Skip to main content

gemm_blk_rows

Function gemm_blk_rows 

Source
pub fn gemm_blk_rows(f: u32, two_pass: bool) -> usize
Expand description

The activation rows a dense tensor-core GEMM of a matrix in format f (its GGUF type id) takes a block: 64 for a Q8_0 matrix under two words, so two blocks of its streamed path, both words of each stage, share a multiprocessor; GEMM_BLK_TILE_M otherwise.