CUDA（Ⅷ）：纹理内存（Texture Memory）

其他 2019-04-30 23:52:18 阅读次数: 0
版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/qq_24990189/article/details/89606221

1.学习目标：

了解纹理内存的性能特性

了解如何在CUDA C中使用一维纹理内存

了解如何在CUDA C中使用二维纹理内存

2.背景提出：

特定环境中使用特殊的内存将极大地提升应用程序的性能。与Constant Memory相似，Texture Memory同样缓存在芯片上，只读内存，在某些情况下，能减少对内存的请求并提供更高效的内存带宽。Texture Mmeory是专门为那些在内存访问模式中存在大量空间局部性（Spatial Locality）的图形应用程序而设计的。意思是，在某个计算应用程序中，这意味着一个Thread读取的位置可能与邻近Thread读取的位置“非常接近”。

从数学角度，图中四个地址并非连续，在一般的CPU缓存模式中，这些地址将不会缓存。GPU纹理缓存是为了加速这种访问模式而设设计的，因此如果在这种情况中使用纹理内存而不是全局内存，那么将会获得性能提升。

3.案例分析：二维热传导模拟

构造：

格网中随机散步“热源”，他们有着不同的固定的温度。在给定了矩形格网以及热源分布后，我们可以计算格网中每个单元的温度随时间的变化情况。为了简单，热源单元本身的温度将保持不变。但是，热量从更热的单元传导到更冷的单元

温度更新计算：

 

4.代码复现-一维纹理内存版本-- texture、cudaBindTexture()

#include "cuda.h"
#include "../common/book.h"
#include "../common/cpu_anim.h"

#define DIM 1024
#define PI 3.1415926535897932f
#define MAX_TEMP 1.0f
#define MIN_TEMP 0.0001f
#define SPEED   0.25f

// these exist on the GPU side
// 这些变量将位与GPU上
texture<float>  texConstSrc;
texture<float>  texIn;
texture<float>  texOut;



// this kernel takes in a 2-d array of floats
// it updates the value-of-interest by a scaled value based
// on itself and its nearest neighbors
//给定一个输入格网，据等式7.2中的更新公式计算输出温度格网。
__global__ void blend_kernel( float *dst,
                              bool dstOut ) {
    // map from threadIdx/BlockIdx to pixel position
    int x = threadIdx.x + blockIdx.x * blockDim.x;
    int y = threadIdx.y + blockIdx.y * blockDim.y;
    int offset = x + y * blockDim.x * gridDim.x;

    int left = offset - 1;
    int right = offset + 1;
    if (x == 0)   left++;
    if (x == DIM-1) right--; 

    int top = offset - DIM;
    int bottom = offset + DIM;
    if (y == 0)   top += DIM;
    if (y == DIM-1) bottom -= DIM;

    float   t, l, c, r, b;
    if (dstOut) {
        t = tex1Dfetch(texIn,top);
        l = tex1Dfetch(texIn,left);
        c = tex1Dfetch(texIn,offset);
        r = tex1Dfetch(texIn,right);
        b = tex1Dfetch(texIn,bottom);

    } else {
        t = tex1Dfetch(texOut,top);
        l = tex1Dfetch(texOut,left);
        c = tex1Dfetch(texOut,offset);
        r = tex1Dfetch(texOut,right);
        b = tex1Dfetch(texOut,bottom);
    }
    dst[offset] = c + SPEED * (t + b + r + l - 4 * c);
}

// NOTE - texOffsetConstSrc could either be passed as a
// parameter to this function, or passed in __constant__ memory
// if we declared it as a global above, it would be
// a parameter here: 
//给定一个包含初始输入温度的格网，将其中作为热源的单元温度值复制到格网相应单元中。这将覆盖这些大院之前计算出的温度。
// __global__ void copy_const_kernel( float *iptr,
//                                    size_t texOffset )
__global__ void copy_const_kernel( float *iptr ) {
    // map from threadIdx/BlockIdx to pixel position
    int x = threadIdx.x + blockIdx.x * blockDim.x;
    int y = threadIdx.y + blockIdx.y * blockDim.y;
    int offset = x + y * blockDim.x * gridDim.x;

    float c = tex1Dfetch(texConstSrc,offset);
    if (c != 0)
        iptr[offset] = c;
}

// globals needed by the update routine
struct DataBlock {
    unsigned char   *output_bitmap;
    float           *dev_inSrc;
    float           *dev_outSrc;
    float           *dev_constSrc;
    CPUAnimBitmap  *bitmap;

    cudaEvent_t     start, stop;
    float           totalTime;
    float           frames;
};

void anim_gpu( DataBlock *d, int ticks ) {
    HANDLE_ERROR( cudaEventRecord( d->start, 0 ) );
    dim3    blocks(DIM/16,DIM/16);
    dim3    threads(16,16);
    CPUAnimBitmap  *bitmap = d->bitmap;

    // since tex is global and bound, we have to use a flag to
    // select which is in/out per iteration
    volatile bool dstOut = true;
    for (int i=0; i<90; i++) {
        float   *in, *out;
        if (dstOut) {
            in  = d->dev_inSrc;
            out = d->dev_outSrc;
        } else {
            out = d->dev_inSrc;
            in  = d->dev_outSrc;
        }
        copy_const_kernel<<<blocks,threads>>>( in );
        blend_kernel<<<blocks,threads>>>( out, dstOut );
        dstOut = !dstOut;
    }
    float_to_color<<<blocks,threads>>>( d->output_bitmap,
                                        d->dev_inSrc );

    HANDLE_ERROR( cudaMemcpy( bitmap->get_ptr(),
                              d->output_bitmap,
                              bitmap->image_size(),
                              cudaMemcpyDeviceToHost ) );

    HANDLE_ERROR( cudaEventRecord( d->stop, 0 ) );
    HANDLE_ERROR( cudaEventSynchronize( d->stop ) );
    float   elapsedTime;
    HANDLE_ERROR( cudaEventElapsedTime( &elapsedTime,
                                        d->start, d->stop ) );
    d->totalTime += elapsedTime;
    ++d->frames;
    printf( "Average Time per frame:  %3.1f ms\n",
            d->totalTime/d->frames  );
}

// clean up memory allocated on the GPU
void anim_exit( DataBlock *d ) {
    cudaUnbindTexture( texIn );
    cudaUnbindTexture( texOut );
    cudaUnbindTexture( texConstSrc );
    HANDLE_ERROR( cudaFree( d->dev_inSrc ) );
    HANDLE_ERROR( cudaFree( d->dev_outSrc ) );
    HANDLE_ERROR( cudaFree( d->dev_constSrc ) );

    HANDLE_ERROR( cudaEventDestroy( d->start ) );
    HANDLE_ERROR( cudaEventDestroy( d->stop ) );
}


int main( void ) {
    DataBlock   data;
    CPUAnimBitmap bitmap( DIM, DIM, &data );
    data.bitmap = &bitmap;
    data.totalTime = 0;
    data.frames = 0;
    HANDLE_ERROR( cudaEventCreate( &data.start ) );
    HANDLE_ERROR( cudaEventCreate( &data.stop ) );

    int imageSize = bitmap.image_size();

    HANDLE_ERROR( cudaMalloc( (void**)&data.output_bitmap,
                               imageSize ) );

    // assume float == 4 chars in size (ie rgba)
    HANDLE_ERROR( cudaMalloc( (void**)&data.dev_inSrc,
                              imageSize ) );
    HANDLE_ERROR( cudaMalloc( (void**)&data.dev_outSrc,
                              imageSize ) );
    HANDLE_ERROR( cudaMalloc( (void**)&data.dev_constSrc,
                              imageSize ) );

    HANDLE_ERROR( cudaBindTexture( NULL, texConstSrc,
                                   data.dev_constSrc,
                                   imageSize ) );

    HANDLE_ERROR( cudaBindTexture( NULL, texIn,
                                   data.dev_inSrc,
                                   imageSize ) );

    HANDLE_ERROR( cudaBindTexture( NULL, texOut,
                                   data.dev_outSrc,
                                   imageSize ) );

    // intialize the constant data
    float *temp = (float*)malloc( imageSize );
    for (int i=0; i<DIM*DIM; i++) {
        temp[i] = 0;
        int x = i % DIM;
        int y = i / DIM;
        if ((x>300) && (x<600) && (y>310) && (y<601))
            temp[i] = MAX_TEMP;
    }
    temp[DIM*100+100] = (MAX_TEMP + MIN_TEMP)/2;
    temp[DIM*700+100] = MIN_TEMP;
    temp[DIM*300+300] = MIN_TEMP;
    temp[DIM*200+700] = MIN_TEMP;
    for (int y=800; y<900; y++) {
        for (int x=400; x<500; x++) {
            temp[x+y*DIM] = MIN_TEMP;
        }
    }
    HANDLE_ERROR( cudaMemcpy( data.dev_constSrc, temp,
                              imageSize,
                              cudaMemcpyHostToDevice ) );    

    // initialize the input data
    for (int y=800; y<DIM; y++) {
        for (int x=0; x<200; x++) {
            temp[x+y*DIM] = MAX_TEMP;
        }
    }
    HANDLE_ERROR( cudaMemcpy( data.dev_inSrc, temp,
                              imageSize,
                              cudaMemcpyHostToDevice ) );
    free( temp );

    bitmap.anim_and_exit( (void (*)(void*,int))anim_gpu,
                           (void (*)(void*))anim_exit );
}

 

wlsh@wlsh-ThinkStation:~/Desktop/GPU高性能编程CUDA实战—示例代码/chapter07$ nvcc -o heat heat.cu -lglut -lGL -lGLU
../common/cpu_anim.h(61): warning: conversion from a string literal to "char *" is deprecated

../common/cpu_anim.h(61): warning: conversion from a string literal to "char *" is deprecated

wlsh@wlsh-ThinkStation:~/Desktop/GPU高性能编程CUDA实战—示例代码/chapter07$ ./heat 
wlsh@wlsh-ThinkStation:~/Desktop/GPU高性能编程CUDA实战—示例代码/chapter07$ ./heat 
Average Time per frame:  9.2 ms
Average Time per frame:  8.6 ms
Average Time per frame:  8.5 ms
Average Time per frame:  8.5 ms
Average Time per frame:  8.3 ms
Average Time per frame:  8.3 ms
Average Time per frame:  8.3 ms
Average Time per frame:  8.2 ms
Average Time per frame:  8.2 ms
Average Time per frame:  8.1 ms
Average Time per frame:  8.2 ms
Average Time per frame:  8.2 ms
Average Time per frame:  8.3 ms
Average Time per frame:  8.7 ms
Average Time per frame:  9.0 ms
Average Time per frame:  8.9 ms
Average Time per frame:  8.9 ms

5.代码复现-二维纹理内存版本

#include "cuda.h"
#include "../common/book.h"
#include "../common/cpu_anim.h"

#define DIM 1024
#define PI 3.1415926535897932f
#define MAX_TEMP 1.0f
#define MIN_TEMP 0.0001f
#define SPEED   0.25f

// these exist on the GPU side
texture<float,2>  texConstSrc;
texture<float,2>  texIn;
texture<float,2>  texOut;

__global__ void blend_kernel( float *dst,
                              bool dstOut ) {
    // map from threadIdx/BlockIdx to pixel position
    int x = threadIdx.x + blockIdx.x * blockDim.x;
    int y = threadIdx.y + blockIdx.y * blockDim.y;
    int offset = x + y * blockDim.x * gridDim.x;

    float   t, l, c, r, b;
    if (dstOut) {
        t = tex2D(texIn,x,y-1);
        l = tex2D(texIn,x-1,y);
        c = tex2D(texIn,x,y);
        r = tex2D(texIn,x+1,y);
        b = tex2D(texIn,x,y+1);
    } else {
        t = tex2D(texOut,x,y-1);
        l = tex2D(texOut,x-1,y);
        c = tex2D(texOut,x,y);
        r = tex2D(texOut,x+1,y);
        b = tex2D(texOut,x,y+1);
    }
    dst[offset] = c + SPEED * (t + b + r + l - 4 * c);
}

__global__ void copy_const_kernel( float *iptr ) {
    // map from threadIdx/BlockIdx to pixel position
    int x = threadIdx.x + blockIdx.x * blockDim.x;
    int y = threadIdx.y + blockIdx.y * blockDim.y;
    int offset = x + y * blockDim.x * gridDim.x;

    float c = tex2D(texConstSrc,x,y);
    if (c != 0)
        iptr[offset] = c;
}

// globals needed by the update routine
struct DataBlock {
    unsigned char   *output_bitmap;
    float           *dev_inSrc;
    float           *dev_outSrc;
    float           *dev_constSrc;
    CPUAnimBitmap  *bitmap;

    cudaEvent_t     start, stop;
    float           totalTime;
    float           frames;
};

void anim_gpu( DataBlock *d, int ticks ) {
    HANDLE_ERROR( cudaEventRecord( d->start, 0 ) );
    dim3    blocks(DIM/16,DIM/16);
    dim3    threads(16,16);
    CPUAnimBitmap  *bitmap = d->bitmap;

    // since tex is global and bound, we have to use a flag to
    // select which is in/out per iteration
    volatile bool dstOut = true;
    for (int i=0; i<90; i++) {
        float   *in, *out;
        if (dstOut) {
            in  = d->dev_inSrc;
            out = d->dev_outSrc;
        } else {
            out = d->dev_inSrc;
            in  = d->dev_outSrc;
        }
        copy_const_kernel<<<blocks,threads>>>( in );
        blend_kernel<<<blocks,threads>>>( out, dstOut );
        dstOut = !dstOut;
    }
    float_to_color<<<blocks,threads>>>( d->output_bitmap,
                                        d->dev_inSrc );

    HANDLE_ERROR( cudaMemcpy( bitmap->get_ptr(),
                              d->output_bitmap,
                              bitmap->image_size(),
                              cudaMemcpyDeviceToHost ) );

    HANDLE_ERROR( cudaEventRecord( d->stop, 0 ) );
    HANDLE_ERROR( cudaEventSynchronize( d->stop ) );
    float   elapsedTime;
    HANDLE_ERROR( cudaEventElapsedTime( &elapsedTime,
                                        d->start, d->stop ) );
    d->totalTime += elapsedTime;
    ++d->frames;
    printf( "Average Time per frame:  %3.1f ms\n",
            d->totalTime/d->frames  );
}

// clean up memory allocated on the GPU
void anim_exit( DataBlock *d ) {
    cudaUnbindTexture( texIn );
    cudaUnbindTexture( texOut );
    cudaUnbindTexture( texConstSrc );
    HANDLE_ERROR( cudaFree( d->dev_inSrc ) );
    HANDLE_ERROR( cudaFree( d->dev_outSrc ) );
    HANDLE_ERROR( cudaFree( d->dev_constSrc ) );

    HANDLE_ERROR( cudaEventDestroy( d->start ) );
    HANDLE_ERROR( cudaEventDestroy( d->stop ) );
}


int main( void ) {
    DataBlock   data;
    CPUAnimBitmap bitmap( DIM, DIM, &data );
    data.bitmap = &bitmap;
    data.totalTime = 0;
    data.frames = 0;
    HANDLE_ERROR( cudaEventCreate( &data.start ) );
    HANDLE_ERROR( cudaEventCreate( &data.stop ) );

    int imageSize = bitmap.image_size();

    HANDLE_ERROR( cudaMalloc( (void**)&data.output_bitmap,
                               imageSize ) );

    // assume float == 4 chars in size (ie rgba)
    HANDLE_ERROR( cudaMalloc( (void**)&data.dev_inSrc,
                              imageSize ) );
    HANDLE_ERROR( cudaMalloc( (void**)&data.dev_outSrc,
                              imageSize ) );
    HANDLE_ERROR( cudaMalloc( (void**)&data.dev_constSrc,
                              imageSize ) );

    cudaChannelFormatDesc desc = cudaCreateChannelDesc<float>();
    HANDLE_ERROR( cudaBindTexture2D( NULL, texConstSrc,
                                   data.dev_constSrc,
                                   desc, DIM, DIM,
                                   sizeof(float) * DIM ) );

    HANDLE_ERROR( cudaBindTexture2D( NULL, texIn,
                                   data.dev_inSrc,
                                   desc, DIM, DIM,
                                   sizeof(float) * DIM ) );

    HANDLE_ERROR( cudaBindTexture2D( NULL, texOut,
                                   data.dev_outSrc,
                                   desc, DIM, DIM,
                                   sizeof(float) * DIM ) );

    // initialize the constant data
    float *temp = (float*)malloc( imageSize );
    for (int i=0; i<DIM*DIM; i++) {
        temp[i] = 0;
        int x = i % DIM;
        int y = i / DIM;
        if ((x>300) && (x<600) && (y>310) && (y<601))
            temp[i] = MAX_TEMP;
    }
    temp[DIM*100+100] = (MAX_TEMP + MIN_TEMP)/2;
    temp[DIM*700+100] = MIN_TEMP;
    temp[DIM*300+300] = MIN_TEMP;
    temp[DIM*200+700] = MIN_TEMP;
    for (int y=800; y<900; y++) {
        for (int x=400; x<500; x++) {
            temp[x+y*DIM] = MIN_TEMP;
        }
    }
    HANDLE_ERROR( cudaMemcpy( data.dev_constSrc, temp,
                              imageSize,
                              cudaMemcpyHostToDevice ) );    

    // initialize the input data
    for (int y=800; y<DIM; y++) {
        for (int x=0; x<200; x++) {
            temp[x+y*DIM] = MAX_TEMP;
        }
    }
    HANDLE_ERROR( cudaMemcpy( data.dev_inSrc, temp,
                              imageSize,
                              cudaMemcpyHostToDevice ) );
    free( temp );

    bitmap.anim_and_exit( (void (*)(void*,int))anim_gpu,
                           (void (*)(void*))anim_exit );
}

wlsh@wlsh-ThinkStation:~/Desktop/GPU高性能编程CUDA实战—示例代码/chapter07$ nvcc -o heat_2d heat_2d.cu -lglut -lGL -lGLU

wlsh@wlsh-ThinkStation:~/Desktop/GPU高性能编程CUDA实战—示例代码/chapter07$ ./heat_2d

 

猜你喜欢

转载自blog.csdn.net/qq_24990189/article/details/89606221
今日推荐
火速冲上 GitHub 热榜 —— 开源编程语言、框架哪有这么可爱？
北京人形机器人创新中心发布全球首个纯电驱拟人奔跑的全尺寸人形机器人“天工”
LFOSSA 源来如此公开课 | 掌握云原生未来：CNCF 认证全面攻略与备考秘籍
国产云输入法——仅华为无云端数据上传安全问题
开源日报 | 工业开源项目OGG 1.0；姐姐，你要和我一起配置火狐吗；苹果AI遥遥落后？Fedora 40
开放签电子签章：停止新增，优化体验，前进更进（五一假期前工作）
开源日报 | 中学生开源前端动画引擎；全球首个Llama3 8B中文版开源模型；联想电脑恐出局；Linus讽刺AI炒作
周排行
浏览器对同一域名进行请求的最大并发连接数
React Hook之自定义Hook
【转】MyBatis缓存机制
-Java-泛型
自动化测试常用脚本-发送邮件
LeetCode#859: Buddy Strings
java、Python处理字符串
第二篇の博客
Hadoop伪分布式环境安装
SQL Server进阶（十一）临时表、表变量
每日归档
更多
2024-04-27(56)
2024-04-26(39)
2024-04-25(22)
2024-04-24(36)
2024-04-23(26)
2024-04-22(39)
2024-04-21(0)
2024-04-20(6)
2024-04-19(5)
2024-04-18(0)

代码天地 © 2018 codetd.com

境外服务器   最新电视剧2022