Activity
Today
dmitshur opened an issue go/types: TestFixedbugs/issue70549.go fails on linux-s390x-ibm builder21m
The `TestFixedbugs/issue70549.go` test recently added in [CL 662775](https://go.dev/cl/662775) is consistently failing on linux-s390x-ibm and linux-s390x-ibm-race builders:
```
--- FAIL: TestFixedbu…dmitshur commented on cmd/go: confusing situation with 'go run' and '//go:build ignore'42m
CC @matloob, @samthanawalla (via [owners](https://dev.golang.org/owners)).dmitshur closed an issue build: build failure on gotip-windows-amd6449m
dmitshur commented on os/exec: ocassional fork/exec failures with "The parameter is incorrect." on Windows51m
@qmuntal is [looking into this](https://go-review.googlesource.com/c/go/+/653856/5#message-195160c33eb703a9b3cd6740dec47a275005ac3e). CC @golang/windows.dmitshur closed an issue runtime:mayMoreStackMove: TestSyscallN/arg-32 failures57m
dmitshur closed an issue runtime:mayMoreStackMove: TestSyscallN/arg-15 failures57m
dmitshur closed an issue runtime:mayMoreStackMove: TestSyscallN/arg-13 failures57m
dmitshur closed an issue cmd/vet: TestVet/stdversion failures57m
dmitshur closed an issue runtime:mayMoreStackMove: TestSyscallN/arg-33 failures58m
dmitshur closed an issue runtime:mayMoreStackMove: TestSyscallN/arg-3 failures58m
dmitshur closed an issue net/textproto: unrecognized failures58m
dmitshur closed an issue internal/coverage/cfile: TestIssue59563TruncatedCoverPkgAll failures58m
dmitshur closed an issue runtime:mayMoreStackMove: TestSyscallN/arg-20 failures58m
dmitshur closed an issue os: TestStartProcess/absolute failures59m
dmitshur commented on golang.org/x/sys/windows: some syscalls are creating dangling pointers1h
CC @golang/windows.dmitshur reviewed +2 on doc/go1.25: document macOS requirements4h
Thanks. Trybot failure looks unrelated to the change; bypassing.dmitshur commented on golang.org/x/crypto/acme: return error from pre-authorization when unsupported5h
(1 comment)
If this error variable is exported, isn't this change in scope of the proposal process (https://go.dev/s/proposal#scope)?
If it doesn't need to be exported, then making it exported can …dmitshur reviewed +2 on go.dev/wiki/Gomote.md: update Windows paths5h
Thanks.Yesterday
dmitshur commented on os: avoid symlink races in RemoveAll on Windows1d
Note that x/review/git-codereview.TestHooks and 5 more related tests are failing as of this change on GOOS=windows as seen on the [x/review (gotip) by go commit](https://ci.chromium.org/p/golang/g/x-…dmitshur commented on cmd/compile: stack allocate variable-sized makeslice1d
Note that x/sys/windows.TestBuildSecurityDescriptor is failing as of this change on GOOS=windows as seen on the [x/sys (gotip) by go commit](https://ci.chromium.org/p/golang/g/x-sys-gotip-by-go/conso…dmitshur reviewed +2 on doc/next: fix link1d
Thanks.Last Week
dmitshur commented on golang.org/x/sys/...: simplify code by using modern Go constructs4d
(1 comment)
Noting here that if [for range](https://go.dev/ref/spec#For_range) is being used, this could also be written as:
```
for i := range sa.Addr {
```
Which has the same behavior but is arg…dmitshur commented on crypto/tls: ECH decodeInnerClientHello incorrectly rejects ClientHello with GREASE values in supportedVersions [1.24 backport]4d
Can you please clarify if the request to backport to the 1.24 branch is because the problem only exists in 1.24 but not 1.23? Or does this problem affect 1.23 too? We need this information because th…dmitshur reviewed +2 on api: move go1.25 to next/702504d
dmitshur reviewed +2 on api: move go1.25 to next/702504d
(1 comment)
Noting that [normally](https://cs.opensource.google/go/go/+/master:doc/README.md) each proposal gets its own individual fragment file. If you're interested in updating the test to detect…dmitshur reviewed +2 on api: move go1.25 to next/702504d
dmitshur reviewed +2 on api: move go1.25 to next/702504d
Thanks. As TestCheckAPIFragments reports, new APIs need a corresponding release note.dmitshur commented on cmd/link/internal/ld: introduce -funcalign=N option5d
(1 comment)
Oh, based on https://go.dev/issue/72130#issuecomment-2710961286, maybe this just needs to say "For" instead of "Fixes".dmitshur commented on cmd/link/internal/ld: introduce -funcalign=N option5d
(1 comment)
Issue #72130 is a proposal in 'Incoming' state. Should this CL wait for the proposal to be accepted before being submitted? Or should the issue be taken out of the proposal process?dmitshur commented on golang.org/x/vuln: expose vulnerability feed as rss5d
@zpavlinovic This issue is closed as "completed", is that the right status? Or should it be closed as "not planned"?dmitshur reviewed +2 on golang.org/x/build/internal/buildgo: fix use of wrong error5d
Thanks.dmitshur reviewed +2 on cmd/vet: add waitgroup analyzer5d
Thanks.dmitshur reviewed +2 on cmd/vendor: update golang.org/x/tools to v0.31.1-0.20250328151535-a857356d5cc55d
(1 comment)
Thanks. It works out well since we want to update all golang.org/x dependencies regularly anyway (#36905).dmitshur reviewed +2 on cmd/vendor: update golang.org/x/tools to v0.31.1-0.20250328151535-a857356d5cc55d
(1 comment)
Is this revert to v0.30.0 unintentional? If intentional, it needs to also be done for the cmd module to keep them both using the same x/sys version.dmitshur commented on golang.org/x/build/buildgo: fix use wrong error6d
In your other comments you said "Fix applied". In order to apply that change, it does require adding a new "patch set". A patch set is what Gerrit refers to revisions, and a new one is needed to make…dmitshur reviewed +2 on golang.org/x/website/internal/history: document Go 1.24.2 and 1.23.8 releases6d
Thanks.
"net/http" is already listed under Security.Packages for the #72011 security fix.
But "net/http" also has a bug (non-security) fix for issue #72103. Add it to the Bug.Packages list too?
No…dmitshur commented on golang.org/x/build/buildgo: fix use wrong error6d
No problem. Note there's still no patch set 2 available. If you have it as a draft, it there's a button you need to press to publish it.dmitshur fixed an issue golang.org/x/text: test results fail to be uploaded to ResultDB6d
Fixed via [crrev.com/c/6414930](https://crrev.com/c/6414930), which adds escaping for this character on the side of result_adapter.
Filed #73116 in case there's something the go command can do bette…dmitshur commented on golang.org/x/text: test results fail to be uploaded to ResultDB6d
Fixed via [crrev.com/c/6414930](https://crrev.com/c/6414930), which adds escaping for this character on the side of result_adapter.
Filed #73116 in case there's something the go command can do bette…dmitshur opened an issue cmd/go: consider shrinking set of characters allowed/passed through in subtest names6d
This issue is meant to capture a small finding from #73086, and it's related to the cluster of proposals like #67562 (and its earlier version #45549), #27896, possibly more like #44970, #20209, #2021…dmitshur commented on golang.org/x/build/buildgo: fix use wrong error1w
(1 comment)
Did you mean to upload a new patch set, or do you mean something different by "Acknowledged"?dmitshur reviewed +2 on golang.org/x/build/buildgo: fix use wrong error1w
Thanks.
```suggestion
internal/buildgo: fix use of wrong error
```
```suggestion
err has been checked, it is nil now, here should call sp.Done with
remoteErr.
```
(See go.dev/wiki/CommitMessage.)dmitshur commented on cmd/compile/internal/ssa: replace uses of interface{} with Sym/Aux1w
Thanks.Earlier
dmitshur reviewed +2 on cmd/vet: add waitgroup analyzer1w
Thanks.
This CL enables the implementation for this feature, documents it, and closes the issue in the Go1.25 milestone (via 'Fixes #18022'). That's 3 out of 4 things we need to be able to release G…dmitshur reviewed +2 on cmd/vendor: update golang.org/x/tools to v0.31.1-0.20250328151535-a857356d5cc51w
Yes, both of those approaches would work.
Since the golang.org/x/tools/go/analysis/passes/waitgroup package is small (2 .go files), to me it seems fine to get it vendored in the next CL (i.e., the s…dmitshur commented on golang.org/x/build/env/freebsd-amd64: let swarming user call /sbin/shutdown -r now1w
Thanks.dmitshur commented on golang.org/x/build/env/freebsd-amd64: upgrade to FreeBSD 14.21w
Thanks.dmitshur commented on golang.org/x/mobile: gobind adds local paths to .so file even with -trimpath1w
CC @hajimehoshi.dmitshur commented on golang.org/x/build/main.star: enable freebsd-amd64_14.2 for x/ repos1w
Ready to take this step now, dropping hold.dmitshur opened a change golang.org/x/build/env/freebsd-amd64: let swarming user call /sbin/shutdown -r now1w
One of Swarming Bot requirements is to have passwordless
sudo reboot enabled.
While here, also install unzip for x/build/cmd/racebuild,
and nano for some interactive debugging convenience.
For gola…dmitshur opened a change golang.org/x/text/message: quote fmt and val in subtest names1w
For golang/go#73086.dmitshur opened an issue golang.org/x/text: test results fail to be uploaded to ResultDB1w
The x/text dashboard has turned red:
https://ci.chromium.org/p/golang/g/x-text-gotip-by-go/console
The text execution itself passes, but LUCI builds (for example, https://ci.chromium.org/b/87191676…dmitshur commented on runtime/cgo: naming a DLL exported function as "CloseHandle" causes all DLL exported functions to deadlock1w
CC @golang/runtime, @golang/windows.dmitshur commented on net: ResolveTCPAddr fails with error when resolving a link-local IPv4 address containing a zone1w
CC @ianlancetaylor, @neild.dmitshur starred github.com/venkat1017/Opengit1w
dmitshur reviewed +2 on internal/syscall/windows: run go generate1w
dmitshur reviewed +2 on internal/syscall/windows: run go generate1w
Thanks.dmitshur commented on debug/buildinfo: fall back code for when there's no .go.buildinfo is wrong1w
CC @matloob, @samthanawalla.dmitshur commented on crypto/x509: accepts CRL with two AKI extension1w
CC @golang/security.dmitshur commented on crypto/x509: unexpected handling of escape characters1w
Thanks. The value of `crl.Issuer.Organization` contains only one backslash character:
```
Organization:[]string{`VeriSign\`}
```
And the String method prints that one backslash in escaped format. H…dmitshur reviewed +2 on log/slog: Handler doc points to handler guide1w
Thanks.
go.dev is newer and shorter compared to golang.org. While they both work, it seems we should prefer writing down the former in new code and documentation.
```suggestion
// Before implementi…dmitshur reviewed +2 on golang.org/x/example/slog-handler-guide: warn against embedding1w
Thanks.dmitshur commented on golang.org/x/build/main.star: enable freebsd-amd64_14.2 for x/ repos1w
Hold for now. This is meant to be submitted later, after we're done iterating on the image and ready to proceed.dmitshur commented on golang.org/x/build/main.star: add freebsd-amd64_14.21w
Thanks.dmitshur commented on golang.org/x/tools/gopls/internal/settings: enable more staticcheck passes1w
> What is that subset?
As noted at https://staticcheck.dev/docs/configuration/, "Staticcheck tries to provide a good out-of-the-box experience, [...]", that is, this question is already being consid…dmitshur commented on golang.org/x/build/env/freebsd-amd64: upgrade to FreeBSD 14.21w
This change is ready for review.dmitshur opened a change golang.org/x/build/main.star: finish upgrading freebsd-amd64 to FreeBSD 14.21w
For golang/go#72030.dmitshur opened a change golang.org/x/build/main.star: add freebsd-amd64_14.21w
For golang/go#72030.dmitshur opened a change golang.org/x/build/main.star: enable freebsd-amd64_14.2 for x/ repos1w
For golang/go#72030.dmitshur commented on cmd/internal/testdir: testenv.go:166: skipping test: 'go build' not compatible with setting $GO_GCFLAGS1w
Is this the same as issue #18222?dmitshur opened a change golang.org/x/build/env/freebsd-amd64: upgrade to FreeBSD 14.21w
For golang/go#72030.dmitshur commented on cmd/go: cache confuses package with different build tags when using CGO with pkg-config1w
CC @matloob, @samthanawalla.dmitshur commented on golang.org/x/vuln: expose vulnerability feed as rss1w
CC @golang/vulndb.dmitshur commented on crypto/x509: accepts invalid AKI extension in CRL1w
CC @golang/security.dmitshur commented on crypto/x509: incorrect handling of the CRL Number field1w
CC @golang/security.dmitshur commented on crypto/x509: unexpected handling of escape characters1w
In the issue report you mention "I use Go to parse this CRL file, Go interprets". Can you please provide a snippet of code for that? In particular, that code will help determine if the problem is in …dmitshur commented on testing: b.StopTimer breaks b.Loop [1.24 backport]1w
@JunyangShao Since you authored the original CL, would you like to create the cherry-pick CL for it too? The process is documented at https://go.dev/wiki/MinorReleases#making-cherry-pick-cls. Thanks.dmitshur commented on crypto/x509: accepts DN with all empty values1w
CC @golang/security.dmitshur starred github.com/rudransh61/Physix-go2w
dmitshur commented on net/http/httputil: ReverseProxy calls WriteHeader after hijacking connection2w
@neild Both this issue (#72954) and issue #72140 say:
> This is a regression in 1.24 caused by CL 637939 for #35892.
And they both request a backport to 1.24, but as far I can tell, CL 637939 is on…dmitshur commented on testing: structured output for test attributes2w
To clarify, is it intentional that the test2json line would exclude remaining TestEvent fields like "Time", "Package" (in contrast to https://github.com/golang/go/issues/71287#issuecomment-2737759645…dmitshur closed an issue golang.org/x/build/env: host-freebsd-amd64-12_3 (VM Image: freebsd-amd64-123-stable-20211230)2w
FreeBSD 12.3 is EOL'ed as noted in https://github.com/golang/go/issues/61095#issuecomment-2334940821. We ended up adding a newer FreeBSD 14.1 as part of that issue, so there's nothing more to do for …dmitshur commented on golang.org/x/build/env: host-openbsd-386-* (VM Image: openbsd-386-*)2w
It was possible to successfully build the image, and its swarming bot client successfully connects with `cipd_platform:openbsd-386`:
https://chromium-swarm.appspot.com/botlist?f=cipd_platform%3Aopen…dmitshur commented on golang.org/x/tools/gopls/internal/test/integration: 10 tests failing on openbsd/amd64 after OpenBSD 7.2 → 7.6 upgrade3w
The flakes in #54461 seem to happen on OpenBSD 7.2 builders only. [CL 654676](https://go.dev/cl/654676) replaces the previous OpenBSD 7.2 builders with the new 7.6 ones, so those those will go away.
…dmitshur fixed an issue golang.org/x/build: bot netbsd-arm64-bsiegert reported as broken3w
The builder is currently operational. Thanks for resolving this @bsiegert. Closing.dmitshur fixed an issue golang.org/x/build: bot solaris-amd64 reported as broken3w
dmitshur commented on golang.org/x/build: bot solaris-amd64 reported as broken3w
The builder is currently operational. Thanks for resolving this. Closing.dmitshur fixed an issue golang.org/x/build: bot openbsd-ppc64-n2vi reported as broken3w
The builder is currently operational. Thanks for resolving this @n2vi. Closing.dmitshur closed an issue golang.org/x/build: bot aix-ppc64-power8 reported as broken3w
This bot doesn't exist now. It existed briefly (see https://github.com/golang/go/issues/67299#issuecomment-2276610972) but has since been deleted because its cipd_platform was incorrectly computed.
…dmitshur closed an issue golang.org/x/build: bot linux-ppc64le-power8--03 reported as broken3w
Fixed alongside #72851.dmitshur closed an issue golang.org/x/build: bot linux-ppc64le-power8--05 reported as broken3w
CC @golang/release.
Fixed alongside #72851.dmitshur closed an issue golang.org/x/build: bot linux-ppc64le-power8--04 reported as broken3w
Fixed alongside #72851.dmitshur closed an issue golang.org/x/build: bot linux-ppc64le-power8--02 reported as broken3w
Fixed alongside #72851.dmitshur fixed an issue golang.org/x/build: bot linux-ppc64le-power8--01 reported as broken3w
Thanks!dmitshur deleted branch in rsc.io/rf3w
drop-go116dmitshur pushed to
main in rsc.io/rf3w
dmitshur merged a change rsc.io/rf/git-generate: modernize slightly3w
dmitshur commented on rsc.io/rf/git-generate: modernize slightly3w
Thanks.