proc - process information pseudo-filesystem
/proc is a pseudo-filesystem which is used as an interface to kernel data structures rather than reading and interpreting /dev/kmem. Most of it is read-only, but some
files allow kernel variables to be changed.
The following outline gives a quick tour through the /proc hierarchy.
- There is a numerical subdirectory for each running process; the subdirectory is named by the process ID. Each contains the following pseudo-files and directories.
- This holds the complete command line for the process, unless the whole process has been swapped out, or unless the process is a zombie. In either of these later cases, there is nothing in this file: i.e. a read on this file will return as having read 0 characters. This file is
null-terminated, but not newline-terminated.
- This is a link to the current working directory of the process. To find out the cwd of process 20, for instance, you can do this:
cd /proc/20/cwd; /bin/pwd
- Note that the pwd command is often a shell builtin, and might not work properly in this context.
- This file contains the environment for the process. The entries are separated by null characters, and there may be a null character at the end. Thus, to print out the environment of process 1, you would do:
(cat /proc/1/environ; echo) | tr
- (For a reason why one should want to do this, see lilo(8).)
- a pointer to the binary which was executed, and appears as a symbolic link. A readlink(2) call on the exe special file returns under Linux 2.0 and earlier a string in the format:
- For example, [0301?:1502 would be inode 1502 on device major 03 (IDE, MFM, etc. drives) minor 01 (first partition on the first drive). Under Linux 2.2 the link contains the actual path name of the command.
- Also, the symbolic link can be dereferenced normally - attempting to open /proc/[number?/exe'' to run another copy of the same process as [number?.
- find(1) with the -inum option can be used to locate the file.
- This is a subdirectory containing one entry for each file which the process has open, named by its file descriptor, and which is a symbolic link to the actual file (as the exe entry does). Thus, 0 is standard input, 1 standard output, 2 standard error, etc.
- Programs that will take a filename, but will not take the standard input, and which write to a file, but will not send their output to standard output, can be effectively foiled this way, assuming that -i is the flag designating an input file and -o is the flag designating an output file:
foobar -i /proc/self/fd/0 -o /proc/self/fd/1 ...
- and you have a working filter. Note that this will not work for programs that seek on their files, as the files in the fd directory are not seekable.
- /proc/self/fd/N is approximately the same as /dev/fd/N in some UNIX and UNIX-like systems. Most Linux MAKEDEV scripts symbolically link /dev/fd to [..?/proc/self/fd, in fact.
- A file containing the currently mapped memory regions and their access permissions.
- The format is:
address perms offset dev inode
00000000-0002f000 r-x-- 00000400 03:03 1401
0002f000-00032000 rwx-p 0002f400 03:03 1401
00032000-0005b000 rwx-p 00000000 00:00 0
60000000-60098000 rwx-p 00000400 03:03 215
60098000-600c7000 rwx-p 00000000 00:00 0
bfffa000-c0000000 rwx-p 00000000 00:00 0
- 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 file/whatever, 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.
- Under Linux 2.2 there is an additional field giving a pathname where applicable.
- This is not the same as the mem (1:1) device, despite the fact that it has the same device numbers. The /dev/mem device is the physical memory before any address translation is done, but the mem file here is the memory of the process that accesses it. This cannot be mmap(2)'ed currently, and will not be until a general mmap(2) is added to the kernel. (This might have happened by the time you read this.)
- Directory of maps by mmap(2) which are symbolic links like exe, fd/, etc. Note that maps includes a superset of this information, so /proc//mmap should be considered obsolete.
- /proc/*/mmap was removed in Linux kernel version 1.1.40. (It really was obsolete!)
- Unix and linux support the idea of a per-process root of the filesystem, set by the chroot(2) system call. Root points to the file system root, and behaves as exe, fd/*, etc. do.
- Status information about the process. This is used by ps(1).
- The fields, in order, with their proper scanf(3) format specifiers, are:
|pid|%d|The process id.
|comm|%s| The filename of the executable, in parentheses. This is visible whether or not the executable is swapped out.
|state|%c| One character from the string
|ppid|%d| The PID of the parent.
|pgrp|%d| The process group ID of the process.
|session|%d| The session ID of the process.
|tty|%d| The tty the process uses.
|tpgid|%d| The process group ID of the process which currently owns the tty that the process is connected to.
|flags|%u|The flags of the process. Currently, every flag has the math bit set, because crt0.s checks for math emulation, so this is not included in the output. This is probably a bug, as not every process is a compiled C program. The math bit should be a decimal 4, and the traced bit is decimal 10.
|minflt|%u| The number of minor faults the process has made, those which have not required loading a memory page from disk.
|cminflt|%u| The number of minor faults that the process and its children have made.
|majflt|%u| The number of major faults the process has made, those which have required loading a memory page from disk.
|cmajflt|%u| The number of major faults that the process and its children have made.
|utime|%d| The number of jiffies that this process has been scheduled in user mode.
|stime|%d| The number of jiffies that this process has been scheduled in kernel mode.
|cutime|%d| The number of jiffies that this process and its children have been scheduled in user mode.
|cstime|%d| The number of jiffies that this process and its children have been scheduled in kernel mode.
|counter|%d| The current maximum size in jiffies of the process's next timeslice, or what is currently left of its current timeslice, if it is the currently running
|priority|%d| The standard nice value, plus fifteen. The value is never negative in the kernel.
|timeout|%u| The time in jiffies of the process's next timeout.
|itrealvalue|%u|The time (in jiffies) before the next SIGALRM is sent to the process due to an interval timer.
|starttime|%d| Time the process started in jiffies after system boot.
|vsize|%u| Virtual memory size
|rss|%u| Resident Set Size: number of pages the process has in real memory, minus 3 for administrative purposes. This is just the pages which count towards text, data, or stack space. This does not include pages which have not been demand-loaded in, or which are swapped out.
|rlim|%u Current limit in bytes on the rss of the process (usually 2,147,483,647).
|startcode|%u| The address above which program text can run.
|endcode|%u| The address below which program text can run.
|startstack|%u| The address of the start of the stack.
|kstkesp|%u|The current value of esp (32-bit stack pointer), as found in the kernel stack page for the process.
|kstkeip|%u| The current EIP (32-bit instruction pointer).
|signal|%d| The bitmap of pending signals (usually 0).
|blocked|%d| The bitmap of blocked signals (usually 0, 2 for shells).
|sigignore|%d| The bitmap of ignored signals.
|sigcatch|%d| The bitmap of catched signals.
|wchan|%u| This is the "channel" in which the process is waiting. It is the address of a system call, and can be looked up in a namelist if you need a textual name. (If you have an up to date /etc/psdatabase, then try ps -l to see the WCHAN field in action)
- This is a collection of CPU and system architecture dependent items, for each supported architecture a different list. The only two common entries are
cpu which is (guess what) the CPU currently in use and BogoMIPS a system constant which is calculated during kernel initialization.
- Text listing of major numbers and device groups. This can be used by MAKEDEV scripts for consistency with the kernel.
- This is a list of the registered ISA DMA (direct memory access) channels in use.
- A text listing of the filesystems which were compiled into the kernel. Incidentally, this is used by mount(1)? to cycle through different filesystems when none is specified.
- This is used to record the number of interrupts per each IRQ on (at least) the i386 architechure. Very easy to read formatting, done in ASCII.
- This is a list of currently registered Input-Output port regions that are in use.
- This file represents the physical memory of the system and is stored in the core file format. With this pseudo-file, and an unstripped kernel (/usr/src/linux/tools/zSystem) binary, GDB can be used to examine the current state of any kernel data structures.
- The total length of the file is the size of physical memory (RAM) plus 4KB.
- This file can be used instead of the syslog(2) system call to log kernel messages. A process must have superuser privileges to read this file, and only one process should read this file. This file should not be read if a syslog process is running which uses the syslog(2) system call facility to log kernel messages.
- Information in this file is retrieved with the dmesg(8) program).
- This holds the kernel exported symbol definitions used by the modules(X) tools to dynamically link and bind loadable modules.
- The load average numbers give the number of jobs in the run queue (state R) or waiting for disk I/O (state D) averaged over 1, 5 and 15 minutes. They are the same as the load average numbers given by uptime(1) and other programs.
- This file shows current file locks.
- This file is only present if CONFIGDEBUGMALLOC was defined during compilation.
- This is used by free(1) to report the amount of free and used memory (both physical and swap) on the system as well as the shared memory and buffers used by the kernel.
- It is in the same format as free(1), except in bytes rather than KB.
- A text list of the modules that have been loaded by the system.
- various net pseudo-files, all of which give the status of some part of the networking layer. These files contain ASCII structures, and are therefore readable with cat. However, the standard netstat(8) suite provides much cleaner access to these files.
- This holds an ASCII readable dump of the kernel ARP table used for address resolutions. It will show both dynamically learned and pre-programmed ARP entries. The format is:
IP address HW type Flags HW address
10.11.100.129 0x1 0x6 00:20:8A:00:0C:5A
10.11.100.5 0x1 0x2 00:C0:EA:00:00:4E
22.214.171.124 0x3 0x2 GW4PTS
- Where 'IP address' is the IPv4 address of the machine, the 'HW type' is the hardware type of the address from RFC 826. The flags are the internal flags of the ARP structure (as defined in /usr/include/linux/if_arp.h) and the 'HW address' is the physical layer mapping for that IP address if it is known.
- The dev pseudo-file contains network device status information. This gives the number of received and sent packets, the number of errors and collisions and other basic statistics. These are used by the ifconfig(8) program to report device status. The format is:
Inter-| Receive | Transmit
face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
lo: 2776770 11307 0 0 0 0 0 0 2776770 11307 0 0 0 0 0 0
eth0: 1215645 2751 0 0 0 0 0 0 1782404 4324 0 0 0 427 0 0
ppp0: 1622270 5552 1 0 0 0 0 0 354130 5669 0 0 0 0 0 0
tap0: 7714 81 0 0 0 0 0 0 7714 81 0 0 0 0 0 0
- No information.
- No information.
- This file uses the same format as the arp file and contains the current reverse mapping database used to provide rarp(8) reverse address lookup services. If RARP is not configured into the kernel this file will not be present.
- Holds a dump of the RAW socket table. Much of the information is not of use apart from debugging. The 'sl' value is the kernel hash slot for the socket, the 'local address' is the local address and protocol number pair.
- No information, but looks similar to route(8)
- This file holds the ASCII data needed for the IP, ICMP, TCP and UDP management information bases for an snmp agent. As of writing the TCP mib is incomplete. It is hoped to have it completed by 1.2.0.
- Holds a dump of the TCP socket table. Much of the information is not of use apart from debugging. The "sl" value is the kernel hash slot for the socket, the "local address" is the local address and port number pair. The "remote address" is the remote address and port number pair (if connected). 'St' is the internal status of the socket. The 'tx_queue' and 'rx_queue' are the outgoing and incoming data queue in terms of kernel memory usage. The "tr", "tm->when", and "rexmits" fields hold internal information of the kernel socket state and are only useful for debugging. The uid field holds the creator euid of the socket.
- Holds a dump of the UDP socket table. Much of the information is not of use apart from debugging. The "sl" value is the kernel hash slot for the socket, the "local address" is the local address and port number pair. The "remote address" is the remote address and port number pair (if connected). "St" is the internal status of the socket. The "tx_queue" and "rx_queue" are the outgoing and incoming data queue in terms of kernel memory usage. The "tr", "tm->when", and "rexmits" fields are not used by UDP. The uid field holds the creator euid of the socket. The format is:
sl local_address rem_address st tx_queue rx_queue tr rexmits tm->when uid
1: 01642C89:0201 0C642C89:03FF 01 00000000:00000001 01:000071BA 00000000 0
1: 00000000:0801 00000000:0000 0A 00000000:00000000 00:00000000 6F000100 0
1: 00000000:0201 00000000:0000 0A 00000000:00000000 00:00000000 00000000 0
- Lists the UNIX domain sockets present within the system and their status. The format is:
Num !RefCount? Protocol Flags Type St Path
0: 00000002 00000000 00000000 0001 03
1: 00000001 00000000 00010000 0001 01 /dev/printer
- Where 'Num' is the kernel table slot number, '!RefCount?' is the number of users of the socket, 'Protocol' is currently always 0, 'Flags' represent the internal kernel flags holding the status of the socket. Type is always '1' currently (Unix domain datagram sockets are not yet supported in the kernel). 'St' is the internal state of the
socket and Path is the bound path (if any) of the socket.
- This is a listing of all PCI devices found during kernel initialization and their configuration.
- A directory with the scsi midlevel pseudo-file and various SCSI lowlevel driver directories, which contain a file for each SCSI host in this system, all of which give the status of some part of the SCSI IO subsystem. These files contain ASCII structures, and are therefore readable with cat.
- You can also write to some of the files to reconfigure the subsystem or switch certain features on or off.
- This is a listing of all SCSI devices known to the kernel. The listing is similar to the one seen during bootup. scsi currently supports only the add-single-device command which allows root to add a hotplugged device to the list of known devices.
echo 'scsi add-single-device 1 0 5 0' >/proc/scsi/scsi
- will cause host scsi1 to scan on SCSI channel 0 for a device on ID 5 LUN 0. If there is already a device known on this address or the address is invalid an error will be returned.
- drivername can currently be: NCR53c7xx, aha152x, aha1542, aha1740, aic7xxx, buslogic, eata_dma, eata_pio, fdomain, in2000, pas16, qlogic, scsi_debug, seagate, t128, u15-24f, ultrastore or wd7000. These directories show up for all drivers which registered at least one SCSI HBA. Every directory contains one file per registered host. Every host-file is named after the number the host got assigned during initilization.
- Reading these files will usually show driver and host configuration, statistics etc.
Writing to these files allows different things on different hosts. For example with the latency and nolatency commands root can switch on and off command
latency measurement code in the eata_dma driver. With the lockup and unlock commands root can control bus lockups simulated by the scsi_debug driver.
- This directory refers to the process accessing the /proc filesystem, and is identical to the /proc directory named by the process ID of the same process.
- kernel/system statistics
cpu 3357 0 4313 1362393
- The number of jiffies (1/100ths of a second) that the system spent in user mode, user mode with low priority (nice), system mode, and the idle task, respectively. The last value should be 100 times the second entry in the uptime pseudo-file.
disk 0 0 0 0
- The four disk entries are not implemented at this time. I'm not even sure what this should be, since kernel statistics on other machines usually track both transfer rate and I/Os per second and this only allows for one field per drive.
page 5741 1808
- The number of pages the system paged in and the number that were paged out (from disk).
swap 1 0
- The number of swap pages that have been brought in and out.
- The number of interrupts received from the system boot.
- The number of context switches that the system underwent.
- boot time, in seconds since the epoch (January 1, 1970).
- This directory (present since 1.3.57) contains a number of files and subdirectories corresponding to kernel variables. These variables can be read and sometimes modified using the proc file system, and using the sysctl(2) system call. Presently, there are subdirectories kernel, net, vm that each contain more
files and subdirectories.
- This contains files domainname, file-max, file-nr, hostname, inode-max, inode-nr, osrelease, ostype,
panic, real-root-dev, securelevel, version, with function fairly clear from the name.
- The (read-only) file file-nr gives the number of files presently opened.
- The file file-max gives the maximum number of open files the kernel is willing to handle. If 1024 is not enough for you, try
echo 4096 >file-max
- Similarly, the files inode-nr and inode-max indicate the present and the maximum number of inodes.
- The files ostype, osrelease, version give substrings of /proc/version.
- The file panic gives r/w access to the kernel variable panic_timeout. If this is zero, the kernel will loop on a panic; if nonzero it indicates that the
kernel should autoreboot after this number of seconds.
- The file securelevel seems rather meaningless at present - root is just too powerful.
- This file contains two numbers: the uptime of the system (seconds), and the amount of time spent in idle process (seconds).
- This strings identifies the kernel version that is currently running. For instance:
Linux version 1.0.9 (quinlan@phaze) #1 Sat May 14 01:51:54 EDT 1994
cat(1), find(1), free(1), mount(1)?, ps(1), tr(1), uptime(1), chroot(2), mmap(2), readlink(2), syslog(2), slabinfo(5), hier(7), arp(8), dmesg(8), netstat(8), ifconfig(8), procinfo(8), route(8), and much more
This roughly conforms to a Linux 1.3.11 kernel. Please update this as necessary!
Last updated for Linux 1.3.11.
Note that many strings (i.e., the environment and command line) are in the internal format, with sub-fields terminated by NUL bytes, so you may find that things are more readable if you use od -c or tr "\000" "\n" to read them.
This manual page is incomplete, possibly inaccurate, and is the kind of thing that needs to be updated very often.
The /proc file system may introduce security holes into processes running with chroot(2). For example, if /proc is mounted in the chroot hierarchy, a
chdir(2) to /proc/1/root will return to the original root of the file system. This may be considered a feature instead of a bug, since Linux does not yet support
the fchroot(2)? call.