x/perf/benchstat: no output when parsing go test results on windows

[brandondube]

What version of Go are you using (go version)?

$ go version
go version go1.20.2 windows/amd64

and

$ go install golang.org/x/perf/cmd/benchstat@latest
go: downloading golang.org/x/perf v0.0.0-20230427221525-d343f6398b76

Does this issue reproduce with the latest release?

did not test

What operating system and processor architecture are you using (go env)?

go env Output

$ go env
set GO111MODULE=
set GOARCH=amd64
set GOBIN=
set GOCACHE=C:\Users\brand\AppData\Local\go-build
set GOENV=C:\Users\brand\AppData\Roaming\go\env
set GOEXE=.exe
set GOEXPERIMENT=
set GOFLAGS=
set GOHOSTARCH=amd64
set GOHOSTOS=windows
set GOINSECURE=
set GOMODCACHE=C:\Users\brand\go\pkg\mod
set GONOPROXY=
set GONOSUMDB=
set GOOS=windows
set GOPATH=C:\Users\brand\go
set GOPRIVATE=
set GOPROXY=https://proxy.golang.org,direct
set GOROOT=C:\Program Files\Go
set GOSUMDB=sum.golang.org
set GOTMPDIR=
set GOTOOLDIR=C:\Program Files\Go\pkg\tool\windows_amd64
set GOVCS=
set GOVERSION=go1.20.2
set GCCGO=gccgo
set GOAMD64=v1
set AR=ar
set CC=gcc
set CXX=g++
set CGO_ENABLED=0
set GOMOD=C:\Users\brand\src\pctl\go.mod
set GOWORK=
set CGO_CFLAGS=-O2 -g
set CGO_CPPFLAGS=
set CGO_CXXFLAGS=-O2 -g
set CGO_FFLAGS=-O2 -g
set CGO_LDFLAGS=-O2 -g
set PKG_CONFIG=pkg-config
set GOGCCFLAGS=-m64 -fno-caret-diagnostics -Qunused-arguments -Wl,--no-gc-sections -fmessage-length=0 -fdebug-prefix-map=C:\Users\brand\AppData\Local\Temp\go-build2102171391=/tmp/go-build -gno-record-gcc-switches

What did you do?

clone https://github.com/brandondube/pctl

in the root pctl folder, go test -count=10 -bench . > bench.txt

This produces the file:

bench.txt

goos: windows
goarch: amd64
pkg: github.com/brandondube/pctl
cpu: AMD Ryzen 9 7950X 16-Core Processor
BenchmarkPIDLoop-32       	556910445	         2.134 ns/op
BenchmarkPIDLoop-32       	536436079	         2.128 ns/op
BenchmarkPIDLoop-32       	543744961	         2.150 ns/op
BenchmarkPIDLoop-32       	571188943	         2.144 ns/op
BenchmarkPIDLoop-32       	553921231	         2.130 ns/op
BenchmarkPIDLoop-32       	560300208	         2.133 ns/op
BenchmarkPIDLoop-32       	557006866	         2.137 ns/op
BenchmarkPIDLoop-32       	568716604	         2.163 ns/op
BenchmarkPIDLoop-32       	548276173	         2.141 ns/op
BenchmarkPIDLoop-32       	560737706	         2.141 ns/op
BenchmarkLPF-32           	372365629	         3.234 ns/op
BenchmarkLPF-32           	365019788	         3.225 ns/op
BenchmarkLPF-32           	362279024	         3.224 ns/op
BenchmarkLPF-32           	368677864	         3.257 ns/op
BenchmarkLPF-32           	363842319	         3.297 ns/op
BenchmarkLPF-32           	363315544	         3.260 ns/op
BenchmarkLPF-32           	372020348	         3.296 ns/op
BenchmarkLPF-32           	366008430	         3.324 ns/op
BenchmarkLPF-32           	362339845	         3.277 ns/op
BenchmarkLPF-32           	362009732	         3.238 ns/op
BenchmarkHPF-32           	451668122	         2.683 ns/op
BenchmarkHPF-32           	436226106	         2.693 ns/op
BenchmarkHPF-32           	462930814	         2.696 ns/op
BenchmarkHPF-32           	451433809	         2.685 ns/op
BenchmarkHPF-32           	450776010	         2.718 ns/op
BenchmarkHPF-32           	440844540	         2.689 ns/op
BenchmarkHPF-32           	440125185	         2.668 ns/op
BenchmarkHPF-32           	445756618	         2.670 ns/op
BenchmarkHPF-32           	440521033	         2.686 ns/op
BenchmarkHPF-32           	443193980	         2.692 ns/op
BenchmarkBiquad-32        	368877667	         3.200 ns/op
BenchmarkBiquad-32        	372806614	         3.208 ns/op
BenchmarkBiquad-32        	374070783	         3.192 ns/op
BenchmarkBiquad-32        	374509742	         3.212 ns/op
BenchmarkBiquad-32        	372645230	         3.209 ns/op
BenchmarkBiquad-32        	373663568	         3.216 ns/op
BenchmarkBiquad-32        	374019950	         3.244 ns/op
BenchmarkBiquad-32        	375003984	         3.215 ns/op
BenchmarkBiquad-32        	362629900	         3.225 ns/op
BenchmarkBiquad-32        	362400795	         3.225 ns/op
BenchmarkStateSpace-32    	163445652	         7.359 ns/op
BenchmarkStateSpace-32    	163384921	         7.364 ns/op
BenchmarkStateSpace-32    	160790014	         7.354 ns/op
BenchmarkStateSpace-32    	163293565	         7.367 ns/op
BenchmarkStateSpace-32    	161149338	         7.352 ns/op
BenchmarkStateSpace-32    	163366044	         7.362 ns/op
BenchmarkStateSpace-32    	162739160	         7.326 ns/op
BenchmarkStateSpace-32    	161796784	         7.374 ns/op
BenchmarkStateSpace-32    	162368193	         7.358 ns/op
BenchmarkStateSpace-32    	161850598	         7.452 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1794 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1817 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1905 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1809 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1886 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1805 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1835 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1915 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1816 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1825 ns/op
BenchmarkFIRFilter-32     	175447726	         6.813 ns/op
BenchmarkFIRFilter-32     	176195204	         6.780 ns/op
BenchmarkFIRFilter-32     	178611482	         6.745 ns/op
BenchmarkFIRFilter-32     	180126871	         6.712 ns/op
BenchmarkFIRFilter-32     	176791880	         6.778 ns/op
BenchmarkFIRFilter-32     	175960251	         6.728 ns/op
BenchmarkFIRFilter-32     	177647713	         6.894 ns/op
BenchmarkFIRFilter-32     	176275880	         6.847 ns/op
BenchmarkFIRFilter-32     	173582355	         6.729 ns/op
BenchmarkFIRFilter-32     	175104055	         6.705 ns/op
PASS
ok  	github.com/brandondube/pctl	99.668s

The line endings are CRLF, but changing them to LF does not change the behavior of benchstat.

After doing so, invoke benchstat:

benchstat bench.txt

This produces a null output, and benchstat's return code is 0. I expected to see the usual benchstat output.

Running the same benchmark on windows subsystem for linux, with go 1.20.5, produces the file

bench.txt

goos: linux
goarch: amd64
pkg: github.com/brandondube/pctl
cpu: AMD Ryzen 9 7950X 16-Core Processor            
BenchmarkPIDLoop-32       	556950556	         2.167 ns/op
BenchmarkPIDLoop-32       	557676615	         2.193 ns/op
BenchmarkPIDLoop-32       	553675562	         2.170 ns/op
BenchmarkPIDLoop-32       	546509467	         2.147 ns/op
BenchmarkPIDLoop-32       	557190387	         2.201 ns/op
BenchmarkPIDLoop-32       	543512101	         2.203 ns/op
BenchmarkPIDLoop-32       	548559962	         2.174 ns/op
BenchmarkPIDLoop-32       	557450202	         2.139 ns/op
BenchmarkPIDLoop-32       	564419833	         2.128 ns/op
BenchmarkPIDLoop-32       	561758073	         2.146 ns/op
BenchmarkLPF-32           	377603649	         3.202 ns/op
BenchmarkLPF-32           	374204203	         3.190 ns/op
BenchmarkLPF-32           	371718177	         3.194 ns/op
BenchmarkLPF-32           	373730676	         3.205 ns/op
BenchmarkLPF-32           	374568943	         3.181 ns/op
BenchmarkLPF-32           	375112218	         3.185 ns/op
BenchmarkLPF-32           	376773606	         3.186 ns/op
BenchmarkLPF-32           	376786018	         3.187 ns/op
BenchmarkLPF-32           	377499621	         3.189 ns/op
BenchmarkLPF-32           	376156272	         3.192 ns/op
BenchmarkHPF-32           	452945294	         2.656 ns/op
BenchmarkHPF-32           	450845560	         2.643 ns/op
BenchmarkHPF-32           	452651097	         2.653 ns/op
BenchmarkHPF-32           	453425650	         2.648 ns/op
BenchmarkHPF-32           	452481432	         2.654 ns/op
BenchmarkHPF-32           	451396864	         2.654 ns/op
BenchmarkHPF-32           	451599208	         2.656 ns/op
BenchmarkHPF-32           	451884454	         2.651 ns/op
BenchmarkHPF-32           	451845573	         2.644 ns/op
BenchmarkHPF-32           	449351928	         2.656 ns/op
BenchmarkBiquad-32        	377297011	         3.183 ns/op
BenchmarkBiquad-32        	374881530	         3.189 ns/op
BenchmarkBiquad-32        	376649973	         3.194 ns/op
BenchmarkBiquad-32        	374645312	         3.192 ns/op
BenchmarkBiquad-32        	377104418	         3.184 ns/op
BenchmarkBiquad-32        	374657930	         3.185 ns/op
BenchmarkBiquad-32        	377005658	         3.182 ns/op
BenchmarkBiquad-32        	375266190	         3.193 ns/op
BenchmarkBiquad-32        	375875184	         3.197 ns/op
BenchmarkBiquad-32        	374370884	         3.248 ns/op
BenchmarkStateSpace-32    	163193721	         7.396 ns/op
BenchmarkStateSpace-32    	164258181	         7.300 ns/op
BenchmarkStateSpace-32    	164072041	         7.315 ns/op
BenchmarkStateSpace-32    	163204105	         7.285 ns/op
BenchmarkStateSpace-32    	164398299	         7.293 ns/op
BenchmarkStateSpace-32    	164617305	         7.301 ns/op
BenchmarkStateSpace-32    	164586385	         7.280 ns/op
BenchmarkStateSpace-32    	164607597	         7.317 ns/op
BenchmarkStateSpace-32    	163880760	         7.356 ns/op
BenchmarkStateSpace-32    	164462554	         7.301 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1800 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1826 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1817 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1794 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1803 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1824 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1813 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1795 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1797 ns/op
BenchmarkSetpoint-32      	1000000000	         0.1797 ns/op
BenchmarkFIRFilter-32     	175677933	         7.127 ns/op
BenchmarkFIRFilter-32     	162187520	         6.941 ns/op
BenchmarkFIRFilter-32     	164319968	         7.097 ns/op
BenchmarkFIRFilter-32     	170883248	         6.911 ns/op
BenchmarkFIRFilter-32     	171958609	         6.944 ns/op
BenchmarkFIRFilter-32     	176170941	         6.868 ns/op
BenchmarkFIRFilter-32     	176384228	         6.824 ns/op
BenchmarkFIRFilter-32     	176330742	         6.989 ns/op
BenchmarkFIRFilter-32     	176533140	         7.012 ns/op
BenchmarkFIRFilter-32     	162067396	         6.942 ns/op
PASS
ok  	github.com/brandondube/pctl	99.844s

with LF line endings. Benchstat produces the expected output in this case, regardless of whether it is called from Windows or WSL:

$ benchstat bench.txt # no output, returns 0
$ benchstat benchwsl.txt
goos: linux
goarch: amd64
pkg: github.com/brandondube/pctl
cpu: AMD Ryzen 9 7950X 16-Core Processor
              │ benchwsl.txt │
              │    sec/op    │
PIDLoop-32       2.168n ± 1%
LPF-32           3.190n ± 0%
HPF-32           2.654n ± 0%
Biquad-32        3.191n ± 0%
StateSpace-32    7.301n ± 1%
Setpoint-32     0.1801n ± 1%
FIRFilter-32     6.943n ± 2%
geomean          2.453n

[seankhliao]

is this #58576

[bcmills]

(CC @golang/windows)

[brandondube]

It appears to be the same as #58576 yes; thanks!