A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 6, 15:45 UTC
[Compile] Fix compile warning `st256_cs` in `cuda_vec_utils.cuh` (#36693)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Wentao Ye committed
Mar 16, 2026 at 11:12 UTC
ce8cf9161d2228745aa40135f6e427b603572597
1 file changed
+1
csrc/cuda_vec_utils.cuh
+1
@@ -196,6 +196,7 @@ __forceinline__ __device__ u32x8_t ld256_cs(const u32x8_t* addr) {
196
return val;
197
#else
198
assert(false && "ld256_cs requires SM100+ with CUDA 12.9+");
199
+ return u32x8_t{};
200
#endif
201
}
202