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 3.13 /proc/<pid>/fd - List of symlinks to open files 51 52 4 Configuring procfs 53 4.1 Mount options 54 55 5 Filesystem behavior 56 57Preface 58======= 59 600.1 Introduction/Credits 61------------------------ 62 63This documentation is part of a soon (or so we hope) to be released book on 64the SuSE Linux distribution. As there is no complete documentation for the 65/proc file system and we've used many freely available sources to write these 66chapters, it seems only fair to give the work back to the Linux community. 67This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 68afraid it's still far from complete, but we hope it will be useful. As far as 69we know, it is the first 'all-in-one' document about the /proc file system. It 70is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 71SPARC, AXP, etc., features, you probably won't find what you are looking for. 72It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 73additions and patches are welcome and will be added to this document if you 74mail them to Bodo. 75 76We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 77other people for help compiling this documentation. We'd also like to extend a 78special thank you to Andi Kleen for documentation, which we relied on heavily 79to create this document, as well as the additional information he provided. 80Thanks to everybody else who contributed source or docs to the Linux kernel 81and helped create a great piece of software... :) 82 83If you have any comments, corrections or additions, please don't hesitate to 84contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 85document. 86 87The latest version of this document is available online at 88https://www.kernel.org/doc/html/latest/filesystems/proc.html 89 90If the above direction does not works for you, you could try the kernel 91mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 92comandante@zaralinux.com. 93 940.2 Legal Stuff 95--------------- 96 97We don't guarantee the correctness of this document, and if you come to us 98complaining about how you screwed up your system because of incorrect 99documentation, we won't feel responsible... 100 101Chapter 1: Collecting System Information 102======================================== 103 104In This Chapter 105--------------- 106* Investigating the properties of the pseudo file system /proc and its 107 ability to provide information on the running Linux system 108* Examining /proc's structure 109* Uncovering various information about the kernel and the processes running 110 on the system 111 112------------------------------------------------------------------------------ 113 114The proc file system acts as an interface to internal data structures in the 115kernel. It can be used to obtain information about the system and to change 116certain kernel parameters at runtime (sysctl). 117 118First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 119show you how you can use /proc/sys to change settings. 120 1211.1 Process-Specific Subdirectories 122----------------------------------- 123 124The directory /proc contains (among other things) one subdirectory for each 125process running on the system, which is named after the process ID (PID). 126 127The link 'self' points to the process reading the file system. Each process 128subdirectory has the entries listed in Table 1-1. 129 130Note that an open file descriptor to /proc/<pid> or to any of its 131contained files or subdirectories does not prevent <pid> being reused 132for some other process in the event that <pid> exits. Operations on 133open /proc/<pid> file descriptors corresponding to dead processes 134never act on any new process that the kernel may, through chance, have 135also assigned the process ID <pid>. Instead, operations on these FDs 136usually fail with ESRCH. 137 138.. table:: Table 1-1: Process specific entries in /proc 139 140 ============= =============================================================== 141 File Content 142 ============= =============================================================== 143 clear_refs Clears page referenced bits shown in smaps output 144 cmdline Command line arguments 145 cpu Current and last cpu in which it was executed (2.4)(smp) 146 cwd Link to the current working directory 147 environ Values of environment variables 148 exe Link to the executable of this process 149 fd Directory, which contains all file descriptors 150 maps Memory maps to executables and library files (2.4) 151 mem Memory held by this process 152 root Link to the root directory of this process 153 stat Process status 154 statm Process memory status information 155 status Process status in human readable form 156 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 157 symbol the task is blocked in - or "0" if not blocked. 158 pagemap Page table 159 stack Report full stack trace, enable via CONFIG_STACKTRACE 160 smaps An extension based on maps, showing the memory consumption of 161 each mapping and flags associated with it 162 smaps_rollup Accumulated smaps stats for all mappings of the process. This 163 can be derived from smaps, but is faster and more convenient 164 numa_maps An extension based on maps, showing the memory locality and 165 binding policy as well as mem usage (in pages) of each mapping. 166 ============= =============================================================== 167 168For example, to get the status information of a process, all you have to do is 169read the file /proc/PID/status:: 170 171 >cat /proc/self/status 172 Name: cat 173 State: R (running) 174 Tgid: 5452 175 Pid: 5452 176 PPid: 743 177 TracerPid: 0 (2.4) 178 Uid: 501 501 501 501 179 Gid: 100 100 100 100 180 FDSize: 256 181 Groups: 100 14 16 182 Kthread: 0 183 VmPeak: 5004 kB 184 VmSize: 5004 kB 185 VmLck: 0 kB 186 VmHWM: 476 kB 187 VmRSS: 476 kB 188 RssAnon: 352 kB 189 RssFile: 120 kB 190 RssShmem: 4 kB 191 VmData: 156 kB 192 VmStk: 88 kB 193 VmExe: 68 kB 194 VmLib: 1412 kB 195 VmPTE: 20 kb 196 VmSwap: 0 kB 197 HugetlbPages: 0 kB 198 CoreDumping: 0 199 THP_enabled: 1 200 Threads: 1 201 SigQ: 0/28578 202 SigPnd: 0000000000000000 203 ShdPnd: 0000000000000000 204 SigBlk: 0000000000000000 205 SigIgn: 0000000000000000 206 SigCgt: 0000000000000000 207 CapInh: 00000000fffffeff 208 CapPrm: 0000000000000000 209 CapEff: 0000000000000000 210 CapBnd: ffffffffffffffff 211 CapAmb: 0000000000000000 212 NoNewPrivs: 0 213 Seccomp: 0 214 Speculation_Store_Bypass: thread vulnerable 215 SpeculationIndirectBranch: conditional enabled 216 voluntary_ctxt_switches: 0 217 nonvoluntary_ctxt_switches: 1 218 219This shows you nearly the same information you would get if you viewed it with 220the ps command. In fact, ps uses the proc file system to obtain its 221information. But you get a more detailed view of the process by reading the 222file /proc/PID/status. It fields are described in table 1-2. 223 224The statm file contains more detailed information about the process 225memory usage. Its seven fields are explained in Table 1-3. The stat file 226contains detailed information about the process itself. Its fields are 227explained in Table 1-4. 228 229(for SMP CONFIG users) 230 231For making accounting scalable, RSS related information are handled in an 232asynchronous manner and the value may not be very precise. To see a precise 233snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 234It's slow but very precise. 235 236.. table:: Table 1-2: Contents of the status fields (as of 4.19) 237 238 ========================== =================================================== 239 Field Content 240 ========================== =================================================== 241 Name filename of the executable 242 Umask file mode creation mask 243 State state (R is running, S is sleeping, D is sleeping 244 in an uninterruptible wait, Z is zombie, 245 T is traced or stopped) 246 Tgid thread group ID 247 Ngid NUMA group ID (0 if none) 248 Pid process id 249 PPid process id of the parent process 250 TracerPid PID of process tracing this process (0 if not, or 251 the tracer is outside of the current pid namespace) 252 Uid Real, effective, saved set, and file system UIDs 253 Gid Real, effective, saved set, and file system GIDs 254 FDSize number of file descriptor slots currently allocated 255 Groups supplementary group list 256 NStgid descendant namespace thread group ID hierarchy 257 NSpid descendant namespace process ID hierarchy 258 NSpgid descendant namespace process group ID hierarchy 259 NSsid descendant namespace session ID hierarchy 260 Kthread kernel thread flag, 1 is yes, 0 is no 261 VmPeak peak virtual memory size 262 VmSize total program size 263 VmLck locked memory size 264 VmPin pinned memory size 265 VmHWM peak resident set size ("high water mark") 266 VmRSS size of memory portions. It contains the three 267 following parts 268 (VmRSS = RssAnon + RssFile + RssShmem) 269 RssAnon size of resident anonymous memory 270 RssFile size of resident file mappings 271 RssShmem size of resident shmem memory (includes SysV shm, 272 mapping of tmpfs and shared anonymous mappings) 273 VmData size of private data segments 274 VmStk size of stack segments 275 VmExe size of text segment 276 VmLib size of shared library code 277 VmPTE size of page table entries 278 VmSwap amount of swap used by anonymous private data 279 (shmem swap usage is not included) 280 HugetlbPages size of hugetlb memory portions 281 CoreDumping process's memory is currently being dumped 282 (killing the process may lead to a corrupted core) 283 THP_enabled process is allowed to use THP (returns 0 when 284 PR_SET_THP_DISABLE is set on the process 285 Threads number of threads 286 SigQ number of signals queued/max. number for queue 287 SigPnd bitmap of pending signals for the thread 288 ShdPnd bitmap of shared pending signals for the process 289 SigBlk bitmap of blocked signals 290 SigIgn bitmap of ignored signals 291 SigCgt bitmap of caught signals 292 CapInh bitmap of inheritable capabilities 293 CapPrm bitmap of permitted capabilities 294 CapEff bitmap of effective capabilities 295 CapBnd bitmap of capabilities bounding set 296 CapAmb bitmap of ambient capabilities 297 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) 298 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 299 Speculation_Store_Bypass speculative store bypass mitigation status 300 SpeculationIndirectBranch indirect branch speculation mode 301 Cpus_allowed mask of CPUs on which this process may run 302 Cpus_allowed_list Same as previous, but in "list format" 303 Mems_allowed mask of memory nodes allowed to this process 304 Mems_allowed_list Same as previous, but in "list format" 305 voluntary_ctxt_switches number of voluntary context switches 306 nonvoluntary_ctxt_switches number of non voluntary context switches 307 ========================== =================================================== 308 309 310.. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3) 311 312 ======== =============================== ============================== 313 Field Content 314 ======== =============================== ============================== 315 size total program size (pages) (same as VmSize in status) 316 resident size of memory portions (pages) (same as VmRSS in status) 317 shared number of pages that are shared (i.e. backed by a file, same 318 as RssFile+RssShmem in status) 319 trs number of pages that are 'code' (not including libs; broken, 320 includes data segment) 321 lrs number of pages of library (always 0 on 2.6) 322 drs number of pages of data/stack (including libs; broken, 323 includes library text) 324 dt number of dirty pages (always 0 on 2.6) 325 ======== =============================== ============================== 326 327 328.. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7) 329 330 ============= =============================================================== 331 Field Content 332 ============= =============================================================== 333 pid process id 334 tcomm filename of the executable 335 state state (R is running, S is sleeping, D is sleeping in an 336 uninterruptible wait, Z is zombie, T is traced or stopped) 337 ppid process id of the parent process 338 pgrp pgrp of the process 339 sid session id 340 tty_nr tty the process uses 341 tty_pgrp pgrp of the tty 342 flags task flags 343 min_flt number of minor faults 344 cmin_flt number of minor faults with child's 345 maj_flt number of major faults 346 cmaj_flt number of major faults with child's 347 utime user mode jiffies 348 stime kernel mode jiffies 349 cutime user mode jiffies with child's 350 cstime kernel mode jiffies with child's 351 priority priority level 352 nice nice level 353 num_threads number of threads 354 it_real_value (obsolete, always 0) 355 start_time time the process started after system boot 356 vsize virtual memory size 357 rss resident set memory size 358 rsslim current limit in bytes on the rss 359 start_code address above which program text can run 360 end_code address below which program text can run 361 start_stack address of the start of the main process stack 362 esp current value of ESP 363 eip current value of EIP 364 pending bitmap of pending signals 365 blocked bitmap of blocked signals 366 sigign bitmap of ignored signals 367 sigcatch bitmap of caught signals 368 0 (place holder, used to be the wchan address, 369 use /proc/PID/wchan instead) 370 0 (place holder) 371 0 (place holder) 372 exit_signal signal to send to parent thread on exit 373 task_cpu which CPU the task is scheduled on 374 rt_priority realtime priority 375 policy scheduling policy (man sched_setscheduler) 376 blkio_ticks time spent waiting for block IO 377 gtime guest time of the task in jiffies 378 cgtime guest time of the task children in jiffies 379 start_data address above which program data+bss is placed 380 end_data address below which program data+bss is placed 381 start_brk address above which program heap can be expanded with brk() 382 arg_start address above which program command line is placed 383 arg_end address below which program command line is placed 384 env_start address above which program environment is placed 385 env_end address below which program environment is placed 386 exit_code the thread's exit_code in the form reported by the waitpid 387 system call 388 ============= =============================================================== 389 390The /proc/PID/maps file contains the currently mapped memory regions and 391their access permissions. 392 393The format is:: 394 395 address perms offset dev inode pathname 396 397 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 398 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 399 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] 400 a7cb1000-a7cb2000 ---p 00000000 00:00 0 401 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 402 a7eb2000-a7eb3000 ---p 00000000 00:00 0 403 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 404 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 405 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 406 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 407 a800b000-a800e000 rw-p 00000000 00:00 0 408 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 409 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 410 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 411 a8024000-a8027000 rw-p 00000000 00:00 0 412 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 413 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 414 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 415 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 416 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 417 418where "address" is the address space in the process that it occupies, "perms" 419is a set of permissions:: 420 421 r = read 422 w = write 423 x = execute 424 s = shared 425 p = private (copy on write) 426 427"offset" is the offset into the mapping, "dev" is the device (major:minor), and 428"inode" is the inode on that device. 0 indicates that no inode is associated 429with the memory region, as the case would be with BSS (uninitialized data). 430The "pathname" shows the name associated file for this mapping. If the mapping 431is not associated with a file: 432 433 =================== =========================================== 434 [heap] the heap of the program 435 [stack] the stack of the main process 436 [vdso] the "virtual dynamic shared object", 437 the kernel system call handler 438 [anon:<name>] a private anonymous mapping that has been 439 named by userspace 440 [anon_shmem:<name>] an anonymous shared memory mapping that has 441 been named by userspace 442 =================== =========================================== 443 444 or if empty, the mapping is anonymous. 445 446The /proc/PID/smaps is an extension based on maps, showing the memory 447consumption for each of the process's mappings. For each mapping (aka Virtual 448Memory Area, or VMA) there is a series of lines such as the following:: 449 450 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 451 452 Size: 1084 kB 453 KernelPageSize: 4 kB 454 MMUPageSize: 4 kB 455 Rss: 892 kB 456 Pss: 374 kB 457 Pss_Dirty: 0 kB 458 Shared_Clean: 892 kB 459 Shared_Dirty: 0 kB 460 Private_Clean: 0 kB 461 Private_Dirty: 0 kB 462 Referenced: 892 kB 463 Anonymous: 0 kB 464 KSM: 0 kB 465 LazyFree: 0 kB 466 AnonHugePages: 0 kB 467 ShmemPmdMapped: 0 kB 468 Shared_Hugetlb: 0 kB 469 Private_Hugetlb: 0 kB 470 Swap: 0 kB 471 SwapPss: 0 kB 472 KernelPageSize: 4 kB 473 MMUPageSize: 4 kB 474 Locked: 0 kB 475 THPeligible: 0 476 VmFlags: rd ex mr mw me dw 477 478The first of these lines shows the same information as is displayed for the 479mapping in /proc/PID/maps. Following lines show the size of the mapping 480(size); the size of each page allocated when backing a VMA (KernelPageSize), 481which is usually the same as the size in the page table entries; the page size 482used by the MMU when backing a VMA (in most cases, the same as KernelPageSize); 483the amount of the mapping that is currently resident in RAM (RSS); the 484process' proportional share of this mapping (PSS); and the number of clean and 485dirty shared and private pages in the mapping. 486 487The "proportional set size" (PSS) of a process is the count of pages it has 488in memory, where each page is divided by the number of processes sharing it. 489So if a process has 1000 pages all to itself, and 1000 shared with one other 490process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which 491consists of dirty pages. ("Pss_Clean" is not included, but it can be 492calculated by subtracting "Pss_Dirty" from "Pss".) 493 494Note that even a page which is part of a MAP_SHARED mapping, but has only 495a single pte mapped, i.e. is currently used by only one process, is accounted 496as private and not as shared. 497 498"Referenced" indicates the amount of memory currently marked as referenced or 499accessed. 500 501"Anonymous" shows the amount of memory that does not belong to any file. Even 502a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 503and a page is modified, the file page is replaced by a private anonymous copy. 504 505"KSM" reports how many of the pages are KSM pages. Note that KSM-placed zeropages 506are not included, only actual KSM pages. 507 508"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 509The memory isn't freed immediately with madvise(). It's freed in memory 510pressure if the memory is clean. Please note that the printed value might 511be lower than the real value due to optimizations used in the current 512implementation. If this is not desirable please file a bug report. 513 514"AnonHugePages" shows the amount of memory backed by transparent hugepage. 515 516"ShmemPmdMapped" shows the amount of shared (shmem/tmpfs) memory backed by 517huge pages. 518 519"Shared_Hugetlb" and "Private_Hugetlb" show the amounts of memory backed by 520hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 521reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 522 523"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 524 525For shmem mappings, "Swap" includes also the size of the mapped (and not 526replaced by copy-on-write) part of the underlying shmem object out on swap. 527"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 528does not take into account swapped out page of underlying shmem objects. 529"Locked" indicates whether the mapping is locked in memory or not. 530 531"THPeligible" indicates whether the mapping is eligible for allocating 532naturally aligned THP pages of any currently enabled size. 1 if true, 0 533otherwise. 534 535"VmFlags" field deserves a separate description. This member represents the 536kernel flags associated with the particular virtual memory area in two letter 537encoded manner. The codes are the following: 538 539 == ======================================= 540 rd readable 541 wr writeable 542 ex executable 543 sh shared 544 mr may read 545 mw may write 546 me may execute 547 ms may share 548 gd stack segment growns down 549 pf pure PFN range 550 dw disabled write to the mapped file 551 lo pages are locked in memory 552 io memory mapped I/O area 553 sr sequential read advise provided 554 rr random read advise provided 555 dc do not copy area on fork 556 de do not expand area on remapping 557 ac area is accountable 558 nr swap space is not reserved for the area 559 ht area uses huge tlb pages 560 sf synchronous page fault 561 ar architecture specific flag 562 wf wipe on fork 563 dd do not include area into core dump 564 sd soft dirty flag 565 mm mixed map area 566 hg huge page advise flag 567 nh no huge page advise flag 568 mg mergeable advise flag 569 bt arm64 BTI guarded page 570 mt arm64 MTE allocation tags are enabled 571 um userfaultfd missing tracking 572 uw userfaultfd wr-protect tracking 573 ss shadow stack page 574 sl sealed 575 == ======================================= 576 577Note that there is no guarantee that every flag and associated mnemonic will 578be present in all further kernel releases. Things get changed, the flags may 579be vanished or the reverse -- new added. Interpretation of their meaning 580might change in future as well. So each consumer of these flags has to 581follow each specific kernel version for the exact semantic. 582 583This file is only present if the CONFIG_MMU kernel configuration option is 584enabled. 585 586Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 587output can be achieved only in the single read call). 588 589This typically manifests when doing partial reads of these files while the 590memory map is being modified. Despite the races, we do provide the following 591guarantees: 592 5931) The mapped addresses never go backwards, which implies no two 594 regions will ever overlap. 5952) If there is something at a given vaddr during the entirety of the 596 life of the smaps/maps walk, there will be some output for it. 597 598The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 599but their values are the sums of the corresponding values for all mappings of 600the process. Additionally, it contains these fields: 601 602- Pss_Anon 603- Pss_File 604- Pss_Shmem 605 606They represent the proportional shares of anonymous, file, and shmem pages, as 607described for smaps above. These fields are omitted in smaps since each 608mapping identifies the type (anon, file, or shmem) of all pages it contains. 609Thus all information in smaps_rollup can be derived from smaps, but at a 610significantly higher cost. 611 612The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 613bits on both physical and virtual pages associated with a process, and the 614soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 615for details). 616To clear the bits for all the pages associated with the process:: 617 618 > echo 1 > /proc/PID/clear_refs 619 620To clear the bits for the anonymous pages associated with the process:: 621 622 > echo 2 > /proc/PID/clear_refs 623 624To clear the bits for the file mapped pages associated with the process:: 625 626 > echo 3 > /proc/PID/clear_refs 627 628To clear the soft-dirty bit:: 629 630 > echo 4 > /proc/PID/clear_refs 631 632To reset the peak resident set size ("high water mark") to the process's 633current value:: 634 635 > echo 5 > /proc/PID/clear_refs 636 637Any other value written to /proc/PID/clear_refs will have no effect. 638 639The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 640using /proc/kpageflags and number of times a page is mapped using 641/proc/kpagecount. For detailed explanation, see 642Documentation/admin-guide/mm/pagemap.rst. 643 644The /proc/pid/numa_maps is an extension based on maps, showing the memory 645locality and binding policy, as well as the memory usage (in pages) of 646each mapping. The output follows a general format where mapping details get 647summarized separated by blank spaces, one mapping per each file line:: 648 649 address policy mapping details 650 651 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 652 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 653 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 654 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 655 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 656 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 657 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 658 320698b000 default file=/lib64/libc-2.12.so 659 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 660 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 661 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 662 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 663 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 664 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 665 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 666 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 667 668Where: 669 670"address" is the starting address for the mapping; 671 672"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 673 674"mapping details" summarizes mapping data such as mapping type, page usage counters, 675node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 676size, in KB, that is backing the mapping up. 677 6781.2 Kernel data 679--------------- 680 681Similar to the process entries, the kernel data files give information about 682the running kernel. The files used to obtain this information are contained in 683/proc and are listed in Table 1-5. Not all of these will be present in your 684system. It depends on the kernel configuration and the loaded modules, which 685files are there, and which are missing. 686 687.. table:: Table 1-5: Kernel info in /proc 688 689 ============ =============================================================== 690 File Content 691 ============ =============================================================== 692 allocinfo Memory allocations profiling information 693 apm Advanced power management info 694 bootconfig Kernel command line obtained from boot config, 695 and, if there were kernel parameters from the 696 boot loader, a "# Parameters from bootloader:" 697 line followed by a line containing those 698 parameters prefixed by "# ". (5.5) 699 buddyinfo Kernel memory allocator information (see text) (2.5) 700 bus Directory containing bus specific information 701 cmdline Kernel command line, both from bootloader and embedded 702 in the kernel image 703 cpuinfo Info about the CPU 704 devices Available devices (block and character) 705 dma Used DMS channels 706 filesystems Supported filesystems 707 driver Various drivers grouped here, currently rtc (2.4) 708 execdomains Execdomains, related to security (2.4) 709 fb Frame Buffer devices (2.4) 710 fs File system parameters, currently nfs/exports (2.4) 711 ide Directory containing info about the IDE subsystem 712 interrupts Interrupt usage 713 iomem Memory map (2.4) 714 ioports I/O port usage 715 irq Masks for irq to cpu affinity (2.4)(smp?) 716 isapnp ISA PnP (Plug&Play) Info (2.4) 717 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 718 kmsg Kernel messages 719 ksyms Kernel symbol table 720 loadavg Load average of last 1, 5 & 15 minutes; 721 number of processes currently runnable (running or on ready queue); 722 total number of processes in system; 723 last pid created. 724 All fields are separated by one space except "number of 725 processes currently runnable" and "total number of processes 726 in system", which are separated by a slash ('/'). Example: 727 0.61 0.61 0.55 3/828 22084 728 locks Kernel locks 729 meminfo Memory info 730 misc Miscellaneous 731 modules List of loaded modules 732 mounts Mounted filesystems 733 net Networking info (see text) 734 pagetypeinfo Additional page allocator information (see text) (2.5) 735 partitions Table of partitions known to the system 736 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 737 decoupled by lspci (2.4) 738 rtc Real time clock 739 scsi SCSI info (see text) 740 slabinfo Slab pool info 741 softirqs softirq usage 742 stat Overall statistics 743 swaps Swap space utilization 744 sys See chapter 2 745 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 746 tty Info of tty drivers 747 uptime Wall clock since boot, combined idle time of all cpus 748 version Kernel version 749 video bttv info of video resources (2.4) 750 vmallocinfo Show vmalloced areas 751 ============ =============================================================== 752 753You can, for example, check which interrupts are currently in use and what 754they are used for by looking in the file /proc/interrupts:: 755 756 > cat /proc/interrupts 757 CPU0 758 0: 8728810 XT-PIC timer 759 1: 895 XT-PIC keyboard 760 2: 0 XT-PIC cascade 761 3: 531695 XT-PIC aha152x 762 4: 2014133 XT-PIC serial 763 5: 44401 XT-PIC pcnet_cs 764 8: 2 XT-PIC rtc 765 11: 8 XT-PIC i82365 766 12: 182918 XT-PIC PS/2 Mouse 767 13: 1 XT-PIC fpu 768 14: 1232265 XT-PIC ide0 769 15: 7 XT-PIC ide1 770 NMI: 0 771 772In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 773output of a SMP machine):: 774 775 > cat /proc/interrupts 776 777 CPU0 CPU1 778 0: 1243498 1214548 IO-APIC-edge timer 779 1: 8949 8958 IO-APIC-edge keyboard 780 2: 0 0 XT-PIC cascade 781 5: 11286 10161 IO-APIC-edge soundblaster 782 8: 1 0 IO-APIC-edge rtc 783 9: 27422 27407 IO-APIC-edge 3c503 784 12: 113645 113873 IO-APIC-edge PS/2 Mouse 785 13: 0 0 XT-PIC fpu 786 14: 22491 24012 IO-APIC-edge ide0 787 15: 2183 2415 IO-APIC-edge ide1 788 17: 30564 30414 IO-APIC-level eth0 789 18: 177 164 IO-APIC-level bttv 790 NMI: 2457961 2457959 791 LOC: 2457882 2457881 792 ERR: 2155 793 794NMI is incremented in this case because every timer interrupt generates a NMI 795(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 796 797LOC is the local interrupt counter of the internal APIC of every CPU. 798 799ERR is incremented in the case of errors in the IO-APIC bus (the bus that 800connects the CPUs in a SMP system. This means that an error has been detected, 801the IO-APIC automatically retry the transmission, so it should not be a big 802problem, but you should read the SMP-FAQ. 803 804In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 805/proc/interrupts to display every IRQ vector in use by the system, not 806just those considered 'most important'. The new vectors are: 807 808THR 809 interrupt raised when a machine check threshold counter 810 (typically counting ECC corrected errors of memory or cache) exceeds 811 a configurable threshold. Only available on some systems. 812 813TRM 814 a thermal event interrupt occurs when a temperature threshold 815 has been exceeded for the CPU. This interrupt may also be generated 816 when the temperature drops back to normal. 817 818SPU 819 a spurious interrupt is some interrupt that was raised then lowered 820 by some IO device before it could be fully processed by the APIC. Hence 821 the APIC sees the interrupt but does not know what device it came from. 822 For this case the APIC will generate the interrupt with a IRQ vector 823 of 0xff. This might also be generated by chipset bugs. 824 825RES, CAL, TLB 826 rescheduling, call and TLB flush interrupts are 827 sent from one CPU to another per the needs of the OS. Typically, 828 their statistics are used by kernel developers and interested users to 829 determine the occurrence of interrupts of the given type. 830 831The above IRQ vectors are displayed only when relevant. For example, 832the threshold vector does not exist on x86_64 platforms. Others are 833suppressed when the system is a uniprocessor. As of this writing, only 834i386 and x86_64 platforms support the new IRQ vector displays. 835 836Of some interest is the introduction of the /proc/irq directory to 2.4. 837It could be used to set IRQ to CPU affinity. This means that you can "hook" an 838IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 839irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 840prof_cpu_mask. 841 842For example:: 843 844 > ls /proc/irq/ 845 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 846 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 847 > ls /proc/irq/0/ 848 smp_affinity 849 850smp_affinity is a bitmask, in which you can specify which CPUs can handle the 851IRQ. You can set it by doing:: 852 853 > echo 1 > /proc/irq/10/smp_affinity 854 855This means that only the first CPU will handle the IRQ, but you can also echo 8565 which means that only the first and third CPU can handle the IRQ. 857 858The contents of each smp_affinity file is the same by default:: 859 860 > cat /proc/irq/0/smp_affinity 861 ffffffff 862 863There is an alternate interface, smp_affinity_list which allows specifying 864a CPU range instead of a bitmask:: 865 866 > cat /proc/irq/0/smp_affinity_list 867 1024-1031 868 869The default_smp_affinity mask applies to all non-active IRQs, which are the 870IRQs which have not yet been allocated/activated, and hence which lack a 871/proc/irq/[0-9]* directory. 872 873The node file on an SMP system shows the node to which the device using the IRQ 874reports itself as being attached. This hardware locality information does not 875include information about any possible driver locality preference. 876 877prof_cpu_mask specifies which CPUs are to be profiled by the system wide 878profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 879 880The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 881between all the CPUs which are allowed to handle it. As usual the kernel has 882more info than you and does a better job than you, so the defaults are the 883best choice for almost everyone. [Note this applies only to those IO-APIC's 884that support "Round Robin" interrupt distribution.] 885 886There are three more important subdirectories in /proc: net, scsi, and sys. 887The general rule is that the contents, or even the existence of these 888directories, depend on your kernel configuration. If SCSI is not enabled, the 889directory scsi may not exist. The same is true with the net, which is there 890only when networking support is present in the running kernel. 891 892The slabinfo file gives information about memory usage at the slab level. 893Linux uses slab pools for memory management above page level in version 2.2. 894Commonly used objects have their own slab pool (such as network buffers, 895directory cache, and so on). 896 897:: 898 899 > cat /proc/buddyinfo 900 901 Node 0, zone DMA 0 4 5 4 4 3 ... 902 Node 0, zone Normal 1 0 0 1 101 8 ... 903 Node 0, zone HighMem 2 0 0 1 1 0 ... 904 905External fragmentation is a problem under some workloads, and buddyinfo is a 906useful tool for helping diagnose these problems. Buddyinfo will give you a 907clue as to how big an area you can safely allocate, or why a previous 908allocation failed. 909 910Each column represents the number of pages of a certain order which are 911available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 912ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 913available in ZONE_NORMAL, etc... 914 915More information relevant to external fragmentation can be found in 916pagetypeinfo:: 917 918 > cat /proc/pagetypeinfo 919 Page block order: 9 920 Pages per block: 512 921 922 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 923 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 924 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 925 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 926 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 927 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 928 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 929 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 930 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 931 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 932 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 933 934 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 935 Node 0, zone DMA 2 0 5 1 0 936 Node 0, zone DMA32 41 6 967 2 0 937 938Fragmentation avoidance in the kernel works by grouping pages of different 939migrate types into the same contiguous regions of memory called page blocks. 940A page block is typically the size of the default hugepage size, e.g. 2MB on 941X86-64. By keeping pages grouped based on their ability to move, the kernel 942can reclaim pages within a page block to satisfy a high-order allocation. 943 944The pagetypinfo begins with information on the size of a page block. It 945then gives the same type of information as buddyinfo except broken down 946by migrate-type and finishes with details on how many page blocks of each 947type exist. 948 949If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 950from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 951make an estimate of the likely number of huge pages that can be allocated 952at a given point in time. All the "Movable" blocks should be allocatable 953unless memory has been mlock()'d. Some of the Reclaimable blocks should 954also be allocatable although a lot of filesystem metadata may have to be 955reclaimed to achieve this. 956 957 958allocinfo 959~~~~~~~~~ 960 961Provides information about memory allocations at all locations in the code 962base. Each allocation in the code is identified by its source file, line 963number, module (if originates from a loadable module) and the function calling 964the allocation. The number of bytes allocated and number of calls at each 965location are reported. The first line indicates the version of the file, the 966second line is the header listing fields in the file. 967 968Example output. 969 970:: 971 972 > tail -n +3 /proc/allocinfo | sort -rn 973 127664128 31168 mm/page_ext.c:270 func:alloc_page_ext 974 56373248 4737 mm/slub.c:2259 func:alloc_slab_page 975 14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded 976 14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash 977 13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs 978 11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio 979 9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node 980 4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable 981 4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start 982 3940352 962 mm/memory.c:4214 func:alloc_anon_folio 983 2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node 984 ... 985 986 987meminfo 988~~~~~~~ 989 990Provides information about distribution and utilization of memory. This 991varies by architecture and compile options. Some of the counters reported 992here overlap. The memory reported by the non overlapping counters may not 993add up to the overall memory usage and the difference for some workloads 994can be substantial. In many cases there are other means to find out 995additional memory using subsystem specific interfaces, for instance 996/proc/net/sockstat for TCP memory allocations. 997 998Example output. You may not have all of these fields. 999 1000:: 1001 1002 > cat /proc/meminfo 1003 1004 MemTotal: 32858820 kB 1005 MemFree: 21001236 kB 1006 MemAvailable: 27214312 kB 1007 Buffers: 581092 kB 1008 Cached: 5587612 kB 1009 SwapCached: 0 kB 1010 Active: 3237152 kB 1011 Inactive: 7586256 kB 1012 Active(anon): 94064 kB 1013 Inactive(anon): 4570616 kB 1014 Active(file): 3143088 kB 1015 Inactive(file): 3015640 kB 1016 Unevictable: 0 kB 1017 Mlocked: 0 kB 1018 SwapTotal: 0 kB 1019 SwapFree: 0 kB 1020 Zswap: 1904 kB 1021 Zswapped: 7792 kB 1022 Dirty: 12 kB 1023 Writeback: 0 kB 1024 AnonPages: 4654780 kB 1025 Mapped: 266244 kB 1026 Shmem: 9976 kB 1027 KReclaimable: 517708 kB 1028 Slab: 660044 kB 1029 SReclaimable: 517708 kB 1030 SUnreclaim: 142336 kB 1031 KernelStack: 11168 kB 1032 PageTables: 20540 kB 1033 SecPageTables: 0 kB 1034 NFS_Unstable: 0 kB 1035 Bounce: 0 kB 1036 WritebackTmp: 0 kB 1037 CommitLimit: 16429408 kB 1038 Committed_AS: 7715148 kB 1039 VmallocTotal: 34359738367 kB 1040 VmallocUsed: 40444 kB 1041 VmallocChunk: 0 kB 1042 Percpu: 29312 kB 1043 EarlyMemtestBad: 0 kB 1044 HardwareCorrupted: 0 kB 1045 AnonHugePages: 4149248 kB 1046 ShmemHugePages: 0 kB 1047 ShmemPmdMapped: 0 kB 1048 FileHugePages: 0 kB 1049 FilePmdMapped: 0 kB 1050 CmaTotal: 0 kB 1051 CmaFree: 0 kB 1052 HugePages_Total: 0 1053 HugePages_Free: 0 1054 HugePages_Rsvd: 0 1055 HugePages_Surp: 0 1056 Hugepagesize: 2048 kB 1057 Hugetlb: 0 kB 1058 DirectMap4k: 401152 kB 1059 DirectMap2M: 10008576 kB 1060 DirectMap1G: 24117248 kB 1061 1062MemTotal 1063 Total usable RAM (i.e. physical RAM minus a few reserved 1064 bits and the kernel binary code) 1065MemFree 1066 Total free RAM. On highmem systems, the sum of LowFree+HighFree 1067MemAvailable 1068 An estimate of how much memory is available for starting new 1069 applications, without swapping. Calculated from MemFree, 1070 SReclaimable, the size of the file LRU lists, and the low 1071 watermarks in each zone. 1072 The estimate takes into account that the system needs some 1073 page cache to function well, and that not all reclaimable 1074 slab will be reclaimable, due to items being in use. The 1075 impact of those factors will vary from system to system. 1076Buffers 1077 Relatively temporary storage for raw disk blocks 1078 shouldn't get tremendously large (20MB or so) 1079Cached 1080 In-memory cache for files read from the disk (the 1081 pagecache) as well as tmpfs & shmem. 1082 Doesn't include SwapCached. 1083SwapCached 1084 Memory that once was swapped out, is swapped back in but 1085 still also is in the swapfile (if memory is needed it 1086 doesn't need to be swapped out AGAIN because it is already 1087 in the swapfile. This saves I/O) 1088Active 1089 Memory that has been used more recently and usually not 1090 reclaimed unless absolutely necessary. 1091Inactive 1092 Memory which has been less recently used. It is more 1093 eligible to be reclaimed for other purposes 1094Unevictable 1095 Memory allocated for userspace which cannot be reclaimed, such 1096 as mlocked pages, ramfs backing pages, secret memfd pages etc. 1097Mlocked 1098 Memory locked with mlock(). 1099HighTotal, HighFree 1100 Highmem is all memory above ~860MB of physical memory. 1101 Highmem areas are for use by userspace programs, or 1102 for the pagecache. The kernel must use tricks to access 1103 this memory, making it slower to access than lowmem. 1104LowTotal, LowFree 1105 Lowmem is memory which can be used for everything that 1106 highmem can be used for, but it is also available for the 1107 kernel's use for its own data structures. Among many 1108 other things, it is where everything from the Slab is 1109 allocated. Bad things happen when you're out of lowmem. 1110SwapTotal 1111 total amount of swap space available 1112SwapFree 1113 Memory which has been evicted from RAM, and is temporarily 1114 on the disk 1115Zswap 1116 Memory consumed by the zswap backend (compressed size) 1117Zswapped 1118 Amount of anonymous memory stored in zswap (original size) 1119Dirty 1120 Memory which is waiting to get written back to the disk 1121Writeback 1122 Memory which is actively being written back to the disk 1123AnonPages 1124 Non-file backed pages mapped into userspace page tables 1125Mapped 1126 files which have been mmapped, such as libraries 1127Shmem 1128 Total memory used by shared memory (shmem) and tmpfs 1129KReclaimable 1130 Kernel allocations that the kernel will attempt to reclaim 1131 under memory pressure. Includes SReclaimable (below), and other 1132 direct allocations with a shrinker. 1133Slab 1134 in-kernel data structures cache 1135SReclaimable 1136 Part of Slab, that might be reclaimed, such as caches 1137SUnreclaim 1138 Part of Slab, that cannot be reclaimed on memory pressure 1139KernelStack 1140 Memory consumed by the kernel stacks of all tasks 1141PageTables 1142 Memory consumed by userspace page tables 1143SecPageTables 1144 Memory consumed by secondary page tables, this currently includes 1145 KVM mmu and IOMMU allocations on x86 and arm64. 1146NFS_Unstable 1147 Always zero. Previous counted pages which had been written to 1148 the server, but has not been committed to stable storage. 1149Bounce 1150 Memory used for block device "bounce buffers" 1151WritebackTmp 1152 Memory used by FUSE for temporary writeback buffers 1153CommitLimit 1154 Based on the overcommit ratio ('vm.overcommit_ratio'), 1155 this is the total amount of memory currently available to 1156 be allocated on the system. This limit is only adhered to 1157 if strict overcommit accounting is enabled (mode 2 in 1158 'vm.overcommit_memory'). 1159 1160 The CommitLimit is calculated with the following formula:: 1161 1162 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1163 overcommit_ratio / 100 + [total swap pages] 1164 1165 For example, on a system with 1G of physical RAM and 7G 1166 of swap with a `vm.overcommit_ratio` of 30 it would 1167 yield a CommitLimit of 7.3G. 1168 1169 For more details, see the memory overcommit documentation 1170 in mm/overcommit-accounting. 1171Committed_AS 1172 The amount of memory presently allocated on the system. 1173 The committed memory is a sum of all of the memory which 1174 has been allocated by processes, even if it has not been 1175 "used" by them as of yet. A process which malloc()'s 1G 1176 of memory, but only touches 300M of it will show up as 1177 using 1G. This 1G is memory which has been "committed" to 1178 by the VM and can be used at any time by the allocating 1179 application. With strict overcommit enabled on the system 1180 (mode 2 in 'vm.overcommit_memory'), allocations which would 1181 exceed the CommitLimit (detailed above) will not be permitted. 1182 This is useful if one needs to guarantee that processes will 1183 not fail due to lack of memory once that memory has been 1184 successfully allocated. 1185VmallocTotal 1186 total size of vmalloc virtual address space 1187VmallocUsed 1188 amount of vmalloc area which is used 1189VmallocChunk 1190 largest contiguous block of vmalloc area which is free 1191Percpu 1192 Memory allocated to the percpu allocator used to back percpu 1193 allocations. This stat excludes the cost of metadata. 1194EarlyMemtestBad 1195 The amount of RAM/memory in kB, that was identified as corrupted 1196 by early memtest. If memtest was not run, this field will not 1197 be displayed at all. Size is never rounded down to 0 kB. 1198 That means if 0 kB is reported, you can safely assume 1199 there was at least one pass of memtest and none of the passes 1200 found a single faulty byte of RAM. 1201HardwareCorrupted 1202 The amount of RAM/memory in KB, the kernel identifies as 1203 corrupted. 1204AnonHugePages 1205 Non-file backed huge pages mapped into userspace page tables 1206ShmemHugePages 1207 Memory used by shared memory (shmem) and tmpfs allocated 1208 with huge pages 1209ShmemPmdMapped 1210 Shared memory mapped into userspace with huge pages 1211FileHugePages 1212 Memory used for filesystem data (page cache) allocated 1213 with huge pages 1214FilePmdMapped 1215 Page cache mapped into userspace with huge pages 1216CmaTotal 1217 Memory reserved for the Contiguous Memory Allocator (CMA) 1218CmaFree 1219 Free remaining memory in the CMA reserves 1220HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb 1221 See Documentation/admin-guide/mm/hugetlbpage.rst. 1222DirectMap4k, DirectMap2M, DirectMap1G 1223 Breakdown of page table sizes used in the kernel's 1224 identity mapping of RAM 1225 1226vmallocinfo 1227~~~~~~~~~~~ 1228 1229Provides information about vmalloced/vmaped areas. One line per area, 1230containing the virtual address range of the area, size in bytes, 1231caller information of the creator, and optional information depending 1232on the kind of area: 1233 1234 ========== =================================================== 1235 pages=nr number of pages 1236 phys=addr if a physical address was specified 1237 ioremap I/O mapping (ioremap() and friends) 1238 vmalloc vmalloc() area 1239 vmap vmap()ed pages 1240 user VM_USERMAP area 1241 vpages buffer for pages pointers was vmalloced (huge area) 1242 N<node>=nr (Only on NUMA kernels) 1243 Number of pages allocated on memory node <node> 1244 ========== =================================================== 1245 1246:: 1247 1248 > cat /proc/vmallocinfo 1249 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1250 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1251 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1252 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1253 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1254 phys=7fee8000 ioremap 1255 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1256 phys=7fee7000 ioremap 1257 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1258 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1259 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1260 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1261 pages=2 vmalloc N1=2 1262 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1263 /0x130 [x_tables] pages=4 vmalloc N0=4 1264 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1265 pages=14 vmalloc N2=14 1266 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1267 pages=4 vmalloc N1=4 1268 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1269 pages=2 vmalloc N1=2 1270 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1271 pages=10 vmalloc N0=10 1272 1273 1274softirqs 1275~~~~~~~~ 1276 1277Provides counts of softirq handlers serviced since boot time, for each CPU. 1278 1279:: 1280 1281 > cat /proc/softirqs 1282 CPU0 CPU1 CPU2 CPU3 1283 HI: 0 0 0 0 1284 TIMER: 27166 27120 27097 27034 1285 NET_TX: 0 0 0 17 1286 NET_RX: 42 0 0 39 1287 BLOCK: 0 0 107 1121 1288 TASKLET: 0 0 0 290 1289 SCHED: 27035 26983 26971 26746 1290 HRTIMER: 0 0 0 0 1291 RCU: 1678 1769 2178 2250 1292 12931.3 Networking info in /proc/net 1294-------------------------------- 1295 1296The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1297additional values you get for IP version 6 if you configure the kernel to 1298support this. Table 1-9 lists the files and their meaning. 1299 1300 1301.. table:: Table 1-8: IPv6 info in /proc/net 1302 1303 ========== ===================================================== 1304 File Content 1305 ========== ===================================================== 1306 udp6 UDP sockets (IPv6) 1307 tcp6 TCP sockets (IPv6) 1308 raw6 Raw device statistics (IPv6) 1309 igmp6 IP multicast addresses, which this host joined (IPv6) 1310 if_inet6 List of IPv6 interface addresses 1311 ipv6_route Kernel routing table for IPv6 1312 rt6_stats Global IPv6 routing tables statistics 1313 sockstat6 Socket statistics (IPv6) 1314 snmp6 Snmp data (IPv6) 1315 ========== ===================================================== 1316 1317.. table:: Table 1-9: Network info in /proc/net 1318 1319 ============= ================================================================ 1320 File Content 1321 ============= ================================================================ 1322 arp Kernel ARP table 1323 dev network devices with statistics 1324 dev_mcast the Layer2 multicast groups a device is listening too 1325 (interface index, label, number of references, number of bound 1326 addresses). 1327 dev_stat network device status 1328 ip_fwchains Firewall chain linkage 1329 ip_fwnames Firewall chain names 1330 ip_masq Directory containing the masquerading tables 1331 ip_masquerade Major masquerading table 1332 netstat Network statistics 1333 raw raw device statistics 1334 route Kernel routing table 1335 rpc Directory containing rpc info 1336 rt_cache Routing cache 1337 snmp SNMP data 1338 sockstat Socket statistics 1339 softnet_stat Per-CPU incoming packets queues statistics of online CPUs 1340 tcp TCP sockets 1341 udp UDP sockets 1342 unix UNIX domain sockets 1343 wireless Wireless interface data (Wavelan etc) 1344 igmp IP multicast addresses, which this host joined 1345 psched Global packet scheduler parameters. 1346 netlink List of PF_NETLINK sockets 1347 ip_mr_vifs List of multicast virtual interfaces 1348 ip_mr_cache List of multicast routing cache 1349 ============= ================================================================ 1350 1351You can use this information to see which network devices are available in 1352your system and how much traffic was routed over those devices:: 1353 1354 > cat /proc/net/dev 1355 Inter-|Receive |[... 1356 face |bytes packets errs drop fifo frame compressed multicast|[... 1357 lo: 908188 5596 0 0 0 0 0 0 [... 1358 ppp0:15475140 20721 410 0 0 410 0 0 [... 1359 eth0: 614530 7085 0 0 0 0 0 1 [... 1360 1361 ...] Transmit 1362 ...] bytes packets errs drop fifo colls carrier compressed 1363 ...] 908188 5596 0 0 0 0 0 0 1364 ...] 1375103 17405 0 0 0 0 0 0 1365 ...] 1703981 5535 0 0 0 3 0 0 1366 1367In addition, each Channel Bond interface has its own directory. For 1368example, the bond0 device will have a directory called /proc/net/bond0/. 1369It will contain information that is specific to that bond, such as the 1370current slaves of the bond, the link status of the slaves, and how 1371many times the slaves link has failed. 1372 13731.4 SCSI info 1374------------- 1375 1376If you have a SCSI or ATA host adapter in your system, you'll find a 1377subdirectory named after the driver for this adapter in /proc/scsi. 1378You'll also see a list of all recognized SCSI devices in /proc/scsi:: 1379 1380 >cat /proc/scsi/scsi 1381 Attached devices: 1382 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1383 Vendor: IBM Model: DGHS09U Rev: 03E0 1384 Type: Direct-Access ANSI SCSI revision: 03 1385 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1386 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1387 Type: CD-ROM ANSI SCSI revision: 02 1388 1389 1390The directory named after the driver has one file for each adapter found in 1391the system. These files contain information about the controller, including 1392the used IRQ and the IO address range. The amount of information shown is 1393dependent on the adapter you use. The example shows the output for an Adaptec 1394AHA-2940 SCSI adapter:: 1395 1396 > cat /proc/scsi/aic7xxx/0 1397 1398 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1399 Compile Options: 1400 TCQ Enabled By Default : Disabled 1401 AIC7XXX_PROC_STATS : Disabled 1402 AIC7XXX_RESET_DELAY : 5 1403 Adapter Configuration: 1404 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1405 Ultra Wide Controller 1406 PCI MMAPed I/O Base: 0xeb001000 1407 Adapter SEEPROM Config: SEEPROM found and used. 1408 Adaptec SCSI BIOS: Enabled 1409 IRQ: 10 1410 SCBs: Active 0, Max Active 2, 1411 Allocated 15, HW 16, Page 255 1412 Interrupts: 160328 1413 BIOS Control Word: 0x18b6 1414 Adapter Control Word: 0x005b 1415 Extended Translation: Enabled 1416 Disconnect Enable Flags: 0xffff 1417 Ultra Enable Flags: 0x0001 1418 Tag Queue Enable Flags: 0x0000 1419 Ordered Queue Tag Flags: 0x0000 1420 Default Tag Queue Depth: 8 1421 Tagged Queue By Device array for aic7xxx host instance 0: 1422 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1423 Actual queue depth per device for aic7xxx host instance 0: 1424 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1425 Statistics: 1426 (scsi0:0:0:0) 1427 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1428 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1429 Total transfers 160151 (74577 reads and 85574 writes) 1430 (scsi0:0:6:0) 1431 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1432 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1433 Total transfers 0 (0 reads and 0 writes) 1434 1435 14361.5 Parallel port info in /proc/parport 1437--------------------------------------- 1438 1439The directory /proc/parport contains information about the parallel ports of 1440your system. It has one subdirectory for each port, named after the port 1441number (0,1,2,...). 1442 1443These directories contain the four files shown in Table 1-10. 1444 1445 1446.. table:: Table 1-10: Files in /proc/parport 1447 1448 ========= ==================================================================== 1449 File Content 1450 ========= ==================================================================== 1451 autoprobe Any IEEE-1284 device ID information that has been acquired. 1452 devices list of the device drivers using that port. A + will appear by the 1453 name of the device currently using the port (it might not appear 1454 against any). 1455 hardware Parallel port's base address, IRQ line and DMA channel. 1456 irq IRQ that parport is using for that port. This is in a separate 1457 file to allow you to alter it by writing a new value in (IRQ 1458 number or none). 1459 ========= ==================================================================== 1460 14611.6 TTY info in /proc/tty 1462------------------------- 1463 1464Information about the available and actually used tty's can be found in the 1465directory /proc/tty. You'll find entries for drivers and line disciplines in 1466this directory, as shown in Table 1-11. 1467 1468 1469.. table:: Table 1-11: Files in /proc/tty 1470 1471 ============= ============================================== 1472 File Content 1473 ============= ============================================== 1474 drivers list of drivers and their usage 1475 ldiscs registered line disciplines 1476 driver/serial usage statistic and status of single tty lines 1477 ============= ============================================== 1478 1479To see which tty's are currently in use, you can simply look into the file 1480/proc/tty/drivers:: 1481 1482 > cat /proc/tty/drivers 1483 pty_slave /dev/pts 136 0-255 pty:slave 1484 pty_master /dev/ptm 128 0-255 pty:master 1485 pty_slave /dev/ttyp 3 0-255 pty:slave 1486 pty_master /dev/pty 2 0-255 pty:master 1487 serial /dev/cua 5 64-67 serial:callout 1488 serial /dev/ttyS 4 64-67 serial 1489 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1490 /dev/ptmx /dev/ptmx 5 2 system 1491 /dev/console /dev/console 5 1 system:console 1492 /dev/tty /dev/tty 5 0 system:/dev/tty 1493 unknown /dev/tty 4 1-63 console 1494 1495 14961.7 Miscellaneous kernel statistics in /proc/stat 1497------------------------------------------------- 1498 1499Various pieces of information about kernel activity are available in the 1500/proc/stat file. All of the numbers reported in this file are aggregates 1501since the system first booted. For a quick look, simply cat the file:: 1502 1503 > cat /proc/stat 1504 cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0 1505 cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0 1506 cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0 1507 cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0 1508 cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0 1509 intr 8688370575 8 3373 0 0 0 0 0 0 1 40791 0 0 353317 0 0 0 0 224789828 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 190974333 41958554 123983334 43 0 224593 0 0 0 <more 0's deleted> 1510 ctxt 22848221062 1511 btime 1605316999 1512 processes 746787147 1513 procs_running 2 1514 procs_blocked 0 1515 softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354 1516 1517The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1518lines. These numbers identify the amount of time the CPU has spent performing 1519different kinds of work. Time units are in USER_HZ (typically hundredths of a 1520second). The meanings of the columns are as follows, from left to right: 1521 1522- user: normal processes executing in user mode 1523- nice: niced processes executing in user mode 1524- system: processes executing in kernel mode 1525- idle: twiddling thumbs 1526- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1527 are several problems: 1528 1529 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1530 waiting for I/O to complete. When CPU goes into idle state for 1531 outstanding task I/O, another task will be scheduled on this CPU. 1532 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1533 on any CPU, so the iowait of each CPU is difficult to calculate. 1534 3. The value of iowait field in /proc/stat will decrease in certain 1535 conditions. 1536 1537 So, the iowait is not reliable by reading from /proc/stat. 1538- irq: servicing interrupts 1539- softirq: servicing softirqs 1540- steal: involuntary wait 1541- guest: running a normal guest 1542- guest_nice: running a niced guest 1543 1544The "intr" line gives counts of interrupts serviced since boot time, for each 1545of the possible system interrupts. The first column is the total of all 1546interrupts serviced including unnumbered architecture specific interrupts; 1547each subsequent column is the total for that particular numbered interrupt. 1548Unnumbered interrupts are not shown, only summed into the total. 1549 1550The "ctxt" line gives the total number of context switches across all CPUs. 1551 1552The "btime" line gives the time at which the system booted, in seconds since 1553the Unix epoch. 1554 1555The "processes" line gives the number of processes and threads created, which 1556includes (but is not limited to) those created by calls to the fork() and 1557clone() system calls. 1558 1559The "procs_running" line gives the total number of threads that are 1560running or ready to run (i.e., the total number of runnable threads). 1561 1562The "procs_blocked" line gives the number of processes currently blocked, 1563waiting for I/O to complete. 1564 1565The "softirq" line gives counts of softirqs serviced since boot time, for each 1566of the possible system softirqs. The first column is the total of all 1567softirqs serviced; each subsequent column is the total for that particular 1568softirq. 1569 1570 15711.8 Ext4 file system parameters 1572------------------------------- 1573 1574Information about mounted ext4 file systems can be found in 1575/proc/fs/ext4. Each mounted filesystem will have a directory in 1576/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1577/proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device 1578directory are shown in Table 1-12, below. 1579 1580.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1581 1582 ============== ========================================================== 1583 File Content 1584 mb_groups details of multiblock allocator buddy cache of free blocks 1585 ============== ========================================================== 1586 15871.9 /proc/consoles 1588------------------- 1589Shows registered system console lines. 1590 1591To see which character device lines are currently used for the system console 1592/dev/console, you may simply look into the file /proc/consoles:: 1593 1594 > cat /proc/consoles 1595 tty0 -WU (ECp) 4:7 1596 ttyS0 -W- (Ep) 4:64 1597 1598The columns are: 1599 1600+--------------------+-------------------------------------------------------+ 1601| device | name of the device | 1602+====================+=======================================================+ 1603| operations | * R = can do read operations | 1604| | * W = can do write operations | 1605| | * U = can do unblank | 1606+--------------------+-------------------------------------------------------+ 1607| flags | * E = it is enabled | 1608| | * C = it is preferred console | 1609| | * B = it is primary boot console | 1610| | * p = it is used for printk buffer | 1611| | * b = it is not a TTY but a Braille device | 1612| | * a = it is safe to use when cpu is offline | 1613+--------------------+-------------------------------------------------------+ 1614| major:minor | major and minor number of the device separated by a | 1615| | colon | 1616+--------------------+-------------------------------------------------------+ 1617 1618Summary 1619------- 1620 1621The /proc file system serves information about the running system. It not only 1622allows access to process data but also allows you to request the kernel status 1623by reading files in the hierarchy. 1624 1625The directory structure of /proc reflects the types of information and makes 1626it easy, if not obvious, where to look for specific data. 1627 1628Chapter 2: Modifying System Parameters 1629====================================== 1630 1631In This Chapter 1632--------------- 1633 1634* Modifying kernel parameters by writing into files found in /proc/sys 1635* Exploring the files which modify certain parameters 1636* Review of the /proc/sys file tree 1637 1638------------------------------------------------------------------------------ 1639 1640A very interesting part of /proc is the directory /proc/sys. This is not only 1641a source of information, it also allows you to change parameters within the 1642kernel. Be very careful when attempting this. You can optimize your system, 1643but you can also cause it to crash. Never alter kernel parameters on a 1644production system. Set up a development machine and test to make sure that 1645everything works the way you want it to. You may have no alternative but to 1646reboot the machine once an error has been made. 1647 1648To change a value, simply echo the new value into the file. 1649You need to be root to do this. You can create your own boot script 1650to perform this every time your system boots. 1651 1652The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1653general things in the operation of the Linux kernel. Since some of the files 1654can inadvertently disrupt your system, it is advisable to read both 1655documentation and source before actually making adjustments. In any case, be 1656very careful when writing to any of these files. The entries in /proc may 1657change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1658review the kernel documentation in the directory linux/Documentation. 1659This chapter is heavily based on the documentation included in the pre 2.2 1660kernels, and became part of it in version 2.2.1 of the Linux kernel. 1661 1662Please see: Documentation/admin-guide/sysctl/ directory for descriptions of 1663these entries. 1664 1665Summary 1666------- 1667 1668Certain aspects of kernel behavior can be modified at runtime, without the 1669need to recompile the kernel, or even to reboot the system. The files in the 1670/proc/sys tree can not only be read, but also modified. You can use the echo 1671command to write value into these files, thereby changing the default settings 1672of the kernel. 1673 1674 1675Chapter 3: Per-process Parameters 1676================================= 1677 16783.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1679-------------------------------------------------------------------------------- 1680 1681These files can be used to adjust the badness heuristic used to select which 1682process gets killed in out of memory (oom) conditions. 1683 1684The badness heuristic assigns a value to each candidate task ranging from 0 1685(never kill) to 1000 (always kill) to determine which process is targeted. The 1686units are roughly a proportion along that range of allowed memory the process 1687may allocate from based on an estimation of its current memory and swap use. 1688For example, if a task is using all allowed memory, its badness score will be 16891000. If it is using half of its allowed memory, its score will be 500. 1690 1691The amount of "allowed" memory depends on the context in which the oom killer 1692was called. If it is due to the memory assigned to the allocating task's cpuset 1693being exhausted, the allowed memory represents the set of mems assigned to that 1694cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1695memory represents the set of mempolicy nodes. If it is due to a memory 1696limit (or swap limit) being reached, the allowed memory is that configured 1697limit. Finally, if it is due to the entire system being out of memory, the 1698allowed memory represents all allocatable resources. 1699 1700The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1701is used to determine which task to kill. Acceptable values range from -1000 1702(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1703polarize the preference for oom killing either by always preferring a certain 1704task or completely disabling it. The lowest possible value, -1000, is 1705equivalent to disabling oom killing entirely for that task since it will always 1706report a badness score of 0. 1707 1708Consequently, it is very simple for userspace to define the amount of memory to 1709consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1710example, is roughly equivalent to allowing the remainder of tasks sharing the 1711same system, cpuset, mempolicy, or memory controller resources to use at least 171250% more memory. A value of -500, on the other hand, would be roughly 1713equivalent to discounting 50% of the task's allowed memory from being considered 1714as scoring against the task. 1715 1716For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1717be used to tune the badness score. Its acceptable values range from -16 1718(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1719(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1720scaled linearly with /proc/<pid>/oom_score_adj. 1721 1722The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1723value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1724requires CAP_SYS_RESOURCE. 1725 1726 17273.2 /proc/<pid>/oom_score - Display current oom-killer score 1728------------------------------------------------------------- 1729 1730This file can be used to check the current score used by the oom-killer for 1731any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1732process should be killed in an out-of-memory situation. 1733 1734Please note that the exported value includes oom_score_adj so it is 1735effectively in range [0,2000]. 1736 1737 17383.3 /proc/<pid>/io - Display the IO accounting fields 1739------------------------------------------------------- 1740 1741This file contains IO statistics for each running process. 1742 1743Example 1744~~~~~~~ 1745 1746:: 1747 1748 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1749 [1] 3828 1750 1751 test:/tmp # cat /proc/3828/io 1752 rchar: 323934931 1753 wchar: 323929600 1754 syscr: 632687 1755 syscw: 632675 1756 read_bytes: 0 1757 write_bytes: 323932160 1758 cancelled_write_bytes: 0 1759 1760 1761Description 1762~~~~~~~~~~~ 1763 1764rchar 1765^^^^^ 1766 1767I/O counter: chars read 1768The number of bytes which this task has caused to be read from storage. This 1769is simply the sum of bytes which this process passed to read() and pread(). 1770It includes things like tty IO and it is unaffected by whether or not actual 1771physical disk IO was required (the read might have been satisfied from 1772pagecache). 1773 1774 1775wchar 1776^^^^^ 1777 1778I/O counter: chars written 1779The number of bytes which this task has caused, or shall cause to be written 1780to disk. Similar caveats apply here as with rchar. 1781 1782 1783syscr 1784^^^^^ 1785 1786I/O counter: read syscalls 1787Attempt to count the number of read I/O operations, i.e. syscalls like read() 1788and pread(). 1789 1790 1791syscw 1792^^^^^ 1793 1794I/O counter: write syscalls 1795Attempt to count the number of write I/O operations, i.e. syscalls like 1796write() and pwrite(). 1797 1798 1799read_bytes 1800^^^^^^^^^^ 1801 1802I/O counter: bytes read 1803Attempt to count the number of bytes which this process really did cause to 1804be fetched from the storage layer. Done at the submit_bio() level, so it is 1805accurate for block-backed filesystems. <please add status regarding NFS and 1806CIFS at a later time> 1807 1808 1809write_bytes 1810^^^^^^^^^^^ 1811 1812I/O counter: bytes written 1813Attempt to count the number of bytes which this process caused to be sent to 1814the storage layer. This is done at page-dirtying time. 1815 1816 1817cancelled_write_bytes 1818^^^^^^^^^^^^^^^^^^^^^ 1819 1820The big inaccuracy here is truncate. If a process writes 1MB to a file and 1821then deletes the file, it will in fact perform no writeout. But it will have 1822been accounted as having caused 1MB of write. 1823In other words: The number of bytes which this process caused to not happen, 1824by truncating pagecache. A task can cause "negative" IO too. If this task 1825truncates some dirty pagecache, some IO which another task has been accounted 1826for (in its write_bytes) will not be happening. We _could_ just subtract that 1827from the truncating task's write_bytes, but there is information loss in doing 1828that. 1829 1830 1831.. Note:: 1832 1833 At its current implementation state, this is a bit racy on 32-bit machines: 1834 if process A reads process B's /proc/pid/io while process B is updating one 1835 of those 64-bit counters, process A could see an intermediate result. 1836 1837 1838More information about this can be found within the taskstats documentation in 1839Documentation/accounting. 1840 18413.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1842--------------------------------------------------------------- 1843When a process is dumped, all anonymous memory is written to a core file as 1844long as the size of the core file isn't limited. But sometimes we don't want 1845to dump some memory segments, for example, huge shared memory or DAX. 1846Conversely, sometimes we want to save file-backed memory segments into a core 1847file, not only the individual files. 1848 1849/proc/<pid>/coredump_filter allows you to customize which memory segments 1850will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1851of memory types. If a bit of the bitmask is set, memory segments of the 1852corresponding memory type are dumped, otherwise they are not dumped. 1853 1854The following 9 memory types are supported: 1855 1856 - (bit 0) anonymous private memory 1857 - (bit 1) anonymous shared memory 1858 - (bit 2) file-backed private memory 1859 - (bit 3) file-backed shared memory 1860 - (bit 4) ELF header pages in file-backed private memory areas (it is 1861 effective only if the bit 2 is cleared) 1862 - (bit 5) hugetlb private memory 1863 - (bit 6) hugetlb shared memory 1864 - (bit 7) DAX private memory 1865 - (bit 8) DAX shared memory 1866 1867 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1868 are always dumped regardless of the bitmask status. 1869 1870 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1871 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1872 1873The default value of coredump_filter is 0x33; this means all anonymous memory 1874segments, ELF header pages and hugetlb private memory are dumped. 1875 1876If you don't want to dump all shared memory segments attached to pid 1234, 1877write 0x31 to the process's proc file:: 1878 1879 $ echo 0x31 > /proc/1234/coredump_filter 1880 1881When a new process is created, the process inherits the bitmask status from its 1882parent. It is useful to set up coredump_filter before the program runs. 1883For example:: 1884 1885 $ echo 0x7 > /proc/self/coredump_filter 1886 $ ./some_program 1887 18883.5 /proc/<pid>/mountinfo - Information about mounts 1889-------------------------------------------------------- 1890 1891This file contains lines of the form:: 1892 1893 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1894 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4) 1895 1896 (1) mount ID: unique identifier of the mount (may be reused after umount) 1897 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1898 (3) major:minor: value of st_dev for files on filesystem 1899 (4) root: root of the mount within the filesystem 1900 (5) mount point: mount point relative to the process's root 1901 (6) mount options: per mount options 1902 (n…m) optional fields: zero or more fields of the form "tag[:value]" 1903 (m+1) separator: marks the end of the optional fields 1904 (m+2) filesystem type: name of filesystem of the form "type[.subtype]" 1905 (m+3) mount source: filesystem specific information or "none" 1906 (m+4) super options: per super block options 1907 1908Parsers should ignore all unrecognised optional fields. Currently the 1909possible optional fields are: 1910 1911================ ============================================================== 1912shared:X mount is shared in peer group X 1913master:X mount is slave to peer group X 1914propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1915unbindable mount is unbindable 1916================ ============================================================== 1917 1918.. [#] X is the closest dominant peer group under the process's root. If 1919 X is the immediate master of the mount, or if there's no dominant peer 1920 group under the same root, then only the "master:X" field is present 1921 and not the "propagate_from:X" field. 1922 1923For more information on mount propagation see: 1924 1925 Documentation/filesystems/sharedsubtree.rst 1926 1927 19283.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1929-------------------------------------------------------- 1930These files provide a method to access a task's comm value. It also allows for 1931a task to set its own or one of its thread siblings comm value. The comm value 1932is limited in size compared to the cmdline value, so writing anything longer 1933then the kernel's TASK_COMM_LEN (currently 16 chars, including the NUL 1934terminator) will result in a truncated comm value. 1935 1936 19373.7 /proc/<pid>/task/<tid>/children - Information about task children 1938------------------------------------------------------------------------- 1939This file provides a fast way to retrieve first level children pids 1940of a task pointed by <pid>/<tid> pair. The format is a space separated 1941stream of pids. 1942 1943Note the "first level" here -- if a child has its own children they will 1944not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1945to obtain the descendants. 1946 1947Since this interface is intended to be fast and cheap it doesn't 1948guarantee to provide precise results and some children might be 1949skipped, especially if they've exited right after we printed their 1950pids, so one needs to either stop or freeze processes being inspected 1951if precise results are needed. 1952 1953 19543.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1955--------------------------------------------------------------- 1956This file provides information associated with an opened file. The regular 1957files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. 1958The 'pos' represents the current offset of the opened file in decimal 1959form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the 1960file has been created with [see open(2) for details] and 'mnt_id' represents 1961mount ID of the file system containing the opened file [see 3.5 1962/proc/<pid>/mountinfo for details]. 'ino' represents the inode number of 1963the file. 1964 1965A typical output is:: 1966 1967 pos: 0 1968 flags: 0100002 1969 mnt_id: 19 1970 ino: 63107 1971 1972All locks associated with a file descriptor are shown in its fdinfo too:: 1973 1974 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1975 1976The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1977pair provide additional information particular to the objects they represent. 1978 1979Eventfd files 1980~~~~~~~~~~~~~ 1981 1982:: 1983 1984 pos: 0 1985 flags: 04002 1986 mnt_id: 9 1987 ino: 63107 1988 eventfd-count: 5a 1989 1990where 'eventfd-count' is hex value of a counter. 1991 1992Signalfd files 1993~~~~~~~~~~~~~~ 1994 1995:: 1996 1997 pos: 0 1998 flags: 04002 1999 mnt_id: 9 2000 ino: 63107 2001 sigmask: 0000000000000200 2002 2003where 'sigmask' is hex value of the signal mask associated 2004with a file. 2005 2006Epoll files 2007~~~~~~~~~~~ 2008 2009:: 2010 2011 pos: 0 2012 flags: 02 2013 mnt_id: 9 2014 ino: 63107 2015 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 2016 2017where 'tfd' is a target file descriptor number in decimal form, 2018'events' is events mask being watched and the 'data' is data 2019associated with a target [see epoll(7) for more details]. 2020 2021The 'pos' is current offset of the target file in decimal form 2022[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 2023where target file resides, all in hex format. 2024 2025Fsnotify files 2026~~~~~~~~~~~~~~ 2027For inotify files the format is the following:: 2028 2029 pos: 0 2030 flags: 02000000 2031 mnt_id: 9 2032 ino: 63107 2033 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 2034 2035where 'wd' is a watch descriptor in decimal form, i.e. a target file 2036descriptor number, 'ino' and 'sdev' are inode and device where the 2037target file resides and the 'mask' is the mask of events, all in hex 2038form [see inotify(7) for more details]. 2039 2040If the kernel was built with exportfs support, the path to the target 2041file is encoded as a file handle. The file handle is provided by three 2042fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 2043format. 2044 2045If the kernel is built without exportfs support the file handle won't be 2046printed out. 2047 2048If there is no inotify mark attached yet the 'inotify' line will be omitted. 2049 2050For fanotify files the format is:: 2051 2052 pos: 0 2053 flags: 02 2054 mnt_id: 9 2055 ino: 63107 2056 fanotify flags:10 event-flags:0 2057 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 2058 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 2059 2060where fanotify 'flags' and 'event-flags' are values used in fanotify_init 2061call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 2062flags associated with mark which are tracked separately from events 2063mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 2064mask and 'ignored_mask' is the mask of events which are to be ignored. 2065All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2066provide information about flags and mask used in fanotify_mark 2067call [see fsnotify manpage for details]. 2068 2069While the first three lines are mandatory and always printed, the rest is 2070optional and may be omitted if no marks created yet. 2071 2072Timerfd files 2073~~~~~~~~~~~~~ 2074 2075:: 2076 2077 pos: 0 2078 flags: 02 2079 mnt_id: 9 2080 ino: 63107 2081 clockid: 0 2082 ticks: 0 2083 settime flags: 01 2084 it_value: (0, 49406829) 2085 it_interval: (1, 0) 2086 2087where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2088that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2089flags in octal form been used to setup the timer [see timerfd_settime(2) for 2090details]. 'it_value' is remaining time until the timer expiration. 2091'it_interval' is the interval for the timer. Note the timer might be set up 2092with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2093still exhibits timer's remaining time. 2094 2095DMA Buffer files 2096~~~~~~~~~~~~~~~~ 2097 2098:: 2099 2100 pos: 0 2101 flags: 04002 2102 mnt_id: 9 2103 ino: 63107 2104 size: 32768 2105 count: 2 2106 exp_name: system-heap 2107 2108where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of 2109the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. 2110 21113.9 /proc/<pid>/map_files - Information about memory mapped files 2112--------------------------------------------------------------------- 2113This directory contains symbolic links which represent memory mapped files 2114the process is maintaining. Example output:: 2115 2116 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2117 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2118 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2119 | ... 2120 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2121 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2122 2123The name of a link represents the virtual memory bounds of a mapping, i.e. 2124vm_area_struct::vm_start-vm_area_struct::vm_end. 2125 2126The main purpose of the map_files is to retrieve a set of memory mapped 2127files in a fast way instead of parsing /proc/<pid>/maps or 2128/proc/<pid>/smaps, both of which contain many more records. At the same 2129time one can open(2) mappings from the listings of two processes and 2130comparing their inode numbers to figure out which anonymous memory areas 2131are actually shared. 2132 21333.10 /proc/<pid>/timerslack_ns - Task timerslack value 2134--------------------------------------------------------- 2135This file provides the value of the task's timerslack value in nanoseconds. 2136This value specifies an amount of time that normal timers may be deferred 2137in order to coalesce timers and avoid unnecessary wakeups. 2138 2139This allows a task's interactivity vs power consumption tradeoff to be 2140adjusted. 2141 2142Writing 0 to the file will set the task's timerslack to the default value. 2143 2144Valid values are from 0 - ULLONG_MAX 2145 2146An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2147permissions on the task specified to change its timerslack_ns value. 2148 21493.11 /proc/<pid>/patch_state - Livepatch patch operation state 2150----------------------------------------------------------------- 2151When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2152patch state for the task. 2153 2154A value of '-1' indicates that no patch is in transition. 2155 2156A value of '0' indicates that a patch is in transition and the task is 2157unpatched. If the patch is being enabled, then the task hasn't been 2158patched yet. If the patch is being disabled, then the task has already 2159been unpatched. 2160 2161A value of '1' indicates that a patch is in transition and the task is 2162patched. If the patch is being enabled, then the task has already been 2163patched. If the patch is being disabled, then the task hasn't been 2164unpatched yet. 2165 21663.12 /proc/<pid>/arch_status - task architecture specific status 2167------------------------------------------------------------------- 2168When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2169architecture specific status of the task. 2170 2171Example 2172~~~~~~~ 2173 2174:: 2175 2176 $ cat /proc/6753/arch_status 2177 AVX512_elapsed_ms: 8 2178 2179Description 2180~~~~~~~~~~~ 2181 2182x86 specific entries 2183~~~~~~~~~~~~~~~~~~~~~ 2184 2185AVX512_elapsed_ms 2186^^^^^^^^^^^^^^^^^^ 2187 2188 If AVX512 is supported on the machine, this entry shows the milliseconds 2189 elapsed since the last time AVX512 usage was recorded. The recording 2190 happens on a best effort basis when a task is scheduled out. This means 2191 that the value depends on two factors: 2192 2193 1) The time which the task spent on the CPU without being scheduled 2194 out. With CPU isolation and a single runnable task this can take 2195 several seconds. 2196 2197 2) The time since the task was scheduled out last. Depending on the 2198 reason for being scheduled out (time slice exhausted, syscall ...) 2199 this can be arbitrary long time. 2200 2201 As a consequence the value cannot be considered precise and authoritative 2202 information. The application which uses this information has to be aware 2203 of the overall scenario on the system in order to determine whether a 2204 task is a real AVX512 user or not. Precise information can be obtained 2205 with performance counters. 2206 2207 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2208 the task is unlikely an AVX512 user, but depends on the workload and the 2209 scheduling scenario, it also could be a false negative mentioned above. 2210 22113.13 /proc/<pid>/fd - List of symlinks to open files 2212------------------------------------------------------- 2213This directory contains symbolic links which represent open files 2214the process is maintaining. Example output:: 2215 2216 lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null 2217 l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null 2218 lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]' 2219 lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]' 2220 lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]' 2221 2222The number of open files for the process is stored in 'size' member 2223of stat() output for /proc/<pid>/fd for fast access. 2224------------------------------------------------------- 2225 2226 2227Chapter 4: Configuring procfs 2228============================= 2229 22304.1 Mount options 2231--------------------- 2232 2233The following mount options are supported: 2234 2235 ========= ======================================================== 2236 hidepid= Set /proc/<pid>/ access mode. 2237 gid= Set the group authorized to learn processes information. 2238 subset= Show only the specified subset of procfs. 2239 ========= ======================================================== 2240 2241hidepid=off or hidepid=0 means classic mode - everybody may access all 2242/proc/<pid>/ directories (default). 2243 2244hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2245directories but their own. Sensitive files like cmdline, sched*, status are now 2246protected against other users. This makes it impossible to learn whether any 2247user runs specific program (given the program doesn't reveal itself by its 2248behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2249other users, poorly written programs passing sensitive information via program 2250arguments are now protected against local eavesdroppers. 2251 2252hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2253fully invisible to other users. It doesn't mean that it hides a fact whether a 2254process with a specific pid value exists (it can be learned by other means, e.g. 2255by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by 2256stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2257gathering information about running processes, whether some daemon runs with 2258elevated privileges, whether other user runs some sensitive program, whether 2259other users run any program at all, etc. 2260 2261hidepid=ptraceable or hidepid=4 means that procfs should only contain 2262/proc/<pid>/ directories that the caller can ptrace. 2263 2264gid= defines a group authorized to learn processes information otherwise 2265prohibited by hidepid=. If you use some daemon like identd which needs to learn 2266information about processes information, just add identd to this group. 2267 2268subset=pid hides all top level files and directories in the procfs that 2269are not related to tasks. 2270 2271Chapter 5: Filesystem behavior 2272============================== 2273 2274Originally, before the advent of pid namespace, procfs was a global file 2275system. It means that there was only one procfs instance in the system. 2276 2277When pid namespace was added, a separate procfs instance was mounted in 2278each pid namespace. So, procfs mount options are global among all 2279mountpoints within the same namespace:: 2280 2281 # grep ^proc /proc/mounts 2282 proc /proc proc rw,relatime,hidepid=2 0 0 2283 2284 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2285 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2286 +++ exited with 0 +++ 2287 2288 # grep ^proc /proc/mounts 2289 proc /proc proc rw,relatime,hidepid=2 0 0 2290 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2291 2292and only after remounting procfs mount options will change at all 2293mountpoints:: 2294 2295 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2296 2297 # grep ^proc /proc/mounts 2298 proc /proc proc rw,relatime,hidepid=1 0 0 2299 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2300 2301This behavior is different from the behavior of other filesystems. 2302 2303The new procfs behavior is more like other filesystems. Each procfs mount 2304creates a new procfs instance. Mount options affect own procfs instance. 2305It means that it became possible to have several procfs instances 2306displaying tasks with different filtering options in one pid namespace:: 2307 2308 # mount -o hidepid=invisible -t proc proc /proc 2309 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2310 # grep ^proc /proc/mounts 2311 proc /proc proc rw,relatime,hidepid=invisible 0 0 2312 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2313