xfs
[Top] [All Lists]

Re: [PATCH 2/2] xfs_repair: Properly set lost+found inode as used

To: Carlos Maiolino <cmaiolino@xxxxxxxxxx>
Subject: Re: [PATCH 2/2] xfs_repair: Properly set lost+found inode as used
From: Eric Sandeen <sandeen@xxxxxxxxxxx>
Date: Tue, 08 Nov 2011 12:55:54 -0600
Cc: xfs@xxxxxxxxxxx
In-reply-to: <1320777990-6556-3-git-send-email-cmaiolino@xxxxxxxxxx>
References: <1320777990-6556-1-git-send-email-cmaiolino@xxxxxxxxxx> <1320777990-6556-3-git-send-email-cmaiolino@xxxxxxxxxx>
User-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1
On 11/8/11 12:46 PM, Carlos Maiolino wrote:
> This patch makes mk_orphanage() to properly set the inode link count of
> the recently allocated inode in the AVL tree, avoiding the lost+found
> directory to be bypass the link count check in phase7 and possibly leaving
> lost+found directory with a wrong link count.

Looks pretty good to me.  I think maybe add_inode_reached shouldn't
be called until the transaction to do it is complete, though?
Otherwise that might fail and we'd still have marked it as reached;
I think that could be a problem.

Thanks,
-Eric

> ---
>  repair/phase6.c |   17 ++++++++++++++++-
>  1 files changed, 16 insertions(+), 1 deletions(-)
> 
> diff --git a/repair/phase6.c b/repair/phase6.c
> index adad61d..0e0e294 100644
> --- a/repair/phase6.c
> +++ b/repair/phase6.c
> @@ -823,6 +823,8 @@ mk_orphanage(xfs_mount_t *mp)
>       xfs_inode_t     *ip;
>       xfs_inode_t     *pip;
>       xfs_fsblock_t   first;
> +     ino_tree_node_t *irec;
> +     int             ino_offset = 0;
>       int             i;
>       int             committed;
>       int             error;
> @@ -875,6 +877,18 @@ mk_orphanage(xfs_mount_t *mp)
>                       ORPHANAGE, error);
>       }
>       ip->i_d.di_nlink++;             /* account for . */
> +     ino = ip->i_ino;
> +
> +     irec = find_inode_rec(mp,
> +                     XFS_INO_TO_AGNO(mp, ino),
> +                     XFS_INO_TO_AGINO(mp,ino));
> +     ino_offset = get_inode_offset(mp, ino, irec);
> +
> +     /* Set the inode allocated to lost+found as used in the AVL
> +      * tree, so it is not bypassed in phase 7
> +      */
> +     set_inode_used(irec,ino_offset);
> +     add_inode_ref(irec,ino_offset);
>  
>       /*
>        * now that we know the transaction will stay around,
> @@ -902,6 +916,8 @@ mk_orphanage(xfs_mount_t *mp)
>                               XFS_INO_TO_AGINO(mp, mp->m_sb.sb_rootino)), 0);
>  
>  
> +     add_inode_reached(irec,ino_offset);
> +
>       libxfs_trans_log_inode(tp, pip, XFS_ILOG_CORE);
>       libxfs_dir_init(tp, ip, pip);
>       libxfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
> @@ -912,7 +928,6 @@ mk_orphanage(xfs_mount_t *mp)
>                       ORPHANAGE, error);
>       }
>  
> -     ino = ip->i_ino;
>  
>       libxfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_SYNC);
>  

<Prev in Thread] Current Thread [Next in Thread>