On Sun 13-01-13 18:44:55, Wang shilong wrote:
> From: Wang Shilong <wangsl-fnst@xxxxxxxxxxxxxx>
>
> It will be better to have ENOMEM return rather than EIO,because
> the only reason that sb_getblk fails is allocation fails.
Again one case is wrong. See below.
>
> Signed-off-by: Wang Shilong <wangsl-fnst@xxxxxxxxxxxxxx>
> ---
> fs/ext3/inode.c | 6 +++---
> fs/ext3/resize.c | 6 +++---
> fs/ext3/xattr.c | 2 +-
> 3 files changed, 7 insertions(+), 7 deletions(-)
>
> diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
> index c91f8bf..d5315d5 100644
> --- a/fs/ext3/inode.c
> +++ b/fs/ext3/inode.c
> @@ -1084,7 +1084,7 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode,
> struct buffer_head *bh;
> bh = sb_getblk(inode->i_sb, dummy.b_blocknr);
> if (unlikely(!bh)) {
> - *errp = -EIO;
> + *errp = -ENOMEM;
> goto err;
> }
> if (buffer_new(&dummy)) {
> @@ -2739,7 +2739,7 @@ static int __ext3_get_inode_loc(struct inode *inode,
> "unable to read inode block - "
> "inode=%lu, block="E3FSBLK,
> inode->i_ino, block);
> - return -EIO;
> + return -ENOMEM;
> }
> if (!buffer_uptodate(bh)) {
> lock_buffer(bh);
> @@ -2833,7 +2833,7 @@ make_io:
> "inode=%lu, block="E3FSBLK,
> inode->i_ino, block);
> brelse(bh);
> - return -EIO;
> + return -ENOMEM;
This is wrong. Here we really failed because of IO error (buffer was not
uptodate).
Honza
--
Jan Kara <jack@xxxxxxx>
SUSE Labs, CR
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
