This patch introduces the new syscall pipe2 which is like pipe but it also
takes an additional parameter which takes a flag value. This patch implements
the handling of O_CLOEXEC for the flag. I did not add support for the new
syscall for the architectures which have a special sys_pipe implementation. I
think the maintainers of those archs have the chance to go with the unified
implementation but that's up to them.
The implementation introduces do_pipe_flags. I did that instead of changing
all callers of do_pipe because some of the callers are written in assembler.
I would probably screw up changing the assembly code. To avoid breaking code
do_pipe is now a small wrapper around do_pipe_flags. Once all callers are
changed over to do_pipe_flags the old do_pipe function can be removed.
The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>
#ifndef __NR_pipe2
# ifdef __x86_64__
# define __NR_pipe2 293
# elif defined __i386__
# define __NR_pipe2 331
# else
# error "need __NR_pipe2"
# endif
#endif
int
main (void)
{
int fd[2];
if (syscall (__NR_pipe2, fd, 0) != 0)
{
puts ("pipe2(0) failed");
return 1;
}
for (int i = 0; i < 2; ++i)
{
int coe = fcntl (fd[i], F_GETFD);
if (coe == -1)
{
puts ("fcntl failed");
return 1;
}
if (coe & FD_CLOEXEC)
{
printf ("pipe2(0) set close-on-exit for fd[%d]\n", i);
return 1;
}
}
close (fd[0]);
close (fd[1]);
if (syscall (__NR_pipe2, fd, O_CLOEXEC) != 0)
{
puts ("pipe2(O_CLOEXEC) failed");
return 1;
}
for (int i = 0; i < 2; ++i)
{
int coe = fcntl (fd[i], F_GETFD);
if (coe == -1)
{
puts ("fcntl failed");
return 1;
}
if ((coe & FD_CLOEXEC) == 0)
{
printf ("pipe2(O_CLOEXEC) does not set close-on-exit for fd[%d]\n", i);
return 1;
}
}
close (fd[0]);
close (fd[1]);
puts ("OK");
return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 files changed:
+ retval = do_pipe_flags(fds, 0);
if (retval)
goto out;
if (copy_to_user(fd, fds, sizeof(fds)))
if (retval)
goto out;
if (copy_to_user(fd, fds, sizeof(fds)))
+ retval = do_pipe_flags(fd, 0);
if (retval)
goto out;
retval = fd[0];
if (retval)
goto out;
retval = fd[0];
int fd[2];
int error, res;
int fd[2];
int error, res;
+ error = do_pipe_flags(fd, 0);
if (error) {
res = error;
goto out;
if (error) {
res = error;
goto out;
int error;
lock_kernel();
int error;
lock_kernel();
- error = do_pipe(kstack_fildes);
+ error = do_pipe_flags(kstack_fildes, 0);
unlock_kernel();
return error;
}
unlock_kernel();
return error;
}
+ error = do_pipe_flags(fd, 0);
if (!error) {
regs->regs[1] = fd[1];
return fd[0];
if (!error) {
regs->regs[1] = fd[1];
return fd[0];
+ error = do_pipe_flags(fd, 0);
if (error)
goto out;
regs->u_regs[UREG_I1] = fd[1];
if (error)
goto out;
regs->u_regs[UREG_I1] = fd[1];
+ error = do_pipe_flags(fd, 0);
if (error)
goto out;
regs->u_regs[UREG_I1] = fd[1];
if (error)
goto out;
regs->u_regs[UREG_I1] = fd[1];
.quad sys_eventfd2
.quad sys_epoll_create2
.quad sys_dup3 /* 330 */
.quad sys_eventfd2
.quad sys_epoll_create2
.quad sys_dup3 /* 330 */
+ retval = do_pipe_flags(fds, 0);
if (retval)
goto out;
if (copy_to_user(fd, fds, sizeof(fds)))
if (retval)
goto out;
if (copy_to_user(fd, fds, sizeof(fds)))
.long sys_eventfd2
.long sys_epoll_create2
.long sys_dup3 /* 330 */
.long sys_eventfd2
.long sys_epoll_create2
.long sys_dup3 /* 330 */
+ error = do_pipe_flags(fd, 0);
if (!error) {
if (copy_to_user(userfds, fd, 2 * sizeof(int)))
error = -EFAULT;
if (!error) {
if (copy_to_user(userfds, fd, 2 * sizeof(int)))
error = -EFAULT;
+int do_pipe_flags(int *fd, int flags)
{
struct file *fw, *fr;
int error;
int fdw, fdr;
{
struct file *fw, *fr;
int error;
int fdw, fdr;
+ if (flags & ~O_CLOEXEC)
+ return -EINVAL;
+
fw = create_write_pipe();
if (IS_ERR(fw))
return PTR_ERR(fw);
fw = create_write_pipe();
if (IS_ERR(fw))
return PTR_ERR(fw);
if (IS_ERR(fr))
goto err_write_pipe;
if (IS_ERR(fr))
goto err_write_pipe;
- error = get_unused_fd();
+ error = get_unused_fd_flags(flags);
if (error < 0)
goto err_read_pipe;
fdr = error;
if (error < 0)
goto err_read_pipe;
fdr = error;
- error = get_unused_fd();
+ error = get_unused_fd_flags(flags);
if (error < 0)
goto err_fdr;
fdw = error;
if (error < 0)
goto err_fdr;
fdw = error;
+int do_pipe(int *fd)
+{
+ return do_pipe_flags(fd, 0);
+}
+
/*
* sys_pipe() is the normal C calling standard for creating
* a pipe. It's not the way Unix traditionally does this, though.
*/
/*
* sys_pipe() is the normal C calling standard for creating
* a pipe. It's not the way Unix traditionally does this, though.
*/
-asmlinkage long __weak sys_pipe(int __user *fildes)
+asmlinkage long __weak sys_pipe2(int __user *fildes, int flags)
+ error = do_pipe_flags(fd, flags);
if (!error) {
if (copy_to_user(fildes, fd, sizeof(fd))) {
sys_close(fd[0]);
if (!error) {
if (copy_to_user(fildes, fd, sizeof(fd))) {
sys_close(fd[0]);
+asmlinkage long __weak sys_pipe(int __user *fildes)
+{
+ return sys_pipe2(fildes, 0);
+}
+
/*
* pipefs should _never_ be mounted by userland - too much of security hassle,
* no real gain from having the whole whorehouse mounted. So we don't need
/*
* pipefs should _never_ be mounted by userland - too much of security hassle,
* no real gain from having the whole whorehouse mounted. So we don't need
#define __NR_eventfd2 328
#define __NR_epoll_create2 329
#define __NR_dup3 330
#define __NR_eventfd2 328
#define __NR_epoll_create2 329
#define __NR_dup3 330
__SYSCALL(__NR_epoll_create2, sys_epoll_create2)
#define __NR_dup3 292
__SYSCALL(__NR_dup3, sys_dup3)
__SYSCALL(__NR_epoll_create2, sys_epoll_create2)
#define __NR_dup3 292
__SYSCALL(__NR_dup3, sys_dup3)
+#define __NR_pipe2 293
+__SYSCALL(__NR_pipe2, sys_pipe2)
atomic_inc(&file->f_path.dentry->d_inode->i_writecount);
}
extern int do_pipe(int *);
atomic_inc(&file->f_path.dentry->d_inode->i_writecount);
}
extern int do_pipe(int *);
+extern int do_pipe_flags(int *, int);
extern struct file *create_read_pipe(struct file *f);
extern struct file *create_write_pipe(void);
extern void free_write_pipe(struct file *);
extern struct file *create_read_pipe(struct file *f);
extern struct file *create_write_pipe(void);
extern void free_write_pipe(struct file *);