About Kernel Documentation Linux Kernel Contact Linux Resources Linux Blog

Documentation / filesystems / proc.txt




Custom Search

Based on kernel version 4.13.3. Page generated on 2017-09-23 13:55 EST.

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>
6	
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
14	
15	Table of Contents
16	-----------------
17	
18	  0     Preface
19	  0.1	Introduction/Credits
20	  0.2	Legal Stuff
21	
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
32	
33	  2	Modifying System Parameters
34	
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
43	  3.7   /proc/<pid>/task/<tid>/children - Information about task children
44	  3.8   /proc/<pid>/fdinfo/<fd> - Information about opened file
45	  3.9   /proc/<pid>/map_files - Information about memory mapped files
46	  3.10  /proc/<pid>/timerslack_ns - Task timerslack value
47	  3.11	/proc/<pid>/patch_state - Livepatch patch operation state
48	
49	  4	Configuring procfs
50	  4.1	Mount options
51	
52	------------------------------------------------------------------------------
53	Preface
54	------------------------------------------------------------------------------
55	
56	0.1 Introduction/Credits
57	------------------------
58	
59	This documentation is  part of a soon (or  so we hope) to be  released book on
60	the SuSE  Linux distribution. As  there is  no complete documentation  for the
61	/proc file system and we've used  many freely available sources to write these
62	chapters, it  seems only fair  to give the work  back to the  Linux community.
63	This work is  based on the 2.2.*  kernel version and the  upcoming 2.4.*. I'm
64	afraid it's still far from complete, but we  hope it will be useful. As far as
65	we know, it is the first 'all-in-one' document about the /proc file system. It
66	is focused  on the Intel  x86 hardware,  so if you  are looking for  PPC, ARM,
67	SPARC, AXP, etc., features, you probably  won't find what you are looking for.
68	It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
69	additions and patches  are welcome and will  be added to this  document if you
70	mail them to Bodo.
71	
72	We'd like  to  thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
73	other people for help compiling this documentation. We'd also like to extend a
74	special thank  you to Andi Kleen for documentation, which we relied on heavily
75	to create  this  document,  as well as the additional information he provided.
76	Thanks to  everybody  else  who contributed source or docs to the Linux kernel
77	and helped create a great piece of software... :)
78	
79	If you  have  any comments, corrections or additions, please don't hesitate to
80	contact Bodo  Bauer  at  bb@ricochet.net.  We'll  be happy to add them to this
81	document.
82	
83	The   latest   version    of   this   document   is    available   online   at
84	http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
85	
86	If  the above  direction does  not works  for you,  you could  try the  kernel
87	mailing  list  at  linux-kernel@vger.kernel.org  and/or try  to  reach  me  at
88	comandante@zaralinux.com.
89	
90	0.2 Legal Stuff
91	---------------
92	
93	We don't  guarantee  the  correctness  of this document, and if you come to us
94	complaining about  how  you  screwed  up  your  system  because  of  incorrect
95	documentation, we won't feel responsible...
96	
97	------------------------------------------------------------------------------
98	CHAPTER 1: COLLECTING SYSTEM INFORMATION
99	------------------------------------------------------------------------------
100	
101	------------------------------------------------------------------------------
102	In This Chapter
103	------------------------------------------------------------------------------
104	* Investigating  the  properties  of  the  pseudo  file  system  /proc and its
105	  ability to provide information on the running Linux system
106	* Examining /proc's structure
107	* Uncovering  various  information  about the kernel and the processes running
108	  on the system
109	------------------------------------------------------------------------------
110	
111	
112	The proc  file  system acts as an interface to internal data structures in the
113	kernel. It  can  be  used to obtain information about the system and to change
114	certain kernel parameters at runtime (sysctl).
115	
116	First, we'll  take  a  look  at the read-only parts of /proc. In Chapter 2, we
117	show you how you can use /proc/sys to change settings.
118	
119	1.1 Process-Specific Subdirectories
120	-----------------------------------
121	
122	The directory  /proc  contains  (among other things) one subdirectory for each
123	process running on the system, which is named after the process ID (PID).
124	
125	The link  self  points  to  the  process reading the file system. Each process
126	subdirectory has the entries listed in Table 1-1.
127	
128	
129	Table 1-1: Process specific entries in /proc
130	..............................................................................
131	 File		Content
132	 clear_refs	Clears page referenced bits shown in smaps output
133	 cmdline	Command line arguments
134	 cpu		Current and last cpu in which it was executed	(2.4)(smp)
135	 cwd		Link to the current working directory
136	 environ	Values of environment variables
137	 exe		Link to the executable of this process
138	 fd		Directory, which contains all file descriptors
139	 maps		Memory maps to executables and library files	(2.4)
140	 mem		Memory held by this process
141	 root		Link to the root directory of this process
142	 stat		Process status
143	 statm		Process memory status information
144	 status		Process status in human readable form
145	 wchan		Present with CONFIG_KALLSYMS=y: it shows the kernel function
146			symbol the task is blocked in - or "0" if not blocked.
147	 pagemap	Page table
148	 stack		Report full stack trace, enable via CONFIG_STACKTRACE
149	 smaps		an extension based on maps, showing the memory consumption of
150			each mapping and flags associated with it
151	 numa_maps	an extension based on maps, showing the memory locality and
152			binding policy as well as mem usage (in pages) of each mapping.
153	..............................................................................
154	
155	For example, to get the status information of a process, all you have to do is
156	read the file /proc/PID/status:
157	
158	  >cat /proc/self/status
159	  Name:   cat
160	  State:  R (running)
161	  Tgid:   5452
162	  Pid:    5452
163	  PPid:   743
164	  TracerPid:      0						(2.4)
165	  Uid:    501     501     501     501
166	  Gid:    100     100     100     100
167	  FDSize: 256
168	  Groups: 100 14 16
169	  VmPeak:     5004 kB
170	  VmSize:     5004 kB
171	  VmLck:         0 kB
172	  VmHWM:       476 kB
173	  VmRSS:       476 kB
174	  RssAnon:             352 kB
175	  RssFile:             120 kB
176	  RssShmem:              4 kB
177	  VmData:      156 kB
178	  VmStk:        88 kB
179	  VmExe:        68 kB
180	  VmLib:      1412 kB
181	  VmPTE:        20 kb
182	  VmSwap:        0 kB
183	  HugetlbPages:          0 kB
184	  Threads:        1
185	  SigQ:   0/28578
186	  SigPnd: 0000000000000000
187	  ShdPnd: 0000000000000000
188	  SigBlk: 0000000000000000
189	  SigIgn: 0000000000000000
190	  SigCgt: 0000000000000000
191	  CapInh: 00000000fffffeff
192	  CapPrm: 0000000000000000
193	  CapEff: 0000000000000000
194	  CapBnd: ffffffffffffffff
195	  NoNewPrivs:     0
196	  Seccomp:        0
197	  voluntary_ctxt_switches:        0
198	  nonvoluntary_ctxt_switches:     1
199	
200	This shows you nearly the same information you would get if you viewed it with
201	the ps  command.  In  fact,  ps  uses  the  proc  file  system  to  obtain its
202	information.  But you get a more detailed  view of the  process by reading the
203	file /proc/PID/status. It fields are described in table 1-2.
204	
205	The  statm  file  contains  more  detailed  information about the process
206	memory usage. Its seven fields are explained in Table 1-3.  The stat file
207	contains details information about the process itself.  Its fields are
208	explained in Table 1-4.
209	
210	(for SMP CONFIG users)
211	For making accounting scalable, RSS related information are handled in an
212	asynchronous manner and the value may not be very precise. To see a precise
213	snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
214	It's slow but very precise.
215	
216	Table 1-2: Contents of the status files (as of 4.8)
217	..............................................................................
218	 Field                       Content
219	 Name                        filename of the executable
220	 Umask                       file mode creation mask
221	 State                       state (R is running, S is sleeping, D is sleeping
222	                             in an uninterruptible wait, Z is zombie,
223				     T is traced or stopped)
224	 Tgid                        thread group ID
225	 Ngid                        NUMA group ID (0 if none)
226	 Pid                         process id
227	 PPid                        process id of the parent process
228	 TracerPid                   PID of process tracing this process (0 if not)
229	 Uid                         Real, effective, saved set, and  file system UIDs
230	 Gid                         Real, effective, saved set, and  file system GIDs
231	 FDSize                      number of file descriptor slots currently allocated
232	 Groups                      supplementary group list
233	 NStgid                      descendant namespace thread group ID hierarchy
234	 NSpid                       descendant namespace process ID hierarchy
235	 NSpgid                      descendant namespace process group ID hierarchy
236	 NSsid                       descendant namespace session ID hierarchy
237	 VmPeak                      peak virtual memory size
238	 VmSize                      total program size
239	 VmLck                       locked memory size
240	 VmPin                       pinned memory size
241	 VmHWM                       peak resident set size ("high water mark")
242	 VmRSS                       size of memory portions. It contains the three
243	                             following parts (VmRSS = RssAnon + RssFile + RssShmem)
244	 RssAnon                     size of resident anonymous memory
245	 RssFile                     size of resident file mappings
246	 RssShmem                    size of resident shmem memory (includes SysV shm,
247	                             mapping of tmpfs and shared anonymous mappings)
248	 VmData                      size of private data segments
249	 VmStk                       size of stack segments
250	 VmExe                       size of text segment
251	 VmLib                       size of shared library code
252	 VmPTE                       size of page table entries
253	 VmPMD                       size of second level page tables
254	 VmSwap                      amount of swap used by anonymous private data
255	                             (shmem swap usage is not included)
256	 HugetlbPages                size of hugetlb memory portions
257	 Threads                     number of threads
258	 SigQ                        number of signals queued/max. number for queue
259	 SigPnd                      bitmap of pending signals for the thread
260	 ShdPnd                      bitmap of shared pending signals for the process
261	 SigBlk                      bitmap of blocked signals
262	 SigIgn                      bitmap of ignored signals
263	 SigCgt                      bitmap of caught signals
264	 CapInh                      bitmap of inheritable capabilities
265	 CapPrm                      bitmap of permitted capabilities
266	 CapEff                      bitmap of effective capabilities
267	 CapBnd                      bitmap of capabilities bounding set
268	 NoNewPrivs                  no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
269	 Seccomp                     seccomp mode, like prctl(PR_GET_SECCOMP, ...)
270	 Cpus_allowed                mask of CPUs on which this process may run
271	 Cpus_allowed_list           Same as previous, but in "list format"
272	 Mems_allowed                mask of memory nodes allowed to this process
273	 Mems_allowed_list           Same as previous, but in "list format"
274	 voluntary_ctxt_switches     number of voluntary context switches
275	 nonvoluntary_ctxt_switches  number of non voluntary context switches
276	..............................................................................
277	
278	Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
279	..............................................................................
280	 Field    Content
281	 size     total program size (pages)		(same as VmSize in status)
282	 resident size of memory portions (pages)	(same as VmRSS in status)
283	 shared   number of pages that are shared	(i.e. backed by a file, same
284							as RssFile+RssShmem in status)
285	 trs      number of pages that are 'code'	(not including libs; broken,
286								includes data segment)
287	 lrs      number of pages of library		(always 0 on 2.6)
288	 drs      number of pages of data/stack		(including libs; broken,
289								includes library text)
290	 dt       number of dirty pages			(always 0 on 2.6)
291	..............................................................................
292	
293	
294	Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
295	..............................................................................
296	 Field          Content
297	  pid           process id
298	  tcomm         filename of the executable
299	  state         state (R is running, S is sleeping, D is sleeping in an
300	                uninterruptible wait, Z is zombie, T is traced or stopped)
301	  ppid          process id of the parent process
302	  pgrp          pgrp of the process
303	  sid           session id
304	  tty_nr        tty the process uses
305	  tty_pgrp      pgrp of the tty
306	  flags         task flags
307	  min_flt       number of minor faults
308	  cmin_flt      number of minor faults with child's
309	  maj_flt       number of major faults
310	  cmaj_flt      number of major faults with child's
311	  utime         user mode jiffies
312	  stime         kernel mode jiffies
313	  cutime        user mode jiffies with child's
314	  cstime        kernel mode jiffies with child's
315	  priority      priority level
316	  nice          nice level
317	  num_threads   number of threads
318	  it_real_value	(obsolete, always 0)
319	  start_time    time the process started after system boot
320	  vsize         virtual memory size
321	  rss           resident set memory size
322	  rsslim        current limit in bytes on the rss
323	  start_code    address above which program text can run
324	  end_code      address below which program text can run
325	  start_stack   address of the start of the main process stack
326	  esp           current value of ESP
327	  eip           current value of EIP
328	  pending       bitmap of pending signals
329	  blocked       bitmap of blocked signals
330	  sigign        bitmap of ignored signals
331	  sigcatch      bitmap of caught signals
332	  0		(place holder, used to be the wchan address, use /proc/PID/wchan instead)
333	  0             (place holder)
334	  0             (place holder)
335	  exit_signal   signal to send to parent thread on exit
336	  task_cpu      which CPU the task is scheduled on
337	  rt_priority   realtime priority
338	  policy        scheduling policy (man sched_setscheduler)
339	  blkio_ticks   time spent waiting for block IO
340	  gtime         guest time of the task in jiffies
341	  cgtime        guest time of the task children in jiffies
342	  start_data    address above which program data+bss is placed
343	  end_data      address below which program data+bss is placed
344	  start_brk     address above which program heap can be expanded with brk()
345	  arg_start     address above which program command line is placed
346	  arg_end       address below which program command line is placed
347	  env_start     address above which program environment is placed
348	  env_end       address below which program environment is placed
349	  exit_code     the thread's exit_code in the form reported by the waitpid system call
350	..............................................................................
351	
352	The /proc/PID/maps file containing the currently mapped memory regions and
353	their access permissions.
354	
355	The format is:
356	
357	address           perms offset  dev   inode      pathname
358	
359	08048000-08049000 r-xp 00000000 03:00 8312       /opt/test
360	08049000-0804a000 rw-p 00001000 03:00 8312       /opt/test
361	0804a000-0806b000 rw-p 00000000 00:00 0          [heap]
362	a7cb1000-a7cb2000 ---p 00000000 00:00 0
363	a7cb2000-a7eb2000 rw-p 00000000 00:00 0
364	a7eb2000-a7eb3000 ---p 00000000 00:00 0
365	a7eb3000-a7ed5000 rw-p 00000000 00:00 0
366	a7ed5000-a8008000 r-xp 00000000 03:00 4222       /lib/libc.so.6
367	a8008000-a800a000 r--p 00133000 03:00 4222       /lib/libc.so.6
368	a800a000-a800b000 rw-p 00135000 03:00 4222       /lib/libc.so.6
369	a800b000-a800e000 rw-p 00000000 00:00 0
370	a800e000-a8022000 r-xp 00000000 03:00 14462      /lib/libpthread.so.0
371	a8022000-a8023000 r--p 00013000 03:00 14462      /lib/libpthread.so.0
372	a8023000-a8024000 rw-p 00014000 03:00 14462      /lib/libpthread.so.0
373	a8024000-a8027000 rw-p 00000000 00:00 0
374	a8027000-a8043000 r-xp 00000000 03:00 8317       /lib/ld-linux.so.2
375	a8043000-a8044000 r--p 0001b000 03:00 8317       /lib/ld-linux.so.2
376	a8044000-a8045000 rw-p 0001c000 03:00 8317       /lib/ld-linux.so.2
377	aff35000-aff4a000 rw-p 00000000 00:00 0          [stack]
378	ffffe000-fffff000 r-xp 00000000 00:00 0          [vdso]
379	
380	where "address" is the address space in the process that it occupies, "perms"
381	is a set of permissions:
382	
383	 r = read
384	 w = write
385	 x = execute
386	 s = shared
387	 p = private (copy on write)
388	
389	"offset" is the offset into the mapping, "dev" is the device (major:minor), and
390	"inode" is the inode  on that device.  0 indicates that  no inode is associated
391	with the memory region, as the case would be with BSS (uninitialized data).
392	The "pathname" shows the name associated file for this mapping.  If the mapping
393	is not associated with a file:
394	
395	 [heap]                   = the heap of the program
396	 [stack]                  = the stack of the main process
397	 [vdso]                   = the "virtual dynamic shared object",
398	                            the kernel system call handler
399	
400	 or if empty, the mapping is anonymous.
401	
402	The /proc/PID/smaps is an extension based on maps, showing the memory
403	consumption for each of the process's mappings. For each of mappings there
404	is a series of lines such as the following:
405	
406	08048000-080bc000 r-xp 00000000 03:02 13130      /bin/bash
407	Size:               1084 kB
408	Rss:                 892 kB
409	Pss:                 374 kB
410	Shared_Clean:        892 kB
411	Shared_Dirty:          0 kB
412	Private_Clean:         0 kB
413	Private_Dirty:         0 kB
414	Referenced:          892 kB
415	Anonymous:             0 kB
416	LazyFree:              0 kB
417	AnonHugePages:         0 kB
418	ShmemPmdMapped:        0 kB
419	Shared_Hugetlb:        0 kB
420	Private_Hugetlb:       0 kB
421	Swap:                  0 kB
422	SwapPss:               0 kB
423	KernelPageSize:        4 kB
424	MMUPageSize:           4 kB
425	Locked:                0 kB
426	VmFlags: rd ex mr mw me dw
427	
428	the first of these lines shows the same information as is displayed for the
429	mapping in /proc/PID/maps.  The remaining lines show the size of the mapping
430	(size), the amount of the mapping that is currently resident in RAM (RSS), the
431	process' proportional share of this mapping (PSS), the number of clean and
432	dirty private pages in the mapping.
433	
434	The "proportional set size" (PSS) of a process is the count of pages it has
435	in memory, where each page is divided by the number of processes sharing it.
436	So if a process has 1000 pages all to itself, and 1000 shared with one other
437	process, its PSS will be 1500.
438	Note that even a page which is part of a MAP_SHARED mapping, but has only
439	a single pte mapped, i.e.  is currently used by only one process, is accounted
440	as private and not as shared.
441	"Referenced" indicates the amount of memory currently marked as referenced or
442	accessed.
443	"Anonymous" shows the amount of memory that does not belong to any file.  Even
444	a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
445	and a page is modified, the file page is replaced by a private anonymous copy.
446	"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
447	The memory isn't freed immediately with madvise(). It's freed in memory
448	pressure if the memory is clean. Please note that the printed value might
449	be lower than the real value due to optimizations used in the current
450	implementation. If this is not desirable please file a bug report.
451	"AnonHugePages" shows the ammount of memory backed by transparent hugepage.
452	"ShmemPmdMapped" shows the ammount of shared (shmem/tmpfs) memory backed by
453	huge pages.
454	"Shared_Hugetlb" and "Private_Hugetlb" show the ammounts of memory backed by
455	hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
456	reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.
457	"Swap" shows how much would-be-anonymous memory is also used, but out on swap.
458	For shmem mappings, "Swap" includes also the size of the mapped (and not
459	replaced by copy-on-write) part of the underlying shmem object out on swap.
460	"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
461	does not take into account swapped out page of underlying shmem objects.
462	"Locked" indicates whether the mapping is locked in memory or not.
463	
464	"VmFlags" field deserves a separate description. This member represents the kernel
465	flags associated with the particular virtual memory area in two letter encoded
466	manner. The codes are the following:
467	    rd  - readable
468	    wr  - writeable
469	    ex  - executable
470	    sh  - shared
471	    mr  - may read
472	    mw  - may write
473	    me  - may execute
474	    ms  - may share
475	    gd  - stack segment growns down
476	    pf  - pure PFN range
477	    dw  - disabled write to the mapped file
478	    lo  - pages are locked in memory
479	    io  - memory mapped I/O area
480	    sr  - sequential read advise provided
481	    rr  - random read advise provided
482	    dc  - do not copy area on fork
483	    de  - do not expand area on remapping
484	    ac  - area is accountable
485	    nr  - swap space is not reserved for the area
486	    ht  - area uses huge tlb pages
487	    ar  - architecture specific flag
488	    dd  - do not include area into core dump
489	    sd  - soft-dirty flag
490	    mm  - mixed map area
491	    hg  - huge page advise flag
492	    nh  - no-huge page advise flag
493	    mg  - mergable advise flag
494	
495	Note that there is no guarantee that every flag and associated mnemonic will
496	be present in all further kernel releases. Things get changed, the flags may
497	be vanished or the reverse -- new added.
498	
499	This file is only present if the CONFIG_MMU kernel configuration option is
500	enabled.
501	
502	Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
503	output can be achieved only in the single read call).
504	This typically manifests when doing partial reads of these files while the
505	memory map is being modified.  Despite the races, we do provide the following
506	guarantees:
507	
508	1) The mapped addresses never go backwards, which implies no two
509	   regions will ever overlap.
510	2) If there is something at a given vaddr during the entirety of the
511	   life of the smaps/maps walk, there will be some output for it.
512	
513	
514	The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
515	bits on both physical and virtual pages associated with a process, and the
516	soft-dirty bit on pte (see Documentation/vm/soft-dirty.txt for details).
517	To clear the bits for all the pages associated with the process
518	    > echo 1 > /proc/PID/clear_refs
519	
520	To clear the bits for the anonymous pages associated with the process
521	    > echo 2 > /proc/PID/clear_refs
522	
523	To clear the bits for the file mapped pages associated with the process
524	    > echo 3 > /proc/PID/clear_refs
525	
526	To clear the soft-dirty bit
527	    > echo 4 > /proc/PID/clear_refs
528	
529	To reset the peak resident set size ("high water mark") to the process's
530	current value:
531	    > echo 5 > /proc/PID/clear_refs
532	
533	Any other value written to /proc/PID/clear_refs will have no effect.
534	
535	The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
536	using /proc/kpageflags and number of times a page is mapped using
537	/proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
538	
539	The /proc/pid/numa_maps is an extension based on maps, showing the memory
540	locality and binding policy, as well as the memory usage (in pages) of
541	each mapping. The output follows a general format where mapping details get
542	summarized separated by blank spaces, one mapping per each file line:
543	
544	address   policy    mapping details
545	
546	00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
547	00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
548	3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
549	320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
550	3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
551	3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
552	3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
553	320698b000 default file=/lib64/libc-2.12.so
554	3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
555	3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
556	3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
557	7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
558	7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
559	7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
560	7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
561	7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4
562	
563	Where:
564	"address" is the starting address for the mapping;
565	"policy" reports the NUMA memory policy set for the mapping (see vm/numa_memory_policy.txt);
566	"mapping details" summarizes mapping data such as mapping type, page usage counters,
567	node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
568	size, in KB, that is backing the mapping up.
569	
570	1.2 Kernel data
571	---------------
572	
573	Similar to  the  process entries, the kernel data files give information about
574	the running kernel. The files used to obtain this information are contained in
575	/proc and  are  listed  in Table 1-5. Not all of these will be present in your
576	system. It  depends  on the kernel configuration and the loaded modules, which
577	files are there, and which are missing.
578	
579	Table 1-5: Kernel info in /proc
580	..............................................................................
581	 File        Content                                           
582	 apm         Advanced power management info                    
583	 buddyinfo   Kernel memory allocator information (see text)	(2.5)
584	 bus         Directory containing bus specific information     
585	 cmdline     Kernel command line                               
586	 cpuinfo     Info about the CPU                                
587	 devices     Available devices (block and character)           
588	 dma         Used DMS channels                                 
589	 filesystems Supported filesystems                             
590	 driver	     Various drivers grouped here, currently rtc (2.4)
591	 execdomains Execdomains, related to security			(2.4)
592	 fb	     Frame Buffer devices				(2.4)
593	 fs	     File system parameters, currently nfs/exports	(2.4)
594	 ide         Directory containing info about the IDE subsystem 
595	 interrupts  Interrupt usage                                   
596	 iomem	     Memory map						(2.4)
597	 ioports     I/O port usage                                    
598	 irq	     Masks for irq to cpu affinity			(2.4)(smp?)
599	 isapnp	     ISA PnP (Plug&Play) Info				(2.4)
600	 kcore       Kernel core image (can be ELF or A.OUT(deprecated in 2.4))   
601	 kmsg        Kernel messages                                   
602	 ksyms       Kernel symbol table                               
603	 loadavg     Load average of last 1, 5 & 15 minutes                
604	 locks       Kernel locks                                      
605	 meminfo     Memory info                                       
606	 misc        Miscellaneous                                     
607	 modules     List of loaded modules                            
608	 mounts      Mounted filesystems                               
609	 net         Networking info (see text)                        
610	 pagetypeinfo Additional page allocator information (see text)  (2.5)
611	 partitions  Table of partitions known to the system           
612	 pci	     Deprecated info of PCI bus (new way -> /proc/bus/pci/,
613	             decoupled by lspci					(2.4)
614	 rtc         Real time clock                                   
615	 scsi        SCSI info (see text)                              
616	 slabinfo    Slab pool info                                    
617	 softirqs    softirq usage
618	 stat        Overall statistics                                
619	 swaps       Swap space utilization                            
620	 sys         See chapter 2                                     
621	 sysvipc     Info of SysVIPC Resources (msg, sem, shm)		(2.4)
622	 tty	     Info of tty drivers
623	 uptime      Wall clock since boot, combined idle time of all cpus
624	 version     Kernel version                                    
625	 video	     bttv info of video resources			(2.4)
626	 vmallocinfo Show vmalloced areas
627	..............................................................................
628	
629	You can,  for  example,  check  which interrupts are currently in use and what
630	they are used for by looking in the file /proc/interrupts:
631	
632	  > cat /proc/interrupts 
633	             CPU0        
634	    0:    8728810          XT-PIC  timer 
635	    1:        895          XT-PIC  keyboard 
636	    2:          0          XT-PIC  cascade 
637	    3:     531695          XT-PIC  aha152x 
638	    4:    2014133          XT-PIC  serial 
639	    5:      44401          XT-PIC  pcnet_cs 
640	    8:          2          XT-PIC  rtc 
641	   11:          8          XT-PIC  i82365 
642	   12:     182918          XT-PIC  PS/2 Mouse 
643	   13:          1          XT-PIC  fpu 
644	   14:    1232265          XT-PIC  ide0 
645	   15:          7          XT-PIC  ide1 
646	  NMI:          0 
647	
648	In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
649	output of a SMP machine):
650	
651	  > cat /proc/interrupts 
652	
653	             CPU0       CPU1       
654	    0:    1243498    1214548    IO-APIC-edge  timer
655	    1:       8949       8958    IO-APIC-edge  keyboard
656	    2:          0          0          XT-PIC  cascade
657	    5:      11286      10161    IO-APIC-edge  soundblaster
658	    8:          1          0    IO-APIC-edge  rtc
659	    9:      27422      27407    IO-APIC-edge  3c503
660	   12:     113645     113873    IO-APIC-edge  PS/2 Mouse
661	   13:          0          0          XT-PIC  fpu
662	   14:      22491      24012    IO-APIC-edge  ide0
663	   15:       2183       2415    IO-APIC-edge  ide1
664	   17:      30564      30414   IO-APIC-level  eth0
665	   18:        177        164   IO-APIC-level  bttv
666	  NMI:    2457961    2457959 
667	  LOC:    2457882    2457881 
668	  ERR:       2155
669	
670	NMI is incremented in this case because every timer interrupt generates a NMI
671	(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
672	
673	LOC is the local interrupt counter of the internal APIC of every CPU.
674	
675	ERR is incremented in the case of errors in the IO-APIC bus (the bus that
676	connects the CPUs in a SMP system. This means that an error has been detected,
677	the IO-APIC automatically retry the transmission, so it should not be a big
678	problem, but you should read the SMP-FAQ.
679	
680	In 2.6.2* /proc/interrupts was expanded again.  This time the goal was for
681	/proc/interrupts to display every IRQ vector in use by the system, not
682	just those considered 'most important'.  The new vectors are:
683	
684	  THR -- interrupt raised when a machine check threshold counter
685	  (typically counting ECC corrected errors of memory or cache) exceeds
686	  a configurable threshold.  Only available on some systems.
687	
688	  TRM -- a thermal event interrupt occurs when a temperature threshold
689	  has been exceeded for the CPU.  This interrupt may also be generated
690	  when the temperature drops back to normal.
691	
692	  SPU -- a spurious interrupt is some interrupt that was raised then lowered
693	  by some IO device before it could be fully processed by the APIC.  Hence
694	  the APIC sees the interrupt but does not know what device it came from.
695	  For this case the APIC will generate the interrupt with a IRQ vector
696	  of 0xff. This might also be generated by chipset bugs.
697	
698	  RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
699	  sent from one CPU to another per the needs of the OS.  Typically,
700	  their statistics are used by kernel developers and interested users to
701	  determine the occurrence of interrupts of the given type.
702	
703	The above IRQ vectors are displayed only when relevant.  For example,
704	the threshold vector does not exist on x86_64 platforms.  Others are
705	suppressed when the system is a uniprocessor.  As of this writing, only
706	i386 and x86_64 platforms support the new IRQ vector displays.
707	
708	Of some interest is the introduction of the /proc/irq directory to 2.4.
709	It could be used to set IRQ to CPU affinity, this means that you can "hook" an
710	IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
711	irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
712	prof_cpu_mask.
713	
714	For example 
715	  > ls /proc/irq/
716	  0  10  12  14  16  18  2  4  6  8  prof_cpu_mask
717	  1  11  13  15  17  19  3  5  7  9  default_smp_affinity
718	  > ls /proc/irq/0/
719	  smp_affinity
720	
721	smp_affinity is a bitmask, in which you can specify which CPUs can handle the
722	IRQ, you can set it by doing:
723	
724	  > echo 1 > /proc/irq/10/smp_affinity
725	
726	This means that only the first CPU will handle the IRQ, but you can also echo
727	5 which means that only the first and third CPU can handle the IRQ.
728	
729	The contents of each smp_affinity file is the same by default:
730	
731	  > cat /proc/irq/0/smp_affinity
732	  ffffffff
733	
734	There is an alternate interface, smp_affinity_list which allows specifying
735	a cpu range instead of a bitmask:
736	
737	  > cat /proc/irq/0/smp_affinity_list
738	  1024-1031
739	
740	The default_smp_affinity mask applies to all non-active IRQs, which are the
741	IRQs which have not yet been allocated/activated, and hence which lack a
742	/proc/irq/[0-9]* directory.
743	
744	The node file on an SMP system shows the node to which the device using the IRQ
745	reports itself as being attached. This hardware locality information does not
746	include information about any possible driver locality preference.
747	
748	prof_cpu_mask specifies which CPUs are to be profiled by the system wide
749	profiler. Default value is ffffffff (all cpus if there are only 32 of them).
750	
751	The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
752	between all the CPUs which are allowed to handle it. As usual the kernel has
753	more info than you and does a better job than you, so the defaults are the
754	best choice for almost everyone.  [Note this applies only to those IO-APIC's
755	that support "Round Robin" interrupt distribution.]
756	
757	There are  three  more  important subdirectories in /proc: net, scsi, and sys.
758	The general  rule  is  that  the  contents,  or  even  the  existence of these
759	directories, depend  on your kernel configuration. If SCSI is not enabled, the
760	directory scsi  may  not  exist. The same is true with the net, which is there
761	only when networking support is present in the running kernel.
762	
763	The slabinfo  file  gives  information  about  memory usage at the slab level.
764	Linux uses  slab  pools for memory management above page level in version 2.2.
765	Commonly used  objects  have  their  own  slab  pool (such as network buffers,
766	directory cache, and so on).
767	
768	..............................................................................
769	
770	> cat /proc/buddyinfo
771	
772	Node 0, zone      DMA      0      4      5      4      4      3 ...
773	Node 0, zone   Normal      1      0      0      1    101      8 ...
774	Node 0, zone  HighMem      2      0      0      1      1      0 ...
775	
776	External fragmentation is a problem under some workloads, and buddyinfo is a
777	useful tool for helping diagnose these problems.  Buddyinfo will give you a 
778	clue as to how big an area you can safely allocate, or why a previous
779	allocation failed.
780	
781	Each column represents the number of pages of a certain order which are 
782	available.  In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 
783	ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 
784	available in ZONE_NORMAL, etc... 
785	
786	More information relevant to external fragmentation can be found in
787	pagetypeinfo.
788	
789	> cat /proc/pagetypeinfo
790	Page block order: 9
791	Pages per block:  512
792	
793	Free pages count per migrate type at order       0      1      2      3      4      5      6      7      8      9     10
794	Node    0, zone      DMA, type    Unmovable      0      0      0      1      1      1      1      1      1      1      0
795	Node    0, zone      DMA, type  Reclaimable      0      0      0      0      0      0      0      0      0      0      0
796	Node    0, zone      DMA, type      Movable      1      1      2      1      2      1      1      0      1      0      2
797	Node    0, zone      DMA, type      Reserve      0      0      0      0      0      0      0      0      0      1      0
798	Node    0, zone      DMA, type      Isolate      0      0      0      0      0      0      0      0      0      0      0
799	Node    0, zone    DMA32, type    Unmovable    103     54     77      1      1      1     11      8      7      1      9
800	Node    0, zone    DMA32, type  Reclaimable      0      0      2      1      0      0      0      0      1      0      0
801	Node    0, zone    DMA32, type      Movable    169    152    113     91     77     54     39     13      6      1    452
802	Node    0, zone    DMA32, type      Reserve      1      2      2      2      2      0      1      1      1      1      0
803	Node    0, zone    DMA32, type      Isolate      0      0      0      0      0      0      0      0      0      0      0
804	
805	Number of blocks type     Unmovable  Reclaimable      Movable      Reserve      Isolate
806	Node 0, zone      DMA            2            0            5            1            0
807	Node 0, zone    DMA32           41            6          967            2            0
808	
809	Fragmentation avoidance in the kernel works by grouping pages of different
810	migrate types into the same contiguous regions of memory called page blocks.
811	A page block is typically the size of the default hugepage size e.g. 2MB on
812	X86-64. By keeping pages grouped based on their ability to move, the kernel
813	can reclaim pages within a page block to satisfy a high-order allocation.
814	
815	The pagetypinfo begins with information on the size of a page block. It
816	then gives the same type of information as buddyinfo except broken down
817	by migrate-type and finishes with details on how many page blocks of each
818	type exist.
819	
820	If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
821	from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
822	make an estimate of the likely number of huge pages that can be allocated
823	at a given point in time. All the "Movable" blocks should be allocatable
824	unless memory has been mlock()'d. Some of the Reclaimable blocks should
825	also be allocatable although a lot of filesystem metadata may have to be
826	reclaimed to achieve this.
827	
828	..............................................................................
829	
830	meminfo:
831	
832	Provides information about distribution and utilization of memory.  This
833	varies by architecture and compile options.  The following is from a
834	16GB PIII, which has highmem enabled.  You may not have all of these fields.
835	
836	> cat /proc/meminfo
837	
838	MemTotal:     16344972 kB
839	MemFree:      13634064 kB
840	MemAvailable: 14836172 kB
841	Buffers:          3656 kB
842	Cached:        1195708 kB
843	SwapCached:          0 kB
844	Active:         891636 kB
845	Inactive:      1077224 kB
846	HighTotal:    15597528 kB
847	HighFree:     13629632 kB
848	LowTotal:       747444 kB
849	LowFree:          4432 kB
850	SwapTotal:           0 kB
851	SwapFree:            0 kB
852	Dirty:             968 kB
853	Writeback:           0 kB
854	AnonPages:      861800 kB
855	Mapped:         280372 kB
856	Shmem:             644 kB
857	Slab:           284364 kB
858	SReclaimable:   159856 kB
859	SUnreclaim:     124508 kB
860	PageTables:      24448 kB
861	NFS_Unstable:        0 kB
862	Bounce:              0 kB
863	WritebackTmp:        0 kB
864	CommitLimit:   7669796 kB
865	Committed_AS:   100056 kB
866	VmallocTotal:   112216 kB
867	VmallocUsed:       428 kB
868	VmallocChunk:   111088 kB
869	AnonHugePages:   49152 kB
870	ShmemHugePages:      0 kB
871	ShmemPmdMapped:      0 kB
872	
873	
874	    MemTotal: Total usable ram (i.e. physical ram minus a few reserved
875	              bits and the kernel binary code)
876	     MemFree: The sum of LowFree+HighFree
877	MemAvailable: An estimate of how much memory is available for starting new
878	              applications, without swapping. Calculated from MemFree,
879	              SReclaimable, the size of the file LRU lists, and the low
880	              watermarks in each zone.
881	              The estimate takes into account that the system needs some
882	              page cache to function well, and that not all reclaimable
883	              slab will be reclaimable, due to items being in use. The
884	              impact of those factors will vary from system to system.
885	     Buffers: Relatively temporary storage for raw disk blocks
886	              shouldn't get tremendously large (20MB or so)
887	      Cached: in-memory cache for files read from the disk (the
888	              pagecache).  Doesn't include SwapCached
889	  SwapCached: Memory that once was swapped out, is swapped back in but
890	              still also is in the swapfile (if memory is needed it
891	              doesn't need to be swapped out AGAIN because it is already
892	              in the swapfile. This saves I/O)
893	      Active: Memory that has been used more recently and usually not
894	              reclaimed unless absolutely necessary.
895	    Inactive: Memory which has been less recently used.  It is more
896	              eligible to be reclaimed for other purposes
897	   HighTotal:
898	    HighFree: Highmem is all memory above ~860MB of physical memory
899	              Highmem areas are for use by userspace programs, or
900	              for the pagecache.  The kernel must use tricks to access
901	              this memory, making it slower to access than lowmem.
902	    LowTotal:
903	     LowFree: Lowmem is memory which can be used for everything that
904	              highmem can be used for, but it is also available for the
905	              kernel's use for its own data structures.  Among many
906	              other things, it is where everything from the Slab is
907	              allocated.  Bad things happen when you're out of lowmem.
908	   SwapTotal: total amount of swap space available
909	    SwapFree: Memory which has been evicted from RAM, and is temporarily
910	              on the disk
911	       Dirty: Memory which is waiting to get written back to the disk
912	   Writeback: Memory which is actively being written back to the disk
913	   AnonPages: Non-file backed pages mapped into userspace page tables
914	AnonHugePages: Non-file backed huge pages mapped into userspace page tables
915	      Mapped: files which have been mmaped, such as libraries
916	       Shmem: Total memory used by shared memory (shmem) and tmpfs
917	ShmemHugePages: Memory used by shared memory (shmem) and tmpfs allocated
918	              with huge pages
919	ShmemPmdMapped: Shared memory mapped into userspace with huge pages
920	        Slab: in-kernel data structures cache
921	SReclaimable: Part of Slab, that might be reclaimed, such as caches
922	  SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
923	  PageTables: amount of memory dedicated to the lowest level of page
924	              tables.
925	NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
926		      storage
927	      Bounce: Memory used for block device "bounce buffers"
928	WritebackTmp: Memory used by FUSE for temporary writeback buffers
929	 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
930	              this is the total amount of  memory currently available to
931	              be allocated on the system. This limit is only adhered to
932	              if strict overcommit accounting is enabled (mode 2 in
933	              'vm.overcommit_memory').
934	              The CommitLimit is calculated with the following formula:
935	              CommitLimit = ([total RAM pages] - [total huge TLB pages]) *
936	                             overcommit_ratio / 100 + [total swap pages]
937	              For example, on a system with 1G of physical RAM and 7G
938	              of swap with a `vm.overcommit_ratio` of 30 it would
939	              yield a CommitLimit of 7.3G.
940	              For more details, see the memory overcommit documentation
941	              in vm/overcommit-accounting.
942	Committed_AS: The amount of memory presently allocated on the system.
943	              The committed memory is a sum of all of the memory which
944	              has been allocated by processes, even if it has not been
945	              "used" by them as of yet. A process which malloc()'s 1G
946	              of memory, but only touches 300M of it will show up as
947		      using 1G. This 1G is memory which has been "committed" to
948	              by the VM and can be used at any time by the allocating
949	              application. With strict overcommit enabled on the system
950	              (mode 2 in 'vm.overcommit_memory'),allocations which would
951	              exceed the CommitLimit (detailed above) will not be permitted.
952	              This is useful if one needs to guarantee that processes will
953	              not fail due to lack of memory once that memory has been
954	              successfully allocated.
955	VmallocTotal: total size of vmalloc memory area
956	 VmallocUsed: amount of vmalloc area which is used
957	VmallocChunk: largest contiguous block of vmalloc area which is free
958	
959	..............................................................................
960	
961	vmallocinfo:
962	
963	Provides information about vmalloced/vmaped areas. One line per area,
964	containing the virtual address range of the area, size in bytes,
965	caller information of the creator, and optional information depending
966	on the kind of area :
967	
968	 pages=nr    number of pages
969	 phys=addr   if a physical address was specified
970	 ioremap     I/O mapping (ioremap() and friends)
971	 vmalloc     vmalloc() area
972	 vmap        vmap()ed pages
973	 user        VM_USERMAP area
974	 vpages      buffer for pages pointers was vmalloced (huge area)
975	 N<node>=nr  (Only on NUMA kernels)
976	             Number of pages allocated on memory node <node>
977	
978	> cat /proc/vmallocinfo
979	0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
980	  /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
981	0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
982	  /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
983	0xffffc20000302000-0xffffc20000304000    8192 acpi_tb_verify_table+0x21/0x4f...
984	  phys=7fee8000 ioremap
985	0xffffc20000304000-0xffffc20000307000   12288 acpi_tb_verify_table+0x21/0x4f...
986	  phys=7fee7000 ioremap
987	0xffffc2000031d000-0xffffc2000031f000    8192 init_vdso_vars+0x112/0x210
988	0xffffc2000031f000-0xffffc2000032b000   49152 cramfs_uncompress_init+0x2e ...
989	  /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
990	0xffffc2000033a000-0xffffc2000033d000   12288 sys_swapon+0x640/0xac0      ...
991	  pages=2 vmalloc N1=2
992	0xffffc20000347000-0xffffc2000034c000   20480 xt_alloc_table_info+0xfe ...
993	  /0x130 [x_tables] pages=4 vmalloc N0=4
994	0xffffffffa0000000-0xffffffffa000f000   61440 sys_init_module+0xc27/0x1d00 ...
995	   pages=14 vmalloc N2=14
996	0xffffffffa000f000-0xffffffffa0014000   20480 sys_init_module+0xc27/0x1d00 ...
997	   pages=4 vmalloc N1=4
998	0xffffffffa0014000-0xffffffffa0017000   12288 sys_init_module+0xc27/0x1d00 ...
999	   pages=2 vmalloc N1=2
1000	0xffffffffa0017000-0xffffffffa0022000   45056 sys_init_module+0xc27/0x1d00 ...
1001	   pages=10 vmalloc N0=10
1002	
1003	..............................................................................
1004	
1005	softirqs:
1006	
1007	Provides counts of softirq handlers serviced since boot time, for each cpu.
1008	
1009	> cat /proc/softirqs
1010	                CPU0       CPU1       CPU2       CPU3
1011	      HI:          0          0          0          0
1012	   TIMER:      27166      27120      27097      27034
1013	  NET_TX:          0          0          0         17
1014	  NET_RX:         42          0          0         39
1015	   BLOCK:          0          0        107       1121
1016	 TASKLET:          0          0          0        290
1017	   SCHED:      27035      26983      26971      26746
1018	 HRTIMER:          0          0          0          0
1019	     RCU:       1678       1769       2178       2250
1020	
1021	
1022	1.3 IDE devices in /proc/ide
1023	----------------------------
1024	
1025	The subdirectory /proc/ide contains information about all IDE devices of which
1026	the kernel  is  aware.  There is one subdirectory for each IDE controller, the
1027	file drivers  and a link for each IDE device, pointing to the device directory
1028	in the controller specific subtree.
1029	
1030	The file  drivers  contains general information about the drivers used for the
1031	IDE devices:
1032	
1033	  > cat /proc/ide/drivers
1034	  ide-cdrom version 4.53
1035	  ide-disk version 1.08
1036	
1037	More detailed  information  can  be  found  in  the  controller  specific
1038	subdirectories. These  are  named  ide0,  ide1  and  so  on.  Each  of  these
1039	directories contains the files shown in table 1-6.
1040	
1041	
1042	Table 1-6: IDE controller info in  /proc/ide/ide?
1043	..............................................................................
1044	 File    Content                                 
1045	 channel IDE channel (0 or 1)                    
1046	 config  Configuration (only for PCI/IDE bridge) 
1047	 mate    Mate name                               
1048	 model   Type/Chipset of IDE controller          
1049	..............................................................................
1050	
1051	Each device  connected  to  a  controller  has  a separate subdirectory in the
1052	controllers directory.  The  files  listed in table 1-7 are contained in these
1053	directories.
1054	
1055	
1056	Table 1-7: IDE device information
1057	..............................................................................
1058	 File             Content                                    
1059	 cache            The cache                                  
1060	 capacity         Capacity of the medium (in 512Byte blocks) 
1061	 driver           driver and version                         
1062	 geometry         physical and logical geometry              
1063	 identify         device identify block                      
1064	 media            media type                                 
1065	 model            device identifier                          
1066	 settings         device setup                               
1067	 smart_thresholds IDE disk management thresholds             
1068	 smart_values     IDE disk management values                 
1069	..............................................................................
1070	
1071	The most  interesting  file is settings. This file contains a nice overview of
1072	the drive parameters:
1073	
1074	  # cat /proc/ide/ide0/hda/settings 
1075	  name                    value           min             max             mode 
1076	  ----                    -----           ---             ---             ---- 
1077	  bios_cyl                526             0               65535           rw 
1078	  bios_head               255             0               255             rw 
1079	  bios_sect               63              0               63              rw 
1080	  breada_readahead        4               0               127             rw 
1081	  bswap                   0               0               1               r 
1082	  file_readahead          72              0               2097151         rw 
1083	  io_32bit                0               0               3               rw 
1084	  keepsettings            0               0               1               rw 
1085	  max_kb_per_request      122             1               127             rw 
1086	  multcount               0               0               8               rw 
1087	  nice1                   1               0               1               rw 
1088	  nowerr                  0               0               1               rw 
1089	  pio_mode                write-only      0               255             w 
1090	  slow                    0               0               1               rw 
1091	  unmaskirq               0               0               1               rw 
1092	  using_dma               0               0               1               rw 
1093	
1094	
1095	1.4 Networking info in /proc/net
1096	--------------------------------
1097	
1098	The subdirectory  /proc/net  follows  the  usual  pattern. Table 1-8 shows the
1099	additional values  you  get  for  IP  version 6 if you configure the kernel to
1100	support this. Table 1-9 lists the files and their meaning.
1101	
1102	
1103	Table 1-8: IPv6 info in /proc/net
1104	..............................................................................
1105	 File       Content                                               
1106	 udp6       UDP sockets (IPv6)                                    
1107	 tcp6       TCP sockets (IPv6)                                    
1108	 raw6       Raw device statistics (IPv6)                          
1109	 igmp6      IP multicast addresses, which this host joined (IPv6) 
1110	 if_inet6   List of IPv6 interface addresses                      
1111	 ipv6_route Kernel routing table for IPv6                         
1112	 rt6_stats  Global IPv6 routing tables statistics                 
1113	 sockstat6  Socket statistics (IPv6)                              
1114	 snmp6      Snmp data (IPv6)                                      
1115	..............................................................................
1116	
1117	
1118	Table 1-9: Network info in /proc/net
1119	..............................................................................
1120	 File          Content                                                         
1121	 arp           Kernel  ARP table                                               
1122	 dev           network devices with statistics                                 
1123	 dev_mcast     the Layer2 multicast groups a device is listening too
1124	               (interface index, label, number of references, number of bound
1125	               addresses). 
1126	 dev_stat      network device status                                           
1127	 ip_fwchains   Firewall chain linkage                                          
1128	 ip_fwnames    Firewall chain names                                            
1129	 ip_masq       Directory containing the masquerading tables                    
1130	 ip_masquerade Major masquerading table                                        
1131	 netstat       Network statistics                                              
1132	 raw           raw device statistics                                           
1133	 route         Kernel routing table                                            
1134	 rpc           Directory containing rpc info                                   
1135	 rt_cache      Routing cache                                                   
1136	 snmp          SNMP data                                                       
1137	 sockstat      Socket statistics                                               
1138	 tcp           TCP  sockets                                                    
1139	 udp           UDP sockets                                                     
1140	 unix          UNIX domain sockets                                             
1141	 wireless      Wireless interface data (Wavelan etc)                           
1142	 igmp          IP multicast addresses, which this host joined                  
1143	 psched        Global packet scheduler parameters.                             
1144	 netlink       List of PF_NETLINK sockets                                      
1145	 ip_mr_vifs    List of multicast virtual interfaces                            
1146	 ip_mr_cache   List of multicast routing cache                                 
1147	..............................................................................
1148	
1149	You can  use  this  information  to see which network devices are available in
1150	your system and how much traffic was routed over those devices:
1151	
1152	  > cat /proc/net/dev 
1153	  Inter-|Receive                                                   |[... 
1154	   face |bytes    packets errs drop fifo frame compressed multicast|[... 
1155	      lo:  908188   5596     0    0    0     0          0         0 [...         
1156	    ppp0:15475140  20721   410    0    0   410          0         0 [...  
1157	    eth0:  614530   7085     0    0    0     0          0         1 [... 
1158	   
1159	  ...] Transmit 
1160	  ...] bytes    packets errs drop fifo colls carrier compressed 
1161	  ...]  908188     5596    0    0    0     0       0          0 
1162	  ...] 1375103    17405    0    0    0     0       0          0 
1163	  ...] 1703981     5535    0    0    0     3       0          0 
1164	
1165	In addition, each Channel Bond interface has its own directory.  For
1166	example, the bond0 device will have a directory called /proc/net/bond0/.
1167	It will contain information that is specific to that bond, such as the
1168	current slaves of the bond, the link status of the slaves, and how
1169	many times the slaves link has failed.
1170	
1171	1.5 SCSI info
1172	-------------
1173	
1174	If you  have  a  SCSI  host adapter in your system, you'll find a subdirectory
1175	named after  the driver for this adapter in /proc/scsi. You'll also see a list
1176	of all recognized SCSI devices in /proc/scsi:
1177	
1178	  >cat /proc/scsi/scsi 
1179	  Attached devices: 
1180	  Host: scsi0 Channel: 00 Id: 00 Lun: 00 
1181	    Vendor: IBM      Model: DGHS09U          Rev: 03E0 
1182	    Type:   Direct-Access                    ANSI SCSI revision: 03 
1183	  Host: scsi0 Channel: 00 Id: 06 Lun: 00 
1184	    Vendor: PIONEER  Model: CD-ROM DR-U06S   Rev: 1.04 
1185	    Type:   CD-ROM                           ANSI SCSI revision: 02 
1186	
1187	
1188	The directory  named  after  the driver has one file for each adapter found in
1189	the system.  These  files  contain information about the controller, including
1190	the used  IRQ  and  the  IO  address range. The amount of information shown is
1191	dependent on  the adapter you use. The example shows the output for an Adaptec
1192	AHA-2940 SCSI adapter:
1193	
1194	  > cat /proc/scsi/aic7xxx/0 
1195	   
1196	  Adaptec AIC7xxx driver version: 5.1.19/3.2.4 
1197	  Compile Options: 
1198	    TCQ Enabled By Default : Disabled 
1199	    AIC7XXX_PROC_STATS     : Disabled 
1200	    AIC7XXX_RESET_DELAY    : 5 
1201	  Adapter Configuration: 
1202	             SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 
1203	                             Ultra Wide Controller 
1204	      PCI MMAPed I/O Base: 0xeb001000 
1205	   Adapter SEEPROM Config: SEEPROM found and used. 
1206	        Adaptec SCSI BIOS: Enabled 
1207	                      IRQ: 10 
1208	                     SCBs: Active 0, Max Active 2, 
1209	                           Allocated 15, HW 16, Page 255 
1210	               Interrupts: 160328 
1211	        BIOS Control Word: 0x18b6 
1212	     Adapter Control Word: 0x005b 
1213	     Extended Translation: Enabled 
1214	  Disconnect Enable Flags: 0xffff 
1215	       Ultra Enable Flags: 0x0001 
1216	   Tag Queue Enable Flags: 0x0000 
1217	  Ordered Queue Tag Flags: 0x0000 
1218	  Default Tag Queue Depth: 8 
1219	      Tagged Queue By Device array for aic7xxx host instance 0: 
1220	        {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 
1221	      Actual queue depth per device for aic7xxx host instance 0: 
1222	        {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 
1223	  Statistics: 
1224	  (scsi0:0:0:0) 
1225	    Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 
1226	    Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 
1227	    Total transfers 160151 (74577 reads and 85574 writes) 
1228	  (scsi0:0:6:0) 
1229	    Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 
1230	    Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 
1231	    Total transfers 0 (0 reads and 0 writes) 
1232	
1233	
1234	1.6 Parallel port info in /proc/parport
1235	---------------------------------------
1236	
1237	The directory  /proc/parport  contains information about the parallel ports of
1238	your system.  It  has  one  subdirectory  for  each port, named after the port
1239	number (0,1,2,...).
1240	
1241	These directories contain the four files shown in Table 1-10.
1242	
1243	
1244	Table 1-10: Files in /proc/parport
1245	..............................................................................
1246	 File      Content                                                             
1247	 autoprobe Any IEEE-1284 device ID information that has been acquired.         
1248	 devices   list of the device drivers using that port. A + will appear by the
1249	           name of the device currently using the port (it might not appear
1250	           against any). 
1251	 hardware  Parallel port's base address, IRQ line and DMA channel.             
1252	 irq       IRQ that parport is using for that port. This is in a separate
1253	           file to allow you to alter it by writing a new value in (IRQ
1254	           number or none). 
1255	..............................................................................
1256	
1257	1.7 TTY info in /proc/tty
1258	-------------------------
1259	
1260	Information about  the  available  and actually used tty's can be found in the
1261	directory /proc/tty.You'll  find  entries  for drivers and line disciplines in
1262	this directory, as shown in Table 1-11.
1263	
1264	
1265	Table 1-11: Files in /proc/tty
1266	..............................................................................
1267	 File          Content                                        
1268	 drivers       list of drivers and their usage                
1269	 ldiscs        registered line disciplines                    
1270	 driver/serial usage statistic and status of single tty lines 
1271	..............................................................................
1272	
1273	To see  which  tty's  are  currently in use, you can simply look into the file
1274	/proc/tty/drivers:
1275	
1276	  > cat /proc/tty/drivers 
1277	  pty_slave            /dev/pts      136   0-255 pty:slave 
1278	  pty_master           /dev/ptm      128   0-255 pty:master 
1279	  pty_slave            /dev/ttyp       3   0-255 pty:slave 
1280	  pty_master           /dev/pty        2   0-255 pty:master 
1281	  serial               /dev/cua        5   64-67 serial:callout 
1282	  serial               /dev/ttyS       4   64-67 serial 
1283	  /dev/tty0            /dev/tty0       4       0 system:vtmaster 
1284	  /dev/ptmx            /dev/ptmx       5       2 system 
1285	  /dev/console         /dev/console    5       1 system:console 
1286	  /dev/tty             /dev/tty        5       0 system:/dev/tty 
1287	  unknown              /dev/tty        4    1-63 console 
1288	
1289	
1290	1.8 Miscellaneous kernel statistics in /proc/stat
1291	-------------------------------------------------
1292	
1293	Various pieces   of  information about  kernel activity  are  available in the
1294	/proc/stat file.  All  of  the numbers reported  in  this file are  aggregates
1295	since the system first booted.  For a quick look, simply cat the file:
1296	
1297	  > cat /proc/stat
1298	  cpu  2255 34 2290 22625563 6290 127 456 0 0 0
1299	  cpu0 1132 34 1441 11311718 3675 127 438 0 0 0
1300	  cpu1 1123 0 849 11313845 2614 0 18 0 0 0
1301	  intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1302	  ctxt 1990473
1303	  btime 1062191376
1304	  processes 2915
1305	  procs_running 1
1306	  procs_blocked 0
1307	  softirq 183433 0 21755 12 39 1137 231 21459 2263
1308	
1309	The very first  "cpu" line aggregates the  numbers in all  of the other "cpuN"
1310	lines.  These numbers identify the amount of time the CPU has spent performing
1311	different kinds of work.  Time units are in USER_HZ (typically hundredths of a
1312	second).  The meanings of the columns are as follows, from left to right:
1313	
1314	- user: normal processes executing in user mode
1315	- nice: niced processes executing in user mode
1316	- system: processes executing in kernel mode
1317	- idle: twiddling thumbs
1318	- iowait: In a word, iowait stands for waiting for I/O to complete. But there
1319	  are several problems:
1320	  1. Cpu will not wait for I/O to complete, iowait is the time that a task is
1321	     waiting for I/O to complete. When cpu goes into idle state for
1322	     outstanding task io, another task will be scheduled on this CPU.
1323	  2. In a multi-core CPU, the task waiting for I/O to complete is not running
1324	     on any CPU, so the iowait of each CPU is difficult to calculate.
1325	  3. The value of iowait field in /proc/stat will decrease in certain
1326	     conditions.
1327	  So, the iowait is not reliable by reading from /proc/stat.
1328	- irq: servicing interrupts
1329	- softirq: servicing softirqs
1330	- steal: involuntary wait
1331	- guest: running a normal guest
1332	- guest_nice: running a niced guest
1333	
1334	The "intr" line gives counts of interrupts  serviced since boot time, for each
1335	of the  possible system interrupts.   The first  column  is the  total of  all
1336	interrupts serviced  including  unnumbered  architecture specific  interrupts;
1337	each  subsequent column is the  total for that particular numbered interrupt.
1338	Unnumbered interrupts are not shown, only summed into the total.
1339	
1340	The "ctxt" line gives the total number of context switches across all CPUs.
1341	
1342	The "btime" line gives  the time at which the  system booted, in seconds since
1343	the Unix epoch.
1344	
1345	The "processes" line gives the number  of processes and threads created, which
1346	includes (but  is not limited  to) those  created by  calls to the  fork() and
1347	clone() system calls.
1348	
1349	The "procs_running" line gives the total number of threads that are
1350	running or ready to run (i.e., the total number of runnable threads).
1351	
1352	The   "procs_blocked" line gives  the  number of  processes currently blocked,
1353	waiting for I/O to complete.
1354	
1355	The "softirq" line gives counts of softirqs serviced since boot time, for each
1356	of the possible system softirqs. The first column is the total of all
1357	softirqs serviced; each subsequent column is the total for that particular
1358	softirq.
1359	
1360	
1361	1.9 Ext4 file system parameters
1362	-------------------------------
1363	
1364	Information about mounted ext4 file systems can be found in
1365	/proc/fs/ext4.  Each mounted filesystem will have a directory in
1366	/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1367	/proc/fs/ext4/dm-0).   The files in each per-device directory are shown
1368	in Table 1-12, below.
1369	
1370	Table 1-12: Files in /proc/fs/ext4/<devname>
1371	..............................................................................
1372	 File            Content                                        
1373	 mb_groups       details of multiblock allocator buddy cache of free blocks
1374	..............................................................................
1375	
1376	2.0 /proc/consoles
1377	------------------
1378	Shows registered system console lines.
1379	
1380	To see which character device lines are currently used for the system console
1381	/dev/console, you may simply look into the file /proc/consoles:
1382	
1383	  > cat /proc/consoles
1384	  tty0                 -WU (ECp)       4:7
1385	  ttyS0                -W- (Ep)        4:64
1386	
1387	The columns are:
1388	
1389	  device               name of the device
1390	  operations           R = can do read operations
1391	                       W = can do write operations
1392	                       U = can do unblank
1393	  flags                E = it is enabled
1394	                       C = it is preferred console
1395	                       B = it is primary boot console
1396	                       p = it is used for printk buffer
1397	                       b = it is not a TTY but a Braille device
1398	                       a = it is safe to use when cpu is offline
1399	  major:minor          major and minor number of the device separated by a colon
1400	
1401	------------------------------------------------------------------------------
1402	Summary
1403	------------------------------------------------------------------------------
1404	The /proc file system serves information about the running system. It not only
1405	allows access to process data but also allows you to request the kernel status
1406	by reading files in the hierarchy.
1407	
1408	The directory  structure  of /proc reflects the types of information and makes
1409	it easy, if not obvious, where to look for specific data.
1410	------------------------------------------------------------------------------
1411	
1412	------------------------------------------------------------------------------
1413	CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1414	------------------------------------------------------------------------------
1415	
1416	------------------------------------------------------------------------------
1417	In This Chapter
1418	------------------------------------------------------------------------------
1419	* Modifying kernel parameters by writing into files found in /proc/sys
1420	* Exploring the files which modify certain parameters
1421	* Review of the /proc/sys file tree
1422	------------------------------------------------------------------------------
1423	
1424	
1425	A very  interesting part of /proc is the directory /proc/sys. This is not only
1426	a source  of  information,  it also allows you to change parameters within the
1427	kernel. Be  very  careful  when attempting this. You can optimize your system,
1428	but you  can  also  cause  it  to  crash.  Never  alter kernel parameters on a
1429	production system.  Set  up  a  development machine and test to make sure that
1430	everything works  the  way  you want it to. You may have no alternative but to
1431	reboot the machine once an error has been made.
1432	
1433	To change  a  value,  simply  echo  the new value into the file. An example is
1434	given below  in the section on the file system data. You need to be root to do
1435	this. You  can  create  your  own  boot script to perform this every time your
1436	system boots.
1437	
1438	The files  in /proc/sys can be used to fine tune and monitor miscellaneous and
1439	general things  in  the operation of the Linux kernel. Since some of the files
1440	can inadvertently  disrupt  your  system,  it  is  advisable  to  read  both
1441	documentation and  source  before actually making adjustments. In any case, be
1442	very careful  when  writing  to  any  of these files. The entries in /proc may
1443	change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1444	review the kernel documentation in the directory /usr/src/linux/Documentation.
1445	This chapter  is  heavily  based  on the documentation included in the pre 2.2
1446	kernels, and became part of it in version 2.2.1 of the Linux kernel.
1447	
1448	Please see: Documentation/sysctl/ directory for descriptions of these
1449	entries.
1450	
1451	------------------------------------------------------------------------------
1452	Summary
1453	------------------------------------------------------------------------------
1454	Certain aspects  of  kernel  behavior  can be modified at runtime, without the
1455	need to  recompile  the kernel, or even to reboot the system. The files in the
1456	/proc/sys tree  can  not only be read, but also modified. You can use the echo
1457	command to write value into these files, thereby changing the default settings
1458	of the kernel.
1459	------------------------------------------------------------------------------
1460	
1461	------------------------------------------------------------------------------
1462	CHAPTER 3: PER-PROCESS PARAMETERS
1463	------------------------------------------------------------------------------
1464	
1465	3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1466	--------------------------------------------------------------------------------
1467	
1468	These file can be used to adjust the badness heuristic used to select which
1469	process gets killed in out of memory conditions.
1470	
1471	The badness heuristic assigns a value to each candidate task ranging from 0
1472	(never kill) to 1000 (always kill) to determine which process is targeted.  The
1473	units are roughly a proportion along that range of allowed memory the process
1474	may allocate from based on an estimation of its current memory and swap use.
1475	For example, if a task is using all allowed memory, its badness score will be
1476	1000.  If it is using half of its allowed memory, its score will be 500.
1477	
1478	There is an additional factor included in the badness score: the current memory
1479	and swap usage is discounted by 3% for root processes.
1480	
1481	The amount of "allowed" memory depends on the context in which the oom killer
1482	was called.  If it is due to the memory assigned to the allocating task's cpuset
1483	being exhausted, the allowed memory represents the set of mems assigned to that
1484	cpuset.  If it is due to a mempolicy's node(s) being exhausted, the allowed
1485	memory represents the set of mempolicy nodes.  If it is due to a memory
1486	limit (or swap limit) being reached, the allowed memory is that configured
1487	limit.  Finally, if it is due to the entire system being out of memory, the
1488	allowed memory represents all allocatable resources.
1489	
1490	The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1491	is used to determine which task to kill.  Acceptable values range from -1000
1492	(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX).  This allows userspace to
1493	polarize the preference for oom killing either by always preferring a certain
1494	task or completely disabling it.  The lowest possible value, -1000, is
1495	equivalent to disabling oom killing entirely for that task since it will always
1496	report a badness score of 0.
1497	
1498	Consequently, it is very simple for userspace to define the amount of memory to
1499	consider for each task.  Setting a /proc/<pid>/oom_score_adj value of +500, for
1500	example, is roughly equivalent to allowing the remainder of tasks sharing the
1501	same system, cpuset, mempolicy, or memory controller resources to use at least
1502	50% more memory.  A value of -500, on the other hand, would be roughly
1503	equivalent to discounting 50% of the task's allowed memory from being considered
1504	as scoring against the task.
1505	
1506	For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1507	be used to tune the badness score.  Its acceptable values range from -16
1508	(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1509	(OOM_DISABLE) to disable oom killing entirely for that task.  Its value is
1510	scaled linearly with /proc/<pid>/oom_score_adj.
1511	
1512	The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1513	value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1514	requires CAP_SYS_RESOURCE.
1515	
1516	Caveat: when a parent task is selected, the oom killer will sacrifice any first
1517	generation children with separate address spaces instead, if possible.  This
1518	avoids servers and important system daemons from being killed and loses the
1519	minimal amount of work.
1520	
1521	
1522	3.2 /proc/<pid>/oom_score - Display current oom-killer score
1523	-------------------------------------------------------------
1524	
1525	This file can be used to check the current score used by the oom-killer is for
1526	any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which
1527	process should be killed in an out-of-memory situation.
1528	
1529	
1530	3.3  /proc/<pid>/io - Display the IO accounting fields
1531	-------------------------------------------------------
1532	
1533	This file contains IO statistics for each running process
1534	
1535	Example
1536	-------
1537	
1538	test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1539	[1] 3828
1540	
1541	test:/tmp # cat /proc/3828/io
1542	rchar: 323934931
1543	wchar: 323929600
1544	syscr: 632687
1545	syscw: 632675
1546	read_bytes: 0
1547	write_bytes: 323932160
1548	cancelled_write_bytes: 0
1549	
1550	
1551	Description
1552	-----------
1553	
1554	rchar
1555	-----
1556	
1557	I/O counter: chars read
1558	The number of bytes which this task has caused to be read from storage. This
1559	is simply the sum of bytes which this process passed to read() and pread().
1560	It includes things like tty IO and it is unaffected by whether or not actual
1561	physical disk IO was required (the read might have been satisfied from
1562	pagecache)
1563	
1564	
1565	wchar
1566	-----
1567	
1568	I/O counter: chars written
1569	The number of bytes which this task has caused, or shall cause to be written
1570	to disk. Similar caveats apply here as with rchar.
1571	
1572	
1573	syscr
1574	-----
1575	
1576	I/O counter: read syscalls
1577	Attempt to count the number of read I/O operations, i.e. syscalls like read()
1578	and pread().
1579	
1580	
1581	syscw
1582	-----
1583	
1584	I/O counter: write syscalls
1585	Attempt to count the number of write I/O operations, i.e. syscalls like
1586	write() and pwrite().
1587	
1588	
1589	read_bytes
1590	----------
1591	
1592	I/O counter: bytes read
1593	Attempt to count the number of bytes which this process really did cause to
1594	be fetched from the storage layer. Done at the submit_bio() level, so it is
1595	accurate for block-backed filesystems. <please add status regarding NFS and
1596	CIFS at a later time>
1597	
1598	
1599	write_bytes
1600	-----------
1601	
1602	I/O counter: bytes written
1603	Attempt to count the number of bytes which this process caused to be sent to
1604	the storage layer. This is done at page-dirtying time.
1605	
1606	
1607	cancelled_write_bytes
1608	---------------------
1609	
1610	The big inaccuracy here is truncate. If a process writes 1MB to a file and
1611	then deletes the file, it will in fact perform no writeout. But it will have
1612	been accounted as having caused 1MB of write.
1613	In other words: The number of bytes which this process caused to not happen,
1614	by truncating pagecache. A task can cause "negative" IO too. If this task
1615	truncates some dirty pagecache, some IO which another task has been accounted
1616	for (in its write_bytes) will not be happening. We _could_ just subtract that
1617	from the truncating task's write_bytes, but there is information loss in doing
1618	that.
1619	
1620	
1621	Note
1622	----
1623	
1624	At its current implementation state, this is a bit racy on 32-bit machines: if
1625	process A reads process B's /proc/pid/io while process B is updating one of
1626	those 64-bit counters, process A could see an intermediate result.
1627	
1628	
1629	More information about this can be found within the taskstats documentation in
1630	Documentation/accounting.
1631	
1632	3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1633	---------------------------------------------------------------
1634	When a process is dumped, all anonymous memory is written to a core file as
1635	long as the size of the core file isn't limited. But sometimes we don't want
1636	to dump some memory segments, for example, huge shared memory or DAX.
1637	Conversely, sometimes we want to save file-backed memory segments into a core
1638	file, not only the individual files.
1639	
1640	/proc/<pid>/coredump_filter allows you to customize which memory segments
1641	will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1642	of memory types. If a bit of the bitmask is set, memory segments of the
1643	corresponding memory type are dumped, otherwise they are not dumped.
1644	
1645	The following 9 memory types are supported:
1646	  - (bit 0) anonymous private memory
1647	  - (bit 1) anonymous shared memory
1648	  - (bit 2) file-backed private memory
1649	  - (bit 3) file-backed shared memory
1650	  - (bit 4) ELF header pages in file-backed private memory areas (it is
1651	            effective only if the bit 2 is cleared)
1652	  - (bit 5) hugetlb private memory
1653	  - (bit 6) hugetlb shared memory
1654	  - (bit 7) DAX private memory
1655	  - (bit 8) DAX shared memory
1656	
1657	  Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1658	  are always dumped regardless of the bitmask status.
1659	
1660	  Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is
1661	  only affected by bit 5-6, and DAX is only affected by bits 7-8.
1662	
1663	The default value of coredump_filter is 0x33; this means all anonymous memory
1664	segments, ELF header pages and hugetlb private memory are dumped.
1665	
1666	If you don't want to dump all shared memory segments attached to pid 1234,
1667	write 0x31 to the process's proc file.
1668	
1669	  $ echo 0x31 > /proc/1234/coredump_filter
1670	
1671	When a new process is created, the process inherits the bitmask status from its
1672	parent. It is useful to set up coredump_filter before the program runs.
1673	For example:
1674	
1675	  $ echo 0x7 > /proc/self/coredump_filter
1676	  $ ./some_program
1677	
1678	3.5	/proc/<pid>/mountinfo - Information about mounts
1679	--------------------------------------------------------
1680	
1681	This file contains lines of the form:
1682	
1683	36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1684	(1)(2)(3)   (4)   (5)      (6)      (7)   (8) (9)   (10)         (11)
1685	
1686	(1) mount ID:  unique identifier of the mount (may be reused after umount)
1687	(2) parent ID:  ID of parent (or of self for the top of the mount tree)
1688	(3) major:minor:  value of st_dev for files on filesystem
1689	(4) root:  root of the mount within the filesystem
1690	(5) mount point:  mount point relative to the process's root
1691	(6) mount options:  per mount options
1692	(7) optional fields:  zero or more fields of the form "tag[:value]"
1693	(8) separator:  marks the end of the optional fields
1694	(9) filesystem type:  name of filesystem of the form "type[.subtype]"
1695	(10) mount source:  filesystem specific information or "none"
1696	(11) super options:  per super block options
1697	
1698	Parsers should ignore all unrecognised optional fields.  Currently the
1699	possible optional fields are:
1700	
1701	shared:X  mount is shared in peer group X
1702	master:X  mount is slave to peer group X
1703	propagate_from:X  mount is slave and receives propagation from peer group X (*)
1704	unbindable  mount is unbindable
1705	
1706	(*) X is the closest dominant peer group under the process's root.  If
1707	X is the immediate master of the mount, or if there's no dominant peer
1708	group under the same root, then only the "master:X" field is present
1709	and not the "propagate_from:X" field.
1710	
1711	For more information on mount propagation see:
1712	
1713	  Documentation/filesystems/sharedsubtree.txt
1714	
1715	
1716	3.6	/proc/<pid>/comm  & /proc/<pid>/task/<tid>/comm
1717	--------------------------------------------------------
1718	These files provide a method to access a tasks comm value. It also allows for
1719	a task to set its own or one of its thread siblings comm value. The comm value
1720	is limited in size compared to the cmdline value, so writing anything longer
1721	then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1722	comm value.
1723	
1724	
1725	3.7	/proc/<pid>/task/<tid>/children - Information about task children
1726	-------------------------------------------------------------------------
1727	This file provides a fast way to retrieve first level children pids
1728	of a task pointed by <pid>/<tid> pair. The format is a space separated
1729	stream of pids.
1730	
1731	Note the "first level" here -- if a child has own children they will
1732	not be listed here, one needs to read /proc/<children-pid>/task/<tid>/children
1733	to obtain the descendants.
1734	
1735	Since this interface is intended to be fast and cheap it doesn't
1736	guarantee to provide precise results and some children might be
1737	skipped, especially if they've exited right after we printed their
1738	pids, so one need to either stop or freeze processes being inspected
1739	if precise results are needed.
1740	
1741	
1742	3.8	/proc/<pid>/fdinfo/<fd> - Information about opened file
1743	---------------------------------------------------------------
1744	This file provides information associated with an opened file. The regular
1745	files have at least three fields -- 'pos', 'flags' and mnt_id. The 'pos'
1746	represents the current offset of the opened file in decimal form [see lseek(2)
1747	for details], 'flags' denotes the octal O_xxx mask the file has been
1748	created with [see open(2) for details] and 'mnt_id' represents mount ID of
1749	the file system containing the opened file [see 3.5 /proc/<pid>/mountinfo
1750	for details].
1751	
1752	A typical output is
1753	
1754		pos:	0
1755		flags:	0100002
1756		mnt_id:	19
1757	
1758	All locks associated with a file descriptor are shown in its fdinfo too.
1759	
1760	lock:       1: FLOCK  ADVISORY  WRITE 359 00:13:11691 0 EOF
1761	
1762	The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags
1763	pair provide additional information particular to the objects they represent.
1764	
1765		Eventfd files
1766		~~~~~~~~~~~~~
1767		pos:	0
1768		flags:	04002
1769		mnt_id:	9
1770		eventfd-count:	5a
1771	
1772		where 'eventfd-count' is hex value of a counter.
1773	
1774		Signalfd files
1775		~~~~~~~~~~~~~~
1776		pos:	0
1777		flags:	04002
1778		mnt_id:	9
1779		sigmask:	0000000000000200
1780	
1781		where 'sigmask' is hex value of the signal mask associated
1782		with a file.
1783	
1784		Epoll files
1785		~~~~~~~~~~~
1786		pos:	0
1787		flags:	02
1788		mnt_id:	9
1789		tfd:        5 events:       1d data: ffffffffffffffff pos:0 ino:61af sdev:7
1790	
1791		where 'tfd' is a target file descriptor number in decimal form,
1792		'events' is events mask being watched and the 'data' is data
1793		associated with a target [see epoll(7) for more details].
1794	
1795		The 'pos' is current offset of the target file in decimal form
1796		[see lseek(2)], 'ino' and 'sdev' are inode and device numbers
1797		where target file resides, all in hex format.
1798	
1799		Fsnotify files
1800		~~~~~~~~~~~~~~
1801		For inotify files the format is the following
1802	
1803		pos:	0
1804		flags:	02000000
1805		inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
1806	
1807		where 'wd' is a watch descriptor in decimal form, ie a target file
1808		descriptor number, 'ino' and 'sdev' are inode and device where the
1809		target file resides and the 'mask' is the mask of events, all in hex
1810		form [see inotify(7) for more details].
1811	
1812		If the kernel was built with exportfs support, the path to the target
1813		file is encoded as a file handle.  The file handle is provided by three
1814		fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex
1815		format.
1816	
1817		If the kernel is built without exportfs support the file handle won't be
1818		printed out.
1819	
1820		If there is no inotify mark attached yet the 'inotify' line will be omitted.
1821	
1822		For fanotify files the format is
1823	
1824		pos:	0
1825		flags:	02
1826		mnt_id:	9
1827		fanotify flags:10 event-flags:0
1828		fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
1829		fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
1830	
1831		where fanotify 'flags' and 'event-flags' are values used in fanotify_init
1832		call, 'mnt_id' is the mount point identifier, 'mflags' is the value of
1833		flags associated with mark which are tracked separately from events
1834		mask. 'ino', 'sdev' are target inode and device, 'mask' is the events
1835		mask and 'ignored_mask' is the mask of events which are to be ignored.
1836		All in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask'
1837		does provide information about flags and mask used in fanotify_mark
1838		call [see fsnotify manpage for details].
1839	
1840		While the first three lines are mandatory and always printed, the rest is
1841		optional and may be omitted if no marks created yet.
1842	
1843		Timerfd files
1844		~~~~~~~~~~~~~
1845	
1846		pos:	0
1847		flags:	02
1848		mnt_id:	9
1849		clockid: 0
1850		ticks: 0
1851		settime flags: 01
1852		it_value: (0, 49406829)
1853		it_interval: (1, 0)
1854	
1855		where 'clockid' is the clock type and 'ticks' is the number of the timer expirations
1856		that have occurred [see timerfd_create(2) for details]. 'settime flags' are
1857		flags in octal form been used to setup the timer [see timerfd_settime(2) for
1858		details]. 'it_value' is remaining time until the timer exiration.
1859		'it_interval' is the interval for the timer. Note the timer might be set up
1860		with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value'
1861		still exhibits timer's remaining time.
1862	
1863	3.9	/proc/<pid>/map_files - Information about memory mapped files
1864	---------------------------------------------------------------------
1865	This directory contains symbolic links which represent memory mapped files
1866	the process is maintaining.  Example output:
1867	
1868	     | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so
1869	     | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so
1870	     | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so
1871	     | ...
1872	     | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1
1873	     | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls
1874	
1875	The name of a link represents the virtual memory bounds of a mapping, i.e.
1876	vm_area_struct::vm_start-vm_area_struct::vm_end.
1877	
1878	The main purpose of the map_files is to retrieve a set of memory mapped
1879	files in a fast way instead of parsing /proc/<pid>/maps or
1880	/proc/<pid>/smaps, both of which contain many more records.  At the same
1881	time one can open(2) mappings from the listings of two processes and
1882	comparing their inode numbers to figure out which anonymous memory areas
1883	are actually shared.
1884	
1885	3.10	/proc/<pid>/timerslack_ns - Task timerslack value
1886	---------------------------------------------------------
1887	This file provides the value of the task's timerslack value in nanoseconds.
1888	This value specifies a amount of time that normal timers may be deferred
1889	in order to coalesce timers and avoid unnecessary wakeups.
1890	
1891	This allows a task's interactivity vs power consumption trade off to be
1892	adjusted.
1893	
1894	Writing 0 to the file will set the tasks timerslack to the default value.
1895	
1896	Valid values are from 0 - ULLONG_MAX
1897	
1898	An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
1899	permissions on the task specified to change its timerslack_ns value.
1900	
1901	3.11	/proc/<pid>/patch_state - Livepatch patch operation state
1902	-----------------------------------------------------------------
1903	When CONFIG_LIVEPATCH is enabled, this file displays the value of the
1904	patch state for the task.
1905	
1906	A value of '-1' indicates that no patch is in transition.
1907	
1908	A value of '0' indicates that a patch is in transition and the task is
1909	unpatched.  If the patch is being enabled, then the task hasn't been
1910	patched yet.  If the patch is being disabled, then the task has already
1911	been unpatched.
1912	
1913	A value of '1' indicates that a patch is in transition and the task is
1914	patched.  If the patch is being enabled, then the task has already been
1915	patched.  If the patch is being disabled, then the task hasn't been
1916	unpatched yet.
1917	
1918	
1919	------------------------------------------------------------------------------
1920	Configuring procfs
1921	------------------------------------------------------------------------------
1922	
1923	4.1	Mount options
1924	---------------------
1925	
1926	The following mount options are supported:
1927	
1928		hidepid=	Set /proc/<pid>/ access mode.
1929		gid=		Set the group authorized to learn processes information.
1930	
1931	hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1932	(default).
1933	
1934	hidepid=1 means users may not access any /proc/<pid>/ directories but their
1935	own.  Sensitive files like cmdline, sched*, status are now protected against
1936	other users.  This makes it impossible to learn whether any user runs
1937	specific program (given the program doesn't reveal itself by its behaviour).
1938	As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1939	poorly written programs passing sensitive information via program arguments are
1940	now protected against local eavesdroppers.
1941	
1942	hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1943	users.  It doesn't mean that it hides a fact whether a process with a specific
1944	pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1945	but it hides process' uid and gid, which may be learned by stat()'ing
1946	/proc/<pid>/ otherwise.  It greatly complicates an intruder's task of gathering
1947	information about running processes, whether some daemon runs with elevated
1948	privileges, whether other user runs some sensitive program, whether other users
1949	run any program at all, etc.
1950	
1951	gid= defines a group authorized to learn processes information otherwise
1952	prohibited by hidepid=.  If you use some daemon like identd which needs to learn
1953	information about processes information, just add identd to this group.
Hide Line Numbers
About Kernel Documentation Linux Kernel Contact Linux Resources Linux Blog

Information is copyright its respective author. All material is available from the Linux Kernel Source distributed under a GPL License. This page is provided as a free service by mjmwired.net.