Bind CUDA Texture to a float Image

Question

I have a 1 channel, float image in C side like the following: ``` int width, height; float* img; ``` I want to pass this image to a CUDA texture. I'm reading the NVIDIA CUDA C Programming Guide (page 42-43) and using the tutorial, wrote a code like the following: main.cpp: ``` int main() { int width, height; float* h_Input; ReadImage(&h_Input, &width, &height); // My function which reads the image. WriteImage(h_Input, width, height); // works perfectly... float* h_Output = (float*) malloc(sizeof(float) * width * height); CalculateWithCuda(h_Input, h_Output, width,height); WriteImage(h_Output, width, height); // writes an empty-gray colored image.... *WHY???* } ``` kernel.cu: ``` texture<float, cudaTextureType2D, cudaReadModeElementType> texRef; // 2D float texture __global__ void Kernel(float* output, int width, int height) { int i = blockIdx.y * blockDim.y + threadIdx.y; // row number int j = blockIdx.x * blockDim.x + threadIdx.x; // col number if(i < height && j < width) { float tem…

Accepted Answer

When you use the normalized coordinates the texture is accessed via the coordinates from 0 to 1 (exclusive). You forgot translate your integer threadIdx-based coordinates into normalized.

```
unsigned int x = blockIdx.x * blockDim.x + threadIdx.x;
unsigned int y = blockIdx.y * blockDim.y + threadIdx.y; 
float u = x / (float)width;  
float v = y / (float)height;
```

Bind CUDA Texture to a float Image

Solution

Problem