Merge tag 'for-5.12-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "This is the first batch of fixes that usually arrive during the merge
  window code freeze. Regressions and stable material.

  Regressions:

   - fix deadlock in log sync in zoned mode

   - fix bugs in subpage mode still wrongly assuming sectorsize == page
     size

  Fixes:

   - fix missing kunmap of the Q stripe in RAID6

   - block group fixes:
      - fix race between extent freeing/allocation when using bitmaps
      - avoid double put of block group when emptying cluster

   - swapfile fixes:
      - fix swapfile writes vs running scrub
      - fix swapfile activation vs snapshot creation

   - fix stale data exposure after cloning a hole with NO_HOLES enabled

   - remove tree-checker check that does not work in case information
     from other leaves is necessary"

* tag 'for-5.12-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: zoned: fix deadlock on log sync
  btrfs: avoid double put of block group when emptying cluster
  btrfs: fix stale data exposure after cloning a hole with NO_HOLES enabled
  btrfs: tree-checker: do not error out if extent ref hash doesn't match
  btrfs: fix race between swap file activation and snapshot creation
  btrfs: fix race between writes to swap files and scrub
  btrfs: avoid checking for RO block group twice during nocow writeback
  btrfs: fix race between extent freeing/allocation when using bitmaps
  btrfs: make check_compressed_csum() to be subpage compatible
  btrfs: make btrfs_submit_compressed_read() subpage compatible
  btrfs: fix raid6 qstripe kmap

fs/btrfs/block-group.c

@@ -1162,6 +1162,11 @@ static int inc_block_group_ro(struct btrfs_block_group *cache, int force)
 	spin_lock(&sinfo->lock);
 	spin_lock(&cache->lock);
 
+	if (cache->swap_extents) {
+		ret = -ETXTBSY;
+		goto out;
+	}
+
 	if (cache->ro) {
 		cache->ro++;
 		ret = 0;
@@ -2307,7 +2312,7 @@ int btrfs_inc_block_group_ro(struct btrfs_block_group *cache,
 	}
 
 	ret = inc_block_group_ro(cache, 0);
-	if (!do_chunk_alloc)
+	if (!do_chunk_alloc || ret == -ETXTBSY)
 		goto unlock_out;
 	if (!ret)
 		goto out;
@@ -2316,6 +2321,8 @@ int btrfs_inc_block_group_ro(struct btrfs_block_group *cache,
 	if (ret < 0)
 		goto out;
 	ret = inc_block_group_ro(cache, 0);
+	if (ret == -ETXTBSY)
+		goto unlock_out;
 out:
 	if (cache->flags & BTRFS_BLOCK_GROUP_SYSTEM) {
 		alloc_flags = btrfs_get_alloc_profile(fs_info, cache->flags);
@@ -3406,6 +3413,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		ASSERT(list_empty(&block_group->io_list));
 		ASSERT(list_empty(&block_group->bg_list));
 		ASSERT(refcount_read(&block_group->refs) == 1);
+		ASSERT(block_group->swap_extents == 0);
 		btrfs_put_block_group(block_group);
 
 		spin_lock(&info->block_group_cache_lock);
@@ -3472,3 +3480,26 @@ void btrfs_unfreeze_block_group(struct btrfs_block_group *block_group)
 		__btrfs_remove_free_space_cache(block_group->free_space_ctl);
 	}
 }
+
+bool btrfs_inc_block_group_swap_extents(struct btrfs_block_group *bg)
+{
+	bool ret = true;
+
+	spin_lock(&bg->lock);
+	if (bg->ro)
+		ret = false;
+	else
+		bg->swap_extents++;
+	spin_unlock(&bg->lock);
+
+	return ret;
+}
+
+void btrfs_dec_block_group_swap_extents(struct btrfs_block_group *bg, int amount)
+{
+	spin_lock(&bg->lock);
+	ASSERT(!bg->ro);
+	ASSERT(bg->swap_extents >= amount);
+	bg->swap_extents -= amount;
+	spin_unlock(&bg->lock);
+}

fs/btrfs/block-group.h

@@ -186,6 +186,12 @@ struct btrfs_block_group {
 	/* Flag indicating this block group is placed on a sequential zone */
 	bool seq_zone;
 
+	/*
+	 * Number of extents in this block group used for swap files.
+	 * All accesses protected by the spinlock 'lock'.
+	 */
+	int swap_extents;
+
 	/* Record locked full stripes for RAID5/6 block group */
 	struct btrfs_full_stripe_locks_tree full_stripe_locks_root;
 
@@ -312,4 +318,7 @@ static inline int btrfs_block_group_done(struct btrfs_block_group *cache)
 void btrfs_freeze_block_group(struct btrfs_block_group *cache);
 void btrfs_unfreeze_block_group(struct btrfs_block_group *cache);
 
+bool btrfs_inc_block_group_swap_extents(struct btrfs_block_group *bg);
+void btrfs_dec_block_group_swap_extents(struct btrfs_block_group *bg, int amount);
+
 #endif /* BTRFS_BLOCK_GROUP_H */

fs/btrfs/compression.c

@@ -141,6 +141,7 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio,
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
 	const u32 csum_size = fs_info->csum_size;
+	const u32 sectorsize = fs_info->sectorsize;
 	struct page *page;
 	unsigned long i;
 	char *kaddr;
@@ -154,13 +155,23 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio,
 	shash->tfm = fs_info->csum_shash;
 
 	for (i = 0; i < cb->nr_pages; i++) {
+		u32 pg_offset;
+		u32 bytes_left = PAGE_SIZE;
 		page = cb->compressed_pages[i];
 
+		/* Determine the remaining bytes inside the page first */
+		if (i == cb->nr_pages - 1)
+			bytes_left = cb->compressed_len - i * PAGE_SIZE;
+
+		/* Hash through the page sector by sector */
+		for (pg_offset = 0; pg_offset < bytes_left;
+		     pg_offset += sectorsize) {
 			kaddr = kmap_atomic(page);
-		crypto_shash_digest(shash, kaddr, PAGE_SIZE, csum);
+			crypto_shash_digest(shash, kaddr + pg_offset,
+					    sectorsize, csum);
 			kunmap_atomic(kaddr);
 
-		if (memcmp(&csum, cb_sum, csum_size)) {
+			if (memcmp(&csum, cb_sum, csum_size) != 0) {
 				btrfs_print_data_csum_error(inode, disk_start,
 						csum, cb_sum, cb->mirror_num);
 				if (btrfs_io_bio(bio)->device)
@@ -170,6 +181,8 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio,
 				return -EIO;
 			}
 			cb_sum += csum_size;
+			disk_start += sectorsize;
+		}
 	}
 	return 0;
 }
@@ -640,7 +653,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree,
 				   page_offset(bio_first_page_all(bio)),
-				   PAGE_SIZE);
+				   fs_info->sectorsize);
 	read_unlock(&em_tree->lock);
 	if (!em)
 		return BLK_STS_IOERR;
@@ -698,19 +711,30 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	refcount_set(&cb->pending_bios, 1);
 
 	for (pg_index = 0; pg_index < nr_pages; pg_index++) {
+		u32 pg_len = PAGE_SIZE;
 		int submit = 0;
 
+		/*
+		 * To handle subpage case, we need to make sure the bio only
+		 * covers the range we need.
+		 *
+		 * If we're at the last page, truncate the length to only cover
+		 * the remaining part.
+		 */
+		if (pg_index == nr_pages - 1)
+			pg_len = min_t(u32, PAGE_SIZE,
+					compressed_len - pg_index * PAGE_SIZE);
+
 		page = cb->compressed_pages[pg_index];
 		page->mapping = inode->i_mapping;
 		page->index = em_start >> PAGE_SHIFT;
 
 		if (comp_bio->bi_iter.bi_size)
-			submit = btrfs_bio_fits_in_stripe(page, PAGE_SIZE,
+			submit = btrfs_bio_fits_in_stripe(page, pg_len,
 							  comp_bio, 0);
 
 		page->mapping = NULL;
-		if (submit || bio_add_page(comp_bio, page, PAGE_SIZE, 0) <
-		    PAGE_SIZE) {
+		if (submit || bio_add_page(comp_bio, page, pg_len, 0) < pg_len) {
 			unsigned int nr_sectors;
 
 			ret = btrfs_bio_wq_end_io(fs_info, comp_bio,
@@ -743,9 +767,9 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 			comp_bio->bi_private = cb;
 			comp_bio->bi_end_io = end_compressed_bio_read;
 
-			bio_add_page(comp_bio, page, PAGE_SIZE, 0);
+			bio_add_page(comp_bio, page, pg_len, 0);
 		}
-		cur_disk_byte += PAGE_SIZE;
+		cur_disk_byte += pg_len;
 	}
 
 	ret = btrfs_bio_wq_end_io(fs_info, comp_bio, BTRFS_WQ_ENDIO_DATA);

fs/btrfs/ctree.h

@@ -524,6 +524,11 @@ struct btrfs_swapfile_pin {
 	 * points to a struct btrfs_device.
 	 */
 	bool is_block_group;
+	/*
+	 * Only used when 'is_block_group' is true and it is the number of
+	 * extents used by a swapfile for this block group ('ptr' field).
+	 */
+	int bg_extent_count;
 };
 
 bool btrfs_pinned_by_swapfile(struct btrfs_fs_info *fs_info, void *ptr);

fs/btrfs/free-space-cache.c

@@ -2801,8 +2801,10 @@ static void __btrfs_return_cluster_to_free_space(
 	struct rb_node *node;
 
 	spin_lock(&cluster->lock);
-	if (cluster->block_group != block_group)
-		goto out;
+	if (cluster->block_group != block_group) {
+		spin_unlock(&cluster->lock);
+		return;
+	}
 
 	cluster->block_group = NULL;
 	cluster->window_start = 0;
@@ -2840,8 +2842,6 @@ static void __btrfs_return_cluster_to_free_space(
 				   entry->offset, &entry->offset_index, bitmap);
 	}
 	cluster->root = RB_ROOT;
-
-out:
 	spin_unlock(&cluster->lock);
 	btrfs_put_block_group(block_group);
 }
@@ -3125,8 +3125,6 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group,
 			entry->bytes -= bytes;
 		}
 
-		if (entry->bytes == 0)
-			rb_erase(&entry->offset_index, &cluster->root);
 		break;
 	}
 out:
@@ -3143,7 +3141,10 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group,
 	ctl->free_space -= bytes;
 	if (!entry->bitmap && !btrfs_free_space_trimmed(entry))
 		ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes;
+
+	spin_lock(&cluster->lock);
 	if (entry->bytes == 0) {
+		rb_erase(&entry->offset_index, &cluster->root);
 		ctl->free_extents--;
 		if (entry->bitmap) {
 			kmem_cache_free(btrfs_free_space_bitmap_cachep,
@@ -3156,6 +3157,7 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group,
 		kmem_cache_free(btrfs_free_space_cachep, entry);
 	}
 
+	spin_unlock(&cluster->lock);
 	spin_unlock(&ctl->tree_lock);
 
 	return ret;

fs/btrfs/inode.c

@@ -1674,9 +1674,6 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 			 */
 			btrfs_release_path(path);
 
-			/* If extent is RO, we must COW it */
-			if (btrfs_extent_readonly(fs_info, disk_bytenr))
-				goto out_check;
 			ret = btrfs_cross_ref_exist(root, ino,
 						    found_key.offset -
 						    extent_offset, disk_bytenr, false);
@@ -1723,6 +1720,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 				WARN_ON_ONCE(freespace_inode);
 				goto out_check;
 			}
+			/* If the extent's block group is RO, we must COW */
 			if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr))
 				goto out_check;
 			nocow = true;
@@ -10200,6 +10198,7 @@ static int btrfs_add_swapfile_pin(struct inode *inode, void *ptr,
 	sp->ptr = ptr;
 	sp->inode = inode;
 	sp->is_block_group = is_block_group;
+	sp->bg_extent_count = 1;
 
 	spin_lock(&fs_info->swapfile_pins_lock);
 	p = &fs_info->swapfile_pins.rb_node;
@@ -10213,6 +10212,8 @@ static int btrfs_add_swapfile_pin(struct inode *inode, void *ptr,
 			   (sp->ptr == entry->ptr && sp->inode > entry->inode)) {
 			p = &(*p)->rb_right;
 		} else {
+			if (is_block_group)
+				entry->bg_extent_count++;
 			spin_unlock(&fs_info->swapfile_pins_lock);
 			kfree(sp);
 			return 1;
@@ -10238,8 +10239,11 @@ static void btrfs_free_swapfile_pins(struct inode *inode)
 		sp = rb_entry(node, struct btrfs_swapfile_pin, node);
 		if (sp->inode == inode) {
 			rb_erase(&sp->node, &fs_info->swapfile_pins);
-			if (sp->is_block_group)
+			if (sp->is_block_group) {
+				btrfs_dec_block_group_swap_extents(sp->ptr,
+							   sp->bg_extent_count);
 				btrfs_put_block_group(sp->ptr);
+			}
 			kfree(sp);
 		}
 		node = next;
@@ -10300,7 +10304,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 			       sector_t *span)
 {
 	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct extent_state *cached_state = NULL;
 	struct extent_map *em = NULL;
@@ -10351,13 +10356,27 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 	   "cannot activate swapfile while exclusive operation is running");
 		return -EBUSY;
 	}
+
+	/*
+	 * Prevent snapshot creation while we are activating the swap file.
+	 * We do not want to race with snapshot creation. If snapshot creation
+	 * already started before we bumped nr_swapfiles from 0 to 1 and
+	 * completes before the first write into the swap file after it is
+	 * activated, than that write would fallback to COW.
+	 */
+	if (!btrfs_drew_try_write_lock(&root->snapshot_lock)) {
+		btrfs_exclop_finish(fs_info);
+		btrfs_warn(fs_info,
+	   "cannot activate swapfile because snapshot creation is in progress");
+		return -EINVAL;
+	}
 	/*
 	 * Snapshots can create extents which require COW even if NODATACOW is
 	 * set. We use this counter to prevent snapshots. We must increment it
 	 * before walking the extents because we don't want a concurrent
 	 * snapshot to run after we've already checked the extents.
 	 */
-	atomic_inc(&BTRFS_I(inode)->root->nr_swapfiles);
+	atomic_inc(&root->nr_swapfiles);
 
 	isize = ALIGN_DOWN(inode->i_size, fs_info->sectorsize);
 
@@ -10454,6 +10473,17 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 			goto out;
 		}
 
+		if (!btrfs_inc_block_group_swap_extents(bg)) {
+			btrfs_warn(fs_info,
+			   "block group for swapfile at %llu is read-only%s",
+			   bg->start,
+			   atomic_read(&fs_info->scrubs_running) ?
+				       " (scrub running)" : "");
+			btrfs_put_block_group(bg);
+			ret = -EINVAL;
+			goto out;
+		}
+
 		ret = btrfs_add_swapfile_pin(inode, bg, true);
 		if (ret) {
 			btrfs_put_block_group(bg);
@@ -10492,6 +10522,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 	if (ret)
 		btrfs_swap_deactivate(file);
 
+	btrfs_drew_write_unlock(&root->snapshot_lock);
+
 	btrfs_exclop_finish(fs_info);
 
 	if (ret)

fs/btrfs/raid56.c

@@ -2359,16 +2359,21 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 	SetPageUptodate(p_page);
 
 	if (has_qstripe) {
+		/* RAID6, allocate and map temp space for the Q stripe */
 		q_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
 		if (!q_page) {
 			__free_page(p_page);
 			goto cleanup;
 		}
 		SetPageUptodate(q_page);
+		pointers[rbio->real_stripes - 1] = kmap(q_page);
 	}
 
 	atomic_set(&rbio->error, 0);
 
+	/* Map the parity stripe just once */
+	pointers[nr_data] = kmap(p_page);
+
 	for_each_set_bit(pagenr, rbio->dbitmap, rbio->stripe_npages) {
 		struct page *p;
 		void *parity;
@@ -2378,16 +2383,8 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 			pointers[stripe] = kmap(p);
 		}
 
-		/* then add the parity stripe */
-		pointers[stripe++] = kmap(p_page);
-
 		if (has_qstripe) {
-			/*
-			 * raid6, add the qstripe and call the
-			 * library function to fill in our p/q
-			 */
-			pointers[stripe++] = kmap(q_page);
-
+			/* RAID6, call the library function to fill in our P/Q */
 			raid6_call.gen_syndrome(rbio->real_stripes, PAGE_SIZE,
 						pointers);
 		} else {
@@ -2408,12 +2405,14 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 
 		for (stripe = 0; stripe < nr_data; stripe++)
 			kunmap(page_in_rbio(rbio, stripe, pagenr, 0));
-		kunmap(p_page);
 	}
 
+	kunmap(p_page);
 	__free_page(p_page);
-	if (q_page)
+	if (q_page) {
+		kunmap(q_page);
 		__free_page(q_page);
+	}
 
 writeback:
 	/*

fs/btrfs/reflink.c

@@ -553,6 +553,24 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 		 */
 		btrfs_release_path(path);
 
+		/*
+		 * When using NO_HOLES and we are cloning a range that covers
+		 * only a hole (no extents) into a range beyond the current
+		 * i_size, punching a hole in the target range will not create
+		 * an extent map defining a hole, because the range starts at or
+		 * beyond current i_size. If the file previously had an i_size
+		 * greater than the new i_size set by this clone operation, we
+		 * need to make sure the next fsync is a full fsync, so that it
+		 * detects and logs a hole covering a range from the current
+		 * i_size to the new i_size. If the clone range covers extents,
+		 * besides a hole, then we know the full sync flag was already
+		 * set by previous calls to btrfs_replace_file_extents() that
+		 * replaced file extent items.
+		 */
+		if (last_dest_end >= i_size_read(inode))
+			set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+				&BTRFS_I(inode)->runtime_flags);
+
 		ret = btrfs_replace_file_extents(inode, path, last_dest_end,
 				destoff + len - 1, NULL, &trans);
 		if (ret)

fs/btrfs/scrub.c

@@ -3767,6 +3767,13 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			 * commit_transactions.
 			 */
 			ro_set = 0;
+		} else if (ret == -ETXTBSY) {
+			btrfs_warn(fs_info,
+		   "skipping scrub of block group %llu due to active swapfile",
+				   cache->start);
+			scrub_pause_off(fs_info);
+			ret = 0;
+			goto skip_unfreeze;
 		} else {
 			btrfs_warn(fs_info,
 				   "failed setting block group ro: %d", ret);
@@ -3862,7 +3869,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		} else {
 			spin_unlock(&cache->lock);
 		}
-
+skip_unfreeze:
 		btrfs_unfreeze_block_group(cache);
 		btrfs_put_block_group(cache);
 		if (ret)

fs/btrfs/tree-checker.c

@@ -1453,22 +1453,14 @@ static int check_extent_data_ref(struct extent_buffer *leaf,
 		return -EUCLEAN;
 	}
 	for (; ptr < end; ptr += sizeof(*dref)) {
-		u64 root_objectid;
-		u64 owner;
 		u64 offset;
-		u64 hash;
 
+		/*
+		 * We cannot check the extent_data_ref hash due to possible
+		 * overflow from the leaf due to hash collisions.
+		 */
 		dref = (struct btrfs_extent_data_ref *)ptr;
-		root_objectid = btrfs_extent_data_ref_root(leaf, dref);
-		owner = btrfs_extent_data_ref_objectid(leaf, dref);
 		offset = btrfs_extent_data_ref_offset(leaf, dref);
-		hash = hash_extent_data_ref(root_objectid, owner, offset);
-		if (unlikely(hash != key->offset)) {
-			extent_err(leaf, slot,
-	"invalid extent data ref hash, item has 0x%016llx key has 0x%016llx",
-				   hash, key->offset);
-			return -EUCLEAN;
-		}
 		if (unlikely(!IS_ALIGNED(offset, leaf->fs_info->sectorsize))) {
 			extent_err(leaf, slot,
 	"invalid extent data backref offset, have %llu expect aligned to %u",

fs/btrfs/tree-log.c

@@ -3174,16 +3174,13 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	root_log_ctx.log_transid = log_root_tree->log_transid;
 
 	if (btrfs_is_zoned(fs_info)) {
-		mutex_lock(&fs_info->tree_root->log_mutex);
 		if (!log_root_tree->node) {
 			ret = btrfs_alloc_log_tree_node(trans, log_root_tree);
 			if (ret) {
-				mutex_unlock(&fs_info->tree_log_mutex);
 				mutex_unlock(&log_root_tree->log_mutex);
 				goto out;
 			}
 		}
-		mutex_unlock(&fs_info->tree_root->log_mutex);
 	}
 
 	/*