xfs: resample the data fork mapping after cycling ILOCK
Conflicts:
fs/xfs/xfs_reflink.c - The variables around seq_before in
xfs_reflink_fill_cow_hole are slightly different. We adjust the
patch so that it still just adds the seq_before declaration.
[ Upstream commit 2f4acd0fcd862e22eab45690ec2c08c80b6ef2e7 ]
xfs_reflink_fill_{cow_hole,delalloc} are both presented with an inode,
a data fork mapping, and a cow fork mapping. Unfortunately, these two
helpers cycle the ILOCK to grab a transaction, which means that the
mappings are stale as soon as we reacquire the ILOCK. Currently we
refresh the cow fork mapping by re-calling xfs_find_trim_cow_extent, but
we don't refresh the data fork mapping beforehand, which means that the
xfs_bmap_trim_cow in that function queries the refcount btree about the
wrong physical blocks and returns an inaccurate value in *shared.
If *shared is now false, the directio write proceeds with a stale data
fork mapping. Fix this by querying the data fork mapping if the
sequence counter changes across the ILOCK cycle.
BUG=b/535631885
TEST=presubmit, compile
RELEASE_NOTE=Fixed an important bug for xfs file system users.
cos-patch: bug
Fixes: 3c68d44a2b49a0 ("xfs: allocate direct I/O COW blocks in iomap_begin")
Signed-off-by: Kevin Berry <kpberry@google.com>
Change-Id: If6c9c2c82fba19110a2f610e11d2ad9c3cd51bc6
Reviewed-on: https://cos-review.googlesource.com/c/third_party/kernel/+/169925
Tested-by: Cusky Presubmit Bot <presubmit@cos-infra-prod.iam.gserviceaccount.com>
Main-Branch-Verified: Cusky Presubmit Bot <presubmit@cos-infra-prod.iam.gserviceaccount.com>
Reviewed-by: Robert Kolchmeyer <rkolchmeyer@google.com>
Reviewed-on: https://cos-review.googlesource.com/c/third_party/kernel/+/170124
diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
index 4058cf3..23d1650 100644
--- a/fs/xfs/xfs_reflink.c
+++ b/fs/xfs/xfs_reflink.c
@@ -389,6 +389,7 @@ xfs_reflink_fill_cow_hole(
struct xfs_trans *tp;
xfs_filblks_t resaligned;
xfs_extlen_t resblks;
+ unsigned int seq_before = READ_ONCE(ip->i_df.if_seq);
int nimaps;
int error;
bool found;
@@ -407,6 +408,22 @@ xfs_reflink_fill_cow_hole(
*lockmode = XFS_ILOCK_EXCL;
+ /*
+ * The data fork mapping may have changed while we dropped the ILOCK
+ * (a racing O_DIRECT writer under IOLOCK_SHARED can complete a full
+ * CoW cycle including xfs_reflink_end_cow(), which remaps this offset
+ * and drops the refcount of the old shared block). Re-read it so the
+ * shared-status recheck below and the caller's in-place iomap both
+ * operate on the current mapping rather than a stale physical block.
+ */
+ if (seq_before != READ_ONCE(ip->i_df.if_seq)) {
+ nimaps = 1;
+ error = xfs_bmapi_read(ip, imap->br_startoff,
+ imap->br_blockcount, imap, &nimaps, 0);
+ if (error)
+ goto out_trans_cancel;
+ }
+
error = xfs_find_trim_cow_extent(ip, imap, cmap, shared, &found);
if (error || !*shared)
goto out_trans_cancel;
@@ -453,6 +470,8 @@ xfs_reflink_fill_delalloc(
bool found;
do {
+ unsigned int seq_before = READ_ONCE(ip->i_df.if_seq);
+
xfs_iunlock(ip, *lockmode);
*lockmode = 0;
@@ -463,6 +482,23 @@ xfs_reflink_fill_delalloc(
*lockmode = XFS_ILOCK_EXCL;
+ /*
+ * The data fork mapping may have changed while we dropped the
+ * ILOCK (a racing O_DIRECT writer under IOLOCK_SHARED can
+ * complete a full CoW cycle including xfs_reflink_end_cow(),
+ * which remaps this offset and drops the refcount of the old
+ * shared block). Re-read it so the shared-status recheck
+ * below and the caller's in-place iomap both operate on the
+ * current mapping rather than a stale physical block.
+ */
+ if (seq_before != READ_ONCE(ip->i_df.if_seq)) {
+ nimaps = 1;
+ error = xfs_bmapi_read(ip, imap->br_startoff,
+ imap->br_blockcount, imap, &nimaps, 0);
+ if (error)
+ goto out_trans_cancel;
+ }
+
error = xfs_find_trim_cow_extent(ip, imap, cmap, shared,
&found);
if (error || !*shared)
