Skip to main content

gemm_f32_splits

logicaffeine_compile::codegen_cuda::lane::kernels

Function gemm_f32_splits 

Source 
pub fn gemm_f32_splits(tiles: usize, sms: usize, chunks: usize) -> usize
Expand description

The parts an f32 router GEMM (lk_gemm_f32) of tiles 64 × 64 tiles cuts each tile’s chunks runs of 32 k into: as many as give the chip’s sms two blocks each, whole runs of chunks every one.