cuda
RtZero
这个作者很懒,什么都没留下…
展开
-
cuda,day-13,tensorcalc张量计算
//--------------------------//Tensor calculate//--------------------------#include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include #include #include原创 2016-05-18 12:00:48 · 584 阅读 · 0 评论 -
day-3,cdpSimplePrint
/*** Copyright 1993-2015 NVIDIA Corporation. All rights reserved.** Please refer to the NVIDIA end user license agreement (EULA) associated* with this source code for terms and conditions tha转载 2016-03-17 17:11:20 · 367 阅读 · 0 评论 -
CUDA 函数
CUDA里 A << B 是什么意思啊?相当于 A * 2^B那A >>B 呢?A / 2^B原创 2016-03-17 15:36:37 · 366 阅读 · 0 评论 -
CUDA,标准文件
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include using namespace std;#define u32 unsigned intint main(){return 0;}原创 2016-03-17 12:03:50 · 287 阅读 · 0 评论 -
CUDA,day-2,统计直方图
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include #include #include using namespace std;__global__ void func1(const unsig原创 2016-03-17 12:01:44 · 367 阅读 · 0 评论 -
CUDA,day-2,寄存器用法
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include using namespace std;__device__ static unsigned int d_tmp[NUM_ELEM];__global__ void test_gpu_gmem(unsigne原创 2016-03-17 09:34:07 · 377 阅读 · 0 评论 -
CUDA,day-2,合并列表
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include #include using namespace std;#define u32 unsigned int__global__ void gpu_sort_原创 2016-03-17 09:32:53 · 366 阅读 · 0 评论 -
CUDA,day-2,共享内存-排序算法
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include using namespace std;#define u32 unsigned int__device__ void func1(u32 * const sort_tmp,const u32原创 2016-03-17 09:32:15 · 371 阅读 · 0 评论 -
CUDA,day-2,二维数组操作
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func2(int * block_x,int * block_y,int * th原创 2016-03-17 09:31:53 · 1334 阅读 · 0 评论 -
CUDA,day-1,一维数组
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func1(int* block,int* thread,int* warp,int*原创 2016-03-17 09:30:51 · 992 阅读 · 0 评论 -
CUDA 内置变量
内置变量1.gridDim 此变量的类型为 dim3,包含网格的维度。2. blockIdx此变量的类型为 uint3,包含网格内的块索引。3.blockDim此变量的类型为 dim3,包含块的维度。4. threadIdx此变量的类型为 uint3,包含块内的线程索引。5.warpSize此变量的类型为 int,包含以线程为单位的 warp 块原创 2016-03-20 18:06:13 · 930 阅读 · 0 评论 -
CUDA 数学函数
C.1 数学函数C.1.1 atomicAdd()int atomicAdd(int* address, int val);unsigned int atomicAdd(unsigned int* address, unsigned int val);unsigned long long int atomicAdd(un原创 2016-03-20 18:10:31 · 3044 阅读 · 0 评论 -
cuda-矩阵乘法
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include "iostream"using namespace std;#define Mx 64#define My 32#define Nx 32#define Ny 48#define Px 64#defi原创 2016-04-24 08:59:52 · 350 阅读 · 0 评论 -
Cuda,多层感知机-cuda框架
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;void f(float *dst, float src[], int h, int w);void df(float *dst, float src[], int h原创 2016-04-12 11:38:34 · 401 阅读 · 0 评论 -
CUDA,day-9,大规模矩阵乘法
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;#define u32 unsigned int #define ARRAY_SIZE_X 8#define ARRAY_SIZE_Y 8u32原创 2016-04-04 16:30:37 · 342 阅读 · 0 评论 -
cuda,day-9,矩阵乘法
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;#define u32 unsigned int #define ARRAY_SIZE_X 2#define ARRAY_SIZE_Y 3u32原创 2016-04-04 16:17:42 · 247 阅读 · 0 评论 -
cuda,常量内存使用2
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include "assert.h"#include "conio.h"using namespace std;#define u16 unsigned short int#define u32 unsigned int原创 2016-03-24 16:53:35 · 17843 阅读 · 0 评论 -
cuda ,常量内存使用
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include "assert.h"#include "conio.h"using namespace std;#define CUDA_CALL(x){const cudaError_t a=(x);if(a!=cuda原创 2016-03-24 16:40:37 · 11865 阅读 · 0 评论 -
cuda,内存变量定义
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func1(int* data1, int* data2, int* data3);__dev原创 2016-03-24 16:13:57 · 736 阅读 · 0 评论 -
cuda 常量内存,头文件不知道有什么gui问题
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include "device_functions.h"#include "device_atomic_functions.h"#include #include #include "crt\func_macro.h"#include "asse原创 2016-03-24 15:43:32 · 1144 阅读 · 0 评论 -
CUDA 常数与指针的传递
#include #include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;__global__ void func3(int * const data, int num_elements);#define原创 2016-03-22 18:19:52 · 1840 阅读 · 0 评论 -
day-12,多层感知机-矩阵计算 <C版>
#include "cuda_runtime.h"#include "device_launch_parameters.h"#include #include #include using namespace std;void f(float *dst, float src[], int h, int w);void df(float *dst, float src[], int h原创 2016-04-11 09:42:33 · 464 阅读 · 0 评论