Re: [PATCH 6.1 000/600] 6.1.53-rc1 review

[Naresh Kamboju]

Hi Greg,

> > > > > For what it's worth, bisection points to 903b487b5ba6 ("udf: Handle
> > > > > error when adding extent to a file").
> > > >
> > > > I see the following commit is fixing the reported problem.
> > > >
> > > > commit 6d5ab7c2f7cf90877dab8f2bb06eb5ca8edc73ef
> > > > Author: Tom Rix <trix@xxxxxxxxxx>
> > > > Date:   Fri Dec 30 12:53:41 2022 -0500
> > > >
> > > >     udf: initialize newblock to 0
> > > >
> > > >     The clang build reports this error
> > > >     fs/udf/inode.c:805:6: error: variable 'newblock' is used
> > > > uninitialized whenever 'if' condition is true
> > > > [-Werror,-Wsometimes-uninitialized]
> > > >             if (*err < 0)
> > > >                 ^~~~~~~~
> > > >     newblock is never set before error handling jump.
> > > >     Initialize newblock to 0 and remove redundant settings.
> > > >
> > > >     Fixes: d8b39db5fab8 ("udf: Handle error when adding extent to a file")
> > > >     Reported-by: Nathan Chancellor <nathan@xxxxxxxxxx>
> > > >     Signed-off-by: Tom Rix <trix@xxxxxxxxxx>
> > > >     Signed-off-by: Jan Kara <jack@xxxxxxx>
> > > >     Message-Id: <20221230175341.1629734-1-trix@xxxxxxxxxx>
> > >
> > > Wait, where is this commit?  I don't see it in Linus's tree either, nor
> > > in linux-next.  Where did you find it?
> >
> > Can you find this commit id ?
> >
> > Commit id: 23970a1c9475b305770fd37bebfec7a10f263787
> > subject: ("udf: initialize newblock to 0")
>
> Yes, that is in 6.2.  Where did the id you used above come from?

While looking around for the fix commit i found this on Linux next,

$ git log --oneline next-20221226..next-20230106  -- fs/udf/inode.c
e86812bfac97e udf: Detect system inodes linked into directory hierarchy
453bc25de0a55 udf: Preserve link count of system files
6d5ab7c2f7cf9 udf: initialize newblock to 0

>
> confused,

Sorry for the confusion.

> greg k-h

- Naresh