From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Mon, 21 Jan 2008 19:58:12 -0800 (PST) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m0M3w2Nw002054 for ; Mon, 21 Jan 2008 19:58:06 -0800 Date: Tue, 22 Jan 2008 14:58:17 +1100 From: David Chinner Subject: Re: [patch] reduce log I/O latency V2 Message-ID: <20080122035817.GT155259@sgi.com> References: <20080121041649.GA155259@sgi.com> <47956615.90700@sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <47956615.90700@sgi.com> Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: Lachlan McIlroy Cc: David Chinner , xfs-dev , xfs-oss On Tue, Jan 22, 2008 at 02:42:13PM +1100, Lachlan McIlroy wrote: > Looks good Dave. Shouldn't we also tag the second buffer in the > split log write case? Hmm - I thought it was being tagged. looks like I missed a reject on my last set of patch re-orderings - updated patch attached. --- Reduce log I/O latency To ensure that log I/O is issued as the highest priority I/O, set the I/O priority of the log I/O to the highest possible. This will ensure that log I/O is not held up behind bulk data or other metadata I/O as delaying log I/O can pause the entire transaction subsystem. Introduce a new buffer flag to allow us to tag the log buffers so we can discriminate when issuing the I/O. Signed-off-by: Dave Chinner --- fs/xfs/linux-2.6/xfs_buf.c | 3 +++ fs/xfs/linux-2.6/xfs_buf.h | 5 ++++- fs/xfs/xfs_log.c | 3 +++ 3 files changed, 10 insertions(+), 1 deletion(-) Index: 2.6.x-xfs-new/fs/xfs/linux-2.6/xfs_buf.c =================================================================== --- 2.6.x-xfs-new.orig/fs/xfs/linux-2.6/xfs_buf.c 2008-01-18 19:13:07.499131222 +1100 +++ 2.6.x-xfs-new/fs/xfs/linux-2.6/xfs_buf.c 2008-01-22 14:56:58.738098626 +1100 @@ -1198,6 +1198,9 @@ next_chunk: submit_io: if (likely(bio->bi_size)) { + /* log I/O should not be delayed by anything except realtime. */ + if (bp->b_flags & XBF_LOG_BUFFER) + bio_set_prio(bio, IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 0)); submit_bio(rw, bio); if (size) goto next_chunk; Index: 2.6.x-xfs-new/fs/xfs/linux-2.6/xfs_buf.h =================================================================== --- 2.6.x-xfs-new.orig/fs/xfs/linux-2.6/xfs_buf.h 2008-01-18 19:13:07.503130710 +1100 +++ 2.6.x-xfs-new/fs/xfs/linux-2.6/xfs_buf.h 2008-01-18 19:13:29.448320166 +1100 @@ -53,7 +53,8 @@ typedef enum { XBF_DELWRI = (1 << 6), /* buffer has dirty pages */ XBF_STALE = (1 << 7), /* buffer has been staled, do not find it */ XBF_FS_MANAGED = (1 << 8), /* filesystem controls freeing memory */ - XBF_ORDERED = (1 << 11), /* use ordered writes */ + XBF_LOG_BUFFER = (1 << 9), /* Buffer issued by the log */ + XBF_ORDERED = (1 << 11), /* use ordered writes */ XBF_READ_AHEAD = (1 << 12), /* asynchronous read-ahead */ /* flags used only as arguments to access routines */ @@ -339,6 +340,8 @@ extern void xfs_buf_trace(xfs_buf_t *, c #define XFS_BUF_TARGET(bp) ((bp)->b_target) #define XFS_BUFTARG_NAME(target) xfs_buf_target_name(target) +#define XFS_BUF_SET_LOGBUF(bp) ((bp)->b_flags |= XBF_LOG_BUFFER) + static inline int xfs_bawrite(void *mp, xfs_buf_t *bp) { bp->b_fspriv3 = mp; Index: 2.6.x-xfs-new/fs/xfs/xfs_log.c =================================================================== --- 2.6.x-xfs-new.orig/fs/xfs/xfs_log.c 2008-01-18 19:13:07.519128661 +1100 +++ 2.6.x-xfs-new/fs/xfs/xfs_log.c 2008-01-22 14:56:57.550673091 +1100 @@ -1462,6 +1462,8 @@ xlog_sync(xlog_t *log, XFS_BUF_ZEROFLAGS(bp); XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); + XFS_BUF_SET_LOGBUF(bp); + /* * Do an ordered write for the log block. * Its unnecessary to flush the first split block in the log wrap case. @@ -1499,6 +1501,7 @@ xlog_sync(xlog_t *log, XFS_BUF_ZEROFLAGS(bp); XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); + XFS_BUF_SET_LOGBUF(bp); if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) XFS_BUF_ORDERED(bp); dptr = XFS_BUF_PTR(bp);