1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 9 Bodo Bauer <bb@ricochet.net> 102.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 11move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 12fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 13===================== ======================================= ================ 14 15 16 17.. Table of Contents 18 19 0 Preface 20 0.1 Introduction/Credits 21 0.2 Legal Stuff 22 23 1 Collecting System Information 24 1.1 Process-Specific Subdirectories 25 1.2 Kernel data 26 1.3 IDE devices in /proc/ide 27 1.4 Networking info in /proc/net 28 1.5 SCSI info 29 1.6 Parallel port info in /proc/parport 30 1.7 TTY info in /proc/tty 31 1.8 Miscellaneous kernel statistics in /proc/stat 32 1.9 Ext4 file system parameters 33 34 2 Modifying System Parameters 35 36 3 Per-Process Parameters 37 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 38 score 39 3.2 /proc/<pid>/oom_score - Display current oom-killer score 40 3.3 /proc/<pid>/io - Display the IO accounting fields 41 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings 42 3.5 /proc/<pid>/mountinfo - Information about mounts 43 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 44 3.7 /proc/<pid>/task/<tid>/children - Information about task children 45 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 46 3.9 /proc/<pid>/map_files - Information about memory mapped files 47 3.10 /proc/<pid>/timerslack_ns - Task timerslack value 48 3.11 /proc/<pid>/patch_state - Livepatch patch operation state 49 3.12 /proc/<pid>/arch_status - Task architecture specific information 50 3.13 /proc/<pid>/fd - List of symlinks to open files 51 52 4 Configuring procfs 53 4.1 Mount options 54 55 5 Filesystem behavior 56 57Preface 58======= 59 600.1 Introduction/Credits 61------------------------ 62 63This documentation is part of a soon (or so we hope) to be released book on 64the SuSE Linux distribution. As there is no complete documentation for the 65/proc file system and we've used many freely available sources to write these 66chapters, it seems only fair to give the work back to the Linux community. 67This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 68afraid it's still far from complete, but we hope it will be useful. As far as 69we know, it is the first 'all-in-one' document about the /proc file system. It 70is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 71SPARC, AXP, etc., features, you probably won't find what you are looking for. 72It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 73additions and patches are welcome and will be added to this document if you 74mail them to Bodo. 75 76We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 77other people for help compiling this documentation. We'd also like to extend a 78special thank you to Andi Kleen for documentation, which we relied on heavily 79to create this document, as well as the additional information he provided. 80Thanks to everybody else who contributed source or docs to the Linux kernel 81and helped create a great piece of software... :) 82 83If you have any comments, corrections or additions, please don't hesitate to 84contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 85document. 86 87The latest version of this document is available online at 88https://www.kernel.org/doc/html/latest/filesystems/proc.html 89 90If the above direction does not works for you, you could try the kernel 91mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 92comandante@zaralinux.com. 93 940.2 Legal Stuff 95--------------- 96 97We don't guarantee the correctness of this document, and if you come to us 98complaining about how you screwed up your system because of incorrect 99documentation, we won't feel responsible... 100 101Chapter 1: Collecting System Information 102======================================== 103 104In This Chapter 105--------------- 106* Investigating the properties of the pseudo file system /proc and its 107 ability to provide information on the running Linux system 108* Examining /proc's structure 109* Uncovering various information about the kernel and the processes running 110 on the system 111 112------------------------------------------------------------------------------ 113 114The proc file system acts as an interface to internal data structures in the 115kernel. It can be used to obtain information about the system and to change 116certain kernel parameters at runtime (sysctl). 117 118First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 119show you how you can use /proc/sys to change settings. 120 1211.1 Process-Specific Subdirectories 122----------------------------------- 123 124The directory /proc contains (among other things) one subdirectory for each 125process running on the system, which is named after the process ID (PID). 126 127The link 'self' points to the process reading the file system. Each process 128subdirectory has the entries listed in Table 1-1. 129 130Note that an open file descriptor to /proc/<pid> or to any of its 131contained files or subdirectories does not prevent <pid> being reused 132for some other process in the event that <pid> exits. Operations on 133open /proc/<pid> file descriptors corresponding to dead processes 134never act on any new process that the kernel may, through chance, have 135also assigned the process ID <pid>. Instead, operations on these FDs 136usually fail with ESRCH. 137 138.. table:: Table 1-1: Process specific entries in /proc 139 140 ============= =============================================================== 141 File Content 142 ============= =============================================================== 143 clear_refs Clears page referenced bits shown in smaps output 144 cmdline Command line arguments 145 cpu Current and last cpu in which it was executed (2.4)(smp) 146 cwd Link to the current working directory 147 environ Values of environment variables 148 exe Link to the executable of this process 149 fd Directory, which contains all file descriptors 150 maps Memory maps to executables and library files (2.4) 151 mem Memory held by this process 152 root Link to the root directory of this process 153 stat Process status 154 statm Process memory status information 155 status Process status in human readable form 156 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 157 symbol the task is blocked in - or "0" if not blocked. 158 pagemap Page table 159 stack Report full stack trace, enable via CONFIG_STACKTRACE 160 smaps An extension based on maps, showing the memory consumption of 161 each mapping and flags associated with it 162 smaps_rollup Accumulated smaps stats for all mappings of the process. This 163 can be derived from smaps, but is faster and more convenient 164 numa_maps An extension based on maps, showing the memory locality and 165 binding policy as well as mem usage (in pages) of each mapping. 166 ============= =============================================================== 167 168For example, to get the status information of a process, all you have to do is 169read the file /proc/PID/status:: 170 171 >cat /proc/self/status 172 Name: cat 173 State: R (running) 174 Tgid: 5452 175 Pid: 5452 176 PPid: 743 177 TracerPid: 0 (2.4) 178 Uid: 501 501 501 501 179 Gid: 100 100 100 100 180 FDSize: 256 181 Groups: 100 14 16 182 Kthread: 0 183 VmPeak: 5004 kB 184 VmSize: 5004 kB 185 VmLck: 0 kB 186 VmHWM: 476 kB 187 VmRSS: 476 kB 188 RssAnon: 352 kB 189 RssFile: 120 kB 190 RssShmem: 4 kB 191 VmData: 156 kB 192 VmStk: 88 kB 193 VmExe: 68 kB 194 VmLib: 1412 kB 195 VmPTE: 20 kb 196 VmSwap: 0 kB 197 HugetlbPages: 0 kB 198 CoreDumping: 0 199 THP_enabled: 1 200 Threads: 1 201 SigQ: 0/28578 202 SigPnd: 0000000000000000 203 ShdPnd: 0000000000000000 204 SigBlk: 0000000000000000 205 SigIgn: 0000000000000000 206 SigCgt: 0000000000000000 207 CapInh: 00000000fffffeff 208 CapPrm: 0000000000000000 209 CapEff: 0000000000000000 210 CapBnd: ffffffffffffffff 211 CapAmb: 0000000000000000 212 NoNewPrivs: 0 213 Seccomp: 0 214 Speculation_Store_Bypass: thread vulnerable 215 SpeculationIndirectBranch: conditional enabled 216 voluntary_ctxt_switches: 0 217 nonvoluntary_ctxt_switches: 1 218 219This shows you nearly the same information you would get if you viewed it with 220the ps command. In fact, ps uses the proc file system to obtain its 221information. But you get a more detailed view of the process by reading the 222file /proc/PID/status. It fields are described in table 1-2. 223 224The statm file contains more detailed information about the process 225memory usage. Its seven fields are explained in Table 1-3. The stat file 226contains detailed information about the process itself. Its fields are 227explained in Table 1-4. 228 229(for SMP CONFIG users) 230 231For making accounting scalable, RSS related information are handled in an 232asynchronous manner and the value may not be very precise. To see a precise 233snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 234It's slow but very precise. 235 236.. table:: Table 1-2: Contents of the status fields (as of 4.19) 237 238 ========================== =================================================== 239 Field Content 240 ========================== =================================================== 241 Name filename of the executable 242 Umask file mode creation mask 243 State state (R is running, S is sleeping, D is sleeping 244 in an uninterruptible wait, Z is zombie, 245 T is traced or stopped) 246 Tgid thread group ID 247 Ngid NUMA group ID (0 if none) 248 Pid process id 249 PPid process id of the parent process 250 TracerPid PID of process tracing this process (0 if not, or 251 the tracer is outside of the current pid namespace) 252 Uid Real, effective, saved set, and file system UIDs 253 Gid Real, effective, saved set, and file system GIDs 254 FDSize number of file descriptor slots currently allocated 255 Groups supplementary group list 256 NStgid descendant namespace thread group ID hierarchy 257 NSpid descendant namespace process ID hierarchy 258 NSpgid descendant namespace process group ID hierarchy 259 NSsid descendant namespace session ID hierarchy 260 Kthread kernel thread flag, 1 is yes, 0 is no 261 VmPeak peak virtual memory size 262 VmSize total program size 263 VmLck locked memory size 264 VmPin pinned memory size 265 VmHWM peak resident set size ("high water mark") 266 VmRSS size of memory portions. It contains the three 267 following parts 268 (VmRSS = RssAnon + RssFile + RssShmem) 269 RssAnon size of resident anonymous memory 270 RssFile size of resident file mappings 271 RssShmem size of resident shmem memory (includes SysV shm, 272 mapping of tmpfs and shared anonymous mappings) 273 VmData size of private data segments 274 VmStk size of stack segments 275 VmExe size of text segment 276 VmLib size of shared library code 277 VmPTE size of page table entries 278 VmSwap amount of swap used by anonymous private data 279 (shmem swap usage is not included) 280 HugetlbPages size of hugetlb memory portions 281 CoreDumping process's memory is currently being dumped 282 (killing the process may lead to a corrupted core) 283 THP_enabled process is allowed to use THP (returns 0 when 284 PR_SET_THP_DISABLE is set on the process 285 Threads number of threads 286 SigQ number of signals queued/max. number for queue 287 SigPnd bitmap of pending signals for the thread 288 ShdPnd bitmap of shared pending signals for the process 289 SigBlk bitmap of blocked signals 290 SigIgn bitmap of ignored signals 291 SigCgt bitmap of caught signals 292 CapInh bitmap of inheritable capabilities 293 CapPrm bitmap of permitted capabilities 294 CapEff bitmap of effective capabilities 295 CapBnd bitmap of capabilities bounding set 296 CapAmb bitmap of ambient capabilities 297 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) 298 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 299 Speculation_Store_Bypass speculative store bypass mitigation status 300 SpeculationIndirectBranch indirect branch speculation mode 301 Cpus_allowed mask of CPUs on which this process may run 302 Cpus_allowed_list Same as previous, but in "list format" 303 Mems_allowed mask of memory nodes allowed to this process 304 Mems_allowed_list Same as previous, but in "list format" 305 voluntary_ctxt_switches number of voluntary context switches 306 nonvoluntary_ctxt_switches number of non voluntary context switches 307 ========================== =================================================== 308 309 310.. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3) 311 312 ======== =============================== ============================== 313 Field Content 314 ======== =============================== ============================== 315 size total program size (pages) (same as VmSize in status) 316 resident size of memory portions (pages) (same as VmRSS in status) 317 shared number of pages that are shared (i.e. backed by a file, same 318 as RssFile+RssShmem in status) 319 trs number of pages that are 'code' (not including libs; broken, 320 includes data segment) 321 lrs number of pages of library (always 0 on 2.6) 322 drs number of pages of data/stack (including libs; broken, 323 includes library text) 324 dt number of dirty pages (always 0 on 2.6) 325 ======== =============================== ============================== 326 327 328.. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7) 329 330 ============= =============================================================== 331 Field Content 332 ============= =============================================================== 333 pid process id 334 tcomm filename of the executable 335 state state (R is running, S is sleeping, D is sleeping in an 336 uninterruptible wait, Z is zombie, T is traced or stopped) 337 ppid process id of the parent process 338 pgrp pgrp of the process 339 sid session id 340 tty_nr tty the process uses 341 tty_pgrp pgrp of the tty 342 flags task flags 343 min_flt number of minor faults 344 cmin_flt number of minor faults with child's 345 maj_flt number of major faults 346 cmaj_flt number of major faults with child's 347 utime user mode jiffies 348 stime kernel mode jiffies 349 cutime user mode jiffies with child's 350 cstime kernel mode jiffies with child's 351 priority priority level 352 nice nice level 353 num_threads number of threads 354 it_real_value (obsolete, always 0) 355 start_time time the process started after system boot 356 vsize virtual memory size 357 rss resident set memory size 358 rsslim current limit in bytes on the rss 359 start_code address above which program text can run 360 end_code address below which program text can run 361 start_stack address of the start of the main process stack 362 esp current value of ESP 363 eip current value of EIP 364 pending bitmap of pending signals 365 blocked bitmap of blocked signals 366 sigign bitmap of ignored signals 367 sigcatch bitmap of caught signals 368 0 (place holder, used to be the wchan address, 369 use /proc/PID/wchan instead) 370 0 (place holder) 371 0 (place holder) 372 exit_signal signal to send to parent thread on exit 373 task_cpu which CPU the task is scheduled on 374 rt_priority realtime priority 375 policy scheduling policy (man sched_setscheduler) 376 blkio_ticks time spent waiting for block IO 377 gtime guest time of the task in jiffies 378 cgtime guest time of the task children in jiffies 379 start_data address above which program data+bss is placed 380 end_data address below which program data+bss is placed 381 start_brk address above which program heap can be expanded with brk() 382 arg_start address above which program command line is placed 383 arg_end address below which program command line is placed 384 env_start address above which program environment is placed 385 env_end address below which program environment is placed 386 exit_code the thread's exit_code in the form reported by the waitpid 387 system call 388 ============= =============================================================== 389 390The /proc/PID/maps file contains the currently mapped memory regions and 391their access permissions. 392 393The format is:: 394 395 address perms offset dev inode pathname 396 397 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 398 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 399 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] 400 a7cb1000-a7cb2000 ---p 00000000 00:00 0 401 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 402 a7eb2000-a7eb3000 ---p 00000000 00:00 0 403 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 404 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 405 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 406 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 407 a800b000-a800e000 rw-p 00000000 00:00 0 408 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 409 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 410 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 411 a8024000-a8027000 rw-p 00000000 00:00 0 412 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 413 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 414 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 415 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 416 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 417 418where "address" is the address space in the process that it occupies, "perms" 419is a set of permissions:: 420 421 r = read 422 w = write 423 x = execute 424 s = shared 425 p = private (copy on write) 426 427"offset" is the offset into the mapping, "dev" is the device (major:minor), and 428"inode" is the inode on that device. 0 indicates that no inode is associated 429with the memory region, as the case would be with BSS (uninitialized data). 430The "pathname" shows the name associated file for this mapping. If the mapping 431is not associated with a file: 432 433 =================== =========================================== 434 [heap] the heap of the program 435 [stack] the stack of the main process 436 [vdso] the "virtual dynamic shared object", 437 the kernel system call handler 438 [anon:<name>] a private anonymous mapping that has been 439 named by userspace 440 [anon_shmem:<name>] an anonymous shared memory mapping that has 441 been named by userspace 442 =================== =========================================== 443 444 or if empty, the mapping is anonymous. 445 446The /proc/PID/smaps is an extension based on maps, showing the memory 447consumption for each of the process's mappings. For each mapping (aka Virtual 448Memory Area, or VMA) there is a series of lines such as the following:: 449 450 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 451 452 Size: 1084 kB 453 KernelPageSize: 4 kB 454 MMUPageSize: 4 kB 455 Rss: 892 kB 456 Pss: 374 kB 457 Pss_Dirty: 0 kB 458 Shared_Clean: 892 kB 459 Shared_Dirty: 0 kB 460 Private_Clean: 0 kB 461 Private_Dirty: 0 kB 462 Referenced: 892 kB 463 Anonymous: 0 kB 464 LazyFree: 0 kB 465 AnonHugePages: 0 kB 466 ShmemPmdMapped: 0 kB 467 Shared_Hugetlb: 0 kB 468 Private_Hugetlb: 0 kB 469 Swap: 0 kB 470 SwapPss: 0 kB 471 KernelPageSize: 4 kB 472 MMUPageSize: 4 kB 473 Locked: 0 kB 474 THPeligible: 0 475 VmFlags: rd ex mr mw me dw 476 477The first of these lines shows the same information as is displayed for the 478mapping in /proc/PID/maps. Following lines show the size of the mapping 479(size); the size of each page allocated when backing a VMA (KernelPageSize), 480which is usually the same as the size in the page table entries; the page size 481used by the MMU when backing a VMA (in most cases, the same as KernelPageSize); 482the amount of the mapping that is currently resident in RAM (RSS); the 483process' proportional share of this mapping (PSS); and the number of clean and 484dirty shared and private pages in the mapping. 485 486The "proportional set size" (PSS) of a process is the count of pages it has 487in memory, where each page is divided by the number of processes sharing it. 488So if a process has 1000 pages all to itself, and 1000 shared with one other 489process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which 490consists of dirty pages. ("Pss_Clean" is not included, but it can be 491calculated by subtracting "Pss_Dirty" from "Pss".) 492 493Note that even a page which is part of a MAP_SHARED mapping, but has only 494a single pte mapped, i.e. is currently used by only one process, is accounted 495as private and not as shared. 496 497"Referenced" indicates the amount of memory currently marked as referenced or 498accessed. 499 500"Anonymous" shows the amount of memory that does not belong to any file. Even 501a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 502and a page is modified, the file page is replaced by a private anonymous copy. 503 504"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 505The memory isn't freed immediately with madvise(). It's freed in memory 506pressure if the memory is clean. Please note that the printed value might 507be lower than the real value due to optimizations used in the current 508implementation. If this is not desirable please file a bug report. 509 510"AnonHugePages" shows the amount of memory backed by transparent hugepage. 511 512"ShmemPmdMapped" shows the amount of shared (shmem/tmpfs) memory backed by 513huge pages. 514 515"Shared_Hugetlb" and "Private_Hugetlb" show the amounts of memory backed by 516hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 517reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 518 519"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 520 521For shmem mappings, "Swap" includes also the size of the mapped (and not 522replaced by copy-on-write) part of the underlying shmem object out on swap. 523"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 524does not take into account swapped out page of underlying shmem objects. 525"Locked" indicates whether the mapping is locked in memory or not. 526 527"THPeligible" indicates whether the mapping is eligible for allocating THP 528pages as well as the THP is PMD mappable or not - 1 if true, 0 otherwise. 529It just shows the current status. 530 531"VmFlags" field deserves a separate description. This member represents the 532kernel flags associated with the particular virtual memory area in two letter 533encoded manner. The codes are the following: 534 535 == ======================================= 536 rd readable 537 wr writeable 538 ex executable 539 sh shared 540 mr may read 541 mw may write 542 me may execute 543 ms may share 544 gd stack segment growns down 545 pf pure PFN range 546 dw disabled write to the mapped file 547 lo pages are locked in memory 548 io memory mapped I/O area 549 sr sequential read advise provided 550 rr random read advise provided 551 dc do not copy area on fork 552 de do not expand area on remapping 553 ac area is accountable 554 nr swap space is not reserved for the area 555 ht area uses huge tlb pages 556 sf synchronous page fault 557 ar architecture specific flag 558 wf wipe on fork 559 dd do not include area into core dump 560 sd soft dirty flag 561 mm mixed map area 562 hg huge page advise flag 563 nh no huge page advise flag 564 mg mergeable advise flag 565 bt arm64 BTI guarded page 566 mt arm64 MTE allocation tags are enabled 567 um userfaultfd missing tracking 568 uw userfaultfd wr-protect tracking 569 ss shadow stack page 570 == ======================================= 571 572Note that there is no guarantee that every flag and associated mnemonic will 573be present in all further kernel releases. Things get changed, the flags may 574be vanished or the reverse -- new added. Interpretation of their meaning 575might change in future as well. So each consumer of these flags has to 576follow each specific kernel version for the exact semantic. 577 578This file is only present if the CONFIG_MMU kernel configuration option is 579enabled. 580 581Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 582output can be achieved only in the single read call). 583 584This typically manifests when doing partial reads of these files while the 585memory map is being modified. Despite the races, we do provide the following 586guarantees: 587 5881) The mapped addresses never go backwards, which implies no two 589 regions will ever overlap. 5902) If there is something at a given vaddr during the entirety of the 591 life of the smaps/maps walk, there will be some output for it. 592 593The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 594but their values are the sums of the corresponding values for all mappings of 595the process. Additionally, it contains these fields: 596 597- Pss_Anon 598- Pss_File 599- Pss_Shmem 600 601They represent the proportional shares of anonymous, file, and shmem pages, as 602described for smaps above. These fields are omitted in smaps since each 603mapping identifies the type (anon, file, or shmem) of all pages it contains. 604Thus all information in smaps_rollup can be derived from smaps, but at a 605significantly higher cost. 606 607The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 608bits on both physical and virtual pages associated with a process, and the 609soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 610for details). 611To clear the bits for all the pages associated with the process:: 612 613 > echo 1 > /proc/PID/clear_refs 614 615To clear the bits for the anonymous pages associated with the process:: 616 617 > echo 2 > /proc/PID/clear_refs 618 619To clear the bits for the file mapped pages associated with the process:: 620 621 > echo 3 > /proc/PID/clear_refs 622 623To clear the soft-dirty bit:: 624 625 > echo 4 > /proc/PID/clear_refs 626 627To reset the peak resident set size ("high water mark") to the process's 628current value:: 629 630 > echo 5 > /proc/PID/clear_refs 631 632Any other value written to /proc/PID/clear_refs will have no effect. 633 634The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 635using /proc/kpageflags and number of times a page is mapped using 636/proc/kpagecount. For detailed explanation, see 637Documentation/admin-guide/mm/pagemap.rst. 638 639The /proc/pid/numa_maps is an extension based on maps, showing the memory 640locality and binding policy, as well as the memory usage (in pages) of 641each mapping. The output follows a general format where mapping details get 642summarized separated by blank spaces, one mapping per each file line:: 643 644 address policy mapping details 645 646 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 647 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 648 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 649 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 650 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 651 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 652 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 653 320698b000 default file=/lib64/libc-2.12.so 654 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 655 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 656 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 657 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 658 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 659 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 660 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 661 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 662 663Where: 664 665"address" is the starting address for the mapping; 666 667"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 668 669"mapping details" summarizes mapping data such as mapping type, page usage counters, 670node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 671size, in KB, that is backing the mapping up. 672 6731.2 Kernel data 674--------------- 675 676Similar to the process entries, the kernel data files give information about 677the running kernel. The files used to obtain this information are contained in 678/proc and are listed in Table 1-5. Not all of these will be present in your 679system. It depends on the kernel configuration and the loaded modules, which 680files are there, and which are missing. 681 682.. table:: Table 1-5: Kernel info in /proc 683 684 ============ =============================================================== 685 File Content 686 ============ =============================================================== 687 apm Advanced power management info 688 buddyinfo Kernel memory allocator information (see text) (2.5) 689 bus Directory containing bus specific information 690 cmdline Kernel command line 691 cpuinfo Info about the CPU 692 devices Available devices (block and character) 693 dma Used DMS channels 694 filesystems Supported filesystems 695 driver Various drivers grouped here, currently rtc (2.4) 696 execdomains Execdomains, related to security (2.4) 697 fb Frame Buffer devices (2.4) 698 fs File system parameters, currently nfs/exports (2.4) 699 ide Directory containing info about the IDE subsystem 700 interrupts Interrupt usage 701 iomem Memory map (2.4) 702 ioports I/O port usage 703 irq Masks for irq to cpu affinity (2.4)(smp?) 704 isapnp ISA PnP (Plug&Play) Info (2.4) 705 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 706 kmsg Kernel messages 707 ksyms Kernel symbol table 708 loadavg Load average of last 1, 5 & 15 minutes; 709 number of processes currently runnable (running or on ready queue); 710 total number of processes in system; 711 last pid created. 712 All fields are separated by one space except "number of 713 processes currently runnable" and "total number of processes 714 in system", which are separated by a slash ('/'). Example: 715 0.61 0.61 0.55 3/828 22084 716 locks Kernel locks 717 meminfo Memory info 718 misc Miscellaneous 719 modules List of loaded modules 720 mounts Mounted filesystems 721 net Networking info (see text) 722 pagetypeinfo Additional page allocator information (see text) (2.5) 723 partitions Table of partitions known to the system 724 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 725 decoupled by lspci (2.4) 726 rtc Real time clock 727 scsi SCSI info (see text) 728 slabinfo Slab pool info 729 softirqs softirq usage 730 stat Overall statistics 731 swaps Swap space utilization 732 sys See chapter 2 733 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 734 tty Info of tty drivers 735 uptime Wall clock since boot, combined idle time of all cpus 736 version Kernel version 737 video bttv info of video resources (2.4) 738 vmallocinfo Show vmalloced areas 739 ============ =============================================================== 740 741You can, for example, check which interrupts are currently in use and what 742they are used for by looking in the file /proc/interrupts:: 743 744 > cat /proc/interrupts 745 CPU0 746 0: 8728810 XT-PIC timer 747 1: 895 XT-PIC keyboard 748 2: 0 XT-PIC cascade 749 3: 531695 XT-PIC aha152x 750 4: 2014133 XT-PIC serial 751 5: 44401 XT-PIC pcnet_cs 752 8: 2 XT-PIC rtc 753 11: 8 XT-PIC i82365 754 12: 182918 XT-PIC PS/2 Mouse 755 13: 1 XT-PIC fpu 756 14: 1232265 XT-PIC ide0 757 15: 7 XT-PIC ide1 758 NMI: 0 759 760In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 761output of a SMP machine):: 762 763 > cat /proc/interrupts 764 765 CPU0 CPU1 766 0: 1243498 1214548 IO-APIC-edge timer 767 1: 8949 8958 IO-APIC-edge keyboard 768 2: 0 0 XT-PIC cascade 769 5: 11286 10161 IO-APIC-edge soundblaster 770 8: 1 0 IO-APIC-edge rtc 771 9: 27422 27407 IO-APIC-edge 3c503 772 12: 113645 113873 IO-APIC-edge PS/2 Mouse 773 13: 0 0 XT-PIC fpu 774 14: 22491 24012 IO-APIC-edge ide0 775 15: 2183 2415 IO-APIC-edge ide1 776 17: 30564 30414 IO-APIC-level eth0 777 18: 177 164 IO-APIC-level bttv 778 NMI: 2457961 2457959 779 LOC: 2457882 2457881 780 ERR: 2155 781 782NMI is incremented in this case because every timer interrupt generates a NMI 783(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 784 785LOC is the local interrupt counter of the internal APIC of every CPU. 786 787ERR is incremented in the case of errors in the IO-APIC bus (the bus that 788connects the CPUs in a SMP system. This means that an error has been detected, 789the IO-APIC automatically retry the transmission, so it should not be a big 790problem, but you should read the SMP-FAQ. 791 792In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 793/proc/interrupts to display every IRQ vector in use by the system, not 794just those considered 'most important'. The new vectors are: 795 796THR 797 interrupt raised when a machine check threshold counter 798 (typically counting ECC corrected errors of memory or cache) exceeds 799 a configurable threshold. Only available on some systems. 800 801TRM 802 a thermal event interrupt occurs when a temperature threshold 803 has been exceeded for the CPU. This interrupt may also be generated 804 when the temperature drops back to normal. 805 806SPU 807 a spurious interrupt is some interrupt that was raised then lowered 808 by some IO device before it could be fully processed by the APIC. Hence 809 the APIC sees the interrupt but does not know what device it came from. 810 For this case the APIC will generate the interrupt with a IRQ vector 811 of 0xff. This might also be generated by chipset bugs. 812 813RES, CAL, TLB 814 rescheduling, call and TLB flush interrupts are 815 sent from one CPU to another per the needs of the OS. Typically, 816 their statistics are used by kernel developers and interested users to 817 determine the occurrence of interrupts of the given type. 818 819The above IRQ vectors are displayed only when relevant. For example, 820the threshold vector does not exist on x86_64 platforms. Others are 821suppressed when the system is a uniprocessor. As of this writing, only 822i386 and x86_64 platforms support the new IRQ vector displays. 823 824Of some interest is the introduction of the /proc/irq directory to 2.4. 825It could be used to set IRQ to CPU affinity. This means that you can "hook" an 826IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 827irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 828prof_cpu_mask. 829 830For example:: 831 832 > ls /proc/irq/ 833 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 834 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 835 > ls /proc/irq/0/ 836 smp_affinity 837 838smp_affinity is a bitmask, in which you can specify which CPUs can handle the 839IRQ. You can set it by doing:: 840 841 > echo 1 > /proc/irq/10/smp_affinity 842 843This means that only the first CPU will handle the IRQ, but you can also echo 8445 which means that only the first and third CPU can handle the IRQ. 845 846The contents of each smp_affinity file is the same by default:: 847 848 > cat /proc/irq/0/smp_affinity 849 ffffffff 850 851There is an alternate interface, smp_affinity_list which allows specifying 852a CPU range instead of a bitmask:: 853 854 > cat /proc/irq/0/smp_affinity_list 855 1024-1031 856 857The default_smp_affinity mask applies to all non-active IRQs, which are the 858IRQs which have not yet been allocated/activated, and hence which lack a 859/proc/irq/[0-9]* directory. 860 861The node file on an SMP system shows the node to which the device using the IRQ 862reports itself as being attached. This hardware locality information does not 863include information about any possible driver locality preference. 864 865prof_cpu_mask specifies which CPUs are to be profiled by the system wide 866profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 867 868The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 869between all the CPUs which are allowed to handle it. As usual the kernel has 870more info than you and does a better job than you, so the defaults are the 871best choice for almost everyone. [Note this applies only to those IO-APIC's 872that support "Round Robin" interrupt distribution.] 873 874There are three more important subdirectories in /proc: net, scsi, and sys. 875The general rule is that the contents, or even the existence of these 876directories, depend on your kernel configuration. If SCSI is not enabled, the 877directory scsi may not exist. The same is true with the net, which is there 878only when networking support is present in the running kernel. 879 880The slabinfo file gives information about memory usage at the slab level. 881Linux uses slab pools for memory management above page level in version 2.2. 882Commonly used objects have their own slab pool (such as network buffers, 883directory cache, and so on). 884 885:: 886 887 > cat /proc/buddyinfo 888 889 Node 0, zone DMA 0 4 5 4 4 3 ... 890 Node 0, zone Normal 1 0 0 1 101 8 ... 891 Node 0, zone HighMem 2 0 0 1 1 0 ... 892 893External fragmentation is a problem under some workloads, and buddyinfo is a 894useful tool for helping diagnose these problems. Buddyinfo will give you a 895clue as to how big an area you can safely allocate, or why a previous 896allocation failed. 897 898Each column represents the number of pages of a certain order which are 899available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 900ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 901available in ZONE_NORMAL, etc... 902 903More information relevant to external fragmentation can be found in 904pagetypeinfo:: 905 906 > cat /proc/pagetypeinfo 907 Page block order: 9 908 Pages per block: 512 909 910 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 911 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 912 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 913 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 914 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 915 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 916 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 917 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 918 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 919 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 920 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 921 922 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 923 Node 0, zone DMA 2 0 5 1 0 924 Node 0, zone DMA32 41 6 967 2 0 925 926Fragmentation avoidance in the kernel works by grouping pages of different 927migrate types into the same contiguous regions of memory called page blocks. 928A page block is typically the size of the default hugepage size, e.g. 2MB on 929X86-64. By keeping pages grouped based on their ability to move, the kernel 930can reclaim pages within a page block to satisfy a high-order allocation. 931 932The pagetypinfo begins with information on the size of a page block. It 933then gives the same type of information as buddyinfo except broken down 934by migrate-type and finishes with details on how many page blocks of each 935type exist. 936 937If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 938from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 939make an estimate of the likely number of huge pages that can be allocated 940at a given point in time. All the "Movable" blocks should be allocatable 941unless memory has been mlock()'d. Some of the Reclaimable blocks should 942also be allocatable although a lot of filesystem metadata may have to be 943reclaimed to achieve this. 944 945 946meminfo 947~~~~~~~ 948 949Provides information about distribution and utilization of memory. This 950varies by architecture and compile options. Some of the counters reported 951here overlap. The memory reported by the non overlapping counters may not 952add up to the overall memory usage and the difference for some workloads 953can be substantial. In many cases there are other means to find out 954additional memory using subsystem specific interfaces, for instance 955/proc/net/sockstat for TCP memory allocations. 956 957Example output. You may not have all of these fields. 958 959:: 960 961 > cat /proc/meminfo 962 963 MemTotal: 32858820 kB 964 MemFree: 21001236 kB 965 MemAvailable: 27214312 kB 966 Buffers: 581092 kB 967 Cached: 5587612 kB 968 SwapCached: 0 kB 969 Active: 3237152 kB 970 Inactive: 7586256 kB 971 Active(anon): 94064 kB 972 Inactive(anon): 4570616 kB 973 Active(file): 3143088 kB 974 Inactive(file): 3015640 kB 975 Unevictable: 0 kB 976 Mlocked: 0 kB 977 SwapTotal: 0 kB 978 SwapFree: 0 kB 979 Zswap: 1904 kB 980 Zswapped: 7792 kB 981 Dirty: 12 kB 982 Writeback: 0 kB 983 AnonPages: 4654780 kB 984 Mapped: 266244 kB 985 Shmem: 9976 kB 986 KReclaimable: 517708 kB 987 Slab: 660044 kB 988 SReclaimable: 517708 kB 989 SUnreclaim: 142336 kB 990 KernelStack: 11168 kB 991 PageTables: 20540 kB 992 SecPageTables: 0 kB 993 NFS_Unstable: 0 kB 994 Bounce: 0 kB 995 WritebackTmp: 0 kB 996 CommitLimit: 16429408 kB 997 Committed_AS: 7715148 kB 998 VmallocTotal: 34359738367 kB 999 VmallocUsed: 40444 kB 1000 VmallocChunk: 0 kB 1001 Percpu: 29312 kB 1002 EarlyMemtestBad: 0 kB 1003 HardwareCorrupted: 0 kB 1004 AnonHugePages: 4149248 kB 1005 ShmemHugePages: 0 kB 1006 ShmemPmdMapped: 0 kB 1007 FileHugePages: 0 kB 1008 FilePmdMapped: 0 kB 1009 CmaTotal: 0 kB 1010 CmaFree: 0 kB 1011 HugePages_Total: 0 1012 HugePages_Free: 0 1013 HugePages_Rsvd: 0 1014 HugePages_Surp: 0 1015 Hugepagesize: 2048 kB 1016 Hugetlb: 0 kB 1017 DirectMap4k: 401152 kB 1018 DirectMap2M: 10008576 kB 1019 DirectMap1G: 24117248 kB 1020 1021MemTotal 1022 Total usable RAM (i.e. physical RAM minus a few reserved 1023 bits and the kernel binary code) 1024MemFree 1025 Total free RAM. On highmem systems, the sum of LowFree+HighFree 1026MemAvailable 1027 An estimate of how much memory is available for starting new 1028 applications, without swapping. Calculated from MemFree, 1029 SReclaimable, the size of the file LRU lists, and the low 1030 watermarks in each zone. 1031 The estimate takes into account that the system needs some 1032 page cache to function well, and that not all reclaimable 1033 slab will be reclaimable, due to items being in use. The 1034 impact of those factors will vary from system to system. 1035Buffers 1036 Relatively temporary storage for raw disk blocks 1037 shouldn't get tremendously large (20MB or so) 1038Cached 1039 In-memory cache for files read from the disk (the 1040 pagecache) as well as tmpfs & shmem. 1041 Doesn't include SwapCached. 1042SwapCached 1043 Memory that once was swapped out, is swapped back in but 1044 still also is in the swapfile (if memory is needed it 1045 doesn't need to be swapped out AGAIN because it is already 1046 in the swapfile. This saves I/O) 1047Active 1048 Memory that has been used more recently and usually not 1049 reclaimed unless absolutely necessary. 1050Inactive 1051 Memory which has been less recently used. It is more 1052 eligible to be reclaimed for other purposes 1053Unevictable 1054 Memory allocated for userspace which cannot be reclaimed, such 1055 as mlocked pages, ramfs backing pages, secret memfd pages etc. 1056Mlocked 1057 Memory locked with mlock(). 1058HighTotal, HighFree 1059 Highmem is all memory above ~860MB of physical memory. 1060 Highmem areas are for use by userspace programs, or 1061 for the pagecache. The kernel must use tricks to access 1062 this memory, making it slower to access than lowmem. 1063LowTotal, LowFree 1064 Lowmem is memory which can be used for everything that 1065 highmem can be used for, but it is also available for the 1066 kernel's use for its own data structures. Among many 1067 other things, it is where everything from the Slab is 1068 allocated. Bad things happen when you're out of lowmem. 1069SwapTotal 1070 total amount of swap space available 1071SwapFree 1072 Memory which has been evicted from RAM, and is temporarily 1073 on the disk 1074Zswap 1075 Memory consumed by the zswap backend (compressed size) 1076Zswapped 1077 Amount of anonymous memory stored in zswap (original size) 1078Dirty 1079 Memory which is waiting to get written back to the disk 1080Writeback 1081 Memory which is actively being written back to the disk 1082AnonPages 1083 Non-file backed pages mapped into userspace page tables 1084Mapped 1085 files which have been mmapped, such as libraries 1086Shmem 1087 Total memory used by shared memory (shmem) and tmpfs 1088KReclaimable 1089 Kernel allocations that the kernel will attempt to reclaim 1090 under memory pressure. Includes SReclaimable (below), and other 1091 direct allocations with a shrinker. 1092Slab 1093 in-kernel data structures cache 1094SReclaimable 1095 Part of Slab, that might be reclaimed, such as caches 1096SUnreclaim 1097 Part of Slab, that cannot be reclaimed on memory pressure 1098KernelStack 1099 Memory consumed by the kernel stacks of all tasks 1100PageTables 1101 Memory consumed by userspace page tables 1102SecPageTables 1103 Memory consumed by secondary page tables, this currently 1104 currently includes KVM mmu allocations on x86 and arm64. 1105NFS_Unstable 1106 Always zero. Previous counted pages which had been written to 1107 the server, but has not been committed to stable storage. 1108Bounce 1109 Memory used for block device "bounce buffers" 1110WritebackTmp 1111 Memory used by FUSE for temporary writeback buffers 1112CommitLimit 1113 Based on the overcommit ratio ('vm.overcommit_ratio'), 1114 this is the total amount of memory currently available to 1115 be allocated on the system. This limit is only adhered to 1116 if strict overcommit accounting is enabled (mode 2 in 1117 'vm.overcommit_memory'). 1118 1119 The CommitLimit is calculated with the following formula:: 1120 1121 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1122 overcommit_ratio / 100 + [total swap pages] 1123 1124 For example, on a system with 1G of physical RAM and 7G 1125 of swap with a `vm.overcommit_ratio` of 30 it would 1126 yield a CommitLimit of 7.3G. 1127 1128 For more details, see the memory overcommit documentation 1129 in mm/overcommit-accounting. 1130Committed_AS 1131 The amount of memory presently allocated on the system. 1132 The committed memory is a sum of all of the memory which 1133 has been allocated by processes, even if it has not been 1134 "used" by them as of yet. A process which malloc()'s 1G 1135 of memory, but only touches 300M of it will show up as 1136 using 1G. This 1G is memory which has been "committed" to 1137 by the VM and can be used at any time by the allocating 1138 application. With strict overcommit enabled on the system 1139 (mode 2 in 'vm.overcommit_memory'), allocations which would 1140 exceed the CommitLimit (detailed above) will not be permitted. 1141 This is useful if one needs to guarantee that processes will 1142 not fail due to lack of memory once that memory has been 1143 successfully allocated. 1144VmallocTotal 1145 total size of vmalloc virtual address space 1146VmallocUsed 1147 amount of vmalloc area which is used 1148VmallocChunk 1149 largest contiguous block of vmalloc area which is free 1150Percpu 1151 Memory allocated to the percpu allocator used to back percpu 1152 allocations. This stat excludes the cost of metadata. 1153EarlyMemtestBad 1154 The amount of RAM/memory in kB, that was identified as corrupted 1155 by early memtest. If memtest was not run, this field will not 1156 be displayed at all. Size is never rounded down to 0 kB. 1157 That means if 0 kB is reported, you can safely assume 1158 there was at least one pass of memtest and none of the passes 1159 found a single faulty byte of RAM. 1160HardwareCorrupted 1161 The amount of RAM/memory in KB, the kernel identifies as 1162 corrupted. 1163AnonHugePages 1164 Non-file backed huge pages mapped into userspace page tables 1165ShmemHugePages 1166 Memory used by shared memory (shmem) and tmpfs allocated 1167 with huge pages 1168ShmemPmdMapped 1169 Shared memory mapped into userspace with huge pages 1170FileHugePages 1171 Memory used for filesystem data (page cache) allocated 1172 with huge pages 1173FilePmdMapped 1174 Page cache mapped into userspace with huge pages 1175CmaTotal 1176 Memory reserved for the Contiguous Memory Allocator (CMA) 1177CmaFree 1178 Free remaining memory in the CMA reserves 1179HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb 1180 See Documentation/admin-guide/mm/hugetlbpage.rst. 1181DirectMap4k, DirectMap2M, DirectMap1G 1182 Breakdown of page table sizes used in the kernel's 1183 identity mapping of RAM 1184 1185vmallocinfo 1186~~~~~~~~~~~ 1187 1188Provides information about vmalloced/vmaped areas. One line per area, 1189containing the virtual address range of the area, size in bytes, 1190caller information of the creator, and optional information depending 1191on the kind of area: 1192 1193 ========== =================================================== 1194 pages=nr number of pages 1195 phys=addr if a physical address was specified 1196 ioremap I/O mapping (ioremap() and friends) 1197 vmalloc vmalloc() area 1198 vmap vmap()ed pages 1199 user VM_USERMAP area 1200 vpages buffer for pages pointers was vmalloced (huge area) 1201 N<node>=nr (Only on NUMA kernels) 1202 Number of pages allocated on memory node <node> 1203 ========== =================================================== 1204 1205:: 1206 1207 > cat /proc/vmallocinfo 1208 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1209 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1210 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1211 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1212 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1213 phys=7fee8000 ioremap 1214 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1215 phys=7fee7000 ioremap 1216 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1217 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1218 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1219 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1220 pages=2 vmalloc N1=2 1221 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1222 /0x130 [x_tables] pages=4 vmalloc N0=4 1223 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1224 pages=14 vmalloc N2=14 1225 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1226 pages=4 vmalloc N1=4 1227 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1228 pages=2 vmalloc N1=2 1229 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1230 pages=10 vmalloc N0=10 1231 1232 1233softirqs 1234~~~~~~~~ 1235 1236Provides counts of softirq handlers serviced since boot time, for each CPU. 1237 1238:: 1239 1240 > cat /proc/softirqs 1241 CPU0 CPU1 CPU2 CPU3 1242 HI: 0 0 0 0 1243 TIMER: 27166 27120 27097 27034 1244 NET_TX: 0 0 0 17 1245 NET_RX: 42 0 0 39 1246 BLOCK: 0 0 107 1121 1247 TASKLET: 0 0 0 290 1248 SCHED: 27035 26983 26971 26746 1249 HRTIMER: 0 0 0 0 1250 RCU: 1678 1769 2178 2250 1251 12521.3 Networking info in /proc/net 1253-------------------------------- 1254 1255The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1256additional values you get for IP version 6 if you configure the kernel to 1257support this. Table 1-9 lists the files and their meaning. 1258 1259 1260.. table:: Table 1-8: IPv6 info in /proc/net 1261 1262 ========== ===================================================== 1263 File Content 1264 ========== ===================================================== 1265 udp6 UDP sockets (IPv6) 1266 tcp6 TCP sockets (IPv6) 1267 raw6 Raw device statistics (IPv6) 1268 igmp6 IP multicast addresses, which this host joined (IPv6) 1269 if_inet6 List of IPv6 interface addresses 1270 ipv6_route Kernel routing table for IPv6 1271 rt6_stats Global IPv6 routing tables statistics 1272 sockstat6 Socket statistics (IPv6) 1273 snmp6 Snmp data (IPv6) 1274 ========== ===================================================== 1275 1276.. table:: Table 1-9: Network info in /proc/net 1277 1278 ============= ================================================================ 1279 File Content 1280 ============= ================================================================ 1281 arp Kernel ARP table 1282 dev network devices with statistics 1283 dev_mcast the Layer2 multicast groups a device is listening too 1284 (interface index, label, number of references, number of bound 1285 addresses). 1286 dev_stat network device status 1287 ip_fwchains Firewall chain linkage 1288 ip_fwnames Firewall chain names 1289 ip_masq Directory containing the masquerading tables 1290 ip_masquerade Major masquerading table 1291 netstat Network statistics 1292 raw raw device statistics 1293 route Kernel routing table 1294 rpc Directory containing rpc info 1295 rt_cache Routing cache 1296 snmp SNMP data 1297 sockstat Socket statistics 1298 softnet_stat Per-CPU incoming packets queues statistics of online CPUs 1299 tcp TCP sockets 1300 udp UDP sockets 1301 unix UNIX domain sockets 1302 wireless Wireless interface data (Wavelan etc) 1303 igmp IP multicast addresses, which this host joined 1304 psched Global packet scheduler parameters. 1305 netlink List of PF_NETLINK sockets 1306 ip_mr_vifs List of multicast virtual interfaces 1307 ip_mr_cache List of multicast routing cache 1308 ============= ================================================================ 1309 1310You can use this information to see which network devices are available in 1311your system and how much traffic was routed over those devices:: 1312 1313 > cat /proc/net/dev 1314 Inter-|Receive |[... 1315 face |bytes packets errs drop fifo frame compressed multicast|[... 1316 lo: 908188 5596 0 0 0 0 0 0 [... 1317 ppp0:15475140 20721 410 0 0 410 0 0 [... 1318 eth0: 614530 7085 0 0 0 0 0 1 [... 1319 1320 ...] Transmit 1321 ...] bytes packets errs drop fifo colls carrier compressed 1322 ...] 908188 5596 0 0 0 0 0 0 1323 ...] 1375103 17405 0 0 0 0 0 0 1324 ...] 1703981 5535 0 0 0 3 0 0 1325 1326In addition, each Channel Bond interface has its own directory. For 1327example, the bond0 device will have a directory called /proc/net/bond0/. 1328It will contain information that is specific to that bond, such as the 1329current slaves of the bond, the link status of the slaves, and how 1330many times the slaves link has failed. 1331 13321.4 SCSI info 1333------------- 1334 1335If you have a SCSI or ATA host adapter in your system, you'll find a 1336subdirectory named after the driver for this adapter in /proc/scsi. 1337You'll also see a list of all recognized SCSI devices in /proc/scsi:: 1338 1339 >cat /proc/scsi/scsi 1340 Attached devices: 1341 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1342 Vendor: IBM Model: DGHS09U Rev: 03E0 1343 Type: Direct-Access ANSI SCSI revision: 03 1344 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1345 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1346 Type: CD-ROM ANSI SCSI revision: 02 1347 1348 1349The directory named after the driver has one file for each adapter found in 1350the system. These files contain information about the controller, including 1351the used IRQ and the IO address range. The amount of information shown is 1352dependent on the adapter you use. The example shows the output for an Adaptec 1353AHA-2940 SCSI adapter:: 1354 1355 > cat /proc/scsi/aic7xxx/0 1356 1357 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1358 Compile Options: 1359 TCQ Enabled By Default : Disabled 1360 AIC7XXX_PROC_STATS : Disabled 1361 AIC7XXX_RESET_DELAY : 5 1362 Adapter Configuration: 1363 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1364 Ultra Wide Controller 1365 PCI MMAPed I/O Base: 0xeb001000 1366 Adapter SEEPROM Config: SEEPROM found and used. 1367 Adaptec SCSI BIOS: Enabled 1368 IRQ: 10 1369 SCBs: Active 0, Max Active 2, 1370 Allocated 15, HW 16, Page 255 1371 Interrupts: 160328 1372 BIOS Control Word: 0x18b6 1373 Adapter Control Word: 0x005b 1374 Extended Translation: Enabled 1375 Disconnect Enable Flags: 0xffff 1376 Ultra Enable Flags: 0x0001 1377 Tag Queue Enable Flags: 0x0000 1378 Ordered Queue Tag Flags: 0x0000 1379 Default Tag Queue Depth: 8 1380 Tagged Queue By Device array for aic7xxx host instance 0: 1381 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1382 Actual queue depth per device for aic7xxx host instance 0: 1383 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1384 Statistics: 1385 (scsi0:0:0:0) 1386 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1387 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1388 Total transfers 160151 (74577 reads and 85574 writes) 1389 (scsi0:0:6:0) 1390 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1391 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1392 Total transfers 0 (0 reads and 0 writes) 1393 1394 13951.5 Parallel port info in /proc/parport 1396--------------------------------------- 1397 1398The directory /proc/parport contains information about the parallel ports of 1399your system. It has one subdirectory for each port, named after the port 1400number (0,1,2,...). 1401 1402These directories contain the four files shown in Table 1-10. 1403 1404 1405.. table:: Table 1-10: Files in /proc/parport 1406 1407 ========= ==================================================================== 1408 File Content 1409 ========= ==================================================================== 1410 autoprobe Any IEEE-1284 device ID information that has been acquired. 1411 devices list of the device drivers using that port. A + will appear by the 1412 name of the device currently using the port (it might not appear 1413 against any). 1414 hardware Parallel port's base address, IRQ line and DMA channel. 1415 irq IRQ that parport is using for that port. This is in a separate 1416 file to allow you to alter it by writing a new value in (IRQ 1417 number or none). 1418 ========= ==================================================================== 1419 14201.6 TTY info in /proc/tty 1421------------------------- 1422 1423Information about the available and actually used tty's can be found in the 1424directory /proc/tty. You'll find entries for drivers and line disciplines in 1425this directory, as shown in Table 1-11. 1426 1427 1428.. table:: Table 1-11: Files in /proc/tty 1429 1430 ============= ============================================== 1431 File Content 1432 ============= ============================================== 1433 drivers list of drivers and their usage 1434 ldiscs registered line disciplines 1435 driver/serial usage statistic and status of single tty lines 1436 ============= ============================================== 1437 1438To see which tty's are currently in use, you can simply look into the file 1439/proc/tty/drivers:: 1440 1441 > cat /proc/tty/drivers 1442 pty_slave /dev/pts 136 0-255 pty:slave 1443 pty_master /dev/ptm 128 0-255 pty:master 1444 pty_slave /dev/ttyp 3 0-255 pty:slave 1445 pty_master /dev/pty 2 0-255 pty:master 1446 serial /dev/cua 5 64-67 serial:callout 1447 serial /dev/ttyS 4 64-67 serial 1448 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1449 /dev/ptmx /dev/ptmx 5 2 system 1450 /dev/console /dev/console 5 1 system:console 1451 /dev/tty /dev/tty 5 0 system:/dev/tty 1452 unknown /dev/tty 4 1-63 console 1453 1454 14551.7 Miscellaneous kernel statistics in /proc/stat 1456------------------------------------------------- 1457 1458Various pieces of information about kernel activity are available in the 1459/proc/stat file. All of the numbers reported in this file are aggregates 1460since the system first booted. For a quick look, simply cat the file:: 1461 1462 > cat /proc/stat 1463 cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0 1464 cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0 1465 cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0 1466 cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0 1467 cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0 1468 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> 1469 ctxt 22848221062 1470 btime 1605316999 1471 processes 746787147 1472 procs_running 2 1473 procs_blocked 0 1474 softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354 1475 1476The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1477lines. These numbers identify the amount of time the CPU has spent performing 1478different kinds of work. Time units are in USER_HZ (typically hundredths of a 1479second). The meanings of the columns are as follows, from left to right: 1480 1481- user: normal processes executing in user mode 1482- nice: niced processes executing in user mode 1483- system: processes executing in kernel mode 1484- idle: twiddling thumbs 1485- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1486 are several problems: 1487 1488 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1489 waiting for I/O to complete. When CPU goes into idle state for 1490 outstanding task I/O, another task will be scheduled on this CPU. 1491 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1492 on any CPU, so the iowait of each CPU is difficult to calculate. 1493 3. The value of iowait field in /proc/stat will decrease in certain 1494 conditions. 1495 1496 So, the iowait is not reliable by reading from /proc/stat. 1497- irq: servicing interrupts 1498- softirq: servicing softirqs 1499- steal: involuntary wait 1500- guest: running a normal guest 1501- guest_nice: running a niced guest 1502 1503The "intr" line gives counts of interrupts serviced since boot time, for each 1504of the possible system interrupts. The first column is the total of all 1505interrupts serviced including unnumbered architecture specific interrupts; 1506each subsequent column is the total for that particular numbered interrupt. 1507Unnumbered interrupts are not shown, only summed into the total. 1508 1509The "ctxt" line gives the total number of context switches across all CPUs. 1510 1511The "btime" line gives the time at which the system booted, in seconds since 1512the Unix epoch. 1513 1514The "processes" line gives the number of processes and threads created, which 1515includes (but is not limited to) those created by calls to the fork() and 1516clone() system calls. 1517 1518The "procs_running" line gives the total number of threads that are 1519running or ready to run (i.e., the total number of runnable threads). 1520 1521The "procs_blocked" line gives the number of processes currently blocked, 1522waiting for I/O to complete. 1523 1524The "softirq" line gives counts of softirqs serviced since boot time, for each 1525of the possible system softirqs. The first column is the total of all 1526softirqs serviced; each subsequent column is the total for that particular 1527softirq. 1528 1529 15301.8 Ext4 file system parameters 1531------------------------------- 1532 1533Information about mounted ext4 file systems can be found in 1534/proc/fs/ext4. Each mounted filesystem will have a directory in 1535/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1536/proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device 1537directory are shown in Table 1-12, below. 1538 1539.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1540 1541 ============== ========================================================== 1542 File Content 1543 mb_groups details of multiblock allocator buddy cache of free blocks 1544 ============== ========================================================== 1545 15461.9 /proc/consoles 1547------------------- 1548Shows registered system console lines. 1549 1550To see which character device lines are currently used for the system console 1551/dev/console, you may simply look into the file /proc/consoles:: 1552 1553 > cat /proc/consoles 1554 tty0 -WU (ECp) 4:7 1555 ttyS0 -W- (Ep) 4:64 1556 1557The columns are: 1558 1559+--------------------+-------------------------------------------------------+ 1560| device | name of the device | 1561+====================+=======================================================+ 1562| operations | * R = can do read operations | 1563| | * W = can do write operations | 1564| | * U = can do unblank | 1565+--------------------+-------------------------------------------------------+ 1566| flags | * E = it is enabled | 1567| | * C = it is preferred console | 1568| | * B = it is primary boot console | 1569| | * p = it is used for printk buffer | 1570| | * b = it is not a TTY but a Braille device | 1571| | * a = it is safe to use when cpu is offline | 1572+--------------------+-------------------------------------------------------+ 1573| major:minor | major and minor number of the device separated by a | 1574| | colon | 1575+--------------------+-------------------------------------------------------+ 1576 1577Summary 1578------- 1579 1580The /proc file system serves information about the running system. It not only 1581allows access to process data but also allows you to request the kernel status 1582by reading files in the hierarchy. 1583 1584The directory structure of /proc reflects the types of information and makes 1585it easy, if not obvious, where to look for specific data. 1586 1587Chapter 2: Modifying System Parameters 1588====================================== 1589 1590In This Chapter 1591--------------- 1592 1593* Modifying kernel parameters by writing into files found in /proc/sys 1594* Exploring the files which modify certain parameters 1595* Review of the /proc/sys file tree 1596 1597------------------------------------------------------------------------------ 1598 1599A very interesting part of /proc is the directory /proc/sys. This is not only 1600a source of information, it also allows you to change parameters within the 1601kernel. Be very careful when attempting this. You can optimize your system, 1602but you can also cause it to crash. Never alter kernel parameters on a 1603production system. Set up a development machine and test to make sure that 1604everything works the way you want it to. You may have no alternative but to 1605reboot the machine once an error has been made. 1606 1607To change a value, simply echo the new value into the file. 1608You need to be root to do this. You can create your own boot script 1609to perform this every time your system boots. 1610 1611The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1612general things in the operation of the Linux kernel. Since some of the files 1613can inadvertently disrupt your system, it is advisable to read both 1614documentation and source before actually making adjustments. In any case, be 1615very careful when writing to any of these files. The entries in /proc may 1616change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1617review the kernel documentation in the directory linux/Documentation. 1618This chapter is heavily based on the documentation included in the pre 2.2 1619kernels, and became part of it in version 2.2.1 of the Linux kernel. 1620 1621Please see: Documentation/admin-guide/sysctl/ directory for descriptions of 1622these entries. 1623 1624Summary 1625------- 1626 1627Certain aspects of kernel behavior can be modified at runtime, without the 1628need to recompile the kernel, or even to reboot the system. The files in the 1629/proc/sys tree can not only be read, but also modified. You can use the echo 1630command to write value into these files, thereby changing the default settings 1631of the kernel. 1632 1633 1634Chapter 3: Per-process Parameters 1635================================= 1636 16373.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1638-------------------------------------------------------------------------------- 1639 1640These files can be used to adjust the badness heuristic used to select which 1641process gets killed in out of memory (oom) conditions. 1642 1643The badness heuristic assigns a value to each candidate task ranging from 0 1644(never kill) to 1000 (always kill) to determine which process is targeted. The 1645units are roughly a proportion along that range of allowed memory the process 1646may allocate from based on an estimation of its current memory and swap use. 1647For example, if a task is using all allowed memory, its badness score will be 16481000. If it is using half of its allowed memory, its score will be 500. 1649 1650The amount of "allowed" memory depends on the context in which the oom killer 1651was called. If it is due to the memory assigned to the allocating task's cpuset 1652being exhausted, the allowed memory represents the set of mems assigned to that 1653cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1654memory represents the set of mempolicy nodes. If it is due to a memory 1655limit (or swap limit) being reached, the allowed memory is that configured 1656limit. Finally, if it is due to the entire system being out of memory, the 1657allowed memory represents all allocatable resources. 1658 1659The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1660is used to determine which task to kill. Acceptable values range from -1000 1661(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1662polarize the preference for oom killing either by always preferring a certain 1663task or completely disabling it. The lowest possible value, -1000, is 1664equivalent to disabling oom killing entirely for that task since it will always 1665report a badness score of 0. 1666 1667Consequently, it is very simple for userspace to define the amount of memory to 1668consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1669example, is roughly equivalent to allowing the remainder of tasks sharing the 1670same system, cpuset, mempolicy, or memory controller resources to use at least 167150% more memory. A value of -500, on the other hand, would be roughly 1672equivalent to discounting 50% of the task's allowed memory from being considered 1673as scoring against the task. 1674 1675For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1676be used to tune the badness score. Its acceptable values range from -16 1677(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1678(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1679scaled linearly with /proc/<pid>/oom_score_adj. 1680 1681The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1682value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1683requires CAP_SYS_RESOURCE. 1684 1685 16863.2 /proc/<pid>/oom_score - Display current oom-killer score 1687------------------------------------------------------------- 1688 1689This file can be used to check the current score used by the oom-killer for 1690any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1691process should be killed in an out-of-memory situation. 1692 1693Please note that the exported value includes oom_score_adj so it is 1694effectively in range [0,2000]. 1695 1696 16973.3 /proc/<pid>/io - Display the IO accounting fields 1698------------------------------------------------------- 1699 1700This file contains IO statistics for each running process. 1701 1702Example 1703~~~~~~~ 1704 1705:: 1706 1707 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1708 [1] 3828 1709 1710 test:/tmp # cat /proc/3828/io 1711 rchar: 323934931 1712 wchar: 323929600 1713 syscr: 632687 1714 syscw: 632675 1715 read_bytes: 0 1716 write_bytes: 323932160 1717 cancelled_write_bytes: 0 1718 1719 1720Description 1721~~~~~~~~~~~ 1722 1723rchar 1724^^^^^ 1725 1726I/O counter: chars read 1727The number of bytes which this task has caused to be read from storage. This 1728is simply the sum of bytes which this process passed to read() and pread(). 1729It includes things like tty IO and it is unaffected by whether or not actual 1730physical disk IO was required (the read might have been satisfied from 1731pagecache). 1732 1733 1734wchar 1735^^^^^ 1736 1737I/O counter: chars written 1738The number of bytes which this task has caused, or shall cause to be written 1739to disk. Similar caveats apply here as with rchar. 1740 1741 1742syscr 1743^^^^^ 1744 1745I/O counter: read syscalls 1746Attempt to count the number of read I/O operations, i.e. syscalls like read() 1747and pread(). 1748 1749 1750syscw 1751^^^^^ 1752 1753I/O counter: write syscalls 1754Attempt to count the number of write I/O operations, i.e. syscalls like 1755write() and pwrite(). 1756 1757 1758read_bytes 1759^^^^^^^^^^ 1760 1761I/O counter: bytes read 1762Attempt to count the number of bytes which this process really did cause to 1763be fetched from the storage layer. Done at the submit_bio() level, so it is 1764accurate for block-backed filesystems. <please add status regarding NFS and 1765CIFS at a later time> 1766 1767 1768write_bytes 1769^^^^^^^^^^^ 1770 1771I/O counter: bytes written 1772Attempt to count the number of bytes which this process caused to be sent to 1773the storage layer. This is done at page-dirtying time. 1774 1775 1776cancelled_write_bytes 1777^^^^^^^^^^^^^^^^^^^^^ 1778 1779The big inaccuracy here is truncate. If a process writes 1MB to a file and 1780then deletes the file, it will in fact perform no writeout. But it will have 1781been accounted as having caused 1MB of write. 1782In other words: The number of bytes which this process caused to not happen, 1783by truncating pagecache. A task can cause "negative" IO too. If this task 1784truncates some dirty pagecache, some IO which another task has been accounted 1785for (in its write_bytes) will not be happening. We _could_ just subtract that 1786from the truncating task's write_bytes, but there is information loss in doing 1787that. 1788 1789 1790.. Note:: 1791 1792 At its current implementation state, this is a bit racy on 32-bit machines: 1793 if process A reads process B's /proc/pid/io while process B is updating one 1794 of those 64-bit counters, process A could see an intermediate result. 1795 1796 1797More information about this can be found within the taskstats documentation in 1798Documentation/accounting. 1799 18003.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1801--------------------------------------------------------------- 1802When a process is dumped, all anonymous memory is written to a core file as 1803long as the size of the core file isn't limited. But sometimes we don't want 1804to dump some memory segments, for example, huge shared memory or DAX. 1805Conversely, sometimes we want to save file-backed memory segments into a core 1806file, not only the individual files. 1807 1808/proc/<pid>/coredump_filter allows you to customize which memory segments 1809will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1810of memory types. If a bit of the bitmask is set, memory segments of the 1811corresponding memory type are dumped, otherwise they are not dumped. 1812 1813The following 9 memory types are supported: 1814 1815 - (bit 0) anonymous private memory 1816 - (bit 1) anonymous shared memory 1817 - (bit 2) file-backed private memory 1818 - (bit 3) file-backed shared memory 1819 - (bit 4) ELF header pages in file-backed private memory areas (it is 1820 effective only if the bit 2 is cleared) 1821 - (bit 5) hugetlb private memory 1822 - (bit 6) hugetlb shared memory 1823 - (bit 7) DAX private memory 1824 - (bit 8) DAX shared memory 1825 1826 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1827 are always dumped regardless of the bitmask status. 1828 1829 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1830 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1831 1832The default value of coredump_filter is 0x33; this means all anonymous memory 1833segments, ELF header pages and hugetlb private memory are dumped. 1834 1835If you don't want to dump all shared memory segments attached to pid 1234, 1836write 0x31 to the process's proc file:: 1837 1838 $ echo 0x31 > /proc/1234/coredump_filter 1839 1840When a new process is created, the process inherits the bitmask status from its 1841parent. It is useful to set up coredump_filter before the program runs. 1842For example:: 1843 1844 $ echo 0x7 > /proc/self/coredump_filter 1845 $ ./some_program 1846 18473.5 /proc/<pid>/mountinfo - Information about mounts 1848-------------------------------------------------------- 1849 1850This file contains lines of the form:: 1851 1852 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1853 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4) 1854 1855 (1) mount ID: unique identifier of the mount (may be reused after umount) 1856 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1857 (3) major:minor: value of st_dev for files on filesystem 1858 (4) root: root of the mount within the filesystem 1859 (5) mount point: mount point relative to the process's root 1860 (6) mount options: per mount options 1861 (n…m) optional fields: zero or more fields of the form "tag[:value]" 1862 (m+1) separator: marks the end of the optional fields 1863 (m+2) filesystem type: name of filesystem of the form "type[.subtype]" 1864 (m+3) mount source: filesystem specific information or "none" 1865 (m+4) super options: per super block options 1866 1867Parsers should ignore all unrecognised optional fields. Currently the 1868possible optional fields are: 1869 1870================ ============================================================== 1871shared:X mount is shared in peer group X 1872master:X mount is slave to peer group X 1873propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1874unbindable mount is unbindable 1875================ ============================================================== 1876 1877.. [#] X is the closest dominant peer group under the process's root. If 1878 X is the immediate master of the mount, or if there's no dominant peer 1879 group under the same root, then only the "master:X" field is present 1880 and not the "propagate_from:X" field. 1881 1882For more information on mount propagation see: 1883 1884 Documentation/filesystems/sharedsubtree.rst 1885 1886 18873.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1888-------------------------------------------------------- 1889These files provide a method to access a task's comm value. It also allows for 1890a task to set its own or one of its thread siblings comm value. The comm value 1891is limited in size compared to the cmdline value, so writing anything longer 1892then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated 1893comm value. 1894 1895 18963.7 /proc/<pid>/task/<tid>/children - Information about task children 1897------------------------------------------------------------------------- 1898This file provides a fast way to retrieve first level children pids 1899of a task pointed by <pid>/<tid> pair. The format is a space separated 1900stream of pids. 1901 1902Note the "first level" here -- if a child has its own children they will 1903not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1904to obtain the descendants. 1905 1906Since this interface is intended to be fast and cheap it doesn't 1907guarantee to provide precise results and some children might be 1908skipped, especially if they've exited right after we printed their 1909pids, so one needs to either stop or freeze processes being inspected 1910if precise results are needed. 1911 1912 19133.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1914--------------------------------------------------------------- 1915This file provides information associated with an opened file. The regular 1916files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. 1917The 'pos' represents the current offset of the opened file in decimal 1918form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the 1919file has been created with [see open(2) for details] and 'mnt_id' represents 1920mount ID of the file system containing the opened file [see 3.5 1921/proc/<pid>/mountinfo for details]. 'ino' represents the inode number of 1922the file. 1923 1924A typical output is:: 1925 1926 pos: 0 1927 flags: 0100002 1928 mnt_id: 19 1929 ino: 63107 1930 1931All locks associated with a file descriptor are shown in its fdinfo too:: 1932 1933 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1934 1935The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1936pair provide additional information particular to the objects they represent. 1937 1938Eventfd files 1939~~~~~~~~~~~~~ 1940 1941:: 1942 1943 pos: 0 1944 flags: 04002 1945 mnt_id: 9 1946 ino: 63107 1947 eventfd-count: 5a 1948 1949where 'eventfd-count' is hex value of a counter. 1950 1951Signalfd files 1952~~~~~~~~~~~~~~ 1953 1954:: 1955 1956 pos: 0 1957 flags: 04002 1958 mnt_id: 9 1959 ino: 63107 1960 sigmask: 0000000000000200 1961 1962where 'sigmask' is hex value of the signal mask associated 1963with a file. 1964 1965Epoll files 1966~~~~~~~~~~~ 1967 1968:: 1969 1970 pos: 0 1971 flags: 02 1972 mnt_id: 9 1973 ino: 63107 1974 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 1975 1976where 'tfd' is a target file descriptor number in decimal form, 1977'events' is events mask being watched and the 'data' is data 1978associated with a target [see epoll(7) for more details]. 1979 1980The 'pos' is current offset of the target file in decimal form 1981[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 1982where target file resides, all in hex format. 1983 1984Fsnotify files 1985~~~~~~~~~~~~~~ 1986For inotify files the format is the following:: 1987 1988 pos: 0 1989 flags: 02000000 1990 mnt_id: 9 1991 ino: 63107 1992 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 1993 1994where 'wd' is a watch descriptor in decimal form, i.e. a target file 1995descriptor number, 'ino' and 'sdev' are inode and device where the 1996target file resides and the 'mask' is the mask of events, all in hex 1997form [see inotify(7) for more details]. 1998 1999If the kernel was built with exportfs support, the path to the target 2000file is encoded as a file handle. The file handle is provided by three 2001fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 2002format. 2003 2004If the kernel is built without exportfs support the file handle won't be 2005printed out. 2006 2007If there is no inotify mark attached yet the 'inotify' line will be omitted. 2008 2009For fanotify files the format is:: 2010 2011 pos: 0 2012 flags: 02 2013 mnt_id: 9 2014 ino: 63107 2015 fanotify flags:10 event-flags:0 2016 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 2017 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 2018 2019where fanotify 'flags' and 'event-flags' are values used in fanotify_init 2020call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 2021flags associated with mark which are tracked separately from events 2022mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 2023mask and 'ignored_mask' is the mask of events which are to be ignored. 2024All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2025provide information about flags and mask used in fanotify_mark 2026call [see fsnotify manpage for details]. 2027 2028While the first three lines are mandatory and always printed, the rest is 2029optional and may be omitted if no marks created yet. 2030 2031Timerfd files 2032~~~~~~~~~~~~~ 2033 2034:: 2035 2036 pos: 0 2037 flags: 02 2038 mnt_id: 9 2039 ino: 63107 2040 clockid: 0 2041 ticks: 0 2042 settime flags: 01 2043 it_value: (0, 49406829) 2044 it_interval: (1, 0) 2045 2046where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2047that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2048flags in octal form been used to setup the timer [see timerfd_settime(2) for 2049details]. 'it_value' is remaining time until the timer expiration. 2050'it_interval' is the interval for the timer. Note the timer might be set up 2051with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2052still exhibits timer's remaining time. 2053 2054DMA Buffer files 2055~~~~~~~~~~~~~~~~ 2056 2057:: 2058 2059 pos: 0 2060 flags: 04002 2061 mnt_id: 9 2062 ino: 63107 2063 size: 32768 2064 count: 2 2065 exp_name: system-heap 2066 2067where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of 2068the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. 2069 20703.9 /proc/<pid>/map_files - Information about memory mapped files 2071--------------------------------------------------------------------- 2072This directory contains symbolic links which represent memory mapped files 2073the process is maintaining. Example output:: 2074 2075 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2076 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2077 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2078 | ... 2079 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2080 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2081 2082The name of a link represents the virtual memory bounds of a mapping, i.e. 2083vm_area_struct::vm_start-vm_area_struct::vm_end. 2084 2085The main purpose of the map_files is to retrieve a set of memory mapped 2086files in a fast way instead of parsing /proc/<pid>/maps or 2087/proc/<pid>/smaps, both of which contain many more records. At the same 2088time one can open(2) mappings from the listings of two processes and 2089comparing their inode numbers to figure out which anonymous memory areas 2090are actually shared. 2091 20923.10 /proc/<pid>/timerslack_ns - Task timerslack value 2093--------------------------------------------------------- 2094This file provides the value of the task's timerslack value in nanoseconds. 2095This value specifies an amount of time that normal timers may be deferred 2096in order to coalesce timers and avoid unnecessary wakeups. 2097 2098This allows a task's interactivity vs power consumption tradeoff to be 2099adjusted. 2100 2101Writing 0 to the file will set the task's timerslack to the default value. 2102 2103Valid values are from 0 - ULLONG_MAX 2104 2105An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2106permissions on the task specified to change its timerslack_ns value. 2107 21083.11 /proc/<pid>/patch_state - Livepatch patch operation state 2109----------------------------------------------------------------- 2110When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2111patch state for the task. 2112 2113A value of '-1' indicates that no patch is in transition. 2114 2115A value of '0' indicates that a patch is in transition and the task is 2116unpatched. If the patch is being enabled, then the task hasn't been 2117patched yet. If the patch is being disabled, then the task has already 2118been unpatched. 2119 2120A value of '1' indicates that a patch is in transition and the task is 2121patched. If the patch is being enabled, then the task has already been 2122patched. If the patch is being disabled, then the task hasn't been 2123unpatched yet. 2124 21253.12 /proc/<pid>/arch_status - task architecture specific status 2126------------------------------------------------------------------- 2127When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2128architecture specific status of the task. 2129 2130Example 2131~~~~~~~ 2132 2133:: 2134 2135 $ cat /proc/6753/arch_status 2136 AVX512_elapsed_ms: 8 2137 2138Description 2139~~~~~~~~~~~ 2140 2141x86 specific entries 2142~~~~~~~~~~~~~~~~~~~~~ 2143 2144AVX512_elapsed_ms 2145^^^^^^^^^^^^^^^^^^ 2146 2147 If AVX512 is supported on the machine, this entry shows the milliseconds 2148 elapsed since the last time AVX512 usage was recorded. The recording 2149 happens on a best effort basis when a task is scheduled out. This means 2150 that the value depends on two factors: 2151 2152 1) The time which the task spent on the CPU without being scheduled 2153 out. With CPU isolation and a single runnable task this can take 2154 several seconds. 2155 2156 2) The time since the task was scheduled out last. Depending on the 2157 reason for being scheduled out (time slice exhausted, syscall ...) 2158 this can be arbitrary long time. 2159 2160 As a consequence the value cannot be considered precise and authoritative 2161 information. The application which uses this information has to be aware 2162 of the overall scenario on the system in order to determine whether a 2163 task is a real AVX512 user or not. Precise information can be obtained 2164 with performance counters. 2165 2166 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2167 the task is unlikely an AVX512 user, but depends on the workload and the 2168 scheduling scenario, it also could be a false negative mentioned above. 2169 21703.13 /proc/<pid>/fd - List of symlinks to open files 2171------------------------------------------------------- 2172This directory contains symbolic links which represent open files 2173the process is maintaining. Example output:: 2174 2175 lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null 2176 l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null 2177 lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]' 2178 lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]' 2179 lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]' 2180 2181The number of open files for the process is stored in 'size' member 2182of stat() output for /proc/<pid>/fd for fast access. 2183------------------------------------------------------- 2184 2185 2186Chapter 4: Configuring procfs 2187============================= 2188 21894.1 Mount options 2190--------------------- 2191 2192The following mount options are supported: 2193 2194 ========= ======================================================== 2195 hidepid= Set /proc/<pid>/ access mode. 2196 gid= Set the group authorized to learn processes information. 2197 subset= Show only the specified subset of procfs. 2198 ========= ======================================================== 2199 2200hidepid=off or hidepid=0 means classic mode - everybody may access all 2201/proc/<pid>/ directories (default). 2202 2203hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2204directories but their own. Sensitive files like cmdline, sched*, status are now 2205protected against other users. This makes it impossible to learn whether any 2206user runs specific program (given the program doesn't reveal itself by its 2207behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2208other users, poorly written programs passing sensitive information via program 2209arguments are now protected against local eavesdroppers. 2210 2211hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2212fully invisible to other users. It doesn't mean that it hides a fact whether a 2213process with a specific pid value exists (it can be learned by other means, e.g. 2214by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by 2215stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2216gathering information about running processes, whether some daemon runs with 2217elevated privileges, whether other user runs some sensitive program, whether 2218other users run any program at all, etc. 2219 2220hidepid=ptraceable or hidepid=4 means that procfs should only contain 2221/proc/<pid>/ directories that the caller can ptrace. 2222 2223gid= defines a group authorized to learn processes information otherwise 2224prohibited by hidepid=. If you use some daemon like identd which needs to learn 2225information about processes information, just add identd to this group. 2226 2227subset=pid hides all top level files and directories in the procfs that 2228are not related to tasks. 2229 2230Chapter 5: Filesystem behavior 2231============================== 2232 2233Originally, before the advent of pid namespace, procfs was a global file 2234system. It means that there was only one procfs instance in the system. 2235 2236When pid namespace was added, a separate procfs instance was mounted in 2237each pid namespace. So, procfs mount options are global among all 2238mountpoints within the same namespace:: 2239 2240 # grep ^proc /proc/mounts 2241 proc /proc proc rw,relatime,hidepid=2 0 0 2242 2243 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2244 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2245 +++ exited with 0 +++ 2246 2247 # grep ^proc /proc/mounts 2248 proc /proc proc rw,relatime,hidepid=2 0 0 2249 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2250 2251and only after remounting procfs mount options will change at all 2252mountpoints:: 2253 2254 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2255 2256 # grep ^proc /proc/mounts 2257 proc /proc proc rw,relatime,hidepid=1 0 0 2258 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2259 2260This behavior is different from the behavior of other filesystems. 2261 2262The new procfs behavior is more like other filesystems. Each procfs mount 2263creates a new procfs instance. Mount options affect own procfs instance. 2264It means that it became possible to have several procfs instances 2265displaying tasks with different filtering options in one pid namespace:: 2266 2267 # mount -o hidepid=invisible -t proc proc /proc 2268 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2269 # grep ^proc /proc/mounts 2270 proc /proc proc rw,relatime,hidepid=invisible 0 0 2271 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2272