Revert "[Kernel Slimming] Migrate NVFP4 kernels to JIT" (#20005)

This commit is contained in:
Baizhou Zhang
2026-03-05 19:40:00 -08:00
committed by GitHub
parent 6e5a2de354
commit 51e5dc845a
27 changed files with 987 additions and 2456 deletions
+6
View File
@@ -286,6 +286,11 @@ set(SOURCES
"csrc/gemm/fp8_blockwise_gemm_kernel.cu"
"csrc/gemm/fp8_gemm_kernel.cu"
"csrc/gemm/int8_gemm_kernel.cu"
"csrc/gemm/nvfp4_expert_quant.cu"
"csrc/gemm/nvfp4_quant_entry.cu"
"csrc/gemm/nvfp4_quant_kernels.cu"
"csrc/gemm/nvfp4_scaled_mm_entry.cu"
"csrc/gemm/nvfp4_scaled_mm_kernels.cu"
"csrc/gemm/per_tensor_quant_fp8.cu"
"csrc/gemm/per_token_group_quant_8bit.cu"
"csrc/gemm/per_token_group_quant_8bit_v2.cu"
@@ -311,6 +316,7 @@ set(SOURCES
"csrc/moe/moe_sum_reduce.cu"
"csrc/moe/moe_topk_softmax_kernels.cu"
"csrc/moe/moe_topk_sigmoid_kernels.cu"
"csrc/moe/nvfp4_blockwise_moe.cu"
"csrc/moe/fp8_blockwise_moe_kernel.cu"
"csrc/moe/prepare_moe_input.cu"