NVIDIA / NVIDIA/Megatron-LM

🐛 CI failure: test_paged_stashing.py::TestPagedStashingOverBudget::test_overload_factor_and_over_budget

Open
#4,345 1 comment 0 reactions 1 assignee Claimed by @nanz-nv View on GitHub
bug
Dominant language
Python
Stars
17.9k
Forks
4.5k
Avg merge
4d 6h
Merged PRs (30d)
271

Description

**Describe the bug**

CI test \`tests/unit_tests/transformer/moe/test_paged_stashing.py::TestPagedStashingOverBudget::test_overload_factor_and_over_budget\` failed in job [tests/unit_tests/transformer/moe/**/*.py - latest](https://github.com/NVIDIA/Megatron-LM/actions/runs/24514641750/job/71660734791).
Tag the [@mcore-oncall](https://github.com/orgs/NVIDIA/teams/mcore-oncall) to get oncall's attention to this issue.

Same root cause as #4339 — \`TEGroupedMLP\` asserts \`_is_fused_impl_supported()\` when \`use_transformer_engine_op_fuser=True\`, but the combination with \`moe_paged_stash=True\` makes that check return \`False\`.

**Failing run**

| Field | Value |
|-------|-------|
| PR | [#4332: build: bump TransformerEngine to release_v2.14 (dev)](https://github.com/NVIDIA/Megatron-LM/pull/4332) |
| Run | [24514641750](https://github.com/NVIDIA/Megatron-LM/actions/runs/24514641750) |
| Job | [71660734791](https://github.com/NVIDIA/Megatron-LM/actions/runs/24514641750/job/71660734791) |

**Error**

```
AssertionError: Fused GroupedMLP is not supported for this configuration.
megatron/core/transformer/moe/experts.py:730: AssertionError

FAILED tests/unit_tests/transformer/moe/test_paged_stashing.py::TestPagedStashingOverBudget::test_overload_factor_and_over_budget
```

**Steps/Code to reproduce bug**

```bash
pytest tests/unit_tests/transformer/moe/test_paged_stashing.py::TestPagedStashingOverBudget::test_overload_factor_and_over_budget
```

**Additional context**

Triaged automatically via \`/triage-issue\`.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.