Run the MoE planner memory tests in the Megatron CI lane - #958
Merged
Merged
Conversation
The root unit-test environment lacks Megatron, so importorskip skipped 28 of the module's tests there. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
tests/unit/test_trainer_rank_moe_memory.pyneeds Megatron. It came from #877 and was extended by #954's HybridEP pricing and buffer-growth tests. The module importsart.megatron.lorathroughpytest.importorskip. CI's root unit-test step has no Megatron, so onmain(aab5d5416) 52 of its 57 tests were skipped and 5 ran. The skipped ones include the HybridEP tests.This moves the module into the "Megatron lightweight tests" step, which runs in
megatron_runtime/.venv, and ignores it in the root step. That is the pattern the other trainer-rank memory tests already use. Several of those modules import helpers from this one, so it was already importable in that lane; it just wasn't collected. The tests are CPU-only by design. Locally, with CUDA hidden, all 57 pass.The edits sit in different hunks from #955's workflow change, with unchanged lines between them. #955 wires in that PR's own new test the same way. A three-way merge of the two branches shows no conflicts.
🤖 Generated with Claude Code