qutaoandClaude Fable 5 f681a83779 Slim tilelang/quantize/nvfp4.py ...

Two structural reductions, no behavior change:

- Merge the three TMA-staged quantize kernel variants (16/32/64x256)
  into one _tilelang_nvfp4_blockscaled_quantize_kernel_tiled factory:
  their inner loops were line-identical, differing only in rows_per_cta,
  stage count and the store path (TMA store vs direct global stores),
  all of which are trace-time parameters now. Outputs verified
  bitwise-identical to the previous per-variant kernels for every
  dispatch configuration (quant bytes and scale source).
- Move the 14 SM120 OMMA scale-selector contract oracles into the
  contract test file that is their only consumer; the quantize module
  stays focused on quantization and layout packing.

1334 -> 978 lines. Quantize suite: 32 passed.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

2026-07-10 15:21:00 +08:00

..

test_tilelang_quantize_nvfp4_scale_layout.py

Slim tilelang/quantize/nvfp4.py

2026-07-10 15:21:00 +08:00