diff options
| author | Meng Zhuo <mzh@golangcn.org> | 2021-09-30 22:46:09 +0800 |
|---|---|---|
| committer | Meng Zhuo <mzh@golangcn.org> | 2021-10-07 14:01:52 +0000 |
| commit | ecb2f231fa41b581319505139f8d5ac779763bee (patch) | |
| tree | 9892a8ece179edd0a1a88273e271255d18295b62 /src/runtime | |
| parent | 6f74ed06c5b0e1d69fb70e89f31f002f18554c79 (diff) | |
| download | go-ecb2f231fa41b581319505139f8d5ac779763bee.tar.xz | |
runtime,sync: using fastrandn instead of modulo reduction
fastrandn is ~50% faster than fastrand() % n.
`ack -v 'fastrand\(\)\s?\%'` finds all modulo on fastrand()
name old time/op new time/op delta
Fastrandn/2 2.86ns ± 0% 1.59ns ± 0% -44.35% (p=0.000 n=9+10)
Fastrandn/3 2.87ns ± 1% 1.59ns ± 0% -44.41% (p=0.000 n=10+9)
Fastrandn/4 2.87ns ± 1% 1.58ns ± 1% -45.10% (p=0.000 n=10+10)
Fastrandn/5 2.86ns ± 1% 1.58ns ± 1% -44.84% (p=0.000 n=10+10)
Change-Id: Ic91f5ca9b9e3b65127bc34792b62fd64fbd13b5c
Reviewed-on: https://go-review.googlesource.com/c/go/+/353269
Trust: Meng Zhuo <mzh@golangcn.org>
Run-TryBot: Meng Zhuo <mzh@golangcn.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Diffstat (limited to 'src/runtime')
| -rw-r--r-- | src/runtime/malloc.go | 4 | ||||
| -rw-r--r-- | src/runtime/mbitmap.go | 2 | ||||
| -rw-r--r-- | src/runtime/proc.go | 2 | ||||
| -rw-r--r-- | src/runtime/stubs.go | 4 | ||||
| -rw-r--r-- | src/runtime/symtab.go | 2 |
5 files changed, 7 insertions, 7 deletions
diff --git a/src/runtime/malloc.go b/src/runtime/malloc.go index 7affe244a2..c389cb1e45 100644 --- a/src/runtime/malloc.go +++ b/src/runtime/malloc.go @@ -1301,7 +1301,7 @@ func fastexprand(mean int) int32 { // x = -log_e(q) * mean // x = log_2(q) * (-log_e(2)) * mean ; Using log_2 for efficiency const randomBitCount = 26 - q := fastrand()%(1<<randomBitCount) + 1 + q := fastrandn(1<<randomBitCount) + 1 qlog := fastlog2(float64(q)) - randomBitCount if qlog > 0 { qlog = 0 @@ -1319,7 +1319,7 @@ func nextSampleNoFP() uintptr { rate = 0x3fffffff } if rate != 0 { - return uintptr(fastrand() % uint32(2*rate)) + return uintptr(fastrandn(uint32(2 * rate))) } return 0 } diff --git a/src/runtime/mbitmap.go b/src/runtime/mbitmap.go index 9363409e36..daf1fcfbc0 100644 --- a/src/runtime/mbitmap.go +++ b/src/runtime/mbitmap.go @@ -974,7 +974,7 @@ func heapBitsSetType(x, size, dataSize uintptr, typ *_type) { // machine instructions. outOfPlace := false - if arenaIndex(x+size-1) != arenaIdx(h.arena) || (doubleCheck && fastrand()%2 == 0) { + if arenaIndex(x+size-1) != arenaIdx(h.arena) || (doubleCheck && fastrandn(2) == 0) { // This object spans heap arenas, so the bitmap may be // discontiguous. Unroll it into the object instead // and then copy it out. diff --git a/src/runtime/proc.go b/src/runtime/proc.go index aa2ba96c8f..c14a1f143c 100644 --- a/src/runtime/proc.go +++ b/src/runtime/proc.go @@ -5840,7 +5840,7 @@ const randomizeScheduler = raceenabled // If the run queue is full, runnext puts g on the global queue. // Executed only by the owner P. func runqput(_p_ *p, gp *g, next bool) { - if randomizeScheduler && next && fastrand()%2 == 0 { + if randomizeScheduler && next && fastrandn(2) == 0 { next = false } diff --git a/src/runtime/stubs.go b/src/runtime/stubs.go index 91380a0d51..ad78363bb6 100644 --- a/src/runtime/stubs.go +++ b/src/runtime/stubs.go @@ -156,8 +156,8 @@ func fastrandn(n uint32) uint32 { return uint32(uint64(fastrand()) * uint64(n) >> 32) } -//go:linkname sync_fastrand sync.fastrand -func sync_fastrand() uint32 { return fastrand() } +//go:linkname sync_fastrandn sync.fastrandn +func sync_fastrandn(n uint32) uint32 { return fastrandn(n) } //go:linkname net_fastrand net.fastrand func net_fastrand() uint32 { return fastrand() } diff --git a/src/runtime/symtab.go b/src/runtime/symtab.go index 7641c491f1..2de518d2e6 100644 --- a/src/runtime/symtab.go +++ b/src/runtime/symtab.go @@ -921,7 +921,7 @@ func pcvalue(f funcInfo, off uint32, targetpc uintptr, cache *pcvalueCache, stri if cache != nil { x := pcvalueCacheKey(targetpc) e := &cache.entries[x] - ci := fastrand() % uint32(len(cache.entries[x])) + ci := fastrandn(uint32(len(cache.entries[x]))) e[ci] = e[0] e[0] = pcvalueCacheEnt{ targetpc: targetpc, |