Re: [PATCH 06/32] xfs: don't assert fail on bad inode numbers

[Eric Sandeen <sandeen@xxxxxxxxxxx>]

On 9/29/13 10:15 PM, Dave Chinner wrote:
> From: Dave Chinner <dchinner@xxxxxxxxxx>
> 
> Let the inode verifier do it's work by returning an error when we
> fail to find correct magic numbers in an inode buffer.

Makes sense & matches kernel 74ffa796e127906883cacedcf3871494192c9e42

Reviewed-by: Eric Sandeen <sandeen@xxxxxxxxxx>

> Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
> ---
>  libxfs/xfs_inode_buf.c | 8 +++-----
>  1 file changed, 3 insertions(+), 5 deletions(-)
> 
> diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
> index 6205318..b096f77 100644
> --- a/libxfs/xfs_inode_buf.c
> +++ b/libxfs/xfs_inode_buf.c
> @@ -39,9 +39,8 @@ xfs_inobp_check(
>                                       i * mp->m_sb.sb_inodesize);
>               if (!dip->di_next_unlinked)  {
>                       xfs_alert(mp,
> -     "Detected bogus zero next_unlinked field in incore inode buffer 0x%p.",
> -                             bp);
> -                     ASSERT(dip->di_next_unlinked);
> +     "Detected bogus zero next_unlinked field in inode %d buffer 0x%llx.",
> +                             i, (long long)bp->b_bn);
>               }
>       }
>  }
> @@ -74,11 +73,10 @@ xfs_inode_buf_verify(
>                       XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_HIGH,
>                                            mp, dip);
>  #ifdef DEBUG
> -                     xfs_emerg(mp,
> +                     xfs_alert(mp,
>                               "bad inode magic/vsn daddr %lld #%d (magic=%x)",
>                               (unsigned long long)bp->b_bn, i,
>                               be16_to_cpu(dip->di_magic));
> -                     ASSERT(0);
>  #endif
>               }
>       }
>