// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Linux system calls.
// This file is compiled as ordinary Go code,
// but it is also input to mksyscall,
// which parses the //sys lines and generates system call stubs.
// Note that sometimes we use a lowercase //sys name and
// wrap it in our own nicer implementation.
package syscall
import "unsafe"
const OS = "linux"
/*
* Wrapped
*/
//sys pipe(p *[2]_C_int) (errno int)
func Pipe(p []int) (errno int) {
if len(p) != 2 {
return EINVAL
}
var pp [2]_C_int;
errno = pipe(&pp);
p[0] = int(pp[0]);
p[1] = int(pp[1]);
return;
}
//sys utimes(path string, times *[2]Timeval) (errno int)
func Utimes(path string, tv []Timeval) (errno int) {
if len(tv) != 2 {
return EINVAL
}
return utimes(path, (*[2]Timeval)(unsafe.Pointer(&tv[0])));
}
//sys futimesat(dirfd int, path string, times *[2]Timeval) (errno int)
func Futimesat(dirfd int, path string, tv []Timeval) (errno int) {
if len(tv) != 2 {
return EINVAL
}
return futimesat(dirfd, path, (*[2]Timeval)(unsafe.Pointer(&tv[0])));
}
const ImplementsGetwd = true
//sys Getcwd(buf []byte) (n int, errno int)
func Getwd() (wd string, errno int) {
var buf [PathMax]byte;
n, err := Getcwd(&buf);
if err != 0 {
return "", err
}
// Getcwd returns the number of bytes written to buf, including the NUL.
if n < 1 || n > len(buf) || buf[n-1] != 0 {
return "", EINVAL
}
return string(buf[0 : n-1]), 0;
}
func Getgroups() (gids []int, errno int) {
n, err := getgroups(0, nil);
if err != 0 {
return nil, errno
}
if n == 0 {
return nil, 0
}
// Sanity check group count. Max is 1<<16 on Linux.
if n < 0 || n > 1<<20 {
return nil, EINVAL
}
a := make([]_Gid_t, n);
n, err = getgroups(n, &a[0]);
if err != 0 {
return nil, errno
}
gids = make([]int, n);
for i, v := range a[0:n] {
gids[i] = int(v)
}
return;
}
func Setgroups(gids []int) (errno int) {
if len(gids) == 0 {
return setgroups(0, nil)
}
a := make([]_Gid_t, len(gids));
for i, v := range gids {
a[i] = _Gid_t(v)
}
return setgroups(len(a), &a[0]);
}
type WaitStatus uint32
// Wait status is 7 bits at bottom, either 0 (exited),
// 0x7F (stopped), or a signal number that caused an exit.
// The 0x80 bit is whether there was a core dump.
// An extra number (exit code, signal causing a stop)
// is in the high bits. At least that's the idea.
// There are various irregularities. For example, the
// "continued" status is 0xFFFF, distinguishing itself
// from stopped via the core dump bit.
const (
mask = 0x7F;
core = 0x80;
exited = 0x00;
stopped = 0x7F;
shift = 8;
)
func (w WaitStatus) Exited() bool { return w&mask == exited }
func (w WaitStatus) Signaled() bool { return w&mask != stopped && w&mask != exited }
func (w WaitStatus) Stopped() bool { return w&0xFF == stopped }
func (w WaitStatus) Continued() bool { return w == 0xFFFF }
func (w WaitStatus) CoreDump() bool { return w.Signaled() && w&core != 0 }
func (w WaitStatus) ExitStatus() int {
if !w.Exited() {
return -1
}
return int(w>>shift) & 0xFF;
}
func (w WaitStatus) Signal() int {
if !w.Signaled() {
return -1
}
return int(w & mask);
}
func (w WaitStatus) StopSignal() int {
if !w.Stopped() {
return -1
}
return int(w>>shift) & 0xFF;
}
func (w WaitStatus) TrapCause() int {
if w.StopSignal() != SIGTRAP {
return -1
}
return int(w>>shift) >> 8;
}
//sys wait4(pid int, wstatus *_C_int, options int, rusage *Rusage) (wpid int, errno int)
func Wait4(pid int, wstatus *WaitStatus, options int, rusage *Rusage) (wpid int, errno int) {
var status _C_int;
wpid, errno = wait4(pid, &status, options, rusage);
if wstatus != nil {
*wstatus = WaitStatus(status)
}
return;
}
func Sleep(nsec int64) (errno int) {
tv := NsecToTimeval(nsec);
_, err := Select(0, nil, nil, nil, &tv);
return err;
}
// For testing: clients can set this flag to force
// creation of IPv6 sockets to return EAFNOSUPPORT.
var SocketDisableIPv6 bool
type Sockaddr interface {
sockaddr() (ptr uintptr, len _Socklen, errno int); // lowercase; only we can define Sockaddrs
}
type SockaddrInet4 struct {
Port int;
Addr [4]byte;
raw RawSockaddrInet4;
}
func (sa *SockaddrInet4) sockaddr() (uintptr, _Socklen, int) {
if sa.Port < 0 || sa.Port > 0xFFFF {
return 0, 0, EINVAL
}
sa.raw.Family = AF_INET;
p := (*[2]byte)(unsafe.Pointer(&sa.raw.Port));
p[0] = byte(sa.Port >> 8);
p[1] = byte(sa.Port);
for i := 0; i < len(sa.Addr); i++ {
sa.raw.Addr[i] = sa.Addr[i]
}
return uintptr(unsafe.Pointer(&sa.raw)), SizeofSockaddrInet4, 0;
}
type SockaddrInet6 struct {
Port int;
Addr [16]byte;
raw RawSockaddrInet6;
}
func (sa *SockaddrInet6) sockaddr() (uintptr, _Socklen, int) {
if sa.Port < 0 || sa.Port > 0xFFFF {
return 0, 0, EINVAL
}
sa.raw.Family = AF_INET6;
p := (*[2]byte)(unsafe.Pointer(&sa.raw.Port));
p[0] = byte(sa.Port >> 8);
p[1] = byte(sa.Port);
for i := 0; i < len(sa.Addr); i++ {
sa.raw.Addr[i] = sa.Addr[i]
}
return uintptr(unsafe.Pointer(&sa.raw)), SizeofSockaddrInet6, 0;
}
type SockaddrUnix struct {
Name string;
raw RawSockaddrUnix;
}
func (sa *SockaddrUnix) sockaddr() (uintptr, _Socklen, int) {
name := sa.Name;
n := len(name);
if n >= len(sa.raw.Path) || n == 0 {
return 0, 0, EINVAL
}
sa.raw.Family = AF_UNIX;
for i := 0; i < n; i++ {
sa.raw.Path[i] = int8(name[i])
}
if sa.raw.Path[0] == '@' {
sa.raw.Path[0] = 0
}
// length is family, name, NUL.
return uintptr(unsafe.Pointer(&sa.raw)), 1 + _Socklen(n) + 1, 0;
}
func anyToSockaddr(rsa *RawSockaddrAny) (Sockaddr, int) {
switch rsa.Addr.Family {
case AF_UNIX:
pp := (*RawSockaddrUnix)(unsafe.Pointer(rsa));
sa := new(SockaddrUnix);
if pp.Path[0] == 0 {
// "Abstract" Unix domain socket.
// Rewrite leading NUL as @ for textual display.
// (This is the standard convention.)
// Not friendly to overwrite in place,
// but the callers below don't care.
pp.Path[0] = '@'
}
// Assume path ends at NUL.
// This is not technically the Linux semantics for
// abstract Unix domain sockets--they are supposed
// to be uninterpreted fixed-size binary blobs--but
// everyone uses this convention.
n := 0;
for n < len(pp.Path) && pp.Path[n] != 0 {
n++
}
bytes := (*[len(pp.Path)]byte)(unsafe.Pointer(&pp.Path[0]));
sa.Name = string(bytes[0:n]);
return sa, 0;
case AF_INET:
pp := (*RawSockaddrInet4)(unsafe.Pointer(rsa));
sa := new(SockaddrInet4);
p := (*[2]byte)(unsafe.Pointer(&pp.Port));
sa.Port = int(p[0])<<8 + int(p[1]);
for i := 0; i < len(sa.Addr); i++ {
sa.Addr[i] = pp.Addr[i]
}
return sa, 0;
case AF_INET6:
pp := (*RawSockaddrInet6)(unsafe.Pointer(rsa));
sa := new(SockaddrInet6);
p := (*[2]byte)(unsafe.Pointer(&pp.Port));
sa.Port = int(p[0])<<8 + int(p[1]);
for i := 0; i < len(sa.Addr); i++ {
sa.Addr[i] = pp.Addr[i]
}
return sa, 0;
}
return nil, EAFNOSUPPORT;
}
func Accept(fd int) (nfd int, sa Sockaddr, errno int) {
var rsa RawSockaddrAny;
var len _Socklen = SizeofSockaddrAny;
nfd, errno = accept(fd, &rsa, &len);
if errno != 0 {
return
}
sa, errno = anyToSockaddr(&rsa);
if errno != 0 {
Close(nfd);
nfd = 0;
}
return;
}
func Getsockname(fd int) (sa Sockaddr, errno int) {
var rsa RawSockaddrAny;
var len _Socklen = SizeofSockaddrAny;
if errno = getsockname(fd, &rsa, &len); errno != 0 {
return
}
return anyToSockaddr(&rsa);
}
func Getpeername(fd int) (sa Sockaddr, errno int) {
var rsa RawSockaddrAny;
var len _Socklen = SizeofSockaddrAny;
if errno = getpeername(fd, &rsa, &len); errno != 0 {
return
}
return anyToSockaddr(&rsa);
}
func Bind(fd int, sa Sockaddr) (errno int) {
ptr, n, err := sa.sockaddr();
if err != 0 {
return err
}
return bind(fd, ptr, n);
}
func Connect(fd int, sa Sockaddr) (errno int) {
ptr, n, err := sa.sockaddr();
if err != 0 {
return err
}
return connect(fd, ptr, n);
}
func Socket(domain, typ, proto int) (fd, errno int) {
if domain == AF_INET6 && SocketDisableIPv6 {
return -1, EAFNOSUPPORT
}
fd, errno = socket(domain, typ, proto);
return;
}
func SetsockoptInt(fd, level, opt int, value int) (errno int) {
var n = int32(value);
return setsockopt(fd, level, opt, uintptr(unsafe.Pointer(&n)), 4);
}
func SetsockoptTimeval(fd, level, opt int, tv *Timeval) (errno int) {
return setsockopt(fd, level, opt, uintptr(unsafe.Pointer(tv)), unsafe.Sizeof(*tv))
}
func SetsockoptLinger(fd, level, opt int, l *Linger) (errno int) {
return setsockopt(fd, level, opt, uintptr(unsafe.Pointer(l)), unsafe.Sizeof(*l))
}
func Recvfrom(fd int, p []byte, flags int) (n int, from Sockaddr, errno int) {
var rsa RawSockaddrAny;
var len _Socklen = SizeofSockaddrAny;
if n, errno = recvfrom(fd, p, flags, &rsa, &len); errno != 0 {
return
}
from, errno = anyToSockaddr(&rsa);
return;
}
func Sendto(fd int, p []byte, flags int, to Sockaddr) (errno int) {
ptr, n, err := to.sockaddr();
if err != 0 {
return err
}
return sendto(fd, p, flags, ptr, n);
}
//sys ptrace(request int, pid int, addr uintptr, data uintptr) (errno int)
func ptracePeek(req int, pid int, addr uintptr, out []byte) (count int, errno int) {
// The peek requests are machine-size oriented, so we wrap it
// to retrieve arbitrary-length data.
// The ptrace syscall differs from glibc's ptrace.
// Peeks returns the word in *data, not as the return value.
var buf [sizeofPtr]byte;
// Leading edge. PEEKTEXT/PEEKDATA don't require aligned
// access (PEEKUSER warns that it might), but if we don't
// align our reads, we might straddle an unmapped page
// boundary and not get the bytes leading up to the page
// boundary.
n := 0;
if addr%sizeofPtr != 0 {
errno = ptrace(req, pid, addr-addr%sizeofPtr, uintptr(unsafe.Pointer(&buf[0])));
if errno != 0 {
return 0, errno
}
n += copy(out, buf[addr%sizeofPtr:]);
out = out[n:];
}
// Remainder.
for len(out) > 0 {
// We use an internal buffer to gaurantee alignment.
// It's not documented if this is necessary, but we're paranoid.
errno = ptrace(req, pid, addr+uintptr(n), uintptr(unsafe.Pointer(&buf[0])));
if errno != 0 {
return n, errno
}
copied := copy(out, &buf);
n += copied;
out = out[copied:];
}
return n, 0;
}
func PtracePeekText(pid int, addr uintptr, out []byte) (count int, errno int) {
return ptracePeek(PTRACE_PEEKTEXT, pid, addr, out)
}
func PtracePeekData(pid int, addr uintptr, out []byte) (count int, errno int) {
return ptracePeek(PTRACE_PEEKDATA, pid, addr, out)
}
func ptracePoke(pokeReq int, peekReq int, pid int, addr uintptr, data []byte) (count int, errno int) {
// As for ptracePeek, we need to align our accesses to deal
// with the possibility of straddling an invalid page.
// Leading edge.
n := 0;
if addr%sizeofPtr != 0 {
var buf [sizeofPtr]byte;
errno = ptrace(peekReq, pid, addr-addr%sizeofPtr, uintptr(unsafe.Pointer(&buf[0])));
if errno != 0 {
return 0, errno
}
n += copy(buf[addr%sizeofPtr:], data);
word := *((*uintptr)(unsafe.Pointer(&buf[0])));
errno = ptrace(pokeReq, pid, addr-addr%sizeofPtr, word);
if errno != 0 {
return 0, errno
}
data = data[n:];
}
// Interior.
for len(data) > sizeofPtr {
word := *((*uintptr)(unsafe.Pointer(&data[0])));
errno = ptrace(pokeReq, pid, addr+uintptr(n), word);
if errno != 0 {
return n, errno
}
n += sizeofPtr;
data = data[sizeofPtr:];
}
// Trailing edge.
if len(data) > 0 {
var buf [sizeofPtr]byte;
errno = ptrace(peekReq, pid, addr+uintptr(n), uintptr(unsafe.Pointer(&buf[0])));
if errno != 0 {
return n, errno
}
copy(&buf, data);
word := *((*uintptr)(unsafe.Pointer(&buf[0])));
errno = ptrace(pokeReq, pid, addr+uintptr(n), word);
if errno != 0 {
return n, errno
}
n += len(data);
}
return n, 0;
}
func PtracePokeText(pid int, addr uintptr, data []byte) (count int, errno int) {
return ptracePoke(PTRACE_POKETEXT, PTRACE_PEEKTEXT, pid, addr, data)
}
func PtracePokeData(pid int, addr uintptr, data []byte) (count int, errno int) {
return ptracePoke(PTRACE_POKEDATA, PTRACE_PEEKDATA, pid, addr, data)
}
func PtraceGetRegs(pid int, regsout *PtraceRegs) (errno int) {
return ptrace(PTRACE_GETREGS, pid, 0, uintptr(unsafe.Pointer(regsout)))
}
func PtraceSetRegs(pid int, regs *PtraceRegs) (errno int) {
return ptrace(PTRACE_SETREGS, pid, 0, uintptr(unsafe.Pointer(regs)))
}
func PtraceSetOptions(pid int, options int) (errno int) {
return ptrace(PTRACE_SETOPTIONS, pid, 0, uintptr(options))
}
func PtraceGetEventMsg(pid int) (msg uint, errno int) {
var data _C_long;
errno = ptrace(PTRACE_GETEVENTMSG, pid, 0, uintptr(unsafe.Pointer(&data)));
msg = uint(data);
return;
}
func PtraceCont(pid int, signal int) (errno int) {
return ptrace(PTRACE_CONT, pid, 0, uintptr(signal))
}
func PtraceSingleStep(pid int) (errno int) { return ptrace(PTRACE_SINGLESTEP, pid, 0, 0) }
func PtraceAttach(pid int) (errno int) { return ptrace(PTRACE_ATTACH, pid, 0, 0) }
func PtraceDetach(pid int) (errno int) { return ptrace(PTRACE_DETACH, pid, 0, 0) }
// Sendto
// Recvfrom
// Sendmsg
// Recvmsg
// Socketpair
// Getsockopt
/*
* Direct access
*/
//sys Access(path string, mode int) (errno int)
//sys Acct(path string) (errno int)
//sys Adjtimex(buf *Timex) (state int, errno int)
//sys Chdir(path string) (errno int)
//sys Chmod(path string, mode int) (errno int)
//sys Chroot(path string) (errno int)
//sys Close(fd int) (errno int)
//sys Creat(path string, mode int) (fd int, errno int)
//sys Dup(oldfd int) (fd int, errno int)
//sys Dup2(oldfd int, newfd int) (fd int, errno int)
//sys EpollCreate(size int) (fd int, errno int)
//sys EpollCtl(epfd int, op int, fd int, event *EpollEvent) (errno int)
//sys EpollWait(epfd int, events []EpollEvent, msec int) (n int, errno int)
//sys Exit(code int) = SYS_EXIT_GROUP
//sys Faccessat(dirfd int, path string, mode int, flags int) (errno int)
//sys Fallocate(fd int, mode int, off int64, len int64) (errno int)
//sys Fchdir(fd int) (errno int)
//sys Fchmod(fd int, mode int) (errno int)
//sys Fchmodat(dirfd int, path string, mode int, flags int) (errno int)
//sys Fchownat(dirfd int, path string, uid int, gid int, flags int) (errno int)
//sys fcntl(fd int, cmd int, arg int) (val int, errno int)
//sys Fdatasync(fd int) (errno int)
//sys Fsync(fd int) (errno int)
//sys Ftruncate(fd int, length int64) (errno int)
//sys Getdents(fd int, buf []byte) (n int, errno int) = SYS_GETDENTS64
//sys Getpgid(pid int) (pgid int, errno int)
//sys Getpgrp() (pid int)
//sys Getpid() (pid int)
//sys Getppid() (ppid int)
//sys Getrlimit(resource int, rlim *Rlimit) (errno int)
//sys Getrusage(who int, rusage *Rusage) (errno int)
//sys Gettid() (tid int)
//sys Gettimeofday(tv *Timeval) (errno int)
//sys Kill(pid int, sig int) (errno int)
//sys Klogctl(typ int, buf []byte) (n int, errno int) = SYS_SYSLOG
//sys Link(oldpath string, newpath string) (errno int)
//sys Mkdir(path string, mode int) (errno int)
//sys Mkdirat(dirfd int, path string, mode int) (errno int)
//sys Mknod(path string, mode int, dev int) (errno int)
//sys Mknodat(dirfd int, path string, mode int, dev int) (errno int)
//sys Nanosleep(time *Timespec, leftover *Timespec) (errno int)
//sys Open(path string, mode int, perm int) (fd int, errno int)
//sys Openat(dirfd int, path string, flags int, mode int) (fd int, errno int)
//sys Pause() (errno int)
//sys PivotRoot(newroot string, putold string) (errno int) = SYS_PIVOT_ROOT
//sys Pread(fd int, p []byte, offset int64) (n int, errno int) = SYS_PREAD64
//sys Pwrite(fd int, p []byte, offset int64) (n int, errno int) = SYS_PWRITE64
//sys Read(fd int, p []byte) (n int, errno int)
//sys Readlink(path string, buf []byte) (n int, errno int)
//sys Rename(oldpath string, newpath string) (errno int)
//sys Renameat(olddirfd int, oldpath string, newdirfd int, newpath string) (errno int)
//sys Rmdir(path string) (errno int)
//sys Setdomainname(p []byte) (errno int)
//sys Sethostname(p []byte) (errno int)
//sys Setpgid(pid int, pgid int) (errno int)
//sys Setrlimit(resource int, rlim *Rlimit) (errno int)
//sys Setsid() (pid int)
//sys Settimeofday(tv *Timeval) (errno int)
//sys Setuid(uid int) (errno int)
//sys Splice(rfd int, roff *int64, wfd int, woff *int64, len int, flags int) (n int64, errno int)
//sys Symlink(oldpath string, newpath string) (errno int)
//sys Sync()
//sys Sysinfo(info *Sysinfo_t) (errno int)
//sys Tee(rfd int, wfd int, len int, flags int) (n int64, errno int)
//sys Tgkill(tgid int, tid int, sig int) (errno int)
//sys Time(t *Time_t) (tt Time_t, errno int)
//sys Times(tms *Tms) (ticks uintptr, errno int)
//sys Truncate(path string, length int64) (errno int)
//sys Umask(mask int) (oldmask int)
//sys Uname(buf *Utsname) (errno int)
//sys Unlink(path string) (errno int)
//sys Unlinkat(dirfd int, path string) (errno int)
//sys Unshare(flags int) (errno int)
//sys Ustat(dev int, ubuf *Ustat_t) (errno int)
//sys Utime(path string, buf *Utimbuf) (errno int)
//sys Write(fd int, p []byte) (n int, errno int)
//sys exitThread(code int) (errno int) = SYS_EXIT
//sys read(fd int, p *byte, np int) (n int, errno int)
//sys write(fd int, p *byte, np int) (n int, errno int)
/*
* Unimplemented
*/
// AddKey
// AfsSyscall
// Alarm
// ArchPrctl
// Brk
// Capget
// Capset
// ClockGetres
// ClockGettime
// ClockNanosleep
// ClockSettime
// Clone
// CreateModule
// DeleteModule
// EpollCtlOld
// EpollPwait
// EpollWaitOld
// Eventfd
// Execve
// Fadvise64
// Fgetxattr
// Flistxattr
// Flock
// Fork
// Fremovexattr
// Fsetxattr
// Futex
// GetKernelSyms
// GetMempolicy
// GetRobustList
// GetThreadArea
// Getitimer
// Getpmsg
// Getpriority
// Getxattr
// InotifyAddWatch
// InotifyInit
// InotifyRmWatch
// IoCancel
// IoDestroy
// IoGetevents
// IoSetup
// IoSubmit
// Ioctl
// IoprioGet
// IoprioSet
// KexecLoad
// Keyctl
// Lgetxattr
// Listxattr
// Llistxattr
// LookupDcookie
// Lremovexattr
// Lsetxattr
// Madvise
// Mbind
// MigratePages
// Mincore
// Mlock
// Mmap
// ModifyLdt
// Mount
// MovePages
// Mprotect
// MqGetsetattr
// MqNotify
// MqOpen
// MqTimedreceive
// MqTimedsend
// MqUnlink
// Mremap
// Msgctl
// Msgget
// Msgrcv
// Msgsnd
// Msync
// Munlock
// Munlockall
// Munmap
// Newfstatat
// Nfsservctl
// Personality
// Poll
// Ppoll
// Prctl
// Pselect6
// Ptrace
// Putpmsg
// QueryModule
// Quotactl
// Readahead
// Readv
// Reboot
// RemapFilePages
// Removexattr
// RequestKey
// RestartSyscall
// RtSigaction
// RtSigpending
// RtSigprocmask
// RtSigqueueinfo
// RtSigreturn
// RtSigsuspend
// RtSigtimedwait
// SchedGetPriorityMax
// SchedGetPriorityMin
// SchedGetaffinity
// SchedGetparam
// SchedGetscheduler
// SchedRrGetInterval
// SchedSetaffinity
// SchedSetparam
// SchedYield
// Security
// Semctl
// Semget
// Semop
// Semtimedop
// Sendfile
// SetMempolicy
// SetRobustList
// SetThreadArea
// SetTidAddress
// Setpriority
// Setxattr
// Shmat
// Shmctl
// Shmdt
// Shmget
// Sigaltstack
// Signalfd
// Swapoff
// Swapon
// Sysfs
// TimerCreate
// TimerDelete
// TimerGetoverrun
// TimerGettime
// TimerSettime
// Timerfd
// Tkill (obsolete)
// Tuxcall
// Umount2
// Uselib
// Utimensat
// Vfork
// Vhangup
// Vmsplice
// Vserver
// Waitid
// Writev
// _Sysctl
|