ptxas gives DEPBAR.LE SB0, n (cp.async.wait_group) 4 on sm_86 and sm_121
(every wait read off nvdisasm -hex on the DGX Spark, 2026-09-26, research
p4-hmma/cp_async*.cu); the realization keeps it.
547def sm86WaitAsyncGroupsStall : Nat = 4The compiler supplied declaration spans and resolved links from this source snapshot. This page does not assert that this file belongs to a checked closure.