From ee36c709c3d5f7040e1bd11f5c75318aa03e789f Mon Sep 17 00:00:00 2001 From: Gvozden Neskovic Date: Sat, 27 Aug 2016 20:12:53 +0200 Subject: Performance optimization of AVL tree comparator functions perf: 2.75x faster ddt_entry_compare() First 256bits of ddt_key_t is a block checksum, which are expected to be close to random data. Hence, on average, comparison only needs to look at first few bytes of the keys. To reduce number of conditional jump instructions, the result is computed as: sign(memcmp(k1, k2)). Sign of an integer 'a' can be obtained as: `(0 < a) - (a < 0)` := {-1, 0, 1} , which is computed efficiently. Synthetic performance evaluation of original and new algorithm over 1G random keys on 2.6GHz Intel(R) Xeon(R) CPU E5-2660 v3: old 6.85789 s new 2.49089 s perf: 2.8x faster vdev_queue_offset_compare() and vdev_queue_timestamp_compare() Compute the result directly instead of using conditionals perf: zfs_range_compare() Speedup between 1.1x - 2.5x, depending on compiler version and optimization level. perf: spa_error_entry_compare() `bcmp()` is not suitable for comparator use. Use `memcmp()` instead. perf: 2.8x faster metaslab_compare() and metaslab_rangesize_compare() perf: 2.8x faster zil_bp_compare() perf: 2.8x faster mze_compare() perf: faster dbuf_compare() perf: faster compares in spa_misc perf: 2.8x faster layout_hash_compare() perf: 2.8x faster space_reftree_compare() perf: libzfs: faster avl tree comparators perf: guid_compare() perf: dsl_deadlist_compare() perf: perm_set_compare() perf: 2x faster range_tree_seg_compare() perf: faster unique_compare() perf: faster vdev_cache _compare() perf: faster vdev_uberblock_compare() perf: faster fuid _compare() perf: faster zfs_znode_hold_compare() Signed-off-by: Gvozden Neskovic Signed-off-by: Richard Elling Signed-off-by: Brian Behlendorf Closes #5033 --- module/zfs/dnode.c | 26 +++++++------------------- 1 file changed, 7 insertions(+), 19 deletions(-) (limited to 'module/zfs/dnode.c') diff --git a/module/zfs/dnode.c b/module/zfs/dnode.c index 8015f54ed..abc004bd4 100644 --- a/module/zfs/dnode.c +++ b/module/zfs/dnode.c @@ -69,19 +69,13 @@ dbuf_compare(const void *x1, const void *x2) const dmu_buf_impl_t *d1 = x1; const dmu_buf_impl_t *d2 = x2; - if (d1->db_level < d2->db_level) { - return (-1); - } - if (d1->db_level > d2->db_level) { - return (1); - } + int cmp = AVL_CMP(d1->db_level, d2->db_level); + if (likely(cmp)) + return (cmp); - if (d1->db_blkid < d2->db_blkid) { - return (-1); - } - if (d1->db_blkid > d2->db_blkid) { - return (1); - } + cmp = AVL_CMP(d1->db_blkid, d2->db_blkid); + if (likely(cmp)) + return (cmp); if (d1->db_state == DB_SEARCH) { ASSERT3S(d2->db_state, !=, DB_SEARCH); @@ -91,13 +85,7 @@ dbuf_compare(const void *x1, const void *x2) return (1); } - if ((uintptr_t)d1 < (uintptr_t)d2) { - return (-1); - } - if ((uintptr_t)d1 > (uintptr_t)d2) { - return (1); - } - return (0); + return (AVL_PCMP(d1, d2)); } /* ARGSUSED */ -- cgit v1.2.3