Skip to main content

Mojo module

blockwise_fp8_smem

Shared memory layout for blockwise FP8 SM100 matmul.

This module provides the SMEM struct for blockwise FP8 matmul kernels where:

  • A-scales are loaded via TMA and stored in SMEM (1D: 1 x BM per stage)
  • B-scales are read directly from global memory (not stored in SMEM)
  • Scaling is applied post-MMA in CUDA cores, not within the MMA unit

Unlike block-scaled matmul, blockwise FP8 uses register-based accumulation across K iterations, with scales applied per-iteration.

Structs

  • BlockwiseFP8Smem: SMEM struct for blockwise FP8 matmul: A/B tiles, A-scales, C output, barriers.

Was this page helpful?