[PATCH][RFC] don't bother with path_get()/path_put() in unix_open_file()

[PATCH][RFC] don't bother with path_get()/path_put() in unix_open_file()

[Al Viro]

Once unix_sock ->path is set, we are guaranteed that its ->path will remain
unchanged (and pinned) until the socket is closed.  OTOH, dentry_open()
does not modify the path passed to it.

IOW, there's no need to copy unix_sk(sk)->path in unix_open_file() - we
can just pass it to dentry_open() and be done with that.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 52b155123985..019ba2609b66 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -3191,7 +3191,6 @@ EXPORT_SYMBOL_GPL(unix_outq_len);
 
 static int unix_open_file(struct sock *sk)
 {
-	struct path path;
 	struct file *f;
 	int fd;
 
@@ -3201,27 +3200,20 @@ static int unix_open_file(struct sock *sk)
 	if (!smp_load_acquire(&unix_sk(sk)->addr))
 		return -ENOENT;
 
-	path = unix_sk(sk)->path;
-	if (!path.dentry)
+	if (!unix_sk(sk)->path.dentry)
 		return -ENOENT;
 
-	path_get(&path);
-
 	fd = get_unused_fd_flags(O_CLOEXEC);
 	if (fd < 0)
-		goto out;
+		return fd;
 
-	f = dentry_open(&path, O_PATH, current_cred());
+	f = dentry_open(&unix_sk(sk)->path, O_PATH, current_cred());
 	if (IS_ERR(f)) {
 		put_unused_fd(fd);
-		fd = PTR_ERR(f);
-		goto out;
+		return PTR_ERR(f);
 	}
 
 	fd_install(fd, f);
-out:
-	path_put(&path);
-
 	return fd;
 }
 

Re: [PATCH][RFC] don't bother with path_get()/path_put() in unix_open_file()

[Kuniyuki Iwashima]

From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 12 Jul 2025 06:41:57 +0100
> Once unix_sock ->path is set, we are guaranteed that its ->path will remain
> unchanged (and pinned) until the socket is closed.  OTOH, dentry_open()
> does not modify the path passed to it.
> 
> IOW, there's no need to copy unix_sk(sk)->path in unix_open_file() - we
> can just pass it to dentry_open() and be done with that.
> 
> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Sounds good.  I confirmed vfs_open() copies the passed const path ptr.

Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>


> diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
> index 52b155123985..019ba2609b66 100644
> --- a/net/unix/af_unix.c
> +++ b/net/unix/af_unix.c
> @@ -3191,7 +3191,6 @@ EXPORT_SYMBOL_GPL(unix_outq_len);
>  
>  static int unix_open_file(struct sock *sk)
>  {
> -	struct path path;
>  	struct file *f;
>  	int fd;
>  
> @@ -3201,27 +3200,20 @@ static int unix_open_file(struct sock *sk)
>  	if (!smp_load_acquire(&unix_sk(sk)->addr))
>  		return -ENOENT;
>  
> -	path = unix_sk(sk)->path;
> -	if (!path.dentry)
> +	if (!unix_sk(sk)->path.dentry)
>  		return -ENOENT;
>  
> -	path_get(&path);
> -
>  	fd = get_unused_fd_flags(O_CLOEXEC);
>  	if (fd < 0)
> -		goto out;
> +		return fd;
>  
> -	f = dentry_open(&path, O_PATH, current_cred());
> +	f = dentry_open(&unix_sk(sk)->path, O_PATH, current_cred());
>  	if (IS_ERR(f)) {
>  		put_unused_fd(fd);
> -		fd = PTR_ERR(f);
> -		goto out;
> +		return PTR_ERR(f);
>  	}
>  
>  	fd_install(fd, f);
> -out:
> -	path_put(&path);
> -
>  	return fd;
>  }

Re: [PATCH][RFC] don't bother with path_get()/path_put() in unix_open_file()

[Christian Brauner]

On Sat, Jul 12, 2025 at 06:38:33AM +0000, Kuniyuki Iwashima wrote:
> From: Al Viro <viro@zeniv.linux.org.uk>
> Date: Sat, 12 Jul 2025 06:41:57 +0100
> > Once unix_sock ->path is set, we are guaranteed that its ->path will remain
> > unchanged (and pinned) until the socket is closed.  OTOH, dentry_open()
> > does not modify the path passed to it.
> > 
> > IOW, there's no need to copy unix_sk(sk)->path in unix_open_file() - we
> > can just pass it to dentry_open() and be done with that.
> > 
> > Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> 
> Sounds good.  I confirmed vfs_open() copies the passed const path ptr.
> 
> Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>

I can just throw that into the SCM_PIDFD branch?

Re: [PATCH][RFC] don't bother with path_get()/path_put() in unix_open_file()

[Al Viro]

On Mon, Jul 14, 2025 at 10:24:11AM +0200, Christian Brauner wrote:
> On Sat, Jul 12, 2025 at 06:38:33AM +0000, Kuniyuki Iwashima wrote:
> > From: Al Viro <viro@zeniv.linux.org.uk>
> > Date: Sat, 12 Jul 2025 06:41:57 +0100
> > > Once unix_sock ->path is set, we are guaranteed that its ->path will remain
> > > unchanged (and pinned) until the socket is closed.  OTOH, dentry_open()
> > > does not modify the path passed to it.
> > > 
> > > IOW, there's no need to copy unix_sk(sk)->path in unix_open_file() - we
> > > can just pass it to dentry_open() and be done with that.
> > > 
> > > Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> > 
> > Sounds good.  I confirmed vfs_open() copies the passed const path ptr.
> > 
> > Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
> 
> I can just throw that into the SCM_PIDFD branch?

Fine by me; the thing is, I don't have anything else in the area at the moment
(and won't until -rc1 - CLASS(get_unused_fd) series will stray there, but
it's not settled enough yet, so it's definitely the next cycle fodder).

So if you (or netdev folks) already have anything going on in the af_unix.c,
I've no problem with that thing going there.

Re: [PATCH][RFC] don't bother with path_get()/path_put() in unix_open_file()

[Kuniyuki Iwashima]

On Mon, Jul 14, 2025 at 8:04 AM Al Viro <viro@zeniv.linux.org.uk> wrote:
>
> On Mon, Jul 14, 2025 at 10:24:11AM +0200, Christian Brauner wrote:
> > On Sat, Jul 12, 2025 at 06:38:33AM +0000, Kuniyuki Iwashima wrote:
> > > From: Al Viro <viro@zeniv.linux.org.uk>
> > > Date: Sat, 12 Jul 2025 06:41:57 +0100
> > > > Once unix_sock ->path is set, we are guaranteed that its ->path will remain
> > > > unchanged (and pinned) until the socket is closed.  OTOH, dentry_open()
> > > > does not modify the path passed to it.
> > > >
> > > > IOW, there's no need to copy unix_sk(sk)->path in unix_open_file() - we
> > > > can just pass it to dentry_open() and be done with that.
> > > >
> > > > Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> > >
> > > Sounds good.  I confirmed vfs_open() copies the passed const path ptr.
> > >
> > > Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
> >
> > I can just throw that into the SCM_PIDFD branch?
>
> Fine by me; the thing is, I don't have anything else in the area at the moment
> (and won't until -rc1 - CLASS(get_unused_fd) series will stray there, but
> it's not settled enough yet, so it's definitely the next cycle fodder).
>
> So if you (or netdev folks) already have anything going on in the af_unix.c,
> I've no problem with that thing going there.

AFAIK, there's no conflicting changes around unix_open_file() in
net-next, and this is more of vfs stuff, so whichever is fine to me.

Re: [PATCH][RFC] don't bother with path_get()/path_put() in unix_open_file()

[Christian Brauner]

On Sat, 12 Jul 2025 06:41:57 +0100, Al Viro wrote:
> Once unix_sock ->path is set, we are guaranteed that its ->path will remain
> unchanged (and pinned) until the socket is closed.  OTOH, dentry_open()
> does not modify the path passed to it.
> 
> IOW, there's no need to copy unix_sk(sk)->path in unix_open_file() - we
> can just pass it to dentry_open() and be done with that.
> 
> [...]

Applied to the vfs-6.17.pidfs branch of the vfs/vfs.git tree.
Patches in the vfs-6.17.pidfs branch should appear in linux-next soon.

Please report any outstanding bugs that were missed during review in a
new review to the original patch series allowing us to drop it.

It's encouraged to provide Acked-bys and Reviewed-bys even though the
patch has now been applied. If possible patch trailers will be updated.

Note that commit hashes shown below are subject to change due to rebase,
trailer updates or similar. If in doubt, please check the listed branch.

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git
branch: vfs-6.17.pidfs

[1/1] don't bother with path_get()/path_put() in unix_open_file()
      https://git.kernel.org/vfs/vfs/c/1f531e35c146