Overview of Linux syscalls

A - B - C - D - E - F - G - H - I - J - K - L - M - N - O - P - Q - R - S - T - U - V - W - X - Y - Z

A

accept: Accept a connection on a socket
access: Checks whether the calling process can access the pathname, dereferenced when it is a symbolic link
add_key: Create or update a key for kernel key management facility
adjtimex: Reads and optionally sets adjustment parameters for clock adjustment algorithm used on Linux (RFC 5905)
alarm: Schedule an alarm; it lets the system generate a SIGALRM signal for the process after a specified time

B

bind: Assigns address to a socket that was created with socket()
brk: Change the location of program break, specifically the end of the process's data segment

C

cacheflush: Flushes contents of cache(s) for user addresses in specified range
capget: Retrieve thread capabilities
capset: Set thread capabilities
chdir: Change work directory
chmod: Change mode of the file, dereferenced for symbolic links
chown: Changes ownership of file specified by pathname, dereferenced if file is a symbolic link
chroot: The root directory (which is normally /) of the calling process will be changed to the one specified in the path to sandbox a process
clock_adjtime: Reads and optionally sets adjustment parameters for clock adjustment algorithm used on Linux (RFC 5905). It behaves like adjtimex(), but takes an additional clk_id argument to define the clock
clock_getres: Retrieve the resolution (precision) of a specified clock
clock_gettime: Retrieve time from specified clock
clock_settime: Set time of specified clock
clone: Similar to fork() to create a child process, with more fine-grained options to define what is shared between calling process and child. This system call can also make a new process part of newly created namespace by specifying a flag.
clone2: Specific for ia64 platform, similar to clone() but with different handling of stack assignment
clone3: Provides superset of the functionality of the older clone() interface to create child process
close: Close file descriptor
close_range: Close the file descriptors of the selected range
connect: Initiate connection on a defined socket
creat: Like open(), but sets flags O_CREAT|O_WRONLY|O_TRUNC
create_module: Deprecated syscall to create an entry to load a kernel module and reserve the kernel memory for it

D

delete_module: Tries to remove an unused loadable module entry which is related currently loaded Linux kernel module (LKM)
dup: Duplicate file descriptor; more specifically it allocates a new file descriptor that also refers to open file description oldfd
dup2: Same as dup(), duplicate file descriptor; difference is that it uses file descriptor number specified in newfd
dup3: Same as dup2(); difference is that caller can force close-on-exec flag (O_CLOEXEC) to be set

E

epoll_ctl: Manage (add, modify, remove) entries in epoll instance, which is used to monitor if I/O is allowed on the defined set of file descriptors. Similar to poll(), with additional benefits.
epoll_wait: Waits for events on an epoll instance which is defined by a file descriptor (epfd)
execve: Executes the program referred to by specified pathname
exit: Terminates the calling process, parent process will receive a SIGCHLD signal

F

faccessat: Similar to access(), works slightly different when pathname is relative
faccessat2: Closely similar to faccessat() but implements flags argument to correct incorrect implementation in faccessat()
fallocate: Allows the caller to manipulate allocated disk space for a file
fchdir: Similar to chdir, but uses open file descriptor
fchmod: Same as chmod, but used file by open file descriptor fd
fchmodat: Similar to chmod(), works slightly different when pathname is relative
fchown: Changes ownership of file, referred to by open file descriptor (fd)
fchownat: Similar to fchown(), but deals differently with relative paths
fcntl: Performs an action on file defined by a file descriptor, such as setting flags
fgetxattr: Retrieves xattr (extended attributes) by using a file descriptor returned by open()
finit_module: Similar to init_module(); loads image (ELF) but refers to a file description
flistxattr: Like listxattr() it retrieves the list of extended attributes (xattr) but associated with an open file descriptor
flock: Apply or remove advisory lock on file
fork: Create a new child process by duplicating the calling process, with caller becoming the parent process
fstat: Similar to stat(), but uses file descriptor fd
fstatat: More generic interface to receive information that is similar to stat(), fstat(), and lstat()
fstatfs: Returns information about an open file
ftruncate: Truncate a file open for writing to specified number of bytes, which may fill it with null bytes (\0) or decrease its size and losing data
futex: Provides a method for waiting until certain condition becomes true

G

get_current_dir_name: Like getcwd(); will retrieve current working directory but creates an array to store the value
getcwd: Copies the absolute pathname of current working directory to a buffer
getdents: Retrieve entries from a directory
getegid: Returns effective group ID of the calling process
geteuid: Retrieve effective user ID of calling process
getgid: Returns real group ID of the calling process
getgroups: Returns supplementary group IDs of calling process
getpeername: Receive address of the peer connected to a socket
getpgid: Retrieve process group ID (PGID)
getpgrp: Retrieves process group ID (PGID) of the calling process
getpid: Returns process ID (PID) of calling process
getppid: Returns process ID (PID) of parent of the calling process
getrandom: Receive random bytes
getrlimit: Get resource limits
getsid: Receive session ID of a defined process
getsockname: Retrieve current address of defined socket
getsockopt: Get options for socket
gettid: Returns thread ID (TID) of caller. Same as process ID (PID) for single-threaded applications, otherwise different
gettimeofday: Get time or timezone
getuid: Retrieve real user ID of calling process
getwd: Retrieves current working directory in specified pointer to a related buffer; deprecated for security and portability reasons
getxattr: Retrieves xattr (extended attributes) from a given file defined by its path and name

I

init_module: Load image (ELF) into the kernel space including the required steps to initialize it, including triggering the init() function of the module
inotify_add_watch: Adds a new watch or change an existing watch for a file
inotify_init: Initializes new inotify instance and returns file descriptor that is associated with a new inotify event queue
inotify_init1: Like inotify_init() it initializes a new inotify instance and returns file descriptor that is associated with a new inotify event queue, with additional flags
inotify_rm_watch: Removes a watch with a watch descriptor (wd) from an inotify instance specified by its file descriptor (fd)
io_cancel: Attempts to cancel asynchronous I/O operation that was submitted by io_submit()
io_submit: Submit asynchronous I/O blocks for processing, can be cancelled with io_cancel()

K

kexec_file_load: Similar to kexec_load(), but uses file descriptor for kernel and initrd (initial ram disk)
kexec_load: Load new kernel for later execution
keyctl: Allow user-space programs to take actions on keys, such as updating, revocation, ownership

L

lchown: Like chown(), does not dereference symbolic links
lgetxattr: Retrieves xattr (extended attributes) for symbolic links itself, not the destination it links to
link: Create new link (hard link) to existing file
linkat: Similar to link(), but deals differently with relative paths
listen: Marks socket as a passive to allow it accepting incoming connections with accept()
listxattr: Retrieves the list of extended attributes (xattr) associated with the path
llistxattr: Like listxattr, it retrieves the list of extended attributes (xattr) associated with the path, but if it is a symbolic link it will give information about the link, not the destination target
lseek: Reposition file offset for read/write
lstat: Similar to stat(), but if pathname is symbolic link, return information about link and not the file that symbolic link points to

M

mkdir: Create directory
mkdirat: Similar to mkdir() but deals differently with relative paths
mknod: Create filesystem node (file, device special file, or named pipe) named pathname
mknodat: Similar to mknod, works slightly different when pathname is relative
mlock: Lock pages in a specified address range, so they are guaranteed to stay in memory instead of being swapped to disk
mlock2: Same as mlock() if flags is 0. With flag MLOCK_ONFAULT is locks the current resident pages, the mark the range so currently nonresident pages are locked later when they are used (page fault)
mlockall: Similar to mlock, but tries to lock all the memory pages of the calling process to prevent swapping
mmap: Create new mapping in the virtual address space of the calling process
mprotect: Sets protection on a defined region of memory
mseal: Provides memory protection by allow developers to mark specific memory regions immutable, which helps with exploit mitigation
munlock: Opposite of mlock() to release lock on memory area, so it can be swapped to disk if needed
munlockall: Unlocks all memory pages of calling process so it can be swapped to disk again by the kernel
munmap: Deletes the mappings for specified address range and marks range to generate invalid memory references

N

nanosleep: Suspends the execution of the thread calling the request
nice: Change process priority, with +19 (lowest priority) up to to -20 (high priority)
ntp_adjtime: Doing same task as adjtimex(), but is a more portable interface

O

open: Opens file specified by pathname to allow reading or writing data
open_by_handle_at: The open_by_handle_at opens a file handle returned by a name_to_handle_at() and returns an open file descriptor.
openat: Similar to open(), but uses dirfd argument and deals differently with path

P

pause: Changes the thread or process that is calling the request to sleep until the moment a signal is received
personality: Set or get the personality for a process, a kernel execution domain
pidfd_open: Obtain a file descriptor referring to a process
pipe: Create a pipe that allows unidirectional communication between processes
pipe2: Similar to pipe(), to create a channel between two processes. With flag O_DIRECT it will use packet-style communication instead of a stream
poll: Similar task to select(2), which is waiting for a set of file descriptors to become available for I/O.
ppol: Let an application wait until file descriptor is available or signal is caught
prctl: Perform operations on a process or thread, such as changing its capabilities, set name of the calling thread, set the secure computing mode (seccomp), and more.
pread64: Similar to pread() but with an offtype of type off64_t that allows changing file positions in files larger than two gigabytes
preadv: Reads data into multiple buffers as readv(), with option to set an offset where in the file the read operation is to be performed.
preadv2: Reads data into multiple buffers like preadv(), with additional flags
prlimit: Get and sets resource limit, combines getrlimit() and setrlimit()
ptrace: Process tracing; usually for breakpoint debugging and system call tracing
pwrite64: Similar to pwrite() but with an offtype of type off64_t that allows changing file positions in files larger than two gigabytes
pwritev: Write to file description like writev(), except that multiple buffers are written
pwritev2: Write to file description like pwritev() with multiple buffers, additionally has extra flags

R

read: Read from file descriptor
readdir: Read a directory
readlink: Places a copy of the symbolic link referenced by its path into a buffer
readlinkat: Like readlinkat() it places a copy of the symbolic link referenced by its path into a buffer, but can use relative path
readv: Read buffers from file
reboot: Reboots the system, or enables/disables reboot keystroke (default: Ctrl+Alt+Delete; changed using loadkeys(1))
recv: Like read(), but normally only used on a socket and has additional flags that can be set
recvfrom: Receives a message on a socket, close to recv(), but with additional flags related to receiving source
recvmsg: Receives a message on a socket with a predefined structure to minimize the number of arguments
rename: Rename a file, move it between directories if required
renameat: Similar to rename(), with deals differently with relative paths
renameat2: Similar to renameat() when no flags are provided, otherwise it has additional options
request_key: Request a key from kernel key management facility
rmdir: Delete directory

S

sbrk: Change the location of program break, specifically it increments the program's data space
sched_getaffinity: Retrieves the mask on which CPUs the process thread can run
sched_setaffinity: Defines by using a mask on which CPUs the process thread can run
sched_yield: Request by the calling thread to free up itself from the CPU and move it to the very end of the queue, so the next thread can run
seccomp: Perform operations on the secure computing (seccomp) state of a process
select: Let a program monitor multiple file descriptors until one or more become available for I/O actions. This system call has limitations and typically poll or epoll is used.
sendfile: Copies data between one file descriptor and another
sendto: Transmits a message to another socket
setdomainname: Sets the NIS domain name to the defined value
setfsgid: Sets the group identity used for performing file system checks
setgid: Set effective group ID of calling process, with CAP_SETGID capability it also sets real GID and saved set-group-ID
setgroups: Defines supplementary group IDs of calling process
sethostname: Sets the hostname to the defined value
setns: Allows calling thread to switch to a different namespace
setpgid: Sets the process group ID (PGID) of the process
setrlimit: Set resource limits
setsockopt: Set options on socket
settimeofday: Set time or timezone
setuid: Set effective user ID of calling process, with CAP_SETUID capability it also sets real UID and saved set-user-ID
sigreturn: Allow implementation of signal handlers
socket: Create endpoint for communication and return file descriptor
socketpair: Create a pair of connected sockets, for example for communication between parent and child process
stat: Get information about file
statfs: Returns information about mounted file system
symlink: Create symbolic link
symlinkat: Similar to symlink() but deals differently with relative paths

T

tee: Duplicate pipe content, does not consume the data
time: Return time; as number of seconds since the Epoch (1970-01-01 00:00:00 +0000 (UTC))
times: Get process and child process times, including CPU time in userspace and by the system for the calling process, and similar for the child processes
truncate: Truncate a writable file to specified number of bytes, which may fill it with null bytes (\0) or decrease its size and losing data

U

umask: Set file mode creation mask
uname: Retrieve name and information about the current kernel
unlink: Delete name from filesystem
unlinkat: Similar to unlink() but deals differently with relative paths
unshare: Allows a process to unshare parts of its execution context, such as mount namespace, from other processes. Parts of the execution context are automatically shared with other processes when fork(2), vfork(2) or clone(2) are used. With this syscall it does not have not to create a new process.
utime: Change access and modification times of inode
utimes: Similar to utime(), but uses array instead of a structure

W

wait: Suspend the execution of the calling process thread until one of its child processes terminates.
waitpid: Suspend the execution of the calling process thread until one of the specified child processes (by PID) terminates. By using specific options, also other actions like termination can be waited for. The functionality of this system call is similar to wait(), yet with more control over the children and states.
write: Write to file descriptor
writev: Writes buffers to file