{
    "mode": "man",
    "parameter": "procfs",
    "section": "5",
    "url": "https://www.chedong.com/phpMan.php/man/procfs/5/json",
    "generated": "2026-06-13T13:55:35Z",
    "sections": {
        "NAME": {
            "content": "proc - process information pseudo-filesystem\n",
            "subsections": []
        },
        "DESCRIPTION": {
            "content": "The  proc filesystem is a pseudo-filesystem which provides an interface to kernel data struc‐\ntures.  It is commonly mounted at /proc.  Typically, it is mounted automatically by the  sys‐\ntem, but it can also be mounted manually using a command such as:\n\nmount -t proc proc /proc\n\nMost of the files in the proc filesystem are read-only, but some files are writable, allowing\nkernel variables to be changed.\n",
            "subsections": [
                {
                    "name": "Mount options",
                    "content": "The proc filesystem supports the following mount options:\n\nhidepid=n (since Linux 3.3)\nThis option controls who can access the information in /proc/[pid]  directories.   The\nargument, n, is one of the following values:\n\n0   Everybody  may access all /proc/[pid] directories.  This is the traditional behav‐\nior, and the default if this mount option is not specified.\n\n1   Users may not access files and subdirectories inside any  /proc/[pid]  directories\nbut  their own (the /proc/[pid] directories themselves remain visible).  Sensitive\nfiles such as /proc/[pid]/cmdline and /proc/[pid]/status are now protected against\nother users.  This makes it impossible to learn whether any user is running a spe‐\ncific program (so long as the program doesn't otherwise reveal itself by  its  be‐\nhavior).\n\n2   As  for  mode  1,  but  in addition the /proc/[pid] directories belonging to other\nusers become invisible.  This means that /proc/[pid] entries can no longer be used\nto  discover  the  PIDs  on the system.  This doesn't hide the fact that a process\nwith a specific PID value exists (it can be learned by other means,  for  example,\nby \"kill -0 $PID\"), but it hides a process's UID and GID, which could otherwise be\nlearned by employing stat(2) on a /proc/[pid] directory.  This greatly complicates\nan attacker's task of gathering information about running processes (e.g., discov‐\nering whether some daemon is running with  elevated  privileges,  whether  another\nuser  is  running some sensitive program, whether other users are running any pro‐\ngram at all, and so on).\n\ngid=gid (since Linux 3.3)\nSpecifies the ID of a group whose members are authorized to learn process  information\notherwise  prohibited by hidepid (i.e., users in this group behave as though /proc was\nmounted with hidepid=0).  This group should be used  instead  of  approaches  such  as\nputting nonroot users into the sudoers(5) file.\n"
                },
                {
                    "name": "Overview",
                    "content": "Underneath /proc, there are the following general groups of files and subdirectories:\n\n/proc/[pid] subdirectories\nEach  one  of these subdirectories contains files and subdirectories exposing informa‐\ntion about the process with the corresponding process ID.\n\nUnderneath each of the /proc/[pid] directories, a task subdirectory contains subdirec‐\ntories  of  the form task/[tid], which contain corresponding information about each of\nthe threads in the process, where tid is the kernel thread ID of the thread.\n\nThe /proc/[pid] subdirectories are visible when  iterating  through  /proc  with  get‐‐\ndents(2) (and thus are visible when one uses ls(1) to view the contents of /proc).\n\n/proc/[tid] subdirectories\nEach  one  of these subdirectories contains files and subdirectories exposing informa‐\ntion about the thread with the corresponding thread ID.  The contents of these  direc‐\ntories are the same as the corresponding /proc/[pid]/task/[tid] directories.\n\nThe  /proc/[tid] subdirectories are not visible when iterating through /proc with get‐‐\ndents(2) (and thus are not visible when one uses ls(1) to view the contents of /proc).\n\n/proc/self\nWhen a process accesses this magic symbolic link, it resolves  to  the  process's  own\n/proc/[pid] directory.\n\n/proc/thread-self\nWhen  a  thread  accesses  this  magic symbolic link, it resolves to the process's own\n/proc/self/task/[tid] directory.\n\n/proc/[a-z]*\nVarious other files and subdirectories under /proc expose system-wide information.\n\nAll of the above are described in more detail below.\n"
                },
                {
                    "name": "Files and directories",
                    "content": "The following list provides details of many of the files and directories under the /proc  hi‐\nerarchy.\n\n/proc/[pid]\nThere  is a numerical subdirectory for each running process; the subdirectory is named\nby the process ID.  Each /proc/[pid] subdirectory contains the pseudo-files and direc‐\ntories described below.\n\nThe  files  inside each /proc/[pid] directory are normally owned by the effective user\nand effective group ID of the process.  However, as a security measure, the  ownership\nis  made  root:root if the process's \"dumpable\" attribute is set to a value other than\n1.\n\nBefore Linux 4.11, root:root meant the \"global\" root user ID and group ID (i.e., UID 0\nand  GID  0  in the initial user namespace).  Since Linux 4.11, if the process is in a\nnoninitial user namespace that has a valid mapping for user (group) ID  0  inside  the\nnamespace,  then  the user (group) ownership of the files under /proc/[pid] is instead\nmade the same as the root user (group) ID of the namespace.  This means that inside  a\ncontainer, things work as expected for the container \"root\" user.\n\nThe process's \"dumpable\" attribute may change for the following reasons:\n\n*  The attribute was explicitly set via the prctl(2) PRSETDUMPABLE operation.\n\n*  The  attribute  was  reset to the value in the file /proc/sys/fs/suiddumpable (de‐\nscribed below), for the reasons described in prctl(2).\n\nResetting the \"dumpable\" attribute to 1 reverts the  ownership  of  the  /proc/[pid]/*\nfiles  to  the  process's effective UID and GID.  Note, however, that if the effective\nUID or GID is subsequently modified, then the \"dumpable\" attribute may  be  reset,  as\ndescribed  in prctl(2).  Therefore, it may be desirable to reset the \"dumpable\" attri‐\nbute after making any desired changes to the process's effective UID or GID.\n\n/proc/[pid]/attr\nThe files in this directory provide an API for security modules.  The contents of this\ndirectory  are files that can be read and written in order to set security-related at‐\ntributes.  This directory was added to support SELinux, but the intention was that the\nAPI  be general enough to support other security modules.  For the purpose of explana‐\ntion, examples of how SELinux uses these files are provided below.\n\nThis directory is present only if the kernel was configured with CONFIGSECURITY.\n\n/proc/[pid]/attr/current (since Linux 2.6.0)\nThe contents of this file represent the current security attributes of the process.\n\nIn SELinux, this file is used to get the security context  of  a  process.   Prior  to\nLinux 2.6.11, this file could not be used to set the security context (a write was al‐\nways denied), since SELinux limited process security transitions to execve(2) (see the\ndescription of /proc/[pid]/attr/exec, below).  Since Linux 2.6.11, SELinux lifted this\nrestriction and began supporting \"set\" operations via writes to this  node  if  autho‐\nrized by policy, although use of this operation is only suitable for applications that\nare trusted to maintain any desired separation between the old and new  security  con‐\ntexts.\n\nPrior to Linux 2.6.28, SELinux did not allow threads within a multithreaded process to\nset their security context via this node as it would yield an inconsistency among  the\nsecurity  contexts  of the threads sharing the same memory space.  Since Linux 2.6.28,\nSELinux lifted this restriction and began  supporting  \"set\"  operations  for  threads\nwithin a multithreaded process if the new security context is bounded by the old secu‐\nrity context, where the bounded relation is defined in policy and guarantees that  the\nnew security context has a subset of the permissions of the old security context.\n\nOther security modules may choose to support \"set\" operations via writes to this node.\n\n/proc/[pid]/attr/exec (since Linux 2.6.0)\nThis  file  represents  the  attributes to assign to the process upon a subsequent ex‐‐\necve(2).\n\nIn SELinux, this is needed to support role/domain transitions, and  execve(2)  is  the\npreferred  point  to  make  such transitions because it offers better control over the\ninitialization of the process in the new security label and the inheritance of  state.\nIn  SELinux,  this  attribute is reset on execve(2) so that the new program reverts to\nthe default behavior for any execve(2) calls that it may make.  In SELinux, a  process\ncan set only its own /proc/[pid]/attr/exec attribute.\n\n/proc/[pid]/attr/fscreate (since Linux 2.6.0)\nThis  file represents the attributes to assign to files created by subsequent calls to\nopen(2), mkdir(2), symlink(2), and mknod(2)\n\nSELinux employs this file to support creation of a file (using the aforementioned sys‐\ntem  calls)  in a secure state, so that there is no risk of inappropriate access being\nobtained between the time of creation and  the  time  that  attributes  are  set.   In\nSELinux,  this attribute is reset on execve(2), so that the new program reverts to the\ndefault behavior for any file creation calls it may make, but the attribute will  per‐\nsist  across multiple file creation calls within a program unless it is explicitly re‐\nset.  In SELinux, a process can set only its own /proc/[pid]/attr/fscreate attribute.\n\n/proc/[pid]/attr/keycreate (since Linux 2.6.18)\nIf a process writes a security context into this file, all subsequently  created  keys\n(addkey(2)) will be labeled with this context.  For further information, see the ker‐\nnel source  file  Documentation/security/keys/core.rst  (or  file  Documentation/secu‐\nrity/keys.txt  on  Linux  between 3.0 and 4.13, or Documentation/keys.txt before Linux\n3.0).\n\n/proc/[pid]/attr/prev (since Linux 2.6.0)\nThis file contains the security context of the process before the last execve(2); that\nis, the previous value of /proc/[pid]/attr/current.\n\n/proc/[pid]/attr/socketcreate (since Linux 2.6.18)\nIf  a process writes a security context into this file, all subsequently created sock‐\nets will be labeled with this context.\n\n/proc/[pid]/autogroup (since Linux 2.6.38)\nSee sched(7).\n\n/proc/[pid]/auxv (since 2.6.0)\nThis contains the contents of the ELF interpreter information passed to the process at\nexec  time.   The format is one unsigned long ID plus one unsigned long value for each\nentry.  The last entry contains two zeros.  See also getauxval(3).\n\nPermission  to  access   this   file   is   governed   by   a   ptrace   access   mode\nPTRACEMODEREADFSCREDS check; see ptrace(2).\n\n/proc/[pid]/cgroup (since Linux 2.6.24)\nSee cgroups(7).\n\n/proc/[pid]/clearrefs (since Linux 2.6.22)\n\nThis is a write-only file, writable only by owner of the process.\n\nThe following values may be written to the file:\n\n1 (since Linux 2.6.22)\nReset  the  PGReferenced  and ACCESSED/YOUNG bits for all the pages associated\nwith the process.  (Before kernel 2.6.32, writing any  nonzero  value  to  this\nfile had this effect.)\n\n2 (since Linux 2.6.32)\nReset the PGReferenced and ACCESSED/YOUNG bits for all anonymous pages associ‐\nated with the process.\n\n3 (since Linux 2.6.32)\nReset the PGReferenced and ACCESSED/YOUNG bits for all file-mapped pages asso‐\nciated with the process.\n\nClearing  the  PGReferenced  and ACCESSED/YOUNG bits provides a method to measure ap‐\nproximately how much memory a process is using.  One first inspects the values in  the\n\"Referenced\" fields for the VMAs shown in /proc/[pid]/smaps to get an idea of the mem‐\nory footprint of the process.  One then clears the  PGReferenced  and  ACCESSED/YOUNG\nbits  and,  after  some  measured time interval, once again inspects the values in the\n\"Referenced\" fields to get an idea of the change in memory footprint  of  the  process\nduring  the  measured  interval.  If one is interested only in inspecting the selected\nmapping types, then the value 2 or 3 can be used instead of 1.\n\nFurther values can be written to affect different properties:\n\n4 (since Linux 3.11)\nClear the soft-dirty bit for all the pages associated with the  process.   This\nis  used  (in  conjunction with /proc/[pid]/pagemap) by the check-point restore\nsystem to discover which pages of a process have been dirtied  since  the  file\n/proc/[pid]/clearrefs was written to.\n\n5 (since Linux 4.0)\nReset  the  peak resident set size (\"high water mark\") to the process's current\nresident set size value.\n\nWriting any value to /proc/[pid]/clearrefs other than those listed above has  no  ef‐\nfect.\n\nThe /proc/[pid]/clearrefs file is present only if the CONFIGPROCPAGEMONITOR kernel\nconfiguration option is enabled.\n\n/proc/[pid]/cmdline\nThis read-only file holds the complete  command  line  for  the  process,  unless  the\nprocess  is  a  zombie.  In the latter case, there is nothing in this file: that is, a\nread on this file will return 0 characters.  The command-line arguments appear in this\nfile  as a set of strings separated by null bytes ('\\0'), with a further null byte af‐\nter the last string.\n\nIf, after an execve(2), the process modifies its argv strings, those changes will show\nup here.  This is not the same thing as modifying the argv array.\n\nFurthermore,  a  process  may  change  the  memory  location that this file refers via\nprctl(2) operations such as PRSETMMARGSTART.\n\nThink of this file as the command line that the process wants you to see.\n\n/proc/[pid]/comm (since Linux 2.6.33)\nThis file exposes the process's comm value—that is, the command name  associated  with\nthe  process.   Different  threads in the same process may have different comm values,\naccessible via /proc/[pid]/task/[tid]/comm.  A thread may modify its  comm  value,  or\nthat  of  any  of  other  thread  in  the  same  thread  group  (see the discussion of\nCLONETHREAD in clone(2)), by writing to the file /proc/self/task/[tid]/comm.  Strings\nlonger  than  TASKCOMMLEN  (16) characters (including the terminating null byte) are\nsilently truncated.\n\nThis file provides a superset of the prctl(2) PRSETNAME and PRGETNAME  operations,\nand  is  employed  by pthreadsetnamenp(3) when used to rename threads other than the\ncaller.  The value in this file  is  used  for  the  %e  specifier  in  /proc/sys/ker‐\nnel/corepattern; see core(5).\n\n/proc/[pid]/coredumpfilter (since Linux 2.6.23)\nSee core(5).\n\n/proc/[pid]/cpuset (since Linux 2.6.12)\nSee cpuset(7).\n\n/proc/[pid]/cwd\nThis  is a symbolic link to the current working directory of the process.  To find out\nthe current working directory of process 20, for instance, you can do this:\n\n$ cd /proc/20/cwd; pwd -P\n\nIn a multithreaded process, the contents of this symbolic link are  not  available  if\nthe main thread has already terminated (typically by calling pthreadexit(3)).\n\nPermission  to  dereference  or read (readlink(2)) this symbolic link is governed by a\nptrace access mode PTRACEMODEREADFSCREDS check; see ptrace(2).\n\n/proc/[pid]/environ\nThis file contains the initial environment that was set when the  currently  executing\nprogram  was  started  via execve(2).  The entries are separated by null bytes ('\\0'),\nand there may be a null byte at the end.   Thus,  to  print  out  the  environment  of\nprocess 1, you would do:\n\n$ cat /proc/1/environ | tr '\\000' '\\n'\n\nIf,  after  an execve(2), the process modifies its environment (e.g., by calling func‐\ntions such as putenv(3) or modifying the environ(7) variable directly), this file will\nnot reflect those changes.\n\nFurthermore,  a  process  may  change  the  memory  location that this file refers via\nprctl(2) operations such as PRSETMMENVSTART.\n\nPermission  to  access   this   file   is   governed   by   a   ptrace   access   mode\nPTRACEMODEREADFSCREDS check; see ptrace(2).\n\n/proc/[pid]/exe\nUnder Linux 2.2 and later, this file is a symbolic link containing the actual pathname\nof the executed command.  This symbolic link can be dereferenced normally;  attempting\nto open it will open the executable.  You can even type /proc/[pid]/exe to run another\ncopy of the same executable that is being run by process [pid].  If the  pathname  has\nbeen  unlinked,  the symbolic link will contain the string '(deleted)' appended to the\noriginal pathname.  In a multithreaded process, the contents of this symbolic link are\nnot  available  if  the  main  thread  has  already  terminated  (typically by calling\npthreadexit(3)).\n\nPermission to dereference or read (readlink(2)) this symbolic link is  governed  by  a\nptrace access mode PTRACEMODEREADFSCREDS check; see ptrace(2).\n\nUnder Linux 2.0 and earlier, /proc/[pid]/exe is a pointer to the binary which was exe‐\ncuted, and appears as a symbolic link.  A readlink(2) call on this  file  under  Linux\n2.0 returns a string in the format:\n\n[device]:inode\n\nFor  example,  [0301]:1502  would  be  inode  1502  on device major 03 (IDE, MFM, etc.\ndrives) minor 01 (first partition on the first drive).\n\nfind(1) with the -inum option can be used to locate the file.\n\n/proc/[pid]/fd/\nThis is a subdirectory containing one entry for each file which the process has  open,\nnamed  by its file descriptor, and which is a symbolic link to the actual file.  Thus,\n0 is standard input, 1 standard output, 2 standard error, and so on.\n\nFor file descriptors for pipes and sockets, the entries will be symbolic  links  whose\ncontent  is  the  file type with the inode.  A readlink(2) call on this file returns a\nstring in the format:\n\ntype:[inode]\n\nFor example, socket:[2248868] will be a socket and its inode is 2248868.  For sockets,\nthat inode can be used to find more information in one of the files under /proc/net/.\n\nFor file descriptors that have no corresponding inode (e.g., file descriptors produced\nby bpf(2),  epollcreate(2),  eventfd(2),  inotifyinit(2),  perfeventopen(2),  sig‐‐\nnalfd(2),  timerfdcreate(2),  and  userfaultfd(2)), the entry will be a symbolic link\nwith contents of the form\n\nanoninode:<file-type>\n\nIn many cases (but not all), the file-type is surrounded by square brackets.\n\nFor example, an epoll file descriptor will have a symbolic link whose content  is  the\nstring anoninode:[eventpoll].\n\nIn  a  multithreaded  process, the contents of this directory are not available if the\nmain thread has already terminated (typically by calling pthreadexit(3)).\n\nPrograms that take a filename as a command-line argument, but don't  take  input  from\nstandard  input if no argument is supplied, and programs that write to a file named as\na command-line argument, but don't send their output to standard output if no argument\nis  supplied, can nevertheless be made to use standard input or standard output by us‐\ning /proc/[pid]/fd files as command-line arguments.  For example, assuming that -i  is\nthe flag designating an input file and -o is the flag designating an output file:\n\n$ foobar -i /proc/self/fd/0 -o /proc/self/fd/1 ...\n\nand you have a working filter.\n\n/proc/self/fd/N is approximately the same as /dev/fd/N in some UNIX and UNIX-like sys‐\ntems.  Most Linux MAKEDEV scripts symbolically link /dev/fd to /proc/self/fd, in fact.\n\nMost systems provide symbolic links /dev/stdin, /dev/stdout,  and  /dev/stderr,  which\nrespectively link to the files 0, 1, and 2 in /proc/self/fd.  Thus the example command\nabove could be written as:\n\n$ foobar -i /dev/stdin -o /dev/stdout ...\n\nPermission to dereference or read (readlink(2)) the symbolic links in  this  directory\nis governed by a ptrace access mode PTRACEMODEREADFSCREDS check; see ptrace(2).\n\nNote  that  for  file  descriptors referring to inodes (pipes and sockets, see above),\nthose inodes still have permission bits and ownership information distinct from  those\nof the /proc/[pid]/fd entry, and that the owner may differ from the user and group IDs\nof the process.  An unprivileged process may lack permissions to open them, as in this\nexample:\n\n$ echo test | sudo -u nobody cat\ntest\n$ echo test | sudo -u nobody cat /proc/self/fd/0\ncat: /proc/self/fd/0: Permission denied\n\nFile  descriptor  0  refers to the pipe created by the shell and owned by that shell's\nuser, which is not nobody, so cat does not have permission to create a  new  file  de‐\nscriptor to read from that inode, even though it can still read from its existing file\ndescriptor 0.\n\n/proc/[pid]/fdinfo/ (since Linux 2.6.22)\nThis is a subdirectory containing one entry for each file which the process has  open,\nnamed  by  its  file descriptor.  The files in this directory are readable only by the\nowner of the process.  The contents of each file can be  read  to  obtain  information\nabout  the corresponding file descriptor.  The content depends on the type of file re‐\nferred to by the corresponding file descriptor.\n\nFor regular files and directories, we see something like:\n\n$ cat /proc/12015/fdinfo/4\npos:    1000\nflags:  01002002\nmntid: 21\n\nThe fields are as follows:\n\npos    This is a decimal number showing the file offset.\n\nflags  This is an octal number that displays the file  access  mode  and  file  status\nflags  (see  open(2)).   If the close-on-exec file descriptor flag is set, then\nflags will also include the value OCLOEXEC.\n\nBefore Linux 3.1, this field incorrectly displayed the setting of OCLOEXEC  at\nthe  time the file was opened, rather than the current setting of the close-on-\nexec flag.\n\nmntid This field, present since Linux 3.15, is the ID of the mount  point  containing\nthis file.  See the description of /proc/[pid]/mountinfo.\n\nFor  eventfd file descriptors (see eventfd(2)), we see (since Linux 3.8) the following\nfields:\n\npos: 0\nflags:    02\nmntid:   10\neventfd-count:               40\n\neventfd-count is the current value of the eventfd counter, in hexadecimal.\n\nFor epoll file descriptors (see epoll(7)), we see  (since  Linux  3.8)  the  following\nfields:\n\npos: 0\nflags:    02\nmntid:   10\ntfd:        9 events:       19 data: 74253d2500000009\ntfd:        7 events:       19 data: 74253d2500000007\n\nEach  of the lines beginning tfd describes one of the file descriptors being monitored\nvia the epoll file descriptor (see epollctl(2) for some details).  The tfd  field  is\nthe  number  of  the  file  descriptor.  The events field is a hexadecimal mask of the\nevents being monitored for this file descriptor.  The data field is the data value as‐\nsociated with this file descriptor.\n\nFor  signalfd file descriptors (see signalfd(2)), we see (since Linux 3.8) the follow‐\ning fields:\n\npos: 0\nflags:    02\nmntid:   10\nsigmask:  0000000000000006\n\nsigmask is the hexadecimal mask of signals that are accepted via  this  signalfd  file\ndescriptor.  (In this example, bits 2 and 3 are set, corresponding to the signals SIG‐‐\nINT and SIGQUIT; see signal(7).)\n\nFor inotify file descriptors (see inotify(7)), we see (since Linux 3.8) the  following\nfields:\n\npos: 0\nflags:    00\nmntid:   11\ninotify wd:2 ino:7ef82a sdev:800001 mask:800afff ignoredmask:0 fhandle-bytes:8 fhandle-type:1 fhandle:2af87e00220ffd73\ninotify wd:1 ino:192627 sdev:800001 mask:800afff ignoredmask:0 fhandle-bytes:8 fhandle-type:1 fhandle:27261900802dfd73\n\nEach  of the lines beginning with \"inotify\" displays information about one file or di‐\nrectory that is being monitored.  The fields in this line are as follows:\n\nwd     A watch descriptor number (in decimal).\n\nino    The inode number of the target file (in hexadecimal).\n\nsdev   The ID of the device where the target file resides (in hexadecimal).\n\nmask   The mask of events being monitored for the target file (in hexadecimal).\n\nIf the kernel was built with exportfs support, the path to the target file is  exposed\nas  a  file  handle,  via  three  hexadecimal fields: fhandle-bytes, fhandle-type, and\nfhandle.\n\nFor fanotify file descriptors (see fanotify(7)), we see (since Linux 3.8) the  follow‐\ning fields:\n\npos: 0\nflags:    02\nmntid:   11\nfanotify flags:0 event-flags:88002\nfanotify ino:19264f sdev:800001 mflags:0 mask:1 ignoredmask:0 fhandle-bytes:8 fhandle-type:1 fhandle:4f261900a82dfd73\n\nThe  fourth  line displays information defined when the fanotify group was created via\nfanotifyinit(2):\n\nflags  The flags argument given to fanotifyinit(2) (expressed in hexadecimal).\n\nevent-flags\nThe eventfflags argument given to fanotifyinit(2)  (expressed  in  hexadeci‐\nmal).\n\nEach  additional line shown in the file contains information about one of the marks in\nthe fanotify group.  Most of these fields are as for inotify, except:\n\nmflags The flags associated with the mark (expressed in hexadecimal).\n\nmask   The events mask for this mark (expressed in hexadecimal).\n\nignoredmask\nThe mask of events that are ignored for this mark (expressed in hexadecimal).\n\nFor details on these fields, see fanotifymark(2).\n\nFor timerfd file descriptors (see timerfd(2)), we see (since Linux 3.17) the following\nfields:\n\npos:    0\nflags:  02004002\nmntid: 13\nclockid: 0\nticks: 0\nsettime flags: 03\nitvalue: (7695568592, 640020877)\nitinterval: (0, 0)\n\nclockid\nThis  is the numeric value of the clock ID (corresponding to one of the CLOCK*\nconstants defined via <time.h>) that is used to mark the progress of the  timer\n(in this example, 0 is CLOCKREALTIME).\n\nticks  This  is  the  number of timer expirations that have occurred, (i.e., the value\nthat read(2) on it would return).\n\nsettime flags\nThis field lists  the  flags  with  which  the  timerfd  was  last  armed  (see\ntimerfdsettime(2)),  in  octal  (in  this  example, both TFDTIMERABSTIME and\nTFDTIMERCANCELONSET are set).\n\nitvalue\nThis field contains the amount of time until the timer will  next  expire,  ex‐\npressed  in  seconds  and  nanoseconds.  This is always expressed as a relative\nvalue, regardless of whether the timer was created using the  TFDTIMERABSTIME\nflag.\n\nitinterval\nThis  field  contains  the  interval  of the timer, in seconds and nanoseconds.\n(The itvalue and itinterval fields contain the values that timerfdgettime(2)\non this file descriptor would return.)\n\n/proc/[pid]/gidmap (since Linux 3.5)\nSee usernamespaces(7).\n\n/proc/[pid]/io (since kernel 2.6.20)\nThis file contains I/O statistics for the process, for example:\n\n# cat /proc/3828/io\nrchar: 323934931\nwchar: 323929600\nsyscr: 632687\nsyscw: 632675\nreadbytes: 0\nwritebytes: 323932160\ncancelledwritebytes: 0\n\nThe fields are as follows:\n\nrchar: characters read\nThe  number  of bytes which this task has caused to be read from storage.  This\nis simply the sum of bytes which this process passed  to  read(2)  and  similar\nsystem  calls.   It  includes  things such as terminal I/O and is unaffected by\nwhether or not actual physical disk I/O was required (the read might have  been\nsatisfied from pagecache).\n\nwchar: characters written\nThe number of bytes which this task has caused, or shall cause to be written to\ndisk.  Similar caveats apply here as with rchar.\n\nsyscr: read syscalls\nAttempt to count the number of read I/O operations—that is, system  calls  such\nas read(2) and pread(2).\n\nsyscw: write syscalls\nAttempt  to count the number of write I/O operations—that is, system calls such\nas write(2) and pwrite(2).\n\nreadbytes: bytes read\nAttempt to count the number of bytes which this process really did cause to  be\nfetched from the storage layer.  This is accurate for block-backed filesystems.\n\nwritebytes: bytes written\nAttempt  to  count  the number of bytes which this process caused to be sent to\nthe storage layer.\n\ncancelledwritebytes:\nThe big inaccuracy here is truncate.  If a process writes 1 MB to  a  file  and\nthen  deletes  the file, it will in fact perform no writeout.  But it will have\nbeen accounted as having caused 1 MB of write.  In other words: this field rep‐\nresents  the  number of bytes which this process caused to not happen, by trun‐\ncating pagecache.  A task can cause \"negative\" I/O too.  If this task truncates\nsome  dirty  pagecache,  some I/O which another task has been accounted for (in\nits writebytes) will not be happening.\n\nNote: In the current implementation, things are a  bit  racy  on  32-bit  systems:  if\nprocess  A  reads  process B's /proc/[pid]/io while process B is updating one of these\n64-bit counters, process A could see an intermediate result.\n\nPermission  to  access   this   file   is   governed   by   a   ptrace   access   mode\nPTRACEMODEREADFSCREDS check; see ptrace(2).\n\n/proc/[pid]/limits (since Linux 2.6.24)\nThis  file  displays  the soft limit, hard limit, and units of measurement for each of\nthe process's resource limits (see getrlimit(2)).  Up to and including  Linux  2.6.35,\nthis  file  is  protected to allow reading only by the real UID of the process.  Since\nLinux 2.6.36, this file is readable by all users on the system.\n\n/proc/[pid]/mapfiles/ (since kernel 3.3)\nThis subdirectory contains entries corresponding to memory-mapped files (see mmap(2)).\nEntries  are named by memory region start and end address pair (expressed as hexadeci‐\nmal numbers), and are symbolic links to the mapped files themselves.  Here is an exam‐\nple, with the output wrapped and reformatted to fit on an 80-column display:\n\n# ls -l /proc/self/mapfiles/\nlr--------. 1 root root 64 Apr 16 21:31\n3252e00000-3252e20000 -> /usr/lib64/ld-2.15.so\n...\n\nAlthough  these  entries  are  present  for  memory  regions that were mapped with the\nMAPFILE flag, the way anonymous shared memory (regions created with  the  MAPANON  |\nMAPSHARED  flags) is implemented in Linux means that such regions also appear on this\ndirectory.  Here is an example where the target file is the deleted /dev/zero one:\n\nlrw-------. 1 root root 64 Apr 16 21:33\n7fc075d2f000-7fc075e6f000 -> /dev/zero (deleted)\n\nPermission  to  access   this   file   is   governed   by   a   ptrace   access   mode\nPTRACEMODEREADFSCREDS check; see ptrace(2).\n\nUntil  kernel  version  4.3, this directory appeared only if the CONFIGCHECKPOINTRE‐‐\nSTORE kernel configuration option was enabled.\n\nCapabilities are required to read the contents of the symbolic links  in  this  direc‐\ntory: before Linux 5.9, the reading process requires CAPSYSADMIN in the initial user\nnamespace; since Linux 5.9, the reading process  must  have  either  CAPSYSADMIN  or\nCAPCHECKPOINTRESTORE in the user namespace where it resides.\n\n/proc/[pid]/maps\nA  file  containing  the currently mapped memory regions and their access permissions.\nSee mmap(2) for some further information about memory mappings.\n\nPermission  to  access   this   file   is   governed   by   a   ptrace   access   mode\nPTRACEMODEREADFSCREDS check; see ptrace(2).\n\nThe format of the file is:\n\naddress           perms offset  dev   inode       pathname\n00400000-00452000 r-xp 00000000 08:02 173521      /usr/bin/dbus-daemon\n00651000-00652000 r--p 00051000 08:02 173521      /usr/bin/dbus-daemon\n00652000-00655000 rw-p 00052000 08:02 173521      /usr/bin/dbus-daemon\n00e03000-00e24000 rw-p 00000000 00:00 0           [heap]\n00e24000-011f7000 rw-p 00000000 00:00 0           [heap]\n...\n35b1800000-35b1820000 r-xp 00000000 08:02 135522  /usr/lib64/ld-2.15.so\n35b1a1f000-35b1a20000 r--p 0001f000 08:02 135522  /usr/lib64/ld-2.15.so\n35b1a20000-35b1a21000 rw-p 00020000 08:02 135522  /usr/lib64/ld-2.15.so\n35b1a21000-35b1a22000 rw-p 00000000 00:00 0\n35b1c00000-35b1dac000 r-xp 00000000 08:02 135870  /usr/lib64/libc-2.15.so\n35b1dac000-35b1fac000 ---p 001ac000 08:02 135870  /usr/lib64/libc-2.15.so\n35b1fac000-35b1fb0000 r--p 001ac000 08:02 135870  /usr/lib64/libc-2.15.so\n35b1fb0000-35b1fb2000 rw-p 001b0000 08:02 135870  /usr/lib64/libc-2.15.so\n...\nf2c6ff8c000-7f2c7078c000 rw-p 00000000 00:00 0    [stack:986]\n...\n7fffb2c0d000-7fffb2c2e000 rw-p 00000000 00:00 0   [stack]\n7fffb2d48000-7fffb2d49000 r-xp 00000000 00:00 0   [vdso]\n\nThe  address field is the address space in the process that the mapping occupies.  The\nperms field is a set of permissions:\n\nr = read\nw = write\nx = execute\ns = shared\np = private (copy on write)\n\nThe offset field is the offset into the file/whatever; dev is  the  device  (major:mi‐\nnor); inode is the inode on that device.  0 indicates that no inode is associated with\nthe memory region, as would be the case with BSS (uninitialized data).\n\nThe pathname field will usually be the file that is  backing  the  mapping.   For  ELF\nfiles,  you can easily coordinate with the offset field by looking at the Offset field\nin the ELF program headers (readelf -l).\n\nThere are additional helpful pseudo-paths:\n\n[stack]\nThe initial process's (also known as the main thread's) stack.\n\n[stack:<tid>] (from Linux 3.4 to 4.4)\nA thread's stack (where the <tid> is a  thread  ID).   It  corresponds  to  the\n/proc/[pid]/task/[tid]/  path.  This field was removed in Linux 4.5, since pro‐\nviding this information for a process with large numbers of threads  is  expen‐\nsive.\n\n[vdso] The virtual dynamically linked shared object.  See vdso(7).\n\n[heap] The process's heap.\n\nIf  the pathname field is blank, this is an anonymous mapping as obtained via mmap(2).\nThere is no easy way to coordinate this back to a process's source, short  of  running\nit through gdb(1), strace(1), or similar.\n\npathname  is shown unescaped except for newline characters, which are replaced with an\noctal escape sequence.  As a result, it is not possible to determine whether the orig‐\ninal pathname contained a newline character or the literal \\012 character sequence.\n\nIf  the  mapping is file-backed and the file has been deleted, the string \" (deleted)\"\nis appended to the pathname.  Note that this is ambiguous too.\n\nUnder Linux 2.0, there is no field giving pathname.\n\n/proc/[pid]/mem\nThis file can be used to access the pages  of  a  process's  memory  through  open(2),\nread(2), and lseek(2).\n\nPermission  to  access  this  file is governed by a ptrace access mode PTRACEMODEAT‐‐\nTACHFSCREDS check; see ptrace(2).\n\n/proc/[pid]/mountinfo (since Linux 2.6.26)\nThis file contains information about mount points in  the  process's  mount  namespace\n(see  mountnamespaces(7)).  It supplies various information (e.g., propagation state,\nroot of mount for bind mounts, identifier for each mount and its parent) that is miss‐\ning  from  the  (older) /proc/[pid]/mounts file, and fixes various other problems with\nthat file (e.g., nonextensibility, failure to  distinguish  per-mount  versus  per-su‐\nperblock options).\n\nThe file contains lines of the form:\n\n36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue\n(1)(2)(3)   (4)   (5)      (6)      (7)   (8) (9)   (10)         (11)\n\nThe numbers in parentheses are labels for the descriptions below:\n\n(1)  mount ID: a unique ID for the mount (may be reused after umount(2)).\n\n(2)  parent  ID:  the  ID  of  the parent mount (or of self for the root of this mount\nnamespace's mount tree).\n\nIf a new mount is stacked on top of a previous existing mount (so that  it  hides\nthe existing mount) at pathname P, then the parent of the new mount is the previ‐\nous mount at that location.  Thus, when looking at all the mounts  stacked  at  a\nparticular  location, the top-most mount is the one that is not the parent of any\nother mount at the same location.  (Note, however, that this top-most mount  will\nbe  accessible  only  if the longest path subprefix of P that is a mount point is\nnot itself hidden by a stacked mount.)\n\nIf the parent mount point lies outside the  process's  root  directory  (see  ch‐‐\nroot(2)),  the ID shown here won't have a corresponding record in mountinfo whose\nmount ID (field 1) matches this parent mount ID (because mount  points  that  lie\noutside  the  process's root directory are not shown in mountinfo).  As a special\ncase of this point, the process's root mount point may have a parent  mount  (for\nthe  initramfs filesystem) that lies outside the process's root directory, and an\nentry for that mount point will not appear in mountinfo.\n\n(3)  major:minor: the value of stdev for files on this filesystem (see stat(2)).\n\n(4)  root: the pathname of the directory in the filesystem which  forms  the  root  of\nthis mount.\n\n(5)  mount  point:  the pathname of the mount point relative to the process's root di‐\nrectory.\n\n(6)  mount options: per-mount options (see mount(2)).\n\n(7)  optional fields: zero or more fields of the form \"tag[:value]\"; see below.\n\n(8)  separator: the end of the optional fields is marked by a single hyphen.\n\n(9)  filesystem type: the filesystem type in the form \"type[.subtype]\".\n\n(10) mount source: filesystem-specific information or \"none\".\n\n(11) super options: per-superblock options (see mount(2)).\n\nCurrently, the possible optional fields are shared, master,  propagatefrom,  and  un‐\nbindable.   See mountnamespaces(7) for a description of these fields.  Parsers should\nignore all unrecognized optional fields.\n\nFor more information on mount  propagation  see:  Documentation/filesystems/sharedsub‐\ntree.txt in the Linux kernel source tree.\n\n/proc/[pid]/mounts (since Linux 2.4.19)\nThis file lists all the filesystems currently mounted in the process's mount namespace\n(see mountnamespaces(7)).  The format of this file is documented in fstab(5).\n\nSince kernel version 2.6.15, this file is pollable: after opening the file  for  read‐\ning,  a  change in this file (i.e., a filesystem mount or unmount) causes select(2) to\nmark the  file  descriptor  as  having  an  exceptional  condition,  and  poll(2)  and\nepollwait(2)  mark  the  file  as  having  a priority event (POLLPRI).  (Before Linux\n2.6.30, a change in this file was indicated by the file  descriptor  being  marked  as\nreadable  for select(2), and being marked as having an error condition for poll(2) and\nepollwait(2).)\n\n/proc/[pid]/mountstats (since Linux 2.6.17)\nThis file exports information (statistics, configuration information) about the  mount\npoints in the process's mount namespace (see mountnamespaces(7)).  Lines in this file\nhave the form:\n\ndevice /dev/sda7 mounted on /home with fstype ext3 [stats]\n(       1      )            ( 2 )             (3 ) (  4  )\n\nThe fields in each line are:\n\n(1)  The name of the mounted device (or \"nodevice\" if there is  no  corresponding  de‐\nvice).\n\n(2)  The mount point within the filesystem tree.\n\n(3)  The filesystem type.\n\n(4)  Optional  statistics  and  configuration  information.   Currently  (as  at Linux\n2.6.26), only NFS filesystems export information via this field.\n\nThis file is readable only by the owner of the process.\n\n/proc/[pid]/net (since Linux 2.6.25)\nSee the description of /proc/net.\n\n/proc/[pid]/ns/ (since Linux 3.0)\nThis is a subdirectory containing one entry for each namespace that supports being ma‐\nnipulated by setns(2).  For more information, see namespaces(7).\n\n/proc/[pid]/numamaps (since Linux 2.6.14)\nSee numa(7).\n\n/proc/[pid]/oomadj (since Linux 2.6.11)\nThis  file  can  be  used  to  adjust the score used to select which process should be\nkilled in an out-of-memory (OOM) situation.  The kernel uses this  value  for  a  bit-\nshift operation of the process's oomscore value: valid values are in the range -16 to\n+15, plus the special value  -17,  which  disables  OOM-killing  altogether  for  this\nprocess.   A  positive  score increases the likelihood of this process being killed by\nthe OOM-killer; a negative score decreases the likelihood.\n\nThe default value for this file is 0; a new process inherits its parent's oomadj set‐\nting.  A process must be privileged (CAPSYSRESOURCE) to update this file.\n\nSince    Linux    2.6.36,   use   of   this   file   is   deprecated   in   favor   of\n/proc/[pid]/oomscoreadj.\n\n/proc/[pid]/oomscore (since Linux 2.6.11)\nThis file displays the current score that the kernel gives to  this  process  for  the\npurpose  of  selecting  a  process  for the OOM-killer.  A higher score means that the\nprocess is more likely to be selected by the OOM-killer.  The basis for this score  is\nthe amount of memory used by the process, with increases (+) or decreases (-) for fac‐\ntors including:\n\n* whether the process is privileged (-).\n\nBefore kernel 2.6.36 the following factors  were  also  used  in  the  calculation  of\noomscore:\n\n* whether the process creates a lot of children using fork(2) (+);\n\n* whether the process has been running a long time, or has used a lot of CPU time (-);\n\n* whether the process has a low nice value (i.e., > 0) (+); and\n\n* whether the process is making direct hardware access (-).\n\nThe  oomscore  also reflects the adjustment specified by the oomscoreadj or oomadj\nsetting for the process.\n\n/proc/[pid]/oomscoreadj (since Linux 2.6.36)\nThis file can be used to adjust the badness heuristic used  to  select  which  process\ngets killed in out-of-memory conditions.\n\nThe  badness  heuristic  assigns  a value to each candidate task ranging from 0 (never\nkill) to 1000 (always kill) to determine which process is  targeted.   The  units  are\nroughly a proportion along that range of allowed memory the process may allocate from,\nbased on an estimation of its current memory and swap use.  For example, if a task  is\nusing  all allowed memory, its badness score will be 1000.  If it is using half of its\nallowed memory, its score will be 500.\n\nThere is an additional factor included in the badness score: root processes are  given\n3% extra memory over other tasks.\n\nThe  amount  of  \"allowed\"  memory  depends on the context in which the OOM-killer was\ncalled.  If it is due to the memory assigned to the allocating task's cpuset being ex‐\nhausted,  the  allowed  memory represents the set of mems assigned to that cpuset (see\ncpuset(7)).  If it is due to a mempolicy's node(s) being exhausted, the allowed memory\nrepresents the set of mempolicy nodes.  If it is due to a memory limit (or swap limit)\nbeing reached, the allowed memory is that configured limit.  Finally, if it is due  to\nthe  entire  system being out of memory, the allowed memory represents all allocatable\nresources.\n\nThe value of oomscoreadj is added to the badness score before it is used  to  deter‐\nmine  which  task  to kill.  Acceptable values range from -1000 (OOMSCOREADJMIN) to\n+1000 (OOMSCOREADJMAX).  This allows user space to control the preference for  OOM-\nkilling, ranging from always preferring a certain task or completely disabling it from\nOOM killing.  The lowest possible value, -1000, is equivalent to disabling OOM-killing\nentirely for that task, since it will always report a badness score of 0.\n\nConsequently,  it is very simple for user space to define the amount of memory to con‐\nsider for each task.  Setting an oomscoreadj value of +500, for example, is  roughly\nequivalent to allowing the remainder of tasks sharing the same system, cpuset, mempol‐\nicy, or memory controller resources to use at least 50% more memory.  A value of -500,\non  the  other  hand, would be roughly equivalent to discounting 50% of the task's al‐\nlowed memory from being considered as scoring against the task.\n\nFor backward compatibility with previous kernels,  /proc/[pid]/oomadj  can  still  be\nused to tune the badness score.  Its value is scaled linearly with oomscoreadj.\n\nWriting to /proc/[pid]/oomscoreadj or /proc/[pid]/oomadj will change the other with\nits scaled value.\n\nThe choom(1) program provides a command-line interface for adjusting the oomscoreadj\nvalue of a running process or a newly executed command.\n\n/proc/[pid]/pagemap (since Linux 2.6.25)\nThis  file shows the mapping of each of the process's virtual pages into physical page\nframes or swap area.  It contains one 64-bit value for each  virtual  page,  with  the\nbits set as follows:\n\n63     If set, the page is present in RAM.\n\n62     If set, the page is in swap space\n\n61 (since Linux 3.5)\nThe page is a file-mapped page or a shared anonymous page.\n\n60–57 (since Linux 3.11)\nZero\n\n56 (since Linux 4.2)\nThe page is exclusively mapped.\n\n55 (since Linux 3.11)\nPTE   is   soft-dirty   (see   the   kernel   source   file   Documentation/ad‐\nmin-guide/mm/soft-dirty.rst).\n\n54–0   If the page is present in RAM (bit 63), then these bits provide the page  frame\nnumber,  which  can be used to index /proc/kpageflags and /proc/kpagecount.  If\nthe page is present in swap (bit 62), then bits 4–0 give  the  swap  type,  and\nbits 54–5 encode the swap offset.\n\nBefore Linux 3.11, bits 60–55 were used to encode the base-2 log of the page size.\n\nTo employ /proc/[pid]/pagemap efficiently, use /proc/[pid]/maps to determine which ar‐\neas of memory are actually mapped and seek to skip over unmapped regions.\n\nThe /proc/[pid]/pagemap file is present only if  the  CONFIGPROCPAGEMONITOR  kernel\nconfiguration option is enabled.\n\nPermission   to   access   this   file   is   governed   by   a   ptrace  access  mode\nPTRACEMODEREADFSCREDS check; see ptrace(2).\n\n/proc/[pid]/personality (since Linux 2.6.28)\nThis read-only file exposes the process's execution domain, as set by  personality(2).\nThe value is displayed in hexadecimal notation.\n\nPermission  to  access  this  file is governed by a ptrace access mode PTRACEMODEAT‐‐\nTACHFSCREDS check; see ptrace(2).\n\n/proc/[pid]/root\nUNIX and Linux support the idea of a per-process root of the filesystem,  set  by  the\nchroot(2) system call.  This file is a symbolic link that points to the process's root\ndirectory, and behaves in the same way as exe, and fd/*.\n\nNote however that this file is not merely a symbolic link.  It provides the same  view\nof  the  filesystem  (including  namespaces  and the set of per-process mounts) as the\nprocess itself.  An example illustrates this point.  In one terminal, we start a shell\nin new user and mount namespaces, and in that shell we create some new mount points:\n\n$ PS1='sh1# ' unshare -Urnm\nsh1# mount -t tmpfs tmpfs /etc  # Mount empty tmpfs at /etc\nsh1# mount --bind /usr /dev     # Mount /usr at /dev\nsh1# echo $$\n27123\n\nIn  a  second terminal window, in the initial mount namespace, we look at the contents\nof the corresponding mounts in the initial and new namespaces:\n\n$ PS1='sh2# ' sudo sh\nsh2# ls /etc | wc -l                  # In initial NS\n309\nsh2# ls /proc/27123/root/etc | wc -l  # /etc in other NS\n0                                     # The empty tmpfs dir\nsh2# ls /dev | wc -l                  # In initial NS\n205\nsh2# ls /proc/27123/root/dev | wc -l  # /dev in other NS\n11                                    # Actually bind\n# mounted to /usr\nsh2# ls /usr | wc -l                  # /usr in initial NS\n11\n\nIn a multithreaded process, the contents of the /proc/[pid]/root symbolic link are not\navailable   if   the   main  thread  has  already  terminated  (typically  by  calling\npthreadexit(3)).\n\nPermission to dereference or read (readlink(2)) this symbolic link is  governed  by  a\nptrace access mode PTRACEMODEREADFSCREDS check; see ptrace(2).\n\n/proc/[pid]/seccomp (Linux 2.6.12 to 2.6.22)\nThis file can be used to read and change the process's secure computing (seccomp) mode\nsetting.  It contains the value 0 if the process is not in seccomp mode, and 1 if  the\nprocess is in strict seccomp mode (see seccomp(2)).  Writing 1 to this file places the\nprocess irreversibly in strict seccomp mode.  (Further attempts to write to  the  file\nfail with the EPERM error.)\n\nIn  Linux  2.6.23,  this file went away, to be replaced by the prctl(2) PRGETSECCOMP\nand PRSETSECCOMP operations (and later  by  seccomp(2)  and  the  Seccomp  field  in\n/proc/[pid]/status).\n\n/proc/[pid]/setgroups (since Linux 3.19)\nSee usernamespaces(7).\n\n/proc/[pid]/smaps (since Linux 2.6.14)\nThis  file  shows memory consumption for each of the process's mappings.  (The pmap(1)\ncommand displays similar information, in a form that may be easier for parsing.)   For\neach mapping there is a series of lines such as the following:\n\n00400000-0048a000 r-xp 00000000 fd:03 960637       /bin/bash\nSize:                552 kB\nRss:                 460 kB\nPss:                 100 kB\nSharedClean:        452 kB\nSharedDirty:          0 kB\nPrivateClean:         8 kB\nPrivateDirty:         0 kB\nReferenced:          460 kB\nAnonymous:             0 kB\nAnonHugePages:         0 kB\nShmemHugePages:        0 kB\nShmemPmdMapped:        0 kB\nSwap:                  0 kB\nKernelPageSize:        4 kB\nMMUPageSize:           4 kB\nKernelPageSize:        4 kB\nMMUPageSize:           4 kB\nLocked:                0 kB\nProtectionKey:         0\nVmFlags: rd ex mr mw me dw\n\nThe first of these lines shows the same information as is displayed for the mapping in\n/proc/[pid]/maps.  The following lines show the size of the mapping, the amount of the\nmapping that is currently resident in RAM (\"Rss\"), the process's proportional share of\nthis mapping (\"Pss\"), the number of clean and dirty shared pages in the  mapping,  and\nthe  number  of  clean and dirty private pages in the mapping.  \"Referenced\" indicates\nthe amount of memory currently marked as referenced or  accessed.   \"Anonymous\"  shows\nthe  amount  of memory that does not belong to any file.  \"Swap\" shows how much would-\nbe-anonymous memory is also used, but out on swap.\n\nThe \"KernelPageSize\" line (available since Linux 2.6.29) is the page size used by  the\nkernel  to back the virtual memory area.  This matches the size used by the MMU in the\nmajority of cases.  However, one counter-example occurs on  PPC64  kernels  whereby  a\nkernel  using  64 kB as a base page size may still use 4 kB pages for the MMU on older\nprocessors.  To distinguish the two attributes, the \"MMUPageSize\" line (also available\nsince Linux 2.6.29) reports the page size used by the MMU.\n\nThe \"Locked\" indicates whether the mapping is locked in memory or not.\n\nThe  \"ProtectionKey\" line (available since Linux 4.9, on x86 only) contains the memory\nprotection key (see pkeys(7)) associated with the virtual memory area.  This entry  is\npresent  only if the kernel was built with the CONFIGX86INTELMEMORYPROTECTIONKEYS\nconfiguration option (since Linux 4.6).\n\nThe \"VmFlags\" line (available since Linux 3.8) represents the kernel flags  associated\nwith the virtual memory area, encoded using the following two-letter codes:\n\nrd  - readable\nwr  - writable\nex  - executable\nsh  - shared\nmr  - may read\nmw  - may write\nme  - may execute\nms  - may share\ngd  - stack segment grows down\npf  - pure PFN range\ndw  - disabled write to the mapped file\nlo  - pages are locked in memory\nio  - memory mapped I/O area\nsr  - sequential read advise provided\nrr  - random read advise provided\ndc  - do not copy area on fork\nde  - do not expand area on remapping\nac  - area is accountable\nnr  - swap space is not reserved for the area\nht  - area uses huge tlb pages\nsf  - perform synchronous page faults (since Linux 4.15)\nnl  - non-linear mapping (removed in Linux 4.0)\nar  - architecture specific flag\nwf  - wipe on fork (since Linux 4.14)\ndd  - do not include area into core dump\nsd  - soft-dirty flag (since Linux 3.13)\nmm  - mixed map area\nhg  - huge page advise flag\nnh  - no-huge page advise flag\nmg  - mergeable advise flag\num  - userfaultfd missing pages tracking (since Linux 4.3)\nuw  - userfaultfd wprotect pages tracking (since Linux 4.3)\n\nThe /proc/[pid]/smaps file is present only if the CONFIGPROCPAGEMONITOR kernel con‐\nfiguration option is enabled.\n\n/proc/[pid]/stack (since Linux 2.6.29)\nThis file provides a symbolic trace of the function calls  in  this  process's  kernel\nstack.   This file is provided only if the kernel was built with the CONFIGSTACKTRACE\nconfiguration option.\n\nPermission to access this file is governed by a  ptrace  access  mode  PTRACEMODEAT‐‐\nTACHFSCREDS check; see ptrace(2).\n\n/proc/[pid]/stat\nStatus  information  about  the process.  This is used by ps(1).  It is defined in the\nkernel source file fs/proc/array.c.\n\nThe fields, in order, with their proper scanf(3) format specifiers, are listed  below.\nWhether  or  not  certain  of  these fields display valid information is governed by a\nptrace access  mode  PTRACEMODEREADFSCREDS | PTRACEMODENOAUDIT  check  (refer  to\nptrace(2)).   If the check denies access, then the field value is displayed as 0.  The\naffected fields are indicated with the marking [PT].\n\n(1) pid  %d\nThe process ID.\n\n(2) comm  %s\nThe  filename  of  the  executable,  in  parentheses.   Strings   longer   than\nTASKCOMMLEN  (16)  characters  (including  the  terminating  null  byte)  are\nsilently truncated.  This is visible whether or not the executable  is  swapped\nout.\n\n(3) state  %c\nOne of the following characters, indicating process state:\n\nR  Running\n\nS  Sleeping in an interruptible wait\n\nD  Waiting in uninterruptible disk sleep\n\nZ  Zombie\n\nT  Stopped (on a signal) or (before Linux 2.6.33) trace stopped\n\nt  Tracing stop (Linux 2.6.33 onward)\n\nW  Paging (only before Linux 2.6.0)\n\nX  Dead (from Linux 2.6.0 onward)\n\nx  Dead (Linux 2.6.33 to 3.13 only)\n\nK  Wakekill (Linux 2.6.33 to 3.13 only)\n\nW  Waking (Linux 2.6.33 to 3.13 only)\n\nP  Parked (Linux 3.9 to 3.13 only)\n\n(4) ppid  %d\nThe PID of the parent of this process.\n\n(5) pgrp  %d\nThe process group ID of the process.\n\n(6) session  %d\nThe session ID of the process.\n\n(7) ttynr  %d\nThe controlling terminal of the process.  (The minor device number is contained\nin the combination of bits 31 to 20 and 7 to 0; the major device number  is  in\nbits 15 to 8.)\n\n(8) tpgid  %d\nThe  ID  of  the  foreground  process  group of the controlling terminal of the\nprocess.\n\n(9) flags  %u\nThe kernel flags word of the process.  For bit meanings, see the  PF*  defines\nin  the  Linux kernel source file include/linux/sched.h.  Details depend on the\nkernel version.\n\nThe format for this field was %lu before Linux 2.6.\n\n(10) minflt  %lu\nThe number of minor faults the process has made which have not required loading\na memory page from disk.\n\n(11) cminflt  %lu\nThe number of minor faults that the process's waited-for children have made.\n\n(12) majflt  %lu\nThe  number  of major faults the process has made which have required loading a\nmemory page from disk.\n\n(13) cmajflt  %lu\nThe number of major faults that the process's waited-for children have made.\n\n(14) utime  %lu\nAmount of time that this process has been scheduled in user mode,  measured  in\nclock  ticks  (divide  by  sysconf(SCCLKTCK)).   This  includes  guest time,\nguesttime (time spent running a virtual CPU, see below), so that  applications\nthat  are  not  aware  of the guest time field do not lose that time from their\ncalculations.\n\n(15) stime  %lu\nAmount of time that this process has been scheduled in kernel mode, measured in\nclock ticks (divide by sysconf(SCCLKTCK)).\n\n(16) cutime  %ld\nAmount  of  time that this process's waited-for children have been scheduled in\nuser mode, measured in clock ticks (divide by sysconf(SCCLKTCK)).  (See also\ntimes(2).)  This includes guest time, cguesttime (time spent running a virtual\nCPU, see below).\n\n(17) cstime  %ld\nAmount of time that this process's waited-for children have been  scheduled  in\nkernel mode, measured in clock ticks (divide by sysconf(SCCLKTCK)).\n\n(18) priority  %ld\n(Explanation for Linux 2.6) For processes running a real-time scheduling policy\n(policy below; see schedsetscheduler(2)), this is the negated scheduling  pri‐\nority,  minus  one; that is, a number in the range -2 to -100, corresponding to\nreal-time priorities 1 to 99.  For  processes  running  under  a  non-real-time\nscheduling  policy,  this is the raw nice value (setpriority(2)) as represented\nin the kernel.  The kernel stores nice values as numbers in the range 0  (high)\nto 39 (low), corresponding to the user-visible nice range of -20 to 19.\n\nBefore  Linux  2.6,  this  was  a scaled value based on the scheduler weighting\ngiven to this process.\n\n(19) nice  %ld\nThe nice value (see setpriority(2)), a value in the range 19 (low priority)  to\n-20 (high priority).\n\n(20) numthreads  %ld\nNumber  of  threads in this process (since Linux 2.6).  Before kernel 2.6, this\nfield was hard coded to 0 as a placeholder for an earlier removed field.\n\n(21) itrealvalue  %ld\nThe time in jiffies before the next SIGALRM is sent to the process  due  to  an\ninterval  timer.   Since kernel 2.6.17, this field is no longer maintained, and\nis hard coded as 0.\n\n(22) starttime  %llu\nThe time the process started after system boot.  In kernels before  Linux  2.6,\nthis  value  was expressed in jiffies.  Since Linux 2.6, the value is expressed\nin clock ticks (divide by sysconf(SCCLKTCK)).\n\nThe format for this field was %lu before Linux 2.6.\n\n(23) vsize  %lu\nVirtual memory size in bytes.\n\n(24) rss  %ld\nResident Set Size: number of pages the process has in  real  memory.   This  is\njust  the  pages  which count toward text, data, or stack space.  This does not\ninclude pages which have not been demand-loaded in, or which are  swapped  out.\nThis value is inaccurate; see /proc/[pid]/statm below.\n\n(25) rsslim  %lu\nCurrent  soft  limit in bytes on the rss of the process; see the description of\nRLIMITRSS in getrlimit(2).\n\n(26) startcode  %lu  [PT]\nThe address above which program text can run.\n\n(27) endcode  %lu  [PT]\nThe address below which program text can run.\n\n(28) startstack  %lu  [PT]\nThe address of the start (i.e., bottom) of the stack.\n\n(29) kstkesp  %lu  [PT]\nThe current value of ESP (stack pointer), as found in the kernel stack page for\nthe process.\n\n(30) kstkeip  %lu  [PT]\nThe current EIP (instruction pointer).\n\n(31) signal  %lu\nThe  bitmap  of  pending signals, displayed as a decimal number.  Obsolete, be‐\ncause  it  does   not   provide   information   on   real-time   signals;   use\n/proc/[pid]/status instead.\n\n(32) blocked  %lu\nThe  bitmap  of  blocked signals, displayed as a decimal number.  Obsolete, be‐\ncause  it  does   not   provide   information   on   real-time   signals;   use\n/proc/[pid]/status instead.\n\n(33) sigignore  %lu\nThe  bitmap  of  ignored signals, displayed as a decimal number.  Obsolete, be‐\ncause  it  does   not   provide   information   on   real-time   signals;   use\n/proc/[pid]/status instead.\n\n(34) sigcatch  %lu\nThe bitmap of caught signals, displayed as a decimal number.  Obsolete, because\nit does not provide information on real-time  signals;  use  /proc/[pid]/status\ninstead.\n\n(35) wchan  %lu  [PT]\nThis  is the \"channel\" in which the process is waiting.  It is the address of a\nlocation in the kernel where the process is sleeping.  The  corresponding  sym‐\nbolic name can be found in /proc/[pid]/wchan.\n\n(36) nswap  %lu\nNumber of pages swapped (not maintained).\n\n(37) cnswap  %lu\nCumulative nswap for child processes (not maintained).\n\n(38) exitsignal  %d  (since Linux 2.1.22)\nSignal to be sent to parent when we die.\n\n(39) processor  %d  (since Linux 2.2.8)\nCPU number last executed on.\n\n(40) rtpriority  %u  (since Linux 2.5.19)\nReal-time  scheduling  priority,  a  number  in the range 1 to 99 for processes\nscheduled under a real-time policy, or  0,  for  non-real-time  processes  (see\nschedsetscheduler(2)).\n\n(41) policy  %u  (since Linux 2.5.19)\nScheduling  policy  (see schedsetscheduler(2)).  Decode using the SCHED* con‐\nstants in linux/sched.h.\n\nThe format for this field was %lu before Linux 2.6.22.\n\n(42) delayacctblkioticks  %llu  (since Linux 2.6.18)\nAggregated block I/O delays, measured in clock ticks (centiseconds).\n\n(43) guesttime  %lu  (since Linux 2.6.24)\nGuest time of the process (time spent running a virtual CPU for a guest operat‐\ning system), measured in clock ticks (divide by sysconf(SCCLKTCK)).\n\n(44) cguesttime  %ld  (since Linux 2.6.24)\nGuest  time  of  the  process's  children,  measured  in clock ticks (divide by\nsysconf(SCCLKTCK)).\n\n(45) startdata  %lu  (since Linux 3.3)  [PT]\nAddress above which  program  initialized  and  uninitialized  (BSS)  data  are\nplaced.\n\n(46) enddata  %lu  (since Linux 3.3)  [PT]\nAddress  below  which  program  initialized  and  uninitialized  (BSS) data are\nplaced.\n\n(47) startbrk  %lu  (since Linux 3.3)  [PT]\nAddress above which program heap can be expanded with brk(2).\n\n(48) argstart  %lu  (since Linux 3.5)  [PT]\nAddress above which program command-line arguments (argv) are placed.\n\n(49) argend  %lu  (since Linux 3.5)  [PT]\nAddress below program command-line arguments (argv) are placed.\n\n(50) envstart  %lu  (since Linux 3.5)  [PT]\nAddress above which program environment is placed.\n\n(51) envend  %lu  (since Linux 3.5)  [PT]\nAddress below which program environment is placed.\n\n(52) exitcode  %d  (since Linux 3.5)  [PT]\nThe thread's exit status in the form reported by waitpid(2).\n\n/proc/[pid]/statm\nProvides information about memory usage, measured in pages.  The columns are:\n\nsize       (1) total program size\n(same as VmSize in /proc/[pid]/status)\nresident   (2) resident set size\n(inaccurate; same as VmRSS in /proc/[pid]/status)\nshared     (3) number of resident shared pages\n(i.e., backed by a file)\n(inaccurate; same as RssFile+RssShmem in\n/proc/[pid]/status)\ntext       (4) text (code)\nlib        (5) library (unused since Linux 2.6; always 0)\ndata       (6) data + stack\ndt         (7) dirty pages (unused since Linux 2.6; always 0)\n\nSome of these values are inaccurate because of a kernel-internal scalability optimiza‐\ntion.     If    accurate    values    are    required,    use   /proc/[pid]/smaps   or\n/proc/[pid]/smapsrollup instead, which are much slower but provide accurate, detailed\ninformation.\n\n/proc/[pid]/status\nProvides much of the information in /proc/[pid]/stat and /proc/[pid]/statm in a format\nthat's easier for humans to parse.  Here's an example:\n\n$ cat /proc/$$/status\nName:   bash\nUmask:  0022\nState:  S (sleeping)\nTgid:   17248\nNgid:   0\nPid:    17248\nPPid:   17200\nTracerPid:      0\nUid:    1000    1000    1000    1000\nGid:    100     100     100     100\nFDSize: 256\nGroups: 16 33 100\nNStgid: 17248\nNSpid:  17248\nNSpgid: 17248\nNSsid:  17200\nVmPeak:     131168 kB\nVmSize:     131168 kB\nVmLck:           0 kB\nVmPin:           0 kB\nVmHWM:       13484 kB\nVmRSS:       13484 kB\nRssAnon:     10264 kB\nRssFile:      3220 kB\nRssShmem:        0 kB\nVmData:      10332 kB\nVmStk:         136 kB\nVmExe:         992 kB\nVmLib:        2104 kB\nVmPTE:          76 kB\nVmPMD:          12 kB\nVmSwap:          0 kB\nHugetlbPages:          0 kB        # 4.4\nCoreDumping:   0                       # 4.15\nThreads:        1\nSigQ:   0/3067\nSigPnd: 0000000000000000\nShdPnd: 0000000000000000\nSigBlk: 0000000000010000\nSigIgn: 0000000000384004\nSigCgt: 000000004b813efb\nCapInh: 0000000000000000\nCapPrm: 0000000000000000\nCapEff: 0000000000000000\nCapBnd: ffffffffffffffff\nCapAmb:   0000000000000000\nNoNewPrivs:     0\nSeccomp:        0\nSpeculationStoreBypass:       vulnerable\nCpusallowed:   00000001\nCpusallowedlist:      0\nMemsallowed:   1\nMemsallowedlist:      0\nvoluntaryctxtswitches:        150\nnonvoluntaryctxtswitches:     545\n\nThe fields are as follows:\n\nName   Command run by this process.  Strings longer than TASKCOMMLEN (16) characters\n(including the terminating null byte) are silently truncated.\n\nUmask  Process  umask,  expressed  in octal with a leading zero; see umask(2).  (Since\nLinux 4.7.)\n\nState  Current state of the process.  One of \"R (running)\", \"S (sleeping)\",  \"D  (disk\nsleep)\", \"T (stopped)\", \"t (tracing stop)\", \"Z (zombie)\", or \"X (dead)\".\n\nTgid   Thread group ID (i.e., Process ID).\n\nNgid   NUMA group ID (0 if none; since Linux 3.13).\n\nPid    Thread ID (see gettid(2)).\n\nPPid   PID of parent process.\n\nTracerPid\nPID of process tracing this process (0 if not being traced).\n\nUid, Gid\nReal, effective, saved set, and filesystem UIDs (GIDs).\n\nFDSize Number of file descriptor slots currently allocated.\n\nGroups Supplementary group list.\n\nNStgid Thread  group  ID (i.e., PID) in each of the PID namespaces of which [pid] is a\nmember.  The leftmost entry shows the value with respect to the  PID  namespace\nof  the  process  that mounted this procfs (or the root namespace if mounted by\nthe kernel), followed by the value in  successively  nested  inner  namespaces.\n(Since Linux 4.1.)\n\nNSpid  Thread ID in each of the PID namespaces of which [pid] is a member.  The fields\nare ordered as for NStgid.  (Since Linux 4.1.)\n\nNSpgid Process group ID in each of the PID namespaces of which [pid] is a member.  The\nfields are ordered as for NStgid.  (Since Linux 4.1.)\n\nNSsid  descendant  namespace  session ID hierarchy Session ID in each of the PID name‐\nspaces of which [pid] is a member.  The  fields  are  ordered  as  for  NStgid.\n(Since Linux 4.1.)\n\nVmPeak Peak virtual memory size.\n\nVmSize Virtual memory size.\n\nVmLck  Locked memory size (see mlock(2)).\n\nVmPin  Pinned  memory size (since Linux 3.2).  These are pages that can't be moved be‐\ncause something needs to directly access physical memory.\n\nVmHWM  Peak resident set size (\"high water mark\").   This  value  is  inaccurate;  see\n/proc/[pid]/statm above.\n\nVmRSS  Resident  set  size.   Note that the value here is the sum of RssAnon, RssFile,\nand RssShmem.  This value is inaccurate; see /proc/[pid]/statm above.\n\nRssAnon\nSize of resident anonymous memory.  (since Linux 4.5).  This value  is  inaccu‐\nrate; see /proc/[pid]/statm above.\n\nRssFile\nSize  of resident file mappings.  (since Linux 4.5).  This value is inaccurate;\nsee /proc/[pid]/statm above.\n\nRssShmem\nSize of resident shared memory (includes System V shared memory, mappings  from\ntmpfs(5), and shared anonymous mappings).  (since Linux 4.5).\n\nVmData, VmStk, VmExe\nSize  of  data,  stack,  and  text  segments.   This  value  is inaccurate; see\n/proc/[pid]/statm above.\n\nVmLib  Shared library code size.\n\nVmPTE  Page table entries size (since Linux 2.6.10).\n\nVmPMD  Size of second-level page tables (added in Linux 4.0; removed in Linux 4.15).\n\nVmSwap Swapped-out virtual memory size by anonymous private pages; shmem swap usage is\nnot   included   (since   Linux   2.6.34).    This  value  is  inaccurate;  see\n/proc/[pid]/statm above.\n\nHugetlbPages\nSize of hugetlb memory portions (since Linux 4.4).\n\nCoreDumping\nContains the value 1 if the process is currently dumping core, and 0 if  it  is\nnot  (since  Linux 4.15).  This information can be used by a monitoring process\nto avoid killing a process that is currently dumping core, which  could  result\nin a corrupted core dump file.\n\nThreads\nNumber of threads in process containing this thread.\n\nSigQ   This  field  contains two slash-separated numbers that relate to queued signals\nfor the real user ID of this process.  The first of these is the number of cur‐\nrently  queued  signals  for  this real user ID, and the second is the resource\nlimit on the number of queued signals for this process (see the description  of\nRLIMITSIGPENDING in getrlimit(2)).\n\nSigPnd, ShdPnd\nMask  (expressed  in hexadecimal) of signals pending for thread and for process\nas a whole (see pthreads(7) and signal(7)).\n\nSigBlk, SigIgn, SigCgt\nMasks (expressed in hexadecimal) indicating signals being blocked, ignored, and\ncaught (see signal(7)).\n\nCapInh, CapPrm, CapEff\nMasks  (expressed  in hexadecimal) of capabilities enabled in inheritable, per‐\nmitted, and effective sets (see capabilities(7)).\n\nCapBnd Capability bounding set, expressed in hexadecimal (since Linux 2.6.26, see  ca‐‐\npabilities(7)).\n\nCapAmb Ambient capability set, expressed in hexadecimal (since Linux 4.3, see capabil‐‐\nities(7)).\n\nNoNewPrivs\nValue of the nonewprivs bit (since Linux 4.10, see prctl(2)).\n\nSeccomp\nSeccomp mode of the process (since Linux 3.8, see seccomp(2)).   0  means  SEC‐‐\nCOMPMODEDISABLED;  1  means SECCOMPMODESTRICT; 2 means SECCOMPMODEFILTER.\nThis field is provided only if the kernel was  built  with  the  CONFIGSECCOMP\nkernel configuration option enabled.\n\nSpeculationStoreBypass\nSpeculation flaw mitigation state (since Linux 4.17, see prctl(2)).\n\nCpusallowed\nHexadecimal mask of CPUs on which this process may run (since Linux 2.6.24, see\ncpuset(7)).\n\nCpusallowedlist\nSame as previous, but in \"list format\" (since Linux 2.6.26, see cpuset(7)).\n\nMemsallowed\nMask of  memory  nodes  allowed  to  this  process  (since  Linux  2.6.24,  see\ncpuset(7)).\n\nMemsallowedlist\nSame as previous, but in \"list format\" (since Linux 2.6.26, see cpuset(7)).\n\nvoluntaryctxtswitches, nonvoluntaryctxtswitches\nNumber of voluntary and involuntary context switches (since Linux 2.6.23).\n\n/proc/[pid]/syscall (since Linux 2.6.27)\nThis  file  exposes  the system call number and argument registers for the system call\ncurrently being executed by the process, followed by the values of the  stack  pointer\nand  program counter registers.  The values of all six argument registers are exposed,\nalthough most system calls use fewer registers.\n\nIf the process is blocked, but not in a system call, then  the  file  displays  -1  in\nplace  of the system call number, followed by just the values of the stack pointer and\nprogram counter.  If process is not blocked, then the file contains  just  the  string\n\"running\".\n\nThis  file  is  present only if the kernel was configured with CONFIGHAVEARCHTRACE‐‐\nHOOK.\n\nPermission to access this file is governed by a  ptrace  access  mode  PTRACEMODEAT‐‐\nTACHFSCREDS check; see ptrace(2).\n\n/proc/[pid]/task (since Linux 2.6.0)\nThis  is  a  directory  that contains one subdirectory for each thread in the process.\nThe name of each subdirectory is the numerical thread ID ([tid]) of  the  thread  (see\ngettid(2)).\n\nWithin  each  of these subdirectories, there is a set of files with the same names and\ncontents as under the /proc/[pid] directories.  For attributes that are shared by  all\nthreads,  the  contents for each of the files under the task/[tid] subdirectories will\nbe the same as in the corresponding file in the parent /proc/[pid] directory (e.g., in\na  multithreaded  process, all of the task/[tid]/cwd files will have the same value as\nthe /proc/[pid]/cwd file in the parent directory,  since  all  of  the  threads  in  a\nprocess share a working directory).  For attributes that are distinct for each thread,\nthe corresponding files under task/[tid] may  have  different  values  (e.g.,  various\nfields  in  each  of the task/[tid]/status files may be different for each thread), or\nthey might not exist in /proc/[pid] at all.\n\nIn a multithreaded process, the contents of the  /proc/[pid]/task  directory  are  not\navailable   if   the   main  thread  has  already  terminated  (typically  by  calling\npthreadexit(3)).\n\n/proc/[pid]/task/[tid]/children (since Linux 3.5)\nA space-separated list of child tasks of this task.  Each child task is represented by\nits TID.\n\nThis  option is intended for use by the checkpoint-restore (CRIU) system, and reliably\nprovides a list of children only if all of the child processes are stopped or  frozen.\nIt  does not work properly if children of the target task exit while the file is being\nread!  Exiting children may cause non-exiting children to be omitted  from  the  list.\nThis  makes  this  interface even more unreliable than classic PID-based approaches if\nthe inspected task and its children aren't frozen, and most code should  probably  not\nuse this interface.\n\nUntil  Linux  4.2, the presence of this file was governed by the CONFIGCHECKPOINTRE‐‐\nSTORE kernel configuration option.  Since Linux  4.2,  it  is  governed  by  the  CON‐‐\nFIGPROCCHILDREN option.\n\n/proc/[pid]/timers (since Linux 3.10)\nA  list  of  the POSIX timers for this process.  Each timer is listed with a line that\nstarts with the string \"ID:\".  For example:\n\nID: 1\nsignal: 60/00007fff86e452a8\nnotify: signal/pid.2634\nClockID: 0\nID: 0\nsignal: 60/00007fff86e452a8\nnotify: signal/pid.2634\nClockID: 1\n\nThe lines shown for each timer have the following meanings:\n\nID     The ID for this timer.  This is not the  same  as  the  timer  ID  returned  by\ntimercreate(2);  rather,  it  is the same kernel-internal ID that is available\nvia the sitimerid field of the siginfot structure (see sigaction(2)).\n\nsignal This is the signal number that this timer uses to  deliver  notifications  fol‐\nlowed  by  a  slash, and then the sigevvalue value supplied to the signal han‐\ndler.  Valid only for timers that notify via a signal.\n\nnotify The part before the slash specifies the mechanism that this timer uses  to  de‐\nliver  notifications, and is one of \"thread\", \"signal\", or \"none\".  Immediately\nfollowing the slash is either the string \"tid\" for timers with  SIGEVTHREADID\nnotification,  or  \"pid\" for timers that notify by other mechanisms.  Following\nthe \".\" is the PID of the process (or the kernel thread ID of the thread)  that\nwill be delivered a signal if the timer delivers notifications via a signal.\n\nClockID\nThis  field  identifies  the clock that the timer uses for measuring time.  For\nmost clocks, this is a number that matches one of the user-space  CLOCK*  con‐\nstants  exposed  via  <time.h>.  CLOCKPROCESSCPUTIMEID timers display with a\nvalue of -6 in this field.  CLOCKTHREADCPUTIMEID timers display with a value\nof -2 in this field.\n\nThis  file is available only when the kernel was configured with CONFIGCHECKPOINTRE‐‐\nSTORE.\n\n/proc/[pid]/timerslackns (since Linux 4.6)\nThis file exposes the process's \"current\" timer slack value, expressed in nanoseconds.\nThe file is writable, allowing the process's timer slack value to be changed.  Writing\n0 to this file resets the \"current\" timer slack to the \"default\"  timer  slack  value.\nFor further details, see the discussion of PRSETTIMERSLACK in prctl(2).\n\nInitially,  permission  to  access  this  file  was  governed  by a ptrace access mode\nPTRACEMODEATTACHFSCREDS check (see  ptrace(2)).   However,  this  was  subsequently\ndeemed  too  strict a requirement (and had the side effect that requiring a process to\nhave the CAPSYSPTRACE capability  would  also  allow  it  to  view  and  change  any\nprocess's  memory).   Therefore, since Linux 4.9, only the (weaker) CAPSYSNICE capa‐\nbility is required to access this file.\n\n/proc/[pid]/uidmap, /proc/[pid]/gidmap (since Linux 3.5)\nSee usernamespaces(7).\n\n/proc/[pid]/wchan (since Linux 2.6.0)\nThe symbolic name corresponding to the location in the kernel  where  the  process  is\nsleeping.\n\nPermission   to   access   this   file   is   governed   by   a   ptrace  access  mode\nPTRACEMODEREADFSCREDS check; see ptrace(2).\n\n/proc/[tid]\nThere  is a numerical subdirectory for each running thread that is not a thread  group\nleader  (i.e., a thread whose thread ID is not the same as its process ID); the subdi‐\nrectory is named by the thread ID.  Each one of these  subdirectories  contains  files\nand  subdirectories exposing information about the thread with the thread ID tid.  The\ncontents of these directories are the same as the corresponding /proc/[pid]/task/[tid]\ndirectories.\n\nThe  /proc/[tid] subdirectories are not visible when iterating through /proc with get‐‐\ndents(2) (and thus are not visible when one uses ls(1) to view the contents of /proc).\nHowever,  the pathnames of these directories are visible to (i.e., usable as arguments\nin) system calls that operate on pathnames.\n\n/proc/apm\nAdvanced power management version and battery information when CONFIGAPM  is  defined\nat kernel compilation time.\n\n/proc/buddyinfo\nThis  file  contains information which is used for diagnosing memory fragmentation is‐\nsues.  Each line starts with the identification of the node and the name of  the  zone\nwhich together identify a memory region.  This is then followed by the count of avail‐\nable chunks of a certain order in which these zones are split.  The size in bytes of a\ncertain order is given by the formula:\n\n(2^order) * PAGESIZE\n\nThe  binary  buddy allocator algorithm inside the kernel will split one chunk into two\nchunks of a smaller order (thus with half the size) or combine two  contiguous  chunks\ninto one larger chunk of a higher order (thus with double the size) to satisfy alloca‐\ntion requests and to counter memory fragmentation.  The order matches the column  num‐\nber, when starting to count at zero.\n\nFor example on an x86-64 system:\nNode 0, zone     DMA     1    1    1    0    2    1    1    0    1    1    3\nNode 0, zone   DMA32    65   47    4   81   52   28   13   10    5    1  404\nNode 0, zone  Normal   216   55  189  101   84   38   37   27    5    3  587\n\nIn  this  example, there is one node containing three zones and there are 11 different\nchunk sizes.  If the page size is 4 kilobytes, then the first zone called DMA (on  x86\nthe  first  16  megabyte of memory) has 1 chunk of 4 kilobytes (order 0) available and\nhas 3 chunks of 4 megabytes (order 10) available.\n\nIf the memory is heavily fragmented, the counters for higher order chunks will be zero\nand allocation of large contiguous areas will fail.\n\nFurther information about the zones can be found in /proc/zoneinfo.\n\n/proc/bus\nContains subdirectories for installed busses.\n\n/proc/bus/pccard\nSubdirectory for PCMCIA devices when CONFIGPCMCIA is set at kernel compilation time.\n\n/proc/bus/pccard/drivers\n\n/proc/bus/pci\nContains  various bus subdirectories and pseudo-files containing information about PCI\nbusses, installed devices, and device drivers.  Some of these files are not ASCII.\n\n/proc/bus/pci/devices\nInformation about PCI devices.  They may be accessed through lspci(8) and setpci(8).\n\n/proc/cgroups (since Linux 2.6.24)\nSee cgroups(7).\n\n/proc/cmdline\nArguments passed to the Linux kernel at boot time.  Often done via a boot manager such\nas lilo(8) or grub(8).\n\n/proc/config.gz (since Linux 2.6)\nThis file exposes the configuration options that were used to build the currently run‐\nning kernel, in the same format as they would be shown in the .config  file  that  re‐\nsulted when configuring the kernel (using make xconfig, make config, or similar).  The\nfile contents are compressed; view or search them using zcat(1) and zgrep(1).  As long\nas  no  changes  have been made to the following file, the contents of /proc/config.gz\nare the same as those provided by:\n\ncat /lib/modules/$(uname -r)/build/.config\n\n/proc/config.gz is provided only  if  the  kernel  is  configured  with  CONFIGIKCON‐‐\nFIGPROC.\n\n/proc/crypto\nA  list of the ciphers provided by the kernel crypto API.  For details, see the kernel\nLinux Kernel Crypto API documentation available under the kernel source directory Doc‐\numentation/crypto/  (or  Documentation/DocBook  before  4.10; the documentation can be\nbuilt using a command such as make htmldocs in the root directory of the kernel source\ntree).\n\n/proc/cpuinfo\nThis  is  a  collection  of CPU and system architecture dependent items, for each sup‐\nported architecture a different list.  Two common entries are  processor  which  gives\nCPU  number  and bogomips; a system constant that is calculated during kernel initial‐\nization.  SMP machines have information for each CPU.  The  lscpu(1)  command  gathers\nits information from this file.\n\n/proc/devices\nText  listing of major numbers and device groups.  This can be used by MAKEDEV scripts\nfor consistency with the kernel.\n\n/proc/diskstats (since Linux 2.5.69)\nThis file contains disk I/O statistics for each disk device.   See  the  Linux  kernel\nsource file Documentation/iostats.txt for further information.\n\n/proc/dma\nThis is a list of the registered ISA DMA (direct memory access) channels in use.\n\n/proc/driver\nEmpty subdirectory.\n\n/proc/execdomains\nList of the execution domains (ABI personalities).\n\n/proc/fb\nFrame buffer information when CONFIGFB is defined during kernel compilation.\n\n/proc/filesystems\nA  text  listing of the filesystems which are supported by the kernel, namely filesys‐\ntems which were compiled into the kernel or whose kernel modules are currently loaded.\n(See also filesystems(5).)  If a filesystem is marked with \"nodev\", this means that it\ndoes not require a block device to  be  mounted  (e.g.,  virtual  filesystem,  network\nfilesystem).\n\nIncidentally, this file may be used by mount(8) when no filesystem is specified and it\ndidn't manage to determine the filesystem type.  Then filesystems  contained  in  this\nfile are tried (excepted those that are marked with \"nodev\").\n\n/proc/fs\nContains  subdirectories  that  in turn contain files with information about (certain)\nmounted filesystems.\n\n/proc/ide\nThis directory exists on systems with the IDE bus.  There are directories for each IDE\nchannel and attached device.  Files include:\n\ncache              buffer size in KB\ncapacity           number of sectors\ndriver             driver version\ngeometry           physical and logical geometry\nidentify           in hexadecimal\nmedia              media type\nmodel              manufacturer's model number\nsettings           drive settings\nsmartthresholds   IDE disk management thresholds (in hex)\nsmartvalues       IDE disk management values (in hex)\n\nThe hdparm(8) utility provides access to this information in a friendly format.\n\n/proc/interrupts\nThis  is  used  to record the number of interrupts per CPU per IO device.  Since Linux\n2.6.24, for the i386 and x86-64 architectures, at least, this also includes interrupts\ninternal  to  the  system (that is, not associated with a device as such), such as NMI\n(nonmaskable interrupt), LOC (local timer interrupt), and for SMP  systems,  TLB  (TLB\nflush  interrupt), RES (rescheduling interrupt), CAL (remote function call interrupt),\nand possibly others.  Very easy to read formatting, done in ASCII.\n\n/proc/iomem\nI/O memory map in Linux 2.4.\n\n/proc/ioports\nThis is a list of currently registered Input-Output port regions that are in use.\n\n/proc/kallsyms (since Linux 2.5.71)\nThis holds the kernel exported symbol definitions used by the modules(X) tools to  dy‐\nnamically link and bind loadable modules.  In Linux 2.5.47 and earlier, a similar file\nwith slightly different syntax was named ksyms.\n\n/proc/kcore\nThis file represents the physical memory of the system and is stored in the  ELF  core\nfile format.  With this pseudo-file, and an unstripped kernel (/usr/src/linux/vmlinux)\nbinary, GDB can be used to examine the current state of any kernel data structures.\n\nThe total length of the file is the size of physical memory (RAM) plus 4 KiB.\n\n/proc/keys (since Linux 2.6.10)\nSee keyrings(7).\n\n/proc/key-users (since Linux 2.6.10)\nSee keyrings(7).\n\n/proc/kmsg\nThis file can be used instead of the syslog(2) system call to read kernel messages.  A\nprocess  must have superuser privileges to read this file, and only one process should\nread this file.  This file should not be read if a syslog  process  is  running  which\nuses the syslog(2) system call facility to log kernel messages.\n\nInformation in this file is retrieved with the dmesg(1) program.\n\n/proc/kpagecgroup (since Linux 4.3)\nThis file contains a 64-bit inode number of the memory cgroup each page is charged to,\nindexed by page frame number (see the discussion of /proc/[pid]/pagemap).\n\nThe /proc/kpagecgroup file is present only if the  CONFIGMEMCG  kernel  configuration\noption is enabled.\n\n/proc/kpagecount (since Linux 2.6.25)\nThis  file  contains a 64-bit count of the number of times each physical page frame is\nmapped, indexed by page frame number (see the discussion of /proc/[pid]/pagemap).\n\nThe /proc/kpagecount file is present only if the CONFIGPROCPAGEMONITOR kernel  con‐\nfiguration option is enabled.\n\n/proc/kpageflags (since Linux 2.6.25)\nThis  file  contains 64-bit masks corresponding to each physical page frame; it is in‐\ndexed by page frame number (see the discussion of /proc/[pid]/pagemap).  The bits  are\nas follows:\n\n0 - KPFLOCKED\n1 - KPFERROR\n2 - KPFREFERENCED\n3 - KPFUPTODATE\n4 - KPFDIRTY\n5 - KPFLRU\n6 - KPFACTIVE\n7 - KPFSLAB\n8 - KPFWRITEBACK\n9 - KPFRECLAIM\n10 - KPFBUDDY\n11 - KPFMMAP           (since Linux 2.6.31)\n12 - KPFANON           (since Linux 2.6.31)\n13 - KPFSWAPCACHE      (since Linux 2.6.31)\n14 - KPFSWAPBACKED     (since Linux 2.6.31)\n15 - KPFCOMPOUNDHEAD  (since Linux 2.6.31)\n16 - KPFCOMPOUNDTAIL  (since Linux 2.6.31)\n17 - KPFHUGE           (since Linux 2.6.31)\n18 - KPFUNEVICTABLE    (since Linux 2.6.31)\n19 - KPFHWPOISON       (since Linux 2.6.31)\n20 - KPFNOPAGE         (since Linux 2.6.31)\n21 - KPFKSM            (since Linux 2.6.32)\n22 - KPFTHP            (since Linux 3.4)\n23 - KPFBALLOON        (since Linux 3.18)\n24 - KPFZEROPAGE      (since Linux 4.0)\n25 - KPFIDLE           (since Linux 4.3)\n\nFor further details on the meanings of these bits, see the kernel source file Documen‐\ntation/admin-guide/mm/pagemap.rst.  Before kernel 2.6.29, KPFWRITEBACK,  KPFRECLAIM,\nKPFBUDDY, and KPFLOCKED did not report correctly.\n\nThe  /proc/kpageflags file is present only if the CONFIGPROCPAGEMONITOR kernel con‐\nfiguration option is enabled.\n\n/proc/ksyms (Linux 1.1.23–2.5.47)\nSee /proc/kallsyms.\n\n/proc/loadavg\nThe first three fields in this file are load average figures giving the number of jobs\nin  the  run queue (state R) or waiting for disk I/O (state D) averaged over 1, 5, and\n15 minutes.  They are the same as the load average  numbers  given  by  uptime(1)  and\nother  programs.   The  fourth field consists of two numbers separated by a slash (/).\nThe first of these is the number of  currently  runnable  kernel  scheduling  entities\n(processes,  threads).   The  value after the slash is the number of kernel scheduling\nentities that currently exist on the system.  The  fifth  field  is  the  PID  of  the\nprocess that was most recently created on the system.\n\n/proc/locks\nThis file shows current file locks (flock(2) and fcntl(2)) and leases (fcntl(2)).\n\nAn example of the content shown in this file is the following:\n\n1: POSIX  ADVISORY  READ  5433 08:01:7864448 128 128\n2: FLOCK  ADVISORY  WRITE 2001 08:01:7864554 0 EOF\n3: FLOCK  ADVISORY  WRITE 1568 00:2f:32388 0 EOF\n4: POSIX  ADVISORY  WRITE 699 00:16:28457 0 EOF\n5: POSIX  ADVISORY  WRITE 764 00:16:21448 0 0\n6: POSIX  ADVISORY  READ  3548 08:01:7867240 1 1\n7: POSIX  ADVISORY  READ  3548 08:01:7865567 1826 2335\n8: OFDLCK ADVISORY  WRITE -1 08:01:8713209 128 191\n\nThe fields shown in each line are as follows:\n\n(1) The ordinal position of the lock in the list.\n\n(2) The lock type.  Values that may appear here include:\n\nFLOCK  This is a BSD file lock created using flock(2).\n\nOFDLCK This is an open file description (OFD) lock created using fcntl(2).\n\nPOSIX  This is a POSIX byte-range lock created using fcntl(2).\n\n(3) Among the strings that can appear here are the following:\n\nADVISORY\nThis is an advisory lock.\n\nMANDATORY\nThis is a mandatory lock.\n\n(4) The type of lock.  Values that can appear here are:\n\nREAD   This is a POSIX or OFD read lock, or a BSD shared lock.\n\nWRITE  This is a POSIX or OFD write lock, or a BSD exclusive lock.\n\n(5) The PID of the process that owns the lock.\n\nBecause  OFD locks are not owned by a single process (since multiple processes may\nhave file descriptors that refer to the same open file description), the value  -1\nis  displayed  in this field for OFD locks.  (Before kernel 4.14, a bug meant that\nthe PID of the process that initially acquired the lock was displayed  instead  of\nthe value -1.)\n\n(6) Three colon-separated subfields that identify the major and minor device ID of the\ndevice containing the filesystem where the locked file resides,  followed  by  the\ninode number of the locked file.\n\n(7) The  byte  offset of the first byte of the lock.  For BSD locks, this value is al‐\nways 0.\n\n(8) The byte offset of the last byte of the lock.  EOF in this field  means  that  the\nlock  extends  to  the  end of the file.  For BSD locks, the value shown is always\nEOF.\n\nSince Linux 4.9, the list of locks shown in /proc/locks is filtered to show  just  the\nlocks  for  the  processes  in the PID namespace (see pidnamespaces(7)) for which the\n/proc filesystem was mounted.  (In the initial PID namespace, there is no filtering of\nthe records shown in this file.)\n\nThe lslocks(8) command provides a bit more information about each lock.\n\n/proc/malloc (only up to and including Linux 2.2)\nThis file is present only if CONFIGDEBUGMALLOC was defined during compilation.\n\n/proc/meminfo\nThis  file reports statistics about memory usage on the system.  It is used by free(1)\nto report the amount of free and used memory (both physical and swap) on the system as\nwell  as the shared memory and buffers used by the kernel.  Each line of the file con‐\nsists of a parameter name, followed by a colon, the value of the parameter, and an op‐\ntion  unit  of measurement (e.g., \"kB\").  The list below describes the parameter names\nand the format specifier required to read the field value.  Except as noted below, all\nof the fields have been present since at least Linux 2.6.0.  Some fields are displayed\nonly if the kernel was configured with various options; those dependencies  are  noted\nin the list.\n\nMemTotal %lu\nTotal  usable  RAM (i.e., physical RAM minus a few reserved bits and the kernel\nbinary code).\n\nMemFree %lu\nThe sum of LowFree+HighFree.\n\nMemAvailable %lu (since Linux 3.14)\nAn estimate of how much memory is  available  for  starting  new  applications,\nwithout swapping.\n\nBuffers %lu\nRelatively  temporary  storage  for  raw disk blocks that shouldn't get tremen‐\ndously large (20 MB or so).\n\nCached %lu\nIn-memory cache for files read from the disk (the page cache).  Doesn't include\nSwapCached.\n\nSwapCached %lu\nMemory  that  once was swapped out, is swapped back in but still also is in the\nswap file.  (If memory pressure is high, these pages don't need to  be  swapped\nout again because they are already in the swap file.  This saves I/O.)\n\nActive %lu\nMemory  that has been used more recently and usually not reclaimed unless abso‐\nlutely necessary.\n\nInactive %lu\nMemory which has been less recently used.  It is more eligible to be  reclaimed\nfor other purposes.\n\nActive(anon) %lu (since Linux 2.6.28)\n[To be documented.]\n\nInactive(anon) %lu (since Linux 2.6.28)\n[To be documented.]\n\nActive(file) %lu (since Linux 2.6.28)\n[To be documented.]\n\nInactive(file) %lu (since Linux 2.6.28)\n[To be documented.]\n\nUnevictable %lu (since Linux 2.6.28)\n(From  Linux  2.6.28  to  2.6.30, CONFIGUNEVICTABLELRU was required.)  [To be\ndocumented.]\n\nMlocked %lu (since Linux 2.6.28)\n(From Linux 2.6.28 to 2.6.30, CONFIGUNEVICTABLELRU  was  required.)   [To  be\ndocumented.]\n\nHighTotal %lu\n(Starting  with  Linux  2.6.19,  CONFIGHIGHMEM  is required.)  Total amount of\nhighmem.  Highmem is all memory above ~860 MB of physical memory.  Highmem  ar‐\neas are for use by user-space programs, or for the page cache.  The kernel must\nuse tricks to access this memory, making it slower to access than lowmem.\n\nHighFree %lu\n(Starting with Linux 2.6.19, CONFIGHIGHMEM is required.)  Amount of free high‐\nmem.\n\nLowTotal %lu\n(Starting  with  Linux  2.6.19,  CONFIGHIGHMEM  is required.)  Total amount of\nlowmem.  Lowmem is memory which can be used for everything that highmem can  be\nused for, but it is also available for the kernel's use for its own data struc‐\ntures.  Among many other things, it is where everything from Slab is allocated.\nBad things happen when you're out of lowmem.\n\nLowFree %lu\n(Starting  with  Linux  2.6.19,  CONFIGHIGHMEM  is  required.)  Amount of free\nlowmem.\n\nMmapCopy %lu (since Linux 2.6.29)\n(CONFIGMMU is required.)  [To be documented.]\n\nSwapTotal %lu\nTotal amount of swap space available.\n\nSwapFree %lu\nAmount of swap space that is currently unused.\n\nDirty %lu\nMemory which is waiting to get written back to the disk.\n\nWriteback %lu\nMemory which is actively being written back to the disk.\n\nAnonPages %lu (since Linux 2.6.18)\nNon-file backed pages mapped into user-space page tables.\n\nMapped %lu\nFiles which have been mapped into memory (with mmap(2)), such as libraries.\n\nShmem %lu (since Linux 2.6.32)\nAmount of memory consumed in tmpfs(5) filesystems.\n\nKReclaimable %lu (since Linux 4.20)\nKernel allocations that the kernel will attempt to reclaim under  memory  pres‐\nsure.   Includes  SReclaimable  (below),  and  other  direct allocations with a\nshrinker.\n\nSlab %lu\nIn-kernel data structures cache.  (See slabinfo(5).)\n\nSReclaimable %lu (since Linux 2.6.19)\nPart of Slab, that might be reclaimed, such as caches.\n\nSUnreclaim %lu (since Linux 2.6.19)\nPart of Slab, that cannot be reclaimed on memory pressure.\n\nKernelStack %lu (since Linux 2.6.32)\nAmount of memory allocated to kernel stacks.\n\nPageTables %lu (since Linux 2.6.18)\nAmount of memory dedicated to the lowest level of page tables.\n\nQuicklists %lu (since Linux 2.6.27)\n(CONFIGQUICKLIST is required.)  [To be documented.]\n\nNFSUnstable %lu (since Linux 2.6.18)\nNFS pages sent to the server, but not yet committed to stable storage.\n\nBounce %lu (since Linux 2.6.18)\nMemory used for block device \"bounce buffers\".\n\nWritebackTmp %lu (since Linux 2.6.26)\nMemory used by FUSE for temporary writeback buffers.\n\nCommitLimit %lu (since Linux 2.6.10)\nThis is the total amount of memory currently available to be allocated  on  the\nsystem,  expressed in kilobytes.  This limit is adhered to only if strict over‐\ncommit accounting is enabled (mode 2 in  /proc/sys/vm/overcommitmemory).   The\nlimit is calculated according to the formula described under /proc/sys/vm/over‐\ncommitmemory.  For further details, see  the  kernel  source  file  Documenta‐\ntion/vm/overcommit-accounting.rst.\n\nCommittedAS %lu\nThe  amount  of memory presently allocated on the system.  The committed memory\nis a sum of all of the memory which has been allocated by processes, even if it\nhas  not been \"used\" by them as of yet.  A process which allocates 1 GB of mem‐\nory (using malloc(3) or similar), but touches only 300 MB of that  memory  will\nshow  up  as using only 300 MB of memory even if it has the address space allo‐\ncated for the entire 1 GB.\n\nThis 1 GB is memory which has been \"committed\" to by the VM and can be used  at\nany  time by the allocating application.  With strict overcommit enabled on the\nsystem (mode 2 in /proc/sys/vm/overcommitmemory), allocations which would  ex‐\nceed  the  CommitLimit  will  not be permitted.  This is useful if one needs to\nguarantee that processes will not fail due to lack of memory once  that  memory\nhas been successfully allocated.\n\nVmallocTotal %lu\nTotal size of vmalloc memory area.\n\nVmallocUsed %lu\nAmount of vmalloc area which is used.  Since Linux 4.4, this field is no longer\ncalculated, and is hard coded as 0.  See /proc/vmallocinfo.\n\nVmallocChunk %lu\nLargest contiguous block of vmalloc area which is free.  Since Linux 4.4,  this\nfield is no longer calculated and is hard coded as 0.  See /proc/vmallocinfo.\n\nHardwareCorrupted %lu (since Linux 2.6.32)\n(CONFIGMEMORYFAILURE is required.)  [To be documented.]\n\nLazyFree %lu (since Linux 4.12)\nShows the amount of memory marked by madvise(2) MADVFREE.\n\nAnonHugePages %lu (since Linux 2.6.38)\n(CONFIGTRANSPARENTHUGEPAGE  is  required.)  Non-file backed huge pages mapped\ninto user-space page tables.\n\nShmemHugePages %lu (since Linux 4.8)\n(CONFIGTRANSPARENTHUGEPAGE  is  required.)   Memory  used  by  shared  memory\n(shmem) and tmpfs(5) allocated with huge pages.\n\nShmemPmdMapped %lu (since Linux 4.8)\n(CONFIGTRANSPARENTHUGEPAGE  is  required.)   Shared  memory  mapped into user\nspace with huge pages.\n\nCmaTotal %lu (since Linux 3.1)\nTotal CMA (Contiguous Memory Allocator) pages.  (CONFIGCMA is required.)\n\nCmaFree %lu (since Linux 3.1)\nFree CMA (Contiguous Memory Allocator) pages.  (CONFIGCMA is required.)\n\nHugePagesTotal %lu\n(CONFIGHUGETLBPAGE is required.)  The size of the pool of huge pages.\n\nHugePagesFree %lu\n(CONFIGHUGETLBPAGE is required.)  The number of huge pages in the  pool  that\nare not yet allocated.\n\nHugePagesRsvd %lu (since Linux 2.6.17)\n(CONFIGHUGETLBPAGE  is required.)  This is the number of huge pages for which\na commitment to allocate from the pool has been made, but no allocation has yet\nbeen  made.   These  reserved  huge pages guarantee that an application will be\nable to allocate a huge page from the pool of huge pages at fault time.\n\nHugePagesSurp %lu (since Linux 2.6.24)\n(CONFIGHUGETLBPAGE is required.)  This is the number of  huge  pages  in  the\npool  above the value in /proc/sys/vm/nrhugepages.  The maximum number of sur‐\nplus huge pages is controlled by /proc/sys/vm/nrovercommithugepages.\n\nHugepagesize %lu\n(CONFIGHUGETLBPAGE is required.)  The size of huge pages.\n\nDirectMap4k %lu (since Linux 2.6.27)\nNumber of bytes of RAM linearly mapped by kernel in 4 kB pages.  (x86.)\n\nDirectMap4M %lu (since Linux 2.6.27)\nNumber of bytes of RAM linearly mapped by kernel in 4 MB pages.  (x86 with CON‐‐\nFIGX8664 or CONFIGX86PAE enabled.)\n\nDirectMap2M %lu (since Linux 2.6.27)\nNumber of bytes of RAM linearly mapped by kernel in 2 MB pages.  (x86 with nei‐\nther CONFIGX8664 nor CONFIGX86PAE enabled.)\n\nDirectMap1G %lu (since Linux 2.6.27)\n(x86 with CONFIGX8664 and CONFIGX86DIRECTGBPAGES enabled.)\n\n/proc/modules\nA text list of the modules that have been loaded by the system.  See also lsmod(8).\n\n/proc/mounts\nBefore kernel 2.4.19, this file was a list of all the filesystems currently mounted on\nthe  system.   With  the  introduction of per-process mount namespaces in Linux 2.4.19\n(see mountnamespaces(7)), this file became a link to /proc/self/mounts,  which  lists\nthe  mount  points  of  the process's own mount namespace.  The format of this file is\ndocumented in fstab(5).\n\n/proc/mtrr\nMemory  Type  Range  Registers.   See  the  Linux  kernel   source   file   Documenta‐\ntion/x86/mtrr.txt (or Documentation/mtrr.txt before Linux 2.6.28) for details.\n\n/proc/net\nThis  directory contains various files and subdirectories containing information about\nthe networking layer.  The files contain ASCII structures and are, therefore, readable\nwith  cat(1).   However, the standard netstat(8) suite provides much cleaner access to\nthese files.\n\nWith the advent of network namespaces, various information  relating  to  the  network\nstack is virtualized (see networknamespaces(7)).  Thus, since Linux 2.6.25, /proc/net\nis a symbolic link to the directory /proc/self/net, which contains the same files  and\ndirectories as listed below.  However, these files and directories now expose informa‐\ntion for the network namespace of which the process is a member.\n\n/proc/net/arp\nThis holds an ASCII readable dump of the kernel ARP table  used  for  address  resolu‐\ntions.  It will show both dynamically learned and preprogrammed ARP entries.  The for‐\nmat is:\n\nIP address     HW type   Flags     HW address          Mask   Device\n192.168.0.50   0x1       0x2       00:50:BF:25:68:F3   *      eth0\n192.168.0.250  0x1       0xc       00:00:00:00:00:00   *      eth0\n\nHere \"IP address\" is the IPv4 address of the machine and the \"HW type\" is the hardware\ntype  of the address from RFC 826.  The flags are the internal flags of the ARP struc‐\nture (as defined in /usr/include/linux/ifarp.h) and the \"HW address\" is the data link\nlayer mapping for that IP address if it is known.\n\n/proc/net/dev\nThe dev pseudo-file contains network device status information.  This gives the number\nof received and sent packets, the number of errors and collisions and other basic sta‐\ntistics.  These are used by the ifconfig(8) program to report device status.  The for‐\nmat is:\n\nInter-|   Receive                                                |  Transmit\nface |bytes    packets errs drop fifo frame compressed multicast|bytes    packets errs drop fifo colls carrier compressed\nlo: 2776770   11307    0    0    0     0          0         0  2776770   11307    0    0    0     0       0          0\neth0: 1215645    2751    0    0    0     0          0         0  1782404    4324    0    0    0   427       0          0\nppp0: 1622270    5552    1    0    0     0          0         0   354130    5669    0    0    0     0       0          0\ntap0:    7714      81    0    0    0     0          0         0     7714      81    0    0    0     0       0          0\n\n/proc/net/devmcast\nDefined in /usr/src/linux/net/core/devmcast.c:\n\nindx interfacename  dmiu dmig dmiaddress\n2    eth0            1     0     01005e000001\n3    eth1            1     0     01005e000001\n4    eth2            1     0     01005e000001\n\n/proc/net/igmp\nInternet Group Management Protocol.  Defined in /usr/src/linux/net/core/igmp.c.\n\n/proc/net/rarp\nThis file uses the same format as the arp file and contains the current  reverse  map‐\nping database used to provide rarp(8) reverse address lookup services.  If RARP is not\nconfigured into the kernel, this file will not be present.\n\n/proc/net/raw\nHolds a dump of the RAW socket table.  Much of the information is  not  of  use  apart\nfrom debugging.  The \"sl\" value is the kernel hash slot for the socket, the \"localad‐\ndress\" is the local address and protocol number pair.  \"St\" is the internal status  of\nthe socket.  The \"txqueue\" and \"rxqueue\" are the outgoing and incoming data queue in\nterms of kernel memory usage.  The \"tr\", \"tm->when\", and \"rexmits\" fields are not used\nby RAW.  The \"uid\" field holds the effective UID of the creator of the socket.\n\n/proc/net/snmp\nThis file holds the ASCII data needed for the IP, ICMP, TCP, and UDP management infor‐\nmation bases for an SNMP agent.\n\n/proc/net/tcp\nHolds a dump of the TCP socket table.  Much of the information is  not  of  use  apart\nfrom debugging.  The \"sl\" value is the kernel hash slot for the socket, the \"localad‐\ndress\" is the local address and port number pair.  The \"remaddress\" is the remote ad‐\ndress and port number pair (if connected).  \"St\" is the internal status of the socket.\nThe \"txqueue\" and \"rxqueue\" are the outgoing and incoming data  queue  in  terms  of\nkernel  memory usage.  The \"tr\", \"tm->when\", and \"rexmits\" fields hold internal infor‐\nmation of the kernel socket state and are useful only for debugging.  The \"uid\"  field\nholds the effective UID of the creator of the socket.\n\n/proc/net/udp\nHolds  a  dump  of  the UDP socket table.  Much of the information is not of use apart\nfrom debugging.  The \"sl\" value is the kernel hash slot for the socket, the \"localad‐\ndress\" is the local address and port number pair.  The \"remaddress\" is the remote ad‐\ndress and port number pair (if connected).  \"St\" is the internal status of the socket.\nThe  \"txqueue\"  and  \"rxqueue\"  are the outgoing and incoming data queue in terms of\nkernel memory usage.  The \"tr\", \"tm->when\", and \"rexmits\" fields are not used by  UDP.\nThe \"uid\" field holds the effective UID of the creator of the socket.  The format is:\n\nsl  localaddress remaddress   st txqueue rxqueue tr rexmits  tm->when uid\n1: 01642C89:0201 0C642C89:03FF 01 00000000:00000001 01:000071BA 00000000 0\n1: 00000000:0801 00000000:0000 0A 00000000:00000000 00:00000000 6F000100 0\n1: 00000000:0201 00000000:0000 0A 00000000:00000000 00:00000000 00000000 0\n\n/proc/net/unix\nLists  the UNIX domain sockets present within the system and their status.  The format\nis:\n\nNum RefCount Protocol Flags    Type St Inode Path\n0: 00000002 00000000 00000000 0001 03    42\n1: 00000001 00000000 00010000 0001 01  1948 /dev/printer\n\nThe fields are as follows:\n\nNum:      the kernel table slot number.\n\nRefCount: the number of users of the socket.\n\nProtocol: currently always 0.\n\nFlags:    the internal kernel flags holding the status of the socket.\n\nType:     the socket type.  For SOCKSTREAM sockets,  this  is  0001;  for  SOCKDGRAM\nsockets, it is 0002; and for SOCKSEQPACKET sockets, it is 0005.\n\nSt:       the internal state of the socket.\n\nInode:    the inode number of the socket.\n\nPath:     the  bound  pathname  (if any) of the socket.  Sockets in the abstract name‐\nspace are included in the list, and are shown with  a  Path  that  commences\nwith the character '@'.\n\n/proc/net/netfilter/nfnetlinkqueue\nThis  file  contains  information  about netfilter user-space queueing, if used.  Each\nline represents a queue.  Queues that have not been subscribed to by  user  space  are\nnot shown.\n\n1   4207     0  2 65535     0     0        0  1\n(1)   (2)    (3)(4)  (5)    (6)   (7)      (8)\n\nThe fields in each line are:\n\n(1)  The  ID  of  the  queue.   This  matches  what is specified in the --queue-num or\n--queue-balance options to the iptables(8) NFQUEUE target.   See  iptables-exten‐‐\nsions(8) for more information.\n\n(2)  The netlink port ID subscribed to the queue.\n\n(3)  The  number of packets currently queued and waiting to be processed by the appli‐\ncation.\n\n(4)  The copy mode of the queue.  It is either 1 (metadata only) or 2 (also copy  pay‐\nload data to user space).\n\n(5)  Copy  range;  that  is, how many bytes of packet payload should be copied to user\nspace at most.\n\n(6)  queue dropped.  Number of packets that had to be dropped by  the  kernel  because\ntoo  many  packets  are already waiting for user space to send back the mandatory\naccept/drop verdicts.\n\n(7)  queue user dropped.  Number of packets that were dropped within the netlink  sub‐\nsystem.   Such drops usually happen when the corresponding socket buffer is full;\nthat is, user space is not able to read messages fast enough.\n\n(8)  sequence number.  Every queued packet is associated with a (32-bit) monotonically\nincreasing sequence number.  This shows the ID of the most recent packet queued.\n\nThe last number exists only for compatibility reasons and is always 1.\n\n/proc/partitions\nContains  the  major  and  minor  numbers  of  each partition as well as the number of\n1024-byte blocks and the partition name.\n\n/proc/pci\nThis is a listing of all PCI devices found during kernel initialization and their con‐\nfiguration.\n\nThis   file   has  been  deprecated  in  favor  of  a  new  /proc  interface  for  PCI\n(/proc/bus/pci).  It became optional in Linux 2.2 (available with  CONFIGPCIOLDPROC\nset  at  kernel compilation).  It became once more nonoptionally enabled in Linux 2.4.\nNext, it was deprecated in Linux  2.6  (still  available  with  CONFIGPCILEGACYPROC\nset), and finally removed altogether since Linux 2.6.17.\n\n/proc/profile (since Linux 2.4)\nThis file is present only if the kernel was booted with the profile=1 command-line op‐\ntion.  It exposes kernel profiling information in a binary format for use by  readpro‐‐\nfile(1).   Writing (e.g., an empty string) to this file resets the profiling counters;\non some architectures,  writing  a  binary  integer  \"profiling  multiplier\"  of  size\nsizeof(int) sets the profiling interrupt frequency.\n\n/proc/scsi\nA  directory with the scsi mid-level pseudo-file and various SCSI low-level driver di‐\nrectories, which contain a file for each SCSI host in this system, all of  which  give\nthe  status  of  some part of the SCSI IO subsystem.  These files contain ASCII struc‐\ntures and are, therefore, readable with cat(1).\n\nYou can also write to some of the files to reconfigure the subsystem or switch certain\nfeatures on or off.\n\n/proc/scsi/scsi\nThis  is a listing of all SCSI devices known to the kernel.  The listing is similar to\nthe one seen during bootup.  scsi currently supports only the  add-single-device  com‐\nmand which allows root to add a hotplugged device to the list of known devices.\n\nThe command\n\necho 'scsi add-single-device 1 0 5 0' > /proc/scsi/scsi\n\nwill  cause host scsi1 to scan on SCSI channel 0 for a device on ID 5 LUN 0.  If there\nis already a device known on this address or the address is invalid, an error will  be\nreturned.\n\n/proc/scsi/[drivername]\n[drivername] can currently be NCR53c7xx, aha152x, aha1542, aha1740, aic7xxx, buslogic,\neatadma,  eatapio,  fdomain,  in2000,  pas16,  qlogic,  scsidebug,  seagate,  t128,\nu15-24f, ultrastore, or wd7000.  These directories show up for all drivers that regis‐\ntered at least one SCSI HBA.  Every directory contains one file per  registered  host.\nEvery host-file is named after the number the host was assigned during initialization.\n\nReading  these  files will usually show driver and host configuration, statistics, and\nso on.\n\nWriting to these files allows different things on different hosts.  For example,  with\nthe  latency  and  nolatency commands, root can switch on and off command latency mea‐\nsurement code in the eatadma driver.  With the lockup and unlock commands,  root  can\ncontrol bus lockups simulated by the scsidebug driver.\n\n/proc/self\nThis  directory refers to the process accessing the /proc filesystem, and is identical\nto the /proc directory named by the process ID of the same process.\n\n/proc/slabinfo\nInformation about kernel caches.  See slabinfo(5) for details.\n\n/proc/stat\nkernel/system statistics.  Varies with architecture.  Common entries include:\n\ncpu 10132153 290696 3084719 46828483 16683 0 25195 0 175628 0\ncpu0 1393280 32966 572056 13343292 6130 0 17875 0 23933 0\nThe amount of time, measured in units of USERHZ (1/100ths of a second on  most\narchitectures,  use  sysconf(SCCLKTCK)  to obtain the right value), that the\nsystem (\"cpu\" line) or the specific CPU (\"cpuN\" line) spent in various states:\n\nuser   (1) Time spent in user mode.\n\nnice   (2) Time spent in user mode with low priority (nice).\n\nsystem (3) Time spent in system mode.\n\nidle   (4) Time spent in the idle task.  This value should be USERHZ times the\nsecond entry in the /proc/uptime pseudo-file.\n\niowait (since Linux 2.5.41)\n(5)  Time  waiting for I/O to complete.  This value is not reliable, for\nthe following reasons:\n\n1. The CPU will not wait for I/O to complete; iowait is the time that  a\ntask is waiting for I/O to complete.  When a CPU goes into idle state\nfor outstanding task I/O, another task will be scheduled on this CPU.\n\n2. On a multi-core CPU, the task waiting for I/O to complete is not run‐\nning on any CPU, so the iowait of each CPU is difficult to calculate.\n\n3. The value in this field may decrease in certain conditions.\n\nirq (since Linux 2.6.0)\n(6) Time servicing interrupts.\n\nsoftirq (since Linux 2.6.0)\n(7) Time servicing softirqs.\n\nsteal (since Linux 2.6.11)\n(8) Stolen time, which is the time spent in other operating systems when\nrunning in a virtualized environment\n\nguest (since Linux 2.6.24)\n(9) Time spent running a virtual CPU for guest operating  systems  under\nthe control of the Linux kernel.\n\nguestnice (since Linux 2.6.33)\n(10)  Time  spent running a niced guest (virtual CPU for guest operating\nsystems under the control of the Linux kernel).\n\npage 5741 1808\nThe number of pages the system paged in and the  number  that  were  paged  out\n(from disk).\n\nswap 1 0\nThe number of swap pages that have been brought in and out.\n\nintr 1462898\nThis  line shows counts of interrupts serviced since boot time, for each of the\npossible system interrupts.  The first column is the total  of  all  interrupts\nserviced including unnumbered architecture specific interrupts; each subsequent\ncolumn is the total for that particular numbered interrupt.  Unnumbered  inter‐\nrupts are not shown, only summed into the total.\n\ndiskio: (2,0):(31,30,5764,1,2) (3,0):...\n(major,diskidx):(noinfo, readioops, blksread, writeioops, blkswritten)\n(Linux 2.4 only)\n\nctxt 115315\nThe number of context switches that the system underwent.\n\nbtime 769041601\nboot time, in seconds since the Epoch, 1970-01-01 00:00:00 +0000 (UTC).\n\nprocesses 86031\nNumber of forks since boot.\n\nprocsrunning 6\nNumber of processes in runnable state.  (Linux 2.5.45 onward.)\n\nprocsblocked 2\nNumber  of  processes  blocked  waiting for I/O to complete.  (Linux 2.5.45 on‐\nward.)\n\nsoftirq 229245889 94 60001584 13619 5175704 2471304 28 51212741 59130143 0 51240672\nThis line shows the number of softirq for all CPUs.  The first  column  is  the\ntotal  of  all  softirqs and each subsequent column is the total for particular\nsoftirq.  (Linux 2.6.31 onward.)\n\n/proc/swaps\nSwap areas in use.  See also swapon(8).\n\n/proc/sys\nThis directory (present since 1.3.57) contains a number of  files  and  subdirectories\ncorresponding to kernel variables.  These variables can be read and in some cases mod‐\nified using the /proc filesystem, and the (deprecated) sysctl(2) system call.\n\nString values may be terminated by either '\\0' or '\\n'.\n\nInteger and long values may be written either in decimal or  in  hexadecimal  notation\n(e.g.,  0x3FFF).  When writing multiple integer or long values, these may be separated\nby any of the following whitespace characters: ' ', '\\t', or '\\n'.  Using other  sepa‐\nrators leads to the error EINVAL.\n\n/proc/sys/abi (since Linux 2.4.10)\nThis  directory  may contain files with application binary information.  See the Linux\nkernel source file Documentation/sysctl/abi.txt for more information.\n\n/proc/sys/debug\nThis directory may be empty.\n\n/proc/sys/dev\nThis directory contains device-specific information (e.g., dev/cdrom/info).   On  some\nsystems, it may be empty.\n\n/proc/sys/fs\nThis  directory  contains the files and subdirectories for kernel variables related to\nfilesystems.\n\n/proc/sys/fs/aio-max-nr and /proc/sys/fs/aio-nr (since Linux 2.6.4)\naio-nr is the running total of the number of events specified by iosetup(2) calls for\nall  currently  active  AIO  contexts.  If aio-nr reaches aio-max-nr, then iosetup(2)\nwill fail with the error EAGAIN.  Raising aio-max-nr does not result in the prealloca‐\ntion or resizing of any kernel data structures.\n\n/proc/sys/fs/binfmtmisc\nDocumentation  for  files in this directory can be found in the Linux kernel source in\nthe    file    Documentation/admin-guide/binfmt-misc.rst     (or     in     Documenta‐\ntion/binfmtmisc.txt on older kernels).\n\n/proc/sys/fs/dentry-state (since Linux 2.2)\nThis  file contains information about the status of the directory cache (dcache).  The\nfile  contains  six  numbers,  nrdentry,  nrunused,  agelimit  (age  in   seconds),\nwantpages (pages requested by system) and two dummy values.\n\n* nrdentry  is  the number of allocated dentries (dcache entries).  This field is un‐\nused in Linux 2.2.\n\n* nrunused is the number of unused dentries.\n\n* agelimit is the age in seconds after which dcache entries  can  be  reclaimed  when\nmemory is short.\n\n* wantpages  is  nonzero  when  the  kernel  has called shrinkdcachepages() and the\ndcache isn't pruned yet.\n\n/proc/sys/fs/dir-notify-enable\nThis file can be used to disable or enable the dnotify interface described in fcntl(2)\non a system-wide basis.  A value of 0 in this file disables the interface, and a value\nof 1 enables it.\n\n/proc/sys/fs/dquot-max\nThis file shows the maximum number of cached disk quota entries.  On some  (2.4)  sys‐\ntems,  it is not present.  If the number of free cached disk quota entries is very low\nand you have some awesome number of simultaneous system users, you might want to raise\nthe limit.\n\n/proc/sys/fs/dquot-nr\nThis file shows the number of allocated disk quota entries and the number of free disk\nquota entries.\n\n/proc/sys/fs/epoll (since Linux 2.6.28)\nThis directory contains the file maxuserwatches, which can  be  used  to  limit  the\namount  of  kernel  memory  consumed by the epoll interface.  For further details, see\nepoll(7).\n\n/proc/sys/fs/file-max\nThis file defines a system-wide limit on the number of open files for  all  processes.\nSystem  calls that fail when encountering this limit fail with the error ENFILE.  (See\nalso setrlimit(2), which can be used by  a  process  to  set  the  per-process  limit,\nRLIMITNOFILE, on the number of files it may open.)  If you get lots of error messages\nin the kernel log about running out of file handles (open file descriptions) (look for\n\"VFS: file-max limit <number> reached\"), try increasing this value:\n\necho 100000 > /proc/sys/fs/file-max\n\nPrivileged processes (CAPSYSADMIN) can override the file-max limit.\n\n/proc/sys/fs/file-nr\nThis  (read-only)  file  contains  three numbers: the number of allocated file handles\n(i.e., the number of open file descriptions; see open(2)); the  number  of  free  file\nhandles;   and   the  maximum  number  of  file  handles  (i.e.,  the  same  value  as\n/proc/sys/fs/file-max).  If the number of allocated file handles is close to the maxi‐\nmum,  you  should consider increasing the maximum.  Before Linux 2.6, the kernel allo‐\ncated file handles dynamically, but it didn't free them again.  Instead the free  file\nhandles  were kept in a list for reallocation; the \"free file handles\" value indicates\nthe size of that list.  A large number of free file handles indicates that there was a\npast peak in the usage of open file handles.  Since Linux 2.6, the kernel does deallo‐\ncate freed file handles, and the \"free file handles\" value is always zero.\n\n/proc/sys/fs/inode-max (only present until Linux 2.2)\nThis file contains the maximum number of in-memory inodes.  This value should  be  3–4\ntimes  larger than the value in file-max, since stdin, stdout and network sockets also\nneed an inode to handle them.  When you regularly run out of inodes, you need  to  in‐\ncrease this value.\n\nStarting  with  Linux  2.4, there is no longer a static limit on the number of inodes,\nand this file is removed.\n\n/proc/sys/fs/inode-nr\nThis file contains the first two values from inode-state.\n\n/proc/sys/fs/inode-state\nThis file contains seven numbers: nrinodes, nrfreeinodes, preshrink, and four dummy\nvalues (always zero).\n\nnrinodes is the number of inodes the system has allocated.  nrfreeinodes represents\nthe number of free inodes.\n\npreshrink is nonzero when the nrinodes > inode-max and the system needs to prune  the\ninode  list  instead  of allocating more; since Linux 2.4, this field is a dummy value\n(always zero).\n\n/proc/sys/fs/inotify (since Linux 2.6.13)\nThis   directory   contains   files   maxqueuedevents,    maxuserinstances,    and\nmaxuserwatches,  that  can  be used to limit the amount of kernel memory consumed by\nthe inotify interface.  For further details, see inotify(7).\n\n/proc/sys/fs/lease-break-time\nThis file specifies the grace period that the kernel grants to  a  process  holding  a\nfile lease (fcntl(2)) after it has sent a signal to that process notifying it that an‐\nother process is waiting to open the file.  If the lease holder  does  not  remove  or\ndowngrade the lease within this grace period, the kernel forcibly breaks the lease.\n\n/proc/sys/fs/leases-enable\nThis file can be used to enable or disable file leases (fcntl(2)) on a system-wide ba‐\nsis.  If this file contains the value 0, leases are disabled.  A nonzero value enables\nleases.\n\n/proc/sys/fs/mount-max (since Linux 4.9)\nThe  value  in  this  file  specifies the maximum number of mounts that may exist in a\nmount namespace.  The default value in this file is 100,000.\n\n/proc/sys/fs/mqueue (since Linux 2.6.6)\nThis directory contains files msgmax, msgsizemax, and  queuesmax,  controlling  the\nresources used by POSIX message queues.  See mqoverview(7) for details.\n\n/proc/sys/fs/nropen (since Linux 2.6.25)\nThis file imposes a ceiling on the value to which the RLIMITNOFILE resource limit can\nbe raised (see getrlimit(2)).  This ceiling is  enforced  for  both  unprivileged  and\nprivileged process.  The default value in this file is 1048576.  (Before Linux 2.6.25,\nthe ceiling for RLIMITNOFILE was hard-coded to the same value.)\n\n/proc/sys/fs/overflowgid and /proc/sys/fs/overflowuid\nThese files allow you to change the value of the fixed UID and GID.   The  default  is\n65534.  Some filesystems support only 16-bit UIDs and GIDs, although in Linux UIDs and\nGIDs are 32 bits.  When one of these filesystems is mounted with writes  enabled,  any\nUID  or  GID  that would exceed 65535 is translated to the overflow value before being\nwritten to disk.\n\n/proc/sys/fs/pipe-max-size (since Linux 2.6.35)\nSee pipe(7).\n\n/proc/sys/fs/pipe-user-pages-hard (since Linux 4.5)\nSee pipe(7).\n\n/proc/sys/fs/pipe-user-pages-soft (since Linux 4.5)\nSee pipe(7).\n\n/proc/sys/fs/protectedfifos (since Linux 4.19)\nThe value in this file is/can be set to one of the following:\n\n0   Writing to FIFOs is unrestricted.\n\n1   Don't allow OCREAT open(2) on FIFOs that the caller doesn't own in world-writable\nsticky directories, unless the FIFO is owned by the owner of the directory.\n\n2   As  for the value 1, but the restriction also applies to group-writable sticky di‐\nrectories.\n\nThe intent of the above protections is to avoid unintentional writes to  an  attacker-\ncontrolled FIFO when a program expected to create a regular file.\n\n/proc/sys/fs/protectedhardlinks (since Linux 3.6)\nWhen  the  value in this file is 0, no restrictions are placed on the creation of hard\nlinks (i.e., this is the historical behavior before Linux 3.6).   When  the  value  in\nthis file is 1, a hard link can be created to a target file only if one of the follow‐\ning conditions is true:\n\n*  The calling process has the CAPFOWNER capability in its  user  namespace  and  the\nfile UID has a mapping in the namespace.\n\n*  The  filesystem UID of the process creating the link matches the owner (UID) of the\ntarget file (as described in credentials(7), a process's filesystem UID is normally\nthe same as its effective UID).\n\n*  All of the following conditions are true:\n\n•  the target is a regular file;\n\n•  the target file does not have its set-user-ID mode bit enabled;\n\n•  the  target  file does not have both its set-group-ID and group-executable mode\nbits enabled; and\n\n•  the caller has permission to read and write the target  file  (either  via  the\nfile's permissions mask or because it has suitable capabilities).\n\nThe  default  value in this file is 0.  Setting the value to 1 prevents a longstanding\nclass of security issues caused by hard-link-based time-of-check,  time-of-use  races,\nmost  commonly  seen in world-writable directories such as /tmp.  The common method of\nexploiting this flaw is to cross privilege boundaries when following a given hard link\n(i.e.,  a root process follows a hard link created by another user).  Additionally, on\nsystems without separated partitions, this stops  unauthorized  users  from  \"pinning\"\nvulnerable  set-user-ID  and set-group-ID files against being upgraded by the adminis‐\ntrator, or linking to special files.\n\n/proc/sys/fs/protectedregular (since Linux 4.19)\nThe value in this file is/can be set to one of the following:\n\n0   Writing to regular files is unrestricted.\n\n1   Don't allow OCREAT open(2) on regular files that the caller doesn't own in world-\nwritable  sticky directories, unless the regular file is owned by the owner of the\ndirectory.\n\n2   As for the value 1, but the restriction also applies to group-writable sticky  di‐\nrectories.\n\nThe  intent  of the above protections is similar to protectedfifos, but allows an ap‐\nplication to avoid writes to an attacker-controlled regular file, where  the  applica‐\ntion expected to create one.\n\n/proc/sys/fs/protectedsymlinks (since Linux 3.6)\nWhen  the  value  in  this file is 0, no restrictions are placed on following symbolic\nlinks (i.e., this is the historical behavior before Linux 3.6).   When  the  value  in\nthis file is 1, symbolic links are followed only in the following circumstances:\n\n*  the filesystem UID of the process following the link matches the owner (UID) of the\nsymbolic link (as described in credentials(7), a process's filesystem UID  is  nor‐\nmally the same as its effective UID);\n\n*  the link is not in a sticky world-writable directory; or\n\n*  the symbolic link and its parent directory have the same owner (UID)\n\nA  system  call that fails to follow a symbolic link because of the above restrictions\nreturns the error EACCES in errno.\n\nThe default value in this file is 0.  Setting the value to  1  avoids  a  longstanding\nclass of security issues based on time-of-check, time-of-use races when accessing sym‐\nbolic links.\n\n/proc/sys/fs/suiddumpable (since Linux 2.6.13)\nThe value in this file is assigned to a process's \"dumpable\" flag in the circumstances\ndescribed in prctl(2).  In effect, the value in this file determines whether core dump\nfiles are produced for  set-user-ID  or  otherwise  protected/tainted  binaries.   The\n\"dumpable\"  setting also affects the ownership of files in a process's /proc/[pid] di‐\nrectory, as described above.\n\nThree different integer values can be specified:\n\n0 (default)\nThis provides the traditional (pre-Linux 2.6.13) behavior.  A  core  dump  will\nnot  be  produced  for  a process which has changed credentials (by calling se‐‐\nteuid(2), setgid(2), or similar, or by executing a set-user-ID or  set-group-ID\nprogram) or whose binary does not have read permission enabled.\n\n1 (\"debug\")\nAll  processes dump core when possible.  (Reasons why a process might neverthe‐\nless not dump core are described in core(5).)  The core dump is  owned  by  the\nfilesystem  user ID of the dumping process and no security is applied.  This is\nintended for system debugging situations only: this mode is insecure because it\nallows  unprivileged  users  to  examine the memory contents of privileged pro‐\ncesses.\n\n2 (\"suidsafe\")\nAny binary which normally would not be dumped (see \"0\" above) is  dumped  read‐\nable  by  root only.  This allows the user to remove the core dump file but not\nto read it.  For security reasons core dumps in this mode  will  not  overwrite\none  another  or other files.  This mode is appropriate when administrators are\nattempting to debug problems in a normal environment.\n\nAdditionally, since Linux 3.6, /proc/sys/kernel/corepattern must either be  an\nabsolute  pathname or a pipe command, as detailed in core(5).  Warnings will be\nwritten to the kernel log if corepattern does not follow these rules,  and  no\ncore dump will be produced.\n\nFor  details  of  the  effect  of a process's \"dumpable\" setting on ptrace access mode\nchecking, see ptrace(2).\n\n/proc/sys/fs/super-max\nThis file controls the maximum number of superblocks, and thus the maximum  number  of\nmounted filesystems the kernel can have.  You need increase only super-max if you need\nto mount more filesystems than the current value in super-max allows you to.\n\n/proc/sys/fs/super-nr\nThis file contains the number of filesystems currently mounted.\n\n/proc/sys/kernel\nThis directory contains files controlling a range of kernel parameters,  as  described\nbelow.\n\n/proc/sys/kernel/acct\nThis  file  contains  three numbers: highwater, lowwater, and frequency.  If BSD-style\nprocess accounting is enabled, these values control its behavior.  If  free  space  on\nfilesystem  where  the log lives goes below lowwater percent, accounting suspends.  If\nfree space gets above highwater percent, accounting resumes.  frequency determines how\noften  the kernel checks the amount of free space (value is in seconds).  Default val‐\nues are 4, 2 and 30.  That is, suspend accounting if 2% or less space is free;  resume\nit  if 4% or more space is free; consider information about amount of free space valid\nfor 30 seconds.\n\n/proc/sys/kernel/automsgmni (Linux 2.6.27 to 3.18)\nFrom Linux 2.6.27 to 3.18, this file was used to control recomputing of the  value  in\n/proc/sys/kernel/msgmni  upon  the addition or removal of memory or upon IPC namespace\ncreation/removal.  Echoing \"1\" into this file  enabled  msgmni  automatic  recomputing\n(and triggered a recomputation of msgmni based on the current amount of available mem‐\nory and number of IPC namespaces).  Echoing \"0\" disabled automatic recomputing.   (Au‐\ntomatic  recomputing  was  also  disabled  if  a  value  was  explicitly  assigned  to\n/proc/sys/kernel/msgmni.)  The default value in automsgmni was 1.\n\nSince Linux 3.19, the content of this file has no effect (because msgmni  defaults  to\nnear  the  maximum  value  possible), and reads from this file always return the value\n\"0\".\n\n/proc/sys/kernel/caplastcap (since Linux 3.2)\nSee capabilities(7).\n\n/proc/sys/kernel/cap-bound (from Linux 2.2 to 2.6.24)\nThis file holds the value of the kernel capability bounding set (expressed as a signed\ndecimal  number).   This  set is ANDed against the capabilities permitted to a process\nduring execve(2).  Starting with Linux 2.6.25, the system-wide capability bounding set\ndisappeared, and was replaced by a per-thread bounding set; see capabilities(7).\n\n/proc/sys/kernel/corepattern\nSee core(5).\n\n/proc/sys/kernel/corepipelimit\nSee core(5).\n\n/proc/sys/kernel/coreusespid\nSee core(5).\n\n/proc/sys/kernel/ctrl-alt-del\nThis  file controls the handling of Ctrl-Alt-Del from the keyboard.  When the value in\nthis file is 0, Ctrl-Alt-Del is trapped and sent to the init(1) program  to  handle  a\ngraceful  restart.   When the value is greater than zero, Linux's reaction to a Vulcan\nNerve Pinch (tm) will be an immediate reboot, without even syncing its dirty  buffers.\nNote: when a program (like dosemu) has the keyboard in \"raw\" mode, the ctrl-alt-del is\nintercepted by the program before it ever reaches the kernel tty layer, and it's up to\nthe program to decide what to do with it.\n\n/proc/sys/kernel/dmesgrestrict (since Linux 2.6.37)\nThe value in this file determines who can see kernel syslog contents.  A value of 0 in\nthis file imposes no restrictions.  If the value is 1, only privileged users can  read\nthe  kernel  syslog.   (See  syslog(2) for more details.)  Since Linux 3.4, only users\nwith the CAPSYSADMIN capability may change the value in this file.\n\n/proc/sys/kernel/domainname and /proc/sys/kernel/hostname\ncan be used to set the NIS/YP domainname and the hostname of your box in  exactly  the\nsame way as the commands domainname(1) and hostname(1), that is:\n\n# echo 'darkstar' > /proc/sys/kernel/hostname\n# echo 'mydomain' > /proc/sys/kernel/domainname\n\nhas the same effect as\n\n# hostname 'darkstar'\n# domainname 'mydomain'\n\nNote,  however, that the classic darkstar.frop.org has the hostname \"darkstar\" and DNS\n(Internet Domain Name Server) domainname \"frop.org\", not to be confused with  the  NIS\n(Network Information Service) or YP (Yellow Pages) domainname.  These two domain names\nare in general different.  For a detailed discussion see the hostname(1) man page.\n\n/proc/sys/kernel/hotplug\nThis file contains the pathname for the hotplug policy agent.  The  default  value  in\nthis file is /sbin/hotplug.\n\n/proc/sys/kernel/htab-reclaim (before Linux 2.4.9.2)\n(PowerPC  only)  If  this file is set to a nonzero value, the PowerPC htab (see kernel\nfile Documentation/powerpc/ppchtab.txt) is pruned each time the system hits the  idle\nloop.\n\n/proc/sys/kernel/keys/*\nThis  directory  contains various files that define parameters and limits for the key-\nmanagement facility.  These files are described in keyrings(7).\n\n/proc/sys/kernel/kptrrestrict (since Linux 2.6.38)\nThe value in this file determines whether kernel addresses are exposed via /proc files\nand  other  interfaces.   A  value  of 0 in this file imposes no restrictions.  If the\nvalue is 1, kernel pointers printed using the %pK format specifier  will  be  replaced\nwith  zeros  unless the user has the CAPSYSLOG capability.  If the value is 2, kernel\npointers printed using the %pK format specifier will be replaced with zeros regardless\nof  the  user's  capabilities.  The initial default value for this file was 1, but the\ndefault was changed to 0 in Linux 2.6.39.   Since  Linux  3.4,  only  users  with  the\nCAPSYSADMIN capability can change the value in this file.\n\n/proc/sys/kernel/l2cr\n(PowerPC  only)  This  file contains a flag that controls the L2 cache of G3 processor\nboards.  If 0, the cache is disabled.  Enabled if nonzero.\n\n/proc/sys/kernel/modprobe\nThis file contains the pathname for the kernel module loader.  The  default  value  is\n/sbin/modprobe.   The file is present only if the kernel is built with the CONFIGMOD‐‐\nULES (CONFIGKMOD in Linux 2.6.26 and earlier) option enabled.  It is described by the\nLinux  kernel  source file Documentation/kmod.txt (present only in kernel 2.4 and ear‐\nlier).\n\n/proc/sys/kernel/modulesdisabled (since Linux 2.6.31)\nA toggle value indicating if modules are allowed to be loaded in an otherwise  modular\nkernel.  This toggle defaults to off (0), but can be set true (1).  Once true, modules\ncan be neither loaded nor unloaded, and the toggle cannot be set back to  false.   The\nfile is present only if the kernel is built with the CONFIGMODULES option enabled.\n\n/proc/sys/kernel/msgmax (since Linux 2.2)\nThis file defines a system-wide limit specifying the maximum number of bytes in a sin‐\ngle message written on a System V message queue.\n\n/proc/sys/kernel/msgmni (since Linux 2.4)\nThis file defines the system-wide limit on the number of  message  queue  identifiers.\nSee also /proc/sys/kernel/automsgmni.\n\n/proc/sys/kernel/msgmnb (since Linux 2.2)\nThis  file  defines  a system-wide parameter used to initialize the msgqbytes setting\nfor subsequently created message queues.  The msgqbytes setting specifies the maximum\nnumber of bytes that may be written to the message queue.\n\n/proc/sys/kernel/ngroupsmax (since Linux 2.6.4)\nThis  is  a  read-only file that displays the upper limit on the number of a process's\ngroup memberships.\n\n/proc/sys/kernel/nslastpid (since Linux 3.3)\nSee pidnamespaces(7).\n\n/proc/sys/kernel/ostype and /proc/sys/kernel/osrelease\nThese files give substrings of /proc/version.\n\n/proc/sys/kernel/overflowgid and /proc/sys/kernel/overflowuid\nThese files duplicate the files /proc/sys/fs/overflowgid and /proc/sys/fs/overflowuid.\n\n/proc/sys/kernel/panic\nThis file gives read/write access to the kernel variable panictimeout.   If  this  is\nzero, the kernel will loop on a panic; if nonzero, it indicates that the kernel should\nautoreboot after this number of seconds.  When you use the  software  watchdog  device\ndriver, the recommended setting is 60.\n\n/proc/sys/kernel/paniconoops (since Linux 2.5.68)\nThis  file controls the kernel's behavior when an oops or BUG is encountered.  If this\nfile contains 0, then the system tries to continue operation.  If it contains 1,  then\nthe  system  delays  a  few seconds (to give klogd time to record the oops output) and\nthen panics.  If the /proc/sys/kernel/panic file is also  nonzero,  then  the  machine\nwill be rebooted.\n\n/proc/sys/kernel/pidmax (since Linux 2.5.34)\nThis  file specifies the value at which PIDs wrap around (i.e., the value in this file\nis one greater than the maximum PID).  PIDs greater than this value are not allocated;\nthus,  the  value in this file also acts as a system-wide limit on the total number of\nprocesses and threads.  The default value for this file, 32768, results  in  the  same\nrange  of PIDs as on earlier kernels.  On 32-bit platforms, 32768 is the maximum value\nfor pidmax.  On 64-bit  systems,  pidmax  can  be  set  to  any  value  up  to  2^22\n(PIDMAXLIMIT, approximately 4 million).\n\n/proc/sys/kernel/powersave-nap (PowerPC only)\nThis  file contains a flag.  If set, Linux-PPC will use the \"nap\" mode of powersaving,\notherwise the \"doze\" mode will be used.\n\n/proc/sys/kernel/printk\nSee syslog(2).\n\n/proc/sys/kernel/pty (since Linux 2.6.4)\nThis directory contains two files relating to the number of  UNIX  98  pseudoterminals\n(see pts(4)) on the system.\n\n/proc/sys/kernel/pty/max\nThis file defines the maximum number of pseudoterminals.\n\n/proc/sys/kernel/pty/nr\nThis read-only file indicates how many pseudoterminals are currently in use.\n\n/proc/sys/kernel/random\nThis  directory  contains  various  parameters  controlling  the operation of the file\n/dev/random.  See random(4) for further information.\n\n/proc/sys/kernel/random/uuid (since Linux 2.4)\nEach read from this read-only file returns a randomly generated  128-bit  UUID,  as  a\nstring in the standard UUID format.\n\n/proc/sys/kernel/randomizevaspace (since Linux 2.6.12)\nSelect  the address space layout randomization (ASLR) policy for the system (on archi‐\ntectures that support ASLR).  Three values are supported for this file:\n\n0  Turn ASLR off.  This is the default for architectures that don't support ASLR,  and\nwhen the kernel is booted with the norandmaps parameter.\n\n1  Make the addresses of mmap(2) allocations, the stack, and the VDSO page randomized.\nAmong other things, this means that shared libraries will be loaded  at  randomized\naddresses.   The  text segment of PIE-linked binaries will also be loaded at a ran‐\ndomized address.  This value is the default if the kernel was configured with  CON‐‐\nFIGCOMPATBRK.\n\n2  (Since Linux 2.6.25) Also support heap randomization.  This value is the default if\nthe kernel was not configured with CONFIGCOMPATBRK.\n\n/proc/sys/kernel/real-root-dev\nThis file is documented in the Linux kernel source file Documentation/admin-guide/ini‐\ntrd.rst (or Documentation/initrd.txt before Linux 4.10).\n\n/proc/sys/kernel/reboot-cmd (Sparc only)\nThis  file  seems  to be a way to give an argument to the SPARC ROM/Flash boot loader.\nMaybe to tell it what to do after rebooting?\n\n/proc/sys/kernel/rtsig-max\n(Only in kernels up to and including 2.6.7; see setrlimit(2)) This file can be used to\ntune the maximum number of POSIX real-time (queued) signals that can be outstanding in\nthe system.\n\n/proc/sys/kernel/rtsig-nr\n(Only in kernels up to and including 2.6.7.)  This file  shows  the  number  of  POSIX\nreal-time signals currently queued.\n\n/proc/[pid]/schedautogroupenabled (since Linux 2.6.38)\nSee sched(7).\n\n/proc/sys/kernel/schedchildrunsfirst (since Linux 2.6.23)\nIf  this  file  contains  the  value  zero, then, after a fork(2), the parent is first\nscheduled on the CPU.  If the file contains a nonzero value, then the child is  sched‐\nuled  first  on  the  CPU.  (Of course, on a multiprocessor system, the parent and the\nchild might both immediately be scheduled on a CPU.)\n\n/proc/sys/kernel/schedrrtimeslicems (since Linux 3.9)\nSee schedrrgetinterval(2).\n\n/proc/sys/kernel/schedrtperiodus (since Linux 2.6.25)\nSee sched(7).\n\n/proc/sys/kernel/schedrtruntimeus (since Linux 2.6.25)\nSee sched(7).\n\n/proc/sys/kernel/seccomp (since Linux 4.14)\nThis directory provides additional seccomp information and  configuration.   See  sec‐‐\ncomp(2) for further details.\n\n/proc/sys/kernel/sem (since Linux 2.4)\nThis  file  contains  4  numbers  defining  limits for System V IPC semaphores.  These\nfields are, in order:\n\nSEMMSL  The maximum semaphores per semaphore set.\n\nSEMMNS  A system-wide limit on the number of semaphores in all semaphore sets.\n\nSEMOPM  The maximum number of operations that may be specified in a semop(2) call.\n\nSEMMNI  A system-wide limit on the maximum number of semaphore identifiers.\n\n/proc/sys/kernel/sg-big-buff\nThis file shows the size of the generic SCSI device (sg) buffer.  You  can't  tune  it\njust  yet,  but  you  could change it at compile time by editing include/scsi/sg.h and\nchanging the value of SGBIGBUFF.  However, there shouldn't be any reason  to  change\nthis value.\n\n/proc/sys/kernel/shmrmidforced (since Linux 3.1)\nIf  this  file is set to 1, all System V shared memory segments will be marked for de‐\nstruction as soon as the number of attached processes falls to zero; in  other  words,\nit  is no longer possible to create shared memory segments that exist independently of\nany attached process.\n\nThe effect is as though a shmctl(2) IPCRMID is performed on all existing segments  as\nwell as all segments created in the future (until this file is reset to 0).  Note that\nexisting segments that are attached to no process will be immediately  destroyed  when\nthis  file is set to 1.  Setting this option will also destroy segments that were cre‐\nated, but never attached, upon termination of the process  that  created  the  segment\nwith shmget(2).\n\nSetting this file to 1 provides a way of ensuring that all System V shared memory seg‐\nments are counted against the resource usage and resource limits (see the  description\nof RLIMITAS in getrlimit(2)) of at least one process.\n\nBecause  setting  this  file to 1 produces behavior that is nonstandard and could also\nbreak existing applications, the default value in this file is 0.  Set this file to  1\nonly  if you have a good understanding of the semantics of the applications using Sys‐\ntem V shared memory on your system.\n\n/proc/sys/kernel/shmall (since Linux 2.2)\nThis file contains the system-wide limit on the total number  of  pages  of  System  V\nshared memory.\n\n/proc/sys/kernel/shmmax (since Linux 2.2)\nThis  file  can  be  used to query and set the run-time limit on the maximum (System V\nIPC) shared memory segment size that can be created.  Shared memory segments up  to  1\nGB are now supported in the kernel.  This value defaults to SHMMAX.\n\n/proc/sys/kernel/shmmni (since Linux 2.4)\nThis  file specifies the system-wide maximum number of System V shared memory segments\nthat can be created.\n\n/proc/sys/kernel/sysctlwritesstrict (since Linux 3.16)\nThe value in this file determines how the file offset affects the behavior of updating\nentries in files under /proc/sys.  The file has three possible values:\n\n-1  This  provides legacy handling, with no printk warnings.  Each write(2) must fully\ncontain the value to be written, and multiple writes on the same  file  descriptor\nwill overwrite the entire value, regardless of the file position.\n\n0   (default) This provides the same behavior as for -1, but printk warnings are writ‐\nten for processes that perform writes when the file offset is not 0.\n\n1   Respect the file offset when  writing  strings  into  /proc/sys  files.   Multiple\nwrites  will  append  to  the  value  buffer.  Anything written beyond the maximum\nlength of the value buffer will be ignored.  Writes to numeric  /proc/sys  entries\nmust  always be at file offset 0 and the value must be fully contained in the buf‐\nfer provided to write(2).\n\n/proc/sys/kernel/sysrq\nThis file controls the functions allowed to be invoked by the SysRq key.  By  default,\nthe  file  contains  1  meaning that every possible SysRq request is allowed (in older\nkernel versions, SysRq was disabled by default, and you were required to  specifically\nenable  it  at  run-time, but this is not the case any more).  Possible values in this\nfile are:\n\n0    Disable sysrq completely\n\n1    Enable all functions of sysrq\n\n> 1  Bit mask of allowed sysrq functions, as follows:\n2  Enable control of console logging level\n4  Enable control of keyboard (SAK, unraw)\n8  Enable debugging dumps of processes etc.\n16  Enable sync command\n32  Enable remount read-only\n64  Enable signaling of processes (term, kill, oom-kill)\n128  Allow reboot/poweroff\n256  Allow nicing of all real-time tasks\n\nThis file is present only if the CONFIGMAGICSYSRQ kernel configuration option is en‐\nabled.   For  further  details  see  the  Linux  kernel  source file Documentation/ad‐\nmin-guide/sysrq.rst (or Documentation/sysrq.txt before Linux 4.10).\n\n/proc/sys/kernel/version\nThis file contains a string such as:\n\n#5 Wed Feb 25 21:49:24 MET 1998\n\nThe \"#5\" means that this is the fifth kernel built from this source base and the  date\nfollowing it indicates the time the kernel was built.\n\n/proc/sys/kernel/threads-max (since Linux 2.3.11)\nThis file specifies the system-wide limit on the number of threads (tasks) that can be\ncreated on the system.\n\nSince Linux 4.1, the value that can be written to threads-max is bounded.  The minimum\nvalue  that  can  be written is 20.  The maximum value that can be written is given by\nthe constant FUTEXTIDMASK (0x3fffffff).  If a value outside of this range is written\nto threads-max, the error EINVAL occurs.\n\nThe  value  written  is checked against the available RAM pages.  If the thread struc‐\ntures would occupy too much (more than 1/8th) of the available RAM pages,  threads-max\nis reduced accordingly.\n\n/proc/sys/kernel/yama/ptracescope (since Linux 3.5)\nSee ptrace(2).\n\n/proc/sys/kernel/zero-paged (PowerPC only)\nThis  file  contains a flag.  When enabled (nonzero), Linux-PPC will pre-zero pages in\nthe idle loop, possibly speeding up getfreepages.\n\n/proc/sys/net\nThis directory contains networking stuff.  Explanations for some of  the  files  under\nthis directory can be found in tcp(7) and ip(7).\n\n/proc/sys/net/core/bpfjitenable\nSee bpf(2).\n\n/proc/sys/net/core/somaxconn\nThis  file defines a ceiling value for the backlog argument of listen(2); see the lis‐‐\nten(2) manual page for details.\n\n/proc/sys/proc\nThis directory may be empty.\n\n/proc/sys/sunrpc\nThis directory supports Sun remote procedure call for network  filesystem  (NFS).   On\nsome systems, it is not present.\n\n/proc/sys/user (since Linux 4.9)\nSee namespaces(7).\n\n/proc/sys/vm\nThis  directory  contains files for memory management tuning, buffer and cache manage‐\nment.\n\n/proc/sys/vm/adminreservekbytes (since Linux 3.10)\nThis file defines the amount of free memory (in KiB) on the system that should be  re‐\nserved for users with the capability CAPSYSADMIN.\n\nThe default value in this file is the minimum of [3% of free pages, 8MiB] expressed as\nKiB.  The default is intended to provide enough for the superuser to log in and kill a\nprocess,  if  necessary,  under  the  default  overcommit  'guess'  mode  (i.e.,  0 in\n/proc/sys/vm/overcommitmemory).\n\nSystems running in \"overcommit never\" mode (i.e., 2 in /proc/sys/vm/overcommitmemory)\nshould  increase the value in this file to account for the full virtual memory size of\nthe programs used to recover (e.g., login(1) ssh(1), and top(1)) Otherwise, the  supe‐\nruser may not be able to log in to recover the system.  For example, on x86-64 a suit‐\nable value is 131072 (128MiB reserved).\n\nChanging the value in this file takes effect whenever an application requests memory.\n\n/proc/sys/vm/compactmemory (since Linux 2.6.35)\nWhen 1 is written to this file, all zones are  compacted  such  that  free  memory  is\navailable  in contiguous blocks where possible.  The effect of this action can be seen\nby examining /proc/buddyinfo.\n\nPresent only if the kernel was configured with CONFIGCOMPACTION.\n\n/proc/sys/vm/dropcaches (since Linux 2.6.16)\nWriting to this file causes the kernel to drop clean caches, dentries, and inodes from\nmemory,  causing that memory to become free.  This can be useful for memory management\ntesting and performing reproducible filesystem benchmarks.  Because  writing  to  this\nfile  causes the benefits of caching to be lost, it can degrade overall system perfor‐\nmance.\n\nTo free pagecache, use:\n\necho 1 > /proc/sys/vm/dropcaches\n\nTo free dentries and inodes, use:\n\necho 2 > /proc/sys/vm/dropcaches\n\nTo free pagecache, dentries and inodes, use:\n\necho 3 > /proc/sys/vm/dropcaches\n\nBecause writing to this file is a nondestructive operation and dirty objects  are  not\nfreeable, the user should run sync(1) first.\n\n/proc/sys/vm/legacyvalayout (since Linux 2.6.9)\nIf  nonzero,  this  disables the new 32-bit memory-mapping layout; the kernel will use\nthe legacy (2.4) layout for all processes.\n\n/proc/sys/vm/memoryfailureearlykill (since Linux 2.6.32)\nControl how to kill processes when an uncorrected memory error (typically a 2-bit  er‐\nror  in a memory module) that cannot be handled by the kernel is detected in the back‐\nground by hardware.  In some cases (like the page still having a valid copy on  disk),\nthe  kernel  will handle the failure transparently without affecting any applications.\nBut if there is no other up-to-date copy of the data, it will kill processes  to  pre‐\nvent any data corruptions from propagating.\n\nThe file has one of the following values:\n\n1:  Kill  all processes that have the corrupted-and-not-reloadable page mapped as soon\nas the corruption is detected.  Note that this is not supported for a few types of\npages,  such  as kernel internally allocated data or the swap cache, but works for\nthe majority of user pages.\n\n0:  Unmap the corrupted page from all processes and kill a process only if it tries to\naccess the page.\n\nThe  kill  is performed using a SIGBUS signal with sicode set to BUSMCEERRAO.  Pro‐\ncesses can handle this if they want to; see sigaction(2) for more details.\n\nThis feature is active only on architectures/platforms  with  advanced  machine  check\nhandling and depends on the hardware capabilities.\n\nApplications  can override the memoryfailureearlykill setting individually with the\nprctl(2) PRMCEKILL operation.\n\nPresent only if the kernel was configured with CONFIGMEMORYFAILURE.\n\n/proc/sys/vm/memoryfailurerecovery (since Linux 2.6.32)\nEnable memory failure recovery (when supported by the platform).\n\n1:  Attempt recovery.\n\n0:  Always panic on a memory failure.\n\nPresent only if the kernel was configured with CONFIGMEMORYFAILURE.\n\n/proc/sys/vm/oomdumptasks (since Linux 2.6.25)\nEnables a system-wide task dump (excluding kernel threads) to  be  produced  when  the\nkernel  performs an OOM-killing.  The dump includes the following information for each\ntask (thread, process): thread ID, real user ID, thread group ID (process ID), virtual\nmemory  size,  resident set size, the CPU that the task is scheduled on, oomadj score\n(see the description of /proc/[pid]/oomadj), and command name.  This  is  helpful  to\ndetermine  why  the  OOM-killer was invoked and to identify the rogue task that caused\nit.\n\nIf this contains the value zero, this information is suppressed.  On very  large  sys‐\ntems with thousands of tasks, it may not be feasible to dump the memory state informa‐\ntion for each one.  Such systems should not be forced to incur a  performance  penalty\nin OOM situations when the information may not be desired.\n\nIf  this is set to nonzero, this information is shown whenever the OOM-killer actually\nkills a memory-hogging task.\n\nThe default value is 0.\n\n/proc/sys/vm/oomkillallocatingtask (since Linux 2.6.24)\nThis enables or disables killing the OOM-triggering task in out-of-memory situations.\n\nIf this is set to zero, the OOM-killer will scan through the entire tasklist  and  se‐\nlect a task based on heuristics to kill.  This normally selects a rogue memory-hogging\ntask that frees up a large amount of memory when killed.\n\nIf this is set to nonzero, the OOM-killer simply kills the  task  that  triggered  the\nout-of-memory condition.  This avoids a possibly expensive tasklist scan.\n\nIf  /proc/sys/vm/paniconoom  is  nonzero, it takes precedence over whatever value is\nused in /proc/sys/vm/oomkillallocatingtask.\n\nThe default value is 0.\n\n/proc/sys/vm/overcommitkbytes (since Linux 3.14)\nThis writable file provides an alternative to /proc/sys/vm/overcommitratio  for  con‐\ntrolling  the CommitLimit when /proc/sys/vm/overcommitmemory has the value 2.  It al‐\nlows the amount of memory overcommitting to be specified as an absolute value (in kB),\nrather than as a percentage, as is done with overcommitratio.  This allows for finer-\ngrained control of CommitLimit on systems with extremely large memory sizes.\n\nOnly one of overcommitkbytes or overcommitratio can  have  an  effect:  if  overcom‐\nmitkbytes  has  a  nonzero value, then it is used to calculate CommitLimit, otherwise\novercommitratio is used.  Writing a value to either of these files causes  the  value\nin the other file to be set to zero.\n\n/proc/sys/vm/overcommitmemory\nThis file contains the kernel virtual memory accounting mode.  Values are:\n\n0: heuristic overcommit (this is the default)\n1: always overcommit, never check\n2: always check, never overcommit\n\nIn  mode 0, calls of mmap(2) with MAPNORESERVE are not checked, and the default check\nis very weak, leading to the risk of getting a process \"OOM-killed\".\n\nIn mode 1, the kernel pretends there is always enough memory,  until  memory  actually\nruns out.  One use case for this mode is scientific computing applications that employ\nlarge sparse arrays.  In Linux kernel versions before 2.6.0, any nonzero value implies\nmode 1.\n\nIn mode 2 (available since Linux 2.6), the total virtual address space that can be al‐\nlocated (CommitLimit in /proc/meminfo) is calculated as\n\nCommitLimit = (totalRAM - totalhugeTLB) *\novercommitratio / 100 + totalswap\n\nwhere:\n\n*  totalRAM is the total amount of RAM on the system;\n\n*  totalhugeTLB is the amount of memory set aside for huge pages;\n\n*  overcommitratio is the value in /proc/sys/vm/overcommitratio; and\n\n*  totalswap is the amount of swap space.\n\nFor example, on a system with 16 GB of physical RAM, 16 GB of swap, no space dedicated\nto  huge pages, and an overcommitratio of 50, this formula yields a CommitLimit of 24\nGB.\n\nSince Linux 3.14, if the value in /proc/sys/vm/overcommitkbytes is nonzero, then Com‐\nmitLimit is instead calculated as:\n\nCommitLimit = overcommitkbytes + totalswap\n\nSee     also     the     description    of    /proc/sys/vm/adminreservekbytes    and\n/proc/sys/vm/userreservekbytes.\n\n/proc/sys/vm/overcommitratio (since Linux 2.6.0)\nThis writable file defines a percentage by which memory can be overcommitted.  The de‐\nfault value in the file is 50.  See the description of /proc/sys/vm/overcommitmemory.\n\n/proc/sys/vm/paniconoom (since Linux 2.6.18)\nThis enables or disables a kernel panic in an out-of-memory situation.\n\nIf  this  file  is  set  to  the value 0, the kernel's OOM-killer will kill some rogue\nprocess.  Usually, the OOM-killer is able to kill a rogue process and the system  will\nsurvive.\n\nIf this file is set to the value 1, then the kernel normally panics when out-of-memory\nhappens.  However, if a process limits allocations to certain nodes using memory poli‐\ncies  (mbind(2) MPOLBIND) or cpusets (cpuset(7)) and those nodes reach memory exhaus‐\ntion status, one process may be killed by the OOM-killer.  No  panic  occurs  in  this\ncase:  because  other  nodes' memory may be free, this means the system as a whole may\nnot have reached an out-of-memory situation yet.\n\nIf this file is set to the value 2, the kernel always  panics  when  an  out-of-memory\ncondition occurs.\n\nThe  default  value  is 0.  1 and 2 are for failover of clustering.  Select either ac‐\ncording to your policy of failover.\n\n/proc/sys/vm/swappiness\nThe value in this file controls how aggressively the kernel will  swap  memory  pages.\nHigher  values increase aggressiveness, lower values decrease aggressiveness.  The de‐\nfault value is 60.\n\n/proc/sys/vm/userreservekbytes (since Linux 3.10)\nSpecifies an amount of memory (in KiB) to reserve for user  processes.   This  is  in‐\ntended to prevent a user from starting a single memory hogging process, such that they\ncannot recover (kill the hog).  The value  in  this  file  has  an  effect  only  when\n/proc/sys/vm/overcommitmemory  is  set to 2 (\"overcommit never\" mode).  In this case,\nthe system reserves an amount of memory that is the minimum of [3% of current  process\nsize, userreservekbytes].\n\nThe  default value in this file is the minimum of [3% of free pages, 128MiB] expressed\nas KiB.\n\nIf the value in this file is set to zero, then a user will be allowed to allocate  all\nfree memory with a single process (minus the amount reserved by /proc/sys/vm/adminre‐\nservekbytes).  Any subsequent attempts to execute a command  will  result  in  \"fork:\nCannot allocate memory\".\n\nChanging the value in this file takes effect whenever an application requests memory.\n\n/proc/sys/vm/unprivilegeduserfaultfd (since Linux 5.2)\nThis  (writable)  file exposes a flag that controls whether unprivileged processes are\nallowed to employ userfaultfd(2).  If this file has the  value  1,  then  unprivileged\nprocesses  may  use userfaultfd(2).  If this file has the value 0, then only processes\nthat have the CAPSYSPTRACE capability may employ userfaultfd(2).  The default  value\nin this file is 1.\n\n/proc/sysrq-trigger (since Linux 2.4.21)\nWriting  a  character  to  this  file  triggers the same SysRq function as typing ALT-\nSysRq-<character> (see the description of /proc/sys/kernel/sysrq).  This file is  nor‐\nmally  writable  only  by  root.  For further details see the Linux kernel source file\nDocumentation/admin-guide/sysrq.rst (or Documentation/sysrq.txt before Linux 4.10).\n\n/proc/sysvipc\nSubdirectory containing the pseudo-files msg, sem and shm.  These files list the  Sys‐\ntem  V  Interprocess  Communication (IPC) objects (respectively: message queues, sema‐\nphores, and shared memory) that currently exist on the system, providing  similar  in‐\nformation  to  that available via ipcs(1).  These files have headers and are formatted\n(one IPC object per line) for easy understanding.  sysvipc(7) provides  further  back‐\nground on the information shown by these files.\n\n/proc/thread-self (since Linux 3.17)\nThis  directory  refers to the thread accessing the /proc filesystem, and is identical\nto the /proc/self/task/[tid] directory named by the process thread ID ([tid])  of  the\nsame thread.\n\n/proc/timerlist (since Linux 2.6.21)\nThis  read-only file exposes a list of all currently pending (high-resolution) timers,\nall clock-event sources, and their parameters in a human-readable form.\n\n/proc/timerstats (from  Linux 2.6.21 until Linux 4.10)\nThis is a debugging facility to make timer (ab)use in a Linux system visible to kernel\nand user-space developers.  It can be used by kernel and user-space developers to ver‐\nify that their code does not make undue use of timers.  The goal is to avoid  unneces‐\nsary wakeups, thereby optimizing power consumption.\n\nIf  enabled  in the kernel (CONFIGTIMERSTATS), but not used, it has almost zero run-\ntime overhead and a relatively small data-structure overhead.  Even if  collection  is\nenabled at run time, overhead is low: all the locking is per-CPU and lookup is hashed.\n\nThe  /proc/timerstats  file is used both to control sampling facility and to read out\nthe sampled information.\n\nThe timerstats functionality is inactive on bootup.  A sampling period can be started\nusing the following command:\n\n# echo 1 > /proc/timerstats\n\nThe following command stops a sampling period:\n\n# echo 0 > /proc/timerstats\n\nThe statistics can be retrieved by:\n\n$ cat /proc/timerstats\n\nWhile  sampling is enabled, each readout from /proc/timerstats will see newly updated\nstatistics.  Once sampling is disabled, the sampled information is kept  until  a  new\nsample period is started.  This allows multiple readouts.\n\nSample output from /proc/timerstats:\n\n$ cat /proc/timerstats\nTimer Stats Version: v0.3\nSample period: 1.764 s\nCollection: active\n255,     0 swapper/3        hrtimerstartrangens (tickschedtimer)\n71,     0 swapper/1        hrtimerstartrangens (tickschedtimer)\n58,     0 swapper/0        hrtimerstartrangens (tickschedtimer)\n4,  1694 gnome-shell      moddelayedworkon (delayedworktimerfn)\n17,     7 rcusched        rcugpkthread (processtimeout)\n...\n1,  4911 kworker/u16:0    moddelayedworkon (delayedworktimerfn)\n1D,  2522 kworker/0:0      queuedelayedworkon (delayedworktimerfn)\n1029 total events, 583.333 events/sec\n\nThe output columns are:\n\n*  a  count  of  the number of events, optionally (since Linux 2.6.23) followed by the\nletter 'D' if this is a deferrable timer;\n\n*  the PID of the process that initialized the timer;\n\n*  the name of the process that initialized the timer;\n\n*  the function where the timer was initialized; and\n\n*  (in parentheses) the callback function that is associated with the timer.\n\nDuring the Linux 4.11 development cycle, this file  was removed  because  of  security\nconcerns, as it exposes information across namespaces.  Furthermore, it is possible to\nobtain the same information via in-kernel tracing facilities such as ftrace.\n\n/proc/tty\nSubdirectory containing the pseudo-files and subdirectories for tty drivers  and  line\ndisciplines.\n\n/proc/uptime\nThis  file contains two numbers (values in seconds): the uptime of the system (includ‐\ning time spent in suspend) and the amount of time spent in the idle process.\n\n/proc/version\nThis string identifies the kernel version that is currently running.  It includes  the\ncontents  of  /proc/sys/kernel/ostype,  /proc/sys/kernel/osrelease, and /proc/sys/ker‐\nnel/version.  For example:\n\nLinux version 1.0.9 (quinlan@phaze) #1 Sat May 14 01:51:54 EDT 1994\n\n/proc/vmstat (since Linux 2.6.0)\nThis file displays various virtual memory statistics.  Each line of this file contains\na  single  name-value  pair, delimited by white space.  Some lines are present only if\nthe kernel was configured with suitable options.  (In some cases, the options required\nfor particular files have changed across kernel versions, so they are not listed here.\nDetails can be found by consulting the kernel source code.)  The following fields  may\nbe present:\n\nnrfreepages (since Linux 2.6.31)\n\nnrallocbatch (since Linux 3.12)\n\nnrinactiveanon (since Linux 2.6.28)\n\nnractiveanon (since Linux 2.6.28)\n\nnrinactivefile (since Linux 2.6.28)\n\nnractivefile (since Linux 2.6.28)\n\nnrunevictable (since Linux 2.6.28)\n\nnrmlock (since Linux 2.6.28)\n\nnranonpages (since Linux 2.6.18)\n\nnrmapped (since Linux 2.6.0)\n\nnrfilepages (since Linux 2.6.18)\n\nnrdirty (since Linux 2.6.0)\n\nnrwriteback (since Linux 2.6.0)\n\nnrslabreclaimable (since Linux 2.6.19)\n\nnrslabunreclaimable (since Linux 2.6.19)\n\nnrpagetablepages (since Linux 2.6.0)\n\nnrkernelstack (since Linux 2.6.32)\nAmount of memory allocated to kernel stacks.\n\nnrunstable (since Linux 2.6.0)\n\nnrbounce (since Linux 2.6.12)\n\nnrvmscanwrite (since Linux 2.6.19)\n\nnrvmscanimmediatereclaim (since Linux 3.2)\n\nnrwritebacktemp (since Linux 2.6.26)\n\nnrisolatedanon (since Linux 2.6.32)\n\nnrisolatedfile (since Linux 2.6.32)\n\nnrshmem (since Linux 2.6.32)\nPages used by shmem and tmpfs(5).\n\nnrdirtied (since Linux 2.6.37)\n\nnrwritten (since Linux 2.6.37)\n\nnrpagesscanned (since Linux 3.17)\n\nnumahit (since Linux 2.6.18)\n\nnumamiss (since Linux 2.6.18)\n\nnumaforeign (since Linux 2.6.18)\n\nnumainterleave (since Linux 2.6.18)\n\nnumalocal (since Linux 2.6.18)\n\nnumaother (since Linux 2.6.18)\n\nworkingsetrefault (since Linux 3.15)\n\nworkingsetactivate (since Linux 3.15)\n\nworkingsetnodereclaim (since Linux 3.15)\n\nnranontransparenthugepages (since Linux 2.6.38)\n\nnrfreecma (since Linux 3.7)\nNumber of free CMA (Contiguous Memory Allocator) pages.\n\nnrdirtythreshold (since Linux 2.6.37)\n\nnrdirtybackgroundthreshold (since Linux 2.6.37)\n\npgpgin (since Linux 2.6.0)\n\npgpgout (since Linux 2.6.0)\n\npswpin (since Linux 2.6.0)\n\npswpout (since Linux 2.6.0)\n\npgallocdma (since Linux 2.6.5)\n\npgallocdma32 (since Linux 2.6.16)\n\npgallocnormal (since Linux 2.6.5)\n\npgallochigh (since Linux 2.6.5)\n\npgallocmovable (since Linux 2.6.23)\n\npgfree (since Linux 2.6.0)\n\npgactivate (since Linux 2.6.0)\n\npgdeactivate (since Linux 2.6.0)\n\npgfault (since Linux 2.6.0)\n\npgmajfault (since Linux 2.6.0)\n\npgrefilldma (since Linux 2.6.5)\n\npgrefilldma32 (since Linux 2.6.16)\n\npgrefillnormal (since Linux 2.6.5)\n\npgrefillhigh (since Linux 2.6.5)\n\npgrefillmovable (since Linux 2.6.23)\n\npgstealkswapddma (since Linux 3.4)\n\npgstealkswapddma32 (since Linux 3.4)\n\npgstealkswapdnormal (since Linux 3.4)\n\npgstealkswapdhigh (since Linux 3.4)\n\npgstealkswapdmovable (since Linux 3.4)\n\npgstealdirectdma\n\npgstealdirectdma32 (since Linux 3.4)\n\npgstealdirectnormal (since Linux 3.4)\n\npgstealdirecthigh (since Linux 3.4)\n\npgstealdirectmovable (since Linux 2.6.23)\n\npgscankswapddma\n\npgscankswapddma32 (since Linux 2.6.16)\n\npgscankswapdnormal (since Linux 2.6.5)\n\npgscankswapdhigh\n\npgscankswapdmovable (since Linux 2.6.23)\n\npgscandirectdma\n\npgscandirectdma32 (since Linux 2.6.16)\n\npgscandirectnormal\n\npgscandirecthigh\n\npgscandirectmovable (since Linux 2.6.23)\n\npgscandirectthrottle (since Linux 3.6)\n\nzonereclaimfailed (since linux 2.6.31)\n\npginodesteal (since linux 2.6.0)\n\nslabsscanned (since linux 2.6.5)\n\nkswapdinodesteal (since linux 2.6.0)\n\nkswapdlowwmarkhitquickly (since 2.6.33)\n\nkswapdhighwmarkhitquickly (since 2.6.33)\n\npageoutrun (since Linux 2.6.0)\n\nallocstall (since Linux 2.6.0)\n\npgrotated (since Linux 2.6.0)\n\ndroppagecache (since Linux 3.15)\n\ndropslab (since Linux 3.15)\n\nnumapteupdates (since Linux 3.8)\n\nnumahugepteupdates (since Linux 3.13)\n\nnumahintfaults (since Linux 3.8)\n\nnumahintfaultslocal (since Linux 3.8)\n\nnumapagesmigrated (since Linux 3.8)\n\npgmigratesuccess (since Linux 3.8)\n\npgmigratefail (since Linux 3.8)\n\ncompactmigratescanned (since Linux 3.8)\n\ncompactfreescanned (since Linux 3.8)\n\ncompactisolated (since Linux 3.8)\n\ncompactstall (since Linux 2.6.35)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\ncompactfail (since Linux 2.6.35)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\ncompactsuccess (since Linux 2.6.35)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nhtlbbuddyallocsuccess (since Linux 2.6.26)\n\nhtlbbuddyallocfail (since Linux 2.6.26)\n\nunevictablepgsculled (since Linux 2.6.28)\n\nunevictablepgsscanned (since Linux 2.6.28)\n\nunevictablepgsrescued (since Linux 2.6.28)\n\nunevictablepgsmlocked (since Linux 2.6.28)\n\nunevictablepgsmunlocked (since Linux 2.6.28)\n\nunevictablepgscleared (since Linux 2.6.28)\n\nunevictablepgsstranded (since Linux 2.6.28)\n\nthpfaultalloc (since Linux 2.6.39)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nthpfaultfallback (since Linux 2.6.39)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nthpcollapsealloc (since Linux 2.6.39)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nthpcollapseallocfailed (since Linux 2.6.39)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nthpsplit (since Linux 2.6.39)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nthpzeropagealloc (since Linux 3.8)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nthpzeropageallocfailed (since Linux 3.8)\nSee the kernel source file Documentation/admin-guide/mm/transhuge.rst.\n\nballooninflate (since Linux 3.18)\n\nballoondeflate (since Linux 3.18)\n\nballoonmigrate (since Linux 3.18)\n\nnrtlbremoteflush (since Linux 3.12)\n\nnrtlbremoteflushreceived (since Linux 3.12)\n\nnrtlblocalflushall (since Linux 3.12)\n\nnrtlblocalflushone (since Linux 3.12)\n\nvmacachefindcalls (since Linux 3.16)\n\nvmacachefindhits (since Linux 3.16)\n\nvmacachefullflushes (since Linux 3.19)\n\n/proc/zoneinfo (since Linux 2.6.13)\nThis  file displays information about memory zones.  This is useful for analyzing vir‐\ntual memory behavior.\n"
                }
            ]
        },
        "NOTES": {
            "content": "Many files contain strings (e.g., the environment and command line) that are in the  internal\nformat,  with subfields terminated by null bytes ('\\0').  When inspecting such files, you may\nfind that the results are more readable if you use a command of the following form to display\nthem:\n\n$ cat file | tr '\\000' '\\n'\n\nThis  manual  page is incomplete, possibly inaccurate, and is the kind of thing that needs to\nbe updated very often.\n",
            "subsections": []
        },
        "SEE ALSO": {
            "content": "cat(1), dmesg(1), find(1), free(1), htop(1), init(1), ps(1), pstree(1), tr(1), uptime(1), ch‐‐\nroot(2),  mmap(2),  readlink(2),  syslog(2),  slabinfo(5),  sysfs(5), hier(7), namespaces(7),\ntime(7),  arp(8),  hdparm(8),  ifconfig(8),   lsmod(8),   lspci(8),   mount(8),   netstat(8),\nprocinfo(8), route(8), sysctl(8)\n\nThe    Linux    kernel    source    files:   Documentation/filesystems/proc.txt,   Documenta‐\ntion/sysctl/fs.txt, Documentation/sysctl/kernel.txt, Documentation/sysctl/net.txt, and  Docu‐\nmentation/sysctl/vm.txt.\n",
            "subsections": []
        },
        "COLOPHON": {
            "content": "This  page  is  part  of  release  5.10 of the Linux man-pages project.  A description of the\nproject, information about reporting bugs, and the latest version of this page, can be  found\nat https://www.kernel.org/doc/man-pages/.\n\n\n\nLinux                                        2020-12-21                                      PROC(5)",
            "subsections": []
        }
    },
    "summary": "proc - process information pseudo-filesystem",
    "flags": [],
    "examples": [],
    "see_also": [
        {
            "name": "cat",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/cat/1/json"
        },
        {
            "name": "dmesg",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/dmesg/1/json"
        },
        {
            "name": "find",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/find/1/json"
        },
        {
            "name": "free",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/free/1/json"
        },
        {
            "name": "htop",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/htop/1/json"
        },
        {
            "name": "init",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/init/1/json"
        },
        {
            "name": "ps",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/ps/1/json"
        },
        {
            "name": "pstree",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/pstree/1/json"
        },
        {
            "name": "tr",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/tr/1/json"
        },
        {
            "name": "uptime",
            "section": "1",
            "url": "https://www.chedong.com/phpMan.php/man/uptime/1/json"
        },
        {
            "name": "root",
            "section": "2",
            "url": "https://www.chedong.com/phpMan.php/man/root/2/json"
        },
        {
            "name": "mmap",
            "section": "2",
            "url": "https://www.chedong.com/phpMan.php/man/mmap/2/json"
        },
        {
            "name": "readlink",
            "section": "2",
            "url": "https://www.chedong.com/phpMan.php/man/readlink/2/json"
        },
        {
            "name": "syslog",
            "section": "2",
            "url": "https://www.chedong.com/phpMan.php/man/syslog/2/json"
        },
        {
            "name": "slabinfo",
            "section": "5",
            "url": "https://www.chedong.com/phpMan.php/man/slabinfo/5/json"
        },
        {
            "name": "sysfs",
            "section": "5",
            "url": "https://www.chedong.com/phpMan.php/man/sysfs/5/json"
        },
        {
            "name": "hier",
            "section": "7",
            "url": "https://www.chedong.com/phpMan.php/man/hier/7/json"
        },
        {
            "name": "namespaces",
            "section": "7",
            "url": "https://www.chedong.com/phpMan.php/man/namespaces/7/json"
        },
        {
            "name": "time",
            "section": "7",
            "url": "https://www.chedong.com/phpMan.php/man/time/7/json"
        },
        {
            "name": "arp",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/arp/8/json"
        },
        {
            "name": "hdparm",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/hdparm/8/json"
        },
        {
            "name": "ifconfig",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/ifconfig/8/json"
        },
        {
            "name": "lsmod",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/lsmod/8/json"
        },
        {
            "name": "lspci",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/lspci/8/json"
        },
        {
            "name": "mount",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/mount/8/json"
        },
        {
            "name": "netstat",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/netstat/8/json"
        },
        {
            "name": "procinfo",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/procinfo/8/json"
        },
        {
            "name": "route",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/route/8/json"
        },
        {
            "name": "sysctl",
            "section": "8",
            "url": "https://www.chedong.com/phpMan.php/man/sysctl/8/json"
        }
    ]
}