From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id oAT1BR7A076997 for ; Sun, 28 Nov 2010 19:11:27 -0600 Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id A6BE313F659E for ; Sun, 28 Nov 2010 17:13:07 -0800 (PST) Received: from mail.internode.on.net (bld-mail14.adl6.internode.on.net [150.101.137.99]) by cuda.sgi.com with ESMTP id QZNMGPf4BMQeQksD for ; Sun, 28 Nov 2010 17:13:07 -0800 (PST) Received: from dastard (unverified [121.44.88.148]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 48418095-1927428 for ; Mon, 29 Nov 2010 11:43:06 +1030 (CDT) Received: from chute ([192.168.1.1] helo=disappointment) by dastard with esmtp (Exim 4.71) (envelope-from ) id 1PMsIY-0000aj-Tv for xfs@oss.sgi.com; Mon, 29 Nov 2010 12:12:54 +1100 Received: from dave by disappointment with local (Exim 4.72) (envelope-from ) id 1PMsIE-0005Uv-Gd for xfs@oss.sgi.com; Mon, 29 Nov 2010 12:12:34 +1100 From: Dave Chinner Subject: [PATCH 5/8] xfs: consume iodone callback items on buffers as they are processed Date: Mon, 29 Nov 2010 12:12:29 +1100 Message-Id: <1290993152-20999-6-git-send-email-david@fromorbit.com> In-Reply-To: <1290993152-20999-1-git-send-email-david@fromorbit.com> References: <1290993152-20999-1-git-send-email-david@fromorbit.com> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com From: Dave Chinner To allow buffer iodone callbacks to consume multiple items off the callback list, first we need to convert the xfs_buf_do_callbacks() to consume items and always pull the next item from the head of the list. The means the item list walk is never dependent on knowing the next item on the list and hence allows callbacks to remove itesm from the list as well. This allows callbacks to do bulk operations by scanning the list for identical callbacks, consuming them all and then processing them in bulk, negating the need for multiple callbacks of that type. Signed-off-by: Dave Chinner --- fs/xfs/xfs_buf_item.c | 32 +++++++++++++++++++++----------- 1 files changed, 21 insertions(+), 11 deletions(-) diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 2686d0d..ed2b65f 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -142,7 +142,7 @@ xfs_buf_item_log_check( #endif STATIC void xfs_buf_error_relse(xfs_buf_t *bp); -STATIC void xfs_buf_do_callbacks(xfs_buf_t *bp, xfs_log_item_t *lip); +STATIC void xfs_buf_do_callbacks(struct xfs_buf *bp); /* * This returns the number of log iovecs needed to log the @@ -450,7 +450,7 @@ xfs_buf_item_unpin( * xfs_trans_ail_delete() drops the AIL lock. */ if (bip->bli_flags & XFS_BLI_STALE_INODE) { - xfs_buf_do_callbacks(bp, (xfs_log_item_t *)bip); + xfs_buf_do_callbacks(bp); XFS_BUF_SET_FSPRIVATE(bp, NULL); XFS_BUF_CLR_IODONE_FUNC(bp); } else { @@ -918,15 +918,26 @@ xfs_buf_attach_iodone( XFS_BUF_SET_IODONE_FUNC(bp, xfs_buf_iodone_callbacks); } +/* + * We can have many callbacks on a buffer. Running the callbacks individually + * can cause a lot of contention on the AIL lock, so we allow for a single + * callback to be able to scan the remaining lip->li_bio_list for other items + * of the same type and callback to be processed in the first call. + * + * As a result, the loop walking the callback list below will also modify the + * list. it removes the first item from the list and then runs the callback. + * The loop then restarts from the new head of the list. This allows the + * callback to scan and modify the list attached to the buffer and we don't + * have to care about maintaining a next item pointer. + */ STATIC void xfs_buf_do_callbacks( - xfs_buf_t *bp, - xfs_log_item_t *lip) + struct xfs_buf *bp) { - xfs_log_item_t *nlip; + struct xfs_log_item *lip; - while (lip != NULL) { - nlip = lip->li_bio_list; + while ((lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *)) != NULL) { + XFS_BUF_SET_FSPRIVATE(bp, lip->li_bio_list); ASSERT(lip->li_cb != NULL); /* * Clear the next pointer so we don't have any @@ -936,7 +947,6 @@ xfs_buf_do_callbacks( */ lip->li_bio_list = NULL; lip->li_cb(bp, lip); - lip = nlip; } } @@ -970,7 +980,7 @@ xfs_buf_iodone_callbacks( ASSERT(XFS_BUF_TARGET(bp) == mp->m_ddev_targp); XFS_BUF_SUPER_STALE(bp); trace_xfs_buf_item_iodone(bp, _RET_IP_); - xfs_buf_do_callbacks(bp, lip); + xfs_buf_do_callbacks(bp); XFS_BUF_SET_FSPRIVATE(bp, NULL); XFS_BUF_CLR_IODONE_FUNC(bp); xfs_buf_ioend(bp, 0); @@ -1029,7 +1039,7 @@ xfs_buf_iodone_callbacks( return; } - xfs_buf_do_callbacks(bp, lip); + xfs_buf_do_callbacks(bp); XFS_BUF_SET_FSPRIVATE(bp, NULL); XFS_BUF_CLR_IODONE_FUNC(bp); xfs_buf_ioend(bp, 0); @@ -1063,7 +1073,7 @@ xfs_buf_error_relse( * We have to unpin the pinned buffers so do the * callbacks. */ - xfs_buf_do_callbacks(bp, lip); + xfs_buf_do_callbacks(bp); XFS_BUF_SET_FSPRIVATE(bp, NULL); XFS_BUF_CLR_IODONE_FUNC(bp); XFS_BUF_SET_BRELSE_FUNC(bp,NULL); -- 1.7.2.3 _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs