Re: [PATCH] xfs: change some error-less functions to void types

[Dave Chinner <david@fromorbit.com>]

On Wed, May 01, 2019 at 10:55:54AM -0500, Eric Sandeen wrote:
> There are several functions which have no opportunity to retun
> an error, and don't contain any ASSERTs which could be argued
> to be better constructed as error cases.  So, make them voids
> to simplify the callers.
> 
> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
> ---
....
> @@ -1840,7 +1840,6 @@ xlog_sync(
>  	uint		count_init;	/* initial count before roundup */
>  	int		roundoff;       /* roundoff to BB or stripe */
>  	int		split = 0;	/* split write into two regions */
> -	int		error;
>  	int		v2 = xfs_sb_version_haslogv2(&log->l_mp->m_sb);
>  	int		size;
>  
> @@ -1959,11 +1958,8 @@ xlog_sync(
>  	 * Don't call xfs_bwrite here. We do log-syncs even when the filesystem
>  	 * is shutting down.
>  	 */
> -	error = xlog_bdstrat(bp);
> -	if (error) {
> -		xfs_buf_ioerror_alert(bp, "xlog_sync");
> -		return error;
> -	}
> +	xlog_bdstrat(bp);

Shouldn't this be checking bp->b_error rather than completely
ignoring buffer submit errors?

> +
>  	if (split) {
>  		bp = iclog->ic_log->l_xbuf;
>  		XFS_BUF_SET_ADDR(bp, 0);	     /* logical 0 */
> @@ -1978,11 +1974,7 @@ xlog_sync(
>  
>  		/* account for internal log which doesn't start at block #0 */
>  		XFS_BUF_SET_ADDR(bp, XFS_BUF_ADDR(bp) + log->l_logBBstart);
> -		error = xlog_bdstrat(bp);
> -		if (error) {
> -			xfs_buf_ioerror_alert(bp, "xlog_sync (split)");
> -			return error;
> -		}
> +		xlog_bdstrat(bp);

Ditto.

Otherwise it looks fine.

-Dave.
-- 
Dave Chinner
david@fromorbit.com