fix: add cuda_runtime.h to hgemm_bind.cpp for cudaStream_t

2026-08-14 16:33:29 +00:00
parent 09e5261ba6
commit 3bee73207e
1 changed files with 1 additions and 0 deletions
--- a/ex_engine/xllm_kernels/cuda/bindings/hgemm_bind.cpp
+++ b/ex_engine/xllm_kernels/cuda/bindings/hgemm_bind.cpp
@@ -5,6 +5,7 @@
 //   moe_expert_gemm(input, weights, expert_counts) → output

 #include <torch/extension.h>
+#include <cuda_runtime.h>
 #include <cuda_fp16.h>
 #include <vector>