1
0
mirror of https://github.com/golang/go synced 2024-11-18 08:04:40 -07:00
go/test/codegen/condmove.go

454 lines
6.2 KiB
Go
Raw Normal View History

cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
// asmcheck
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
package codegen
func cmovint(c int) int {
x := c + 4
if x < 0 {
x = 182
}
// amd64:"CMOVQLT"
// arm64:"CSEL\tLT"
// ppc64x:"ISEL\t[$]0"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return x
}
func cmovchan(x, y chan int) chan int {
if x != y {
x = y
}
// amd64:"CMOVQNE"
// arm64:"CSEL\tNE"
// ppc64x:"ISEL\t[$]2"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return x
}
func cmovuintptr(x, y uintptr) uintptr {
if x < y {
x = -y
}
// amd64:"CMOVQ(HI|CS)"
// arm64:"CSNEG\tLS"
// ppc64x:"ISEL\t[$]1"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return x
}
func cmov32bit(x, y uint32) uint32 {
if x < y {
x = -y
}
// amd64:"CMOVL(HI|CS)"
// arm64:"CSNEG\t(LS|HS)"
// ppc64x:"ISEL\t[$]1"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return x
}
func cmov16bit(x, y uint16) uint16 {
if x < y {
x = -y
}
// amd64:"CMOVW(HI|CS)"
// arm64:"CSNEG\t(LS|HS)"
// ppc64x:"ISEL\t[$]0"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return x
}
// Floating point comparison. For EQ/NE, we must
// generate special code to handle NaNs.
func cmovfloateq(x, y float64) int {
a := 128
if x == y {
a = 256
}
// amd64:"CMOVQNE","CMOVQPC"
// arm64:"CSEL\tEQ"
// ppc64x:"ISEL\t[$]2"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return a
}
func cmovfloatne(x, y float64) int {
a := 128
if x != y {
a = 256
}
// amd64:"CMOVQNE","CMOVQPS"
// arm64:"CSEL\tNE"
// ppc64x:"ISEL\t[$]2"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return a
}
//go:noinline
func frexp(f float64) (frac float64, exp int) {
return 1.0, 4
}
//go:noinline
func ldexp(frac float64, exp int) float64 {
return 1.0
}
// Generate a CMOV with a floating comparison and integer move.
func cmovfloatint2(x, y float64) float64 {
yfr, yexp := 4.0, 5
r := x
for r >= y {
rfr, rexp := frexp(r)
if rfr < yfr {
rexp = rexp - 1
}
// amd64:"CMOVQHI"
// arm64:"CSEL\tMI"
// ppc64x:"ISEL\t[$]0"
// wasm:"Select"
r = r - ldexp(y, rexp-yexp)
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
}
return r
}
func cmovloaded(x [4]int, y int) int {
if x[2] != 0 {
y = x[2]
} else {
y = y >> 2
}
// amd64:"CMOVQNE"
// arm64:"CSEL\tNE"
// ppc64x:"ISEL\t[$]2"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return y
}
func cmovuintptr2(x, y uintptr) uintptr {
a := x * 2
if a == 0 {
a = 256
}
// amd64:"CMOVQEQ"
// arm64:"CSEL\tEQ"
// ppc64x:"ISEL\t[$]2"
// wasm:"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return a
}
// Floating point CMOVs are not supported by amd64/arm64/ppc64x
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
func cmovfloatmove(x, y int) float64 {
a := 1.0
if x <= y {
a = 2.0
}
// amd64:-"CMOV"
// arm64:-"CSEL"
// ppc64x:-"ISEL"
// wasm:-"Select"
cmd/compile: implement CMOV on amd64 This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <rasky@develer.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2018-03-05 12:59:40 -07:00
return a
}
// On amd64, the following patterns trigger comparison inversion.
// Test that we correctly invert the CMOV condition
var gsink int64
var gusink uint64
func cmovinvert1(x, y int64) int64 {
if x < gsink {
y = -y
}
// amd64:"CMOVQGT"
return y
}
func cmovinvert2(x, y int64) int64 {
if x <= gsink {
y = -y
}
// amd64:"CMOVQGE"
return y
}
func cmovinvert3(x, y int64) int64 {
if x == gsink {
y = -y
}
// amd64:"CMOVQEQ"
return y
}
func cmovinvert4(x, y int64) int64 {
if x != gsink {
y = -y
}
// amd64:"CMOVQNE"
return y
}
func cmovinvert5(x, y uint64) uint64 {
if x > gusink {
y = -y
}
// amd64:"CMOVQCS"
return y
}
func cmovinvert6(x, y uint64) uint64 {
if x >= gusink {
y = -y
}
// amd64:"CMOVQLS"
return y
}
func cmovload(a []int, i int, b bool) int {
if b {
i++
}
// See issue 26306
// amd64:-"CMOVQNE"
return a[i]
}
func cmovstore(a []int, i int, b bool) {
if b {
i++
}
// amd64:"CMOVQNE"
a[i] = 7
}
var r0, r1, r2, r3, r4, r5 int
func cmovinc(cond bool, a, b, c int) {
var x0, x1 int
if cond {
x0 = a
} else {
x0 = b + 1
}
// arm64:"CSINC\tNE", -"CSEL"
r0 = x0
if cond {
x1 = b + 1
} else {
x1 = a
}
// arm64:"CSINC\tEQ", -"CSEL"
r1 = x1
if cond {
c++
}
// arm64:"CSINC\tEQ", -"CSEL"
r2 = c
}
func cmovinv(cond bool, a, b int) {
var x0, x1 int
if cond {
x0 = a
} else {
x0 = ^b
}
// arm64:"CSINV\tNE", -"CSEL"
r0 = x0
if cond {
x1 = ^b
} else {
x1 = a
}
// arm64:"CSINV\tEQ", -"CSEL"
r1 = x1
}
func cmovneg(cond bool, a, b, c int) {
var x0, x1 int
if cond {
x0 = a
} else {
x0 = -b
}
// arm64:"CSNEG\tNE", -"CSEL"
r0 = x0
if cond {
x1 = -b
} else {
x1 = a
}
// arm64:"CSNEG\tEQ", -"CSEL"
r1 = x1
}
func cmovsetm(cond bool, x int) {
var x0, x1 int
if cond {
x0 = -1
} else {
x0 = 0
}
// arm64:"CSETM\tNE", -"CSEL"
r0 = x0
if cond {
x1 = 0
} else {
x1 = -1
}
// arm64:"CSETM\tEQ", -"CSEL"
r1 = x1
}
func cmovFcmp0(s, t float64, a, b int) {
var x0, x1, x2, x3, x4, x5 int
if s < t {
x0 = a
} else {
x0 = b + 1
}
// arm64:"CSINC\tMI", -"CSEL"
r0 = x0
if s <= t {
x1 = a
} else {
x1 = ^b
}
// arm64:"CSINV\tLS", -"CSEL"
r1 = x1
if s > t {
x2 = a
} else {
x2 = -b
}
// arm64:"CSNEG\tMI", -"CSEL"
r2 = x2
if s >= t {
x3 = -1
} else {
x3 = 0
}
// arm64:"CSETM\tLS", -"CSEL"
r3 = x3
if s == t {
x4 = a
} else {
x4 = b + 1
}
// arm64:"CSINC\tEQ", -"CSEL"
r4 = x4
if s != t {
x5 = a
} else {
x5 = b + 1
}
// arm64:"CSINC\tNE", -"CSEL"
r5 = x5
}
func cmovFcmp1(s, t float64, a, b int) {
var x0, x1, x2, x3, x4, x5 int
if s < t {
x0 = b + 1
} else {
x0 = a
}
// arm64:"CSINC\tPL", -"CSEL"
r0 = x0
if s <= t {
x1 = ^b
} else {
x1 = a
}
// arm64:"CSINV\tHI", -"CSEL"
r1 = x1
if s > t {
x2 = -b
} else {
x2 = a
}
// arm64:"CSNEG\tPL", -"CSEL"
r2 = x2
if s >= t {
x3 = 0
} else {
x3 = -1
}
// arm64:"CSETM\tHI", -"CSEL"
r3 = x3
if s == t {
x4 = b + 1
} else {
x4 = a
}
// arm64:"CSINC\tNE", -"CSEL"
r4 = x4
if s != t {
x5 = b + 1
} else {
x5 = a
}
// arm64:"CSINC\tEQ", -"CSEL"
r5 = x5
}
func cmovzero1(c bool) int {
var x int
if c {
x = 182
}
// loong64:"MASKEQZ", -"MASKNEZ"
return x
}
func cmovzero2(c bool) int {
var x int
if !c {
x = 182
}
// loong64:"MASKNEZ", -"MASKEQZ"
return x
}
// Conditionally selecting between a value or 0 can be done without
// an extra load of 0 to a register on PPC64 by using R0 (which always
// holds the value $0) instead. Verify both cases where either arg1
// or arg2 is zero.
func cmovzeroreg0(a, b int) int {
x := 0
if a == b {
x = a
}
// ppc64x:"ISEL\t[$]2, R[0-9]+, R0, R[0-9]+"
return x
}
func cmovzeroreg1(a, b int) int {
x := a
if a == b {
x = 0
}
// ppc64x:"ISEL\t[$]2, R0, R[0-9]+, R[0-9]+"
return x
}