xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-03 23:53:10 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix][ROCm] Fix typo: is_linear_fp8_enaled -> is_linear_fp8_enabled (#31109)

Browse Source 
Signed-off-by: c0de128 <kevin.mckay@outlook.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
This commit is contained in:
Kevin McKay 2025-12-21 23:14:58 -06:00 committed by GitHub
parent 44ae85f725
commit cf8eed7bef
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/_aiter_ops.py
View File

@ -761,7 +761,7 @@ class rocm_aiter_ops:
@classmethod
@if_aiter_supported
def is_linear_fp8_enaled(cls) -> bool:
def is_linear_fp8_enabled(cls) -> bool:
return cls.is_linear_enabled()
@classmethod

2
vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py
View File

@ -61,7 +61,7 @@ class CompressedTensorsW8A8Fp8(CompressedTensorsScheme):
)
self.cutlass_block_fp8_supported = cutlass_block_fp8_supported()
self.use_aiter_and_is_supported = rocm_aiter_ops.is_linear_fp8_enaled()
self.use_aiter_and_is_supported = rocm_aiter_ops.is_linear_fp8_enabled()
if self.weight_block_size is not None:
assert not self.is_static_input_scheme

2
vllm/model_executor/layers/quantization/fp8.py
View File

@ -414,7 +414,7 @@ class Fp8LinearMethod(LinearMethodBase):
if vllm_is_batch_invariant():
self.use_marlin = False
self.use_aiter_and_is_supported = rocm_aiter_ops.is_linear_fp8_enaled()
self.use_aiter_and_is_supported = rocm_aiter_ops.is_linear_fp8_enabled()
self.use_deep_gemm = is_deep_gemm_supported()
self.weight_block_size = self.quant_config.weight_block_size

2
vllm/model_executor/layers/quantization/input_quant_fp8.py
View File

@ -51,7 +51,7 @@ class QuantFP8(CustomOp):
self.column_major_scales = column_major_scales
self.use_ue8m0 = use_ue8m0
self.use_aiter = rocm_aiter_ops.is_linear_fp8_enaled()
self.use_aiter = rocm_aiter_ops.is_linear_fp8_enabled()
self.is_group_quant = group_shape.is_per_group()
if self.is_group_quant:

2
vllm/platforms/rocm.py
View File

@ -408,7 +408,7 @@ class RocmPlatform(Platform):
parallel_config = vllm_config.parallel_config
is_eager_execution = compilation_config == CUDAGraphMode.NONE
use_aiter_rms_norm = rocm_aiter_ops.is_rmsnorm_enabled()
use_aiter_fp8_linear = rocm_aiter_ops.is_linear_fp8_enaled()
use_aiter_fp8_linear = rocm_aiter_ops.is_linear_fp8_enabled()
if compilation_config.cudagraph_mode.has_full_cudagraphs():
# decode context parallel does not support full cudagraphs