NVIDIA/cutlass
1
0
Fork 0
mirror of https://github.com/NVIDIA/cutlass.git synced 2026-05-11 08:50:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

cutlass 3.9 update (#2255)

Browse Source 
* cutlass 3.9 update

* rebase

* fixes out of shared memory for blockwise Blackwell

* doc format

* fix issue 2253

* disable host ref by default

* fix sm120 smem capacity

---------

Co-authored-by: yuzhai <yuzhai@nvidia.com>
Co-authored-by: Haicheng Wu <haichengw@nvidia.com>
This commit is contained in:
Yujia Zhai
2025-04-24 12:42:40 -07:00
committed by GitHub
parent 8e345c5c5b
commit 331a1f5b3f
143 changed files with 18089 additions and 5935 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/04_tile_iterator/tile_iterator.cu
View File

@@ -34,7 +34,7 @@
addressable memory, and then store it back into addressable memory.
TileIterator is a core concept in CUTLASS that enables efficient loading and storing of data to
and from addressable memory. The PredicateTileIterator accepts a ThreadMap type, which defines
and from addressable memory. The PredicatedTileIterator accepts a ThreadMap type, which defines
the mapping of threads to a "tile" in memory. This separation of concerns enables user-defined
thread mappings to be specified.