Skip to main content

emit_gemm_program

logicaffeine_compile::codegen_cuda::gemm_tile

Function emit_gemm_program 

Source 
pub fn emit_gemm_program(
    m: usize,
    k: usize,
    n: usize,
    has_tensor_cores: bool,
    tile: &CudaTile,
) -> String
Expand description

Emit a full runnable GEMM program, choosing the compute primitive from the problem + hardware. When the GPU has tensor cores and the shape is 16-aligned, this emits the WMMA (tensor-core) kernel; otherwise the tiled DP4A (CUDA-core) kernel driven by tile. Both paths are byte-exact vs the scalar oracle.