Skip to content

P
Paddle

BaiXuePrincess / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
提交 f2aa2747 编写于 作者: F Fan Yang
浏览文件
操作

fixed build issue of double definition of atomicAdd on modern GPUs

上级 df826899
显示空白变更内容
内联 并排
Showing with 4 addition and 0 deletion
paddle/cuda/include/hl_device_functions.cuh 100644 → 100755
浏览文件 @ f2aa2747
...@@ -16,6 +16,8 @@ limitations under the License. */ ...@@ -16,6 +16,8 @@ limitations under the License. */
#ifndef HL_DEVICE_FUNCTIONS_CUH_ #ifndef HL_DEVICE_FUNCTIONS_CUH_
#define HL_DEVICE_FUNCTIONS_CUH_ #define HL_DEVICE_FUNCTIONS_CUH_
#if defined(__CUDA_ARCH__) && __CUDA_ARCH__ < 600
namespace hppl { namespace hppl {
static __inline__ __device__ double atomicAdd(double* address, double val) { static __inline__ __device__ double atomicAdd(double* address, double val) {
...@@ -38,4 +40,6 @@ static __inline__ __device__ double atomicAdd(double* address, double val) { ...@@ -38,4 +40,6 @@ static __inline__ __device__ double atomicAdd(double* address, double val) {
using hppl::atomicAdd; using hppl::atomicAdd;
#endif
#endif /* HL_DEVICE_FUNCTIONS_CUH_ */ #endif /* HL_DEVICE_FUNCTIONS_CUH_ */
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或