Activity
There was a problem fetching the latest activity.
Yesterday
dmitshur commented on golang.org/x/build/cmd/relui: bypass trybots for VERSION file CLs18h
Permission change applied.dmitshur fixed an issue golang.org/x/build/cmd/relui: bypass trybots for VERSION file CLs18h
This Week
dmitshur fixed an issue os: TestRootRemoveDot failing on Plan 91d
dmitshur reopened an issue golang.org/x/build/cmd/relui: bypass trybots for VERSION file CLs1d
Need to grant the gobot account a permission for the TryBot-Bypass label, since it is being used by it now. Reopening for that.dmitshur opened an issue golang.org/x/build: extend main Go repo's "every vet check is forcibly made high-confidence" behavior to golang.org/x packages that are vendored (or exclude vendored packages from it)1d
The `go test` command normally runs a select subset of vet checks, and fails the test if any of them fail.
In the main Go repo, to aid development, that behavior is forcibly modified to have all vet…dmitshur commented on golang.org/x/build/cmd/watchflakes: no reports after May 19, 20252d
[CL 678816](https://go.dev/cl/678816) is working, and watchflakes has started posting again. I'll keep an eye on it in the short term, and issue #70743 still remains open to track unexpected restarts.dmitshur opened an issue golang.org/x/exp/rand: isolate into nested module, tag, delete3d
This issue tracks a remaining subset of the accepted proposal #61716. Specifically:
> Once math/rand/v2 has shipped, we would tag and delete x/exp/rand. This would keep programs that already use x/e…Last Week
dmitshur commented on testing: add missing Output method to testing.TB6d
Indeed—freeze exceptions would be needed for changes that aren't in scope of the freeze (and ideally they need to be filed and approved before the freeze), but finding and resolving issues in new-t…dmitshur commented on access: please add new email to existing account1w
Great! I've also updated your email in the approvers group now, so everything should continue to work.
Just as a reminder, please be familiar with the rules for voting and submitting changes as desc…dmitshur commented on cmd/internal/moddeps: TestAllDependencies/std(thorough) failures [consistent failure]1w
Latest instance fixed in [CL 663975](https://go.dev/cl/663975).dmitshur fixed an issue cmd/internal/moddeps: TestAllDependencies/std(thorough) failures [consistent failure]1w
dmitshur commented on golang.org/x/build: add darwin-arm64 longtest and race builders1w
The darwin-arm64-longtest builder is passing at tip and on the Go 1.24 release branch, but it has one failing test on the Go 1.23 release branch. I filed #73922 for it.
The darwin-arm64-race builder…dmitshur opened an issue cmd/go: TestScript/build_trimpath_cgo fails to decode dwarf on release-branch.go1.231w
### Go version
This affects Go 1.23.x only. The test passes on Go 1.24.x and at tip.
### Output of `go env` in your module/workspace:
```shell
GOOS='darwin'
GOARCH='arm64'
```
### What did you do…dmitshur commented on golang.org/x/build/cmd/watchflakes: no reports after May 19, 20251w
The problem appears to be due to a regression in the result_adapter command in [crrev.com/c/6554882](https://crrev.com/c/6554882) from May 19, 2025, which has a pending fix in [crrev.com/c/6598395](h…dmitshur opened an issue golang.org/x/build/cmd/watchflakes: no reports after May 19, 20251w
The most recent new issue watchflakes opened was #73775 on May 19, 2025. It appears to have gotten almost entirely stuck as of that date, and isn't producing reports. This is the tracking issue.
The…dmitshur commented on access: please add new email to existing account1w
@ysmolski Glad to hear that you were able to update the email in your Gerrit account! It indeed shows up as expected at https://go-review.googlesource.com/dashboard/26536.
About the CLA page, in the…dmitshur commented on access: approvers1w
CC @4a6f656c, @Rongronggg9, @milanknezevic, @mengzhuo Since you're also members of [golang/mips](https://github.com/orgs/golang/teams/mips) and [golang/riscv64](https://github.com/orgs/golang/teams/r…dmitshur fixed an issue github.com/shurcooL/graphql: Please make "constructQuery" exported1w
dmitshur commented on github.com/shurcooL/graphql: Please make "constructQuery" exported1w
The current implementation of `constructQuery` is very tailored to the needs of the rest of the graphql package. It doesn't take any options, like the ability to include spaces in formatting, and doe…Earlier
dmitshur commented on text/template: index/element switched for iter.Seq2 in range1w
Since it's not already mentioned here, the proposal for this was #66107, and the accepted plan there was:
> The proposal is to allow range over func and int the same as in Go.
The current implement…dmitshur commented on golang.org/x/tour/tour: Missing a period at the end of the last sentence.1w
For clarity, it looks like "Methods and interfaces" was referring to the section title:
dmitshur commented on golang.org/x/build/internal/releasetargets: update for each upcoming major Go release2w
[CL 664215](https://go.dev/cl/664215) generated a preliminary list for Go 1.25. It's the same list as for Go 1.24 except the macOS minimum version became 12, and that's noted at https://tip.golang.or…dmitshur commented on all: announce end of support for old Linux versions2w
There isn't anything to pre-announce in Go 1.25 release notes, moving to the next.dmitshur commented on all: announce end of support for old Windows releases2w
There isn't anything to pre-announce in Go 1.25 release notes, moving to the next.dmitshur commented on github.com/go-gl/gl: Problem with getting v2.1 gl2w
The package requires cgo, so it’s possible to get that error if you’ve disabled cgo. Does anything change if you use CGO_ENABLED=1?dmitshur commented on access: please add new email to existing account2w
Hello.
If you're still able to login to your Gerrit account 26536, you should be able to add and verify a new email (and then make it your new primary preferred email, if desired). The reason you're…dmitshur commented on cmd/go: build and run tools that are not necessary for builds as needed and don't include in binary distribution2w
Does this need a release note (#71661)?dmitshur fixed an issue github.com/shurcooL/vfsgen: Possible to pack SQLite file?2w
dmitshur commented on golang.org/x/build: bot linux-riscv64-mengzhuo--megrez-1 reported as broken3w
Thanks for addressing this so quickly. I've closed this since the builder is back online and working. If it's useful to have an open issue, we can make a new one.dmitshur fixed an issue golang.org/x/build: bot linux-riscv64-mengzhuo--megrez-1 reported as broken3w
dmitshur closed a change golang.org/x/build: Rename README.md to -github.com-binance-exchange-binance-official-/ap…3w
dmitshur commented on os: new test TestOpenFileCreateExclDanglingSymlink fails on Plan 93w
CC @golang/plan9, @neild.dmitshur closed an issue github.com/gopherjs/gopherjs: Arrays internalized as maps3w
Perhaps the "not planned" reason is more applicable here, since there was no change.dmitshur commented on golang.org/x/crypto/x509roots/fallback: high, unskippable, init cost3w
CC @rolandshoemaker, @FiloSottile.dmitshur opened an issue golang.org/x/arch/riscv64/riscv64asm: TestObjdumpRISCV64TestDecodeGNUSyntaxdata stalls on linux/s390x3w
See https://ci.chromium.org/b/8715046913413654785 for example:
```
=== RUN TestObjdumpRISCV64TestDecodeGNUSyntaxdata
panic: test timed out after 10m0s
running tests:
TestObjdumpRISCV64TestDecod…dmitshur opened an issue golang.org/x/crypto/sha3: TestMarshalUnmarshal/SHA3-224 fails on linux/s390x3w
See https://ci.chromium.org/b/8715046724760256497 for example:
```
=== RUN TestMarshalUnmarshal
--- FAIL: TestMarshalUnmarshal (0.00s)
panic: interface conversion: *sha3.asmState is not encoding.B…dmitshur commented on golang.org/x/build: add LUCI linux-s390x builder3w
Great, thank you.
I think all that's left on our side is to adjust the builder configuration to let the builder run on applicable golang.org/x repositories (it was set to try only the main Go repo w…dmitshur commented on golang.org/x/build: add LUCI linux-s390x builder4w
Glad to hear that, and congrats on getting to this point!
The swarming bot generally takes care of using all of the CPU cores and machine capacity available to it, so the typical setup is to have on…dmitshur closed an issue go/version: IsValid reports false on inputs like "go1.8.5rc5", "go1.9.2rc2"4w
I've had a chance to talk with Russ about this, and he confirmed that these "go1.8.5rc5" and "go1.9.2rc2" versions aren't really proper Go releases—they're side-effects that happened to be produced…dmitshur commented on go/version: doesn't handle devel toolchain versions4w
We've taken the alternative route of making the generated development versions be compatible with the existing https://go.dev/doc/toolchain#name format (as a "non-standard toolchain") in #73372. Is t…dmitshur commented on github.com/andygrunwald/go-gerrit/fix: handle changes/<change>/patch JSON marshalling error4w
I left some review comments from a quick look, I hope they're helpful. I haven't tried to test the changes locally so I may have missed something.
In general, the test seems to be useful, but I th…dmitshur commented on github.com/go-delve/delve/goversion: fix development version parsing4w
Thank you!dmitshur fixed an issue runtime: timeouts on netbsd-arm64-bsiegert builder1mo
dmitshur commented on runtime: timeouts on netbsd-arm64-bsiegert builder1mo
This issue has stopped matching because this builder in LUCI no longer has -bsiegert in the name (it's still there in the underlying bot name, but the pattern uses builder name), but there've been so…dmitshur closed an issue golang.org/x/build: netbsd-amd64-8_0 builder timing out frequently1mo
The netbsd-amd64-8_0 builder doesn't exist by now, closing.dmitshur opened an issue golang.org/x/build: timeout on netbsd-arm64 builder1mo
```
#!watchflakes
default <- builder ~ `netbsd-arm64` && "panic: test timed out after 10m0s"
```
Filing this to hopefully catch future instances of issues like #73487, where the underlying problem w…dmitshur fixed an issue golang.org/x/build: LUCI support for plan91mo
dmitshur commented on golang.org/x/build: LUCI support for plan91mo
I believe by now we have resolved the general approach to support Plan 9 testing on LUCI. The plan9 arm and 386 builders seem to be generally working okay for a while now. The amd64 builder seems to …dmitshur commented on golang.org/x/build: bot mac-330-h526 reported as broken1mo
I'm not sure why this issue was opened 9 hours ago. Given #72857 is now resolved, this should happen only if the builder is dead for 24 hrs, but from looking at [its task history](https://chromium-sw…dmitshur commented on golang.org/x/build/internal/task: update postTweet to Twitter API v2 when possible1mo
Noting that we ran into the same error today. (CC @cherrymui.) The app once again ended up becoming "standalone" for reasons unknown to us. We re-added it to the project.dmitshur opened an issue golang.org/x/build/cmd/relui: bypass trybots for VERSION file CLs1mo
This is an enhancement to speed up the release process and slightly reduce its resource use, by adding some complexity to its implementation that lets it avoid duplicate work.
The Go build system us…dmitshur closed an issue golang.org/x/build: bot mac-331-h526 reported as broken1mo
By now the bot is back and idle, having run the last 25+ builds successfully. Not sure that there's anything to do here.dmitshur opened an issue golang.org/x/build: gotip-openbsd-riscv64 builder is missing1mo
There are currently no bots running at https://ci.chromium.org/ui/p/golang/builders/ci/gotip-openbsd-riscv64.
CC @4a6f656c, @golang/riscv64.dmitshur opened an issue golang.org/x/build: gotip-freebsd-riscv64 builder is missing1mo
There are currently no bots running at https://ci.chromium.org/ui/p/golang/builders/ci/gotip-freebsd-riscv64.
CC @mengzhuo, @golang/riscv64.dmitshur commented on golang.org/x/build: add LUCI linux-s390x builder1mo
From looking at the source for those functions ([here](https://source.chromium.org/chromium/infra/infra_superproject/+/main:infra/luci/appengine/swarming/swarming_bot/api/platforms/linux.py?q=%22%5B%…dmitshur commented on math/big: broken mips and mips64 builders1mo
CC @golang/mips.dmitshur commented on cmd/dist: switch to a devel version format that satisfies go/version.Lang and go/version.IsValid in more cases1mo
@xnox Including additional fields about the toolchain and its possible modifications in the build information (in addition to the go version string) like that is an interesting idea. It has some conn…dmitshur commented on internal/fuzz: enable fuzz testing on OpenBSD 1mo
It would be good to mention this in Go 1.25 release notes (#71661), otherwise users may not know about it. See [doc/README.md](https://cs.opensource.google/go/go/+/master:doc/README.md).dmitshur commented on golang.org/x/oauth2/google: findDefaultCredentials fallback to ComputeTokenSource should be predicated on availability on service account1mo
@seankhliao Can you please clarify what you mean with that comment? Why is this issue closed as not planned?dmitshur commented on golang.org/x/build: add darwin-arm64 longtest and race builders1mo
CC @JunyangShao, @prattmic.dmitshur commented on build: build failure on x/crypto GOARCH=loong64 with Go 1.23/1.24 [consistent failure]1mo
These build failures are fixed via revert CLs above. Closing.dmitshur fixed an issue build: build failure on x/crypto GOARCH=loong64 with Go 1.23/1.24 [consistent failure]1mo
dmitshur commented on golang.org/x/build: gotip-linux-amd64-misccompile builder fails with no space left on device1mo
We haven't seen this happen for a while, right?
I believe this was a combination of some short-lived builder VMs inadvertently staying around for much longer than intended, and some cache directorie…dmitshur fixed an issue golang.org/x/build: gotip-linux-amd64-misccompile builder fails with no space left on device1mo
dmitshur fixed an issue golang.org/x/build: set a maintenance window for the kubernetes instance1mo
dmitshur commented on golang.org/x/build: set a maintenance window for the kubernetes instance1mo
It seems to be generally working well. Let's call this done; we can reopen if more is needed.dmitshur commented on cmd/compile: infinite loop in the inliner1mo
We'll need to create a backport candidate issue for the Go1.24.3 milestone; moving this to Go1.25.dmitshur commented on golang.org/x/build: add LUCI linux-s390x builder1mo
@srinivas-pokala I believe the change in [crrev.com/c/6439429](https://crrev.com/c/6439429) is rolled out and you should be able to try starting up the builder again.dmitshur commented on golang.org/x/build/internal/relui: TestRelease/major failures [consistent failure]1mo
Fixed by [CL 664495](https://go.dev/cl/664495).dmitshur fixed an issue golang.org/x/build/internal/relui: TestRelease/major failures [consistent failure]1mo
dmitshur fixed an issue golang.org/x/build/env: host-linux-amd64-sid (Container Image: linux-x86-sid:latest)1mo
dmitshur commented on golang.org/x/build/env: host-linux-amd64-sid (Container Image: linux-x86-sid:latest)1mo
Closed by merging [CL 665856](https://go.dev/cl/665856) (commit golang/build@48b27723f5bd8d92de3f0fff371a73debc7d8feb) to `luci-config`.dmitshur fixed an issue golang.org/x/build: consider adding a linux builder with a newer kernel1mo
dmitshur commented on golang.org/x/build: consider adding a linux builder with a newer kernel1mo
The new issue #73405 (CC @prattmic) is a continuation of this, folding this into that.dmitshur fixed an issue golang.org/x/build/env/linux: add bleeding edge kernel builder1mo
dmitshur commented on golang.org/x/build/env/linux: add bleeding edge kernel builder1mo
The new issue #73405 (CC @prattmic) is a continuation of this, folding this into that.dmitshur commented on golang.org/x/build: recent Linux kernel builder1mo
We can try moving the sid builder (#61112) to cos-dev (up from cos-113-lts). If that ends up being too new and too noisy to be useful, we can move down to something less bleeding edge like cos-beta o…dmitshur fixed an issue build: build failure on gotip-linux-amd64-misccompile [consistent failure]1mo
dmitshur commented on math/big: wrong parameter names for mulAddVWW in loong64 assembly [consistent failure]1mo
Fixed by [CL 665717](https://go.dev/cl/665717).dmitshur fixed an issue math/big: wrong parameter names for mulAddVWW in loong64 assembly [consistent failure]1mo
dmitshur commented on golang.org/x/build/cmd/golangbuild: switch to generating a valid (per go/version.IsValid) version string1mo
Yes, agreed, we should apply this to cmd/dist too. Retitled to capture that too.
As for why golangbuild (and coordinator, previously) generates a VERSION file, it does it for a few related reasons:
…dmitshur commented on golang.org/x/review: `~/go/bin/git-codereview change` fails with `git-codereview: not found`1mo
It is intended for git-codereview to be available in PATH—see https://go.dev/doc/contribute#git-codereview_install. If you prefer not to add $HOME/go/bin to $PATH, it should still work to place the…dmitshur opened an issue golang.org/x/build/cmd/golangbuild: switch to generating a valid (per go/version.IsValid) version string1mo
golangbuild currently generates VERSION files with strings like this, a behavior that the Go build system had for a while:
```
devel <commit>
devel <change>/<patchset>
```
That satisfies the proper…dmitshur commented on go/version: doesn't handle devel toolchain versions1mo
The ability to be able to tell that a given Go version isn't a released version can be useful, and quite a bit of code relies on the presence of the "devel" substring as you mentioned. Instead of rem…dmitshur commented on golang.org/x/website: automatic URL fragments for sub-sections of 'Minor changes to the library' are overly repetitive1mo
I don't think it's viable to rely on manually specifying IDs as in that CL. People will forget to do it in some cases, and changing anchors afterwards on recent release notes might break links.dmitshur fixed an issue golang.org/x/build: migrate low-capacity (reverse) builders to LUCI1mo
dmitshur commented on golang.org/x/build: migrate low-capacity (reverse) builders to LUCI1mo
This umbrella issue is similar to the #60440 umbrella issue, we can merge them together. Closing this one. Thanks everyone.dmitshur reopened an issue golang.org/x/build/env: host-linux-amd64-sid (Container Image: linux-x86-sid:latest)1mo
The build script was updated and a new LUCI-capable image was indeed built, though it's missing python3 (needed for the swarming bot). But we still need to add it to luci-config. Reopening for that.dmitshur fixed an issue golang.org/x/build/cmd/racebuild: Error running create: failed to create buildlet on gotip-linux-loong641mo
dmitshur commented on golang.org/x/build/cmd/racebuild: Error running create: failed to create buildlet on gotip-linux-loong641mo
Yes, there is some similarity to #67468, but it's not the same issue. Over there it is a problem on the side of Gerrit, where for some reason triggering optional trybots doesn't use your primary Gerr…dmitshur opened an issue golang.org/x/review/git-codereview: TestHooks and 5 more related tests are failing as of "os: avoid symlink races in RemoveAll" on Windows1mo
As seen on the [x/review (gotip) by go commit](https://ci.chromium.org/p/golang/g/x-review-gotip-by-go/console) view, a number of tests have begun to fail starting with [CL 661575](https://go.dev/cl/…dmitshur commented on golang.org/x/review/git-codereview: `change -n` still writes the hooks without announcing1mo
Thanks for reporting.
This should be a matter of adding a check for `*noRun` inside `installHook`, before it starts to do `os.Mkdir` and `os.WriteFile`.dmitshur commented on golang.org/x/website: clarify Linux installation instructions regarding root/sudo permissions1mo
Your suggested alternative sounds good. Please go ahead and send the change.
One minor suggestion, since that is a complete sentence inside the parentheses, let's also move the period to be inside.dmitshur commented on cmd/compile: internal compiler error: assertion failed with type aliases1mo
CC @golang/compiler.dmitshur starred github.com/prattmic/benchfmt-convert4mo
dmitshur commented on net/http: TestTransportIdleConnTimeout failures4mo
Looks like [CL 637897](https://go.dev/cl/637897) indeed fixed this. Closing, watchflakes will follow up if it still happens.dmitshur fixed an issue net/http: TestTransportIdleConnTimeout failures4mo
dmitshur commented on golang.org/x/build/cmd/coordinator: older dashboard pages omit LUCI builders in detailed views (page 2, x/ repo commits) but don't make it apparent it's WAI4mo
Thanks for reporting this.
The missing builders on those pages are all the new LUCI builders. In #65913 we provided limited support for displaying LUCI build results on build.golang.org, but that su…dmitshur closed an issue golang.org/x/build: Tricium should not enforce Go style rules on testdata/*.go4mo
The implementation of the check we were using previously is fully gone as of [CL 643916](https://go.dev/cl/643916). I'll close this since the original problem is no longer happening and so the issue …dmitshur commented on golang.org/x/build: run platform/arch specific builders when platform/arch constrained files are changed4mo
To connect some dots: in #66009, @mknyszek has a paragraph that mentions a "proxy" builder, which is similar (if not identical) to the "spawning an additional set of builders" description here.
At a…