NotificationsYou must be signed in to change notification settings
Fork5.2k
Star17.2k

Add pgo testing to outerloop#53301

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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Jump to bottom

Merged

AndyAyersMS merged 3 commits intodotnet:mainfromAndyAyersMS:PgoTesting

May 27, 2021

Merged

Add pgo testing to outerloop#53301

AndyAyersMS merged 3 commits intodotnet:mainfromAndyAyersMS:PgoTesting

May 27, 2021

Conversation

Copy link

Member

AndyAyersMS commentedMay 26, 2021

Also update jit-experimental to no longer run PGO or EH WriteThru.

Add pgo testing to outerloop

04ed8f2

Also update jit-experimental to no longer run PGO or EH WriteThru.

Copy link

ghost commentedMay 26, 2021

I couldn't figure out the best area label to add to this PR. If you have write-permissions please help me learn by adding exactly onearea label.

Copy link

MemberAuthor

AndyAyersMS commentedMay 26, 2021

cc @dotnet/jit-contrib

fix syntax

4940538

Copy link

MemberAuthor

AndyAyersMS commentedMay 26, 2021

Ignoredotnet.runtime (1) as I didnt get the name right initially.

BruceForstall approved these changes

May 26, 2021

View reviewed changes

Copy link

Contributor

BruceForstall left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

I think you should rename the pipelines "runtime-coreclr pgo" and "runtime-coreclr libraries-pgo"

eng/pipelines/coreclr/libraries-pgo.yml Outdated


		schedules:
		- cron: "0 16 * * 0,6"
		displayName: Sun at 8:00 AM (UTC-8:00)

Copy link

Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

nit: looks like the 'cron' is Sat and Sun

eng/pipelines/coreclr/pgo.yml


		schedules:
		- cron: "0 0 * * *"
		displayName: Mon through Sun at 12:00 AM (UTC-8:00)

Copy link

Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

cron "0 0" is midnight UTC, so 4pm PDT?

Copy link

MemberAuthor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

Will update these.

Copy link

MemberAuthor

AndyAyersMS commentedMay 26, 2021

Libraries failures are unrelated.

Linux debug x64 libraries test failure is in post-test reporting.

Searching '/root/helix/work/workitem/uploads' for test results filesFatal Python error: could not acquire lock for <_io.BufferedWriter name='<stdout>'> at interpreter shutdown, possibly due to daemon threadsPython runtime state: finalizing (tstate=0x55dc30e0ac00)

the other two legs fail this test (seems like a novel failure)

    System.IO.Tests.SymbolicLink_Changed_Tests.FileSystemWatcher_DirectorySymbolicLink_TargetsFile_Fails [FAIL]      Assert.Throws() Failure      Expected: typeof(System.IO.FileNotFoundException)      Actual:   (No exception was thrown)

Copy link

MemberAuthor

AndyAyersMS commentedMay 27, 2021

71 test failures in libraries tests, 10 in jit tests.

update display name/time

7d6d2a5

Copy link

MemberAuthor

AndyAyersMS commentedMay 27, 2021

Scanned through the various failing tests and they're all getting the expected complus settings.

AndyAyersMS added the area-CodeGen-coreclrCLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label

May 27, 2021

AndyAyersMS self-assigned this

May 27, 2021

Copy link

MemberAuthor

AndyAyersMS commentedMay 27, 2021

Not entirely sure why the pgo jobs got retriggered on push .... anybody know?

Copy link

Contributor

BruceForstall commentedMay 27, 2021

Not entirely sure why the pgo jobs got retriggered on push .... anybody know?

Seems unexpected.

Copy link

MemberAuthor

AndyAyersMS commentedMay 27, 2021

I guess what I'm wondering is if these new pipelines are somehow attached to PRs/Innerloop -- how can I check that?

Copy link

Contributor

BruceForstall commentedMay 27, 2021

My reading of the YAML is that shouldn't happen: it has "trigger: none" and no "pr:" entry. You can go to the AzDO pipeline settings page for your branch:

https://dev.azure.com/dnceng/public/_apps/hub/ms.vss-build-web.ci-designer-hub?pipelineId=1000&nonce=CWUo1CVo5%2F9j6DaKlnWjbw%3D%3D&branch=refs%2Fpull%2F53301%2Fhead

but it just tells you the same thing. Maybe after it gets merged to main the behavior will be different? i.e., AzDO glitch?

Copy link

MemberAuthor

AndyAyersMS commentedMay 27, 2021

Thanks for double-checking. I'll submit a trial PR after I merge this just to make sure.

Copy link

MemberAuthor

AndyAyersMS commentedMay 27, 2021

FYI those retriggered runs all failed with

.packages/microsoft.dotnet.helix.sdk/6.0.0-beta.21271.3/tools/azure-pipelines/AzurePipelines.MultiQueue.targets(16,5): error : (NETCORE_ENGINEERING_TELEMETRY=Build) A call to an Azure DevOps api returned 401, which may indicate a bad 'System.AccessToken' value.Please Check the 'Make secrets available to builds of forks' in the pipeline pull request validation trigger settings.

Copy link

MemberAuthor

AndyAyersMS commentedMay 27, 2021

Remaining legs unrelated, so I'm going to merge this.

AndyAyersMS merged commitde6b2d9 intodotnet:main

May 27, 2021

AndyAyersMS deleted the PgoTesting branch

May 27, 2021 03:01

Copy link

Member

I think this change might be causing all PRs to fail. I haven't seen a single pass of this build definition and it's running on all PRs:https://dev.azure.com/dnceng/public/_build?definitionId=1002&_a=summary