From 1a0a5dad67b60250dce151c9533ccbecdfd822d4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 9 Aug 2023 15:05:39 -0700 Subject: [PATCH] xfs: document the invalidate_bdev call in invalidate_bdev Copy and paste the commit message from Darrick into a comment to explain the seemingly odd invalidate_bdev in xfs_shutdown_devices. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Message-Id: <20230809220545.1308228-8-hch@lst.de> Signed-off-by: Christian Brauner --- fs/xfs/xfs_super.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 18d34e8..46d0cbe 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -410,6 +410,32 @@ STATIC void xfs_shutdown_devices( struct xfs_mount *mp) { + /* + * Udev is triggered whenever anyone closes a block device or unmounts + * a file systemm on a block device. + * The default udev rules invoke blkid to read the fs super and create + * symlinks to the bdev under /dev/disk. For this, it uses buffered + * reads through the page cache. + * + * xfs_db also uses buffered reads to examine metadata. There is no + * coordination between xfs_db and udev, which means that they can run + * concurrently. Note there is no coordination between the kernel and + * blkid either. + * + * On a system with 64k pages, the page cache can cache the superblock + * and the root inode (and hence the root directory) with the same 64k + * page. If udev spawns blkid after the mkfs and the system is busy + * enough that it is still running when xfs_db starts up, they'll both + * read from the same page in the pagecache. + * + * The unmount writes updated inode metadata to disk directly. The XFS + * buffer cache does not use the bdev pagecache, so it needs to + * invalidate that pagecache on unmount. If the above scenario occurs, + * the pagecache no longer reflects what's on disk, xfs_db reads the + * stale metadata, and fails to find /a. Most of the time this succeeds + * because closing a bdev invalidates the page cache, but when processes + * race, everyone loses. + */ if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) { blkdev_issue_flush(mp->m_logdev_targp->bt_bdev); invalidate_bdev(mp->m_logdev_targp->bt_bdev); -- 2.7.4