apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 00:17:25 +08:00
Code Issues Actions 19 Packages Projects Releases Wiki Activity

Revert cuda check (#5915)

Browse Source 
* commit

* commit
This commit is contained in:
周周周
2026-01-07 14:40:18 +08:00
committed by GitHub
parent 29898372e9
commit f15df1ec89
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
custom_ops/gpu_ops/moe/deepgemm_preprocess.cu
+2 -2
View File
@@ -51,7 +51,7 @@ std::vector<paddle::Tensor> count_tokens_per_expert_func(
auto stream = topk_ids.stream();
using scalar_t = int64_t;
CUDA_CHECK(cudaGetLastError());
// CUDA_CHECK(cudaGetLastError());
cuda_kernel<<<1, 1024, num_experts * sizeof(int32_t), stream>>>(
topk_ids.data<scalar_t>(),
token_nums_per_expert.data<int32_t>(),
@@ -59,7 +59,7 @@ std::vector<paddle::Tensor> count_tokens_per_expert_func(
topk_ids_numel,
num_experts);
CUDA_CHECK(cudaGetLastError());
// CUDA_CHECK(cudaGetLastError());
return {token_nums_per_expert};
}