diff options
author | Filipe Manana <fdmanana@gmail.com> | 2014-02-25 14:15:13 +0000 |
---|---|---|
committer | Josef Bacik <jbacik@fb.com> | 2014-03-10 15:16:57 -0400 |
commit | 176840b3aa3cb795ddec4fc665ffbd707abff906 (patch) | |
tree | c709c05160ab4a4d104d16460e59ecc4cc3cd7a4 /fs/btrfs/file.c | |
parent | f2071b21553bf8f1eae583e32b9068393f61cbe9 (diff) |
Btrfs: more efficient btrfs_drop_extent_cache
While droping extent map structures from the extent cache that cover our
target range, we would remove each extent map structure from the red black
tree and then add either 1 or 2 new extent map structures if the former
extent map covered sections outside our target range.
This change simply attempts to replace the existing extent map structure
with a new one that covers the subsection we're not interested in, instead
of doing a red black remove operation followed by an insertion operation.
The number of elements in an inode's extent map tree can get very high for large
files under random writes. For example, while running the following test:
sysbench --test=fileio --file-num=1 --file-total-size=10G \
--file-test-mode=rndrw --num-threads=32 --file-block-size=32768 \
--max-requests=500000 --file-rw-ratio=2 [prepare|run]
I captured the following histogram capturing the number of extent_map items
in the red black tree while that test was running:
Count: 122462
Range: 1.000 - 172231.000; Mean: 96415.831; Median: 101855.000; Stddev: 49700.981
Percentiles: 90th: 160120.000; 95th: 166335.000; 99th: 171070.000
1.000 - 5.231: 452 |
5.231 - 187.392: 87 |
187.392 - 585.911: 206 |
585.911 - 1827.438: 623 |
1827.438 - 5695.245: 1962 #
5695.245 - 17744.861: 6204 ####
17744.861 - 55283.764: 21115 ############
55283.764 - 172231.000: 91813 #####################################################
Benchmark:
sysbench --test=fileio --file-num=1 --file-total-size=10G --file-test-mode=rndwr \
--num-threads=64 --file-block-size=32768 --max-requests=0 --max-time=60 \
--file-io-mode=sync --file-fsync-freq=0 [prepare|run]
Before this change: 122.1Mb/sec
After this change: 125.07Mb/sec
(averages of 5 test runs)
Test machine: quad core intel i5-3570K, 32Gb of ram, SSD
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Diffstat (limited to 'fs/btrfs/file.c')
-rw-r--r-- | fs/btrfs/file.c | 16 |
1 files changed, 11 insertions, 5 deletions
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 762ca32bd988..31e48b947060 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -591,7 +591,6 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, clear_bit(EXTENT_FLAG_PINNED, &em->flags); clear_bit(EXTENT_FLAG_LOGGING, &flags); modified = !list_empty(&em->list); - remove_extent_mapping(em_tree, em); if (no_splits) goto next; @@ -622,8 +621,7 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, split->bdev = em->bdev; split->flags = flags; split->compress_type = em->compress_type; - ret = add_extent_mapping(em_tree, split, modified); - BUG_ON(ret); /* Logic error */ + replace_extent_mapping(em_tree, em, split, modified); free_extent_map(split); split = split2; split2 = NULL; @@ -661,12 +659,20 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, split->orig_block_len = 0; } - ret = add_extent_mapping(em_tree, split, modified); - BUG_ON(ret); /* Logic error */ + if (extent_map_in_tree(em)) { + replace_extent_mapping(em_tree, em, split, + modified); + } else { + ret = add_extent_mapping(em_tree, split, + modified); + ASSERT(ret == 0); /* Logic error */ + } free_extent_map(split); split = NULL; } next: + if (extent_map_in_tree(em)) + remove_extent_mapping(em_tree, em); write_unlock(&em_tree->lock); /* once for us */ |