ext4: add prefetching for block allocation bitmaps

This should significantly improve bitmap loading, especially for flex
groups as it tries to load all bitmaps within a flex.group instead of
one by one synchronously.

Prefetching is done in 8 * flex_bg groups, so it should be 8
read-ahead reads for a single allocating thread. At the end of
allocation the thread waits for read-ahead completion and initializes
buddy information so that read-aheads are not lost in case of memory
pressure.

At cr=0 the number of prefetching IOs is limited per allocation
context to prevent a situation when mballoc loads thousands of bitmaps
looking for a perfect group and ignoring groups with good chunks.

Together with the patch "ext4: limit scanning of uninitialized groups"
the mount time (which includes few tiny allocations) of a 1PB
filesystem is reduced significantly:

               0% full    50%-full unpatched    patched
  mount time       33s                9279s       563s

[ Restructured by tytso; removed the state flags in the allocation
  context, so it can be used to lazily prefetch the allocation bitmaps
  immediately after the file system is mounted.  Skip prefetching
  block groups which are uninitialized.  Finally pass in the
  REQ_RAHEAD flag to the block layer while prefetching. ]

Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>

1 files changed, 11 insertions, 3 deletions

diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 1ba46d87cdf1..1e2b1b4093aa 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -413,7 +413,8 @@ verified:
  * Return buffer_head on success or an ERR_PTR in case of failure.
  */
 struct buffer_head *
-ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group)
+ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group,
+			      bool ignore_locked)
 {
 	struct ext4_group_desc *desc;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -441,6 +442,12 @@ ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group)
 		return ERR_PTR(-ENOMEM);
 	}
 
+	if (ignore_locked && buffer_locked(bh)) {
+		/* buffer under IO already, return if called for prefetching */
+		put_bh(bh);
+		return NULL;
+	}
+
 	if (bitmap_uptodate(bh))
 		goto verify;
 
@@ -490,7 +497,8 @@ ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group)
 	trace_ext4_read_block_bitmap_load(sb, block_group);
 	bh->b_end_io = ext4_end_bitmap_read;
 	get_bh(bh);
-	submit_bh(REQ_OP_READ, REQ_META | REQ_PRIO, bh);
+	submit_bh(REQ_OP_READ, REQ_META | REQ_PRIO |
+		  (ignore_locked ? REQ_RAHEAD : 0), bh);
 	return bh;
 verify:
 	err = ext4_validate_block_bitmap(sb, desc, block_group, bh);
@@ -534,7 +542,7 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 	struct buffer_head *bh;
 	int err;
 
-	bh = ext4_read_block_bitmap_nowait(sb, block_group);
+	bh = ext4_read_block_bitmap_nowait(sb, block_group, false);
 	if (IS_ERR(bh))
 		return bh;
 	err = ext4_wait_block_bitmap(sb, block_group, bh);