1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 9 Bodo Bauer <bb@ricochet.net> 102.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 11move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 12fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 13===================== ======================================= ================ 14 15 16 17.. Table of Contents 18 19 0 Preface 20 0.1 Introduction/Credits 21 0.2 Legal Stuff 22 23 1 Collecting System Information 24 1.1 Process-Specific Subdirectories 25 1.2 Kernel data 26 1.3 IDE devices in /proc/ide 27 1.4 Networking info in /proc/net 28 1.5 SCSI info 29 1.6 Parallel port info in /proc/parport 30 1.7 TTY info in /proc/tty 31 1.8 Miscellaneous kernel statistics in /proc/stat 32 1.9 Ext4 file system parameters 33 34 2 Modifying System Parameters 35 36 3 Per-Process Parameters 37 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 38 score 39 3.2 /proc/<pid>/oom_score - Display current oom-killer score 40 3.3 /proc/<pid>/io - Display the IO accounting fields 41 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings 42 3.5 /proc/<pid>/mountinfo - Information about mounts 43 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 44 3.7 /proc/<pid>/task/<tid>/children - Information about task children 45 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 46 3.9 /proc/<pid>/map_files - Information about memory mapped files 47 3.10 /proc/<pid>/timerslack_ns - Task timerslack value 48 3.11 /proc/<pid>/patch_state - Livepatch patch operation state 49 3.12 /proc/<pid>/arch_status - Task architecture specific information 50 51 4 Configuring procfs 52 4.1 Mount options 53 54 5 Filesystem behavior 55 56Preface 57======= 58 590.1 Introduction/Credits 60------------------------ 61 62This documentation is part of a soon (or so we hope) to be released book on 63the SuSE Linux distribution. As there is no complete documentation for the 64/proc file system and we've used many freely available sources to write these 65chapters, it seems only fair to give the work back to the Linux community. 66This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 67afraid it's still far from complete, but we hope it will be useful. As far as 68we know, it is the first 'all-in-one' document about the /proc file system. It 69is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 70SPARC, AXP, etc., features, you probably won't find what you are looking for. 71It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 72additions and patches are welcome and will be added to this document if you 73mail them to Bodo. 74 75We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 76other people for help compiling this documentation. We'd also like to extend a 77special thank you to Andi Kleen for documentation, which we relied on heavily 78to create this document, as well as the additional information he provided. 79Thanks to everybody else who contributed source or docs to the Linux kernel 80and helped create a great piece of software... :) 81 82If you have any comments, corrections or additions, please don't hesitate to 83contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 84document. 85 86The latest version of this document is available online at 87http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html 88 89If the above direction does not works for you, you could try the kernel 90mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 91comandante@zaralinux.com. 92 930.2 Legal Stuff 94--------------- 95 96We don't guarantee the correctness of this document, and if you come to us 97complaining about how you screwed up your system because of incorrect 98documentation, we won't feel responsible... 99 100Chapter 1: Collecting System Information 101======================================== 102 103In This Chapter 104--------------- 105* Investigating the properties of the pseudo file system /proc and its 106 ability to provide information on the running Linux system 107* Examining /proc's structure 108* Uncovering various information about the kernel and the processes running 109 on the system 110 111------------------------------------------------------------------------------ 112 113The proc file system acts as an interface to internal data structures in the 114kernel. It can be used to obtain information about the system and to change 115certain kernel parameters at runtime (sysctl). 116 117First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 118show you how you can use /proc/sys to change settings. 119 1201.1 Process-Specific Subdirectories 121----------------------------------- 122 123The directory /proc contains (among other things) one subdirectory for each 124process running on the system, which is named after the process ID (PID). 125 126The link 'self' points to the process reading the file system. Each process 127subdirectory has the entries listed in Table 1-1. 128 129Note that an open file descriptor to /proc/<pid> or to any of its 130contained files or subdirectories does not prevent <pid> being reused 131for some other process in the event that <pid> exits. Operations on 132open /proc/<pid> file descriptors corresponding to dead processes 133never act on any new process that the kernel may, through chance, have 134also assigned the process ID <pid>. Instead, operations on these FDs 135usually fail with ESRCH. 136 137.. table:: Table 1-1: Process specific entries in /proc 138 139 ============= =============================================================== 140 File Content 141 ============= =============================================================== 142 clear_refs Clears page referenced bits shown in smaps output 143 cmdline Command line arguments 144 cpu Current and last cpu in which it was executed (2.4)(smp) 145 cwd Link to the current working directory 146 environ Values of environment variables 147 exe Link to the executable of this process 148 fd Directory, which contains all file descriptors 149 maps Memory maps to executables and library files (2.4) 150 mem Memory held by this process 151 root Link to the root directory of this process 152 stat Process status 153 statm Process memory status information 154 status Process status in human readable form 155 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 156 symbol the task is blocked in - or "0" if not blocked. 157 pagemap Page table 158 stack Report full stack trace, enable via CONFIG_STACKTRACE 159 smaps An extension based on maps, showing the memory consumption of 160 each mapping and flags associated with it 161 smaps_rollup Accumulated smaps stats for all mappings of the process. This 162 can be derived from smaps, but is faster and more convenient 163 numa_maps An extension based on maps, showing the memory locality and 164 binding policy as well as mem usage (in pages) of each mapping. 165 ============= =============================================================== 166 167For example, to get the status information of a process, all you have to do is 168read the file /proc/PID/status:: 169 170 >cat /proc/self/status 171 Name: cat 172 State: R (running) 173 Tgid: 5452 174 Pid: 5452 175 PPid: 743 176 TracerPid: 0 (2.4) 177 Uid: 501 501 501 501 178 Gid: 100 100 100 100 179 FDSize: 256 180 Groups: 100 14 16 181 VmPeak: 5004 kB 182 VmSize: 5004 kB 183 VmLck: 0 kB 184 VmHWM: 476 kB 185 VmRSS: 476 kB 186 RssAnon: 352 kB 187 RssFile: 120 kB 188 RssShmem: 4 kB 189 VmData: 156 kB 190 VmStk: 88 kB 191 VmExe: 68 kB 192 VmLib: 1412 kB 193 VmPTE: 20 kb 194 VmSwap: 0 kB 195 HugetlbPages: 0 kB 196 CoreDumping: 0 197 THP_enabled: 1 198 Threads: 1 199 SigQ: 0/28578 200 SigPnd: 0000000000000000 201 ShdPnd: 0000000000000000 202 SigBlk: 0000000000000000 203 SigIgn: 0000000000000000 204 SigCgt: 0000000000000000 205 CapInh: 00000000fffffeff 206 CapPrm: 0000000000000000 207 CapEff: 0000000000000000 208 CapBnd: ffffffffffffffff 209 CapAmb: 0000000000000000 210 NoNewPrivs: 0 211 Seccomp: 0 212 Speculation_Store_Bypass: thread vulnerable 213 SpeculationIndirectBranch: conditional enabled 214 voluntary_ctxt_switches: 0 215 nonvoluntary_ctxt_switches: 1 216 217This shows you nearly the same information you would get if you viewed it with 218the ps command. In fact, ps uses the proc file system to obtain its 219information. But you get a more detailed view of the process by reading the 220file /proc/PID/status. It fields are described in table 1-2. 221 222The statm file contains more detailed information about the process 223memory usage. Its seven fields are explained in Table 1-3. The stat file 224contains detailed information about the process itself. Its fields are 225explained in Table 1-4. 226 227(for SMP CONFIG users) 228 229For making accounting scalable, RSS related information are handled in an 230asynchronous manner and the value may not be very precise. To see a precise 231snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 232It's slow but very precise. 233 234.. table:: Table 1-2: Contents of the status files (as of 4.19) 235 236 ========================== =================================================== 237 Field Content 238 ========================== =================================================== 239 Name filename of the executable 240 Umask file mode creation mask 241 State state (R is running, S is sleeping, D is sleeping 242 in an uninterruptible wait, Z is zombie, 243 T is traced or stopped) 244 Tgid thread group ID 245 Ngid NUMA group ID (0 if none) 246 Pid process id 247 PPid process id of the parent process 248 TracerPid PID of process tracing this process (0 if not) 249 Uid Real, effective, saved set, and file system UIDs 250 Gid Real, effective, saved set, and file system GIDs 251 FDSize number of file descriptor slots currently allocated 252 Groups supplementary group list 253 NStgid descendant namespace thread group ID hierarchy 254 NSpid descendant namespace process ID hierarchy 255 NSpgid descendant namespace process group ID hierarchy 256 NSsid descendant namespace session ID hierarchy 257 VmPeak peak virtual memory size 258 VmSize total program size 259 VmLck locked memory size 260 VmPin pinned memory size 261 VmHWM peak resident set size ("high water mark") 262 VmRSS size of memory portions. It contains the three 263 following parts 264 (VmRSS = RssAnon + RssFile + RssShmem) 265 RssAnon size of resident anonymous memory 266 RssFile size of resident file mappings 267 RssShmem size of resident shmem memory (includes SysV shm, 268 mapping of tmpfs and shared anonymous mappings) 269 VmData size of private data segments 270 VmStk size of stack segments 271 VmExe size of text segment 272 VmLib size of shared library code 273 VmPTE size of page table entries 274 VmSwap amount of swap used by anonymous private data 275 (shmem swap usage is not included) 276 HugetlbPages size of hugetlb memory portions 277 CoreDumping process's memory is currently being dumped 278 (killing the process may lead to a corrupted core) 279 THP_enabled process is allowed to use THP (returns 0 when 280 PR_SET_THP_DISABLE is set on the process 281 Threads number of threads 282 SigQ number of signals queued/max. number for queue 283 SigPnd bitmap of pending signals for the thread 284 ShdPnd bitmap of shared pending signals for the process 285 SigBlk bitmap of blocked signals 286 SigIgn bitmap of ignored signals 287 SigCgt bitmap of caught signals 288 CapInh bitmap of inheritable capabilities 289 CapPrm bitmap of permitted capabilities 290 CapEff bitmap of effective capabilities 291 CapBnd bitmap of capabilities bounding set 292 CapAmb bitmap of ambient capabilities 293 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) 294 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 295 Speculation_Store_Bypass speculative store bypass mitigation status 296 SpeculationIndirectBranch indirect branch speculation mode 297 Cpus_allowed mask of CPUs on which this process may run 298 Cpus_allowed_list Same as previous, but in "list format" 299 Mems_allowed mask of memory nodes allowed to this process 300 Mems_allowed_list Same as previous, but in "list format" 301 voluntary_ctxt_switches number of voluntary context switches 302 nonvoluntary_ctxt_switches number of non voluntary context switches 303 ========================== =================================================== 304 305 306.. table:: Table 1-3: Contents of the statm files (as of 2.6.8-rc3) 307 308 ======== =============================== ============================== 309 Field Content 310 ======== =============================== ============================== 311 size total program size (pages) (same as VmSize in status) 312 resident size of memory portions (pages) (same as VmRSS in status) 313 shared number of pages that are shared (i.e. backed by a file, same 314 as RssFile+RssShmem in status) 315 trs number of pages that are 'code' (not including libs; broken, 316 includes data segment) 317 lrs number of pages of library (always 0 on 2.6) 318 drs number of pages of data/stack (including libs; broken, 319 includes library text) 320 dt number of dirty pages (always 0 on 2.6) 321 ======== =============================== ============================== 322 323 324.. table:: Table 1-4: Contents of the stat files (as of 2.6.30-rc7) 325 326 ============= =============================================================== 327 Field Content 328 ============= =============================================================== 329 pid process id 330 tcomm filename of the executable 331 state state (R is running, S is sleeping, D is sleeping in an 332 uninterruptible wait, Z is zombie, T is traced or stopped) 333 ppid process id of the parent process 334 pgrp pgrp of the process 335 sid session id 336 tty_nr tty the process uses 337 tty_pgrp pgrp of the tty 338 flags task flags 339 min_flt number of minor faults 340 cmin_flt number of minor faults with child's 341 maj_flt number of major faults 342 cmaj_flt number of major faults with child's 343 utime user mode jiffies 344 stime kernel mode jiffies 345 cutime user mode jiffies with child's 346 cstime kernel mode jiffies with child's 347 priority priority level 348 nice nice level 349 num_threads number of threads 350 it_real_value (obsolete, always 0) 351 start_time time the process started after system boot 352 vsize virtual memory size 353 rss resident set memory size 354 rsslim current limit in bytes on the rss 355 start_code address above which program text can run 356 end_code address below which program text can run 357 start_stack address of the start of the main process stack 358 esp current value of ESP 359 eip current value of EIP 360 pending bitmap of pending signals 361 blocked bitmap of blocked signals 362 sigign bitmap of ignored signals 363 sigcatch bitmap of caught signals 364 0 (place holder, used to be the wchan address, 365 use /proc/PID/wchan instead) 366 0 (place holder) 367 0 (place holder) 368 exit_signal signal to send to parent thread on exit 369 task_cpu which CPU the task is scheduled on 370 rt_priority realtime priority 371 policy scheduling policy (man sched_setscheduler) 372 blkio_ticks time spent waiting for block IO 373 gtime guest time of the task in jiffies 374 cgtime guest time of the task children in jiffies 375 start_data address above which program data+bss is placed 376 end_data address below which program data+bss is placed 377 start_brk address above which program heap can be expanded with brk() 378 arg_start address above which program command line is placed 379 arg_end address below which program command line is placed 380 env_start address above which program environment is placed 381 env_end address below which program environment is placed 382 exit_code the thread's exit_code in the form reported by the waitpid 383 system call 384 ============= =============================================================== 385 386The /proc/PID/maps file contains the currently mapped memory regions and 387their access permissions. 388 389The format is:: 390 391 address perms offset dev inode pathname 392 393 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 394 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 395 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] 396 a7cb1000-a7cb2000 ---p 00000000 00:00 0 397 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 398 a7eb2000-a7eb3000 ---p 00000000 00:00 0 399 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 400 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 401 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 402 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 403 a800b000-a800e000 rw-p 00000000 00:00 0 404 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 405 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 406 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 407 a8024000-a8027000 rw-p 00000000 00:00 0 408 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 409 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 410 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 411 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 412 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 413 414where "address" is the address space in the process that it occupies, "perms" 415is a set of permissions:: 416 417 r = read 418 w = write 419 x = execute 420 s = shared 421 p = private (copy on write) 422 423"offset" is the offset into the mapping, "dev" is the device (major:minor), and 424"inode" is the inode on that device. 0 indicates that no inode is associated 425with the memory region, as the case would be with BSS (uninitialized data). 426The "pathname" shows the name associated file for this mapping. If the mapping 427is not associated with a file: 428 429 ============= ==================================== 430 [heap] the heap of the program 431 [stack] the stack of the main process 432 [vdso] the "virtual dynamic shared object", 433 the kernel system call handler 434 [anon:<name>] an anonymous mapping that has been 435 named by userspace 436 ============= ==================================== 437 438 or if empty, the mapping is anonymous. 439 440The /proc/PID/smaps is an extension based on maps, showing the memory 441consumption for each of the process's mappings. For each mapping (aka Virtual 442Memory Area, or VMA) there is a series of lines such as the following:: 443 444 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 445 446 Size: 1084 kB 447 KernelPageSize: 4 kB 448 MMUPageSize: 4 kB 449 Rss: 892 kB 450 Pss: 374 kB 451 Pss_Dirty: 0 kB 452 Shared_Clean: 892 kB 453 Shared_Dirty: 0 kB 454 Private_Clean: 0 kB 455 Private_Dirty: 0 kB 456 Referenced: 892 kB 457 Anonymous: 0 kB 458 LazyFree: 0 kB 459 AnonHugePages: 0 kB 460 ShmemPmdMapped: 0 kB 461 Shared_Hugetlb: 0 kB 462 Private_Hugetlb: 0 kB 463 Swap: 0 kB 464 SwapPss: 0 kB 465 KernelPageSize: 4 kB 466 MMUPageSize: 4 kB 467 Locked: 0 kB 468 THPeligible: 0 469 VmFlags: rd ex mr mw me dw 470 471The first of these lines shows the same information as is displayed for the 472mapping in /proc/PID/maps. Following lines show the size of the mapping 473(size); the size of each page allocated when backing a VMA (KernelPageSize), 474which is usually the same as the size in the page table entries; the page size 475used by the MMU when backing a VMA (in most cases, the same as KernelPageSize); 476the amount of the mapping that is currently resident in RAM (RSS); the 477process' proportional share of this mapping (PSS); and the number of clean and 478dirty shared and private pages in the mapping. 479 480The "proportional set size" (PSS) of a process is the count of pages it has 481in memory, where each page is divided by the number of processes sharing it. 482So if a process has 1000 pages all to itself, and 1000 shared with one other 483process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which 484consists of dirty pages. ("Pss_Clean" is not included, but it can be 485calculated by subtracting "Pss_Dirty" from "Pss".) 486 487Note that even a page which is part of a MAP_SHARED mapping, but has only 488a single pte mapped, i.e. is currently used by only one process, is accounted 489as private and not as shared. 490 491"Referenced" indicates the amount of memory currently marked as referenced or 492accessed. 493 494"Anonymous" shows the amount of memory that does not belong to any file. Even 495a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 496and a page is modified, the file page is replaced by a private anonymous copy. 497 498"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 499The memory isn't freed immediately with madvise(). It's freed in memory 500pressure if the memory is clean. Please note that the printed value might 501be lower than the real value due to optimizations used in the current 502implementation. If this is not desirable please file a bug report. 503 504"AnonHugePages" shows the ammount of memory backed by transparent hugepage. 505 506"ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by 507huge pages. 508 509"Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by 510hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 511reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 512 513"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 514 515For shmem mappings, "Swap" includes also the size of the mapped (and not 516replaced by copy-on-write) part of the underlying shmem object out on swap. 517"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 518does not take into account swapped out page of underlying shmem objects. 519"Locked" indicates whether the mapping is locked in memory or not. 520"THPeligible" indicates whether the mapping is eligible for allocating THP 521pages - 1 if true, 0 otherwise. It just shows the current status. 522 523"VmFlags" field deserves a separate description. This member represents the 524kernel flags associated with the particular virtual memory area in two letter 525encoded manner. The codes are the following: 526 527 == ======================================= 528 rd readable 529 wr writeable 530 ex executable 531 sh shared 532 mr may read 533 mw may write 534 me may execute 535 ms may share 536 gd stack segment growns down 537 pf pure PFN range 538 dw disabled write to the mapped file 539 lo pages are locked in memory 540 io memory mapped I/O area 541 sr sequential read advise provided 542 rr random read advise provided 543 dc do not copy area on fork 544 de do not expand area on remapping 545 ac area is accountable 546 nr swap space is not reserved for the area 547 ht area uses huge tlb pages 548 sf synchronous page fault 549 ar architecture specific flag 550 wf wipe on fork 551 dd do not include area into core dump 552 sd soft dirty flag 553 mm mixed map area 554 hg huge page advise flag 555 nh no huge page advise flag 556 mg mergable advise flag 557 bt arm64 BTI guarded page 558 mt arm64 MTE allocation tags are enabled 559 um userfaultfd missing tracking 560 uw userfaultfd wr-protect tracking 561 == ======================================= 562 563Note that there is no guarantee that every flag and associated mnemonic will 564be present in all further kernel releases. Things get changed, the flags may 565be vanished or the reverse -- new added. Interpretation of their meaning 566might change in future as well. So each consumer of these flags has to 567follow each specific kernel version for the exact semantic. 568 569This file is only present if the CONFIG_MMU kernel configuration option is 570enabled. 571 572Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 573output can be achieved only in the single read call). 574 575This typically manifests when doing partial reads of these files while the 576memory map is being modified. Despite the races, we do provide the following 577guarantees: 578 5791) The mapped addresses never go backwards, which implies no two 580 regions will ever overlap. 5812) If there is something at a given vaddr during the entirety of the 582 life of the smaps/maps walk, there will be some output for it. 583 584The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 585but their values are the sums of the corresponding values for all mappings of 586the process. Additionally, it contains these fields: 587 588- Pss_Anon 589- Pss_File 590- Pss_Shmem 591 592They represent the proportional shares of anonymous, file, and shmem pages, as 593described for smaps above. These fields are omitted in smaps since each 594mapping identifies the type (anon, file, or shmem) of all pages it contains. 595Thus all information in smaps_rollup can be derived from smaps, but at a 596significantly higher cost. 597 598The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 599bits on both physical and virtual pages associated with a process, and the 600soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 601for details). 602To clear the bits for all the pages associated with the process:: 603 604 > echo 1 > /proc/PID/clear_refs 605 606To clear the bits for the anonymous pages associated with the process:: 607 608 > echo 2 > /proc/PID/clear_refs 609 610To clear the bits for the file mapped pages associated with the process:: 611 612 > echo 3 > /proc/PID/clear_refs 613 614To clear the soft-dirty bit:: 615 616 > echo 4 > /proc/PID/clear_refs 617 618To reset the peak resident set size ("high water mark") to the process's 619current value:: 620 621 > echo 5 > /proc/PID/clear_refs 622 623Any other value written to /proc/PID/clear_refs will have no effect. 624 625The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 626using /proc/kpageflags and number of times a page is mapped using 627/proc/kpagecount. For detailed explanation, see 628Documentation/admin-guide/mm/pagemap.rst. 629 630The /proc/pid/numa_maps is an extension based on maps, showing the memory 631locality and binding policy, as well as the memory usage (in pages) of 632each mapping. The output follows a general format where mapping details get 633summarized separated by blank spaces, one mapping per each file line:: 634 635 address policy mapping details 636 637 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 638 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 639 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 640 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 641 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 642 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 643 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 644 320698b000 default file=/lib64/libc-2.12.so 645 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 646 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 647 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 648 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 649 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 650 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 651 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 652 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 653 654Where: 655 656"address" is the starting address for the mapping; 657 658"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 659 660"mapping details" summarizes mapping data such as mapping type, page usage counters, 661node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 662size, in KB, that is backing the mapping up. 663 6641.2 Kernel data 665--------------- 666 667Similar to the process entries, the kernel data files give information about 668the running kernel. The files used to obtain this information are contained in 669/proc and are listed in Table 1-5. Not all of these will be present in your 670system. It depends on the kernel configuration and the loaded modules, which 671files are there, and which are missing. 672 673.. table:: Table 1-5: Kernel info in /proc 674 675 ============ =============================================================== 676 File Content 677 ============ =============================================================== 678 apm Advanced power management info 679 buddyinfo Kernel memory allocator information (see text) (2.5) 680 bus Directory containing bus specific information 681 cmdline Kernel command line 682 cpuinfo Info about the CPU 683 devices Available devices (block and character) 684 dma Used DMS channels 685 filesystems Supported filesystems 686 driver Various drivers grouped here, currently rtc (2.4) 687 execdomains Execdomains, related to security (2.4) 688 fb Frame Buffer devices (2.4) 689 fs File system parameters, currently nfs/exports (2.4) 690 ide Directory containing info about the IDE subsystem 691 interrupts Interrupt usage 692 iomem Memory map (2.4) 693 ioports I/O port usage 694 irq Masks for irq to cpu affinity (2.4)(smp?) 695 isapnp ISA PnP (Plug&Play) Info (2.4) 696 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 697 kmsg Kernel messages 698 ksyms Kernel symbol table 699 loadavg Load average of last 1, 5 & 15 minutes; 700 number of processes currently runnable (running or on ready queue); 701 total number of processes in system; 702 last pid created. 703 All fields are separated by one space except "number of 704 processes currently runnable" and "total number of processes 705 in system", which are separated by a slash ('/'). Example: 706 0.61 0.61 0.55 3/828 22084 707 locks Kernel locks 708 meminfo Memory info 709 misc Miscellaneous 710 modules List of loaded modules 711 mounts Mounted filesystems 712 net Networking info (see text) 713 pagetypeinfo Additional page allocator information (see text) (2.5) 714 partitions Table of partitions known to the system 715 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 716 decoupled by lspci (2.4) 717 rtc Real time clock 718 scsi SCSI info (see text) 719 slabinfo Slab pool info 720 softirqs softirq usage 721 stat Overall statistics 722 swaps Swap space utilization 723 sys See chapter 2 724 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 725 tty Info of tty drivers 726 uptime Wall clock since boot, combined idle time of all cpus 727 version Kernel version 728 video bttv info of video resources (2.4) 729 vmallocinfo Show vmalloced areas 730 ============ =============================================================== 731 732You can, for example, check which interrupts are currently in use and what 733they are used for by looking in the file /proc/interrupts:: 734 735 > cat /proc/interrupts 736 CPU0 737 0: 8728810 XT-PIC timer 738 1: 895 XT-PIC keyboard 739 2: 0 XT-PIC cascade 740 3: 531695 XT-PIC aha152x 741 4: 2014133 XT-PIC serial 742 5: 44401 XT-PIC pcnet_cs 743 8: 2 XT-PIC rtc 744 11: 8 XT-PIC i82365 745 12: 182918 XT-PIC PS/2 Mouse 746 13: 1 XT-PIC fpu 747 14: 1232265 XT-PIC ide0 748 15: 7 XT-PIC ide1 749 NMI: 0 750 751In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 752output of a SMP machine):: 753 754 > cat /proc/interrupts 755 756 CPU0 CPU1 757 0: 1243498 1214548 IO-APIC-edge timer 758 1: 8949 8958 IO-APIC-edge keyboard 759 2: 0 0 XT-PIC cascade 760 5: 11286 10161 IO-APIC-edge soundblaster 761 8: 1 0 IO-APIC-edge rtc 762 9: 27422 27407 IO-APIC-edge 3c503 763 12: 113645 113873 IO-APIC-edge PS/2 Mouse 764 13: 0 0 XT-PIC fpu 765 14: 22491 24012 IO-APIC-edge ide0 766 15: 2183 2415 IO-APIC-edge ide1 767 17: 30564 30414 IO-APIC-level eth0 768 18: 177 164 IO-APIC-level bttv 769 NMI: 2457961 2457959 770 LOC: 2457882 2457881 771 ERR: 2155 772 773NMI is incremented in this case because every timer interrupt generates a NMI 774(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 775 776LOC is the local interrupt counter of the internal APIC of every CPU. 777 778ERR is incremented in the case of errors in the IO-APIC bus (the bus that 779connects the CPUs in a SMP system. This means that an error has been detected, 780the IO-APIC automatically retry the transmission, so it should not be a big 781problem, but you should read the SMP-FAQ. 782 783In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 784/proc/interrupts to display every IRQ vector in use by the system, not 785just those considered 'most important'. The new vectors are: 786 787THR 788 interrupt raised when a machine check threshold counter 789 (typically counting ECC corrected errors of memory or cache) exceeds 790 a configurable threshold. Only available on some systems. 791 792TRM 793 a thermal event interrupt occurs when a temperature threshold 794 has been exceeded for the CPU. This interrupt may also be generated 795 when the temperature drops back to normal. 796 797SPU 798 a spurious interrupt is some interrupt that was raised then lowered 799 by some IO device before it could be fully processed by the APIC. Hence 800 the APIC sees the interrupt but does not know what device it came from. 801 For this case the APIC will generate the interrupt with a IRQ vector 802 of 0xff. This might also be generated by chipset bugs. 803 804RES, CAL, TLB 805 rescheduling, call and TLB flush interrupts are 806 sent from one CPU to another per the needs of the OS. Typically, 807 their statistics are used by kernel developers and interested users to 808 determine the occurrence of interrupts of the given type. 809 810The above IRQ vectors are displayed only when relevant. For example, 811the threshold vector does not exist on x86_64 platforms. Others are 812suppressed when the system is a uniprocessor. As of this writing, only 813i386 and x86_64 platforms support the new IRQ vector displays. 814 815Of some interest is the introduction of the /proc/irq directory to 2.4. 816It could be used to set IRQ to CPU affinity. This means that you can "hook" an 817IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 818irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 819prof_cpu_mask. 820 821For example:: 822 823 > ls /proc/irq/ 824 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 825 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 826 > ls /proc/irq/0/ 827 smp_affinity 828 829smp_affinity is a bitmask, in which you can specify which CPUs can handle the 830IRQ. You can set it by doing:: 831 832 > echo 1 > /proc/irq/10/smp_affinity 833 834This means that only the first CPU will handle the IRQ, but you can also echo 8355 which means that only the first and third CPU can handle the IRQ. 836 837The contents of each smp_affinity file is the same by default:: 838 839 > cat /proc/irq/0/smp_affinity 840 ffffffff 841 842There is an alternate interface, smp_affinity_list which allows specifying 843a CPU range instead of a bitmask:: 844 845 > cat /proc/irq/0/smp_affinity_list 846 1024-1031 847 848The default_smp_affinity mask applies to all non-active IRQs, which are the 849IRQs which have not yet been allocated/activated, and hence which lack a 850/proc/irq/[0-9]* directory. 851 852The node file on an SMP system shows the node to which the device using the IRQ 853reports itself as being attached. This hardware locality information does not 854include information about any possible driver locality preference. 855 856prof_cpu_mask specifies which CPUs are to be profiled by the system wide 857profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 858 859The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 860between all the CPUs which are allowed to handle it. As usual the kernel has 861more info than you and does a better job than you, so the defaults are the 862best choice for almost everyone. [Note this applies only to those IO-APIC's 863that support "Round Robin" interrupt distribution.] 864 865There are three more important subdirectories in /proc: net, scsi, and sys. 866The general rule is that the contents, or even the existence of these 867directories, depend on your kernel configuration. If SCSI is not enabled, the 868directory scsi may not exist. The same is true with the net, which is there 869only when networking support is present in the running kernel. 870 871The slabinfo file gives information about memory usage at the slab level. 872Linux uses slab pools for memory management above page level in version 2.2. 873Commonly used objects have their own slab pool (such as network buffers, 874directory cache, and so on). 875 876:: 877 878 > cat /proc/buddyinfo 879 880 Node 0, zone DMA 0 4 5 4 4 3 ... 881 Node 0, zone Normal 1 0 0 1 101 8 ... 882 Node 0, zone HighMem 2 0 0 1 1 0 ... 883 884External fragmentation is a problem under some workloads, and buddyinfo is a 885useful tool for helping diagnose these problems. Buddyinfo will give you a 886clue as to how big an area you can safely allocate, or why a previous 887allocation failed. 888 889Each column represents the number of pages of a certain order which are 890available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 891ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 892available in ZONE_NORMAL, etc... 893 894More information relevant to external fragmentation can be found in 895pagetypeinfo:: 896 897 > cat /proc/pagetypeinfo 898 Page block order: 9 899 Pages per block: 512 900 901 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 902 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 903 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 904 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 905 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 906 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 907 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 908 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 909 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 910 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 911 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 912 913 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 914 Node 0, zone DMA 2 0 5 1 0 915 Node 0, zone DMA32 41 6 967 2 0 916 917Fragmentation avoidance in the kernel works by grouping pages of different 918migrate types into the same contiguous regions of memory called page blocks. 919A page block is typically the size of the default hugepage size, e.g. 2MB on 920X86-64. By keeping pages grouped based on their ability to move, the kernel 921can reclaim pages within a page block to satisfy a high-order allocation. 922 923The pagetypinfo begins with information on the size of a page block. It 924then gives the same type of information as buddyinfo except broken down 925by migrate-type and finishes with details on how many page blocks of each 926type exist. 927 928If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 929from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 930make an estimate of the likely number of huge pages that can be allocated 931at a given point in time. All the "Movable" blocks should be allocatable 932unless memory has been mlock()'d. Some of the Reclaimable blocks should 933also be allocatable although a lot of filesystem metadata may have to be 934reclaimed to achieve this. 935 936 937meminfo 938~~~~~~~ 939 940Provides information about distribution and utilization of memory. This 941varies by architecture and compile options. Some of the counters reported 942here overlap. The memory reported by the non overlapping counters may not 943add up to the overall memory usage and the difference for some workloads 944can be substantial. In many cases there are other means to find out 945additional memory using subsystem specific interfaces, for instance 946/proc/net/sockstat for TCP memory allocations. 947 948Example output. You may not have all of these fields. 949 950:: 951 952 > cat /proc/meminfo 953 954 MemTotal: 32858820 kB 955 MemFree: 21001236 kB 956 MemAvailable: 27214312 kB 957 Buffers: 581092 kB 958 Cached: 5587612 kB 959 SwapCached: 0 kB 960 Active: 3237152 kB 961 Inactive: 7586256 kB 962 Active(anon): 94064 kB 963 Inactive(anon): 4570616 kB 964 Active(file): 3143088 kB 965 Inactive(file): 3015640 kB 966 Unevictable: 0 kB 967 Mlocked: 0 kB 968 SwapTotal: 0 kB 969 SwapFree: 0 kB 970 Zswap: 1904 kB 971 Zswapped: 7792 kB 972 Dirty: 12 kB 973 Writeback: 0 kB 974 AnonPages: 4654780 kB 975 Mapped: 266244 kB 976 Shmem: 9976 kB 977 KReclaimable: 517708 kB 978 Slab: 660044 kB 979 SReclaimable: 517708 kB 980 SUnreclaim: 142336 kB 981 KernelStack: 11168 kB 982 PageTables: 20540 kB 983 NFS_Unstable: 0 kB 984 Bounce: 0 kB 985 WritebackTmp: 0 kB 986 CommitLimit: 16429408 kB 987 Committed_AS: 7715148 kB 988 VmallocTotal: 34359738367 kB 989 VmallocUsed: 40444 kB 990 VmallocChunk: 0 kB 991 Percpu: 29312 kB 992 HardwareCorrupted: 0 kB 993 AnonHugePages: 4149248 kB 994 ShmemHugePages: 0 kB 995 ShmemPmdMapped: 0 kB 996 FileHugePages: 0 kB 997 FilePmdMapped: 0 kB 998 CmaTotal: 0 kB 999 CmaFree: 0 kB 1000 HugePages_Total: 0 1001 HugePages_Free: 0 1002 HugePages_Rsvd: 0 1003 HugePages_Surp: 0 1004 Hugepagesize: 2048 kB 1005 Hugetlb: 0 kB 1006 DirectMap4k: 401152 kB 1007 DirectMap2M: 10008576 kB 1008 DirectMap1G: 24117248 kB 1009 1010MemTotal 1011 Total usable RAM (i.e. physical RAM minus a few reserved 1012 bits and the kernel binary code) 1013MemFree 1014 Total free RAM. On highmem systems, the sum of LowFree+HighFree 1015MemAvailable 1016 An estimate of how much memory is available for starting new 1017 applications, without swapping. Calculated from MemFree, 1018 SReclaimable, the size of the file LRU lists, and the low 1019 watermarks in each zone. 1020 The estimate takes into account that the system needs some 1021 page cache to function well, and that not all reclaimable 1022 slab will be reclaimable, due to items being in use. The 1023 impact of those factors will vary from system to system. 1024Buffers 1025 Relatively temporary storage for raw disk blocks 1026 shouldn't get tremendously large (20MB or so) 1027Cached 1028 In-memory cache for files read from the disk (the 1029 pagecache) as well as tmpfs & shmem. 1030 Doesn't include SwapCached. 1031SwapCached 1032 Memory that once was swapped out, is swapped back in but 1033 still also is in the swapfile (if memory is needed it 1034 doesn't need to be swapped out AGAIN because it is already 1035 in the swapfile. This saves I/O) 1036Active 1037 Memory that has been used more recently and usually not 1038 reclaimed unless absolutely necessary. 1039Inactive 1040 Memory which has been less recently used. It is more 1041 eligible to be reclaimed for other purposes 1042Unevictable 1043 Memory allocated for userspace which cannot be reclaimed, such 1044 as mlocked pages, ramfs backing pages, secret memfd pages etc. 1045Mlocked 1046 Memory locked with mlock(). 1047HighTotal, HighFree 1048 Highmem is all memory above ~860MB of physical memory. 1049 Highmem areas are for use by userspace programs, or 1050 for the pagecache. The kernel must use tricks to access 1051 this memory, making it slower to access than lowmem. 1052LowTotal, LowFree 1053 Lowmem is memory which can be used for everything that 1054 highmem can be used for, but it is also available for the 1055 kernel's use for its own data structures. Among many 1056 other things, it is where everything from the Slab is 1057 allocated. Bad things happen when you're out of lowmem. 1058SwapTotal 1059 total amount of swap space available 1060SwapFree 1061 Memory which has been evicted from RAM, and is temporarily 1062 on the disk 1063Zswap 1064 Memory consumed by the zswap backend (compressed size) 1065Zswapped 1066 Amount of anonymous memory stored in zswap (original size) 1067Dirty 1068 Memory which is waiting to get written back to the disk 1069Writeback 1070 Memory which is actively being written back to the disk 1071AnonPages 1072 Non-file backed pages mapped into userspace page tables 1073Mapped 1074 files which have been mmaped, such as libraries 1075Shmem 1076 Total memory used by shared memory (shmem) and tmpfs 1077KReclaimable 1078 Kernel allocations that the kernel will attempt to reclaim 1079 under memory pressure. Includes SReclaimable (below), and other 1080 direct allocations with a shrinker. 1081Slab 1082 in-kernel data structures cache 1083SReclaimable 1084 Part of Slab, that might be reclaimed, such as caches 1085SUnreclaim 1086 Part of Slab, that cannot be reclaimed on memory pressure 1087KernelStack 1088 Memory consumed by the kernel stacks of all tasks 1089PageTables 1090 Memory consumed by userspace page tables 1091NFS_Unstable 1092 Always zero. Previous counted pages which had been written to 1093 the server, but has not been committed to stable storage. 1094Bounce 1095 Memory used for block device "bounce buffers" 1096WritebackTmp 1097 Memory used by FUSE for temporary writeback buffers 1098CommitLimit 1099 Based on the overcommit ratio ('vm.overcommit_ratio'), 1100 this is the total amount of memory currently available to 1101 be allocated on the system. This limit is only adhered to 1102 if strict overcommit accounting is enabled (mode 2 in 1103 'vm.overcommit_memory'). 1104 1105 The CommitLimit is calculated with the following formula:: 1106 1107 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1108 overcommit_ratio / 100 + [total swap pages] 1109 1110 For example, on a system with 1G of physical RAM and 7G 1111 of swap with a `vm.overcommit_ratio` of 30 it would 1112 yield a CommitLimit of 7.3G. 1113 1114 For more details, see the memory overcommit documentation 1115 in mm/overcommit-accounting. 1116Committed_AS 1117 The amount of memory presently allocated on the system. 1118 The committed memory is a sum of all of the memory which 1119 has been allocated by processes, even if it has not been 1120 "used" by them as of yet. A process which malloc()'s 1G 1121 of memory, but only touches 300M of it will show up as 1122 using 1G. This 1G is memory which has been "committed" to 1123 by the VM and can be used at any time by the allocating 1124 application. With strict overcommit enabled on the system 1125 (mode 2 in 'vm.overcommit_memory'), allocations which would 1126 exceed the CommitLimit (detailed above) will not be permitted. 1127 This is useful if one needs to guarantee that processes will 1128 not fail due to lack of memory once that memory has been 1129 successfully allocated. 1130VmallocTotal 1131 total size of vmalloc virtual address space 1132VmallocUsed 1133 amount of vmalloc area which is used 1134VmallocChunk 1135 largest contiguous block of vmalloc area which is free 1136Percpu 1137 Memory allocated to the percpu allocator used to back percpu 1138 allocations. This stat excludes the cost of metadata. 1139HardwareCorrupted 1140 The amount of RAM/memory in KB, the kernel identifies as 1141 corrupted. 1142AnonHugePages 1143 Non-file backed huge pages mapped into userspace page tables 1144ShmemHugePages 1145 Memory used by shared memory (shmem) and tmpfs allocated 1146 with huge pages 1147ShmemPmdMapped 1148 Shared memory mapped into userspace with huge pages 1149FileHugePages 1150 Memory used for filesystem data (page cache) allocated 1151 with huge pages 1152FilePmdMapped 1153 Page cache mapped into userspace with huge pages 1154CmaTotal 1155 Memory reserved for the Contiguous Memory Allocator (CMA) 1156CmaFree 1157 Free remaining memory in the CMA reserves 1158HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb 1159 See Documentation/admin-guide/mm/hugetlbpage.rst. 1160DirectMap4k, DirectMap2M, DirectMap1G 1161 Breakdown of page table sizes used in the kernel's 1162 identity mapping of RAM 1163 1164vmallocinfo 1165~~~~~~~~~~~ 1166 1167Provides information about vmalloced/vmaped areas. One line per area, 1168containing the virtual address range of the area, size in bytes, 1169caller information of the creator, and optional information depending 1170on the kind of area: 1171 1172 ========== =================================================== 1173 pages=nr number of pages 1174 phys=addr if a physical address was specified 1175 ioremap I/O mapping (ioremap() and friends) 1176 vmalloc vmalloc() area 1177 vmap vmap()ed pages 1178 user VM_USERMAP area 1179 vpages buffer for pages pointers was vmalloced (huge area) 1180 N<node>=nr (Only on NUMA kernels) 1181 Number of pages allocated on memory node <node> 1182 ========== =================================================== 1183 1184:: 1185 1186 > cat /proc/vmallocinfo 1187 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1188 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1189 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1190 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1191 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1192 phys=7fee8000 ioremap 1193 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1194 phys=7fee7000 ioremap 1195 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1196 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1197 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1198 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1199 pages=2 vmalloc N1=2 1200 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1201 /0x130 [x_tables] pages=4 vmalloc N0=4 1202 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1203 pages=14 vmalloc N2=14 1204 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1205 pages=4 vmalloc N1=4 1206 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1207 pages=2 vmalloc N1=2 1208 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1209 pages=10 vmalloc N0=10 1210 1211 1212softirqs 1213~~~~~~~~ 1214 1215Provides counts of softirq handlers serviced since boot time, for each CPU. 1216 1217:: 1218 1219 > cat /proc/softirqs 1220 CPU0 CPU1 CPU2 CPU3 1221 HI: 0 0 0 0 1222 TIMER: 27166 27120 27097 27034 1223 NET_TX: 0 0 0 17 1224 NET_RX: 42 0 0 39 1225 BLOCK: 0 0 107 1121 1226 TASKLET: 0 0 0 290 1227 SCHED: 27035 26983 26971 26746 1228 HRTIMER: 0 0 0 0 1229 RCU: 1678 1769 2178 2250 1230 12311.3 Networking info in /proc/net 1232-------------------------------- 1233 1234The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1235additional values you get for IP version 6 if you configure the kernel to 1236support this. Table 1-9 lists the files and their meaning. 1237 1238 1239.. table:: Table 1-8: IPv6 info in /proc/net 1240 1241 ========== ===================================================== 1242 File Content 1243 ========== ===================================================== 1244 udp6 UDP sockets (IPv6) 1245 tcp6 TCP sockets (IPv6) 1246 raw6 Raw device statistics (IPv6) 1247 igmp6 IP multicast addresses, which this host joined (IPv6) 1248 if_inet6 List of IPv6 interface addresses 1249 ipv6_route Kernel routing table for IPv6 1250 rt6_stats Global IPv6 routing tables statistics 1251 sockstat6 Socket statistics (IPv6) 1252 snmp6 Snmp data (IPv6) 1253 ========== ===================================================== 1254 1255.. table:: Table 1-9: Network info in /proc/net 1256 1257 ============= ================================================================ 1258 File Content 1259 ============= ================================================================ 1260 arp Kernel ARP table 1261 dev network devices with statistics 1262 dev_mcast the Layer2 multicast groups a device is listening too 1263 (interface index, label, number of references, number of bound 1264 addresses). 1265 dev_stat network device status 1266 ip_fwchains Firewall chain linkage 1267 ip_fwnames Firewall chain names 1268 ip_masq Directory containing the masquerading tables 1269 ip_masquerade Major masquerading table 1270 netstat Network statistics 1271 raw raw device statistics 1272 route Kernel routing table 1273 rpc Directory containing rpc info 1274 rt_cache Routing cache 1275 snmp SNMP data 1276 sockstat Socket statistics 1277 tcp TCP sockets 1278 udp UDP sockets 1279 unix UNIX domain sockets 1280 wireless Wireless interface data (Wavelan etc) 1281 igmp IP multicast addresses, which this host joined 1282 psched Global packet scheduler parameters. 1283 netlink List of PF_NETLINK sockets 1284 ip_mr_vifs List of multicast virtual interfaces 1285 ip_mr_cache List of multicast routing cache 1286 ============= ================================================================ 1287 1288You can use this information to see which network devices are available in 1289your system and how much traffic was routed over those devices:: 1290 1291 > cat /proc/net/dev 1292 Inter-|Receive |[... 1293 face |bytes packets errs drop fifo frame compressed multicast|[... 1294 lo: 908188 5596 0 0 0 0 0 0 [... 1295 ppp0:15475140 20721 410 0 0 410 0 0 [... 1296 eth0: 614530 7085 0 0 0 0 0 1 [... 1297 1298 ...] Transmit 1299 ...] bytes packets errs drop fifo colls carrier compressed 1300 ...] 908188 5596 0 0 0 0 0 0 1301 ...] 1375103 17405 0 0 0 0 0 0 1302 ...] 1703981 5535 0 0 0 3 0 0 1303 1304In addition, each Channel Bond interface has its own directory. For 1305example, the bond0 device will have a directory called /proc/net/bond0/. 1306It will contain information that is specific to that bond, such as the 1307current slaves of the bond, the link status of the slaves, and how 1308many times the slaves link has failed. 1309 13101.4 SCSI info 1311------------- 1312 1313If you have a SCSI host adapter in your system, you'll find a subdirectory 1314named after the driver for this adapter in /proc/scsi. You'll also see a list 1315of all recognized SCSI devices in /proc/scsi:: 1316 1317 >cat /proc/scsi/scsi 1318 Attached devices: 1319 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1320 Vendor: IBM Model: DGHS09U Rev: 03E0 1321 Type: Direct-Access ANSI SCSI revision: 03 1322 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1323 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1324 Type: CD-ROM ANSI SCSI revision: 02 1325 1326 1327The directory named after the driver has one file for each adapter found in 1328the system. These files contain information about the controller, including 1329the used IRQ and the IO address range. The amount of information shown is 1330dependent on the adapter you use. The example shows the output for an Adaptec 1331AHA-2940 SCSI adapter:: 1332 1333 > cat /proc/scsi/aic7xxx/0 1334 1335 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1336 Compile Options: 1337 TCQ Enabled By Default : Disabled 1338 AIC7XXX_PROC_STATS : Disabled 1339 AIC7XXX_RESET_DELAY : 5 1340 Adapter Configuration: 1341 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1342 Ultra Wide Controller 1343 PCI MMAPed I/O Base: 0xeb001000 1344 Adapter SEEPROM Config: SEEPROM found and used. 1345 Adaptec SCSI BIOS: Enabled 1346 IRQ: 10 1347 SCBs: Active 0, Max Active 2, 1348 Allocated 15, HW 16, Page 255 1349 Interrupts: 160328 1350 BIOS Control Word: 0x18b6 1351 Adapter Control Word: 0x005b 1352 Extended Translation: Enabled 1353 Disconnect Enable Flags: 0xffff 1354 Ultra Enable Flags: 0x0001 1355 Tag Queue Enable Flags: 0x0000 1356 Ordered Queue Tag Flags: 0x0000 1357 Default Tag Queue Depth: 8 1358 Tagged Queue By Device array for aic7xxx host instance 0: 1359 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1360 Actual queue depth per device for aic7xxx host instance 0: 1361 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1362 Statistics: 1363 (scsi0:0:0:0) 1364 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1365 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1366 Total transfers 160151 (74577 reads and 85574 writes) 1367 (scsi0:0:6:0) 1368 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1369 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1370 Total transfers 0 (0 reads and 0 writes) 1371 1372 13731.5 Parallel port info in /proc/parport 1374--------------------------------------- 1375 1376The directory /proc/parport contains information about the parallel ports of 1377your system. It has one subdirectory for each port, named after the port 1378number (0,1,2,...). 1379 1380These directories contain the four files shown in Table 1-10. 1381 1382 1383.. table:: Table 1-10: Files in /proc/parport 1384 1385 ========= ==================================================================== 1386 File Content 1387 ========= ==================================================================== 1388 autoprobe Any IEEE-1284 device ID information that has been acquired. 1389 devices list of the device drivers using that port. A + will appear by the 1390 name of the device currently using the port (it might not appear 1391 against any). 1392 hardware Parallel port's base address, IRQ line and DMA channel. 1393 irq IRQ that parport is using for that port. This is in a separate 1394 file to allow you to alter it by writing a new value in (IRQ 1395 number or none). 1396 ========= ==================================================================== 1397 13981.6 TTY info in /proc/tty 1399------------------------- 1400 1401Information about the available and actually used tty's can be found in the 1402directory /proc/tty. You'll find entries for drivers and line disciplines in 1403this directory, as shown in Table 1-11. 1404 1405 1406.. table:: Table 1-11: Files in /proc/tty 1407 1408 ============= ============================================== 1409 File Content 1410 ============= ============================================== 1411 drivers list of drivers and their usage 1412 ldiscs registered line disciplines 1413 driver/serial usage statistic and status of single tty lines 1414 ============= ============================================== 1415 1416To see which tty's are currently in use, you can simply look into the file 1417/proc/tty/drivers:: 1418 1419 > cat /proc/tty/drivers 1420 pty_slave /dev/pts 136 0-255 pty:slave 1421 pty_master /dev/ptm 128 0-255 pty:master 1422 pty_slave /dev/ttyp 3 0-255 pty:slave 1423 pty_master /dev/pty 2 0-255 pty:master 1424 serial /dev/cua 5 64-67 serial:callout 1425 serial /dev/ttyS 4 64-67 serial 1426 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1427 /dev/ptmx /dev/ptmx 5 2 system 1428 /dev/console /dev/console 5 1 system:console 1429 /dev/tty /dev/tty 5 0 system:/dev/tty 1430 unknown /dev/tty 4 1-63 console 1431 1432 14331.7 Miscellaneous kernel statistics in /proc/stat 1434------------------------------------------------- 1435 1436Various pieces of information about kernel activity are available in the 1437/proc/stat file. All of the numbers reported in this file are aggregates 1438since the system first booted. For a quick look, simply cat the file:: 1439 1440 > cat /proc/stat 1441 cpu 2255 34 2290 22625563 6290 127 456 0 0 0 1442 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0 1443 cpu1 1123 0 849 11313845 2614 0 18 0 0 0 1444 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...] 1445 ctxt 1990473 1446 btime 1062191376 1447 processes 2915 1448 procs_running 1 1449 procs_blocked 0 1450 softirq 183433 0 21755 12 39 1137 231 21459 2263 1451 1452The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1453lines. These numbers identify the amount of time the CPU has spent performing 1454different kinds of work. Time units are in USER_HZ (typically hundredths of a 1455second). The meanings of the columns are as follows, from left to right: 1456 1457- user: normal processes executing in user mode 1458- nice: niced processes executing in user mode 1459- system: processes executing in kernel mode 1460- idle: twiddling thumbs 1461- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1462 are several problems: 1463 1464 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1465 waiting for I/O to complete. When CPU goes into idle state for 1466 outstanding task I/O, another task will be scheduled on this CPU. 1467 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1468 on any CPU, so the iowait of each CPU is difficult to calculate. 1469 3. The value of iowait field in /proc/stat will decrease in certain 1470 conditions. 1471 1472 So, the iowait is not reliable by reading from /proc/stat. 1473- irq: servicing interrupts 1474- softirq: servicing softirqs 1475- steal: involuntary wait 1476- guest: running a normal guest 1477- guest_nice: running a niced guest 1478 1479The "intr" line gives counts of interrupts serviced since boot time, for each 1480of the possible system interrupts. The first column is the total of all 1481interrupts serviced including unnumbered architecture specific interrupts; 1482each subsequent column is the total for that particular numbered interrupt. 1483Unnumbered interrupts are not shown, only summed into the total. 1484 1485The "ctxt" line gives the total number of context switches across all CPUs. 1486 1487The "btime" line gives the time at which the system booted, in seconds since 1488the Unix epoch. 1489 1490The "processes" line gives the number of processes and threads created, which 1491includes (but is not limited to) those created by calls to the fork() and 1492clone() system calls. 1493 1494The "procs_running" line gives the total number of threads that are 1495running or ready to run (i.e., the total number of runnable threads). 1496 1497The "procs_blocked" line gives the number of processes currently blocked, 1498waiting for I/O to complete. 1499 1500The "softirq" line gives counts of softirqs serviced since boot time, for each 1501of the possible system softirqs. The first column is the total of all 1502softirqs serviced; each subsequent column is the total for that particular 1503softirq. 1504 1505 15061.8 Ext4 file system parameters 1507------------------------------- 1508 1509Information about mounted ext4 file systems can be found in 1510/proc/fs/ext4. Each mounted filesystem will have a directory in 1511/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1512/proc/fs/ext4/dm-0). The files in each per-device directory are shown 1513in Table 1-12, below. 1514 1515.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1516 1517 ============== ========================================================== 1518 File Content 1519 mb_groups details of multiblock allocator buddy cache of free blocks 1520 ============== ========================================================== 1521 15221.9 /proc/consoles 1523------------------- 1524Shows registered system console lines. 1525 1526To see which character device lines are currently used for the system console 1527/dev/console, you may simply look into the file /proc/consoles:: 1528 1529 > cat /proc/consoles 1530 tty0 -WU (ECp) 4:7 1531 ttyS0 -W- (Ep) 4:64 1532 1533The columns are: 1534 1535+--------------------+-------------------------------------------------------+ 1536| device | name of the device | 1537+====================+=======================================================+ 1538| operations | * R = can do read operations | 1539| | * W = can do write operations | 1540| | * U = can do unblank | 1541+--------------------+-------------------------------------------------------+ 1542| flags | * E = it is enabled | 1543| | * C = it is preferred console | 1544| | * B = it is primary boot console | 1545| | * p = it is used for printk buffer | 1546| | * b = it is not a TTY but a Braille device | 1547| | * a = it is safe to use when cpu is offline | 1548+--------------------+-------------------------------------------------------+ 1549| major:minor | major and minor number of the device separated by a | 1550| | colon | 1551+--------------------+-------------------------------------------------------+ 1552 1553Summary 1554------- 1555 1556The /proc file system serves information about the running system. It not only 1557allows access to process data but also allows you to request the kernel status 1558by reading files in the hierarchy. 1559 1560The directory structure of /proc reflects the types of information and makes 1561it easy, if not obvious, where to look for specific data. 1562 1563Chapter 2: Modifying System Parameters 1564====================================== 1565 1566In This Chapter 1567--------------- 1568 1569* Modifying kernel parameters by writing into files found in /proc/sys 1570* Exploring the files which modify certain parameters 1571* Review of the /proc/sys file tree 1572 1573------------------------------------------------------------------------------ 1574 1575A very interesting part of /proc is the directory /proc/sys. This is not only 1576a source of information, it also allows you to change parameters within the 1577kernel. Be very careful when attempting this. You can optimize your system, 1578but you can also cause it to crash. Never alter kernel parameters on a 1579production system. Set up a development machine and test to make sure that 1580everything works the way you want it to. You may have no alternative but to 1581reboot the machine once an error has been made. 1582 1583To change a value, simply echo the new value into the file. 1584You need to be root to do this. You can create your own boot script 1585to perform this every time your system boots. 1586 1587The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1588general things in the operation of the Linux kernel. Since some of the files 1589can inadvertently disrupt your system, it is advisable to read both 1590documentation and source before actually making adjustments. In any case, be 1591very careful when writing to any of these files. The entries in /proc may 1592change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1593review the kernel documentation in the directory /usr/src/linux/Documentation. 1594This chapter is heavily based on the documentation included in the pre 2.2 1595kernels, and became part of it in version 2.2.1 of the Linux kernel. 1596 1597Please see: Documentation/admin-guide/sysctl/ directory for descriptions of these 1598entries. 1599 1600Summary 1601------- 1602 1603Certain aspects of kernel behavior can be modified at runtime, without the 1604need to recompile the kernel, or even to reboot the system. The files in the 1605/proc/sys tree can not only be read, but also modified. You can use the echo 1606command to write value into these files, thereby changing the default settings 1607of the kernel. 1608 1609 1610Chapter 3: Per-process Parameters 1611================================= 1612 16133.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1614-------------------------------------------------------------------------------- 1615 1616These files can be used to adjust the badness heuristic used to select which 1617process gets killed in out of memory (oom) conditions. 1618 1619The badness heuristic assigns a value to each candidate task ranging from 0 1620(never kill) to 1000 (always kill) to determine which process is targeted. The 1621units are roughly a proportion along that range of allowed memory the process 1622may allocate from based on an estimation of its current memory and swap use. 1623For example, if a task is using all allowed memory, its badness score will be 16241000. If it is using half of its allowed memory, its score will be 500. 1625 1626The amount of "allowed" memory depends on the context in which the oom killer 1627was called. If it is due to the memory assigned to the allocating task's cpuset 1628being exhausted, the allowed memory represents the set of mems assigned to that 1629cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1630memory represents the set of mempolicy nodes. If it is due to a memory 1631limit (or swap limit) being reached, the allowed memory is that configured 1632limit. Finally, if it is due to the entire system being out of memory, the 1633allowed memory represents all allocatable resources. 1634 1635The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1636is used to determine which task to kill. Acceptable values range from -1000 1637(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1638polarize the preference for oom killing either by always preferring a certain 1639task or completely disabling it. The lowest possible value, -1000, is 1640equivalent to disabling oom killing entirely for that task since it will always 1641report a badness score of 0. 1642 1643Consequently, it is very simple for userspace to define the amount of memory to 1644consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1645example, is roughly equivalent to allowing the remainder of tasks sharing the 1646same system, cpuset, mempolicy, or memory controller resources to use at least 164750% more memory. A value of -500, on the other hand, would be roughly 1648equivalent to discounting 50% of the task's allowed memory from being considered 1649as scoring against the task. 1650 1651For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1652be used to tune the badness score. Its acceptable values range from -16 1653(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1654(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1655scaled linearly with /proc/<pid>/oom_score_adj. 1656 1657The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1658value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1659requires CAP_SYS_RESOURCE. 1660 1661 16623.2 /proc/<pid>/oom_score - Display current oom-killer score 1663------------------------------------------------------------- 1664 1665This file can be used to check the current score used by the oom-killer for 1666any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1667process should be killed in an out-of-memory situation. 1668 1669Please note that the exported value includes oom_score_adj so it is 1670effectively in range [0,2000]. 1671 1672 16733.3 /proc/<pid>/io - Display the IO accounting fields 1674------------------------------------------------------- 1675 1676This file contains IO statistics for each running process. 1677 1678Example 1679~~~~~~~ 1680 1681:: 1682 1683 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1684 [1] 3828 1685 1686 test:/tmp # cat /proc/3828/io 1687 rchar: 323934931 1688 wchar: 323929600 1689 syscr: 632687 1690 syscw: 632675 1691 read_bytes: 0 1692 write_bytes: 323932160 1693 cancelled_write_bytes: 0 1694 1695 1696Description 1697~~~~~~~~~~~ 1698 1699rchar 1700^^^^^ 1701 1702I/O counter: chars read 1703The number of bytes which this task has caused to be read from storage. This 1704is simply the sum of bytes which this process passed to read() and pread(). 1705It includes things like tty IO and it is unaffected by whether or not actual 1706physical disk IO was required (the read might have been satisfied from 1707pagecache). 1708 1709 1710wchar 1711^^^^^ 1712 1713I/O counter: chars written 1714The number of bytes which this task has caused, or shall cause to be written 1715to disk. Similar caveats apply here as with rchar. 1716 1717 1718syscr 1719^^^^^ 1720 1721I/O counter: read syscalls 1722Attempt to count the number of read I/O operations, i.e. syscalls like read() 1723and pread(). 1724 1725 1726syscw 1727^^^^^ 1728 1729I/O counter: write syscalls 1730Attempt to count the number of write I/O operations, i.e. syscalls like 1731write() and pwrite(). 1732 1733 1734read_bytes 1735^^^^^^^^^^ 1736 1737I/O counter: bytes read 1738Attempt to count the number of bytes which this process really did cause to 1739be fetched from the storage layer. Done at the submit_bio() level, so it is 1740accurate for block-backed filesystems. <please add status regarding NFS and 1741CIFS at a later time> 1742 1743 1744write_bytes 1745^^^^^^^^^^^ 1746 1747I/O counter: bytes written 1748Attempt to count the number of bytes which this process caused to be sent to 1749the storage layer. This is done at page-dirtying time. 1750 1751 1752cancelled_write_bytes 1753^^^^^^^^^^^^^^^^^^^^^ 1754 1755The big inaccuracy here is truncate. If a process writes 1MB to a file and 1756then deletes the file, it will in fact perform no writeout. But it will have 1757been accounted as having caused 1MB of write. 1758In other words: The number of bytes which this process caused to not happen, 1759by truncating pagecache. A task can cause "negative" IO too. If this task 1760truncates some dirty pagecache, some IO which another task has been accounted 1761for (in its write_bytes) will not be happening. We _could_ just subtract that 1762from the truncating task's write_bytes, but there is information loss in doing 1763that. 1764 1765 1766.. Note:: 1767 1768 At its current implementation state, this is a bit racy on 32-bit machines: 1769 if process A reads process B's /proc/pid/io while process B is updating one 1770 of those 64-bit counters, process A could see an intermediate result. 1771 1772 1773More information about this can be found within the taskstats documentation in 1774Documentation/accounting. 1775 17763.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1777--------------------------------------------------------------- 1778When a process is dumped, all anonymous memory is written to a core file as 1779long as the size of the core file isn't limited. But sometimes we don't want 1780to dump some memory segments, for example, huge shared memory or DAX. 1781Conversely, sometimes we want to save file-backed memory segments into a core 1782file, not only the individual files. 1783 1784/proc/<pid>/coredump_filter allows you to customize which memory segments 1785will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1786of memory types. If a bit of the bitmask is set, memory segments of the 1787corresponding memory type are dumped, otherwise they are not dumped. 1788 1789The following 9 memory types are supported: 1790 1791 - (bit 0) anonymous private memory 1792 - (bit 1) anonymous shared memory 1793 - (bit 2) file-backed private memory 1794 - (bit 3) file-backed shared memory 1795 - (bit 4) ELF header pages in file-backed private memory areas (it is 1796 effective only if the bit 2 is cleared) 1797 - (bit 5) hugetlb private memory 1798 - (bit 6) hugetlb shared memory 1799 - (bit 7) DAX private memory 1800 - (bit 8) DAX shared memory 1801 1802 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1803 are always dumped regardless of the bitmask status. 1804 1805 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1806 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1807 1808The default value of coredump_filter is 0x33; this means all anonymous memory 1809segments, ELF header pages and hugetlb private memory are dumped. 1810 1811If you don't want to dump all shared memory segments attached to pid 1234, 1812write 0x31 to the process's proc file:: 1813 1814 $ echo 0x31 > /proc/1234/coredump_filter 1815 1816When a new process is created, the process inherits the bitmask status from its 1817parent. It is useful to set up coredump_filter before the program runs. 1818For example:: 1819 1820 $ echo 0x7 > /proc/self/coredump_filter 1821 $ ./some_program 1822 18233.5 /proc/<pid>/mountinfo - Information about mounts 1824-------------------------------------------------------- 1825 1826This file contains lines of the form:: 1827 1828 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1829 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4) 1830 1831 (1) mount ID: unique identifier of the mount (may be reused after umount) 1832 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1833 (3) major:minor: value of st_dev for files on filesystem 1834 (4) root: root of the mount within the filesystem 1835 (5) mount point: mount point relative to the process's root 1836 (6) mount options: per mount options 1837 (n…m) optional fields: zero or more fields of the form "tag[:value]" 1838 (m+1) separator: marks the end of the optional fields 1839 (m+2) filesystem type: name of filesystem of the form "type[.subtype]" 1840 (m+3) mount source: filesystem specific information or "none" 1841 (m+4) super options: per super block options 1842 1843Parsers should ignore all unrecognised optional fields. Currently the 1844possible optional fields are: 1845 1846================ ============================================================== 1847shared:X mount is shared in peer group X 1848master:X mount is slave to peer group X 1849propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1850unbindable mount is unbindable 1851================ ============================================================== 1852 1853.. [#] X is the closest dominant peer group under the process's root. If 1854 X is the immediate master of the mount, or if there's no dominant peer 1855 group under the same root, then only the "master:X" field is present 1856 and not the "propagate_from:X" field. 1857 1858For more information on mount propagation see: 1859 1860 Documentation/filesystems/sharedsubtree.rst 1861 1862 18633.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1864-------------------------------------------------------- 1865These files provide a method to access a task's comm value. It also allows for 1866a task to set its own or one of its thread siblings comm value. The comm value 1867is limited in size compared to the cmdline value, so writing anything longer 1868then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated 1869comm value. 1870 1871 18723.7 /proc/<pid>/task/<tid>/children - Information about task children 1873------------------------------------------------------------------------- 1874This file provides a fast way to retrieve first level children pids 1875of a task pointed by <pid>/<tid> pair. The format is a space separated 1876stream of pids. 1877 1878Note the "first level" here -- if a child has its own children they will 1879not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1880to obtain the descendants. 1881 1882Since this interface is intended to be fast and cheap it doesn't 1883guarantee to provide precise results and some children might be 1884skipped, especially if they've exited right after we printed their 1885pids, so one needs to either stop or freeze processes being inspected 1886if precise results are needed. 1887 1888 18893.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1890--------------------------------------------------------------- 1891This file provides information associated with an opened file. The regular 1892files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. 1893The 'pos' represents the current offset of the opened file in decimal 1894form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the 1895file has been created with [see open(2) for details] and 'mnt_id' represents 1896mount ID of the file system containing the opened file [see 3.5 1897/proc/<pid>/mountinfo for details]. 'ino' represents the inode number of 1898the file. 1899 1900A typical output is:: 1901 1902 pos: 0 1903 flags: 0100002 1904 mnt_id: 19 1905 ino: 63107 1906 1907All locks associated with a file descriptor are shown in its fdinfo too:: 1908 1909 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1910 1911The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1912pair provide additional information particular to the objects they represent. 1913 1914Eventfd files 1915~~~~~~~~~~~~~ 1916 1917:: 1918 1919 pos: 0 1920 flags: 04002 1921 mnt_id: 9 1922 ino: 63107 1923 eventfd-count: 5a 1924 1925where 'eventfd-count' is hex value of a counter. 1926 1927Signalfd files 1928~~~~~~~~~~~~~~ 1929 1930:: 1931 1932 pos: 0 1933 flags: 04002 1934 mnt_id: 9 1935 ino: 63107 1936 sigmask: 0000000000000200 1937 1938where 'sigmask' is hex value of the signal mask associated 1939with a file. 1940 1941Epoll files 1942~~~~~~~~~~~ 1943 1944:: 1945 1946 pos: 0 1947 flags: 02 1948 mnt_id: 9 1949 ino: 63107 1950 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 1951 1952where 'tfd' is a target file descriptor number in decimal form, 1953'events' is events mask being watched and the 'data' is data 1954associated with a target [see epoll(7) for more details]. 1955 1956The 'pos' is current offset of the target file in decimal form 1957[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 1958where target file resides, all in hex format. 1959 1960Fsnotify files 1961~~~~~~~~~~~~~~ 1962For inotify files the format is the following:: 1963 1964 pos: 0 1965 flags: 02000000 1966 mnt_id: 9 1967 ino: 63107 1968 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 1969 1970where 'wd' is a watch descriptor in decimal form, i.e. a target file 1971descriptor number, 'ino' and 'sdev' are inode and device where the 1972target file resides and the 'mask' is the mask of events, all in hex 1973form [see inotify(7) for more details]. 1974 1975If the kernel was built with exportfs support, the path to the target 1976file is encoded as a file handle. The file handle is provided by three 1977fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 1978format. 1979 1980If the kernel is built without exportfs support the file handle won't be 1981printed out. 1982 1983If there is no inotify mark attached yet the 'inotify' line will be omitted. 1984 1985For fanotify files the format is:: 1986 1987 pos: 0 1988 flags: 02 1989 mnt_id: 9 1990 ino: 63107 1991 fanotify flags:10 event-flags:0 1992 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 1993 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 1994 1995where fanotify 'flags' and 'event-flags' are values used in fanotify_init 1996call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 1997flags associated with mark which are tracked separately from events 1998mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 1999mask and 'ignored_mask' is the mask of events which are to be ignored. 2000All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2001provide information about flags and mask used in fanotify_mark 2002call [see fsnotify manpage for details]. 2003 2004While the first three lines are mandatory and always printed, the rest is 2005optional and may be omitted if no marks created yet. 2006 2007Timerfd files 2008~~~~~~~~~~~~~ 2009 2010:: 2011 2012 pos: 0 2013 flags: 02 2014 mnt_id: 9 2015 ino: 63107 2016 clockid: 0 2017 ticks: 0 2018 settime flags: 01 2019 it_value: (0, 49406829) 2020 it_interval: (1, 0) 2021 2022where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2023that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2024flags in octal form been used to setup the timer [see timerfd_settime(2) for 2025details]. 'it_value' is remaining time until the timer expiration. 2026'it_interval' is the interval for the timer. Note the timer might be set up 2027with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2028still exhibits timer's remaining time. 2029 2030DMA Buffer files 2031~~~~~~~~~~~~~~~~ 2032 2033:: 2034 2035 pos: 0 2036 flags: 04002 2037 mnt_id: 9 2038 ino: 63107 2039 size: 32768 2040 count: 2 2041 exp_name: system-heap 2042 2043where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of 2044the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. 2045 20463.9 /proc/<pid>/map_files - Information about memory mapped files 2047--------------------------------------------------------------------- 2048This directory contains symbolic links which represent memory mapped files 2049the process is maintaining. Example output:: 2050 2051 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2052 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2053 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2054 | ... 2055 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2056 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2057 2058The name of a link represents the virtual memory bounds of a mapping, i.e. 2059vm_area_struct::vm_start-vm_area_struct::vm_end. 2060 2061The main purpose of the map_files is to retrieve a set of memory mapped 2062files in a fast way instead of parsing /proc/<pid>/maps or 2063/proc/<pid>/smaps, both of which contain many more records. At the same 2064time one can open(2) mappings from the listings of two processes and 2065comparing their inode numbers to figure out which anonymous memory areas 2066are actually shared. 2067 20683.10 /proc/<pid>/timerslack_ns - Task timerslack value 2069--------------------------------------------------------- 2070This file provides the value of the task's timerslack value in nanoseconds. 2071This value specifies an amount of time that normal timers may be deferred 2072in order to coalesce timers and avoid unnecessary wakeups. 2073 2074This allows a task's interactivity vs power consumption tradeoff to be 2075adjusted. 2076 2077Writing 0 to the file will set the task's timerslack to the default value. 2078 2079Valid values are from 0 - ULLONG_MAX 2080 2081An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2082permissions on the task specified to change its timerslack_ns value. 2083 20843.11 /proc/<pid>/patch_state - Livepatch patch operation state 2085----------------------------------------------------------------- 2086When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2087patch state for the task. 2088 2089A value of '-1' indicates that no patch is in transition. 2090 2091A value of '0' indicates that a patch is in transition and the task is 2092unpatched. If the patch is being enabled, then the task hasn't been 2093patched yet. If the patch is being disabled, then the task has already 2094been unpatched. 2095 2096A value of '1' indicates that a patch is in transition and the task is 2097patched. If the patch is being enabled, then the task has already been 2098patched. If the patch is being disabled, then the task hasn't been 2099unpatched yet. 2100 21013.12 /proc/<pid>/arch_status - task architecture specific status 2102------------------------------------------------------------------- 2103When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2104architecture specific status of the task. 2105 2106Example 2107~~~~~~~ 2108 2109:: 2110 2111 $ cat /proc/6753/arch_status 2112 AVX512_elapsed_ms: 8 2113 2114Description 2115~~~~~~~~~~~ 2116 2117x86 specific entries 2118~~~~~~~~~~~~~~~~~~~~~ 2119 2120AVX512_elapsed_ms 2121^^^^^^^^^^^^^^^^^^ 2122 2123 If AVX512 is supported on the machine, this entry shows the milliseconds 2124 elapsed since the last time AVX512 usage was recorded. The recording 2125 happens on a best effort basis when a task is scheduled out. This means 2126 that the value depends on two factors: 2127 2128 1) The time which the task spent on the CPU without being scheduled 2129 out. With CPU isolation and a single runnable task this can take 2130 several seconds. 2131 2132 2) The time since the task was scheduled out last. Depending on the 2133 reason for being scheduled out (time slice exhausted, syscall ...) 2134 this can be arbitrary long time. 2135 2136 As a consequence the value cannot be considered precise and authoritative 2137 information. The application which uses this information has to be aware 2138 of the overall scenario on the system in order to determine whether a 2139 task is a real AVX512 user or not. Precise information can be obtained 2140 with performance counters. 2141 2142 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2143 the task is unlikely an AVX512 user, but depends on the workload and the 2144 scheduling scenario, it also could be a false negative mentioned above. 2145 2146Chapter 4: Configuring procfs 2147============================= 2148 21494.1 Mount options 2150--------------------- 2151 2152The following mount options are supported: 2153 2154 ========= ======================================================== 2155 hidepid= Set /proc/<pid>/ access mode. 2156 gid= Set the group authorized to learn processes information. 2157 subset= Show only the specified subset of procfs. 2158 ========= ======================================================== 2159 2160hidepid=off or hidepid=0 means classic mode - everybody may access all 2161/proc/<pid>/ directories (default). 2162 2163hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2164directories but their own. Sensitive files like cmdline, sched*, status are now 2165protected against other users. This makes it impossible to learn whether any 2166user runs specific program (given the program doesn't reveal itself by its 2167behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2168other users, poorly written programs passing sensitive information via program 2169arguments are now protected against local eavesdroppers. 2170 2171hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2172fully invisible to other users. It doesn't mean that it hides a fact whether a 2173process with a specific pid value exists (it can be learned by other means, e.g. 2174by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by 2175stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2176gathering information about running processes, whether some daemon runs with 2177elevated privileges, whether other user runs some sensitive program, whether 2178other users run any program at all, etc. 2179 2180hidepid=ptraceable or hidepid=4 means that procfs should only contain 2181/proc/<pid>/ directories that the caller can ptrace. 2182 2183gid= defines a group authorized to learn processes information otherwise 2184prohibited by hidepid=. If you use some daemon like identd which needs to learn 2185information about processes information, just add identd to this group. 2186 2187subset=pid hides all top level files and directories in the procfs that 2188are not related to tasks. 2189 2190Chapter 5: Filesystem behavior 2191============================== 2192 2193Originally, before the advent of pid namepsace, procfs was a global file 2194system. It means that there was only one procfs instance in the system. 2195 2196When pid namespace was added, a separate procfs instance was mounted in 2197each pid namespace. So, procfs mount options are global among all 2198mountpoints within the same namespace:: 2199 2200 # grep ^proc /proc/mounts 2201 proc /proc proc rw,relatime,hidepid=2 0 0 2202 2203 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2204 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2205 +++ exited with 0 +++ 2206 2207 # grep ^proc /proc/mounts 2208 proc /proc proc rw,relatime,hidepid=2 0 0 2209 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2210 2211and only after remounting procfs mount options will change at all 2212mountpoints:: 2213 2214 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2215 2216 # grep ^proc /proc/mounts 2217 proc /proc proc rw,relatime,hidepid=1 0 0 2218 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2219 2220This behavior is different from the behavior of other filesystems. 2221 2222The new procfs behavior is more like other filesystems. Each procfs mount 2223creates a new procfs instance. Mount options affect own procfs instance. 2224It means that it became possible to have several procfs instances 2225displaying tasks with different filtering options in one pid namespace:: 2226 2227 # mount -o hidepid=invisible -t proc proc /proc 2228 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2229 # grep ^proc /proc/mounts 2230 proc /proc proc rw,relatime,hidepid=invisible 0 0 2231 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2232