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 446Starting with 6.11 kernel, /proc/PID/maps provides an alternative 447ioctl()-based API that gives ability to flexibly and efficiently query and 448filter individual VMAs. This interface is binary and is meant for more 449efficient and easy programmatic use. `struct procmap_query`, defined in 450linux/fs.h UAPI header, serves as an input/output argument to the 451`PROCMAP_QUERY` ioctl() command. See comments in linus/fs.h UAPI header for 452details on query semantics, supported flags, data returned, and general API 453usage information. 454 455The /proc/PID/smaps is an extension based on maps, showing the memory 456consumption for each of the process's mappings. For each mapping (aka Virtual 457Memory Area, or VMA) there is a series of lines such as the following:: 458 459 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 460 461 Size: 1084 kB 462 KernelPageSize: 4 kB 463 MMUPageSize: 4 kB 464 Rss: 892 kB 465 Pss: 374 kB 466 Pss_Dirty: 0 kB 467 Shared_Clean: 892 kB 468 Shared_Dirty: 0 kB 469 Private_Clean: 0 kB 470 Private_Dirty: 0 kB 471 Referenced: 892 kB 472 Anonymous: 0 kB 473 KSM: 0 kB 474 LazyFree: 0 kB 475 AnonHugePages: 0 kB 476 ShmemPmdMapped: 0 kB 477 Shared_Hugetlb: 0 kB 478 Private_Hugetlb: 0 kB 479 Swap: 0 kB 480 SwapPss: 0 kB 481 KernelPageSize: 4 kB 482 MMUPageSize: 4 kB 483 Locked: 0 kB 484 THPeligible: 0 485 VmFlags: rd ex mr mw me dw 486 487The first of these lines shows the same information as is displayed for the 488mapping in /proc/PID/maps. Following lines show the size of the mapping 489(size); the size of each page allocated when backing a VMA (KernelPageSize), 490which is usually the same as the size in the page table entries; the page size 491used by the MMU when backing a VMA (in most cases, the same as KernelPageSize); 492the amount of the mapping that is currently resident in RAM (RSS); the 493process' proportional share of this mapping (PSS); and the number of clean and 494dirty shared and private pages in the mapping. 495 496The "proportional set size" (PSS) of a process is the count of pages it has 497in memory, where each page is divided by the number of processes sharing it. 498So if a process has 1000 pages all to itself, and 1000 shared with one other 499process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which 500consists of dirty pages. ("Pss_Clean" is not included, but it can be 501calculated by subtracting "Pss_Dirty" from "Pss".) 502 503Note that even a page which is part of a MAP_SHARED mapping, but has only 504a single pte mapped, i.e. is currently used by only one process, is accounted 505as private and not as shared. 506 507"Referenced" indicates the amount of memory currently marked as referenced or 508accessed. 509 510"Anonymous" shows the amount of memory that does not belong to any file. Even 511a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 512and a page is modified, the file page is replaced by a private anonymous copy. 513 514"KSM" reports how many of the pages are KSM pages. Note that KSM-placed zeropages 515are not included, only actual KSM pages. 516 517"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 518The memory isn't freed immediately with madvise(). It's freed in memory 519pressure if the memory is clean. Please note that the printed value might 520be lower than the real value due to optimizations used in the current 521implementation. If this is not desirable please file a bug report. 522 523"AnonHugePages" shows the amount of memory backed by transparent hugepage. 524 525"ShmemPmdMapped" shows the amount of shared (shmem/tmpfs) memory backed by 526huge pages. 527 528"Shared_Hugetlb" and "Private_Hugetlb" show the amounts of memory backed by 529hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 530reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 531 532"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 533 534For shmem mappings, "Swap" includes also the size of the mapped (and not 535replaced by copy-on-write) part of the underlying shmem object out on swap. 536"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 537does not take into account swapped out page of underlying shmem objects. 538"Locked" indicates whether the mapping is locked in memory or not. 539 540"THPeligible" indicates whether the mapping is eligible for allocating 541naturally aligned THP pages of any currently enabled size. 1 if true, 0 542otherwise. 543 544"VmFlags" field deserves a separate description. This member represents the 545kernel flags associated with the particular virtual memory area in two letter 546encoded manner. The codes are the following: 547 548 == ======================================= 549 rd readable 550 wr writeable 551 ex executable 552 sh shared 553 mr may read 554 mw may write 555 me may execute 556 ms may share 557 gd stack segment growns down 558 pf pure PFN range 559 dw disabled write to the mapped file 560 lo pages are locked in memory 561 io memory mapped I/O area 562 sr sequential read advise provided 563 rr random read advise provided 564 dc do not copy area on fork 565 de do not expand area on remapping 566 ac area is accountable 567 nr swap space is not reserved for the area 568 ht area uses huge tlb pages 569 sf synchronous page fault 570 ar architecture specific flag 571 wf wipe on fork 572 dd do not include area into core dump 573 sd soft dirty flag 574 mm mixed map area 575 hg huge page advise flag 576 nh no huge page advise flag 577 mg mergeable advise flag 578 bt arm64 BTI guarded page 579 mt arm64 MTE allocation tags are enabled 580 um userfaultfd missing tracking 581 uw userfaultfd wr-protect tracking 582 ss shadow stack page 583 sl sealed 584 == ======================================= 585 586Note that there is no guarantee that every flag and associated mnemonic will 587be present in all further kernel releases. Things get changed, the flags may 588be vanished or the reverse -- new added. Interpretation of their meaning 589might change in future as well. So each consumer of these flags has to 590follow each specific kernel version for the exact semantic. 591 592This file is only present if the CONFIG_MMU kernel configuration option is 593enabled. 594 595Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 596output can be achieved only in the single read call). 597 598This typically manifests when doing partial reads of these files while the 599memory map is being modified. Despite the races, we do provide the following 600guarantees: 601 6021) The mapped addresses never go backwards, which implies no two 603 regions will ever overlap. 6042) If there is something at a given vaddr during the entirety of the 605 life of the smaps/maps walk, there will be some output for it. 606 607The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 608but their values are the sums of the corresponding values for all mappings of 609the process. Additionally, it contains these fields: 610 611- Pss_Anon 612- Pss_File 613- Pss_Shmem 614 615They represent the proportional shares of anonymous, file, and shmem pages, as 616described for smaps above. These fields are omitted in smaps since each 617mapping identifies the type (anon, file, or shmem) of all pages it contains. 618Thus all information in smaps_rollup can be derived from smaps, but at a 619significantly higher cost. 620 621The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 622bits on both physical and virtual pages associated with a process, and the 623soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 624for details). 625To clear the bits for all the pages associated with the process:: 626 627 > echo 1 > /proc/PID/clear_refs 628 629To clear the bits for the anonymous pages associated with the process:: 630 631 > echo 2 > /proc/PID/clear_refs 632 633To clear the bits for the file mapped pages associated with the process:: 634 635 > echo 3 > /proc/PID/clear_refs 636 637To clear the soft-dirty bit:: 638 639 > echo 4 > /proc/PID/clear_refs 640 641To reset the peak resident set size ("high water mark") to the process's 642current value:: 643 644 > echo 5 > /proc/PID/clear_refs 645 646Any other value written to /proc/PID/clear_refs will have no effect. 647 648The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 649using /proc/kpageflags and number of times a page is mapped using 650/proc/kpagecount. For detailed explanation, see 651Documentation/admin-guide/mm/pagemap.rst. 652 653The /proc/pid/numa_maps is an extension based on maps, showing the memory 654locality and binding policy, as well as the memory usage (in pages) of 655each mapping. The output follows a general format where mapping details get 656summarized separated by blank spaces, one mapping per each file line:: 657 658 address policy mapping details 659 660 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 661 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 662 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 663 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 664 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 665 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 666 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 667 320698b000 default file=/lib64/libc-2.12.so 668 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 669 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 670 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 671 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 672 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 673 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 674 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 675 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 676 677Where: 678 679"address" is the starting address for the mapping; 680 681"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 682 683"mapping details" summarizes mapping data such as mapping type, page usage counters, 684node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 685size, in KB, that is backing the mapping up. 686 6871.2 Kernel data 688--------------- 689 690Similar to the process entries, the kernel data files give information about 691the running kernel. The files used to obtain this information are contained in 692/proc and are listed in Table 1-5. Not all of these will be present in your 693system. It depends on the kernel configuration and the loaded modules, which 694files are there, and which are missing. 695 696.. table:: Table 1-5: Kernel info in /proc 697 698 ============ =============================================================== 699 File Content 700 ============ =============================================================== 701 allocinfo Memory allocations profiling information 702 apm Advanced power management info 703 bootconfig Kernel command line obtained from boot config, 704 and, if there were kernel parameters from the 705 boot loader, a "# Parameters from bootloader:" 706 line followed by a line containing those 707 parameters prefixed by "# ". (5.5) 708 buddyinfo Kernel memory allocator information (see text) (2.5) 709 bus Directory containing bus specific information 710 cmdline Kernel command line, both from bootloader and embedded 711 in the kernel image 712 cpuinfo Info about the CPU 713 devices Available devices (block and character) 714 dma Used DMS channels 715 filesystems Supported filesystems 716 driver Various drivers grouped here, currently rtc (2.4) 717 execdomains Execdomains, related to security (2.4) 718 fb Frame Buffer devices (2.4) 719 fs File system parameters, currently nfs/exports (2.4) 720 ide Directory containing info about the IDE subsystem 721 interrupts Interrupt usage 722 iomem Memory map (2.4) 723 ioports I/O port usage 724 irq Masks for irq to cpu affinity (2.4)(smp?) 725 isapnp ISA PnP (Plug&Play) Info (2.4) 726 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 727 kmsg Kernel messages 728 ksyms Kernel symbol table 729 loadavg Load average of last 1, 5 & 15 minutes; 730 number of processes currently runnable (running or on ready queue); 731 total number of processes in system; 732 last pid created. 733 All fields are separated by one space except "number of 734 processes currently runnable" and "total number of processes 735 in system", which are separated by a slash ('/'). Example: 736 0.61 0.61 0.55 3/828 22084 737 locks Kernel locks 738 meminfo Memory info 739 misc Miscellaneous 740 modules List of loaded modules 741 mounts Mounted filesystems 742 net Networking info (see text) 743 pagetypeinfo Additional page allocator information (see text) (2.5) 744 partitions Table of partitions known to the system 745 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 746 decoupled by lspci (2.4) 747 rtc Real time clock 748 scsi SCSI info (see text) 749 slabinfo Slab pool info 750 softirqs softirq usage 751 stat Overall statistics 752 swaps Swap space utilization 753 sys See chapter 2 754 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 755 tty Info of tty drivers 756 uptime Wall clock since boot, combined idle time of all cpus 757 version Kernel version 758 video bttv info of video resources (2.4) 759 vmallocinfo Show vmalloced areas 760 ============ =============================================================== 761 762You can, for example, check which interrupts are currently in use and what 763they are used for by looking in the file /proc/interrupts:: 764 765 > cat /proc/interrupts 766 CPU0 767 0: 8728810 XT-PIC timer 768 1: 895 XT-PIC keyboard 769 2: 0 XT-PIC cascade 770 3: 531695 XT-PIC aha152x 771 4: 2014133 XT-PIC serial 772 5: 44401 XT-PIC pcnet_cs 773 8: 2 XT-PIC rtc 774 11: 8 XT-PIC i82365 775 12: 182918 XT-PIC PS/2 Mouse 776 13: 1 XT-PIC fpu 777 14: 1232265 XT-PIC ide0 778 15: 7 XT-PIC ide1 779 NMI: 0 780 781In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 782output of a SMP machine):: 783 784 > cat /proc/interrupts 785 786 CPU0 CPU1 787 0: 1243498 1214548 IO-APIC-edge timer 788 1: 8949 8958 IO-APIC-edge keyboard 789 2: 0 0 XT-PIC cascade 790 5: 11286 10161 IO-APIC-edge soundblaster 791 8: 1 0 IO-APIC-edge rtc 792 9: 27422 27407 IO-APIC-edge 3c503 793 12: 113645 113873 IO-APIC-edge PS/2 Mouse 794 13: 0 0 XT-PIC fpu 795 14: 22491 24012 IO-APIC-edge ide0 796 15: 2183 2415 IO-APIC-edge ide1 797 17: 30564 30414 IO-APIC-level eth0 798 18: 177 164 IO-APIC-level bttv 799 NMI: 2457961 2457959 800 LOC: 2457882 2457881 801 ERR: 2155 802 803NMI is incremented in this case because every timer interrupt generates a NMI 804(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 805 806LOC is the local interrupt counter of the internal APIC of every CPU. 807 808ERR is incremented in the case of errors in the IO-APIC bus (the bus that 809connects the CPUs in a SMP system. This means that an error has been detected, 810the IO-APIC automatically retry the transmission, so it should not be a big 811problem, but you should read the SMP-FAQ. 812 813In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 814/proc/interrupts to display every IRQ vector in use by the system, not 815just those considered 'most important'. The new vectors are: 816 817THR 818 interrupt raised when a machine check threshold counter 819 (typically counting ECC corrected errors of memory or cache) exceeds 820 a configurable threshold. Only available on some systems. 821 822TRM 823 a thermal event interrupt occurs when a temperature threshold 824 has been exceeded for the CPU. This interrupt may also be generated 825 when the temperature drops back to normal. 826 827SPU 828 a spurious interrupt is some interrupt that was raised then lowered 829 by some IO device before it could be fully processed by the APIC. Hence 830 the APIC sees the interrupt but does not know what device it came from. 831 For this case the APIC will generate the interrupt with a IRQ vector 832 of 0xff. This might also be generated by chipset bugs. 833 834RES, CAL, TLB 835 rescheduling, call and TLB flush interrupts are 836 sent from one CPU to another per the needs of the OS. Typically, 837 their statistics are used by kernel developers and interested users to 838 determine the occurrence of interrupts of the given type. 839 840The above IRQ vectors are displayed only when relevant. For example, 841the threshold vector does not exist on x86_64 platforms. Others are 842suppressed when the system is a uniprocessor. As of this writing, only 843i386 and x86_64 platforms support the new IRQ vector displays. 844 845Of some interest is the introduction of the /proc/irq directory to 2.4. 846It could be used to set IRQ to CPU affinity. This means that you can "hook" an 847IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 848irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 849prof_cpu_mask. 850 851For example:: 852 853 > ls /proc/irq/ 854 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 855 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 856 > ls /proc/irq/0/ 857 smp_affinity 858 859smp_affinity is a bitmask, in which you can specify which CPUs can handle the 860IRQ. You can set it by doing:: 861 862 > echo 1 > /proc/irq/10/smp_affinity 863 864This means that only the first CPU will handle the IRQ, but you can also echo 8655 which means that only the first and third CPU can handle the IRQ. 866 867The contents of each smp_affinity file is the same by default:: 868 869 > cat /proc/irq/0/smp_affinity 870 ffffffff 871 872There is an alternate interface, smp_affinity_list which allows specifying 873a CPU range instead of a bitmask:: 874 875 > cat /proc/irq/0/smp_affinity_list 876 1024-1031 877 878The default_smp_affinity mask applies to all non-active IRQs, which are the 879IRQs which have not yet been allocated/activated, and hence which lack a 880/proc/irq/[0-9]* directory. 881 882The node file on an SMP system shows the node to which the device using the IRQ 883reports itself as being attached. This hardware locality information does not 884include information about any possible driver locality preference. 885 886prof_cpu_mask specifies which CPUs are to be profiled by the system wide 887profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 888 889The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 890between all the CPUs which are allowed to handle it. As usual the kernel has 891more info than you and does a better job than you, so the defaults are the 892best choice for almost everyone. [Note this applies only to those IO-APIC's 893that support "Round Robin" interrupt distribution.] 894 895There are three more important subdirectories in /proc: net, scsi, and sys. 896The general rule is that the contents, or even the existence of these 897directories, depend on your kernel configuration. If SCSI is not enabled, the 898directory scsi may not exist. The same is true with the net, which is there 899only when networking support is present in the running kernel. 900 901The slabinfo file gives information about memory usage at the slab level. 902Linux uses slab pools for memory management above page level in version 2.2. 903Commonly used objects have their own slab pool (such as network buffers, 904directory cache, and so on). 905 906:: 907 908 > cat /proc/buddyinfo 909 910 Node 0, zone DMA 0 4 5 4 4 3 ... 911 Node 0, zone Normal 1 0 0 1 101 8 ... 912 Node 0, zone HighMem 2 0 0 1 1 0 ... 913 914External fragmentation is a problem under some workloads, and buddyinfo is a 915useful tool for helping diagnose these problems. Buddyinfo will give you a 916clue as to how big an area you can safely allocate, or why a previous 917allocation failed. 918 919Each column represents the number of pages of a certain order which are 920available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 921ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 922available in ZONE_NORMAL, etc... 923 924More information relevant to external fragmentation can be found in 925pagetypeinfo:: 926 927 > cat /proc/pagetypeinfo 928 Page block order: 9 929 Pages per block: 512 930 931 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 932 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 933 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 934 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 935 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 936 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 937 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 938 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 939 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 940 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 941 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 942 943 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 944 Node 0, zone DMA 2 0 5 1 0 945 Node 0, zone DMA32 41 6 967 2 0 946 947Fragmentation avoidance in the kernel works by grouping pages of different 948migrate types into the same contiguous regions of memory called page blocks. 949A page block is typically the size of the default hugepage size, e.g. 2MB on 950X86-64. By keeping pages grouped based on their ability to move, the kernel 951can reclaim pages within a page block to satisfy a high-order allocation. 952 953The pagetypinfo begins with information on the size of a page block. It 954then gives the same type of information as buddyinfo except broken down 955by migrate-type and finishes with details on how many page blocks of each 956type exist. 957 958If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 959from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 960make an estimate of the likely number of huge pages that can be allocated 961at a given point in time. All the "Movable" blocks should be allocatable 962unless memory has been mlock()'d. Some of the Reclaimable blocks should 963also be allocatable although a lot of filesystem metadata may have to be 964reclaimed to achieve this. 965 966 967allocinfo 968~~~~~~~~~ 969 970Provides information about memory allocations at all locations in the code 971base. Each allocation in the code is identified by its source file, line 972number, module (if originates from a loadable module) and the function calling 973the allocation. The number of bytes allocated and number of calls at each 974location are reported. The first line indicates the version of the file, the 975second line is the header listing fields in the file. 976 977Example output. 978 979:: 980 981 > tail -n +3 /proc/allocinfo | sort -rn 982 127664128 31168 mm/page_ext.c:270 func:alloc_page_ext 983 56373248 4737 mm/slub.c:2259 func:alloc_slab_page 984 14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded 985 14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash 986 13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs 987 11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio 988 9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node 989 4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable 990 4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start 991 3940352 962 mm/memory.c:4214 func:alloc_anon_folio 992 2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node 993 ... 994 995 996meminfo 997~~~~~~~ 998 999Provides information about distribution and utilization of memory. This 1000varies by architecture and compile options. Some of the counters reported 1001here overlap. The memory reported by the non overlapping counters may not 1002add up to the overall memory usage and the difference for some workloads 1003can be substantial. In many cases there are other means to find out 1004additional memory using subsystem specific interfaces, for instance 1005/proc/net/sockstat for TCP memory allocations. 1006 1007Example output. You may not have all of these fields. 1008 1009:: 1010 1011 > cat /proc/meminfo 1012 1013 MemTotal: 32858820 kB 1014 MemFree: 21001236 kB 1015 MemAvailable: 27214312 kB 1016 Buffers: 581092 kB 1017 Cached: 5587612 kB 1018 SwapCached: 0 kB 1019 Active: 3237152 kB 1020 Inactive: 7586256 kB 1021 Active(anon): 94064 kB 1022 Inactive(anon): 4570616 kB 1023 Active(file): 3143088 kB 1024 Inactive(file): 3015640 kB 1025 Unevictable: 0 kB 1026 Mlocked: 0 kB 1027 SwapTotal: 0 kB 1028 SwapFree: 0 kB 1029 Zswap: 1904 kB 1030 Zswapped: 7792 kB 1031 Dirty: 12 kB 1032 Writeback: 0 kB 1033 AnonPages: 4654780 kB 1034 Mapped: 266244 kB 1035 Shmem: 9976 kB 1036 KReclaimable: 517708 kB 1037 Slab: 660044 kB 1038 SReclaimable: 517708 kB 1039 SUnreclaim: 142336 kB 1040 KernelStack: 11168 kB 1041 PageTables: 20540 kB 1042 SecPageTables: 0 kB 1043 NFS_Unstable: 0 kB 1044 Bounce: 0 kB 1045 WritebackTmp: 0 kB 1046 CommitLimit: 16429408 kB 1047 Committed_AS: 7715148 kB 1048 VmallocTotal: 34359738367 kB 1049 VmallocUsed: 40444 kB 1050 VmallocChunk: 0 kB 1051 Percpu: 29312 kB 1052 EarlyMemtestBad: 0 kB 1053 HardwareCorrupted: 0 kB 1054 AnonHugePages: 4149248 kB 1055 ShmemHugePages: 0 kB 1056 ShmemPmdMapped: 0 kB 1057 FileHugePages: 0 kB 1058 FilePmdMapped: 0 kB 1059 CmaTotal: 0 kB 1060 CmaFree: 0 kB 1061 HugePages_Total: 0 1062 HugePages_Free: 0 1063 HugePages_Rsvd: 0 1064 HugePages_Surp: 0 1065 Hugepagesize: 2048 kB 1066 Hugetlb: 0 kB 1067 DirectMap4k: 401152 kB 1068 DirectMap2M: 10008576 kB 1069 DirectMap1G: 24117248 kB 1070 1071MemTotal 1072 Total usable RAM (i.e. physical RAM minus a few reserved 1073 bits and the kernel binary code) 1074MemFree 1075 Total free RAM. On highmem systems, the sum of LowFree+HighFree 1076MemAvailable 1077 An estimate of how much memory is available for starting new 1078 applications, without swapping. Calculated from MemFree, 1079 SReclaimable, the size of the file LRU lists, and the low 1080 watermarks in each zone. 1081 The estimate takes into account that the system needs some 1082 page cache to function well, and that not all reclaimable 1083 slab will be reclaimable, due to items being in use. The 1084 impact of those factors will vary from system to system. 1085Buffers 1086 Relatively temporary storage for raw disk blocks 1087 shouldn't get tremendously large (20MB or so) 1088Cached 1089 In-memory cache for files read from the disk (the 1090 pagecache) as well as tmpfs & shmem. 1091 Doesn't include SwapCached. 1092SwapCached 1093 Memory that once was swapped out, is swapped back in but 1094 still also is in the swapfile (if memory is needed it 1095 doesn't need to be swapped out AGAIN because it is already 1096 in the swapfile. This saves I/O) 1097Active 1098 Memory that has been used more recently and usually not 1099 reclaimed unless absolutely necessary. 1100Inactive 1101 Memory which has been less recently used. It is more 1102 eligible to be reclaimed for other purposes 1103Unevictable 1104 Memory allocated for userspace which cannot be reclaimed, such 1105 as mlocked pages, ramfs backing pages, secret memfd pages etc. 1106Mlocked 1107 Memory locked with mlock(). 1108HighTotal, HighFree 1109 Highmem is all memory above ~860MB of physical memory. 1110 Highmem areas are for use by userspace programs, or 1111 for the pagecache. The kernel must use tricks to access 1112 this memory, making it slower to access than lowmem. 1113LowTotal, LowFree 1114 Lowmem is memory which can be used for everything that 1115 highmem can be used for, but it is also available for the 1116 kernel's use for its own data structures. Among many 1117 other things, it is where everything from the Slab is 1118 allocated. Bad things happen when you're out of lowmem. 1119SwapTotal 1120 total amount of swap space available 1121SwapFree 1122 Memory which has been evicted from RAM, and is temporarily 1123 on the disk 1124Zswap 1125 Memory consumed by the zswap backend (compressed size) 1126Zswapped 1127 Amount of anonymous memory stored in zswap (original size) 1128Dirty 1129 Memory which is waiting to get written back to the disk 1130Writeback 1131 Memory which is actively being written back to the disk 1132AnonPages 1133 Non-file backed pages mapped into userspace page tables 1134Mapped 1135 files which have been mmapped, such as libraries 1136Shmem 1137 Total memory used by shared memory (shmem) and tmpfs 1138KReclaimable 1139 Kernel allocations that the kernel will attempt to reclaim 1140 under memory pressure. Includes SReclaimable (below), and other 1141 direct allocations with a shrinker. 1142Slab 1143 in-kernel data structures cache 1144SReclaimable 1145 Part of Slab, that might be reclaimed, such as caches 1146SUnreclaim 1147 Part of Slab, that cannot be reclaimed on memory pressure 1148KernelStack 1149 Memory consumed by the kernel stacks of all tasks 1150PageTables 1151 Memory consumed by userspace page tables 1152SecPageTables 1153 Memory consumed by secondary page tables, this currently includes 1154 KVM mmu and IOMMU allocations on x86 and arm64. 1155NFS_Unstable 1156 Always zero. Previous counted pages which had been written to 1157 the server, but has not been committed to stable storage. 1158Bounce 1159 Memory used for block device "bounce buffers" 1160WritebackTmp 1161 Memory used by FUSE for temporary writeback buffers 1162CommitLimit 1163 Based on the overcommit ratio ('vm.overcommit_ratio'), 1164 this is the total amount of memory currently available to 1165 be allocated on the system. This limit is only adhered to 1166 if strict overcommit accounting is enabled (mode 2 in 1167 'vm.overcommit_memory'). 1168 1169 The CommitLimit is calculated with the following formula:: 1170 1171 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1172 overcommit_ratio / 100 + [total swap pages] 1173 1174 For example, on a system with 1G of physical RAM and 7G 1175 of swap with a `vm.overcommit_ratio` of 30 it would 1176 yield a CommitLimit of 7.3G. 1177 1178 For more details, see the memory overcommit documentation 1179 in mm/overcommit-accounting. 1180Committed_AS 1181 The amount of memory presently allocated on the system. 1182 The committed memory is a sum of all of the memory which 1183 has been allocated by processes, even if it has not been 1184 "used" by them as of yet. A process which malloc()'s 1G 1185 of memory, but only touches 300M of it will show up as 1186 using 1G. This 1G is memory which has been "committed" to 1187 by the VM and can be used at any time by the allocating 1188 application. With strict overcommit enabled on the system 1189 (mode 2 in 'vm.overcommit_memory'), allocations which would 1190 exceed the CommitLimit (detailed above) will not be permitted. 1191 This is useful if one needs to guarantee that processes will 1192 not fail due to lack of memory once that memory has been 1193 successfully allocated. 1194VmallocTotal 1195 total size of vmalloc virtual address space 1196VmallocUsed 1197 amount of vmalloc area which is used 1198VmallocChunk 1199 largest contiguous block of vmalloc area which is free 1200Percpu 1201 Memory allocated to the percpu allocator used to back percpu 1202 allocations. This stat excludes the cost of metadata. 1203EarlyMemtestBad 1204 The amount of RAM/memory in kB, that was identified as corrupted 1205 by early memtest. If memtest was not run, this field will not 1206 be displayed at all. Size is never rounded down to 0 kB. 1207 That means if 0 kB is reported, you can safely assume 1208 there was at least one pass of memtest and none of the passes 1209 found a single faulty byte of RAM. 1210HardwareCorrupted 1211 The amount of RAM/memory in KB, the kernel identifies as 1212 corrupted. 1213AnonHugePages 1214 Non-file backed huge pages mapped into userspace page tables 1215ShmemHugePages 1216 Memory used by shared memory (shmem) and tmpfs allocated 1217 with huge pages 1218ShmemPmdMapped 1219 Shared memory mapped into userspace with huge pages 1220FileHugePages 1221 Memory used for filesystem data (page cache) allocated 1222 with huge pages 1223FilePmdMapped 1224 Page cache mapped into userspace with huge pages 1225CmaTotal 1226 Memory reserved for the Contiguous Memory Allocator (CMA) 1227CmaFree 1228 Free remaining memory in the CMA reserves 1229HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb 1230 See Documentation/admin-guide/mm/hugetlbpage.rst. 1231DirectMap4k, DirectMap2M, DirectMap1G 1232 Breakdown of page table sizes used in the kernel's 1233 identity mapping of RAM 1234 1235vmallocinfo 1236~~~~~~~~~~~ 1237 1238Provides information about vmalloced/vmaped areas. One line per area, 1239containing the virtual address range of the area, size in bytes, 1240caller information of the creator, and optional information depending 1241on the kind of area: 1242 1243 ========== =================================================== 1244 pages=nr number of pages 1245 phys=addr if a physical address was specified 1246 ioremap I/O mapping (ioremap() and friends) 1247 vmalloc vmalloc() area 1248 vmap vmap()ed pages 1249 user VM_USERMAP area 1250 vpages buffer for pages pointers was vmalloced (huge area) 1251 N<node>=nr (Only on NUMA kernels) 1252 Number of pages allocated on memory node <node> 1253 ========== =================================================== 1254 1255:: 1256 1257 > cat /proc/vmallocinfo 1258 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1259 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1260 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1261 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1262 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1263 phys=7fee8000 ioremap 1264 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1265 phys=7fee7000 ioremap 1266 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1267 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1268 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1269 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1270 pages=2 vmalloc N1=2 1271 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1272 /0x130 [x_tables] pages=4 vmalloc N0=4 1273 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1274 pages=14 vmalloc N2=14 1275 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1276 pages=4 vmalloc N1=4 1277 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1278 pages=2 vmalloc N1=2 1279 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1280 pages=10 vmalloc N0=10 1281 1282 1283softirqs 1284~~~~~~~~ 1285 1286Provides counts of softirq handlers serviced since boot time, for each CPU. 1287 1288:: 1289 1290 > cat /proc/softirqs 1291 CPU0 CPU1 CPU2 CPU3 1292 HI: 0 0 0 0 1293 TIMER: 27166 27120 27097 27034 1294 NET_TX: 0 0 0 17 1295 NET_RX: 42 0 0 39 1296 BLOCK: 0 0 107 1121 1297 TASKLET: 0 0 0 290 1298 SCHED: 27035 26983 26971 26746 1299 HRTIMER: 0 0 0 0 1300 RCU: 1678 1769 2178 2250 1301 13021.3 Networking info in /proc/net 1303-------------------------------- 1304 1305The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1306additional values you get for IP version 6 if you configure the kernel to 1307support this. Table 1-9 lists the files and their meaning. 1308 1309 1310.. table:: Table 1-8: IPv6 info in /proc/net 1311 1312 ========== ===================================================== 1313 File Content 1314 ========== ===================================================== 1315 udp6 UDP sockets (IPv6) 1316 tcp6 TCP sockets (IPv6) 1317 raw6 Raw device statistics (IPv6) 1318 igmp6 IP multicast addresses, which this host joined (IPv6) 1319 if_inet6 List of IPv6 interface addresses 1320 ipv6_route Kernel routing table for IPv6 1321 rt6_stats Global IPv6 routing tables statistics 1322 sockstat6 Socket statistics (IPv6) 1323 snmp6 Snmp data (IPv6) 1324 ========== ===================================================== 1325 1326.. table:: Table 1-9: Network info in /proc/net 1327 1328 ============= ================================================================ 1329 File Content 1330 ============= ================================================================ 1331 arp Kernel ARP table 1332 dev network devices with statistics 1333 dev_mcast the Layer2 multicast groups a device is listening too 1334 (interface index, label, number of references, number of bound 1335 addresses). 1336 dev_stat network device status 1337 ip_fwchains Firewall chain linkage 1338 ip_fwnames Firewall chain names 1339 ip_masq Directory containing the masquerading tables 1340 ip_masquerade Major masquerading table 1341 netstat Network statistics 1342 raw raw device statistics 1343 route Kernel routing table 1344 rpc Directory containing rpc info 1345 rt_cache Routing cache 1346 snmp SNMP data 1347 sockstat Socket statistics 1348 softnet_stat Per-CPU incoming packets queues statistics of online CPUs 1349 tcp TCP sockets 1350 udp UDP sockets 1351 unix UNIX domain sockets 1352 wireless Wireless interface data (Wavelan etc) 1353 igmp IP multicast addresses, which this host joined 1354 psched Global packet scheduler parameters. 1355 netlink List of PF_NETLINK sockets 1356 ip_mr_vifs List of multicast virtual interfaces 1357 ip_mr_cache List of multicast routing cache 1358 ============= ================================================================ 1359 1360You can use this information to see which network devices are available in 1361your system and how much traffic was routed over those devices:: 1362 1363 > cat /proc/net/dev 1364 Inter-|Receive |[... 1365 face |bytes packets errs drop fifo frame compressed multicast|[... 1366 lo: 908188 5596 0 0 0 0 0 0 [... 1367 ppp0:15475140 20721 410 0 0 410 0 0 [... 1368 eth0: 614530 7085 0 0 0 0 0 1 [... 1369 1370 ...] Transmit 1371 ...] bytes packets errs drop fifo colls carrier compressed 1372 ...] 908188 5596 0 0 0 0 0 0 1373 ...] 1375103 17405 0 0 0 0 0 0 1374 ...] 1703981 5535 0 0 0 3 0 0 1375 1376In addition, each Channel Bond interface has its own directory. For 1377example, the bond0 device will have a directory called /proc/net/bond0/. 1378It will contain information that is specific to that bond, such as the 1379current slaves of the bond, the link status of the slaves, and how 1380many times the slaves link has failed. 1381 13821.4 SCSI info 1383------------- 1384 1385If you have a SCSI or ATA host adapter in your system, you'll find a 1386subdirectory named after the driver for this adapter in /proc/scsi. 1387You'll also see a list of all recognized SCSI devices in /proc/scsi:: 1388 1389 >cat /proc/scsi/scsi 1390 Attached devices: 1391 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1392 Vendor: IBM Model: DGHS09U Rev: 03E0 1393 Type: Direct-Access ANSI SCSI revision: 03 1394 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1395 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1396 Type: CD-ROM ANSI SCSI revision: 02 1397 1398 1399The directory named after the driver has one file for each adapter found in 1400the system. These files contain information about the controller, including 1401the used IRQ and the IO address range. The amount of information shown is 1402dependent on the adapter you use. The example shows the output for an Adaptec 1403AHA-2940 SCSI adapter:: 1404 1405 > cat /proc/scsi/aic7xxx/0 1406 1407 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1408 Compile Options: 1409 TCQ Enabled By Default : Disabled 1410 AIC7XXX_PROC_STATS : Disabled 1411 AIC7XXX_RESET_DELAY : 5 1412 Adapter Configuration: 1413 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1414 Ultra Wide Controller 1415 PCI MMAPed I/O Base: 0xeb001000 1416 Adapter SEEPROM Config: SEEPROM found and used. 1417 Adaptec SCSI BIOS: Enabled 1418 IRQ: 10 1419 SCBs: Active 0, Max Active 2, 1420 Allocated 15, HW 16, Page 255 1421 Interrupts: 160328 1422 BIOS Control Word: 0x18b6 1423 Adapter Control Word: 0x005b 1424 Extended Translation: Enabled 1425 Disconnect Enable Flags: 0xffff 1426 Ultra Enable Flags: 0x0001 1427 Tag Queue Enable Flags: 0x0000 1428 Ordered Queue Tag Flags: 0x0000 1429 Default Tag Queue Depth: 8 1430 Tagged Queue By Device array for aic7xxx host instance 0: 1431 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1432 Actual queue depth per device for aic7xxx host instance 0: 1433 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1434 Statistics: 1435 (scsi0:0:0:0) 1436 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1437 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1438 Total transfers 160151 (74577 reads and 85574 writes) 1439 (scsi0:0:6:0) 1440 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1441 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1442 Total transfers 0 (0 reads and 0 writes) 1443 1444 14451.5 Parallel port info in /proc/parport 1446--------------------------------------- 1447 1448The directory /proc/parport contains information about the parallel ports of 1449your system. It has one subdirectory for each port, named after the port 1450number (0,1,2,...). 1451 1452These directories contain the four files shown in Table 1-10. 1453 1454 1455.. table:: Table 1-10: Files in /proc/parport 1456 1457 ========= ==================================================================== 1458 File Content 1459 ========= ==================================================================== 1460 autoprobe Any IEEE-1284 device ID information that has been acquired. 1461 devices list of the device drivers using that port. A + will appear by the 1462 name of the device currently using the port (it might not appear 1463 against any). 1464 hardware Parallel port's base address, IRQ line and DMA channel. 1465 irq IRQ that parport is using for that port. This is in a separate 1466 file to allow you to alter it by writing a new value in (IRQ 1467 number or none). 1468 ========= ==================================================================== 1469 14701.6 TTY info in /proc/tty 1471------------------------- 1472 1473Information about the available and actually used tty's can be found in the 1474directory /proc/tty. You'll find entries for drivers and line disciplines in 1475this directory, as shown in Table 1-11. 1476 1477 1478.. table:: Table 1-11: Files in /proc/tty 1479 1480 ============= ============================================== 1481 File Content 1482 ============= ============================================== 1483 drivers list of drivers and their usage 1484 ldiscs registered line disciplines 1485 driver/serial usage statistic and status of single tty lines 1486 ============= ============================================== 1487 1488To see which tty's are currently in use, you can simply look into the file 1489/proc/tty/drivers:: 1490 1491 > cat /proc/tty/drivers 1492 pty_slave /dev/pts 136 0-255 pty:slave 1493 pty_master /dev/ptm 128 0-255 pty:master 1494 pty_slave /dev/ttyp 3 0-255 pty:slave 1495 pty_master /dev/pty 2 0-255 pty:master 1496 serial /dev/cua 5 64-67 serial:callout 1497 serial /dev/ttyS 4 64-67 serial 1498 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1499 /dev/ptmx /dev/ptmx 5 2 system 1500 /dev/console /dev/console 5 1 system:console 1501 /dev/tty /dev/tty 5 0 system:/dev/tty 1502 unknown /dev/tty 4 1-63 console 1503 1504 15051.7 Miscellaneous kernel statistics in /proc/stat 1506------------------------------------------------- 1507 1508Various pieces of information about kernel activity are available in the 1509/proc/stat file. All of the numbers reported in this file are aggregates 1510since the system first booted. For a quick look, simply cat the file:: 1511 1512 > cat /proc/stat 1513 cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0 1514 cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0 1515 cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0 1516 cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0 1517 cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0 1518 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> 1519 ctxt 22848221062 1520 btime 1605316999 1521 processes 746787147 1522 procs_running 2 1523 procs_blocked 0 1524 softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354 1525 1526The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1527lines. These numbers identify the amount of time the CPU has spent performing 1528different kinds of work. Time units are in USER_HZ (typically hundredths of a 1529second). The meanings of the columns are as follows, from left to right: 1530 1531- user: normal processes executing in user mode 1532- nice: niced processes executing in user mode 1533- system: processes executing in kernel mode 1534- idle: twiddling thumbs 1535- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1536 are several problems: 1537 1538 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1539 waiting for I/O to complete. When CPU goes into idle state for 1540 outstanding task I/O, another task will be scheduled on this CPU. 1541 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1542 on any CPU, so the iowait of each CPU is difficult to calculate. 1543 3. The value of iowait field in /proc/stat will decrease in certain 1544 conditions. 1545 1546 So, the iowait is not reliable by reading from /proc/stat. 1547- irq: servicing interrupts 1548- softirq: servicing softirqs 1549- steal: involuntary wait 1550- guest: running a normal guest 1551- guest_nice: running a niced guest 1552 1553The "intr" line gives counts of interrupts serviced since boot time, for each 1554of the possible system interrupts. The first column is the total of all 1555interrupts serviced including unnumbered architecture specific interrupts; 1556each subsequent column is the total for that particular numbered interrupt. 1557Unnumbered interrupts are not shown, only summed into the total. 1558 1559The "ctxt" line gives the total number of context switches across all CPUs. 1560 1561The "btime" line gives the time at which the system booted, in seconds since 1562the Unix epoch. 1563 1564The "processes" line gives the number of processes and threads created, which 1565includes (but is not limited to) those created by calls to the fork() and 1566clone() system calls. 1567 1568The "procs_running" line gives the total number of threads that are 1569running or ready to run (i.e., the total number of runnable threads). 1570 1571The "procs_blocked" line gives the number of processes currently blocked, 1572waiting for I/O to complete. 1573 1574The "softirq" line gives counts of softirqs serviced since boot time, for each 1575of the possible system softirqs. The first column is the total of all 1576softirqs serviced; each subsequent column is the total for that particular 1577softirq. 1578 1579 15801.8 Ext4 file system parameters 1581------------------------------- 1582 1583Information about mounted ext4 file systems can be found in 1584/proc/fs/ext4. Each mounted filesystem will have a directory in 1585/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1586/proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device 1587directory are shown in Table 1-12, below. 1588 1589.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1590 1591 ============== ========================================================== 1592 File Content 1593 mb_groups details of multiblock allocator buddy cache of free blocks 1594 ============== ========================================================== 1595 15961.9 /proc/consoles 1597------------------- 1598Shows registered system console lines. 1599 1600To see which character device lines are currently used for the system console 1601/dev/console, you may simply look into the file /proc/consoles:: 1602 1603 > cat /proc/consoles 1604 tty0 -WU (ECp) 4:7 1605 ttyS0 -W- (Ep) 4:64 1606 1607The columns are: 1608 1609+--------------------+-------------------------------------------------------+ 1610| device | name of the device | 1611+====================+=======================================================+ 1612| operations | * R = can do read operations | 1613| | * W = can do write operations | 1614| | * U = can do unblank | 1615+--------------------+-------------------------------------------------------+ 1616| flags | * E = it is enabled | 1617| | * C = it is preferred console | 1618| | * B = it is primary boot console | 1619| | * p = it is used for printk buffer | 1620| | * b = it is not a TTY but a Braille device | 1621| | * a = it is safe to use when cpu is offline | 1622+--------------------+-------------------------------------------------------+ 1623| major:minor | major and minor number of the device separated by a | 1624| | colon | 1625+--------------------+-------------------------------------------------------+ 1626 1627Summary 1628------- 1629 1630The /proc file system serves information about the running system. It not only 1631allows access to process data but also allows you to request the kernel status 1632by reading files in the hierarchy. 1633 1634The directory structure of /proc reflects the types of information and makes 1635it easy, if not obvious, where to look for specific data. 1636 1637Chapter 2: Modifying System Parameters 1638====================================== 1639 1640In This Chapter 1641--------------- 1642 1643* Modifying kernel parameters by writing into files found in /proc/sys 1644* Exploring the files which modify certain parameters 1645* Review of the /proc/sys file tree 1646 1647------------------------------------------------------------------------------ 1648 1649A very interesting part of /proc is the directory /proc/sys. This is not only 1650a source of information, it also allows you to change parameters within the 1651kernel. Be very careful when attempting this. You can optimize your system, 1652but you can also cause it to crash. Never alter kernel parameters on a 1653production system. Set up a development machine and test to make sure that 1654everything works the way you want it to. You may have no alternative but to 1655reboot the machine once an error has been made. 1656 1657To change a value, simply echo the new value into the file. 1658You need to be root to do this. You can create your own boot script 1659to perform this every time your system boots. 1660 1661The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1662general things in the operation of the Linux kernel. Since some of the files 1663can inadvertently disrupt your system, it is advisable to read both 1664documentation and source before actually making adjustments. In any case, be 1665very careful when writing to any of these files. The entries in /proc may 1666change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1667review the kernel documentation in the directory linux/Documentation. 1668This chapter is heavily based on the documentation included in the pre 2.2 1669kernels, and became part of it in version 2.2.1 of the Linux kernel. 1670 1671Please see: Documentation/admin-guide/sysctl/ directory for descriptions of 1672these entries. 1673 1674Summary 1675------- 1676 1677Certain aspects of kernel behavior can be modified at runtime, without the 1678need to recompile the kernel, or even to reboot the system. The files in the 1679/proc/sys tree can not only be read, but also modified. You can use the echo 1680command to write value into these files, thereby changing the default settings 1681of the kernel. 1682 1683 1684Chapter 3: Per-process Parameters 1685================================= 1686 16873.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1688-------------------------------------------------------------------------------- 1689 1690These files can be used to adjust the badness heuristic used to select which 1691process gets killed in out of memory (oom) conditions. 1692 1693The badness heuristic assigns a value to each candidate task ranging from 0 1694(never kill) to 1000 (always kill) to determine which process is targeted. The 1695units are roughly a proportion along that range of allowed memory the process 1696may allocate from based on an estimation of its current memory and swap use. 1697For example, if a task is using all allowed memory, its badness score will be 16981000. If it is using half of its allowed memory, its score will be 500. 1699 1700The amount of "allowed" memory depends on the context in which the oom killer 1701was called. If it is due to the memory assigned to the allocating task's cpuset 1702being exhausted, the allowed memory represents the set of mems assigned to that 1703cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1704memory represents the set of mempolicy nodes. If it is due to a memory 1705limit (or swap limit) being reached, the allowed memory is that configured 1706limit. Finally, if it is due to the entire system being out of memory, the 1707allowed memory represents all allocatable resources. 1708 1709The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1710is used to determine which task to kill. Acceptable values range from -1000 1711(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1712polarize the preference for oom killing either by always preferring a certain 1713task or completely disabling it. The lowest possible value, -1000, is 1714equivalent to disabling oom killing entirely for that task since it will always 1715report a badness score of 0. 1716 1717Consequently, it is very simple for userspace to define the amount of memory to 1718consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1719example, is roughly equivalent to allowing the remainder of tasks sharing the 1720same system, cpuset, mempolicy, or memory controller resources to use at least 172150% more memory. A value of -500, on the other hand, would be roughly 1722equivalent to discounting 50% of the task's allowed memory from being considered 1723as scoring against the task. 1724 1725For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1726be used to tune the badness score. Its acceptable values range from -16 1727(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1728(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1729scaled linearly with /proc/<pid>/oom_score_adj. 1730 1731The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1732value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1733requires CAP_SYS_RESOURCE. 1734 1735 17363.2 /proc/<pid>/oom_score - Display current oom-killer score 1737------------------------------------------------------------- 1738 1739This file can be used to check the current score used by the oom-killer for 1740any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1741process should be killed in an out-of-memory situation. 1742 1743Please note that the exported value includes oom_score_adj so it is 1744effectively in range [0,2000]. 1745 1746 17473.3 /proc/<pid>/io - Display the IO accounting fields 1748------------------------------------------------------- 1749 1750This file contains IO statistics for each running process. 1751 1752Example 1753~~~~~~~ 1754 1755:: 1756 1757 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1758 [1] 3828 1759 1760 test:/tmp # cat /proc/3828/io 1761 rchar: 323934931 1762 wchar: 323929600 1763 syscr: 632687 1764 syscw: 632675 1765 read_bytes: 0 1766 write_bytes: 323932160 1767 cancelled_write_bytes: 0 1768 1769 1770Description 1771~~~~~~~~~~~ 1772 1773rchar 1774^^^^^ 1775 1776I/O counter: chars read 1777The number of bytes which this task has caused to be read from storage. This 1778is simply the sum of bytes which this process passed to read() and pread(). 1779It includes things like tty IO and it is unaffected by whether or not actual 1780physical disk IO was required (the read might have been satisfied from 1781pagecache). 1782 1783 1784wchar 1785^^^^^ 1786 1787I/O counter: chars written 1788The number of bytes which this task has caused, or shall cause to be written 1789to disk. Similar caveats apply here as with rchar. 1790 1791 1792syscr 1793^^^^^ 1794 1795I/O counter: read syscalls 1796Attempt to count the number of read I/O operations, i.e. syscalls like read() 1797and pread(). 1798 1799 1800syscw 1801^^^^^ 1802 1803I/O counter: write syscalls 1804Attempt to count the number of write I/O operations, i.e. syscalls like 1805write() and pwrite(). 1806 1807 1808read_bytes 1809^^^^^^^^^^ 1810 1811I/O counter: bytes read 1812Attempt to count the number of bytes which this process really did cause to 1813be fetched from the storage layer. Done at the submit_bio() level, so it is 1814accurate for block-backed filesystems. <please add status regarding NFS and 1815CIFS at a later time> 1816 1817 1818write_bytes 1819^^^^^^^^^^^ 1820 1821I/O counter: bytes written 1822Attempt to count the number of bytes which this process caused to be sent to 1823the storage layer. This is done at page-dirtying time. 1824 1825 1826cancelled_write_bytes 1827^^^^^^^^^^^^^^^^^^^^^ 1828 1829The big inaccuracy here is truncate. If a process writes 1MB to a file and 1830then deletes the file, it will in fact perform no writeout. But it will have 1831been accounted as having caused 1MB of write. 1832In other words: The number of bytes which this process caused to not happen, 1833by truncating pagecache. A task can cause "negative" IO too. If this task 1834truncates some dirty pagecache, some IO which another task has been accounted 1835for (in its write_bytes) will not be happening. We _could_ just subtract that 1836from the truncating task's write_bytes, but there is information loss in doing 1837that. 1838 1839 1840.. Note:: 1841 1842 At its current implementation state, this is a bit racy on 32-bit machines: 1843 if process A reads process B's /proc/pid/io while process B is updating one 1844 of those 64-bit counters, process A could see an intermediate result. 1845 1846 1847More information about this can be found within the taskstats documentation in 1848Documentation/accounting. 1849 18503.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1851--------------------------------------------------------------- 1852When a process is dumped, all anonymous memory is written to a core file as 1853long as the size of the core file isn't limited. But sometimes we don't want 1854to dump some memory segments, for example, huge shared memory or DAX. 1855Conversely, sometimes we want to save file-backed memory segments into a core 1856file, not only the individual files. 1857 1858/proc/<pid>/coredump_filter allows you to customize which memory segments 1859will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1860of memory types. If a bit of the bitmask is set, memory segments of the 1861corresponding memory type are dumped, otherwise they are not dumped. 1862 1863The following 9 memory types are supported: 1864 1865 - (bit 0) anonymous private memory 1866 - (bit 1) anonymous shared memory 1867 - (bit 2) file-backed private memory 1868 - (bit 3) file-backed shared memory 1869 - (bit 4) ELF header pages in file-backed private memory areas (it is 1870 effective only if the bit 2 is cleared) 1871 - (bit 5) hugetlb private memory 1872 - (bit 6) hugetlb shared memory 1873 - (bit 7) DAX private memory 1874 - (bit 8) DAX shared memory 1875 1876 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1877 are always dumped regardless of the bitmask status. 1878 1879 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1880 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1881 1882The default value of coredump_filter is 0x33; this means all anonymous memory 1883segments, ELF header pages and hugetlb private memory are dumped. 1884 1885If you don't want to dump all shared memory segments attached to pid 1234, 1886write 0x31 to the process's proc file:: 1887 1888 $ echo 0x31 > /proc/1234/coredump_filter 1889 1890When a new process is created, the process inherits the bitmask status from its 1891parent. It is useful to set up coredump_filter before the program runs. 1892For example:: 1893 1894 $ echo 0x7 > /proc/self/coredump_filter 1895 $ ./some_program 1896 18973.5 /proc/<pid>/mountinfo - Information about mounts 1898-------------------------------------------------------- 1899 1900This file contains lines of the form:: 1901 1902 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1903 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4) 1904 1905 (1) mount ID: unique identifier of the mount (may be reused after umount) 1906 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1907 (3) major:minor: value of st_dev for files on filesystem 1908 (4) root: root of the mount within the filesystem 1909 (5) mount point: mount point relative to the process's root 1910 (6) mount options: per mount options 1911 (n…m) optional fields: zero or more fields of the form "tag[:value]" 1912 (m+1) separator: marks the end of the optional fields 1913 (m+2) filesystem type: name of filesystem of the form "type[.subtype]" 1914 (m+3) mount source: filesystem specific information or "none" 1915 (m+4) super options: per super block options 1916 1917Parsers should ignore all unrecognised optional fields. Currently the 1918possible optional fields are: 1919 1920================ ============================================================== 1921shared:X mount is shared in peer group X 1922master:X mount is slave to peer group X 1923propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1924unbindable mount is unbindable 1925================ ============================================================== 1926 1927.. [#] X is the closest dominant peer group under the process's root. If 1928 X is the immediate master of the mount, or if there's no dominant peer 1929 group under the same root, then only the "master:X" field is present 1930 and not the "propagate_from:X" field. 1931 1932For more information on mount propagation see: 1933 1934 Documentation/filesystems/sharedsubtree.rst 1935 1936 19373.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1938-------------------------------------------------------- 1939These files provide a method to access a task's comm value. It also allows for 1940a task to set its own or one of its thread siblings comm value. The comm value 1941is limited in size compared to the cmdline value, so writing anything longer 1942then the kernel's TASK_COMM_LEN (currently 16 chars, including the NUL 1943terminator) will result in a truncated comm value. 1944 1945 19463.7 /proc/<pid>/task/<tid>/children - Information about task children 1947------------------------------------------------------------------------- 1948This file provides a fast way to retrieve first level children pids 1949of a task pointed by <pid>/<tid> pair. The format is a space separated 1950stream of pids. 1951 1952Note the "first level" here -- if a child has its own children they will 1953not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1954to obtain the descendants. 1955 1956Since this interface is intended to be fast and cheap it doesn't 1957guarantee to provide precise results and some children might be 1958skipped, especially if they've exited right after we printed their 1959pids, so one needs to either stop or freeze processes being inspected 1960if precise results are needed. 1961 1962 19633.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1964--------------------------------------------------------------- 1965This file provides information associated with an opened file. The regular 1966files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. 1967The 'pos' represents the current offset of the opened file in decimal 1968form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the 1969file has been created with [see open(2) for details] and 'mnt_id' represents 1970mount ID of the file system containing the opened file [see 3.5 1971/proc/<pid>/mountinfo for details]. 'ino' represents the inode number of 1972the file. 1973 1974A typical output is:: 1975 1976 pos: 0 1977 flags: 0100002 1978 mnt_id: 19 1979 ino: 63107 1980 1981All locks associated with a file descriptor are shown in its fdinfo too:: 1982 1983 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1984 1985The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1986pair provide additional information particular to the objects they represent. 1987 1988Eventfd files 1989~~~~~~~~~~~~~ 1990 1991:: 1992 1993 pos: 0 1994 flags: 04002 1995 mnt_id: 9 1996 ino: 63107 1997 eventfd-count: 5a 1998 1999where 'eventfd-count' is hex value of a counter. 2000 2001Signalfd files 2002~~~~~~~~~~~~~~ 2003 2004:: 2005 2006 pos: 0 2007 flags: 04002 2008 mnt_id: 9 2009 ino: 63107 2010 sigmask: 0000000000000200 2011 2012where 'sigmask' is hex value of the signal mask associated 2013with a file. 2014 2015Epoll files 2016~~~~~~~~~~~ 2017 2018:: 2019 2020 pos: 0 2021 flags: 02 2022 mnt_id: 9 2023 ino: 63107 2024 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 2025 2026where 'tfd' is a target file descriptor number in decimal form, 2027'events' is events mask being watched and the 'data' is data 2028associated with a target [see epoll(7) for more details]. 2029 2030The 'pos' is current offset of the target file in decimal form 2031[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 2032where target file resides, all in hex format. 2033 2034Fsnotify files 2035~~~~~~~~~~~~~~ 2036For inotify files the format is the following:: 2037 2038 pos: 0 2039 flags: 02000000 2040 mnt_id: 9 2041 ino: 63107 2042 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 2043 2044where 'wd' is a watch descriptor in decimal form, i.e. a target file 2045descriptor number, 'ino' and 'sdev' are inode and device where the 2046target file resides and the 'mask' is the mask of events, all in hex 2047form [see inotify(7) for more details]. 2048 2049If the kernel was built with exportfs support, the path to the target 2050file is encoded as a file handle. The file handle is provided by three 2051fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 2052format. 2053 2054If the kernel is built without exportfs support the file handle won't be 2055printed out. 2056 2057If there is no inotify mark attached yet the 'inotify' line will be omitted. 2058 2059For fanotify files the format is:: 2060 2061 pos: 0 2062 flags: 02 2063 mnt_id: 9 2064 ino: 63107 2065 fanotify flags:10 event-flags:0 2066 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 2067 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 2068 2069where fanotify 'flags' and 'event-flags' are values used in fanotify_init 2070call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 2071flags associated with mark which are tracked separately from events 2072mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 2073mask and 'ignored_mask' is the mask of events which are to be ignored. 2074All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2075provide information about flags and mask used in fanotify_mark 2076call [see fsnotify manpage for details]. 2077 2078While the first three lines are mandatory and always printed, the rest is 2079optional and may be omitted if no marks created yet. 2080 2081Timerfd files 2082~~~~~~~~~~~~~ 2083 2084:: 2085 2086 pos: 0 2087 flags: 02 2088 mnt_id: 9 2089 ino: 63107 2090 clockid: 0 2091 ticks: 0 2092 settime flags: 01 2093 it_value: (0, 49406829) 2094 it_interval: (1, 0) 2095 2096where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2097that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2098flags in octal form been used to setup the timer [see timerfd_settime(2) for 2099details]. 'it_value' is remaining time until the timer expiration. 2100'it_interval' is the interval for the timer. Note the timer might be set up 2101with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2102still exhibits timer's remaining time. 2103 2104DMA Buffer files 2105~~~~~~~~~~~~~~~~ 2106 2107:: 2108 2109 pos: 0 2110 flags: 04002 2111 mnt_id: 9 2112 ino: 63107 2113 size: 32768 2114 count: 2 2115 exp_name: system-heap 2116 2117where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of 2118the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. 2119 21203.9 /proc/<pid>/map_files - Information about memory mapped files 2121--------------------------------------------------------------------- 2122This directory contains symbolic links which represent memory mapped files 2123the process is maintaining. Example output:: 2124 2125 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2126 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2127 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2128 | ... 2129 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2130 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2131 2132The name of a link represents the virtual memory bounds of a mapping, i.e. 2133vm_area_struct::vm_start-vm_area_struct::vm_end. 2134 2135The main purpose of the map_files is to retrieve a set of memory mapped 2136files in a fast way instead of parsing /proc/<pid>/maps or 2137/proc/<pid>/smaps, both of which contain many more records. At the same 2138time one can open(2) mappings from the listings of two processes and 2139comparing their inode numbers to figure out which anonymous memory areas 2140are actually shared. 2141 21423.10 /proc/<pid>/timerslack_ns - Task timerslack value 2143--------------------------------------------------------- 2144This file provides the value of the task's timerslack value in nanoseconds. 2145This value specifies an amount of time that normal timers may be deferred 2146in order to coalesce timers and avoid unnecessary wakeups. 2147 2148This allows a task's interactivity vs power consumption tradeoff to be 2149adjusted. 2150 2151Writing 0 to the file will set the task's timerslack to the default value. 2152 2153Valid values are from 0 - ULLONG_MAX 2154 2155An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2156permissions on the task specified to change its timerslack_ns value. 2157 21583.11 /proc/<pid>/patch_state - Livepatch patch operation state 2159----------------------------------------------------------------- 2160When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2161patch state for the task. 2162 2163A value of '-1' indicates that no patch is in transition. 2164 2165A value of '0' indicates that a patch is in transition and the task is 2166unpatched. If the patch is being enabled, then the task hasn't been 2167patched yet. If the patch is being disabled, then the task has already 2168been unpatched. 2169 2170A value of '1' indicates that a patch is in transition and the task is 2171patched. If the patch is being enabled, then the task has already been 2172patched. If the patch is being disabled, then the task hasn't been 2173unpatched yet. 2174 21753.12 /proc/<pid>/arch_status - task architecture specific status 2176------------------------------------------------------------------- 2177When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2178architecture specific status of the task. 2179 2180Example 2181~~~~~~~ 2182 2183:: 2184 2185 $ cat /proc/6753/arch_status 2186 AVX512_elapsed_ms: 8 2187 2188Description 2189~~~~~~~~~~~ 2190 2191x86 specific entries 2192~~~~~~~~~~~~~~~~~~~~~ 2193 2194AVX512_elapsed_ms 2195^^^^^^^^^^^^^^^^^^ 2196 2197 If AVX512 is supported on the machine, this entry shows the milliseconds 2198 elapsed since the last time AVX512 usage was recorded. The recording 2199 happens on a best effort basis when a task is scheduled out. This means 2200 that the value depends on two factors: 2201 2202 1) The time which the task spent on the CPU without being scheduled 2203 out. With CPU isolation and a single runnable task this can take 2204 several seconds. 2205 2206 2) The time since the task was scheduled out last. Depending on the 2207 reason for being scheduled out (time slice exhausted, syscall ...) 2208 this can be arbitrary long time. 2209 2210 As a consequence the value cannot be considered precise and authoritative 2211 information. The application which uses this information has to be aware 2212 of the overall scenario on the system in order to determine whether a 2213 task is a real AVX512 user or not. Precise information can be obtained 2214 with performance counters. 2215 2216 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2217 the task is unlikely an AVX512 user, but depends on the workload and the 2218 scheduling scenario, it also could be a false negative mentioned above. 2219 22203.13 /proc/<pid>/fd - List of symlinks to open files 2221------------------------------------------------------- 2222This directory contains symbolic links which represent open files 2223the process is maintaining. Example output:: 2224 2225 lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null 2226 l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null 2227 lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]' 2228 lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]' 2229 lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]' 2230 2231The number of open files for the process is stored in 'size' member 2232of stat() output for /proc/<pid>/fd for fast access. 2233------------------------------------------------------- 2234 2235 2236Chapter 4: Configuring procfs 2237============================= 2238 22394.1 Mount options 2240--------------------- 2241 2242The following mount options are supported: 2243 2244 ========= ======================================================== 2245 hidepid= Set /proc/<pid>/ access mode. 2246 gid= Set the group authorized to learn processes information. 2247 subset= Show only the specified subset of procfs. 2248 ========= ======================================================== 2249 2250hidepid=off or hidepid=0 means classic mode - everybody may access all 2251/proc/<pid>/ directories (default). 2252 2253hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2254directories but their own. Sensitive files like cmdline, sched*, status are now 2255protected against other users. This makes it impossible to learn whether any 2256user runs specific program (given the program doesn't reveal itself by its 2257behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2258other users, poorly written programs passing sensitive information via program 2259arguments are now protected against local eavesdroppers. 2260 2261hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2262fully invisible to other users. It doesn't mean that it hides a fact whether a 2263process with a specific pid value exists (it can be learned by other means, e.g. 2264by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by 2265stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2266gathering information about running processes, whether some daemon runs with 2267elevated privileges, whether other user runs some sensitive program, whether 2268other users run any program at all, etc. 2269 2270hidepid=ptraceable or hidepid=4 means that procfs should only contain 2271/proc/<pid>/ directories that the caller can ptrace. 2272 2273gid= defines a group authorized to learn processes information otherwise 2274prohibited by hidepid=. If you use some daemon like identd which needs to learn 2275information about processes information, just add identd to this group. 2276 2277subset=pid hides all top level files and directories in the procfs that 2278are not related to tasks. 2279 2280Chapter 5: Filesystem behavior 2281============================== 2282 2283Originally, before the advent of pid namespace, procfs was a global file 2284system. It means that there was only one procfs instance in the system. 2285 2286When pid namespace was added, a separate procfs instance was mounted in 2287each pid namespace. So, procfs mount options are global among all 2288mountpoints within the same namespace:: 2289 2290 # grep ^proc /proc/mounts 2291 proc /proc proc rw,relatime,hidepid=2 0 0 2292 2293 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2294 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2295 +++ exited with 0 +++ 2296 2297 # grep ^proc /proc/mounts 2298 proc /proc proc rw,relatime,hidepid=2 0 0 2299 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2300 2301and only after remounting procfs mount options will change at all 2302mountpoints:: 2303 2304 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2305 2306 # grep ^proc /proc/mounts 2307 proc /proc proc rw,relatime,hidepid=1 0 0 2308 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2309 2310This behavior is different from the behavior of other filesystems. 2311 2312The new procfs behavior is more like other filesystems. Each procfs mount 2313creates a new procfs instance. Mount options affect own procfs instance. 2314It means that it became possible to have several procfs instances 2315displaying tasks with different filtering options in one pid namespace:: 2316 2317 # mount -o hidepid=invisible -t proc proc /proc 2318 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2319 # grep ^proc /proc/mounts 2320 proc /proc proc rw,relatime,hidepid=invisible 0 0 2321 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2322