pidfd: cleanup the usage of __pidfd_prepare's flags

- make pidfd_create() static.

- Don't pass O_RDWR | O_CLOEXEC to __pidfd_prepare() in copy_process(),
  __pidfd_prepare() adds these flags unconditionally.

- Kill the flags check in __pidfd_prepare(). sys_pidfd_open() checks the
  flags itself, all other users of pidfd_prepare() pass flags = 0.

  If we need a sanity check for those other in kernel users then
  WARN_ON_ONCE(flags & ~PIDFD_NONBLOCK) makes more sense.

- Don't pass O_RDWR to get_unused_fd_flags(), it ignores everything except
  O_CLOEXEC.

- Don't pass O_CLOEXEC to anon_inode_getfile(), it ignores everything except
  O_ACCMODE | O_NONBLOCK.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20240125161734.GA778@redhat.comSigned-off-by: Christian Brauner <brauner@kernel.org>

include/linux/pid.h

@@ -73,7 +73,6 @@ struct file;
 extern struct pid *pidfd_pid(const struct file *file);
 struct pid *pidfd_get_pid(unsigned int fd, unsigned int *flags);
 struct task_struct *pidfd_get_task(int pidfd, unsigned int *flags);
-int pidfd_create(struct pid *pid, unsigned int flags);
 int pidfd_prepare(struct pid *pid, unsigned int flags, struct file **ret);
 
 static inline struct pid *get_pid(struct pid *pid)

kernel/fork.c

@@ -2130,15 +2130,12 @@ static int __pidfd_prepare(struct pid *pid, unsigned int flags, struct file **re
 	int pidfd;
 	struct file *pidfd_file;
 
-	if (flags & ~(O_NONBLOCK | O_RDWR | O_CLOEXEC))
-		return -EINVAL;
-
-	pidfd = get_unused_fd_flags(O_RDWR | O_CLOEXEC);
+	pidfd = get_unused_fd_flags(O_CLOEXEC);
 	if (pidfd < 0)
 		return pidfd;
 
 	pidfd_file = anon_inode_getfile("[pidfd]", &pidfd_fops, pid,
-					flags | O_RDWR | O_CLOEXEC);
+					flags | O_RDWR);
 	if (IS_ERR(pidfd_file)) {
 		put_unused_fd(pidfd);
 		return PTR_ERR(pidfd_file);
@@ -2524,7 +2521,7 @@ __latent_entropy struct task_struct *copy_process(
 	 */
 	if (clone_flags & CLONE_PIDFD) {
 		/* Note that no task has been attached to @pid yet. */
-		retval = __pidfd_prepare(pid, O_RDWR | O_CLOEXEC, &pidfile);
+		retval = __pidfd_prepare(pid, 0, &pidfile);
 		if (retval < 0)
 			goto bad_fork_free_pid;
 		pidfd = retval;

kernel/pid.c

@@ -595,7 +595,7 @@ struct task_struct *pidfd_get_task(int pidfd, unsigned int *flags)
  * Return: On success, a cloexec pidfd is returned.
  *         On error, a negative errno number will be returned.
  */
-int pidfd_create(struct pid *pid, unsigned int flags)
+static int pidfd_create(struct pid *pid, unsigned int flags)
 {
 	int pidfd;
 	struct file *pidfd_file;