Skip to main content

emit_wmma_i8

logicaffeine_compile::codegen_cuda::gemm_tile

Function emit_wmma_i8 

Source 
pub fn emit_wmma_i8() -> String
Expand description

Emit a WMMA int8 tensor-core GEMM kernel: one warp per 16×16 output tile, accumulating over K in 16-wide fragments with wmma::mma_sync (the IMMA path). Row-major A(M×K)·B(K×N) → int32 C(M×N).