fixed splitK

example/15_grouped_gemm/grouped_gemm_xdl_bfp16.cpp

@@ -22,7 +22,7 @@ template <ck::index_t... Is>
 using S = ck::Sequence<Is...>;
 
 using BF16 = ck::bhalf_t;
-using F32 = float;
+using F32  = float;
 
 using Row = ck::tensor_layout::gemm::RowMajor;
 using Col = ck::tensor_layout::gemm::ColumnMajor;

example/24_batched_gemm/batched_gemm_xdl_bfp16.cpp

@@ -19,7 +19,7 @@ template <ck::index_t... Is>
 using S = ck::Sequence<Is...>;
 
 using BF16 = ck::bhalf_t;
-using F32 = float;
+using F32  = float;
 
 using Row = ck::tensor_layout::gemm::RowMajor;
 using Col = ck::tensor_layout::gemm::ColumnMajor;

include/ck/tensor_operation/gpu/device/device_gemm_xdl_splitk_c_shuffle.hpp

@@ -95,7 +95,7 @@ struct DeviceGemmXdlSplitKCShuffle : public DeviceGemmSplitK<ALayout,
 
         const auto a_grid_desc_m_kpad = transform_tensor_descriptor(
             a_grid_desc_m_k,
-            make_tuple(make_right_pad_transform(K, KPad - K), make_pass_through_transform(M)),
+            make_tuple(make_pass_through_transform(M), make_right_pad_transform(K, KPad - K)),
             make_tuple(Sequence<0>{}, Sequence<1>{}),
             make_tuple(Sequence<0>{}, Sequence<1>{}));
 

include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4r2.hpp

@@ -547,11 +547,6 @@ struct GridwiseGemm_bk0mk1_bk0nk1_mn_xdlops_v2r4r2
                 static_cast<FloatC*>(p_shared_block),
                 c_block_desc_mblock_mperblock_nblock_nperblock.GetElementSpaceSize());
 
-            static_assert(M1 == MWave, "");
-            static_assert(N1 == NWave, "");
-            static_assert(M2 * M3 * M4 == MPerXDL, "");
-            static_assert(N2 == NPerXDL, "");
-
             constexpr auto c_block_desc_m0_n0_m1_n1_m2_m3_m4_n2 = transform_tensor_descriptor(
                 c_block_desc_mblock_mperblock_nblock_nperblock,
                 make_tuple(