Unable to debug kernel code with cuda-gdb

I’m trying to get VSCode + WSL setup as my dev environment so that I can use cuda-gdb for debugging kernels. I’ve managed to get everything installed and I can build and run Cuda applications, but I can only debug CPU code using cuda-gdb. This isn’t just limited to VSCode either. If I run my app through cuda-gdb on the command line and add a break point inside the kernel code it will set the breakpoint on the closing bracket of the kernel instead of the line I picked

    "version": "2.0.0",
    "tasks": [
            "type": "cmake",
            "label": "CMake: configure",
            "command": "configure",
            "problemMatcher": [],
            "detail": "CMake template configure task"
            "type": "shell",
            "label": "Build",
            "command": "make dbg=1",
            "group": {
                "kind": "build",
                "isDefault": true,
            "problemMatcher": ["$nvcc"],
            "dependsOn": ["CMake: configure"],
            "options": {
                "cwd": "${workspaceFolder}/build"

    "configurations": [
            "name": "CUDA C++: Launch",
            "type": "cuda-gdb",
            "request": "launch",
            "program": "${workspaceFolder}/build/ISP_AMF",

The output of cuda-gdb when used via the command line:

(cuda-gdb) break main.cu:6
Breakpoint 1 at 0xb0a7: file /mnt/c/Dev/ISP-AMF/main.cu, line 8.

The default saxpy kernel code

#include <stdio.h>

void saxpy(int n, float a, float *x, float *y)
  int i = blockIdx.x*blockDim.x + threadIdx.x;
  if (i < n) y[i] = a*x[i] + y[i];

int main(void)
  int N = 1<<20;
  float *x, *y, *d_x, *d_y;
  x = (float*)malloc(N*sizeof(float));
  y = (float*)malloc(N*sizeof(float));

  cudaMalloc(&d_x, N*sizeof(float)); 
  cudaMalloc(&d_y, N*sizeof(float));

  for (int i = 0; i < N; i++) {
    x[i] = 1.0f;
    y[i] = 2.0f;

  cudaMemcpy(d_x, x, N*sizeof(float), cudaMemcpyHostToDevice);
  cudaMemcpy(d_y, y, N*sizeof(float), cudaMemcpyHostToDevice);

  // Perform SAXPY on 1M elements
  saxpy<<<(N+255)/256, 256>>>(N, 2.0f, d_x, d_y);

  cudaMemcpy(y, d_y, N*sizeof(float), cudaMemcpyDeviceToHost);

  float maxError = 0.0f;
  for (int i = 0; i < N; i++)
    maxError = max(maxError, abs(y[i]-4.0f));
  printf("Max error: %f\n", maxError);


Based on some other posts similar I have already checked Windows regedit for the key that is commonly missing or set to false and it is set to 1. So I am at a bit of a loss of what to do to get kernel debugging working. Any help would be appreciated