使用 OpenGL 运行 cuda 程序时出现分段错误

Question

我正在编写一个 cuda 代码，可在http://code.google.com/p/snp-gpgpu/source/browse/trunk/cuda_by_example_codes/chapter08/basic_interop.cu?r=4

我

通过明确提及标题来替换标题，我提到它们如下：

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <GL/gl.h>
#include <GL/glut.h>
#include <cuda.h>
#include <cuda_runtime.h>
#include "cutil.h"
#include <cuda_gl_interop.h>

我正在编译发出命令的代码nvcc -o test_cuda basic_prog_num_1.cu -lGL -lGLU -lglut该程序正在编译而没有任何错误test_cuda并创建了可执行文件。但是当我尝试运行可执行文件时发出./test_cuda然后它给了我分割错误。我想提的一件事是，以前当我使用"cutil.h"编译时，发生了一个错误，即没有像cutil.h这样的文件或目录。所以我明确下载了"cutil.h"并包含在与程序相同的文件夹中。程序正在编译时没有任何错误，但在运行时返回分段错误。

代码给出如下：

PFNGLBINDBUFFERARBPROC    glBindBuffer     = NULL;
PFNGLDELETEBUFFERSARBPROC glDeleteBuffers  = NULL;
PFNGLGENBUFFERSARBPROC    glGenBuffers     = NULL;
PFNGLBUFFERDATAARBPROC    glBufferData     = NULL;
#define     DIM    512
GLuint  bufferObj;
cudaGraphicsResource *resource;
__global__ void kernel( uchar4 *ptr ) {
    // 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;
    // now calculate the value at that position
    float fx = x/(float)DIM - 0.5f;
    float fy = y/(float)DIM - 0.5f;
    unsigned char   green = 128 + 127 *
                            sin( abs(fx*100) - abs(fy*100) );
    // accessing uchar4 vs unsigned char*
    ptr[offset].x = 0;
    ptr[offset].y = green;
    ptr[offset].z = 0;
    ptr[offset].w = 255;
}
static void key_func( unsigned char key, int x, int y ) {
    switch (key) {
        case 27:
            HANDLE_ERROR( cudaGraphicsUnregisterResource( resource ) );
            glBindBuffer( GL_PIXEL_UNPACK_BUFFER_ARB, 0 );
            glDeleteBuffers( 1, &bufferObj );
            exit(0);
    }
}
static void draw_func( void ) {
    glDrawPixels( DIM, DIM, GL_RGBA, GL_UNSIGNED_BYTE, 0 );
    glutSwapBuffers();
}

int main( int argc, char **argv ) {
    cudaDeviceProp  prop;
    int dev;
    memset( &prop, 0, sizeof( cudaDeviceProp ) );
    prop.major = 1;
    prop.minor = 0;
    HANDLE_ERROR( cudaChooseDevice( &dev, &prop ) );
    cudaGLSetGLDevice( dev );

    glutInit( &argc, argv );
    glutInitDisplayMode( GLUT_DOU@harrism : BLE | GLUT_RGBA );
    glutInitWindowSize( DIM, DIM );
    glutCreateWindow( "bitmap" );
    /*glBindBuffer    = (PFNGLBINDBUFFERARBPROC)GET_PROC_ADDRESS("glBindBuffer");
    glDeleteBuffers = (PFNGLDELETEBUFFERSARBPROC)GET_PROC_ADDRESS("glDeleteBuffers");
    glGenBuffers    = (PFNGLGENBUFFERSARBPROC)GET_PROC_ADDRESS("glGenBuffers");
    glBufferData    = (PFNGLBUFFERDATAARBPROC)GET_PROC_ADDRESS("glBufferData");*/

    glGenBuffers( 1, &bufferObj );
    glBindBuffer( GL_PIXEL_UNPACK_BUFFER_ARB, bufferObj );
    glBufferData( GL_PIXEL_UNPACK_BUFFER_ARB, DIM * DIM * 4,
                  NULL, GL_DYNAMIC_DRAW_ARB );

        cudaGraphicsGLRegisterBuf@harrism : fer( &resource, 
                                      bufferObj, 
                                      cudaGraphicsMapFlagsNone );

   cudaGraphicsMapResources( 1, &resource, NULL );
    uchar4* devPtr;
    size_t  size;
        cudaGraphicsResourceGetMappedPointer( (void**)&devPtr, 
                                              &size, 
                                              resource);
    dim3    grids(DIM/16,DIM/16);
    dim3    threads(16,16);
    kernel<<<grids,threads>>>( devPtr );
  cudaGraphicsUnmapResources( 1, &resource, NULL );
    // set up GLUT and kick off main loop
    glutKeyboardFunc( key_func );
    glutDisplayFunc( draw_func );
    glutMainLoop();
}

Answer 1

你在代码中注释掉了原书示例中的几行关键行。 你为什么这样做，我不知道，但净效果是你在正确创建openGL渲染上下文之前尝试了某些OGL调用（glGenBuffers是seg错误的来源）。

取消注释这 4 行会导致显示代码和标头中缺少"GET_PROC_ADDRESS"宏。 这让我发现您没有正确包含原始书头。 您的代码中还包含一些垃圾，例如@harrism在各个地方。

无论如何，我

接受了您的代码并向其添加了一些内容，以下内容对我有用：

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <GL/gl.h>
#include <GL/glut.h>
#include <cuda.h>
#include <cuda_runtime.h>
// #include "cutil.h"
#include <cuda_gl_interop.h>
#include <GL/glext.h>
#include <GL/glx.h>
#define GET_PROC_ADDRESS( str ) glXGetProcAddress( (const GLubyte *)str )
static void HandleError( cudaError_t err, const char *file,  int line ) {
    if (err != cudaSuccess) {
            printf( "%s in %s at line %dn", cudaGetErrorString( err ),  file, line );
            exit( EXIT_FAILURE );
    }
}
#define HANDLE_ERROR( err ) (HandleError( err, __FILE__, __LINE__ ))

PFNGLBINDBUFFERARBPROC    glBindBuffer     = NULL;
PFNGLDELETEBUFFERSARBPROC glDeleteBuffers  = NULL;
PFNGLGENBUFFERSARBPROC    glGenBuffers     = NULL;
PFNGLBUFFERDATAARBPROC    glBufferData     = NULL;
#define     DIM    512
GLuint  bufferObj;
cudaGraphicsResource *resource;
__global__ void kernel( uchar4 *ptr ) {
// 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;
// now calculate the value at that position
float fx = x/(float)DIM - 0.5f;
float fy = y/(float)DIM - 0.5f;
unsigned char   green = 128 + 127 *
                        sin( abs(fx*100) - abs(fy*100) );
// accessing uchar4 vs unsigned char*
ptr[offset].x = 0;
ptr[offset].y = green;
ptr[offset].z = 0;
ptr[offset].w = 255;
}
static void key_func( unsigned char key, int x, int y ) {
  switch (key) {
    case 27:
        HANDLE_ERROR( cudaGraphicsUnregisterResource( resource ) );
        glBindBuffer( GL_PIXEL_UNPACK_BUFFER_ARB, 0 );
        glDeleteBuffers( 1, &bufferObj );
        exit(0);
  }
}
static void draw_func( void ) {
glDrawPixels( DIM, DIM, GL_RGBA, GL_UNSIGNED_BYTE, 0 );
glutSwapBuffers();
}

int main( int argc, char **argv ) {
cudaDeviceProp  prop;
int dev;
memset( &prop, 0, sizeof( cudaDeviceProp ) );
prop.major = 1;
prop.minor = 0;
HANDLE_ERROR( cudaChooseDevice( &dev, &prop ) );
cudaGLSetGLDevice( dev );
glutInit( &argc, argv );
glutInitDisplayMode( GLUT_DOUBLE | GLUT_RGBA );
glutInitWindowSize( DIM, DIM );
glutCreateWindow( "bitmap" );
glBindBuffer    = (PFNGLBINDBUFFERARBPROC)GET_PROC_ADDRESS("glBindBuffer");
glDeleteBuffers = (PFNGLDELETEBUFFERSARBPROC)GET_PROC_ADDRESS("glDeleteBuffers");
glGenBuffers    = (PFNGLGENBUFFERSARBPROC)GET_PROC_ADDRESS("glGenBuffers");
glBufferData    = (PFNGLBUFFERDATAARBPROC)GET_PROC_ADDRESS("glBufferData");

glGenBuffers( 1, &bufferObj );
glBindBuffer( GL_PIXEL_UNPACK_BUFFER_ARB, bufferObj );
glBufferData( GL_PIXEL_UNPACK_BUFFER_ARB, DIM * DIM * 4,
              NULL, GL_DYNAMIC_DRAW_ARB );

cudaGraphicsGLRegisterBuffer( &resource,
                                  bufferObj,
                                  cudaGraphicsMapFlagsNone );

cudaGraphicsMapResources( 1, &resource, NULL );
uchar4* devPtr;
size_t  size;
cudaGraphicsResourceGetMappedPointer( (void**)&devPtr,
                                          &size,
                                          resource);
dim3    grids(DIM/16,DIM/16);
dim3    threads(16,16);
kernel<<<grids,threads>>>( devPtr );
cudaGraphicsUnmapResources( 1, &resource, NULL );
// set up GLUT and kick off main loop
glutKeyboardFunc( key_func );
glutDisplayFunc( draw_func );
glutMainLoop();
}