aboutsummaryrefslogtreecommitdiffstats
path: root/fs/btrfs/volumes.h
diff options
context:
space:
mode:
authorQu Wenruo <wqu@suse.com>2021-12-14 21:01:45 +0800
committerDavid Sterba <dsterba@suse.com>2022-01-07 14:18:26 +0100
commitf26c92386028563a1e988bb277c6c5ce2e8010ee (patch)
tree37d9c052b6eb5360e7dc911f50f18694f2d7a23e /fs/btrfs/volumes.h
parentbtrfs: scrub: use btrfs_path::reada for extent tree readahead (diff)
downloadlinux-dev-f26c92386028563a1e988bb277c6c5ce2e8010ee.tar.xz
linux-dev-f26c92386028563a1e988bb277c6c5ce2e8010ee.zip
btrfs: remove reada infrastructure
Currently there is only one user for btrfs metadata readahead, and that's scrub. But even for the single user, it's not providing the correct functionality it needs, as scrub needs reada for commit root, which current readahead can't provide. (Although it's pretty easy to add such feature). Despite this, there are some extra problems related to metadata readahead: - Duplicated feature with btrfs_path::reada - Partly duplicated feature of btrfs_fs_info::buffer_radix Btrfs already caches its metadata in buffer_radix, while readahead tries to read the tree block no matter if it's already cached. - Poor layer separation Metadata readahead works kinda at device level. This is definitely not the correct layer it should be, since metadata is at btrfs logical address space, it should not bother device at all. This brings extra chance for bugs to sneak in, while brings unnecessary complexity. - Dead code In the very beginning of scrub.c we have #undef DEBUG, rendering all the debug related code useless and unable to test. Thus here I purpose to remove the metadata readahead mechanism completely. [BENCHMARK] There is a full benchmark for the scrub performance difference using the old btrfs_reada_add() and btrfs_path::reada. For the worst case (no dirty metadata, slow HDD), there could be a 5% performance drop for scrub. For other cases (even SATA SSD), there is no distinguishable performance difference. The number is reported scrub speed, in MiB/s. The resolution is limited by the reported duration, which only has a resolution of 1 second. Old New Diff SSD 455.3 466.332 +2.42% HDD 103.927 98.012 -5.69% Comprehensive test methodology is in the cover letter of the patch. Signed-off-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Diffstat (limited to '')
-rw-r--r--fs/btrfs/volumes.h7
1 files changed, 0 insertions, 7 deletions
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 9cf1d93a3d66..005c9e2a491a 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -123,13 +123,6 @@ struct btrfs_device {
/* per-device scrub information */
struct scrub_ctx *scrub_ctx;
- /* readahead state */
- atomic_t reada_in_flight;
- u64 reada_next;
- struct reada_zone *reada_curr_zone;
- struct radix_tree_root reada_zones;
- struct radix_tree_root reada_extents;
-
/* disk I/O failure stats. For detailed description refer to
* enum btrfs_dev_stat_values in ioctl.h */
int dev_stats_valid;