remove passthrough check

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

@@ -539,14 +539,10 @@ struct GridwiseGemmDlMultipleD_km_kn_mn
                 blockwise_gemm.CalculateCThreadOriginOnBlock_BM0_BM1_BN0_BN1(
                     get_thread_local_1d_id());
 
-            if constexpr(!is_same_v<CDEElementwiseOperation,
-                                    ck::tensor_operation::element_wise::PassThrough>)
-            {
             const auto ds_grid_buf = generate_tuple(
                 [&](auto i) {
                     return make_dynamic_buffer<AddressSpaceEnum::Global>(
-                            p_ds_grid[i],
-                            ds_grid_desc_m0_m10_m11_n0_n10_n11[i].GetElementSpaceSize());
+                        p_ds_grid[i], ds_grid_desc_m0_m10_m11_n0_n10_n11[i].GetElementSpaceSize());
                 },
                 Number<NumDTensor>{});
 
@@ -580,8 +576,7 @@ struct GridwiseGemmDlMultipleD_km_kn_mn
                         CThreadTransferSrcDstVectorDim,
                         CThreadTransferDstScalarPerVector,
                         1,
-                            false>(
-                            ds_grid_desc_m0_m10_m11_n0_n10_n11[i],
+                        false>(ds_grid_desc_m0_m10_m11_n0_n10_n11[i],
                                make_multi_index(im0,
                                                 c_m10_m11_n10_n11_thread_origin_idx_on_block[I0],
                                                 c_m10_m11_n10_n11_thread_origin_idx_on_block[I1],
@@ -593,8 +588,7 @@ struct GridwiseGemmDlMultipleD_km_kn_mn
 
             static_for<0, c_m10_m11_n10_n11_thread_tensor_lengths[I0], 1>{}([&](auto m10) {
                 static_for<0, c_m10_m11_n10_n11_thread_tensor_lengths[I1], 1>{}([&](auto m11) {
-                        static_for<0, c_m10_m11_n10_n11_thread_tensor_lengths[I2], 1>{}(
-                            [&](auto n10) {
+                    static_for<0, c_m10_m11_n10_n11_thread_tensor_lengths[I2], 1>{}([&](auto n10) {
                         // load d matrix data
                         static_for<0, NumDTensor, 1>{}([&](auto i) {
                             ds_threadwise_copy(i).Run(ds_grid_desc_m0_m10_m11_n0_n10_n11[i],
@@ -620,9 +614,7 @@ struct GridwiseGemmDlMultipleD_km_kn_mn
                                         make_tuple(0, m10, m11, 0, n10, i));
                                 auto dst_data_refs = generate_tie(
                                     // return type should be lvalue
-                                            [&](auto) -> auto& {
-                                                return c_thread_buf(Number<c_offset>{});
-                                            },
+                                    [&](auto) -> auto& { return c_thread_buf(Number<c_offset>{}); },
                                     Number<2>{});
 
                                 unpack2(cde_element_op, dst_data_refs, src_data_refs);
@@ -648,7 +640,6 @@ struct GridwiseGemmDlMultipleD_km_kn_mn
                             0, 1, -c_m10_m11_n10_n11_thread_tensor_lengths[I1], 0, 0, 0));
                 });
             });
-            }
 
             ThreadwiseTensorSliceTransfer_v1r3<
                 FloatAcc,