iHore/vLLM
8
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

[ROCm][Test] Fix stale test_gfx950_moe MXFP4 oracle tests (#46260)

Browse Source 
Signed-off-by: Spandan Tiwari <23646532+spandantiwari@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
Spandan Tiwari
2026-06-23 23:07:46 +08:00
committed by GitHub
co-authored by Claude mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent 156b12667c
commit 2aaaf3febd
1 changed files with 23 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/quantization/test_gfx950_moe.py
+23 -8
View File
@@ -42,7 +42,7 @@ def _make_w4a4_moe_config(moe_backend: str = "auto") -> FusedMoEConfig:
num_experts=8,
experts_per_token=2,
hidden_dim=256,
intermediate_size_per_partition=256,
intermediate_size=256,
num_local_experts=8,
num_logical_experts=8,
moe_parallel_config=FusedMoEParallelConfig.make_no_parallel(),
@@ -54,9 +54,22 @@ def _make_w4a4_moe_config(moe_backend: str = "auto") -> FusedMoEConfig:
)
@pytest.fixture
def mxfp4_oracle_config():
"""Stub the config the oracle reads (``model_config.quantization_config``)
so backend dispatch resolves without a real model / user override."""
from unittest.mock import patch
with patch(
"vllm.model_executor.layers.fused_moe.oracle.mxfp4.get_current_vllm_config"
) as mock_get_config:
mock_get_config.return_value.model_config.quantization_config = None
yield
@pytest.mark.skipif(not ROCM_GFX950, reason="Requires GFX950 (mi355x)")
@pytest.mark.skipif(not ROCM_AITER_AVAILABLE, reason="Requires AITER enabled")
def test_w4a4_dispatches_to_aiter():
def test_w4a4_dispatches_to_aiter(mxfp4_oracle_config):
"""With AITER enabled + GFX950, W4A4 selects AITER_MXFP4_MXFP4."""
config = _make_w4a4_moe_config()
backend, experts_cls = select_mxfp4_moe_backend(
@@ -71,16 +84,18 @@ def test_w4a4_dispatches_to_aiter():
ROCM_AITER_AVAILABLE,
reason="Test requires AITER disabled (unset VLLM_ROCM_USE_AITER)",
)
def test_w4a4_raises_without_aiter_and_no_moe_backend():
"""Without AITER and no --moe-backend, raises NotImplementedError
with hint to use --moe-backend emulation."""
def test_w4a4_falls_back_to_triton_unfused_without_aiter(mxfp4_oracle_config):
"""Without AITER and no --moe-backend, ROCm falls back to TRITON_UNFUSED."""
config = _make_w4a4_moe_config()
with pytest.raises(NotImplementedError, match="--moe-backend emulation"):
select_mxfp4_moe_backend(config, activation_key=kMxfp4Dynamic)
backend, experts_cls = select_mxfp4_moe_backend(
config, activation_key=kMxfp4Dynamic
)
assert backend == Mxfp4MoeBackend.TRITON_UNFUSED
assert experts_cls is not None
@pytest.mark.skipif(not ROCM_GFX950, reason="Requires GFX950 (mi355x)")
def test_w4a4_dispatches_to_emulation_with_moe_backend():
def test_w4a4_dispatches_to_emulation_with_moe_backend(mxfp4_oracle_config):
"""With --moe-backend emulation, W4A4 selects EMULATION."""
config = _make_w4a4_moe_config(moe_backend="emulation")
backend, experts_cls = select_mxfp4_moe_backend(