mirror_zfs/module/zfs/space_reftree.c
Gvozden Neskovic ee36c709c3 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 <neskovic@gmail.com>
Signed-off-by: Richard Elling <richard.elling@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #5033
2016-08-31 14:35:34 -07:00

154 lines
4.2 KiB
C

/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* or http://www.opensolaris.org/os/licensing.
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
* Copyright 2009 Sun Microsystems, Inc. All rights reserved.
* Use is subject to license terms.
*/
/*
* Copyright (c) 2013 by Delphix. All rights reserved.
*/
#include <sys/zfs_context.h>
#include <sys/range_tree.h>
#include <sys/space_reftree.h>
/*
* Space reference trees.
*
* A range tree is a collection of integers. Every integer is either
* in the tree, or it's not. A space reference tree generalizes
* the idea: it allows its members to have arbitrary reference counts,
* as opposed to the implicit reference count of 0 or 1 in a range tree.
* This representation comes in handy when computing the union or
* intersection of multiple space maps. For example, the union of
* N range trees is the subset of the reference tree with refcnt >= 1.
* The intersection of N range trees is the subset with refcnt >= N.
*
* [It's very much like a Fourier transform. Unions and intersections
* are hard to perform in the 'range tree domain', so we convert the trees
* into the 'reference count domain', where it's trivial, then invert.]
*
* vdev_dtl_reassess() uses computations of this form to determine
* DTL_MISSING and DTL_OUTAGE for interior vdevs -- e.g. a RAID-Z vdev
* has an outage wherever refcnt >= vdev_nparity + 1, and a mirror vdev
* has an outage wherever refcnt >= vdev_children.
*/
static int
space_reftree_compare(const void *x1, const void *x2)
{
const space_ref_t *sr1 = (const space_ref_t *)x1;
const space_ref_t *sr2 = (const space_ref_t *)x2;
int cmp = AVL_CMP(sr1->sr_offset, sr2->sr_offset);
if (likely(cmp))
return (cmp);
return (AVL_PCMP(sr1, sr2));
}
void
space_reftree_create(avl_tree_t *t)
{
avl_create(t, space_reftree_compare,
sizeof (space_ref_t), offsetof(space_ref_t, sr_node));
}
void
space_reftree_destroy(avl_tree_t *t)
{
space_ref_t *sr;
void *cookie = NULL;
while ((sr = avl_destroy_nodes(t, &cookie)) != NULL)
kmem_free(sr, sizeof (*sr));
avl_destroy(t);
}
static void
space_reftree_add_node(avl_tree_t *t, uint64_t offset, int64_t refcnt)
{
space_ref_t *sr;
sr = kmem_alloc(sizeof (*sr), KM_SLEEP);
sr->sr_offset = offset;
sr->sr_refcnt = refcnt;
avl_add(t, sr);
}
void
space_reftree_add_seg(avl_tree_t *t, uint64_t start, uint64_t end,
int64_t refcnt)
{
space_reftree_add_node(t, start, refcnt);
space_reftree_add_node(t, end, -refcnt);
}
/*
* Convert (or add) a range tree into a reference tree.
*/
void
space_reftree_add_map(avl_tree_t *t, range_tree_t *rt, int64_t refcnt)
{
range_seg_t *rs;
ASSERT(MUTEX_HELD(rt->rt_lock));
for (rs = avl_first(&rt->rt_root); rs; rs = AVL_NEXT(&rt->rt_root, rs))
space_reftree_add_seg(t, rs->rs_start, rs->rs_end, refcnt);
}
/*
* Convert a reference tree into a range tree. The range tree will contain
* all members of the reference tree for which refcnt >= minref.
*/
void
space_reftree_generate_map(avl_tree_t *t, range_tree_t *rt, int64_t minref)
{
uint64_t start = -1ULL;
int64_t refcnt = 0;
space_ref_t *sr;
ASSERT(MUTEX_HELD(rt->rt_lock));
range_tree_vacate(rt, NULL, NULL);
for (sr = avl_first(t); sr != NULL; sr = AVL_NEXT(t, sr)) {
refcnt += sr->sr_refcnt;
if (refcnt >= minref) {
if (start == -1ULL) {
start = sr->sr_offset;
}
} else {
if (start != -1ULL) {
uint64_t end = sr->sr_offset;
ASSERT(start <= end);
if (end > start)
range_tree_add(rt, start, end - start);
start = -1ULL;
}
}
}
ASSERT(refcnt == 0);
ASSERT(start == -1ULL);
}