change dv thread copy

include/ck/tensor_operation/gpu/grid/gridwise_batched_multihead_attention_backward_xdl_cshuffle_v1.hpp

@@ -742,6 +742,28 @@ struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle
             1, // DstScalarStrideInVector
             true>;
         
+        using ABlockwiseCopy_dV = ThreadwiseTensorSliceTransfer_v1r3<
+            FloatGemmAcc,
+            DataType,
+            decltype(a_src_thread_desc_m0_n0_m1_n1_m2_n2_n3_n4),
+            decltype(a_block_desc_m0_n0_m1_n1_m2_n2_n3_n4),
+            tensor_operation::element_wise::Relu,
+            Sequence<Gemm2Params_N_O_M::ABlockSliceLengths_M0_N0_M1_N1::At(
+                         I0), // ThreadSliceLengths
+                     Gemm2Params_N_O_M::ABlockSliceLengths_M0_N0_M1_N1::At(I1),
+                     I1,
+                     I1,
+                     I1,
+                     N2,
+                     I1,
+                     N4>,
+            Sequence<0, 1, 2, 3, 4, 5, 6, 7>,
+            7, // DstVectorDim
+            1, // DstScalarPerVector
+            InMemoryDataOperationEnum::Set,
+            1, // DstScalarStrideInVector
+            true>;
+
         template <typename GridDesc_M0_O_M1>
         using BBlockwiseCopy = ThreadGroupTensorSliceTransfer_v4r1<
             ThisThreadBlock,
@@ -1379,10 +1401,10 @@ struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle
             Gemm2::MakeCGridDesc_N0_O0_N1_O1_N2_O2_O3_O4(vgrad_grid_desc_n_o);
 
         // dV: A matrix VGPR-to-LDS blockwise copy
-        auto vgrad_gemm_tile_p_thread_copy_vgpr_to_lds = typename Gemm2::ABlockwiseCopy{
+        auto vgrad_gemm_tile_p_thread_copy_vgpr_to_lds = typename Gemm2::ABlockwiseCopy_dV{
             Gemm2::a_block_desc_m0_n0_m1_n1_m2_n2_n3_n4,
             Gemm2::MakeAThreadOriginOnBlock_M0_N0_M1_N1_M2_N2_N3_N4(),
-            tensor_operation::element_wise::PassThrough{}};
+            tensor_operation::element_wise::Relu{}};   //relu(P-dropped)
 
         // dV: B matrix global-to-LDS blockwise copy
         auto vgrad_gemm_tile_ygrad_blockwise_copy =