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

Update tile_iterator.cu (#2204)

Browse Source 
Some typos in comments
This commit is contained in:
Haisheng Chen
2025-04-10 11:49:58 -07:00
committed by GitHub
parent 9e1b649827
commit b3f3c7758c
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/04_tile_iterator/tile_iterator.cu
View File

@@ -124,7 +124,7 @@ __global__ void copy(
cudaError_t TestTileIterator(int M, int K) {
// For this example, we chose a <64, 4> tile shape. The PredicateTileIterator expects
// For this example, we chose a <64, 4> tile shape. The PredicatedTileIterator expects
// PitchLinearShape and PitchLinear layout.
using Shape = cutlass::layout::PitchLinearShape<64, 4>;
using Layout = cutlass::layout::PitchLinear;
@@ -136,7 +136,7 @@ cudaError_t TestTileIterator(int M, int K) {
// dimension then along the strided dimension.
using ThreadMap = cutlass::transform::PitchLinearStripminedThreadMap<Shape, kThreads>;
// Define the PredicateTileIterator, using TileShape, Element, Layout, and ThreadMap types
// Define the PredicatedTileIterator, using TileShape, Element, Layout, and ThreadMap types
using Iterator = cutlass::transform::threadblock::PredicatedTileIterator<
Shape, Element, Layout, 1, ThreadMap>;