addKernel launch failed: no kernel image is available for execution on the device addWithCuda failed

visualstudio2019環境でCUDAプログラミングを行おうとしているのですが、プロジェクト開始時に自動生成されるkernel.cuファイルの実行をしようとするとタイトルのようなエラーとなってしまいます。使用しているCUDAのバージョンは11.0で、GPUはGTX680です。x64Nativetoolsコマンドプロンプトなどでnvccやclの確認はすでに通っています。そちらのほうでコンパイルして実行しても、CPU部分はうごいても、GPUの部分が動いていないようなのです。デバイスマネージャーなどで確認するとGPU自体は正常に作動しているようなのでなぜなのかがわからないです。よろしくお願いします。

cu
1#include "cuda_runtime.h"
2#include "device_launch_parameters.h"
3#include "cuda_runtime.h"
4#include "cuda_runtime_api.h"
5#include <stdio.h>
6
7cudaError_t addWithCuda(int *c, const int *a, const int *b, unsigned int size);
8
9__global__ void addKernel(int *c, const int *a, const int *b)
10{
11    int i = threadIdx.x;
12    c[i] = a[i] + b[i];
13}
14
15int main()
16{
17    const int arraySize = 5;
18    const int a[arraySize] = { 1, 2, 3, 4, 5 };
19    const int b[arraySize] = { 10, 20, 30, 40, 50 };
20    int c[arraySize] = { 0 };
21
22    // Add vectors in parallel.
23    cudaError_t cudaStatus = addWithCuda(c, a, b, arraySize);
24    if (cudaStatus != cudaSuccess) {
25        fprintf(stderr, "addWithCuda failed!");
26        return 1;
27    }
28
29    printf("{1,2,3,4,5} + {10,20,30,40,50} = {%d,%d,%d,%d,%d}\n",
30        c[0], c[1], c[2], c[3], c[4]);
31
32    // cudaDeviceReset must be called before exiting in order for profiling and
33    // tracing tools such as Nsight and Visual Profiler to show complete traces.
34    cudaStatus = cudaDeviceReset();
35    if (cudaStatus != cudaSuccess) {
36        fprintf(stderr, "cudaDeviceReset failed!");
37        return 1;
38    }
39
40    return 0;
41}
42
43// Helper function for using CUDA to add vectors in parallel.
44cudaError_t addWithCuda(int *c, const int *a, const int *b, unsigned int size)
45{
46    int *dev_a = 0;
47    int *dev_b = 0;
48    int *dev_c = 0;
49    cudaError_t cudaStatus;
50
51    // Choose which GPU to run on, change this on a multi-GPU system.
52    cudaStatus = cudaSetDevice(0);
53    if (cudaStatus != cudaSuccess) {
54        fprintf(stderr, "cudaSetDevice failed!  Do you have a CUDA-capable GPU installed?");
55        goto Error;
56    }
57
58    // Allocate GPU buffers for three vectors (two input, one output)    .
59    cudaStatus = cudaMalloc((void**)&dev_c, size * sizeof(int));
60    if (cudaStatus != cudaSuccess) {
61        fprintf(stderr, "cudaMalloc failed!");
62        goto Error;
63    }
64
65    cudaStatus = cudaMalloc((void**)&dev_a, size * sizeof(int));
66    if (cudaStatus != cudaSuccess) {
67        fprintf(stderr, "cudaMalloc failed!");
68        goto Error;
69    }
70
71    cudaStatus = cudaMalloc((void**)&dev_b, size * sizeof(int));
72    if (cudaStatus != cudaSuccess) {
73        fprintf(stderr, "cudaMalloc failed!");
74        goto Error;
75    }
76
77    // Copy input vectors from host memory to GPU buffers.
78    cudaStatus = cudaMemcpy(dev_a, a, size * sizeof(int), cudaMemcpyHostToDevice);
79    if (cudaStatus != cudaSuccess) {
80        fprintf(stderr, "cudaMemcpy failed!");
81        goto Error;
82    }
83
84    cudaStatus = cudaMemcpy(dev_b, b, size * sizeof(int), cudaMemcpyHostToDevice);
85    if (cudaStatus != cudaSuccess) {
86        fprintf(stderr, "cudaMemcpy failed!");
87        goto Error;
88    }
89
90    // Launch a kernel on the GPU with one thread for each element.
91    addKernel<<<1, size>>>(dev_c, dev_a, dev_b);
92
93    // Check for any errors launching the kernel
94    cudaStatus = cudaGetLastError();
95    if (cudaStatus != cudaSuccess) {
96        fprintf(stderr, "addKernel launch failed: %s\n", cudaGetErrorString(cudaStatus));
97        goto Error;
98    }
99    
100    // cudaDeviceSynchronize waits for the kernel to finish, and returns
101    // any errors encountered during the launch.
102    cudaStatus = cudaDeviceSynchronize();
103    if (cudaStatus != cudaSuccess) {
104        fprintf(stderr, "cudaDeviceSynchronize returned error code %d after launching addKernel!\n", cudaStatus);
105        goto Error;
106    }
107
108    // Copy output vector from GPU buffer to host memory.
109    cudaStatus = cudaMemcpy(c, dev_c, size * sizeof(int), cudaMemcpyDeviceToHost);
110    if (cudaStatus != cudaSuccess) {
111        fprintf(stderr, "cudaMemcpy failed!");
112        goto Error;
113    }
114
115Error:
116    cudaFree(dev_c);
117    cudaFree(dev_a);
118    cudaFree(dev_b);
119    
120    return cudaStatus;
121}