ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 06:49:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK TILE][AICK-439] Fix cshuffle epilogue wave per shuffle (#3364)

Browse Source 
* [CK TILE] Fix cshufle epligoue wave per shuffle

* Align shuffle per tile with smem

* fixes

* Fixes for double smem

* fix
This commit is contained in:
Bartłomiej Kocot
2025-12-15 12:59:48 +01:00
committed by GitHub
parent 3143a5a480
commit 3b773109e5
11 changed files with 73 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
test/ck_tile/gemm/test_gemm_pipeline_util.hpp
View File

@@ -202,7 +202,13 @@ class TestCkTileGemmPipeline : public ::testing::Test
N_Warp_Tile,
K_Warp_Tile,
UniversalGemmProblem::TransposeC,
memory_operation>>;
memory_operation,
1, /*kNumWaveGroups_*/
false, /*FixedVectorSize_*/
1, /*VectorSizeC_*/
false, /*TiledMMAPermuteN_*/
1, /*BlockedXDLN_PerWarp_*/
DoubleSmemBuffer /*DoubleSmemBuffer*/>>;
using Kernel = ck_tile::GemmKernel<TilePartitioner, GemmPipeline, GemmEpilogue>;
auto kargs = Kernel::MakeKernelArgs(args);