ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-20 04:49:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fixes to "General 2D Reduction Kernel" (#2535) (#2656)

Browse Source 
* fix reduce2d

- revret the combine_partial_results() chnages
- remove auto from function def

* clang-format

[ROCm/composable_kernel commit: 191c62967b]
This commit is contained in:
Yashvardhan Agarwal
2025-08-11 16:01:33 +03:00
committed by GitHub
parent ffaa6ba3b3
commit 78662c8ad5
4 changed files with 7 additions and 63 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
include/ck_tile/ops/reduce/kernel/reduce2d_kernel.hpp
View File

@@ -189,7 +189,9 @@ struct Reduce
/// @note Requirements:
/// - y_continous_dim % ThreadTile_N == 0 (for proper thread distribution)
/// - input_strides[-1] == 1 (for contiguous memory access)
CK_TILE_HOST static bool IsSupportedArgument(index_t y_continous_dim, auto input_strides)
template <typename InputStrides>
CK_TILE_HOST static bool IsSupportedArgument(index_t y_continous_dim,
InputStrides input_strides)
{
using S = typename Problem::BlockShape;