CUDA "warpSize"?

April 19, 2012 - 1 min read

GPU Hacks

In CUDA C, the built-in variable warpSize is initially treated as a variable at compile-time and doesn’t appear to be recognized as a constant until the PTX generation phase. This could be an issue if the warp width is part of some tricky preprocessing early in the compilation.

The simple line:

    const unsigned int w99 = warpSize * 99;`

is resolved to the following PTX:

    mov.u32    %r5, WARP_SZ;
    mul.lo.s32 %r6, %r5, 99;

Yes, the stage after PTX will do a great job folding/propagating away WARP_SZ but sometimes you need to resolve logic in the preprocessor.

Including a #define WARP_SIZE 32 in your kernel is a workaround until NVIDIA tells us otherwise.