[Triton MoE] Add optimized Gluon kernel for AMD CDNA3 with K-dimension unrolling#2277

Open

jwu10003 wants to merge 7 commits intoROCm:mainfrom

jwu10003:issue1638-moe

Commits on Feb 27, 2026

[Triton] Fix triton tests fail due to the API changes from Triton upstream
jwu10003
committed

Commits on Mar 16, 2026

Merge branch 'main' into issue1638-moe
gyohuangxin
authored