mirror of
https://github.com/FlashML-org/FreeToken.git
synced 2026-11-22 05:49:20 +08:00
- pass maxnreg=255 to multi-row stage-1 launches on CUDA: without .maxnreg ptxas holds these tiles at 80-128 registers for occupancy and spills 8-16 B on sm_80 to sm_120 - pick block_n and warps from a launch sweep under maxnreg, among the variants that do not spill on sm_80, 86, 89, 90, 100 and 120; quantized head_dim 512 takes 16 rows x heads again