Skip to content

Commit b722c0b

Browse files
committed
cuda:移除无用cpu代码
1 parent 71d7983 commit b722c0b

4 files changed

Lines changed: 4 additions & 6 deletions

File tree

excuter/op-mem-cuda/CMakeLists.txt

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -32,7 +32,8 @@ file(GLOB_RECURSE CLIENT_SOURCES "src/client/*.cpp")
3232
# include(FetchContent)
3333

3434
find_package(CUDAToolkit REQUIRED COMPONENTS cublas)
35-
set(CUDA_TOOLKIT_ROOT_DIR /usr/local/cuda-12.6) # 根据您的安装路径进行调整
35+
include_directories(${CUDAToolkit_INCLUDE_DIRS})
36+
3637
set(CMAKE_CUDA_STANDARD 17)
3738
set(CMAKE_CUDA_ARCHITECTURES 61) # 根据您的 GPU 计算能力进行调整
3839
set(CMAKE_CUDA_SEPARABLE_COMPILATION ON) #确保 CMake 能够正确识别 CUDA 文件并将其编译为目标

excuter/op-mem-cuda/src/deepx/tensorfunc/equal.hpp

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,7 @@
11
#ifndef DEEPX_TENSORFUNC_EQUAL_HPP
22
#define DEEPX_TENSORFUNC_EQUAL_HPP
33
#include <cmath>
4-
#include <omp.h>
5-
4+
65
#include "deepx/tensor.hpp"
76
#include "deepx/shape.hpp"
87
namespace deepx::tensorfunc

excuter/op-mem-cuda/src/deepx/tensorfunc/matmul.hpp

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,6 @@
11
#ifndef DEEPX_TENSORFUNC_MATMUL_HPP
22
#define DEEPX_TENSORFUNC_MATMUL_HPP
3-
4-
#include <cblas.h> // 如果使用 OpenBLAS
3+
54
#include "deepx/tensor.hpp"
65

76
namespace deepx::tensorfunc

excuter/op-mem-cuda/src/deepx/tensorfunc/reduce.hpp

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,6 @@
55
#include <algorithm>
66
#include <stdexcept>
77
#include <deepx/vector_combination.hpp>
8-
#include <hwy/highway.h>
98

109
#include "deepx/tensor.hpp"
1110
#include "deepx/shape_reduce.hpp"

0 commit comments

Comments
 (0)