[patch 5/11] btrfs: remove unneeded null check in btrfs_rename()

[patch 5/11] btrfs: remove unneeded null check in btrfs_rename()

[Dan Carpenter]

"old_inode" cannot be null here, because we dereference it
unconditionally throughout the function.

Signed-off-by: Dan Carpenter <error27@gmail.com>

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index fa6ccc1..0bc29be 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6487,10 +6487,8 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	 * make sure the inode gets flushed if it is replacing
 	 * something.
 	 */
-	if (new_inode && new_inode->i_size &&
-	    old_inode && S_ISREG(old_inode->i_mode)) {
+	if (new_inode && new_inode->i_size && S_ISREG(old_inode->i_mode))
 		btrfs_add_ordered_operation(trans, root, old_inode);
-	}
 
 	old_dir->i_ctime = old_dir->i_mtime = ctime;
 	new_dir->i_ctime = new_dir->i_mtime = ctime;

Re: [patch 5/11] btrfs: remove unneeded null check in btrfs_rename()

[Mike Fedyk]

On Sat, May 29, 2010 at 2:45 AM, Dan Carpenter <error27@gmail.com> wrot=
e:
> "old_inode" cannot be null here, because we dereference it
> unconditionally throughout the function.
>
> Signed-off-by: Dan Carpenter <error27@gmail.com>
>
> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
> index fa6ccc1..0bc29be 100644
> --- a/fs/btrfs/inode.c
> +++ b/fs/btrfs/inode.c
> @@ -6487,10 +6487,8 @@ static int btrfs_rename(struct inode *old_dir,=
 struct dentry *old_dentry,
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 * make sure the inode gets flushed if it =
is replacing
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 * something.
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 */
> - =C2=A0 =C2=A0 =C2=A0 if (new_inode && new_inode->i_size &&
> - =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 old_inode && S_ISREG(old_inode->=
i_mode)) {
> + =C2=A0 =C2=A0 =C2=A0 if (new_inode && new_inode->i_size && S_ISREG(=
old_inode->i_mode))
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0btrfs_add_orde=
red_operation(trans, root, old_inode);
> - =C2=A0 =C2=A0 =C2=A0 }

I think code like this is here because there are still a lot of
features that are being added to btrfs and it's easier to have the
additional checks than continually adding and removing them as the
code changes.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [patch 5/11] btrfs: remove unneeded null check in btrfs_rename()

[Al Viro]

On Sat, May 29, 2010 at 11:01:56AM -0700, Mike Fedyk wrote:

> I think code like this is here because there are still a lot of
> features that are being added to btrfs and it's easier to have the
> additional checks than continually adding and removing them as the
> code changes.

_What_ features?  Check for ->rename() argument not being negative is
done outside of btrfs.

Re: [patch 5/11] btrfs: remove unneeded null check in btrfs_rename()

[Dan Carpenter]

On Sat, May 29, 2010 at 11:01:56AM -0700, Mike Fedyk wrote:
> On Sat, May 29, 2010 at 2:45 AM, Dan Carpenter <error27@gmail.com> wr=
ote:
> > "old_inode" cannot be null here, because we dereference it
> > unconditionally throughout the function.
> >
> > Signed-off-by: Dan Carpenter <error27@gmail.com>
> >
> > diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
> > index fa6ccc1..0bc29be 100644
> > --- a/fs/btrfs/inode.c
> > +++ b/fs/btrfs/inode.c
> > @@ -6487,10 +6487,8 @@ static int btrfs_rename(struct inode *old_di=
r, struct dentry *old_dentry,
> > =A0 =A0 =A0 =A0 * make sure the inode gets flushed if it is replaci=
ng
> > =A0 =A0 =A0 =A0 * something.
> > =A0 =A0 =A0 =A0 */
> > - =A0 =A0 =A0 if (new_inode && new_inode->i_size &&
> > - =A0 =A0 =A0 =A0 =A0 old_inode && S_ISREG(old_inode->i_mode)) {
> > + =A0 =A0 =A0 if (new_inode && new_inode->i_size && S_ISREG(old_ino=
de->i_mode))
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0btrfs_add_ordered_operation(trans, r=
oot, old_inode);
> > - =A0 =A0 =A0 }
>=20
> I think code like this is here because there are still a lot of
> features that are being added to btrfs and it's easier to have the
> additional checks than continually adding and removing them as the
> code changes.

Right right.  I understand about extra checks and api changes.  But in
this case that doesn't aply.  There is no way this particular check hel=
ps
now or in the future, and it just wastes time when someone is auditing
for inconsistent null checking.

regards,
dan carpenter
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html