[310p]: add rmsnorm gated fallback and unit test (#7424)
### What this PR does / why we need it?
RFC #7394
310P cannot use the fused `rmsnormgated` operator and must fall back to
the native implementation.
### Does this PR introduce _any_ user-facing change?
NO
### How was this patch tested?
ut
- vLLM version: v0.17.0
- vLLM main:
4497431df6
---------
Signed-off-by: Tflowers-0129 <2906339855@qq.com>
This commit is contained in:
41
tests/ut/_310p/ops/test_layernorm_310.py
Normal file
41
tests/ut/_310p/ops/test_layernorm_310.py
Normal file
@@ -0,0 +1,41 @@
|
||||
from unittest.mock import MagicMock, patch
|
||||
|
||||
import pytest
|
||||
import torch
|
||||
from vllm.config import set_current_vllm_config
|
||||
from vllm.model_executor.layers.layernorm import RMSNormGated
|
||||
|
||||
from vllm_ascend._310p.ops.layernorm import AscendRMSNormGated310
|
||||
|
||||
|
||||
@pytest.fixture(autouse=True)
|
||||
def default_vllm_config():
|
||||
mock_config = MagicMock()
|
||||
mock_config.compilation_config.custom_ops = ["all"]
|
||||
with set_current_vllm_config(mock_config):
|
||||
yield mock_config
|
||||
|
||||
|
||||
def test_rmsnorm_gated_310_forward_oot_uses_forward_native():
|
||||
layer = AscendRMSNormGated310(hidden_size=8, eps=1e-5)
|
||||
x = torch.randn(2, 8, dtype=torch.float32)
|
||||
z = torch.randn(2, 8, dtype=torch.float32)
|
||||
expected = torch.randn(2, 8, dtype=torch.float32)
|
||||
|
||||
with patch.object(RMSNormGated, "forward_native", autospec=True, return_value=expected) as mock_forward_native:
|
||||
out = layer.forward_oot(x, z)
|
||||
|
||||
mock_forward_native.assert_called_once_with(layer, x, z)
|
||||
assert out is expected
|
||||
|
||||
|
||||
def test_rmsnorm_gated_310_forward_oot_uses_forward_native_without_gate():
|
||||
layer = AscendRMSNormGated310(hidden_size=8, eps=1e-5)
|
||||
x = torch.randn(2, 8, dtype=torch.float32)
|
||||
expected = torch.randn(2, 8, dtype=torch.float32)
|
||||
|
||||
with patch.object(RMSNormGated, "forward_native", autospec=True, return_value=expected) as mock_forward_native:
|
||||
out = layer.forward_oot(x, None)
|
||||
|
||||
mock_forward_native.assert_called_once_with(layer, x, None)
|
||||
assert out is expected
|
||||
Reference in New Issue
Block a user