1 ------------------------------------------------------------------------------
2 T H E /proc F I L E S Y S T E M
3 ------------------------------------------------------------------------------
4 /proc/sys Terrehon Bowden <terrehon@pacbell.net> October 7 1999
5 Bodo Bauer <bb@ricochet.net>
7 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8 ------------------------------------------------------------------------------
9 Version 1.3 Kernel version 2.2.12
10 Kernel version 2.4.0-test11-pre4
11 ------------------------------------------------------------------------------
17 0.1 Introduction/Credits
20 1 Collecting System Information
21 1.1 Process-Specific Subdirectories
23 1.3 IDE devices in /proc/ide
24 1.4 Networking info in /proc/net
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
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
41 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
42 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
43 2.13 /proc/<pid>/oom_score - Display current oom-killer score
44 2.14 /proc/<pid>/io - Display the IO accounting fields
45 2.15 /proc/<pid>/coredump_filter - Core dump filtering settings
46 2.16 /proc/<pid>/mountinfo - Information about mounts
48 ------------------------------------------------------------------------------
50 ------------------------------------------------------------------------------
52 0.1 Introduction/Credits
53 ------------------------
55 This documentation is part of a soon (or so we hope) to be released book on
56 the SuSE Linux distribution. As there is no complete documentation for the
57 /proc file system and we've used many freely available sources to write these
58 chapters, it seems only fair to give the work back to the Linux community.
59 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
60 afraid it's still far from complete, but we hope it will be useful. As far as
61 we know, it is the first 'all-in-one' document about the /proc file system. It
62 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
63 SPARC, AXP, etc., features, you probably won't find what you are looking for.
64 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
65 additions and patches are welcome and will be added to this document if you
68 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
69 other people for help compiling this documentation. We'd also like to extend a
70 special thank you to Andi Kleen for documentation, which we relied on heavily
71 to create this document, as well as the additional information he provided.
72 Thanks to everybody else who contributed source or docs to the Linux kernel
73 and helped create a great piece of software... :)
75 If you have any comments, corrections or additions, please don't hesitate to
76 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
79 The latest version of this document is available online at
80 http://skaro.nightcrawler.com/~bb/Docs/Proc as HTML version.
82 If the above direction does not works for you, ypu could try the kernel
83 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
84 comandante@zaralinux.com.
89 We don't guarantee the correctness of this document, and if you come to us
90 complaining about how you screwed up your system because of incorrect
91 documentation, we won't feel responsible...
93 ------------------------------------------------------------------------------
94 CHAPTER 1: COLLECTING SYSTEM INFORMATION
95 ------------------------------------------------------------------------------
97 ------------------------------------------------------------------------------
99 ------------------------------------------------------------------------------
100 * Investigating the properties of the pseudo file system /proc and its
101 ability to provide information on the running Linux system
102 * Examining /proc's structure
103 * Uncovering various information about the kernel and the processes running
105 ------------------------------------------------------------------------------
108 The proc file system acts as an interface to internal data structures in the
109 kernel. It can be used to obtain information about the system and to change
110 certain kernel parameters at runtime (sysctl).
112 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
113 show you how you can use /proc/sys to change settings.
115 1.1 Process-Specific Subdirectories
116 -----------------------------------
118 The directory /proc contains (among other things) one subdirectory for each
119 process running on the system, which is named after the process ID (PID).
121 The link self points to the process reading the file system. Each process
122 subdirectory has the entries listed in Table 1-1.
125 Table 1-1: Process specific entries in /proc
126 ..............................................................................
128 clear_refs Clears page referenced bits shown in smaps output
129 cmdline Command line arguments
130 cpu Current and last cpu in which it was executed (2.4)(smp)
131 cwd Link to the current working directory
132 environ Values of environment variables
133 exe Link to the executable of this process
134 fd Directory, which contains all file descriptors
135 maps Memory maps to executables and library files (2.4)
136 mem Memory held by this process
137 root Link to the root directory of this process
139 statm Process memory status information
140 status Process status in human readable form
141 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
142 smaps Extension based on maps, the rss size for each mapped file
143 ..............................................................................
145 For example, to get the status information of a process, all you have to do is
146 read the file /proc/PID/status:
148 >cat /proc/self/status
164 SigPnd: 0000000000000000
165 SigBlk: 0000000000000000
166 SigIgn: 0000000000000000
167 SigCgt: 0000000000000000
168 CapInh: 00000000fffffeff
169 CapPrm: 0000000000000000
170 CapEff: 0000000000000000
173 This shows you nearly the same information you would get if you viewed it with
174 the ps command. In fact, ps uses the proc file system to obtain its
175 information. The statm file contains more detailed information about the
176 process memory usage. Its seven fields are explained in Table 1-2. The stat
177 file contains details information about the process itself. Its fields are
178 explained in Table 1-3.
181 Table 1-2: Contents of the statm files (as of 2.6.8-rc3)
182 ..............................................................................
184 size total program size (pages) (same as VmSize in status)
185 resident size of memory portions (pages) (same as VmRSS in status)
186 shared number of pages that are shared (i.e. backed by a file)
187 trs number of pages that are 'code' (not including libs; broken,
188 includes data segment)
189 lrs number of pages of library (always 0 on 2.6)
190 drs number of pages of data/stack (including libs; broken,
191 includes library text)
192 dt number of dirty pages (always 0 on 2.6)
193 ..............................................................................
196 Table 1-3: Contents of the stat files (as of 2.6.22-rc3)
197 ..............................................................................
200 tcomm filename of the executable
201 state state (R is running, S is sleeping, D is sleeping in an
202 uninterruptible wait, Z is zombie, T is traced or stopped)
203 ppid process id of the parent process
204 pgrp pgrp of the process
206 tty_nr tty the process uses
207 tty_pgrp pgrp of the tty
209 min_flt number of minor faults
210 cmin_flt number of minor faults with child's
211 maj_flt number of major faults
212 cmaj_flt number of major faults with child's
213 utime user mode jiffies
214 stime kernel mode jiffies
215 cutime user mode jiffies with child's
216 cstime kernel mode jiffies with child's
217 priority priority level
219 num_threads number of threads
220 it_real_value (obsolete, always 0)
221 start_time time the process started after system boot
222 vsize virtual memory size
223 rss resident set memory size
224 rsslim current limit in bytes on the rss
225 start_code address above which program text can run
226 end_code address below which program text can run
227 start_stack address of the start of the stack
228 esp current value of ESP
229 eip current value of EIP
230 pending bitmap of pending signals (obsolete)
231 blocked bitmap of blocked signals (obsolete)
232 sigign bitmap of ignored signals (obsolete)
233 sigcatch bitmap of catched signals (obsolete)
234 wchan address where process went to sleep
237 exit_signal signal to send to parent thread on exit
238 task_cpu which CPU the task is scheduled on
239 rt_priority realtime priority
240 policy scheduling policy (man sched_setscheduler)
241 blkio_ticks time spent waiting for block IO
242 ..............................................................................
248 Similar to the process entries, the kernel data files give information about
249 the running kernel. The files used to obtain this information are contained in
250 /proc and are listed in Table 1-4. Not all of these will be present in your
251 system. It depends on the kernel configuration and the loaded modules, which
252 files are there, and which are missing.
254 Table 1-4: Kernel info in /proc
255 ..............................................................................
257 apm Advanced power management info
258 buddyinfo Kernel memory allocator information (see text) (2.5)
259 bus Directory containing bus specific information
260 cmdline Kernel command line
261 cpuinfo Info about the CPU
262 devices Available devices (block and character)
263 dma Used DMS channels
264 filesystems Supported filesystems
265 driver Various drivers grouped here, currently rtc (2.4)
266 execdomains Execdomains, related to security (2.4)
267 fb Frame Buffer devices (2.4)
268 fs File system parameters, currently nfs/exports (2.4)
269 ide Directory containing info about the IDE subsystem
270 interrupts Interrupt usage
271 iomem Memory map (2.4)
272 ioports I/O port usage
273 irq Masks for irq to cpu affinity (2.4)(smp?)
274 isapnp ISA PnP (Plug&Play) Info (2.4)
275 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
277 ksyms Kernel symbol table
278 loadavg Load average of last 1, 5 & 15 minutes
282 modules List of loaded modules
283 mounts Mounted filesystems
284 net Networking info (see text)
285 partitions Table of partitions known to the system
286 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
287 decoupled by lspci (2.4)
289 scsi SCSI info (see text)
290 slabinfo Slab pool info
291 stat Overall statistics
292 swaps Swap space utilization
294 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
295 tty Info of tty drivers
297 version Kernel version
298 video bttv info of video resources (2.4)
299 ..............................................................................
301 You can, for example, check which interrupts are currently in use and what
302 they are used for by looking in the file /proc/interrupts:
304 > cat /proc/interrupts
306 0: 8728810 XT-PIC timer
307 1: 895 XT-PIC keyboard
309 3: 531695 XT-PIC aha152x
310 4: 2014133 XT-PIC serial
311 5: 44401 XT-PIC pcnet_cs
314 12: 182918 XT-PIC PS/2 Mouse
316 14: 1232265 XT-PIC ide0
320 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
321 output of a SMP machine):
323 > cat /proc/interrupts
326 0: 1243498 1214548 IO-APIC-edge timer
327 1: 8949 8958 IO-APIC-edge keyboard
328 2: 0 0 XT-PIC cascade
329 5: 11286 10161 IO-APIC-edge soundblaster
330 8: 1 0 IO-APIC-edge rtc
331 9: 27422 27407 IO-APIC-edge 3c503
332 12: 113645 113873 IO-APIC-edge PS/2 Mouse
334 14: 22491 24012 IO-APIC-edge ide0
335 15: 2183 2415 IO-APIC-edge ide1
336 17: 30564 30414 IO-APIC-level eth0
337 18: 177 164 IO-APIC-level bttv
342 NMI is incremented in this case because every timer interrupt generates a NMI
343 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
345 LOC is the local interrupt counter of the internal APIC of every CPU.
347 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
348 connects the CPUs in a SMP system. This means that an error has been detected,
349 the IO-APIC automatically retry the transmission, so it should not be a big
350 problem, but you should read the SMP-FAQ.
352 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
353 /proc/interrupts to display every IRQ vector in use by the system, not
354 just those considered 'most important'. The new vectors are:
356 THR -- interrupt raised when a machine check threshold counter
357 (typically counting ECC corrected errors of memory or cache) exceeds
358 a configurable threshold. Only available on some systems.
360 TRM -- a thermal event interrupt occurs when a temperature threshold
361 has been exceeded for the CPU. This interrupt may also be generated
362 when the temperature drops back to normal.
364 SPU -- a spurious interrupt is some interrupt that was raised then lowered
365 by some IO device before it could be fully processed by the APIC. Hence
366 the APIC sees the interrupt but does not know what device it came from.
367 For this case the APIC will generate the interrupt with a IRQ vector
368 of 0xff. This might also be generated by chipset bugs.
370 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
371 sent from one CPU to another per the needs of the OS. Typically,
372 their statistics are used by kernel developers and interested users to
373 determine the occurance of interrupt of the given type.
375 The above IRQ vectors are displayed only when relevent. For example,
376 the threshold vector does not exist on x86_64 platforms. Others are
377 suppressed when the system is a uniprocessor. As of this writing, only
378 i386 and x86_64 platforms support the new IRQ vector displays.
380 Of some interest is the introduction of the /proc/irq directory to 2.4.
381 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
382 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
383 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
388 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
389 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
393 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
394 IRQ, you can set it by doing:
396 > echo 1 > /proc/irq/10/smp_affinity
398 This means that only the first CPU will handle the IRQ, but you can also echo
399 5 which means that only the first and fourth CPU can handle the IRQ.
401 The contents of each smp_affinity file is the same by default:
403 > cat /proc/irq/0/smp_affinity
406 The default_smp_affinity mask applies to all non-active IRQs, which are the
407 IRQs which have not yet been allocated/activated, and hence which lack a
408 /proc/irq/[0-9]* directory.
410 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
411 profiler. Default value is ffffffff (all cpus).
413 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
414 between all the CPUs which are allowed to handle it. As usual the kernel has
415 more info than you and does a better job than you, so the defaults are the
416 best choice for almost everyone.
418 There are three more important subdirectories in /proc: net, scsi, and sys.
419 The general rule is that the contents, or even the existence of these
420 directories, depend on your kernel configuration. If SCSI is not enabled, the
421 directory scsi may not exist. The same is true with the net, which is there
422 only when networking support is present in the running kernel.
424 The slabinfo file gives information about memory usage at the slab level.
425 Linux uses slab pools for memory management above page level in version 2.2.
426 Commonly used objects have their own slab pool (such as network buffers,
427 directory cache, and so on).
429 ..............................................................................
431 > cat /proc/buddyinfo
433 Node 0, zone DMA 0 4 5 4 4 3 ...
434 Node 0, zone Normal 1 0 0 1 101 8 ...
435 Node 0, zone HighMem 2 0 0 1 1 0 ...
437 Memory fragmentation is a problem under some workloads, and buddyinfo is a
438 useful tool for helping diagnose these problems. Buddyinfo will give you a
439 clue as to how big an area you can safely allocate, or why a previous
442 Each column represents the number of pages of a certain order which are
443 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
444 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
445 available in ZONE_NORMAL, etc...
447 ..............................................................................
451 Provides information about distribution and utilization of memory. This
452 varies by architecture and compile options. The following is from a
453 16GB PIII, which has highmem enabled. You may not have all of these fields.
458 MemTotal: 16344972 kB
465 HighTotal: 15597528 kB
466 HighFree: 13629632 kB
476 SReclaimable: 159856 kB
477 SUnreclaim: 124508 kB
482 CommitLimit: 7669796 kB
483 Committed_AS: 100056 kB
484 VmallocTotal: 112216 kB
486 VmallocChunk: 111088 kB
488 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
489 bits and the kernel binary code)
490 MemFree: The sum of LowFree+HighFree
491 Buffers: Relatively temporary storage for raw disk blocks
492 shouldn't get tremendously large (20MB or so)
493 Cached: in-memory cache for files read from the disk (the
494 pagecache). Doesn't include SwapCached
495 SwapCached: Memory that once was swapped out, is swapped back in but
496 still also is in the swapfile (if memory is needed it
497 doesn't need to be swapped out AGAIN because it is already
498 in the swapfile. This saves I/O)
499 Active: Memory that has been used more recently and usually not
500 reclaimed unless absolutely necessary.
501 Inactive: Memory which has been less recently used. It is more
502 eligible to be reclaimed for other purposes
504 HighFree: Highmem is all memory above ~860MB of physical memory
505 Highmem areas are for use by userspace programs, or
506 for the pagecache. The kernel must use tricks to access
507 this memory, making it slower to access than lowmem.
509 LowFree: Lowmem is memory which can be used for everything that
510 highmem can be used for, but it is also available for the
511 kernel's use for its own data structures. Among many
512 other things, it is where everything from the Slab is
513 allocated. Bad things happen when you're out of lowmem.
514 SwapTotal: total amount of swap space available
515 SwapFree: Memory which has been evicted from RAM, and is temporarily
517 Dirty: Memory which is waiting to get written back to the disk
518 Writeback: Memory which is actively being written back to the disk
519 AnonPages: Non-file backed pages mapped into userspace page tables
520 Mapped: files which have been mmaped, such as libraries
521 Slab: in-kernel data structures cache
522 SReclaimable: Part of Slab, that might be reclaimed, such as caches
523 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
524 PageTables: amount of memory dedicated to the lowest level of page
526 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
528 Bounce: Memory used for block device "bounce buffers"
529 WritebackTmp: Memory used by FUSE for temporary writeback buffers
530 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
531 this is the total amount of memory currently available to
532 be allocated on the system. This limit is only adhered to
533 if strict overcommit accounting is enabled (mode 2 in
534 'vm.overcommit_memory').
535 The CommitLimit is calculated with the following formula:
536 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
537 For example, on a system with 1G of physical RAM and 7G
538 of swap with a `vm.overcommit_ratio` of 30 it would
539 yield a CommitLimit of 7.3G.
540 For more details, see the memory overcommit documentation
541 in vm/overcommit-accounting.
542 Committed_AS: The amount of memory presently allocated on the system.
543 The committed memory is a sum of all of the memory which
544 has been allocated by processes, even if it has not been
545 "used" by them as of yet. A process which malloc()'s 1G
546 of memory, but only touches 300M of it will only show up
547 as using 300M of memory even if it has the address space
548 allocated for the entire 1G. This 1G is memory which has
549 been "committed" to by the VM and can be used at any time
550 by the allocating application. With strict overcommit
551 enabled on the system (mode 2 in 'vm.overcommit_memory'),
552 allocations which would exceed the CommitLimit (detailed
553 above) will not be permitted. This is useful if one needs
554 to guarantee that processes will not fail due to lack of
555 memory once that memory has been successfully allocated.
556 VmallocTotal: total size of vmalloc memory area
557 VmallocUsed: amount of vmalloc area which is used
558 VmallocChunk: largest contigious block of vmalloc area which is free
561 1.3 IDE devices in /proc/ide
562 ----------------------------
564 The subdirectory /proc/ide contains information about all IDE devices of which
565 the kernel is aware. There is one subdirectory for each IDE controller, the
566 file drivers and a link for each IDE device, pointing to the device directory
567 in the controller specific subtree.
569 The file drivers contains general information about the drivers used for the
572 > cat /proc/ide/drivers
573 ide-cdrom version 4.53
574 ide-disk version 1.08
576 More detailed information can be found in the controller specific
577 subdirectories. These are named ide0, ide1 and so on. Each of these
578 directories contains the files shown in table 1-5.
581 Table 1-5: IDE controller info in /proc/ide/ide?
582 ..............................................................................
584 channel IDE channel (0 or 1)
585 config Configuration (only for PCI/IDE bridge)
587 model Type/Chipset of IDE controller
588 ..............................................................................
590 Each device connected to a controller has a separate subdirectory in the
591 controllers directory. The files listed in table 1-6 are contained in these
595 Table 1-6: IDE device information
596 ..............................................................................
599 capacity Capacity of the medium (in 512Byte blocks)
600 driver driver and version
601 geometry physical and logical geometry
602 identify device identify block
604 model device identifier
605 settings device setup
606 smart_thresholds IDE disk management thresholds
607 smart_values IDE disk management values
608 ..............................................................................
610 The most interesting file is settings. This file contains a nice overview of
611 the drive parameters:
613 # cat /proc/ide/ide0/hda/settings
614 name value min max mode
615 ---- ----- --- --- ----
616 bios_cyl 526 0 65535 rw
617 bios_head 255 0 255 rw
619 breada_readahead 4 0 127 rw
621 file_readahead 72 0 2097151 rw
623 keepsettings 0 0 1 rw
624 max_kb_per_request 122 1 127 rw
628 pio_mode write-only 0 255 w
634 1.4 Networking info in /proc/net
635 --------------------------------
637 The subdirectory /proc/net follows the usual pattern. Table 1-6 shows the
638 additional values you get for IP version 6 if you configure the kernel to
639 support this. Table 1-7 lists the files and their meaning.
642 Table 1-6: IPv6 info in /proc/net
643 ..............................................................................
645 udp6 UDP sockets (IPv6)
646 tcp6 TCP sockets (IPv6)
647 raw6 Raw device statistics (IPv6)
648 igmp6 IP multicast addresses, which this host joined (IPv6)
649 if_inet6 List of IPv6 interface addresses
650 ipv6_route Kernel routing table for IPv6
651 rt6_stats Global IPv6 routing tables statistics
652 sockstat6 Socket statistics (IPv6)
653 snmp6 Snmp data (IPv6)
654 ..............................................................................
657 Table 1-7: Network info in /proc/net
658 ..............................................................................
661 dev network devices with statistics
662 dev_mcast the Layer2 multicast groups a device is listening too
663 (interface index, label, number of references, number of bound
665 dev_stat network device status
666 ip_fwchains Firewall chain linkage
667 ip_fwnames Firewall chain names
668 ip_masq Directory containing the masquerading tables
669 ip_masquerade Major masquerading table
670 netstat Network statistics
671 raw raw device statistics
672 route Kernel routing table
673 rpc Directory containing rpc info
674 rt_cache Routing cache
676 sockstat Socket statistics
678 tr_rif Token ring RIF routing table
680 unix UNIX domain sockets
681 wireless Wireless interface data (Wavelan etc)
682 igmp IP multicast addresses, which this host joined
683 psched Global packet scheduler parameters.
684 netlink List of PF_NETLINK sockets
685 ip_mr_vifs List of multicast virtual interfaces
686 ip_mr_cache List of multicast routing cache
687 ..............................................................................
689 You can use this information to see which network devices are available in
690 your system and how much traffic was routed over those devices:
694 face |bytes packets errs drop fifo frame compressed multicast|[...
695 lo: 908188 5596 0 0 0 0 0 0 [...
696 ppp0:15475140 20721 410 0 0 410 0 0 [...
697 eth0: 614530 7085 0 0 0 0 0 1 [...
700 ...] bytes packets errs drop fifo colls carrier compressed
701 ...] 908188 5596 0 0 0 0 0 0
702 ...] 1375103 17405 0 0 0 0 0 0
703 ...] 1703981 5535 0 0 0 3 0 0
705 In addition, each Channel Bond interface has it's own directory. For
706 example, the bond0 device will have a directory called /proc/net/bond0/.
707 It will contain information that is specific to that bond, such as the
708 current slaves of the bond, the link status of the slaves, and how
709 many times the slaves link has failed.
714 If you have a SCSI host adapter in your system, you'll find a subdirectory
715 named after the driver for this adapter in /proc/scsi. You'll also see a list
716 of all recognized SCSI devices in /proc/scsi:
720 Host: scsi0 Channel: 00 Id: 00 Lun: 00
721 Vendor: IBM Model: DGHS09U Rev: 03E0
722 Type: Direct-Access ANSI SCSI revision: 03
723 Host: scsi0 Channel: 00 Id: 06 Lun: 00
724 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
725 Type: CD-ROM ANSI SCSI revision: 02
728 The directory named after the driver has one file for each adapter found in
729 the system. These files contain information about the controller, including
730 the used IRQ and the IO address range. The amount of information shown is
731 dependent on the adapter you use. The example shows the output for an Adaptec
732 AHA-2940 SCSI adapter:
734 > cat /proc/scsi/aic7xxx/0
736 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
738 TCQ Enabled By Default : Disabled
739 AIC7XXX_PROC_STATS : Disabled
740 AIC7XXX_RESET_DELAY : 5
741 Adapter Configuration:
742 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
743 Ultra Wide Controller
744 PCI MMAPed I/O Base: 0xeb001000
745 Adapter SEEPROM Config: SEEPROM found and used.
746 Adaptec SCSI BIOS: Enabled
748 SCBs: Active 0, Max Active 2,
749 Allocated 15, HW 16, Page 255
751 BIOS Control Word: 0x18b6
752 Adapter Control Word: 0x005b
753 Extended Translation: Enabled
754 Disconnect Enable Flags: 0xffff
755 Ultra Enable Flags: 0x0001
756 Tag Queue Enable Flags: 0x0000
757 Ordered Queue Tag Flags: 0x0000
758 Default Tag Queue Depth: 8
759 Tagged Queue By Device array for aic7xxx host instance 0:
760 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
761 Actual queue depth per device for aic7xxx host instance 0:
762 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
765 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
766 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
767 Total transfers 160151 (74577 reads and 85574 writes)
769 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
770 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
771 Total transfers 0 (0 reads and 0 writes)
774 1.6 Parallel port info in /proc/parport
775 ---------------------------------------
777 The directory /proc/parport contains information about the parallel ports of
778 your system. It has one subdirectory for each port, named after the port
781 These directories contain the four files shown in Table 1-8.
784 Table 1-8: Files in /proc/parport
785 ..............................................................................
787 autoprobe Any IEEE-1284 device ID information that has been acquired.
788 devices list of the device drivers using that port. A + will appear by the
789 name of the device currently using the port (it might not appear
791 hardware Parallel port's base address, IRQ line and DMA channel.
792 irq IRQ that parport is using for that port. This is in a separate
793 file to allow you to alter it by writing a new value in (IRQ
795 ..............................................................................
797 1.7 TTY info in /proc/tty
798 -------------------------
800 Information about the available and actually used tty's can be found in the
801 directory /proc/tty.You'll find entries for drivers and line disciplines in
802 this directory, as shown in Table 1-9.
805 Table 1-9: Files in /proc/tty
806 ..............................................................................
808 drivers list of drivers and their usage
809 ldiscs registered line disciplines
810 driver/serial usage statistic and status of single tty lines
811 ..............................................................................
813 To see which tty's are currently in use, you can simply look into the file
816 > cat /proc/tty/drivers
817 pty_slave /dev/pts 136 0-255 pty:slave
818 pty_master /dev/ptm 128 0-255 pty:master
819 pty_slave /dev/ttyp 3 0-255 pty:slave
820 pty_master /dev/pty 2 0-255 pty:master
821 serial /dev/cua 5 64-67 serial:callout
822 serial /dev/ttyS 4 64-67 serial
823 /dev/tty0 /dev/tty0 4 0 system:vtmaster
824 /dev/ptmx /dev/ptmx 5 2 system
825 /dev/console /dev/console 5 1 system:console
826 /dev/tty /dev/tty 5 0 system:/dev/tty
827 unknown /dev/tty 4 1-63 console
830 1.8 Miscellaneous kernel statistics in /proc/stat
831 -------------------------------------------------
833 Various pieces of information about kernel activity are available in the
834 /proc/stat file. All of the numbers reported in this file are aggregates
835 since the system first booted. For a quick look, simply cat the file:
838 cpu 2255 34 2290 22625563 6290 127 456 0
839 cpu0 1132 34 1441 11311718 3675 127 438 0
840 cpu1 1123 0 849 11313845 2614 0 18 0
841 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
848 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
849 lines. These numbers identify the amount of time the CPU has spent performing
850 different kinds of work. Time units are in USER_HZ (typically hundredths of a
851 second). The meanings of the columns are as follows, from left to right:
853 - user: normal processes executing in user mode
854 - nice: niced processes executing in user mode
855 - system: processes executing in kernel mode
856 - idle: twiddling thumbs
857 - iowait: waiting for I/O to complete
858 - irq: servicing interrupts
859 - softirq: servicing softirqs
860 - steal: involuntary wait
862 The "intr" line gives counts of interrupts serviced since boot time, for each
863 of the possible system interrupts. The first column is the total of all
864 interrupts serviced; each subsequent column is the total for that particular
867 The "ctxt" line gives the total number of context switches across all CPUs.
869 The "btime" line gives the time at which the system booted, in seconds since
872 The "processes" line gives the number of processes and threads created, which
873 includes (but is not limited to) those created by calls to the fork() and
874 clone() system calls.
876 The "procs_running" line gives the number of processes currently running on
879 The "procs_blocked" line gives the number of processes currently blocked,
880 waiting for I/O to complete.
882 1.9 Ext4 file system parameters
883 ------------------------------
884 Ext4 file system have one directory per partition under /proc/fs/ext4/
885 # ls /proc/fs/ext4/hdc/
886 group_prealloc max_to_scan mb_groups mb_history min_to_scan order2_req
890 This file gives the details of mutiblock allocator buddy cache of free blocks
893 Multiblock allocation history.
896 This file indicate whether the multiblock allocator should start collecting
897 statistics. The statistics are shown during unmount
900 The multiblock allocator normalize the block allocation request to
901 group_prealloc filesystem blocks if we don't have strip value set.
902 The stripe value can be specified at mount time or during mke2fs.
905 How long multiblock allocator can look for a best extent (in found extents)
908 How long multiblock allocator must look for a best extent
911 Multiblock allocator use 2^N search using buddies only for requests greater
912 than or equal to order2_req. The request size is specfied in file system
913 blocks. A value of 2 indicate only if the requests are greater than or equal
917 Files smaller than stream_req are served by the stream allocator, whose
918 purpose is to pack requests as close each to other as possible to
919 produce smooth I/O traffic. Avalue of 16 indicate that file smaller than 16
920 filesystem block size will use group based preallocation.
922 ------------------------------------------------------------------------------
924 ------------------------------------------------------------------------------
925 The /proc file system serves information about the running system. It not only
926 allows access to process data but also allows you to request the kernel status
927 by reading files in the hierarchy.
929 The directory structure of /proc reflects the types of information and makes
930 it easy, if not obvious, where to look for specific data.
931 ------------------------------------------------------------------------------
933 ------------------------------------------------------------------------------
934 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
935 ------------------------------------------------------------------------------
937 ------------------------------------------------------------------------------
939 ------------------------------------------------------------------------------
940 * Modifying kernel parameters by writing into files found in /proc/sys
941 * Exploring the files which modify certain parameters
942 * Review of the /proc/sys file tree
943 ------------------------------------------------------------------------------
946 A very interesting part of /proc is the directory /proc/sys. This is not only
947 a source of information, it also allows you to change parameters within the
948 kernel. Be very careful when attempting this. You can optimize your system,
949 but you can also cause it to crash. Never alter kernel parameters on a
950 production system. Set up a development machine and test to make sure that
951 everything works the way you want it to. You may have no alternative but to
952 reboot the machine once an error has been made.
954 To change a value, simply echo the new value into the file. An example is
955 given below in the section on the file system data. You need to be root to do
956 this. You can create your own boot script to perform this every time your
959 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
960 general things in the operation of the Linux kernel. Since some of the files
961 can inadvertently disrupt your system, it is advisable to read both
962 documentation and source before actually making adjustments. In any case, be
963 very careful when writing to any of these files. The entries in /proc may
964 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
965 review the kernel documentation in the directory /usr/src/linux/Documentation.
966 This chapter is heavily based on the documentation included in the pre 2.2
967 kernels, and became part of it in version 2.2.1 of the Linux kernel.
969 2.1 /proc/sys/fs - File system data
970 -----------------------------------
972 This subdirectory contains specific file system, file handle, inode, dentry
973 and quota information.
975 Currently, these files are in /proc/sys/fs:
980 Status of the directory cache. Since directory entries are dynamically
981 allocated and deallocated, this file indicates the current status. It holds
982 six values, in which the last two are not used and are always zero. The others
983 are listed in table 2-1.
986 Table 2-1: Status files of the directory cache
987 ..............................................................................
989 nr_dentry Almost always zero
990 nr_unused Number of unused cache entries
992 in seconds after the entry may be reclaimed, when memory is short
993 want_pages internally
994 ..............................................................................
996 dquot-nr and dquot-max
997 ----------------------
999 The file dquot-max shows the maximum number of cached disk quota entries.
1001 The file dquot-nr shows the number of allocated disk quota entries and the
1002 number of free disk quota entries.
1004 If the number of available cached disk quotas is very low and you have a large
1005 number of simultaneous system users, you might want to raise the limit.
1007 file-nr and file-max
1008 --------------------
1010 The kernel allocates file handles dynamically, but doesn't free them again at
1013 The value in file-max denotes the maximum number of file handles that the
1014 Linux kernel will allocate. When you get a lot of error messages about running
1015 out of file handles, you might want to raise this limit. The default value is
1016 10% of RAM in kilobytes. To change it, just write the new number into the
1019 # cat /proc/sys/fs/file-max
1021 # echo 8192 > /proc/sys/fs/file-max
1022 # cat /proc/sys/fs/file-max
1026 This method of revision is useful for all customizable parameters of the
1027 kernel - simply echo the new value to the corresponding file.
1029 Historically, the three values in file-nr denoted the number of allocated file
1030 handles, the number of allocated but unused file handles, and the maximum
1031 number of file handles. Linux 2.6 always reports 0 as the number of free file
1032 handles -- this is not an error, it just means that the number of allocated
1033 file handles exactly matches the number of used file handles.
1035 Attempts to allocate more file descriptors than file-max are reported with
1036 printk, look for "VFS: file-max limit <number> reached".
1038 inode-state and inode-nr
1039 ------------------------
1041 The file inode-nr contains the first two items from inode-state, so we'll skip
1044 inode-state contains two actual numbers and five dummy values. The numbers
1045 are nr_inodes and nr_free_inodes (in order of appearance).
1050 Denotes the number of inodes the system has allocated. This number will
1051 grow and shrink dynamically.
1056 Denotes the maximum number of file-handles a process can
1057 allocate. Default value is 1024*1024 (1048576) which should be
1058 enough for most machines. Actual limit depends on RLIMIT_NOFILE
1064 Represents the number of free inodes. Ie. The number of inuse inodes is
1065 (nr_inodes - nr_free_inodes).
1067 aio-nr and aio-max-nr
1068 ---------------------
1070 aio-nr is the running total of the number of events specified on the
1071 io_setup system call for all currently active aio contexts. If aio-nr
1072 reaches aio-max-nr then io_setup will fail with EAGAIN. Note that
1073 raising aio-max-nr does not result in the pre-allocation or re-sizing
1074 of any kernel data structures.
1076 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
1077 -----------------------------------------------------------
1079 Besides these files, there is the subdirectory /proc/sys/fs/binfmt_misc. This
1080 handles the kernel support for miscellaneous binary formats.
1082 Binfmt_misc provides the ability to register additional binary formats to the
1083 Kernel without compiling an additional module/kernel. Therefore, binfmt_misc
1084 needs to know magic numbers at the beginning or the filename extension of the
1087 It works by maintaining a linked list of structs that contain a description of
1088 a binary format, including a magic with size (or the filename extension),
1089 offset and mask, and the interpreter name. On request it invokes the given
1090 interpreter with the original program as argument, as binfmt_java and
1091 binfmt_em86 and binfmt_mz do. Since binfmt_misc does not define any default
1092 binary-formats, you have to register an additional binary-format.
1094 There are two general files in binfmt_misc and one file per registered format.
1095 The two general files are register and status.
1097 Registering a new binary format
1098 -------------------------------
1100 To register a new binary format you have to issue the command
1102 echo :name:type:offset:magic:mask:interpreter: > /proc/sys/fs/binfmt_misc/register
1106 with appropriate name (the name for the /proc-dir entry), offset (defaults to
1107 0, if omitted), magic, mask (which can be omitted, defaults to all 0xff) and
1108 last but not least, the interpreter that is to be invoked (for example and
1109 testing /bin/echo). Type can be M for usual magic matching or E for filename
1110 extension matching (give extension in place of magic).
1112 Check or reset the status of the binary format handler
1113 ------------------------------------------------------
1115 If you do a cat on the file /proc/sys/fs/binfmt_misc/status, you will get the
1116 current status (enabled/disabled) of binfmt_misc. Change the status by echoing
1117 0 (disables) or 1 (enables) or -1 (caution: this clears all previously
1118 registered binary formats) to status. For example echo 0 > status to disable
1119 binfmt_misc (temporarily).
1121 Status of a single handler
1122 --------------------------
1124 Each registered handler has an entry in /proc/sys/fs/binfmt_misc. These files
1125 perform the same function as status, but their scope is limited to the actual
1126 binary format. By cating this file, you also receive all related information
1127 about the interpreter/magic of the binfmt.
1129 Example usage of binfmt_misc (emulate binfmt_java)
1130 --------------------------------------------------
1132 cd /proc/sys/fs/binfmt_misc
1133 echo ':Java:M::\xca\xfe\xba\xbe::/usr/local/java/bin/javawrapper:' > register
1134 echo ':HTML:E::html::/usr/local/java/bin/appletviewer:' > register
1135 echo ':Applet:M::<!--applet::/usr/local/java/bin/appletviewer:' > register
1136 echo ':DEXE:M::\x0eDEX::/usr/bin/dosexec:' > register
1139 These four lines add support for Java executables and Java applets (like
1140 binfmt_java, additionally recognizing the .html extension with no need to put
1141 <!--applet> to every applet file). You have to install the JDK and the
1142 shell-script /usr/local/java/bin/javawrapper too. It works around the
1143 brokenness of the Java filename handling. To add a Java binary, just create a
1144 link to the class-file somewhere in the path.
1146 2.3 /proc/sys/kernel - general kernel parameters
1147 ------------------------------------------------
1149 This directory reflects general kernel behaviors. As I've said before, the
1150 contents depend on your configuration. Here you'll find the most important
1151 files, along with descriptions of what they mean and how to use them.
1156 The file contains three values; highwater, lowwater, and frequency.
1158 It exists only when BSD-style process accounting is enabled. These values
1159 control its behavior. If the free space on the file system where the log lives
1160 goes below lowwater percentage, accounting suspends. If it goes above
1161 highwater percentage, accounting resumes. Frequency determines how often you
1162 check the amount of free space (value is in seconds). Default settings are: 4,
1163 2, and 30. That is, suspend accounting if there is less than 2 percent free;
1164 resume it if we have a value of 3 or more percent; consider information about
1165 the amount of free space valid for 30 seconds
1170 When the value in this file is 0, ctrl-alt-del is trapped and sent to the init
1171 program to handle a graceful restart. However, when the value is greater that
1172 zero, Linux's reaction to this key combination will be an immediate reboot,
1173 without syncing its dirty buffers.
1176 When a program (like dosemu) has the keyboard in raw mode, the
1177 ctrl-alt-del is intercepted by the program before it ever reaches the
1178 kernel tty layer, and it is up to the program to decide what to do with
1181 domainname and hostname
1182 -----------------------
1184 These files can be controlled to set the NIS domainname and hostname of your
1185 box. For the classic darkstar.frop.org a simple:
1187 # echo "darkstar" > /proc/sys/kernel/hostname
1188 # echo "frop.org" > /proc/sys/kernel/domainname
1191 would suffice to set your hostname and NIS domainname.
1193 osrelease, ostype and version
1194 -----------------------------
1196 The names make it pretty obvious what these fields contain:
1198 > cat /proc/sys/kernel/osrelease
1201 > cat /proc/sys/kernel/ostype
1204 > cat /proc/sys/kernel/version
1205 #4 Fri Oct 1 12:41:14 PDT 1999
1208 The files osrelease and ostype should be clear enough. Version needs a little
1209 more clarification. The #4 means that this is the 4th kernel built from this
1210 source base and the date after it indicates the time the kernel was built. The
1211 only way to tune these values is to rebuild the kernel.
1216 The value in this file represents the number of seconds the kernel waits
1217 before rebooting on a panic. When you use the software watchdog, the
1218 recommended setting is 60. If set to 0, the auto reboot after a kernel panic
1219 is disabled, which is the default setting.
1224 The four values in printk denote
1226 * default_message_loglevel,
1227 * minimum_console_loglevel and
1228 * default_console_loglevel
1231 These values influence printk() behavior when printing or logging error
1232 messages, which come from inside the kernel. See syslog(2) for more
1233 information on the different log levels.
1238 Messages with a higher priority than this will be printed to the console.
1240 default_message_level
1241 ---------------------
1243 Messages without an explicit priority will be printed with this priority.
1245 minimum_console_loglevel
1246 ------------------------
1248 Minimum (highest) value to which the console_loglevel can be set.
1250 default_console_loglevel
1251 ------------------------
1253 Default value for console_loglevel.
1258 This file shows the size of the generic SCSI (sg) buffer. At this point, you
1259 can't tune it yet, but you can change it at compile time by editing
1260 include/scsi/sg.h and changing the value of SG_BIG_BUFF.
1262 If you use a scanner with SANE (Scanner Access Now Easy) you might want to set
1263 this to a higher value. Refer to the SANE documentation on this issue.
1268 The location where the modprobe binary is located. The kernel uses this
1269 program to load modules on demand.
1274 The value in this file affects behavior of handling NMI. When the value is
1275 non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
1276 debugging information is displayed on console.
1278 NMI switch that most IA32 servers have fires unknown NMI up, for example.
1279 If a system hangs up, try pressing the NMI switch.
1284 Enables/Disables the NMI watchdog on x86 systems. When the value is non-zero
1285 the NMI watchdog is enabled and will continuously test all online cpus to
1286 determine whether or not they are still functioning properly.
1288 Because the NMI watchdog shares registers with oprofile, by disabling the NMI
1289 watchdog, oprofile may have more registers to utilize.
1294 Enables/Disables the protection of the per-process proc entries "maps" and
1295 "smaps". When enabled, the contents of these files are visible only to
1296 readers that are allowed to ptrace() the given process.
1299 2.4 /proc/sys/vm - The virtual memory subsystem
1300 -----------------------------------------------
1302 The files in this directory can be used to tune the operation of the virtual
1303 memory (VM) subsystem of the Linux kernel.
1308 Controls the tendency of the kernel to reclaim the memory which is used for
1309 caching of directory and inode objects.
1311 At the default value of vfs_cache_pressure=100 the kernel will attempt to
1312 reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1313 swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1314 to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100
1315 causes the kernel to prefer to reclaim dentries and inodes.
1317 dirty_background_ratio
1318 ----------------------
1320 Contains, as a percentage of total system memory, the number of pages at which
1321 the pdflush background writeback daemon will start writing out dirty data.
1326 Contains, as a percentage of total system memory, the number of pages at which
1327 a process which is generating disk writes will itself start writing out dirty
1330 dirty_writeback_centisecs
1331 -------------------------
1333 The pdflush writeback daemons will periodically wake up and write `old' data
1334 out to disk. This tunable expresses the interval between those wakeups, in
1335 100'ths of a second.
1337 Setting this to zero disables periodic writeback altogether.
1339 dirty_expire_centisecs
1340 ----------------------
1342 This tunable is used to define when dirty data is old enough to be eligible
1343 for writeout by the pdflush daemons. It is expressed in 100'ths of a second.
1344 Data which has been dirty in-memory for longer than this interval will be
1345 written out next time a pdflush daemon wakes up.
1347 highmem_is_dirtyable
1348 --------------------
1350 Only present if CONFIG_HIGHMEM is set.
1352 This defaults to 0 (false), meaning that the ratios set above are calculated
1353 as a percentage of lowmem only. This protects against excessive scanning
1354 in page reclaim, swapping and general VM distress.
1356 Setting this to 1 can be useful on 32 bit machines where you want to make
1357 random changes within an MMAPed file that is larger than your available
1358 lowmem without causing large quantities of random IO. Is is safe if the
1359 behavior of all programs running on the machine is known and memory will
1360 not be otherwise stressed.
1365 If non-zero, this sysctl disables the new 32-bit mmap mmap layout - the kernel
1366 will use the legacy (2.4) layout for all processes.
1368 lowmem_reserve_ratio
1369 ---------------------
1371 For some specialised workloads on highmem machines it is dangerous for
1372 the kernel to allow process memory to be allocated from the "lowmem"
1373 zone. This is because that memory could then be pinned via the mlock()
1374 system call, or by unavailability of swapspace.
1376 And on large highmem machines this lack of reclaimable lowmem memory
1379 So the Linux page allocator has a mechanism which prevents allocations
1380 which _could_ use highmem from using too much lowmem. This means that
1381 a certain amount of lowmem is defended from the possibility of being
1382 captured into pinned user memory.
1384 (The same argument applies to the old 16 megabyte ISA DMA region. This
1385 mechanism will also defend that region from allocations which could use
1388 The `lowmem_reserve_ratio' tunable determines how aggressive the kernel is
1389 in defending these lower zones.
1391 If you have a machine which uses highmem or ISA DMA and your
1392 applications are using mlock(), or if you are running with no swap then
1393 you probably should change the lowmem_reserve_ratio setting.
1395 The lowmem_reserve_ratio is an array. You can see them by reading this file.
1397 % cat /proc/sys/vm/lowmem_reserve_ratio
1400 Note: # of this elements is one fewer than number of zones. Because the highest
1401 zone's value is not necessary for following calculation.
1403 But, these values are not used directly. The kernel calculates # of protection
1404 pages for each zones from them. These are shown as array of protection pages
1405 in /proc/zoneinfo like followings. (This is an example of x86-64 box).
1406 Each zone has an array of protection pages like this.
1417 protection: (0, 2004, 2004, 2004)
1418 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1423 These protections are added to score to judge whether this zone should be used
1424 for page allocation or should be reclaimed.
1426 In this example, if normal pages (index=2) are required to this DMA zone and
1427 pages_high is used for watermark, the kernel judges this zone should not be
1428 used because pages_free(1355) is smaller than watermark + protection[2]
1429 (4 + 2004 = 2008). If this protection value is 0, this zone would be used for
1430 normal page requirement. If requirement is DMA zone(index=0), protection[0]
1433 zone[i]'s protection[j] is calculated by following exprssion.
1436 zone[i]->protection[j]
1437 = (total sums of present_pages from zone[i+1] to zone[j] on the node)
1438 / lowmem_reserve_ratio[i];
1440 (should not be protected. = 0;
1442 (not necessary, but looks 0)
1444 The default values of lowmem_reserve_ratio[i] are
1445 256 (if zone[i] means DMA or DMA32 zone)
1447 As above expression, they are reciprocal number of ratio.
1448 256 means 1/256. # of protection pages becomes about "0.39%" of total present
1449 pages of higher zones on the node.
1451 If you would like to protect more pages, smaller values are effective.
1452 The minimum value is 1 (1/1 -> 100%).
1457 page-cluster controls the number of pages which are written to swap in
1458 a single attempt. The swap I/O size.
1460 It is a logarithmic value - setting it to zero means "1 page", setting
1461 it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
1463 The default value is three (eight pages at a time). There may be some
1464 small benefits in tuning this to a different value if your workload is
1470 Controls overcommit of system memory, possibly allowing processes
1471 to allocate (but not use) more memory than is actually available.
1474 0 - Heuristic overcommit handling. Obvious overcommits of
1475 address space are refused. Used for a typical system. It
1476 ensures a seriously wild allocation fails while allowing
1477 overcommit to reduce swap usage. root is allowed to
1478 allocate slightly more memory in this mode. This is the
1481 1 - Always overcommit. Appropriate for some scientific
1484 2 - Don't overcommit. The total address space commit
1485 for the system is not permitted to exceed swap plus a
1486 configurable percentage (default is 50) of physical RAM.
1487 Depending on the percentage you use, in most situations
1488 this means a process will not be killed while attempting
1489 to use already-allocated memory but will receive errors
1490 on memory allocation as appropriate.
1495 Percentage of physical memory size to include in overcommit calculations
1498 Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100)
1500 swapspace = total size of all swap areas
1501 physmem = size of physical memory in system
1503 nr_hugepages and hugetlb_shm_group
1504 ----------------------------------
1506 nr_hugepages configures number of hugetlb page reserved for the system.
1508 hugetlb_shm_group contains group id that is allowed to create SysV shared
1509 memory segment using hugetlb page.
1511 hugepages_treat_as_movable
1512 --------------------------
1514 This parameter is only useful when kernelcore= is specified at boot time to
1515 create ZONE_MOVABLE for pages that may be reclaimed or migrated. Huge pages
1516 are not movable so are not normally allocated from ZONE_MOVABLE. A non-zero
1517 value written to hugepages_treat_as_movable allows huge pages to be allocated
1520 Once enabled, the ZONE_MOVABLE is treated as an area of memory the huge
1521 pages pool can easily grow or shrink within. Assuming that applications are
1522 not running that mlock() a lot of memory, it is likely the huge pages pool
1523 can grow to the size of ZONE_MOVABLE by repeatedly entering the desired value
1524 into nr_hugepages and triggering page reclaim.
1529 laptop_mode is a knob that controls "laptop mode". All the things that are
1530 controlled by this knob are discussed in Documentation/laptops/laptop-mode.txt.
1535 block_dump enables block I/O debugging when set to a nonzero value. More
1536 information on block I/O debugging is in Documentation/laptops/laptop-mode.txt.
1541 This file contains valid hold time of swap out protection token. The Linux
1542 VM has token based thrashing control mechanism and uses the token to prevent
1543 unnecessary page faults in thrashing situation. The unit of the value is
1544 second. The value would be useful to tune thrashing behavior.
1549 Writing to this will cause the kernel to drop clean caches, dentries and
1550 inodes from memory, causing that memory to become free.
1553 echo 1 > /proc/sys/vm/drop_caches
1554 To free dentries and inodes:
1555 echo 2 > /proc/sys/vm/drop_caches
1556 To free pagecache, dentries and inodes:
1557 echo 3 > /proc/sys/vm/drop_caches
1559 As this is a non-destructive operation and dirty objects are not freeable, the
1560 user should run `sync' first.
1563 2.5 /proc/sys/dev - Device specific parameters
1564 ----------------------------------------------
1566 Currently there is only support for CDROM drives, and for those, there is only
1567 one read-only file containing information about the CD-ROM drives attached to
1570 >cat /proc/sys/dev/cdrom/info
1571 CD-ROM information, Id: cdrom.c 2.55 1999/04/25
1575 drive # of slots: 1 0
1579 Can change speed: 1 1
1580 Can select disk: 0 1
1581 Can read multisession: 1 1
1583 Reports media changed: 1 1
1587 You see two drives, sr0 and hdb, along with a list of their features.
1589 2.6 /proc/sys/sunrpc - Remote procedure calls
1590 ---------------------------------------------
1592 This directory contains four files, which enable or disable debugging for the
1593 RPC functions NFS, NFS-daemon, RPC and NLM. The default values are 0. They can
1594 be set to one to turn debugging on. (The default value is 0 for each)
1596 2.7 /proc/sys/net - Networking stuff
1597 ------------------------------------
1599 The interface to the networking parts of the kernel is located in
1600 /proc/sys/net. Table 2-3 shows all possible subdirectories. You may see only
1601 some of them, depending on your kernel's configuration.
1604 Table 2-3: Subdirectories in /proc/sys/net
1605 ..............................................................................
1606 Directory Content Directory Content
1607 core General parameter appletalk Appletalk protocol
1608 unix Unix domain sockets netrom NET/ROM
1609 802 E802 protocol ax25 AX25
1610 ethernet Ethernet protocol rose X.25 PLP layer
1611 ipv4 IP version 4 x25 X.25 protocol
1612 ipx IPX token-ring IBM token ring
1613 bridge Bridging decnet DEC net
1615 ..............................................................................
1617 We will concentrate on IP networking here. Since AX15, X.25, and DEC Net are
1618 only minor players in the Linux world, we'll skip them in this chapter. You'll
1619 find some short info on Appletalk and IPX further on in this chapter. Review
1620 the online documentation and the kernel source to get a detailed view of the
1621 parameters for those protocols. In this section we'll discuss the
1622 subdirectories printed in bold letters in the table above. As default values
1623 are suitable for most needs, there is no need to change these values.
1625 /proc/sys/net/core - Network core options
1626 -----------------------------------------
1631 The default setting of the socket receive buffer in bytes.
1636 The maximum receive socket buffer size in bytes.
1641 The default setting (in bytes) of the socket send buffer.
1646 The maximum send socket buffer size in bytes.
1648 message_burst and message_cost
1649 ------------------------------
1651 These parameters are used to limit the warning messages written to the kernel
1652 log from the networking code. They enforce a rate limit to make a
1653 denial-of-service attack impossible. A higher message_cost factor, results in
1654 fewer messages that will be written. Message_burst controls when messages will
1655 be dropped. The default settings limit warning messages to one every five
1661 This controls console messages from the networking stack that can occur because
1662 of problems on the network like duplicate address or bad checksums. Normally,
1663 this should be enabled, but if the problem persists the messages can be
1670 Maximum number of packets, queued on the INPUT side, when the interface
1671 receives packets faster than kernel can process them.
1676 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
1677 of struct cmsghdr structures with appended data.
1679 /proc/sys/net/unix - Parameters for Unix domain sockets
1680 -------------------------------------------------------
1682 There are only two files in this subdirectory. They control the delays for
1683 deleting and destroying socket descriptors.
1685 2.8 /proc/sys/net/ipv4 - IPV4 settings
1686 --------------------------------------
1688 IP version 4 is still the most used protocol in Unix networking. It will be
1689 replaced by IP version 6 in the next couple of years, but for the moment it's
1690 the de facto standard for the internet and is used in most networking
1691 environments around the world. Because of the importance of this protocol,
1692 we'll have a deeper look into the subtree controlling the behavior of the IPv4
1693 subsystem of the Linux kernel.
1695 Let's start with the entries in /proc/sys/net/ipv4.
1700 icmp_echo_ignore_all and icmp_echo_ignore_broadcasts
1701 ----------------------------------------------------
1703 Turn on (1) or off (0), if the kernel should ignore all ICMP ECHO requests, or
1704 just those to broadcast and multicast addresses.
1706 Please note that if you accept ICMP echo requests with a broadcast/multi\-cast
1707 destination address your network may be used as an exploder for denial of
1708 service packet flooding attacks to other hosts.
1710 icmp_destunreach_rate, icmp_echoreply_rate, icmp_paramprob_rate and icmp_timeexeed_rate
1711 ---------------------------------------------------------------------------------------
1713 Sets limits for sending ICMP packets to specific targets. A value of zero
1714 disables all limiting. Any positive value sets the maximum package rate in
1715 hundredth of a second (on Intel systems).
1723 This file contains the number one if the host received its IP configuration by
1724 RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero.
1729 TTL (Time To Live) for IPv4 interfaces. This is simply the maximum number of
1730 hops a packet may travel.
1735 Enable dynamic socket address rewriting on interface address change. This is
1736 useful for dialup interface with changing IP addresses.
1741 Enable or disable forwarding of IP packages between interfaces. Changing this
1742 value resets all other parameters to their default values. They differ if the
1743 kernel is configured as host or router.
1748 Range of ports used by TCP and UDP to choose the local port. Contains two
1749 numbers, the first number is the lowest port, the second number the highest
1750 local port. Default is 1024-4999. Should be changed to 32768-61000 for
1756 Global switch to turn path MTU discovery off. It can also be set on a per
1757 socket basis by the applications or on a per route basis.
1762 Enable/disable debugging of IP masquerading.
1764 IP fragmentation settings
1765 -------------------------
1767 ipfrag_high_trash and ipfrag_low_trash
1768 --------------------------------------
1770 Maximum memory used to reassemble IP fragments. When ipfrag_high_thresh bytes
1771 of memory is allocated for this purpose, the fragment handler will toss
1772 packets until ipfrag_low_thresh is reached.
1777 Time in seconds to keep an IP fragment in memory.
1785 This file controls the use of the ECN bit in the IPv4 headers. This is a new
1786 feature about Explicit Congestion Notification, but some routers and firewalls
1787 block traffic that has this bit set, so it could be necessary to echo 0 to
1788 /proc/sys/net/ipv4/tcp_ecn if you want to talk to these sites. For more info
1789 you could read RFC2481.
1791 tcp_retrans_collapse
1792 --------------------
1794 Bug-to-bug compatibility with some broken printers. On retransmit, try to send
1795 larger packets to work around bugs in certain TCP stacks. Can be turned off by
1798 tcp_keepalive_probes
1799 --------------------
1801 Number of keep alive probes TCP sends out, until it decides that the
1802 connection is broken.
1807 How often TCP sends out keep alive messages, when keep alive is enabled. The
1813 Number of times initial SYNs for a TCP connection attempt will be
1814 retransmitted. Should not be higher than 255. This is only the timeout for
1815 outgoing connections, for incoming connections the number of retransmits is
1816 defined by tcp_retries1.
1821 Enable select acknowledgments after RFC2018.
1826 Enable timestamps as defined in RFC1323.
1831 Enable the strict RFC793 interpretation of the TCP urgent pointer field. The
1832 default is to use the BSD compatible interpretation of the urgent pointer
1833 pointing to the first byte after the urgent data. The RFC793 interpretation is
1834 to have it point to the last byte of urgent data. Enabling this option may
1835 lead to interoperability problems. Disabled by default.
1840 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES. Send out
1841 syncookies when the syn backlog queue of a socket overflows. This is to ward
1842 off the common 'syn flood attack'. Disabled by default.
1844 Note that the concept of a socket backlog is abandoned. This means the peer
1845 may not receive reliable error messages from an over loaded server with
1851 Enable window scaling as defined in RFC1323.
1856 The length of time in seconds it takes to receive a final FIN before the
1857 socket is always closed. This is strictly a violation of the TCP
1858 specification, but required to prevent denial-of-service attacks.
1863 Indicates how many keep alive probes are sent per slow timer run. Should not
1864 be set too high to prevent bursts.
1869 Length of the per socket backlog queue. Since Linux 2.2 the backlog specified
1870 in listen(2) only specifies the length of the backlog queue of already
1871 established sockets. When more connection requests arrive Linux starts to drop
1872 packets. When syncookies are enabled the packets are still answered and the
1873 maximum queue is effectively ignored.
1878 Defines how often an answer to a TCP connection request is retransmitted
1884 Defines how often a TCP packet is retransmitted before giving up.
1886 Interface specific settings
1887 ---------------------------
1889 In the directory /proc/sys/net/ipv4/conf you'll find one subdirectory for each
1890 interface the system knows about and one directory calls all. Changes in the
1891 all subdirectory affect all interfaces, whereas changes in the other
1892 subdirectories affect only one interface. All directories have the same
1898 This switch decides if the kernel accepts ICMP redirect messages or not. The
1899 default is 'yes' if the kernel is configured for a regular host and 'no' for a
1900 router configuration.
1905 Should source routed packages be accepted or declined. The default is
1906 dependent on the kernel configuration. It's 'yes' for routers and 'no' for
1912 Accept packets with source address 0.b.c.d with destinations not to this host
1913 as local ones. It is supposed that a BOOTP relay daemon will catch and forward
1916 The default is 0, since this feature is not implemented yet (kernel version
1922 Enable or disable IP forwarding on this interface.
1927 Log packets with source addresses with no known route to kernel log.
1932 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE and a
1933 multicast routing daemon is required.
1938 Does (1) or does not (0) perform proxy ARP.
1943 Integer value determines if a source validation should be made. 1 means yes, 0
1944 means no. Disabled by default, but local/broadcast address spoofing is always
1947 If you set this to 1 on a router that is the only connection for a network to
1948 the net, it will prevent spoofing attacks against your internal networks
1949 (external addresses can still be spoofed), without the need for additional
1955 Accept ICMP redirect messages only for gateways, listed in default gateway
1956 list. Enabled by default.
1961 If it is not set the kernel does not assume that different subnets on this
1962 device can communicate directly. Default setting is 'yes'.
1967 Determines whether to send ICMP redirects to other hosts.
1972 The directory /proc/sys/net/ipv4/route contains several file to control
1975 error_burst and error_cost
1976 --------------------------
1978 These parameters are used to limit how many ICMP destination unreachable to
1979 send from the host in question. ICMP destination unreachable messages are
1980 sent when we cannot reach the next hop while trying to transmit a packet.
1981 It will also print some error messages to kernel logs if someone is ignoring
1982 our ICMP redirects. The higher the error_cost factor is, the fewer
1983 destination unreachable and error messages will be let through. Error_burst
1984 controls when destination unreachable messages and error messages will be
1985 dropped. The default settings limit warning messages to five every second.
1990 Writing to this file results in a flush of the routing cache.
1992 gc_elasticity, gc_interval, gc_min_interval_ms, gc_timeout, gc_thresh
1993 ---------------------------------------------------------------------
1995 Values to control the frequency and behavior of the garbage collection
1996 algorithm for the routing cache. gc_min_interval is deprecated and replaced
1997 by gc_min_interval_ms.
2003 Maximum size of the routing cache. Old entries will be purged once the cache
2004 reached has this size.
2006 redirect_load, redirect_number
2007 ------------------------------
2009 Factors which determine if more ICPM redirects should be sent to a specific
2010 host. No redirects will be sent once the load limit or the maximum number of
2011 redirects has been reached.
2016 Timeout for redirects. After this period redirects will be sent again, even if
2017 this has been stopped, because the load or number limit has been reached.
2019 Network Neighbor handling
2020 -------------------------
2022 Settings about how to handle connections with direct neighbors (nodes attached
2023 to the same link) can be found in the directory /proc/sys/net/ipv4/neigh.
2025 As we saw it in the conf directory, there is a default subdirectory which
2026 holds the default values, and one directory for each interface. The contents
2027 of the directories are identical, with the single exception that the default
2028 settings contain additional options to set garbage collection parameters.
2030 In the interface directories you'll find the following entries:
2032 base_reachable_time, base_reachable_time_ms
2033 -------------------------------------------
2035 A base value used for computing the random reachable time value as specified
2038 Expression of base_reachable_time, which is deprecated, is in seconds.
2039 Expression of base_reachable_time_ms is in milliseconds.
2041 retrans_time, retrans_time_ms
2042 -----------------------------
2044 The time between retransmitted Neighbor Solicitation messages.
2045 Used for address resolution and to determine if a neighbor is
2048 Expression of retrans_time, which is deprecated, is in 1/100 seconds (for
2049 IPv4) or in jiffies (for IPv6).
2050 Expression of retrans_time_ms is in milliseconds.
2055 Maximum queue length for a pending arp request - the number of packets which
2056 are accepted from other layers while the ARP address is still resolved.
2061 Maximum for random delay of answers to neighbor solicitation messages in
2062 jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support
2068 Maximum number of retries for unicast solicitation.
2073 Maximum number of retries for multicast solicitation.
2075 delay_first_probe_time
2076 ----------------------
2078 Delay for the first time probe if the neighbor is reachable. (see
2084 An ARP/neighbor entry is only replaced with a new one if the old is at least
2085 locktime old. This prevents ARP cache thrashing.
2090 Maximum time (real time is random [0..proxytime]) before answering to an ARP
2091 request for which we have an proxy ARP entry. In some cases, this is used to
2092 prevent network flooding.
2097 Maximum queue length of the delayed proxy arp timer. (see proxy_delay).
2102 Determines the number of requests to send to the user level ARP daemon. Use 0
2108 Determines how often to check for stale ARP entries. After an ARP entry is
2109 stale it will be resolved again (which is useful when an IP address migrates
2110 to another machine). When ucast_solicit is greater than 0 it first tries to
2111 send an ARP packet directly to the known host When that fails and
2112 mcast_solicit is greater than 0, an ARP request is broadcasted.
2117 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
2118 when Appletalk is loaded. The configurable parameters are:
2123 The amount of time we keep an ARP entry before expiring it. Used to age out
2129 The amount of time we will spend trying to resolve an Appletalk address.
2131 aarp-retransmit-limit
2132 ---------------------
2134 The number of times we will retransmit a query before giving up.
2139 Controls the rate at which expires are checked.
2141 The directory /proc/net/appletalk holds the list of active Appletalk sockets
2144 The fields indicate the DDP type, the local address (in network:node format)
2145 the remote address, the size of the transmit pending queue, the size of the
2146 received queue (bytes waiting for applications to read) the state and the uid
2149 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
2150 shows the name of the interface, its Appletalk address, the network range on
2151 that address (or network number for phase 1 networks), and the status of the
2154 /proc/net/atalk_route lists each known network route. It lists the target
2155 (network) that the route leads to, the router (may be directly connected), the
2156 route flags, and the device the route is using.
2161 The IPX protocol has no tunable values in proc/sys/net.
2163 The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
2164 socket giving the local and remote addresses in Novell format (that is
2165 network:node:port). In accordance with the strange Novell tradition,
2166 everything but the port is in hex. Not_Connected is displayed for sockets that
2167 are not tied to a specific remote address. The Tx and Rx queue sizes indicate
2168 the number of bytes pending for transmission and reception. The state
2169 indicates the state the socket is in and the uid is the owning uid of the
2172 The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
2173 it gives the network number, the node number, and indicates if the network is
2174 the primary network. It also indicates which device it is bound to (or
2175 Internal for internal networks) and the Frame Type if appropriate. Linux
2176 supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
2179 The /proc/net/ipx_route table holds a list of IPX routes. For each route it
2180 gives the destination network, the router node (or Directly) and the network
2181 address of the router (or Connected) for internal networks.
2183 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
2184 ----------------------------------------------------------
2186 The "mqueue" filesystem provides the necessary kernel features to enable the
2187 creation of a user space library that implements the POSIX message queues
2188 API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
2189 Interfaces specification.)
2191 The "mqueue" filesystem contains values for determining/setting the amount of
2192 resources used by the file system.
2194 /proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
2195 maximum number of message queues allowed on the system.
2197 /proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
2198 maximum number of messages in a queue value. In fact it is the limiting value
2199 for another (user) limit which is set in mq_open invocation. This attribute of
2200 a queue must be less or equal then msg_max.
2202 /proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
2203 maximum message size value (it is every message queue's attribute set during
2206 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
2207 ------------------------------------------------------
2209 This file can be used to adjust the score used to select which processes
2210 should be killed in an out-of-memory situation. Giving it a high score will
2211 increase the likelihood of this process being killed by the oom-killer. Valid
2212 values are in the range -16 to +15, plus the special value -17, which disables
2213 oom-killing altogether for this process.
2215 2.13 /proc/<pid>/oom_score - Display current oom-killer score
2216 -------------------------------------------------------------
2218 ------------------------------------------------------------------------------
2219 This file can be used to check the current score used by the oom-killer is for
2220 any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
2221 process should be killed in an out-of-memory situation.
2223 ------------------------------------------------------------------------------
2225 ------------------------------------------------------------------------------
2226 Certain aspects of kernel behavior can be modified at runtime, without the
2227 need to recompile the kernel, or even to reboot the system. The files in the
2228 /proc/sys tree can not only be read, but also modified. You can use the echo
2229 command to write value into these files, thereby changing the default settings
2231 ------------------------------------------------------------------------------
2233 2.14 /proc/<pid>/io - Display the IO accounting fields
2234 -------------------------------------------------------
2236 This file contains IO statistics for each running process
2241 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
2244 test:/tmp # cat /proc/3828/io
2250 write_bytes: 323932160
2251 cancelled_write_bytes: 0
2260 I/O counter: chars read
2261 The number of bytes which this task has caused to be read from storage. This
2262 is simply the sum of bytes which this process passed to read() and pread().
2263 It includes things like tty IO and it is unaffected by whether or not actual
2264 physical disk IO was required (the read might have been satisfied from
2271 I/O counter: chars written
2272 The number of bytes which this task has caused, or shall cause to be written
2273 to disk. Similar caveats apply here as with rchar.
2279 I/O counter: read syscalls
2280 Attempt to count the number of read I/O operations, i.e. syscalls like read()
2287 I/O counter: write syscalls
2288 Attempt to count the number of write I/O operations, i.e. syscalls like
2289 write() and pwrite().
2295 I/O counter: bytes read
2296 Attempt to count the number of bytes which this process really did cause to
2297 be fetched from the storage layer. Done at the submit_bio() level, so it is
2298 accurate for block-backed filesystems. <please add status regarding NFS and
2299 CIFS at a later time>
2305 I/O counter: bytes written
2306 Attempt to count the number of bytes which this process caused to be sent to
2307 the storage layer. This is done at page-dirtying time.
2310 cancelled_write_bytes
2311 ---------------------
2313 The big inaccuracy here is truncate. If a process writes 1MB to a file and
2314 then deletes the file, it will in fact perform no writeout. But it will have
2315 been accounted as having caused 1MB of write.
2316 In other words: The number of bytes which this process caused to not happen,
2317 by truncating pagecache. A task can cause "negative" IO too. If this task
2318 truncates some dirty pagecache, some IO which another task has been accounted
2319 for (in it's write_bytes) will not be happening. We _could_ just subtract that
2320 from the truncating task's write_bytes, but there is information loss in doing
2327 At its current implementation state, this is a bit racy on 32-bit machines: if
2328 process A reads process B's /proc/pid/io while process B is updating one of
2329 those 64-bit counters, process A could see an intermediate result.
2332 More information about this can be found within the taskstats documentation in
2333 Documentation/accounting.
2335 2.15 /proc/<pid>/coredump_filter - Core dump filtering settings
2336 ---------------------------------------------------------------
2337 When a process is dumped, all anonymous memory is written to a core file as
2338 long as the size of the core file isn't limited. But sometimes we don't want
2339 to dump some memory segments, for example, huge shared memory. Conversely,
2340 sometimes we want to save file-backed memory segments into a core file, not
2341 only the individual files.
2343 /proc/<pid>/coredump_filter allows you to customize which memory segments
2344 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
2345 of memory types. If a bit of the bitmask is set, memory segments of the
2346 corresponding memory type are dumped, otherwise they are not dumped.
2348 The following 4 memory types are supported:
2349 - (bit 0) anonymous private memory
2350 - (bit 1) anonymous shared memory
2351 - (bit 2) file-backed private memory
2352 - (bit 3) file-backed shared memory
2354 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
2355 are always dumped regardless of the bitmask status.
2357 Default value of coredump_filter is 0x3; this means all anonymous memory
2358 segments are dumped.
2360 If you don't want to dump all shared memory segments attached to pid 1234,
2361 write 1 to the process's proc file.
2363 $ echo 0x1 > /proc/1234/coredump_filter
2365 When a new process is created, the process inherits the bitmask status from its
2366 parent. It is useful to set up coredump_filter before the program runs.
2369 $ echo 0x7 > /proc/self/coredump_filter
2372 2.16 /proc/<pid>/mountinfo - Information about mounts
2373 --------------------------------------------------------
2375 This file contains lines of the form:
2377 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
2378 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
2380 (1) mount ID: unique identifier of the mount (may be reused after umount)
2381 (2) parent ID: ID of parent (or of self for the top of the mount tree)
2382 (3) major:minor: value of st_dev for files on filesystem
2383 (4) root: root of the mount within the filesystem
2384 (5) mount point: mount point relative to the process's root
2385 (6) mount options: per mount options
2386 (7) optional fields: zero or more fields of the form "tag[:value]"
2387 (8) separator: marks the end of the optional fields
2388 (9) filesystem type: name of filesystem of the form "type[.subtype]"
2389 (10) mount source: filesystem specific information or "none"
2390 (11) super options: per super block options
2392 Parsers should ignore all unrecognised optional fields. Currently the
2393 possible optional fields are:
2395 shared:X mount is shared in peer group X
2396 master:X mount is slave to peer group X
2397 propagate_from:X mount is slave and receives propagation from peer group X (*)
2398 unbindable mount is unbindable
2400 (*) X is the closest dominant peer group under the process's root. If
2401 X is the immediate master of the mount, or if there's no dominant peer
2402 group under the same root, then only the "master:X" field is present
2403 and not the "propagate_from:X" field.
2405 For more information on mount propagation see:
2407 Documentation/filesystems/sharedsubtree.txt
2409 ------------------------------------------------------------------------------
projects / linux-2.6 / blob