Developing C Programs in Visual Studio

A Simple C Project

1. Open Visual Studio

   

2. Choose Create a new project

   

3. Choose Blank Solution and click Next

   

4. Name the solution ctutorial and click Create

   

5. Right-click the solution node in Solution Explorer and choose Add -> New Project

   

6. Choose project type Console App C++ and click Next

   

7. Name the project capp and click Create

   

8. Right-click the file capp.cpp in Solution Explorer and choose Remove

   

9. Right-click the Source Files folder and choose Add -> New Item

   

10. Choose file type C++ File (cpp) , name it main.c, and click Add

    • Note! Visual Studio knows this is a C source file due to the .c file extension
    • If we had kept the .cpp file extension, it would be treated as a C++ file.
    • Don’t forget to change this from .cpp to .c for a C file!

    

11. Add the code below to main.c.

    #include <stdio.h>
    
    int main(int argc, char* argv[])
    {
        // Print out command line arguments
        printf("Number of command line arguments: %d\n", argc);
        for (int i = 0; i < argc; ++i)
        {
            printf("Command line argument %d: %s\n", i, argv[i]);
        }
        puts("");
    
        // Return 0 (program executed successfully) to operating system
        return 0;
    }

    

12. Compile (build) and run (debug) the program with F5.

    • Notice the text written to the terminal window.
      • It shows the number of command line arguments sent to the application.
      • It also shows each command line argument (currently only one).
        • The first command line argument is the full path to the executable.
    • Then Press any key to close this window . . .

    

13. Right-click the project node in Solution Explorer and choose Properties

    

14. In Configuration Properties notice:

    • Under General:
      • Output Directory: $(SolutionDir)$(Platform)\$(Configuration)\
        • E.g. C:\Users\patrick\source\repos\ctutorial\x64\Debug\
          • $(SolutionDir) is C:\Users\patrick\source\repos\ctutorial\
          • $(Platform) is x64
          • $(Configuration) is Debug
        • If you right-click the solution node in Solution Explorer and choose Open Folder in File Explorer, it will take you to $(SolutionDir)
      • Configuration Type: Application (.exe)
        • The project compiles to an executable file (.exe)
      • C Language Standard: Default (Legacy MSVC)
        • You can choose a C Language Standard version here
    • Under Debugging:
      • Command Arguments
        • This is where we can pass command line arguments duing debugging
        • Enter: One Two Three
    • Under C/C++ -> General
      • Additional Include Directories
        • This is where we can enter directory path(s) to additional header files (.h)
    • Under C/C++ -> Preprocessor
      • Preprocessor Definitions
        • This is where we can set additional #defines for the preprocessor
    • Under C/C++ -> Precompiled Headers
      • Precompiled Header
        • For all projects we want this to be Not Using Precompiled Headers
    • Under C/C++ -> Command Line
      • Additional Options
        • This is where we can add additional flags for the cl.exe compiler
    • Under Linker -> General
      • Additional Library Directories
        • This is where we can enter directory path(s) to additional library files (.lib)
    • Under Linker -> Input
      • Additional Dependencies
        • This is where we can enter names of additional libraries files, e.g. OpenCL.lib

15. Click Apply and then OK to cloae the Project Properties window.

16. Set a breakpoint on the row with the second printf() statement and debug again

    • Notice program execution stops at the breakpoint.
    • You can view local variables, etc. in the window below the code editor.
      • Notice the variable argc is set to 4 since we added 3 command line arguments in the Project Properties (the full path to the executable is always the first argument, hence 1+3=4)

    

17. Let the program run to completion.

    • Notice the text written to the terminal window.
      • It shows the number of command line arguments sent to the application.
      • It also shows each command line argument (currently four).
        • The first command line argument is the full path to the executable.
        • The other three are the ones we set in the Project Properties
    • Finally Press any key to close this window . . .

    

18. Show all Files

    • Click the project node in Solution Explorer
    • Then click the Show All Files icon to the left of the spanner icon
    • Notice the project file structure shown in Solution Explorer
      • This is the actual file/folder structrue in the project directory
    • Click the Show All Files icon again
      • This is the default view that uses filters to sort file types into virtual folders such as Source Files and Header FIles

19. Right-click the solution node in Solution Explorer and choose Open Folder in File Explorer

    

20. In File Explorer, notice the file/folder structure

    • ctutorial is the solution folder that contains:
      • .vs which is a subfolder that contains config settings
      • capp which is the project subfolder which contains:
        • capp.vcxproj which is the capp Project file
        • main.c which is the capp project’s only source file
      • x64 which is the target platform (64 bit) that contains:
        • Debug which is the build configuration (debug) that contains:
          • capp.exe which is the capp project executable
          • capp.pdb which contains the capp project debug symbols
            
      • ctutorial.sln which is the Solution file

Add a Module to the C Project

21. Add a simplemath module, starting with the header file.
    • Right-click Header Files and choose Add -> New Item.

22. Choose file type Header File (.h) , name it simplemath.h, and click Add

    

23. Add the code below to simplemath.h.

    #pragma once
    
    double hypotenuse(double a, double b);

    Note that #pragma once is equivalent to an include guard around the code, i.e.:

    #ifndef SIMPLEMATH_H
    #define SIMPLEMATH_H
    
    double hypotenuse(double a, double b);
    
    #endif // SIMPLEMATH_H

    

24. Add the simplemath module’s implementation file.

    • Right-click Source Files and choose Add -> New Item.

25. Choose file type C++ File (cpp) , name it simplemath.c, and click Add

    • Remember to change the file extension from .cpp to .c!

    

26. Add the code below to simplemath.c.

    #include <math.h>
    #include "simplemath.h"
    
    double hypotenuse(double a, double b)
    {
        return sqrt(a * a + b * b);
    }

    

27. Use the simplemath module in main.c.

    • Open main.c and replace its contents with the code below.

    #include <stdio.h>
    #include "simplemath.h"
    
    int main(int argc, char* argv[])
    {
        // Print out command line arguments
        printf("Number of command line arguments: %d\n", argc);
        for (int i = 0; i < argc; ++i)
        {
            printf("Command line argument %d: %s\n", i, argv[i]);
        }
        puts("");
    
        // Print out the hypotenuse of the sides in a right triangle
        double a = 3.0;
        double b = 4.0;
        printf("hypotenuse(%.2f, %.2f) = %.2f.\n", a, b, hypotenuse(a, b));
    
        // Return 0 (program executed successfully) to operating system
        return 0;
    }

    

28. Compile (build) and run (debug) the program with F5.

    • Notice the text hypotenuse(3.00, 4.00) = 5.00 written to the terminal window.
    • Then Press any key to close this window . . .

    

29. Notice the final structure of our C application:

    • Header files (.h) are in the Header Files folder.
      • Each header file (.h) has an include guard, e.g. #pragma once.
    • Source code/implementation files (.c) are in the Source Files folder.
    • Only one source code file (.c) contains a main() function.
      • This source code file doesn’t have a corresponding header file (.h).
    • Modules have at least one source code file (.c) and a one header file (.h).
      • For example, the simplemath module has simplemath.c and simplemath.h.
      • The module’s source code file (.c) #includes the module’s header file (.h).
      • main.c #includes the module’s header file (.h), if it wants to use the module.

Add a Static Library Project to the Solution

30. Right-click the solution node in Solution Explorer and choose Add -> New Project

    

31. Choose project type Static Library and click Next

    

32. Name the project cstatic and click Create

    

33. Select the files framework.h, pch.h, cstatic.cpp, and pch.cpp, then right-click the selection and choose Remove to delete them.

    

34. Add the file simplemath2.h with the contents below.

    • Note! Do not copy-and-paste the simplemath.h file from the capp project via Solution Explorer, since it will only copy a reference to the file (use Windows File Explorer for that).

    #pragma once
    
    double hypotenuse2(double a, double b);

    

35. Add the file simplemath2.c with the contents below.

    • Note! Do not copy-and-paste the simplemath.c file from the capp project via Solution Explorer, since it will only copy a reference to the file (use Windows File Explorer for that).

    #include <math.h>
    #include "simplemath2.h"
    
    double hypotenuse2(double a, double b)
    {
        return sqrt(a * a + b * b);
    }

    

36. Right-click the cstatic project node in Solution Explorer and choose Properties

    • Under Configuration Properties -> General notice:
      • Output Directory has the same setting as the capp project
      • Configuration Type is Static Library (.lib)

    

    • Under Configuration Properties -> C/C++ -> Precompiled Headers set:
      • Precompiled Header to Not Using Precompiled Headers

    

37. Click the Apply button and then the OK button to close the project properties.

38. Build the solution with Ctrl + Shift + B.

39. Right-click the solution node in Solution Explorer and choose Open Folder in File Explorer which will open the solution folder ctutorial, then:

    • Navigate to the subfolder ctutorial\cstatic and notice it contains the file simplemath2.h (and the file simplemath2.c).
    • Navigate to the subfolder ctutorial\x64\Debug and notice it contains the compiled static library file cstatic.lib (and the file capp.exe).

Use the Static Library from the capp Project

40. The capp program needs to know where the files simplemath2.h and cstatic.lib are, so right-click the capp project node in Solution Explorer and choose Properties.

    • Under Configuration Properties -> C/C++ -> General:
      • To the right of Additional Include Directories, click the drop-down combobox.
      • Then click <Edit...>.
      • Click the folder icon in the pop-up window.
      • Then, add the path to the folder containing simplemath2.h
      • Finally, click the OK button to close the pop-up window.
    • Under Configuration Properties -> Linker -> General:
      • To the right of Additional Library Directories, click the drop-down combobox.
      • Then click <Edit...>.
      • Click the folder icon in the pop-up window.
      • Then, add the path to the folder containing cstatic.lib
      • Finally, click the OK button to close the pop-up window.
    • Under Configuration Properties -> Linker -> Input:
      • To the right of Additional Dependencies, click the drop-down combobox.
      • Then click <Edit...>.
      • In the pop-up window, enter the name of the static library cstatic.lib.
      • Click the OK button to close the pop-up window.
    • Finally, click the Apply button and then the OK button to close the project properties.

41. Open the file main.c and replace its contents with the code below.

    • We have only added one #include and one printf().

    #include <stdio.h>
    #include "simplemath.h"
    #include "simplemath2.h"
    
    int main(int argc, char* argv[])
    {
        // Print out command line arguments
        printf("Number of command line arguments: %d\n", argc);
        for (int i = 0; i < argc; ++i)
        {
            printf("Command line argument %d: %s\n", i, argv[i]);
        }
        puts("");
    
        // Print out the hypotenuse of the sides in a right triangle
        double a = 3.0;
        double b = 4.0;
        printf("hypotenuse(%.2f, %.2f) = %.2f.\n", a, b, hypotenuse(a, b));
        printf("hypotenuse2(%.2f, %.2f) = %.2f.\n", a, b, hypotenuse2(a, b));
    
        // Return 0 (program executed successfully) to operating system
        return 0;
    }

    

42. Compile (build) and run (debug) the program with F5.

    • Notice the text hypotenuse2(3.00, 4.00) = 5.00 written to the terminal window.
    • Then Press any key to close this window . . .

    

Add a Dynamic Library Project to the Solution

43. Right-click the solution node in Solution Explorer and choose Add -> New Project

    

44. Choose project type Dynamic-Link Library (DLL) and click Next

    

45. Name the project cdynamic and click Create

    

46. Select the files framework.h, pch.h, dllmain.cpp, and pch.cpp, then right-click the selection and choose Remove to delete them.

    

47. Add the file simplemath3.h with the contents below.

    • Note! Do not copy-and-paste the simplemath.h file from the capp project via Solution Explorer, since it will only copy a reference to the file (use Windows File Explorer for that).

    # pragma once
    
    #ifdef MYDLL_EXPORTS
    #define MYDLL_API __declspec(dllexport)
    #else
    #define MYDLL_API __declspec(dllimport)
    #endif
    
    MYDLL_API double hypotenuse3(double a, double b);

    

48. Add the file simplemath3.c with the contents below.

    • Note! Do not copy-and-paste the simplemath.c file from the capp project via Solution Explorer, since it will only copy a reference to the file (use Windows File Explorer for that).

    #include <math.h>
    #include "simplemath3.h"
    
    double hypotenuse3(double a, double b)
    {
      return sqrt(a * a + b * b);
    }

    

49. Right-click the cdynamic project node in Solution Explorer and choose Properties

    • Under Configuration Properties -> General notice:
      • Output Directory has the same setting as the capp project
      • Configuration Type is Dynamic Library (.dll)

    

    • Under Configuration Properties -> C/C++ -> Preprocessor:
      • To the right of Preprocessor Definitions, click the drop-down combobox.
      • Then click <Edit...>.
      • In the pop-up window, add the entry MYDLL_EXPORTS.
      • Finally, click the OK button to close the pop-up window.
    • Under Configuration Properties -> C/C++ -> Precompiled Headers set:
      • Precompiled Header to Not Using Precompiled Headers
    • Click the Apply button and then the OK button to close the project properties.

50. Build the solution with Ctrl + Shift + B.

51. Right-click the solution node in Solution Explorer and choose Open Folder in File Explorer which will open the solution folder ctutorial, then:

    • Navigate to the subfolder ctutorial\cdynamic and notice it contains the file simplemath3.h (and the file simplemath3.c).
    • Navigate to the subfolder ctutorial\x64\Debug and notice it contains the compiled dynamic library file cdynamic.dll together with its import library cdynamic.lib (and the file capp.exe).

Use the Dynamic Library from the capp Project

52. The capp program needs to know where the files simplemath3.h and cdynamic.lib are (cdynamic.dll is in the same folder as capp.exe which is where it should be), so right-click the capp project node in Solution Explorer and choose Properties.

    • Under Configuration Properties -> C/C++ -> General:
      • To the right of Additional Include Directories, click the drop-down combobox.
      • Then click <Edit...>.
      • Click the folder icon in the pop-up window.
      • Then, add the path to the folder containing simplemath3.h
      • Finally, click the OK button to close the pop-up window.
    • Under Configuration Properties -> Linker -> General:
      • Notice Additional Library Directories already has an entry to the correct directory (we added this when we used the static library).
    • Under Configuration Properties -> Linker -> Input:
      • To the right of Additional Dependencies, click the drop-down combobox.
      • Then click <Edit...>.
      • In the pop-up window, enter the name of the import library cdynamic.lib.
      • Click the OK button to close the pop-up window.
    • Finally, click the Apply button and then the OK button to close the project properties.

53. Open the file main.c and replace its contents with the code below.

    • We have only added one #include and one printf().

    #include <stdio.h>
    #include "simplemath.h"
    #include "simplemath2.h"
    #include "simplemath3.h"
    
    int main(int argc, char* argv[])
    {
        // Print out command line arguments
        printf("Number of command line arguments: %d\n", argc);
        for (int i = 0; i < argc; ++i)
        {
            printf("Command line argument %d: %s\n", i, argv[i]);
        }
        puts("");
    
        // Print out the hypotenuse of the sides in a right triangle
        double a = 3.0;
        double b = 4.0;
        printf("hypotenuse(%.2f, %.2f) = %.2f.\n", a, b, hypotenuse(a, b));
        printf("hypotenuse2(%.2f, %.2f) = %.2f.\n", a, b, hypotenuse2(a, b));
        printf("hypotenuse3(%.2f, %.2f) = %.2f.\n", a, b, hypotenuse3(a, b));
    
        // Return 0 (program executed successfully) to operating system
        return 0;
    }

    

54. Compile (build) and run (debug) the program with F5.

    • Notice the text hypotenuse3(3.00, 4.00) = 5.00 written to the terminal window.
    • Then Press any key to close this window . . .

    

Vector Addition in C

55. From the main menu in Visual Studio, choose File -> New Project

    

56. Choose project type Console App C++ and click Next

    

57. Name the project cvec and click Create

58. Right-click the file cvec.cpp in Solution Explorer and choose Remove

59. Right-click the Source Files folder and choose Add -> New Item

60. Choose file type C++ File (cpp) , name it main.c, and click Add

    • Note! Visual Studio knows this is a C source file due to the .c file extension
    • If we had kept the .cpp file extension, it would be treated as a C++ file.
    • Don’t forget to change this from .cpp to .c for a C file!

    

61. Add the code below to main.c.

    #include <stdio.h>
    #include <stdlib.h>
    
    #define N 1024
    
    void vec_add(const float* A, const float* B, float* C)
    {
        for (int id = 0; id < N; ++id)
        {
            C[id] = A[id] + B[id];
        }
    }
    
    int main(void)
    {
        float *h_A, *h_B, *h_C;
    
        size_t size = N * sizeof(float);
    
        // Allocate host memory
        h_A = (float*)malloc(size);
        h_B = (float*)malloc(size);
        h_C = (float*)malloc(size);
    
        // Initialize host arrays
        for (int i = 0; i < N; ++i) {
            h_A[i] = (float)i;
            h_B[i] = (float)i * 2;
            h_C[i] = (float)0;
        }
    
        // Launch calculation
        vec_add(h_A, h_B, h_C);
    
        // Display results
        for (int i = 0; i < 10; ++i)
            printf("%f + %f = %f\n", h_A[i], h_B[i], h_C[i]);
    
        // Cleanup
        free(h_A); free(h_B); free(h_C);
    
        return 0;
    }

    

62. Compile (build) and run (debug) the program with F5.

    • Notice the text written to the terminal window.
      • It shows the first 10 vector additions (of 1024 in total).
    • Then Press any key to close this window . . .

    

Vector Addition in OpenCL

63. From the main menu in Visual Studio, choose File -> New Project

    

64. Choose project type Console App C++ and click Next

    

65. Name the project openclvec and click Create

66. Right-click the file openclvec.cpp in Solution Explorer and choose Remove

67. Right-click the Source Files folder and choose Add -> New Item

68. Choose file type C++ File (cpp) , name it main.c, and click Add

    • Note! Visual Studio knows this is a C source file due to the .c file extension
    • If we had kept the .cpp file extension, it would be treated as a C++ file.
    • Don’t forget to change this from .cpp to .c for a C file!

    

69. Add the code below to main.c.

    #include <CL/cl.h>
    #include <stdio.h>
    #include <stdlib.h>
    
    #define N 1024
    #define CL_TARGET_OPENCL_VERSION 300
    #define KERNEL_FILE "kernel.cl"
    
    char* read_kernel_source(const char* filename)
    {
        FILE* fp = fopen(filename, "r");
        if (!fp) {
            fprintf(stderr, "Failed to open kernel file: %s\n", filename);
            exit(1);
        }
        fseek(fp, 0, SEEK_END);
        long size = ftell(fp);
        rewind(fp);
        char* src = (char*)malloc(size + 1);
        fread(src, 1, size, fp);
        src[size] = '\0';
        fclose(fp);
        return src;
    }
    
    int main()
    {
        float A[N], B[N], C[N];
        for (int i = 0; i < N; i++) {
            A[i] = (float)i;
            B[i] = (float)i * 2;
        }
    
        // OpenCL setup
        cl_platform_id platform;
        cl_device_id device;
        cl_context context;
        cl_command_queue queue;
        cl_program program;
        cl_kernel kernel;
    
        cl_mem bufA, bufB, bufC;
        cl_int err;
    
        err = clGetPlatformIDs(1, &platform, NULL);
        err |= clGetDeviceIDs(platform, CL_DEVICE_TYPE_DEFAULT, 1, &device, NULL);
        context = clCreateContext(NULL, 1, &device, NULL, NULL, &err);
        queue = clCreateCommandQueueWithProperties(context, device, 0, &err);
    
        // Read kernel source
        char* source = read_kernel_source(KERNEL_FILE);
    
        // Create program and build
        program = clCreateProgramWithSource(context, 1, (const char**)&source, NULL, &err);
        err = clBuildProgram(program, 1, &device, NULL, NULL, NULL);
        if (err != CL_SUCCESS) {
            // Show build log on error
            char log[2048];
            clGetProgramBuildInfo(program, device, CL_PROGRAM_BUILD_LOG, sizeof(log), log, NULL);
            printf("Build error:\n%s\n", log);
            exit(1);
        }
    
        kernel = clCreateKernel(program, "vec_add", &err);
    
        // Buffers
        bufA = clCreateBuffer(context, CL_MEM_READ_ONLY, sizeof(float) * N, NULL, NULL);
        bufB = clCreateBuffer(context, CL_MEM_READ_ONLY, sizeof(float) * N, NULL, NULL);
        bufC = clCreateBuffer(context, CL_MEM_WRITE_ONLY, sizeof(float) * N, NULL, NULL);
    
        clEnqueueWriteBuffer(queue, bufA, CL_TRUE, 0, sizeof(float) * N, A, 0, NULL, NULL);
        clEnqueueWriteBuffer(queue, bufB, CL_TRUE, 0, sizeof(float) * N, B, 0, NULL, NULL);
    
        clSetKernelArg(kernel, 0, sizeof(cl_mem), &bufA);
        clSetKernelArg(kernel, 1, sizeof(cl_mem), &bufB);
        clSetKernelArg(kernel, 2, sizeof(cl_mem), &bufC);
    
        size_t global_size = N;
        clEnqueueNDRangeKernel(queue, kernel, 1, NULL, &global_size, NULL, 0, NULL, NULL);
        clEnqueueReadBuffer(queue, bufC, CL_TRUE, 0, sizeof(float) * N, C, 0, NULL, NULL);
    
        for (int i = 0; i < 10; i++)
            printf("%f + %f = %f\n", A[i], B[i], C[i]);
    
        free(source);
        clReleaseMemObject(bufA);
        clReleaseMemObject(bufB);
        clReleaseMemObject(bufC);
        clReleaseProgram(program);
        clReleaseKernel(kernel);
        clReleaseCommandQueue(queue);
        clReleaseContext(context);
    
        return 0;
    }

    

70. Right-click the Source Files folder and choose Add -> New Item

71. Choose file type C++ File (cpp) , name it kernel.cl, and click Add

    • Note! Make sure the file extension is .cl.

    

72. Add the code below to kernel.cl.

    __kernel void vec_add(__global const float* A, __global const float* B, __global float* C)
    {
        int id = get_global_id(0);
        C[id] = A[id] + B[id];
    }

    

73. Compile (build) and run (debug) the program with F5.

    • Notice the text written to the terminal window.
      • It shows the first 10 vector additions (of 1024 in total).
    • Then Press any key to close this window . . .

74. In the previous step, if you get a compilation error is not valid UTF-8:

    • Open the file kernel.cl in Visual Studio Code (VSCode).
    • In the lower right of VSCode, click CRLF.
    • In the Command Palette (top of VSCode), choose LF.
    • Save the file with Ctrl + S.
    • Now compile and run the Visual Studio Project again.

75. If you get preprocessor or linker errors:

    • You probably didn’t integrate VCVPK with Visual Studio during your environment setup.
    • Right-click the project node in Solution Explorer and choose Properties.
    • Under Configuration Properties -> C/C++ -> General:
      • Additional Include Directories: Add path to OpenCL include directory which might be akin to:
        • %LOCALAPPDATA%\Programs\vcpkg\installed\x64-windows\include
        • C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\include
    • Under Configuration Properties -> Linker -> General:
      • Additional Library Directories: Add path to OpenCL library directory which might be akin to:
        • %LOCALAPPDATA%\Programs\vcpkg\installed\x64-windows\lib
        • C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\lib\x64
    • Under Configuration Properties -> Linker -> Input:
      • Additional Dependencies: Add OpenCL.lib
    • Now compile and run the Visual Studio Project again.

Vector Addition in CUDA

76. From the main menu in Visual Studio, choose File -> New Project

    

77. Choose project type CUDA xx.x Runtime and click Next

    • Replace xx.x with your installed version of the CUDA Toolkit.

    

78. Name the project cudavec and click Create

79. Replace the contents in the file kernel.cu with the code below:

    #include "cuda_runtime.h"
    #include "device_launch_parameters.h"
    #include <stdio.h>
    #include <stdlib.h>
    
    #define N 1024
    
    __global__ void vec_add(const float* A, const float* B, float* C)
    {
        int id = threadIdx.x + blockDim.x * blockIdx.x;
        if (id < N) {
            C[id] = A[id] + B[id];
        }
    }
    
    int main(void)
    {
        float* h_A, * h_B, * h_C;
        float* d_A, * d_B, * d_C;
    
        size_t size = N * sizeof(float);
    
        // Allocate host memory
        h_A = (float*)malloc(size);
        h_B = (float*)malloc(size);
        h_C = (float*)malloc(size);
    
        // Initialize host arrays
        for (int i = 0; i < N; ++i) {
            h_A[i] = (float)i;
            h_B[i] = (float)i * 2;
        }
    
        // Allocate device memory
        cudaMalloc((void**)&d_A, size);
        cudaMalloc((void**)&d_B, size);
        cudaMalloc((void**)&d_C, size);
    
        // Copy data to device
        cudaMemcpy(d_A, h_A, size, cudaMemcpyHostToDevice);
        cudaMemcpy(d_B, h_B, size, cudaMemcpyHostToDevice);
    
        // Launch kernel
        int threadsPerBlock = 256;
        int blocksPerGrid = (N + threadsPerBlock - 1) / threadsPerBlock;
    
        vec_add<<<blocksPerGrid, threadsPerBlock>>>(d_A, d_B, d_C);
    
        // Copy result back to host
        cudaMemcpy(h_C, d_C, size, cudaMemcpyDeviceToHost);
    
        // Display results
        for (int i = 0; i < 10; ++i)
            printf("%f + %f = %f\n", h_A[i], h_B[i], h_C[i]);
    
        // Cleanup
        free(h_A); free(h_B); free(h_C);
        cudaFree(d_A); cudaFree(d_B); cudaFree(d_C);
    
        return 0;
    }

    

80. Compile (build) and run (debug) the program with F5.

    • Notice the text written to the terminal window.
      • It shows the first 10 vector additions (of 1024 in total).
    • Then Press any key to close this window . . .

81. In the previous step, if you get preprocessor or linker errors:

    • You’re CUDA Toolkit installation was probably not successful.
    • Right-click the project node in Solution Explorer and choose Properties.
    • Under Configuration Properties -> C/C++ -> General:
      • Additional Include Directories: Add path to CUDA include directory which might be akin to:
        • C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\include
    • Under Configuration Properties -> Linker -> General:
      • Additional Library Directories: Add path to CUDA library directory which might be akin to:
        • C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\lib\x64
    • Under Configuration Properties -> Linker -> Input:
      • Additional Dependencies: Add cudart.lib or cudart_static.lib
    • Now compile and run the Visual Studio Project again.