CUTLASS 3.0.0 (#786)

* CUTLASS 3.0.0
This commit is contained in:
Vijay Thakkar
2023-01-23 20:55:28 -05:00
committed by GitHub
parent 66d9cddc83
commit 277bd6e537
377 changed files with 76396 additions and 1186 deletions
@@ -481,7 +481,7 @@ struct AttentionKernel {
SharedStorage& shared_storage = *((SharedStorage*)smem_buffer);
auto& m_prime = shared_storage.m_prime;
auto& s_prime = shared_storage.s_prime;
auto& si = shared_storage.after_mm0.si;
[[maybe_unused]] auto& si = shared_storage.after_mm0.si;
auto& mi = shared_storage.mi;
static_assert(kQueriesPerBlock < kNumWarpsPerBlock * kWarpSize, "");