youngkingdom/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

v4.1 release update v2. (#2481)

Browse Source
This commit is contained in:
Junkai-Wu
2025-07-22 10:03:55 +08:00
committed by GitHub
parent 9baa06dd57
commit fd6cfe1ed0
179 changed files with 7878 additions and 1286 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/53_hopper_gemm_permute/53_hopper_gemm_permute.cu
View File

@ -65,7 +65,7 @@
The approach relies on two things:
- The ability of CUTLASS 3 to naturally perform general tensor contractions (GETT) owing to the
flexibility of CuTe's hierarchical layouts (see example 51_hopper_gett for more details).
- The harware capabilities of Hopper TMA units that allow for loading multidimensional tensors with
- The hardware capabilities of Hopper TMA units that allow for loading multidimensional tensors with
(almost) arbitrary strides, which can be used to represent a permuted view of the data.
In this example we reuse the permutation classes of examples 39_gemm_permute as operation tags.