From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay1.corp.sgi.com [137.38.102.111]) by oss.sgi.com (Postfix) with ESMTP id C2E867F47 for ; Fri, 14 Nov 2014 12:00:26 -0600 (CST) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by relay1.corp.sgi.com (Postfix) with ESMTP id B09A38F8050 for ; Fri, 14 Nov 2014 10:00:26 -0800 (PST) Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by cuda.sgi.com with ESMTP id hFGaX2rXWrCd9LrQ (version=TLSv1 cipher=AES256-SHA bits=256 verify=NO) for ; Fri, 14 Nov 2014 10:00:25 -0800 (PST) Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id sAEGH9Yl006340 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=FAIL) for ; Fri, 14 Nov 2014 11:17:09 -0500 Received: from bfoster.bfoster ([10.18.41.237]) by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id sAEGH81i012331 for ; Fri, 14 Nov 2014 11:17:08 -0500 From: Brian Foster Subject: [PATCH 3/3] xfs: allow lazy sb counter sync during filesystem freeze sequence Date: Fri, 14 Nov 2014 11:17:07 -0500 Message-Id: <1415981827-52993-4-git-send-email-bfoster@redhat.com> In-Reply-To: <1415981827-52993-1-git-send-email-bfoster@redhat.com> References: <1415981827-52993-1-git-send-email-bfoster@redhat.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 Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com The expectation since the introduction the lazy superblock counters is that the counters are synced and superblock logged appropriately as part of the filesystem freeze sequence. This does not occur, however, due to the logic in xfs_fs_writable() that prevents progress when the fs is in any state other than SB_UNFROZEN. While this is a bug, it has not been exposed to date because the last thing XFS does during freeze is dirty the log. The log recovery process recalculates the counters from AGI/AGF metadata to ensure everything is correct. Therefore should a crash occur while an fs is frozen, the subsequent log recovery puts everything back in order. See the following commit for reference: 92821e2b [XFS] Lazy Superblock Counters We might not always want to rely on dirtying the log on a frozen fs. Modify xfs_log_sbcount() to proceed when the filesystem is freezing but not once the freeze process has completed. Modify xfs_fs_writable() to accept the minimum freeze level for which modifications should be blocked to support various codepaths. Signed-off-by: Brian Foster --- fs/xfs/xfs_log.c | 2 +- fs/xfs/xfs_mount.c | 29 +++++++++++++++++++++-------- fs/xfs/xfs_mount.h | 2 +- 3 files changed, 23 insertions(+), 10 deletions(-) diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index fe88ef6..e810e9d 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1031,7 +1031,7 @@ xfs_log_need_covered(xfs_mount_t *mp) struct xlog *log = mp->m_log; int needed = 0; - if (!xfs_fs_writable(mp)) + if (!xfs_fs_writable(mp, SB_FREEZE_WRITE)) return 0; if (!xlog_cil_empty(log)) diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 51435db..13d1170 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1074,11 +1074,23 @@ xfs_unmountfs( xfs_sysfs_del(&mp->m_kobj); } -int -xfs_fs_writable(xfs_mount_t *mp) +/* + * Determine whether modifications can proceed. The caller specifies the minimum + * freeze level for which modifications should not be allowed. This allows + * certain operations to proceed while the freeze sequence is in progress, if + * necessary. + */ +bool +xfs_fs_writable( + struct xfs_mount *mp, + int level) { - return !(mp->m_super->s_writers.frozen || XFS_FORCED_SHUTDOWN(mp) || - (mp->m_flags & XFS_MOUNT_RDONLY)); + ASSERT(level > SB_UNFROZEN); + if ((mp->m_super->s_writers.frozen >= level) || + XFS_FORCED_SHUTDOWN(mp) || (mp->m_flags & XFS_MOUNT_RDONLY)) + return false; + + return true; } /* @@ -1086,9 +1098,9 @@ xfs_fs_writable(xfs_mount_t *mp) * * Sync the superblock counters to disk. * - * Note this code can be called during the process of freezing, so - * we may need to use the transaction allocator which does not - * block when the transaction subsystem is in its frozen state. + * Note this code can be called during the process of freezing, so we use the + * transaction allocator that does not block when the transaction subsystem is + * in its frozen state. */ int xfs_log_sbcount(xfs_mount_t *mp) @@ -1096,7 +1108,8 @@ xfs_log_sbcount(xfs_mount_t *mp) xfs_trans_t *tp; int error; - if (!xfs_fs_writable(mp)) + /* allow this to proceed during the freeze sequence... */ + if (!xfs_fs_writable(mp, SB_FREEZE_COMPLETE)) return 0; xfs_icsb_sync_counters(mp, 0); diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index b0447c8..8852418 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -384,7 +384,7 @@ extern int xfs_mount_log_sb(xfs_mount_t *, __int64_t); extern struct xfs_buf *xfs_getsb(xfs_mount_t *, int); extern int xfs_readsb(xfs_mount_t *, int); extern void xfs_freesb(xfs_mount_t *); -extern int xfs_fs_writable(xfs_mount_t *); +extern bool xfs_fs_writable(struct xfs_mount *mp, int level); extern int xfs_sb_validate_fsb_count(struct xfs_sb *, __uint64_t); extern int xfs_dev_is_read_only(struct xfs_mount *, char *); -- 1.8.3.1 _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs