Because 'inode' is being initialised before checking if 'dentry' is negative
it looks like an extra iput() on 'inode' may happen since the ihold() is
done only if the dentry is *not* negative.  In reality this doesn't happen
because d_is_negative() is never true if ->d_inode is NULL.  This patch only
makes the code easier to understand, as I was initially mislead by it.

Fixes: b18825a7c8e3 ("VFS: Put a small type field into struct dentry::d_flags")
Signed-off-by: Luís Henriques <lhenriques@suse.de>
---
Changes since v1:
Rephrased commit message to make it clear there isn't a real bug

 fs/namei.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index 567ee547492b..156a570d7831 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -4386,11 +4386,9 @@ int do_unlinkat(int dfd, struct filename *name)
 	if (!IS_ERR(dentry)) {
 
 		/* Why not before? Because we want correct error value */
-		if (last.name[last.len])
+		if (last.name[last.len] || d_is_negative(dentry))
 			goto slashes;
 		inode = dentry->d_inode;
-		if (d_is_negative(dentry))
-			goto slashes;
 		ihold(inode);
 		error = security_path_unlink(&path, dentry);
 		if (error)

On Thu, 28 Sep 2023 16:23:41 +0100, Luís Henriques wrote:
> Because 'inode' is being initialised before checking if 'dentry' is negative
> it looks like an extra iput() on 'inode' may happen since the ihold() is
> done only if the dentry is *not* negative.  In reality this doesn't happen
> because d_is_negative() is never true if ->d_inode is NULL.  This patch only
> makes the code easier to understand, as I was initially mislead by it.
> 
> 
> [...]

Applied to the vfs.misc branch of the vfs/vfs.git tree.
Patches in the vfs.misc 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.misc

[1/1] fs: simplify misleading code to remove ambiguity regarding ihold()/iput()
      https://git.kernel.org/vfs/vfs/c/5c29bcfaa4cf

Hi Christian,

Christian Brauner <brauner@kernel.org> writes:

> On Thu, 28 Sep 2023 16:23:41 +0100, Luís Henriques wrote:
>> Because 'inode' is being initialised before checking if 'dentry' is negative
>> it looks like an extra iput() on 'inode' may happen since the ihold() is
>> done only if the dentry is *not* negative.  In reality this doesn't happen
>> because d_is_negative() is never true if ->d_inode is NULL.  This patch only
>> makes the code easier to understand, as I was initially mislead by it.
>> 
>> 
>> [...]
>
> Applied to the vfs.misc branch of the vfs/vfs.git tree.
> Patches in the vfs.misc 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.misc
>
> [1/1] fs: simplify misleading code to remove ambiguity regarding ihold()/iput()
>       https://git.kernel.org/vfs/vfs/c/5c29bcfaa4cf

Could you please double-check this was indeed applied?  I can't see it
anywhere.  Maybe I'm looking at the wrong place, but since your scripts
seem to have messed-up my email address, something else may have went
wrong.

Cheers,
-- 
Luís

> Could you please double-check this was indeed applied?  I can't see it
> anywhere.  Maybe I'm looking at the wrong place, but since your scripts
> seem to have messed-up my email address, something else may have went
> wrong.

It was applied it's just not pushed out yet because of another patch
discussion. It should show up in the next 30 minutes though.

Christian Brauner <brauner@kernel.org> writes:

>> Could you please double-check this was indeed applied?  I can't see it
>> anywhere.  Maybe I'm looking at the wrong place, but since your scripts
>> seem to have messed-up my email address, something else may have went
>> wrong.
>
> It was applied it's just not pushed out yet because of another patch
> discussion. It should show up in the next 30 minutes though.

Awesome, thanks for confirming!  I guess I should have waited a bit longer
before asking.

Cheers,
-- 
Luís