[ROCm] Fixes for GPTQ on ROCm (#2180)

This commit is contained in:
kliuae
2023-12-18 10:41:04 -08:00
committed by GitHub
parent bbe4466fd9
commit 1b7c791d60
4 changed files with 23 additions and 16 deletions
+1 -1
View File
@@ -219,13 +219,13 @@ vllm_extension_sources = [
"csrc/activation_kernels.cu",
"csrc/layernorm_kernels.cu",
"csrc/quantization/squeezellm/quant_cuda_kernel.cu",
"csrc/quantization/gptq/q_gemm.cu",
"csrc/cuda_utils_kernels.cu",
"csrc/pybind.cpp",
]
if _is_cuda():
vllm_extension_sources.append("csrc/quantization/awq/gemm_kernels.cu")
vllm_extension_sources.append("csrc/quantization/gptq/q_gemm.cu")
vllm_extension = CUDAExtension(
name="vllm._C",