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

Doc typo

Browse Source 
Signed-off-by: Peter Han <fujun.han@iluvatar.ai>
This commit is contained in:
Peter Han 2021-07-29 08:43:27 +08:00
parent 1ac4559d12
commit 64dd1e1915
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
media/docs/efficient_gemm.md
View File

@ -163,7 +163,7 @@ consecutively launched threadblocks to packed two-dimensional regions of the par
problem to increase the probability that these will access the same tiles of global memory at problem to increase the probability that these will access the same tiles of global memory at
approximately the same time. approximately the same time.
Several functions are defined in [cutlass/gemm/threadblock_swizzle.h](cutlass/gemm/threadblock/threadblock_swizzle.h). Several functions are defined in [cutlass/gemm/threadblock_swizzle.h](/include/cutlass/gemm/threadblock/threadblock_swizzle.h).
### Parallelized Reductions ### Parallelized Reductions