Re: xfs: support inode btree blockcounts in online repair

["Darrick J. Wong" <darrick.wong@oracle.com>]

On Fri, Sep 11, 2020 at 12:16:33PM +0100, Colin Ian King wrote:
> Hi,
> 
> Static analysis with Coverity has detected an issue with the following
> commit:
> 
> commit 30deae31eab501f568aadea45cfb3258b9e522f5
> Author: Darrick J. Wong <darrick.wong@oracle.com>
> Date:   Wed Aug 26 10:48:50 2020 -0700
> 
>     xfs: support inode btree blockcounts in online repair
> 
> the analysis is as follows:
> 
> 830                cur = xfs_inobt_init_cursor(mp, sc->tp, agi_bp,
> sc->sa.agno,
> 831                                XFS_BTNUM_FINO);

xfs_inobt_init_cursor can't return NULL (because we guarantee the
allocation cannot fail) so all that really needs to be done here is to
remove the pointless error check in lines 832-833.

--D

> const: At condition error, the value of error must be equal to 0.
> dead_error_condition: The condition error cannot be true.
> 
>  832                if (error)
> 
> CID: Logically dead code (DEADCODE)dead_error_line: Execution cannot
> reach this statement: goto err;.
> 
>  833                        goto err;
> 
> While it is tempting to change the if (error) check to if (cur), the
> exit error path uses the errnoeous cur as follows:
> 
>  842 err:
>  843       xfs_btree_del_cursor(cur, error);
>  844       return error;
> 
> so the error exit path needs some sorting out too.