[PATCH 3/6] fs: Convert block_read_full_page to be synchronous

[Matthew Wilcox (Oracle)]

Use the new blk_completion infrastructure to wait for multiple I/Os.
Also coalesce adjacent buffer heads into a single BIO instead of
submitting one BIO per buffer head.  This doesn't work for fscrypt yet,
so keep the old code around for now.

Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
---
 fs/buffer.c | 90 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 90 insertions(+)

diff --git a/fs/buffer.c b/fs/buffer.c
index 1b0ba1d59966..ccb90081117c 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -2249,6 +2249,87 @@ int block_is_partially_uptodate(struct page *page, unsigned long from,
 }
 EXPORT_SYMBOL(block_is_partially_uptodate);
 
+static void readpage_end_bio(struct bio *bio)
+{
+	struct bio_vec *bvec;
+	struct page *page;
+	struct buffer_head *bh;
+	int i, nr = 0;
+
+	bio_for_each_bvec_all(bvec, bio, i) {
+		size_t offset = 0;
+		size_t max = bvec->bv_offset + bvec->bv_len;
+
+		page = bvec->bv_page;
+		bh = page_buffers(page);
+
+		for (offset = 0; offset < max; offset += bh->b_size,
+				bh = bh->b_this_page) {
+			if (offset < bvec->bv_offset)
+				continue;
+			BUG_ON(bh_offset(bh) != offset);
+			nr++;
+			if (unlikely(bio_flagged(bio, BIO_QUIET)))
+				set_bit(BH_Quiet, &bh->b_state);
+			if (bio->bi_status == BLK_STS_OK)
+				set_buffer_uptodate(bh);
+			else
+				buffer_io_error(bh, ", async page read");
+			unlock_buffer(bh);
+		}
+	}
+
+	if (blk_completion_sub(bio->bi_private, bio->bi_status, nr) < 0)
+		unlock_page(page);
+	bio_put(bio);
+}
+
+static int readpage_submit_bhs(struct page *page, struct blk_completion *cmpl,
+		unsigned int nr, struct buffer_head **bhs)
+{
+	struct bio *bio = NULL;
+	unsigned int i;
+	int err;
+
+	blk_completion_init(cmpl, nr);
+
+	for (i = 0; i < nr; i++) {
+		struct buffer_head *bh = bhs[i];
+		sector_t sector = bh->b_blocknr * (bh->b_size >> 9);
+		bool same_page;
+
+		if (buffer_uptodate(bh)) {
+			end_buffer_async_read(bh, 1);
+			blk_completion_sub(cmpl, BLK_STS_OK, 1);
+			continue;
+		}
+		if (bio) {
+			if (bio_end_sector(bio) == sector &&
+			    __bio_try_merge_page(bio, bh->b_page, bh->b_size,
+					bh_offset(bh), &same_page))
+				continue;
+			submit_bio(bio);
+		}
+		bio = bio_alloc(GFP_NOIO, 1);
+		bio_set_dev(bio, bh->b_bdev);
+		bio->bi_iter.bi_sector = sector;
+		bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh));
+		bio->bi_end_io = readpage_end_bio;
+		bio->bi_private = cmpl;
+		/* Take care of bh's that straddle the end of the device */
+		guard_bio_eod(bio);
+	}
+
+	if (bio)
+		submit_bio(bio);
+
+	err = blk_completion_wait_killable(cmpl);
+	if (!err)
+		return AOP_UPDATED_PAGE;
+	unlock_page(page);
+	return err;
+}
+
 /*
  * Generic "read page" function for block devices that have the normal
  * get_block functionality. This is most of the block device filesystems.
@@ -2258,6 +2339,7 @@ EXPORT_SYMBOL(block_is_partially_uptodate);
  */
 int block_read_full_page(struct page *page, get_block_t *get_block)
 {
+	struct blk_completion *cmpl = kmalloc(sizeof(*cmpl), GFP_NOIO);
 	struct inode *inode = page->mapping->host;
 	sector_t iblock, lblock;
 	struct buffer_head *bh, *head, *arr[MAX_BUF_PER_PAGE];
@@ -2265,6 +2347,9 @@ int block_read_full_page(struct page *page, get_block_t *get_block)
 	int nr, i, err = 0;
 	int fully_mapped = 1;
 
+	if (!cmpl)
+		return -ENOMEM;
+
 	head = create_page_buffers(page, inode, 0);
 	blocksize = head->b_size;
 	bbits = block_size_bits(blocksize);
@@ -2303,6 +2388,7 @@ int block_read_full_page(struct page *page, get_block_t *get_block)
 	} while (i++, iblock++, (bh = bh->b_this_page) != head);
 
 	if (err) {
+		kfree(cmpl);
 		unlock_page(page);
 		return err;
 	}
@@ -2322,6 +2408,10 @@ int block_read_full_page(struct page *page, get_block_t *get_block)
 		mark_buffer_async_read(bh);
 	}
 
+	if (!fscrypt_inode_uses_fs_layer_crypto(inode))
+		return readpage_submit_bhs(page, cmpl, nr, arr);
+	kfree(cmpl);
+
 	/*
 	 * Stage 3: start the IO.  Check for uptodateness
 	 * inside the buffer lock in case another process reading
-- 
2.28.0