Move libcontainer to x/sys/unix

Since syscall is outdated and broken for some architectures,
use x/sys/unix instead.

There are still some dependencies on the syscall package that will
remain in syscall for the forseeable future:

Errno
Signal
SysProcAttr

Additionally:
- os still uses syscall, so it needs to be kept for anything
returning *os.ProcessState, such as process.Wait.

Signed-off-by: Christy Perez <christy@linux.vnet.ibm.com>

libcontainer/utils/utils.go

@@ -8,8 +8,9 @@ import (
 	"os"
 	"path/filepath"
 	"strings"
-	"syscall"
 	"unsafe"
+
+	"golang.org/x/sys/unix"
 )
 
 const (
@@ -41,7 +42,7 @@ func ResolveRootfs(uncleanRootfs string) (string, error) {
 
 // ExitStatus returns the correct exit status for a process based on if it
 // was signaled or exited cleanly
-func ExitStatus(status syscall.WaitStatus) int {
+func ExitStatus(status unix.WaitStatus) int {
 	if status.Signaled() {
 		return exitSignalOffset + int(status.Signal())
 	}

libcontainer/utils/utils_test.go

@@ -5,8 +5,9 @@ import (
 	"fmt"
 	"os"
 	"path/filepath"
-	"syscall"
 	"testing"
+
+	"golang.org/x/sys/unix"
 )
 
 func TestGenerateName(t *testing.T) {
@@ -94,7 +95,7 @@ func TestResolveRootfsWithNonExistingDir(t *testing.T) {
 }
 
 func TestExitStatus(t *testing.T) {
-	status := syscall.WaitStatus(0)
+	status := unix.WaitStatus(0)
 	ex := ExitStatus(status)
 	if ex != 0 {
 		t.Errorf("expected exit status to equal 0 and received %d", ex)
@@ -102,7 +103,7 @@ func TestExitStatus(t *testing.T) {
 }
 
 func TestExitStatusSignaled(t *testing.T) {
-	status := syscall.WaitStatus(2)
+	status := unix.WaitStatus(2)
 	ex := ExitStatus(status)
 	if ex != 130 {
 		t.Errorf("expected exit status to equal 130 and received %d", ex)

libcontainer/utils/utils_unix.go

@@ -6,7 +6,8 @@ import (
 	"io/ioutil"
 	"os"
 	"strconv"
-	"syscall"
+
+	"golang.org/x/sys/unix"
 )
 
 func CloseExecFrom(minFd int) error {
@@ -26,8 +27,8 @@ func CloseExecFrom(minFd int) error {
 			continue
 		}
 
-		// intentionally ignore errors from syscall.CloseOnExec
-		syscall.CloseOnExec(fd)
+		// intentionally ignore errors from unix.CloseOnExec
+		unix.CloseOnExec(fd)
 		// the cases where this might fail are basically file descriptors that have already been closed (including and especially the one that was created when ioutil.ReadDir did the "opendir" syscall)
 	}
 	return nil
@@ -35,7 +36,7 @@ func CloseExecFrom(minFd int) error {
 
 // NewSockPair returns a new unix socket pair
 func NewSockPair(name string) (parent *os.File, child *os.File, err error) {
-	fds, err := syscall.Socketpair(syscall.AF_LOCAL, syscall.SOCK_STREAM|syscall.SOCK_CLOEXEC, 0)
+	fds, err := unix.Socketpair(unix.AF_LOCAL, unix.SOCK_STREAM|unix.SOCK_CLOEXEC, 0)
 	if err != nil {
 		return nil, nil, err
 	}