 11851.3 IDE devices in /proc/ide 1186---------------------------- 1187 1188The subdirectory /proc/ide contains information about all IDE devices of which 1189the kernel is aware. There is one subdirectory for each IDE controller, the 1190file drivers and a link for each IDE device, pointing to the device directory 1191in the controller specific subtree. 1192 1193The file 'drivers' contains general information about the drivers used for the 1194IDE devices:: 1195 1196 > cat /proc/ide/drivers 1197 ide-cdrom version 4.53 1198 ide-disk version 1.08 1199 1200More detailed information can be found in the controller specific 1201subdirectories. These are named ide0, ide1 and so on. Each of these 1202directories contains the files shown in table 1-6. 1203 1204 1205.. table:: Table 1-6: IDE controller info in /proc/ide/ide? 1206 1207 ======= ======================================= 1208 File Content 1209 ======= ======================================= 1210 channel IDE channel (0 or 1) 1211 config Configuration (only for PCI/IDE bridge) 1212 mate Mate name 1213 model Type/Chipset of IDE controller 1214 ======= ======================================= 1215 1216Each device connected to a controller has a separate subdirectory in the 1217controllers directory. The files listed in table 1-7 are contained in these 1218directories. 1219 1220 1221.. table:: Table 1-7: IDE device information 1222 1223 ================ ========================================== 1224 File Content 1225 ================ ========================================== 1226 cache The cache 1227 capacity Capacity of the medium (in 512Byte blocks) 1228 driver driver and version 1229 geometry physical and logical geometry 1230 identify device identify block 1231 media media type 1232 model device identifier 1233 settings device setup 1234 smart_thresholds IDE disk management thresholds 1235 smart_values IDE disk management values 1236 ================ ========================================== 1237 1238The most interesting file is ``settings``. This file contains a nice 1239overview of the drive parameters:: 1240 1241 # cat /proc/ide/ide0/hda/settings 1242 name value min max mode 1243 ---- ----- --- --- ---- 1244 bios_cyl 526 0 65535 rw 1245 bios_head 255 0 255 rw 1246 bios_sect 63 0 63 rw 1247 breada_readahead 4 0 127 rw 1248 bswap 0 0 1 r 1249 file_readahead 72 0 2097151 rw 1250 io_32bit 0 0 3 rw 1251 keepsettings 0 0 1 rw 1252 max_kb_per_request 122 1 127 rw 1253 multcount 0 0 8 rw 1254 nice1 1 0 1 rw 1255 nowerr 0 0 1 rw 1256 pio_mode write-only 0 255 w 1257 slow 0 0 1 rw 1258 unmaskirq 0 0 1 rw 1259 using_dma 0 0 1 rw 1260 1261 12621.4 Networking info in /proc/net 1263-------------------------------- 1264 1265The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1266additional values you get for IP version 6 if you configure the kernel to 1267support this. Table 1-9 lists the files and their meaning. 1268 1269 1270.. table:: Table 1-8: IPv6 info in /proc/net 1271 1272 ========== ===================================================== 1273 File Content 1274 ========== ===================================================== 1275 udp6 UDP sockets (IPv6) 1276 tcp6 TCP sockets (IPv6) 1277 raw6 Raw device statistics (IPv6) 1278 igmp6 IP multicast addresses, which this host joined (IPv6) 1279 if_inet6 List of IPv6 interface addresses 1280 ipv6_route Kernel routing table for IPv6 1281 rt6_stats Global IPv6 routing tables statistics 1282 sockstat6 Socket statistics (IPv6) 1283 snmp6 Snmp data (IPv6) 1284 ========== ===================================================== 1285 1286.. table:: Table 1-9: Network info in /proc/net 1287 1288 ============= ================================================================ 1289 File Content 1290 ============= ================================================================ 1291 arp Kernel ARP table 1292 dev network devices with statistics 1293 dev_mcast the Layer2 multicast groups a device is listening too 1294 (interface index, label, number of references, number of bound 1295 addresses). 1296 dev_stat network device status 1297 ip_fwchains Firewall chain linkage 1298 ip_fwnames Firewall chain names 1299 ip_masq Directory containing the masquerading tables 1300 ip_masquerade Major masquerading table 1301 netstat Network statistics 1302 raw raw device statistics 1303 route Kernel routing table 1304 rpc Directory containing rpc info 1305 rt_cache Routing cache 1306 snmp SNMP data 1307 sockstat Socket statistics 1308 tcp TCP sockets 1309 udp UDP sockets 1310 unix UNIX domain sockets 1311 wireless Wireless interface data (Wavelan etc) 1312 igmp IP multicast addresses, which this host joined 1313 psched Global packet scheduler parameters. 1314 netlink List of PF_NETLINK sockets 1315 ip_mr_vifs List of multicast virtual interfaces 1316 ip_mr_cache List of multicast routing cache 1317 ============= ================================================================ 1318 1319You can use this information to see which network devices are available in 1320your system and how much traffic was routed over those devices:: 1321 1322 > cat /proc/net/dev 1323 Inter-|Receive |[... 1324 face |bytes packets errs drop fifo frame compressed multicast|[... 1325 lo: 908188 5596 0 0 0 0 0 0 [... 1326 ppp0:15475140 20721 410 0 0 410 0 0 [... 1327 eth0: 614530 7085 0 0 0 0 0 1 [... 1328 1329 ...] Transmit 1330 ...] bytes packets errs drop fifo colls carrier compressed 1331 ...] 908188 5596 0 0 0 0 0 0 1332 ...] 1375103 17405 0 0 0 0 0 0 1333 ...] 1703981 5535 0 0 0 3 0 0 1334 1335In addition, each Channel Bond interface has its own directory. For 1336example, the bond0 device will have a directory called /proc/net/bond0/. 1337It will contain information that is specific to that bond, such as the 1338current slaves of the bond, the link status of the slaves, and how 1339many times the slaves link has failed. 1340 13411.5 SCSI info 1342------------- 1343 1344If you have a SCSI host adapter in your system, you'll find a subdirectory 1345named after the driver for this adapter in /proc/scsi. You'll also see a list 1346of all recognized SCSI devices in /proc/scsi:: 1347 1348 >cat /proc/scsi/scsi 1349 Attached devices: 1350 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1351 Vendor: IBM Model: DGHS09U Rev: 03E0 1352 Type: Direct-Access ANSI SCSI revision: 03 1353 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1354 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1355 Type: CD-ROM ANSI SCSI revision: 02 1356 1357 1358The directory named after the driver has one file for each adapter found in 1359the system. These files contain information about the controller, including 1360the used IRQ and the IO address range. The amount of information shown is 1361dependent on the adapter you use. The example shows the output for an Adaptec 1362AHA-2940 SCSI adapter:: 1363 1364 > cat /proc/scsi/aic7xxx/0 1365 1366 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1367 Compile Options: 1368 TCQ Enabled By Default : Disabled 1369 AIC7XXX_PROC_STATS : Disabled 1370 AIC7XXX_RESET_DELAY : 5 1371 Adapter Configuration: 1372 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1373 Ultra Wide Controller 1374 PCI MMAPed I/O Base: 0xeb001000 1375 Adapter SEEPROM Config: SEEPROM found and used. 1376 Adaptec SCSI BIOS: Enabled 1377 IRQ: 10 1378 SCBs: Active 0, Max Active 2, 1379 Allocated 15, HW 16, Page 255 1380 Interrupts: 160328 1381 BIOS Control Word: 0x18b6 1382 Adapter Control Word: 0x005b 1383 Extended Translation: Enabled 1384 Disconnect Enable Flags: 0xffff 1385 Ultra Enable Flags: 0x0001 1386 Tag Queue Enable Flags: 0x0000 1387 Ordered Queue Tag Flags: 0x0000 1388 Default Tag Queue Depth: 8 1389 Tagged Queue By Device array for aic7xxx host instance 0: 1390 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1391 Actual queue depth per device for aic7xxx host instance 0: 1392 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1393 Statistics: 1394 (scsi0:0:0:0) 1395 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1396 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1397 Total transfers 160151 (74577 reads and 85574 writes) 1398 (scsi0:0:6:0) 1399 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1400 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1401 Total transfers 0 (0 reads and 0 writes) 1402 1403 14041.6 Parallel port info in /proc/parport 1405--------------------------------------- 1406 1407The directory /proc/parport contains information about the parallel ports of 1408your system. It has one subdirectory for each port, named after the port 1409number (0,1,2,...). 1410 1411These directories contain the four files shown in Table 1-10. 1412 1413 1414.. table:: Table 1-10: Files in /proc/parport 1415 1416 ========= ==================================================================== 1417 File Content 1418 ========= ==================================================================== 1419 autoprobe Any IEEE-1284 device ID information that has been acquired. 1420 devices list of the device drivers using that port. A + will appear by the 1421 name of the device currently using the port (it might not appear 1422 against any). 1423 hardware Parallel port's base address, IRQ line and DMA channel. 1424 irq IRQ that parport is using for that port. This is in a separate 1425 file to allow you to alter it by writing a new value in (IRQ 1426 number or none). 1427 ========= ==================================================================== 1428 14291.7 TTY info in /proc/tty 1430------------------------- 1431 1432Information about the available and actually used tty's can be found in the 1433directory /proc/tty. You'll find entries for drivers and line disciplines in 1434this directory, as shown in Table 1-11. 1435 1436 1437.. table:: Table 1-11: Files in /proc/tty 1438 1439 ============= ============================================== 1440 File Content 1441 ============= ============================================== 1442 drivers list of drivers and their usage 1443 ldiscs registered line disciplines 1444 driver/serial usage statistic and status of single tty lines 1445 ============= ============================================== 1446 1447To see which tty's are currently in use, you can simply look into the file 1448/proc/tty/drivers:: 1449 1450 > cat /proc/tty/drivers 1451 pty_slave /dev/pts 136 0-255 pty:slave 1452 pty_master /dev/ptm 128 0-255 pty:master 1453 pty_slave /dev/ttyp 3 0-255 pty:slave 1454 pty_master /dev/pty 2 0-255 pty:master 1455 serial /dev/cua 5 64-67 serial:callout 1456 serial /dev/ttyS 4 64-67 serial 1457 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1458 /dev/ptmx /dev/ptmx 5 2 system 1459 /dev/console /dev/console 5 1 system:console 1460 /dev/tty /dev/tty 5 0 system:/dev/tty 1461 unknown /dev/tty 4 1-63 console 1462 1463 14641.8 Miscellaneous kernel statistics in /proc/stat 1465------------------------------------------------- 1466 1467Various pieces of information about kernel activity are available in the 1468/proc/stat file. All of the numbers reported in this file are aggregates 1469since the system first booted. For a quick look, simply cat the file:: 1470 1471 > cat /proc/stat 1472 cpu 2255 34 2290 22625563 6290 127 456 0 0 0 1473 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0 1474 cpu1 1123 0 849 11313845 2614 0 18 0 0 0 1475 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...] 1476 ctxt 1990473 1477 btime 1062191376 1478 processes 2915 1479 procs_running 1 1480 procs_blocked 0 1481 softirq 183433 0 21755 12 39 1137 231 21459 2263 1482 1483The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1484lines. These numbers identify the amount of time the CPU has spent performing 1485different kinds of work. Time units are in USER_HZ (typically hundredths of a 1486second). The meanings of the columns are as follows, from left to right: 1487 1488- user: normal processes executing in user mode 1489- nice: niced processes executing in user mode 1490- system: processes executing in kernel mode 1491- idle: twiddling thumbs 1492- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1493 are several problems: 1494 1495 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1496 waiting for I/O to complete. When CPU goes into idle state for 1497 outstanding task I/O, another task will be scheduled on this CPU. 1498 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1499 on any CPU, so the iowait of each CPU is difficult to calculate. 1500 3. The value of iowait field in /proc/stat will decrease in certain 1501 conditions. 1502 1503 So, the iowait is not reliable by reading from /proc/stat. 1504- irq: servicing interrupts 1505- softirq: servicing softirqs 1506- steal: involuntary wait 1507- guest: running a normal guest 1508- guest_nice: running a niced guest 1509 1510The "intr" line gives counts of interrupts serviced since boot time, for each 1511of the possible system interrupts. The first column is the total of all 1512interrupts serviced including unnumbered architecture specific interrupts; 1513each subsequent column is the total for that particular numbered interrupt. 1514Unnumbered interrupts are not shown, only summed into the total. 1515 1516The "ctxt" line gives the total number of context switches across all CPUs. 1517 1518The "btime" line gives the time at which the system booted, in seconds since 1519the Unix epoch. 1520 1521The "processes" line gives the number of processes and threads created, which 1522includes (but is not limited to) those created by calls to the fork() and 1523clone() system calls. 1524 1525The "procs_running" line gives the total number of threads that are 1526running or ready to run (i.e., the total number of runnable threads). 1527 1528The "procs_blocked" line gives the number of processes currently blocked, 1529waiting for I/O to complete. 1530 1531The "softirq" line gives counts of softirqs serviced since boot time, for each 1532of the possible system softirqs. The first column is the total of all 1533softirqs serviced; each subsequent column is the total for that particular 1534softirq. 1535 1536 15371.9 Ext4 file system parameters 1538------------------------------- 1539 1540Information about mounted ext4 file systems can be found in 1541/proc/fs/ext4. Each mounted filesystem will have a directory in 1542/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1543/proc/fs/ext4/dm-0). The files in each per-device directory are shown 1544in Table 1-12, below. 1545 1546.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1547 1548 ============== ========================================================== 1549 File Content 1550 mb_groups details of multiblock allocator buddy cache of free blocks 1551 ============== ========================================================== 1552 15531.10 /proc/consoles 1554------------------- 1555Shows registered system console lines. 1556 1557To see which character device lines are currently used for the system console 1558/dev/console, you may simply look into the file /proc/consoles:: 1559 1560 > cat /proc/consoles 1561 tty0 -WU (ECp) 4:7 1562 ttyS0 -W- (Ep) 4:64 1563 1564The columns are: 1565 1566+--------------------+-------------------------------------------------------+ 1567| device | name of the device | 1568+====================+=======================================================+ 1569| operations | * R = can do read operations | 1570| | * W = can do write operations | 1571| | * U = can do unblank | 1572+--------------------+-------------------------------------------------------+ 1573| flags | * E = it is enabled | 1574| | * C = it is preferred console | 1575| | * B = it is primary boot console | 1576| | * p = it is used for printk buffer | 1577| | * b = it is not a TTY but a Braille device | 1578| | * a = it is safe to use when cpu is offline | 1579+--------------------+-------------------------------------------------------+ 1580| major:minor | major and minor number of the device separated by a | 1581| | colon | 1582+--------------------+-------------------------------------------------------+ 1583 1584Summary 1585------- 1586 1587The /proc file system serves information about the running system. It not only 1588allows access to process data but also allows you to request the kernel status 1589by reading files in the hierarchy. 1590 1591The directory structure of /proc reflects the types of information and makes 1592it easy, if not obvious, where to look for specific data. 1593 1594Chapter 2: Modifying System Parameters 1595====================================== 1596 1597In This Chapter 1598--------------- 1599 1600* Modifying kernel parameters by writing into files found in /proc/sys 1601* Exploring the files which modify certain parameters 1602* Review of the /proc/sys file tree 1603 1604------------------------------------------------------------------------------ 1605 1606A very interesting part of /proc is the directory /proc/sys. This is not only 1607a source of information, it also allows you to change parameters within the 1608kernel. Be very careful when attempting this. You can optimize your system, 1609but you can also cause it to crash. Never alter kernel parameters on a 1610production system. Set up a development machine and test to make sure that 1611everything works the way you want it to. You may have no alternative but to 1612reboot the machine once an error has been made. 1613 1614To change a value, simply echo the new value into the file. 1615You need to be root to do this. You can create your own boot script 1616to perform this every time your system boots. 1617 1618The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1619general things in the operation of the Linux kernel. Since some of the files 1620can inadvertently disrupt your system, it is advisable to read both 1621documentation and source before actually making adjustments. In any case, be 1622very careful when writing to any of these files. The entries in /proc may 1623change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1624review the kernel documentation in the directory /usr/src/linux/Documentation. 1625This chapter is heavily based on the documentation included in the pre 2.2 1626kernels, and became part of it in version 2.2.1 of the Linux kernel. 1627 1628Please see: Documentation/admin-guide/sysctl/ directory for descriptions of these 1629entries. 1630 1631Summary 1632------- 1633 1634Certain aspects of kernel behavior can be modified at runtime, without the 1635need to recompile the kernel, or even to reboot the system. The files in the 1636/proc/sys tree can not only be read, but also modified. You can use the echo 1637command to write value into these files, thereby changing the default settings 1638of the kernel. 1639 1640 1641Chapter 3: Per-process Parameters 1642================================= 1643 16443.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1645-------------------------------------------------------------------------------- 1646 1647These files can be used to adjust the badness heuristic used to select which 1648process gets killed in out of memory (oom) conditions. 1649 1650The badness heuristic assigns a value to each candidate task ranging from 0 1651(never kill) to 1000 (always kill) to determine which process is targeted. The 1652units are roughly a proportion along that range of allowed memory the process 1653may allocate from based on an estimation of its current memory and swap use. 1654For example, if a task is using all allowed memory, its badness score will be 16551000. If it is using half of its allowed memory, its score will be 500. 1656 1657The amount of "allowed" memory depends on the context in which the oom killer 1658was called. If it is due to the memory assigned to the allocating task's cpuset 1659being exhausted, the allowed memory represents the set of mems assigned to that 1660cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1661memory represents the set of mempolicy nodes. If it is due to a memory 1662limit (or swap limit) being reached, the allowed memory is that configured 1663limit. Finally, if it is due to the entire system being out of memory, the 1664allowed memory represents all allocatable resources. 1665 1666The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1667is used to determine which task to kill. Acceptable values range from -1000 1668(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1669polarize the preference for oom killing either by always preferring a certain 1670task or completely disabling it. The lowest possible value, -1000, is 1671equivalent to disabling oom killing entirely for that task since it will always 1672report a badness score of 0. 1673 1674Consequently, it is very simple for userspace to define the amount of memory to 1675consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1676example, is roughly equivalent to allowing the remainder of tasks sharing the 1677same system, cpuset, mempolicy, or memory controller resources to use at least 167850% more memory. A value of -500, on the other hand, would be roughly 1679equivalent to discounting 50% of the task's allowed memory from being considered 1680as scoring against the task. 1681 1682For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1683be used to tune the badness score. Its acceptable values range from -16 1684(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1685(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1686scaled linearly with /proc/<pid>/oom_score_adj. 1687 1688The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1689value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1690requires CAP_SYS_RESOURCE. 1691 1692 16933.2 /proc/<pid>/oom_score - Display current oom-killer score 1694------------------------------------------------------------- 1695 1696This file can be used to check the current score used by the oom-killer for 1697any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1698process should be killed in an out-of-memory situation. 1699 1700Please note that the exported value includes oom_score_adj so it is 1701effectively in range [0,2000]. 1702 1703 17043.3 /proc/<pid>/io - Display the IO accounting fields 1705------------------------------------------------------- 1706 1707This file contains IO statistics for each running process. 1708 1709Example 1710~~~~~~~ 1711 1712:: 1713 1714 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1715 [1] 3828 1716 1717 test:/tmp # cat /proc/3828/io 1718 rchar: 323934931 1719 wchar: 323929600 1720 syscr: 632687 1721 syscw: 632675 1722 read_bytes: 0 1723 write_bytes: 323932160 1724 cancelled_write_bytes: 0 1725 1726 1727Description 1728~~~~~~~~~~~ 1729 1730rchar 1731^^^^^ 1732 1733I/O counter: chars read 1734The number of bytes which this task has caused to be read from storage. This 1735is simply the sum of bytes which this process passed to read() and pread(). 1736It includes things like tty IO and it is unaffected by whether or not actual 1737physical disk IO was required (the read might have been satisfied from 1738pagecache). 1739 1740 1741wchar 1742^^^^^ 1743 1744I/O counter: chars written 1745The number of bytes which this task has caused, or shall cause to be written 1746to disk. Similar caveats apply here as with rchar. 1747 1748 1749syscr 1750^^^^^ 1751 1752I/O counter: read syscalls 1753Attempt to count the number of read I/O operations, i.e. syscalls like read() 1754and pread(). 1755 1756 1757syscw 1758^^^^^ 1759 1760I/O counter: write syscalls 1761Attempt to count the number of write I/O operations, i.e. syscalls like 1762write() and pwrite(). 1763 1764 1765read_bytes 1766^^^^^^^^^^ 1767 1768I/O counter: bytes read 1769Attempt to count the number of bytes which this process really did cause to 1770be fetched from the storage layer. Done at the submit_bio() level, so it is 1771accurate for block-backed filesystems. <please add status regarding NFS and 1772CIFS at a later time> 1773 1774 1775write_bytes 1776^^^^^^^^^^^ 1777 1778I/O counter: bytes written 1779Attempt to count the number of bytes which this process caused to be sent to 1780the storage layer. This is done at page-dirtying time. 1781 1782 1783cancelled_write_bytes 1784^^^^^^^^^^^^^^^^^^^^^ 1785 1786The big inaccuracy here is truncate. If a process writes 1MB to a file and 1787then deletes the file, it will in fact perform no writeout. But it will have 1788been accounted as having caused 1MB of write. 1789In other words: The number of bytes which this process caused to not happen, 1790by truncating pagecache. A task can cause "negative" IO too. If this task 1791truncates some dirty pagecache, some IO which another task has been accounted 1792for (in its write_bytes) will not be happening. We _could_ just subtract that 1793from the truncating task's write_bytes, but there is information loss in doing 1794that. 1795 1796 1797.. Note:: 1798 1799 At its current implementation state, this is a bit racy on 32-bit machines: 1800 if process A reads process B's /proc/pid/io while process B is updating one 1801 of those 64-bit counters, process A could see an intermediate result. 1802 1803 1804More information about this can be found within the taskstats documentation in 1805Documentation/accounting. 1806 18073.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1808--------------------------------------------------------------- 1809When a process is dumped, all anonymous memory is written to a core file as 1810long as the size of the core file isn't limited. But sometimes we don't want 1811to dump some memory segments, for example, huge shared memory or DAX. 1812Conversely, sometimes we want to save file-backed memory segments into a core 1813file, not only the individual files. 1814 1815/proc/<pid>/coredump_filter allows you to customize which memory segments 1816will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1817of memory types. If a bit of the bitmask is set, memory segments of the 1818corresponding memory type are dumped, otherwise they are not dumped. 1819 1820The following 9 memory types are supported: 1821 1822 - (bit 0) anonymous private memory 1823 - (bit 1) anonymous shared memory 1824 - (bit 2) file-backed private memory 1825 - (bit 3) file-backed shared memory 1826 - (bit 4) ELF header pages in file-backed private memory areas (it is 1827 effective only if the bit 2 is cleared) 1828 - (bit 5) hugetlb private memory 1829 - (bit 6) hugetlb shared memory 1830 - (bit 7) DAX private memory 1831 - (bit 8) DAX shared memory 1832 1833 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1834 are always dumped regardless of the bitmask status. 1835 1836 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1837 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1838 1839The default value of coredump_filter is 0x33; this means all anonymous memory 1840segments, ELF header pages and hugetlb private memory are dumped. 1841 1842If you don't want to dump all shared memory segments attached to pid 1234, 1843write 0x31 to the process's proc file:: 1844 1845 $ echo 0x31 > /proc/1234/coredump_filter 1846 1847When a new process is created, the process inherits the bitmask status from its 1848parent. It is useful to set up coredump_filter before the program runs. 1849For example:: 1850 1851 $ echo 0x7 > /proc/self/coredump_filter 1852 $ ./some_program 1853 18543.5 /proc/<pid>/mountinfo - Information about mounts 1855-------------------------------------------------------- 1856 1857This file contains lines of the form:: 1858 1859 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1860 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11) 1861 1862 (1) mount ID: unique identifier of the mount (may be reused after umount) 1863 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1864 (3) major:minor: value of st_dev for files on filesystem 1865 (4) root: root of the mount within the filesystem 1866 (5) mount point: mount point relative to the process's root 1867 (6) mount options: per mount options 1868 (7) optional fields: zero or more fields of the form "tag[:value]" 1869 (8) separator: marks the end of the optional fields 1870 (9) filesystem type: name of filesystem of the form "type[.subtype]" 1871 (10) mount source: filesystem specific information or "none" 1872 (11) super options: per super block options 1873 1874Parsers should ignore all unrecognised optional fields. Currently the 1875possible optional fields are: 1876 1877================ ============================================================== 1878shared:X mount is shared in peer group X 1879master:X mount is slave to peer group X 1880propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1881unbindable mount is unbindable 1882================ ============================================================== 1883 1884.. [#] X is the closest dominant peer group under the process's root. If 1885 X is the immediate master of the mount, or if there's no dominant peer 1886 group under the same root, then only the "master:X" field is present 1887 and not the "propagate_from:X" field. 1888 1889For more information on mount propagation see: 1890 1891 Documentation/filesystems/sharedsubtree.rst 1892 1893 18943.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1895-------------------------------------------------------- 1896These files provide a method to access a task's comm value. It also allows for 1897a task to set its own or one of its thread siblings comm value. The comm value 1898is limited in size compared to the cmdline value, so writing anything longer 1899then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated 1900comm value. 1901 1902 19033.7 /proc/<pid>/task/<tid>/children - Information about task children 1904------------------------------------------------------------------------- 1905This file provides a fast way to retrieve first level children pids 1906of a task pointed by <pid>/<tid> pair. The format is a space separated 1907stream of pids. 1908 1909Note the "first level" here -- if a child has its own children they will 1910not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1911to obtain the descendants. 1912 1913Since this interface is intended to be fast and cheap it doesn't 1914guarantee to provide precise results and some children might be 1915skipped, especially if they've exited right after we printed their 1916pids, so one needs to either stop or freeze processes being inspected 1917if precise results are needed. 1918 1919 19203.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1921--------------------------------------------------------------- 1922This file provides information associated with an opened file. The regular 1923files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. 1924The 'pos' represents the current offset of the opened file in decimal 1925form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the 1926file has been created with [see open(2) for details] and 'mnt_id' represents 1927mount ID of the file system containing the opened file [see 3.5 1928/proc/<pid>/mountinfo for details]. 'ino' represents the inode number of 1929the file. 1930 1931A typical output is:: 1932 1933 pos: 0 1934 flags: 0100002 1935 mnt_id: 19 1936 ino: 63107 1937 1938All locks associated with a file descriptor are shown in its fdinfo too:: 1939 1940 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1941 1942The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1943pair provide additional information particular to the objects they represent. 1944 1945Eventfd files 1946~~~~~~~~~~~~~ 1947 1948:: 1949 1950 pos: 0 1951 flags: 04002 1952 mnt_id: 9 1953 ino: 63107 1954 eventfd-count: 5a 1955 1956where 'eventfd-count' is hex value of a counter. 1957 1958Signalfd files 1959~~~~~~~~~~~~~~ 1960 1961:: 1962 1963 pos: 0 1964 flags: 04002 1965 mnt_id: 9 1966 ino: 63107 1967 sigmask: 0000000000000200 1968 1969where 'sigmask' is hex value of the signal mask associated 1970with a file. 1971 1972Epoll files 1973~~~~~~~~~~~ 1974 1975:: 1976 1977 pos: 0 1978 flags: 02 1979 mnt_id: 9 1980 ino: 63107 1981 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 1982 1983where 'tfd' is a target file descriptor number in decimal form, 1984'events' is events mask being watched and the 'data' is data 1985associated with a target [see epoll(7) for more details]. 1986 1987The 'pos' is current offset of the target file in decimal form 1988[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 1989where target file resides, all in hex format. 1990 1991Fsnotify files 1992~~~~~~~~~~~~~~ 1993For inotify files the format is the following:: 1994 1995 pos: 0 1996 flags: 02000000 1997 mnt_id: 9 1998 ino: 63107 1999 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 2000 2001where 'wd' is a watch descriptor in decimal form, i.e. a target file 2002descriptor number, 'ino' and 'sdev' are inode and device where the 2003target file resides and the 'mask' is the mask of events, all in hex 2004form [see inotify(7) for more details]. 2005 2006If the kernel was built with exportfs support, the path to the target 2007file is encoded as a file handle. The file handle is provided by three 2008fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 2009format. 2010 2011If the kernel is built without exportfs support the file handle won't be 2012printed out. 2013 2014If there is no inotify mark attached yet the 'inotify' line will be omitted. 2015 2016For fanotify files the format is:: 2017 2018 pos: 0 2019 flags: 02 2020 mnt_id: 9 2021 ino: 63107 2022 fanotify flags:10 event-flags:0 2023 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 2024 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 2025 2026where fanotify 'flags' and 'event-flags' are values used in fanotify_init 2027call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 2028flags associated with mark which are tracked separately from events 2029mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 2030mask and 'ignored_mask' is the mask of events which are to be ignored. 2031All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2032provide information about flags and mask used in fanotify_mark 2033call [see fsnotify manpage for details]. 2034 2035While the first three lines are mandatory and always printed, the rest is 2036optional and may be omitted if no marks created yet. 2037 2038Timerfd files 2039~~~~~~~~~~~~~ 2040 2041:: 2042 2043 pos: 0 2044 flags: 02 2045 mnt_id: 9 2046 ino: 63107 2047 clockid: 0 2048 ticks: 0 2049 settime flags: 01 2050 it_value: (0, 49406829) 2051 it_interval: (1, 0) 2052 2053where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2054that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2055flags in octal form been used to setup the timer [see timerfd_settime(2) for 2056details]. 'it_value' is remaining time until the timer expiration. 2057'it_interval' is the interval for the timer. Note the timer might be set up 2058with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2059still exhibits timer's remaining time. 2060 2061DMA Buffer files 2062~~~~~~~~~~~~~~~~ 2063 2064:: 2065 2066 pos: 0 2067 flags: 04002 2068 mnt_id: 9 2069 ino: 63107 2070 size: 32768 2071 count: 2 2072 exp_name: system-heap 2073 2074where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of 2075the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. 2076 20773.9 /proc/<pid>/map_files - Information about memory mapped files 2078--------------------------------------------------------------------- 2079This directory contains symbolic links which represent memory mapped files 2080the process is maintaining. Example output:: 2081 2082 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2083 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2084 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2085 | ... 2086 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2087 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2088 2089The name of a link represents the virtual memory bounds of a mapping, i.e. 2090vm_area_struct::vm_start-vm_area_struct::vm_end. 2091 2092The main purpose of the map_files is to retrieve a set of memory mapped 2093files in a fast way instead of parsing /proc/<pid>/maps or 2094/proc/<pid>/smaps, both of which contain many more records. At the same 2095time one can open(2) mappings from the listings of two processes and 2096comparing their inode numbers to figure out which anonymous memory areas 2097are actually shared. 2098 20993.10 /proc/<pid>/timerslack_ns - Task timerslack value 2100--------------------------------------------------------- 2101This file provides the value of the task's timerslack value in nanoseconds. 2102This value specifies an amount of time that normal timers may be deferred 2103in order to coalesce timers and avoid unnecessary wakeups. 2104 2105This allows a task's interactivity vs power consumption tradeoff to be 2106adjusted. 2107 2108Writing 0 to the file will set the task's timerslack to the default value. 2109 2110Valid values are from 0 - ULLONG_MAX 2111 2112An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2113permissions on the task specified to change its timerslack_ns value. 2114 21153.11 /proc/<pid>/patch_state - Livepatch patch operation state 2116----------------------------------------------------------------- 2117When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2118patch state for the task. 2119 2120A value of '-1' indicates that no patch is in transition. 2121 2122A value of '0' indicates that a patch is in transition and the task is 2123unpatched. If the patch is being enabled, then the task hasn't been 2124patched yet. If the patch is being disabled, then the task has already 2125been unpatched. 2126 2127A value of '1' indicates that a patch is in transition and the task is 2128patched. If the patch is being enabled, then the task has already been 2129patched. If the patch is being disabled, then the task hasn't been 2130unpatched yet. 2131 21323.12 /proc/<pid>/arch_status - task architecture specific status 2133------------------------------------------------------------------- 2134When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2135architecture specific status of the task. 2136 2137Example 2138~~~~~~~ 2139 2140:: 2141 2142 $ cat /proc/6753/arch_status 2143 AVX512_elapsed_ms: 8 2144 2145Description 2146~~~~~~~~~~~ 2147 2148x86 specific entries 2149~~~~~~~~~~~~~~~~~~~~~ 2150 2151AVX512_elapsed_ms 2152^^^^^^^^^^^^^^^^^^ 2153 2154 If AVX512 is supported on the machine, this entry shows the milliseconds 2155 elapsed since the last time AVX512 usage was recorded. The recording 2156 happens on a best effort basis when a task is scheduled out. This means 2157 that the value depends on two factors: 2158 2159 1) The time which the task spent on the CPU without being scheduled 2160 out. With CPU isolation and a single runnable task this can take 2161 several seconds. 2162 2163 2) The time since the task was scheduled out last. Depending on the 2164 reason for being scheduled out (time slice exhausted, syscall ...) 2165 this can be arbitrary long time. 2166 2167 As a consequence the value cannot be considered precise and authoritative 2168 information. The application which uses this information has to be aware 2169 of the overall scenario on the system in order to determine whether a 2170 task is a real AVX512 user or not. Precise information can be obtained 2171 with performance counters. 2172 2173 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2174 the task is unlikely an AVX512 user, but depends on the workload and the 2175 scheduling scenario, it also could be a false negative mentioned above. 2176 2177Chapter 4: Configuring procfs 2178============================= 2179 21804.1 Mount options 2181--------------------- 2182 2183The following mount options are supported: 2184 2185 ========= ======================================================== 2186 hidepid= Set /proc/<pid>/ access mode. 2187 gid= Set the group authorized to learn processes information. 2188 subset= Show only the specified subset of procfs. 2189 ========= ======================================================== 2190 2191hidepid=off or hidepid=0 means classic mode - everybody may access all 2192/proc/<pid>/ directories (default). 2193 2194hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2195directories but their own. Sensitive files like cmdline, sched*, status are now 2196protected against other users. This makes it impossible to learn whether any 2197user runs specific program (given the program doesn't reveal itself by its 2198behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2199other users, poorly written programs passing sensitive information via program 2200arguments are now protected against local eavesdroppers. 2201 2202hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2203fully invisible to other users. It doesn't mean that it hides a fact whether a 2204process with a specific pid value exists (it can be learned by other means, e.g. 2205by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by 2206stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2207gathering information about running processes, whether some daemon runs with 2208elevated privileges, whether other user runs some sensitive program, whether 2209other users run any program at all, etc. 2210 2211hidepid=ptraceable or hidepid=4 means that procfs should only contain 2212/proc/<pid>/ directories that the caller can ptrace. 2213 2214gid= defines a group authorized to learn processes information otherwise 2215prohibited by hidepid=. If you use some daemon like identd which needs to learn 2216information about processes information, just add identd to this group. 2217 2218subset=pid hides all top level files and directories in the procfs that 2219are not related to tasks. 2220 2221Chapter 5: Filesystem behavior 2222============================== 2223 2224Originally, before the advent of pid namepsace, procfs was a global file 2225system. It means that there was only one procfs instance in the system. 2226 2227When pid namespace was added, a separate procfs instance was mounted in 2228each pid namespace. So, procfs mount options are global among all 2229mountpoints within the same namespace:: 2230 2231 # grep ^proc /proc/mounts 2232 proc /proc proc rw,relatime,hidepid=2 0 0 2233 2234 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2235 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2236 +++ exited with 0 +++ 2237 2238 # grep ^proc /proc/mounts 2239 proc /proc proc rw,relatime,hidepid=2 0 0 2240 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2241 2242and only after remounting procfs mount options will change at all 2243mountpoints:: 2244 2245 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2246 2247 # grep ^proc /proc/mounts 2248 proc /proc proc rw,relatime,hidepid=1 0 0 2249 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2250 2251This behavior is different from the behavior of other filesystems. 2252 2253The new procfs behavior is more like other filesystems. Each procfs mount 2254creates a new procfs instance. Mount options affect own procfs instance. 2255It means that it became possible to have several procfs instances 2256displaying tasks with different filtering options in one pid namespace:: 2257 2258 # mount -o hidepid=invisible -t proc proc /proc 2259 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2260 # grep ^proc /proc/mounts 2261 proc /proc proc rw,relatime,hidepid=invisible 0 0 2262 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2263