1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 9 Bodo Bauer <bb@ricochet.net> 102.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 11move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 12fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 13===================== ======================================= ================ 14 15 16 17.. Table of Contents 18 19 0 Preface 20 0.1 Introduction/Credits 21 0.2 Legal Stuff 22 23 1 Collecting System Information 24 1.1 Process-Specific Subdirectories 25 1.2 Kernel data 26 1.3 IDE devices in /proc/ide 27 1.4 Networking info in /proc/net 28 1.5 SCSI info 29 1.6 Parallel port info in /proc/parport 30 1.7 TTY info in /proc/tty 31 1.8 Miscellaneous kernel statistics in /proc/stat 32 1.9 Ext4 file system parameters 33 34 2 Modifying System Parameters 35 36 3 Per-Process Parameters 37 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 38 score 39 3.2 /proc/<pid>/oom_score - Display current oom-killer score 40 3.3 /proc/<pid>/io - Display the IO accounting fields 41 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings 42 3.5 /proc/<pid>/mountinfo - Information about mounts 43 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 44 3.7 /proc/<pid>/task/<tid>/children - Information about task children 45 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 46 3.9 /proc/<pid>/map_files - Information about memory mapped files 47 3.10 /proc/<pid>/timerslack_ns - Task timerslack value 48 3.11 /proc/<pid>/patch_state - Livepatch patch operation state 49 3.12 /proc/<pid>/arch_status - Task architecture specific information 50 51 4 Configuring procfs 52 4.1 Mount options 53 54 5 Filesystem behavior 55 56Preface 57======= 58 590.1 Introduction/Credits 60------------------------ 61 62This documentation is part of a soon (or so we hope) to be released book on 63the SuSE Linux distribution. As there is no complete documentation for the 64/proc file system and we've used many freely available sources to write these 65chapters, it seems only fair to give the work back to the Linux community. 66This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 67afraid it's still far from complete, but we hope it will be useful. As far as 68we know, it is the first 'all-in-one' document about the /proc file system. It 69is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 70SPARC, AXP, etc., features, you probably won't find what you are looking for. 71It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 72additions and patches are welcome and will be added to this document if you 73mail them to Bodo. 74 75We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 76other people for help compiling this documentation. We'd also like to extend a 77special thank you to Andi Kleen for documentation, which we relied on heavily 78to create this document, as well as the additional information he provided. 79Thanks to everybody else who contributed source or docs to the Linux kernel 80and helped create a great piece of software... :) 81 82If you have any comments, corrections or additions, please don't hesitate to 83contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 84document. 85 86The latest version of this document is available online at 87http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html 88 89If the above direction does not works for you, you could try the kernel 90mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 91comandante@zaralinux.com. 92 930.2 Legal Stuff 94--------------- 95 96We don't guarantee the correctness of this document, and if you come to us 97complaining about how you screwed up your system because of incorrect 98documentation, we won't feel responsible... 99 100Chapter 1: Collecting System Information 101======================================== 102 103In This Chapter 104--------------- 105* Investigating the properties of the pseudo file system /proc and its 106 ability to provide information on the running Linux system 107* Examining /proc's structure 108* Uncovering various information about the kernel and the processes running 109 on the system 110 111------------------------------------------------------------------------------ 112 113The proc file system acts as an interface to internal data structures in the 114kernel. It can be used to obtain information about the system and to change 115certain kernel parameters at runtime (sysctl). 116 117First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 118show you how you can use /proc/sys to change settings. 119 1201.1 Process-Specific Subdirectories 121----------------------------------- 122 123The directory /proc contains (among other things) one subdirectory for each 124process running on the system, which is named after the process ID (PID). 125 126The link 'self' points to the process reading the file system. Each process 127subdirectory has the entries listed in Table 1-1. 128 129Note that an open file descriptor to /proc/<pid> or to any of its 130contained files or subdirectories does not prevent <pid> being reused 131for some other process in the event that <pid> exits. Operations on 132open /proc/<pid> file descriptors corresponding to dead processes 133never act on any new process that the kernel may, through chance, have 134also assigned the process ID <pid>. Instead, operations on these FDs 135usually fail with ESRCH. 136 137.. table:: Table 1-1: Process specific entries in /proc 138 139 ============= =============================================================== 140 File Content 141 ============= =============================================================== 142 clear_refs Clears page referenced bits shown in smaps output 143 cmdline Command line arguments 144 cpu Current and last cpu in which it was executed (2.4)(smp) 145 cwd Link to the current working directory 146 environ Values of environment variables 147 exe Link to the executable of this process 148 fd Directory, which contains all file descriptors 149 maps Memory maps to executables and library files (2.4) 150 mem Memory held by this process 151 root Link to the root directory of this process 152 stat Process status 153 statm Process memory status information 154 status Process status in human readable form 155 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 156 symbol the task is blocked in - or "0" if not blocked. 157 pagemap Page table 158 stack Report full stack trace, enable via CONFIG_STACKTRACE 159 smaps An extension based on maps, showing the memory consumption of 160 each mapping and flags associated with it 161 smaps_rollup Accumulated smaps stats for all mappings of the process. This 162 can be derived from smaps, but is faster and more convenient 163 numa_maps An extension based on maps, showing the memory locality and 164 binding policy as well as mem usage (in pages) of each mapping. 165 ============= =============================================================== 166 167For example, to get the status information of a process, all you have to do is 168read the file /proc/PID/status:: 169 170 >cat /proc/self/status 171 Name: cat 172 State: R (running) 173 Tgid: 5452 174 Pid: 5452 175 PPid: 743 176 TracerPid: 0 (2.4) 177 Uid: 501 501 501 501 178 Gid: 100 100 100 100 179 FDSize: 256 180 Groups: 100 14 16 181 VmPeak: 5004 kB 182 VmSize: 5004 kB 183 VmLck: 0 kB 184 VmHWM: 476 kB 185 VmRSS: 476 kB 186 RssAnon: 352 kB 187 RssFile: 120 kB 188 RssShmem: 4 kB 189 VmData: 156 kB 190 VmStk: 88 kB 191 VmExe: 68 kB 192 VmLib: 1412 kB 193 VmPTE: 20 kb 194 VmSwap: 0 kB 195 HugetlbPages: 0 kB 196 CoreDumping: 0 197 THP_enabled: 1 198 Threads: 1 199 SigQ: 0/28578 200 SigPnd: 0000000000000000 201 ShdPnd: 0000000000000000 202 SigBlk: 0000000000000000 203 SigIgn: 0000000000000000 204 SigCgt: 0000000000000000 205 CapInh: 00000000fffffeff 206 CapPrm: 0000000000000000 207 CapEff: 0000000000000000 208 CapBnd: ffffffffffffffff 209 CapAmb: 0000000000000000 210 NoNewPrivs: 0 211 Seccomp: 0 212 Speculation_Store_Bypass: thread vulnerable 213 voluntary_ctxt_switches: 0 214 nonvoluntary_ctxt_switches: 1 215 216This shows you nearly the same information you would get if you viewed it with 217the ps command. In fact, ps uses the proc file system to obtain its 218information. But you get a more detailed view of the process by reading the 219file /proc/PID/status. It fields are described in table 1-2. 220 221The statm file contains more detailed information about the process 222memory usage. Its seven fields are explained in Table 1-3. The stat file 223contains detailed information about the process itself. Its fields are 224explained in Table 1-4. 225 226(for SMP CONFIG users) 227 228For making accounting scalable, RSS related information are handled in an 229asynchronous manner and the value may not be very precise. To see a precise 230snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 231It's slow but very precise. 232 233.. table:: Table 1-2: Contents of the status files (as of 4.19) 234 235 ========================== =================================================== 236 Field Content 237 ========================== =================================================== 238 Name filename of the executable 239 Umask file mode creation mask 240 State state (R is running, S is sleeping, D is sleeping 241 in an uninterruptible wait, Z is zombie, 242 T is traced or stopped) 243 Tgid thread group ID 244 Ngid NUMA group ID (0 if none) 245 Pid process id 246 PPid process id of the parent process 247 TracerPid PID of process tracing this process (0 if not) 248 Uid Real, effective, saved set, and file system UIDs 249 Gid Real, effective, saved set, and file system GIDs 250 FDSize number of file descriptor slots currently allocated 251 Groups supplementary group list 252 NStgid descendant namespace thread group ID hierarchy 253 NSpid descendant namespace process ID hierarchy 254 NSpgid descendant namespace process group ID hierarchy 255 NSsid descendant namespace session ID hierarchy 256 VmPeak peak virtual memory size 257 VmSize total program size 258 VmLck locked memory size 259 VmPin pinned memory size 260 VmHWM peak resident set size ("high water mark") 261 VmRSS size of memory portions. It contains the three 262 following parts 263 (VmRSS = RssAnon + RssFile + RssShmem) 264 RssAnon size of resident anonymous memory 265 RssFile size of resident file mappings 266 RssShmem size of resident shmem memory (includes SysV shm, 267 mapping of tmpfs and shared anonymous mappings) 268 VmData size of private data segments 269 VmStk size of stack segments 270 VmExe size of text segment 271 VmLib size of shared library code 272 VmPTE size of page table entries 273 VmSwap amount of swap used by anonymous private data 274 (shmem swap usage is not included) 275 HugetlbPages size of hugetlb memory portions 276 CoreDumping process's memory is currently being dumped 277 (killing the process may lead to a corrupted core) 278 THP_enabled process is allowed to use THP (returns 0 when 279 PR_SET_THP_DISABLE is set on the process 280 Threads number of threads 281 SigQ number of signals queued/max. number for queue 282 SigPnd bitmap of pending signals for the thread 283 ShdPnd bitmap of shared pending signals for the process 284 SigBlk bitmap of blocked signals 285 SigIgn bitmap of ignored signals 286 SigCgt bitmap of caught signals 287 CapInh bitmap of inheritable capabilities 288 CapPrm bitmap of permitted capabilities 289 CapEff bitmap of effective capabilities 290 CapBnd bitmap of capabilities bounding set 291 CapAmb bitmap of ambient capabilities 292 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) 293 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 294 Speculation_Store_Bypass speculative store bypass mitigation status 295 Cpus_allowed mask of CPUs on which this process may run 296 Cpus_allowed_list Same as previous, but in "list format" 297 Mems_allowed mask of memory nodes allowed to this process 298 Mems_allowed_list Same as previous, but in "list format" 299 voluntary_ctxt_switches number of voluntary context switches 300 nonvoluntary_ctxt_switches number of non voluntary context switches 301 ========================== =================================================== 302 303 304.. table:: Table 1-3: Contents of the statm files (as of 2.6.8-rc3) 305 306 ======== =============================== ============================== 307 Field Content 308 ======== =============================== ============================== 309 size total program size (pages) (same as VmSize in status) 310 resident size of memory portions (pages) (same as VmRSS in status) 311 shared number of pages that are shared (i.e. backed by a file, same 312 as RssFile+RssShmem in status) 313 trs number of pages that are 'code' (not including libs; broken, 314 includes data segment) 315 lrs number of pages of library (always 0 on 2.6) 316 drs number of pages of data/stack (including libs; broken, 317 includes library text) 318 dt number of dirty pages (always 0 on 2.6) 319 ======== =============================== ============================== 320 321 322.. table:: Table 1-4: Contents of the stat files (as of 2.6.30-rc7) 323 324 ============= =============================================================== 325 Field Content 326 ============= =============================================================== 327 pid process id 328 tcomm filename of the executable 329 state state (R is running, S is sleeping, D is sleeping in an 330 uninterruptible wait, Z is zombie, T is traced or stopped) 331 ppid process id of the parent process 332 pgrp pgrp of the process 333 sid session id 334 tty_nr tty the process uses 335 tty_pgrp pgrp of the tty 336 flags task flags 337 min_flt number of minor faults 338 cmin_flt number of minor faults with child's 339 maj_flt number of major faults 340 cmaj_flt number of major faults with child's 341 utime user mode jiffies 342 stime kernel mode jiffies 343 cutime user mode jiffies with child's 344 cstime kernel mode jiffies with child's 345 priority priority level 346 nice nice level 347 num_threads number of threads 348 it_real_value (obsolete, always 0) 349 start_time time the process started after system boot 350 vsize virtual memory size 351 rss resident set memory size 352 rsslim current limit in bytes on the rss 353 start_code address above which program text can run 354 end_code address below which program text can run 355 start_stack address of the start of the main process stack 356 esp current value of ESP 357 eip current value of EIP 358 pending bitmap of pending signals 359 blocked bitmap of blocked signals 360 sigign bitmap of ignored signals 361 sigcatch bitmap of caught signals 362 0 (place holder, used to be the wchan address, 363 use /proc/PID/wchan instead) 364 0 (place holder) 365 0 (place holder) 366 exit_signal signal to send to parent thread on exit 367 task_cpu which CPU the task is scheduled on 368 rt_priority realtime priority 369 policy scheduling policy (man sched_setscheduler) 370 blkio_ticks time spent waiting for block IO 371 gtime guest time of the task in jiffies 372 cgtime guest time of the task children in jiffies 373 start_data address above which program data+bss is placed 374 end_data address below which program data+bss is placed 375 start_brk address above which program heap can be expanded with brk() 376 arg_start address above which program command line is placed 377 arg_end address below which program command line is placed 378 env_start address above which program environment is placed 379 env_end address below which program environment is placed 380 exit_code the thread's exit_code in the form reported by the waitpid 381 system call 382 ============= =============================================================== 383 384The /proc/PID/maps file contains the currently mapped memory regions and 385their access permissions. 386 387The format is:: 388 389 address perms offset dev inode pathname 390 391 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 392 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 393 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] 394 a7cb1000-a7cb2000 ---p 00000000 00:00 0 395 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 396 a7eb2000-a7eb3000 ---p 00000000 00:00 0 397 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 398 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 399 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 400 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 401 a800b000-a800e000 rw-p 00000000 00:00 0 402 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 403 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 404 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 405 a8024000-a8027000 rw-p 00000000 00:00 0 406 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 407 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 408 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 409 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 410 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 411 412where "address" is the address space in the process that it occupies, "perms" 413is a set of permissions:: 414 415 r = read 416 w = write 417 x = execute 418 s = shared 419 p = private (copy on write) 420 421"offset" is the offset into the mapping, "dev" is the device (major:minor), and 422"inode" is the inode on that device. 0 indicates that no inode is associated 423with the memory region, as the case would be with BSS (uninitialized data). 424The "pathname" shows the name associated file for this mapping. If the mapping 425is not associated with a file: 426 427 ======= ==================================== 428 [heap] the heap of the program 429 [stack] the stack of the main process 430 [vdso] the "virtual dynamic shared object", 431 the kernel system call handler 432[anon:<name>] an anonymous mapping that has been 433 named by userspace 434 ======= ==================================== 435 436 or if empty, the mapping is anonymous. 437 438The /proc/PID/smaps is an extension based on maps, showing the memory 439consumption for each of the process's mappings. For each mapping (aka Virtual 440Memory Area, or VMA) there is a series of lines such as the following:: 441 442 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 443 444 Size: 1084 kB 445 KernelPageSize: 4 kB 446 MMUPageSize: 4 kB 447 Rss: 892 kB 448 Pss: 374 kB 449 Shared_Clean: 892 kB 450 Shared_Dirty: 0 kB 451 Private_Clean: 0 kB 452 Private_Dirty: 0 kB 453 Referenced: 892 kB 454 Anonymous: 0 kB 455 LazyFree: 0 kB 456 AnonHugePages: 0 kB 457 ShmemPmdMapped: 0 kB 458 Shared_Hugetlb: 0 kB 459 Private_Hugetlb: 0 kB 460 Swap: 0 kB 461 SwapPss: 0 kB 462 KernelPageSize: 4 kB 463 MMUPageSize: 4 kB 464 Locked: 0 kB 465 THPeligible: 0 466 VmFlags: rd ex mr mw me dw 467 468The first of these lines shows the same information as is displayed for the 469mapping in /proc/PID/maps. Following lines show the size of the mapping 470(size); the size of each page allocated when backing a VMA (KernelPageSize), 471which is usually the same as the size in the page table entries; the page size 472used by the MMU when backing a VMA (in most cases, the same as KernelPageSize); 473the amount of the mapping that is currently resident in RAM (RSS); the 474process' proportional share of this mapping (PSS); and the number of clean and 475dirty shared and private pages in the mapping. 476 477The "proportional set size" (PSS) of a process is the count of pages it has 478in memory, where each page is divided by the number of processes sharing it. 479So if a process has 1000 pages all to itself, and 1000 shared with one other 480process, its PSS will be 1500. 481 482Note that even a page which is part of a MAP_SHARED mapping, but has only 483a single pte mapped, i.e. is currently used by only one process, is accounted 484as private and not as shared. 485 486"Referenced" indicates the amount of memory currently marked as referenced or 487accessed. 488 489"Anonymous" shows the amount of memory that does not belong to any file. Even 490a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 491and a page is modified, the file page is replaced by a private anonymous copy. 492 493"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 494The memory isn't freed immediately with madvise(). It's freed in memory 495pressure if the memory is clean. Please note that the printed value might 496be lower than the real value due to optimizations used in the current 497implementation. If this is not desirable please file a bug report. 498 499"AnonHugePages" shows the ammount of memory backed by transparent hugepage. 500 501"ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by 502huge pages. 503 504"Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by 505hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 506reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 507 508"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 509 510For shmem mappings, "Swap" includes also the size of the mapped (and not 511replaced by copy-on-write) part of the underlying shmem object out on swap. 512"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 513does not take into account swapped out page of underlying shmem objects. 514"Locked" indicates whether the mapping is locked in memory or not. 515"THPeligible" indicates whether the mapping is eligible for allocating THP 516pages - 1 if true, 0 otherwise. It just shows the current status. 517 518"VmFlags" field deserves a separate description. This member represents the 519kernel flags associated with the particular virtual memory area in two letter 520encoded manner. The codes are the following: 521 522 == ======================================= 523 rd readable 524 wr writeable 525 ex executable 526 sh shared 527 mr may read 528 mw may write 529 me may execute 530 ms may share 531 gd stack segment growns down 532 pf pure PFN range 533 dw disabled write to the mapped file 534 lo pages are locked in memory 535 io memory mapped I/O area 536 sr sequential read advise provided 537 rr random read advise provided 538 dc do not copy area on fork 539 de do not expand area on remapping 540 ac area is accountable 541 nr swap space is not reserved for the area 542 ht area uses huge tlb pages 543 ar architecture specific flag 544 dd do not include area into core dump 545 sd soft dirty flag 546 mm mixed map area 547 hg huge page advise flag 548 nh no huge page advise flag 549 mg mergable advise flag 550 bt arm64 BTI guarded page 551 == ======================================= 552 553Note that there is no guarantee that every flag and associated mnemonic will 554be present in all further kernel releases. Things get changed, the flags may 555be vanished or the reverse -- new added. Interpretation of their meaning 556might change in future as well. So each consumer of these flags has to 557follow each specific kernel version for the exact semantic. 558 559This file is only present if the CONFIG_MMU kernel configuration option is 560enabled. 561 562Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 563output can be achieved only in the single read call). 564 565This typically manifests when doing partial reads of these files while the 566memory map is being modified. Despite the races, we do provide the following 567guarantees: 568 5691) The mapped addresses never go backwards, which implies no two 570 regions will ever overlap. 5712) If there is something at a given vaddr during the entirety of the 572 life of the smaps/maps walk, there will be some output for it. 573 574The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 575but their values are the sums of the corresponding values for all mappings of 576the process. Additionally, it contains these fields: 577 578- Pss_Anon 579- Pss_File 580- Pss_Shmem 581 582They represent the proportional shares of anonymous, file, and shmem pages, as 583described for smaps above. These fields are omitted in smaps since each 584mapping identifies the type (anon, file, or shmem) of all pages it contains. 585Thus all information in smaps_rollup can be derived from smaps, but at a 586significantly higher cost. 587 588The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 589bits on both physical and virtual pages associated with a process, and the 590soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 591for details). 592To clear the bits for all the pages associated with the process:: 593 594 > echo 1 > /proc/PID/clear_refs 595 596To clear the bits for the anonymous pages associated with the process:: 597 598 > echo 2 > /proc/PID/clear_refs 599 600To clear the bits for the file mapped pages associated with the process:: 601 602 > echo 3 > /proc/PID/clear_refs 603 604To clear the soft-dirty bit:: 605 606 > echo 4 > /proc/PID/clear_refs 607 608To reset the peak resident set size ("high water mark") to the process's 609current value:: 610 611 > echo 5 > /proc/PID/clear_refs 612 613Any other value written to /proc/PID/clear_refs will have no effect. 614 615The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 616using /proc/kpageflags and number of times a page is mapped using 617/proc/kpagecount. For detailed explanation, see 618Documentation/admin-guide/mm/pagemap.rst. 619 620The /proc/pid/numa_maps is an extension based on maps, showing the memory 621locality and binding policy, as well as the memory usage (in pages) of 622each mapping. The output follows a general format where mapping details get 623summarized separated by blank spaces, one mapping per each file line:: 624 625 address policy mapping details 626 627 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 628 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 629 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 630 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 631 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 632 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 633 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 634 320698b000 default file=/lib64/libc-2.12.so 635 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 636 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 637 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 638 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 639 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 640 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 641 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 642 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 643 644Where: 645 646"address" is the starting address for the mapping; 647 648"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 649 650"mapping details" summarizes mapping data such as mapping type, page usage counters, 651node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 652size, in KB, that is backing the mapping up. 653 6541.2 Kernel data 655--------------- 656 657Similar to the process entries, the kernel data files give information about 658the running kernel. The files used to obtain this information are contained in 659/proc and are listed in Table 1-5. Not all of these will be present in your 660system. It depends on the kernel configuration and the loaded modules, which 661files are there, and which are missing. 662 663.. table:: Table 1-5: Kernel info in /proc 664 665 ============ =============================================================== 666 File Content 667 ============ =============================================================== 668 apm Advanced power management info 669 buddyinfo Kernel memory allocator information (see text) (2.5) 670 bus Directory containing bus specific information 671 cmdline Kernel command line 672 cpuinfo Info about the CPU 673 devices Available devices (block and character) 674 dma Used DMS channels 675 filesystems Supported filesystems 676 driver Various drivers grouped here, currently rtc (2.4) 677 execdomains Execdomains, related to security (2.4) 678 fb Frame Buffer devices (2.4) 679 fs File system parameters, currently nfs/exports (2.4) 680 ide Directory containing info about the IDE subsystem 681 interrupts Interrupt usage 682 iomem Memory map (2.4) 683 ioports I/O port usage 684 irq Masks for irq to cpu affinity (2.4)(smp?) 685 isapnp ISA PnP (Plug&Play) Info (2.4) 686 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 687 kmsg Kernel messages 688 ksyms Kernel symbol table 689 loadavg Load average of last 1, 5 & 15 minutes 690 locks Kernel locks 691 meminfo Memory info 692 misc Miscellaneous 693 modules List of loaded modules 694 mounts Mounted filesystems 695 net Networking info (see text) 696 pagetypeinfo Additional page allocator information (see text) (2.5) 697 partitions Table of partitions known to the system 698 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 699 decoupled by lspci (2.4) 700 rtc Real time clock 701 scsi SCSI info (see text) 702 slabinfo Slab pool info 703 softirqs softirq usage 704 stat Overall statistics 705 swaps Swap space utilization 706 sys See chapter 2 707 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 708 tty Info of tty drivers 709 uptime Wall clock since boot, combined idle time of all cpus 710 version Kernel version 711 video bttv info of video resources (2.4) 712 vmallocinfo Show vmalloced areas 713 ============ =============================================================== 714 715You can, for example, check which interrupts are currently in use and what 716they are used for by looking in the file /proc/interrupts:: 717 718 > cat /proc/interrupts 719 CPU0 720 0: 8728810 XT-PIC timer 721 1: 895 XT-PIC keyboard 722 2: 0 XT-PIC cascade 723 3: 531695 XT-PIC aha152x 724 4: 2014133 XT-PIC serial 725 5: 44401 XT-PIC pcnet_cs 726 8: 2 XT-PIC rtc 727 11: 8 XT-PIC i82365 728 12: 182918 XT-PIC PS/2 Mouse 729 13: 1 XT-PIC fpu 730 14: 1232265 XT-PIC ide0 731 15: 7 XT-PIC ide1 732 NMI: 0 733 734In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 735output of a SMP machine):: 736 737 > cat /proc/interrupts 738 739 CPU0 CPU1 740 0: 1243498 1214548 IO-APIC-edge timer 741 1: 8949 8958 IO-APIC-edge keyboard 742 2: 0 0 XT-PIC cascade 743 5: 11286 10161 IO-APIC-edge soundblaster 744 8: 1 0 IO-APIC-edge rtc 745 9: 27422 27407 IO-APIC-edge 3c503 746 12: 113645 113873 IO-APIC-edge PS/2 Mouse 747 13: 0 0 XT-PIC fpu 748 14: 22491 24012 IO-APIC-edge ide0 749 15: 2183 2415 IO-APIC-edge ide1 750 17: 30564 30414 IO-APIC-level eth0 751 18: 177 164 IO-APIC-level bttv 752 NMI: 2457961 2457959 753 LOC: 2457882 2457881 754 ERR: 2155 755 756NMI is incremented in this case because every timer interrupt generates a NMI 757(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 758 759LOC is the local interrupt counter of the internal APIC of every CPU. 760 761ERR is incremented in the case of errors in the IO-APIC bus (the bus that 762connects the CPUs in a SMP system. This means that an error has been detected, 763the IO-APIC automatically retry the transmission, so it should not be a big 764problem, but you should read the SMP-FAQ. 765 766In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 767/proc/interrupts to display every IRQ vector in use by the system, not 768just those considered 'most important'. The new vectors are: 769 770THR 771 interrupt raised when a machine check threshold counter 772 (typically counting ECC corrected errors of memory or cache) exceeds 773 a configurable threshold. Only available on some systems. 774 775TRM 776 a thermal event interrupt occurs when a temperature threshold 777 has been exceeded for the CPU. This interrupt may also be generated 778 when the temperature drops back to normal. 779 780SPU 781 a spurious interrupt is some interrupt that was raised then lowered 782 by some IO device before it could be fully processed by the APIC. Hence 783 the APIC sees the interrupt but does not know what device it came from. 784 For this case the APIC will generate the interrupt with a IRQ vector 785 of 0xff. This might also be generated by chipset bugs. 786 787RES, CAL, TLB 788 rescheduling, call and TLB flush interrupts are 789 sent from one CPU to another per the needs of the OS. Typically, 790 their statistics are used by kernel developers and interested users to 791 determine the occurrence of interrupts of the given type. 792 793The above IRQ vectors are displayed only when relevant. For example, 794the threshold vector does not exist on x86_64 platforms. Others are 795suppressed when the system is a uniprocessor. As of this writing, only 796i386 and x86_64 platforms support the new IRQ vector displays. 797 798Of some interest is the introduction of the /proc/irq directory to 2.4. 799It could be used to set IRQ to CPU affinity. This means that you can "hook" an 800IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 801irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 802prof_cpu_mask. 803 804For example:: 805 806 > ls /proc/irq/ 807 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 808 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 809 > ls /proc/irq/0/ 810 smp_affinity 811 812smp_affinity is a bitmask, in which you can specify which CPUs can handle the 813IRQ. You can set it by doing:: 814 815 > echo 1 > /proc/irq/10/smp_affinity 816 817This means that only the first CPU will handle the IRQ, but you can also echo 8185 which means that only the first and third CPU can handle the IRQ. 819 820The contents of each smp_affinity file is the same by default:: 821 822 > cat /proc/irq/0/smp_affinity 823 ffffffff 824 825There is an alternate interface, smp_affinity_list which allows specifying 826a CPU range instead of a bitmask:: 827 828 > cat /proc/irq/0/smp_affinity_list 829 1024-1031 830 831The default_smp_affinity mask applies to all non-active IRQs, which are the 832IRQs which have not yet been allocated/activated, and hence which lack a 833/proc/irq/[0-9]* directory. 834 835The node file on an SMP system shows the node to which the device using the IRQ 836reports itself as being attached. This hardware locality information does not 837include information about any possible driver locality preference. 838 839prof_cpu_mask specifies which CPUs are to be profiled by the system wide 840profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 841 842The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 843between all the CPUs which are allowed to handle it. As usual the kernel has 844more info than you and does a better job than you, so the defaults are the 845best choice for almost everyone. [Note this applies only to those IO-APIC's 846that support "Round Robin" interrupt distribution.] 847 848There are three more important subdirectories in /proc: net, scsi, and sys. 849The general rule is that the contents, or even the existence of these 850directories, depend on your kernel configuration. If SCSI is not enabled, the 851directory scsi may not exist. The same is true with the net, which is there 852only when networking support is present in the running kernel. 853 854The slabinfo file gives information about memory usage at the slab level. 855Linux uses slab pools for memory management above page level in version 2.2. 856Commonly used objects have their own slab pool (such as network buffers, 857directory cache, and so on). 858 859:: 860 861 > cat /proc/buddyinfo 862 863 Node 0, zone DMA 0 4 5 4 4 3 ... 864 Node 0, zone Normal 1 0 0 1 101 8 ... 865 Node 0, zone HighMem 2 0 0 1 1 0 ... 866 867External fragmentation is a problem under some workloads, and buddyinfo is a 868useful tool for helping diagnose these problems. Buddyinfo will give you a 869clue as to how big an area you can safely allocate, or why a previous 870allocation failed. 871 872Each column represents the number of pages of a certain order which are 873available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 874ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 875available in ZONE_NORMAL, etc... 876 877More information relevant to external fragmentation can be found in 878pagetypeinfo:: 879 880 > cat /proc/pagetypeinfo 881 Page block order: 9 882 Pages per block: 512 883 884 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 885 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 886 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 887 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 888 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 889 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 890 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 891 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 892 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 893 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 894 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 895 896 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 897 Node 0, zone DMA 2 0 5 1 0 898 Node 0, zone DMA32 41 6 967 2 0 899 900Fragmentation avoidance in the kernel works by grouping pages of different 901migrate types into the same contiguous regions of memory called page blocks. 902A page block is typically the size of the default hugepage size, e.g. 2MB on 903X86-64. By keeping pages grouped based on their ability to move, the kernel 904can reclaim pages within a page block to satisfy a high-order allocation. 905 906The pagetypinfo begins with information on the size of a page block. It 907then gives the same type of information as buddyinfo except broken down 908by migrate-type and finishes with details on how many page blocks of each 909type exist. 910 911If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 912from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 913make an estimate of the likely number of huge pages that can be allocated 914at a given point in time. All the "Movable" blocks should be allocatable 915unless memory has been mlock()'d. Some of the Reclaimable blocks should 916also be allocatable although a lot of filesystem metadata may have to be 917reclaimed to achieve this. 918 919 920meminfo 921~~~~~~~ 922 923Provides information about distribution and utilization of memory. This 924varies by architecture and compile options. The following is from a 92516GB PIII, which has highmem enabled. You may not have all of these fields. 926 927:: 928 929 > cat /proc/meminfo 930 931 MemTotal: 16344972 kB 932 MemFree: 13634064 kB 933 MemAvailable: 14836172 kB 934 Buffers: 3656 kB 935 Cached: 1195708 kB 936 SwapCached: 0 kB 937 Active: 891636 kB 938 Inactive: 1077224 kB 939 HighTotal: 15597528 kB 940 HighFree: 13629632 kB 941 LowTotal: 747444 kB 942 LowFree: 4432 kB 943 SwapTotal: 0 kB 944 SwapFree: 0 kB 945 Dirty: 968 kB 946 Writeback: 0 kB 947 AnonPages: 861800 kB 948 Mapped: 280372 kB 949 Shmem: 644 kB 950 KReclaimable: 168048 kB 951 Slab: 284364 kB 952 SReclaimable: 159856 kB 953 SUnreclaim: 124508 kB 954 PageTables: 24448 kB 955 NFS_Unstable: 0 kB 956 Bounce: 0 kB 957 WritebackTmp: 0 kB 958 CommitLimit: 7669796 kB 959 Committed_AS: 100056 kB 960 VmallocTotal: 112216 kB 961 VmallocUsed: 428 kB 962 VmallocChunk: 111088 kB 963 Percpu: 62080 kB 964 HardwareCorrupted: 0 kB 965 AnonHugePages: 49152 kB 966 ShmemHugePages: 0 kB 967 ShmemPmdMapped: 0 kB 968 969MemTotal 970 Total usable RAM (i.e. physical RAM minus a few reserved 971 bits and the kernel binary code) 972MemFree 973 The sum of LowFree+HighFree 974MemAvailable 975 An estimate of how much memory is available for starting new 976 applications, without swapping. Calculated from MemFree, 977 SReclaimable, the size of the file LRU lists, and the low 978 watermarks in each zone. 979 The estimate takes into account that the system needs some 980 page cache to function well, and that not all reclaimable 981 slab will be reclaimable, due to items being in use. The 982 impact of those factors will vary from system to system. 983Buffers 984 Relatively temporary storage for raw disk blocks 985 shouldn't get tremendously large (20MB or so) 986Cached 987 in-memory cache for files read from the disk (the 988 pagecache). Doesn't include SwapCached 989SwapCached 990 Memory that once was swapped out, is swapped back in but 991 still also is in the swapfile (if memory is needed it 992 doesn't need to be swapped out AGAIN because it is already 993 in the swapfile. This saves I/O) 994Active 995 Memory that has been used more recently and usually not 996 reclaimed unless absolutely necessary. 997Inactive 998 Memory which has been less recently used. It is more 999 eligible to be reclaimed for other purposes 1000HighTotal, HighFree 1001 Highmem is all memory above ~860MB of physical memory. 1002 Highmem areas are for use by userspace programs, or 1003 for the pagecache. The kernel must use tricks to access 1004 this memory, making it slower to access than lowmem. 1005LowTotal, LowFree 1006 Lowmem is memory which can be used for everything that 1007 highmem can be used for, but it is also available for the 1008 kernel's use for its own data structures. Among many 1009 other things, it is where everything from the Slab is 1010 allocated. Bad things happen when you're out of lowmem. 1011SwapTotal 1012 total amount of swap space available 1013SwapFree 1014 Memory which has been evicted from RAM, and is temporarily 1015 on the disk 1016Dirty 1017 Memory which is waiting to get written back to the disk 1018Writeback 1019 Memory which is actively being written back to the disk 1020AnonPages 1021 Non-file backed pages mapped into userspace page tables 1022HardwareCorrupted 1023 The amount of RAM/memory in KB, the kernel identifies as 1024 corrupted. 1025AnonHugePages 1026 Non-file backed huge pages mapped into userspace page tables 1027Mapped 1028 files which have been mmaped, such as libraries 1029Shmem 1030 Total memory used by shared memory (shmem) and tmpfs 1031ShmemHugePages 1032 Memory used by shared memory (shmem) and tmpfs allocated 1033 with huge pages 1034ShmemPmdMapped 1035 Shared memory mapped into userspace with huge pages 1036KReclaimable 1037 Kernel allocations that the kernel will attempt to reclaim 1038 under memory pressure. Includes SReclaimable (below), and other 1039 direct allocations with a shrinker. 1040Slab 1041 in-kernel data structures cache 1042SReclaimable 1043 Part of Slab, that might be reclaimed, such as caches 1044SUnreclaim 1045 Part of Slab, that cannot be reclaimed on memory pressure 1046PageTables 1047 amount of memory dedicated to the lowest level of page 1048 tables. 1049NFS_Unstable 1050 Always zero. Previous counted pages which had been written to 1051 the server, but has not been committed to stable storage. 1052Bounce 1053 Memory used for block device "bounce buffers" 1054WritebackTmp 1055 Memory used by FUSE for temporary writeback buffers 1056CommitLimit 1057 Based on the overcommit ratio ('vm.overcommit_ratio'), 1058 this is the total amount of memory currently available to 1059 be allocated on the system. This limit is only adhered to 1060 if strict overcommit accounting is enabled (mode 2 in 1061 'vm.overcommit_memory'). 1062 1063 The CommitLimit is calculated with the following formula:: 1064 1065 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1066 overcommit_ratio / 100 + [total swap pages] 1067 1068 For example, on a system with 1G of physical RAM and 7G 1069 of swap with a `vm.overcommit_ratio` of 30 it would 1070 yield a CommitLimit of 7.3G. 1071 1072 For more details, see the memory overcommit documentation 1073 in vm/overcommit-accounting. 1074Committed_AS 1075 The amount of memory presently allocated on the system. 1076 The committed memory is a sum of all of the memory which 1077 has been allocated by processes, even if it has not been 1078 "used" by them as of yet. A process which malloc()'s 1G 1079 of memory, but only touches 300M of it will show up as 1080 using 1G. This 1G is memory which has been "committed" to 1081 by the VM and can be used at any time by the allocating 1082 application. With strict overcommit enabled on the system 1083 (mode 2 in 'vm.overcommit_memory'), allocations which would 1084 exceed the CommitLimit (detailed above) will not be permitted. 1085 This is useful if one needs to guarantee that processes will 1086 not fail due to lack of memory once that memory has been 1087 successfully allocated. 1088VmallocTotal 1089 total size of vmalloc memory area 1090VmallocUsed 1091 amount of vmalloc area which is used 1092VmallocChunk 1093 largest contiguous block of vmalloc area which is free 1094Percpu 1095 Memory allocated to the percpu allocator used to back percpu 1096 allocations. This stat excludes the cost of metadata. 1097 1098vmallocinfo 1099~~~~~~~~~~~ 1100 1101Provides information about vmalloced/vmaped areas. One line per area, 1102containing the virtual address range of the area, size in bytes, 1103caller information of the creator, and optional information depending 1104on the kind of area: 1105 1106 ========== =================================================== 1107 pages=nr number of pages 1108 phys=addr if a physical address was specified 1109 ioremap I/O mapping (ioremap() and friends) 1110 vmalloc vmalloc() area 1111 vmap vmap()ed pages 1112 user VM_USERMAP area 1113 vpages buffer for pages pointers was vmalloced (huge area) 1114 N<node>=nr (Only on NUMA kernels) 1115 Number of pages allocated on memory node <node> 1116 ========== =================================================== 1117 1118:: 1119 1120 > cat /proc/vmallocinfo 1121 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1122 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1123 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1124 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1125 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1126 phys=7fee8000 ioremap 1127 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1128 phys=7fee7000 ioremap 1129 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1130 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1131 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1132 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1133 pages=2 vmalloc N1=2 1134 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1135 /0x130 [x_tables] pages=4 vmalloc N0=4 1136 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1137 pages=14 vmalloc N2=14 1138 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1139 pages=4 vmalloc N1=4 1140 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1141 pages=2 vmalloc N1=2 1142 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1143 pages=10 vmalloc N0=10 1144 1145 1146softirqs 1147~~~~~~~~ 1148 1149Provides counts of softirq handlers serviced since boot time, for each CPU. 1150 1151:: 1152 1153 > cat /proc/softirqs 1154 CPU0 CPU1 CPU2 CPU3 1155 HI: 0 0 0 0 1156 TIMER: 27166 27120 27097 27034 1157 NET_TX: 0 0 0 17 1158 NET_RX: 42 0 0 39 1159 BLOCK: 0 0 107 1121 1160 TASKLET: 0 0 0 290 1161 SCHED: 27035 26983 26971 26746 1162 HRTIMER: 0 0 0 0 1163 RCU: 1678 1769 2178 2250 1164 1165 11661.3 IDE devices in /proc/ide 1167---------------------------- 1168 1169The subdirectory /proc/ide contains information about all IDE devices of which 1170the kernel is aware. There is one subdirectory for each IDE controller, the 1171file drivers and a link for each IDE device, pointing to the device directory 1172in the controller specific subtree. 1173 1174The file 'drivers' contains general information about the drivers used for the 1175IDE devices:: 1176 1177 > cat /proc/ide/drivers 1178 ide-cdrom version 4.53 1179 ide-disk version 1.08 1180 1181More detailed information can be found in the controller specific 1182subdirectories. These are named ide0, ide1 and so on. Each of these 1183directories contains the files shown in table 1-6. 1184 1185 1186.. table:: Table 1-6: IDE controller info in /proc/ide/ide? 1187 1188 ======= ======================================= 1189 File Content 1190 ======= ======================================= 1191 channel IDE channel (0 or 1) 1192 config Configuration (only for PCI/IDE bridge) 1193 mate Mate name 1194 model Type/Chipset of IDE controller 1195 ======= ======================================= 1196 1197Each device connected to a controller has a separate subdirectory in the 1198controllers directory. The files listed in table 1-7 are contained in these 1199directories. 1200 1201 1202.. table:: Table 1-7: IDE device information 1203 1204 ================ ========================================== 1205 File Content 1206 ================ ========================================== 1207 cache The cache 1208 capacity Capacity of the medium (in 512Byte blocks) 1209 driver driver and version 1210 geometry physical and logical geometry 1211 identify device identify block 1212 media media type 1213 model device identifier 1214 settings device setup 1215 smart_thresholds IDE disk management thresholds 1216 smart_values IDE disk management values 1217 ================ ========================================== 1218 1219The most interesting file is ``settings``. This file contains a nice 1220overview of the drive parameters:: 1221 1222 # cat /proc/ide/ide0/hda/settings 1223 name value min max mode 1224 ---- ----- --- --- ---- 1225 bios_cyl 526 0 65535 rw 1226 bios_head 255 0 255 rw 1227 bios_sect 63 0 63 rw 1228 breada_readahead 4 0 127 rw 1229 bswap 0 0 1 r 1230 file_readahead 72 0 2097151 rw 1231 io_32bit 0 0 3 rw 1232 keepsettings 0 0 1 rw 1233 max_kb_per_request 122 1 127 rw 1234 multcount 0 0 8 rw 1235 nice1 1 0 1 rw 1236 nowerr 0 0 1 rw 1237 pio_mode write-only 0 255 w 1238 slow 0 0 1 rw 1239 unmaskirq 0 0 1 rw 1240 using_dma 0 0 1 rw 1241 1242 12431.4 Networking info in /proc/net 1244-------------------------------- 1245 1246The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1247additional values you get for IP version 6 if you configure the kernel to 1248support this. Table 1-9 lists the files and their meaning. 1249 1250 1251.. table:: Table 1-8: IPv6 info in /proc/net 1252 1253 ========== ===================================================== 1254 File Content 1255 ========== ===================================================== 1256 udp6 UDP sockets (IPv6) 1257 tcp6 TCP sockets (IPv6) 1258 raw6 Raw device statistics (IPv6) 1259 igmp6 IP multicast addresses, which this host joined (IPv6) 1260 if_inet6 List of IPv6 interface addresses 1261 ipv6_route Kernel routing table for IPv6 1262 rt6_stats Global IPv6 routing tables statistics 1263 sockstat6 Socket statistics (IPv6) 1264 snmp6 Snmp data (IPv6) 1265 ========== ===================================================== 1266 1267.. table:: Table 1-9: Network info in /proc/net 1268 1269 ============= ================================================================ 1270 File Content 1271 ============= ================================================================ 1272 arp Kernel ARP table 1273 dev network devices with statistics 1274 dev_mcast the Layer2 multicast groups a device is listening too 1275 (interface index, label, number of references, number of bound 1276 addresses). 1277 dev_stat network device status 1278 ip_fwchains Firewall chain linkage 1279 ip_fwnames Firewall chain names 1280 ip_masq Directory containing the masquerading tables 1281 ip_masquerade Major masquerading table 1282 netstat Network statistics 1283 raw raw device statistics 1284 route Kernel routing table 1285 rpc Directory containing rpc info 1286 rt_cache Routing cache 1287 snmp SNMP data 1288 sockstat Socket statistics 1289 tcp TCP sockets 1290 udp UDP sockets 1291 unix UNIX domain sockets 1292 wireless Wireless interface data (Wavelan etc) 1293 igmp IP multicast addresses, which this host joined 1294 psched Global packet scheduler parameters. 1295 netlink List of PF_NETLINK sockets 1296 ip_mr_vifs List of multicast virtual interfaces 1297 ip_mr_cache List of multicast routing cache 1298 ============= ================================================================ 1299 1300You can use this information to see which network devices are available in 1301your system and how much traffic was routed over those devices:: 1302 1303 > cat /proc/net/dev 1304 Inter-|Receive |[... 1305 face |bytes packets errs drop fifo frame compressed multicast|[... 1306 lo: 908188 5596 0 0 0 0 0 0 [... 1307 ppp0:15475140 20721 410 0 0 410 0 0 [... 1308 eth0: 614530 7085 0 0 0 0 0 1 [... 1309 1310 ...] Transmit 1311 ...] bytes packets errs drop fifo colls carrier compressed 1312 ...] 908188 5596 0 0 0 0 0 0 1313 ...] 1375103 17405 0 0 0 0 0 0 1314 ...] 1703981 5535 0 0 0 3 0 0 1315 1316In addition, each Channel Bond interface has its own directory. For 1317example, the bond0 device will have a directory called /proc/net/bond0/. 1318It will contain information that is specific to that bond, such as the 1319current slaves of the bond, the link status of the slaves, and how 1320many times the slaves link has failed. 1321 13221.5 SCSI info 1323------------- 1324 1325If you have a SCSI host adapter in your system, you'll find a subdirectory 1326named after the driver for this adapter in /proc/scsi. You'll also see a list 1327of all recognized SCSI devices in /proc/scsi:: 1328 1329 >cat /proc/scsi/scsi 1330 Attached devices: 1331 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1332 Vendor: IBM Model: DGHS09U Rev: 03E0 1333 Type: Direct-Access ANSI SCSI revision: 03 1334 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1335 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1336 Type: CD-ROM ANSI SCSI revision: 02 1337 1338 1339The directory named after the driver has one file for each adapter found in 1340the system. These files contain information about the controller, including 1341the used IRQ and the IO address range. The amount of information shown is 1342dependent on the adapter you use. The example shows the output for an Adaptec 1343AHA-2940 SCSI adapter:: 1344 1345 > cat /proc/scsi/aic7xxx/0 1346 1347 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1348 Compile Options: 1349 TCQ Enabled By Default : Disabled 1350 AIC7XXX_PROC_STATS : Disabled 1351 AIC7XXX_RESET_DELAY : 5 1352 Adapter Configuration: 1353 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1354 Ultra Wide Controller 1355 PCI MMAPed I/O Base: 0xeb001000 1356 Adapter SEEPROM Config: SEEPROM found and used. 1357 Adaptec SCSI BIOS: Enabled 1358 IRQ: 10 1359 SCBs: Active 0, Max Active 2, 1360 Allocated 15, HW 16, Page 255 1361 Interrupts: 160328 1362 BIOS Control Word: 0x18b6 1363 Adapter Control Word: 0x005b 1364 Extended Translation: Enabled 1365 Disconnect Enable Flags: 0xffff 1366 Ultra Enable Flags: 0x0001 1367 Tag Queue Enable Flags: 0x0000 1368 Ordered Queue Tag Flags: 0x0000 1369 Default Tag Queue Depth: 8 1370 Tagged Queue By Device array for aic7xxx host instance 0: 1371 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1372 Actual queue depth per device for aic7xxx host instance 0: 1373 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1374 Statistics: 1375 (scsi0:0:0:0) 1376 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1377 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1378 Total transfers 160151 (74577 reads and 85574 writes) 1379 (scsi0:0:6:0) 1380 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1381 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1382 Total transfers 0 (0 reads and 0 writes) 1383 1384 13851.6 Parallel port info in /proc/parport 1386--------------------------------------- 1387 1388The directory /proc/parport contains information about the parallel ports of 1389your system. It has one subdirectory for each port, named after the port 1390number (0,1,2,...). 1391 1392These directories contain the four files shown in Table 1-10. 1393 1394 1395.. table:: Table 1-10: Files in /proc/parport 1396 1397 ========= ==================================================================== 1398 File Content 1399 ========= ==================================================================== 1400 autoprobe Any IEEE-1284 device ID information that has been acquired. 1401 devices list of the device drivers using that port. A + will appear by the 1402 name of the device currently using the port (it might not appear 1403 against any). 1404 hardware Parallel port's base address, IRQ line and DMA channel. 1405 irq IRQ that parport is using for that port. This is in a separate 1406 file to allow you to alter it by writing a new value in (IRQ 1407 number or none). 1408 ========= ==================================================================== 1409 14101.7 TTY info in /proc/tty 1411------------------------- 1412 1413Information about the available and actually used tty's can be found in the 1414directory /proc/tty. You'll find entries for drivers and line disciplines in 1415this directory, as shown in Table 1-11. 1416 1417 1418.. table:: Table 1-11: Files in /proc/tty 1419 1420 ============= ============================================== 1421 File Content 1422 ============= ============================================== 1423 drivers list of drivers and their usage 1424 ldiscs registered line disciplines 1425 driver/serial usage statistic and status of single tty lines 1426 ============= ============================================== 1427 1428To see which tty's are currently in use, you can simply look into the file 1429/proc/tty/drivers:: 1430 1431 > cat /proc/tty/drivers 1432 pty_slave /dev/pts 136 0-255 pty:slave 1433 pty_master /dev/ptm 128 0-255 pty:master 1434 pty_slave /dev/ttyp 3 0-255 pty:slave 1435 pty_master /dev/pty 2 0-255 pty:master 1436 serial /dev/cua 5 64-67 serial:callout 1437 serial /dev/ttyS 4 64-67 serial 1438 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1439 /dev/ptmx /dev/ptmx 5 2 system 1440 /dev/console /dev/console 5 1 system:console 1441 /dev/tty /dev/tty 5 0 system:/dev/tty 1442 unknown /dev/tty 4 1-63 console 1443 1444 14451.8 Miscellaneous kernel statistics in /proc/stat 1446------------------------------------------------- 1447 1448Various pieces of information about kernel activity are available in the 1449/proc/stat file. All of the numbers reported in this file are aggregates 1450since the system first booted. For a quick look, simply cat the file:: 1451 1452 > cat /proc/stat 1453 cpu 2255 34 2290 22625563 6290 127 456 0 0 0 1454 cpu0 1132 34 1441 11311718 3675 127 438 0 0 0 1455 cpu1 1123 0 849 11313845 2614 0 18 0 0 0 1456 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...] 1457 ctxt 1990473 1458 btime 1062191376 1459 processes 2915 1460 procs_running 1 1461 procs_blocked 0 1462 softirq 183433 0 21755 12 39 1137 231 21459 2263 1463 1464The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1465lines. These numbers identify the amount of time the CPU has spent performing 1466different kinds of work. Time units are in USER_HZ (typically hundredths of a 1467second). The meanings of the columns are as follows, from left to right: 1468 1469- user: normal processes executing in user mode 1470- nice: niced processes executing in user mode 1471- system: processes executing in kernel mode 1472- idle: twiddling thumbs 1473- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1474 are several problems: 1475 1476 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1477 waiting for I/O to complete. When CPU goes into idle state for 1478 outstanding task I/O, another task will be scheduled on this CPU. 1479 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1480 on any CPU, so the iowait of each CPU is difficult to calculate. 1481 3. The value of iowait field in /proc/stat will decrease in certain 1482 conditions. 1483 1484 So, the iowait is not reliable by reading from /proc/stat. 1485- irq: servicing interrupts 1486- softirq: servicing softirqs 1487- steal: involuntary wait 1488- guest: running a normal guest 1489- guest_nice: running a niced guest 1490 1491The "intr" line gives counts of interrupts serviced since boot time, for each 1492of the possible system interrupts. The first column is the total of all 1493interrupts serviced including unnumbered architecture specific interrupts; 1494each subsequent column is the total for that particular numbered interrupt. 1495Unnumbered interrupts are not shown, only summed into the total. 1496 1497The "ctxt" line gives the total number of context switches across all CPUs. 1498 1499The "btime" line gives the time at which the system booted, in seconds since 1500the Unix epoch. 1501 1502The "processes" line gives the number of processes and threads created, which 1503includes (but is not limited to) those created by calls to the fork() and 1504clone() system calls. 1505 1506The "procs_running" line gives the total number of threads that are 1507running or ready to run (i.e., the total number of runnable threads). 1508 1509The "procs_blocked" line gives the number of processes currently blocked, 1510waiting for I/O to complete. 1511 1512The "softirq" line gives counts of softirqs serviced since boot time, for each 1513of the possible system softirqs. The first column is the total of all 1514softirqs serviced; each subsequent column is the total for that particular 1515softirq. 1516 1517 15181.9 Ext4 file system parameters 1519------------------------------- 1520 1521Information about mounted ext4 file systems can be found in 1522/proc/fs/ext4. Each mounted filesystem will have a directory in 1523/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1524/proc/fs/ext4/dm-0). The files in each per-device directory are shown 1525in Table 1-12, below. 1526 1527.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1528 1529 ============== ========================================================== 1530 File Content 1531 mb_groups details of multiblock allocator buddy cache of free blocks 1532 ============== ========================================================== 1533 15341.10 /proc/consoles 1535------------------- 1536Shows registered system console lines. 1537 1538To see which character device lines are currently used for the system console 1539/dev/console, you may simply look into the file /proc/consoles:: 1540 1541 > cat /proc/consoles 1542 tty0 -WU (ECp) 4:7 1543 ttyS0 -W- (Ep) 4:64 1544 1545The columns are: 1546 1547+--------------------+-------------------------------------------------------+ 1548| device | name of the device | 1549+====================+=======================================================+ 1550| operations | * R = can do read operations | 1551| | * W = can do write operations | 1552| | * U = can do unblank | 1553+--------------------+-------------------------------------------------------+ 1554| flags | * E = it is enabled | 1555| | * C = it is preferred console | 1556| | * B = it is primary boot console | 1557| | * p = it is used for printk buffer | 1558| | * b = it is not a TTY but a Braille device | 1559| | * a = it is safe to use when cpu is offline | 1560+--------------------+-------------------------------------------------------+ 1561| major:minor | major and minor number of the device separated by a | 1562| | colon | 1563+--------------------+-------------------------------------------------------+ 1564 1565Summary 1566------- 1567 1568The /proc file system serves information about the running system. It not only 1569allows access to process data but also allows you to request the kernel status 1570by reading files in the hierarchy. 1571 1572The directory structure of /proc reflects the types of information and makes 1573it easy, if not obvious, where to look for specific data. 1574 1575Chapter 2: Modifying System Parameters 1576====================================== 1577 1578In This Chapter 1579--------------- 1580 1581* Modifying kernel parameters by writing into files found in /proc/sys 1582* Exploring the files which modify certain parameters 1583* Review of the /proc/sys file tree 1584 1585------------------------------------------------------------------------------ 1586 1587A very interesting part of /proc is the directory /proc/sys. This is not only 1588a source of information, it also allows you to change parameters within the 1589kernel. Be very careful when attempting this. You can optimize your system, 1590but you can also cause it to crash. Never alter kernel parameters on a 1591production system. Set up a development machine and test to make sure that 1592everything works the way you want it to. You may have no alternative but to 1593reboot the machine once an error has been made. 1594 1595To change a value, simply echo the new value into the file. 1596You need to be root to do this. You can create your own boot script 1597to perform this every time your system boots. 1598 1599The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1600general things in the operation of the Linux kernel. Since some of the files 1601can inadvertently disrupt your system, it is advisable to read both 1602documentation and source before actually making adjustments. In any case, be 1603very careful when writing to any of these files. The entries in /proc may 1604change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1605review the kernel documentation in the directory /usr/src/linux/Documentation. 1606This chapter is heavily based on the documentation included in the pre 2.2 1607kernels, and became part of it in version 2.2.1 of the Linux kernel. 1608 1609Please see: Documentation/admin-guide/sysctl/ directory for descriptions of these 1610entries. 1611 1612Summary 1613------- 1614 1615Certain aspects of kernel behavior can be modified at runtime, without the 1616need to recompile the kernel, or even to reboot the system. The files in the 1617/proc/sys tree can not only be read, but also modified. You can use the echo 1618command to write value into these files, thereby changing the default settings 1619of the kernel. 1620 1621 1622Chapter 3: Per-process Parameters 1623================================= 1624 16253.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1626-------------------------------------------------------------------------------- 1627 1628These files can be used to adjust the badness heuristic used to select which 1629process gets killed in out of memory (oom) conditions. 1630 1631The badness heuristic assigns a value to each candidate task ranging from 0 1632(never kill) to 1000 (always kill) to determine which process is targeted. The 1633units are roughly a proportion along that range of allowed memory the process 1634may allocate from based on an estimation of its current memory and swap use. 1635For example, if a task is using all allowed memory, its badness score will be 16361000. If it is using half of its allowed memory, its score will be 500. 1637 1638The amount of "allowed" memory depends on the context in which the oom killer 1639was called. If it is due to the memory assigned to the allocating task's cpuset 1640being exhausted, the allowed memory represents the set of mems assigned to that 1641cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1642memory represents the set of mempolicy nodes. If it is due to a memory 1643limit (or swap limit) being reached, the allowed memory is that configured 1644limit. Finally, if it is due to the entire system being out of memory, the 1645allowed memory represents all allocatable resources. 1646 1647The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1648is used to determine which task to kill. Acceptable values range from -1000 1649(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1650polarize the preference for oom killing either by always preferring a certain 1651task or completely disabling it. The lowest possible value, -1000, is 1652equivalent to disabling oom killing entirely for that task since it will always 1653report a badness score of 0. 1654 1655Consequently, it is very simple for userspace to define the amount of memory to 1656consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1657example, is roughly equivalent to allowing the remainder of tasks sharing the 1658same system, cpuset, mempolicy, or memory controller resources to use at least 165950% more memory. A value of -500, on the other hand, would be roughly 1660equivalent to discounting 50% of the task's allowed memory from being considered 1661as scoring against the task. 1662 1663For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1664be used to tune the badness score. Its acceptable values range from -16 1665(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1666(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1667scaled linearly with /proc/<pid>/oom_score_adj. 1668 1669The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1670value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1671requires CAP_SYS_RESOURCE. 1672 1673 16743.2 /proc/<pid>/oom_score - Display current oom-killer score 1675------------------------------------------------------------- 1676 1677This file can be used to check the current score used by the oom-killer for 1678any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1679process should be killed in an out-of-memory situation. 1680 1681Please note that the exported value includes oom_score_adj so it is 1682effectively in range [0,2000]. 1683 1684 16853.3 /proc/<pid>/io - Display the IO accounting fields 1686------------------------------------------------------- 1687 1688This file contains IO statistics for each running process. 1689 1690Example 1691~~~~~~~ 1692 1693:: 1694 1695 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1696 [1] 3828 1697 1698 test:/tmp # cat /proc/3828/io 1699 rchar: 323934931 1700 wchar: 323929600 1701 syscr: 632687 1702 syscw: 632675 1703 read_bytes: 0 1704 write_bytes: 323932160 1705 cancelled_write_bytes: 0 1706 1707 1708Description 1709~~~~~~~~~~~ 1710 1711rchar 1712^^^^^ 1713 1714I/O counter: chars read 1715The number of bytes which this task has caused to be read from storage. This 1716is simply the sum of bytes which this process passed to read() and pread(). 1717It includes things like tty IO and it is unaffected by whether or not actual 1718physical disk IO was required (the read might have been satisfied from 1719pagecache). 1720 1721 1722wchar 1723^^^^^ 1724 1725I/O counter: chars written 1726The number of bytes which this task has caused, or shall cause to be written 1727to disk. Similar caveats apply here as with rchar. 1728 1729 1730syscr 1731^^^^^ 1732 1733I/O counter: read syscalls 1734Attempt to count the number of read I/O operations, i.e. syscalls like read() 1735and pread(). 1736 1737 1738syscw 1739^^^^^ 1740 1741I/O counter: write syscalls 1742Attempt to count the number of write I/O operations, i.e. syscalls like 1743write() and pwrite(). 1744 1745 1746read_bytes 1747^^^^^^^^^^ 1748 1749I/O counter: bytes read 1750Attempt to count the number of bytes which this process really did cause to 1751be fetched from the storage layer. Done at the submit_bio() level, so it is 1752accurate for block-backed filesystems. <please add status regarding NFS and 1753CIFS at a later time> 1754 1755 1756write_bytes 1757^^^^^^^^^^^ 1758 1759I/O counter: bytes written 1760Attempt to count the number of bytes which this process caused to be sent to 1761the storage layer. This is done at page-dirtying time. 1762 1763 1764cancelled_write_bytes 1765^^^^^^^^^^^^^^^^^^^^^ 1766 1767The big inaccuracy here is truncate. If a process writes 1MB to a file and 1768then deletes the file, it will in fact perform no writeout. But it will have 1769been accounted as having caused 1MB of write. 1770In other words: The number of bytes which this process caused to not happen, 1771by truncating pagecache. A task can cause "negative" IO too. If this task 1772truncates some dirty pagecache, some IO which another task has been accounted 1773for (in its write_bytes) will not be happening. We _could_ just subtract that 1774from the truncating task's write_bytes, but there is information loss in doing 1775that. 1776 1777 1778.. Note:: 1779 1780 At its current implementation state, this is a bit racy on 32-bit machines: 1781 if process A reads process B's /proc/pid/io while process B is updating one 1782 of those 64-bit counters, process A could see an intermediate result. 1783 1784 1785More information about this can be found within the taskstats documentation in 1786Documentation/accounting. 1787 17883.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1789--------------------------------------------------------------- 1790When a process is dumped, all anonymous memory is written to a core file as 1791long as the size of the core file isn't limited. But sometimes we don't want 1792to dump some memory segments, for example, huge shared memory or DAX. 1793Conversely, sometimes we want to save file-backed memory segments into a core 1794file, not only the individual files. 1795 1796/proc/<pid>/coredump_filter allows you to customize which memory segments 1797will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1798of memory types. If a bit of the bitmask is set, memory segments of the 1799corresponding memory type are dumped, otherwise they are not dumped. 1800 1801The following 9 memory types are supported: 1802 1803 - (bit 0) anonymous private memory 1804 - (bit 1) anonymous shared memory 1805 - (bit 2) file-backed private memory 1806 - (bit 3) file-backed shared memory 1807 - (bit 4) ELF header pages in file-backed private memory areas (it is 1808 effective only if the bit 2 is cleared) 1809 - (bit 5) hugetlb private memory 1810 - (bit 6) hugetlb shared memory 1811 - (bit 7) DAX private memory 1812 - (bit 8) DAX shared memory 1813 1814 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1815 are always dumped regardless of the bitmask status. 1816 1817 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1818 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1819 1820The default value of coredump_filter is 0x33; this means all anonymous memory 1821segments, ELF header pages and hugetlb private memory are dumped. 1822 1823If you don't want to dump all shared memory segments attached to pid 1234, 1824write 0x31 to the process's proc file:: 1825 1826 $ echo 0x31 > /proc/1234/coredump_filter 1827 1828When a new process is created, the process inherits the bitmask status from its 1829parent. It is useful to set up coredump_filter before the program runs. 1830For example:: 1831 1832 $ echo 0x7 > /proc/self/coredump_filter 1833 $ ./some_program 1834 18353.5 /proc/<pid>/mountinfo - Information about mounts 1836-------------------------------------------------------- 1837 1838This file contains lines of the form:: 1839 1840 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1841 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11) 1842 1843 (1) mount ID: unique identifier of the mount (may be reused after umount) 1844 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1845 (3) major:minor: value of st_dev for files on filesystem 1846 (4) root: root of the mount within the filesystem 1847 (5) mount point: mount point relative to the process's root 1848 (6) mount options: per mount options 1849 (7) optional fields: zero or more fields of the form "tag[:value]" 1850 (8) separator: marks the end of the optional fields 1851 (9) filesystem type: name of filesystem of the form "type[.subtype]" 1852 (10) mount source: filesystem specific information or "none" 1853 (11) super options: per super block options 1854 1855Parsers should ignore all unrecognised optional fields. Currently the 1856possible optional fields are: 1857 1858================ ============================================================== 1859shared:X mount is shared in peer group X 1860master:X mount is slave to peer group X 1861propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1862unbindable mount is unbindable 1863================ ============================================================== 1864 1865.. [#] X is the closest dominant peer group under the process's root. If 1866 X is the immediate master of the mount, or if there's no dominant peer 1867 group under the same root, then only the "master:X" field is present 1868 and not the "propagate_from:X" field. 1869 1870For more information on mount propagation see: 1871 1872 Documentation/filesystems/sharedsubtree.rst 1873 1874 18753.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1876-------------------------------------------------------- 1877These files provide a method to access a task's comm value. It also allows for 1878a task to set its own or one of its thread siblings comm value. The comm value 1879is limited in size compared to the cmdline value, so writing anything longer 1880then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated 1881comm value. 1882 1883 18843.7 /proc/<pid>/task/<tid>/children - Information about task children 1885------------------------------------------------------------------------- 1886This file provides a fast way to retrieve first level children pids 1887of a task pointed by <pid>/<tid> pair. The format is a space separated 1888stream of pids. 1889 1890Note the "first level" here -- if a child has its own children they will 1891not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1892to obtain the descendants. 1893 1894Since this interface is intended to be fast and cheap it doesn't 1895guarantee to provide precise results and some children might be 1896skipped, especially if they've exited right after we printed their 1897pids, so one needs to either stop or freeze processes being inspected 1898if precise results are needed. 1899 1900 19013.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1902--------------------------------------------------------------- 1903This file provides information associated with an opened file. The regular 1904files have at least three fields -- 'pos', 'flags' and 'mnt_id'. The 'pos' 1905represents the current offset of the opened file in decimal form [see lseek(2) 1906for details], 'flags' denotes the octal O_xxx mask the file has been 1907created with [see open(2) for details] and 'mnt_id' represents mount ID of 1908the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo 1909for details]. 1910 1911A typical output is:: 1912 1913 pos: 0 1914 flags: 0100002 1915 mnt_id: 19 1916 1917All locks associated with a file descriptor are shown in its fdinfo too:: 1918 1919 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1920 1921The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1922pair provide additional information particular to the objects they represent. 1923 1924Eventfd files 1925~~~~~~~~~~~~~ 1926 1927:: 1928 1929 pos: 0 1930 flags: 04002 1931 mnt_id: 9 1932 eventfd-count: 5a 1933 1934where 'eventfd-count' is hex value of a counter. 1935 1936Signalfd files 1937~~~~~~~~~~~~~~ 1938 1939:: 1940 1941 pos: 0 1942 flags: 04002 1943 mnt_id: 9 1944 sigmask: 0000000000000200 1945 1946where 'sigmask' is hex value of the signal mask associated 1947with a file. 1948 1949Epoll files 1950~~~~~~~~~~~ 1951 1952:: 1953 1954 pos: 0 1955 flags: 02 1956 mnt_id: 9 1957 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 1958 1959where 'tfd' is a target file descriptor number in decimal form, 1960'events' is events mask being watched and the 'data' is data 1961associated with a target [see epoll(7) for more details]. 1962 1963The 'pos' is current offset of the target file in decimal form 1964[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 1965where target file resides, all in hex format. 1966 1967Fsnotify files 1968~~~~~~~~~~~~~~ 1969For inotify files the format is the following:: 1970 1971 pos: 0 1972 flags: 02000000 1973 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 1974 1975where 'wd' is a watch descriptor in decimal form, i.e. a target file 1976descriptor number, 'ino' and 'sdev' are inode and device where the 1977target file resides and the 'mask' is the mask of events, all in hex 1978form [see inotify(7) for more details]. 1979 1980If the kernel was built with exportfs support, the path to the target 1981file is encoded as a file handle. The file handle is provided by three 1982fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 1983format. 1984 1985If the kernel is built without exportfs support the file handle won't be 1986printed out. 1987 1988If there is no inotify mark attached yet the 'inotify' line will be omitted. 1989 1990For fanotify files the format is:: 1991 1992 pos: 0 1993 flags: 02 1994 mnt_id: 9 1995 fanotify flags:10 event-flags:0 1996 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 1997 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 1998 1999where fanotify 'flags' and 'event-flags' are values used in fanotify_init 2000call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 2001flags associated with mark which are tracked separately from events 2002mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 2003mask and 'ignored_mask' is the mask of events which are to be ignored. 2004All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2005provide information about flags and mask used in fanotify_mark 2006call [see fsnotify manpage for details]. 2007 2008While the first three lines are mandatory and always printed, the rest is 2009optional and may be omitted if no marks created yet. 2010 2011Timerfd files 2012~~~~~~~~~~~~~ 2013 2014:: 2015 2016 pos: 0 2017 flags: 02 2018 mnt_id: 9 2019 clockid: 0 2020 ticks: 0 2021 settime flags: 01 2022 it_value: (0, 49406829) 2023 it_interval: (1, 0) 2024 2025where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2026that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2027flags in octal form been used to setup the timer [see timerfd_settime(2) for 2028details]. 'it_value' is remaining time until the timer expiration. 2029'it_interval' is the interval for the timer. Note the timer might be set up 2030with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2031still exhibits timer's remaining time. 2032 20333.9 /proc/<pid>/map_files - Information about memory mapped files 2034--------------------------------------------------------------------- 2035This directory contains symbolic links which represent memory mapped files 2036the process is maintaining. Example output:: 2037 2038 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2039 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2040 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2041 | ... 2042 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2043 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2044 2045The name of a link represents the virtual memory bounds of a mapping, i.e. 2046vm_area_struct::vm_start-vm_area_struct::vm_end. 2047 2048The main purpose of the map_files is to retrieve a set of memory mapped 2049files in a fast way instead of parsing /proc/<pid>/maps or 2050/proc/<pid>/smaps, both of which contain many more records. At the same 2051time one can open(2) mappings from the listings of two processes and 2052comparing their inode numbers to figure out which anonymous memory areas 2053are actually shared. 2054 20553.10 /proc/<pid>/timerslack_ns - Task timerslack value 2056--------------------------------------------------------- 2057This file provides the value of the task's timerslack value in nanoseconds. 2058This value specifies an amount of time that normal timers may be deferred 2059in order to coalesce timers and avoid unnecessary wakeups. 2060 2061This allows a task's interactivity vs power consumption tradeoff to be 2062adjusted. 2063 2064Writing 0 to the file will set the task's timerslack to the default value. 2065 2066Valid values are from 0 - ULLONG_MAX 2067 2068An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2069permissions on the task specified to change its timerslack_ns value. 2070 20713.11 /proc/<pid>/patch_state - Livepatch patch operation state 2072----------------------------------------------------------------- 2073When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2074patch state for the task. 2075 2076A value of '-1' indicates that no patch is in transition. 2077 2078A value of '0' indicates that a patch is in transition and the task is 2079unpatched. If the patch is being enabled, then the task hasn't been 2080patched yet. If the patch is being disabled, then the task has already 2081been unpatched. 2082 2083A value of '1' indicates that a patch is in transition and the task is 2084patched. If the patch is being enabled, then the task has already been 2085patched. If the patch is being disabled, then the task hasn't been 2086unpatched yet. 2087 20883.12 /proc/<pid>/arch_status - task architecture specific status 2089------------------------------------------------------------------- 2090When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2091architecture specific status of the task. 2092 2093Example 2094~~~~~~~ 2095 2096:: 2097 2098 $ cat /proc/6753/arch_status 2099 AVX512_elapsed_ms: 8 2100 2101Description 2102~~~~~~~~~~~ 2103 2104x86 specific entries 2105~~~~~~~~~~~~~~~~~~~~~ 2106 2107AVX512_elapsed_ms 2108^^^^^^^^^^^^^^^^^^ 2109 2110 If AVX512 is supported on the machine, this entry shows the milliseconds 2111 elapsed since the last time AVX512 usage was recorded. The recording 2112 happens on a best effort basis when a task is scheduled out. This means 2113 that the value depends on two factors: 2114 2115 1) The time which the task spent on the CPU without being scheduled 2116 out. With CPU isolation and a single runnable task this can take 2117 several seconds. 2118 2119 2) The time since the task was scheduled out last. Depending on the 2120 reason for being scheduled out (time slice exhausted, syscall ...) 2121 this can be arbitrary long time. 2122 2123 As a consequence the value cannot be considered precise and authoritative 2124 information. The application which uses this information has to be aware 2125 of the overall scenario on the system in order to determine whether a 2126 task is a real AVX512 user or not. Precise information can be obtained 2127 with performance counters. 2128 2129 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2130 the task is unlikely an AVX512 user, but depends on the workload and the 2131 scheduling scenario, it also could be a false negative mentioned above. 2132 2133Chapter 4: Configuring procfs 2134============================= 2135 21364.1 Mount options 2137--------------------- 2138 2139The following mount options are supported: 2140 2141 ========= ======================================================== 2142 hidepid= Set /proc/<pid>/ access mode. 2143 gid= Set the group authorized to learn processes information. 2144 subset= Show only the specified subset of procfs. 2145 ========= ======================================================== 2146 2147hidepid=off or hidepid=0 means classic mode - everybody may access all 2148/proc/<pid>/ directories (default). 2149 2150hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2151directories but their own. Sensitive files like cmdline, sched*, status are now 2152protected against other users. This makes it impossible to learn whether any 2153user runs specific program (given the program doesn't reveal itself by its 2154behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2155other users, poorly written programs passing sensitive information via program 2156arguments are now protected against local eavesdroppers. 2157 2158hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2159fully invisible to other users. It doesn't mean that it hides a fact whether a 2160process with a specific pid value exists (it can be learned by other means, e.g. 2161by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by 2162stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2163gathering information about running processes, whether some daemon runs with 2164elevated privileges, whether other user runs some sensitive program, whether 2165other users run any program at all, etc. 2166 2167hidepid=ptraceable or hidepid=4 means that procfs should only contain 2168/proc/<pid>/ directories that the caller can ptrace. 2169 2170gid= defines a group authorized to learn processes information otherwise 2171prohibited by hidepid=. If you use some daemon like identd which needs to learn 2172information about processes information, just add identd to this group. 2173 2174subset=pid hides all top level files and directories in the procfs that 2175are not related to tasks. 2176 2177Chapter 5: Filesystem behavior 2178============================== 2179 2180Originally, before the advent of pid namepsace, procfs was a global file 2181system. It means that there was only one procfs instance in the system. 2182 2183When pid namespace was added, a separate procfs instance was mounted in 2184each pid namespace. So, procfs mount options are global among all 2185mountpoints within the same namespace:: 2186 2187 # grep ^proc /proc/mounts 2188 proc /proc proc rw,relatime,hidepid=2 0 0 2189 2190 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2191 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2192 +++ exited with 0 +++ 2193 2194 # grep ^proc /proc/mounts 2195 proc /proc proc rw,relatime,hidepid=2 0 0 2196 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2197 2198and only after remounting procfs mount options will change at all 2199mountpoints:: 2200 2201 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2202 2203 # grep ^proc /proc/mounts 2204 proc /proc proc rw,relatime,hidepid=1 0 0 2205 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2206 2207This behavior is different from the behavior of other filesystems. 2208 2209The new procfs behavior is more like other filesystems. Each procfs mount 2210creates a new procfs instance. Mount options affect own procfs instance. 2211It means that it became possible to have several procfs instances 2212displaying tasks with different filtering options in one pid namespace:: 2213 2214 # mount -o hidepid=invisible -t proc proc /proc 2215 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2216 # grep ^proc /proc/mounts 2217 proc /proc proc rw,relatime,hidepid=invisible 0 0 2218 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2219