biondizzle/nvfp4-megamoe-kernel
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Use kt directly as TMA GMEM coordinate

Browse Source
This commit is contained in:
biondizzle
2026-05-22 22:00:46 +00:00
parent 8c9d4eb1ef
commit 12c166245d
1 changed files with 2 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/fmha_v3_stage_c_example9.py
View File

@@ -217,12 +217,10 @@ class FmhaV3StageCMulti:
cute.copy(tma_q, tAgQ[(None, Int32(0))], tAsQ[(None, qh.index)], tma_bar_ptr=qh.barrier)
qp.tail()
kvp.reset()
kv_coord = Int32(0)
for kt in cutlass.range(0, n_kv_tiles, 1, unroll=1):
kvh = kvp.acquire_and_advance()
cute.copy(tma_k, tBgK[(None, kv_coord)], tBsK[(None, kvh.index)], tma_bar_ptr=kvh.barrier)
cute.copy(tma_v, tVgV[(None, kv_coord)], tVsV[(None, kvh.index)], tma_bar_ptr=kvh.barrier)
kv_coord += 1
cute.copy(tma_k, tBgK[(None, kt)], tBsK[(None, kvh.index)], tma_bar_ptr=kvh.barrier)
cute.copy(tma_v, tVgV[(None, kt)], tVsV[(None, kvh.index)], tma_bar_ptr=kvh.barrier)
kvp.tail()
# ===== MMA warp =====