biondizzle 016edbcc97 D5c: add row_sum output for proper external normalization ...

The kernel's O_unnorm is max-shifted (divided by 2^row_max), so
O_norm != O_unnorm * exp(-LSE). Instead, O_norm = O_unnorm / row_sum.
Added mRowSums output tensor to enable correct normalization.

2026-05-26 15:07:22 +00:00

..

__init__.py

Restructure: cutedsl/ -> dsv4/ with proper layering

2026-05-21 17:30:44 +00:00

fmha_backup_pre_epilog.py

D1.5: Replace TMEM round-trip normalize with correction epilog (one-way: TMEM→reg→SMEM→GMEM)

2026-05-24 00:24:24 +00:00

fmha.py

D5c: add row_sum output for proper external normalization

2026-05-26 15:07:22 +00:00

fmha.py.backup

auto: pre-test commit

2026-05-23 20:08:31 +00:00