xfs: store in-progress CoW allocations in the refcount btree

Due to the way the CoW algorithm in XFS works, there's an interval
during which blocks allocated to handle a CoW can be lost -- if the FS
goes down after the blocks are allocated but before the block
remapping takes place.  This is exacerbated by the cowextsz hint --
allocated reservations can sit around for a while, waiting to get
used.

Since the refcount btree doesn't normally store records with refcount
of 1, we can use it to record these in-progress extents.  In-progress
blocks cannot be shared because they're not user-visible, so there
shouldn't be any conflicts with other programs.  This is a better
solution than holding EFIs during writeback because (a) EFIs can't be
relogged currently, (b) even if they could, EFIs are bound by
available log space, which puts an unnecessary upper bound on how much
CoW we can have in flight, and (c) we already have a mechanism to
track blocks.

At mount time, read the refcount records and free anything we find
with a refcount of 1 because those were in-progress when the FS went
down.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>

1 files changed, 13 insertions, 1 deletions

diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 8b826102dfe3..777fa974a319 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -1375,7 +1375,8 @@ struct xfs_owner_info {
 #define XFS_RMAP_OWN_INOBT	(-6ULL)	/* Inode btree blocks */
 #define XFS_RMAP_OWN_INODES	(-7ULL)	/* Inode chunk */
 #define XFS_RMAP_OWN_REFC	(-8ULL) /* refcount tree */
-#define XFS_RMAP_OWN_MIN	(-9ULL) /* guard */
+#define XFS_RMAP_OWN_COW	(-9ULL) /* cow allocations */
+#define XFS_RMAP_OWN_MIN	(-10ULL) /* guard */
 
 #define XFS_RMAP_NON_INODE_OWNER(owner)	(!!((owner) & (1ULL << 63)))
 
@@ -1477,6 +1478,17 @@ unsigned int xfs_refc_block(struct xfs_mount *mp);
  * data) are not tracked here.  Free space is also not tracked here.
  * This is consistent with pre-reflink XFS.
  */
+
+/*
+ * Extents that are being used to stage a copy on write are stored
+ * in the refcount btree with a refcount of 1 and the upper bit set
+ * on the startblock.  This speeds up mount time deletion of stale
+ * staging extents because they're all at the right side of the tree.
+ */
+#define XFS_REFC_COW_START		((xfs_agblock_t)(1U << 31))
+#define REFCNTBT_COWFLAG_BITLEN		1
+#define REFCNTBT_AGBLOCK_BITLEN		31
+
 struct xfs_refcount_rec {
 	__be32		rc_startblock;	/* starting block number */
 	__be32		rc_blockcount;	/* count of blocks */