NVIDIA/cutlass
1
0
Fork 0
mirror of https://github.com/NVIDIA/cutlass.git synced 2026-04-20 06:48:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

[Doc]Fix Mode Name and Stride in 0t_mma_atom.md (#2910)

Browse Source 
* Fix wrong stride in 0t_mma_atom.md

* Fix mode name.

* Update media/docs/cpp/cute/0t_mma_atom.md

Co-authored-by: Cris Cecka <ccecka@users.noreply.github.com>

---------

Co-authored-by: Cris Cecka <ccecka@users.noreply.github.com>
This commit is contained in:
Qi Yuhang
2026-01-23 12:53:30 +08:00
committed by GitHub
parent 3f5bafb326
commit 667446a9dd
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
media/docs/cpp/cute/0t_mma_atom.md
View File

@@ -403,7 +403,7 @@ using CLayout = Layout<Shape <Shape < _4, _8, ...>, Shape < _2, _2>>,
Stride<Stride<_128, _1, ...>, Stride<_64, _8>>>;
```
Finally, we get this entire pattern repeating four times, once for each warp, down the `M`-mode starting at `(m,n) = (16,0) = 16`. where two core matrices that belong to the same warp are stacked on top of each other. This makes the size of the final sub-mode of M 4. As for the stride, this time we go to `(T32, V0)`, which makes it a stride of 32.
Finally, we get this entire pattern repeating four times, once for each warp, down the `M`-mode starting at `(m,n) = (16,0) = 16`, where four core matrices that belong to the same warp are stacked on top of each other. This makes the size of the final sub-mode of `thrID` 4 (there are four warps) with a stride of `16` (to take us to coordinate `(16,0) = 16`).
```cpp
// (T128,V4) -> (M64,N8)