Mojo package
blockwise_fp8
Blockwise FP8 matmul kernel for SM100.
Modules
-
blockwise_fp8_accumulator: Register-based accumulator for blockwise FP8 matmul. -
blockwise_fp8_matmul: CPU entry points for blockwise FP8 SM100 matmul. -
blockwise_fp8_matmul_kernel: Blockwise FP8 SM100 matmul kernel - Structured kernel with register accumulation. -
blockwise_fp8_output_writer: Output writer for blockwise FP8 SM100 matmul. -
blockwise_fp8_smem: Shared memory layout for blockwise FP8 SM100 matmul.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!