1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 9 Bodo Bauer <bb@ricochet.net> 102.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 11move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 12fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 13===================== ======================================= ================ 14 15 16 17.. Table of Contents 18 19 0 Preface 20 0.1 Introduction/Credits 21 0.2 Legal Stuff 22 23 1 Collecting System Information 24 1.1 Process-Specific Subdirectories 25 1.2 Kernel data 26 1.3 IDE devices in /proc/ide 27 1.4 Networking info in /proc/net 28 1.5 SCSI info 29 1.6 Parallel port info in /proc/parport 30 1.7 TTY info in /proc/tty 31 1.8 Miscellaneous kernel statistics in /proc/stat 32 1.9 Ext4 file system parameters 33 34 2 Modifying System Parameters 35 36 3 Per-Process Parameters 37 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 38 score 39 3.2 /proc/<pid>/oom_score - Display current oom-killer score 40 3.3 /proc/<pid>/io - Display the IO accounting fields 41 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings 42 3.5 /proc/<pid>/mountinfo - Information about mounts 43 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 44 3.7 /proc/<pid>/task/<tid>/children - Information about task children 45 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 46 3.9 /proc/<pid>/map_files - Information about memory mapped files 47 3.10 /proc/<pid>/timerslack_ns - Task timerslack value 48 3.11 /proc/<pid>/patch_state - Livepatch patch operation state 49 3.12 /proc/<pid>/arch_status - Task architecture specific information 50 3.13 /proc/<pid>/fd - List of symlinks to open files 51 3.14 /proc/<pid/ksm_stat - Information about the process's ksm status. 52 53 4 Configuring procfs 54 4.1 Mount options 55 56 5 Filesystem behavior 57 58Preface 59======= 60 610.1 Introduction/Credits 62------------------------ 63 64We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 65other people for help compiling this documentation. We'd also like to extend a 66special thank you to Andi Kleen for documentation, which we relied on heavily 67to create this document, as well as the additional information he provided. 68Thanks to everybody else who contributed source or docs to the Linux kernel 69and helped create a great piece of software... :) 70 71The latest version of this document is available online at 72https://www.kernel.org/doc/html/latest/filesystems/proc.html 73 740.2 Legal Stuff 75--------------- 76 77We don't guarantee the correctness of this document, and if you come to us 78complaining about how you screwed up your system because of incorrect 79documentation, we won't feel responsible... 80 81Chapter 1: Collecting System Information 82======================================== 83 84In This Chapter 85--------------- 86* Investigating the properties of the pseudo file system /proc and its 87 ability to provide information on the running Linux system 88* Examining /proc's structure 89* Uncovering various information about the kernel and the processes running 90 on the system 91 92------------------------------------------------------------------------------ 93 94The proc file system acts as an interface to internal data structures in the 95kernel. It can be used to obtain information about the system and to change 96certain kernel parameters at runtime (sysctl). 97 98First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 99show you how you can use /proc/sys to change settings. 100 1011.1 Process-Specific Subdirectories 102----------------------------------- 103 104The directory /proc contains (among other things) one subdirectory for each 105process running on the system, which is named after the process ID (PID). 106 107The link 'self' points to the process reading the file system. Each process 108subdirectory has the entries listed in Table 1-1. 109 110A process can read its own information from /proc/PID/* with no extra 111permissions. When reading /proc/PID/* information for other processes, reading 112process is required to have either CAP_SYS_PTRACE capability with 113PTRACE_MODE_READ access permissions, or, alternatively, CAP_PERFMON 114capability. This applies to all read-only information like `maps`, `environ`, 115`pagemap`, etc. The only exception is `mem` file due to its read-write nature, 116which requires CAP_SYS_PTRACE capabilities with more elevated 117PTRACE_MODE_ATTACH permissions; CAP_PERFMON capability does not grant access 118to /proc/PID/mem for other processes. 119 120Note that an open file descriptor to /proc/<pid> or to any of its 121contained files or subdirectories does not prevent <pid> being reused 122for some other process in the event that <pid> exits. Operations on 123open /proc/<pid> file descriptors corresponding to dead processes 124never act on any new process that the kernel may, through chance, have 125also assigned the process ID <pid>. Instead, operations on these FDs 126usually fail with ESRCH. 127 128.. table:: Table 1-1: Process specific entries in /proc 129 130 ============= =============================================================== 131 File Content 132 ============= =============================================================== 133 clear_refs Clears page referenced bits shown in smaps output 134 cmdline Command line arguments 135 cpu Current and last cpu in which it was executed (2.4)(smp) 136 cwd Link to the current working directory 137 environ Values of environment variables 138 exe Link to the executable of this process 139 fd Directory, which contains all file descriptors 140 maps Memory maps to executables and library files (2.4) 141 mem Memory held by this process 142 root Link to the root directory of this process 143 stat Process status 144 statm Process memory status information 145 status Process status in human readable form 146 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 147 symbol the task is blocked in - or "0" if not blocked. 148 pagemap Page table 149 stack Report full stack trace, enable via CONFIG_STACKTRACE 150 smaps An extension based on maps, showing the memory consumption of 151 each mapping and flags associated with it 152 smaps_rollup Accumulated smaps stats for all mappings of the process. This 153 can be derived from smaps, but is faster and more convenient 154 numa_maps An extension based on maps, showing the memory locality and 155 binding policy as well as mem usage (in pages) of each mapping. 156 ============= =============================================================== 157 158For example, to get the status information of a process, all you have to do is 159read the file /proc/PID/status:: 160 161 >cat /proc/self/status 162 Name: cat 163 State: R (running) 164 Tgid: 5452 165 Pid: 5452 166 PPid: 743 167 TracerPid: 0 (2.4) 168 Uid: 501 501 501 501 169 Gid: 100 100 100 100 170 FDSize: 256 171 Groups: 100 14 16 172 Kthread: 0 173 VmPeak: 5004 kB 174 VmSize: 5004 kB 175 VmLck: 0 kB 176 VmHWM: 476 kB 177 VmRSS: 476 kB 178 RssAnon: 352 kB 179 RssFile: 120 kB 180 RssShmem: 4 kB 181 VmData: 156 kB 182 VmStk: 88 kB 183 VmExe: 68 kB 184 VmLib: 1412 kB 185 VmPTE: 20 kb 186 VmSwap: 0 kB 187 HugetlbPages: 0 kB 188 CoreDumping: 0 189 THP_enabled: 1 190 Threads: 1 191 SigQ: 0/28578 192 SigPnd: 0000000000000000 193 ShdPnd: 0000000000000000 194 SigBlk: 0000000000000000 195 SigIgn: 0000000000000000 196 SigCgt: 0000000000000000 197 CapInh: 00000000fffffeff 198 CapPrm: 0000000000000000 199 CapEff: 0000000000000000 200 CapBnd: ffffffffffffffff 201 CapAmb: 0000000000000000 202 NoNewPrivs: 0 203 Seccomp: 0 204 Speculation_Store_Bypass: thread vulnerable 205 SpeculationIndirectBranch: conditional enabled 206 voluntary_ctxt_switches: 0 207 nonvoluntary_ctxt_switches: 1 208 209This shows you nearly the same information you would get if you viewed it with 210the ps command. In fact, ps uses the proc file system to obtain its 211information. But you get a more detailed view of the process by reading the 212file /proc/PID/status. It fields are described in table 1-2. 213 214The statm file contains more detailed information about the process 215memory usage. Its seven fields are explained in Table 1-3. The stat file 216contains detailed information about the process itself. Its fields are 217explained in Table 1-4. 218 219(for SMP CONFIG users) 220 221For making accounting scalable, RSS related information are handled in an 222asynchronous manner and the value may not be very precise. To see a precise 223snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 224It's slow but very precise. 225 226.. table:: Table 1-2: Contents of the status fields (as of 4.19) 227 228 ========================== =================================================== 229 Field Content 230 ========================== =================================================== 231 Name filename of the executable 232 Umask file mode creation mask 233 State state (R is running, S is sleeping, D is sleeping 234 in an uninterruptible wait, Z is zombie, 235 T is traced or stopped) 236 Tgid thread group ID 237 Ngid NUMA group ID (0 if none) 238 Pid process id 239 PPid process id of the parent process 240 TracerPid PID of process tracing this process (0 if not, or 241 the tracer is outside of the current pid namespace) 242 Uid Real, effective, saved set, and file system UIDs 243 Gid Real, effective, saved set, and file system GIDs 244 FDSize number of file descriptor slots currently allocated 245 Groups supplementary group list 246 NStgid descendant namespace thread group ID hierarchy 247 NSpid descendant namespace process ID hierarchy 248 NSpgid descendant namespace process group ID hierarchy 249 NSsid descendant namespace session ID hierarchy 250 Kthread kernel thread flag, 1 is yes, 0 is no 251 VmPeak peak virtual memory size 252 VmSize total program size 253 VmLck locked memory size 254 VmPin pinned memory size 255 VmHWM peak resident set size ("high water mark") 256 VmRSS size of memory portions. It contains the three 257 following parts 258 (VmRSS = RssAnon + RssFile + RssShmem) 259 RssAnon size of resident anonymous memory 260 RssFile size of resident file mappings 261 RssShmem size of resident shmem memory (includes SysV shm, 262 mapping of tmpfs and shared anonymous mappings) 263 VmData size of private data segments 264 VmStk size of stack segments 265 VmExe size of text segment 266 VmLib size of shared library code 267 VmPTE size of page table entries 268 VmSwap amount of swap used by anonymous private data 269 (shmem swap usage is not included) 270 HugetlbPages size of hugetlb memory portions 271 CoreDumping process's memory is currently being dumped 272 (killing the process may lead to a corrupted core) 273 THP_enabled process is allowed to use THP (returns 0 when 274 PR_SET_THP_DISABLE is set on the process to disable 275 THP completely, not just partially) 276 Threads number of threads 277 SigQ number of signals queued/max. number for queue 278 SigPnd bitmap of pending signals for the thread 279 ShdPnd bitmap of shared pending signals for the process 280 SigBlk bitmap of blocked signals 281 SigIgn bitmap of ignored signals 282 SigCgt bitmap of caught signals 283 CapInh bitmap of inheritable capabilities 284 CapPrm bitmap of permitted capabilities 285 CapEff bitmap of effective capabilities 286 CapBnd bitmap of capabilities bounding set 287 CapAmb bitmap of ambient capabilities 288 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) 289 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 290 Speculation_Store_Bypass speculative store bypass mitigation status 291 SpeculationIndirectBranch indirect branch speculation mode 292 Cpus_allowed mask of CPUs on which this process may run 293 Cpus_allowed_list Same as previous, but in "list format" 294 Mems_allowed mask of memory nodes allowed to this process 295 Mems_allowed_list Same as previous, but in "list format" 296 voluntary_ctxt_switches number of voluntary context switches 297 nonvoluntary_ctxt_switches number of non voluntary context switches 298 ========================== =================================================== 299 300 301.. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3) 302 303 ======== =============================== ============================== 304 Field Content 305 ======== =============================== ============================== 306 size total program size (pages) (same as VmSize in status) 307 resident size of memory portions (pages) (same as VmRSS in status) 308 shared number of pages that are shared (i.e. backed by a file, same 309 as RssFile+RssShmem in status) 310 trs number of pages that are 'code' (not including libs; broken, 311 includes data segment) 312 lrs number of pages of library (always 0 on 2.6) 313 drs number of pages of data/stack (including libs; broken, 314 includes library text) 315 dt number of dirty pages (always 0 on 2.6) 316 ======== =============================== ============================== 317 318 319.. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7) 320 321 ============= =============================================================== 322 Field Content 323 ============= =============================================================== 324 pid process id 325 tcomm filename of the executable 326 state state (R is running, S is sleeping, D is sleeping in an 327 uninterruptible wait, Z is zombie, T is traced or stopped) 328 ppid process id of the parent process 329 pgrp pgrp of the process 330 sid session id 331 tty_nr tty the process uses 332 tty_pgrp pgrp of the tty 333 flags task flags 334 min_flt number of minor faults 335 cmin_flt number of minor faults with child's 336 maj_flt number of major faults 337 cmaj_flt number of major faults with child's 338 utime user mode jiffies 339 stime kernel mode jiffies 340 cutime user mode jiffies with child's 341 cstime kernel mode jiffies with child's 342 priority priority level 343 nice nice level 344 num_threads number of threads 345 it_real_value (obsolete, always 0) 346 start_time time the process started after system boot 347 vsize virtual memory size 348 rss resident set memory size 349 rsslim current limit in bytes on the rss 350 start_code address above which program text can run 351 end_code address below which program text can run 352 start_stack address of the start of the main process stack 353 esp current value of ESP 354 eip current value of EIP 355 pending bitmap of pending signals 356 blocked bitmap of blocked signals 357 sigign bitmap of ignored signals 358 sigcatch bitmap of caught signals 359 0 (place holder, used to be the wchan address, 360 use /proc/PID/wchan instead) 361 0 (place holder) 362 0 (place holder) 363 exit_signal signal to send to parent thread on exit 364 task_cpu which CPU the task is scheduled on 365 rt_priority realtime priority 366 policy scheduling policy (man sched_setscheduler) 367 blkio_ticks time spent waiting for block IO 368 gtime guest time of the task in jiffies 369 cgtime guest time of the task children in jiffies 370 start_data address above which program data+bss is placed 371 end_data address below which program data+bss is placed 372 start_brk address above which program heap can be expanded with brk() 373 arg_start address above which program command line is placed 374 arg_end address below which program command line is placed 375 env_start address above which program environment is placed 376 env_end address below which program environment is placed 377 exit_code the thread's exit_code in the form reported by the waitpid 378 system call 379 ============= =============================================================== 380 381The /proc/PID/maps file contains the currently mapped memory regions and 382their access permissions. 383 384The format is:: 385 386 address perms offset dev inode pathname 387 388 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 389 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 390 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] 391 a7cb1000-a7cb2000 ---p 00000000 00:00 0 392 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 393 a7eb2000-a7eb3000 ---p 00000000 00:00 0 394 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 395 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 396 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 397 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 398 a800b000-a800e000 rw-p 00000000 00:00 0 399 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 400 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 401 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 402 a8024000-a8027000 rw-p 00000000 00:00 0 403 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 404 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 405 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 406 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 407 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 408 409where "address" is the address space in the process that it occupies, "perms" 410is a set of permissions:: 411 412 r = read 413 w = write 414 x = execute 415 s = shared 416 p = private (copy on write) 417 418"offset" is the offset into the mapping, "dev" is the device (major:minor), and 419"inode" is the inode on that device. 0 indicates that no inode is associated 420with the memory region, as the case would be with BSS (uninitialized data). 421The "pathname" shows the name associated file for this mapping. If the mapping 422is not associated with a file: 423 424 =================== =========================================== 425 [heap] the heap of the program 426 [stack] the stack of the main process 427 [vdso] the "virtual dynamic shared object", 428 the kernel system call handler 429 [anon:<name>] a private anonymous mapping that has been 430 named by userspace 431 [anon_shmem:<name>] an anonymous shared memory mapping that has 432 been named by userspace 433 =================== =========================================== 434 435 or if empty, the mapping is anonymous. 436 437Starting with 6.11 kernel, /proc/PID/maps provides an alternative 438ioctl()-based API that gives ability to flexibly and efficiently query and 439filter individual VMAs. This interface is binary and is meant for more 440efficient and easy programmatic use. `struct procmap_query`, defined in 441linux/fs.h UAPI header, serves as an input/output argument to the 442`PROCMAP_QUERY` ioctl() command. See comments in linus/fs.h UAPI header for 443details on query semantics, supported flags, data returned, and general API 444usage information. 445 446The /proc/PID/smaps is an extension based on maps, showing the memory 447consumption for each of the process's mappings. For each mapping (aka Virtual 448Memory Area, or VMA) there is a series of lines such as the following:: 449 450 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 451 452 Size: 1084 kB 453 KernelPageSize: 4 kB 454 MMUPageSize: 4 kB 455 Rss: 892 kB 456 Pss: 374 kB 457 Pss_Dirty: 0 kB 458 Shared_Clean: 892 kB 459 Shared_Dirty: 0 kB 460 Private_Clean: 0 kB 461 Private_Dirty: 0 kB 462 Referenced: 892 kB 463 Anonymous: 0 kB 464 KSM: 0 kB 465 LazyFree: 0 kB 466 AnonHugePages: 0 kB 467 ShmemPmdMapped: 0 kB 468 Shared_Hugetlb: 0 kB 469 Private_Hugetlb: 0 kB 470 Swap: 0 kB 471 SwapPss: 0 kB 472 KernelPageSize: 4 kB 473 MMUPageSize: 4 kB 474 Locked: 0 kB 475 THPeligible: 0 476 VmFlags: rd ex mr mw me dw 477 478The first of these lines shows the same information as is displayed for 479the mapping in /proc/PID/maps. Following lines show the size of the 480mapping (size); the size of each page allocated when backing a VMA 481(KernelPageSize), which is usually the same as the size in the page table 482entries; the page size used by the MMU when backing a VMA (in most cases, 483the same as KernelPageSize); the amount of the mapping that is currently 484resident in RAM (RSS); the process's proportional share of this mapping 485(PSS); and the number of clean and dirty shared and private pages in the 486mapping. 487 488The "proportional set size" (PSS) of a process is the count of pages it has 489in memory, where each page is divided by the number of processes sharing it. 490So if a process has 1000 pages all to itself, and 1000 shared with one other 491process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which 492consists of dirty pages. ("Pss_Clean" is not included, but it can be 493calculated by subtracting "Pss_Dirty" from "Pss".) 494 495Traditionally, a page is accounted as "private" if it is mapped exactly once, 496and a page is accounted as "shared" when mapped multiple times, even when 497mapped in the same process multiple times. Note that this accounting is 498independent of MAP_SHARED. 499 500In some kernel configurations, the semantics of pages part of a larger 501allocation (e.g., THP) can differ: a page is accounted as "private" if all 502pages part of the corresponding large allocation are *certainly* mapped in the 503same process, even if the page is mapped multiple times in that process. A 504page is accounted as "shared" if any page page of the larger allocation 505is *maybe* mapped in a different process. In some cases, a large allocation 506might be treated as "maybe mapped by multiple processes" even though this 507is no longer the case. 508 509Some kernel configurations do not track the precise number of times a page part 510of a larger allocation is mapped. In this case, when calculating the PSS, the 511average number of mappings per page in this larger allocation might be used 512as an approximation for the number of mappings of a page. The PSS calculation 513will be imprecise in this case. 514 515"Referenced" indicates the amount of memory currently marked as referenced or 516accessed. 517 518"Anonymous" shows the amount of memory that does not belong to any file. Even 519a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 520and a page is modified, the file page is replaced by a private anonymous copy. 521 522"KSM" reports how many of the pages are KSM pages. Note that KSM-placed zeropages 523are not included, only actual KSM pages. 524 525"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 526The memory isn't freed immediately with madvise(). It's freed in memory 527pressure if the memory is clean. Please note that the printed value might 528be lower than the real value due to optimizations used in the current 529implementation. If this is not desirable please file a bug report. 530 531"AnonHugePages" shows the amount of memory backed by transparent hugepage. 532 533"ShmemPmdMapped" shows the amount of shared (shmem/tmpfs) memory backed by 534huge pages. 535 536"Shared_Hugetlb" and "Private_Hugetlb" show the amounts of memory backed by 537hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 538reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 539 540"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 541 542For shmem mappings, "Swap" includes also the size of the mapped (and not 543replaced by copy-on-write) part of the underlying shmem object out on swap. 544"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 545does not take into account swapped out page of underlying shmem objects. 546"Locked" indicates whether the mapping is locked in memory or not. 547 548"THPeligible" indicates whether the mapping is eligible for allocating 549naturally aligned THP pages of any currently enabled size. 1 if true, 0 550otherwise. 551 552"VmFlags" field deserves a separate description. This member represents the 553kernel flags associated with the particular virtual memory area in two letter 554encoded manner. The codes are the following: 555 556 == ======================================= 557 rd readable 558 wr writeable 559 ex executable 560 sh shared 561 mr may read 562 mw may write 563 me may execute 564 ms may share 565 gd stack segment growns down 566 pf pure PFN range 567 lo pages are locked in memory 568 io memory mapped I/O area 569 sr sequential read advise provided 570 rr random read advise provided 571 dc do not copy area on fork 572 de do not expand area on remapping 573 ac area is accountable 574 nr swap space is not reserved for the area 575 ht area uses huge tlb pages 576 sf synchronous page fault 577 ar architecture specific flag 578 wf wipe on fork 579 dd do not include area into core dump 580 sd soft dirty flag 581 mm mixed map area 582 hg huge page advise flag 583 nh no huge page advise flag 584 mg mergeable advise flag 585 bt arm64 BTI guarded page 586 mt arm64 MTE allocation tags are enabled 587 um userfaultfd missing tracking 588 uw userfaultfd wr-protect tracking 589 ui userfaultfd minor fault 590 ss shadow/guarded control stack page 591 sl sealed 592 lf lock on fault pages 593 dp always lazily freeable mapping 594 == ======================================= 595 596Note that there is no guarantee that every flag and associated mnemonic will 597be present in all further kernel releases. Things get changed, the flags may 598be vanished or the reverse -- new added. Interpretation of their meaning 599might change in future as well. So each consumer of these flags has to 600follow each specific kernel version for the exact semantic. 601 602This file is only present if the CONFIG_MMU kernel configuration option is 603enabled. 604 605Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 606output can be achieved only in the single read call). 607 608This typically manifests when doing partial reads of these files while the 609memory map is being modified. Despite the races, we do provide the following 610guarantees: 611 6121) The mapped addresses never go backwards, which implies no two 613 regions will ever overlap. 6142) If there is something at a given vaddr during the entirety of the 615 life of the smaps/maps walk, there will be some output for it. 616 617The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 618but their values are the sums of the corresponding values for all mappings of 619the process. Additionally, it contains these fields: 620 621- Pss_Anon 622- Pss_File 623- Pss_Shmem 624 625They represent the proportional shares of anonymous, file, and shmem pages, as 626described for smaps above. These fields are omitted in smaps since each 627mapping identifies the type (anon, file, or shmem) of all pages it contains. 628Thus all information in smaps_rollup can be derived from smaps, but at a 629significantly higher cost. 630 631The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 632bits on both physical and virtual pages associated with a process, and the 633soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 634for details). 635To clear the bits for all the pages associated with the process:: 636 637 > echo 1 > /proc/PID/clear_refs 638 639To clear the bits for the anonymous pages associated with the process:: 640 641 > echo 2 > /proc/PID/clear_refs 642 643To clear the bits for the file mapped pages associated with the process:: 644 645 > echo 3 > /proc/PID/clear_refs 646 647To clear the soft-dirty bit:: 648 649 > echo 4 > /proc/PID/clear_refs 650 651To reset the peak resident set size ("high water mark") to the process's 652current value:: 653 654 > echo 5 > /proc/PID/clear_refs 655 656Any other value written to /proc/PID/clear_refs will have no effect. 657 658The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 659using /proc/kpageflags and number of times a page is mapped using 660/proc/kpagecount. For detailed explanation, see 661Documentation/admin-guide/mm/pagemap.rst. 662 663The /proc/pid/numa_maps is an extension based on maps, showing the memory 664locality and binding policy, as well as the memory usage (in pages) of 665each mapping. The output follows a general format where mapping details get 666summarized separated by blank spaces, one mapping per each file line:: 667 668 address policy mapping details 669 670 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 671 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 672 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 673 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 674 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 675 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 676 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 677 320698b000 default file=/lib64/libc-2.12.so 678 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 679 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 680 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 681 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 682 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 683 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 684 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 685 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 686 687Where: 688 689"address" is the starting address for the mapping; 690 691"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 692 693"mapping details" summarizes mapping data such as mapping type, page usage counters, 694node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 695size, in KB, that is backing the mapping up. 696 697Note that some kernel configurations do not track the precise number of times 698a page part of a larger allocation (e.g., THP) is mapped. In these 699configurations, "mapmax" might corresponds to the average number of mappings 700per page in such a larger allocation instead. 701 7021.2 Kernel data 703--------------- 704 705Similar to the process entries, the kernel data files give information about 706the running kernel. The files used to obtain this information are contained in 707/proc and are listed in Table 1-5. Not all of these will be present in your 708system. It depends on the kernel configuration and the loaded modules, which 709files are there, and which are missing. 710 711.. table:: Table 1-5: Kernel info in /proc 712 713 ============ =============================================================== 714 File Content 715 ============ =============================================================== 716 allocinfo Memory allocations profiling information 717 apm Advanced power management info 718 bootconfig Kernel command line obtained from boot config, 719 and, if there were kernel parameters from the 720 boot loader, a "# Parameters from bootloader:" 721 line followed by a line containing those 722 parameters prefixed by "# ". (5.5) 723 buddyinfo Kernel memory allocator information (see text) (2.5) 724 bus Directory containing bus specific information 725 cmdline Kernel command line, both from bootloader and embedded 726 in the kernel image 727 cpuinfo Info about the CPU 728 devices Available devices (block and character) 729 dma Used DMS channels 730 filesystems Supported filesystems 731 driver Various drivers grouped here, currently rtc (2.4) 732 execdomains Execdomains, related to security (2.4) 733 fb Frame Buffer devices (2.4) 734 fs File system parameters, currently nfs/exports (2.4) 735 ide Directory containing info about the IDE subsystem 736 interrupts Interrupt usage 737 iomem Memory map (2.4) 738 ioports I/O port usage 739 irq Masks for irq to cpu affinity (2.4)(smp?) 740 isapnp ISA PnP (Plug&Play) Info (2.4) 741 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 742 kmsg Kernel messages 743 ksyms Kernel symbol table 744 loadavg Load average of last 1, 5 & 15 minutes; 745 number of processes currently runnable (running or on ready queue); 746 total number of processes in system; 747 last pid created. 748 All fields are separated by one space except "number of 749 processes currently runnable" and "total number of processes 750 in system", which are separated by a slash ('/'). Example: 751 0.61 0.61 0.55 3/828 22084 752 locks Kernel locks 753 meminfo Memory info 754 misc Miscellaneous 755 modules List of loaded modules 756 mounts Mounted filesystems 757 net Networking info (see text) 758 pagetypeinfo Additional page allocator information (see text) (2.5) 759 partitions Table of partitions known to the system 760 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 761 decoupled by lspci (2.4) 762 rtc Real time clock 763 scsi SCSI info (see text) 764 slabinfo Slab pool info 765 softirqs softirq usage 766 stat Overall statistics 767 swaps Swap space utilization 768 sys See chapter 2 769 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 770 tty Info of tty drivers 771 uptime Wall clock since boot, combined idle time of all cpus 772 version Kernel version 773 video bttv info of video resources (2.4) 774 vmallocinfo Show vmalloced areas 775 ============ =============================================================== 776 777You can, for example, check which interrupts are currently in use and what 778they are used for by looking in the file /proc/interrupts:: 779 780 > cat /proc/interrupts 781 CPU0 782 0: 8728810 XT-PIC timer 783 1: 895 XT-PIC keyboard 784 2: 0 XT-PIC cascade 785 3: 531695 XT-PIC aha152x 786 4: 2014133 XT-PIC serial 787 5: 44401 XT-PIC pcnet_cs 788 8: 2 XT-PIC rtc 789 11: 8 XT-PIC i82365 790 12: 182918 XT-PIC PS/2 Mouse 791 13: 1 XT-PIC fpu 792 14: 1232265 XT-PIC ide0 793 15: 7 XT-PIC ide1 794 NMI: 0 795 796In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 797output of a SMP machine):: 798 799 > cat /proc/interrupts 800 801 CPU0 CPU1 802 0: 1243498 1214548 IO-APIC-edge timer 803 1: 8949 8958 IO-APIC-edge keyboard 804 2: 0 0 XT-PIC cascade 805 5: 11286 10161 IO-APIC-edge soundblaster 806 8: 1 0 IO-APIC-edge rtc 807 9: 27422 27407 IO-APIC-edge 3c503 808 12: 113645 113873 IO-APIC-edge PS/2 Mouse 809 13: 0 0 XT-PIC fpu 810 14: 22491 24012 IO-APIC-edge ide0 811 15: 2183 2415 IO-APIC-edge ide1 812 17: 30564 30414 IO-APIC-level eth0 813 18: 177 164 IO-APIC-level bttv 814 NMI: 2457961 2457959 815 LOC: 2457882 2457881 816 ERR: 2155 817 818NMI is incremented in this case because every timer interrupt generates a NMI 819(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 820 821LOC is the local interrupt counter of the internal APIC of every CPU. 822 823ERR is incremented in the case of errors in the IO-APIC bus (the bus that 824connects the CPUs in a SMP system. This means that an error has been detected, 825the IO-APIC automatically retry the transmission, so it should not be a big 826problem, but you should read the SMP-FAQ. 827 828In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 829/proc/interrupts to display every IRQ vector in use by the system, not 830just those considered 'most important'. The new vectors are: 831 832THR 833 interrupt raised when a machine check threshold counter 834 (typically counting ECC corrected errors of memory or cache) exceeds 835 a configurable threshold. Only available on some systems. 836 837TRM 838 a thermal event interrupt occurs when a temperature threshold 839 has been exceeded for the CPU. This interrupt may also be generated 840 when the temperature drops back to normal. 841 842SPU 843 a spurious interrupt is some interrupt that was raised then lowered 844 by some IO device before it could be fully processed by the APIC. Hence 845 the APIC sees the interrupt but does not know what device it came from. 846 For this case the APIC will generate the interrupt with a IRQ vector 847 of 0xff. This might also be generated by chipset bugs. 848 849RES, CAL, TLB 850 rescheduling, call and TLB flush interrupts are 851 sent from one CPU to another per the needs of the OS. Typically, 852 their statistics are used by kernel developers and interested users to 853 determine the occurrence of interrupts of the given type. 854 855The above IRQ vectors are displayed only when relevant. For example, 856the threshold vector does not exist on x86_64 platforms. Others are 857suppressed when the system is a uniprocessor. As of this writing, only 858i386 and x86_64 platforms support the new IRQ vector displays. 859 860Of some interest is the introduction of the /proc/irq directory to 2.4. 861It could be used to set IRQ to CPU affinity. This means that you can "hook" an 862IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 863irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 864prof_cpu_mask. 865 866For example:: 867 868 > ls /proc/irq/ 869 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 870 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 871 > ls /proc/irq/0/ 872 smp_affinity 873 874smp_affinity is a bitmask, in which you can specify which CPUs can handle the 875IRQ. You can set it by doing:: 876 877 > echo 1 > /proc/irq/10/smp_affinity 878 879This means that only the first CPU will handle the IRQ, but you can also echo 8805 which means that only the first and third CPU can handle the IRQ. 881 882The contents of each smp_affinity file is the same by default:: 883 884 > cat /proc/irq/0/smp_affinity 885 ffffffff 886 887There is an alternate interface, smp_affinity_list which allows specifying 888a CPU range instead of a bitmask:: 889 890 > cat /proc/irq/0/smp_affinity_list 891 1024-1031 892 893The default_smp_affinity mask applies to all non-active IRQs, which are the 894IRQs which have not yet been allocated/activated, and hence which lack a 895/proc/irq/[0-9]* directory. 896 897The node file on an SMP system shows the node to which the device using the IRQ 898reports itself as being attached. This hardware locality information does not 899include information about any possible driver locality preference. 900 901prof_cpu_mask specifies which CPUs are to be profiled by the system wide 902profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 903 904The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 905between all the CPUs which are allowed to handle it. As usual the kernel has 906more info than you and does a better job than you, so the defaults are the 907best choice for almost everyone. [Note this applies only to those IO-APIC's 908that support "Round Robin" interrupt distribution.] 909 910There are three more important subdirectories in /proc: net, scsi, and sys. 911The general rule is that the contents, or even the existence of these 912directories, depend on your kernel configuration. If SCSI is not enabled, the 913directory scsi may not exist. The same is true with the net, which is there 914only when networking support is present in the running kernel. 915 916The slabinfo file gives information about memory usage at the slab level. 917Linux uses slab pools for memory management above page level in version 2.2. 918Commonly used objects have their own slab pool (such as network buffers, 919directory cache, and so on). 920 921:: 922 923 > cat /proc/buddyinfo 924 925 Node 0, zone DMA 0 4 5 4 4 3 ... 926 Node 0, zone Normal 1 0 0 1 101 8 ... 927 Node 0, zone HighMem 2 0 0 1 1 0 ... 928 929External fragmentation is a problem under some workloads, and buddyinfo is a 930useful tool for helping diagnose these problems. Buddyinfo will give you a 931clue as to how big an area you can safely allocate, or why a previous 932allocation failed. 933 934Each column represents the number of pages of a certain order which are 935available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 936ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 937available in ZONE_NORMAL, etc... 938 939More information relevant to external fragmentation can be found in 940pagetypeinfo:: 941 942 > cat /proc/pagetypeinfo 943 Page block order: 9 944 Pages per block: 512 945 946 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 947 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 948 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 949 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 950 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 951 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 952 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 953 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 954 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 955 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 956 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 957 958 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 959 Node 0, zone DMA 2 0 5 1 0 960 Node 0, zone DMA32 41 6 967 2 0 961 962Fragmentation avoidance in the kernel works by grouping pages of different 963migrate types into the same contiguous regions of memory called page blocks. 964A page block is typically the size of the default hugepage size, e.g. 2MB on 965X86-64. By keeping pages grouped based on their ability to move, the kernel 966can reclaim pages within a page block to satisfy a high-order allocation. 967 968The pagetypinfo begins with information on the size of a page block. It 969then gives the same type of information as buddyinfo except broken down 970by migrate-type and finishes with details on how many page blocks of each 971type exist. 972 973If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 974from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 975make an estimate of the likely number of huge pages that can be allocated 976at a given point in time. All the "Movable" blocks should be allocatable 977unless memory has been mlock()'d. Some of the Reclaimable blocks should 978also be allocatable although a lot of filesystem metadata may have to be 979reclaimed to achieve this. 980 981 982allocinfo 983~~~~~~~~~ 984 985Provides information about memory allocations at all locations in the code 986base. Each allocation in the code is identified by its source file, line 987number, module (if originates from a loadable module) and the function calling 988the allocation. The number of bytes allocated and number of calls at each 989location are reported. The first line indicates the version of the file, the 990second line is the header listing fields in the file. 991If file version is 2.0 or higher then each line may contain additional 992<key>:<value> pairs representing extra information about the call site. 993For example if the counters are not accurate, the line will be appended with 994"accurate:no" pair. 995 996Supported markers in v2: 997accurate:no 998 999 Absolute values of the counters in this line are not accurate 1000 because of the failure to allocate memory to track some of the 1001 allocations made at this location. Deltas in these counters are 1002 accurate, therefore counters can be used to track allocation size 1003 and count changes. 1004 1005Example output. 1006 1007:: 1008 1009 > tail -n +3 /proc/allocinfo | sort -rn 1010 127664128 31168 mm/page_ext.c:270 func:alloc_page_ext 1011 56373248 4737 mm/slub.c:2259 func:alloc_slab_page 1012 14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded 1013 14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash 1014 13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs 1015 11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio 1016 9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node 1017 4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable 1018 4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start 1019 3940352 962 mm/memory.c:4214 func:alloc_anon_folio 1020 2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node 1021 ... 1022 1023 1024meminfo 1025~~~~~~~ 1026 1027Provides information about distribution and utilization of memory. This 1028varies by architecture and compile options. Some of the counters reported 1029here overlap. The memory reported by the non overlapping counters may not 1030add up to the overall memory usage and the difference for some workloads 1031can be substantial. In many cases there are other means to find out 1032additional memory using subsystem specific interfaces, for instance 1033/proc/net/sockstat for TCP memory allocations. 1034 1035Example output. You may not have all of these fields. 1036 1037:: 1038 1039 > cat /proc/meminfo 1040 1041 MemTotal: 32858820 kB 1042 MemFree: 21001236 kB 1043 MemAvailable: 27214312 kB 1044 Buffers: 581092 kB 1045 Cached: 5587612 kB 1046 SwapCached: 0 kB 1047 Active: 3237152 kB 1048 Inactive: 7586256 kB 1049 Active(anon): 94064 kB 1050 Inactive(anon): 4570616 kB 1051 Active(file): 3143088 kB 1052 Inactive(file): 3015640 kB 1053 Unevictable: 0 kB 1054 Mlocked: 0 kB 1055 SwapTotal: 0 kB 1056 SwapFree: 0 kB 1057 Zswap: 1904 kB 1058 Zswapped: 7792 kB 1059 Dirty: 12 kB 1060 Writeback: 0 kB 1061 AnonPages: 4654780 kB 1062 Mapped: 266244 kB 1063 Shmem: 9976 kB 1064 KReclaimable: 517708 kB 1065 Slab: 660044 kB 1066 SReclaimable: 517708 kB 1067 SUnreclaim: 142336 kB 1068 KernelStack: 11168 kB 1069 PageTables: 20540 kB 1070 SecPageTables: 0 kB 1071 NFS_Unstable: 0 kB 1072 Bounce: 0 kB 1073 WritebackTmp: 0 kB 1074 CommitLimit: 16429408 kB 1075 Committed_AS: 7715148 kB 1076 VmallocTotal: 34359738367 kB 1077 VmallocUsed: 40444 kB 1078 VmallocChunk: 0 kB 1079 Percpu: 29312 kB 1080 EarlyMemtestBad: 0 kB 1081 HardwareCorrupted: 0 kB 1082 AnonHugePages: 4149248 kB 1083 ShmemHugePages: 0 kB 1084 ShmemPmdMapped: 0 kB 1085 FileHugePages: 0 kB 1086 FilePmdMapped: 0 kB 1087 CmaTotal: 0 kB 1088 CmaFree: 0 kB 1089 Unaccepted: 0 kB 1090 Balloon: 0 kB 1091 HugePages_Total: 0 1092 HugePages_Free: 0 1093 HugePages_Rsvd: 0 1094 HugePages_Surp: 0 1095 Hugepagesize: 2048 kB 1096 Hugetlb: 0 kB 1097 DirectMap4k: 401152 kB 1098 DirectMap2M: 10008576 kB 1099 DirectMap1G: 24117248 kB 1100 1101MemTotal 1102 Total usable RAM (i.e. physical RAM minus a few reserved 1103 bits and the kernel binary code) 1104MemFree 1105 Total free RAM. On highmem systems, the sum of LowFree+HighFree 1106MemAvailable 1107 An estimate of how much memory is available for starting new 1108 applications, without swapping. Calculated from MemFree, 1109 SReclaimable, the size of the file LRU lists, and the low 1110 watermarks in each zone. 1111 The estimate takes into account that the system needs some 1112 page cache to function well, and that not all reclaimable 1113 slab will be reclaimable, due to items being in use. The 1114 impact of those factors will vary from system to system. 1115Buffers 1116 Relatively temporary storage for raw disk blocks 1117 shouldn't get tremendously large (20MB or so) 1118Cached 1119 In-memory cache for files read from the disk (the 1120 pagecache) as well as tmpfs & shmem. 1121 Doesn't include SwapCached. 1122SwapCached 1123 Memory that once was swapped out, is swapped back in but 1124 still also is in the swapfile (if memory is needed it 1125 doesn't need to be swapped out AGAIN because it is already 1126 in the swapfile. This saves I/O) 1127Active 1128 Memory that has been used more recently and usually not 1129 reclaimed unless absolutely necessary. 1130Inactive 1131 Memory which has been less recently used. It is more 1132 eligible to be reclaimed for other purposes 1133Unevictable 1134 Memory allocated for userspace which cannot be reclaimed, such 1135 as mlocked pages, ramfs backing pages, secret memfd pages etc. 1136Mlocked 1137 Memory locked with mlock(). 1138HighTotal, HighFree 1139 Highmem is all memory above ~860MB of physical memory. 1140 Highmem areas are for use by userspace programs, or 1141 for the pagecache. The kernel must use tricks to access 1142 this memory, making it slower to access than lowmem. 1143LowTotal, LowFree 1144 Lowmem is memory which can be used for everything that 1145 highmem can be used for, but it is also available for the 1146 kernel's use for its own data structures. Among many 1147 other things, it is where everything from the Slab is 1148 allocated. Bad things happen when you're out of lowmem. 1149SwapTotal 1150 total amount of swap space available 1151SwapFree 1152 Memory which has been evicted from RAM, and is temporarily 1153 on the disk 1154Zswap 1155 Memory consumed by the zswap backend (compressed size) 1156Zswapped 1157 Amount of anonymous memory stored in zswap (original size) 1158Dirty 1159 Memory which is waiting to get written back to the disk 1160Writeback 1161 Memory which is actively being written back to the disk 1162AnonPages 1163 Non-file backed pages mapped into userspace page tables. Note that 1164 some kernel configurations might consider all pages part of a 1165 larger allocation (e.g., THP) as "mapped", as soon as a single 1166 page is mapped. 1167Mapped 1168 files which have been mmapped, such as libraries. Note that some 1169 kernel configurations might consider all pages part of a larger 1170 allocation (e.g., THP) as "mapped", as soon as a single page is 1171 mapped. 1172Shmem 1173 Total memory used by shared memory (shmem) and tmpfs 1174KReclaimable 1175 Kernel allocations that the kernel will attempt to reclaim 1176 under memory pressure. Includes SReclaimable (below), and other 1177 direct allocations with a shrinker. 1178Slab 1179 in-kernel data structures cache 1180SReclaimable 1181 Part of Slab, that might be reclaimed, such as caches 1182SUnreclaim 1183 Part of Slab, that cannot be reclaimed on memory pressure 1184KernelStack 1185 Memory consumed by the kernel stacks of all tasks 1186PageTables 1187 Memory consumed by userspace page tables 1188SecPageTables 1189 Memory consumed by secondary page tables, this currently includes 1190 KVM mmu and IOMMU allocations on x86 and arm64. 1191NFS_Unstable 1192 Always zero. Previously counted pages which had been written to 1193 the server, but has not been committed to stable storage. 1194Bounce 1195 Always zero. Previously memory used for block device 1196 "bounce buffers". 1197WritebackTmp 1198 Always zero. Previously memory used by FUSE for temporary 1199 writeback buffers. 1200CommitLimit 1201 Based on the overcommit ratio ('vm.overcommit_ratio'), 1202 this is the total amount of memory currently available to 1203 be allocated on the system. This limit is only adhered to 1204 if strict overcommit accounting is enabled (mode 2 in 1205 'vm.overcommit_memory'). 1206 1207 The CommitLimit is calculated with the following formula:: 1208 1209 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1210 overcommit_ratio / 100 + [total swap pages] 1211 1212 For example, on a system with 1G of physical RAM and 7G 1213 of swap with a `vm.overcommit_ratio` of 30 it would 1214 yield a CommitLimit of 7.3G. 1215 1216 For more details, see the memory overcommit documentation 1217 in mm/overcommit-accounting. 1218Committed_AS 1219 The amount of memory presently allocated on the system. 1220 The committed memory is a sum of all of the memory which 1221 has been allocated by processes, even if it has not been 1222 "used" by them as of yet. A process which malloc()'s 1G 1223 of memory, but only touches 300M of it will show up as 1224 using 1G. This 1G is memory which has been "committed" to 1225 by the VM and can be used at any time by the allocating 1226 application. With strict overcommit enabled on the system 1227 (mode 2 in 'vm.overcommit_memory'), allocations which would 1228 exceed the CommitLimit (detailed above) will not be permitted. 1229 This is useful if one needs to guarantee that processes will 1230 not fail due to lack of memory once that memory has been 1231 successfully allocated. 1232VmallocTotal 1233 total size of vmalloc virtual address space 1234VmallocUsed 1235 amount of vmalloc area which is used 1236VmallocChunk 1237 largest contiguous block of vmalloc area which is free 1238Percpu 1239 Memory allocated to the percpu allocator used to back percpu 1240 allocations. This stat excludes the cost of metadata. 1241EarlyMemtestBad 1242 The amount of RAM/memory in kB, that was identified as corrupted 1243 by early memtest. If memtest was not run, this field will not 1244 be displayed at all. Size is never rounded down to 0 kB. 1245 That means if 0 kB is reported, you can safely assume 1246 there was at least one pass of memtest and none of the passes 1247 found a single faulty byte of RAM. 1248HardwareCorrupted 1249 The amount of RAM/memory in KB, the kernel identifies as 1250 corrupted. 1251AnonHugePages 1252 Non-file backed huge pages mapped into userspace page tables 1253ShmemHugePages 1254 Memory used by shared memory (shmem) and tmpfs allocated 1255 with huge pages 1256ShmemPmdMapped 1257 Shared memory mapped into userspace with huge pages 1258FileHugePages 1259 Memory used for filesystem data (page cache) allocated 1260 with huge pages 1261FilePmdMapped 1262 Page cache mapped into userspace with huge pages 1263CmaTotal 1264 Memory reserved for the Contiguous Memory Allocator (CMA) 1265CmaFree 1266 Free remaining memory in the CMA reserves 1267Unaccepted 1268 Memory that has not been accepted by the guest 1269Balloon 1270 Memory returned to Host by VM Balloon Drivers 1271HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb 1272 See Documentation/admin-guide/mm/hugetlbpage.rst. 1273DirectMap4k, DirectMap2M, DirectMap1G 1274 Breakdown of page table sizes used in the kernel's 1275 identity mapping of RAM 1276 1277vmallocinfo 1278~~~~~~~~~~~ 1279 1280Provides information about vmalloced/vmaped areas. One line per area, 1281containing the virtual address range of the area, size in bytes, 1282caller information of the creator, and optional information depending 1283on the kind of area: 1284 1285 ========== =================================================== 1286 pages=nr number of pages 1287 phys=addr if a physical address was specified 1288 ioremap I/O mapping (ioremap() and friends) 1289 vmalloc vmalloc() area 1290 vmap vmap()ed pages 1291 user VM_USERMAP area 1292 vpages buffer for pages pointers was vmalloced (huge area) 1293 N<node>=nr (Only on NUMA kernels) 1294 Number of pages allocated on memory node <node> 1295 ========== =================================================== 1296 1297:: 1298 1299 > cat /proc/vmallocinfo 1300 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1301 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1302 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1303 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1304 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1305 phys=7fee8000 ioremap 1306 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1307 phys=7fee7000 ioremap 1308 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1309 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1310 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1311 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1312 pages=2 vmalloc N1=2 1313 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1314 /0x130 [x_tables] pages=4 vmalloc N0=4 1315 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1316 pages=14 vmalloc N2=14 1317 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1318 pages=4 vmalloc N1=4 1319 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1320 pages=2 vmalloc N1=2 1321 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1322 pages=10 vmalloc N0=10 1323 1324 1325softirqs 1326~~~~~~~~ 1327 1328Provides counts of softirq handlers serviced since boot time, for each CPU. 1329 1330:: 1331 1332 > cat /proc/softirqs 1333 CPU0 CPU1 CPU2 CPU3 1334 HI: 0 0 0 0 1335 TIMER: 27166 27120 27097 27034 1336 NET_TX: 0 0 0 17 1337 NET_RX: 42 0 0 39 1338 BLOCK: 0 0 107 1121 1339 TASKLET: 0 0 0 290 1340 SCHED: 27035 26983 26971 26746 1341 HRTIMER: 0 0 0 0 1342 RCU: 1678 1769 2178 2250 1343 13441.3 Networking info in /proc/net 1345-------------------------------- 1346 1347The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1348additional values you get for IP version 6 if you configure the kernel to 1349support this. Table 1-9 lists the files and their meaning. 1350 1351 1352.. table:: Table 1-8: IPv6 info in /proc/net 1353 1354 ========== ===================================================== 1355 File Content 1356 ========== ===================================================== 1357 udp6 UDP sockets (IPv6) 1358 tcp6 TCP sockets (IPv6) 1359 raw6 Raw device statistics (IPv6) 1360 igmp6 IP multicast addresses, which this host joined (IPv6) 1361 if_inet6 List of IPv6 interface addresses 1362 ipv6_route Kernel routing table for IPv6 1363 rt6_stats Global IPv6 routing tables statistics 1364 sockstat6 Socket statistics (IPv6) 1365 snmp6 Snmp data (IPv6) 1366 ========== ===================================================== 1367 1368.. table:: Table 1-9: Network info in /proc/net 1369 1370 ============= ================================================================ 1371 File Content 1372 ============= ================================================================ 1373 arp Kernel ARP table 1374 dev network devices with statistics 1375 dev_mcast the Layer2 multicast groups a device is listening too 1376 (interface index, label, number of references, number of bound 1377 addresses). 1378 dev_stat network device status 1379 ip_fwchains Firewall chain linkage 1380 ip_fwnames Firewall chain names 1381 ip_masq Directory containing the masquerading tables 1382 ip_masquerade Major masquerading table 1383 netstat Network statistics 1384 raw raw device statistics 1385 route Kernel routing table 1386 rpc Directory containing rpc info 1387 rt_cache Routing cache 1388 snmp SNMP data 1389 sockstat Socket statistics 1390 softnet_stat Per-CPU incoming packets queues statistics of online CPUs 1391 tcp TCP sockets 1392 udp UDP sockets 1393 unix UNIX domain sockets 1394 wireless Wireless interface data (Wavelan etc) 1395 igmp IP multicast addresses, which this host joined 1396 psched Global packet scheduler parameters. 1397 netlink List of PF_NETLINK sockets 1398 ip_mr_vifs List of multicast virtual interfaces 1399 ip_mr_cache List of multicast routing cache 1400 ============= ================================================================ 1401 1402You can use this information to see which network devices are available in 1403your system and how much traffic was routed over those devices:: 1404 1405 > cat /proc/net/dev 1406 Inter-|Receive |[... 1407 face |bytes packets errs drop fifo frame compressed multicast|[... 1408 lo: 908188 5596 0 0 0 0 0 0 [... 1409 ppp0:15475140 20721 410 0 0 410 0 0 [... 1410 eth0: 614530 7085 0 0 0 0 0 1 [... 1411 1412 ...] Transmit 1413 ...] bytes packets errs drop fifo colls carrier compressed 1414 ...] 908188 5596 0 0 0 0 0 0 1415 ...] 1375103 17405 0 0 0 0 0 0 1416 ...] 1703981 5535 0 0 0 3 0 0 1417 1418In addition, each Channel Bond interface has its own directory. For 1419example, the bond0 device will have a directory called /proc/net/bond0/. 1420It will contain information that is specific to that bond, such as the 1421current slaves of the bond, the link status of the slaves, and how 1422many times the slaves link has failed. 1423 14241.4 SCSI info 1425------------- 1426 1427If you have a SCSI or ATA host adapter in your system, you'll find a 1428subdirectory named after the driver for this adapter in /proc/scsi. 1429You'll also see a list of all recognized SCSI devices in /proc/scsi:: 1430 1431 >cat /proc/scsi/scsi 1432 Attached devices: 1433 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1434 Vendor: IBM Model: DGHS09U Rev: 03E0 1435 Type: Direct-Access ANSI SCSI revision: 03 1436 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1437 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1438 Type: CD-ROM ANSI SCSI revision: 02 1439 1440 1441The directory named after the driver has one file for each adapter found in 1442the system. These files contain information about the controller, including 1443the used IRQ and the IO address range. The amount of information shown is 1444dependent on the adapter you use. The example shows the output for an Adaptec 1445AHA-2940 SCSI adapter:: 1446 1447 > cat /proc/scsi/aic7xxx/0 1448 1449 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1450 Compile Options: 1451 TCQ Enabled By Default : Disabled 1452 AIC7XXX_PROC_STATS : Disabled 1453 AIC7XXX_RESET_DELAY : 5 1454 Adapter Configuration: 1455 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1456 Ultra Wide Controller 1457 PCI MMAPed I/O Base: 0xeb001000 1458 Adapter SEEPROM Config: SEEPROM found and used. 1459 Adaptec SCSI BIOS: Enabled 1460 IRQ: 10 1461 SCBs: Active 0, Max Active 2, 1462 Allocated 15, HW 16, Page 255 1463 Interrupts: 160328 1464 BIOS Control Word: 0x18b6 1465 Adapter Control Word: 0x005b 1466 Extended Translation: Enabled 1467 Disconnect Enable Flags: 0xffff 1468 Ultra Enable Flags: 0x0001 1469 Tag Queue Enable Flags: 0x0000 1470 Ordered Queue Tag Flags: 0x0000 1471 Default Tag Queue Depth: 8 1472 Tagged Queue By Device array for aic7xxx host instance 0: 1473 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1474 Actual queue depth per device for aic7xxx host instance 0: 1475 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1476 Statistics: 1477 (scsi0:0:0:0) 1478 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1479 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1480 Total transfers 160151 (74577 reads and 85574 writes) 1481 (scsi0:0:6:0) 1482 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1483 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1484 Total transfers 0 (0 reads and 0 writes) 1485 1486 14871.5 Parallel port info in /proc/parport 1488--------------------------------------- 1489 1490The directory /proc/parport contains information about the parallel ports of 1491your system. It has one subdirectory for each port, named after the port 1492number (0,1,2,...). 1493 1494These directories contain the four files shown in Table 1-10. 1495 1496 1497.. table:: Table 1-10: Files in /proc/parport 1498 1499 ========= ==================================================================== 1500 File Content 1501 ========= ==================================================================== 1502 autoprobe Any IEEE-1284 device ID information that has been acquired. 1503 devices list of the device drivers using that port. A + will appear by the 1504 name of the device currently using the port (it might not appear 1505 against any). 1506 hardware Parallel port's base address, IRQ line and DMA channel. 1507 irq IRQ that parport is using for that port. This is in a separate 1508 file to allow you to alter it by writing a new value in (IRQ 1509 number or none). 1510 ========= ==================================================================== 1511 15121.6 TTY info in /proc/tty 1513------------------------- 1514 1515Information about the available and actually used tty's can be found in the 1516directory /proc/tty. You'll find entries for drivers and line disciplines in 1517this directory, as shown in Table 1-11. 1518 1519 1520.. table:: Table 1-11: Files in /proc/tty 1521 1522 ============= ============================================== 1523 File Content 1524 ============= ============================================== 1525 drivers list of drivers and their usage 1526 ldiscs registered line disciplines 1527 driver/serial usage statistic and status of single tty lines 1528 ============= ============================================== 1529 1530To see which tty's are currently in use, you can simply look into the file 1531/proc/tty/drivers:: 1532 1533 > cat /proc/tty/drivers 1534 pty_slave /dev/pts 136 0-255 pty:slave 1535 pty_master /dev/ptm 128 0-255 pty:master 1536 pty_slave /dev/ttyp 3 0-255 pty:slave 1537 pty_master /dev/pty 2 0-255 pty:master 1538 serial /dev/cua 5 64-67 serial:callout 1539 serial /dev/ttyS 4 64-67 serial 1540 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1541 /dev/ptmx /dev/ptmx 5 2 system 1542 /dev/console /dev/console 5 1 system:console 1543 /dev/tty /dev/tty 5 0 system:/dev/tty 1544 unknown /dev/tty 4 1-63 console 1545 1546 15471.7 Miscellaneous kernel statistics in /proc/stat 1548------------------------------------------------- 1549 1550Various pieces of information about kernel activity are available in the 1551/proc/stat file. All of the numbers reported in this file are aggregates 1552since the system first booted. For a quick look, simply cat the file:: 1553 1554 > cat /proc/stat 1555 cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0 1556 cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0 1557 cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0 1558 cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0 1559 cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0 1560 intr 8688370575 8 3373 0 0 0 0 0 0 1 40791 0 0 353317 0 0 0 0 224789828 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 190974333 41958554 123983334 43 0 224593 0 0 0 <more 0's deleted> 1561 ctxt 22848221062 1562 btime 1605316999 1563 processes 746787147 1564 procs_running 2 1565 procs_blocked 0 1566 softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354 1567 1568The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1569lines. These numbers identify the amount of time the CPU has spent performing 1570different kinds of work. Time units are in USER_HZ (typically hundredths of a 1571second). The meanings of the columns are as follows, from left to right: 1572 1573- user: normal processes executing in user mode 1574- nice: niced processes executing in user mode 1575- system: processes executing in kernel mode 1576- idle: twiddling thumbs 1577- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1578 are several problems: 1579 1580 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1581 waiting for I/O to complete. When CPU goes into idle state for 1582 outstanding task I/O, another task will be scheduled on this CPU. 1583 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1584 on any CPU, so the iowait of each CPU is difficult to calculate. 1585 3. The value of iowait field in /proc/stat will decrease in certain 1586 conditions. 1587 1588 So, the iowait is not reliable by reading from /proc/stat. 1589- irq: servicing interrupts 1590- softirq: servicing softirqs 1591- steal: involuntary wait 1592- guest: running a normal guest 1593- guest_nice: running a niced guest 1594 1595The "intr" line gives counts of interrupts serviced since boot time, for each 1596of the possible system interrupts. The first column is the total of all 1597interrupts serviced including unnumbered architecture specific interrupts; 1598each subsequent column is the total for that particular numbered interrupt. 1599Unnumbered interrupts are not shown, only summed into the total. 1600 1601The "ctxt" line gives the total number of context switches across all CPUs. 1602 1603The "btime" line gives the time at which the system booted, in seconds since 1604the Unix epoch. 1605 1606The "processes" line gives the number of processes and threads created, which 1607includes (but is not limited to) those created by calls to the fork() and 1608clone() system calls. 1609 1610The "procs_running" line gives the total number of threads that are 1611running or ready to run (i.e., the total number of runnable threads). 1612 1613The "procs_blocked" line gives the number of processes currently blocked, 1614waiting for I/O to complete. 1615 1616The "softirq" line gives counts of softirqs serviced since boot time, for each 1617of the possible system softirqs. The first column is the total of all 1618softirqs serviced; each subsequent column is the total for that particular 1619softirq. 1620 1621 16221.8 Ext4 file system parameters 1623------------------------------- 1624 1625Information about mounted ext4 file systems can be found in 1626/proc/fs/ext4. Each mounted filesystem will have a directory in 1627/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1628/proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device 1629directory are shown in Table 1-12, below. 1630 1631.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1632 1633 ============== ========================================================== 1634 File Content 1635 mb_groups details of multiblock allocator buddy cache of free blocks 1636 ============== ========================================================== 1637 16381.9 /proc/consoles 1639------------------- 1640Shows registered system console lines. 1641 1642To see which character device lines are currently used for the system console 1643/dev/console, you may simply look into the file /proc/consoles:: 1644 1645 > cat /proc/consoles 1646 tty0 -WU (ECp) 4:7 1647 ttyS0 -W- (Ep) 4:64 1648 1649The columns are: 1650 1651+--------------------+-------------------------------------------------------+ 1652| device | name of the device | 1653+====================+=======================================================+ 1654| operations | * R = can do read operations | 1655| | * W = can do write operations | 1656| | * U = can do unblank | 1657+--------------------+-------------------------------------------------------+ 1658| flags | * E = it is enabled | 1659| | * C = it is preferred console | 1660| | * B = it is primary boot console | 1661| | * p = it is used for printk buffer | 1662| | * b = it is not a TTY but a Braille device | 1663| | * a = it is safe to use when cpu is offline | 1664+--------------------+-------------------------------------------------------+ 1665| major:minor | major and minor number of the device separated by a | 1666| | colon | 1667+--------------------+-------------------------------------------------------+ 1668 1669Summary 1670------- 1671 1672The /proc file system serves information about the running system. It not only 1673allows access to process data but also allows you to request the kernel status 1674by reading files in the hierarchy. 1675 1676The directory structure of /proc reflects the types of information and makes 1677it easy, if not obvious, where to look for specific data. 1678 1679Chapter 2: Modifying System Parameters 1680====================================== 1681 1682In This Chapter 1683--------------- 1684 1685* Modifying kernel parameters by writing into files found in /proc/sys 1686* Exploring the files which modify certain parameters 1687* Review of the /proc/sys file tree 1688 1689------------------------------------------------------------------------------ 1690 1691A very interesting part of /proc is the directory /proc/sys. This is not only 1692a source of information, it also allows you to change parameters within the 1693kernel. Be very careful when attempting this. You can optimize your system, 1694but you can also cause it to crash. Never alter kernel parameters on a 1695production system. Set up a development machine and test to make sure that 1696everything works the way you want it to. You may have no alternative but to 1697reboot the machine once an error has been made. 1698 1699To change a value, simply echo the new value into the file. 1700You need to be root to do this. You can create your own boot script 1701to perform this every time your system boots. 1702 1703The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1704general things in the operation of the Linux kernel. Since some of the files 1705can inadvertently disrupt your system, it is advisable to read both 1706documentation and source before actually making adjustments. In any case, be 1707very careful when writing to any of these files. The entries in /proc may 1708change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1709review the kernel documentation in the directory linux/Documentation. 1710This chapter is heavily based on the documentation included in the pre 2.2 1711kernels, and became part of it in version 2.2.1 of the Linux kernel. 1712 1713Please see: Documentation/admin-guide/sysctl/ directory for descriptions of 1714these entries. 1715 1716Summary 1717------- 1718 1719Certain aspects of kernel behavior can be modified at runtime, without the 1720need to recompile the kernel, or even to reboot the system. The files in the 1721/proc/sys tree can not only be read, but also modified. You can use the echo 1722command to write value into these files, thereby changing the default settings 1723of the kernel. 1724 1725 1726Chapter 3: Per-process Parameters 1727================================= 1728 17293.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1730-------------------------------------------------------------------------------- 1731 1732These files can be used to adjust the badness heuristic used to select which 1733process gets killed in out of memory (oom) conditions. 1734 1735The badness heuristic assigns a value to each candidate task ranging from 0 1736(never kill) to 1000 (always kill) to determine which process is targeted. The 1737units are roughly a proportion along that range of allowed memory the process 1738may allocate from based on an estimation of its current memory and swap use. 1739For example, if a task is using all allowed memory, its badness score will be 17401000. If it is using half of its allowed memory, its score will be 500. 1741 1742The amount of "allowed" memory depends on the context in which the oom killer 1743was called. If it is due to the memory assigned to the allocating task's cpuset 1744being exhausted, the allowed memory represents the set of mems assigned to that 1745cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1746memory represents the set of mempolicy nodes. If it is due to a memory 1747limit (or swap limit) being reached, the allowed memory is that configured 1748limit. Finally, if it is due to the entire system being out of memory, the 1749allowed memory represents all allocatable resources. 1750 1751The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1752is used to determine which task to kill. Acceptable values range from -1000 1753(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1754polarize the preference for oom killing either by always preferring a certain 1755task or completely disabling it. The lowest possible value, -1000, is 1756equivalent to disabling oom killing entirely for that task since it will always 1757report a badness score of 0. 1758 1759Consequently, it is very simple for userspace to define the amount of memory to 1760consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1761example, is roughly equivalent to allowing the remainder of tasks sharing the 1762same system, cpuset, mempolicy, or memory controller resources to use at least 176350% more memory. A value of -500, on the other hand, would be roughly 1764equivalent to discounting 50% of the task's allowed memory from being considered 1765as scoring against the task. 1766 1767For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1768be used to tune the badness score. Its acceptable values range from -16 1769(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1770(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1771scaled linearly with /proc/<pid>/oom_score_adj. 1772 1773The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1774value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1775requires CAP_SYS_RESOURCE. 1776 1777 17783.2 /proc/<pid>/oom_score - Display current oom-killer score 1779------------------------------------------------------------- 1780 1781This file can be used to check the current score used by the oom-killer for 1782any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1783process should be killed in an out-of-memory situation. 1784 1785Please note that the exported value includes oom_score_adj so it is 1786effectively in range [0,2000]. 1787 1788 17893.3 /proc/<pid>/io - Display the IO accounting fields 1790------------------------------------------------------- 1791 1792This file contains IO statistics for each running process. 1793 1794Example 1795~~~~~~~ 1796 1797:: 1798 1799 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1800 [1] 3828 1801 1802 test:/tmp # cat /proc/3828/io 1803 rchar: 323934931 1804 wchar: 323929600 1805 syscr: 632687 1806 syscw: 632675 1807 read_bytes: 0 1808 write_bytes: 323932160 1809 cancelled_write_bytes: 0 1810 1811 1812Description 1813~~~~~~~~~~~ 1814 1815rchar 1816^^^^^ 1817 1818I/O counter: chars read 1819The number of bytes which this task has caused to be read from storage. This 1820is simply the sum of bytes which this process passed to read() and pread(). 1821It includes things like tty IO and it is unaffected by whether or not actual 1822physical disk IO was required (the read might have been satisfied from 1823pagecache). 1824 1825 1826wchar 1827^^^^^ 1828 1829I/O counter: chars written 1830The number of bytes which this task has caused, or shall cause to be written 1831to disk. Similar caveats apply here as with rchar. 1832 1833 1834syscr 1835^^^^^ 1836 1837I/O counter: read syscalls 1838Attempt to count the number of read I/O operations, i.e. syscalls like read() 1839and pread(). 1840 1841 1842syscw 1843^^^^^ 1844 1845I/O counter: write syscalls 1846Attempt to count the number of write I/O operations, i.e. syscalls like 1847write() and pwrite(). 1848 1849 1850read_bytes 1851^^^^^^^^^^ 1852 1853I/O counter: bytes read 1854Attempt to count the number of bytes which this process really did cause to 1855be fetched from the storage layer. Done at the submit_bio() level, so it is 1856accurate for block-backed filesystems. <please add status regarding NFS and 1857CIFS at a later time> 1858 1859 1860write_bytes 1861^^^^^^^^^^^ 1862 1863I/O counter: bytes written 1864Attempt to count the number of bytes which this process caused to be sent to 1865the storage layer. This is done at page-dirtying time. 1866 1867 1868cancelled_write_bytes 1869^^^^^^^^^^^^^^^^^^^^^ 1870 1871The big inaccuracy here is truncate. If a process writes 1MB to a file and 1872then deletes the file, it will in fact perform no writeout. But it will have 1873been accounted as having caused 1MB of write. 1874In other words: The number of bytes which this process caused to not happen, 1875by truncating pagecache. A task can cause "negative" IO too. If this task 1876truncates some dirty pagecache, some IO which another task has been accounted 1877for (in its write_bytes) will not be happening. We _could_ just subtract that 1878from the truncating task's write_bytes, but there is information loss in doing 1879that. 1880 1881 1882.. Note:: 1883 1884 At its current implementation state, this is a bit racy on 32-bit machines: 1885 if process A reads process B's /proc/pid/io while process B is updating one 1886 of those 64-bit counters, process A could see an intermediate result. 1887 1888 1889More information about this can be found within the taskstats documentation in 1890Documentation/accounting. 1891 18923.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1893--------------------------------------------------------------- 1894When a process is dumped, all anonymous memory is written to a core file as 1895long as the size of the core file isn't limited. But sometimes we don't want 1896to dump some memory segments, for example, huge shared memory or DAX. 1897Conversely, sometimes we want to save file-backed memory segments into a core 1898file, not only the individual files. 1899 1900/proc/<pid>/coredump_filter allows you to customize which memory segments 1901will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1902of memory types. If a bit of the bitmask is set, memory segments of the 1903corresponding memory type are dumped, otherwise they are not dumped. 1904 1905The following 9 memory types are supported: 1906 1907 - (bit 0) anonymous private memory 1908 - (bit 1) anonymous shared memory 1909 - (bit 2) file-backed private memory 1910 - (bit 3) file-backed shared memory 1911 - (bit 4) ELF header pages in file-backed private memory areas (it is 1912 effective only if the bit 2 is cleared) 1913 - (bit 5) hugetlb private memory 1914 - (bit 6) hugetlb shared memory 1915 - (bit 7) DAX private memory 1916 - (bit 8) DAX shared memory 1917 1918 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1919 are always dumped regardless of the bitmask status. 1920 1921 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1922 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1923 1924The default value of coredump_filter is 0x33; this means all anonymous memory 1925segments, ELF header pages and hugetlb private memory are dumped. 1926 1927If you don't want to dump all shared memory segments attached to pid 1234, 1928write 0x31 to the process's proc file:: 1929 1930 $ echo 0x31 > /proc/1234/coredump_filter 1931 1932When a new process is created, the process inherits the bitmask status from its 1933parent. It is useful to set up coredump_filter before the program runs. 1934For example:: 1935 1936 $ echo 0x7 > /proc/self/coredump_filter 1937 $ ./some_program 1938 19393.5 /proc/<pid>/mountinfo - Information about mounts 1940-------------------------------------------------------- 1941 1942This file contains lines of the form:: 1943 1944 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1945 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4) 1946 1947 (1) mount ID: unique identifier of the mount (may be reused after umount) 1948 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1949 (3) major:minor: value of st_dev for files on filesystem 1950 (4) root: root of the mount within the filesystem 1951 (5) mount point: mount point relative to the process's root 1952 (6) mount options: per mount options 1953 (n…m) optional fields: zero or more fields of the form "tag[:value]" 1954 (m+1) separator: marks the end of the optional fields 1955 (m+2) filesystem type: name of filesystem of the form "type[.subtype]" 1956 (m+3) mount source: filesystem specific information or "none" 1957 (m+4) super options: per super block options 1958 1959Parsers should ignore all unrecognised optional fields. Currently the 1960possible optional fields are: 1961 1962================ ============================================================== 1963shared:X mount is shared in peer group X 1964master:X mount is slave to peer group X 1965propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1966unbindable mount is unbindable 1967================ ============================================================== 1968 1969.. [#] X is the closest dominant peer group under the process's root. If 1970 X is the immediate master of the mount, or if there's no dominant peer 1971 group under the same root, then only the "master:X" field is present 1972 and not the "propagate_from:X" field. 1973 1974For more information on mount propagation see: 1975 1976 Documentation/filesystems/sharedsubtree.rst 1977 1978 19793.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1980-------------------------------------------------------- 1981These files provide a method to access a task's comm value. It also allows for 1982a task to set its own or one of its thread siblings comm value. The comm value 1983is limited in size compared to the cmdline value, so writing anything longer 1984then the kernel's TASK_COMM_LEN (currently 16 chars, including the NUL 1985terminator) will result in a truncated comm value. 1986 1987 19883.7 /proc/<pid>/task/<tid>/children - Information about task children 1989------------------------------------------------------------------------- 1990This file provides a fast way to retrieve first level children pids 1991of a task pointed by <pid>/<tid> pair. The format is a space separated 1992stream of pids. 1993 1994Note the "first level" here -- if a child has its own children they will 1995not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1996to obtain the descendants. 1997 1998Since this interface is intended to be fast and cheap it doesn't 1999guarantee to provide precise results and some children might be 2000skipped, especially if they've exited right after we printed their 2001pids, so one needs to either stop or freeze processes being inspected 2002if precise results are needed. 2003 2004 20053.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 2006--------------------------------------------------------------- 2007This file provides information associated with an opened file. The regular 2008files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. 2009The 'pos' represents the current offset of the opened file in decimal 2010form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the 2011file has been created with [see open(2) for details] and 'mnt_id' represents 2012mount ID of the file system containing the opened file [see 3.5 2013/proc/<pid>/mountinfo for details]. 'ino' represents the inode number of 2014the file. 2015 2016A typical output is:: 2017 2018 pos: 0 2019 flags: 0100002 2020 mnt_id: 19 2021 ino: 63107 2022 2023All locks associated with a file descriptor are shown in its fdinfo too:: 2024 2025 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 2026 2027The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 2028pair provide additional information particular to the objects they represent. 2029 2030Eventfd files 2031~~~~~~~~~~~~~ 2032 2033:: 2034 2035 pos: 0 2036 flags: 04002 2037 mnt_id: 9 2038 ino: 63107 2039 eventfd-count: 5a 2040 2041where 'eventfd-count' is hex value of a counter. 2042 2043Signalfd files 2044~~~~~~~~~~~~~~ 2045 2046:: 2047 2048 pos: 0 2049 flags: 04002 2050 mnt_id: 9 2051 ino: 63107 2052 sigmask: 0000000000000200 2053 2054where 'sigmask' is hex value of the signal mask associated 2055with a file. 2056 2057Epoll files 2058~~~~~~~~~~~ 2059 2060:: 2061 2062 pos: 0 2063 flags: 02 2064 mnt_id: 9 2065 ino: 63107 2066 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 2067 2068where 'tfd' is a target file descriptor number in decimal form, 2069'events' is events mask being watched and the 'data' is data 2070associated with a target [see epoll(7) for more details]. 2071 2072The 'pos' is current offset of the target file in decimal form 2073[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 2074where target file resides, all in hex format. 2075 2076Fsnotify files 2077~~~~~~~~~~~~~~ 2078For inotify files the format is the following:: 2079 2080 pos: 0 2081 flags: 02000000 2082 mnt_id: 9 2083 ino: 63107 2084 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 2085 2086where 'wd' is a watch descriptor in decimal form, i.e. a target file 2087descriptor number, 'ino' and 'sdev' are inode and device where the 2088target file resides and the 'mask' is the mask of events, all in hex 2089form [see inotify(7) for more details]. 2090 2091If the kernel was built with exportfs support, the path to the target 2092file is encoded as a file handle. The file handle is provided by three 2093fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 2094format. 2095 2096If the kernel is built without exportfs support the file handle won't be 2097printed out. 2098 2099If there is no inotify mark attached yet the 'inotify' line will be omitted. 2100 2101For fanotify files the format is:: 2102 2103 pos: 0 2104 flags: 02 2105 mnt_id: 9 2106 ino: 63107 2107 fanotify flags:10 event-flags:0 2108 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 2109 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 2110 2111where fanotify 'flags' and 'event-flags' are values used in fanotify_init 2112call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 2113flags associated with mark which are tracked separately from events 2114mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 2115mask and 'ignored_mask' is the mask of events which are to be ignored. 2116All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2117provide information about flags and mask used in fanotify_mark 2118call [see fsnotify manpage for details]. 2119 2120While the first three lines are mandatory and always printed, the rest is 2121optional and may be omitted if no marks created yet. 2122 2123Timerfd files 2124~~~~~~~~~~~~~ 2125 2126:: 2127 2128 pos: 0 2129 flags: 02 2130 mnt_id: 9 2131 ino: 63107 2132 clockid: 0 2133 ticks: 0 2134 settime flags: 01 2135 it_value: (0, 49406829) 2136 it_interval: (1, 0) 2137 2138where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2139that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2140flags in octal form been used to setup the timer [see timerfd_settime(2) for 2141details]. 'it_value' is remaining time until the timer expiration. 2142'it_interval' is the interval for the timer. Note the timer might be set up 2143with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2144still exhibits timer's remaining time. 2145 2146DMA Buffer files 2147~~~~~~~~~~~~~~~~ 2148 2149:: 2150 2151 pos: 0 2152 flags: 04002 2153 mnt_id: 9 2154 ino: 63107 2155 size: 32768 2156 count: 2 2157 exp_name: system-heap 2158 2159where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of 2160the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. 2161 2162VFIO Device files 2163~~~~~~~~~~~~~~~~~ 2164 2165:: 2166 2167 pos: 0 2168 flags: 02000002 2169 mnt_id: 17 2170 ino: 5122 2171 vfio-device-syspath: /sys/devices/pci0000:e0/0000:e0:01.1/0000:e1:00.0/0000:e2:05.0/0000:e8:00.0 2172 2173where 'vfio-device-syspath' is the sysfs path corresponding to the VFIO device 2174file. 2175 21763.9 /proc/<pid>/map_files - Information about memory mapped files 2177--------------------------------------------------------------------- 2178This directory contains symbolic links which represent memory mapped files 2179the process is maintaining. Example output:: 2180 2181 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2182 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2183 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2184 | ... 2185 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2186 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2187 2188The name of a link represents the virtual memory bounds of a mapping, i.e. 2189vm_area_struct::vm_start-vm_area_struct::vm_end. 2190 2191The main purpose of the map_files is to retrieve a set of memory mapped 2192files in a fast way instead of parsing /proc/<pid>/maps or 2193/proc/<pid>/smaps, both of which contain many more records. At the same 2194time one can open(2) mappings from the listings of two processes and 2195comparing their inode numbers to figure out which anonymous memory areas 2196are actually shared. 2197 21983.10 /proc/<pid>/timerslack_ns - Task timerslack value 2199--------------------------------------------------------- 2200This file provides the value of the task's timerslack value in nanoseconds. 2201This value specifies an amount of time that normal timers may be deferred 2202in order to coalesce timers and avoid unnecessary wakeups. 2203 2204This allows a task's interactivity vs power consumption tradeoff to be 2205adjusted. 2206 2207Writing 0 to the file will set the task's timerslack to the default value. 2208 2209Valid values are from 0 - ULLONG_MAX 2210 2211An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2212permissions on the task specified to change its timerslack_ns value. 2213 22143.11 /proc/<pid>/patch_state - Livepatch patch operation state 2215----------------------------------------------------------------- 2216When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2217patch state for the task. 2218 2219A value of '-1' indicates that no patch is in transition. 2220 2221A value of '0' indicates that a patch is in transition and the task is 2222unpatched. If the patch is being enabled, then the task hasn't been 2223patched yet. If the patch is being disabled, then the task has already 2224been unpatched. 2225 2226A value of '1' indicates that a patch is in transition and the task is 2227patched. If the patch is being enabled, then the task has already been 2228patched. If the patch is being disabled, then the task hasn't been 2229unpatched yet. 2230 22313.12 /proc/<pid>/arch_status - task architecture specific status 2232------------------------------------------------------------------- 2233When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2234architecture specific status of the task. 2235 2236Example 2237~~~~~~~ 2238 2239:: 2240 2241 $ cat /proc/6753/arch_status 2242 AVX512_elapsed_ms: 8 2243 2244Description 2245~~~~~~~~~~~ 2246 2247x86 specific entries 2248~~~~~~~~~~~~~~~~~~~~~ 2249 2250AVX512_elapsed_ms 2251^^^^^^^^^^^^^^^^^^ 2252 2253 If AVX512 is supported on the machine, this entry shows the milliseconds 2254 elapsed since the last time AVX512 usage was recorded. The recording 2255 happens on a best effort basis when a task is scheduled out. This means 2256 that the value depends on two factors: 2257 2258 1) The time which the task spent on the CPU without being scheduled 2259 out. With CPU isolation and a single runnable task this can take 2260 several seconds. 2261 2262 2) The time since the task was scheduled out last. Depending on the 2263 reason for being scheduled out (time slice exhausted, syscall ...) 2264 this can be arbitrary long time. 2265 2266 As a consequence the value cannot be considered precise and authoritative 2267 information. The application which uses this information has to be aware 2268 of the overall scenario on the system in order to determine whether a 2269 task is a real AVX512 user or not. Precise information can be obtained 2270 with performance counters. 2271 2272 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2273 the task is unlikely an AVX512 user, but depends on the workload and the 2274 scheduling scenario, it also could be a false negative mentioned above. 2275 22763.13 /proc/<pid>/fd - List of symlinks to open files 2277------------------------------------------------------- 2278This directory contains symbolic links which represent open files 2279the process is maintaining. Example output:: 2280 2281 lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null 2282 l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null 2283 lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]' 2284 lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]' 2285 lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]' 2286 2287The number of open files for the process is stored in 'size' member 2288of stat() output for /proc/<pid>/fd for fast access. 2289------------------------------------------------------- 2290 22913.14 /proc/<pid/ksm_stat - Information about the process's ksm status 2292--------------------------------------------------------------------- 2293When CONFIG_KSM is enabled, each process has this file which displays 2294the information of ksm merging status. 2295 2296Example 2297~~~~~~~ 2298 2299:: 2300 2301 / # cat /proc/self/ksm_stat 2302 ksm_rmap_items 0 2303 ksm_zero_pages 0 2304 ksm_merging_pages 0 2305 ksm_process_profit 0 2306 ksm_merge_any: no 2307 ksm_mergeable: no 2308 2309Description 2310~~~~~~~~~~~ 2311 2312ksm_rmap_items 2313^^^^^^^^^^^^^^ 2314 2315The number of ksm_rmap_item structures in use. The structure 2316ksm_rmap_item stores the reverse mapping information for virtual 2317addresses. KSM will generate a ksm_rmap_item for each ksm-scanned page of 2318the process. 2319 2320ksm_zero_pages 2321^^^^^^^^^^^^^^ 2322 2323When /sys/kernel/mm/ksm/use_zero_pages is enabled, it represent how many 2324empty pages are merged with kernel zero pages by KSM. 2325 2326ksm_merging_pages 2327^^^^^^^^^^^^^^^^^ 2328 2329It represents how many pages of this process are involved in KSM merging 2330(not including ksm_zero_pages). It is the same with what 2331/proc/<pid>/ksm_merging_pages shows. 2332 2333ksm_process_profit 2334^^^^^^^^^^^^^^^^^^ 2335 2336The profit that KSM brings (Saved bytes). KSM can save memory by merging 2337identical pages, but also can consume additional memory, because it needs 2338to generate a number of rmap_items to save each scanned page's brief rmap 2339information. Some of these pages may be merged, but some may not be abled 2340to be merged after being checked several times, which are unprofitable 2341memory consumed. 2342 2343ksm_merge_any 2344^^^^^^^^^^^^^ 2345 2346It specifies whether the process's 'mm is added by prctl() into the 2347candidate list of KSM or not, and if KSM scanning is fully enabled at 2348process level. 2349 2350ksm_mergeable 2351^^^^^^^^^^^^^ 2352 2353It specifies whether any VMAs of the process''s mms are currently 2354applicable to KSM. 2355 2356More information about KSM can be found in 2357Documentation/admin-guide/mm/ksm.rst. 2358 2359 2360Chapter 4: Configuring procfs 2361============================= 2362 23634.1 Mount options 2364--------------------- 2365 2366The following mount options are supported: 2367 2368 ========= ======================================================== 2369 hidepid= Set /proc/<pid>/ access mode. 2370 gid= Set the group authorized to learn processes information. 2371 subset= Show only the specified subset of procfs. 2372 pidns= Specify a the namespace used by this procfs. 2373 ========= ======================================================== 2374 2375hidepid=off or hidepid=0 means classic mode - everybody may access all 2376/proc/<pid>/ directories (default). 2377 2378hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2379directories but their own. Sensitive files like cmdline, sched*, status are now 2380protected against other users. This makes it impossible to learn whether any 2381user runs specific program (given the program doesn't reveal itself by its 2382behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2383other users, poorly written programs passing sensitive information via program 2384arguments are now protected against local eavesdroppers. 2385 2386hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2387fully invisible to other users. It doesn't mean that it hides a fact whether a 2388process with a specific pid value exists (it can be learned by other means, e.g. 2389by "kill -0 $PID"), but it hides process's uid and gid, which may be learned by 2390stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2391gathering information about running processes, whether some daemon runs with 2392elevated privileges, whether other user runs some sensitive program, whether 2393other users run any program at all, etc. 2394 2395hidepid=ptraceable or hidepid=4 means that procfs should only contain 2396/proc/<pid>/ directories that the caller can ptrace. 2397 2398gid= defines a group authorized to learn processes information otherwise 2399prohibited by hidepid=. If you use some daemon like identd which needs to learn 2400information about processes information, just add identd to this group. 2401 2402subset=pid hides all top level files and directories in the procfs that 2403are not related to tasks. 2404 2405pidns= specifies a pid namespace (either as a string path to something like 2406`/proc/$pid/ns/pid`, or a file descriptor when using `FSCONFIG_SET_FD`) that 2407will be used by the procfs instance when translating pids. By default, procfs 2408will use the calling process's active pid namespace. Note that the pid 2409namespace of an existing procfs instance cannot be modified (attempting to do 2410so will give an `-EBUSY` error). 2411 2412Chapter 5: Filesystem behavior 2413============================== 2414 2415Originally, before the advent of pid namespace, procfs was a global file 2416system. It means that there was only one procfs instance in the system. 2417 2418When pid namespace was added, a separate procfs instance was mounted in 2419each pid namespace. So, procfs mount options are global among all 2420mountpoints within the same namespace:: 2421 2422 # grep ^proc /proc/mounts 2423 proc /proc proc rw,relatime,hidepid=2 0 0 2424 2425 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2426 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2427 +++ exited with 0 +++ 2428 2429 # grep ^proc /proc/mounts 2430 proc /proc proc rw,relatime,hidepid=2 0 0 2431 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2432 2433and only after remounting procfs mount options will change at all 2434mountpoints:: 2435 2436 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2437 2438 # grep ^proc /proc/mounts 2439 proc /proc proc rw,relatime,hidepid=1 0 0 2440 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2441 2442This behavior is different from the behavior of other filesystems. 2443 2444The new procfs behavior is more like other filesystems. Each procfs mount 2445creates a new procfs instance. Mount options affect own procfs instance. 2446It means that it became possible to have several procfs instances 2447displaying tasks with different filtering options in one pid namespace:: 2448 2449 # mount -o hidepid=invisible -t proc proc /proc 2450 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2451 # grep ^proc /proc/mounts 2452 proc /proc proc rw,relatime,hidepid=invisible 0 0 2453 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2454