[Kernel/Model] Migrate mamba_ssm and causal_conv1d kernels to vLLM (#7651)
This commit is contained in:
@@ -203,6 +203,8 @@ if(VLLM_GPU_LANG STREQUAL "CUDA")
|
||||
FetchContent_MakeAvailable(cutlass)
|
||||
|
||||
list(APPEND VLLM_EXT_SRC
|
||||
"csrc/mamba/mamba_ssm/selective_scan_fwd.cu"
|
||||
"csrc/mamba/causal_conv1d/causal_conv1d.cu"
|
||||
"csrc/quantization/aqlm/gemm_kernels.cu"
|
||||
"csrc/quantization/awq/gemm_kernels.cu"
|
||||
"csrc/quantization/marlin/dense/marlin_cuda_kernel.cu"
|
||||
|
||||
Reference in New Issue
Block a user