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 vmallocinfo Show vmalloced areas
300 ..............................................................................
302 You can, for example, check which interrupts are currently in use and what
303 they are used for by looking in the file /proc/interrupts:
305 > cat /proc/interrupts
307 0: 8728810 XT-PIC timer
308 1: 895 XT-PIC keyboard
310 3: 531695 XT-PIC aha152x
311 4: 2014133 XT-PIC serial
312 5: 44401 XT-PIC pcnet_cs
315 12: 182918 XT-PIC PS/2 Mouse
317 14: 1232265 XT-PIC ide0
321 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
322 output of a SMP machine):
324 > cat /proc/interrupts
327 0: 1243498 1214548 IO-APIC-edge timer
328 1: 8949 8958 IO-APIC-edge keyboard
329 2: 0 0 XT-PIC cascade
330 5: 11286 10161 IO-APIC-edge soundblaster
331 8: 1 0 IO-APIC-edge rtc
332 9: 27422 27407 IO-APIC-edge 3c503
333 12: 113645 113873 IO-APIC-edge PS/2 Mouse
335 14: 22491 24012 IO-APIC-edge ide0
336 15: 2183 2415 IO-APIC-edge ide1
337 17: 30564 30414 IO-APIC-level eth0
338 18: 177 164 IO-APIC-level bttv
343 NMI is incremented in this case because every timer interrupt generates a NMI
344 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
346 LOC is the local interrupt counter of the internal APIC of every CPU.
348 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
349 connects the CPUs in a SMP system. This means that an error has been detected,
350 the IO-APIC automatically retry the transmission, so it should not be a big
351 problem, but you should read the SMP-FAQ.
353 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
354 /proc/interrupts to display every IRQ vector in use by the system, not
355 just those considered 'most important'. The new vectors are:
357 THR -- interrupt raised when a machine check threshold counter
358 (typically counting ECC corrected errors of memory or cache) exceeds
359 a configurable threshold. Only available on some systems.
361 TRM -- a thermal event interrupt occurs when a temperature threshold
362 has been exceeded for the CPU. This interrupt may also be generated
363 when the temperature drops back to normal.
365 SPU -- a spurious interrupt is some interrupt that was raised then lowered
366 by some IO device before it could be fully processed by the APIC. Hence
367 the APIC sees the interrupt but does not know what device it came from.
368 For this case the APIC will generate the interrupt with a IRQ vector
369 of 0xff. This might also be generated by chipset bugs.
371 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
372 sent from one CPU to another per the needs of the OS. Typically,
373 their statistics are used by kernel developers and interested users to
374 determine the occurance of interrupt of the given type.
376 The above IRQ vectors are displayed only when relevent. For example,
377 the threshold vector does not exist on x86_64 platforms. Others are
378 suppressed when the system is a uniprocessor. As of this writing, only
379 i386 and x86_64 platforms support the new IRQ vector displays.
381 Of some interest is the introduction of the /proc/irq directory to 2.4.
382 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
383 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
384 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
389 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
390 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
394 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
395 IRQ, you can set it by doing:
397 > echo 1 > /proc/irq/10/smp_affinity
399 This means that only the first CPU will handle the IRQ, but you can also echo
400 5 which means that only the first and fourth CPU can handle the IRQ.
402 The contents of each smp_affinity file is the same by default:
404 > cat /proc/irq/0/smp_affinity
407 The default_smp_affinity mask applies to all non-active IRQs, which are the
408 IRQs which have not yet been allocated/activated, and hence which lack a
409 /proc/irq/[0-9]* directory.
411 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
412 profiler. Default value is ffffffff (all cpus).
414 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
415 between all the CPUs which are allowed to handle it. As usual the kernel has
416 more info than you and does a better job than you, so the defaults are the
417 best choice for almost everyone.
419 There are three more important subdirectories in /proc: net, scsi, and sys.
420 The general rule is that the contents, or even the existence of these
421 directories, depend on your kernel configuration. If SCSI is not enabled, the
422 directory scsi may not exist. The same is true with the net, which is there
423 only when networking support is present in the running kernel.
425 The slabinfo file gives information about memory usage at the slab level.
426 Linux uses slab pools for memory management above page level in version 2.2.
427 Commonly used objects have their own slab pool (such as network buffers,
428 directory cache, and so on).
430 ..............................................................................
432 > cat /proc/buddyinfo
434 Node 0, zone DMA 0 4 5 4 4 3 ...
435 Node 0, zone Normal 1 0 0 1 101 8 ...
436 Node 0, zone HighMem 2 0 0 1 1 0 ...
438 Memory fragmentation is a problem under some workloads, and buddyinfo is a
439 useful tool for helping diagnose these problems. Buddyinfo will give you a
440 clue as to how big an area you can safely allocate, or why a previous
443 Each column represents the number of pages of a certain order which are
444 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
445 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
446 available in ZONE_NORMAL, etc...
448 ..............................................................................
452 Provides information about distribution and utilization of memory. This
453 varies by architecture and compile options. The following is from a
454 16GB PIII, which has highmem enabled. You may not have all of these fields.
459 MemTotal: 16344972 kB
466 HighTotal: 15597528 kB
467 HighFree: 13629632 kB
477 SReclaimable: 159856 kB
478 SUnreclaim: 124508 kB
483 CommitLimit: 7669796 kB
484 Committed_AS: 100056 kB
485 VmallocTotal: 112216 kB
487 VmallocChunk: 111088 kB
489 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
490 bits and the kernel binary code)
491 MemFree: The sum of LowFree+HighFree
492 Buffers: Relatively temporary storage for raw disk blocks
493 shouldn't get tremendously large (20MB or so)
494 Cached: in-memory cache for files read from the disk (the
495 pagecache). Doesn't include SwapCached
496 SwapCached: Memory that once was swapped out, is swapped back in but
497 still also is in the swapfile (if memory is needed it
498 doesn't need to be swapped out AGAIN because it is already
499 in the swapfile. This saves I/O)
500 Active: Memory that has been used more recently and usually not
501 reclaimed unless absolutely necessary.
502 Inactive: Memory which has been less recently used. It is more
503 eligible to be reclaimed for other purposes
505 HighFree: Highmem is all memory above ~860MB of physical memory
506 Highmem areas are for use by userspace programs, or
507 for the pagecache. The kernel must use tricks to access
508 this memory, making it slower to access than lowmem.
510 LowFree: Lowmem is memory which can be used for everything that
511 highmem can be used for, but it is also available for the
512 kernel's use for its own data structures. Among many
513 other things, it is where everything from the Slab is
514 allocated. Bad things happen when you're out of lowmem.
515 SwapTotal: total amount of swap space available
516 SwapFree: Memory which has been evicted from RAM, and is temporarily
518 Dirty: Memory which is waiting to get written back to the disk
519 Writeback: Memory which is actively being written back to the disk
520 AnonPages: Non-file backed pages mapped into userspace page tables
521 Mapped: files which have been mmaped, such as libraries
522 Slab: in-kernel data structures cache
523 SReclaimable: Part of Slab, that might be reclaimed, such as caches
524 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
525 PageTables: amount of memory dedicated to the lowest level of page
527 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
529 Bounce: Memory used for block device "bounce buffers"
530 WritebackTmp: Memory used by FUSE for temporary writeback buffers
531 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
532 this is the total amount of memory currently available to
533 be allocated on the system. This limit is only adhered to
534 if strict overcommit accounting is enabled (mode 2 in
535 'vm.overcommit_memory').
536 The CommitLimit is calculated with the following formula:
537 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
538 For example, on a system with 1G of physical RAM and 7G
539 of swap with a `vm.overcommit_ratio` of 30 it would
540 yield a CommitLimit of 7.3G.
541 For more details, see the memory overcommit documentation
542 in vm/overcommit-accounting.
543 Committed_AS: The amount of memory presently allocated on the system.
544 The committed memory is a sum of all of the memory which
545 has been allocated by processes, even if it has not been
546 "used" by them as of yet. A process which malloc()'s 1G
547 of memory, but only touches 300M of it will only show up
548 as using 300M of memory even if it has the address space
549 allocated for the entire 1G. This 1G is memory which has
550 been "committed" to by the VM and can be used at any time
551 by the allocating application. With strict overcommit
552 enabled on the system (mode 2 in 'vm.overcommit_memory'),
553 allocations which would exceed the CommitLimit (detailed
554 above) will not be permitted. This is useful if one needs
555 to guarantee that processes will not fail due to lack of
556 memory once that memory has been successfully allocated.
557 VmallocTotal: total size of vmalloc memory area
558 VmallocUsed: amount of vmalloc area which is used
559 VmallocChunk: largest contigious block of vmalloc area which is free
561 ..............................................................................
565 Provides information about vmalloced/vmaped areas. One line per area,
566 containing the virtual address range of the area, size in bytes,
567 caller information of the creator, and optional information depending
568 on the kind of area :
570 pages=nr number of pages
571 phys=addr if a physical address was specified
572 ioremap I/O mapping (ioremap() and friends)
573 vmalloc vmalloc() area
576 vpages buffer for pages pointers was vmalloced (huge area)
577 N<node>=nr (Only on NUMA kernels)
578 Number of pages allocated on memory node <node>
580 > cat /proc/vmallocinfo
581 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
582 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
583 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
584 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
585 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
586 phys=7fee8000 ioremap
587 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
588 phys=7fee7000 ioremap
589 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
590 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
591 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
592 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
594 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
595 /0x130 [x_tables] pages=4 vmalloc N0=4
596 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
597 pages=14 vmalloc N2=14
598 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
600 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
602 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
603 pages=10 vmalloc N0=10
605 1.3 IDE devices in /proc/ide
606 ----------------------------
608 The subdirectory /proc/ide contains information about all IDE devices of which
609 the kernel is aware. There is one subdirectory for each IDE controller, the
610 file drivers and a link for each IDE device, pointing to the device directory
611 in the controller specific subtree.
613 The file drivers contains general information about the drivers used for the
616 > cat /proc/ide/drivers
617 ide-cdrom version 4.53
618 ide-disk version 1.08
620 More detailed information can be found in the controller specific
621 subdirectories. These are named ide0, ide1 and so on. Each of these
622 directories contains the files shown in table 1-5.
625 Table 1-5: IDE controller info in /proc/ide/ide?
626 ..............................................................................
628 channel IDE channel (0 or 1)
629 config Configuration (only for PCI/IDE bridge)
631 model Type/Chipset of IDE controller
632 ..............................................................................
634 Each device connected to a controller has a separate subdirectory in the
635 controllers directory. The files listed in table 1-6 are contained in these
639 Table 1-6: IDE device information
640 ..............................................................................
643 capacity Capacity of the medium (in 512Byte blocks)
644 driver driver and version
645 geometry physical and logical geometry
646 identify device identify block
648 model device identifier
649 settings device setup
650 smart_thresholds IDE disk management thresholds
651 smart_values IDE disk management values
652 ..............................................................................
654 The most interesting file is settings. This file contains a nice overview of
655 the drive parameters:
657 # cat /proc/ide/ide0/hda/settings
658 name value min max mode
659 ---- ----- --- --- ----
660 bios_cyl 526 0 65535 rw
661 bios_head 255 0 255 rw
663 breada_readahead 4 0 127 rw
665 file_readahead 72 0 2097151 rw
667 keepsettings 0 0 1 rw
668 max_kb_per_request 122 1 127 rw
672 pio_mode write-only 0 255 w
678 1.4 Networking info in /proc/net
679 --------------------------------
681 The subdirectory /proc/net follows the usual pattern. Table 1-6 shows the
682 additional values you get for IP version 6 if you configure the kernel to
683 support this. Table 1-7 lists the files and their meaning.
686 Table 1-6: IPv6 info in /proc/net
687 ..............................................................................
689 udp6 UDP sockets (IPv6)
690 tcp6 TCP sockets (IPv6)
691 raw6 Raw device statistics (IPv6)
692 igmp6 IP multicast addresses, which this host joined (IPv6)
693 if_inet6 List of IPv6 interface addresses
694 ipv6_route Kernel routing table for IPv6
695 rt6_stats Global IPv6 routing tables statistics
696 sockstat6 Socket statistics (IPv6)
697 snmp6 Snmp data (IPv6)
698 ..............................................................................
701 Table 1-7: Network info in /proc/net
702 ..............................................................................
705 dev network devices with statistics
706 dev_mcast the Layer2 multicast groups a device is listening too
707 (interface index, label, number of references, number of bound
709 dev_stat network device status
710 ip_fwchains Firewall chain linkage
711 ip_fwnames Firewall chain names
712 ip_masq Directory containing the masquerading tables
713 ip_masquerade Major masquerading table
714 netstat Network statistics
715 raw raw device statistics
716 route Kernel routing table
717 rpc Directory containing rpc info
718 rt_cache Routing cache
720 sockstat Socket statistics
722 tr_rif Token ring RIF routing table
724 unix UNIX domain sockets
725 wireless Wireless interface data (Wavelan etc)
726 igmp IP multicast addresses, which this host joined
727 psched Global packet scheduler parameters.
728 netlink List of PF_NETLINK sockets
729 ip_mr_vifs List of multicast virtual interfaces
730 ip_mr_cache List of multicast routing cache
731 ..............................................................................
733 You can use this information to see which network devices are available in
734 your system and how much traffic was routed over those devices:
738 face |bytes packets errs drop fifo frame compressed multicast|[...
739 lo: 908188 5596 0 0 0 0 0 0 [...
740 ppp0:15475140 20721 410 0 0 410 0 0 [...
741 eth0: 614530 7085 0 0 0 0 0 1 [...
744 ...] bytes packets errs drop fifo colls carrier compressed
745 ...] 908188 5596 0 0 0 0 0 0
746 ...] 1375103 17405 0 0 0 0 0 0
747 ...] 1703981 5535 0 0 0 3 0 0
749 In addition, each Channel Bond interface has it's own directory. For
750 example, the bond0 device will have a directory called /proc/net/bond0/.
751 It will contain information that is specific to that bond, such as the
752 current slaves of the bond, the link status of the slaves, and how
753 many times the slaves link has failed.
758 If you have a SCSI host adapter in your system, you'll find a subdirectory
759 named after the driver for this adapter in /proc/scsi. You'll also see a list
760 of all recognized SCSI devices in /proc/scsi:
764 Host: scsi0 Channel: 00 Id: 00 Lun: 00
765 Vendor: IBM Model: DGHS09U Rev: 03E0
766 Type: Direct-Access ANSI SCSI revision: 03
767 Host: scsi0 Channel: 00 Id: 06 Lun: 00
768 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
769 Type: CD-ROM ANSI SCSI revision: 02
772 The directory named after the driver has one file for each adapter found in
773 the system. These files contain information about the controller, including
774 the used IRQ and the IO address range. The amount of information shown is
775 dependent on the adapter you use. The example shows the output for an Adaptec
776 AHA-2940 SCSI adapter:
778 > cat /proc/scsi/aic7xxx/0
780 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
782 TCQ Enabled By Default : Disabled
783 AIC7XXX_PROC_STATS : Disabled
784 AIC7XXX_RESET_DELAY : 5
785 Adapter Configuration:
786 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
787 Ultra Wide Controller
788 PCI MMAPed I/O Base: 0xeb001000
789 Adapter SEEPROM Config: SEEPROM found and used.
790 Adaptec SCSI BIOS: Enabled
792 SCBs: Active 0, Max Active 2,
793 Allocated 15, HW 16, Page 255
795 BIOS Control Word: 0x18b6
796 Adapter Control Word: 0x005b
797 Extended Translation: Enabled
798 Disconnect Enable Flags: 0xffff
799 Ultra Enable Flags: 0x0001
800 Tag Queue Enable Flags: 0x0000
801 Ordered Queue Tag Flags: 0x0000
802 Default Tag Queue Depth: 8
803 Tagged Queue By Device array for aic7xxx host instance 0:
804 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
805 Actual queue depth per device for aic7xxx host instance 0:
806 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
809 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
810 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
811 Total transfers 160151 (74577 reads and 85574 writes)
813 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
814 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
815 Total transfers 0 (0 reads and 0 writes)
818 1.6 Parallel port info in /proc/parport
819 ---------------------------------------
821 The directory /proc/parport contains information about the parallel ports of
822 your system. It has one subdirectory for each port, named after the port
825 These directories contain the four files shown in Table 1-8.
828 Table 1-8: Files in /proc/parport
829 ..............................................................................
831 autoprobe Any IEEE-1284 device ID information that has been acquired.
832 devices list of the device drivers using that port. A + will appear by the
833 name of the device currently using the port (it might not appear
835 hardware Parallel port's base address, IRQ line and DMA channel.
836 irq IRQ that parport is using for that port. This is in a separate
837 file to allow you to alter it by writing a new value in (IRQ
839 ..............................................................................
841 1.7 TTY info in /proc/tty
842 -------------------------
844 Information about the available and actually used tty's can be found in the
845 directory /proc/tty.You'll find entries for drivers and line disciplines in
846 this directory, as shown in Table 1-9.
849 Table 1-9: Files in /proc/tty
850 ..............................................................................
852 drivers list of drivers and their usage
853 ldiscs registered line disciplines
854 driver/serial usage statistic and status of single tty lines
855 ..............................................................................
857 To see which tty's are currently in use, you can simply look into the file
860 > cat /proc/tty/drivers
861 pty_slave /dev/pts 136 0-255 pty:slave
862 pty_master /dev/ptm 128 0-255 pty:master
863 pty_slave /dev/ttyp 3 0-255 pty:slave
864 pty_master /dev/pty 2 0-255 pty:master
865 serial /dev/cua 5 64-67 serial:callout
866 serial /dev/ttyS 4 64-67 serial
867 /dev/tty0 /dev/tty0 4 0 system:vtmaster
868 /dev/ptmx /dev/ptmx 5 2 system
869 /dev/console /dev/console 5 1 system:console
870 /dev/tty /dev/tty 5 0 system:/dev/tty
871 unknown /dev/tty 4 1-63 console
874 1.8 Miscellaneous kernel statistics in /proc/stat
875 -------------------------------------------------
877 Various pieces of information about kernel activity are available in the
878 /proc/stat file. All of the numbers reported in this file are aggregates
879 since the system first booted. For a quick look, simply cat the file:
882 cpu 2255 34 2290 22625563 6290 127 456 0
883 cpu0 1132 34 1441 11311718 3675 127 438 0
884 cpu1 1123 0 849 11313845 2614 0 18 0
885 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
892 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
893 lines. These numbers identify the amount of time the CPU has spent performing
894 different kinds of work. Time units are in USER_HZ (typically hundredths of a
895 second). The meanings of the columns are as follows, from left to right:
897 - user: normal processes executing in user mode
898 - nice: niced processes executing in user mode
899 - system: processes executing in kernel mode
900 - idle: twiddling thumbs
901 - iowait: waiting for I/O to complete
902 - irq: servicing interrupts
903 - softirq: servicing softirqs
904 - steal: involuntary wait
906 The "intr" line gives counts of interrupts serviced since boot time, for each
907 of the possible system interrupts. The first column is the total of all
908 interrupts serviced; each subsequent column is the total for that particular
911 The "ctxt" line gives the total number of context switches across all CPUs.
913 The "btime" line gives the time at which the system booted, in seconds since
916 The "processes" line gives the number of processes and threads created, which
917 includes (but is not limited to) those created by calls to the fork() and
918 clone() system calls.
920 The "procs_running" line gives the number of processes currently running on
923 The "procs_blocked" line gives the number of processes currently blocked,
924 waiting for I/O to complete.
926 1.9 Ext4 file system parameters
927 ------------------------------
928 Ext4 file system have one directory per partition under /proc/fs/ext4/
929 # ls /proc/fs/ext4/hdc/
930 group_prealloc max_to_scan mb_groups mb_history min_to_scan order2_req
934 This file gives the details of multiblock allocator buddy cache of free blocks
937 Multiblock allocation history.
940 This file indicate whether the multiblock allocator should start collecting
941 statistics. The statistics are shown during unmount
944 The multiblock allocator normalize the block allocation request to
945 group_prealloc filesystem blocks if we don't have strip value set.
946 The stripe value can be specified at mount time or during mke2fs.
949 How long multiblock allocator can look for a best extent (in found extents)
952 How long multiblock allocator must look for a best extent
955 Multiblock allocator use 2^N search using buddies only for requests greater
956 than or equal to order2_req. The request size is specfied in file system
957 blocks. A value of 2 indicate only if the requests are greater than or equal
961 Files smaller than stream_req are served by the stream allocator, whose
962 purpose is to pack requests as close each to other as possible to
963 produce smooth I/O traffic. Avalue of 16 indicate that file smaller than 16
964 filesystem block size will use group based preallocation.
966 ------------------------------------------------------------------------------
968 ------------------------------------------------------------------------------
969 The /proc file system serves information about the running system. It not only
970 allows access to process data but also allows you to request the kernel status
971 by reading files in the hierarchy.
973 The directory structure of /proc reflects the types of information and makes
974 it easy, if not obvious, where to look for specific data.
975 ------------------------------------------------------------------------------
977 ------------------------------------------------------------------------------
978 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
979 ------------------------------------------------------------------------------
981 ------------------------------------------------------------------------------
983 ------------------------------------------------------------------------------
984 * Modifying kernel parameters by writing into files found in /proc/sys
985 * Exploring the files which modify certain parameters
986 * Review of the /proc/sys file tree
987 ------------------------------------------------------------------------------
990 A very interesting part of /proc is the directory /proc/sys. This is not only
991 a source of information, it also allows you to change parameters within the
992 kernel. Be very careful when attempting this. You can optimize your system,
993 but you can also cause it to crash. Never alter kernel parameters on a
994 production system. Set up a development machine and test to make sure that
995 everything works the way you want it to. You may have no alternative but to
996 reboot the machine once an error has been made.
998 To change a value, simply echo the new value into the file. An example is
999 given below in the section on the file system data. You need to be root to do
1000 this. You can create your own boot script to perform this every time your
1003 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1004 general things in the operation of the Linux kernel. Since some of the files
1005 can inadvertently disrupt your system, it is advisable to read both
1006 documentation and source before actually making adjustments. In any case, be
1007 very careful when writing to any of these files. The entries in /proc may
1008 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1009 review the kernel documentation in the directory /usr/src/linux/Documentation.
1010 This chapter is heavily based on the documentation included in the pre 2.2
1011 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1013 2.1 /proc/sys/fs - File system data
1014 -----------------------------------
1016 This subdirectory contains specific file system, file handle, inode, dentry
1017 and quota information.
1019 Currently, these files are in /proc/sys/fs:
1024 Status of the directory cache. Since directory entries are dynamically
1025 allocated and deallocated, this file indicates the current status. It holds
1026 six values, in which the last two are not used and are always zero. The others
1027 are listed in table 2-1.
1030 Table 2-1: Status files of the directory cache
1031 ..............................................................................
1033 nr_dentry Almost always zero
1034 nr_unused Number of unused cache entries
1036 in seconds after the entry may be reclaimed, when memory is short
1037 want_pages internally
1038 ..............................................................................
1040 dquot-nr and dquot-max
1041 ----------------------
1043 The file dquot-max shows the maximum number of cached disk quota entries.
1045 The file dquot-nr shows the number of allocated disk quota entries and the
1046 number of free disk quota entries.
1048 If the number of available cached disk quotas is very low and you have a large
1049 number of simultaneous system users, you might want to raise the limit.
1051 file-nr and file-max
1052 --------------------
1054 The kernel allocates file handles dynamically, but doesn't free them again at
1057 The value in file-max denotes the maximum number of file handles that the
1058 Linux kernel will allocate. When you get a lot of error messages about running
1059 out of file handles, you might want to raise this limit. The default value is
1060 10% of RAM in kilobytes. To change it, just write the new number into the
1063 # cat /proc/sys/fs/file-max
1065 # echo 8192 > /proc/sys/fs/file-max
1066 # cat /proc/sys/fs/file-max
1070 This method of revision is useful for all customizable parameters of the
1071 kernel - simply echo the new value to the corresponding file.
1073 Historically, the three values in file-nr denoted the number of allocated file
1074 handles, the number of allocated but unused file handles, and the maximum
1075 number of file handles. Linux 2.6 always reports 0 as the number of free file
1076 handles -- this is not an error, it just means that the number of allocated
1077 file handles exactly matches the number of used file handles.
1079 Attempts to allocate more file descriptors than file-max are reported with
1080 printk, look for "VFS: file-max limit <number> reached".
1082 inode-state and inode-nr
1083 ------------------------
1085 The file inode-nr contains the first two items from inode-state, so we'll skip
1088 inode-state contains two actual numbers and five dummy values. The numbers
1089 are nr_inodes and nr_free_inodes (in order of appearance).
1094 Denotes the number of inodes the system has allocated. This number will
1095 grow and shrink dynamically.
1100 Denotes the maximum number of file-handles a process can
1101 allocate. Default value is 1024*1024 (1048576) which should be
1102 enough for most machines. Actual limit depends on RLIMIT_NOFILE
1108 Represents the number of free inodes. Ie. The number of inuse inodes is
1109 (nr_inodes - nr_free_inodes).
1111 aio-nr and aio-max-nr
1112 ---------------------
1114 aio-nr is the running total of the number of events specified on the
1115 io_setup system call for all currently active aio contexts. If aio-nr
1116 reaches aio-max-nr then io_setup will fail with EAGAIN. Note that
1117 raising aio-max-nr does not result in the pre-allocation or re-sizing
1118 of any kernel data structures.
1120 2.2 /proc/sys/fs/binfmt_misc - Miscellaneous binary formats
1121 -----------------------------------------------------------
1123 Besides these files, there is the subdirectory /proc/sys/fs/binfmt_misc. This
1124 handles the kernel support for miscellaneous binary formats.
1126 Binfmt_misc provides the ability to register additional binary formats to the
1127 Kernel without compiling an additional module/kernel. Therefore, binfmt_misc
1128 needs to know magic numbers at the beginning or the filename extension of the
1131 It works by maintaining a linked list of structs that contain a description of
1132 a binary format, including a magic with size (or the filename extension),
1133 offset and mask, and the interpreter name. On request it invokes the given
1134 interpreter with the original program as argument, as binfmt_java and
1135 binfmt_em86 and binfmt_mz do. Since binfmt_misc does not define any default
1136 binary-formats, you have to register an additional binary-format.
1138 There are two general files in binfmt_misc and one file per registered format.
1139 The two general files are register and status.
1141 Registering a new binary format
1142 -------------------------------
1144 To register a new binary format you have to issue the command
1146 echo :name:type:offset:magic:mask:interpreter: > /proc/sys/fs/binfmt_misc/register
1150 with appropriate name (the name for the /proc-dir entry), offset (defaults to
1151 0, if omitted), magic, mask (which can be omitted, defaults to all 0xff) and
1152 last but not least, the interpreter that is to be invoked (for example and
1153 testing /bin/echo). Type can be M for usual magic matching or E for filename
1154 extension matching (give extension in place of magic).
1156 Check or reset the status of the binary format handler
1157 ------------------------------------------------------
1159 If you do a cat on the file /proc/sys/fs/binfmt_misc/status, you will get the
1160 current status (enabled/disabled) of binfmt_misc. Change the status by echoing
1161 0 (disables) or 1 (enables) or -1 (caution: this clears all previously
1162 registered binary formats) to status. For example echo 0 > status to disable
1163 binfmt_misc (temporarily).
1165 Status of a single handler
1166 --------------------------
1168 Each registered handler has an entry in /proc/sys/fs/binfmt_misc. These files
1169 perform the same function as status, but their scope is limited to the actual
1170 binary format. By cating this file, you also receive all related information
1171 about the interpreter/magic of the binfmt.
1173 Example usage of binfmt_misc (emulate binfmt_java)
1174 --------------------------------------------------
1176 cd /proc/sys/fs/binfmt_misc
1177 echo ':Java:M::\xca\xfe\xba\xbe::/usr/local/java/bin/javawrapper:' > register
1178 echo ':HTML:E::html::/usr/local/java/bin/appletviewer:' > register
1179 echo ':Applet:M::<!--applet::/usr/local/java/bin/appletviewer:' > register
1180 echo ':DEXE:M::\x0eDEX::/usr/bin/dosexec:' > register
1183 These four lines add support for Java executables and Java applets (like
1184 binfmt_java, additionally recognizing the .html extension with no need to put
1185 <!--applet> to every applet file). You have to install the JDK and the
1186 shell-script /usr/local/java/bin/javawrapper too. It works around the
1187 brokenness of the Java filename handling. To add a Java binary, just create a
1188 link to the class-file somewhere in the path.
1190 2.3 /proc/sys/kernel - general kernel parameters
1191 ------------------------------------------------
1193 This directory reflects general kernel behaviors. As I've said before, the
1194 contents depend on your configuration. Here you'll find the most important
1195 files, along with descriptions of what they mean and how to use them.
1200 The file contains three values; highwater, lowwater, and frequency.
1202 It exists only when BSD-style process accounting is enabled. These values
1203 control its behavior. If the free space on the file system where the log lives
1204 goes below lowwater percentage, accounting suspends. If it goes above
1205 highwater percentage, accounting resumes. Frequency determines how often you
1206 check the amount of free space (value is in seconds). Default settings are: 4,
1207 2, and 30. That is, suspend accounting if there is less than 2 percent free;
1208 resume it if we have a value of 3 or more percent; consider information about
1209 the amount of free space valid for 30 seconds
1214 When the value in this file is 0, ctrl-alt-del is trapped and sent to the init
1215 program to handle a graceful restart. However, when the value is greater that
1216 zero, Linux's reaction to this key combination will be an immediate reboot,
1217 without syncing its dirty buffers.
1220 When a program (like dosemu) has the keyboard in raw mode, the
1221 ctrl-alt-del is intercepted by the program before it ever reaches the
1222 kernel tty layer, and it is up to the program to decide what to do with
1225 domainname and hostname
1226 -----------------------
1228 These files can be controlled to set the NIS domainname and hostname of your
1229 box. For the classic darkstar.frop.org a simple:
1231 # echo "darkstar" > /proc/sys/kernel/hostname
1232 # echo "frop.org" > /proc/sys/kernel/domainname
1235 would suffice to set your hostname and NIS domainname.
1237 osrelease, ostype and version
1238 -----------------------------
1240 The names make it pretty obvious what these fields contain:
1242 > cat /proc/sys/kernel/osrelease
1245 > cat /proc/sys/kernel/ostype
1248 > cat /proc/sys/kernel/version
1249 #4 Fri Oct 1 12:41:14 PDT 1999
1252 The files osrelease and ostype should be clear enough. Version needs a little
1253 more clarification. The #4 means that this is the 4th kernel built from this
1254 source base and the date after it indicates the time the kernel was built. The
1255 only way to tune these values is to rebuild the kernel.
1260 The value in this file represents the number of seconds the kernel waits
1261 before rebooting on a panic. When you use the software watchdog, the
1262 recommended setting is 60. If set to 0, the auto reboot after a kernel panic
1263 is disabled, which is the default setting.
1268 The four values in printk denote
1270 * default_message_loglevel,
1271 * minimum_console_loglevel and
1272 * default_console_loglevel
1275 These values influence printk() behavior when printing or logging error
1276 messages, which come from inside the kernel. See syslog(2) for more
1277 information on the different log levels.
1282 Messages with a higher priority than this will be printed to the console.
1284 default_message_level
1285 ---------------------
1287 Messages without an explicit priority will be printed with this priority.
1289 minimum_console_loglevel
1290 ------------------------
1292 Minimum (highest) value to which the console_loglevel can be set.
1294 default_console_loglevel
1295 ------------------------
1297 Default value for console_loglevel.
1302 This file shows the size of the generic SCSI (sg) buffer. At this point, you
1303 can't tune it yet, but you can change it at compile time by editing
1304 include/scsi/sg.h and changing the value of SG_BIG_BUFF.
1306 If you use a scanner with SANE (Scanner Access Now Easy) you might want to set
1307 this to a higher value. Refer to the SANE documentation on this issue.
1312 The location where the modprobe binary is located. The kernel uses this
1313 program to load modules on demand.
1318 The value in this file affects behavior of handling NMI. When the value is
1319 non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
1320 debugging information is displayed on console.
1322 NMI switch that most IA32 servers have fires unknown NMI up, for example.
1323 If a system hangs up, try pressing the NMI switch.
1328 Enables/Disables the NMI watchdog on x86 systems. When the value is non-zero
1329 the NMI watchdog is enabled and will continuously test all online cpus to
1330 determine whether or not they are still functioning properly.
1332 Because the NMI watchdog shares registers with oprofile, by disabling the NMI
1333 watchdog, oprofile may have more registers to utilize.
1338 Enables/Disables the protection of the per-process proc entries "maps" and
1339 "smaps". When enabled, the contents of these files are visible only to
1340 readers that are allowed to ptrace() the given process.
1345 Maximum number of message queue ids on the system.
1346 This value scales to the amount of lowmem. It is automatically recomputed
1347 upon memory add/remove or ipc namespace creation/removal.
1348 When a value is written into this file, msgmni's value becomes fixed, i.e. it
1349 is not recomputed anymore when one of the above events occurs.
1350 Use auto_msgmni to change this behavior.
1355 Enables/Disables automatic recomputing of msgmni upon memory add/remove or
1356 upon ipc namespace creation/removal (see the msgmni description above).
1357 Echoing "1" into this file enables msgmni automatic recomputing.
1358 Echoing "0" turns it off.
1359 auto_msgmni default value is 1.
1362 2.4 /proc/sys/vm - The virtual memory subsystem
1363 -----------------------------------------------
1365 The files in this directory can be used to tune the operation of the virtual
1366 memory (VM) subsystem of the Linux kernel.
1371 Controls the tendency of the kernel to reclaim the memory which is used for
1372 caching of directory and inode objects.
1374 At the default value of vfs_cache_pressure=100 the kernel will attempt to
1375 reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1376 swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1377 to retain dentry and inode caches. Increasing vfs_cache_pressure beyond 100
1378 causes the kernel to prefer to reclaim dentries and inodes.
1380 dirty_background_ratio
1381 ----------------------
1383 Contains, as a percentage of total system memory, the number of pages at which
1384 the pdflush background writeback daemon will start writing out dirty data.
1389 Contains, as a percentage of total system memory, the number of pages at which
1390 a process which is generating disk writes will itself start writing out dirty
1393 dirty_writeback_centisecs
1394 -------------------------
1396 The pdflush writeback daemons will periodically wake up and write `old' data
1397 out to disk. This tunable expresses the interval between those wakeups, in
1398 100'ths of a second.
1400 Setting this to zero disables periodic writeback altogether.
1402 dirty_expire_centisecs
1403 ----------------------
1405 This tunable is used to define when dirty data is old enough to be eligible
1406 for writeout by the pdflush daemons. It is expressed in 100'ths of a second.
1407 Data which has been dirty in-memory for longer than this interval will be
1408 written out next time a pdflush daemon wakes up.
1410 highmem_is_dirtyable
1411 --------------------
1413 Only present if CONFIG_HIGHMEM is set.
1415 This defaults to 0 (false), meaning that the ratios set above are calculated
1416 as a percentage of lowmem only. This protects against excessive scanning
1417 in page reclaim, swapping and general VM distress.
1419 Setting this to 1 can be useful on 32 bit machines where you want to make
1420 random changes within an MMAPed file that is larger than your available
1421 lowmem without causing large quantities of random IO. Is is safe if the
1422 behavior of all programs running on the machine is known and memory will
1423 not be otherwise stressed.
1428 If non-zero, this sysctl disables the new 32-bit mmap mmap layout - the kernel
1429 will use the legacy (2.4) layout for all processes.
1431 lowmem_reserve_ratio
1432 ---------------------
1434 For some specialised workloads on highmem machines it is dangerous for
1435 the kernel to allow process memory to be allocated from the "lowmem"
1436 zone. This is because that memory could then be pinned via the mlock()
1437 system call, or by unavailability of swapspace.
1439 And on large highmem machines this lack of reclaimable lowmem memory
1442 So the Linux page allocator has a mechanism which prevents allocations
1443 which _could_ use highmem from using too much lowmem. This means that
1444 a certain amount of lowmem is defended from the possibility of being
1445 captured into pinned user memory.
1447 (The same argument applies to the old 16 megabyte ISA DMA region. This
1448 mechanism will also defend that region from allocations which could use
1451 The `lowmem_reserve_ratio' tunable determines how aggressive the kernel is
1452 in defending these lower zones.
1454 If you have a machine which uses highmem or ISA DMA and your
1455 applications are using mlock(), or if you are running with no swap then
1456 you probably should change the lowmem_reserve_ratio setting.
1458 The lowmem_reserve_ratio is an array. You can see them by reading this file.
1460 % cat /proc/sys/vm/lowmem_reserve_ratio
1463 Note: # of this elements is one fewer than number of zones. Because the highest
1464 zone's value is not necessary for following calculation.
1466 But, these values are not used directly. The kernel calculates # of protection
1467 pages for each zones from them. These are shown as array of protection pages
1468 in /proc/zoneinfo like followings. (This is an example of x86-64 box).
1469 Each zone has an array of protection pages like this.
1480 protection: (0, 2004, 2004, 2004)
1481 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1486 These protections are added to score to judge whether this zone should be used
1487 for page allocation or should be reclaimed.
1489 In this example, if normal pages (index=2) are required to this DMA zone and
1490 pages_high is used for watermark, the kernel judges this zone should not be
1491 used because pages_free(1355) is smaller than watermark + protection[2]
1492 (4 + 2004 = 2008). If this protection value is 0, this zone would be used for
1493 normal page requirement. If requirement is DMA zone(index=0), protection[0]
1496 zone[i]'s protection[j] is calculated by following expression.
1499 zone[i]->protection[j]
1500 = (total sums of present_pages from zone[i+1] to zone[j] on the node)
1501 / lowmem_reserve_ratio[i];
1503 (should not be protected. = 0;
1505 (not necessary, but looks 0)
1507 The default values of lowmem_reserve_ratio[i] are
1508 256 (if zone[i] means DMA or DMA32 zone)
1510 As above expression, they are reciprocal number of ratio.
1511 256 means 1/256. # of protection pages becomes about "0.39%" of total present
1512 pages of higher zones on the node.
1514 If you would like to protect more pages, smaller values are effective.
1515 The minimum value is 1 (1/1 -> 100%).
1520 page-cluster controls the number of pages which are written to swap in
1521 a single attempt. The swap I/O size.
1523 It is a logarithmic value - setting it to zero means "1 page", setting
1524 it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
1526 The default value is three (eight pages at a time). There may be some
1527 small benefits in tuning this to a different value if your workload is
1533 Controls overcommit of system memory, possibly allowing processes
1534 to allocate (but not use) more memory than is actually available.
1537 0 - Heuristic overcommit handling. Obvious overcommits of
1538 address space are refused. Used for a typical system. It
1539 ensures a seriously wild allocation fails while allowing
1540 overcommit to reduce swap usage. root is allowed to
1541 allocate slightly more memory in this mode. This is the
1544 1 - Always overcommit. Appropriate for some scientific
1547 2 - Don't overcommit. The total address space commit
1548 for the system is not permitted to exceed swap plus a
1549 configurable percentage (default is 50) of physical RAM.
1550 Depending on the percentage you use, in most situations
1551 this means a process will not be killed while attempting
1552 to use already-allocated memory but will receive errors
1553 on memory allocation as appropriate.
1558 Percentage of physical memory size to include in overcommit calculations
1561 Memory allocation limit = swapspace + physmem * (overcommit_ratio / 100)
1563 swapspace = total size of all swap areas
1564 physmem = size of physical memory in system
1566 nr_hugepages and hugetlb_shm_group
1567 ----------------------------------
1569 nr_hugepages configures number of hugetlb page reserved for the system.
1571 hugetlb_shm_group contains group id that is allowed to create SysV shared
1572 memory segment using hugetlb page.
1574 hugepages_treat_as_movable
1575 --------------------------
1577 This parameter is only useful when kernelcore= is specified at boot time to
1578 create ZONE_MOVABLE for pages that may be reclaimed or migrated. Huge pages
1579 are not movable so are not normally allocated from ZONE_MOVABLE. A non-zero
1580 value written to hugepages_treat_as_movable allows huge pages to be allocated
1583 Once enabled, the ZONE_MOVABLE is treated as an area of memory the huge
1584 pages pool can easily grow or shrink within. Assuming that applications are
1585 not running that mlock() a lot of memory, it is likely the huge pages pool
1586 can grow to the size of ZONE_MOVABLE by repeatedly entering the desired value
1587 into nr_hugepages and triggering page reclaim.
1592 laptop_mode is a knob that controls "laptop mode". All the things that are
1593 controlled by this knob are discussed in Documentation/laptops/laptop-mode.txt.
1598 block_dump enables block I/O debugging when set to a nonzero value. More
1599 information on block I/O debugging is in Documentation/laptops/laptop-mode.txt.
1604 This file contains valid hold time of swap out protection token. The Linux
1605 VM has token based thrashing control mechanism and uses the token to prevent
1606 unnecessary page faults in thrashing situation. The unit of the value is
1607 second. The value would be useful to tune thrashing behavior.
1612 Writing to this will cause the kernel to drop clean caches, dentries and
1613 inodes from memory, causing that memory to become free.
1616 echo 1 > /proc/sys/vm/drop_caches
1617 To free dentries and inodes:
1618 echo 2 > /proc/sys/vm/drop_caches
1619 To free pagecache, dentries and inodes:
1620 echo 3 > /proc/sys/vm/drop_caches
1622 As this is a non-destructive operation and dirty objects are not freeable, the
1623 user should run `sync' first.
1626 2.5 /proc/sys/dev - Device specific parameters
1627 ----------------------------------------------
1629 Currently there is only support for CDROM drives, and for those, there is only
1630 one read-only file containing information about the CD-ROM drives attached to
1633 >cat /proc/sys/dev/cdrom/info
1634 CD-ROM information, Id: cdrom.c 2.55 1999/04/25
1638 drive # of slots: 1 0
1642 Can change speed: 1 1
1643 Can select disk: 0 1
1644 Can read multisession: 1 1
1646 Reports media changed: 1 1
1650 You see two drives, sr0 and hdb, along with a list of their features.
1652 2.6 /proc/sys/sunrpc - Remote procedure calls
1653 ---------------------------------------------
1655 This directory contains four files, which enable or disable debugging for the
1656 RPC functions NFS, NFS-daemon, RPC and NLM. The default values are 0. They can
1657 be set to one to turn debugging on. (The default value is 0 for each)
1659 2.7 /proc/sys/net - Networking stuff
1660 ------------------------------------
1662 The interface to the networking parts of the kernel is located in
1663 /proc/sys/net. Table 2-3 shows all possible subdirectories. You may see only
1664 some of them, depending on your kernel's configuration.
1667 Table 2-3: Subdirectories in /proc/sys/net
1668 ..............................................................................
1669 Directory Content Directory Content
1670 core General parameter appletalk Appletalk protocol
1671 unix Unix domain sockets netrom NET/ROM
1672 802 E802 protocol ax25 AX25
1673 ethernet Ethernet protocol rose X.25 PLP layer
1674 ipv4 IP version 4 x25 X.25 protocol
1675 ipx IPX token-ring IBM token ring
1676 bridge Bridging decnet DEC net
1678 ..............................................................................
1680 We will concentrate on IP networking here. Since AX15, X.25, and DEC Net are
1681 only minor players in the Linux world, we'll skip them in this chapter. You'll
1682 find some short info on Appletalk and IPX further on in this chapter. Review
1683 the online documentation and the kernel source to get a detailed view of the
1684 parameters for those protocols. In this section we'll discuss the
1685 subdirectories printed in bold letters in the table above. As default values
1686 are suitable for most needs, there is no need to change these values.
1688 /proc/sys/net/core - Network core options
1689 -----------------------------------------
1694 The default setting of the socket receive buffer in bytes.
1699 The maximum receive socket buffer size in bytes.
1704 The default setting (in bytes) of the socket send buffer.
1709 The maximum send socket buffer size in bytes.
1711 message_burst and message_cost
1712 ------------------------------
1714 These parameters are used to limit the warning messages written to the kernel
1715 log from the networking code. They enforce a rate limit to make a
1716 denial-of-service attack impossible. A higher message_cost factor, results in
1717 fewer messages that will be written. Message_burst controls when messages will
1718 be dropped. The default settings limit warning messages to one every five
1724 This controls console messages from the networking stack that can occur because
1725 of problems on the network like duplicate address or bad checksums. Normally,
1726 this should be enabled, but if the problem persists the messages can be
1733 Maximum number of packets, queued on the INPUT side, when the interface
1734 receives packets faster than kernel can process them.
1739 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
1740 of struct cmsghdr structures with appended data.
1742 /proc/sys/net/unix - Parameters for Unix domain sockets
1743 -------------------------------------------------------
1745 There are only two files in this subdirectory. They control the delays for
1746 deleting and destroying socket descriptors.
1748 2.8 /proc/sys/net/ipv4 - IPV4 settings
1749 --------------------------------------
1751 IP version 4 is still the most used protocol in Unix networking. It will be
1752 replaced by IP version 6 in the next couple of years, but for the moment it's
1753 the de facto standard for the internet and is used in most networking
1754 environments around the world. Because of the importance of this protocol,
1755 we'll have a deeper look into the subtree controlling the behavior of the IPv4
1756 subsystem of the Linux kernel.
1758 Let's start with the entries in /proc/sys/net/ipv4.
1763 icmp_echo_ignore_all and icmp_echo_ignore_broadcasts
1764 ----------------------------------------------------
1766 Turn on (1) or off (0), if the kernel should ignore all ICMP ECHO requests, or
1767 just those to broadcast and multicast addresses.
1769 Please note that if you accept ICMP echo requests with a broadcast/multi\-cast
1770 destination address your network may be used as an exploder for denial of
1771 service packet flooding attacks to other hosts.
1773 icmp_destunreach_rate, icmp_echoreply_rate, icmp_paramprob_rate and icmp_timeexeed_rate
1774 ---------------------------------------------------------------------------------------
1776 Sets limits for sending ICMP packets to specific targets. A value of zero
1777 disables all limiting. Any positive value sets the maximum package rate in
1778 hundredth of a second (on Intel systems).
1786 This file contains the number one if the host received its IP configuration by
1787 RARP, BOOTP, DHCP or a similar mechanism. Otherwise it is zero.
1792 TTL (Time To Live) for IPv4 interfaces. This is simply the maximum number of
1793 hops a packet may travel.
1798 Enable dynamic socket address rewriting on interface address change. This is
1799 useful for dialup interface with changing IP addresses.
1804 Enable or disable forwarding of IP packages between interfaces. Changing this
1805 value resets all other parameters to their default values. They differ if the
1806 kernel is configured as host or router.
1811 Range of ports used by TCP and UDP to choose the local port. Contains two
1812 numbers, the first number is the lowest port, the second number the highest
1813 local port. Default is 1024-4999. Should be changed to 32768-61000 for
1819 Global switch to turn path MTU discovery off. It can also be set on a per
1820 socket basis by the applications or on a per route basis.
1825 Enable/disable debugging of IP masquerading.
1827 IP fragmentation settings
1828 -------------------------
1830 ipfrag_high_trash and ipfrag_low_trash
1831 --------------------------------------
1833 Maximum memory used to reassemble IP fragments. When ipfrag_high_thresh bytes
1834 of memory is allocated for this purpose, the fragment handler will toss
1835 packets until ipfrag_low_thresh is reached.
1840 Time in seconds to keep an IP fragment in memory.
1848 This file controls the use of the ECN bit in the IPv4 headers. This is a new
1849 feature about Explicit Congestion Notification, but some routers and firewalls
1850 block traffic that has this bit set, so it could be necessary to echo 0 to
1851 /proc/sys/net/ipv4/tcp_ecn if you want to talk to these sites. For more info
1852 you could read RFC2481.
1854 tcp_retrans_collapse
1855 --------------------
1857 Bug-to-bug compatibility with some broken printers. On retransmit, try to send
1858 larger packets to work around bugs in certain TCP stacks. Can be turned off by
1861 tcp_keepalive_probes
1862 --------------------
1864 Number of keep alive probes TCP sends out, until it decides that the
1865 connection is broken.
1870 How often TCP sends out keep alive messages, when keep alive is enabled. The
1876 Number of times initial SYNs for a TCP connection attempt will be
1877 retransmitted. Should not be higher than 255. This is only the timeout for
1878 outgoing connections, for incoming connections the number of retransmits is
1879 defined by tcp_retries1.
1884 Enable select acknowledgments after RFC2018.
1889 Enable timestamps as defined in RFC1323.
1894 Enable the strict RFC793 interpretation of the TCP urgent pointer field. The
1895 default is to use the BSD compatible interpretation of the urgent pointer
1896 pointing to the first byte after the urgent data. The RFC793 interpretation is
1897 to have it point to the last byte of urgent data. Enabling this option may
1898 lead to interoperability problems. Disabled by default.
1903 Only valid when the kernel was compiled with CONFIG_SYNCOOKIES. Send out
1904 syncookies when the syn backlog queue of a socket overflows. This is to ward
1905 off the common 'syn flood attack'. Disabled by default.
1907 Note that the concept of a socket backlog is abandoned. This means the peer
1908 may not receive reliable error messages from an over loaded server with
1914 Enable window scaling as defined in RFC1323.
1919 The length of time in seconds it takes to receive a final FIN before the
1920 socket is always closed. This is strictly a violation of the TCP
1921 specification, but required to prevent denial-of-service attacks.
1926 Indicates how many keep alive probes are sent per slow timer run. Should not
1927 be set too high to prevent bursts.
1932 Length of the per socket backlog queue. Since Linux 2.2 the backlog specified
1933 in listen(2) only specifies the length of the backlog queue of already
1934 established sockets. When more connection requests arrive Linux starts to drop
1935 packets. When syncookies are enabled the packets are still answered and the
1936 maximum queue is effectively ignored.
1941 Defines how often an answer to a TCP connection request is retransmitted
1947 Defines how often a TCP packet is retransmitted before giving up.
1949 Interface specific settings
1950 ---------------------------
1952 In the directory /proc/sys/net/ipv4/conf you'll find one subdirectory for each
1953 interface the system knows about and one directory calls all. Changes in the
1954 all subdirectory affect all interfaces, whereas changes in the other
1955 subdirectories affect only one interface. All directories have the same
1961 This switch decides if the kernel accepts ICMP redirect messages or not. The
1962 default is 'yes' if the kernel is configured for a regular host and 'no' for a
1963 router configuration.
1968 Should source routed packages be accepted or declined. The default is
1969 dependent on the kernel configuration. It's 'yes' for routers and 'no' for
1975 Accept packets with source address 0.b.c.d with destinations not to this host
1976 as local ones. It is supposed that a BOOTP relay daemon will catch and forward
1979 The default is 0, since this feature is not implemented yet (kernel version
1985 Enable or disable IP forwarding on this interface.
1990 Log packets with source addresses with no known route to kernel log.
1995 Do multicast routing. The kernel needs to be compiled with CONFIG_MROUTE and a
1996 multicast routing daemon is required.
2001 Does (1) or does not (0) perform proxy ARP.
2006 Integer value determines if a source validation should be made. 1 means yes, 0
2007 means no. Disabled by default, but local/broadcast address spoofing is always
2010 If you set this to 1 on a router that is the only connection for a network to
2011 the net, it will prevent spoofing attacks against your internal networks
2012 (external addresses can still be spoofed), without the need for additional
2018 Accept ICMP redirect messages only for gateways, listed in default gateway
2019 list. Enabled by default.
2024 If it is not set the kernel does not assume that different subnets on this
2025 device can communicate directly. Default setting is 'yes'.
2030 Determines whether to send ICMP redirects to other hosts.
2035 The directory /proc/sys/net/ipv4/route contains several file to control
2038 error_burst and error_cost
2039 --------------------------
2041 These parameters are used to limit how many ICMP destination unreachable to
2042 send from the host in question. ICMP destination unreachable messages are
2043 sent when we cannot reach the next hop while trying to transmit a packet.
2044 It will also print some error messages to kernel logs if someone is ignoring
2045 our ICMP redirects. The higher the error_cost factor is, the fewer
2046 destination unreachable and error messages will be let through. Error_burst
2047 controls when destination unreachable messages and error messages will be
2048 dropped. The default settings limit warning messages to five every second.
2053 Writing to this file results in a flush of the routing cache.
2055 gc_elasticity, gc_interval, gc_min_interval_ms, gc_timeout, gc_thresh
2056 ---------------------------------------------------------------------
2058 Values to control the frequency and behavior of the garbage collection
2059 algorithm for the routing cache. gc_min_interval is deprecated and replaced
2060 by gc_min_interval_ms.
2066 Maximum size of the routing cache. Old entries will be purged once the cache
2067 reached has this size.
2069 redirect_load, redirect_number
2070 ------------------------------
2072 Factors which determine if more ICPM redirects should be sent to a specific
2073 host. No redirects will be sent once the load limit or the maximum number of
2074 redirects has been reached.
2079 Timeout for redirects. After this period redirects will be sent again, even if
2080 this has been stopped, because the load or number limit has been reached.
2082 Network Neighbor handling
2083 -------------------------
2085 Settings about how to handle connections with direct neighbors (nodes attached
2086 to the same link) can be found in the directory /proc/sys/net/ipv4/neigh.
2088 As we saw it in the conf directory, there is a default subdirectory which
2089 holds the default values, and one directory for each interface. The contents
2090 of the directories are identical, with the single exception that the default
2091 settings contain additional options to set garbage collection parameters.
2093 In the interface directories you'll find the following entries:
2095 base_reachable_time, base_reachable_time_ms
2096 -------------------------------------------
2098 A base value used for computing the random reachable time value as specified
2101 Expression of base_reachable_time, which is deprecated, is in seconds.
2102 Expression of base_reachable_time_ms is in milliseconds.
2104 retrans_time, retrans_time_ms
2105 -----------------------------
2107 The time between retransmitted Neighbor Solicitation messages.
2108 Used for address resolution and to determine if a neighbor is
2111 Expression of retrans_time, which is deprecated, is in 1/100 seconds (for
2112 IPv4) or in jiffies (for IPv6).
2113 Expression of retrans_time_ms is in milliseconds.
2118 Maximum queue length for a pending arp request - the number of packets which
2119 are accepted from other layers while the ARP address is still resolved.
2124 Maximum for random delay of answers to neighbor solicitation messages in
2125 jiffies (1/100 sec). Not yet implemented (Linux does not have anycast support
2131 Maximum number of retries for unicast solicitation.
2136 Maximum number of retries for multicast solicitation.
2138 delay_first_probe_time
2139 ----------------------
2141 Delay for the first time probe if the neighbor is reachable. (see
2147 An ARP/neighbor entry is only replaced with a new one if the old is at least
2148 locktime old. This prevents ARP cache thrashing.
2153 Maximum time (real time is random [0..proxytime]) before answering to an ARP
2154 request for which we have an proxy ARP entry. In some cases, this is used to
2155 prevent network flooding.
2160 Maximum queue length of the delayed proxy arp timer. (see proxy_delay).
2165 Determines the number of requests to send to the user level ARP daemon. Use 0
2171 Determines how often to check for stale ARP entries. After an ARP entry is
2172 stale it will be resolved again (which is useful when an IP address migrates
2173 to another machine). When ucast_solicit is greater than 0 it first tries to
2174 send an ARP packet directly to the known host When that fails and
2175 mcast_solicit is greater than 0, an ARP request is broadcasted.
2180 The /proc/sys/net/appletalk directory holds the Appletalk configuration data
2181 when Appletalk is loaded. The configurable parameters are:
2186 The amount of time we keep an ARP entry before expiring it. Used to age out
2192 The amount of time we will spend trying to resolve an Appletalk address.
2194 aarp-retransmit-limit
2195 ---------------------
2197 The number of times we will retransmit a query before giving up.
2202 Controls the rate at which expires are checked.
2204 The directory /proc/net/appletalk holds the list of active Appletalk sockets
2207 The fields indicate the DDP type, the local address (in network:node format)
2208 the remote address, the size of the transmit pending queue, the size of the
2209 received queue (bytes waiting for applications to read) the state and the uid
2212 /proc/net/atalk_iface lists all the interfaces configured for appletalk.It
2213 shows the name of the interface, its Appletalk address, the network range on
2214 that address (or network number for phase 1 networks), and the status of the
2217 /proc/net/atalk_route lists each known network route. It lists the target
2218 (network) that the route leads to, the router (may be directly connected), the
2219 route flags, and the device the route is using.
2224 The IPX protocol has no tunable values in proc/sys/net.
2226 The IPX protocol does, however, provide proc/net/ipx. This lists each IPX
2227 socket giving the local and remote addresses in Novell format (that is
2228 network:node:port). In accordance with the strange Novell tradition,
2229 everything but the port is in hex. Not_Connected is displayed for sockets that
2230 are not tied to a specific remote address. The Tx and Rx queue sizes indicate
2231 the number of bytes pending for transmission and reception. The state
2232 indicates the state the socket is in and the uid is the owning uid of the
2235 The /proc/net/ipx_interface file lists all IPX interfaces. For each interface
2236 it gives the network number, the node number, and indicates if the network is
2237 the primary network. It also indicates which device it is bound to (or
2238 Internal for internal networks) and the Frame Type if appropriate. Linux
2239 supports 802.3, 802.2, 802.2 SNAP and DIX (Blue Book) ethernet framing for
2242 The /proc/net/ipx_route table holds a list of IPX routes. For each route it
2243 gives the destination network, the router node (or Directly) and the network
2244 address of the router (or Connected) for internal networks.
2246 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
2247 ----------------------------------------------------------
2249 The "mqueue" filesystem provides the necessary kernel features to enable the
2250 creation of a user space library that implements the POSIX message queues
2251 API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
2252 Interfaces specification.)
2254 The "mqueue" filesystem contains values for determining/setting the amount of
2255 resources used by the file system.
2257 /proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
2258 maximum number of message queues allowed on the system.
2260 /proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
2261 maximum number of messages in a queue value. In fact it is the limiting value
2262 for another (user) limit which is set in mq_open invocation. This attribute of
2263 a queue must be less or equal then msg_max.
2265 /proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
2266 maximum message size value (it is every message queue's attribute set during
2269 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score
2270 ------------------------------------------------------
2272 This file can be used to adjust the score used to select which processes
2273 should be killed in an out-of-memory situation. Giving it a high score will
2274 increase the likelihood of this process being killed by the oom-killer. Valid
2275 values are in the range -16 to +15, plus the special value -17, which disables
2276 oom-killing altogether for this process.
2278 2.13 /proc/<pid>/oom_score - Display current oom-killer score
2279 -------------------------------------------------------------
2281 ------------------------------------------------------------------------------
2282 This file can be used to check the current score used by the oom-killer is for
2283 any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
2284 process should be killed in an out-of-memory situation.
2286 ------------------------------------------------------------------------------
2288 ------------------------------------------------------------------------------
2289 Certain aspects of kernel behavior can be modified at runtime, without the
2290 need to recompile the kernel, or even to reboot the system. The files in the
2291 /proc/sys tree can not only be read, but also modified. You can use the echo
2292 command to write value into these files, thereby changing the default settings
2294 ------------------------------------------------------------------------------
2296 2.14 /proc/<pid>/io - Display the IO accounting fields
2297 -------------------------------------------------------
2299 This file contains IO statistics for each running process
2304 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
2307 test:/tmp # cat /proc/3828/io
2313 write_bytes: 323932160
2314 cancelled_write_bytes: 0
2323 I/O counter: chars read
2324 The number of bytes which this task has caused to be read from storage. This
2325 is simply the sum of bytes which this process passed to read() and pread().
2326 It includes things like tty IO and it is unaffected by whether or not actual
2327 physical disk IO was required (the read might have been satisfied from
2334 I/O counter: chars written
2335 The number of bytes which this task has caused, or shall cause to be written
2336 to disk. Similar caveats apply here as with rchar.
2342 I/O counter: read syscalls
2343 Attempt to count the number of read I/O operations, i.e. syscalls like read()
2350 I/O counter: write syscalls
2351 Attempt to count the number of write I/O operations, i.e. syscalls like
2352 write() and pwrite().
2358 I/O counter: bytes read
2359 Attempt to count the number of bytes which this process really did cause to
2360 be fetched from the storage layer. Done at the submit_bio() level, so it is
2361 accurate for block-backed filesystems. <please add status regarding NFS and
2362 CIFS at a later time>
2368 I/O counter: bytes written
2369 Attempt to count the number of bytes which this process caused to be sent to
2370 the storage layer. This is done at page-dirtying time.
2373 cancelled_write_bytes
2374 ---------------------
2376 The big inaccuracy here is truncate. If a process writes 1MB to a file and
2377 then deletes the file, it will in fact perform no writeout. But it will have
2378 been accounted as having caused 1MB of write.
2379 In other words: The number of bytes which this process caused to not happen,
2380 by truncating pagecache. A task can cause "negative" IO too. If this task
2381 truncates some dirty pagecache, some IO which another task has been accounted
2382 for (in it's write_bytes) will not be happening. We _could_ just subtract that
2383 from the truncating task's write_bytes, but there is information loss in doing
2390 At its current implementation state, this is a bit racy on 32-bit machines: if
2391 process A reads process B's /proc/pid/io while process B is updating one of
2392 those 64-bit counters, process A could see an intermediate result.
2395 More information about this can be found within the taskstats documentation in
2396 Documentation/accounting.
2398 2.15 /proc/<pid>/coredump_filter - Core dump filtering settings
2399 ---------------------------------------------------------------
2400 When a process is dumped, all anonymous memory is written to a core file as
2401 long as the size of the core file isn't limited. But sometimes we don't want
2402 to dump some memory segments, for example, huge shared memory. Conversely,
2403 sometimes we want to save file-backed memory segments into a core file, not
2404 only the individual files.
2406 /proc/<pid>/coredump_filter allows you to customize which memory segments
2407 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
2408 of memory types. If a bit of the bitmask is set, memory segments of the
2409 corresponding memory type are dumped, otherwise they are not dumped.
2411 The following 4 memory types are supported:
2412 - (bit 0) anonymous private memory
2413 - (bit 1) anonymous shared memory
2414 - (bit 2) file-backed private memory
2415 - (bit 3) file-backed shared memory
2417 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
2418 are always dumped regardless of the bitmask status.
2420 Default value of coredump_filter is 0x3; this means all anonymous memory
2421 segments are dumped.
2423 If you don't want to dump all shared memory segments attached to pid 1234,
2424 write 1 to the process's proc file.
2426 $ echo 0x1 > /proc/1234/coredump_filter
2428 When a new process is created, the process inherits the bitmask status from its
2429 parent. It is useful to set up coredump_filter before the program runs.
2432 $ echo 0x7 > /proc/self/coredump_filter
2435 2.16 /proc/<pid>/mountinfo - Information about mounts
2436 --------------------------------------------------------
2438 This file contains lines of the form:
2440 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
2441 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
2443 (1) mount ID: unique identifier of the mount (may be reused after umount)
2444 (2) parent ID: ID of parent (or of self for the top of the mount tree)
2445 (3) major:minor: value of st_dev for files on filesystem
2446 (4) root: root of the mount within the filesystem
2447 (5) mount point: mount point relative to the process's root
2448 (6) mount options: per mount options
2449 (7) optional fields: zero or more fields of the form "tag[:value]"
2450 (8) separator: marks the end of the optional fields
2451 (9) filesystem type: name of filesystem of the form "type[.subtype]"
2452 (10) mount source: filesystem specific information or "none"
2453 (11) super options: per super block options
2455 Parsers should ignore all unrecognised optional fields. Currently the
2456 possible optional fields are:
2458 shared:X mount is shared in peer group X
2459 master:X mount is slave to peer group X
2460 propagate_from:X mount is slave and receives propagation from peer group X (*)
2461 unbindable mount is unbindable
2463 (*) X is the closest dominant peer group under the process's root. If
2464 X is the immediate master of the mount, or if there's no dominant peer
2465 group under the same root, then only the "master:X" field is present
2466 and not the "propagate_from:X" field.
2468 For more information on mount propagation see:
2470 Documentation/filesystems/sharedsubtree.txt
2472 ------------------------------------------------------------------------------
Code Review / linux-2.6.git / blob