Merge tag 'xfs-fixes-6.14-rc2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull xfs bug fixes from Carlos Maiolino:
 "A few fixes for XFS, but the most notable one is:

   - xfs: remove xfs_buf_cache.bc_lock

  which has been hit by different persons including syzbot"

* tag 'xfs-fixes-6.14-rc2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: remove xfs_buf_cache.bc_lock
  xfs: Add error handling for xfs_reflink_cancel_cow_range
  xfs: Propagate errors from xfs_reflink_cancel_cow_range in xfs_dax_write_iomap_end
  xfs: don't call remap_verify_area with sb write protection held
  xfs: remove an out of data comment in _xfs_buf_alloc
  xfs: fix the entry condition of exact EOF block allocation optimization

6 files changed, 58 insertions, 76 deletions

diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 40ad22fb808b..0ef19f1469ec 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3563,12 +3563,12 @@ 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;
 
 		/*
@@ -3736,7 +3736,8 @@ 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
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index d1d4a0a22e13..15bb790359f8 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -41,8 +41,7 @@ struct kmem_cache *xfs_buf_cache;
  *
  * xfs_buf_rele:
  *	b_lock
- *	  pag_buf_lock
- *	    lru_lock
+ *	  lru_lock
  *
  * xfs_buftarg_drain_rele
  *	lru_lock
@@ -220,23 +219,25 @@ _xfs_buf_alloc(
 	 */
 	flags &= ~(XBF_UNMAPPED | XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD);
 
-	spin_lock_init(&bp->b_lock);
+	/*
+	 * A new buffer is held and locked by the owner.  This ensures that the
+	 * buffer is owned by the caller and racing RCU lookups right after
+	 * inserting into the hash table are safe (and will have to wait for
+	 * the unlock to do anything non-trivial).
+	 */
 	bp->b_hold = 1;
+	sema_init(&bp->b_sema, 0); /* held, no waiters */
+
+	spin_lock_init(&bp->b_lock);
 	atomic_set(&bp->b_lru_ref, 1);
 	init_completion(&bp->b_iowait);
 	INIT_LIST_HEAD(&bp->b_lru);
 	INIT_LIST_HEAD(&bp->b_list);
 	INIT_LIST_HEAD(&bp->b_li_list);
-	sema_init(&bp->b_sema, 0); /* held, no waiters */
 	bp->b_target = target;
 	bp->b_mount = target->bt_mount;
 	bp->b_flags = flags;
 
-	/*
-	 * Set length and io_length to the same value initially.
-	 * I/O routines should use io_length, which will be the same in
-	 * most cases but may be reset (e.g. XFS recovery).
-	 */
 	error = xfs_buf_get_maps(bp, nmaps);
 	if (error)  {
 		kmem_cache_free(xfs_buf_cache, bp);
@@ -502,7 +503,6 @@ int
 xfs_buf_cache_init(
 	struct xfs_buf_cache	*bch)
 {
-	spin_lock_init(&bch->bc_lock);
 	return rhashtable_init(&bch->bc_hash, &xfs_buf_hash_params);
 }
 
@@ -652,17 +652,20 @@ xfs_buf_find_insert(
 	if (error)
 		goto out_free_buf;
 
-	spin_lock(&bch->bc_lock);
+	/* The new buffer keeps the perag reference until it is freed. */
+	new_bp->b_pag = pag;
+
+	rcu_read_lock();
 	bp = rhashtable_lookup_get_insert_fast(&bch->bc_hash,
 			&new_bp->b_rhash_head, xfs_buf_hash_params);
 	if (IS_ERR(bp)) {
+		rcu_read_unlock();
 		error = PTR_ERR(bp);
-		spin_unlock(&bch->bc_lock);
 		goto out_free_buf;
 	}
 	if (bp && xfs_buf_try_hold(bp)) {
 		/* found an existing buffer */
-		spin_unlock(&bch->bc_lock);
+		rcu_read_unlock();
 		error = xfs_buf_find_lock(bp, flags);
 		if (error)
 			xfs_buf_rele(bp);
@@ -670,10 +673,8 @@ xfs_buf_find_insert(
 			*bpp = bp;
 		goto out_free_buf;
 	}
+	rcu_read_unlock();
 
-	/* The new buffer keeps the perag reference until it is freed. */
-	new_bp->b_pag = pag;
-	spin_unlock(&bch->bc_lock);
 	*bpp = new_bp;
 	return 0;
 
@@ -1090,7 +1091,6 @@ xfs_buf_rele_cached(
 	}
 
 	/* we are asked to drop the last reference */
-	spin_lock(&bch->bc_lock);
 	__xfs_buf_ioacct_dec(bp);
 	if (!(bp->b_flags & XBF_STALE) && atomic_read(&bp->b_lru_ref)) {
 		/*
@@ -1102,7 +1102,6 @@ xfs_buf_rele_cached(
 			bp->b_state &= ~XFS_BSTATE_DISPOSE;
 		else
 			bp->b_hold--;
-		spin_unlock(&bch->bc_lock);
 	} else {
 		bp->b_hold--;
 		/*
@@ -1120,7 +1119,6 @@ xfs_buf_rele_cached(
 		ASSERT(!(bp->b_flags & _XBF_DELWRI_Q));
 		rhashtable_remove_fast(&bch->bc_hash, &bp->b_rhash_head,
 				xfs_buf_hash_params);
-		spin_unlock(&bch->bc_lock);
 		if (pag)
 			xfs_perag_put(pag);
 		freebuf = true;
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index 7e73663c5d4a..3b4ed42e11c0 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -80,7 +80,6 @@ typedef unsigned int xfs_buf_flags_t;
 #define XFS_BSTATE_IN_FLIGHT	 (1 << 1)	/* I/O in flight */
 
 struct xfs_buf_cache {
-	spinlock_t		bc_lock;
 	struct rhashtable	bc_hash;
 };
 
diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c
index f340a2015c4c..0b41bdfecdfb 100644
--- a/fs/xfs/xfs_exchrange.c
+++ b/fs/xfs/xfs_exchrange.c
@@ -329,22 +329,6 @@ out_trans_cancel:
  * 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.
@@ -355,11 +339,13 @@ 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 */
@@ -368,24 +354,25 @@ 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;
 	}
 
 	/*
@@ -402,15 +389,6 @@ xfs_exchange_range_checks(
 		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
 	 * operation.
@@ -747,6 +725,7 @@ 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 &
@@ -779,14 +758,18 @@ 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))
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index c95fe1b1de4e..b1f9f156ec88 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1404,8 +1404,11 @@ xfs_inactive(
 		goto out;
 
 	/* Try to clean out the cow blocks if there are any. */
-	if (xfs_inode_has_cow_data(ip))
-		xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true);
+	if (xfs_inode_has_cow_data(ip)) {
+		error = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true);
+		if (error)
+			goto out;
+	}
 
 	if (VFS_I(ip)->i_nlink != 0) {
 		/*
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 50fa3ef89f6c..d61460309a78 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -976,10 +976,8 @@ xfs_dax_write_iomap_end(
 	if (!xfs_is_cow_inode(ip))
 		return 0;
 
-	if (!written) {
-		xfs_reflink_cancel_cow_range(ip, pos, length, true);
-		return 0;
-	}
+	if (!written)
+		return xfs_reflink_cancel_cow_range(ip, pos, length, true);
 
 	return xfs_reflink_end_cow(ip, pos, written);
 }