Aviral Goel db79fad16f fix(grouped_gemm): pipeline selection when tail_num varies per group and leads to numerical error (#2863) ...

* fix(grouped_gemm): numerical errors on gfx950 by correctly calculating the tail num

* WIP: add temp config to stress test numerical error correction

* refactor: remove comments

2025-09-16 18:43:19 -07:00

..

batched_gemm_kernel.hpp

Support Wave32 in CK_TILE - Part 1 (#2594)

2025-08-18 10:08:31 -07:00

gemm_kernel.hpp

Support Wave32 in CK_TILE - Part 1 (#2594)

2025-08-18 10:08:31 -07:00

gemm_multi_d_kernel.hpp

[CK-TILE] Default2DEpilogue, example and adding nullptr_t type for D (#2752)

2025-08-28 12:45:50 -07:00

gemm_tile_partitioner.hpp

[CK_TILE] Stream-K GEMM Implementation (#2781)

2025-09-16 16:21:47 -06:00

grouped_gemm_kernel.hpp

fix(grouped_gemm): pipeline selection when tail_num varies per group and leads to numerical error (#2863)

2025-09-16 18:43:19 -07:00

streamk_gemm_kernel.hpp

[CK_TILE] Stream-K GEMM Implementation (#2781)

2025-09-16 16:21:47 -06:00

universal_gemm_kernel.hpp

[CK_TILE] Stream-K GEMM Implementation (#2781)

2025-09-16 16:21:47 -06:00