Manish Gupta 757275f279 Adding more Threadblock Tiles for Mixed-input TensorOp (BF16 * S8) in cutlass_library (#1132) ...

* Adding more tiles in the cutlass_library for mixed-input support.

* fix rebase issue

* more tiles to upcast a

2023-10-13 11:33:15 -04:00

..

cluster_launch

CUTLASS 3.1 (#915)

2023-04-14 23:19:34 -04:00

common

Fix typos (#839)

2023-02-27 11:17:58 -05:00

conv

CUTLASS 3.2.1 (#1113)

2023-09-26 17:24:26 -04:00

core

Support for Mixed Input TensorOp (#1084)

2023-09-27 11:18:30 -04:00

cute

CUTLASS 3.2.1 (#1113)

2023-09-26 17:24:26 -04:00

data/hashes

CUTLASS 2.9 (#468)

2022-04-23 15:02:38 -04:00

epilogue

New updates for 2.11 (#775)

2023-01-20 16:32:57 -05:00

gemm

Adding more Threadblock Tiles for Mixed-input TensorOp (BF16 * S8) in cutlass_library (#1132)

2023-10-13 11:33:15 -04:00

layout

New updates for 2.11 (#775)

2023-01-20 16:32:57 -05:00

nvrtc

CUTLASS 3.2 (#1024)

2023-08-07 20:50:32 -04:00

pipeline

CUTLASS 3.2.1 (#1113)

2023-09-26 17:24:26 -04:00

reduction

CUTLASS 3.2 (#1024)

2023-08-07 20:50:32 -04:00

substrate

CUTLASS 3.1 (#915)

2023-04-14 23:19:34 -04:00

transform

New updates for 2.11 (#775)

2023-01-20 16:32:57 -05:00

util

Allow changing epsilon parameter in RMS norm kernel (#1112)

2023-10-02 20:40:28 -04:00

CMakeLists.txt

CUTLASS 3.2 (#1024)

2023-08-07 20:50:32 -04:00

test_unit.cpp

New updates for 2.11 (#775)

2023-01-20 16:32:57 -05:00