handle kernelcore=: generic
[GitHub/mt8127/android_kernel_alcatel_ttab.git] / Documentation / filesystems / proc.txt
CommitLineData
1da177e4
LT
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
72.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8------------------------------------------------------------------------------
9Version 1.3 Kernel version 2.2.12
10 Kernel version 2.4.0-test11-pre4
11------------------------------------------------------------------------------
12
13Table of Contents
14-----------------
15
16 0 Preface
17 0.1 Introduction/Credits
18 0.2 Legal Stuff
19
20 1 Collecting System Information
21 1.1 Process-Specific Subdirectories
22 1.2 Kernel data
23 1.3 IDE devices in /proc/ide
24 1.4 Networking info in /proc/net
25 1.5 SCSI info
26 1.6 Parallel port info in /proc/parport
27 1.7 TTY info in /proc/tty
28 1.8 Miscellaneous kernel statistics in /proc/stat
29
30 2 Modifying System Parameters
31 2.1 /proc/sys/fs - File system data
32 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
33 2.3 /proc/sys/kernel - general kernel parameters
34 2.4 /proc/sys/vm - The virtual memory subsystem
35 2.5 /proc/sys/dev - Device specific parameters
36 2.6 /proc/sys/sunrpc - Remote procedure calls
37 2.7 /proc/sys/net - Networking stuff
38 2.8 /proc/sys/net/ipv4 - IPV4 settings
39 2.9 Appletalk
40 2.10 IPX
41 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
d7ff0dbf
JFM
42 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
43 2.13 /proc/<pid>/oom_score - Display current oom-killer score
f9c99463 44 2.14 /proc/<pid>/io - Display the IO accounting fields
1da177e4
LT
45
46------------------------------------------------------------------------------
47Preface
48------------------------------------------------------------------------------
49
500.1 Introduction/Credits
51------------------------
52
53This documentation is part of a soon (or so we hope) to be released book on
54the SuSE Linux distribution. As there is no complete documentation for the
55/proc file system and we've used many freely available sources to write these
56chapters, it seems only fair to give the work back to the Linux community.
57This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
58afraid it's still far from complete, but we hope it will be useful. As far as
59we know, it is the first 'all-in-one' document about the /proc file system. It
60is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
61SPARC, AXP, etc., features, you probably won't find what you are looking for.
62It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
63additions and patches are welcome and will be added to this document if you
64mail them to Bodo.
65
66We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
67other people for help compiling this documentation. We'd also like to extend a
68special thank you to Andi Kleen for documentation, which we relied on heavily
69to create this document, as well as the additional information he provided.
70Thanks to everybody else who contributed source or docs to the Linux kernel
71and helped create a great piece of software... :)
72
73If you have any comments, corrections or additions, please don't hesitate to
74contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
75document.
76
77The latest version of this document is available online at
78http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
79
80If the above direction does not works for you, ypu could try the kernel
81mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
82comandante@zaralinux.com.
83
840.2 Legal Stuff
85---------------
86
87We don't guarantee the correctness of this document, and if you come to us
88complaining about how you screwed up your system because of incorrect
89documentation, we won't feel responsible...
90
91------------------------------------------------------------------------------
92CHAPTER 1: COLLECTING SYSTEM INFORMATION
93------------------------------------------------------------------------------
94
95------------------------------------------------------------------------------
96In This Chapter
97------------------------------------------------------------------------------
98* Investigating the properties of the pseudo file system /proc and its
99 ability to provide information on the running Linux system
100* Examining /proc's structure
101* Uncovering various information about the kernel and the processes running
102 on the system
103------------------------------------------------------------------------------
104
105
106The proc file system acts as an interface to internal data structures in the
107kernel. It can be used to obtain information about the system and to change
108certain kernel parameters at runtime (sysctl).
109
110First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
111show you how you can use /proc/sys to change settings.
112
1131.1 Process-Specific Subdirectories
114-----------------------------------
115
116The directory /proc contains (among other things) one subdirectory for each
117process running on the system, which is named after the process ID (PID).
118
119The link self points to the process reading the file system. Each process
120subdirectory has the entries listed in Table 1-1.
121
122
123Table 1-1: Process specific entries in /proc
124..............................................................................
b813e931
DR
125 File Content
126 clear_refs Clears page referenced bits shown in smaps output
127 cmdline Command line arguments
128 cpu Current and last cpu in which it was executed (2.4)(smp)
129 cwd Link to the current working directory
130 environ Values of environment variables
131 exe Link to the executable of this process
132 fd Directory, which contains all file descriptors
133 maps Memory maps to executables and library files (2.4)
134 mem Memory held by this process
135 root Link to the root directory of this process
136 stat Process status
137 statm Process memory status information
138 status Process status in human readable form
139 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
140 smaps Extension based on maps, the rss size for each mapped file
1da177e4
LT
141..............................................................................
142
143For example, to get the status information of a process, all you have to do is
144read the file /proc/PID/status:
145
146 >cat /proc/self/status
147 Name: cat
148 State: R (running)
149 Pid: 5452
150 PPid: 743
151 TracerPid: 0 (2.4)
152 Uid: 501 501 501 501
153 Gid: 100 100 100 100
154 Groups: 100 14 16
155 VmSize: 1112 kB
156 VmLck: 0 kB
157 VmRSS: 348 kB
158 VmData: 24 kB
159 VmStk: 12 kB
160 VmExe: 8 kB
161 VmLib: 1044 kB
162 SigPnd: 0000000000000000
163 SigBlk: 0000000000000000
164 SigIgn: 0000000000000000
165 SigCgt: 0000000000000000
166 CapInh: 00000000fffffeff
167 CapPrm: 0000000000000000
168 CapEff: 0000000000000000
169
170
171This shows you nearly the same information you would get if you viewed it with
172the ps command. In fact, ps uses the proc file system to obtain its
173information. The statm file contains more detailed information about the
18d96779
KC
174process memory usage. Its seven fields are explained in Table 1-2. The stat
175file contains details information about the process itself. Its fields are
176explained in Table 1-3.
1da177e4
LT
177
178
179Table 1-2: Contents of the statm files (as of 2.6.8-rc3)
180..............................................................................
181 Field Content
182 size total program size (pages) (same as VmSize in status)
183 resident size of memory portions (pages) (same as VmRSS in status)
184 shared number of pages that are shared (i.e. backed by a file)
185 trs number of pages that are 'code' (not including libs; broken,
186 includes data segment)
187 lrs number of pages of library (always 0 on 2.6)
188 drs number of pages of data/stack (including libs; broken,
189 includes library text)
190 dt number of dirty pages (always 0 on 2.6)
191..............................................................................
192
18d96779
KC
193
194Table 1-3: Contents of the stat files (as of 2.6.22-rc3)
195..............................................................................
196 Field Content
197 pid process id
198 tcomm filename of the executable
199 state state (R is running, S is sleeping, D is sleeping in an
200 uninterruptible wait, Z is zombie, T is traced or stopped)
201 ppid process id of the parent process
202 pgrp pgrp of the process
203 sid session id
204 tty_nr tty the process uses
205 tty_pgrp pgrp of the tty
206 flags task flags
207 min_flt number of minor faults
208 cmin_flt number of minor faults with child's
209 maj_flt number of major faults
210 cmaj_flt number of major faults with child's
211 utime user mode jiffies
212 stime kernel mode jiffies
213 cutime user mode jiffies with child's
214 cstime kernel mode jiffies with child's
215 priority priority level
216 nice nice level
217 num_threads number of threads
218 start_time time the process started after system boot
219 vsize virtual memory size
220 rss resident set memory size
221 rsslim current limit in bytes on the rss
222 start_code address above which program text can run
223 end_code address below which program text can run
224 start_stack address of the start of the stack
225 esp current value of ESP
226 eip current value of EIP
227 pending bitmap of pending signals (obsolete)
228 blocked bitmap of blocked signals (obsolete)
229 sigign bitmap of ignored signals (obsolete)
230 sigcatch bitmap of catched signals (obsolete)
231 wchan address where process went to sleep
232 0 (place holder)
233 0 (place holder)
234 exit_signal signal to send to parent thread on exit
235 task_cpu which CPU the task is scheduled on
236 rt_priority realtime priority
237 policy scheduling policy (man sched_setscheduler)
238 blkio_ticks time spent waiting for block IO
239..............................................................................
240
241
1da177e4
LT
2421.2 Kernel data
243---------------
244
245Similar to the process entries, the kernel data files give information about
246the running kernel. The files used to obtain this information are contained in
18d96779 247/proc and are listed in Table 1-4. Not all of these will be present in your
1da177e4
LT
248system. It depends on the kernel configuration and the loaded modules, which
249files are there, and which are missing.
250
18d96779 251Table 1-4: Kernel info in /proc
1da177e4
LT
252..............................................................................
253 File Content
254 apm Advanced power management info
255 buddyinfo Kernel memory allocator information (see text) (2.5)
256 bus Directory containing bus specific information
257 cmdline Kernel command line
258 cpuinfo Info about the CPU
259 devices Available devices (block and character)
260 dma Used DMS channels
261 filesystems Supported filesystems
262 driver Various drivers grouped here, currently rtc (2.4)
263 execdomains Execdomains, related to security (2.4)
264 fb Frame Buffer devices (2.4)
265 fs File system parameters, currently nfs/exports (2.4)
266 ide Directory containing info about the IDE subsystem
267 interrupts Interrupt usage
268 iomem Memory map (2.4)
269 ioports I/O port usage
270 irq Masks for irq to cpu affinity (2.4)(smp?)
271 isapnp ISA PnP (Plug&Play) Info (2.4)
272 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
273 kmsg Kernel messages
274 ksyms Kernel symbol table
275 loadavg Load average of last 1, 5 & 15 minutes
276 locks Kernel locks
277 meminfo Memory info
278 misc Miscellaneous
279 modules List of loaded modules
280 mounts Mounted filesystems
281 net Networking info (see text)
282 partitions Table of partitions known to the system
8b60756a 283 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
1da177e4
LT
284 decoupled by lspci (2.4)
285 rtc Real time clock
286 scsi SCSI info (see text)
287 slabinfo Slab pool info
288 stat Overall statistics
289 swaps Swap space utilization
290 sys See chapter 2
291 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
292 tty Info of tty drivers
293 uptime System uptime
294 version Kernel version
295 video bttv info of video resources (2.4)
296..............................................................................
297
298You can, for example, check which interrupts are currently in use and what
299they are used for by looking in the file /proc/interrupts:
300
301 > cat /proc/interrupts
302 CPU0
303 0: 8728810 XT-PIC timer
304 1: 895 XT-PIC keyboard
305 2: 0 XT-PIC cascade
306 3: 531695 XT-PIC aha152x
307 4: 2014133 XT-PIC serial
308 5: 44401 XT-PIC pcnet_cs
309 8: 2 XT-PIC rtc
310 11: 8 XT-PIC i82365
311 12: 182918 XT-PIC PS/2 Mouse
312 13: 1 XT-PIC fpu
313 14: 1232265 XT-PIC ide0
314 15: 7 XT-PIC ide1
315 NMI: 0
316
317In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
318output of a SMP machine):
319
320 > cat /proc/interrupts
321
322 CPU0 CPU1
323 0: 1243498 1214548 IO-APIC-edge timer
324 1: 8949 8958 IO-APIC-edge keyboard
325 2: 0 0 XT-PIC cascade
326 5: 11286 10161 IO-APIC-edge soundblaster
327 8: 1 0 IO-APIC-edge rtc
328 9: 27422 27407 IO-APIC-edge 3c503
329 12: 113645 113873 IO-APIC-edge PS/2 Mouse
330 13: 0 0 XT-PIC fpu
331 14: 22491 24012 IO-APIC-edge ide0
332 15: 2183 2415 IO-APIC-edge ide1
333 17: 30564 30414 IO-APIC-level eth0
334 18: 177 164 IO-APIC-level bttv
335 NMI: 2457961 2457959
336 LOC: 2457882 2457881
337 ERR: 2155
338
339NMI is incremented in this case because every timer interrupt generates a NMI
340(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
341
342LOC is the local interrupt counter of the internal APIC of every CPU.
343
344ERR is incremented in the case of errors in the IO-APIC bus (the bus that
345connects the CPUs in a SMP system. This means that an error has been detected,
346the IO-APIC automatically retry the transmission, so it should not be a big
347problem, but you should read the SMP-FAQ.
348
349In this context it could be interesting to note the new irq directory in 2.4.
350It could be used to set IRQ to CPU affinity, this means that you can "hook" an
351IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
352irq subdir is one subdir for each IRQ, and one file; prof_cpu_mask
353
354For example
355 > ls /proc/irq/
356 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
357 1 11 13 15 17 19 3 5 7 9
358 > ls /proc/irq/0/
359 smp_affinity
360
361The contents of the prof_cpu_mask file and each smp_affinity file for each IRQ
362is the same by default:
363
364 > cat /proc/irq/0/smp_affinity
365 ffffffff
366
c30fe7f7 367It's a bitmask, in which you can specify which CPUs can handle the IRQ, you can
1da177e4
LT
368set it by doing:
369
370 > echo 1 > /proc/irq/prof_cpu_mask
371
372This means that only the first CPU will handle the IRQ, but you can also echo 5
c30fe7f7 373which means that only the first and fourth CPU can handle the IRQ.
1da177e4
LT
374
375The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
376between all the CPUs which are allowed to handle it. As usual the kernel has
377more info than you and does a better job than you, so the defaults are the
378best choice for almost everyone.
379
380There are three more important subdirectories in /proc: net, scsi, and sys.
381The general rule is that the contents, or even the existence of these
382directories, depend on your kernel configuration. If SCSI is not enabled, the
383directory scsi may not exist. The same is true with the net, which is there
384only when networking support is present in the running kernel.
385
386The slabinfo file gives information about memory usage at the slab level.
387Linux uses slab pools for memory management above page level in version 2.2.
388Commonly used objects have their own slab pool (such as network buffers,
389directory cache, and so on).
390
391..............................................................................
392
393> cat /proc/buddyinfo
394
395Node 0, zone DMA 0 4 5 4 4 3 ...
396Node 0, zone Normal 1 0 0 1 101 8 ...
397Node 0, zone HighMem 2 0 0 1 1 0 ...
398
399Memory fragmentation is a problem under some workloads, and buddyinfo is a
400useful tool for helping diagnose these problems. Buddyinfo will give you a
401clue as to how big an area you can safely allocate, or why a previous
402allocation failed.
403
404Each column represents the number of pages of a certain order which are
405available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
406ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
407available in ZONE_NORMAL, etc...
408
409..............................................................................
410
411meminfo:
412
413Provides information about distribution and utilization of memory. This
414varies by architecture and compile options. The following is from a
41516GB PIII, which has highmem enabled. You may not have all of these fields.
416
417> cat /proc/meminfo
418
419
420MemTotal: 16344972 kB
421MemFree: 13634064 kB
422Buffers: 3656 kB
423Cached: 1195708 kB
424SwapCached: 0 kB
425Active: 891636 kB
426Inactive: 1077224 kB
427HighTotal: 15597528 kB
428HighFree: 13629632 kB
429LowTotal: 747444 kB
430LowFree: 4432 kB
431SwapTotal: 0 kB
432SwapFree: 0 kB
433Dirty: 968 kB
434Writeback: 0 kB
435Mapped: 280372 kB
436Slab: 684068 kB
437CommitLimit: 7669796 kB
438Committed_AS: 100056 kB
439PageTables: 24448 kB
440VmallocTotal: 112216 kB
441VmallocUsed: 428 kB
442VmallocChunk: 111088 kB
443
444 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
445 bits and the kernel binary code)
446 MemFree: The sum of LowFree+HighFree
447 Buffers: Relatively temporary storage for raw disk blocks
448 shouldn't get tremendously large (20MB or so)
449 Cached: in-memory cache for files read from the disk (the
450 pagecache). Doesn't include SwapCached
451 SwapCached: Memory that once was swapped out, is swapped back in but
452 still also is in the swapfile (if memory is needed it
453 doesn't need to be swapped out AGAIN because it is already
454 in the swapfile. This saves I/O)
455 Active: Memory that has been used more recently and usually not
456 reclaimed unless absolutely necessary.
457 Inactive: Memory which has been less recently used. It is more
458 eligible to be reclaimed for other purposes
459 HighTotal:
460 HighFree: Highmem is all memory above ~860MB of physical memory
461 Highmem areas are for use by userspace programs, or
462 for the pagecache. The kernel must use tricks to access
463 this memory, making it slower to access than lowmem.
464 LowTotal:
465 LowFree: Lowmem is memory which can be used for everything that
3f6dee9b 466 highmem can be used for, but it is also available for the
1da177e4
LT
467 kernel's use for its own data structures. Among many
468 other things, it is where everything from the Slab is
469 allocated. Bad things happen when you're out of lowmem.
470 SwapTotal: total amount of swap space available
471 SwapFree: Memory which has been evicted from RAM, and is temporarily
472 on the disk
473 Dirty: Memory which is waiting to get written back to the disk
474 Writeback: Memory which is actively being written back to the disk
475 Mapped: files which have been mmaped, such as libraries
e82443c0 476 Slab: in-kernel data structures cache
1da177e4
LT
477 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
478 this is the total amount of memory currently available to
479 be allocated on the system. This limit is only adhered to
480 if strict overcommit accounting is enabled (mode 2 in
481 'vm.overcommit_memory').
482 The CommitLimit is calculated with the following formula:
483 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
484 For example, on a system with 1G of physical RAM and 7G
485 of swap with a `vm.overcommit_ratio` of 30 it would
486 yield a CommitLimit of 7.3G.
487 For more details, see the memory overcommit documentation
488 in vm/overcommit-accounting.
489Committed_AS: The amount of memory presently allocated on the system.
490 The committed memory is a sum of all of the memory which
491 has been allocated by processes, even if it has not been
492 "used" by them as of yet. A process which malloc()'s 1G
493 of memory, but only touches 300M of it will only show up
494 as using 300M of memory even if it has the address space
495 allocated for the entire 1G. This 1G is memory which has
496 been "committed" to by the VM and can be used at any time
497 by the allocating application. With strict overcommit
498 enabled on the system (mode 2 in 'vm.overcommit_memory'),
499 allocations which would exceed the CommitLimit (detailed
500 above) will not be permitted. This is useful if one needs
501 to guarantee that processes will not fail due to lack of
502 memory once that memory has been successfully allocated.
503 PageTables: amount of memory dedicated to the lowest level of page
504 tables.
505VmallocTotal: total size of vmalloc memory area
506 VmallocUsed: amount of vmalloc area which is used
507VmallocChunk: largest contigious block of vmalloc area which is free
508
509
5101.3 IDE devices in /proc/ide
511----------------------------
512
513The subdirectory /proc/ide contains information about all IDE devices of which
514the kernel is aware. There is one subdirectory for each IDE controller, the
515file drivers and a link for each IDE device, pointing to the device directory
516in the controller specific subtree.
517
518The file drivers contains general information about the drivers used for the
519IDE devices:
520
521 > cat /proc/ide/drivers
522 ide-cdrom version 4.53
523 ide-disk version 1.08
524
525More detailed information can be found in the controller specific
526subdirectories. These are named ide0, ide1 and so on. Each of these
18d96779 527directories contains the files shown in table 1-5.
1da177e4
LT
528
529
18d96779 530Table 1-5: IDE controller info in /proc/ide/ide?
1da177e4
LT
531..............................................................................
532 File Content
533 channel IDE channel (0 or 1)
534 config Configuration (only for PCI/IDE bridge)
535 mate Mate name
536 model Type/Chipset of IDE controller
537..............................................................................
538
539Each device connected to a controller has a separate subdirectory in the
18d96779 540controllers directory. The files listed in table 1-6 are contained in these
1da177e4
LT
541directories.
542
543
18d96779 544Table 1-6: IDE device information
1da177e4
LT
545..............................................................................
546 File Content
547 cache The cache
548 capacity Capacity of the medium (in 512Byte blocks)
549 driver driver and version
550 geometry physical and logical geometry
551 identify device identify block
552 media media type
553 model device identifier
554 settings device setup
555 smart_thresholds IDE disk management thresholds
556 smart_values IDE disk management values
557..............................................................................
558
559The most interesting file is settings. This file contains a nice overview of
560the drive parameters:
561
562 # cat /proc/ide/ide0/hda/settings
563 name value min max mode
564 ---- ----- --- --- ----
565 bios_cyl 526 0 65535 rw
566 bios_head 255 0 255 rw
567 bios_sect 63 0 63 rw
568 breada_readahead 4 0 127 rw
569 bswap 0 0 1 r
570 file_readahead 72 0 2097151 rw
571 io_32bit 0 0 3 rw
572 keepsettings 0 0 1 rw
573 max_kb_per_request 122 1 127 rw
574 multcount 0 0 8 rw
575 nice1 1 0 1 rw
576 nowerr 0 0 1 rw
577 pio_mode write-only 0 255 w
578 slow 0 0 1 rw
579 unmaskirq 0 0 1 rw
580 using_dma 0 0 1 rw
581
582
5831.4 Networking info in /proc/net
584--------------------------------
585
586The subdirectory /proc/net follows the usual pattern. Table 1-6 shows the
587additional values you get for IP version 6 if you configure the kernel to
588support this. Table 1-7 lists the files and their meaning.
589
590
591Table 1-6: IPv6 info in /proc/net
592..............................................................................
593 File Content
594 udp6 UDP sockets (IPv6)
595 tcp6 TCP sockets (IPv6)
596 raw6 Raw device statistics (IPv6)
597 igmp6 IP multicast addresses, which this host joined (IPv6)
598 if_inet6 List of IPv6 interface addresses
599 ipv6_route Kernel routing table for IPv6
600 rt6_stats Global IPv6 routing tables statistics
601 sockstat6 Socket statistics (IPv6)
602 snmp6 Snmp data (IPv6)
603..............................................................................
604
605
606Table 1-7: Network info in /proc/net
607..............................................................................
608 File Content
609 arp Kernel ARP table
610 dev network devices with statistics
611 dev_mcast the Layer2 multicast groups a device is listening too
612 (interface index, label, number of references, number of bound
613 addresses).
614 dev_stat network device status
615 ip_fwchains Firewall chain linkage
616 ip_fwnames Firewall chain names
617 ip_masq Directory containing the masquerading tables
618 ip_masquerade Major masquerading table
619 netstat Network statistics
620 raw raw device statistics
621 route Kernel routing table
622 rpc Directory containing rpc info
623 rt_cache Routing cache
624 snmp SNMP data
625 sockstat Socket statistics
626 tcp TCP sockets
627 tr_rif Token ring RIF routing table
628 udp UDP sockets
629 unix UNIX domain sockets
630 wireless Wireless interface data (Wavelan etc)
631 igmp IP multicast addresses, which this host joined
632 psched Global packet scheduler parameters.
633 netlink List of PF_NETLINK sockets
634 ip_mr_vifs List of multicast virtual interfaces
635 ip_mr_cache List of multicast routing cache
636..............................................................................
637
638You can use this information to see which network devices are available in
639your system and how much traffic was routed over those devices:
640
641 > cat /proc/net/dev
642 Inter-|Receive |[...
643 face |bytes packets errs drop fifo frame compressed multicast|[...
644 lo: 908188 5596 0 0 0 0 0 0 [...
645 ppp0:15475140 20721 410 0 0 410 0 0 [...
646 eth0: 614530 7085 0 0 0 0 0 1 [...
647
648 ...] Transmit
649 ...] bytes packets errs drop fifo colls carrier compressed
650 ...] 908188 5596 0 0 0 0 0 0
651 ...] 1375103 17405 0 0 0 0 0 0
652 ...] 1703981 5535 0 0 0 3 0 0
653
654In addition, each Channel Bond interface has it's own directory. For
655example, the bond0 device will have a directory called /proc/net/bond0/.
656It will contain information that is specific to that bond, such as the
657current slaves of the bond, the link status of the slaves, and how
658many times the slaves link has failed.
659
6601.5 SCSI info
661-------------
662
663If you have a SCSI host adapter in your system, you'll find a subdirectory
664named after the driver for this adapter in /proc/scsi. You'll also see a list
665of all recognized SCSI devices in /proc/scsi:
666
667 >cat /proc/scsi/scsi
668 Attached devices:
669 Host: scsi0 Channel: 00 Id: 00 Lun: 00
670 Vendor: IBM Model: DGHS09U Rev: 03E0
671 Type: Direct-Access ANSI SCSI revision: 03
672 Host: scsi0 Channel: 00 Id: 06 Lun: 00
673 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
674 Type: CD-ROM ANSI SCSI revision: 02
675
676
677The directory named after the driver has one file for each adapter found in
678the system. These files contain information about the controller, including
679the used IRQ and the IO address range. The amount of information shown is
680dependent on the adapter you use. The example shows the output for an Adaptec
681AHA-2940 SCSI adapter:
682
683 > cat /proc/scsi/aic7xxx/0
684
685 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
686 Compile Options:
687 TCQ Enabled By Default : Disabled
688 AIC7XXX_PROC_STATS : Disabled
689 AIC7XXX_RESET_DELAY : 5
690 Adapter Configuration:
691 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
692 Ultra Wide Controller
693 PCI MMAPed I/O Base: 0xeb001000
694 Adapter SEEPROM Config: SEEPROM found and used.
695 Adaptec SCSI BIOS: Enabled
696 IRQ: 10
697 SCBs: Active 0, Max Active 2,
698 Allocated 15, HW 16, Page 255
699 Interrupts: 160328
700 BIOS Control Word: 0x18b6
701 Adapter Control Word: 0x005b
702 Extended Translation: Enabled
703 Disconnect Enable Flags: 0xffff
704 Ultra Enable Flags: 0x0001
705 Tag Queue Enable Flags: 0x0000
706 Ordered Queue Tag Flags: 0x0000
707 Default Tag Queue Depth: 8
708 Tagged Queue By Device array for aic7xxx host instance 0:
709 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
710 Actual queue depth per device for aic7xxx host instance 0:
711 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
712 Statistics:
713 (scsi0:0:0:0)
714 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
715 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
716 Total transfers 160151 (74577 reads and 85574 writes)
717 (scsi0:0:6:0)
718 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
719 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
720 Total transfers 0 (0 reads and 0 writes)
721
722
7231.6 Parallel port info in /proc/parport
724---------------------------------------
725
726The directory /proc/parport contains information about the parallel ports of
727your system. It has one subdirectory for each port, named after the port
728number (0,1,2,...).
729
730These directories contain the four files shown in Table 1-8.
731
732
733Table 1-8: Files in /proc/parport
734..............................................................................
735 File Content
736 autoprobe Any IEEE-1284 device ID information that has been acquired.
737 devices list of the device drivers using that port. A + will appear by the
738 name of the device currently using the port (it might not appear
739 against any).
740 hardware Parallel port's base address, IRQ line and DMA channel.
741 irq IRQ that parport is using for that port. This is in a separate
742 file to allow you to alter it by writing a new value in (IRQ
743 number or none).
744..............................................................................
745
7461.7 TTY info in /proc/tty
747-------------------------
748
749Information about the available and actually used tty's can be found in the
750directory /proc/tty.You'll find entries for drivers and line disciplines in
751this directory, as shown in Table 1-9.
752
753
754Table 1-9: Files in /proc/tty
755..............................................................................
756 File Content
757 drivers list of drivers and their usage
758 ldiscs registered line disciplines
759 driver/serial usage statistic and status of single tty lines
760..............................................................................
761
762To see which tty's are currently in use, you can simply look into the file
763/proc/tty/drivers:
764
765 > cat /proc/tty/drivers
766 pty_slave /dev/pts 136 0-255 pty:slave
767 pty_master /dev/ptm 128 0-255 pty:master
768 pty_slave /dev/ttyp 3 0-255 pty:slave
769 pty_master /dev/pty 2 0-255 pty:master
770 serial /dev/cua 5 64-67 serial:callout
771 serial /dev/ttyS 4 64-67 serial
772 /dev/tty0 /dev/tty0 4 0 system:vtmaster
773 /dev/ptmx /dev/ptmx 5 2 system
774 /dev/console /dev/console 5 1 system:console
775 /dev/tty /dev/tty 5 0 system:/dev/tty
776 unknown /dev/tty 4 1-63 console
777
778
7791.8 Miscellaneous kernel statistics in /proc/stat
780-------------------------------------------------
781
782Various pieces of information about kernel activity are available in the
783/proc/stat file. All of the numbers reported in this file are aggregates
784since the system first booted. For a quick look, simply cat the file:
785
786 > cat /proc/stat
787 cpu 2255 34 2290 22625563 6290 127 456
788 cpu0 1132 34 1441 11311718 3675 127 438
789 cpu1 1123 0 849 11313845 2614 0 18
790 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
791 ctxt 1990473
792 btime 1062191376
793 processes 2915
794 procs_running 1
795 procs_blocked 0
796
797The very first "cpu" line aggregates the numbers in all of the other "cpuN"
798lines. These numbers identify the amount of time the CPU has spent performing
799different kinds of work. Time units are in USER_HZ (typically hundredths of a
800second). The meanings of the columns are as follows, from left to right:
801
802- user: normal processes executing in user mode
803- nice: niced processes executing in user mode
804- system: processes executing in kernel mode
805- idle: twiddling thumbs
806- iowait: waiting for I/O to complete
807- irq: servicing interrupts
808- softirq: servicing softirqs
809
810The "intr" line gives counts of interrupts serviced since boot time, for each
811of the possible system interrupts. The first column is the total of all
812interrupts serviced; each subsequent column is the total for that particular
813interrupt.
814
815The "ctxt" line gives the total number of context switches across all CPUs.
816
817The "btime" line gives the time at which the system booted, in seconds since
818the Unix epoch.
819
820The "processes" line gives the number of processes and threads created, which
821includes (but is not limited to) those created by calls to the fork() and
822clone() system calls.
823
824The "procs_running" line gives the number of processes currently running on
825CPUs.
826
827The "procs_blocked" line gives the number of processes currently blocked,
828waiting for I/O to complete.
829
830
831------------------------------------------------------------------------------
832Summary
833------------------------------------------------------------------------------
834The /proc file system serves information about the running system. It not only
835allows access to process data but also allows you to request the kernel status
836by reading files in the hierarchy.
837
838The directory structure of /proc reflects the types of information and makes
839it easy, if not obvious, where to look for specific data.
840------------------------------------------------------------------------------
841
842------------------------------------------------------------------------------
843CHAPTER 2: MODIFYING SYSTEM PARAMETERS
844------------------------------------------------------------------------------
845
846------------------------------------------------------------------------------
847In This Chapter
848------------------------------------------------------------------------------
849* Modifying kernel parameters by writing into files found in /proc/sys
850* Exploring the files which modify certain parameters
851* Review of the /proc/sys file tree
852------------------------------------------------------------------------------
853
854
855A very interesting part of /proc is the directory /proc/sys. This is not only
856a source of information, it also allows you to change parameters within the
857kernel. Be very careful when attempting this. You can optimize your system,
858but you can also cause it to crash. Never alter kernel parameters on a
859production system. Set up a development machine and test to make sure that
860everything works the way you want it to. You may have no alternative but to
861reboot the machine once an error has been made.
862
863To change a value, simply echo the new value into the file. An example is
864given below in the section on the file system data. You need to be root to do
865this. You can create your own boot script to perform this every time your
866system boots.
867
868The files in /proc/sys can be used to fine tune and monitor miscellaneous and
869general things in the operation of the Linux kernel. Since some of the files
870can inadvertently disrupt your system, it is advisable to read both
871documentation and source before actually making adjustments. In any case, be
872very careful when writing to any of these files. The entries in /proc may
873change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
874review the kernel documentation in the directory /usr/src/linux/Documentation.
875This chapter is heavily based on the documentation included in the pre 2.2
876kernels, and became part of it in version 2.2.1 of the Linux kernel.
877
8782.1 /proc/sys/fs - File system data
879-----------------------------------
880
881This subdirectory contains specific file system, file handle, inode, dentry
882and quota information.
883
884Currently, these files are in /proc/sys/fs:
885
886dentry-state
887------------
888
889Status of the directory cache. Since directory entries are dynamically
890allocated and deallocated, this file indicates the current status. It holds
891six values, in which the last two are not used and are always zero. The others
892are listed in table 2-1.
893
894
895Table 2-1: Status files of the directory cache
896..............................................................................
897 File Content
898 nr_dentry Almost always zero
899 nr_unused Number of unused cache entries
900 age_limit
901 in seconds after the entry may be reclaimed, when memory is short
902 want_pages internally
903..............................................................................
904
905dquot-nr and dquot-max
906----------------------
907
908The file dquot-max shows the maximum number of cached disk quota entries.
909
910The file dquot-nr shows the number of allocated disk quota entries and the
911number of free disk quota entries.
912
913If the number of available cached disk quotas is very low and you have a large
914number of simultaneous system users, you might want to raise the limit.
915
916file-nr and file-max
917--------------------
918
919The kernel allocates file handles dynamically, but doesn't free them again at
920this time.
921
922The value in file-max denotes the maximum number of file handles that the
923Linux kernel will allocate. When you get a lot of error messages about running
924out of file handles, you might want to raise this limit. The default value is
92510% of RAM in kilobytes. To change it, just write the new number into the
926file:
927
928 # cat /proc/sys/fs/file-max
929 4096
930 # echo 8192 > /proc/sys/fs/file-max
931 # cat /proc/sys/fs/file-max
932 8192
933
934
935This method of revision is useful for all customizable parameters of the
936kernel - simply echo the new value to the corresponding file.
937
938Historically, the three values in file-nr denoted the number of allocated file
939handles, the number of allocated but unused file handles, and the maximum
940number of file handles. Linux 2.6 always reports 0 as the number of free file
941handles -- this is not an error, it just means that the number of allocated
942file handles exactly matches the number of used file handles.
943
944Attempts to allocate more file descriptors than file-max are reported with
945printk, look for "VFS: file-max limit <number> reached".
946
947inode-state and inode-nr
948------------------------
949
950The file inode-nr contains the first two items from inode-state, so we'll skip
951to that file...
952
953inode-state contains two actual numbers and five dummy values. The numbers
954are nr_inodes and nr_free_inodes (in order of appearance).
955
956nr_inodes
957~~~~~~~~~
958
959Denotes the number of inodes the system has allocated. This number will
960grow and shrink dynamically.
961
962nr_free_inodes
963--------------
964
965Represents the number of free inodes. Ie. The number of inuse inodes is
966(nr_inodes - nr_free_inodes).
967
1da177e4
LT
968aio-nr and aio-max-nr
969---------------------
970
971aio-nr is the running total of the number of events specified on the
972io_setup system call for all currently active aio contexts. If aio-nr
973reaches aio-max-nr then io_setup will fail with EAGAIN. Note that
974raising aio-max-nr does not result in the pre-allocation or re-sizing
975of any kernel data structures.
976
9772.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
978-----------------------------------------------------------
979
980Besides these files, there is the subdirectory /proc/sys/fs/binfmt_misc. This
981handles the kernel support for miscellaneous binary formats.
982
983Binfmt_misc provides the ability to register additional binary formats to the
984Kernel without compiling an additional module/kernel. Therefore, binfmt_misc
985needs to know magic numbers at the beginning or the filename extension of the
986binary.
987
988It works by maintaining a linked list of structs that contain a description of
989a binary format, including a magic with size (or the filename extension),
990offset and mask, and the interpreter name. On request it invokes the given
991interpreter with the original program as argument, as binfmt_java and
992binfmt_em86 and binfmt_mz do. Since binfmt_misc does not define any default
993binary-formats, you have to register an additional binary-format.
994
995There are two general files in binfmt_misc and one file per registered format.
996The two general files are register and status.
997
998Registering a new binary format
999-------------------------------
1000
1001To register a new binary format you have to issue the command
1002
1003 echo :name:type:offset:magic:mask:interpreter: > /proc/sys/fs/binfmt_misc/register
1004
1005
1006
1007with appropriate name (the name for the /proc-dir entry), offset (defaults to
10080, if omitted), magic, mask (which can be omitted, defaults to all 0xff) and
1009last but not least, the interpreter that is to be invoked (for example and
1010testing /bin/echo). Type can be M for usual magic matching or E for filename
1011extension matching (give extension in place of magic).
1012
1013Check or reset the status of the binary format handler
1014------------------------------------------------------
1015
1016If you do a cat on the file /proc/sys/fs/binfmt_misc/status, you will get the
1017current status (enabled/disabled) of binfmt_misc. Change the status by echoing
10180 (disables) or 1 (enables) or -1 (caution: this clears all previously
1019registered binary formats) to status. For example echo 0 > status to disable
1020binfmt_misc (temporarily).
1021
1022Status of a single handler
1023--------------------------
1024
1025Each registered handler has an entry in /proc/sys/fs/binfmt_misc. These files
1026perform the same function as status, but their scope is limited to the actual
1027binary format. By cating this file, you also receive all related information
1028about the interpreter/magic of the binfmt.
1029
1030Example usage of binfmt_misc (emulate binfmt_java)
1031--------------------------------------------------
1032
1033 cd /proc/sys/fs/binfmt_misc
1034 echo ':Java:M::\xca\xfe\xba\xbe::/usr/local/java/bin/javawrapper:' > register
1035 echo ':HTML:E::html::/usr/local/java/bin/appletviewer:' > register
1036 echo ':Applet:M::<!--applet::/usr/local/java/bin/appletviewer:' > register
1037 echo ':DEXE:M::\x0eDEX::/usr/bin/dosexec:' > register
1038
1039
1040These four lines add support for Java executables and Java applets (like
1041binfmt_java, additionally recognizing the .html extension with no need to put
1042<!--applet> to every applet file). You have to install the JDK and the
1043shell-script /usr/local/java/bin/javawrapper too. It works around the
1044brokenness of the Java filename handling. To add a Java binary, just create a
1045link to the class-file somewhere in the path.
1046
10472.3 /proc/sys/kernel - general kernel parameters
1048------------------------------------------------
1049
1050This directory reflects general kernel behaviors. As I've said before, the
1051contents depend on your configuration. Here you'll find the most important
1052files, along with descriptions of what they mean and how to use them.
1053
1054acct
1055----
1056
1057The file contains three values; highwater, lowwater, and frequency.
1058
1059It exists only when BSD-style process accounting is enabled. These values
1060control its behavior. If the free space on the file system where the log lives
1061goes below lowwater percentage, accounting suspends. If it goes above
1062highwater percentage, accounting resumes. Frequency determines how often you
1063check the amount of free space (value is in seconds). Default settings are: 4,
10642, and 30. That is, suspend accounting if there is less than 2 percent free;
1065resume it if we have a value of 3 or more percent; consider information about
1066the amount of free space valid for 30 seconds
1067
1068ctrl-alt-del
1069------------
1070
1071When the value in this file is 0, ctrl-alt-del is trapped and sent to the init
1072program to handle a graceful restart. However, when the value is greater that
1073zero, Linux's reaction to this key combination will be an immediate reboot,
1074without syncing its dirty buffers.
1075
1076[NOTE]
1077 When a program (like dosemu) has the keyboard in raw mode, the
1078 ctrl-alt-del is intercepted by the program before it ever reaches the
1079 kernel tty layer, and it is up to the program to decide what to do with
1080 it.
1081
1082domainname and hostname
1083-----------------------
1084
1085These files can be controlled to set the NIS domainname and hostname of your
1086box. For the classic darkstar.frop.org a simple:
1087
1088 # echo "darkstar" > /proc/sys/kernel/hostname
1089 # echo "frop.org" > /proc/sys/kernel/domainname
1090
1091
1092would suffice to set your hostname and NIS domainname.
1093
1094osrelease, ostype and version
1095-----------------------------
1096
1097The names make it pretty obvious what these fields contain:
1098
1099 > cat /proc/sys/kernel/osrelease
1100 2.2.12
1101
1102 > cat /proc/sys/kernel/ostype
1103 Linux
1104
1105 > cat /proc/sys/kernel/version
1106 #4 Fri Oct 1 12:41:14 PDT 1999
1107
1108
1109The files osrelease and ostype should be clear enough. Version needs a little
1110more clarification. The #4 means that this is the 4th kernel built from this
1111source base and the date after it indicates the time the kernel was built. The
1112only way to tune these values is to rebuild the kernel.
1113
1114panic
1115-----
1116
1117The value in this file represents the number of seconds the kernel waits
1118before rebooting on a panic. When you use the software watchdog, the
1119recommended setting is 60. If set to 0, the auto reboot after a kernel panic
1120is disabled, which is the default setting.
1121
1122printk
1123------
1124
1125The four values in printk denote
1126* console_loglevel,
1127* default_message_loglevel,
1128* minimum_console_loglevel and
1129* default_console_loglevel
1130respectively.
1131
1132These values influence printk() behavior when printing or logging error
1133messages, which come from inside the kernel. See syslog(2) for more
1134information on the different log levels.
1135
1136console_loglevel
1137----------------
1138
1139Messages with a higher priority than this will be printed to the console.
1140
1141default_message_level
1142---------------------
1143
1144Messages without an explicit priority will be printed with this priority.
1145
1146minimum_console_loglevel
1147------------------------
1148
1149Minimum (highest) value to which the console_loglevel can be set.
1150
1151default_console_loglevel
1152------------------------
1153
1154Default value for console_loglevel.
1155
1156sg-big-buff
1157-----------
1158
1159This file shows the size of the generic SCSI (sg) buffer. At this point, you
1160can't tune it yet, but you can change it at compile time by editing
1161include/scsi/sg.h and changing the value of SG_BIG_BUFF.
1162
1163If you use a scanner with SANE (Scanner Access Now Easy) you might want to set
1164this to a higher value. Refer to the SANE documentation on this issue.
1165
1166modprobe
1167--------
1168
1169The location where the modprobe binary is located. The kernel uses this
1170program to load modules on demand.
1171
1172unknown_nmi_panic
1173-----------------
1174
1175The value in this file affects behavior of handling NMI. When the value is
1176non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
1177debugging information is displayed on console.
1178
1179NMI switch that most IA32 servers have fires unknown NMI up, for example.
1180If a system hangs up, try pressing the NMI switch.
1181
e33e89ab
DZ
1182nmi_watchdog
1183------------
1184
1185Enables/Disables the NMI watchdog on x86 systems. When the value is non-zero
1186the NMI watchdog is enabled and will continuously test all online cpus to
1187determine whether or not they are still functioning properly.
1188
1189Because the NMI watchdog shares registers with oprofile, by disabling the NMI
1190watchdog, oprofile may have more registers to utilize.
1da177e4 1191
5096add8
KC
1192maps_protect
1193------------
1194
1195Enables/Disables the protection of the per-process proc entries "maps" and
1196"smaps". When enabled, the contents of these files are visible only to
1197readers that are allowed to ptrace() the given process.
1198
1da177e4
LT
1199
12002.4 /proc/sys/vm - The virtual memory subsystem
1201-----------------------------------------------
1202
1203The files in this directory can be used to tune the operation of the virtual
1204memory (VM) subsystem of the Linux kernel.
1205
1206vfs_cache_pressure
1207------------------
1208
1209Controls the tendency of the kernel to reclaim the memory which is used for
1210caching of directory and inode objects.
1211
1212At the default value of vfs_cache_pressure=100 the kernel will attempt to
1213reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1214swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1215to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100
1216causes the kernel to prefer to reclaim dentries and inodes.
1217
1218dirty_background_ratio
1219----------------------
1220
1221Contains, as a percentage of total system memory, the number of pages at which
1222the pdflush background writeback daemon will start writing out dirty data.
1223
1224dirty_ratio
1225-----------------
1226
1227Contains, as a percentage of total system memory, the number of pages at which
1228a process which is generating disk writes will itself start writing out dirty
1229data.
1230
1231dirty_writeback_centisecs
1232-------------------------
1233
1234The pdflush writeback daemons will periodically wake up and write `old' data
1235out to disk. This tunable expresses the interval between those wakeups, in
1236100'ths of a second.
1237
1238Setting this to zero disables periodic writeback altogether.
1239
1240dirty_expire_centisecs
1241----------------------
1242
1243This tunable is used to define when dirty data is old enough to be eligible
1244for writeout by the pdflush daemons. It is expressed in 100'ths of a second.
1245Data which has been dirty in-memory for longer than this interval will be
1246written out next time a pdflush daemon wakes up.
1247
1248legacy_va_layout
1249----------------
1250
1251If non-zero, this sysctl disables the new 32-bit mmap mmap layout - the kernel
1252will use the legacy (2.4) layout for all processes.
1253
1254lower_zone_protection
1255---------------------
1256
1257For some specialised workloads on highmem machines it is dangerous for
1258the kernel to allow process memory to be allocated from the "lowmem"
1259zone. This is because that memory could then be pinned via the mlock()
1260system call, or by unavailability of swapspace.
1261
1262And on large highmem machines this lack of reclaimable lowmem memory
1263can be fatal.
1264
1265So the Linux page allocator has a mechanism which prevents allocations
1266which _could_ use highmem from using too much lowmem. This means that
1267a certain amount of lowmem is defended from the possibility of being
1268captured into pinned user memory.
1269
1270(The same argument applies to the old 16 megabyte ISA DMA region. This
1271mechanism will also defend that region from allocations which could use
1272highmem or lowmem).
1273
1274The `lower_zone_protection' tunable determines how aggressive the kernel is
1275in defending these lower zones. The default value is zero - no
1276protection at all.
1277
1278If you have a machine which uses highmem or ISA DMA and your
1279applications are using mlock(), or if you are running with no swap then
1280you probably should increase the lower_zone_protection setting.
1281
1282The units of this tunable are fairly vague. It is approximately equal
4ae0edc2 1283to "megabytes," so setting lower_zone_protection=100 will protect around 100
1da177e4 1284megabytes of the lowmem zone from user allocations. It will also make
4ae0edc2 1285those 100 megabytes unavailable for use by applications and by
1da177e4
LT
1286pagecache, so there is a cost.
1287
1288The effects of this tunable may be observed by monitoring
1289/proc/meminfo:LowFree. Write a single huge file and observe the point
1290at which LowFree ceases to fall.
1291
1292A reasonable value for lower_zone_protection is 100.
1293
1294page-cluster
1295------------
1296
1297page-cluster controls the number of pages which are written to swap in
1298a single attempt. The swap I/O size.
1299
1300It is a logarithmic value - setting it to zero means "1 page", setting
1301it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
1302
1303The default value is three (eight pages at a time). There may be some
1304small benefits in tuning this to a different value if your workload is
1305swap-intensive.
1306
1307overcommit_memory
1308-----------------
1309
af97c722
CE
1310Controls overcommit of system memory, possibly allowing processes
1311to allocate (but not use) more memory than is actually available.
1312
1313
13140 - Heuristic overcommit handling. Obvious overcommits of
1315 address space are refused. Used for a typical system. It
1316 ensures a seriously wild allocation fails while allowing
1317 overcommit to reduce swap usage. root is allowed to
53cb4726 1318 allocate slightly more memory in this mode. This is the
af97c722
CE
1319 default.
1320
13211 - Always overcommit. Appropriate for some scientific
1322 applications.
1323
13242 - Don't overcommit. The total address space commit
1325 for the system is not permitted to exceed swap plus a
1326 configurable percentage (default is 50) of physical RAM.
1327 Depending on the percentage you use, in most situations
1328 this means a process will not be killed while attempting
1329 to use already-allocated memory but will receive errors
1330 on memory allocation as appropriate.
1331
1332overcommit_ratio
1333----------------
1334
1335Percentage of physical memory size to include in overcommit calculations
1336(see above.)
1337
1338Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100)
1339
1340 swapspace = total size of all swap areas
1341 physmem = size of physical memory in system
1da177e4
LT
1342
1343nr_hugepages and hugetlb_shm_group
1344----------------------------------
1345
1346nr_hugepages configures number of hugetlb page reserved for the system.
1347
1348hugetlb_shm_group contains group id that is allowed to create SysV shared
1349memory segment using hugetlb page.
1350
ed7ed365
MG
1351hugepages_treat_as_movable
1352--------------------------
1353
1354This parameter is only useful when kernelcore= is specified at boot time to
1355create ZONE_MOVABLE for pages that may be reclaimed or migrated. Huge pages
1356are not movable so are not normally allocated from ZONE_MOVABLE. A non-zero
1357value written to hugepages_treat_as_movable allows huge pages to be allocated
1358from ZONE_MOVABLE.
1359
1360Once enabled, the ZONE_MOVABLE is treated as an area of memory the huge
1361pages pool can easily grow or shrink within. Assuming that applications are
1362not running that mlock() a lot of memory, it is likely the huge pages pool
1363can grow to the size of ZONE_MOVABLE by repeatedly entering the desired value
1364into nr_hugepages and triggering page reclaim.
1365
1da177e4
LT
1366laptop_mode
1367-----------
1368
1369laptop_mode is a knob that controls "laptop mode". All the things that are
1370controlled by this knob are discussed in Documentation/laptop-mode.txt.
1371
1372block_dump
1373----------
1374
1375block_dump enables block I/O debugging when set to a nonzero value. More
1376information on block I/O debugging is in Documentation/laptop-mode.txt.
1377
1378swap_token_timeout
1379------------------
1380
1381This file contains valid hold time of swap out protection token. The Linux
1382VM has token based thrashing control mechanism and uses the token to prevent
1383unnecessary page faults in thrashing situation. The unit of the value is
1384second. The value would be useful to tune thrashing behavior.
1385
9d0243bc
AM
1386drop_caches
1387-----------
1388
1389Writing to this will cause the kernel to drop clean caches, dentries and
1390inodes from memory, causing that memory to become free.
1391
1392To free pagecache:
1393 echo 1 > /proc/sys/vm/drop_caches
1394To free dentries and inodes:
1395 echo 2 > /proc/sys/vm/drop_caches
1396To free pagecache, dentries and inodes:
1397 echo 3 > /proc/sys/vm/drop_caches
1398
1399As this is a non-destructive operation and dirty objects are not freeable, the
1400user should run `sync' first.
1401
1402
1da177e4
LT
14032.5 /proc/sys/dev - Device specific parameters
1404----------------------------------------------
1405
1406Currently there is only support for CDROM drives, and for those, there is only
1407one read-only file containing information about the CD-ROM drives attached to
1408the system:
1409
1410 >cat /proc/sys/dev/cdrom/info
1411 CD-ROM information, Id: cdrom.c 2.55 1999/04/25
1412
1413 drive name: sr0 hdb
1414 drive speed: 32 40
1415 drive # of slots: 1 0
1416 Can close tray: 1 1
1417 Can open tray: 1 1
1418 Can lock tray: 1 1
1419 Can change speed: 1 1
1420 Can select disk: 0 1
1421 Can read multisession: 1 1
1422 Can read MCN: 1 1
1423 Reports media changed: 1 1
1424 Can play audio: 1 1
1425
1426
1427You see two drives, sr0 and hdb, along with a list of their features.
1428
14292.6 /proc/sys/sunrpc - Remote procedure calls
1430---------------------------------------------
1431
1432This directory contains four files, which enable or disable debugging for the
1433RPC functions NFS, NFS-daemon, RPC and NLM. The default values are 0. They can
1434be set to one to turn debugging on. (The default value is 0 for each)
1435
14362.7 /proc/sys/net - Networking stuff
1437------------------------------------
1438
1439The interface to the networking parts of the kernel is located in
1440/proc/sys/net. Table 2-3 shows all possible subdirectories. You may see only
1441some of them, depending on your kernel's configuration.
1442
1443
1444Table 2-3: Subdirectories in /proc/sys/net
1445..............................................................................
1446 Directory Content Directory Content
1447 core General parameter appletalk Appletalk protocol
1448 unix Unix domain sockets netrom NET/ROM
1449 802 E802 protocol ax25 AX25
1450 ethernet Ethernet protocol rose X.25 PLP layer
1451 ipv4 IP version 4 x25 X.25 protocol
1452 ipx IPX token-ring IBM token ring
1453 bridge Bridging decnet DEC net
1454 ipv6 IP version 6
1455..............................................................................
1456
1457We will concentrate on IP networking here. Since AX15, X.25, and DEC Net are
1458only minor players in the Linux world, we'll skip them in this chapter. You'll
1459find some short info on Appletalk and IPX further on in this chapter. Review
1460the online documentation and the kernel source to get a detailed view of the
1461parameters for those protocols. In this section we'll discuss the
1462subdirectories printed in bold letters in the table above. As default values
1463are suitable for most needs, there is no need to change these values.
1464
1465/proc/sys/net/core - Network core options
1466-----------------------------------------
1467
1468rmem_default
1469------------
1470
1471The default setting of the socket receive buffer in bytes.
1472
1473rmem_max
1474--------
1475
1476The maximum receive socket buffer size in bytes.
1477
1478wmem_default
1479------------
1480
1481The default setting (in bytes) of the socket send buffer.
1482
1483wmem_max
1484--------
1485
1486The maximum send socket buffer size in bytes.
1487
1488message_burst and message_cost
1489------------------------------
1490
1491These parameters are used to limit the warning messages written to the kernel
1492log from the networking code. They enforce a rate limit to make a
1493denial-of-service attack impossible. A higher message_cost factor, results in
1494fewer messages that will be written. Message_burst controls when messages will
1495be dropped. The default settings limit warning messages to one every five
1496seconds.
1497
a2a316fd
SH
1498warnings
1499--------
1500
1501This controls console messages from the networking stack that can occur because
1502of problems on the network like duplicate address or bad checksums. Normally,
1503this should be enabled, but if the problem persists the messages can be
1504disabled.
1505
1506
1da177e4
LT
1507netdev_max_backlog
1508------------------
1509
1510Maximum number of packets, queued on the INPUT side, when the interface
1511receives packets faster than kernel can process them.
1512
1513optmem_max
1514----------
1515
1516Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
1517of struct cmsghdr structures with appended data.
1518
1519/proc/sys/net/unix - Parameters for Unix domain sockets
1520-------------------------------------------------------
1521
1522There are only two files in this subdirectory. They control the delays for
1523deleting and destroying socket descriptors.
1524
15252.8 /proc/sys/net/ipv4 - IPV4 settings
1526--------------------------------------
1527
1528IP version 4 is still the most used protocol in Unix networking. It will be
1529replaced by IP version 6 in the next couple of years, but for the moment it's
1530the de facto standard for the internet and is used in most networking
1531environments around the world. Because of the importance of this protocol,
1532we'll have a deeper look into the subtree controlling the behavior of the IPv4
1533subsystem of the Linux kernel.
1534
1535Let's start with the entries in /proc/sys/net/ipv4.
1536
1537ICMP settings
1538-------------
1539
1540icmp_echo_ignore_all and icmp_echo_ignore_broadcasts
1541----------------------------------------------------
1542
1543Turn on (1) or off (0), if the kernel should ignore all ICMP ECHO requests, or
1544just those to broadcast and multicast addresses.
1545
1546Please note that if you accept ICMP echo requests with a broadcast/multi\-cast
1547destination address your network may be used as an exploder for denial of
1548service packet flooding attacks to other hosts.
1549
1550icmp_destunreach_rate, icmp_echoreply_rate, icmp_paramprob_rate and icmp_timeexeed_rate
1551---------------------------------------------------------------------------------------
1552
1553Sets limits for sending ICMP packets to specific targets. A value of zero
1554disables all limiting. Any positive value sets the maximum package rate in
1555hundredth of a second (on Intel systems).
1556
1557IP settings
1558-----------
1559
1560ip_autoconfig
1561-------------
1562
1563This file contains the number one if the host received its IP configuration by
1564RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero.
1565
1566ip_default_ttl
1567--------------
1568
1569TTL (Time To Live) for IPv4 interfaces. This is simply the maximum number of
1570hops a packet may travel.
1571
1572ip_dynaddr
1573----------
1574
1575Enable dynamic socket address rewriting on interface address change. This is
1576useful for dialup interface with changing IP addresses.
1577
1578ip_forward
1579----------
1580
1581Enable or disable forwarding of IP packages between interfaces. Changing this
1582value resets all other parameters to their default values. They differ if the
1583kernel is configured as host or router.
1584
1585ip_local_port_range
1586-------------------
1587
1588Range of ports used by TCP and UDP to choose the local port. Contains two
1589numbers, the first number is the lowest port, the second number the highest
1590local port. Default is 1024-4999. Should be changed to 32768-61000 for
1591high-usage systems.
1592
1593ip_no_pmtu_disc
1594---------------
1595
1596Global switch to turn path MTU discovery off. It can also be set on a per
1597socket basis by the applications or on a per route basis.
1598
1599ip_masq_debug
1600-------------
1601
1602Enable/disable debugging of IP masquerading.
1603
1604IP fragmentation settings
1605-------------------------
1606
1607ipfrag_high_trash and ipfrag_low_trash
1608--------------------------------------
1609
1610Maximum memory used to reassemble IP fragments. When ipfrag_high_thresh bytes
1611of memory is allocated for this purpose, the fragment handler will toss
1612packets until ipfrag_low_thresh is reached.
1613
1614ipfrag_time
1615-----------
1616
1617Time in seconds to keep an IP fragment in memory.
1618
1619TCP settings
1620------------
1621
1622tcp_ecn
1623-------
1624
fa00e7e1 1625This file controls the use of the ECN bit in the IPv4 headers. This is a new
1da177e4 1626feature about Explicit Congestion Notification, but some routers and firewalls
fa00e7e1
ML
1627block traffic that has this bit set, so it could be necessary to echo 0 to
1628/proc/sys/net/ipv4/tcp_ecn if you want to talk to these sites. For more info
1da177e4
LT
1629you could read RFC2481.
1630
1631tcp_retrans_collapse
1632--------------------
1633
1634Bug-to-bug compatibility with some broken printers. On retransmit, try to send
1635larger packets to work around bugs in certain TCP stacks. Can be turned off by
1636setting it to zero.
1637
1638tcp_keepalive_probes
1639--------------------
1640
1641Number of keep alive probes TCP sends out, until it decides that the
1642connection is broken.
1643
1644tcp_keepalive_time
1645------------------
1646
1647How often TCP sends out keep alive messages, when keep alive is enabled. The
1648default is 2 hours.
1649
1650tcp_syn_retries
1651---------------
1652
1653Number of times initial SYNs for a TCP connection attempt will be
1654retransmitted. Should not be higher than 255. This is only the timeout for
1655outgoing connections, for incoming connections the number of retransmits is
1656defined by tcp_retries1.
1657
1658tcp_sack
1659--------
1660
1661Enable select acknowledgments after RFC2018.
1662
1663tcp_timestamps
1664--------------
1665
1666Enable timestamps as defined in RFC1323.
1667
1668tcp_stdurg
1669----------
1670
1671Enable the strict RFC793 interpretation of the TCP urgent pointer field. The
1672default is to use the BSD compatible interpretation of the urgent pointer
1673pointing to the first byte after the urgent data. The RFC793 interpretation is
1674to have it point to the last byte of urgent data. Enabling this option may
2fe0ae78 1675lead to interoperability problems. Disabled by default.
1da177e4
LT
1676
1677tcp_syncookies
1678--------------
1679
1680Only valid when the kernel was compiled with CONFIG_SYNCOOKIES. Send out
1681syncookies when the syn backlog queue of a socket overflows. This is to ward
1682off the common 'syn flood attack'. Disabled by default.
1683
1684Note that the concept of a socket backlog is abandoned. This means the peer
1685may not receive reliable error messages from an over loaded server with
1686syncookies enabled.
1687
1688tcp_window_scaling
1689------------------
1690
1691Enable window scaling as defined in RFC1323.
1692
1693tcp_fin_timeout
1694---------------
1695
1696The length of time in seconds it takes to receive a final FIN before the
1697socket is always closed. This is strictly a violation of the TCP
1698specification, but required to prevent denial-of-service attacks.
1699
1700tcp_max_ka_probes
1701-----------------
1702
1703Indicates how many keep alive probes are sent per slow timer run. Should not
1704be set too high to prevent bursts.
1705
1706tcp_max_syn_backlog
1707-------------------
1708
1709Length of the per socket backlog queue. Since Linux 2.2 the backlog specified
1710in listen(2) only specifies the length of the backlog queue of already
1711established sockets. When more connection requests arrive Linux starts to drop
1712packets. When syncookies are enabled the packets are still answered and the
1713maximum queue is effectively ignored.
1714
1715tcp_retries1
1716------------
1717
1718Defines how often an answer to a TCP connection request is retransmitted
1719before giving up.
1720
1721tcp_retries2
1722------------
1723
1724Defines how often a TCP packet is retransmitted before giving up.
1725
1726Interface specific settings
1727---------------------------
1728
1729In the directory /proc/sys/net/ipv4/conf you'll find one subdirectory for each
1730interface the system knows about and one directory calls all. Changes in the
1731all subdirectory affect all interfaces, whereas changes in the other
1732subdirectories affect only one interface. All directories have the same
1733entries:
1734
1735accept_redirects
1736----------------
1737
1738This switch decides if the kernel accepts ICMP redirect messages or not. The
1739default is 'yes' if the kernel is configured for a regular host and 'no' for a
1740router configuration.
1741
1742accept_source_route
1743-------------------
1744
1745Should source routed packages be accepted or declined. The default is
1746dependent on the kernel configuration. It's 'yes' for routers and 'no' for
1747hosts.
1748
1749bootp_relay
1750~~~~~~~~~~~
1751
1752Accept packets with source address 0.b.c.d with destinations not to this host
1753as local ones. It is supposed that a BOOTP relay daemon will catch and forward
1754such packets.
1755
1756The default is 0, since this feature is not implemented yet (kernel version
17572.2.12).
1758
1759forwarding
1760----------
1761
1762Enable or disable IP forwarding on this interface.
1763
1764log_martians
1765------------
1766
1767Log packets with source addresses with no known route to kernel log.
1768
1769mc_forwarding
1770-------------
1771
1772Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE and a
1773multicast routing daemon is required.
1774
1775proxy_arp
1776---------
1777
1778Does (1) or does not (0) perform proxy ARP.
1779
1780rp_filter
1781---------
1782
1783Integer value determines if a source validation should be made. 1 means yes, 0
1784means no. Disabled by default, but local/broadcast address spoofing is always
1785on.
1786
1787If you set this to 1 on a router that is the only connection for a network to
1788the net, it will prevent spoofing attacks against your internal networks
1789(external addresses can still be spoofed), without the need for additional
1790firewall rules.
1791
1792secure_redirects
1793----------------
1794
1795Accept ICMP redirect messages only for gateways, listed in default gateway
1796list. Enabled by default.
1797
1798shared_media
1799------------
1800
1801If it is not set the kernel does not assume that different subnets on this
1802device can communicate directly. Default setting is 'yes'.
1803
1804send_redirects
1805--------------
1806
1807Determines whether to send ICMP redirects to other hosts.
1808
1809Routing settings
1810----------------
1811
1812The directory /proc/sys/net/ipv4/route contains several file to control
1813routing issues.
1814
1815error_burst and error_cost
1816--------------------------
1817
1818These parameters are used to limit how many ICMP destination unreachable to
1819send from the host in question. ICMP destination unreachable messages are
84eb8d06 1820sent when we cannot reach the next hop while trying to transmit a packet.
1da177e4
LT
1821It will also print some error messages to kernel logs if someone is ignoring
1822our ICMP redirects. The higher the error_cost factor is, the fewer
1823destination unreachable and error messages will be let through. Error_burst
1824controls when destination unreachable messages and error messages will be
1825dropped. The default settings limit warning messages to five every second.
1826
1827flush
1828-----
1829
1830Writing to this file results in a flush of the routing cache.
1831
1832gc_elasticity, gc_interval, gc_min_interval_ms, gc_timeout, gc_thresh
1833---------------------------------------------------------------------
1834
1835Values to control the frequency and behavior of the garbage collection
1836algorithm for the routing cache. gc_min_interval is deprecated and replaced
1837by gc_min_interval_ms.
1838
1839
1840max_size
1841--------
1842
1843Maximum size of the routing cache. Old entries will be purged once the cache
1844reached has this size.
1845
1846max_delay, min_delay
1847--------------------
1848
1849Delays for flushing the routing cache.
1850
1851redirect_load, redirect_number
1852------------------------------
1853
1854Factors which determine if more ICPM redirects should be sent to a specific
1855host. No redirects will be sent once the load limit or the maximum number of
1856redirects has been reached.
1857
1858redirect_silence
1859----------------
1860
1861Timeout for redirects. After this period redirects will be sent again, even if
1862this has been stopped, because the load or number limit has been reached.
1863
1864Network Neighbor handling
1865-------------------------
1866
1867Settings about how to handle connections with direct neighbors (nodes attached
1868to the same link) can be found in the directory /proc/sys/net/ipv4/neigh.
1869
1870As we saw it in the conf directory, there is a default subdirectory which
1871holds the default values, and one directory for each interface. The contents
1872of the directories are identical, with the single exception that the default
1873settings contain additional options to set garbage collection parameters.
1874
1875In the interface directories you'll find the following entries:
1876
1877base_reachable_time, base_reachable_time_ms
1878-------------------------------------------
1879
1880A base value used for computing the random reachable time value as specified
1881in RFC2461.
1882
1883Expression of base_reachable_time, which is deprecated, is in seconds.
1884Expression of base_reachable_time_ms is in milliseconds.
1885
1886retrans_time, retrans_time_ms
1887-----------------------------
1888
1889The time between retransmitted Neighbor Solicitation messages.
1890Used for address resolution and to determine if a neighbor is
1891unreachable.
1892
1893Expression of retrans_time, which is deprecated, is in 1/100 seconds (for
1894IPv4) or in jiffies (for IPv6).
1895Expression of retrans_time_ms is in milliseconds.
1896
1897unres_qlen
1898----------
1899
1900Maximum queue length for a pending arp request - the number of packets which
1901are accepted from other layers while the ARP address is still resolved.
1902
1903anycast_delay
1904-------------
1905
1906Maximum for random delay of answers to neighbor solicitation messages in
1907jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support
1908yet).
1909
1910ucast_solicit
1911-------------
1912
1913Maximum number of retries for unicast solicitation.
1914
1915mcast_solicit
1916-------------
1917
1918Maximum number of retries for multicast solicitation.
1919
1920delay_first_probe_time
1921----------------------
1922
1923Delay for the first time probe if the neighbor is reachable. (see
1924gc_stale_time)
1925
1926locktime
1927--------
1928
1929An ARP/neighbor entry is only replaced with a new one if the old is at least
1930locktime old. This prevents ARP cache thrashing.
1931
1932proxy_delay
1933-----------
1934
1935Maximum time (real time is random [0..proxytime]) before answering to an ARP
1936request for which we have an proxy ARP entry. In some cases, this is used to
1937prevent network flooding.
1938
1939proxy_qlen
1940----------
1941
1942Maximum queue length of the delayed proxy arp timer. (see proxy_delay).
1943
53cb4726 1944app_solicit
1da177e4
LT
1945----------
1946
1947Determines the number of requests to send to the user level ARP daemon. Use 0
1948to turn off.
1949
1950gc_stale_time
1951-------------
1952
1953Determines how often to check for stale ARP entries. After an ARP entry is
1954stale it will be resolved again (which is useful when an IP address migrates
1955to another machine). When ucast_solicit is greater than 0 it first tries to
1956send an ARP packet directly to the known host When that fails and
1957mcast_solicit is greater than 0, an ARP request is broadcasted.
1958
19592.9 Appletalk
1960-------------
1961
1962The /proc/sys/net/appletalk directory holds the Appletalk configuration data
1963when Appletalk is loaded. The configurable parameters are:
1964
1965aarp-expiry-time
1966----------------
1967
1968The amount of time we keep an ARP entry before expiring it. Used to age out
1969old hosts.
1970
1971aarp-resolve-time
1972-----------------
1973
1974The amount of time we will spend trying to resolve an Appletalk address.
1975
1976aarp-retransmit-limit
1977---------------------
1978
1979The number of times we will retransmit a query before giving up.
1980
1981aarp-tick-time
1982--------------
1983
1984Controls the rate at which expires are checked.
1985
1986The directory /proc/net/appletalk holds the list of active Appletalk sockets
1987on a machine.
1988
1989The fields indicate the DDP type, the local address (in network:node format)
1990the remote address, the size of the transmit pending queue, the size of the
1991received queue (bytes waiting for applications to read) the state and the uid
1992owning the socket.
1993
1994/proc/net/atalk_iface lists all the interfaces configured for appletalk.It
1995shows the name of the interface, its Appletalk address, the network range on
1996that address (or network number for phase 1 networks), and the status of the
1997interface.
1998
1999/proc/net/atalk_route lists each known network route. It lists the target
2000(network) that the route leads to, the router (may be directly connected), the
2001route flags, and the device the route is using.
2002
20032.10 IPX
2004--------
2005
2006The IPX protocol has no tunable values in proc/sys/net.
2007
2008The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
2009socket giving the local and remote addresses in Novell format (that is
2010network:node:port). In accordance with the strange Novell tradition,
2011everything but the port is in hex. Not_Connected is displayed for sockets that
2012are not tied to a specific remote address. The Tx and Rx queue sizes indicate
2013the number of bytes pending for transmission and reception. The state
2014indicates the state the socket is in and the uid is the owning uid of the
2015socket.
2016
2017The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
2018it gives the network number, the node number, and indicates if the network is
2019the primary network. It also indicates which device it is bound to (or
2020Internal for internal networks) and the Frame Type if appropriate. Linux
2021supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
2022IPX.
2023
2024The /proc/net/ipx_route table holds a list of IPX routes. For each route it
2025gives the destination network, the router node (or Directly) and the network
2026address of the router (or Connected) for internal networks.
2027
20282.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
2029----------------------------------------------------------
2030
2031The "mqueue" filesystem provides the necessary kernel features to enable the
2032creation of a user space library that implements the POSIX message queues
2033API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
2034Interfaces specification.)
2035
2036The "mqueue" filesystem contains values for determining/setting the amount of
2037resources used by the file system.
2038
2039/proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
2040maximum number of message queues allowed on the system.
2041
2042/proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
2043maximum number of messages in a queue value. In fact it is the limiting value
2044for another (user) limit which is set in mq_open invocation. This attribute of
2045a queue must be less or equal then msg_max.
2046
2047/proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
2048maximum message size value (it is every message queue's attribute set during
2049its creation).
2050
d7ff0dbf
JFM
20512.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
2052------------------------------------------------------
2053
2054This file can be used to adjust the score used to select which processes
2055should be killed in an out-of-memory situation. Giving it a high score will
2056increase the likelihood of this process being killed by the oom-killer. Valid
2057values are in the range -16 to +15, plus the special value -17, which disables
2058oom-killing altogether for this process.
2059
20602.13 /proc/<pid>/oom_score - Display current oom-killer score
2061-------------------------------------------------------------
2062
2063------------------------------------------------------------------------------
2064This file can be used to check the current score used by the oom-killer is for
2065any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
2066process should be killed in an out-of-memory situation.
1da177e4
LT
2067
2068------------------------------------------------------------------------------
2069Summary
2070------------------------------------------------------------------------------
2071Certain aspects of kernel behavior can be modified at runtime, without the
2072need to recompile the kernel, or even to reboot the system. The files in the
2073/proc/sys tree can not only be read, but also modified. You can use the echo
2074command to write value into these files, thereby changing the default settings
2075of the kernel.
2076------------------------------------------------------------------------------
f9c99463
RK
2077
20782.14 /proc/<pid>/io - Display the IO accounting fields
2079-------------------------------------------------------
2080
2081This file contains IO statistics for each running process
2082
2083Example
2084-------
2085
2086test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
2087[1] 3828
2088
2089test:/tmp # cat /proc/3828/io
2090rchar: 323934931
2091wchar: 323929600
2092syscr: 632687
2093syscw: 632675
2094read_bytes: 0
2095write_bytes: 323932160
2096cancelled_write_bytes: 0
2097
2098
2099Description
2100-----------
2101
2102rchar
2103-----
2104
2105I/O counter: chars read
2106The number of bytes which this task has caused to be read from storage. This
2107is simply the sum of bytes which this process passed to read() and pread().
2108It includes things like tty IO and it is unaffected by whether or not actual
2109physical disk IO was required (the read might have been satisfied from
2110pagecache)
2111
2112
2113wchar
2114-----
2115
2116I/O counter: chars written
2117The number of bytes which this task has caused, or shall cause to be written
2118to disk. Similar caveats apply here as with rchar.
2119
2120
2121syscr
2122-----
2123
2124I/O counter: read syscalls
2125Attempt to count the number of read I/O operations, i.e. syscalls like read()
2126and pread().
2127
2128
2129syscw
2130-----
2131
2132I/O counter: write syscalls
2133Attempt to count the number of write I/O operations, i.e. syscalls like
2134write() and pwrite().
2135
2136
2137read_bytes
2138----------
2139
2140I/O counter: bytes read
2141Attempt to count the number of bytes which this process really did cause to
2142be fetched from the storage layer. Done at the submit_bio() level, so it is
2143accurate for block-backed filesystems. <please add status regarding NFS and
2144CIFS at a later time>
2145
2146
2147write_bytes
2148-----------
2149
2150I/O counter: bytes written
2151Attempt to count the number of bytes which this process caused to be sent to
2152the storage layer. This is done at page-dirtying time.
2153
2154
2155cancelled_write_bytes
2156---------------------
2157
2158The big inaccuracy here is truncate. If a process writes 1MB to a file and
2159then deletes the file, it will in fact perform no writeout. But it will have
2160been accounted as having caused 1MB of write.
2161In other words: The number of bytes which this process caused to not happen,
2162by truncating pagecache. A task can cause "negative" IO too. If this task
2163truncates some dirty pagecache, some IO which another task has been accounted
2164for (in it's write_bytes) will not be happening. We _could_ just subtract that
2165from the truncating task's write_bytes, but there is information loss in doing
2166that.
2167
2168
2169Note
2170----
2171
2172At its current implementation state, this is a bit racy on 32-bit machines: if
2173process A reads process B's /proc/pid/io while process B is updating one of
2174those 64-bit counters, process A could see an intermediate result.
2175
2176
2177More information about this can be found within the taskstats documentation in
2178Documentation/accounting.
2179
2180------------------------------------------------------------------------------