refactor: dedupe fp8 kernel helpers and trim comments

Browse Source

- merge the quantize launchers into one Tiled template; extract shared cvt_fp8/publish_amax helpers and replace the dtype x format ladder with two-level template dispatch
- fold the gemm interior/generic operand loads into one kInterior template and the fast/generic async loads into load_async<kFast>; Policy carries the smem budget
- compress kernel comments to the load-bearing invariants, dropping measured-number essays; Policy signature and kernel code unchanged

Benchmark: NVIDIA L20, 1.2B model train step fwd+bwd+CE
- M=8192: fp8 532.2 -> 530.4 ms (1.26x, noise); tests/extension 65 passed, quantize layouts byte-exact, NT routing diff 0.0

...

This commit is contained in:

ViperEkura

2026-08-28 16:45:21 +08:00

parent 8a353117ea

commit bf239d194c

4 changed files with 376 additions and 651 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

csrc/kernels/fp8/gemm.cuh

+226 -434

File diff suppressed because it is too large Load Diff