New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
cmd/trace: TestAnalyzeAnnotations failures on linux/riscv64 #46737
Comments
This affected Hifive unmatched too. |
Would it make sense to skip this test on the |
Change https://golang.org/cl/330849 mentions this issue: |
Change https://golang.org/cl/333449 mentions this issue: |
@hyangah, could you take a look at the fix for this? It's at +1, and would be nice to get in a fix for the underlying problem before the Go 1.17 release. |
Unfortunately It start failing again. https://build.golang.org/log/d2f46348571205149b5bb4480bbdc4a202f1f73e It seems differ with original error that goid is some how higher than previous failure logs. I'll take a look into it. |
The new failure mode has a region with a
2021-11-18T22:40:42-2375b6e/linux-riscv64-unmatched |
(This is a release-blocker for Go 1.18 via #11811. If it isn't feasible to fix for the 1.18 release, we should add a skip on |
If this has been bad since 1.14, we should likely add a skip. @4a6f656c |
This is still failing 2021-12-08T18:30:00-61011de/linux-riscv64-unmatched Any update on this? Thanks. (checking in as this is a release blocker) Yeah, if it fails since 1.14, it is probably not a release blocker. |
#11811 is basically “tests should be passing”. I suppose that the main repo is technically not a subrepo, but we still ought to have the builders passing tests before we cut a release. 😅 (Is there a reason not to add a skip in this case?) |
Thanks.
No. Skipping is fine. |
Change https://golang.org/cl/373034 mentions this issue: |
@cherrymui @bcmills @bcmills I've run 10000 TestAnnotationAnalyze on CL 373034 without failure, please take a look, thanks. |
The riscv64 Hifive Unmatched is the only platform that failed on testcase TestAnalyzeAnnotations occasionally after CL 332954 had merged. The failure happens when ticks per second (freq) is over 1e12 which causing the timestamps of two events are same. There are 2 reasons causing big frequency: 1. RDCYCLE is HART based according to the riscv manual which makes negative ticks delta 2. negative float64 -> uint64 is undefined and "lucky" negative float is too big to handle for trace For #46737 Change-Id: I1f3c1ac31aae249969000c719c32aaf5a66d29a5 Reviewed-on: https://go-review.googlesource.com/c/go/+/373034 Trust: Zhuo Meng <mzh@golangcn.org> Run-TryBot: Zhuo Meng <mzh@golangcn.org> TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Michael Knyszek <mknyszek@google.com>
After the CL above is there anything else still needs to be done for this issue? Or we can close it? Thanks. |
There haven't been any more failures since https://golang.org/cl/373034, so I think this can be closed unless failures reoccur. |
2021-06-13T08:17:17-24cff0f/linux-riscv64-unleashed
2021-04-27T15:57:56-bc62887/linux-riscv64-jsing
2021-04-06T14:15:29-b345a30/linux-riscv64-jsing
2021-04-02T01:14:00-97b3ce4/linux-riscv64-jsing
2021-03-22T03:52:31-d8394bf/linux-riscv64-jsing
2020-05-05T05:13:26-9b18968/linux-riscv64-unleashed
2020-04-25T02:19:12-49f10f3/linux-riscv64-unleashed
CC @prattmic @mknyszek @4a6f656c @bradfitz
The text was updated successfully, but these errors were encountered: