1SYSCALLS(2) Linux Programmer's Manual SYSCALLS(2)
2
6 none - list of all system calls
7
9 Linux 2.4 system calls.
10
12 The system call is the fundamental interface between an application and
13 the Linux kernel. As of Linux 2.4.17, there are 1100 system calls
14 listed in /usr/src/linux/include/asm-*/unistd.h. This man page lists
15 those that are common to most platforms.
16 _llseek(2), _newselect(2), _sysctl(2), accept(2), access(2), acct(2),
18 adjtimex(2), afs_syscall, alarm(2), bdflush(2), bind(2), break, brk(2),
19 cacheflush(2), capget(2), capset(2), chdir(2), chmod(2), chown(2),
20 chown32, chroot(2), clone(2), close(2), connect(2), creat(2), cre‐
21 ate_module(2), delete_module(2), dup(2), dup2(2), execve(2), exit(2),
22 fchdir(2), fchmod(2), fchown(2), fchown32, fcntl(2), fcntl64, fdata‐
23 sync(2), flock(2), fork(2), fstat(2), fstat64, fstatfs(2), fsync(2),
24 ftime, ftruncate(2), ftruncate64, get_kernel_syms(2), getcwd(2), get‐
25 dents(2), getdents64, getegid(2), getegid32, geteuid(2), geteuid32,
26 getgid(2), getgid32, getgroups(2), getgroups32, getitimer(2), get‐
27 pagesize(2), getpeername(2), getpmsg, getpgid(2), getpgrp(2), get‐
28 pid(2), getppid(2), getpriority(2), getresgid(2), getresgid32, get‐
29 resuid(2), getresuid32, getrlimit(2), getrusage(2), getsid(2), get‐
30 sockname(2), getsockopt(2), gettid, gettimeofday(2), getuid(2), get‐
31 uid32, gtty, idle, init_module(2), ioctl(2), ioperm(2), iopl(2),
32 ipc(2), kill(2), lchown(2), lchown32, link(2), listen(2), lock,
33 lseek(2), lstat(2), lstat64, madvise(2), mincore(2), mkdir(2),
34 mknod(2), mlock(2), mlockall(2), mmap(2), modify_ldt(2), mount(2),
35 mprotect(2), mpx, mremap(2), msync(2), munlock(2), munlockall(2), mun‐
36 map(2), nanosleep(2), nfsservctl(2), nice(2), oldfstat, oldlstat,
37 oldolduname, oldstat, oldumount, olduname, open(2), pause(2), personal‐
38 ity(2), phys, pipe(2), pivot_root(2), poll(2), prctl(2), pread(2),
39 prof, profil, ptrace(2), putpmsg, pwrite(2), query_module(2), quo‐
40 tactl(2), read(2), readahead, readdir(2), readlink(2), readv(2),
41 reboot(2), recv(2), recvfrom(2), recvmsg(2), rename(2), rmdir(2),
42 rt_sigaction, rt_sigpending, rt_sigprocmask, rt_sigqueueinfo, rt_sigre‐
43 turn, rt_sigsuspend, rt_sigtimedwait, sched_get_priority_max(2),
44 sched_get_priority_min(2), sched_getparam(2), sched_getscheduler(2),
45 sched_rr_get_interval(2), sched_setparam(2), sched_setscheduler(2),
46 sched_yield(2), security, select(2), sendfile(2), send(2), sendmsg(2),
47 sendto(2), setdomainname(2), setfsgid(2), setfsgid32, setfsuid(2), set‐
48 fsuid32, setgid(2), setgid32, setgroups(2), setgroups32, set‐
49 hostname(2), setitimer(2), setpgid(2), setpriority(2), setregid(2),
50 setregid32, setresgid(2), setresgid32, setresuid(2), setresuid32, set‐
51 reuid(2), setreuid32, setrlimit(2), setsid(2), setsockopt(2), set‐
52 timeofday(2), setuid(2), setuid32, setup(2), sgetmask(2), shutdown(2),
53 sigaction(2), sigaltstack(2), signal(2), sigpending(2), sigprocmask(2),
54 sigreturn(2), sigsuspend(2), socket(2), socketcall(2), socketpair(2),
55 ssetmask(2), stat(2), stat64, statfs(2), stime(2), stty, swapoff(2),
56 swapon(2), symlink(2), sync(2), sysfs(2), sysinfo(2), syslog(2),
57 time(2), times(2), truncate(2), truncate64, ulimit, umask(2),
58 umount(2), uname(2), unlink(2), uselib(2), ustat(2), utime(2),
59 vfork(2), vhangup(2), vm86(2), vm86old, wait4(2), waitpid(2), write(2),
60 writev(2).
61 Of the above, 9 are obsolete, namely getrlimit, oldfstat, oldlstat,
63 oldolduname, oldstat, olduname, readdir, select and vm86old (see also
64 obsolete(2)), and 15 are unimplemented in the standard kernel, namely
65 afs_syscall, break, ftime, getpmsg, gtty, idle, lock, mpx, phys, prof,
66 profil, putpmsg, security, stty and ulimit (see also unimplemented(2)).
67 However, ftime(3), profil(3) and ulimit(3) exist as library routines.
68 The slot for phys is in use since 2.1.116 for umount; phys will never
69 be implemented. The getpmsg and putpmsg calls are for kernels patched
70 to support streams, and may never be in the standard kernel. The secu‐
71 rity call is for future use.
72 Roughly speaking, the code belonging to the system call with number
74 __NR_xxx defined in /usr/include/asm/unistd.h can be found in the ker‐
75 nel source in the routine sys_xxx(). (The dispatch table for i386 can
76 be found in /usr/src/linux/arch/i386/kernel/entry.S.) There are many
77 exceptions, however, mostly because older system calls were superseded
78 by newer ones, and this has been treated somewhat unsystematically. On
79 platforms with proprietary OS emulation, such as parisc, sparc, sparc64
80 and alpha, there are many additional system calls; mips64 also contains
81 a full set of 32-bit system calls. Below the details for Linux 2.4.17.
82 The defines __NR_oldstat and __NR_stat refer to the routines sys_stat()
84 and sys_newstat(), and similarly for fstat and lstat. Similarly, the
85 defines __NR_oldolduname, __NR_olduname and __NR_uname refer to the
86 routines sys_olduname(), sys_uname() and sys_newuname(). Thus,
87 __NR_stat and __NR_uname have always referred to the latest version of
88 the system call, and the older ones are for backward compatibility.
89 It is different with select and mmap. These use five or more parame‐
91 ters, and caused problems the way parameter passing on the i386 used to
92 be set up. Thus, while other architectures have sys_select() and
93 sys_mmap() corresponding to __NR_select and __NR_mmap, on i386 one
94 finds old_select() and old_mmap() (routines that use a pointer to a
95 parameter block) instead. These days passing five parameters is not a
96 problem any more, and there is a __NR__newselect (used by libc 6) that
97 corresponds directly to sys_select() and similarly __NR_mmap2.
98 Two other system call numbers, __NR__llseek and __NR__sysctl have an
100 additional underscore absent in sys_llseek() and sys_sysctl().
101 Then there is __NR_readdir corresponding to old_readdir(), which will
103 read at most one directory entry at a time, and is superseded by
104 sys_getdents().
105 On many platforms, including i386, socket calls are all multiplexed
107 through socketcall() and System V IPC calls through ipc().
108 On newer platforms that only have 64-bit file access and 32-bit uids
110 (e.g. alpha, ia64, s390x) there are no *64 or *32 calls. Where the *64
111 and *32 calls exist, the other versions are obsolete.
112 The chown and lchown system calls were swapped in 2.1.81. The *64 and
114 *32 calls were added for kernel 2.4, as were the new versions of getr‐
115 limit and mmap, and the new calls pivot_root, mincore, madvise, secu‐
116 rity, gettid and readahead.
117Linux 2.4 2002-01-07 SYSCALLS(2)