Hi Heiko, On Wed, May 13, 2026 at 7:50 PM Heiko Schocher <[email protected]> wrote: > > Hello Peter, > > On 12.05.26 07:50, Peter Collingbourne wrote: > > This code is dead because it appears after an infinite loop; remove it. > > > > Fixes: 4c4477c77008 ("fs: ubifs: fix bugs involving symlinks in > > ubifs_findfile") > > Wrong commit ID? I find in current HEAD: > """ > commit d5888d509cc43942ec98d993f2d129f5c8ddc432 > Author: Peter Collingbourne <[email protected]> > Date: Tue May 5 03:38:16 2026 -0700 > > fs: ubifs: fix bugs involving symlinks in ubifs_findfile > """
Yeah, I see the same. Looks like I accidentally used my original commit ID instead of the one that was pushed. > > Signed-off-by: Peter Collingbourne <[email protected]> > > --- > > fs/ubifs/ubifs.c | 3 --- > > 1 file changed, 3 deletions(-) > > > > diff --git a/fs/ubifs/ubifs.c b/fs/ubifs/ubifs.c > > index aafbd01a028..3f2e2037745 100644 > > --- a/fs/ubifs/ubifs.c > > +++ b/fs/ubifs/ubifs.c > > @@ -608,9 +608,6 @@ static unsigned long ubifs_findfile(struct super_block > > *sb, char *filename) > > root_inum = inum; > > name = next; > > } > > - > > - kfree(buf); > > - return 0; > > May we need to stay with "return 0;" at the end of the function, as > some source code checking tool will claim if there is none? I'd probably remove it unless some tool complains. If there's code at the end it can mislead readers into thinking it's reachable, which is how I ended up adding the kfree() there in the first place. > Beside of the nitpick: > > Reviewed-by: Heiko Schocher <[email protected]> > > If you find time to send a v2 with updated commit ID I would be happy, > else I can fix it when picking up the patch, if okay for you. Feel free to fix it up, thanks. Peter
