net/http/pprof: occasional failures in TestDeltaProfile

[bcmills]

TestDeltaProfile, added in CL 147598 for #23401, seems to be flaky on some of the arm builders.

(Perhaps it is relying on some property of scheduler latency that does not hold on the very slowest builders?)

2020-04-18T20:01:46-f5291cf/plan9-arm
2020-04-18T18:07:52-2a20f5c/android-arm-corellium

CC @hyangah @odeke-em @bradfitz

Marking as release-blocker for 1.15 since the test appears to be new for that release.

[bcmills]

Not just on ARM:

2020-04-22T16:45:05-141b11d/android-386-emu
2020-04-22T16:45:05-141b11d/darwin-amd64-nocgo
2020-04-20T19:36:50-0239a5c/openbsd-386-62

[hyangah]

any objection to disabling this test in all platforms other than in linux?

[bcmills]

Is there a reason to suspect that it is not flaky on Linux?

(In my experience, the unusual builders are often just more timing-sensitive than the mainline builders — tests that are flaky on more than one of those platforms also tend to be flaky on linux, just at a lower rate.)

[hyangah]

Or, maybe we completely delete this test?
I don't know what to make out of testing this on the platforms where even time.Sleep(1sec) results in sleeping more than 2 sec.

[gopherbot]

Change https://golang.org/cl/229498 mentions this issue: net/http/pprof: make TestDeltaProfile less flaky by retrying

[gopherbot]

Change https://golang.org/cl/229707 mentions this issue: cmd/compile: avoid double-zeroing

[josharian]

Please ignore that last commit referenced by gopherbot. Typo in commit message.

[hyangah]

I don't see any more test failures after the cl. Closing it. Thanks!

[bcmills]

Still appears to be flaky, but removing the release-blocker label because the flake rate seems at least reduced.

2020-07-13T17:48:55-e88ea87/openbsd-arm-jsing
2020-06-23T18:38:32-9f33108/android-386-emu
2020-06-10T20:58:11-ec177e4/openbsd-arm-jsing
2020-05-29T20:55:14-c2d1df6/android-arm64-corellium
2020-05-28T14:43:19-6bf2eea/openbsd-arm-jsing