Use and steps of CUDA texture
4.0
There are two kinds of texture binding, one is to bind to linear memory, that is, to use cudammalloc(); cudamemcpy(); to open up memory space, the other is to bind to cudammallocarray, & nbsp; cudamemcpytoarray to open up two-dimensional array or three-dimensional array. First of all, it is relatively simple to bind to the memory space opened up by cudamalloc. The first is the texture declaration: the declaration and binding of texture should be carried out in the Cu file, which will be more troublesome in other files. The first is to open up the video memory and assign the value to the video memory. The first is the texture declaration: 1. Texture textref; 2. Open up space for display memory, and initialize cudamalloc ((void * *) & amp; rain_ table,256*4*3); cudaMemcpy(rain_ 3. Texture binding: cudabinditexture (0, textref, rain)_ table);