Automatically setting shard_count on go_test based on t.Parallel() calls #1996
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
yushan26
force-pushed
the
divid-gazelle-shards
branch
from
76db7bb to
1f45778
Compare
fmeum
reviewed
Dec 18, 2024
| break | ||
| } | ||
| if info.isTest { | ||
| info.numParallel += countParallel(fdecl) |
Member
There was a problem hiding this comment.
Should we really be summing these up? My understanding has been that individual test functions in a file are not run in parallel, only the t.Parallel funcs in them. In that case taking the maximum would be more appropriate.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What type of PR is this?
What package or component does this PR mostly affect?
language/go
What does this PR do? Why is it needed?
Bazel may not necessarily schedule tests cases under target in parallel and may assign one core to a target. This means that tests running in t.Parallel may run sequentially. To make sure that the test cases are truly parallelized on heavily loaded machines, you should instead set shard_count on the go_test target. We can automatically setting shard_count on go_test based on the number of t.Parallel() calls at the top level function. This matches what people expects under go test.
This PR counts the number of t.Parallel calls in each target of the top level function and set the shard_count to the number of calls to ensure the test target is truly parallelized
Which issues(s) does this PR fix?
Fixes #1997
Other notes for review