添加showFrame

Hu Chunming
1 parent a11c5683
Showing 7 changed files with 61 additions and 59 deletions
FFNvDecoder/FFNvDecoder.vcxproj
FFNvDecoder/FFNvDecoder.vcxproj.filters
FFNvDecoder/NV12ToRGB.cu
FFNvDecoder/cudaHeader.cu
FFNvDecoder/cudaHeader.h
FFNvDecoder/dec.jpg
FFNvDecoder/main.cpp
@@ -48,13 +48,13 @@
       <Optimization>Disabled</Optimization>
       <PreprocessorDefinitions>WIN32;WIN64;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <AdditionalOptions>/utf-8</AdditionalOptions>
-      <AdditionalIncludeDirectories>..\3rdparty\ffmpeg-5.0.1-win64-dev\include;./;./common/inc;./common/UtilNPP;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\3rdparty\ffmpeg-5.0.1-win64-dev\include;./;./common/inc;./common/UtilNPP;D:\win_dev\opencv\build\include;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
     </ClCompile>
     <Link>
       <GenerateDebugInformation>true</GenerateDebugInformation>
       <SubSystem>Console</SubSystem>
-      <AdditionalDependencies>avcodec.lib;avdevice.lib;avfilter.lib;avformat.lib;avutil.lib;postproc.lib;swresample.lib;swscale.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;nvjpeg.lib;%(AdditionalDependencies)</AdditionalDependencies>
-      <AdditionalLibraryDirectories>..\3rdparty\ffmpeg-5.0.1-win64-dev\lib;%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
+      <AdditionalDependencies>avcodec.lib;avdevice.lib;avfilter.lib;avformat.lib;avutil.lib;postproc.lib;swresample.lib;swscale.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;nvjpeg.lib;opencv_world455d.lib;freeglut.lib;glew64.lib;%(AdditionalDependencies)</AdditionalDependencies>
+      <AdditionalLibraryDirectories>..\3rdparty\ffmpeg-5.0.1-win64-dev\lib;D:\win_dev\opencv\build\x64\vc14\lib;../3rdparty/gl;%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
     </Link>
     <CudaCompile>
       <TargetMachinePlatform>64</TargetMachinePlatform>
@@ -80,7 +80,6 @@
     </CudaCompile>
   </ItemDefinitionGroup>
   <ItemGroup>
-    <CudaCompile Include="cudaHeader.cu" />
     <CudaCompile Include="NV12ToRGB.cu" />
   </ItemGroup>
   <ItemGroup>
@@ -94,7 +93,6 @@
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="check_tool.h" />
-    <ClInclude Include="cudaHeader.h" />
     <ClInclude Include="cuda_kernels.h" />
     <ClInclude Include="FFCuContextManager.h" />
     <ClInclude Include="FFNvDecoder.h" />
 <?xml version="1.0" encoding="utf-8"?>
 <Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
   <ItemGroup>
-    <CudaCompile Include="cudaHeader.cu">
-      <Filter>cu_src</Filter>
-    </CudaCompile>
     <CudaCompile Include="NV12ToRGB.cu">
       <Filter>cu_src</Filter>
     </CudaCompile>
@@ -58,9 +55,6 @@
     <ClInclude Include="FrameQueue.h">
       <Filter>include</Filter>
     </ClInclude>
-    <ClInclude Include="cudaHeader.h">
-      <Filter>cu_src</Filter>
-    </ClInclude>
     <ClInclude Include="cuda_kernels.h">
       <Filter>cu_src</Filter>
     </ClInclude>
@@ -52,9 +52,9 @@ namespace cuda_common
 			return x;
 		}
 	}
-
 	// CUDA kernel for outputing the final RGB output from NV12;
-	__global__ void NV12ToRGB_drvapi2(uint32 *srcImage, size_t nSourcePitch, unsigned char *dstImage, int width, int height)
+	extern "C"
+		__global__ void NV12ToRGB_drvapi2(uint32 *srcImage, size_t nSourcePitch, unsigned char *dstImage, int width, int height)
 	{
 		int32 x, y;
@@ -169,8 +169,9 @@ namespace cuda_common
 	}
-	// CUDA kernel for outputing the final RGB output from NV12;
-	__global__ void CUDAToBGR_drvapi(uint32 *dataY, uint32 *dataUV, size_t pitchY, size_t pitchUV, unsigned char *dstImage, int width, int height)
+		// CUDA kernel for outputing the final RGB output from NV12;
+	extern "C"
+		__global__ void CUDAToBGR_drvapi(uint32 *dataY, uint32 *dataUV, size_t pitchY, size_t pitchUV, unsigned char *dstImage, int width, int height)
 	{
 		int32 x, y;
@@ -307,7 +308,7 @@ namespace cuda_common
 	{
 		dim3 block(32, 16, 1);
 		dim3 grid((width + (2 * block.x - 1)) / (2 * block.x), (height + (block.y - 1)) / block.y, 1);
-		NV12ToRGB_drvapi2 <<< grid, block >>>((uint32 *)d_srcNV12, nSourcePitch, d_dstRGB, width, height);
+		NV12ToRGB_drvapi2 << < grid, block >> >((uint32 *)d_srcNV12, nSourcePitch, d_dstRGB, width, height);
 		cudaError_t cudaStatus = cudaGetLastError();
 		if (cudaStatus != cudaSuccess) {
 			fprintf(stderr, "NV12ToRGB_drvapi launch failed: %s\n", cudaGetErrorString(cudaStatus));
@@ -327,7 +328,7 @@ namespace cuda_common
 	{
 		dim3 block(32, 16, 1);
 		dim3 grid((width + (2 * block.x - 1)) / (2 * block.x), (height + (block.y - 1)) / block.y, 1);
-		CUDAToBGR_drvapi <<< grid, block >>>((uint32 *)dataY, (uint32 *)dataUV, pitchY, pitchUV, d_dstRGB, width, height);
+		CUDAToBGR_drvapi << < grid, block >> >((uint32 *)dataY, (uint32 *)dataUV, pitchY, pitchUV, d_dstRGB, width, height);
 		cudaError_t cudaStatus = cudaGetLastError();
 		if (cudaStatus != cudaSuccess) {
 			fprintf(stderr, "NV12ToRGB_drvapi launch failed: %s\n", cudaGetErrorString(cudaStatus));
-#include"cudaHeader.h"
-#include<iostream>
-
-#include "cuda_runtime.h"
-#include <cuda.h>
-
-//核函数，计算a+b
-__global__ void add(int a,int b,int *c)
-{
-    //保存a+b的计算结果
-    *c=a+b;
-}
-
-
-//cuda测试函数的实现
-void cudaTest()
-{
-    int c = 0;
-    //在gpu上开辟一个相同的内存
-    int *deviceC;
-    cudaMalloc((void**)&deviceC,sizeof(int));
-    //调用核函数
-    add<<<1,1>>>(3,7,deviceC);
-    //把计算结果复制到cpu上
-    cudaMemcpy(&c,deviceC,sizeof(int),cudaMemcpyDeviceToHost);
-    //展示计算结果
-    std::cout<<c<<std::endl;
-    //释放内存
-    cudaFree(deviceC);
-}
 \ No newline at end of file
-#ifndef _CUDA_HEADER_H_
-#define _CUDA_HEADER_H_
-//用于测试cuda能否正常工作的函数
-void cudaTest();
-#endif
 \ No newline at end of file
@@ -7,12 +7,13 @@
 #include "check_tool.h"
-#include "cudaHeader.h"
-
 #include "cuda_kernels.h"
 #include "NvJpegEncoder.h"
+#include "opencv2\opencv.hpp"
+
 using namespace std;
+using namespace cv;
 unsigned char *pHwRgb[2] = {nullptr, nullptr};
@@ -44,6 +45,50 @@ void saveFrame(AVFrame * gpuFrame, string file_name) {
 	pHwData = nullptr;
 }
+mutex m_mutex_show;
+void showFrame(AVFrame * gpuFrame) {
+	std::lock_guard<std::mutex> l(m_mutex_show);
+
+	unsigned char *pHwData = nullptr;
+	cudaError_t cudaStatus = cudaMalloc((void **)&pHwData, 3 * gpuFrame->width * gpuFrame->height * sizeof(unsigned char));
+
+	cuda_common::setColorSpace(ITU709, 0);
+	cudaStatus = cuda_common::CUDAToBGR((CUdeviceptr)gpuFrame->data[0], (CUdeviceptr)gpuFrame->data[1], gpuFrame->linesize[0], gpuFrame->linesize[1], pHwData, gpuFrame->width, gpuFrame->height);
+	cudaDeviceSynchronize();
+	if (cudaStatus != cudaSuccess) {
+		cout << "CUDAToBGR failed !!!" << endl;
+		return;
+	}
+
+
+
+	unsigned char * pHwRgb = pHwData;
+	int channel = 3;
+	int width = gpuFrame->width;
+	int height = gpuFrame->height;
+
+	if (pHwRgb != nullptr && channel > 0 && width > 0 && height > 0) {
+		int nSize = channel * height * width;
+		unsigned char* cpu_data = new unsigned char[nSize];
+
+		cudaMemcpy(cpu_data, pHwRgb, nSize * sizeof(unsigned char), cudaMemcpyDeviceToHost);
+		cudaDeviceSynchronize();
+
+		cv::Mat img_(height, width, CV_8UC3, cpu_data);
+		bool bWrite = cv::imwrite("dec0.jpg", img_);
+
+		imshow("show", img_);
+		waitKey(0);
+
+		delete[] cpu_data;
+		cpu_data = nullptr;
+
+	}
+
+	cudaFree(pHwData);
+	pHwData = nullptr;
+}
+
 /**
  * 注意： gpuFrame 在解码器设置的显卡上，后续操作要十分注意这一点，尤其是多线程情况
  * */
@@ -64,6 +109,7 @@ void postDecoded(const void * userPtr, AVFrame * gpuFrame){
 			cudaSetDevice(atoi(decoder->m_cfg.gpuid.c_str()));
 			saveFrame(gpuFrame, decoder->getName());
+			showFrame(gpuFrame);
 		}
     }
 }
@@ -82,7 +128,7 @@ static long long get_cur_time(){
     return tpMicro.time_since_epoch().count();
 }
-static int sum = 0;
+static int suming = 0;
 unsigned char *pHwData = nullptr;
 void postDecoded0(const void * userPtr, AVFrame * gpuFrame){
@@ -101,7 +147,7 @@ void postDecoded0(const void * userPtr, AVFrame * gpuFrame){
                 end_time = start_time = get_cur_time();
             }
             count_num++;
-            sum ++ ;
+            suming ++ ;
             if (count_num >= count_std)
             {
                 // end_time = get_cur_time();
@@ -112,7 +158,7 @@ void postDecoded0(const void * userPtr, AVFrame * gpuFrame){
                 count_flag = false;
             }
-            cout << "帧数：" << sum << endl;
+            cout << "帧数：" << suming << endl;
         }
     }
 }
@@ -163,8 +209,6 @@ void logFF(void *, int level, const char *fmt, va_list ap)
 int main(int argc, char* argv[]) {
-	cudaTest();
-
     printf("start \n");
     if (argc != 3) {
         fprintf(stderr, "./xxx uri gpu_id\n");