Diag: use Python range() unrolling like stage C test

This commit is contained in:
2026-05-22 17:56:59 +00:00
parent 200430bd3f
commit beecc4df47

View File

@@ -167,7 +167,7 @@ class FmhaV3Diag:
cute.copy(tma_q, tAgQ[(None, Int32(0))], tAsQ[(None, qh.index)], tma_bar_ptr=qh.barrier)
qp.tail()
kvp.reset(); pk = kvp.try_acquire()
for kt in cutlass.range(self.n_kv_tiles, unroll=1):
for kt in range(self.n_kv_tiles):
coord = Int32(kt)
kvh = kvp.acquire_and_advance(pk)
cute.copy(tma_k, tBgK[(None, coord)], tBsK[(None, kvh.index)], tma_bar_ptr=kvh.barrier)