Commit History

fix(htm): set NON_PORTABLE_CLUSTER_SIZE_ALLOWED=1 for cluster_size=16
19dee83
verified

icarus112 commited on

perf(htm): Hopper cluster::sync hardware barrier + sm_90a + cluster launch attr
7721a60
verified

icarus112 commited on

perf(htm): DLB software grid barrier + non-cooperative launch (lifts 132-SM cap)
5c751ce
verified

icarus112 commited on

perf(htm): batched cooperative kernel — B=8 regions in ONE launch via blockIdx.y indexing
bd5981d
verified

icarus112 commited on

fix(htm_rust): remove software grid-barrier slow path
c122498
verified

icarus112 commited on

Update Feather H200 training runtime image
36a4397
verified

icarus112 commited on