[Cluster-devel] [PATCH GFS2] GFS2 fatal: filesystem consistency error on rename

[Steven Whitehouse]

Hi,

Now in the -nmw git tree. Thanks,

Steve.

On Thu, 2010-09-30 at 10:34 -0400, Bob Peterson wrote:
> Hi,
> 
> This patch fixes a GFS2 problem whereby the first rename after a
> mount can result in a file system consistency error being flagged
> improperly and cause the file system to withdraw.  The problem is
> that the rename code tries to run the rgrp list with function
> gfs2_blk2rgrpd before the rgrp list is guaranteed to be read in
> from disk.  The patch makes the rename function hold the rindex
> glock (as the gfs2_unlink code does today) which reads in the rgrp
> list if need be.  There were a total of three places in the rename
> code that improperly referenced the rgrp list without the rindex
> glock and this patch fixes all three.
> 
> Regards,
> 
> Bob Peterson
> Red Hat GFS
> 
> Signed-off-by: Bob Peterson <rpeterso at redhat.com> 
> --
>  fs/gfs2/ops_inode.c |    8 ++++++--
>  fs/gfs2/rgrp.c      |   22 +++++++++++++---------
>  fs/gfs2/rgrp.h      |    8 +++++---
>  3 files changed, 24 insertions(+), 14 deletions(-)
> 
> diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
> index fba0017..0534510 100644
> --- a/fs/gfs2/ops_inode.c
> +++ b/fs/gfs2/ops_inode.c
> @@ -736,7 +736,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
>  	struct gfs2_inode *ip = GFS2_I(odentry->d_inode);
>  	struct gfs2_inode *nip = NULL;
>  	struct gfs2_sbd *sdp = GFS2_SB(odir);
> -	struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, };
> +	struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh;
>  	struct gfs2_rgrpd *nrgd;
>  	unsigned int num_gh;
>  	int dir_rename = 0;
> @@ -750,6 +750,9 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
>  			return 0;
>  	}
>  
> +	error = gfs2_rindex_hold(sdp, &ri_gh);
> +	if (error)
> +		return error;
>  
>  	if (odip != ndip) {
>  		error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE,
> @@ -879,7 +882,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
>  
>  		al->al_requested = sdp->sd_max_dirres;
>  
> -		error = gfs2_inplace_reserve(ndip);
> +		error = gfs2_inplace_reserve_ri(ndip);
>  		if (error)
>  			goto out_gunlock_q;
>  
> @@ -961,6 +964,7 @@ out_gunlock_r:
>  	if (r_gh.gh_gl)
>  		gfs2_glock_dq_uninit(&r_gh);
>  out:
> +	gfs2_glock_dq_uninit(&ri_gh);
>  	return error;
>  }
>  
> diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
> index f9ddcf4..fb67f59 100644
> --- a/fs/gfs2/rgrp.c
> +++ b/fs/gfs2/rgrp.c
> @@ -1200,7 +1200,8 @@ out:
>   * Returns: errno
>   */
>  
> -int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file, unsigned int line)
> +int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex,
> +			   char *file, unsigned int line)
>  {
>  	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
>  	struct gfs2_alloc *al = ip->i_alloc;
> @@ -1211,12 +1212,15 @@ int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file, unsigned int line)
>  		return -EINVAL;
>  
>  try_again:
> -	/* We need to hold the rindex unless the inode we're using is
> -	   the rindex itself, in which case it's already held. */
> -	if (ip != GFS2_I(sdp->sd_rindex))
> -		error = gfs2_rindex_hold(sdp, &al->al_ri_gh);
> -	else if (!sdp->sd_rgrps) /* We may not have the rindex read in, so: */
> -		error = gfs2_ri_update_special(ip);
> +	if (hold_rindex) {
> +		/* We need to hold the rindex unless the inode we're using is
> +		   the rindex itself, in which case it's already held. */
> +		if (ip != GFS2_I(sdp->sd_rindex))
> +			error = gfs2_rindex_hold(sdp, &al->al_ri_gh);
> +		else if (!sdp->sd_rgrps) /* We may not have the rindex read
> +					    in, so: */
> +			error = gfs2_ri_update_special(ip);
> +	}
>  
>  	if (error)
>  		return error;
> @@ -1227,7 +1231,7 @@ try_again:
>  	   try to free it, and try the allocation again. */
>  	error = get_local_rgrp(ip, &unlinked, &last_unlinked);
>  	if (error) {
> -		if (ip != GFS2_I(sdp->sd_rindex))
> +		if (hold_rindex && ip != GFS2_I(sdp->sd_rindex))
>  			gfs2_glock_dq_uninit(&al->al_ri_gh);
>  		if (error != -EAGAIN)
>  			return error;
> @@ -1269,7 +1273,7 @@ void gfs2_inplace_release(struct gfs2_inode *ip)
>  	al->al_rgd = NULL;
>  	if (al->al_rgd_gh.gh_gl)
>  		gfs2_glock_dq_uninit(&al->al_rgd_gh);
> -	if (ip != GFS2_I(sdp->sd_rindex))
> +	if (ip != GFS2_I(sdp->sd_rindex) && al->al_ri_gh.gh_gl)
>  		gfs2_glock_dq_uninit(&al->al_ri_gh);
>  }
>  
> diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h
> index f07119d..0e35c04 100644
> --- a/fs/gfs2/rgrp.h
> +++ b/fs/gfs2/rgrp.h
> @@ -39,10 +39,12 @@ static inline void gfs2_alloc_put(struct gfs2_inode *ip)
>  	ip->i_alloc = NULL;
>  }
>  
> -extern int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file,
> -				  unsigned int line);
> +extern int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex,
> +				  char *file, unsigned int line);
>  #define gfs2_inplace_reserve(ip) \
> -gfs2_inplace_reserve_i((ip), __FILE__, __LINE__)
> +	gfs2_inplace_reserve_i((ip), 1, __FILE__, __LINE__)
> +#define gfs2_inplace_reserve_ri(ip) \
> +	gfs2_inplace_reserve_i((ip), 0, __FILE__, __LINE__)
>  
>  extern void gfs2_inplace_release(struct gfs2_inode *ip);
>  
>