The /proc Filesystem

문서 범위, 목차, 서문

이 문서는 `/proc` filesystem의 system·process 정보 수집, `/proc/sys`를 통한 runtime parameter 변경, process별 control file, procfs mount option과 instance 동작을 설명한다. 초판 `/proc/sys` 자료와 2.4.x 갱신, `/proc/sys` 이동 및 후속 수정의 작성자와 날짜를 원문 표에 기록한다.

1장은 process directory, kernel data, networking, SCSI, parallel port, TTY, `/proc/stat`, ext4와 console 정보를 다룬다. 2장은 system parameter 변경, 3장은 OOM·I/O·core dump·mount·fdinfo·KSM 등 process별 parameter, 4장은 mount option, 5장은 filesystem instance 동작을 설명한다.

Alan Cox, Rik van Riel, Alexey Kuznetsov를 비롯한 기여자와 특히 Andi Kleen의 문서를 바탕으로 작성됐으며 최신 문서는 kernel.org HTML에서 볼 수 있다. 정확성을 보증하지 않는다는 법적 고지도 포함한다.

`/proc`은 kernel 내부 data structure에 대한 interface다. 실행 중인 system 정보를 읽을 수 있을 뿐 아니라 sysctl을 통해 특정 kernel parameter를 runtime에 바꿀 수 있다. 1장은 read-only 정보를, 2장은 `/proc/sys` 변경 방법을 다룬다.

.. SPDX-License-Identifier: GPL-2.0

====================
The /proc Filesystem
====================

=====================  =======================================  ================
/proc/sys              Terrehon Bowden <terrehon@pacbell.net>,  October 7 1999
                       Bodo Bauer <bb@ricochet.net>
2.4.x update               Jorge Nerin <comandante@zaralinux.com>   November 14 2000
move /proc/sys               Shen Feng <shen@cn.fujitsu.com>                April 1 2009
fixes/update part 1.1  Stefani Seibold <stefani@seibold.net>    June 9 2009
=====================  =======================================  ================



.. Table of Contents

  0     Preface
  0.1        Introduction/Credits
  0.2        Legal Stuff

  1        Collecting System Information
  1.1        Process-Specific Subdirectories
  1.2        Kernel data
  1.3        IDE devices in /proc/ide
  1.4        Networking info in /proc/net
  1.5        SCSI info
  1.6        Parallel port info in /proc/parport
  1.7        TTY info in /proc/tty
  1.8        Miscellaneous kernel statistics in /proc/stat
  1.9        Ext4 file system parameters

  2        Modifying System Parameters

  3        Per-Process Parameters
  3.1        /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer
                                                                score
  3.2        /proc/<pid>/oom_score - Display current oom-killer score
  3.3        /proc/<pid>/io - Display the IO accounting fields
  3.4        /proc/<pid>/coredump_filter - Core dump filtering settings
  3.5        /proc/<pid>/mountinfo - Information about mounts
  3.6        /proc/<pid>/comm  & /proc/<pid>/task/<tid>/comm
  3.7   /proc/<pid>/task/<tid>/children - Information about task children
  3.8   /proc/<pid>/fdinfo/<fd> - Information about opened file
  3.9   /proc/<pid>/map_files - Information about memory mapped files
  3.10  /proc/<pid>/timerslack_ns - Task timerslack value
  3.11        /proc/<pid>/patch_state - Livepatch patch operation state
  3.12        /proc/<pid>/arch_status - Task architecture specific information
  3.13  /proc/<pid>/fd - List of symlinks to open files
  3.14  /proc/<pid/ksm_stat - Information about the process's ksm status.

  4        Configuring procfs
  4.1        Mount options

  5        Filesystem behavior

Preface
=======

0.1 Introduction/Credits
------------------------

We'd like  to  thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of
other people for help compiling this documentation. We'd also like to extend a
special thank  you to Andi Kleen for documentation, which we relied on heavily
to create  this  document,  as well as the additional information he provided.
Thanks to  everybody  else  who contributed source or docs to the Linux kernel
and helped create a great piece of software... :)

The   latest   version    of   this   document   is    available   online   at
https://www.kernel.org/doc/html/latest/filesystems/proc.html

0.2 Legal Stuff
---------------

We don't  guarantee  the  correctness  of this document, and if you come to us
complaining about  how  you  screwed  up  your  system  because  of  incorrect
documentation, we won't feel responsible...

Chapter 1: Collecting System Information
========================================

In This Chapter
---------------
* Investigating  the  properties  of  the  pseudo  file  system  /proc and its
  ability to provide information on the running Linux system
* Examining /proc's structure
* Uncovering  various  information  about the kernel and the processes running
  on the system

------------------------------------------------------------------------------

The proc  file  system acts as an interface to internal data structures in the
kernel. It  can  be  used to obtain information about the system and to change
certain kernel parameters at runtime (sysctl).

First, we'll  take  a  look  at the read-only parts of /proc. In Chapter 2, we
show you how you can use /proc/sys to change settings.

Process별 directory와 접근 권한

`/proc`에는 실행 중인 각 process ID를 이름으로 한 subdirectory가 있고 `self` symlink는 filesystem을 읽는 process 자신을 가리킨다. 자기 `/proc/PID/*` 정보는 추가 권한 없이 읽을 수 있다. 다른 process의 `maps`, `environ`, `pagemap` 같은 read-only 정보에는 `PTRACE_MODE_READ`가 허용된 `CAP_SYS_PTRACE` 또는 `CAP_PERFMON`이 필요하다.

예외인 `mem`은 read-write 성격 때문에 더 강한 `PTRACE_MODE_ATTACH` 조건의 `CAP_SYS_PTRACE`가 필요하며 `CAP_PERFMON`만으로 다른 process의 `/proc/PID/mem`에 접근할 수 없다.

열린 `/proc/<pid>` descriptor는 process가 종료된 뒤 PID 재사용을 막지 않는다. 하지만 기존 FD의 operation이 우연히 같은 PID를 받은 새 process에 작용하지는 않으며, 보통 `ESRCH`로 실패한다.

`/proc/PID/status`는 `ps`가 procfs에서 얻는 정보보다 자세한 사람이 읽는 view를 제공한다. 예에는 identity, UID/GID, memory, signal, capability, seccomp, speculation mitigation, context switch가 나온다. `statm`은 7개 memory field, `stat`은 process 자체의 저수준 field를 제공한다.

1.1 Process-Specific Subdirectories
-----------------------------------

The directory  /proc  contains  (among other things) one subdirectory for each
process running on the system, which is named after the process ID (PID).

The link  'self'  points to  the process reading the file system. Each process
subdirectory has the entries listed in Table 1-1.

A process can read its own information from /proc/PID/* with no extra
permissions. When reading /proc/PID/* information for other processes, reading
process is required to have either CAP_SYS_PTRACE capability with
PTRACE_MODE_READ access permissions, or, alternatively, CAP_PERFMON
capability. This applies to all read-only information like `maps`, `environ`,
`pagemap`, etc. The only exception is `mem` file due to its read-write nature,
which requires CAP_SYS_PTRACE capabilities with more elevated
PTRACE_MODE_ATTACH permissions; CAP_PERFMON capability does not grant access
to /proc/PID/mem for other processes.

Note that an open file descriptor to /proc/<pid> or to any of its
contained files or subdirectories does not prevent <pid> being reused
for some other process in the event that <pid> exits. Operations on
open /proc/<pid> file descriptors corresponding to dead processes
never act on any new process that the kernel may, through chance, have
also assigned the process ID <pid>. Instead, operations on these FDs
usually fail with ESRCH.

.. table:: Table 1-1: Process specific entries in /proc

 =============  ===============================================================
 File                Content
 =============  ===============================================================
 clear_refs        Clears page referenced bits shown in smaps output
 cmdline        Command line arguments
 cpu                Current and last cpu in which it was executed        (2.4)(smp)
 cwd                Link to the current working directory
 environ        Values of environment variables
 exe                Link to the executable of this process
 fd                Directory, which contains all file descriptors
 maps                Memory maps to executables and library files        (2.4)
 mem                Memory held by this process
 root                Link to the root directory of this process
 stat                Process status
 statm                Process memory status information
 status                Process status in human readable form
 wchan                Present with CONFIG_KALLSYMS=y: it shows the kernel function
                symbol the task is blocked in - or "0" if not blocked.
 pagemap        Page table
 stack                Report full stack trace, enable via CONFIG_STACKTRACE
 smaps                An extension based on maps, showing the memory consumption of
                each mapping and flags associated with it
 smaps_rollup        Accumulated smaps stats for all mappings of the process.  This
                can be derived from smaps, but is faster and more convenient
 numa_maps        An extension based on maps, showing the memory locality and
                binding policy as well as mem usage (in pages) of each mapping.
 =============  ===============================================================

For example, to get the status information of a process, all you have to do is
read the file /proc/PID/status::

  >cat /proc/self/status
  Name:   cat
  State:  R (running)
  Tgid:   5452
  Pid:    5452
  PPid:   743
  TracerPid:      0                                                (2.4)
  Uid:    501     501     501     501
  Gid:    100     100     100     100
  FDSize: 256
  Groups: 100 14 16
  Kthread:    0
  VmPeak:     5004 kB
  VmSize:     5004 kB
  VmLck:         0 kB
  VmHWM:       476 kB
  VmRSS:       476 kB
  RssAnon:             352 kB
  RssFile:             120 kB
  RssShmem:              4 kB
  VmData:      156 kB
  VmStk:        88 kB
  VmExe:        68 kB
  VmLib:      1412 kB
  VmPTE:        20 kb
  VmSwap:        0 kB
  HugetlbPages:          0 kB
  CoreDumping:    0
  THP_enabled:          1
  Threads:        1
  SigQ:   0/28578
  SigPnd: 0000000000000000
  ShdPnd: 0000000000000000
  SigBlk: 0000000000000000
  SigIgn: 0000000000000000
  SigCgt: 0000000000000000
  CapInh: 00000000fffffeff
  CapPrm: 0000000000000000
  CapEff: 0000000000000000
  CapBnd: ffffffffffffffff
  CapAmb: 0000000000000000
  NoNewPrivs:     0
  Seccomp:        0
  Speculation_Store_Bypass:       thread vulnerable
  SpeculationIndirectBranch:      conditional enabled
  voluntary_ctxt_switches:        0
  nonvoluntary_ctxt_switches:     1

This shows you nearly the same information you would get if you viewed it with
the ps  command.  In  fact,  ps  uses  the  proc  file  system  to  obtain its
information.  But you get a more detailed  view of the  process by reading the
file /proc/PID/status. It fields are described in table 1-2.

The  statm  file  contains  more  detailed  information about the process

`status`, `statm`, `stat` field

SMP에서 RSS 관련 accounting은 확장성을 위해 비동기로 처리되므로 값이 정확한 순간 snapshot이 아닐 수 있다. 정확한 값이 필요하면 느리지만 page table을 scan하는 `/proc/<pid>/smaps`를 읽는다.

`statm`의 `size`와 `resident`는 page 단위의 program size와 RSS다. `shared`는 file-backed page로 `RssFile+RssShmem`에 해당한다. `trs`와 `drs`는 각각 code 및 data/stack을 나타내지만 2.6에서 의미가 깨져 있고, `lrs`와 `dt`는 항상 0이다.

`stat`은 `pid`, executable `tcomm`, state, parent·process group·session·TTY ID, task flags, minor/major fault, user·kernel·child time, priority·nice·thread 수, start time, virtual size·RSS·limit, code/data/stack/heap/argument/environment 주소, signal bitmap, exit signal·code, CPU·scheduler policy·realtime priority, block I/O wait와 guest time을 순서대로 제공한다. 옛 `wchan` 위치와 두 항목은 0 placeholder다.

memory usage. Its seven fields are explained in Table 1-3.  The stat file
contains detailed information about the process itself.  Its fields are
explained in Table 1-4.

(for SMP CONFIG users)

For making accounting scalable, RSS related information are handled in an
asynchronous manner and the value may not be very precise. To see a precise
snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table.
It's slow but very precise.

.. table:: Table 1-2: Contents of the status fields (as of 4.19)

 ==========================  ===================================================
 Field                       Content
 ==========================  ===================================================
 Name                        filename of the executable
 Umask                       file mode creation mask
 State                       state (R is running, S is sleeping, D is sleeping
                             in an uninterruptible wait, Z is zombie,
                             T is traced or stopped)
 Tgid                        thread group ID
 Ngid                        NUMA group ID (0 if none)
 Pid                         process id
 PPid                        process id of the parent process
 TracerPid                   PID of process tracing this process (0 if not, or
                             the tracer is outside of the current pid namespace)
 Uid                         Real, effective, saved set, and  file system UIDs
 Gid                         Real, effective, saved set, and  file system GIDs
 FDSize                      number of file descriptor slots currently allocated
 Groups                      supplementary group list
 NStgid                      descendant namespace thread group ID hierarchy
 NSpid                       descendant namespace process ID hierarchy
 NSpgid                      descendant namespace process group ID hierarchy
 NSsid                       descendant namespace session ID hierarchy
 Kthread                     kernel thread flag, 1 is yes, 0 is no
 VmPeak                      peak virtual memory size
 VmSize                      total program size
 VmLck                       locked memory size
 VmPin                       pinned memory size
 VmHWM                       peak resident set size ("high water mark")
 VmRSS                       size of memory portions. It contains the three
                             following parts
                             (VmRSS = RssAnon + RssFile + RssShmem)
 RssAnon                     size of resident anonymous memory
 RssFile                     size of resident file mappings
 RssShmem                    size of resident shmem memory (includes SysV shm,
                             mapping of tmpfs and shared anonymous mappings)
 VmData                      size of private data segments
 VmStk                       size of stack segments
 VmExe                       size of text segment
 VmLib                       size of shared library code
 VmPTE                       size of page table entries
 VmSwap                      amount of swap used by anonymous private data
                             (shmem swap usage is not included)
 HugetlbPages                size of hugetlb memory portions
 CoreDumping                 process's memory is currently being dumped
                             (killing the process may lead to a corrupted core)
 THP_enabled                 process is allowed to use THP (returns 0 when
                             PR_SET_THP_DISABLE is set on the process to disable
                             THP completely, not just partially)
 Threads                     number of threads
 SigQ                        number of signals queued/max. number for queue
 SigPnd                      bitmap of pending signals for the thread
 ShdPnd                      bitmap of shared pending signals for the process
 SigBlk                      bitmap of blocked signals
 SigIgn                      bitmap of ignored signals
 SigCgt                      bitmap of caught signals
 CapInh                      bitmap of inheritable capabilities
 CapPrm                      bitmap of permitted capabilities
 CapEff                      bitmap of effective capabilities
 CapBnd                      bitmap of capabilities bounding set
 CapAmb                      bitmap of ambient capabilities
 NoNewPrivs                  no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...)
 Seccomp                     seccomp mode, like prctl(PR_GET_SECCOMP, ...)
 Speculation_Store_Bypass    speculative store bypass mitigation status
 SpeculationIndirectBranch   indirect branch speculation mode
 Cpus_allowed                mask of CPUs on which this process may run
 Cpus_allowed_list           Same as previous, but in "list format"
 Mems_allowed                mask of memory nodes allowed to this process
 Mems_allowed_list           Same as previous, but in "list format"
 voluntary_ctxt_switches     number of voluntary context switches
 nonvoluntary_ctxt_switches  number of non voluntary context switches
 ==========================  ===================================================


.. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3)

 ======== ===============================        ==============================
 Field    Content
 ======== ===============================        ==============================
 size     total program size (pages)                (same as VmSize in status)
 resident size of memory portions (pages)        (same as VmRSS in status)
 shared   number of pages that are shared        (i.e. backed by a file, same
                                                as RssFile+RssShmem in status)
 trs      number of pages that are 'code'        (not including libs; broken,
                                                includes data segment)
 lrs      number of pages of library                (always 0 on 2.6)
 drs      number of pages of data/stack                (including libs; broken,
                                                includes library text)
 dt       number of dirty pages                        (always 0 on 2.6)
 ======== ===============================        ==============================


.. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7)

  ============= ===============================================================
  Field         Content
  ============= ===============================================================
  pid           process id
  tcomm         filename of the executable
  state         state (R is running, S is sleeping, D is sleeping in an
                uninterruptible wait, Z is zombie, T is traced or stopped)
  ppid          process id of the parent process
  pgrp          pgrp of the process
  sid           session id
  tty_nr        tty the process uses
  tty_pgrp      pgrp of the tty
  flags         task flags
  min_flt       number of minor faults
  cmin_flt      number of minor faults with child's
  maj_flt       number of major faults
  cmaj_flt      number of major faults with child's
  utime         user mode jiffies
  stime         kernel mode jiffies
  cutime        user mode jiffies with child's
  cstime        kernel mode jiffies with child's
  priority      priority level
  nice          nice level
  num_threads   number of threads
  it_real_value        (obsolete, always 0)
  start_time    time the process started after system boot
  vsize         virtual memory size
  rss           resident set memory size
  rsslim        current limit in bytes on the rss
  start_code    address above which program text can run
  end_code      address below which program text can run
  start_stack   address of the start of the main process stack
  esp           current value of ESP
  eip           current value of EIP
  pending       bitmap of pending signals
  blocked       bitmap of blocked signals
  sigign        bitmap of ignored signals
  sigcatch      bitmap of caught signals
  0                (place holder, used to be the wchan address,
                use /proc/PID/wchan instead)
  0             (place holder)
  0             (place holder)
  exit_signal   signal to send to parent thread on exit
  task_cpu      which CPU the task is scheduled on
  rt_priority   realtime priority
  policy        scheduling policy (man sched_setscheduler)
  blkio_ticks   time spent waiting for block IO
  gtime         guest time of the task in jiffies
  cgtime        guest time of the task children in jiffies
  start_data    address above which program data+bss is placed
  end_data      address below which program data+bss is placed
  start_brk     address above which program heap can be expanded with brk()
  arg_start     address above which program command line is placed
  arg_end       address below which program command line is placed
  env_start     address above which program environment is placed
  env_end       address below which program environment is placed
  exit_code     the thread's exit_code in the form reported by the waitpid
                system call
  ============= ===============================================================

`/proc/PID/maps` 형식

`/proc/PID/maps`는 현재 memory mapping region과 access permission을 한 줄씩 표시한다. column은 `address`, `perms`, file `offset`, device `major:minor`, `inode`, `pathname` 순서다.

file 없는 특별 mapping은 `[heap]`, main process `[stack]`, kernel system-call handler인 `[vdso]`로 표시된다. userspace가 이름을 붙인 private anonymous mapping은 `[anon:<name>]`, anonymous shared memory는 `[anon_shmem:<name>]`다. pathname이 비어 있으면 일반 anonymous mapping이다.

Linux 6.11부터 `/proc/PID/maps`에는 VMA를 효율적으로 query·filter하는 binary `ioctl()` API도 있다. UAPI `linux/fs.h`의 `struct procmap_query`가 `PROCMAP_QUERY`의 input/output 인자이며 정확한 flag와 반환 semantics는 해당 header comment를 따른다.

The /proc/PID/maps file contains the currently mapped memory regions and
their access permissions.

The format is::

    address           perms offset  dev   inode      pathname

    08048000-08049000 r-xp 00000000 03:00 8312       /opt/test
    08049000-0804a000 rw-p 00001000 03:00 8312       /opt/test
    0804a000-0806b000 rw-p 00000000 00:00 0          [heap]
    a7cb1000-a7cb2000 ---p 00000000 00:00 0
    a7cb2000-a7eb2000 rw-p 00000000 00:00 0
    a7eb2000-a7eb3000 ---p 00000000 00:00 0
    a7eb3000-a7ed5000 rw-p 00000000 00:00 0
    a7ed5000-a8008000 r-xp 00000000 03:00 4222       /lib/libc.so.6
    a8008000-a800a000 r--p 00133000 03:00 4222       /lib/libc.so.6
    a800a000-a800b000 rw-p 00135000 03:00 4222       /lib/libc.so.6
    a800b000-a800e000 rw-p 00000000 00:00 0
    a800e000-a8022000 r-xp 00000000 03:00 14462      /lib/libpthread.so.0
    a8022000-a8023000 r--p 00013000 03:00 14462      /lib/libpthread.so.0
    a8023000-a8024000 rw-p 00014000 03:00 14462      /lib/libpthread.so.0
    a8024000-a8027000 rw-p 00000000 00:00 0
    a8027000-a8043000 r-xp 00000000 03:00 8317       /lib/ld-linux.so.2
    a8043000-a8044000 r--p 0001b000 03:00 8317       /lib/ld-linux.so.2
    a8044000-a8045000 rw-p 0001c000 03:00 8317       /lib/ld-linux.so.2
    aff35000-aff4a000 rw-p 00000000 00:00 0          [stack]
    ffffe000-fffff000 r-xp 00000000 00:00 0          [vdso]

where "address" is the address space in the process that it occupies, "perms"
is a set of permissions::

 r = read
 w = write
 x = execute
 s = shared
 p = private (copy on write)

"offset" is the offset into the mapping, "dev" is the device (major:minor), and
"inode" is the inode  on that device.  0 indicates that  no inode is associated
with the memory region, as the case would be with BSS (uninitialized data).
The "pathname" shows the name associated file for this mapping.  If the mapping
is not associated with a file:

 ===================        ===========================================
 [heap]                     the heap of the program
 [stack]                    the stack of the main process
 [vdso]                     the "virtual dynamic shared object",
                            the kernel system call handler
 [anon:<name>]              a private anonymous mapping that has been
                            named by userspace
 [anon_shmem:<name>]        an anonymous shared memory mapping that has
                            been named by userspace
 ===================        ===========================================

 or if empty, the mapping is anonymous.

Starting with 6.11 kernel, /proc/PID/maps provides an alternative
ioctl()-based API that gives ability to flexibly and efficiently query and
filter individual VMAs. This interface is binary and is meant for more
efficient and easy programmatic use. `struct procmap_query`, defined in
linux/fs.h UAPI header, serves as an input/output argument to the
`PROCMAP_QUERY` ioctl() command. See comments in linus/fs.h UAPI header for
details on query semantics, supported flags, data returned, and general API
usage information.

`smaps` memory accounting

`/proc/PID/smaps`는 `maps`를 확장해 각 VMA의 memory 소비를 보여 준다. 첫 줄은 `maps`와 같고 뒤에는 mapping size, kernel과 MMU page size, RAM resident size `Rss`, proportional share `Pss`, dirty PSS, shared/private clean·dirty page, reference·anonymous·KSM·lazy-free·hugepage·swap·lock·THP eligibility와 `VmFlags`가 이어진다.

PSS는 각 resident page를 그 page를 공유하는 process 수로 나눠 합한 값이다. 독점 page 1000개와 두 process가 공유하는 page 1000개가 있으면 PSS는 1500 page다. `Pss_Dirty`는 dirty portion이고 clean portion은 `Pss-Pss_Dirty`로 계산한다.

전통적 accounting은 한 번 mapping된 page를 private, 여러 번 mapping된 page를 shared로 분류하며 `MAP_SHARED`와는 독립적이다. THP 같은 큰 allocation은 구성 page가 모두 같은 process에 있다고 확실할 때 private, 다른 process에 있을 가능성이 있으면 shared로 볼 수 있다. mapping 횟수를 정밀하게 추적하지 않는 구성은 큰 allocation의 page별 평균 mapping 수를 사용하므로 PSS가 근사치가 된다.

file mapping도 `MAP_PRIVATE` page가 수정되면 file page가 private anonymous copy로 대체돼 `Anonymous`에 포함될 수 있다. `KSM`은 실제 KSM page만 세며 KSM이 배치한 zero page는 제외한다. `LazyFree`는 구현 최적화 때문에 실제보다 낮게 출력될 수 있다.

The /proc/PID/smaps is an extension based on maps, showing the memory
consumption for each of the process's mappings. For each mapping (aka Virtual
Memory Area, or VMA) there is a series of lines such as the following::

    08048000-080bc000 r-xp 00000000 03:02 13130      /bin/bash

    Size:               1084 kB
    KernelPageSize:        4 kB
    MMUPageSize:           4 kB
    Rss:                 892 kB
    Pss:                 374 kB
    Pss_Dirty:             0 kB
    Shared_Clean:        892 kB
    Shared_Dirty:          0 kB
    Private_Clean:         0 kB
    Private_Dirty:         0 kB
    Referenced:          892 kB
    Anonymous:             0 kB
    KSM:                   0 kB
    LazyFree:              0 kB
    AnonHugePages:         0 kB
    ShmemPmdMapped:        0 kB
    Shared_Hugetlb:        0 kB
    Private_Hugetlb:       0 kB
    Swap:                  0 kB
    SwapPss:               0 kB
    KernelPageSize:        4 kB
    MMUPageSize:           4 kB
    Locked:                0 kB
    THPeligible:           0
    VmFlags: rd ex mr mw me dw

The first of these lines shows the same information as is displayed for
the mapping in /proc/PID/maps.  Following lines show the size of the
mapping (size); the size of each page allocated when backing a VMA
(KernelPageSize), which is usually the same as the size in the page table
entries; the page size used by the MMU when backing a VMA (in most cases,
the same as KernelPageSize); the amount of the mapping that is currently
resident in RAM (RSS); the process's proportional share of this mapping
(PSS); and the number of clean and dirty shared and private pages in the
mapping.

The "proportional set size" (PSS) of a process is the count of pages it has
in memory, where each page is divided by the number of processes sharing it.
So if a process has 1000 pages all to itself, and 1000 shared with one other
process, its PSS will be 1500.  "Pss_Dirty" is the portion of PSS which
consists of dirty pages.  ("Pss_Clean" is not included, but it can be
calculated by subtracting "Pss_Dirty" from "Pss".)

Traditionally, a page is accounted as "private" if it is mapped exactly once,
and a page is accounted as "shared" when mapped multiple times, even when
mapped in the same process multiple times. Note that this accounting is
independent of MAP_SHARED.

In some kernel configurations, the semantics of pages part of a larger
allocation (e.g., THP) can differ: a page is accounted as "private" if all
pages part of the corresponding large allocation are *certainly* mapped in the
same process, even if the page is mapped multiple times in that process. A
page is accounted as "shared" if any page page of the larger allocation
is *maybe* mapped in a different process. In some cases, a large allocation
might be treated as "maybe mapped by multiple processes" even though this
is no longer the case.

Some kernel configurations do not track the precise number of times a page part
of a larger allocation is mapped. In this case, when calculating the PSS, the
average number of mappings per page in this larger allocation might be used
as an approximation for the number of mappings of a page. The PSS calculation
will be imprecise in this case.

"Referenced" indicates the amount of memory currently marked as referenced or
accessed.

"Anonymous" shows the amount of memory that does not belong to any file.  Even
a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE
and a page is modified, the file page is replaced by a private anonymous copy.

"KSM" reports how many of the pages are KSM pages. Note that KSM-placed zeropages
are not included, only actual KSM pages.

"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE).
The memory isn't freed immediately with madvise(). It's freed in memory

`smaps` hugepage·swap·VmFlags와 race

`AnonHugePages`는 THP-backed memory, `ShmemPmdMapped`는 huge page로 뒷받침된 shared shmem/tmpfs다. `Shared_Hugetlb`와 `Private_Hugetlb`는 역사적 이유로 `RSS`, `PSS`, clean/dirty 분해에 포함되지 않는 hugetlbfs memory다.

`Swap`은 swap에 나간 would-be-anonymous memory를 나타낸다. shmem mapping에서는 COW로 대체되지 않은 underlying shmem object의 mapped swap 부분도 포함한다. `SwapPss`는 proportional swap share지만 underlying shmem object의 swapped page는 세지 않는다. `Locked`는 memory lock 여부, `THPeligible`은 현재 활성 크기 중 자연 정렬 THP를 할당할 수 있으면 1이다.

flag와 mnemonic은 kernel release마다 추가·삭제되거나 의미가 바뀔 수 있으므로 consumer는 대상 kernel version의 semantics를 따라야 한다. `smaps`는 `CONFIG_MMU`가 켜진 경우에만 존재한다.

`maps`와 `smaps` 읽기는 본질적으로 race가 있으며 single read call만 일관된 출력을 만들 수 있다. 부분 read 중 memory map이 바뀔 수 있지만 address는 뒤로 가지 않아 region이 중첩되지 않고, walk 전체 생애 동안 특정 virtual address에 mapping이 계속 있었다면 그에 대한 출력은 적어도 하나 제공된다.

pressure if the memory is clean. Please note that the printed value might
be lower than the real value due to optimizations used in the current
implementation. If this is not desirable please file a bug report.

"AnonHugePages" shows the amount of memory backed by transparent hugepage.

"ShmemPmdMapped" shows the amount of shared (shmem/tmpfs) memory backed by
huge pages.

"Shared_Hugetlb" and "Private_Hugetlb" show the amounts of memory backed by
hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical
reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field.

"Swap" shows how much would-be-anonymous memory is also used, but out on swap.

For shmem mappings, "Swap" includes also the size of the mapped (and not
replaced by copy-on-write) part of the underlying shmem object out on swap.
"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this
does not take into account swapped out page of underlying shmem objects.
"Locked" indicates whether the mapping is locked in memory or not.

"THPeligible" indicates whether the mapping is eligible for allocating
naturally aligned THP pages of any currently enabled size. 1 if true, 0
otherwise.

"VmFlags" field deserves a separate description. This member represents the
kernel flags associated with the particular virtual memory area in two letter
encoded manner. The codes are the following:

    ==    =============================================================
    rd    readable
    wr    writeable
    ex    executable
    sh    shared
    mr    may read
    mw    may write
    me    may execute
    ms    may share
    gd    stack segment growns down
    pf    pure PFN range
    lo    pages are locked in memory
    io    memory mapped I/O area
    sr    sequential read advise provided
    rr    random read advise provided
    dc    do not copy area on fork
    de    do not expand area on remapping
    ac    area is accountable
    nr    swap space is not reserved for the area
    ht    area uses huge tlb pages
    sf    synchronous page fault
    ar    architecture specific flag
    wf    wipe on fork
    dd    do not include area into core dump
    sd    soft dirty flag
    mm    mixed map area
    hg    huge page advise flag
    nh    no huge page advise flag
    mg    mergeable advise flag
    bt    arm64 BTI guarded page
    mt    arm64 MTE allocation tags are enabled
    um    userfaultfd missing tracking
    uw    userfaultfd wr-protect tracking
    ui    userfaultfd minor fault
    ss    shadow/guarded control stack page
    sl    sealed
    lf    lock on fault pages
    dp    always lazily freeable mapping
    gu    maybe contains guard regions (if not set, definitely doesn't)
    ==    =============================================================

Note that there is no guarantee that every flag and associated mnemonic will
be present in all further kernel releases. Things get changed, the flags may
be vanished or the reverse -- new added. Interpretation of their meaning
might change in future as well. So each consumer of these flags has to
follow each specific kernel version for the exact semantic.

This file is only present if the CONFIG_MMU kernel configuration option is
enabled.

Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent
output can be achieved only in the single read call).

This typically manifests when doing partial reads of these files while the
memory map is being modified.  Despite the races, we do provide the following
guarantees:

1) The mapped addresses never go backwards, which implies no two
   regions will ever overlap.
2) If there is something at a given vaddr during the entirety of the
   life of the smaps/maps walk, there will be some output for it.

`smaps_rollup`, `clear_refs`, pagemap, NUMA map

`smaps_rollup`은 `smaps`의 동일 field를 process의 모든 mapping에 대해 합산한다. 추가 `Pss_Anon`, `Pss_File`, `Pss_Shmem`은 anonymous·file·shmem page의 proportional share다. 각 mapping type이 이미 명확한 `smaps`에서는 생략되며 rollup 결과를 smaps에서 계산할 수 있지만 비용이 훨씬 크다.

`/proc/pid/pagemap`의 PFN은 `/proc/kpageflags`에서 page flag를, `/proc/kpagecount`에서 mapping 횟수를 찾는 데 쓴다. 자세한 bit layout은 `Documentation/admin-guide/mm/pagemap.rst`에 있다.

`numa_maps`는 mapping별 시작 address, NUMA policy, mapping detail을 한 줄에 표시한다. detail에는 file·stack·huge 같은 type, `mapped`, `anon`, `dirty`, `active`, `mapmax`, node별 `N0`, `N1` page count와 `kernelpagesize_kB`가 들어간다.

THP 같은 큰 allocation의 page별 mapping 수를 정밀 추적하지 않는 구성에서는 `mapmax`가 그 allocation의 page당 평균 mapping 수일 수 있다.

The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps,
but their values are the sums of the corresponding values for all mappings of
the process.  Additionally, it contains these fields:

- Pss_Anon
- Pss_File
- Pss_Shmem

They represent the proportional shares of anonymous, file, and shmem pages, as
described for smaps above.  These fields are omitted in smaps since each
mapping identifies the type (anon, file, or shmem) of all pages it contains.
Thus all information in smaps_rollup can be derived from smaps, but at a
significantly higher cost.

The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG
bits on both physical and virtual pages associated with a process, and the
soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst
for details).
To clear the bits for all the pages associated with the process::

    > echo 1 > /proc/PID/clear_refs

To clear the bits for the anonymous pages associated with the process::

    > echo 2 > /proc/PID/clear_refs

To clear the bits for the file mapped pages associated with the process::

    > echo 3 > /proc/PID/clear_refs

To clear the soft-dirty bit::

    > echo 4 > /proc/PID/clear_refs

To reset the peak resident set size ("high water mark") to the process's
current value::

    > echo 5 > /proc/PID/clear_refs

Any other value written to /proc/PID/clear_refs will have no effect.

The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags
using /proc/kpageflags and number of times a page is mapped using
/proc/kpagecount. For detailed explanation, see
Documentation/admin-guide/mm/pagemap.rst.

The /proc/pid/numa_maps is an extension based on maps, showing the memory
locality and binding policy, as well as the memory usage (in pages) of
each mapping. The output follows a general format where mapping details get
summarized separated by blank spaces, one mapping per each file line::

    address   policy    mapping details

    00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4
    00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4
    3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4
    320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
    3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
    3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4
    3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4
    320698b000 default file=/lib64/libc-2.12.so
    3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4
    3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4
    3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4
    7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4
    7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4
    7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048
    7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4
    7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4

Where:

"address" is the starting address for the mapping;

"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst);

"mapping details" summarizes mapping data such as mapping type, page usage counters,
node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page
size, in KB, that is backing the mapping up.

Note that some kernel configurations do not track the precise number of times
a page part of a larger allocation (e.g., THP) is mapped. In these
configurations, "mapmax" might corresponds to the average number of mappings
per page in such a larger allocation instead.

Kernel data entry와 interrupt 통계

process entry와 마찬가지로 `/proc`의 kernel data file은 실행 중인 kernel 상태를 제공한다. 실제 존재 여부는 kernel configuration과 loaded module에 따라 달라진다.

`loadavg`는 1·5·15분 load average, runnable/전체 process 수, 마지막 생성 PID를 표시한다. runnable과 total만 `/`로 구분하고 나머지는 space로 구분한다.

`/proc/interrupts`는 IRQ별로 각 CPU가 처리한 횟수, controller type과 device를 보여 준다. SMP 출력에는 `NMI`, `LOC`, `ERR`가 추가된다. `NMI`는 NMI watchdog lockup 검출용 non-maskable interrupt, `LOC`는 CPU별 local APIC interrupt, `ERR`는 CPU를 연결하는 IO-APIC bus 오류 수다. IO-APIC는 transmission을 자동 재시도한다.

2.6 계열에서는 모든 사용 중인 IRQ vector를 표시하도록 확장됐다. `THR`은 machine-check threshold, `TRM`은 CPU thermal threshold crossing, `SPU`는 source를 식별하기 전에 내려간 spurious interrupt다. `RES`, `CAL`, `TLB`는 CPU 사이 reschedule, call, TLB flush interrupt다.

1.2 Kernel data
---------------

Similar to  the  process entries, the kernel data files give information about
the running kernel. The files used to obtain this information are contained in
/proc and  are  listed  in Table 1-5. Not all of these will be present in your
system. It  depends  on the kernel configuration and the loaded modules, which
files are there, and which are missing.

.. table:: Table 1-5: Kernel info in /proc

 ============ ===============================================================
 File         Content
 ============ ===============================================================
 allocinfo    Memory allocations profiling information
 apm          Advanced power management info
 bootconfig   Kernel command line obtained from boot config,
               and, if there were kernel parameters from the
              boot loader, a "# Parameters from bootloader:"
              line followed by a line containing those
              parameters prefixed by "# ".                        (5.5)
 buddyinfo    Kernel memory allocator information (see text)        (2.5)
 bus          Directory containing bus specific information
 cmdline      Kernel command line, both from bootloader and embedded
              in the kernel image
 cpuinfo      Info about the CPU
 devices      Available devices (block and character)
 dma          Used DMS channels
 filesystems  Supported filesystems
 driver       Various drivers grouped here, currently rtc        (2.4)
 execdomains  Execdomains, related to security                        (2.4)
 fb               Frame Buffer devices                                (2.4)
 fs               File system parameters, currently nfs/exports        (2.4)
 ide          Directory containing info about the IDE subsystem
 interrupts   Interrupt usage
 iomem               Memory map                                        (2.4)
 ioports      I/O port usage
 irq               Masks for irq to cpu affinity                        (2.4)(smp?)
 isapnp       ISA PnP (Plug&Play) Info                                (2.4)
 kcore        Kernel core image (can be ELF or A.OUT(deprecated in 2.4))
 kmsg         Kernel messages
 ksyms        Kernel symbol table
 loadavg      Load average of last 1, 5 & 15 minutes;
                number of processes currently runnable (running or on ready queue);
                total number of processes in system;
                last pid created.
                All fields are separated by one space except "number of
                processes currently runnable" and "total number of processes
                in system", which are separated by a slash ('/'). Example:
                0.61 0.61 0.55 3/828 22084
 locks        Kernel locks
 meminfo      Memory info
 misc         Miscellaneous
 modules      List of loaded modules
 mounts       Mounted filesystems
 net          Networking info (see text)
 pagetypeinfo Additional page allocator information (see text)  (2.5)
 partitions   Table of partitions known to the system
 pci               Deprecated info of PCI bus (new way -> /proc/bus/pci/,
              decoupled by lspci                                (2.4)
 rtc          Real time clock
 scsi         SCSI info (see text)
 slabinfo     Slab pool info
 softirqs     softirq usage
 stat         Overall statistics
 swaps        Swap space utilization
 sys          See chapter 2
 sysvipc      Info of SysVIPC Resources (msg, sem, shm)                (2.4)
 tty               Info of tty drivers
 uptime       Wall clock since boot, combined idle time of all cpus
 version      Kernel version
 video               bttv info of video resources                        (2.4)
 vmallocinfo  Show vmalloced areas
 ============ ===============================================================

You can,  for  example,  check  which interrupts are currently in use and what
they are used for by looking in the file /proc/interrupts::

  > cat /proc/interrupts
             CPU0
    0:    8728810          XT-PIC  timer
    1:        895          XT-PIC  keyboard
    2:          0          XT-PIC  cascade
    3:     531695          XT-PIC  aha152x
    4:    2014133          XT-PIC  serial
    5:      44401          XT-PIC  pcnet_cs
    8:          2          XT-PIC  rtc
   11:          8          XT-PIC  i82365
   12:     182918          XT-PIC  PS/2 Mouse
   13:          1          XT-PIC  fpu
   14:    1232265          XT-PIC  ide0
   15:          7          XT-PIC  ide1
  NMI:          0

In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the
output of a SMP machine)::

  > cat /proc/interrupts

             CPU0       CPU1
    0:    1243498    1214548    IO-APIC-edge  timer
    1:       8949       8958    IO-APIC-edge  keyboard
    2:          0          0          XT-PIC  cascade
    5:      11286      10161    IO-APIC-edge  soundblaster
    8:          1          0    IO-APIC-edge  rtc
    9:      27422      27407    IO-APIC-edge  3c503
   12:     113645     113873    IO-APIC-edge  PS/2 Mouse
   13:          0          0          XT-PIC  fpu
   14:      22491      24012    IO-APIC-edge  ide0
   15:       2183       2415    IO-APIC-edge  ide1
   17:      30564      30414   IO-APIC-level  eth0
   18:        177        164   IO-APIC-level  bttv
  NMI:    2457961    2457959
  LOC:    2457882    2457881
  ERR:       2155

NMI is incremented in this case because every timer interrupt generates a NMI
(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups.

LOC is the local interrupt counter of the internal APIC of every CPU.

ERR is incremented in the case of errors in the IO-APIC bus (the bus that
connects the CPUs in a SMP system. This means that an error has been detected,
the IO-APIC automatically retry the transmission, so it should not be a big
problem, but you should read the SMP-FAQ.

In 2.6.2* /proc/interrupts was expanded again.  This time the goal was for

IRQ affinity와 proc subdirectory

특수 IRQ vector는 관련 platform에서만 표시된다. 일부 threshold vector는 x86_64에 없고, 일부는 uniprocessor에서 숨겨진다. 문서 작성 당시 새 vector 표시는 i386과 x86_64가 지원했다.

`/proc/irq`에는 IRQ별 subdirectory와 `default_smp_affinity`, `prof_cpu_mask`가 있다. 각 IRQ의 `smp_affinity` bitmask는 처리 가능한 CPU를 지정한다. 예를 들어 1은 첫 CPU만, 5는 첫째와 셋째 CPU를 허용한다. 기본은 보통 `ffffffff`다. `smp_affinity_list`는 bitmask 대신 `1024-1031` 같은 CPU range를 받는다.

`default_smp_affinity`는 아직 allocate·activate되지 않아 `/proc/irq/N` directory가 없는 non-active IRQ에 적용된다. SMP의 `node` file은 device가 보고한 hardware NUMA node를 보여 주며 driver의 locality 선호는 포함하지 않는다. `prof_cpu_mask`는 system-wide profiler가 profile할 CPU를 정한다.

IO-APIC는 허용된 CPU 사이에서 보통 round-robin으로 IRQ를 route한다. 특별한 이유가 없다면 kernel default가 권장된다. `net`, `scsi`, `sys` directory의 존재와 내용도 networking·SCSI 등 kernel configuration에 따라 달라진다.

/proc/interrupts to display every IRQ vector in use by the system, not
just those considered 'most important'.  The new vectors are:

THR
  interrupt raised when a machine check threshold counter
  (typically counting ECC corrected errors of memory or cache) exceeds
  a configurable threshold.  Only available on some systems.

TRM
  a thermal event interrupt occurs when a temperature threshold
  has been exceeded for the CPU.  This interrupt may also be generated
  when the temperature drops back to normal.

SPU
  a spurious interrupt is some interrupt that was raised then lowered
  by some IO device before it could be fully processed by the APIC.  Hence
  the APIC sees the interrupt but does not know what device it came from.
  For this case the APIC will generate the interrupt with a IRQ vector
  of 0xff. This might also be generated by chipset bugs.

RES, CAL, TLB
  rescheduling, call and TLB flush interrupts are
  sent from one CPU to another per the needs of the OS.  Typically,
  their statistics are used by kernel developers and interested users to
  determine the occurrence of interrupts of the given type.

The above IRQ vectors are displayed only when relevant.  For example,
the threshold vector does not exist on x86_64 platforms.  Others are
suppressed when the system is a uniprocessor.  As of this writing, only
i386 and x86_64 platforms support the new IRQ vector displays.

Of some interest is the introduction of the /proc/irq directory to 2.4.
It could be used to set IRQ to CPU affinity. This means that you can "hook" an
IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
prof_cpu_mask.

For example::

  > ls /proc/irq/
  0  10  12  14  16  18  2  4  6  8  prof_cpu_mask
  1  11  13  15  17  19  3  5  7  9  default_smp_affinity
  > ls /proc/irq/0/
  smp_affinity

smp_affinity is a bitmask, in which you can specify which CPUs can handle the
IRQ. You can set it by doing::

  > echo 1 > /proc/irq/10/smp_affinity

This means that only the first CPU will handle the IRQ, but you can also echo
5 which means that only the first and third CPU can handle the IRQ.

The contents of each smp_affinity file is the same by default::

  > cat /proc/irq/0/smp_affinity
  ffffffff

There is an alternate interface, smp_affinity_list which allows specifying
a CPU range instead of a bitmask::

  > cat /proc/irq/0/smp_affinity_list
  1024-1031

The default_smp_affinity mask applies to all non-active IRQs, which are the
IRQs which have not yet been allocated/activated, and hence which lack a
/proc/irq/[0-9]* directory.

The node file on an SMP system shows the node to which the device using the IRQ
reports itself as being attached. This hardware locality information does not
include information about any possible driver locality preference.

prof_cpu_mask specifies which CPUs are to be profiled by the system wide
profiler. Default value is ffffffff (all CPUs if there are only 32 of them).

The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
between all the CPUs which are allowed to handle it. As usual the kernel has
more info than you and does a better job than you, so the defaults are the
best choice for almost everyone.  [Note this applies only to those IO-APIC's
that support "Round Robin" interrupt distribution.]

There are  three  more  important subdirectories in /proc: net, scsi, and sys.
The general  rule  is  that  the  contents,  or  even  the  existence of these
directories, depend  on your kernel configuration. If SCSI is not enabled, the
directory scsi  may  not  exist. The same is true with the net, which is there
only when networking support is present in the running kernel.

Slab, buddy allocator, page type

`slabinfo`는 page보다 큰 수준에서 흔한 kernel object를 관리하는 slab pool의 memory 사용량을 제공한다. network buffer나 directory cache 같은 object가 전용 pool을 갖는다.

`buddyinfo`는 node와 zone별로 order마다 사용할 수 있는 free block 수를 보여 줘 external fragmentation과 high-order allocation 실패를 진단한다. order N의 한 chunk는 `2^N * PAGE_SIZE`다.

`pagetypeinfo`는 page block order와 block당 page 수를 먼저 표시한 뒤 buddyinfo와 같은 free count를 `Unmovable`, `Reclaimable`, `Movable`, `Reserve`, `Isolate` migrate type별로 나누고, 마지막에 type별 page block 수를 보여 준다.

fragmentation 회피는 이동 가능성이 비슷한 page를 hugepage 크기 정도의 contiguous page block에 묶는다. 그러면 kernel이 한 block 안의 movable·reclaimable page를 회수해 high-order allocation을 만족시킬 수 있다.

`min_free_kbytes`가 적절하면 현재 할당 가능한 huge page 수를 추정할 수 있다. `Movable` block은 `mlock()`되지 않았다면 대체로 할당 가능하고, `Reclaimable` 일부도 filesystem metadata 회수 비용을 치르면 가능하다.

The slabinfo  file  gives  information  about  memory usage at the slab level.
Linux uses  slab  pools for memory management above page level in version 2.2.
Commonly used  objects  have  their  own  slab  pool (such as network buffers,
directory cache, and so on).

::

    > cat /proc/buddyinfo

    Node 0, zone      DMA      0      4      5      4      4      3 ...
    Node 0, zone   Normal      1      0      0      1    101      8 ...
    Node 0, zone  HighMem      2      0      0      1      1      0 ...

External fragmentation is a problem under some workloads, and buddyinfo is a
useful tool for helping diagnose these problems.  Buddyinfo will give you a
clue as to how big an area you can safely allocate, or why a previous
allocation failed.

Each column represents the number of pages of a certain order which are
available.  In this case, there are 0 chunks of 2^0*PAGE_SIZE available in
ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE
available in ZONE_NORMAL, etc...

More information relevant to external fragmentation can be found in
pagetypeinfo::

    > cat /proc/pagetypeinfo
    Page block order: 9
    Pages per block:  512

    Free pages count per migrate type at order       0      1      2      3      4      5      6      7      8      9     10
    Node    0, zone      DMA, type    Unmovable      0      0      0      1      1      1      1      1      1      1      0
    Node    0, zone      DMA, type  Reclaimable      0      0      0      0      0      0      0      0      0      0      0
    Node    0, zone      DMA, type      Movable      1      1      2      1      2      1      1      0      1      0      2
    Node    0, zone      DMA, type      Reserve      0      0      0      0      0      0      0      0      0      1      0
    Node    0, zone      DMA, type      Isolate      0      0      0      0      0      0      0      0      0      0      0
    Node    0, zone    DMA32, type    Unmovable    103     54     77      1      1      1     11      8      7      1      9
    Node    0, zone    DMA32, type  Reclaimable      0      0      2      1      0      0      0      0      1      0      0
    Node    0, zone    DMA32, type      Movable    169    152    113     91     77     54     39     13      6      1    452
    Node    0, zone    DMA32, type      Reserve      1      2      2      2      2      0      1      1      1      1      0
    Node    0, zone    DMA32, type      Isolate      0      0      0      0      0      0      0      0      0      0      0

    Number of blocks type     Unmovable  Reclaimable      Movable      Reserve      Isolate
    Node 0, zone      DMA            2            0            5            1            0
    Node 0, zone    DMA32           41            6          967            2            0

Fragmentation avoidance in the kernel works by grouping pages of different
migrate types into the same contiguous regions of memory called page blocks.
A page block is typically the size of the default hugepage size, e.g. 2MB on
X86-64. By keeping pages grouped based on their ability to move, the kernel
can reclaim pages within a page block to satisfy a high-order allocation.

The pagetypinfo begins with information on the size of a page block. It
then gives the same type of information as buddyinfo except broken down
by migrate-type and finishes with details on how many page blocks of each
type exist.

If min_free_kbytes has been tuned correctly (recommendations made by hugeadm
from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can
make an estimate of the likely number of huge pages that can be allocated
at a given point in time. All the "Movable" blocks should be allocatable
unless memory has been mlock()'d. Some of the Reclaimable blocks should
also be allocatable although a lot of filesystem metadata may have to be
reclaimed to achieve this.

`/proc/allocinfo`

`allocinfo`는 code base의 모든 allocation 위치를 source file, line number, loadable module, allocation caller function으로 식별하고 위치별 allocated byte와 call count를 보고한다. 첫 줄은 file version, 둘째 줄은 field header다.

version 2.0 이상에서는 call site 추가 정보를 `<key>:<value>` pair로 붙일 수 있다. 현재 지원 marker `accurate:no`는 tracking용 memory 할당 실패 때문에 absolute counter가 정확하지 않음을 뜻한다. 하지만 delta는 정확하므로 allocation size와 count 변화 추적에는 사용할 수 있다.

예시는 `/proc/allocinfo`의 header 뒤를 byte count 기준 역순 정렬해 `alloc_page_ext`, `alloc_slab_page`, readahead, hash, block request, folio, thread stack 등 큰 allocation site를 찾는다.

allocinfo
~~~~~~~~~

Provides information about memory allocations at all locations in the code
base. Each allocation in the code is identified by its source file, line
number, module (if originates from a loadable module) and the function calling
the allocation. The number of bytes allocated and number of calls at each
location are reported. The first line indicates the version of the file, the
second line is the header listing fields in the file.
If file version is 2.0 or higher then each line may contain additional
<key>:<value> pairs representing extra information about the call site.
For example if the counters are not accurate, the line will be appended with
"accurate:no" pair.

Supported markers in v2:
accurate:no

              Absolute values of the counters in this line are not accurate
              because of the failure to allocate memory to track some of the
              allocations made at this location.  Deltas in these counters are
              accurate, therefore counters can be used to track allocation size
              and count changes.

Example output.

::

    > tail -n +3 /proc/allocinfo | sort -rn
   127664128    31168 mm/page_ext.c:270 func:alloc_page_ext
    56373248     4737 mm/slub.c:2259 func:alloc_slab_page
    14880768     3633 mm/readahead.c:247 func:page_cache_ra_unbounded
    14417920     3520 mm/mm_init.c:2530 func:alloc_large_system_hash
    13377536      234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs
    11718656     2861 mm/filemap.c:1919 func:__filemap_get_folio
     9192960     2800 kernel/fork.c:307 func:alloc_thread_stack_node
     4206592        4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable
     4136960     1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start
     3940352      962 mm/memory.c:4214 func:alloc_anon_folio
     2894464    22613 fs/kernfs/dir.c:615 func:__kernfs_new_node
     ...

`meminfo` 기본 memory 분포

`/proc/meminfo`는 memory 분포와 사용량을 제공하지만 architecture와 build option에 따라 field가 다르고 일부 counter는 서로 겹친다. 겹치지 않는 counter 합도 전체 사용량과 일치하지 않을 수 있으며 workload에 따라 차이가 크다. TCP allocation처럼 subsystem 전용 정보는 `/proc/net/sockstat` 등에서 확인한다.

`MemAvailable`은 `MemFree`, `SReclaimable`, file LRU 크기와 zone low watermark를 바탕으로 계산한다. system 작동에 필요한 page cache와 사용 중이라 실제로는 회수할 수 없는 slab 부분을 고려하므로 system마다 영향이 다르다.

meminfo
~~~~~~~

Provides information about distribution and utilization of memory.  This
varies by architecture and compile options.  Some of the counters reported
here overlap.  The memory reported by the non overlapping counters may not
add up to the overall memory usage and the difference for some workloads
can be substantial.  In many cases there are other means to find out
additional memory using subsystem specific interfaces, for instance
/proc/net/sockstat for TCP memory allocations.

Example output. You may not have all of these fields.

::

    > cat /proc/meminfo

    MemTotal:       32858820 kB
    MemFree:        21001236 kB
    MemAvailable:   27214312 kB
    Buffers:          581092 kB
    Cached:          5587612 kB
    SwapCached:            0 kB
    Active:          3237152 kB
    Inactive:        7586256 kB
    Active(anon):      94064 kB
    Inactive(anon):  4570616 kB
    Active(file):    3143088 kB
    Inactive(file):  3015640 kB
    Unevictable:           0 kB
    Mlocked:               0 kB
    SwapTotal:             0 kB
    SwapFree:              0 kB
    Zswap:              1904 kB
    Zswapped:           7792 kB
    Dirty:                12 kB
    Writeback:             0 kB
    AnonPages:       4654780 kB
    Mapped:           266244 kB
    Shmem:              9976 kB
    KReclaimable:     517708 kB
    Slab:             660044 kB
    SReclaimable:     517708 kB
    SUnreclaim:       142336 kB
    KernelStack:       11168 kB
    PageTables:        20540 kB
    SecPageTables:         0 kB
    NFS_Unstable:          0 kB
    Bounce:                0 kB
    WritebackTmp:          0 kB
    CommitLimit:    16429408 kB
    Committed_AS:    7715148 kB
    VmallocTotal:   34359738367 kB
    VmallocUsed:       40444 kB
    VmallocChunk:          0 kB
    Percpu:            29312 kB
    EarlyMemtestBad:       0 kB
    HardwareCorrupted:     0 kB
    AnonHugePages:   4149248 kB
    ShmemHugePages:        0 kB
    ShmemPmdMapped:        0 kB
    FileHugePages:         0 kB
    FilePmdMapped:         0 kB
    CmaTotal:              0 kB
    CmaFree:               0 kB
    Unaccepted:            0 kB
    Balloon:               0 kB
    HugePages_Total:       0
    HugePages_Free:        0
    HugePages_Rsvd:        0
    HugePages_Surp:        0
    Hugepagesize:       2048 kB
    Hugetlb:               0 kB
    DirectMap4k:      401152 kB
    DirectMap2M:    10008576 kB
    DirectMap1G:    24117248 kB

MemTotal
              Total usable RAM (i.e. physical RAM minus a few reserved
              bits and the kernel binary code)
MemFree
              Total free RAM. On highmem systems, the sum of LowFree+HighFree
MemAvailable
              An estimate of how much memory is available for starting new
              applications, without swapping. Calculated from MemFree,
              SReclaimable, the size of the file LRU lists, and the low

`meminfo` 상세 field

`AnonPages`는 userspace page table에 mapping된 non-file-backed page, `Mapped`는 library 같은 mmap file, `Shmem`은 shmem·tmpfs 전체다. 큰 allocation을 정밀 추적하지 않는 구성에서는 한 page만 mapping돼도 allocation 전체를 mapped로 볼 수 있다.