btrfs: introduce compressed_bio::pending_sectors to trace compressed bio

For btrfs_submit_compressed_read() and btrfs_submit_compressed_write(),
we have a pretty weird dance around compressed_bio::pending_bios:

  btrfs_submit_compressed_read/write()
  {
	cb = kmalloc()
	refcount_set(&cb->pending_bios, 0);
	bio = btrfs_alloc_bio();

	/* NOTE here, we haven't yet submitted any bio */
	refcount_set(&cb->pending_bios, 1);

	for (pg_index = 0; pg_index < cb->nr_pages; pg_index++) {
		if (submit) {
			/* Here we submit bio, but we always have one
			 * extra pending_bios */
			refcount_inc(&cb->pending_bios);
			ret = btrfs_map_bio();
		}
	}

	/* Submit the last bio */
	ret = btrfs_map_bio();
  }

There are two reasons why we do this:

- compressed_bio::pending_bios is a refcount
  Thus if it's reduced to 0, it can not be increased again.

- To ensure the compressed_bio is not freed by some submitted bios
  If the submitted bio is finished before the next bio submitted,
  we can free the compressed_bio completely.

But the above code is sometimes confusing, and we can do it better by
introducing a new member, compressed_bio::pending_sectors.

Now we use compressed_bio::pending_sectors to indicate whether we have
any pending sectors under IO or not yet submitted.

If pending_sectors == 0, we're definitely the last bio of compressed_bio,
and is OK to release the compressed bio.

Now the workflow looks like this:

  btrfs_submit_compressed_read/write()
  {
	cb = kmalloc()
	atomic_set(&cb->pending_bios, 0);
	refcount_set(&cb->pending_sectors,
		     compressed_len >> sectorsize_bits);
	bio = btrfs_alloc_bio();

	for (pg_index = 0; pg_index < cb->nr_pages; pg_index++) {
		if (submit) {
			refcount_inc(&cb->pending_bios);
			ret = btrfs_map_bio();
		}
	}

	/* Submit the last bio */
	refcount_inc(&cb->pending_bios);
	ret = btrfs_map_bio();
  }

For now we still need pending_bios for later error handling, but will
remove pending_bios eventually after properly handling the errors.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>

fs/btrfs/compression.c

@@ -193,6 +193,38 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio,
 	return 0;
 }
 
+/*
+ * Reduce bio and io accounting for a compressed_bio with its corresponding bio.
+ *
+ * Return true if there is no pending bio nor io.
+ * Return false otherwise.
+ */
+static bool dec_and_test_compressed_bio(struct compressed_bio *cb, struct bio *bio)
+{
+	struct btrfs_fs_info *fs_info = btrfs_sb(cb->inode->i_sb);
+	unsigned int bi_size = 0;
+	bool last_io = false;
+	struct bio_vec *bvec;
+	struct bvec_iter_all iter_all;
+
+	/*
+	 * At endio time, bi_iter.bi_size doesn't represent the real bio size.
+	 * Thus here we have to iterate through all segments to grab correct
+	 * bio size.
+	 */
+	bio_for_each_segment_all(bvec, bio, iter_all)
+		bi_size += bvec->bv_len;
+
+	if (bio->bi_status)
+		cb->errors = 1;
+
+	ASSERT(bi_size && bi_size <= cb->compressed_len);
+	last_io = refcount_sub_and_test(bi_size >> fs_info->sectorsize_bits,
+					&cb->pending_sectors);
+	atomic_dec(&cb->pending_bios);
+	return last_io;
+}
+
 /* when we finish reading compressed pages from the disk, we
  * decompress them and then run the bio end_io routines on the
  * decompressed pages (in the inode address space).
@@ -212,13 +244,7 @@ static void end_compressed_bio_read(struct bio *bio)
 	unsigned int mirror = btrfs_bio(bio)->mirror_num;
 	int ret = 0;
 
-	if (bio->bi_status)
-		cb->errors = 1;
-
-	/* if there are more bios still pending for this compressed
-	 * extent, just exit
-	 */
-	if (!refcount_dec_and_test(&cb->pending_bios))
+	if (!dec_and_test_compressed_bio(cb, bio))
 		goto out;
 
 	/*
@@ -336,13 +362,7 @@ static void end_compressed_bio_write(struct bio *bio)
 	struct page *page;
 	unsigned int index;
 
-	if (bio->bi_status)
-		cb->errors = 1;
-
-	/* if there are more bios still pending for this compressed
-	 * extent, just exit
-	 */
-	if (!refcount_dec_and_test(&cb->pending_bios))
+	if (!dec_and_test_compressed_bio(cb, bio))
 		goto out;
 
 	/* ok, we're the last bio for this extent, step one is to
@@ -408,7 +428,8 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 	cb = kmalloc(compressed_bio_size(fs_info, compressed_len), GFP_NOFS);
 	if (!cb)
 		return BLK_STS_RESOURCE;
-	refcount_set(&cb->pending_bios, 0);
+	atomic_set(&cb->pending_bios, 0);
+	refcount_set(&cb->pending_sectors, compressed_len >> fs_info->sectorsize_bits);
 	cb->errors = 0;
 	cb->inode = &inode->vfs_inode;
 	cb->start = start;
@@ -442,7 +463,6 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 		bio->bi_opf |= REQ_CGROUP_PUNT;
 		kthread_associate_blkcg(blkcg_css);
 	}
-	refcount_set(&cb->pending_bios, 1);
 
 	/* create and submit bios for the compressed pages */
 	bytes_left = compressed_len;
@@ -470,13 +490,7 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 
 		page->mapping = NULL;
 		if (submit || len < PAGE_SIZE) {
-			/*
-			 * inc the count before we submit the bio so
-			 * we know the end IO handler won't happen before
-			 * we inc the count.  Otherwise, the cb might get
-			 * freed before we're done setting it up
-			 */
-			refcount_inc(&cb->pending_bios);
+			atomic_inc(&cb->pending_bios);
 			ret = btrfs_bio_wq_end_io(fs_info, bio,
 						  BTRFS_WQ_ENDIO_DATA);
 			BUG_ON(ret); /* -ENOMEM */
@@ -515,6 +529,7 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 		cond_resched();
 	}
 
+	atomic_inc(&cb->pending_bios);
 	ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA);
 	BUG_ON(ret); /* -ENOMEM */
 
@@ -734,7 +749,8 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	if (!cb)
 		goto out;
 
-	refcount_set(&cb->pending_bios, 0);
+	atomic_set(&cb->pending_bios, 0);
+	refcount_set(&cb->pending_sectors, compressed_len >> fs_info->sectorsize_bits);
 	cb->errors = 0;
 	cb->inode = inode;
 	cb->mirror_num = mirror_num;
@@ -779,7 +795,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	comp_bio->bi_opf = REQ_OP_READ;
 	comp_bio->bi_private = cb;
 	comp_bio->bi_end_io = end_compressed_bio_read;
-	refcount_set(&cb->pending_bios, 1);
 
 	for (pg_index = 0; pg_index < nr_pages; pg_index++) {
 		u32 pg_len = PAGE_SIZE;
@@ -808,18 +823,11 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		if (submit || bio_add_page(comp_bio, page, pg_len, 0) < pg_len) {
 			unsigned int nr_sectors;
 
+			atomic_inc(&cb->pending_bios);
 			ret = btrfs_bio_wq_end_io(fs_info, comp_bio,
 						  BTRFS_WQ_ENDIO_DATA);
 			BUG_ON(ret); /* -ENOMEM */
 
-			/*
-			 * inc the count before we submit the bio so
-			 * we know the end IO handler won't happen before
-			 * we inc the count.  Otherwise, the cb might get
-			 * freed before we're done setting it up
-			 */
-			refcount_inc(&cb->pending_bios);
-
 			ret = btrfs_lookup_bio_sums(inode, comp_bio, sums);
 			BUG_ON(ret); /* -ENOMEM */
 
@@ -844,6 +852,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		cur_disk_byte += pg_len;
 	}
 
+	atomic_inc(&cb->pending_bios);
 	ret = btrfs_bio_wq_end_io(fs_info, comp_bio, BTRFS_WQ_ENDIO_DATA);
 	BUG_ON(ret); /* -ENOMEM */
 

fs/btrfs/compression.h

@@ -28,8 +28,11 @@ struct btrfs_inode;
 #define	BTRFS_ZLIB_DEFAULT_LEVEL		3
 
 struct compressed_bio {
-	/* number of bios pending for this compressed extent */
-	refcount_t pending_bios;
+	/* Number of bios pending for this compressed extent */
+	atomic_t pending_bios;
+
+	/* Number of sectors with unfinished IO (unsubmitted or unfinished) */
+	refcount_t pending_sectors;
 
 	/* Number of compressed pages in the array */
 	unsigned int nr_pages;