| blob | e73df2722ff35ccdea2dee328b592ee01d13965a |
1 ------------------------------------------------------------------------------
2 T H E /proc F I L E S Y S T E M
3 ------------------------------------------------------------------------------
4 /proc/sys Terrehon Bowden <terrehon@pacbell.net> October 7 1999
5 Bodo Bauer <bb@ricochet.net>
7 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8 move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009
9 ------------------------------------------------------------------------------
10 Version 1.3 Kernel version 2.2.12
11 Kernel version 2.4.0-test11-pre4
12 ------------------------------------------------------------------------------
13 fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009
15 Table of Contents
16 -----------------
18 0 Preface
19 0.1 Introduction/Credits
20 0.2 Legal Stuff
22 1 Collecting System Information
23 1.1 Process-Specific Subdirectories
24 1.2 Kernel data
25 1.3 IDE devices in /proc/ide
26 1.4 Networking info in /proc/net
27 1.5 SCSI info
28 1.6 Parallel port info in /proc/parport
29 1.7 TTY info in /proc/tty
30 1.8 Miscellaneous kernel statistics in /proc/stat
31 1.9 Ext4 file system parameters
33 2 Modifying System Parameters
35 3 Per-Process Parameters
36 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer
37 score
38 3.2 /proc/<pid>/oom_score - Display current oom-killer score
39 3.3 /proc/<pid>/io - Display the IO accounting fields
40 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
41 3.5 /proc/<pid>/mountinfo - Information about mounts
42 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
45 ------------------------------------------------------------------------------
46 Preface
47 ------------------------------------------------------------------------------
49 0.1 Introduction/Credits
50 ------------------------
52 This documentation is part of a soon (or so we hope) to be released book on
53 the SuSE Linux distribution. As there is no complete documentation for the
54 /proc file system and we've used many freely available sources to write these
55 chapters, it seems only fair to give the work back to the Linux community.
56 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
57 afraid it's still far from complete, but we hope it will be useful. As far as
58 we know, it is the first 'all-in-one' document about the /proc file system. It
59 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
60 SPARC, AXP, etc., features, you probably won't find what you are looking for.
61 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
62 additions and patches are welcome and will be added to this document if you
63 mail them to Bodo.
65 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
66 other people for help compiling this documentation. We'd also like to extend a
67 special thank you to Andi Kleen for documentation, which we relied on heavily
68 to create this document, as well as the additional information he provided.
69 Thanks to everybody else who contributed source or docs to the Linux kernel
70 and helped create a great piece of software... :)
72 If you have any comments, corrections or additions, please don't hesitate to
73 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
74 document.
76 The latest version of this document is available online at
77 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
79 If the above direction does not works for you, you could try the kernel
80 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
81 comandante@zaralinux.com.
83 0.2 Legal Stuff
84 ---------------
86 We don't guarantee the correctness of this document, and if you come to us
87 complaining about how you screwed up your system because of incorrect
88 documentation, we won't feel responsible...
90 ------------------------------------------------------------------------------
91 CHAPTER 1: COLLECTING SYSTEM INFORMATION
92 ------------------------------------------------------------------------------
94 ------------------------------------------------------------------------------
95 In This Chapter
96 ------------------------------------------------------------------------------
97 * Investigating the properties of the pseudo file system /proc and its
98 ability to provide information on the running Linux system
99 * Examining /proc's structure
100 * Uncovering various information about the kernel and the processes running
101 on the system
102 ------------------------------------------------------------------------------
105 The proc file system acts as an interface to internal data structures in the
106 kernel. It can be used to obtain information about the system and to change
107 certain kernel parameters at runtime (sysctl).
109 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
110 show you how you can use /proc/sys to change settings.
112 1.1 Process-Specific Subdirectories
113 -----------------------------------
115 The directory /proc contains (among other things) one subdirectory for each
116 process running on the system, which is named after the process ID (PID).
118 The link self points to the process reading the file system. Each process
119 subdirectory has the entries listed in Table 1-1.
122 Table 1-1: Process specific entries in /proc
123 ..............................................................................
124 File Content
125 clear_refs Clears page referenced bits shown in smaps output
126 cmdline Command line arguments
127 cpu Current and last cpu in which it was executed (2.4)(smp)
128 cwd Link to the current working directory
129 environ Values of environment variables
130 exe Link to the executable of this process
131 fd Directory, which contains all file descriptors
132 maps Memory maps to executables and library files (2.4)
133 mem Memory held by this process
134 root Link to the root directory of this process
135 stat Process status
136 statm Process memory status information
137 status Process status in human readable form
138 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
139 pagemap Page table
140 stack Report full stack trace, enable via CONFIG_STACKTRACE
141 smaps a extension based on maps, showing the memory consumption of
142 each mapping
143 ..............................................................................
145 For example, to get the status information of a process, all you have to do is
146 read the file /proc/PID/status:
148 >cat /proc/self/status
149 Name: cat
150 State: R (running)
151 Tgid: 5452
152 Pid: 5452
153 PPid: 743
154 TracerPid: 0 (2.4)
155 Uid: 501 501 501 501
156 Gid: 100 100 100 100
157 FDSize: 256
158 Groups: 100 14 16
159 VmPeak: 5004 kB
160 VmSize: 5004 kB
161 VmLck: 0 kB
162 VmHWM: 476 kB
163 VmRSS: 476 kB
164 VmData: 156 kB
165 VmStk: 88 kB
166 VmExe: 68 kB
167 VmLib: 1412 kB
168 VmPTE: 20 kb
169 VmSwap: 0 kB
170 Threads: 1
171 SigQ: 0/28578
172 SigPnd: 0000000000000000
173 ShdPnd: 0000000000000000
174 SigBlk: 0000000000000000
175 SigIgn: 0000000000000000
176 SigCgt: 0000000000000000
177 CapInh: 00000000fffffeff
178 CapPrm: 0000000000000000
179 CapEff: 0000000000000000
180 CapBnd: ffffffffffffffff
181 voluntary_ctxt_switches: 0
182 nonvoluntary_ctxt_switches: 1
184 This shows you nearly the same information you would get if you viewed it with
185 the ps command. In fact, ps uses the proc file system to obtain its
186 information. But you get a more detailed view of the process by reading the
187 file /proc/PID/status. It fields are described in table 1-2.
189 The statm file contains more detailed information about the process
190 memory usage. Its seven fields are explained in Table 1-3. The stat file
191 contains details information about the process itself. Its fields are
192 explained in Table 1-4.
194 (for SMP CONFIG users)
195 For making accounting scalable, RSS related information are handled in
196 asynchronous manner and the vaule may not be very precise. To see a precise
197 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
198 It's slow but very precise.
200 Table 1-2: Contents of the status files (as of 2.6.30-rc7)
201 ..............................................................................
202 Field Content
203 Name filename of the executable
204 State state (R is running, S is sleeping, D is sleeping
205 in an uninterruptible wait, Z is zombie,
206 T is traced or stopped)
207 Tgid thread group ID
208 Pid process id
209 PPid process id of the parent process
210 TracerPid PID of process tracing this process (0 if not)
211 Uid Real, effective, saved set, and file system UIDs
212 Gid Real, effective, saved set, and file system GIDs
213 FDSize number of file descriptor slots currently allocated
214 Groups supplementary group list
215 VmPeak peak virtual memory size
216 VmSize total program size
217 VmLck locked memory size
218 VmHWM peak resident set size ("high water mark")
219 VmRSS size of memory portions
220 VmData size of data, stack, and text segments
221 VmStk size of data, stack, and text segments
222 VmExe size of text segment
223 VmLib size of shared library code
224 VmPTE size of page table entries
225 VmSwap size of swap usage (the number of referred swapents)
226 Threads number of threads
227 SigQ number of signals queued/max. number for queue
228 SigPnd bitmap of pending signals for the thread
229 ShdPnd bitmap of shared pending signals for the process
230 SigBlk bitmap of blocked signals
231 SigIgn bitmap of ignored signals
232 SigCgt bitmap of catched signals
233 CapInh bitmap of inheritable capabilities
234 CapPrm bitmap of permitted capabilities
235 CapEff bitmap of effective capabilities
236 CapBnd bitmap of capabilities bounding set
237 Cpus_allowed mask of CPUs on which this process may run
238 Cpus_allowed_list Same as previous, but in "list format"
239 Mems_allowed mask of memory nodes allowed to this process
240 Mems_allowed_list Same as previous, but in "list format"
241 voluntary_ctxt_switches number of voluntary context switches
242 nonvoluntary_ctxt_switches number of non voluntary context switches
243 ..............................................................................
245 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
246 ..............................................................................
247 Field Content
248 size total program size (pages) (same as VmSize in status)
249 resident size of memory portions (pages) (same as VmRSS in status)
250 shared number of pages that are shared (i.e. backed by a file)
251 trs number of pages that are 'code' (not including libs; broken,
252 includes data segment)
253 lrs number of pages of library (always 0 on 2.6)
254 drs number of pages of data/stack (including libs; broken,
255 includes library text)
256 dt number of dirty pages (always 0 on 2.6)
257 ..............................................................................
260 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
261 ..............................................................................
262 Field Content
263 pid process id
264 tcomm filename of the executable
265 state state (R is running, S is sleeping, D is sleeping in an
266 uninterruptible wait, Z is zombie, T is traced or stopped)
267 ppid process id of the parent process
268 pgrp pgrp of the process
269 sid session id
270 tty_nr tty the process uses
271 tty_pgrp pgrp of the tty
272 flags task flags
273 min_flt number of minor faults
274 cmin_flt number of minor faults with child's
275 maj_flt number of major faults
276 cmaj_flt number of major faults with child's
277 utime user mode jiffies
278 stime kernel mode jiffies
279 cutime user mode jiffies with child's
280 cstime kernel mode jiffies with child's
281 priority priority level
282 nice nice level
283 num_threads number of threads
284 it_real_value (obsolete, always 0)
285 start_time time the process started after system boot
286 vsize virtual memory size
287 rss resident set memory size
288 rsslim current limit in bytes on the rss
289 start_code address above which program text can run
290 end_code address below which program text can run
291 start_stack address of the start of the stack
292 esp current value of ESP
293 eip current value of EIP
294 pending bitmap of pending signals
295 blocked bitmap of blocked signals
296 sigign bitmap of ignored signals
297 sigcatch bitmap of catched signals
298 wchan address where process went to sleep
299 0 (place holder)
300 0 (place holder)
301 exit_signal signal to send to parent thread on exit
302 task_cpu which CPU the task is scheduled on
303 rt_priority realtime priority
304 policy scheduling policy (man sched_setscheduler)
305 blkio_ticks time spent waiting for block IO
306 gtime guest time of the task in jiffies
307 cgtime guest time of the task children in jiffies
308 ..............................................................................
310 The /proc/PID/maps file containing the currently mapped memory regions and
311 their access permissions.
313 The format is:
315 address perms offset dev inode pathname
317 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
318 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
319 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
320 a7cb1000-a7cb2000 ---p 00000000 00:00 0
321 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
322 a7eb2000-a7eb3000 ---p 00000000 00:00 0
323 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
324 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
325 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
326 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
327 a800b000-a800e000 rw-p 00000000 00:00 0
328 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
329 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
330 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
331 a8024000-a8027000 rw-p 00000000 00:00 0
332 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
333 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
334 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
335 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
336 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
338 where "address" is the address space in the process that it occupies, "perms"
339 is a set of permissions:
341 r = read
342 w = write
343 x = execute
344 s = shared
345 p = private (copy on write)
347 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
348 "inode" is the inode on that device. 0 indicates that no inode is associated
349 with the memory region, as the case would be with BSS (uninitialized data).
350 The "pathname" shows the name associated file for this mapping. If the mapping
351 is not associated with a file:
353 [heap] = the heap of the program
354 [stack] = the stack of the main process
355 [vdso] = the "virtual dynamic shared object",
356 the kernel system call handler
358 or if empty, the mapping is anonymous.
361 The /proc/PID/smaps is an extension based on maps, showing the memory
362 consumption for each of the process's mappings. For each of mappings there
363 is a series of lines such as the following:
365 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
366 Size: 1084 kB
367 Rss: 892 kB
368 Pss: 374 kB
369 Shared_Clean: 892 kB
370 Shared_Dirty: 0 kB
371 Private_Clean: 0 kB
372 Private_Dirty: 0 kB
373 Referenced: 892 kB
374 Anonymous: 0 kB
375 Swap: 0 kB
376 KernelPageSize: 4 kB
377 MMUPageSize: 4 kB
379 The first of these lines shows the same information as is displayed for the
380 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
381 (size), the amount of the mapping that is currently resident in RAM (RSS), the
382 process' proportional share of this mapping (PSS), the number of clean and
383 dirty private pages in the mapping. Note that even a page which is part of a
384 MAP_SHARED mapping, but has only a single pte mapped, i.e. is currently used
385 by only one process, is accounted as private and not as shared. "Referenced"
386 indicates the amount of memory currently marked as referenced or accessed.
387 "Anonymous" shows the amount of memory that does not belong to any file. Even
388 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
389 and a page is modified, the file page is replaced by a private anonymous copy.
390 "Swap" shows how much would-be-anonymous memory is also used, but out on
391 swap.
393 This file is only present if the CONFIG_MMU kernel configuration option is
394 enabled.
396 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
397 bits on both physical and virtual pages associated with a process.
398 To clear the bits for all the pages associated with the process
399 > echo 1 > /proc/PID/clear_refs
401 To clear the bits for the anonymous pages associated with the process
402 > echo 2 > /proc/PID/clear_refs
404 To clear the bits for the file mapped pages associated with the process
405 > echo 3 > /proc/PID/clear_refs
406 Any other value written to /proc/PID/clear_refs will have no effect.
408 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
409 using /proc/kpageflags and number of times a page is mapped using
410 /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
412 1.2 Kernel data
413 ---------------
415 Similar to the process entries, the kernel data files give information about
416 the running kernel. The files used to obtain this information are contained in
417 /proc and are listed in Table 1-5. Not all of these will be present in your
418 system. It depends on the kernel configuration and the loaded modules, which
419 files are there, and which are missing.
421 Table 1-5: Kernel info in /proc
422 ..............................................................................
423 File Content
424 apm Advanced power management info
425 buddyinfo Kernel memory allocator information (see text) (2.5)
426 bus Directory containing bus specific information
427 cmdline Kernel command line
428 cpuinfo Info about the CPU
429 devices Available devices (block and character)
430 dma Used DMS channels
431 filesystems Supported filesystems
432 driver Various drivers grouped here, currently rtc (2.4)
433 execdomains Execdomains, related to security (2.4)
434 fb Frame Buffer devices (2.4)
435 fs File system parameters, currently nfs/exports (2.4)
436 ide Directory containing info about the IDE subsystem
437 interrupts Interrupt usage
438 iomem Memory map (2.4)
439 ioports I/O port usage
440 irq Masks for irq to cpu affinity (2.4)(smp?)
441 isapnp ISA PnP (Plug&Play) Info (2.4)
442 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
443 kmsg Kernel messages
444 ksyms Kernel symbol table
445 loadavg Load average of last 1, 5 & 15 minutes
446 locks Kernel locks
447 meminfo Memory info
448 misc Miscellaneous
449 modules List of loaded modules
450 mounts Mounted filesystems
451 net Networking info (see text)
452 pagetypeinfo Additional page allocator information (see text) (2.5)
453 partitions Table of partitions known to the system
454 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
455 decoupled by lspci (2.4)
456 rtc Real time clock
457 scsi SCSI info (see text)
458 slabinfo Slab pool info
459 softirqs softirq usage
460 stat Overall statistics
461 swaps Swap space utilization
462 sys See chapter 2
463 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
464 tty Info of tty drivers
465 uptime System uptime
466 version Kernel version
467 video bttv info of video resources (2.4)
468 vmallocinfo Show vmalloced areas
469 ..............................................................................
471 You can, for example, check which interrupts are currently in use and what
472 they are used for by looking in the file /proc/interrupts:
474 > cat /proc/interrupts
475 CPU0
476 0: 8728810 XT-PIC timer
477 1: 895 XT-PIC keyboard
478 2: 0 XT-PIC cascade
479 3: 531695 XT-PIC aha152x
480 4: 2014133 XT-PIC serial
481 5: 44401 XT-PIC pcnet_cs
482 8: 2 XT-PIC rtc
483 11: 8 XT-PIC i82365
484 12: 182918 XT-PIC PS/2 Mouse
485 13: 1 XT-PIC fpu
486 14: 1232265 XT-PIC ide0
487 15: 7 XT-PIC ide1
488 NMI: 0
490 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
491 output of a SMP machine):
493 > cat /proc/interrupts
495 CPU0 CPU1
496 0: 1243498 1214548 IO-APIC-edge timer
497 1: 8949 8958 IO-APIC-edge keyboard
498 2: 0 0 XT-PIC cascade
499 5: 11286 10161 IO-APIC-edge soundblaster
500 8: 1 0 IO-APIC-edge rtc
501 9: 27422 27407 IO-APIC-edge 3c503
502 12: 113645 113873 IO-APIC-edge PS/2 Mouse
503 13: 0 0 XT-PIC fpu
504 14: 22491 24012 IO-APIC-edge ide0
505 15: 2183 2415 IO-APIC-edge ide1
506 17: 30564 30414 IO-APIC-level eth0
507 18: 177 164 IO-APIC-level bttv
508 NMI: 2457961 2457959
509 LOC: 2457882 2457881
510 ERR: 2155
512 NMI is incremented in this case because every timer interrupt generates a NMI
513 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
515 LOC is the local interrupt counter of the internal APIC of every CPU.
517 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
518 connects the CPUs in a SMP system. This means that an error has been detected,
519 the IO-APIC automatically retry the transmission, so it should not be a big
520 problem, but you should read the SMP-FAQ.
522 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
523 /proc/interrupts to display every IRQ vector in use by the system, not
524 just those considered 'most important'. The new vectors are:
526 THR -- interrupt raised when a machine check threshold counter
527 (typically counting ECC corrected errors of memory or cache) exceeds
528 a configurable threshold. Only available on some systems.
530 TRM -- a thermal event interrupt occurs when a temperature threshold
531 has been exceeded for the CPU. This interrupt may also be generated
532 when the temperature drops back to normal.
534 SPU -- a spurious interrupt is some interrupt that was raised then lowered
535 by some IO device before it could be fully processed by the APIC. Hence
536 the APIC sees the interrupt but does not know what device it came from.
537 For this case the APIC will generate the interrupt with a IRQ vector
538 of 0xff. This might also be generated by chipset bugs.
540 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
541 sent from one CPU to another per the needs of the OS. Typically,
542 their statistics are used by kernel developers and interested users to
543 determine the occurrence of interrupts of the given type.
545 The above IRQ vectors are displayed only when relevent. For example,
546 the threshold vector does not exist on x86_64 platforms. Others are
547 suppressed when the system is a uniprocessor. As of this writing, only
548 i386 and x86_64 platforms support the new IRQ vector displays.
550 Of some interest is the introduction of the /proc/irq directory to 2.4.
551 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
552 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
553 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
554 prof_cpu_mask.
556 For example
557 > ls /proc/irq/
558 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
559 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
560 > ls /proc/irq/0/
561 smp_affinity
563 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
564 IRQ, you can set it by doing:
566 > echo 1 > /proc/irq/10/smp_affinity
568 This means that only the first CPU will handle the IRQ, but you can also echo
569 5 which means that only the first and fourth CPU can handle the IRQ.
571 The contents of each smp_affinity file is the same by default:
573 > cat /proc/irq/0/smp_affinity
574 ffffffff
576 The default_smp_affinity mask applies to all non-active IRQs, which are the
577 IRQs which have not yet been allocated/activated, and hence which lack a
578 /proc/irq/[0-9]* directory.
580 The node file on an SMP system shows the node to which the device using the IRQ
581 reports itself as being attached. This hardware locality information does not
582 include information about any possible driver locality preference.
584 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
585 profiler. Default value is ffffffff (all cpus).
587 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
588 between all the CPUs which are allowed to handle it. As usual the kernel has
589 more info than you and does a better job than you, so the defaults are the
590 best choice for almost everyone.
592 There are three more important subdirectories in /proc: net, scsi, and sys.
593 The general rule is that the contents, or even the existence of these
594 directories, depend on your kernel configuration. If SCSI is not enabled, the
595 directory scsi may not exist. The same is true with the net, which is there
596 only when networking support is present in the running kernel.
598 The slabinfo file gives information about memory usage at the slab level.
599 Linux uses slab pools for memory management above page level in version 2.2.
600 Commonly used objects have their own slab pool (such as network buffers,
601 directory cache, and so on).
603 ..............................................................................
605 > cat /proc/buddyinfo
607 Node 0, zone DMA 0 4 5 4 4 3 ...
608 Node 0, zone Normal 1 0 0 1 101 8 ...
609 Node 0, zone HighMem 2 0 0 1 1 0 ...
611 External fragmentation is a problem under some workloads, and buddyinfo is a
612 useful tool for helping diagnose these problems. Buddyinfo will give you a
613 clue as to how big an area you can safely allocate, or why a previous
614 allocation failed.
616 Each column represents the number of pages of a certain order which are
617 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
618 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
619 available in ZONE_NORMAL, etc...
621 More information relevant to external fragmentation can be found in
622 pagetypeinfo.
624 > cat /proc/pagetypeinfo
625 Page block order: 9
626 Pages per block: 512
628 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
629 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
630 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
631 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
632 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
633 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
634 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
635 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
636 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
637 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
638 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
640 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
641 Node 0, zone DMA 2 0 5 1 0
642 Node 0, zone DMA32 41 6 967 2 0
644 Fragmentation avoidance in the kernel works by grouping pages of different
645 migrate types into the same contiguous regions of memory called page blocks.
646 A page block is typically the size of the default hugepage size e.g. 2MB on
647 X86-64. By keeping pages grouped based on their ability to move, the kernel
648 can reclaim pages within a page block to satisfy a high-order allocation.
650 The pagetypinfo begins with information on the size of a page block. It
651 then gives the same type of information as buddyinfo except broken down
652 by migrate-type and finishes with details on how many page blocks of each
653 type exist.
655 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
656 from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can
657 make an estimate of the likely number of huge pages that can be allocated
658 at a given point in time. All the "Movable" blocks should be allocatable
659 unless memory has been mlock()'d. Some of the Reclaimable blocks should
660 also be allocatable although a lot of filesystem metadata may have to be
661 reclaimed to achieve this.
663 ..............................................................................
665 meminfo:
667 Provides information about distribution and utilization of memory. This
668 varies by architecture and compile options. The following is from a
669 16GB PIII, which has highmem enabled. You may not have all of these fields.
671 > cat /proc/meminfo
674 MemTotal: 16344972 kB
675 MemFree: 13634064 kB
676 Buffers: 3656 kB
677 Cached: 1195708 kB
678 SwapCached: 0 kB
679 Active: 891636 kB
680 Inactive: 1077224 kB
681 HighTotal: 15597528 kB
682 HighFree: 13629632 kB
683 LowTotal: 747444 kB
684 LowFree: 4432 kB
685 SwapTotal: 0 kB
686 SwapFree: 0 kB
687 Dirty: 968 kB
688 Writeback: 0 kB
689 AnonPages: 861800 kB
690 Mapped: 280372 kB
691 Slab: 284364 kB
692 SReclaimable: 159856 kB
693 SUnreclaim: 124508 kB
694 PageTables: 24448 kB
695 NFS_Unstable: 0 kB
696 Bounce: 0 kB
697 WritebackTmp: 0 kB
698 CommitLimit: 7669796 kB
699 Committed_AS: 100056 kB
700 VmallocTotal: 112216 kB
701 VmallocUsed: 428 kB
702 VmallocChunk: 111088 kB
704 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
705 bits and the kernel binary code)
706 MemFree: The sum of LowFree+HighFree
707 Buffers: Relatively temporary storage for raw disk blocks
708 shouldn't get tremendously large (20MB or so)
709 Cached: in-memory cache for files read from the disk (the
710 pagecache). Doesn't include SwapCached
711 SwapCached: Memory that once was swapped out, is swapped back in but
712 still also is in the swapfile (if memory is needed it
713 doesn't need to be swapped out AGAIN because it is already
714 in the swapfile. This saves I/O)
715 Active: Memory that has been used more recently and usually not
716 reclaimed unless absolutely necessary.
717 Inactive: Memory which has been less recently used. It is more
718 eligible to be reclaimed for other purposes
719 HighTotal:
720 HighFree: Highmem is all memory above ~860MB of physical memory
721 Highmem areas are for use by userspace programs, or
722 for the pagecache. The kernel must use tricks to access
723 this memory, making it slower to access than lowmem.
724 LowTotal:
725 LowFree: Lowmem is memory which can be used for everything that
726 highmem can be used for, but it is also available for the
727 kernel's use for its own data structures. Among many
728 other things, it is where everything from the Slab is
729 allocated. Bad things happen when you're out of lowmem.
730 SwapTotal: total amount of swap space available
731 SwapFree: Memory which has been evicted from RAM, and is temporarily
732 on the disk
733 Dirty: Memory which is waiting to get written back to the disk
734 Writeback: Memory which is actively being written back to the disk
735 AnonPages: Non-file backed pages mapped into userspace page tables
736 Mapped: files which have been mmaped, such as libraries
737 Slab: in-kernel data structures cache
738 SReclaimable: Part of Slab, that might be reclaimed, such as caches
739 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
740 PageTables: amount of memory dedicated to the lowest level of page
741 tables.
742 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
743 storage
744 Bounce: Memory used for block device "bounce buffers"
745 WritebackTmp: Memory used by FUSE for temporary writeback buffers
746 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
747 this is the total amount of memory currently available to
748 be allocated on the system. This limit is only adhered to
749 if strict overcommit accounting is enabled (mode 2 in
750 'vm.overcommit_memory').
751 The CommitLimit is calculated with the following formula:
752 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
753 For example, on a system with 1G of physical RAM and 7G
754 of swap with a `vm.overcommit_ratio` of 30 it would
755 yield a CommitLimit of 7.3G.
756 For more details, see the memory overcommit documentation
757 in vm/overcommit-accounting.
758 Committed_AS: The amount of memory presently allocated on the system.
759 The committed memory is a sum of all of the memory which
760 has been allocated by processes, even if it has not been
761 "used" by them as of yet. A process which malloc()'s 1G
762 of memory, but only touches 300M of it will only show up
763 as using 300M of memory even if it has the address space
764 allocated for the entire 1G. This 1G is memory which has
765 been "committed" to by the VM and can be used at any time
766 by the allocating application. With strict overcommit
767 enabled on the system (mode 2 in 'vm.overcommit_memory'),
768 allocations which would exceed the CommitLimit (detailed
769 above) will not be permitted. This is useful if one needs
770 to guarantee that processes will not fail due to lack of
771 memory once that memory has been successfully allocated.
772 VmallocTotal: total size of vmalloc memory area
773 VmallocUsed: amount of vmalloc area which is used
774 VmallocChunk: largest contiguous block of vmalloc area which is free
776 ..............................................................................
778 vmallocinfo:
780 Provides information about vmalloced/vmaped areas. One line per area,
781 containing the virtual address range of the area, size in bytes,
782 caller information of the creator, and optional information depending
783 on the kind of area :
785 pages=nr number of pages
786 phys=addr if a physical address was specified
787 ioremap I/O mapping (ioremap() and friends)
788 vmalloc vmalloc() area
789 vmap vmap()ed pages
790 user VM_USERMAP area
791 vpages buffer for pages pointers was vmalloced (huge area)
792 N<node>=nr (Only on NUMA kernels)
793 Number of pages allocated on memory node <node>
795 > cat /proc/vmallocinfo
796 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
797 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
798 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
799 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
800 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
801 phys=7fee8000 ioremap
802 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
803 phys=7fee7000 ioremap
804 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
805 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
806 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
807 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
808 pages=2 vmalloc N1=2
809 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
810 /0x130 [x_tables] pages=4 vmalloc N0=4
811 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
812 pages=14 vmalloc N2=14
813 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
814 pages=4 vmalloc N1=4
815 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
816 pages=2 vmalloc N1=2
817 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
818 pages=10 vmalloc N0=10
820 ..............................................................................
822 softirqs:
824 Provides counts of softirq handlers serviced since boot time, for each cpu.
826 > cat /proc/softirqs
827 CPU0 CPU1 CPU2 CPU3
828 HI: 0 0 0 0
829 TIMER: 27166 27120 27097 27034
830 NET_TX: 0 0 0 17
831 NET_RX: 42 0 0 39
832 BLOCK: 0 0 107 1121
833 TASKLET: 0 0 0 290
834 SCHED: 27035 26983 26971 26746
835 HRTIMER: 0 0 0 0
836 RCU: 1678 1769 2178 2250
839 1.3 IDE devices in /proc/ide
840 ----------------------------
842 The subdirectory /proc/ide contains information about all IDE devices of which
843 the kernel is aware. There is one subdirectory for each IDE controller, the
844 file drivers and a link for each IDE device, pointing to the device directory
845 in the controller specific subtree.
847 The file drivers contains general information about the drivers used for the
848 IDE devices:
850 > cat /proc/ide/drivers
851 ide-cdrom version 4.53
852 ide-disk version 1.08
854 More detailed information can be found in the controller specific
855 subdirectories. These are named ide0, ide1 and so on. Each of these
856 directories contains the files shown in table 1-6.
859 Table 1-6: IDE controller info in /proc/ide/ide?
860 ..............................................................................
861 File Content
862 channel IDE channel (0 or 1)
863 config Configuration (only for PCI/IDE bridge)
864 mate Mate name
865 model Type/Chipset of IDE controller
866 ..............................................................................
868 Each device connected to a controller has a separate subdirectory in the
869 controllers directory. The files listed in table 1-7 are contained in these
870 directories.
873 Table 1-7: IDE device information
874 ..............................................................................
875 File Content
876 cache The cache
877 capacity Capacity of the medium (in 512Byte blocks)
878 driver driver and version
879 geometry physical and logical geometry
880 identify device identify block
881 media media type
882 model device identifier
883 settings device setup
884 smart_thresholds IDE disk management thresholds
885 smart_values IDE disk management values
886 ..............................................................................
888 The most interesting file is settings. This file contains a nice overview of
889 the drive parameters:
891 # cat /proc/ide/ide0/hda/settings
892 name value min max mode
893 ---- ----- --- --- ----
894 bios_cyl 526 0 65535 rw
895 bios_head 255 0 255 rw
896 bios_sect 63 0 63 rw
897 breada_readahead 4 0 127 rw
898 bswap 0 0 1 r
899 file_readahead 72 0 2097151 rw
900 io_32bit 0 0 3 rw
901 keepsettings 0 0 1 rw
902 max_kb_per_request 122 1 127 rw
903 multcount 0 0 8 rw
904 nice1 1 0 1 rw
905 nowerr 0 0 1 rw
906 pio_mode write-only 0 255 w
907 slow 0 0 1 rw
908 unmaskirq 0 0 1 rw
909 using_dma 0 0 1 rw
912 1.4 Networking info in /proc/net
913 --------------------------------
915 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
916 additional values you get for IP version 6 if you configure the kernel to
917 support this. Table 1-9 lists the files and their meaning.
920 Table 1-8: IPv6 info in /proc/net
921 ..............................................................................
922 File Content
923 udp6 UDP sockets (IPv6)
924 tcp6 TCP sockets (IPv6)
925 raw6 Raw device statistics (IPv6)
926 igmp6 IP multicast addresses, which this host joined (IPv6)
927 if_inet6 List of IPv6 interface addresses
928 ipv6_route Kernel routing table for IPv6
929 rt6_stats Global IPv6 routing tables statistics
930 sockstat6 Socket statistics (IPv6)
931 snmp6 Snmp data (IPv6)
932 ..............................................................................
935 Table 1-9: Network info in /proc/net
936 ..............................................................................
937 File Content
938 arp Kernel ARP table
939 dev network devices with statistics
940 dev_mcast the Layer2 multicast groups a device is listening too
941 (interface index, label, number of references, number of bound
942 addresses).
943 dev_stat network device status
944 ip_fwchains Firewall chain linkage
945 ip_fwnames Firewall chain names
946 ip_masq Directory containing the masquerading tables
947 ip_masquerade Major masquerading table
948 netstat Network statistics
949 raw raw device statistics
950 route Kernel routing table
951 rpc Directory containing rpc info
952 rt_cache Routing cache
953 snmp SNMP data
954 sockstat Socket statistics
955 tcp TCP sockets
956 tr_rif Token ring RIF routing table
957 udp UDP sockets
958 unix UNIX domain sockets
959 wireless Wireless interface data (Wavelan etc)
960 igmp IP multicast addresses, which this host joined
961 psched Global packet scheduler parameters.
962 netlink List of PF_NETLINK sockets
963 ip_mr_vifs List of multicast virtual interfaces
964 ip_mr_cache List of multicast routing cache
965 ..............................................................................
967 You can use this information to see which network devices are available in
968 your system and how much traffic was routed over those devices:
970 > cat /proc/net/dev
971 Inter-|Receive |[...
972 face |bytes packets errs drop fifo frame compressed multicast|[...
973 lo: 908188 5596 0 0 0 0 0 0 [...
974 ppp0:15475140 20721 410 0 0 410 0 0 [...
975 eth0: 614530 7085 0 0 0 0 0 1 [...
977 ...] Transmit
978 ...] bytes packets errs drop fifo colls carrier compressed
979 ...] 908188 5596 0 0 0 0 0 0
980 ...] 1375103 17405 0 0 0 0 0 0
981 ...] 1703981 5535 0 0 0 3 0 0
983 In addition, each Channel Bond interface has its own directory. For
984 example, the bond0 device will have a directory called /proc/net/bond0/.
985 It will contain information that is specific to that bond, such as the
986 current slaves of the bond, the link status of the slaves, and how
987 many times the slaves link has failed.
989 1.5 SCSI info
990 -------------
992 If you have a SCSI host adapter in your system, you'll find a subdirectory
993 named after the driver for this adapter in /proc/scsi. You'll also see a list
994 of all recognized SCSI devices in /proc/scsi:
996 >cat /proc/scsi/scsi
997 Attached devices:
998 Host: scsi0 Channel: 00 Id: 00 Lun: 00
999 Vendor: IBM Model: DGHS09U Rev: 03E0
1000 Type: Direct-Access ANSI SCSI revision: 03
1001 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1002 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1003 Type: CD-ROM ANSI SCSI revision: 02
1006 The directory named after the driver has one file for each adapter found in
1007 the system. These files contain information about the controller, including
1008 the used IRQ and the IO address range. The amount of information shown is
1009 dependent on the adapter you use. The example shows the output for an Adaptec
1010 AHA-2940 SCSI adapter:
1012 > cat /proc/scsi/aic7xxx/0
1014 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1015 Compile Options:
1016 TCQ Enabled By Default : Disabled
1017 AIC7XXX_PROC_STATS : Disabled
1018 AIC7XXX_RESET_DELAY : 5
1019 Adapter Configuration:
1020 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1021 Ultra Wide Controller
1022 PCI MMAPed I/O Base: 0xeb001000
1023 Adapter SEEPROM Config: SEEPROM found and used.
1024 Adaptec SCSI BIOS: Enabled
1025 IRQ: 10
1026 SCBs: Active 0, Max Active 2,
1027 Allocated 15, HW 16, Page 255
1028 Interrupts: 160328
1029 BIOS Control Word: 0x18b6
1030 Adapter Control Word: 0x005b
1031 Extended Translation: Enabled
1032 Disconnect Enable Flags: 0xffff
1033 Ultra Enable Flags: 0x0001
1034 Tag Queue Enable Flags: 0x0000
1035 Ordered Queue Tag Flags: 0x0000
1036 Default Tag Queue Depth: 8
1037 Tagged Queue By Device array for aic7xxx host instance 0:
1038 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1039 Actual queue depth per device for aic7xxx host instance 0:
1040 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1041 Statistics:
1042 (scsi0:0:0:0)
1043 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1044 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1045 Total transfers 160151 (74577 reads and 85574 writes)
1046 (scsi0:0:6:0)
1047 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1048 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1049 Total transfers 0 (0 reads and 0 writes)
1052 1.6 Parallel port info in /proc/parport
1053 ---------------------------------------
1055 The directory /proc/parport contains information about the parallel ports of
1056 your system. It has one subdirectory for each port, named after the port
1057 number (0,1,2,...).
1059 These directories contain the four files shown in Table 1-10.
1062 Table 1-10: Files in /proc/parport
1063 ..............................................................................
1064 File Content
1065 autoprobe Any IEEE-1284 device ID information that has been acquired.
1066 devices list of the device drivers using that port. A + will appear by the
1067 name of the device currently using the port (it might not appear
1068 against any).
1069 hardware Parallel port's base address, IRQ line and DMA channel.
1070 irq IRQ that parport is using for that port. This is in a separate
1071 file to allow you to alter it by writing a new value in (IRQ
1072 number or none).
1073 ..............................................................................
1075 1.7 TTY info in /proc/tty
1076 -------------------------
1078 Information about the available and actually used tty's can be found in the
1079 directory /proc/tty.You'll find entries for drivers and line disciplines in
1080 this directory, as shown in Table 1-11.
1083 Table 1-11: Files in /proc/tty
1084 ..............................................................................
1085 File Content
1086 drivers list of drivers and their usage
1087 ldiscs registered line disciplines
1088 driver/serial usage statistic and status of single tty lines
1089 ..............................................................................
1091 To see which tty's are currently in use, you can simply look into the file
1092 /proc/tty/drivers:
1094 > cat /proc/tty/drivers
1095 pty_slave /dev/pts 136 0-255 pty:slave
1096 pty_master /dev/ptm 128 0-255 pty:master
1097 pty_slave /dev/ttyp 3 0-255 pty:slave
1098 pty_master /dev/pty 2 0-255 pty:master
1099 serial /dev/cua 5 64-67 serial:callout
1100 serial /dev/ttyS 4 64-67 serial
1101 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1102 /dev/ptmx /dev/ptmx 5 2 system
1103 /dev/console /dev/console 5 1 system:console
1104 /dev/tty /dev/tty 5 0 system:/dev/tty
1105 unknown /dev/tty 4 1-63 console
1108 1.8 Miscellaneous kernel statistics in /proc/stat
1109 -------------------------------------------------
1111 Various pieces of information about kernel activity are available in the
1112 /proc/stat file. All of the numbers reported in this file are aggregates
1113 since the system first booted. For a quick look, simply cat the file:
1115 > cat /proc/stat
1116 cpu 2255 34 2290 22625563 6290 127 456 0 0
1117 cpu0 1132 34 1441 11311718 3675 127 438 0 0
1118 cpu1 1123 0 849 11313845 2614 0 18 0 0
1119 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1120 ctxt 1990473
1121 btime 1062191376
1122 processes 2915
1123 procs_running 1
1124 procs_blocked 0
1125 softirq 183433 0 21755 12 39 1137 231 21459 2263
1127 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1128 lines. These numbers identify the amount of time the CPU has spent performing
1129 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1130 second). The meanings of the columns are as follows, from left to right:
1132 - user: normal processes executing in user mode
1133 - nice: niced processes executing in user mode
1134 - system: processes executing in kernel mode
1135 - idle: twiddling thumbs
1136 - iowait: waiting for I/O to complete
1137 - irq: servicing interrupts
1138 - softirq: servicing softirqs
1139 - steal: involuntary wait
1140 - guest: running a normal guest
1141 - guest_nice: running a niced guest
1143 The "intr" line gives counts of interrupts serviced since boot time, for each
1144 of the possible system interrupts. The first column is the total of all
1145 interrupts serviced; each subsequent column is the total for that particular
1146 interrupt.
1148 The "ctxt" line gives the total number of context switches across all CPUs.
1150 The "btime" line gives the time at which the system booted, in seconds since
1151 the Unix epoch.
1153 The "processes" line gives the number of processes and threads created, which
1154 includes (but is not limited to) those created by calls to the fork() and
1155 clone() system calls.
1157 The "procs_running" line gives the total number of threads that are
1158 running or ready to run (i.e., the total number of runnable threads).
1160 The "procs_blocked" line gives the number of processes currently blocked,
1161 waiting for I/O to complete.
1163 The "softirq" line gives counts of softirqs serviced since boot time, for each
1164 of the possible system softirqs. The first column is the total of all
1165 softirqs serviced; each subsequent column is the total for that particular
1166 softirq.
1169 1.9 Ext4 file system parameters
1170 ------------------------------
1172 Information about mounted ext4 file systems can be found in
1173 /proc/fs/ext4. Each mounted filesystem will have a directory in
1174 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1175 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1176 in Table 1-12, below.
1178 Table 1-12: Files in /proc/fs/ext4/<devname>
1179 ..............................................................................
1180 File Content
1181 mb_groups details of multiblock allocator buddy cache of free blocks
1182 ..............................................................................
1185 ------------------------------------------------------------------------------
1186 Summary
1187 ------------------------------------------------------------------------------
1188 The /proc file system serves information about the running system. It not only
1189 allows access to process data but also allows you to request the kernel status
1190 by reading files in the hierarchy.
1192 The directory structure of /proc reflects the types of information and makes
1193 it easy, if not obvious, where to look for specific data.
1194 ------------------------------------------------------------------------------
1196 ------------------------------------------------------------------------------
1197 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1198 ------------------------------------------------------------------------------
1200 ------------------------------------------------------------------------------
1201 In This Chapter
1202 ------------------------------------------------------------------------------
1203 * Modifying kernel parameters by writing into files found in /proc/sys
1204 * Exploring the files which modify certain parameters
1205 * Review of the /proc/sys file tree
1206 ------------------------------------------------------------------------------
1209 A very interesting part of /proc is the directory /proc/sys. This is not only
1210 a source of information, it also allows you to change parameters within the
1211 kernel. Be very careful when attempting this. You can optimize your system,
1212 but you can also cause it to crash. Never alter kernel parameters on a
1213 production system. Set up a development machine and test to make sure that
1214 everything works the way you want it to. You may have no alternative but to
1215 reboot the machine once an error has been made.
1217 To change a value, simply echo the new value into the file. An example is
1218 given below in the section on the file system data. You need to be root to do
1219 this. You can create your own boot script to perform this every time your
1220 system boots.
1222 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1223 general things in the operation of the Linux kernel. Since some of the files
1224 can inadvertently disrupt your system, it is advisable to read both
1225 documentation and source before actually making adjustments. In any case, be
1226 very careful when writing to any of these files. The entries in /proc may
1227 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1228 review the kernel documentation in the directory /usr/src/linux/Documentation.
1229 This chapter is heavily based on the documentation included in the pre 2.2
1230 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1232 Please see: Documentation/sysctls/ directory for descriptions of these
1233 entries.
1235 ------------------------------------------------------------------------------
1236 Summary
1237 ------------------------------------------------------------------------------
1238 Certain aspects of kernel behavior can be modified at runtime, without the
1239 need to recompile the kernel, or even to reboot the system. The files in the
1240 /proc/sys tree can not only be read, but also modified. You can use the echo
1241 command to write value into these files, thereby changing the default settings
1242 of the kernel.
1243 ------------------------------------------------------------------------------
1245 ------------------------------------------------------------------------------
1246 CHAPTER 3: PER-PROCESS PARAMETERS
1247 ------------------------------------------------------------------------------
1249 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1250 --------------------------------------------------------------------------------
1252 These file can be used to adjust the badness heuristic used to select which
1253 process gets killed in out of memory conditions.
1255 The badness heuristic assigns a value to each candidate task ranging from 0
1256 (never kill) to 1000 (always kill) to determine which process is targeted. The
1257 units are roughly a proportion along that range of allowed memory the process
1258 may allocate from based on an estimation of its current memory and swap use.
1259 For example, if a task is using all allowed memory, its badness score will be
1260 1000. If it is using half of its allowed memory, its score will be 500.
1262 There is an additional factor included in the badness score: root
1263 processes are given 3% extra memory over other tasks.
1265 The amount of "allowed" memory depends on the context in which the oom killer
1266 was called. If it is due to the memory assigned to the allocating task's cpuset
1267 being exhausted, the allowed memory represents the set of mems assigned to that
1268 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1269 memory represents the set of mempolicy nodes. If it is due to a memory
1270 limit (or swap limit) being reached, the allowed memory is that configured
1271 limit. Finally, if it is due to the entire system being out of memory, the
1272 allowed memory represents all allocatable resources.
1274 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1275 is used to determine which task to kill. Acceptable values range from -1000
1276 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1277 polarize the preference for oom killing either by always preferring a certain
1278 task or completely disabling it. The lowest possible value, -1000, is
1279 equivalent to disabling oom killing entirely for that task since it will always
1280 report a badness score of 0.
1282 Consequently, it is very simple for userspace to define the amount of memory to
1283 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1284 example, is roughly equivalent to allowing the remainder of tasks sharing the
1285 same system, cpuset, mempolicy, or memory controller resources to use at least
1286 50% more memory. A value of -500, on the other hand, would be roughly
1287 equivalent to discounting 50% of the task's allowed memory from being considered
1288 as scoring against the task.
1290 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1291 be used to tune the badness score. Its acceptable values range from -16
1292 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1293 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1294 scaled linearly with /proc/<pid>/oom_score_adj.
1296 Writing to /proc/<pid>/oom_score_adj or /proc/<pid>/oom_adj will change the
1297 other with its scaled value.
1299 NOTICE: /proc/<pid>/oom_adj is deprecated and will be removed, please see
1300 Documentation/feature-removal-schedule.txt.
1302 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1303 generation children with seperate address spaces instead, if possible. This
1304 avoids servers and important system daemons from being killed and loses the
1305 minimal amount of work.
1308 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1309 -------------------------------------------------------------
1311 This file can be used to check the current score used by the oom-killer is for
1312 any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
1313 process should be killed in an out-of-memory situation.
1316 3.3 /proc/<pid>/io - Display the IO accounting fields
1317 -------------------------------------------------------
1319 This file contains IO statistics for each running process
1321 Example
1322 -------
1324 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1325 [1] 3828
1327 test:/tmp # cat /proc/3828/io
1328 rchar: 323934931
1329 wchar: 323929600
1330 syscr: 632687
1331 syscw: 632675
1332 read_bytes: 0
1333 write_bytes: 323932160
1334 cancelled_write_bytes: 0
1337 Description
1338 -----------
1340 rchar
1341 -----
1343 I/O counter: chars read
1344 The number of bytes which this task has caused to be read from storage. This
1345 is simply the sum of bytes which this process passed to read() and pread().
1346 It includes things like tty IO and it is unaffected by whether or not actual
1347 physical disk IO was required (the read might have been satisfied from
1348 pagecache)
1351 wchar
1352 -----
1354 I/O counter: chars written
1355 The number of bytes which this task has caused, or shall cause to be written
1356 to disk. Similar caveats apply here as with rchar.
1359 syscr
1360 -----
1362 I/O counter: read syscalls
1363 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1364 and pread().
1367 syscw
1368 -----
1370 I/O counter: write syscalls
1371 Attempt to count the number of write I/O operations, i.e. syscalls like
1372 write() and pwrite().
1375 read_bytes
1376 ----------
1378 I/O counter: bytes read
1379 Attempt to count the number of bytes which this process really did cause to
1380 be fetched from the storage layer. Done at the submit_bio() level, so it is
1381 accurate for block-backed filesystems. <please add status regarding NFS and
1382 CIFS at a later time>
1385 write_bytes
1386 -----------
1388 I/O counter: bytes written
1389 Attempt to count the number of bytes which this process caused to be sent to
1390 the storage layer. This is done at page-dirtying time.
1393 cancelled_write_bytes
1394 ---------------------
1396 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1397 then deletes the file, it will in fact perform no writeout. But it will have
1398 been accounted as having caused 1MB of write.
1399 In other words: The number of bytes which this process caused to not happen,
1400 by truncating pagecache. A task can cause "negative" IO too. If this task
1401 truncates some dirty pagecache, some IO which another task has been accounted
1402 for (in its write_bytes) will not be happening. We _could_ just subtract that
1403 from the truncating task's write_bytes, but there is information loss in doing
1404 that.
1407 Note
1408 ----
1410 At its current implementation state, this is a bit racy on 32-bit machines: if
1411 process A reads process B's /proc/pid/io while process B is updating one of
1412 those 64-bit counters, process A could see an intermediate result.
1415 More information about this can be found within the taskstats documentation in
1416 Documentation/accounting.
1418 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1419 ---------------------------------------------------------------
1420 When a process is dumped, all anonymous memory is written to a core file as
1421 long as the size of the core file isn't limited. But sometimes we don't want
1422 to dump some memory segments, for example, huge shared memory. Conversely,
1423 sometimes we want to save file-backed memory segments into a core file, not
1424 only the individual files.
1426 /proc/<pid>/coredump_filter allows you to customize which memory segments
1427 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1428 of memory types. If a bit of the bitmask is set, memory segments of the
1429 corresponding memory type are dumped, otherwise they are not dumped.
1431 The following 7 memory types are supported:
1432 - (bit 0) anonymous private memory
1433 - (bit 1) anonymous shared memory
1434 - (bit 2) file-backed private memory
1435 - (bit 3) file-backed shared memory
1436 - (bit 4) ELF header pages in file-backed private memory areas (it is
1437 effective only if the bit 2 is cleared)
1438 - (bit 5) hugetlb private memory
1439 - (bit 6) hugetlb shared memory
1441 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1442 are always dumped regardless of the bitmask status.
1444 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
1445 effected by bit 5-6.
1447 Default value of coredump_filter is 0x23; this means all anonymous memory
1448 segments and hugetlb private memory are dumped.
1450 If you don't want to dump all shared memory segments attached to pid 1234,
1451 write 0x21 to the process's proc file.
1453 $ echo 0x21 > /proc/1234/coredump_filter
1455 When a new process is created, the process inherits the bitmask status from its
1456 parent. It is useful to set up coredump_filter before the program runs.
1457 For example:
1459 $ echo 0x7 > /proc/self/coredump_filter
1460 $ ./some_program
1462 3.5 /proc/<pid>/mountinfo - Information about mounts
1463 --------------------------------------------------------
1465 This file contains lines of the form:
1467 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1468 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1470 (1) mount ID: unique identifier of the mount (may be reused after umount)
1471 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1472 (3) major:minor: value of st_dev for files on filesystem
1473 (4) root: root of the mount within the filesystem
1474 (5) mount point: mount point relative to the process's root
1475 (6) mount options: per mount options
1476 (7) optional fields: zero or more fields of the form "tag[:value]"
1477 (8) separator: marks the end of the optional fields
1478 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1479 (10) mount source: filesystem specific information or "none"
1480 (11) super options: per super block options
1482 Parsers should ignore all unrecognised optional fields. Currently the
1483 possible optional fields are:
1485 shared:X mount is shared in peer group X
1486 master:X mount is slave to peer group X
1487 propagate_from:X mount is slave and receives propagation from peer group X (*)
1488 unbindable mount is unbindable
1490 (*) X is the closest dominant peer group under the process's root. If
1491 X is the immediate master of the mount, or if there's no dominant peer
1492 group under the same root, then only the "master:X" field is present
1493 and not the "propagate_from:X" field.
1495 For more information on mount propagation see:
1497 Documentation/filesystems/sharedsubtree.txt
1500 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1501 --------------------------------------------------------
1502 These files provide a method to access a tasks comm value. It also allows for
1503 a task to set its own or one of its thread siblings comm value. The comm value
1504 is limited in size compared to the cmdline value, so writing anything longer
1505 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1506 comm value.