Commit | Line | Data |
---|---|---|
c33e97ef MCC |
1 | .. SPDX-License-Identifier: GPL-2.0 |
2 | ||
3 | ==================== | |
4 | The /proc Filesystem | |
5 | ==================== | |
6 | ||
7 | ===================== ======================================= ================ | |
8 | /proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 | |
9 | Bodo Bauer <bb@ricochet.net> | |
10 | 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 | |
11 | move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 | |
12 | fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 | |
13 | ===================== ======================================= ================ | |
14 | ||
1da177e4 | 15 | |
1da177e4 | 16 | |
c33e97ef | 17 | .. Table of Contents |
1da177e4 LT |
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 | |
ae96b348 | 32 | 1.9 Ext4 file system parameters |
1da177e4 LT |
33 | |
34 | 2 Modifying System Parameters | |
760df93e SF |
35 | |
36 | 3 Per-Process Parameters | |
fa0cbbf1 | 37 | 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer |
a63d83f4 | 38 | score |
760df93e SF |
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 | |
4614a696 | 43 | 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm |
81841161 | 44 | 3.7 /proc/<pid>/task/<tid>/children - Information about task children |
f1d8c162 | 45 | 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file |
740a5ddb | 46 | 3.9 /proc/<pid>/map_files - Information about memory mapped files |
5de23d43 | 47 | 3.10 /proc/<pid>/timerslack_ns - Task timerslack value |
7c23b330 | 48 | 3.11 /proc/<pid>/patch_state - Livepatch patch operation state |
711486fd | 49 | 3.12 /proc/<pid>/arch_status - Task architecture specific information |
f1f1f256 | 50 | 3.13 /proc/<pid>/fd - List of symlinks to open files |
760df93e | 51 | |
0499680a VK |
52 | 4 Configuring procfs |
53 | 4.1 Mount options | |
1da177e4 | 54 | |
37e7647a AG |
55 | 5 Filesystem behavior |
56 | ||
1da177e4 | 57 | Preface |
c33e97ef | 58 | ======= |
1da177e4 LT |
59 | |
60 | 0.1 Introduction/Credits | |
61 | ------------------------ | |
62 | ||
63 | This documentation is part of a soon (or so we hope) to be released book on | |
64 | the SuSE Linux distribution. As there is no complete documentation for the | |
65 | /proc file system and we've used many freely available sources to write these | |
66 | chapters, it seems only fair to give the work back to the Linux community. | |
67 | This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm | |
68 | afraid it's still far from complete, but we hope it will be useful. As far as | |
69 | we know, it is the first 'all-in-one' document about the /proc file system. It | |
70 | is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, | |
71 | SPARC, AXP, etc., features, you probably won't find what you are looking for. | |
72 | It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But | |
73 | additions and patches are welcome and will be added to this document if you | |
74 | mail them to Bodo. | |
75 | ||
76 | We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of | |
77 | other people for help compiling this documentation. We'd also like to extend a | |
78 | special thank you to Andi Kleen for documentation, which we relied on heavily | |
79 | to create this document, as well as the additional information he provided. | |
80 | Thanks to everybody else who contributed source or docs to the Linux kernel | |
81 | and helped create a great piece of software... :) | |
82 | ||
83 | If you have any comments, corrections or additions, please don't hesitate to | |
84 | contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this | |
85 | document. | |
86 | ||
87 | The latest version of this document is available online at | |
d2ea66a6 | 88 | https://www.kernel.org/doc/html/latest/filesystems/proc.html |
1da177e4 | 89 | |
0ea6e611 | 90 | If the above direction does not works for you, you could try the kernel |
1da177e4 LT |
91 | mailing list at linux-kernel@vger.kernel.org and/or try to reach me at |
92 | comandante@zaralinux.com. | |
93 | ||
94 | 0.2 Legal Stuff | |
95 | --------------- | |
96 | ||
97 | We don't guarantee the correctness of this document, and if you come to us | |
98 | complaining about how you screwed up your system because of incorrect | |
99 | documentation, we won't feel responsible... | |
100 | ||
c33e97ef MCC |
101 | Chapter 1: Collecting System Information |
102 | ======================================== | |
1da177e4 | 103 | |
1da177e4 | 104 | In This Chapter |
c33e97ef | 105 | --------------- |
1da177e4 LT |
106 | * Investigating the properties of the pseudo file system /proc and its |
107 | ability to provide information on the running Linux system | |
108 | * Examining /proc's structure | |
109 | * Uncovering various information about the kernel and the processes running | |
110 | on the system | |
1da177e4 | 111 | |
c33e97ef | 112 | ------------------------------------------------------------------------------ |
1da177e4 LT |
113 | |
114 | The proc file system acts as an interface to internal data structures in the | |
115 | kernel. It can be used to obtain information about the system and to change | |
116 | certain kernel parameters at runtime (sysctl). | |
117 | ||
118 | First, we'll take a look at the read-only parts of /proc. In Chapter 2, we | |
119 | show you how you can use /proc/sys to change settings. | |
120 | ||
121 | 1.1 Process-Specific Subdirectories | |
122 | ----------------------------------- | |
123 | ||
124 | The directory /proc contains (among other things) one subdirectory for each | |
125 | process running on the system, which is named after the process ID (PID). | |
126 | ||
059db434 | 127 | The link 'self' points to the process reading the file system. Each process |
1da177e4 LT |
128 | subdirectory has the entries listed in Table 1-1. |
129 | ||
059db434 | 130 | Note that an open file descriptor to /proc/<pid> or to any of its |
c969eb83 DC |
131 | contained files or subdirectories does not prevent <pid> being reused |
132 | for some other process in the event that <pid> exits. Operations on | |
133 | open /proc/<pid> file descriptors corresponding to dead processes | |
134 | never act on any new process that the kernel may, through chance, have | |
135 | also assigned the process ID <pid>. Instead, operations on these FDs | |
136 | usually fail with ESRCH. | |
1da177e4 | 137 | |
c33e97ef MCC |
138 | .. table:: Table 1-1: Process specific entries in /proc |
139 | ||
140 | ============= =============================================================== | |
b813e931 | 141 | File Content |
c33e97ef | 142 | ============= =============================================================== |
b813e931 DR |
143 | clear_refs Clears page referenced bits shown in smaps output |
144 | cmdline Command line arguments | |
145 | cpu Current and last cpu in which it was executed (2.4)(smp) | |
146 | cwd Link to the current working directory | |
147 | environ Values of environment variables | |
148 | exe Link to the executable of this process | |
149 | fd Directory, which contains all file descriptors | |
150 | maps Memory maps to executables and library files (2.4) | |
151 | mem Memory held by this process | |
152 | root Link to the root directory of this process | |
153 | stat Process status | |
154 | statm Process memory status information | |
155 | status Process status in human readable form | |
b2f73922 IM |
156 | wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function |
157 | symbol the task is blocked in - or "0" if not blocked. | |
03f890f8 | 158 | pagemap Page table |
2ec220e2 | 159 | stack Report full stack trace, enable via CONFIG_STACKTRACE |
ee2ad71b | 160 | smaps An extension based on maps, showing the memory consumption of |
834f82e2 | 161 | each mapping and flags associated with it |
ee2ad71b LS |
162 | smaps_rollup Accumulated smaps stats for all mappings of the process. This |
163 | can be derived from smaps, but is faster and more convenient | |
164 | numa_maps An extension based on maps, showing the memory locality and | |
0c369711 | 165 | binding policy as well as mem usage (in pages) of each mapping. |
c33e97ef | 166 | ============= =============================================================== |
1da177e4 LT |
167 | |
168 | For example, to get the status information of a process, all you have to do is | |
c33e97ef | 169 | read the file /proc/PID/status:: |
1da177e4 | 170 | |
349888ee SS |
171 | >cat /proc/self/status |
172 | Name: cat | |
173 | State: R (running) | |
174 | Tgid: 5452 | |
175 | Pid: 5452 | |
176 | PPid: 743 | |
1da177e4 | 177 | TracerPid: 0 (2.4) |
349888ee SS |
178 | Uid: 501 501 501 501 |
179 | Gid: 100 100 100 100 | |
180 | FDSize: 256 | |
181 | Groups: 100 14 16 | |
522dc4e5 | 182 | Kthread: 0 |
349888ee SS |
183 | VmPeak: 5004 kB |
184 | VmSize: 5004 kB | |
185 | VmLck: 0 kB | |
186 | VmHWM: 476 kB | |
187 | VmRSS: 476 kB | |
8cee852e JM |
188 | RssAnon: 352 kB |
189 | RssFile: 120 kB | |
190 | RssShmem: 4 kB | |
349888ee SS |
191 | VmData: 156 kB |
192 | VmStk: 88 kB | |
193 | VmExe: 68 kB | |
194 | VmLib: 1412 kB | |
195 | VmPTE: 20 kb | |
b084d435 | 196 | VmSwap: 0 kB |
5d317b2b | 197 | HugetlbPages: 0 kB |
c6434012 | 198 | CoreDumping: 0 |
a1400af7 | 199 | THP_enabled: 1 |
349888ee SS |
200 | Threads: 1 |
201 | SigQ: 0/28578 | |
202 | SigPnd: 0000000000000000 | |
203 | ShdPnd: 0000000000000000 | |
204 | SigBlk: 0000000000000000 | |
205 | SigIgn: 0000000000000000 | |
206 | SigCgt: 0000000000000000 | |
207 | CapInh: 00000000fffffeff | |
208 | CapPrm: 0000000000000000 | |
209 | CapEff: 0000000000000000 | |
210 | CapBnd: ffffffffffffffff | |
f8d0dc21 | 211 | CapAmb: 0000000000000000 |
af884cd4 | 212 | NoNewPrivs: 0 |
2f4b3bf6 | 213 | Seccomp: 0 |
f8d0dc21 | 214 | Speculation_Store_Bypass: thread vulnerable |
fe719888 | 215 | SpeculationIndirectBranch: conditional enabled |
349888ee SS |
216 | voluntary_ctxt_switches: 0 |
217 | nonvoluntary_ctxt_switches: 1 | |
1da177e4 LT |
218 | |
219 | This shows you nearly the same information you would get if you viewed it with | |
220 | the ps command. In fact, ps uses the proc file system to obtain its | |
349888ee SS |
221 | information. But you get a more detailed view of the process by reading the |
222 | file /proc/PID/status. It fields are described in table 1-2. | |
223 | ||
224 | The statm file contains more detailed information about the process | |
225 | memory usage. Its seven fields are explained in Table 1-3. The stat file | |
059db434 | 226 | contains detailed information about the process itself. Its fields are |
349888ee | 227 | explained in Table 1-4. |
1da177e4 | 228 | |
34e55232 | 229 | (for SMP CONFIG users) |
c33e97ef | 230 | |
15eb42d6 NS |
231 | For making accounting scalable, RSS related information are handled in an |
232 | asynchronous manner and the value may not be very precise. To see a precise | |
34e55232 KH |
233 | snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. |
234 | It's slow but very precise. | |
235 | ||
d2ea66a6 | 236 | .. table:: Table 1-2: Contents of the status fields (as of 4.19) |
c33e97ef MCC |
237 | |
238 | ========================== =================================================== | |
349888ee | 239 | Field Content |
c33e97ef | 240 | ========================== =================================================== |
349888ee | 241 | Name filename of the executable |
bbd88e1d | 242 | Umask file mode creation mask |
349888ee SS |
243 | State state (R is running, S is sleeping, D is sleeping |
244 | in an uninterruptible wait, Z is zombie, | |
245 | T is traced or stopped) | |
246 | Tgid thread group ID | |
15eb42d6 | 247 | Ngid NUMA group ID (0 if none) |
349888ee SS |
248 | Pid process id |
249 | PPid process id of the parent process | |
df61e945 CL |
250 | TracerPid PID of process tracing this process (0 if not, or |
251 | the tracer is outside of the current pid namespace) | |
349888ee SS |
252 | Uid Real, effective, saved set, and file system UIDs |
253 | Gid Real, effective, saved set, and file system GIDs | |
254 | FDSize number of file descriptor slots currently allocated | |
255 | Groups supplementary group list | |
15eb42d6 NS |
256 | NStgid descendant namespace thread group ID hierarchy |
257 | NSpid descendant namespace process ID hierarchy | |
258 | NSpgid descendant namespace process group ID hierarchy | |
259 | NSsid descendant namespace session ID hierarchy | |
522dc4e5 | 260 | Kthread kernel thread flag, 1 is yes, 0 is no |
349888ee SS |
261 | VmPeak peak virtual memory size |
262 | VmSize total program size | |
263 | VmLck locked memory size | |
bbd88e1d | 264 | VmPin pinned memory size |
349888ee | 265 | VmHWM peak resident set size ("high water mark") |
8cee852e | 266 | VmRSS size of memory portions. It contains the three |
c33e97ef MCC |
267 | following parts |
268 | (VmRSS = RssAnon + RssFile + RssShmem) | |
8cee852e JM |
269 | RssAnon size of resident anonymous memory |
270 | RssFile size of resident file mappings | |
271 | RssShmem size of resident shmem memory (includes SysV shm, | |
272 | mapping of tmpfs and shared anonymous mappings) | |
30bdbb78 KK |
273 | VmData size of private data segments |
274 | VmStk size of stack segments | |
349888ee SS |
275 | VmExe size of text segment |
276 | VmLib size of shared library code | |
277 | VmPTE size of page table entries | |
bf9683d6 VB |
278 | VmSwap amount of swap used by anonymous private data |
279 | (shmem swap usage is not included) | |
5d317b2b | 280 | HugetlbPages size of hugetlb memory portions |
c6434012 RG |
281 | CoreDumping process's memory is currently being dumped |
282 | (killing the process may lead to a corrupted core) | |
a1400af7 MH |
283 | THP_enabled process is allowed to use THP (returns 0 when |
284 | PR_SET_THP_DISABLE is set on the process | |
349888ee SS |
285 | Threads number of threads |
286 | SigQ number of signals queued/max. number for queue | |
287 | SigPnd bitmap of pending signals for the thread | |
288 | ShdPnd bitmap of shared pending signals for the process | |
289 | SigBlk bitmap of blocked signals | |
290 | SigIgn bitmap of ignored signals | |
c98be0c9 | 291 | SigCgt bitmap of caught signals |
349888ee SS |
292 | CapInh bitmap of inheritable capabilities |
293 | CapPrm bitmap of permitted capabilities | |
294 | CapEff bitmap of effective capabilities | |
295 | CapBnd bitmap of capabilities bounding set | |
f8d0dc21 | 296 | CapAmb bitmap of ambient capabilities |
af884cd4 | 297 | NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) |
2f4b3bf6 | 298 | Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) |
f8d0dc21 | 299 | Speculation_Store_Bypass speculative store bypass mitigation status |
fe719888 | 300 | SpeculationIndirectBranch indirect branch speculation mode |
349888ee SS |
301 | Cpus_allowed mask of CPUs on which this process may run |
302 | Cpus_allowed_list Same as previous, but in "list format" | |
303 | Mems_allowed mask of memory nodes allowed to this process | |
304 | Mems_allowed_list Same as previous, but in "list format" | |
305 | voluntary_ctxt_switches number of voluntary context switches | |
306 | nonvoluntary_ctxt_switches number of non voluntary context switches | |
c33e97ef | 307 | ========================== =================================================== |
1da177e4 | 308 | |
c33e97ef | 309 | |
d2ea66a6 | 310 | .. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3) |
c33e97ef MCC |
311 | |
312 | ======== =============================== ============================== | |
1da177e4 | 313 | Field Content |
c33e97ef | 314 | ======== =============================== ============================== |
1da177e4 LT |
315 | size total program size (pages) (same as VmSize in status) |
316 | resident size of memory portions (pages) (same as VmRSS in status) | |
8cee852e JM |
317 | shared number of pages that are shared (i.e. backed by a file, same |
318 | as RssFile+RssShmem in status) | |
1da177e4 | 319 | trs number of pages that are 'code' (not including libs; broken, |
c33e97ef | 320 | includes data segment) |
1da177e4 LT |
321 | lrs number of pages of library (always 0 on 2.6) |
322 | drs number of pages of data/stack (including libs; broken, | |
c33e97ef | 323 | includes library text) |
1da177e4 | 324 | dt number of dirty pages (always 0 on 2.6) |
c33e97ef MCC |
325 | ======== =============================== ============================== |
326 | ||
1da177e4 | 327 | |
d2ea66a6 | 328 | .. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7) |
18d96779 | 329 | |
c33e97ef MCC |
330 | ============= =============================================================== |
331 | Field Content | |
332 | ============= =============================================================== | |
18d96779 KC |
333 | pid process id |
334 | tcomm filename of the executable | |
335 | state state (R is running, S is sleeping, D is sleeping in an | |
336 | uninterruptible wait, Z is zombie, T is traced or stopped) | |
337 | ppid process id of the parent process | |
338 | pgrp pgrp of the process | |
339 | sid session id | |
340 | tty_nr tty the process uses | |
341 | tty_pgrp pgrp of the tty | |
342 | flags task flags | |
343 | min_flt number of minor faults | |
344 | cmin_flt number of minor faults with child's | |
345 | maj_flt number of major faults | |
346 | cmaj_flt number of major faults with child's | |
347 | utime user mode jiffies | |
348 | stime kernel mode jiffies | |
349 | cutime user mode jiffies with child's | |
350 | cstime kernel mode jiffies with child's | |
351 | priority priority level | |
352 | nice nice level | |
353 | num_threads number of threads | |
2e01e00e | 354 | it_real_value (obsolete, always 0) |
18d96779 KC |
355 | start_time time the process started after system boot |
356 | vsize virtual memory size | |
357 | rss resident set memory size | |
358 | rsslim current limit in bytes on the rss | |
359 | start_code address above which program text can run | |
360 | end_code address below which program text can run | |
b7643757 | 361 | start_stack address of the start of the main process stack |
18d96779 KC |
362 | esp current value of ESP |
363 | eip current value of EIP | |
349888ee SS |
364 | pending bitmap of pending signals |
365 | blocked bitmap of blocked signals | |
366 | sigign bitmap of ignored signals | |
c98be0c9 | 367 | sigcatch bitmap of caught signals |
c33e97ef MCC |
368 | 0 (place holder, used to be the wchan address, |
369 | use /proc/PID/wchan instead) | |
18d96779 KC |
370 | 0 (place holder) |
371 | 0 (place holder) | |
372 | exit_signal signal to send to parent thread on exit | |
373 | task_cpu which CPU the task is scheduled on | |
374 | rt_priority realtime priority | |
375 | policy scheduling policy (man sched_setscheduler) | |
376 | blkio_ticks time spent waiting for block IO | |
349888ee SS |
377 | gtime guest time of the task in jiffies |
378 | cgtime guest time of the task children in jiffies | |
b3f7f573 CG |
379 | start_data address above which program data+bss is placed |
380 | end_data address below which program data+bss is placed | |
381 | start_brk address above which program heap can be expanded with brk() | |
5b172087 CG |
382 | arg_start address above which program command line is placed |
383 | arg_end address below which program command line is placed | |
384 | env_start address above which program environment is placed | |
385 | env_end address below which program environment is placed | |
c33e97ef MCC |
386 | exit_code the thread's exit_code in the form reported by the waitpid |
387 | system call | |
388 | ============= =============================================================== | |
18d96779 | 389 | |
ee2ad71b | 390 | The /proc/PID/maps file contains the currently mapped memory regions and |
349888ee SS |
391 | their access permissions. |
392 | ||
c33e97ef MCC |
393 | The format is:: |
394 | ||
395 | address perms offset dev inode pathname | |
396 | ||
397 | 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test | |
398 | 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test | |
399 | 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] | |
400 | a7cb1000-a7cb2000 ---p 00000000 00:00 0 | |
401 | a7cb2000-a7eb2000 rw-p 00000000 00:00 0 | |
402 | a7eb2000-a7eb3000 ---p 00000000 00:00 0 | |
403 | a7eb3000-a7ed5000 rw-p 00000000 00:00 0 | |
404 | a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 | |
405 | a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 | |
406 | a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 | |
407 | a800b000-a800e000 rw-p 00000000 00:00 0 | |
408 | a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 | |
409 | a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 | |
410 | a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 | |
411 | a8024000-a8027000 rw-p 00000000 00:00 0 | |
412 | a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 | |
413 | a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 | |
414 | a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 | |
415 | aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] | |
416 | ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] | |
349888ee SS |
417 | |
418 | where "address" is the address space in the process that it occupies, "perms" | |
c33e97ef | 419 | is a set of permissions:: |
349888ee SS |
420 | |
421 | r = read | |
422 | w = write | |
423 | x = execute | |
424 | s = shared | |
425 | p = private (copy on write) | |
426 | ||
427 | "offset" is the offset into the mapping, "dev" is the device (major:minor), and | |
428 | "inode" is the inode on that device. 0 indicates that no inode is associated | |
429 | with the memory region, as the case would be with BSS (uninitialized data). | |
430 | The "pathname" shows the name associated file for this mapping. If the mapping | |
431 | is not associated with a file: | |
432 | ||
d09e8ca6 | 433 | =================== =========================================== |
c33e97ef MCC |
434 | [heap] the heap of the program |
435 | [stack] the stack of the main process | |
436 | [vdso] the "virtual dynamic shared object", | |
349888ee | 437 | the kernel system call handler |
d09e8ca6 | 438 | [anon:<name>] a private anonymous mapping that has been |
9a10064f | 439 | named by userspace |
d09e8ca6 PT |
440 | [anon_shmem:<name>] an anonymous shared memory mapping that has |
441 | been named by userspace | |
442 | =================== =========================================== | |
349888ee SS |
443 | |
444 | or if empty, the mapping is anonymous. | |
445 | ||
349888ee | 446 | The /proc/PID/smaps is an extension based on maps, showing the memory |
ee2ad71b | 447 | consumption for each of the process's mappings. For each mapping (aka Virtual |
c33e97ef MCC |
448 | Memory Area, or VMA) there is a series of lines such as the following:: |
449 | ||
450 | 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash | |
451 | ||
452 | Size: 1084 kB | |
453 | KernelPageSize: 4 kB | |
454 | MMUPageSize: 4 kB | |
455 | Rss: 892 kB | |
456 | Pss: 374 kB | |
30934843 | 457 | Pss_Dirty: 0 kB |
c33e97ef MCC |
458 | Shared_Clean: 892 kB |
459 | Shared_Dirty: 0 kB | |
460 | Private_Clean: 0 kB | |
461 | Private_Dirty: 0 kB | |
462 | Referenced: 892 kB | |
463 | Anonymous: 0 kB | |
464 | LazyFree: 0 kB | |
465 | AnonHugePages: 0 kB | |
466 | ShmemPmdMapped: 0 kB | |
467 | Shared_Hugetlb: 0 kB | |
468 | Private_Hugetlb: 0 kB | |
469 | Swap: 0 kB | |
470 | SwapPss: 0 kB | |
471 | KernelPageSize: 4 kB | |
472 | MMUPageSize: 4 kB | |
473 | Locked: 0 kB | |
474 | THPeligible: 0 | |
475 | VmFlags: rd ex mr mw me dw | |
349888ee | 476 | |
ee2ad71b LS |
477 | The first of these lines shows the same information as is displayed for the |
478 | mapping in /proc/PID/maps. Following lines show the size of the mapping | |
479 | (size); the size of each page allocated when backing a VMA (KernelPageSize), | |
480 | which is usually the same as the size in the page table entries; the page size | |
481 | used by the MMU when backing a VMA (in most cases, the same as KernelPageSize); | |
482 | the amount of the mapping that is currently resident in RAM (RSS); the | |
483 | process' proportional share of this mapping (PSS); and the number of clean and | |
484 | dirty shared and private pages in the mapping. | |
8334b962 MK |
485 | |
486 | The "proportional set size" (PSS) of a process is the count of pages it has | |
487 | in memory, where each page is divided by the number of processes sharing it. | |
488 | So if a process has 1000 pages all to itself, and 1000 shared with one other | |
30934843 VW |
489 | process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which |
490 | consists of dirty pages. ("Pss_Clean" is not included, but it can be | |
491 | calculated by subtracting "Pss_Dirty" from "Pss".) | |
c33e97ef | 492 | |
8334b962 MK |
493 | Note that even a page which is part of a MAP_SHARED mapping, but has only |
494 | a single pte mapped, i.e. is currently used by only one process, is accounted | |
495 | as private and not as shared. | |
c33e97ef | 496 | |
8334b962 MK |
497 | "Referenced" indicates the amount of memory currently marked as referenced or |
498 | accessed. | |
c33e97ef | 499 | |
b40d4f84 NK |
500 | "Anonymous" shows the amount of memory that does not belong to any file. Even |
501 | a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE | |
502 | and a page is modified, the file page is replaced by a private anonymous copy. | |
c33e97ef | 503 | |
cf8496ea SL |
504 | "LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). |
505 | The memory isn't freed immediately with madvise(). It's freed in memory | |
506 | pressure if the memory is clean. Please note that the printed value might | |
507 | be lower than the real value due to optimizations used in the current | |
508 | implementation. If this is not desirable please file a bug report. | |
c33e97ef | 509 | |
25ee01a2 | 510 | "AnonHugePages" shows the ammount of memory backed by transparent hugepage. |
c33e97ef | 511 | |
1b5946a8 KS |
512 | "ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by |
513 | huge pages. | |
c33e97ef | 514 | |
25ee01a2 NH |
515 | "Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by |
516 | hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical | |
517 | reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. | |
c33e97ef | 518 | |
a5be3563 | 519 | "Swap" shows how much would-be-anonymous memory is also used, but out on swap. |
c33e97ef | 520 | |
c261e7d9 VB |
521 | For shmem mappings, "Swap" includes also the size of the mapped (and not |
522 | replaced by copy-on-write) part of the underlying shmem object out on swap. | |
523 | "SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this | |
524 | does not take into account swapped out page of underlying shmem objects. | |
a5be3563 | 525 | "Locked" indicates whether the mapping is locked in memory or not. |
cb55b838 | 526 | |
c0630669 | 527 | "THPeligible" indicates whether the mapping is eligible for allocating THP |
cb55b838 YS |
528 | pages as well as the THP is PMD mappable or not - 1 if true, 0 otherwise. |
529 | It just shows the current status. | |
25ee01a2 | 530 | |
c33e97ef MCC |
531 | "VmFlags" field deserves a separate description. This member represents the |
532 | kernel flags associated with the particular virtual memory area in two letter | |
533 | encoded manner. The codes are the following: | |
534 | ||
535 | == ======================================= | |
536 | rd readable | |
537 | wr writeable | |
538 | ex executable | |
539 | sh shared | |
540 | mr may read | |
541 | mw may write | |
542 | me may execute | |
543 | ms may share | |
544 | gd stack segment growns down | |
545 | pf pure PFN range | |
546 | dw disabled write to the mapped file | |
547 | lo pages are locked in memory | |
548 | io memory mapped I/O area | |
549 | sr sequential read advise provided | |
550 | rr random read advise provided | |
551 | dc do not copy area on fork | |
552 | de do not expand area on remapping | |
553 | ac area is accountable | |
554 | nr swap space is not reserved for the area | |
555 | ht area uses huge tlb pages | |
1f7faca2 | 556 | sf synchronous page fault |
c33e97ef | 557 | ar architecture specific flag |
1f7faca2 | 558 | wf wipe on fork |
c33e97ef MCC |
559 | dd do not include area into core dump |
560 | sd soft dirty flag | |
561 | mm mixed map area | |
562 | hg huge page advise flag | |
563 | nh no huge page advise flag | |
564 | mg mergable advise flag | |
d5ddc6d9 | 565 | bt arm64 BTI guarded page |
868770c9 | 566 | mt arm64 MTE allocation tags are enabled |
1f7faca2 PX |
567 | um userfaultfd missing tracking |
568 | uw userfaultfd wr-protect tracking | |
c33e97ef | 569 | == ======================================= |
834f82e2 CG |
570 | |
571 | Note that there is no guarantee that every flag and associated mnemonic will | |
572 | be present in all further kernel releases. Things get changed, the flags may | |
7550c607 MH |
573 | be vanished or the reverse -- new added. Interpretation of their meaning |
574 | might change in future as well. So each consumer of these flags has to | |
575 | follow each specific kernel version for the exact semantic. | |
834f82e2 | 576 | |
349888ee SS |
577 | This file is only present if the CONFIG_MMU kernel configuration option is |
578 | enabled. | |
18d96779 | 579 | |
53aeee7a RH |
580 | Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent |
581 | output can be achieved only in the single read call). | |
c33e97ef | 582 | |
53aeee7a RH |
583 | This typically manifests when doing partial reads of these files while the |
584 | memory map is being modified. Despite the races, we do provide the following | |
585 | guarantees: | |
586 | ||
587 | 1) The mapped addresses never go backwards, which implies no two | |
588 | regions will ever overlap. | |
589 | 2) If there is something at a given vaddr during the entirety of the | |
590 | life of the smaps/maps walk, there will be some output for it. | |
591 | ||
ee2ad71b LS |
592 | The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, |
593 | but their values are the sums of the corresponding values for all mappings of | |
594 | the process. Additionally, it contains these fields: | |
595 | ||
c33e97ef MCC |
596 | - Pss_Anon |
597 | - Pss_File | |
598 | - Pss_Shmem | |
ee2ad71b LS |
599 | |
600 | They represent the proportional shares of anonymous, file, and shmem pages, as | |
601 | described for smaps above. These fields are omitted in smaps since each | |
602 | mapping identifies the type (anon, file, or shmem) of all pages it contains. | |
603 | Thus all information in smaps_rollup can be derived from smaps, but at a | |
604 | significantly higher cost. | |
53aeee7a | 605 | |
398499d5 | 606 | The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG |
0f8975ec | 607 | bits on both physical and virtual pages associated with a process, and the |
1ad1335d MR |
608 | soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst |
609 | for details). | |
c33e97ef MCC |
610 | To clear the bits for all the pages associated with the process:: |
611 | ||
398499d5 MB |
612 | > echo 1 > /proc/PID/clear_refs |
613 | ||
c33e97ef MCC |
614 | To clear the bits for the anonymous pages associated with the process:: |
615 | ||
398499d5 MB |
616 | > echo 2 > /proc/PID/clear_refs |
617 | ||
c33e97ef MCC |
618 | To clear the bits for the file mapped pages associated with the process:: |
619 | ||
398499d5 | 620 | > echo 3 > /proc/PID/clear_refs |
0f8975ec | 621 | |
c33e97ef MCC |
622 | To clear the soft-dirty bit:: |
623 | ||
0f8975ec PE |
624 | > echo 4 > /proc/PID/clear_refs |
625 | ||
695f0559 | 626 | To reset the peak resident set size ("high water mark") to the process's |
c33e97ef MCC |
627 | current value:: |
628 | ||
695f0559 PC |
629 | > echo 5 > /proc/PID/clear_refs |
630 | ||
398499d5 MB |
631 | Any other value written to /proc/PID/clear_refs will have no effect. |
632 | ||
03f890f8 NK |
633 | The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags |
634 | using /proc/kpageflags and number of times a page is mapped using | |
1ad1335d MR |
635 | /proc/kpagecount. For detailed explanation, see |
636 | Documentation/admin-guide/mm/pagemap.rst. | |
398499d5 | 637 | |
0c369711 RA |
638 | The /proc/pid/numa_maps is an extension based on maps, showing the memory |
639 | locality and binding policy, as well as the memory usage (in pages) of | |
640 | each mapping. The output follows a general format where mapping details get | |
c33e97ef MCC |
641 | summarized separated by blank spaces, one mapping per each file line:: |
642 | ||
643 | address policy mapping details | |
644 | ||
645 | 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 | |
646 | 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 | |
647 | 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 | |
648 | 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 | |
649 | 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 | |
650 | 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 | |
651 | 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 | |
652 | 320698b000 default file=/lib64/libc-2.12.so | |
653 | 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 | |
654 | 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 | |
655 | 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 | |
656 | 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 | |
657 | 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 | |
658 | 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 | |
659 | 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 | |
660 | 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 | |
0c369711 RA |
661 | |
662 | Where: | |
c33e97ef | 663 | |
0c369711 | 664 | "address" is the starting address for the mapping; |
c33e97ef | 665 | |
3ecf53e4 | 666 | "policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); |
c33e97ef | 667 | |
0c369711 RA |
668 | "mapping details" summarizes mapping data such as mapping type, page usage counters, |
669 | node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page | |
670 | size, in KB, that is backing the mapping up. | |
671 | ||
1da177e4 LT |
672 | 1.2 Kernel data |
673 | --------------- | |
674 | ||
675 | Similar to the process entries, the kernel data files give information about | |
676 | the running kernel. The files used to obtain this information are contained in | |
349888ee | 677 | /proc and are listed in Table 1-5. Not all of these will be present in your |
1da177e4 LT |
678 | system. It depends on the kernel configuration and the loaded modules, which |
679 | files are there, and which are missing. | |
680 | ||
c33e97ef MCC |
681 | .. table:: Table 1-5: Kernel info in /proc |
682 | ||
683 | ============ =============================================================== | |
684 | File Content | |
685 | ============ =============================================================== | |
686 | apm Advanced power management info | |
687 | buddyinfo Kernel memory allocator information (see text) (2.5) | |
688 | bus Directory containing bus specific information | |
689 | cmdline Kernel command line | |
690 | cpuinfo Info about the CPU | |
691 | devices Available devices (block and character) | |
692 | dma Used DMS channels | |
693 | filesystems Supported filesystems | |
694 | driver Various drivers grouped here, currently rtc (2.4) | |
695 | execdomains Execdomains, related to security (2.4) | |
696 | fb Frame Buffer devices (2.4) | |
697 | fs File system parameters, currently nfs/exports (2.4) | |
698 | ide Directory containing info about the IDE subsystem | |
699 | interrupts Interrupt usage | |
700 | iomem Memory map (2.4) | |
701 | ioports I/O port usage | |
702 | irq Masks for irq to cpu affinity (2.4)(smp?) | |
703 | isapnp ISA PnP (Plug&Play) Info (2.4) | |
704 | kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) | |
705 | kmsg Kernel messages | |
706 | ksyms Kernel symbol table | |
4ba1d726 RD |
707 | loadavg Load average of last 1, 5 & 15 minutes; |
708 | number of processes currently runnable (running or on ready queue); | |
709 | total number of processes in system; | |
710 | last pid created. | |
93ea4a0b RD |
711 | All fields are separated by one space except "number of |
712 | processes currently runnable" and "total number of processes | |
713 | in system", which are separated by a slash ('/'). Example: | |
f37a15ea | 714 | 0.61 0.61 0.55 3/828 22084 |
c33e97ef MCC |
715 | locks Kernel locks |
716 | meminfo Memory info | |
717 | misc Miscellaneous | |
718 | modules List of loaded modules | |
719 | mounts Mounted filesystems | |
720 | net Networking info (see text) | |
a1b57ac0 | 721 | pagetypeinfo Additional page allocator information (see text) (2.5) |
c33e97ef MCC |
722 | partitions Table of partitions known to the system |
723 | pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, | |
724 | decoupled by lspci (2.4) | |
725 | rtc Real time clock | |
726 | scsi SCSI info (see text) | |
727 | slabinfo Slab pool info | |
728 | softirqs softirq usage | |
729 | stat Overall statistics | |
730 | swaps Swap space utilization | |
731 | sys See chapter 2 | |
732 | sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) | |
733 | tty Info of tty drivers | |
734 | uptime Wall clock since boot, combined idle time of all cpus | |
735 | version Kernel version | |
736 | video bttv info of video resources (2.4) | |
737 | vmallocinfo Show vmalloced areas | |
738 | ============ =============================================================== | |
1da177e4 LT |
739 | |
740 | You can, for example, check which interrupts are currently in use and what | |
c33e97ef MCC |
741 | they are used for by looking in the file /proc/interrupts:: |
742 | ||
743 | > cat /proc/interrupts | |
744 | CPU0 | |
745 | 0: 8728810 XT-PIC timer | |
746 | 1: 895 XT-PIC keyboard | |
747 | 2: 0 XT-PIC cascade | |
748 | 3: 531695 XT-PIC aha152x | |
749 | 4: 2014133 XT-PIC serial | |
750 | 5: 44401 XT-PIC pcnet_cs | |
751 | 8: 2 XT-PIC rtc | |
752 | 11: 8 XT-PIC i82365 | |
753 | 12: 182918 XT-PIC PS/2 Mouse | |
754 | 13: 1 XT-PIC fpu | |
755 | 14: 1232265 XT-PIC ide0 | |
756 | 15: 7 XT-PIC ide1 | |
757 | NMI: 0 | |
1da177e4 LT |
758 | |
759 | In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the | |
c33e97ef | 760 | output of a SMP machine):: |
1da177e4 | 761 | |
c33e97ef | 762 | > cat /proc/interrupts |
1da177e4 | 763 | |
c33e97ef | 764 | CPU0 CPU1 |
1da177e4 LT |
765 | 0: 1243498 1214548 IO-APIC-edge timer |
766 | 1: 8949 8958 IO-APIC-edge keyboard | |
767 | 2: 0 0 XT-PIC cascade | |
768 | 5: 11286 10161 IO-APIC-edge soundblaster | |
769 | 8: 1 0 IO-APIC-edge rtc | |
770 | 9: 27422 27407 IO-APIC-edge 3c503 | |
771 | 12: 113645 113873 IO-APIC-edge PS/2 Mouse | |
772 | 13: 0 0 XT-PIC fpu | |
773 | 14: 22491 24012 IO-APIC-edge ide0 | |
774 | 15: 2183 2415 IO-APIC-edge ide1 | |
775 | 17: 30564 30414 IO-APIC-level eth0 | |
776 | 18: 177 164 IO-APIC-level bttv | |
c33e97ef MCC |
777 | NMI: 2457961 2457959 |
778 | LOC: 2457882 2457881 | |
1da177e4 LT |
779 | ERR: 2155 |
780 | ||
781 | NMI is incremented in this case because every timer interrupt generates a NMI | |
782 | (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. | |
783 | ||
784 | LOC is the local interrupt counter of the internal APIC of every CPU. | |
785 | ||
786 | ERR is incremented in the case of errors in the IO-APIC bus (the bus that | |
787 | connects the CPUs in a SMP system. This means that an error has been detected, | |
788 | the IO-APIC automatically retry the transmission, so it should not be a big | |
789 | problem, but you should read the SMP-FAQ. | |
790 | ||
38e760a1 JK |
791 | In 2.6.2* /proc/interrupts was expanded again. This time the goal was for |
792 | /proc/interrupts to display every IRQ vector in use by the system, not | |
793 | just those considered 'most important'. The new vectors are: | |
794 | ||
c33e97ef MCC |
795 | THR |
796 | interrupt raised when a machine check threshold counter | |
38e760a1 JK |
797 | (typically counting ECC corrected errors of memory or cache) exceeds |
798 | a configurable threshold. Only available on some systems. | |
799 | ||
c33e97ef MCC |
800 | TRM |
801 | a thermal event interrupt occurs when a temperature threshold | |
38e760a1 JK |
802 | has been exceeded for the CPU. This interrupt may also be generated |
803 | when the temperature drops back to normal. | |
804 | ||
c33e97ef MCC |
805 | SPU |
806 | a spurious interrupt is some interrupt that was raised then lowered | |
38e760a1 JK |
807 | by some IO device before it could be fully processed by the APIC. Hence |
808 | the APIC sees the interrupt but does not know what device it came from. | |
809 | For this case the APIC will generate the interrupt with a IRQ vector | |
810 | of 0xff. This might also be generated by chipset bugs. | |
811 | ||
059db434 | 812 | RES, CAL, TLB |
c33e97ef | 813 | rescheduling, call and TLB flush interrupts are |
38e760a1 JK |
814 | sent from one CPU to another per the needs of the OS. Typically, |
815 | their statistics are used by kernel developers and interested users to | |
19f59460 | 816 | determine the occurrence of interrupts of the given type. |
38e760a1 | 817 | |
25985edc | 818 | The above IRQ vectors are displayed only when relevant. For example, |
38e760a1 JK |
819 | the threshold vector does not exist on x86_64 platforms. Others are |
820 | suppressed when the system is a uniprocessor. As of this writing, only | |
821 | i386 and x86_64 platforms support the new IRQ vector displays. | |
822 | ||
823 | Of some interest is the introduction of the /proc/irq directory to 2.4. | |
059db434 | 824 | It could be used to set IRQ to CPU affinity. This means that you can "hook" an |
1da177e4 | 825 | IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the |
18404756 MK |
826 | irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and |
827 | prof_cpu_mask. | |
1da177e4 | 828 | |
c33e97ef MCC |
829 | For example:: |
830 | ||
1da177e4 LT |
831 | > ls /proc/irq/ |
832 | 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask | |
18404756 | 833 | 1 11 13 15 17 19 3 5 7 9 default_smp_affinity |
1da177e4 LT |
834 | > ls /proc/irq/0/ |
835 | smp_affinity | |
836 | ||
18404756 | 837 | smp_affinity is a bitmask, in which you can specify which CPUs can handle the |
059db434 | 838 | IRQ. You can set it by doing:: |
1da177e4 | 839 | |
18404756 MK |
840 | > echo 1 > /proc/irq/10/smp_affinity |
841 | ||
842 | This means that only the first CPU will handle the IRQ, but you can also echo | |
99e9d958 | 843 | 5 which means that only the first and third CPU can handle the IRQ. |
1da177e4 | 844 | |
c33e97ef | 845 | The contents of each smp_affinity file is the same by default:: |
18404756 MK |
846 | |
847 | > cat /proc/irq/0/smp_affinity | |
848 | ffffffff | |
1da177e4 | 849 | |
4b060420 | 850 | There is an alternate interface, smp_affinity_list which allows specifying |
059db434 | 851 | a CPU range instead of a bitmask:: |
4b060420 MT |
852 | |
853 | > cat /proc/irq/0/smp_affinity_list | |
854 | 1024-1031 | |
855 | ||
18404756 MK |
856 | The default_smp_affinity mask applies to all non-active IRQs, which are the |
857 | IRQs which have not yet been allocated/activated, and hence which lack a | |
858 | /proc/irq/[0-9]* directory. | |
1da177e4 | 859 | |
92d6b71a DS |
860 | The node file on an SMP system shows the node to which the device using the IRQ |
861 | reports itself as being attached. This hardware locality information does not | |
862 | include information about any possible driver locality preference. | |
863 | ||
18404756 | 864 | prof_cpu_mask specifies which CPUs are to be profiled by the system wide |
059db434 | 865 | profiler. Default value is ffffffff (all CPUs if there are only 32 of them). |
1da177e4 LT |
866 | |
867 | The way IRQs are routed is handled by the IO-APIC, and it's Round Robin | |
868 | between all the CPUs which are allowed to handle it. As usual the kernel has | |
869 | more info than you and does a better job than you, so the defaults are the | |
4b060420 MT |
870 | best choice for almost everyone. [Note this applies only to those IO-APIC's |
871 | that support "Round Robin" interrupt distribution.] | |
1da177e4 LT |
872 | |
873 | There are three more important subdirectories in /proc: net, scsi, and sys. | |
874 | The general rule is that the contents, or even the existence of these | |
875 | directories, depend on your kernel configuration. If SCSI is not enabled, the | |
876 | directory scsi may not exist. The same is true with the net, which is there | |
877 | only when networking support is present in the running kernel. | |
878 | ||
879 | The slabinfo file gives information about memory usage at the slab level. | |
880 | Linux uses slab pools for memory management above page level in version 2.2. | |
881 | Commonly used objects have their own slab pool (such as network buffers, | |
882 | directory cache, and so on). | |
883 | ||
c33e97ef | 884 | :: |
1da177e4 | 885 | |
c33e97ef | 886 | > cat /proc/buddyinfo |
1da177e4 | 887 | |
c33e97ef MCC |
888 | Node 0, zone DMA 0 4 5 4 4 3 ... |
889 | Node 0, zone Normal 1 0 0 1 101 8 ... | |
890 | Node 0, zone HighMem 2 0 0 1 1 0 ... | |
1da177e4 | 891 | |
a1b57ac0 | 892 | External fragmentation is a problem under some workloads, and buddyinfo is a |
c33e97ef | 893 | useful tool for helping diagnose these problems. Buddyinfo will give you a |
1da177e4 LT |
894 | clue as to how big an area you can safely allocate, or why a previous |
895 | allocation failed. | |
896 | ||
c33e97ef MCC |
897 | Each column represents the number of pages of a certain order which are |
898 | available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in | |
899 | ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE | |
900 | available in ZONE_NORMAL, etc... | |
1da177e4 | 901 | |
a1b57ac0 | 902 | More information relevant to external fragmentation can be found in |
c33e97ef MCC |
903 | pagetypeinfo:: |
904 | ||
905 | > cat /proc/pagetypeinfo | |
906 | Page block order: 9 | |
907 | Pages per block: 512 | |
908 | ||
909 | Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 | |
910 | Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 | |
911 | Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 | |
912 | Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 | |
913 | Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 | |
914 | Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 | |
915 | Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 | |
916 | Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 | |
917 | Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 | |
918 | Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 | |
919 | Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 | |
920 | ||
921 | Number of blocks type Unmovable Reclaimable Movable Reserve Isolate | |
922 | Node 0, zone DMA 2 0 5 1 0 | |
923 | Node 0, zone DMA32 41 6 967 2 0 | |
a1b57ac0 MG |
924 | |
925 | Fragmentation avoidance in the kernel works by grouping pages of different | |
926 | migrate types into the same contiguous regions of memory called page blocks. | |
059db434 | 927 | A page block is typically the size of the default hugepage size, e.g. 2MB on |
a1b57ac0 MG |
928 | X86-64. By keeping pages grouped based on their ability to move, the kernel |
929 | can reclaim pages within a page block to satisfy a high-order allocation. | |
930 | ||
931 | The pagetypinfo begins with information on the size of a page block. It | |
932 | then gives the same type of information as buddyinfo except broken down | |
933 | by migrate-type and finishes with details on how many page blocks of each | |
934 | type exist. | |
935 | ||
936 | If min_free_kbytes has been tuned correctly (recommendations made by hugeadm | |
ceec86ec | 937 | from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can |
a1b57ac0 MG |
938 | make an estimate of the likely number of huge pages that can be allocated |
939 | at a given point in time. All the "Movable" blocks should be allocatable | |
940 | unless memory has been mlock()'d. Some of the Reclaimable blocks should | |
941 | also be allocatable although a lot of filesystem metadata may have to be | |
942 | reclaimed to achieve this. | |
943 | ||
1da177e4 | 944 | |
c33e97ef MCC |
945 | meminfo |
946 | ~~~~~~~ | |
1da177e4 LT |
947 | |
948 | Provides information about distribution and utilization of memory. This | |
8d719afc MR |
949 | varies by architecture and compile options. Some of the counters reported |
950 | here overlap. The memory reported by the non overlapping counters may not | |
951 | add up to the overall memory usage and the difference for some workloads | |
952 | can be substantial. In many cases there are other means to find out | |
953 | additional memory using subsystem specific interfaces, for instance | |
954 | /proc/net/sockstat for TCP memory allocations. | |
955 | ||
39799b64 | 956 | Example output. You may not have all of these fields. |
1da177e4 | 957 | |
c33e97ef MCC |
958 | :: |
959 | ||
960 | > cat /proc/meminfo | |
961 | ||
39799b64 JW |
962 | MemTotal: 32858820 kB |
963 | MemFree: 21001236 kB | |
964 | MemAvailable: 27214312 kB | |
965 | Buffers: 581092 kB | |
966 | Cached: 5587612 kB | |
967 | SwapCached: 0 kB | |
968 | Active: 3237152 kB | |
969 | Inactive: 7586256 kB | |
970 | Active(anon): 94064 kB | |
971 | Inactive(anon): 4570616 kB | |
972 | Active(file): 3143088 kB | |
973 | Inactive(file): 3015640 kB | |
974 | Unevictable: 0 kB | |
975 | Mlocked: 0 kB | |
976 | SwapTotal: 0 kB | |
977 | SwapFree: 0 kB | |
f6498b77 JW |
978 | Zswap: 1904 kB |
979 | Zswapped: 7792 kB | |
39799b64 JW |
980 | Dirty: 12 kB |
981 | Writeback: 0 kB | |
982 | AnonPages: 4654780 kB | |
983 | Mapped: 266244 kB | |
984 | Shmem: 9976 kB | |
985 | KReclaimable: 517708 kB | |
986 | Slab: 660044 kB | |
987 | SReclaimable: 517708 kB | |
988 | SUnreclaim: 142336 kB | |
989 | KernelStack: 11168 kB | |
990 | PageTables: 20540 kB | |
ebc97a52 | 991 | SecPageTables: 0 kB |
39799b64 JW |
992 | NFS_Unstable: 0 kB |
993 | Bounce: 0 kB | |
994 | WritebackTmp: 0 kB | |
995 | CommitLimit: 16429408 kB | |
996 | Committed_AS: 7715148 kB | |
997 | VmallocTotal: 34359738367 kB | |
998 | VmallocUsed: 40444 kB | |
999 | VmallocChunk: 0 kB | |
1000 | Percpu: 29312 kB | |
bd23024b | 1001 | EarlyMemtestBad: 0 kB |
39799b64 JW |
1002 | HardwareCorrupted: 0 kB |
1003 | AnonHugePages: 4149248 kB | |
1004 | ShmemHugePages: 0 kB | |
1005 | ShmemPmdMapped: 0 kB | |
1006 | FileHugePages: 0 kB | |
1007 | FilePmdMapped: 0 kB | |
1008 | CmaTotal: 0 kB | |
1009 | CmaFree: 0 kB | |
1010 | HugePages_Total: 0 | |
1011 | HugePages_Free: 0 | |
1012 | HugePages_Rsvd: 0 | |
1013 | HugePages_Surp: 0 | |
1014 | Hugepagesize: 2048 kB | |
1015 | Hugetlb: 0 kB | |
1016 | DirectMap4k: 401152 kB | |
1017 | DirectMap2M: 10008576 kB | |
1018 | DirectMap1G: 24117248 kB | |
c33e97ef MCC |
1019 | |
1020 | MemTotal | |
059db434 | 1021 | Total usable RAM (i.e. physical RAM minus a few reserved |
1da177e4 | 1022 | bits and the kernel binary code) |
c33e97ef | 1023 | MemFree |
39799b64 | 1024 | Total free RAM. On highmem systems, the sum of LowFree+HighFree |
c33e97ef MCC |
1025 | MemAvailable |
1026 | An estimate of how much memory is available for starting new | |
34e431b0 RR |
1027 | applications, without swapping. Calculated from MemFree, |
1028 | SReclaimable, the size of the file LRU lists, and the low | |
1029 | watermarks in each zone. | |
1030 | The estimate takes into account that the system needs some | |
1031 | page cache to function well, and that not all reclaimable | |
1032 | slab will be reclaimable, due to items being in use. The | |
1033 | impact of those factors will vary from system to system. | |
c33e97ef MCC |
1034 | Buffers |
1035 | Relatively temporary storage for raw disk blocks | |
1da177e4 | 1036 | shouldn't get tremendously large (20MB or so) |
c33e97ef | 1037 | Cached |
39799b64 JW |
1038 | In-memory cache for files read from the disk (the |
1039 | pagecache) as well as tmpfs & shmem. | |
1040 | Doesn't include SwapCached. | |
c33e97ef MCC |
1041 | SwapCached |
1042 | Memory that once was swapped out, is swapped back in but | |
1da177e4 LT |
1043 | still also is in the swapfile (if memory is needed it |
1044 | doesn't need to be swapped out AGAIN because it is already | |
1045 | in the swapfile. This saves I/O) | |
c33e97ef MCC |
1046 | Active |
1047 | Memory that has been used more recently and usually not | |
1da177e4 | 1048 | reclaimed unless absolutely necessary. |
c33e97ef MCC |
1049 | Inactive |
1050 | Memory which has been less recently used. It is more | |
1da177e4 | 1051 | eligible to be reclaimed for other purposes |
39799b64 JW |
1052 | Unevictable |
1053 | Memory allocated for userspace which cannot be reclaimed, such | |
1054 | as mlocked pages, ramfs backing pages, secret memfd pages etc. | |
1055 | Mlocked | |
1056 | Memory locked with mlock(). | |
c33e97ef | 1057 | HighTotal, HighFree |
059db434 | 1058 | Highmem is all memory above ~860MB of physical memory. |
1da177e4 LT |
1059 | Highmem areas are for use by userspace programs, or |
1060 | for the pagecache. The kernel must use tricks to access | |
1061 | this memory, making it slower to access than lowmem. | |
c33e97ef MCC |
1062 | LowTotal, LowFree |
1063 | Lowmem is memory which can be used for everything that | |
3f6dee9b | 1064 | highmem can be used for, but it is also available for the |
1da177e4 LT |
1065 | kernel's use for its own data structures. Among many |
1066 | other things, it is where everything from the Slab is | |
1067 | allocated. Bad things happen when you're out of lowmem. | |
c33e97ef MCC |
1068 | SwapTotal |
1069 | total amount of swap space available | |
1070 | SwapFree | |
1071 | Memory which has been evicted from RAM, and is temporarily | |
1da177e4 | 1072 | on the disk |
f6498b77 JW |
1073 | Zswap |
1074 | Memory consumed by the zswap backend (compressed size) | |
1075 | Zswapped | |
1076 | Amount of anonymous memory stored in zswap (original size) | |
c33e97ef MCC |
1077 | Dirty |
1078 | Memory which is waiting to get written back to the disk | |
1079 | Writeback | |
1080 | Memory which is actively being written back to the disk | |
1081 | AnonPages | |
1082 | Non-file backed pages mapped into userspace page tables | |
c33e97ef MCC |
1083 | Mapped |
1084 | files which have been mmaped, such as libraries | |
1085 | Shmem | |
1086 | Total memory used by shared memory (shmem) and tmpfs | |
c33e97ef MCC |
1087 | KReclaimable |
1088 | Kernel allocations that the kernel will attempt to reclaim | |
61f94e18 VB |
1089 | under memory pressure. Includes SReclaimable (below), and other |
1090 | direct allocations with a shrinker. | |
c33e97ef MCC |
1091 | Slab |
1092 | in-kernel data structures cache | |
1093 | SReclaimable | |
1094 | Part of Slab, that might be reclaimed, such as caches | |
1095 | SUnreclaim | |
1096 | Part of Slab, that cannot be reclaimed on memory pressure | |
39799b64 JW |
1097 | KernelStack |
1098 | Memory consumed by the kernel stacks of all tasks | |
c33e97ef | 1099 | PageTables |
39799b64 | 1100 | Memory consumed by userspace page tables |
ebc97a52 YA |
1101 | SecPageTables |
1102 | Memory consumed by secondary page tables, this currently | |
1103 | currently includes KVM mmu allocations on x86 and arm64. | |
c33e97ef | 1104 | NFS_Unstable |
8d92890b N |
1105 | Always zero. Previous counted pages which had been written to |
1106 | the server, but has not been committed to stable storage. | |
c33e97ef MCC |
1107 | Bounce |
1108 | Memory used for block device "bounce buffers" | |
1109 | WritebackTmp | |
1110 | Memory used by FUSE for temporary writeback buffers | |
1111 | CommitLimit | |
1112 | Based on the overcommit ratio ('vm.overcommit_ratio'), | |
1da177e4 LT |
1113 | this is the total amount of memory currently available to |
1114 | be allocated on the system. This limit is only adhered to | |
1115 | if strict overcommit accounting is enabled (mode 2 in | |
1116 | 'vm.overcommit_memory'). | |
c33e97ef MCC |
1117 | |
1118 | The CommitLimit is calculated with the following formula:: | |
1119 | ||
1120 | CommitLimit = ([total RAM pages] - [total huge TLB pages]) * | |
1121 | overcommit_ratio / 100 + [total swap pages] | |
1122 | ||
1da177e4 LT |
1123 | For example, on a system with 1G of physical RAM and 7G |
1124 | of swap with a `vm.overcommit_ratio` of 30 it would | |
1125 | yield a CommitLimit of 7.3G. | |
c33e97ef | 1126 | |
1da177e4 | 1127 | For more details, see the memory overcommit documentation |
ee65728e | 1128 | in mm/overcommit-accounting. |
c33e97ef MCC |
1129 | Committed_AS |
1130 | The amount of memory presently allocated on the system. | |
1da177e4 LT |
1131 | The committed memory is a sum of all of the memory which |
1132 | has been allocated by processes, even if it has not been | |
1133 | "used" by them as of yet. A process which malloc()'s 1G | |
46496022 | 1134 | of memory, but only touches 300M of it will show up as |
39799b64 | 1135 | using 1G. This 1G is memory which has been "committed" to |
46496022 MJ |
1136 | by the VM and can be used at any time by the allocating |
1137 | application. With strict overcommit enabled on the system | |
059db434 | 1138 | (mode 2 in 'vm.overcommit_memory'), allocations which would |
46496022 MJ |
1139 | exceed the CommitLimit (detailed above) will not be permitted. |
1140 | This is useful if one needs to guarantee that processes will | |
1141 | not fail due to lack of memory once that memory has been | |
1142 | successfully allocated. | |
c33e97ef | 1143 | VmallocTotal |
39799b64 | 1144 | total size of vmalloc virtual address space |
c33e97ef MCC |
1145 | VmallocUsed |
1146 | amount of vmalloc area which is used | |
1147 | VmallocChunk | |
1148 | largest contiguous block of vmalloc area which is free | |
1149 | Percpu | |
1150 | Memory allocated to the percpu allocator used to back percpu | |
7e8a6304 | 1151 | allocations. This stat excludes the cost of metadata. |
bd23024b TM |
1152 | EarlyMemtestBad |
1153 | The amount of RAM/memory in kB, that was identified as corrupted | |
1154 | by early memtest. If memtest was not run, this field will not | |
1155 | be displayed at all. Size is never rounded down to 0 kB. | |
1156 | That means if 0 kB is reported, you can safely assume | |
1157 | there was at least one pass of memtest and none of the passes | |
1158 | found a single faulty byte of RAM. | |
39799b64 JW |
1159 | HardwareCorrupted |
1160 | The amount of RAM/memory in KB, the kernel identifies as | |
1161 | corrupted. | |
1162 | AnonHugePages | |
1163 | Non-file backed huge pages mapped into userspace page tables | |
1164 | ShmemHugePages | |
1165 | Memory used by shared memory (shmem) and tmpfs allocated | |
1166 | with huge pages | |
1167 | ShmemPmdMapped | |
1168 | Shared memory mapped into userspace with huge pages | |
1169 | FileHugePages | |
1170 | Memory used for filesystem data (page cache) allocated | |
1171 | with huge pages | |
1172 | FilePmdMapped | |
1173 | Page cache mapped into userspace with huge pages | |
1174 | CmaTotal | |
1175 | Memory reserved for the Contiguous Memory Allocator (CMA) | |
1176 | CmaFree | |
1177 | Free remaining memory in the CMA reserves | |
1178 | HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb | |
1179 | See Documentation/admin-guide/mm/hugetlbpage.rst. | |
1180 | DirectMap4k, DirectMap2M, DirectMap1G | |
1181 | Breakdown of page table sizes used in the kernel's | |
1182 | identity mapping of RAM | |
1da177e4 | 1183 | |
c33e97ef MCC |
1184 | vmallocinfo |
1185 | ~~~~~~~~~~~ | |
a47a126a ED |
1186 | |
1187 | Provides information about vmalloced/vmaped areas. One line per area, | |
1188 | containing the virtual address range of the area, size in bytes, | |
1189 | caller information of the creator, and optional information depending | |
059db434 | 1190 | on the kind of area: |
a47a126a | 1191 | |
c33e97ef | 1192 | ========== =================================================== |
a47a126a ED |
1193 | pages=nr number of pages |
1194 | phys=addr if a physical address was specified | |
1195 | ioremap I/O mapping (ioremap() and friends) | |
1196 | vmalloc vmalloc() area | |
1197 | vmap vmap()ed pages | |
1198 | user VM_USERMAP area | |
1199 | vpages buffer for pages pointers was vmalloced (huge area) | |
1200 | N<node>=nr (Only on NUMA kernels) | |
1201 | Number of pages allocated on memory node <node> | |
c33e97ef MCC |
1202 | ========== =================================================== |
1203 | ||
1204 | :: | |
1205 | ||
1206 | > cat /proc/vmallocinfo | |
1207 | 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... | |
1208 | /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 | |
1209 | 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... | |
1210 | /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 | |
1211 | 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... | |
1212 | phys=7fee8000 ioremap | |
1213 | 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... | |
1214 | phys=7fee7000 ioremap | |
1215 | 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 | |
1216 | 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... | |
1217 | /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 | |
1218 | 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... | |
1219 | pages=2 vmalloc N1=2 | |
1220 | 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... | |
1221 | /0x130 [x_tables] pages=4 vmalloc N0=4 | |
1222 | 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... | |
1223 | pages=14 vmalloc N2=14 | |
1224 | 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... | |
1225 | pages=4 vmalloc N1=4 | |
1226 | 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... | |
1227 | pages=2 vmalloc N1=2 | |
1228 | 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... | |
1229 | pages=10 vmalloc N0=10 | |
1230 | ||
1231 | ||
1232 | softirqs | |
1233 | ~~~~~~~~ | |
d3d64df2 | 1234 | |
059db434 | 1235 | Provides counts of softirq handlers serviced since boot time, for each CPU. |
d3d64df2 | 1236 | |
c33e97ef MCC |
1237 | :: |
1238 | ||
1239 | > cat /proc/softirqs | |
059db434 | 1240 | CPU0 CPU1 CPU2 CPU3 |
c33e97ef | 1241 | HI: 0 0 0 0 |
059db434 | 1242 | TIMER: 27166 27120 27097 27034 |
c33e97ef MCC |
1243 | NET_TX: 0 0 0 17 |
1244 | NET_RX: 42 0 0 39 | |
059db434 RD |
1245 | BLOCK: 0 0 107 1121 |
1246 | TASKLET: 0 0 0 290 | |
1247 | SCHED: 27035 26983 26971 26746 | |
1248 | HRTIMER: 0 0 0 0 | |
1249 | RCU: 1678 1769 2178 2250 | |
d3d64df2 | 1250 | |
e24ccaaf | 1251 | 1.3 Networking info in /proc/net |
1da177e4 LT |
1252 | -------------------------------- |
1253 | ||
349888ee | 1254 | The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the |
1da177e4 | 1255 | additional values you get for IP version 6 if you configure the kernel to |
349888ee | 1256 | support this. Table 1-9 lists the files and their meaning. |
1da177e4 LT |
1257 | |
1258 | ||
c33e97ef MCC |
1259 | .. table:: Table 1-8: IPv6 info in /proc/net |
1260 | ||
1261 | ========== ===================================================== | |
1262 | File Content | |
1263 | ========== ===================================================== | |
1264 | udp6 UDP sockets (IPv6) | |
1265 | tcp6 TCP sockets (IPv6) | |
1266 | raw6 Raw device statistics (IPv6) | |
1267 | igmp6 IP multicast addresses, which this host joined (IPv6) | |
1268 | if_inet6 List of IPv6 interface addresses | |
1269 | ipv6_route Kernel routing table for IPv6 | |
1270 | rt6_stats Global IPv6 routing tables statistics | |
1271 | sockstat6 Socket statistics (IPv6) | |
1272 | snmp6 Snmp data (IPv6) | |
1273 | ========== ===================================================== | |
1274 | ||
1275 | .. table:: Table 1-9: Network info in /proc/net | |
1276 | ||
1277 | ============= ================================================================ | |
1278 | File Content | |
1279 | ============= ================================================================ | |
1280 | arp Kernel ARP table | |
1281 | dev network devices with statistics | |
1da177e4 LT |
1282 | dev_mcast the Layer2 multicast groups a device is listening too |
1283 | (interface index, label, number of references, number of bound | |
c33e97ef MCC |
1284 | addresses). |
1285 | dev_stat network device status | |
1286 | ip_fwchains Firewall chain linkage | |
1287 | ip_fwnames Firewall chain names | |
1288 | ip_masq Directory containing the masquerading tables | |
1289 | ip_masquerade Major masquerading table | |
1290 | netstat Network statistics | |
1291 | raw raw device statistics | |
1292 | route Kernel routing table | |
1293 | rpc Directory containing rpc info | |
1294 | rt_cache Routing cache | |
1295 | snmp SNMP data | |
1296 | sockstat Socket statistics | |
8b0a211d | 1297 | softnet_stat Per-CPU incoming packets queues statistics of online CPUs |
c33e97ef MCC |
1298 | tcp TCP sockets |
1299 | udp UDP sockets | |
1300 | unix UNIX domain sockets | |
1301 | wireless Wireless interface data (Wavelan etc) | |
1302 | igmp IP multicast addresses, which this host joined | |
1303 | psched Global packet scheduler parameters. | |
1304 | netlink List of PF_NETLINK sockets | |
1305 | ip_mr_vifs List of multicast virtual interfaces | |
1306 | ip_mr_cache List of multicast routing cache | |
1307 | ============= ================================================================ | |
1da177e4 LT |
1308 | |
1309 | You can use this information to see which network devices are available in | |
c33e97ef MCC |
1310 | your system and how much traffic was routed over those devices:: |
1311 | ||
1312 | > cat /proc/net/dev | |
1313 | Inter-|Receive |[... | |
1314 | face |bytes packets errs drop fifo frame compressed multicast|[... | |
1315 | lo: 908188 5596 0 0 0 0 0 0 [... | |
1316 | ppp0:15475140 20721 410 0 0 410 0 0 [... | |
1317 | eth0: 614530 7085 0 0 0 0 0 1 [... | |
1318 | ||
1319 | ...] Transmit | |
1320 | ...] bytes packets errs drop fifo colls carrier compressed | |
1321 | ...] 908188 5596 0 0 0 0 0 0 | |
1322 | ...] 1375103 17405 0 0 0 0 0 0 | |
1323 | ...] 1703981 5535 0 0 0 3 0 0 | |
1da177e4 | 1324 | |
a33f3224 | 1325 | In addition, each Channel Bond interface has its own directory. For |
1da177e4 LT |
1326 | example, the bond0 device will have a directory called /proc/net/bond0/. |
1327 | It will contain information that is specific to that bond, such as the | |
1328 | current slaves of the bond, the link status of the slaves, and how | |
1329 | many times the slaves link has failed. | |
1330 | ||
e24ccaaf | 1331 | 1.4 SCSI info |
1da177e4 LT |
1332 | ------------- |
1333 | ||
d2ea66a6 RD |
1334 | If you have a SCSI or ATA host adapter in your system, you'll find a |
1335 | subdirectory named after the driver for this adapter in /proc/scsi. | |
1336 | You'll also see a list of all recognized SCSI devices in /proc/scsi:: | |
1da177e4 | 1337 | |
c33e97ef MCC |
1338 | >cat /proc/scsi/scsi |
1339 | Attached devices: | |
1340 | Host: scsi0 Channel: 00 Id: 00 Lun: 00 | |
1341 | Vendor: IBM Model: DGHS09U Rev: 03E0 | |
1342 | Type: Direct-Access ANSI SCSI revision: 03 | |
1343 | Host: scsi0 Channel: 00 Id: 06 Lun: 00 | |
1344 | Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 | |
1345 | Type: CD-ROM ANSI SCSI revision: 02 | |
1da177e4 LT |
1346 | |
1347 | ||
1348 | The directory named after the driver has one file for each adapter found in | |
1349 | the system. These files contain information about the controller, including | |
1350 | the used IRQ and the IO address range. The amount of information shown is | |
1351 | dependent on the adapter you use. The example shows the output for an Adaptec | |
c33e97ef MCC |
1352 | AHA-2940 SCSI adapter:: |
1353 | ||
1354 | > cat /proc/scsi/aic7xxx/0 | |
1355 | ||
1356 | Adaptec AIC7xxx driver version: 5.1.19/3.2.4 | |
1357 | Compile Options: | |
1358 | TCQ Enabled By Default : Disabled | |
1359 | AIC7XXX_PROC_STATS : Disabled | |
1360 | AIC7XXX_RESET_DELAY : 5 | |
1361 | Adapter Configuration: | |
1362 | SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter | |
1363 | Ultra Wide Controller | |
1364 | PCI MMAPed I/O Base: 0xeb001000 | |
1365 | Adapter SEEPROM Config: SEEPROM found and used. | |
1366 | Adaptec SCSI BIOS: Enabled | |
1367 | IRQ: 10 | |
1368 | SCBs: Active 0, Max Active 2, | |
1369 | Allocated 15, HW 16, Page 255 | |
1370 | Interrupts: 160328 | |
1371 | BIOS Control Word: 0x18b6 | |
1372 | Adapter Control Word: 0x005b | |
1373 | Extended Translation: Enabled | |
1374 | Disconnect Enable Flags: 0xffff | |
1375 | Ultra Enable Flags: 0x0001 | |
1376 | Tag Queue Enable Flags: 0x0000 | |
1377 | Ordered Queue Tag Flags: 0x0000 | |
1378 | Default Tag Queue Depth: 8 | |
1379 | Tagged Queue By Device array for aic7xxx host instance 0: | |
1380 | {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} | |
1381 | Actual queue depth per device for aic7xxx host instance 0: | |
1382 | {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} | |
1383 | Statistics: | |
1384 | (scsi0:0:0:0) | |
1385 | Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 | |
1386 | Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) | |
1387 | Total transfers 160151 (74577 reads and 85574 writes) | |
1388 | (scsi0:0:6:0) | |
1389 | Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 | |
1390 | Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) | |
1391 | Total transfers 0 (0 reads and 0 writes) | |
1da177e4 LT |
1392 | |
1393 | ||
e24ccaaf | 1394 | 1.5 Parallel port info in /proc/parport |
1da177e4 LT |
1395 | --------------------------------------- |
1396 | ||
1397 | The directory /proc/parport contains information about the parallel ports of | |
1398 | your system. It has one subdirectory for each port, named after the port | |
1399 | number (0,1,2,...). | |
1400 | ||
349888ee | 1401 | These directories contain the four files shown in Table 1-10. |
1da177e4 LT |
1402 | |
1403 | ||
c33e97ef MCC |
1404 | .. table:: Table 1-10: Files in /proc/parport |
1405 | ||
1406 | ========= ==================================================================== | |
1407 | File Content | |
1408 | ========= ==================================================================== | |
1409 | autoprobe Any IEEE-1284 device ID information that has been acquired. | |
1da177e4 LT |
1410 | devices list of the device drivers using that port. A + will appear by the |
1411 | name of the device currently using the port (it might not appear | |
c33e97ef MCC |
1412 | against any). |
1413 | hardware Parallel port's base address, IRQ line and DMA channel. | |
1da177e4 LT |
1414 | irq IRQ that parport is using for that port. This is in a separate |
1415 | file to allow you to alter it by writing a new value in (IRQ | |
c33e97ef MCC |
1416 | number or none). |
1417 | ========= ==================================================================== | |
1da177e4 | 1418 | |
e24ccaaf | 1419 | 1.6 TTY info in /proc/tty |
1da177e4 LT |
1420 | ------------------------- |
1421 | ||
1422 | Information about the available and actually used tty's can be found in the | |
059db434 | 1423 | directory /proc/tty. You'll find entries for drivers and line disciplines in |
349888ee | 1424 | this directory, as shown in Table 1-11. |
1da177e4 LT |
1425 | |
1426 | ||
c33e97ef MCC |
1427 | .. table:: Table 1-11: Files in /proc/tty |
1428 | ||
1429 | ============= ============================================== | |
1430 | File Content | |
1431 | ============= ============================================== | |
1432 | drivers list of drivers and their usage | |
1433 | ldiscs registered line disciplines | |
1434 | driver/serial usage statistic and status of single tty lines | |
1435 | ============= ============================================== | |
1da177e4 LT |
1436 | |
1437 | To see which tty's are currently in use, you can simply look into the file | |
c33e97ef MCC |
1438 | /proc/tty/drivers:: |
1439 | ||
1440 | > cat /proc/tty/drivers | |
1441 | pty_slave /dev/pts 136 0-255 pty:slave | |
1442 | pty_master /dev/ptm 128 0-255 pty:master | |
1443 | pty_slave /dev/ttyp 3 0-255 pty:slave | |
1444 | pty_master /dev/pty 2 0-255 pty:master | |
1445 | serial /dev/cua 5 64-67 serial:callout | |
1446 | serial /dev/ttyS 4 64-67 serial | |
1447 | /dev/tty0 /dev/tty0 4 0 system:vtmaster | |
1448 | /dev/ptmx /dev/ptmx 5 2 system | |
1449 | /dev/console /dev/console 5 1 system:console | |
1450 | /dev/tty /dev/tty 5 0 system:/dev/tty | |
1451 | unknown /dev/tty 4 1-63 console | |
1da177e4 LT |
1452 | |
1453 | ||
e24ccaaf | 1454 | 1.7 Miscellaneous kernel statistics in /proc/stat |
1da177e4 LT |
1455 | ------------------------------------------------- |
1456 | ||
1457 | Various pieces of information about kernel activity are available in the | |
1458 | /proc/stat file. All of the numbers reported in this file are aggregates | |
c33e97ef | 1459 | since the system first booted. For a quick look, simply cat the file:: |
1da177e4 LT |
1460 | |
1461 | > cat /proc/stat | |
d2ea66a6 RD |
1462 | cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0 |
1463 | cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0 | |
1464 | cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0 | |
1465 | cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0 | |
1466 | cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0 | |
1467 | intr 8688370575 8 3373 0 0 0 0 0 0 1 40791 0 0 353317 0 0 0 0 224789828 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 190974333 41958554 123983334 43 0 224593 0 0 0 <more 0's deleted> | |
1468 | ctxt 22848221062 | |
1469 | btime 1605316999 | |
1470 | processes 746787147 | |
1471 | procs_running 2 | |
1da177e4 | 1472 | procs_blocked 0 |
d2ea66a6 | 1473 | softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354 |
1da177e4 LT |
1474 | |
1475 | The very first "cpu" line aggregates the numbers in all of the other "cpuN" | |
1476 | lines. These numbers identify the amount of time the CPU has spent performing | |
1477 | different kinds of work. Time units are in USER_HZ (typically hundredths of a | |
1478 | second). The meanings of the columns are as follows, from left to right: | |
1479 | ||
1480 | - user: normal processes executing in user mode | |
1481 | - nice: niced processes executing in user mode | |
1482 | - system: processes executing in kernel mode | |
1483 | - idle: twiddling thumbs | |
9c240d75 CF |
1484 | - iowait: In a word, iowait stands for waiting for I/O to complete. But there |
1485 | are several problems: | |
c33e97ef | 1486 | |
059db434 RD |
1487 | 1. CPU will not wait for I/O to complete, iowait is the time that a task is |
1488 | waiting for I/O to complete. When CPU goes into idle state for | |
1489 | outstanding task I/O, another task will be scheduled on this CPU. | |
9c240d75 CF |
1490 | 2. In a multi-core CPU, the task waiting for I/O to complete is not running |
1491 | on any CPU, so the iowait of each CPU is difficult to calculate. | |
1492 | 3. The value of iowait field in /proc/stat will decrease in certain | |
1493 | conditions. | |
c33e97ef | 1494 | |
9c240d75 | 1495 | So, the iowait is not reliable by reading from /proc/stat. |
1da177e4 LT |
1496 | - irq: servicing interrupts |
1497 | - softirq: servicing softirqs | |
b68f2c3a | 1498 | - steal: involuntary wait |
ce0e7b28 RO |
1499 | - guest: running a normal guest |
1500 | - guest_nice: running a niced guest | |
1da177e4 LT |
1501 | |
1502 | The "intr" line gives counts of interrupts serviced since boot time, for each | |
1503 | of the possible system interrupts. The first column is the total of all | |
3568a1db JMM |
1504 | interrupts serviced including unnumbered architecture specific interrupts; |
1505 | each subsequent column is the total for that particular numbered interrupt. | |
1506 | Unnumbered interrupts are not shown, only summed into the total. | |
1da177e4 LT |
1507 | |
1508 | The "ctxt" line gives the total number of context switches across all CPUs. | |
1509 | ||
1510 | The "btime" line gives the time at which the system booted, in seconds since | |
1511 | the Unix epoch. | |
1512 | ||
1513 | The "processes" line gives the number of processes and threads created, which | |
1514 | includes (but is not limited to) those created by calls to the fork() and | |
1515 | clone() system calls. | |
1516 | ||
e3cc2226 LGE |
1517 | The "procs_running" line gives the total number of threads that are |
1518 | running or ready to run (i.e., the total number of runnable threads). | |
1da177e4 LT |
1519 | |
1520 | The "procs_blocked" line gives the number of processes currently blocked, | |
1521 | waiting for I/O to complete. | |
1522 | ||
d3d64df2 KK |
1523 | The "softirq" line gives counts of softirqs serviced since boot time, for each |
1524 | of the possible system softirqs. The first column is the total of all | |
1525 | softirqs serviced; each subsequent column is the total for that particular | |
1526 | softirq. | |
1527 | ||
37515fac | 1528 | |
e24ccaaf | 1529 | 1.8 Ext4 file system parameters |
690b0543 | 1530 | ------------------------------- |
37515fac TT |
1531 | |
1532 | Information about mounted ext4 file systems can be found in | |
1533 | /proc/fs/ext4. Each mounted filesystem will have a directory in | |
1534 | /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or | |
d2ea66a6 RD |
1535 | /proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device |
1536 | directory are shown in Table 1-12, below. | |
37515fac | 1537 | |
c33e97ef MCC |
1538 | .. table:: Table 1-12: Files in /proc/fs/ext4/<devname> |
1539 | ||
1540 | ============== ========================================================== | |
1541 | File Content | |
37515fac | 1542 | mb_groups details of multiblock allocator buddy cache of free blocks |
c33e97ef | 1543 | ============== ========================================================== |
37515fac | 1544 | |
e24ccaaf | 1545 | 1.9 /proc/consoles |
059db434 | 1546 | ------------------- |
23308ba5 JS |
1547 | Shows registered system console lines. |
1548 | ||
1549 | To see which character device lines are currently used for the system console | |
c33e97ef | 1550 | /dev/console, you may simply look into the file /proc/consoles:: |
23308ba5 JS |
1551 | |
1552 | > cat /proc/consoles | |
1553 | tty0 -WU (ECp) 4:7 | |
1554 | ttyS0 -W- (Ep) 4:64 | |
1555 | ||
1556 | The columns are: | |
1557 | ||
c33e97ef MCC |
1558 | +--------------------+-------------------------------------------------------+ |
1559 | | device | name of the device | | |
1560 | +====================+=======================================================+ | |
1561 | | operations | * R = can do read operations | | |
1562 | | | * W = can do write operations | | |
1563 | | | * U = can do unblank | | |
1564 | +--------------------+-------------------------------------------------------+ | |
1565 | | flags | * E = it is enabled | | |
1566 | | | * C = it is preferred console | | |
1567 | | | * B = it is primary boot console | | |
1568 | | | * p = it is used for printk buffer | | |
1569 | | | * b = it is not a TTY but a Braille device | | |
1570 | | | * a = it is safe to use when cpu is offline | | |
1571 | +--------------------+-------------------------------------------------------+ | |
1572 | | major:minor | major and minor number of the device separated by a | | |
1573 | | | colon | | |
1574 | +--------------------+-------------------------------------------------------+ | |
1da177e4 | 1575 | |
1da177e4 | 1576 | Summary |
c33e97ef MCC |
1577 | ------- |
1578 | ||
1da177e4 LT |
1579 | The /proc file system serves information about the running system. It not only |
1580 | allows access to process data but also allows you to request the kernel status | |
1581 | by reading files in the hierarchy. | |
1582 | ||
1583 | The directory structure of /proc reflects the types of information and makes | |
1584 | it easy, if not obvious, where to look for specific data. | |
1da177e4 | 1585 | |
c33e97ef MCC |
1586 | Chapter 2: Modifying System Parameters |
1587 | ====================================== | |
1da177e4 | 1588 | |
1da177e4 | 1589 | In This Chapter |
c33e97ef MCC |
1590 | --------------- |
1591 | ||
1da177e4 LT |
1592 | * Modifying kernel parameters by writing into files found in /proc/sys |
1593 | * Exploring the files which modify certain parameters | |
1594 | * Review of the /proc/sys file tree | |
1da177e4 | 1595 | |
c33e97ef | 1596 | ------------------------------------------------------------------------------ |
1da177e4 LT |
1597 | |
1598 | A very interesting part of /proc is the directory /proc/sys. This is not only | |
1599 | a source of information, it also allows you to change parameters within the | |
1600 | kernel. Be very careful when attempting this. You can optimize your system, | |
1601 | but you can also cause it to crash. Never alter kernel parameters on a | |
1602 | production system. Set up a development machine and test to make sure that | |
1603 | everything works the way you want it to. You may have no alternative but to | |
1604 | reboot the machine once an error has been made. | |
1605 | ||
059db434 RD |
1606 | To change a value, simply echo the new value into the file. |
1607 | You need to be root to do this. You can create your own boot script | |
1608 | to perform this every time your system boots. | |
1da177e4 LT |
1609 | |
1610 | The files in /proc/sys can be used to fine tune and monitor miscellaneous and | |
1611 | general things in the operation of the Linux kernel. Since some of the files | |
1612 | can inadvertently disrupt your system, it is advisable to read both | |
1613 | documentation and source before actually making adjustments. In any case, be | |
1614 | very careful when writing to any of these files. The entries in /proc may | |
1615 | change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt | |
d2ea66a6 | 1616 | review the kernel documentation in the directory linux/Documentation. |
1da177e4 LT |
1617 | This chapter is heavily based on the documentation included in the pre 2.2 |
1618 | kernels, and became part of it in version 2.2.1 of the Linux kernel. | |
1619 | ||
d2ea66a6 RD |
1620 | Please see: Documentation/admin-guide/sysctl/ directory for descriptions of |
1621 | these entries. | |
9d0243bc | 1622 | |
760df93e | 1623 | Summary |
c33e97ef MCC |
1624 | ------- |
1625 | ||
760df93e SF |
1626 | Certain aspects of kernel behavior can be modified at runtime, without the |
1627 | need to recompile the kernel, or even to reboot the system. The files in the | |
1628 | /proc/sys tree can not only be read, but also modified. You can use the echo | |
1629 | command to write value into these files, thereby changing the default settings | |
1630 | of the kernel. | |
9d0243bc | 1631 | |
c33e97ef MCC |
1632 | |
1633 | Chapter 3: Per-process Parameters | |
1634 | ================================= | |
1da177e4 | 1635 | |
fa0cbbf1 | 1636 | 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score |
a63d83f4 DR |
1637 | -------------------------------------------------------------------------------- |
1638 | ||
059db434 RD |
1639 | These files can be used to adjust the badness heuristic used to select which |
1640 | process gets killed in out of memory (oom) conditions. | |
a63d83f4 DR |
1641 | |
1642 | The badness heuristic assigns a value to each candidate task ranging from 0 | |
1643 | (never kill) to 1000 (always kill) to determine which process is targeted. The | |
1644 | units are roughly a proportion along that range of allowed memory the process | |
1645 | may allocate from based on an estimation of its current memory and swap use. | |
1646 | For example, if a task is using all allowed memory, its badness score will be | |
1647 | 1000. If it is using half of its allowed memory, its score will be 500. | |
1648 | ||
a63d83f4 DR |
1649 | The amount of "allowed" memory depends on the context in which the oom killer |
1650 | was called. If it is due to the memory assigned to the allocating task's cpuset | |
1651 | being exhausted, the allowed memory represents the set of mems assigned to that | |
1652 | cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed | |
1653 | memory represents the set of mempolicy nodes. If it is due to a memory | |
1654 | limit (or swap limit) being reached, the allowed memory is that configured | |
1655 | limit. Finally, if it is due to the entire system being out of memory, the | |
1656 | allowed memory represents all allocatable resources. | |
1657 | ||
1658 | The value of /proc/<pid>/oom_score_adj is added to the badness score before it | |
1659 | is used to determine which task to kill. Acceptable values range from -1000 | |
1660 | (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to | |
1661 | polarize the preference for oom killing either by always preferring a certain | |
1662 | task or completely disabling it. The lowest possible value, -1000, is | |
1663 | equivalent to disabling oom killing entirely for that task since it will always | |
1664 | report a badness score of 0. | |
1665 | ||
1666 | Consequently, it is very simple for userspace to define the amount of memory to | |
1667 | consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for | |
1668 | example, is roughly equivalent to allowing the remainder of tasks sharing the | |
1669 | same system, cpuset, mempolicy, or memory controller resources to use at least | |
1670 | 50% more memory. A value of -500, on the other hand, would be roughly | |
1671 | equivalent to discounting 50% of the task's allowed memory from being considered | |
1672 | as scoring against the task. | |
1673 | ||
fa0cbbf1 DR |
1674 | For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also |
1675 | be used to tune the badness score. Its acceptable values range from -16 | |
1676 | (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 | |
1677 | (OOM_DISABLE) to disable oom killing entirely for that task. Its value is | |
1678 | scaled linearly with /proc/<pid>/oom_score_adj. | |
1679 | ||
dabb16f6 MSB |
1680 | The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last |
1681 | value set by a CAP_SYS_RESOURCE process. To reduce the value any lower | |
1682 | requires CAP_SYS_RESOURCE. | |
1683 | ||
9e9e3cbc | 1684 | |
760df93e | 1685 | 3.2 /proc/<pid>/oom_score - Display current oom-killer score |
d7ff0dbf JFM |
1686 | ------------------------------------------------------------- |
1687 | ||
059db434 | 1688 | This file can be used to check the current score used by the oom-killer for |
fa0cbbf1 DR |
1689 | any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which |
1690 | process should be killed in an out-of-memory situation. | |
1691 | ||
b1aa7c93 MH |
1692 | Please note that the exported value includes oom_score_adj so it is |
1693 | effectively in range [0,2000]. | |
1694 | ||
f9c99463 | 1695 | |
760df93e | 1696 | 3.3 /proc/<pid>/io - Display the IO accounting fields |
f9c99463 RK |
1697 | ------------------------------------------------------- |
1698 | ||
059db434 | 1699 | This file contains IO statistics for each running process. |
f9c99463 RK |
1700 | |
1701 | Example | |
c33e97ef MCC |
1702 | ~~~~~~~ |
1703 | ||
1704 | :: | |
f9c99463 | 1705 | |
c33e97ef MCC |
1706 | test:/tmp # dd if=/dev/zero of=/tmp/test.dat & |
1707 | [1] 3828 | |
f9c99463 | 1708 | |
c33e97ef MCC |
1709 | test:/tmp # cat /proc/3828/io |
1710 | rchar: 323934931 | |
1711 | wchar: 323929600 | |
1712 | syscr: 632687 | |
1713 | syscw: 632675 | |
1714 | read_bytes: 0 | |
1715 | write_bytes: 323932160 | |
1716 | cancelled_write_bytes: 0 | |
f9c99463 RK |
1717 | |
1718 | ||
1719 | Description | |
c33e97ef | 1720 | ~~~~~~~~~~~ |
f9c99463 RK |
1721 | |
1722 | rchar | |
c33e97ef | 1723 | ^^^^^ |
f9c99463 RK |
1724 | |
1725 | I/O counter: chars read | |
1726 | The number of bytes which this task has caused to be read from storage. This | |
1727 | is simply the sum of bytes which this process passed to read() and pread(). | |
1728 | It includes things like tty IO and it is unaffected by whether or not actual | |
1729 | physical disk IO was required (the read might have been satisfied from | |
059db434 | 1730 | pagecache). |
f9c99463 RK |
1731 | |
1732 | ||
1733 | wchar | |
c33e97ef | 1734 | ^^^^^ |
f9c99463 RK |
1735 | |
1736 | I/O counter: chars written | |
1737 | The number of bytes which this task has caused, or shall cause to be written | |
1738 | to disk. Similar caveats apply here as with rchar. | |
1739 | ||
1740 | ||
1741 | syscr | |
c33e97ef | 1742 | ^^^^^ |
f9c99463 RK |
1743 | |
1744 | I/O counter: read syscalls | |
1745 | Attempt to count the number of read I/O operations, i.e. syscalls like read() | |
1746 | and pread(). | |
1747 | ||
1748 | ||
1749 | syscw | |
c33e97ef | 1750 | ^^^^^ |
f9c99463 RK |
1751 | |
1752 | I/O counter: write syscalls | |
1753 | Attempt to count the number of write I/O operations, i.e. syscalls like | |
1754 | write() and pwrite(). | |
1755 | ||
1756 | ||
1757 | read_bytes | |
c33e97ef | 1758 | ^^^^^^^^^^ |
f9c99463 RK |
1759 | |
1760 | I/O counter: bytes read | |
1761 | Attempt to count the number of bytes which this process really did cause to | |
1762 | be fetched from the storage layer. Done at the submit_bio() level, so it is | |
1763 | accurate for block-backed filesystems. <please add status regarding NFS and | |
1764 | CIFS at a later time> | |
1765 | ||
1766 | ||
1767 | write_bytes | |
c33e97ef | 1768 | ^^^^^^^^^^^ |
f9c99463 RK |
1769 | |
1770 | I/O counter: bytes written | |
1771 | Attempt to count the number of bytes which this process caused to be sent to | |
1772 | the storage layer. This is done at page-dirtying time. | |
1773 | ||
1774 | ||
1775 | cancelled_write_bytes | |
c33e97ef | 1776 | ^^^^^^^^^^^^^^^^^^^^^ |
f9c99463 RK |
1777 | |
1778 | The big inaccuracy here is truncate. If a process writes 1MB to a file and | |
1779 | then deletes the file, it will in fact perform no writeout. But it will have | |
1780 | been accounted as having caused 1MB of write. | |
1781 | In other words: The number of bytes which this process caused to not happen, | |
1782 | by truncating pagecache. A task can cause "negative" IO too. If this task | |
1783 | truncates some dirty pagecache, some IO which another task has been accounted | |
a33f3224 | 1784 | for (in its write_bytes) will not be happening. We _could_ just subtract that |
f9c99463 RK |
1785 | from the truncating task's write_bytes, but there is information loss in doing |
1786 | that. | |
1787 | ||
1788 | ||
c33e97ef | 1789 | .. Note:: |
f9c99463 | 1790 | |
c33e97ef MCC |
1791 | At its current implementation state, this is a bit racy on 32-bit machines: |
1792 | if process A reads process B's /proc/pid/io while process B is updating one | |
1793 | of those 64-bit counters, process A could see an intermediate result. | |
f9c99463 RK |
1794 | |
1795 | ||
1796 | More information about this can be found within the taskstats documentation in | |
1797 | Documentation/accounting. | |
1798 | ||
760df93e | 1799 | 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings |
bb90110d KH |
1800 | --------------------------------------------------------------- |
1801 | When a process is dumped, all anonymous memory is written to a core file as | |
1802 | long as the size of the core file isn't limited. But sometimes we don't want | |
5037835c RZ |
1803 | to dump some memory segments, for example, huge shared memory or DAX. |
1804 | Conversely, sometimes we want to save file-backed memory segments into a core | |
1805 | file, not only the individual files. | |
bb90110d KH |
1806 | |
1807 | /proc/<pid>/coredump_filter allows you to customize which memory segments | |
1808 | will be dumped when the <pid> process is dumped. coredump_filter is a bitmask | |
1809 | of memory types. If a bit of the bitmask is set, memory segments of the | |
1810 | corresponding memory type are dumped, otherwise they are not dumped. | |
1811 | ||
5037835c | 1812 | The following 9 memory types are supported: |
c33e97ef | 1813 | |
bb90110d KH |
1814 | - (bit 0) anonymous private memory |
1815 | - (bit 1) anonymous shared memory | |
1816 | - (bit 2) file-backed private memory | |
1817 | - (bit 3) file-backed shared memory | |
b261dfea | 1818 | - (bit 4) ELF header pages in file-backed private memory areas (it is |
c33e97ef | 1819 | effective only if the bit 2 is cleared) |
e575f111 KM |
1820 | - (bit 5) hugetlb private memory |
1821 | - (bit 6) hugetlb shared memory | |
5037835c RZ |
1822 | - (bit 7) DAX private memory |
1823 | - (bit 8) DAX shared memory | |
bb90110d KH |
1824 | |
1825 | Note that MMIO pages such as frame buffer are never dumped and vDSO pages | |
1826 | are always dumped regardless of the bitmask status. | |
1827 | ||
5037835c RZ |
1828 | Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is |
1829 | only affected by bit 5-6, and DAX is only affected by bits 7-8. | |
e575f111 | 1830 | |
5037835c RZ |
1831 | The default value of coredump_filter is 0x33; this means all anonymous memory |
1832 | segments, ELF header pages and hugetlb private memory are dumped. | |
bb90110d KH |
1833 | |
1834 | If you don't want to dump all shared memory segments attached to pid 1234, | |
c33e97ef | 1835 | write 0x31 to the process's proc file:: |
bb90110d | 1836 | |
5037835c | 1837 | $ echo 0x31 > /proc/1234/coredump_filter |
bb90110d KH |
1838 | |
1839 | When a new process is created, the process inherits the bitmask status from its | |
1840 | parent. It is useful to set up coredump_filter before the program runs. | |
c33e97ef | 1841 | For example:: |
bb90110d KH |
1842 | |
1843 | $ echo 0x7 > /proc/self/coredump_filter | |
1844 | $ ./some_program | |
1845 | ||
760df93e | 1846 | 3.5 /proc/<pid>/mountinfo - Information about mounts |
2d4d4864 RP |
1847 | -------------------------------------------------------- |
1848 | ||
c33e97ef | 1849 | This file contains lines of the form:: |
2d4d4864 | 1850 | |
c33e97ef | 1851 | 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue |
ff9c3d43 | 1852 | (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4) |
2d4d4864 | 1853 | |
b0b719ce CAM |
1854 | (1) mount ID: unique identifier of the mount (may be reused after umount) |
1855 | (2) parent ID: ID of parent (or of self for the top of the mount tree) | |
1856 | (3) major:minor: value of st_dev for files on filesystem | |
1857 | (4) root: root of the mount within the filesystem | |
1858 | (5) mount point: mount point relative to the process's root | |
1859 | (6) mount options: per mount options | |
1860 | (n…m) optional fields: zero or more fields of the form "tag[:value]" | |
1861 | (m+1) separator: marks the end of the optional fields | |
1862 | (m+2) filesystem type: name of filesystem of the form "type[.subtype]" | |
1863 | (m+3) mount source: filesystem specific information or "none" | |
1864 | (m+4) super options: per super block options | |
2d4d4864 RP |
1865 | |
1866 | Parsers should ignore all unrecognised optional fields. Currently the | |
1867 | possible optional fields are: | |
1868 | ||
c33e97ef MCC |
1869 | ================ ============================================================== |
1870 | shared:X mount is shared in peer group X | |
1871 | master:X mount is slave to peer group X | |
1872 | propagate_from:X mount is slave and receives propagation from peer group X [#]_ | |
1873 | unbindable mount is unbindable | |
1874 | ================ ============================================================== | |
2d4d4864 | 1875 | |
c33e97ef MCC |
1876 | .. [#] X is the closest dominant peer group under the process's root. If |
1877 | X is the immediate master of the mount, or if there's no dominant peer | |
1878 | group under the same root, then only the "master:X" field is present | |
1879 | and not the "propagate_from:X" field. | |
97e7e0f7 | 1880 | |
2d4d4864 RP |
1881 | For more information on mount propagation see: |
1882 | ||
cf06612c | 1883 | Documentation/filesystems/sharedsubtree.rst |
2d4d4864 | 1884 | |
4614a696 | 1885 | |
1886 | 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm | |
1887 | -------------------------------------------------------- | |
059db434 | 1888 | These files provide a method to access a task's comm value. It also allows for |
4614a696 | 1889 | a task to set its own or one of its thread siblings comm value. The comm value |
1890 | is limited in size compared to the cmdline value, so writing anything longer | |
1891 | then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated | |
1892 | comm value. | |
0499680a VK |
1893 | |
1894 | ||
81841161 CG |
1895 | 3.7 /proc/<pid>/task/<tid>/children - Information about task children |
1896 | ------------------------------------------------------------------------- | |
1897 | This file provides a fast way to retrieve first level children pids | |
1898 | of a task pointed by <pid>/<tid> pair. The format is a space separated | |
1899 | stream of pids. | |
1900 | ||
059db434 RD |
1901 | Note the "first level" here -- if a child has its own children they will |
1902 | not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children | |
81841161 CG |
1903 | to obtain the descendants. |
1904 | ||
1905 | Since this interface is intended to be fast and cheap it doesn't | |
1906 | guarantee to provide precise results and some children might be | |
1907 | skipped, especially if they've exited right after we printed their | |
059db434 | 1908 | pids, so one needs to either stop or freeze processes being inspected |
81841161 CG |
1909 | if precise results are needed. |
1910 | ||
1911 | ||
49d063cb | 1912 | 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file |
f1d8c162 CG |
1913 | --------------------------------------------------------------- |
1914 | This file provides information associated with an opened file. The regular | |
3845f256 KS |
1915 | files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. |
1916 | The 'pos' represents the current offset of the opened file in decimal | |
1917 | form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the | |
1918 | file has been created with [see open(2) for details] and 'mnt_id' represents | |
1919 | mount ID of the file system containing the opened file [see 3.5 | |
1920 | /proc/<pid>/mountinfo for details]. 'ino' represents the inode number of | |
1921 | the file. | |
f1d8c162 | 1922 | |
c33e97ef | 1923 | A typical output is:: |
f1d8c162 CG |
1924 | |
1925 | pos: 0 | |
1926 | flags: 0100002 | |
49d063cb | 1927 | mnt_id: 19 |
3845f256 | 1928 | ino: 63107 |
f1d8c162 | 1929 | |
c33e97ef | 1930 | All locks associated with a file descriptor are shown in its fdinfo too:: |
6c8c9031 | 1931 | |
c33e97ef | 1932 | lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF |
6c8c9031 | 1933 | |
f1d8c162 CG |
1934 | The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags |
1935 | pair provide additional information particular to the objects they represent. | |
1936 | ||
c33e97ef MCC |
1937 | Eventfd files |
1938 | ~~~~~~~~~~~~~ | |
1939 | ||
1940 | :: | |
1941 | ||
f1d8c162 CG |
1942 | pos: 0 |
1943 | flags: 04002 | |
49d063cb | 1944 | mnt_id: 9 |
3845f256 | 1945 | ino: 63107 |
f1d8c162 CG |
1946 | eventfd-count: 5a |
1947 | ||
c33e97ef MCC |
1948 | where 'eventfd-count' is hex value of a counter. |
1949 | ||
1950 | Signalfd files | |
1951 | ~~~~~~~~~~~~~~ | |
1952 | ||
1953 | :: | |
f1d8c162 | 1954 | |
f1d8c162 CG |
1955 | pos: 0 |
1956 | flags: 04002 | |
49d063cb | 1957 | mnt_id: 9 |
3845f256 | 1958 | ino: 63107 |
f1d8c162 CG |
1959 | sigmask: 0000000000000200 |
1960 | ||
c33e97ef MCC |
1961 | where 'sigmask' is hex value of the signal mask associated |
1962 | with a file. | |
1963 | ||
1964 | Epoll files | |
1965 | ~~~~~~~~~~~ | |
1966 | ||
1967 | :: | |
f1d8c162 | 1968 | |
f1d8c162 CG |
1969 | pos: 0 |
1970 | flags: 02 | |
49d063cb | 1971 | mnt_id: 9 |
3845f256 | 1972 | ino: 63107 |
77493f04 | 1973 | tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 |
f1d8c162 | 1974 | |
c33e97ef MCC |
1975 | where 'tfd' is a target file descriptor number in decimal form, |
1976 | 'events' is events mask being watched and the 'data' is data | |
1977 | associated with a target [see epoll(7) for more details]. | |
f1d8c162 | 1978 | |
c33e97ef MCC |
1979 | The 'pos' is current offset of the target file in decimal form |
1980 | [see lseek(2)], 'ino' and 'sdev' are inode and device numbers | |
1981 | where target file resides, all in hex format. | |
77493f04 | 1982 | |
c33e97ef MCC |
1983 | Fsnotify files |
1984 | ~~~~~~~~~~~~~~ | |
1985 | For inotify files the format is the following:: | |
f1d8c162 CG |
1986 | |
1987 | pos: 0 | |
1988 | flags: 02000000 | |
3845f256 KS |
1989 | mnt_id: 9 |
1990 | ino: 63107 | |
f1d8c162 CG |
1991 | inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d |
1992 | ||
059db434 | 1993 | where 'wd' is a watch descriptor in decimal form, i.e. a target file |
c33e97ef MCC |
1994 | descriptor number, 'ino' and 'sdev' are inode and device where the |
1995 | target file resides and the 'mask' is the mask of events, all in hex | |
1996 | form [see inotify(7) for more details]. | |
f1d8c162 | 1997 | |
c33e97ef MCC |
1998 | If the kernel was built with exportfs support, the path to the target |
1999 | file is encoded as a file handle. The file handle is provided by three | |
2000 | fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex | |
2001 | format. | |
f1d8c162 | 2002 | |
c33e97ef MCC |
2003 | If the kernel is built without exportfs support the file handle won't be |
2004 | printed out. | |
f1d8c162 | 2005 | |
c33e97ef | 2006 | If there is no inotify mark attached yet the 'inotify' line will be omitted. |
f1d8c162 | 2007 | |
c33e97ef | 2008 | For fanotify files the format is:: |
f1d8c162 CG |
2009 | |
2010 | pos: 0 | |
2011 | flags: 02 | |
49d063cb | 2012 | mnt_id: 9 |
3845f256 | 2013 | ino: 63107 |
e71ec593 CG |
2014 | fanotify flags:10 event-flags:0 |
2015 | fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 | |
2016 | fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 | |
2017 | ||
c33e97ef MCC |
2018 | where fanotify 'flags' and 'event-flags' are values used in fanotify_init |
2019 | call, 'mnt_id' is the mount point identifier, 'mflags' is the value of | |
2020 | flags associated with mark which are tracked separately from events | |
059db434 | 2021 | mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events |
c33e97ef | 2022 | mask and 'ignored_mask' is the mask of events which are to be ignored. |
059db434 RD |
2023 | All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' |
2024 | provide information about flags and mask used in fanotify_mark | |
c33e97ef MCC |
2025 | call [see fsnotify manpage for details]. |
2026 | ||
2027 | While the first three lines are mandatory and always printed, the rest is | |
2028 | optional and may be omitted if no marks created yet. | |
e71ec593 | 2029 | |
c33e97ef MCC |
2030 | Timerfd files |
2031 | ~~~~~~~~~~~~~ | |
f1d8c162 | 2032 | |
c33e97ef | 2033 | :: |
854d06d9 CG |
2034 | |
2035 | pos: 0 | |
2036 | flags: 02 | |
2037 | mnt_id: 9 | |
3845f256 | 2038 | ino: 63107 |
854d06d9 CG |
2039 | clockid: 0 |
2040 | ticks: 0 | |
2041 | settime flags: 01 | |
2042 | it_value: (0, 49406829) | |
2043 | it_interval: (1, 0) | |
2044 | ||
c33e97ef MCC |
2045 | where 'clockid' is the clock type and 'ticks' is the number of the timer expirations |
2046 | that have occurred [see timerfd_create(2) for details]. 'settime flags' are | |
2047 | flags in octal form been used to setup the timer [see timerfd_settime(2) for | |
059db434 | 2048 | details]. 'it_value' is remaining time until the timer expiration. |
c33e97ef MCC |
2049 | 'it_interval' is the interval for the timer. Note the timer might be set up |
2050 | with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' | |
2051 | still exhibits timer's remaining time. | |
f1d8c162 | 2052 | |
3845f256 KS |
2053 | DMA Buffer files |
2054 | ~~~~~~~~~~~~~~~~ | |
2055 | ||
2056 | :: | |
2057 | ||
2058 | pos: 0 | |
2059 | flags: 04002 | |
2060 | mnt_id: 9 | |
2061 | ino: 63107 | |
2062 | size: 32768 | |
2063 | count: 2 | |
2064 | exp_name: system-heap | |
2065 | ||
2066 | where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of | |
2067 | the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. | |
2068 | ||
740a5ddb CG |
2069 | 3.9 /proc/<pid>/map_files - Information about memory mapped files |
2070 | --------------------------------------------------------------------- | |
2071 | This directory contains symbolic links which represent memory mapped files | |
c33e97ef | 2072 | the process is maintaining. Example output:: |
740a5ddb CG |
2073 | |
2074 | | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so | |
2075 | | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so | |
2076 | | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so | |
2077 | | ... | |
2078 | | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 | |
2079 | | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls | |
2080 | ||
2081 | The name of a link represents the virtual memory bounds of a mapping, i.e. | |
2082 | vm_area_struct::vm_start-vm_area_struct::vm_end. | |
2083 | ||
2084 | The main purpose of the map_files is to retrieve a set of memory mapped | |
2085 | files in a fast way instead of parsing /proc/<pid>/maps or | |
2086 | /proc/<pid>/smaps, both of which contain many more records. At the same | |
2087 | time one can open(2) mappings from the listings of two processes and | |
2088 | comparing their inode numbers to figure out which anonymous memory areas | |
2089 | are actually shared. | |
2090 | ||
5de23d43 JS |
2091 | 3.10 /proc/<pid>/timerslack_ns - Task timerslack value |
2092 | --------------------------------------------------------- | |
2093 | This file provides the value of the task's timerslack value in nanoseconds. | |
059db434 | 2094 | This value specifies an amount of time that normal timers may be deferred |
5de23d43 JS |
2095 | in order to coalesce timers and avoid unnecessary wakeups. |
2096 | ||
059db434 | 2097 | This allows a task's interactivity vs power consumption tradeoff to be |
5de23d43 JS |
2098 | adjusted. |
2099 | ||
059db434 | 2100 | Writing 0 to the file will set the task's timerslack to the default value. |
5de23d43 JS |
2101 | |
2102 | Valid values are from 0 - ULLONG_MAX | |
2103 | ||
2104 | An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level | |
2105 | permissions on the task specified to change its timerslack_ns value. | |
2106 | ||
7c23b330 JP |
2107 | 3.11 /proc/<pid>/patch_state - Livepatch patch operation state |
2108 | ----------------------------------------------------------------- | |
2109 | When CONFIG_LIVEPATCH is enabled, this file displays the value of the | |
2110 | patch state for the task. | |
2111 | ||
2112 | A value of '-1' indicates that no patch is in transition. | |
2113 | ||
2114 | A value of '0' indicates that a patch is in transition and the task is | |
2115 | unpatched. If the patch is being enabled, then the task hasn't been | |
2116 | patched yet. If the patch is being disabled, then the task has already | |
2117 | been unpatched. | |
2118 | ||
2119 | A value of '1' indicates that a patch is in transition and the task is | |
2120 | patched. If the patch is being enabled, then the task has already been | |
2121 | patched. If the patch is being disabled, then the task hasn't been | |
2122 | unpatched yet. | |
2123 | ||
711486fd AL |
2124 | 3.12 /proc/<pid>/arch_status - task architecture specific status |
2125 | ------------------------------------------------------------------- | |
2126 | When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the | |
2127 | architecture specific status of the task. | |
2128 | ||
2129 | Example | |
c33e97ef MCC |
2130 | ~~~~~~~ |
2131 | ||
2132 | :: | |
2133 | ||
711486fd AL |
2134 | $ cat /proc/6753/arch_status |
2135 | AVX512_elapsed_ms: 8 | |
2136 | ||
2137 | Description | |
c33e97ef | 2138 | ~~~~~~~~~~~ |
711486fd | 2139 | |
059db434 | 2140 | x86 specific entries |
c33e97ef MCC |
2141 | ~~~~~~~~~~~~~~~~~~~~~ |
2142 | ||
059db434 | 2143 | AVX512_elapsed_ms |
c33e97ef MCC |
2144 | ^^^^^^^^^^^^^^^^^^ |
2145 | ||
711486fd AL |
2146 | If AVX512 is supported on the machine, this entry shows the milliseconds |
2147 | elapsed since the last time AVX512 usage was recorded. The recording | |
2148 | happens on a best effort basis when a task is scheduled out. This means | |
2149 | that the value depends on two factors: | |
2150 | ||
2151 | 1) The time which the task spent on the CPU without being scheduled | |
2152 | out. With CPU isolation and a single runnable task this can take | |
2153 | several seconds. | |
2154 | ||
2155 | 2) The time since the task was scheduled out last. Depending on the | |
2156 | reason for being scheduled out (time slice exhausted, syscall ...) | |
2157 | this can be arbitrary long time. | |
2158 | ||
2159 | As a consequence the value cannot be considered precise and authoritative | |
2160 | information. The application which uses this information has to be aware | |
2161 | of the overall scenario on the system in order to determine whether a | |
2162 | task is a real AVX512 user or not. Precise information can be obtained | |
2163 | with performance counters. | |
2164 | ||
2165 | A special value of '-1' indicates that no AVX512 usage was recorded, thus | |
2166 | the task is unlikely an AVX512 user, but depends on the workload and the | |
2167 | scheduling scenario, it also could be a false negative mentioned above. | |
5de23d43 | 2168 | |
f1f1f256 IB |
2169 | 3.13 /proc/<pid>/fd - List of symlinks to open files |
2170 | ------------------------------------------------------- | |
2171 | This directory contains symbolic links which represent open files | |
2172 | the process is maintaining. Example output:: | |
2173 | ||
2174 | lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null | |
2175 | l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null | |
2176 | lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]' | |
2177 | lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]' | |
2178 | lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]' | |
2179 | ||
2180 | The number of open files for the process is stored in 'size' member | |
2181 | of stat() output for /proc/<pid>/fd for fast access. | |
2182 | ------------------------------------------------------- | |
2183 | ||
2184 | ||
059db434 RD |
2185 | Chapter 4: Configuring procfs |
2186 | ============================= | |
0499680a VK |
2187 | |
2188 | 4.1 Mount options | |
2189 | --------------------- | |
2190 | ||
2191 | The following mount options are supported: | |
2192 | ||
c33e97ef | 2193 | ========= ======================================================== |
0499680a VK |
2194 | hidepid= Set /proc/<pid>/ access mode. |
2195 | gid= Set the group authorized to learn processes information. | |
37e7647a | 2196 | subset= Show only the specified subset of procfs. |
c33e97ef | 2197 | ========= ======================================================== |
0499680a | 2198 | |
1c6c4d11 AG |
2199 | hidepid=off or hidepid=0 means classic mode - everybody may access all |
2200 | /proc/<pid>/ directories (default). | |
2201 | ||
2202 | hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ | |
2203 | directories but their own. Sensitive files like cmdline, sched*, status are now | |
2204 | protected against other users. This makes it impossible to learn whether any | |
2205 | user runs specific program (given the program doesn't reveal itself by its | |
2206 | behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for | |
2207 | other users, poorly written programs passing sensitive information via program | |
2208 | arguments are now protected against local eavesdroppers. | |
2209 | ||
2210 | hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be | |
2211 | fully invisible to other users. It doesn't mean that it hides a fact whether a | |
2212 | process with a specific pid value exists (it can be learned by other means, e.g. | |
2213 | by "kill -0 $PID"), but it hides process' uid and gid, which may be learned by | |
2214 | stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of | |
2215 | gathering information about running processes, whether some daemon runs with | |
2216 | elevated privileges, whether other user runs some sensitive program, whether | |
2217 | other users run any program at all, etc. | |
2218 | ||
2219 | hidepid=ptraceable or hidepid=4 means that procfs should only contain | |
2220 | /proc/<pid>/ directories that the caller can ptrace. | |
0499680a VK |
2221 | |
2222 | gid= defines a group authorized to learn processes information otherwise | |
2223 | prohibited by hidepid=. If you use some daemon like identd which needs to learn | |
2224 | information about processes information, just add identd to this group. | |
37e7647a AG |
2225 | |
2226 | subset=pid hides all top level files and directories in the procfs that | |
2227 | are not related to tasks. | |
2228 | ||
059db434 RD |
2229 | Chapter 5: Filesystem behavior |
2230 | ============================== | |
37e7647a AG |
2231 | |
2232 | Originally, before the advent of pid namepsace, procfs was a global file | |
2233 | system. It means that there was only one procfs instance in the system. | |
2234 | ||
2235 | When pid namespace was added, a separate procfs instance was mounted in | |
2236 | each pid namespace. So, procfs mount options are global among all | |
565dbe72 | 2237 | mountpoints within the same namespace:: |
37e7647a | 2238 | |
565dbe72 MCC |
2239 | # grep ^proc /proc/mounts |
2240 | proc /proc proc rw,relatime,hidepid=2 0 0 | |
37e7647a | 2241 | |
565dbe72 MCC |
2242 | # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc |
2243 | mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 | |
2244 | +++ exited with 0 +++ | |
37e7647a | 2245 | |
565dbe72 MCC |
2246 | # grep ^proc /proc/mounts |
2247 | proc /proc proc rw,relatime,hidepid=2 0 0 | |
2248 | proc /tmp/proc proc rw,relatime,hidepid=2 0 0 | |
37e7647a AG |
2249 | |
2250 | and only after remounting procfs mount options will change at all | |
565dbe72 | 2251 | mountpoints:: |
37e7647a | 2252 | |
565dbe72 | 2253 | # mount -o remount,hidepid=1 -t proc proc /tmp/proc |
37e7647a | 2254 | |
565dbe72 MCC |
2255 | # grep ^proc /proc/mounts |
2256 | proc /proc proc rw,relatime,hidepid=1 0 0 | |
2257 | proc /tmp/proc proc rw,relatime,hidepid=1 0 0 | |
37e7647a AG |
2258 | |
2259 | This behavior is different from the behavior of other filesystems. | |
2260 | ||
2261 | The new procfs behavior is more like other filesystems. Each procfs mount | |
2262 | creates a new procfs instance. Mount options affect own procfs instance. | |
2263 | It means that it became possible to have several procfs instances | |
565dbe72 | 2264 | displaying tasks with different filtering options in one pid namespace:: |
37e7647a | 2265 | |
565dbe72 MCC |
2266 | # mount -o hidepid=invisible -t proc proc /proc |
2267 | # mount -o hidepid=noaccess -t proc proc /tmp/proc | |
2268 | # grep ^proc /proc/mounts | |
2269 | proc /proc proc rw,relatime,hidepid=invisible 0 0 | |
2270 | proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 |