all: ignored builders for test flakes reporting

[rsc]

#!watchflakes
skip <-
	builder ~ `android-.*-emu` && date != "" && date < "2022-12-05" ||
	builder ~ `^darwin-.*-aws` && date != "" && date < "2022-12-05" ||
	builder == "dragonfly-amd64-622" && date != "" && date < "2022-09-01" ||
	builder == "freebsd-arm-paulzhol" && date != "" && date < "2022-12-01"||
	builder == "linux-amd64-alpine" && date != "" && date < "2022-09-01" ||
	builder == "linux-amd64-unified" && date != "" && date < "2022-12-01" ||
	builder == "linux-loong64-3a5000" && date != "" && date < "2022-12-01" ||
	builder ~ `^netbsd.*-9_0` ||
	builder ~ `^netbsd.*` && date != "" && date < "2022-10-12" ||
	builder ~ `^plan9-(386|amd64)` ||
	builder == "plan9-arm" && date != "" && date < "2022-12-01" ||
	builder == "solaris-amd64-oraclerel" && date != "" && date < "2022-11-29" ||
	builder ~ `.*-newcc.*` && date != "" && date < "2022-12-01" ||
	builder == "windows-arm64-11" && date != "" && date < "2022-12-01" ||
	builder == "linux-amd64-wsl" && date != "" && date < "2023-05-01" ||
	builder == "wasip1-wasm-wasmedge" ||
	builder == "wasip1-wasm-wasmer" ||
	builder == "wasip1-wasm-wasmtime" ||
	builder == "openbsd-ppc64-n2vi"

This issue defines policy (above) for which builders are entirely ignored for flake finding.

[gopherbot]

Sorry, but there were parse errors in the watch flakes script.
The script I found was:

#!watchflakes
skip <-
	builder ~ "android-.*-emu" && date != "" && date < "2022-12-05" ||
	builder ~ `^darwin-.*-aws` ||
	builder == "dragonfly-amd64-622" && date != "" && date < "2022-09-01" ||
	builder == "freebsd-arm-paulzhol" ||
	builder == "linux-amd64-alpine" && date != "" && date < "2022-09-01" ||
	builder == "linux-amd64-unified" ||
	builder == "linux-loong64-3a5000" ||
	builder ~ `^netbsd.*-9_0` ||
	builder ~ `^netbsd.*` && date != "" && date < "2022-10-12" ||
	builder == "plan9-386" ||
	builder == "plan9-386-0intro" ||
	builder == "plan9-amd64-0intro" ||
	builder == "plan9-arm" && date != "" && date < "2022-12-01" ||
	builder == "solaris-amd64-oraclerel" && date != "" && date < "2022-11-29" ||
	builder == "windows-386-2008-newcc" ||
	builder == "windows-386-2012-newcc" ||
	builder == "windows-amd64-2008-newcc" ||
	builder == "windows-amd64-2012-newcc" ||
	builder == "windows-amd64-2016-newcc" ||
	builder == "windows-amd64-longtest-newcc" ||
	builder == "windows-amd64-newcc-race" ||
	builder == "windows-arm64-11" && date != "" && date < "2022-12-01"

And the problems were:

script:3.12: ~ requires backquoted regexp
script:4.10: unexpected ~
script:5.10: unexpected ==
script:6.10: unexpected ==
script:7.10: unexpected ==
script:8.10: unexpected ==
script:9.10: unexpected ==
script:10.10: unexpected ~
script:11.10: unexpected ~
script:12.10: unexpected ==
script:13.10: unexpected ==
script:14.10: unexpected ==
script:15.10: unexpected ==
script:16.10: unexpected ==
script:17.10: unexpected ==
script:18.10: unexpected ==
script:19.10: unexpected ==
script:20.10: unexpected ==
script:21.10: unexpected ==
script:22.10: unexpected ==
script:23.10: unexpected ==
script:24.10: unexpected ==

See https://go.dev/wiki/Watchflakes for details.

— watchflakes

[cherrymui]

(Edited the script to fix parse errors.)

[gopherbot]

Sorry, but there were parse errors in the watch flakes script.
The script I found was:

#!watchflakes
skip <-
	builder ~ `android-.*-emu` && date != "" && date < "2022-12-05" ||
	builder ~ `^darwin-.*-aws` ||
	builder == "dragonfly-amd64-622" && date != "" && date < "2022-09-01" ||
	builder == "freebsd-arm-paulzhol" && date != "" && date < "2022-12-01"||
	builder == "linux-amd64-alpine" && date != "" && date < "2022-09-01" ||
	builder == "linux-amd64-unified" && date != "" && date < "2022-12-01" ||
	builder == "linux-loong64-3a5000" && date != "" && date < "2022-12-01" ||
	builder ~ `^netbsd.*-9_0` ||
	builder ~ `^netbsd.*` && date != "" && date < "2022-10-12" ||
	builder ~ `^plan9-.*`
	builder == "solaris-amd64-oraclerel" && date != "" && date < "2022-11-29" ||
	builder ~ `.*-newcc.*` && date != "" && date < "2022-12-01" ||
	builder == "windows-arm64-11" && date != "" && date < "2022-12-01"

And the problems were:

script:13.10: unexpected ==
script:14.10: unexpected ~
script:15.10: unexpected ==

See https://go.dev/wiki/Watchflakes for details.

— watchflakes

[macrombilux]

#!watchflakes
skip <-
	builder ~ `android-.*-emu` && date != "" && date < "2022-12-05" ||
	builder ~ `^darwin-.*-aws` && date != "" && date < "2022-12-05" ||
	builder == "dragonfly-amd64-622" && date != "" && date < "2022-09-01" ||
	builder == "freebsd-arm-paulzhol" && date != "" && date < "2022-12-01"||
	builder == "linux-amd64-alpine" && date != "" && date < "2022-09-01" ||
	builder == "linux-amd64-unified" && date != "" && date < "2022-12-01" ||
	builder == "linux-loong64-3a5000" && date != "" && date < "2022-12-01" ||
	builder ~ `^netbsd.*-9_0` ||
	builder ~ `^netbsd.*` && date != "" && date < "2022-10-12" ||
	builder ~ `^plan9-(386|amd64)` ||
	builder == "plan9-arm" && date != "" && date < "2022-12-01" ||
	builder == "solaris-amd64-oraclerel" && date != "" && date < "2022-11-29" ||
	builder ~ `.*-newcc.*` && date != "" && date < "2022-12-01" ||
	builder == "windows-arm64-11" && date != "" && date < "2022-12-01" ||
	builder == "linux-amd64-wsl" && date != "" && date < "2023-05-01" ||
	builder == "wasip1-wasm-wasmedge" ||
	builder == "wasip1-wasm-wasmer" ||
	builder == "wasip1-wasm-wasmtime" ||
	builder == "openbsd-ppc64-n2vi"

This issue defines policy (above) for which builders are entirely ignored for flake finding.

[Oguidan]

Hi, please can someone explain what you guys are trying to do here? @rsc @seankhliao @macrombilux

[bryanbasor53]

#!watchflakes
skip <-
	builder ~ `android-.*-emu` && date != "" && date < "2022-12-05" ||
	builder ~ `^darwin-.*-aws` && date != "" && date < "2022-12-05" ||
	builder == "dragonfly-amd64-622" && date != "" && date < "2022-09-01" ||
	builder == "freebsd-arm-paulzhol" && date != "" && date < "2022-12-01"||
	builder == "linux-amd64-alpine" && date != "" && date < "2022-09-01" ||
	builder == "linux-amd64-unified" && date != "" && date < "2022-12-01" ||
	builder == "linux-loong64-3a5000" && date != "" && date < "2022-12-01" ||
	builder ~ `^netbsd.*-9_0` ||
	builder ~ `^netbsd.*` && date != "" && date < "2022-10-12" ||
	builder ~ `^plan9-(386|amd64)` ||
	builder == "plan9-arm" && date != "" && date < "2022-12-01" ||
	builder == "solaris-amd64-oraclerel" && date != "" && date < "2022-11-29" ||
	builder ~ `.*-newcc.*` && date != "" && date < "2022-12-01" ||
	builder == "windows-arm64-11" && date != "" && date < "2022-12-01" ||
	builder == "linux-amd64-wsl" && date != "" && date < "2023-05-01" ||
	builder == "wasip1-wasm-wasmedge" ||
	builder == "wasip1-wasm-wasmer" ||
	builder == "wasip1-wasm-wasmtime" ||
	builder == "openbsd-ppc64-n2vi"

This issue defines policy (above) for which builders are entirely ignored for flake finding.

The output you provided is the result of running tests for various packages in the Go programming language. Each line indicates whether the tests for a specific package passed (denoted by "ok") along with the time taken to run those tests in seconds.

Here’s a summary of the output:

1. Package and Time Taken:

   • Various packages are tested, including standard library components like archive/tar, archive/zip, compress/gzip, and several crypto packages.
   • The time taken for each package's tests to run varies, with the fastest being container/list at 0.012 seconds and the slowest being crypto/internal/edwards25519 at 2.633 seconds.

2. Notable Trends:

   • Most packages finished their tests within a second, indicating they are relatively lightweight.
   • The crypto/internal/edwards25519 package and its field operations took significantly longer than others, suggesting that they may be more complex or require more extensive testing.

3. No Tests to Run Warning:

   • The line ok crypto/internal/nistec/fiat 0.017s [no tests to run] indicates that while this package was checked, there were no tests defined to run for it.

Overall, the test results show a good mix of performance times, and it appears all the tests passed successfully without any failures.

If you have any specific questions about these results, need help interpreting them further, or have any other related topics you'd like to discuss, please let me know!