[Ocfs2-devel] [PATCH 1/2] ocfs2: return ENOMEM while sb_getblk failing

[Mark Fasheh]

On Mon, Sep 02, 2013 at 07:15:49PM +0800, Rui Xiang wrote:
> The only reason for sb_getblk() failing is if it can't allocate
> the buffer_head. So return ENOMEM instead when it fails.
> 
> Signed-off-by: Rui Xiang <rui.xiang at huawei.com>
> ---
>  fs/ocfs2/alloc.c        | 2 +-
>  fs/ocfs2/aops.c         | 1 +
>  fs/ocfs2/dir.c          | 8 ++++----
>  fs/ocfs2/namei.c        | 2 +-
>  fs/ocfs2/refcounttree.c | 6 +++---
>  fs/ocfs2/suballoc.c     | 4 ++--
>  fs/ocfs2/super.c        | 4 ++--
>  fs/ocfs2/xattr.c        | 2 +-
>  8 files changed, 15 insertions(+), 14 deletions(-)
> 
> diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
> index 17e6bdd..dc7411f 100644
> --- a/fs/ocfs2/alloc.c
> +++ b/fs/ocfs2/alloc.c
> @@ -1025,7 +1025,7 @@ static int ocfs2_create_new_meta_bhs(handle_t *handle,
>  		for(i = count;  i < (num_got + count); i++) {
>  			bhs[i] = sb_getblk(osb->sb, first_blkno);
>  			if (bhs[i] == NULL) {
> -				status = -EIO;
> +				status = -ENOMEM;
>  				mlog_errno(status);
>  				goto bail;
>  			}
> diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
> index 2abf97b..ffcddfd 100644
> --- a/fs/ocfs2/aops.c
> +++ b/fs/ocfs2/aops.c
> @@ -80,6 +80,7 @@ static int ocfs2_symlink_get_block(struct inode *inode, sector_t iblock,
>  
>  	if ((u64)iblock >= ocfs2_clusters_to_blocks(inode->i_sb,
>  						    le32_to_cpu(fe->i_clusters))) {
> +		err = -ENOMEM;
>  		mlog(ML_ERROR, "block offset is outside the allocated size: "
>  		     "%llu\n", (unsigned long long)iblock);
>  		goto bail;

This chunk looks out of place in this patch - it would be best to resend
without it. Otherwise the patch looks pretty good, thanks.
	--Mark

--
Mark Fasheh