1 ------------------------------------------------------------------------------
2 T H E /proc F I L E S Y S T E M
3 ------------------------------------------------------------------------------
4 /proc/sys Terrehon Bowden <terrehon@pacbell.net> October 7 1999
5 Bodo Bauer <bb@ricochet.net>
7 2.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000
8 move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009
9 ------------------------------------------------------------------------------
10 Version 1.3 Kernel version 2.2.12
11 Kernel version 2.4.0-test11-pre4
12 ------------------------------------------------------------------------------
13 fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009
19 0.1 Introduction/Credits
22 1 Collecting System Information
23 1.1 Process-Specific Subdirectories
25 1.3 IDE devices in /proc/ide
26 1.4 Networking info in /proc/net
28 1.6 Parallel port info in /proc/parport
29 1.7 TTY info in /proc/tty
30 1.8 Miscellaneous kernel statistics in /proc/stat
31 1.9 Ext4 file system parameters
33 2 Modifying System Parameters
35 3 Per-Process Parameters
36 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer
38 3.2 /proc/<pid>/oom_score - Display current oom-killer score
39 3.3 /proc/<pid>/io - Display the IO accounting fields
40 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
41 3.5 /proc/<pid>/mountinfo - Information about mounts
42 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
47 ------------------------------------------------------------------------------
49 ------------------------------------------------------------------------------
51 0.1 Introduction/Credits
52 ------------------------
54 This documentation is part of a soon (or so we hope) to be released book on
55 the SuSE Linux distribution. As there is no complete documentation for the
56 /proc file system and we've used many freely available sources to write these
57 chapters, it seems only fair to give the work back to the Linux community.
58 This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm
59 afraid it's still far from complete, but we hope it will be useful. As far as
60 we know, it is the first 'all-in-one' document about the /proc file system. It
61 is focused on the Intel x86 hardware, so if you are looking for PPC, ARM,
62 SPARC, AXP, etc., features, you probably won't find what you are looking for.
63 It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But
64 additions and patches are welcome and will be added to this document if you
67 We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
68 other people for help compiling this documentation. We'd also like to extend a
69 special thank you to Andi Kleen for documentation, which we relied on heavily
70 to create this document, as well as the additional information he provided.
71 Thanks to everybody else who contributed source or docs to the Linux kernel
72 and helped create a great piece of software... :)
74 If you have any comments, corrections or additions, please don't hesitate to
75 contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this
78 The latest version of this document is available online at
79 http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html
81 If the above direction does not works for you, you could try the kernel
82 mailing list at linux-kernel@vger.kernel.org and/or try to reach me at
83 comandante@zaralinux.com.
88 We don't guarantee the correctness of this document, and if you come to us
89 complaining about how you screwed up your system because of incorrect
90 documentation, we won't feel responsible...
92 ------------------------------------------------------------------------------
93 CHAPTER 1: COLLECTING SYSTEM INFORMATION
94 ------------------------------------------------------------------------------
96 ------------------------------------------------------------------------------
98 ------------------------------------------------------------------------------
99 * Investigating the properties of the pseudo file system /proc and its
100 ability to provide information on the running Linux system
101 * Examining /proc's structure
102 * Uncovering various information about the kernel and the processes running
104 ------------------------------------------------------------------------------
107 The proc file system acts as an interface to internal data structures in the
108 kernel. It can be used to obtain information about the system and to change
109 certain kernel parameters at runtime (sysctl).
111 First, we'll take a look at the read-only parts of /proc. In Chapter 2, we
112 show you how you can use /proc/sys to change settings.
114 1.1 Process-Specific Subdirectories
115 -----------------------------------
117 The directory /proc contains (among other things) one subdirectory for each
118 process running on the system, which is named after the process ID (PID).
120 The link self points to the process reading the file system. Each process
121 subdirectory has the entries listed in Table 1-1.
124 Table 1-1: Process specific entries in /proc
125 ..............................................................................
127 clear_refs Clears page referenced bits shown in smaps output
128 cmdline Command line arguments
129 cpu Current and last cpu in which it was executed (2.4)(smp)
130 cwd Link to the current working directory
131 environ Values of environment variables
132 exe Link to the executable of this process
133 fd Directory, which contains all file descriptors
134 maps Memory maps to executables and library files (2.4)
135 mem Memory held by this process
136 root Link to the root directory of this process
138 statm Process memory status information
139 status Process status in human readable form
140 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan
142 stack Report full stack trace, enable via CONFIG_STACKTRACE
143 smaps a extension based on maps, showing the memory consumption of
145 ..............................................................................
147 For example, to get the status information of a process, all you have to do is
148 read the file /proc/PID/status:
150 >cat /proc/self/status
174 SigPnd: 0000000000000000
175 ShdPnd: 0000000000000000
176 SigBlk: 0000000000000000
177 SigIgn: 0000000000000000
178 SigCgt: 0000000000000000
179 CapInh: 00000000fffffeff
180 CapPrm: 0000000000000000
181 CapEff: 0000000000000000
182 CapBnd: ffffffffffffffff
183 voluntary_ctxt_switches: 0
184 nonvoluntary_ctxt_switches: 1
186 This shows you nearly the same information you would get if you viewed it with
187 the ps command. In fact, ps uses the proc file system to obtain its
188 information. But you get a more detailed view of the process by reading the
189 file /proc/PID/status. It fields are described in table 1-2.
191 The statm file contains more detailed information about the process
192 memory usage. Its seven fields are explained in Table 1-3. The stat file
193 contains details information about the process itself. Its fields are
194 explained in Table 1-4.
196 (for SMP CONFIG users)
197 For making accounting scalable, RSS related information are handled in
198 asynchronous manner and the vaule may not be very precise. To see a precise
199 snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
200 It's slow but very precise.
202 Table 1-2: Contents of the status files (as of 2.6.30-rc7)
203 ..............................................................................
205 Name filename of the executable
206 State state (R is running, S is sleeping, D is sleeping
207 in an uninterruptible wait, Z is zombie,
208 T is traced or stopped)
211 PPid process id of the parent process
212 TracerPid PID of process tracing this process (0 if not)
213 Uid Real, effective, saved set, and file system UIDs
214 Gid Real, effective, saved set, and file system GIDs
215 FDSize number of file descriptor slots currently allocated
216 Groups supplementary group list
217 VmPeak peak virtual memory size
218 VmSize total program size
219 VmLck locked memory size
220 VmHWM peak resident set size ("high water mark")
221 VmRSS size of memory portions
222 VmData size of data, stack, and text segments
223 VmStk size of data, stack, and text segments
224 VmExe size of text segment
225 VmLib size of shared library code
226 VmPTE size of page table entries
227 VmSwap size of swap usage (the number of referred swapents)
228 Threads number of threads
229 SigQ number of signals queued/max. number for queue
230 SigPnd bitmap of pending signals for the thread
231 ShdPnd bitmap of shared pending signals for the process
232 SigBlk bitmap of blocked signals
233 SigIgn bitmap of ignored signals
234 SigCgt bitmap of catched signals
235 CapInh bitmap of inheritable capabilities
236 CapPrm bitmap of permitted capabilities
237 CapEff bitmap of effective capabilities
238 CapBnd bitmap of capabilities bounding set
239 Cpus_allowed mask of CPUs on which this process may run
240 Cpus_allowed_list Same as previous, but in "list format"
241 Mems_allowed mask of memory nodes allowed to this process
242 Mems_allowed_list Same as previous, but in "list format"
243 voluntary_ctxt_switches number of voluntary context switches
244 nonvoluntary_ctxt_switches number of non voluntary context switches
245 ..............................................................................
247 Table 1-3: Contents of the statm files (as of 2.6.8-rc3)
248 ..............................................................................
250 size total program size (pages) (same as VmSize in status)
251 resident size of memory portions (pages) (same as VmRSS in status)
252 shared number of pages that are shared (i.e. backed by a file)
253 trs number of pages that are 'code' (not including libs; broken,
254 includes data segment)
255 lrs number of pages of library (always 0 on 2.6)
256 drs number of pages of data/stack (including libs; broken,
257 includes library text)
258 dt number of dirty pages (always 0 on 2.6)
259 ..............................................................................
262 Table 1-4: Contents of the stat files (as of 2.6.30-rc7)
263 ..............................................................................
266 tcomm filename of the executable
267 state state (R is running, S is sleeping, D is sleeping in an
268 uninterruptible wait, Z is zombie, T is traced or stopped)
269 ppid process id of the parent process
270 pgrp pgrp of the process
272 tty_nr tty the process uses
273 tty_pgrp pgrp of the tty
275 min_flt number of minor faults
276 cmin_flt number of minor faults with child's
277 maj_flt number of major faults
278 cmaj_flt number of major faults with child's
279 utime user mode jiffies
280 stime kernel mode jiffies
281 cutime user mode jiffies with child's
282 cstime kernel mode jiffies with child's
283 priority priority level
285 num_threads number of threads
286 it_real_value (obsolete, always 0)
287 start_time time the process started after system boot
288 vsize virtual memory size
289 rss resident set memory size
290 rsslim current limit in bytes on the rss
291 start_code address above which program text can run
292 end_code address below which program text can run
293 start_stack address of the start of the stack
294 esp current value of ESP
295 eip current value of EIP
296 pending bitmap of pending signals
297 blocked bitmap of blocked signals
298 sigign bitmap of ignored signals
299 sigcatch bitmap of catched signals
300 wchan address where process went to sleep
303 exit_signal signal to send to parent thread on exit
304 task_cpu which CPU the task is scheduled on
305 rt_priority realtime priority
306 policy scheduling policy (man sched_setscheduler)
307 blkio_ticks time spent waiting for block IO
308 gtime guest time of the task in jiffies
309 cgtime guest time of the task children in jiffies
310 ..............................................................................
312 The /proc/PID/maps file containing the currently mapped memory regions and
313 their access permissions.
317 address perms offset dev inode pathname
319 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test
320 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test
321 0804a000-0806b000 rw-p 00000000 00:00 0 [heap]
322 a7cb1000-a7cb2000 ---p 00000000 00:00 0
323 a7cb2000-a7eb2000 rw-p 00000000 00:00 0
324 a7eb2000-a7eb3000 ---p 00000000 00:00 0
325 a7eb3000-a7ed5000 rw-p 00000000 00:00 0
326 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6
327 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6
328 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6
329 a800b000-a800e000 rw-p 00000000 00:00 0
330 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0
331 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0
332 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0
333 a8024000-a8027000 rw-p 00000000 00:00 0
334 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2
335 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2
336 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2
337 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack]
338 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso]
340 where "address" is the address space in the process that it occupies, "perms"
341 is a set of permissions:
347 p = private (copy on write)
349 "offset" is the offset into the mapping, "dev" is the device (major:minor), and
350 "inode" is the inode on that device. 0 indicates that no inode is associated
351 with the memory region, as the case would be with BSS (uninitialized data).
352 The "pathname" shows the name associated file for this mapping. If the mapping
353 is not associated with a file:
355 [heap] = the heap of the program
356 [stack] = the stack of the main process
357 [vdso] = the "virtual dynamic shared object",
358 the kernel system call handler
360 or if empty, the mapping is anonymous.
363 The /proc/PID/smaps is an extension based on maps, showing the memory
364 consumption for each of the process's mappings. For each of mappings there
365 is a series of lines such as the following:
367 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
382 The first of these lines shows the same information as is displayed for the
383 mapping in /proc/PID/maps. The remaining lines show the size of the mapping
384 (size), the amount of the mapping that is currently resident in RAM (RSS), the
385 process' proportional share of this mapping (PSS), the number of clean and
386 dirty private pages in the mapping. Note that even a page which is part of a
387 MAP_SHARED mapping, but has only a single pte mapped, i.e. is currently used
388 by only one process, is accounted as private and not as shared. "Referenced"
389 indicates the amount of memory currently marked as referenced or accessed.
390 "Anonymous" shows the amount of memory that does not belong to any file. Even
391 a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
392 and a page is modified, the file page is replaced by a private anonymous copy.
393 "Swap" shows how much would-be-anonymous memory is also used, but out on
396 This file is only present if the CONFIG_MMU kernel configuration option is
399 The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
400 bits on both physical and virtual pages associated with a process.
401 To clear the bits for all the pages associated with the process
402 > echo 1 > /proc/PID/clear_refs
404 To clear the bits for the anonymous pages associated with the process
405 > echo 2 > /proc/PID/clear_refs
407 To clear the bits for the file mapped pages associated with the process
408 > echo 3 > /proc/PID/clear_refs
409 Any other value written to /proc/PID/clear_refs will have no effect.
411 The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
412 using /proc/kpageflags and number of times a page is mapped using
413 /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt.
418 Similar to the process entries, the kernel data files give information about
419 the running kernel. The files used to obtain this information are contained in
420 /proc and are listed in Table 1-5. Not all of these will be present in your
421 system. It depends on the kernel configuration and the loaded modules, which
422 files are there, and which are missing.
424 Table 1-5: Kernel info in /proc
425 ..............................................................................
427 apm Advanced power management info
428 buddyinfo Kernel memory allocator information (see text) (2.5)
429 bus Directory containing bus specific information
430 cmdline Kernel command line
431 cpuinfo Info about the CPU
432 devices Available devices (block and character)
433 dma Used DMS channels
434 filesystems Supported filesystems
435 driver Various drivers grouped here, currently rtc (2.4)
436 execdomains Execdomains, related to security (2.4)
437 fb Frame Buffer devices (2.4)
438 fs File system parameters, currently nfs/exports (2.4)
439 ide Directory containing info about the IDE subsystem
440 interrupts Interrupt usage
441 iomem Memory map (2.4)
442 ioports I/O port usage
443 irq Masks for irq to cpu affinity (2.4)(smp?)
444 isapnp ISA PnP (Plug&Play) Info (2.4)
445 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
447 ksyms Kernel symbol table
448 loadavg Load average of last 1, 5 & 15 minutes
452 modules List of loaded modules
453 mounts Mounted filesystems
454 net Networking info (see text)
455 pagetypeinfo Additional page allocator information (see text) (2.5)
456 partitions Table of partitions known to the system
457 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/,
458 decoupled by lspci (2.4)
460 scsi SCSI info (see text)
461 slabinfo Slab pool info
462 softirqs softirq usage
463 stat Overall statistics
464 swaps Swap space utilization
466 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4)
467 tty Info of tty drivers
469 version Kernel version
470 video bttv info of video resources (2.4)
471 vmallocinfo Show vmalloced areas
472 ..............................................................................
474 You can, for example, check which interrupts are currently in use and what
475 they are used for by looking in the file /proc/interrupts:
477 > cat /proc/interrupts
479 0: 8728810 XT-PIC timer
480 1: 895 XT-PIC keyboard
482 3: 531695 XT-PIC aha152x
483 4: 2014133 XT-PIC serial
484 5: 44401 XT-PIC pcnet_cs
487 12: 182918 XT-PIC PS/2 Mouse
489 14: 1232265 XT-PIC ide0
493 In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
494 output of a SMP machine):
496 > cat /proc/interrupts
499 0: 1243498 1214548 IO-APIC-edge timer
500 1: 8949 8958 IO-APIC-edge keyboard
501 2: 0 0 XT-PIC cascade
502 5: 11286 10161 IO-APIC-edge soundblaster
503 8: 1 0 IO-APIC-edge rtc
504 9: 27422 27407 IO-APIC-edge 3c503
505 12: 113645 113873 IO-APIC-edge PS/2 Mouse
507 14: 22491 24012 IO-APIC-edge ide0
508 15: 2183 2415 IO-APIC-edge ide1
509 17: 30564 30414 IO-APIC-level eth0
510 18: 177 164 IO-APIC-level bttv
515 NMI is incremented in this case because every timer interrupt generates a NMI
516 (Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.
518 LOC is the local interrupt counter of the internal APIC of every CPU.
520 ERR is incremented in the case of errors in the IO-APIC bus (the bus that
521 connects the CPUs in a SMP system. This means that an error has been detected,
522 the IO-APIC automatically retry the transmission, so it should not be a big
523 problem, but you should read the SMP-FAQ.
525 In 2.6.2* /proc/interrupts was expanded again. This time the goal was for
526 /proc/interrupts to display every IRQ vector in use by the system, not
527 just those considered 'most important'. The new vectors are:
529 THR -- interrupt raised when a machine check threshold counter
530 (typically counting ECC corrected errors of memory or cache) exceeds
531 a configurable threshold. Only available on some systems.
533 TRM -- a thermal event interrupt occurs when a temperature threshold
534 has been exceeded for the CPU. This interrupt may also be generated
535 when the temperature drops back to normal.
537 SPU -- a spurious interrupt is some interrupt that was raised then lowered
538 by some IO device before it could be fully processed by the APIC. Hence
539 the APIC sees the interrupt but does not know what device it came from.
540 For this case the APIC will generate the interrupt with a IRQ vector
541 of 0xff. This might also be generated by chipset bugs.
543 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are
544 sent from one CPU to another per the needs of the OS. Typically,
545 their statistics are used by kernel developers and interested users to
546 determine the occurrence of interrupts of the given type.
548 The above IRQ vectors are displayed only when relevant. For example,
549 the threshold vector does not exist on x86_64 platforms. Others are
550 suppressed when the system is a uniprocessor. As of this writing, only
551 i386 and x86_64 platforms support the new IRQ vector displays.
553 Of some interest is the introduction of the /proc/irq directory to 2.4.
554 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
555 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
556 irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
561 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask
562 1 11 13 15 17 19 3 5 7 9 default_smp_affinity
566 smp_affinity is a bitmask, in which you can specify which CPUs can handle the
567 IRQ, you can set it by doing:
569 > echo 1 > /proc/irq/10/smp_affinity
571 This means that only the first CPU will handle the IRQ, but you can also echo
572 5 which means that only the first and fourth CPU can handle the IRQ.
574 The contents of each smp_affinity file is the same by default:
576 > cat /proc/irq/0/smp_affinity
579 There is an alternate interface, smp_affinity_list which allows specifying
580 a cpu range instead of a bitmask:
582 > cat /proc/irq/0/smp_affinity_list
585 The default_smp_affinity mask applies to all non-active IRQs, which are the
586 IRQs which have not yet been allocated/activated, and hence which lack a
587 /proc/irq/[0-9]* directory.
589 The node file on an SMP system shows the node to which the device using the IRQ
590 reports itself as being attached. This hardware locality information does not
591 include information about any possible driver locality preference.
593 prof_cpu_mask specifies which CPUs are to be profiled by the system wide
594 profiler. Default value is ffffffff (all cpus if there are only 32 of them).
596 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
597 between all the CPUs which are allowed to handle it. As usual the kernel has
598 more info than you and does a better job than you, so the defaults are the
599 best choice for almost everyone. [Note this applies only to those IO-APIC's
600 that support "Round Robin" interrupt distribution.]
602 There are three more important subdirectories in /proc: net, scsi, and sys.
603 The general rule is that the contents, or even the existence of these
604 directories, depend on your kernel configuration. If SCSI is not enabled, the
605 directory scsi may not exist. The same is true with the net, which is there
606 only when networking support is present in the running kernel.
608 The slabinfo file gives information about memory usage at the slab level.
609 Linux uses slab pools for memory management above page level in version 2.2.
610 Commonly used objects have their own slab pool (such as network buffers,
611 directory cache, and so on).
613 ..............................................................................
615 > cat /proc/buddyinfo
617 Node 0, zone DMA 0 4 5 4 4 3 ...
618 Node 0, zone Normal 1 0 0 1 101 8 ...
619 Node 0, zone HighMem 2 0 0 1 1 0 ...
621 External fragmentation is a problem under some workloads, and buddyinfo is a
622 useful tool for helping diagnose these problems. Buddyinfo will give you a
623 clue as to how big an area you can safely allocate, or why a previous
626 Each column represents the number of pages of a certain order which are
627 available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
628 ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
629 available in ZONE_NORMAL, etc...
631 More information relevant to external fragmentation can be found in
634 > cat /proc/pagetypeinfo
638 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
639 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0
640 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0
641 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2
642 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0
643 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0
644 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9
645 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0
646 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452
647 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0
648 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0
650 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate
651 Node 0, zone DMA 2 0 5 1 0
652 Node 0, zone DMA32 41 6 967 2 0
654 Fragmentation avoidance in the kernel works by grouping pages of different
655 migrate types into the same contiguous regions of memory called page blocks.
656 A page block is typically the size of the default hugepage size e.g. 2MB on
657 X86-64. By keeping pages grouped based on their ability to move, the kernel
658 can reclaim pages within a page block to satisfy a high-order allocation.
660 The pagetypinfo begins with information on the size of a page block. It
661 then gives the same type of information as buddyinfo except broken down
662 by migrate-type and finishes with details on how many page blocks of each
665 If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
666 from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can
667 make an estimate of the likely number of huge pages that can be allocated
668 at a given point in time. All the "Movable" blocks should be allocatable
669 unless memory has been mlock()'d. Some of the Reclaimable blocks should
670 also be allocatable although a lot of filesystem metadata may have to be
671 reclaimed to achieve this.
673 ..............................................................................
677 Provides information about distribution and utilization of memory. This
678 varies by architecture and compile options. The following is from a
679 16GB PIII, which has highmem enabled. You may not have all of these fields.
683 The "Locked" indicates whether the mapping is locked in memory or not.
686 MemTotal: 16344972 kB
693 HighTotal: 15597528 kB
694 HighFree: 13629632 kB
704 SReclaimable: 159856 kB
705 SUnreclaim: 124508 kB
710 CommitLimit: 7669796 kB
711 Committed_AS: 100056 kB
712 VmallocTotal: 112216 kB
714 VmallocChunk: 111088 kB
716 MemTotal: Total usable ram (i.e. physical ram minus a few reserved
717 bits and the kernel binary code)
718 MemFree: The sum of LowFree+HighFree
719 Buffers: Relatively temporary storage for raw disk blocks
720 shouldn't get tremendously large (20MB or so)
721 Cached: in-memory cache for files read from the disk (the
722 pagecache). Doesn't include SwapCached
723 SwapCached: Memory that once was swapped out, is swapped back in but
724 still also is in the swapfile (if memory is needed it
725 doesn't need to be swapped out AGAIN because it is already
726 in the swapfile. This saves I/O)
727 Active: Memory that has been used more recently and usually not
728 reclaimed unless absolutely necessary.
729 Inactive: Memory which has been less recently used. It is more
730 eligible to be reclaimed for other purposes
732 HighFree: Highmem is all memory above ~860MB of physical memory
733 Highmem areas are for use by userspace programs, or
734 for the pagecache. The kernel must use tricks to access
735 this memory, making it slower to access than lowmem.
737 LowFree: Lowmem is memory which can be used for everything that
738 highmem can be used for, but it is also available for the
739 kernel's use for its own data structures. Among many
740 other things, it is where everything from the Slab is
741 allocated. Bad things happen when you're out of lowmem.
742 SwapTotal: total amount of swap space available
743 SwapFree: Memory which has been evicted from RAM, and is temporarily
745 Dirty: Memory which is waiting to get written back to the disk
746 Writeback: Memory which is actively being written back to the disk
747 AnonPages: Non-file backed pages mapped into userspace page tables
748 Mapped: files which have been mmaped, such as libraries
749 Slab: in-kernel data structures cache
750 SReclaimable: Part of Slab, that might be reclaimed, such as caches
751 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure
752 PageTables: amount of memory dedicated to the lowest level of page
754 NFS_Unstable: NFS pages sent to the server, but not yet committed to stable
756 Bounce: Memory used for block device "bounce buffers"
757 WritebackTmp: Memory used by FUSE for temporary writeback buffers
758 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'),
759 this is the total amount of memory currently available to
760 be allocated on the system. This limit is only adhered to
761 if strict overcommit accounting is enabled (mode 2 in
762 'vm.overcommit_memory').
763 The CommitLimit is calculated with the following formula:
764 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap
765 For example, on a system with 1G of physical RAM and 7G
766 of swap with a `vm.overcommit_ratio` of 30 it would
767 yield a CommitLimit of 7.3G.
768 For more details, see the memory overcommit documentation
769 in vm/overcommit-accounting.
770 Committed_AS: The amount of memory presently allocated on the system.
771 The committed memory is a sum of all of the memory which
772 has been allocated by processes, even if it has not been
773 "used" by them as of yet. A process which malloc()'s 1G
774 of memory, but only touches 300M of it will only show up
775 as using 300M of memory even if it has the address space
776 allocated for the entire 1G. This 1G is memory which has
777 been "committed" to by the VM and can be used at any time
778 by the allocating application. With strict overcommit
779 enabled on the system (mode 2 in 'vm.overcommit_memory'),
780 allocations which would exceed the CommitLimit (detailed
781 above) will not be permitted. This is useful if one needs
782 to guarantee that processes will not fail due to lack of
783 memory once that memory has been successfully allocated.
784 VmallocTotal: total size of vmalloc memory area
785 VmallocUsed: amount of vmalloc area which is used
786 VmallocChunk: largest contiguous block of vmalloc area which is free
788 ..............................................................................
792 Provides information about vmalloced/vmaped areas. One line per area,
793 containing the virtual address range of the area, size in bytes,
794 caller information of the creator, and optional information depending
795 on the kind of area :
797 pages=nr number of pages
798 phys=addr if a physical address was specified
799 ioremap I/O mapping (ioremap() and friends)
800 vmalloc vmalloc() area
803 vpages buffer for pages pointers was vmalloced (huge area)
804 N<node>=nr (Only on NUMA kernels)
805 Number of pages allocated on memory node <node>
807 > cat /proc/vmallocinfo
808 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
809 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
810 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
811 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
812 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f...
813 phys=7fee8000 ioremap
814 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f...
815 phys=7fee7000 ioremap
816 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210
817 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ...
818 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
819 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ...
821 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ...
822 /0x130 [x_tables] pages=4 vmalloc N0=4
823 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ...
824 pages=14 vmalloc N2=14
825 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ...
827 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ...
829 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ...
830 pages=10 vmalloc N0=10
832 ..............................................................................
836 Provides counts of softirq handlers serviced since boot time, for each cpu.
841 TIMER: 27166 27120 27097 27034
846 SCHED: 27035 26983 26971 26746
848 RCU: 1678 1769 2178 2250
851 1.3 IDE devices in /proc/ide
852 ----------------------------
854 The subdirectory /proc/ide contains information about all IDE devices of which
855 the kernel is aware. There is one subdirectory for each IDE controller, the
856 file drivers and a link for each IDE device, pointing to the device directory
857 in the controller specific subtree.
859 The file drivers contains general information about the drivers used for the
862 > cat /proc/ide/drivers
863 ide-cdrom version 4.53
864 ide-disk version 1.08
866 More detailed information can be found in the controller specific
867 subdirectories. These are named ide0, ide1 and so on. Each of these
868 directories contains the files shown in table 1-6.
871 Table 1-6: IDE controller info in /proc/ide/ide?
872 ..............................................................................
874 channel IDE channel (0 or 1)
875 config Configuration (only for PCI/IDE bridge)
877 model Type/Chipset of IDE controller
878 ..............................................................................
880 Each device connected to a controller has a separate subdirectory in the
881 controllers directory. The files listed in table 1-7 are contained in these
885 Table 1-7: IDE device information
886 ..............................................................................
889 capacity Capacity of the medium (in 512Byte blocks)
890 driver driver and version
891 geometry physical and logical geometry
892 identify device identify block
894 model device identifier
895 settings device setup
896 smart_thresholds IDE disk management thresholds
897 smart_values IDE disk management values
898 ..............................................................................
900 The most interesting file is settings. This file contains a nice overview of
901 the drive parameters:
903 # cat /proc/ide/ide0/hda/settings
904 name value min max mode
905 ---- ----- --- --- ----
906 bios_cyl 526 0 65535 rw
907 bios_head 255 0 255 rw
909 breada_readahead 4 0 127 rw
911 file_readahead 72 0 2097151 rw
913 keepsettings 0 0 1 rw
914 max_kb_per_request 122 1 127 rw
918 pio_mode write-only 0 255 w
924 1.4 Networking info in /proc/net
925 --------------------------------
927 The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the
928 additional values you get for IP version 6 if you configure the kernel to
929 support this. Table 1-9 lists the files and their meaning.
932 Table 1-8: IPv6 info in /proc/net
933 ..............................................................................
935 udp6 UDP sockets (IPv6)
936 tcp6 TCP sockets (IPv6)
937 raw6 Raw device statistics (IPv6)
938 igmp6 IP multicast addresses, which this host joined (IPv6)
939 if_inet6 List of IPv6 interface addresses
940 ipv6_route Kernel routing table for IPv6
941 rt6_stats Global IPv6 routing tables statistics
942 sockstat6 Socket statistics (IPv6)
943 snmp6 Snmp data (IPv6)
944 ..............................................................................
947 Table 1-9: Network info in /proc/net
948 ..............................................................................
951 dev network devices with statistics
952 dev_mcast the Layer2 multicast groups a device is listening too
953 (interface index, label, number of references, number of bound
955 dev_stat network device status
956 ip_fwchains Firewall chain linkage
957 ip_fwnames Firewall chain names
958 ip_masq Directory containing the masquerading tables
959 ip_masquerade Major masquerading table
960 netstat Network statistics
961 raw raw device statistics
962 route Kernel routing table
963 rpc Directory containing rpc info
964 rt_cache Routing cache
966 sockstat Socket statistics
968 tr_rif Token ring RIF routing table
970 unix UNIX domain sockets
971 wireless Wireless interface data (Wavelan etc)
972 igmp IP multicast addresses, which this host joined
973 psched Global packet scheduler parameters.
974 netlink List of PF_NETLINK sockets
975 ip_mr_vifs List of multicast virtual interfaces
976 ip_mr_cache List of multicast routing cache
977 ..............................................................................
979 You can use this information to see which network devices are available in
980 your system and how much traffic was routed over those devices:
984 face |bytes packets errs drop fifo frame compressed multicast|[...
985 lo: 908188 5596 0 0 0 0 0 0 [...
986 ppp0:15475140 20721 410 0 0 410 0 0 [...
987 eth0: 614530 7085 0 0 0 0 0 1 [...
990 ...] bytes packets errs drop fifo colls carrier compressed
991 ...] 908188 5596 0 0 0 0 0 0
992 ...] 1375103 17405 0 0 0 0 0 0
993 ...] 1703981 5535 0 0 0 3 0 0
995 In addition, each Channel Bond interface has its own directory. For
996 example, the bond0 device will have a directory called /proc/net/bond0/.
997 It will contain information that is specific to that bond, such as the
998 current slaves of the bond, the link status of the slaves, and how
999 many times the slaves link has failed.
1004 If you have a SCSI host adapter in your system, you'll find a subdirectory
1005 named after the driver for this adapter in /proc/scsi. You'll also see a list
1006 of all recognized SCSI devices in /proc/scsi:
1008 >cat /proc/scsi/scsi
1010 Host: scsi0 Channel: 00 Id: 00 Lun: 00
1011 Vendor: IBM Model: DGHS09U Rev: 03E0
1012 Type: Direct-Access ANSI SCSI revision: 03
1013 Host: scsi0 Channel: 00 Id: 06 Lun: 00
1014 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04
1015 Type: CD-ROM ANSI SCSI revision: 02
1018 The directory named after the driver has one file for each adapter found in
1019 the system. These files contain information about the controller, including
1020 the used IRQ and the IO address range. The amount of information shown is
1021 dependent on the adapter you use. The example shows the output for an Adaptec
1022 AHA-2940 SCSI adapter:
1024 > cat /proc/scsi/aic7xxx/0
1026 Adaptec AIC7xxx driver version: 5.1.19/3.2.4
1028 TCQ Enabled By Default : Disabled
1029 AIC7XXX_PROC_STATS : Disabled
1030 AIC7XXX_RESET_DELAY : 5
1031 Adapter Configuration:
1032 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter
1033 Ultra Wide Controller
1034 PCI MMAPed I/O Base: 0xeb001000
1035 Adapter SEEPROM Config: SEEPROM found and used.
1036 Adaptec SCSI BIOS: Enabled
1038 SCBs: Active 0, Max Active 2,
1039 Allocated 15, HW 16, Page 255
1041 BIOS Control Word: 0x18b6
1042 Adapter Control Word: 0x005b
1043 Extended Translation: Enabled
1044 Disconnect Enable Flags: 0xffff
1045 Ultra Enable Flags: 0x0001
1046 Tag Queue Enable Flags: 0x0000
1047 Ordered Queue Tag Flags: 0x0000
1048 Default Tag Queue Depth: 8
1049 Tagged Queue By Device array for aic7xxx host instance 0:
1050 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
1051 Actual queue depth per device for aic7xxx host instance 0:
1052 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}
1055 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8
1056 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0)
1057 Total transfers 160151 (74577 reads and 85574 writes)
1059 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15
1060 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0)
1061 Total transfers 0 (0 reads and 0 writes)
1064 1.6 Parallel port info in /proc/parport
1065 ---------------------------------------
1067 The directory /proc/parport contains information about the parallel ports of
1068 your system. It has one subdirectory for each port, named after the port
1071 These directories contain the four files shown in Table 1-10.
1074 Table 1-10: Files in /proc/parport
1075 ..............................................................................
1077 autoprobe Any IEEE-1284 device ID information that has been acquired.
1078 devices list of the device drivers using that port. A + will appear by the
1079 name of the device currently using the port (it might not appear
1081 hardware Parallel port's base address, IRQ line and DMA channel.
1082 irq IRQ that parport is using for that port. This is in a separate
1083 file to allow you to alter it by writing a new value in (IRQ
1085 ..............................................................................
1087 1.7 TTY info in /proc/tty
1088 -------------------------
1090 Information about the available and actually used tty's can be found in the
1091 directory /proc/tty.You'll find entries for drivers and line disciplines in
1092 this directory, as shown in Table 1-11.
1095 Table 1-11: Files in /proc/tty
1096 ..............................................................................
1098 drivers list of drivers and their usage
1099 ldiscs registered line disciplines
1100 driver/serial usage statistic and status of single tty lines
1101 ..............................................................................
1103 To see which tty's are currently in use, you can simply look into the file
1106 > cat /proc/tty/drivers
1107 pty_slave /dev/pts 136 0-255 pty:slave
1108 pty_master /dev/ptm 128 0-255 pty:master
1109 pty_slave /dev/ttyp 3 0-255 pty:slave
1110 pty_master /dev/pty 2 0-255 pty:master
1111 serial /dev/cua 5 64-67 serial:callout
1112 serial /dev/ttyS 4 64-67 serial
1113 /dev/tty0 /dev/tty0 4 0 system:vtmaster
1114 /dev/ptmx /dev/ptmx 5 2 system
1115 /dev/console /dev/console 5 1 system:console
1116 /dev/tty /dev/tty 5 0 system:/dev/tty
1117 unknown /dev/tty 4 1-63 console
1120 1.8 Miscellaneous kernel statistics in /proc/stat
1121 -------------------------------------------------
1123 Various pieces of information about kernel activity are available in the
1124 /proc/stat file. All of the numbers reported in this file are aggregates
1125 since the system first booted. For a quick look, simply cat the file:
1128 cpu 2255 34 2290 22625563 6290 127 456 0 0
1129 cpu0 1132 34 1441 11311718 3675 127 438 0 0
1130 cpu1 1123 0 849 11313845 2614 0 18 0 0
1131 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...]
1137 softirq 183433 0 21755 12 39 1137 231 21459 2263
1139 The very first "cpu" line aggregates the numbers in all of the other "cpuN"
1140 lines. These numbers identify the amount of time the CPU has spent performing
1141 different kinds of work. Time units are in USER_HZ (typically hundredths of a
1142 second). The meanings of the columns are as follows, from left to right:
1144 - user: normal processes executing in user mode
1145 - nice: niced processes executing in user mode
1146 - system: processes executing in kernel mode
1147 - idle: twiddling thumbs
1148 - iowait: waiting for I/O to complete
1149 - irq: servicing interrupts
1150 - softirq: servicing softirqs
1151 - steal: involuntary wait
1152 - guest: running a normal guest
1153 - guest_nice: running a niced guest
1155 The "intr" line gives counts of interrupts serviced since boot time, for each
1156 of the possible system interrupts. The first column is the total of all
1157 interrupts serviced; each subsequent column is the total for that particular
1160 The "ctxt" line gives the total number of context switches across all CPUs.
1162 The "btime" line gives the time at which the system booted, in seconds since
1165 The "processes" line gives the number of processes and threads created, which
1166 includes (but is not limited to) those created by calls to the fork() and
1167 clone() system calls.
1169 The "procs_running" line gives the total number of threads that are
1170 running or ready to run (i.e., the total number of runnable threads).
1172 The "procs_blocked" line gives the number of processes currently blocked,
1173 waiting for I/O to complete.
1175 The "softirq" line gives counts of softirqs serviced since boot time, for each
1176 of the possible system softirqs. The first column is the total of all
1177 softirqs serviced; each subsequent column is the total for that particular
1181 1.9 Ext4 file system parameters
1182 ------------------------------
1184 Information about mounted ext4 file systems can be found in
1185 /proc/fs/ext4. Each mounted filesystem will have a directory in
1186 /proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
1187 /proc/fs/ext4/dm-0). The files in each per-device directory are shown
1188 in Table 1-12, below.
1190 Table 1-12: Files in /proc/fs/ext4/<devname>
1191 ..............................................................................
1193 mb_groups details of multiblock allocator buddy cache of free blocks
1194 ..............................................................................
1198 Shows registered system console lines.
1200 To see which character device lines are currently used for the system console
1201 /dev/console, you may simply look into the file /proc/consoles:
1203 > cat /proc/consoles
1209 device name of the device
1210 operations R = can do read operations
1211 W = can do write operations
1213 flags E = it is enabled
1214 C = it is preferred console
1215 B = it is primary boot console
1216 p = it is used for printk buffer
1217 b = it is not a TTY but a Braille device
1218 a = it is safe to use when cpu is offline
1219 major:minor major and minor number of the device separated by a colon
1221 ------------------------------------------------------------------------------
1223 ------------------------------------------------------------------------------
1224 The /proc file system serves information about the running system. It not only
1225 allows access to process data but also allows you to request the kernel status
1226 by reading files in the hierarchy.
1228 The directory structure of /proc reflects the types of information and makes
1229 it easy, if not obvious, where to look for specific data.
1230 ------------------------------------------------------------------------------
1232 ------------------------------------------------------------------------------
1233 CHAPTER 2: MODIFYING SYSTEM PARAMETERS
1234 ------------------------------------------------------------------------------
1236 ------------------------------------------------------------------------------
1238 ------------------------------------------------------------------------------
1239 * Modifying kernel parameters by writing into files found in /proc/sys
1240 * Exploring the files which modify certain parameters
1241 * Review of the /proc/sys file tree
1242 ------------------------------------------------------------------------------
1245 A very interesting part of /proc is the directory /proc/sys. This is not only
1246 a source of information, it also allows you to change parameters within the
1247 kernel. Be very careful when attempting this. You can optimize your system,
1248 but you can also cause it to crash. Never alter kernel parameters on a
1249 production system. Set up a development machine and test to make sure that
1250 everything works the way you want it to. You may have no alternative but to
1251 reboot the machine once an error has been made.
1253 To change a value, simply echo the new value into the file. An example is
1254 given below in the section on the file system data. You need to be root to do
1255 this. You can create your own boot script to perform this every time your
1258 The files in /proc/sys can be used to fine tune and monitor miscellaneous and
1259 general things in the operation of the Linux kernel. Since some of the files
1260 can inadvertently disrupt your system, it is advisable to read both
1261 documentation and source before actually making adjustments. In any case, be
1262 very careful when writing to any of these files. The entries in /proc may
1263 change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt
1264 review the kernel documentation in the directory /usr/src/linux/Documentation.
1265 This chapter is heavily based on the documentation included in the pre 2.2
1266 kernels, and became part of it in version 2.2.1 of the Linux kernel.
1268 Please see: Documentation/sysctl/ directory for descriptions of these
1271 ------------------------------------------------------------------------------
1273 ------------------------------------------------------------------------------
1274 Certain aspects of kernel behavior can be modified at runtime, without the
1275 need to recompile the kernel, or even to reboot the system. The files in the
1276 /proc/sys tree can not only be read, but also modified. You can use the echo
1277 command to write value into these files, thereby changing the default settings
1279 ------------------------------------------------------------------------------
1281 ------------------------------------------------------------------------------
1282 CHAPTER 3: PER-PROCESS PARAMETERS
1283 ------------------------------------------------------------------------------
1285 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score
1286 --------------------------------------------------------------------------------
1288 These file can be used to adjust the badness heuristic used to select which
1289 process gets killed in out of memory conditions.
1291 The badness heuristic assigns a value to each candidate task ranging from 0
1292 (never kill) to 1000 (always kill) to determine which process is targeted. The
1293 units are roughly a proportion along that range of allowed memory the process
1294 may allocate from based on an estimation of its current memory and swap use.
1295 For example, if a task is using all allowed memory, its badness score will be
1296 1000. If it is using half of its allowed memory, its score will be 500.
1298 There is an additional factor included in the badness score: root
1299 processes are given 3% extra memory over other tasks.
1301 The amount of "allowed" memory depends on the context in which the oom killer
1302 was called. If it is due to the memory assigned to the allocating task's cpuset
1303 being exhausted, the allowed memory represents the set of mems assigned to that
1304 cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed
1305 memory represents the set of mempolicy nodes. If it is due to a memory
1306 limit (or swap limit) being reached, the allowed memory is that configured
1307 limit. Finally, if it is due to the entire system being out of memory, the
1308 allowed memory represents all allocatable resources.
1310 The value of /proc/<pid>/oom_score_adj is added to the badness score before it
1311 is used to determine which task to kill. Acceptable values range from -1000
1312 (OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to
1313 polarize the preference for oom killing either by always preferring a certain
1314 task or completely disabling it. The lowest possible value, -1000, is
1315 equivalent to disabling oom killing entirely for that task since it will always
1316 report a badness score of 0.
1318 Consequently, it is very simple for userspace to define the amount of memory to
1319 consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for
1320 example, is roughly equivalent to allowing the remainder of tasks sharing the
1321 same system, cpuset, mempolicy, or memory controller resources to use at least
1322 50% more memory. A value of -500, on the other hand, would be roughly
1323 equivalent to discounting 50% of the task's allowed memory from being considered
1324 as scoring against the task.
1326 For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also
1327 be used to tune the badness score. Its acceptable values range from -16
1328 (OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17
1329 (OOM_DISABLE) to disable oom killing entirely for that task. Its value is
1330 scaled linearly with /proc/<pid>/oom_score_adj.
1332 Writing to /proc/<pid>/oom_score_adj or /proc/<pid>/oom_adj will change the
1333 other with its scaled value.
1335 The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
1336 value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
1337 requires CAP_SYS_RESOURCE.
1339 NOTICE: /proc/<pid>/oom_adj is deprecated and will be removed, please see
1340 Documentation/feature-removal-schedule.txt.
1342 Caveat: when a parent task is selected, the oom killer will sacrifice any first
1343 generation children with separate address spaces instead, if possible. This
1344 avoids servers and important system daemons from being killed and loses the
1345 minimal amount of work.
1348 3.2 /proc/<pid>/oom_score - Display current oom-killer score
1349 -------------------------------------------------------------
1351 This file can be used to check the current score used by the oom-killer is for
1352 any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which
1353 process should be killed in an out-of-memory situation.
1356 3.3 /proc/<pid>/io - Display the IO accounting fields
1357 -------------------------------------------------------
1359 This file contains IO statistics for each running process
1364 test:/tmp # dd if=/dev/zero of=/tmp/test.dat &
1367 test:/tmp # cat /proc/3828/io
1373 write_bytes: 323932160
1374 cancelled_write_bytes: 0
1383 I/O counter: chars read
1384 The number of bytes which this task has caused to be read from storage. This
1385 is simply the sum of bytes which this process passed to read() and pread().
1386 It includes things like tty IO and it is unaffected by whether or not actual
1387 physical disk IO was required (the read might have been satisfied from
1394 I/O counter: chars written
1395 The number of bytes which this task has caused, or shall cause to be written
1396 to disk. Similar caveats apply here as with rchar.
1402 I/O counter: read syscalls
1403 Attempt to count the number of read I/O operations, i.e. syscalls like read()
1410 I/O counter: write syscalls
1411 Attempt to count the number of write I/O operations, i.e. syscalls like
1412 write() and pwrite().
1418 I/O counter: bytes read
1419 Attempt to count the number of bytes which this process really did cause to
1420 be fetched from the storage layer. Done at the submit_bio() level, so it is
1421 accurate for block-backed filesystems. <please add status regarding NFS and
1422 CIFS at a later time>
1428 I/O counter: bytes written
1429 Attempt to count the number of bytes which this process caused to be sent to
1430 the storage layer. This is done at page-dirtying time.
1433 cancelled_write_bytes
1434 ---------------------
1436 The big inaccuracy here is truncate. If a process writes 1MB to a file and
1437 then deletes the file, it will in fact perform no writeout. But it will have
1438 been accounted as having caused 1MB of write.
1439 In other words: The number of bytes which this process caused to not happen,
1440 by truncating pagecache. A task can cause "negative" IO too. If this task
1441 truncates some dirty pagecache, some IO which another task has been accounted
1442 for (in its write_bytes) will not be happening. We _could_ just subtract that
1443 from the truncating task's write_bytes, but there is information loss in doing
1450 At its current implementation state, this is a bit racy on 32-bit machines: if
1451 process A reads process B's /proc/pid/io while process B is updating one of
1452 those 64-bit counters, process A could see an intermediate result.
1455 More information about this can be found within the taskstats documentation in
1456 Documentation/accounting.
1458 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings
1459 ---------------------------------------------------------------
1460 When a process is dumped, all anonymous memory is written to a core file as
1461 long as the size of the core file isn't limited. But sometimes we don't want
1462 to dump some memory segments, for example, huge shared memory. Conversely,
1463 sometimes we want to save file-backed memory segments into a core file, not
1464 only the individual files.
1466 /proc/<pid>/coredump_filter allows you to customize which memory segments
1467 will be dumped when the <pid> process is dumped. coredump_filter is a bitmask
1468 of memory types. If a bit of the bitmask is set, memory segments of the
1469 corresponding memory type are dumped, otherwise they are not dumped.
1471 The following 7 memory types are supported:
1472 - (bit 0) anonymous private memory
1473 - (bit 1) anonymous shared memory
1474 - (bit 2) file-backed private memory
1475 - (bit 3) file-backed shared memory
1476 - (bit 4) ELF header pages in file-backed private memory areas (it is
1477 effective only if the bit 2 is cleared)
1478 - (bit 5) hugetlb private memory
1479 - (bit 6) hugetlb shared memory
1481 Note that MMIO pages such as frame buffer are never dumped and vDSO pages
1482 are always dumped regardless of the bitmask status.
1484 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only
1485 effected by bit 5-6.
1487 Default value of coredump_filter is 0x23; this means all anonymous memory
1488 segments and hugetlb private memory are dumped.
1490 If you don't want to dump all shared memory segments attached to pid 1234,
1491 write 0x21 to the process's proc file.
1493 $ echo 0x21 > /proc/1234/coredump_filter
1495 When a new process is created, the process inherits the bitmask status from its
1496 parent. It is useful to set up coredump_filter before the program runs.
1499 $ echo 0x7 > /proc/self/coredump_filter
1502 3.5 /proc/<pid>/mountinfo - Information about mounts
1503 --------------------------------------------------------
1505 This file contains lines of the form:
1507 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
1508 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
1510 (1) mount ID: unique identifier of the mount (may be reused after umount)
1511 (2) parent ID: ID of parent (or of self for the top of the mount tree)
1512 (3) major:minor: value of st_dev for files on filesystem
1513 (4) root: root of the mount within the filesystem
1514 (5) mount point: mount point relative to the process's root
1515 (6) mount options: per mount options
1516 (7) optional fields: zero or more fields of the form "tag[:value]"
1517 (8) separator: marks the end of the optional fields
1518 (9) filesystem type: name of filesystem of the form "type[.subtype]"
1519 (10) mount source: filesystem specific information or "none"
1520 (11) super options: per super block options
1522 Parsers should ignore all unrecognised optional fields. Currently the
1523 possible optional fields are:
1525 shared:X mount is shared in peer group X
1526 master:X mount is slave to peer group X
1527 propagate_from:X mount is slave and receives propagation from peer group X (*)
1528 unbindable mount is unbindable
1530 (*) X is the closest dominant peer group under the process's root. If
1531 X is the immediate master of the mount, or if there's no dominant peer
1532 group under the same root, then only the "master:X" field is present
1533 and not the "propagate_from:X" field.
1535 For more information on mount propagation see:
1537 Documentation/filesystems/sharedsubtree.txt
1540 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm
1541 --------------------------------------------------------
1542 These files provide a method to access a tasks comm value. It also allows for
1543 a task to set its own or one of its thread siblings comm value. The comm value
1544 is limited in size compared to the cmdline value, so writing anything longer
1545 then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated
1549 ------------------------------------------------------------------------------
1551 ------------------------------------------------------------------------------
1554 ---------------------
1556 The following mount options are supported:
1558 hidepid= Set /proc/<pid>/ access mode.
1559 gid= Set the group authorized to learn processes information.
1561 hidepid=0 means classic mode - everybody may access all /proc/<pid>/ directories
1564 hidepid=1 means users may not access any /proc/<pid>/ directories but their
1565 own. Sensitive files like cmdline, sched*, status are now protected against
1566 other users. This makes it impossible to learn whether any user runs
1567 specific program (given the program doesn't reveal itself by its behaviour).
1568 As an additional bonus, as /proc/<pid>/cmdline is unaccessible for other users,
1569 poorly written programs passing sensitive information via program arguments are
1570 now protected against local eavesdroppers.
1572 hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be fully invisible to other
1573 users. It doesn't mean that it hides a fact whether a process with a specific
1574 pid value exists (it can be learned by other means, e.g. by "kill -0 $PID"),
1575 but it hides process' uid and gid, which may be learned by stat()'ing
1576 /proc/<pid>/ otherwise. It greatly complicates an intruder's task of gathering
1577 information about running processes, whether some daemon runs with elevated
1578 privileges, whether other user runs some sensitive program, whether other users
1579 run any program at all, etc.
1581 gid= defines a group authorized to learn processes information otherwise
1582 prohibited by hidepid=. If you use some daemon like identd which needs to learn
1583 information about processes information, just add identd to this group.
projects / GitHub / mt8127 / android_kernel_alcatel_ttab.git / blob