Merge: xfs: Updates for 10.1

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-10/-/merge_requests/1180

JIRA: https://issues.redhat.com/browse/RHEL-85590

This is a pro-active update for xfs targeting 10.1

Because RHEL10 is based on Linux v6.12, there are not many patches needed
for now. Just backport some essential ones, as we prepare a more complete
rebase for 10.2.

Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>

Approved-by: Brian Foster <bfoster@redhat.com>
Approved-by: Eric Sandeen <esandeen@redhat.com>
Approved-by: CKI KWF Bot <cki-ci-bot+kwf-gitlab-com@redhat.com>

Merged-by: CKI GitLab Kmaint Pipeline Bot <26919896-cki-kmaint-pipeline-bot@users.noreply.gitlab.com>

Author: CKI KWF Bot

fs/xfs/libxfs/xfs_bmap.c

@@ -3531,12 +3531,12 @@ xfs_bmap_btalloc_at_eof(
 	int			error;
 
 	/*
-	 * If there are already extents in the file, try an exact EOF block
-	 * allocation to extend the file as a contiguous extent. If that fails,
-	 * or it's the first allocation in a file, just try for a stripe aligned
-	 * allocation.
+	 * If there are already extents in the file, and xfs_bmap_adjacent() has
+	 * given a better blkno, try an exact EOF block allocation to extend the
+	 * file as a contiguous extent. If that fails, or it's the first
+	 * allocation in a file, just try for a stripe aligned allocation.
 	 */
-	if (ap->offset) {
+	if (ap->eof) {
 		xfs_extlen_t	nextminlen = 0;
 
 		/*
@@ -3704,7 +3704,8 @@ xfs_bmap_btalloc_best_length(
 	int			error;
 
 	ap->blkno = XFS_INO_TO_FSB(args->mp, ap->ip->i_ino);
-	xfs_bmap_adjacent(ap);
+	if (!xfs_bmap_adjacent(ap))
+		ap->eof = false;
 
 	/*
 	 * Search for an allocation group with a single extent large enough for

fs/xfs/scrub/common.h

@@ -179,7 +179,6 @@ static inline bool xchk_skip_xref(struct xfs_scrub_metadata *sm)
 bool xchk_dir_looks_zapped(struct xfs_inode *dp);
 bool xchk_pptr_looks_zapped(struct xfs_inode *ip);
 
-#ifdef CONFIG_XFS_ONLINE_REPAIR
 /* Decide if a repair is required. */
 static inline bool xchk_needs_repair(const struct xfs_scrub_metadata *sm)
 {
@@ -199,10 +198,6 @@ static inline bool xchk_could_repair(const struct xfs_scrub *sc)
 	return (sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) &&
 		!(sc->flags & XREP_ALREADY_FIXED);
 }
-#else
-# define xchk_needs_repair(sc)		(false)
-# define xchk_could_repair(sc)		(false)
-#endif /* CONFIG_XFS_ONLINE_REPAIR */
 
 int xchk_metadata_inode_forks(struct xfs_scrub *sc);
 

fs/xfs/scrub/repair.h

@@ -163,7 +163,16 @@ bool xrep_buf_verify_struct(struct xfs_buf *bp, const struct xfs_buf_ops *ops);
 #else
 
 #define xrep_ino_dqattach(sc)	(0)
-#define xrep_will_attempt(sc)	(false)
+
+/*
+ * When online repair is not built into the kernel, we still want to attempt
+ * the repair so that the stub xrep_attempt below will return EOPNOTSUPP.
+ */
+static inline bool xrep_will_attempt(const struct xfs_scrub *sc)
+{
+	return (sc->sm->sm_flags & XFS_SCRUB_IFLAG_FORCE_REBUILD) ||
+		xchk_needs_repair(sc->sm);
+}
 
 static inline int
 xrep_attempt(

fs/xfs/scrub/scrub.c

@@ -149,6 +149,18 @@ xchk_probe(
 	if (xchk_should_terminate(sc, &error))
 		return error;
 
+	/*
+	 * If the caller is probing to see if repair works but repair isn't
+	 * built into the kernel, return EOPNOTSUPP because that's the signal
+	 * that userspace expects.  If online repair is built in, set the
+	 * CORRUPT flag (without any of the usual tracing/logging) to force us
+	 * into xrep_probe.
+	 */
+	if (xchk_could_repair(sc)) {
+		if (!IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR))
+			return -EOPNOTSUPP;
+		sc->sm->sm_flags |= XFS_SCRUB_OFLAG_CORRUPT;
+	}
 	return 0;
 }
 

fs/xfs/scrub/trace.h

@@ -601,7 +601,7 @@ TRACE_EVENT(xchk_ifork_btree_op_error,
 	TP_fast_assign(
 		xfs_fsblock_t fsbno = xchk_btree_cur_fsbno(cur, level);
 		__entry->dev = sc->mp->m_super->s_dev;
-		__entry->ino = sc->ip->i_ino;
+		__entry->ino = cur->bc_ino.ip->i_ino;
 		__entry->whichfork = cur->bc_ino.whichfork;
 		__entry->type = sc->sm->sm_type;
 		__assign_str(name);

fs/xfs/xfs_bmap_util.c

@@ -546,10 +546,14 @@ xfs_can_free_eofblocks(
 		return false;
 
 	/*
-	 * Check if there is an post-EOF extent to free.
+	 * Check if there is an post-EOF extent to free.  If there are any
+	 * delalloc blocks attached to the inode (data fork delalloc
+	 * reservations or CoW extents of any kind), we need to free them so
+	 * that inactivation doesn't fail to erase them.
 	 */
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
-	if (xfs_iext_lookup_extent(ip, &ip->i_df, end_fsb, &icur, &imap))
+	if (ip->i_delayed_blks ||
+	    xfs_iext_lookup_extent(ip, &ip->i_df, end_fsb, &icur, &imap))
 		found_blocks = true;
 	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 	return found_blocks;

fs/xfs/xfs_buf.c

@@ -1656,10 +1656,8 @@ _xfs_buf_ioapply(
 	op |= REQ_META;
 
 	/* in-memory targets are directly mapped, no IO required. */
-	if (xfs_buftarg_is_mem(bp->b_target)) {
-		xfs_buf_ioend(bp);
+	if (xfs_buftarg_is_mem(bp->b_target))
 		return;
-	}
 
 	/*
 	 * Walk all the vectors issuing IO on them. Set up the initial offset

fs/xfs/xfs_buf_item_recover.c

@@ -1036,11 +1036,20 @@ xlog_recover_buf_commit_pass2(
 		error = xlog_recover_do_primary_sb_buffer(mp, item, bp, buf_f,
 				current_lsn);
 		if (error)
-			goto out_release;
+			goto out_writebuf;
 	} else {
 		xlog_recover_do_reg_buffer(mp, item, bp, buf_f, current_lsn);
 	}
 
+	/*
+	 * Buffer held by buf log item during 'normal' buffer recovery must
+	 * be committed through buffer I/O submission path to ensure proper
+	 * release. When error occurs during sb buffer recovery, log shutdown
+	 * will be done before submitting buffer list so that buffers can be
+	 * released correctly through ioend failure path.
+	 */
+out_writebuf:
+
 	/*
 	 * Perform delayed write on the buffer.  Asynchronous writes will be
 	 * slower when taking into account all the buffers to be flushed.

fs/xfs/xfs_exchrange.c

@@ -326,22 +326,6 @@ xfs_exchrange_mappings(
  * successfully but before locks are dropped.
  */
 
-/* Verify that we have security clearance to perform this operation. */
-static int
-xfs_exchange_range_verify_area(
-	struct xfs_exchrange	*fxr)
-{
-	int			ret;
-
-	ret = remap_verify_area(fxr->file1, fxr->file1_offset, fxr->length,
-			true);
-	if (ret)
-		return ret;
-
-	return remap_verify_area(fxr->file2, fxr->file2_offset, fxr->length,
-			true);
-}
-
 /*
  * Performs necessary checks before doing a range exchange, having stabilized
  * mutable inode attributes via i_rwsem.
@@ -352,11 +336,13 @@ xfs_exchange_range_checks(
 	unsigned int		alloc_unit)
 {
 	struct inode		*inode1 = file_inode(fxr->file1);
+	loff_t			size1 = i_size_read(inode1);
 	struct inode		*inode2 = file_inode(fxr->file2);
+	loff_t			size2 = i_size_read(inode2);
 	uint64_t		allocmask = alloc_unit - 1;
 	int64_t			test_len;
 	uint64_t		blen;
-	loff_t			size1, size2, tmp;
+	loff_t			tmp;
 	int			error;
 
 	/* Don't touch certain kinds of inodes */
@@ -365,24 +351,25 @@ xfs_exchange_range_checks(
 	if (IS_SWAPFILE(inode1) || IS_SWAPFILE(inode2))
 		return -ETXTBSY;
 
-	size1 = i_size_read(inode1);
-	size2 = i_size_read(inode2);
-
 	/* Ranges cannot start after EOF. */
 	if (fxr->file1_offset > size1 || fxr->file2_offset > size2)
 		return -EINVAL;
 
-	/*
-	 * If the caller said to exchange to EOF, we set the length of the
-	 * request large enough to cover everything to the end of both files.
-	 */
 	if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) {
+		/*
+		 * If the caller said to exchange to EOF, we set the length of
+		 * the request large enough to cover everything to the end of
+		 * both files.
+		 */
 		fxr->length = max_t(int64_t, size1 - fxr->file1_offset,
 					     size2 - fxr->file2_offset);
-
-		error = xfs_exchange_range_verify_area(fxr);
-		if (error)
-			return error;
+	} else {
+		/*
+		 * Otherwise we require both ranges to end within EOF.
+		 */
+		if (fxr->file1_offset + fxr->length > size1 ||
+		    fxr->file2_offset + fxr->length > size2)
+			return -EINVAL;
 	}
 
 	/*
@@ -398,15 +385,6 @@ xfs_exchange_range_checks(
 	    check_add_overflow(fxr->file2_offset, fxr->length, &tmp))
 		return -EINVAL;
 
-	/*
-	 * We require both ranges to end within EOF, unless we're exchanging
-	 * to EOF.
-	 */
-	if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) &&
-	    (fxr->file1_offset + fxr->length > size1 ||
-	     fxr->file2_offset + fxr->length > size2))
-		return -EINVAL;
-
 	/*
 	 * Make sure we don't hit any file size limits.  If we hit any size
 	 * limits such that test_length was adjusted, we abort the whole
@@ -744,6 +722,7 @@ xfs_exchange_range(
 {
 	struct inode		*inode1 = file_inode(fxr->file1);
 	struct inode		*inode2 = file_inode(fxr->file2);
+	loff_t			check_len = fxr->length;
 	int			ret;
 
 	BUILD_BUG_ON(XFS_EXCHANGE_RANGE_ALL_FLAGS &
@@ -776,14 +755,18 @@ xfs_exchange_range(
 		return -EBADF;
 
 	/*
-	 * If we're not exchanging to EOF, we can check the areas before
-	 * stabilizing both files' i_size.
+	 * If we're exchanging to EOF we can't calculate the length until taking
+	 * the iolock.  Pass a 0 length to remap_verify_area similar to the
+	 * FICLONE and FICLONERANGE ioctls that support cloning to EOF as well.
 	 */
-	if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)) {
-		ret = xfs_exchange_range_verify_area(fxr);
-		if (ret)
-			return ret;
-	}
+	if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)
+		check_len = 0;
+	ret = remap_verify_area(fxr->file1, fxr->file1_offset, check_len, true);
+	if (ret)
+		return ret;
+	ret = remap_verify_area(fxr->file2, fxr->file2_offset, check_len, true);
+	if (ret)
+		return ret;
 
 	/* Update cmtime if the fd/inode don't forbid it. */
 	if (!(fxr->file1->f_mode & FMODE_NOCMTIME) && !IS_NOCMTIME(inode1))

fs/xfs/xfs_fsmap.c

@@ -162,7 +162,8 @@ struct xfs_getfsmap_info {
 	xfs_daddr_t		next_daddr;	/* next daddr we expect */
 	/* daddr of low fsmap key when we're using the rtbitmap */
 	xfs_daddr_t		low_daddr;
-	xfs_daddr_t		end_daddr;	/* daddr of high fsmap key */
+	/* daddr of high fsmap key, or the last daddr on the device */
+	xfs_daddr_t		end_daddr;
 	u64			missing_owner;	/* owner of holes */
 	u32			dev;		/* device id */
 	/*
@@ -305,8 +306,8 @@ xfs_getfsmap_helper(
 	 * we calculated from userspace's high key to synthesize the record.
 	 * Note that if the btree query found a mapping, there won't be a gap.
 	 */
-	if (info->last && info->end_daddr != XFS_BUF_DADDR_NULL)
-		rec_daddr = info->end_daddr;
+	if (info->last)
+		rec_daddr = info->end_daddr + 1;
 
 	/* Are we just counting mappings? */
 	if (info->head->fmh_count == 0) {
@@ -898,7 +899,10 @@ xfs_getfsmap(
 	struct xfs_trans		*tp = NULL;
 	struct xfs_fsmap		dkeys[2];	/* per-dev keys */
 	struct xfs_getfsmap_dev		handlers[XFS_GETFSMAP_DEVS];
-	struct xfs_getfsmap_info	info = { NULL };
+	struct xfs_getfsmap_info	info = {
+		.fsmap_recs		= fsmap_recs,
+		.head			= head,
+	};
 	bool				use_rmap;
 	int				i;
 	int				error = 0;
@@ -963,9 +967,6 @@ xfs_getfsmap(
 
 	info.next_daddr = head->fmh_keys[0].fmr_physical +
 			  head->fmh_keys[0].fmr_length;
-	info.end_daddr = XFS_BUF_DADDR_NULL;
-	info.fsmap_recs = fsmap_recs;
-	info.head = head;
 
 	/* For each device we support... */
 	for (i = 0; i < XFS_GETFSMAP_DEVS; i++) {
@@ -978,17 +979,23 @@ xfs_getfsmap(
 			break;
 
 		/*
-		 * If this device number matches the high key, we have
-		 * to pass the high key to the handler to limit the
-		 * query results.  If the device number exceeds the
-		 * low key, zero out the low key so that we get
-		 * everything from the beginning.
+		 * If this device number matches the high key, we have to pass
+		 * the high key to the handler to limit the query results, and
+		 * set the end_daddr so that we can synthesize records at the
+		 * end of the query range or device.
 		 */
 		if (handlers[i].dev == head->fmh_keys[1].fmr_device) {
 			dkeys[1] = head->fmh_keys[1];
 			info.end_daddr = min(handlers[i].nr_sectors - 1,
 					     dkeys[1].fmr_physical);
+		} else {
+			info.end_daddr = handlers[i].nr_sectors - 1;
 		}
+
+		/*
+		 * If the device number exceeds the low key, zero out the low
+		 * key so that we get everything from the beginning.
+		 */
 		if (handlers[i].dev > head->fmh_keys[0].fmr_device)
 			memset(&dkeys[0], 0, sizeof(struct xfs_fsmap));