Fused ring RS: per-tile bitmask + drain-ring with owner-boundary flus… - #1

Open

OpenDarrenlu wants to merge 1 commit into

mpdfdfl:feat/sm100-bf16-gemm-reduce-scatterfrom

OpenDarrenlu:ltz_gemm_rs

Commits on Jul 17, 2026

Fused ring RS: per-tile bitmask + drain-ring with owner-boundary flush (+7-11%)

OpenDarrenlu
authored and
lutingzhan
committed
affe008