Format

src/targets/gpu/kernels/include/migraphx/kernels/ck.hpp

@@ -59,14 +59,15 @@ constexpr auto to_ck_tensor()
     });
 }
 
-template<class F>
+template <class F>
 struct ck_function_adaptor : F
 {
-    template<class... Ts>
+    template <class... Ts>
     constexpr ck_function_adaptor(Ts&&... xs) : F(static_cast<Ts&&>(xs)...)
-    {}
+    {
+    }
 
-    template<class T, class... Ts>
+    template <class T, class... Ts>
     constexpr void operator()(T& out, Ts&&... xs) const
     {
         out = static_cast<const F&>(*this)(static_cast<Ts&&>(xs)...);
@@ -75,9 +76,10 @@ struct ck_function_adaptor : F
 
 struct ck_nop
 {
-    template<class T>
+    template <class T>
     constexpr void operator()(T&) const
-    {}
+    {
+    }
 };
 
 } // namespace migraphx

src/targets/gpu/kernels/include/migraphx/kernels/ck_gemm.hpp

@@ -40,28 +40,30 @@ __device__ void ck_gemm(A a, B b, E e, Ds... ds)
 
     constexpr const auto a_grid_desc_m_k = gemm.matrix_padder.PadADescriptor_M_K(to_ck_tensor<A>());
     constexpr const auto b_grid_desc_n_k = gemm.matrix_padder.PadBDescriptor_N_K(to_ck_tensor<B>());
-    constexpr const auto e_grid_desc_m_n       = gemm.matrix_padder.PadCDescriptor_M_N(to_ck_tensor<E>());
-    constexpr const auto ds_grid_desc_m_n       = ck::make_tuple(gemm.matrix_padder.PadCDescriptor_M_N(to_ck_tensor<Ds>())...);
-    constexpr const auto block_2_etile_map     = gemm.MakeDefaultBlock2ETileMap(e_grid_desc_m_n);
+    constexpr const auto e_grid_desc_m_n = gemm.matrix_padder.PadCDescriptor_M_N(to_ck_tensor<E>());
+    constexpr const auto ds_grid_desc_m_n =
+        ck::make_tuple(gemm.matrix_padder.PadCDescriptor_M_N(to_ck_tensor<Ds>())...);
+    constexpr const auto block_2_etile_map = gemm.MakeDefaultBlock2ETileMap(e_grid_desc_m_n);
 
     using GridwiseGemm = typename G::GridwiseGemm;
 
     // tensor descriptors for block/thread-wise copy
-    constexpr auto a_grid_desc_ak0_m_ak1 = GridwiseGemm::MakeDefaultAGridDescriptor_AK0_M_AK1(a_grid_desc_m_k);
-    constexpr auto b_grid_desc_bk0_n_bk1 = GridwiseGemm::MakeDefaultBGridDescriptor_BK0_N_BK1(b_grid_desc_n_k);
+    constexpr auto a_grid_desc_ak0_m_ak1 =
+        GridwiseGemm::MakeDefaultAGridDescriptor_AK0_M_AK1(a_grid_desc_m_k);
+    constexpr auto b_grid_desc_bk0_n_bk1 =
+        GridwiseGemm::MakeDefaultBGridDescriptor_BK0_N_BK1(b_grid_desc_n_k);
 
     constexpr auto ds_grid_desc_mblock_mperblock_nblock_nperblock =
-        GridwiseGemm::MakeDsGridDescriptor_MBlock_MPerBlock_NBlock_NPerBlock(
-            ds_grid_desc_m_n);
+        GridwiseGemm::MakeDsGridDescriptor_MBlock_MPerBlock_NBlock_NPerBlock(ds_grid_desc_m_n);
 
     constexpr auto e_grid_desc_mblock_mperblock_nblock_nperblock =
-        GridwiseGemm::MakeEGridDescriptor_MBlock_MPerBlock_NBlock_NPerBlock(
-            e_grid_desc_m_n);
+        GridwiseGemm::MakeEGridDescriptor_MBlock_MPerBlock_NBlock_NPerBlock(e_grid_desc_m_n);
 
     __shared__ char p_shared_block[GridwiseGemm::GetSharedMemoryNumberOfByte()];
 
     constexpr const bool HasMainKBlockLoop =
-        GridwiseGemm::CalculateHasMainKBlockLoop(a_grid_desc_ak0_m_ak1.GetLength(ck::Number<0>{}) * a_grid_desc_ak0_m_ak1.GetLength(ck::Number<2>{}));
+        GridwiseGemm::CalculateHasMainKBlockLoop(a_grid_desc_ak0_m_ak1.GetLength(ck::Number<0>{}) *
+                                                 a_grid_desc_ak0_m_ak1.GetLength(ck::Number<2>{}));
     GridwiseGemm::template Run<HasMainKBlockLoop>(a.data(),
                                                   b.data(),
                                                   ck::make_tuple(ds.data()...),

src/targets/gpu/kernels/include/migraphx/kernels/ck_gemm_includes.hpp

@@ -156,7 +156,8 @@ template <typename ALayout,
           ck::LoopScheduler LoopSched = ck::make_default_loop_scheduler()>
 struct CK_DeviceGemmMultipleD
 {
-    ck::tensor_operation::device::MatrixPadder<GemmSpec, ck::index_t, ck::index_t, ck::index_t> matrix_padder {MPerBlock, NPerBlock, KPerBlock};
+    ck::tensor_operation::device::MatrixPadder<GemmSpec, ck::index_t, ck::index_t, ck::index_t>
+        matrix_padder{MPerBlock, NPerBlock, KPerBlock};
 
     // GridwiseGemm
     using GridwiseGemm = ck::GridwiseGemmMultipleD_xdl_cshuffle<
@@ -203,7 +204,7 @@ struct CK_DeviceGemmMultipleD
         LoopSched>;
 
     // return block_id to E matrix tile idx (m0, n0) mapping
-    template<class EGridDesc_M_N>
+    template <class EGridDesc_M_N>
     __device__ static constexpr auto
     MakeDefaultBlock2ETileMap(const EGridDesc_M_N& e_grid_desc_m_n_)
     {