[CK_TILE] Update flatmm related kernels (#3022)

---------

Co-authored-by: Ding, Yi <yi.ding@amd.com>
Co-authored-by: felix <felix.li@amd.com>

include/ck_tile/ops/flatmm/block/block_flatmm_asmem_bsmem_creg_v1.hpp

@@ -113,6 +113,7 @@ struct BlockFlatmmASmemBSmemCRegV1
                         merge_sequences(sequence<mIter, nIter>{}, c_warp_y_index_zeros),
                         merge_sequences(sequence<1, 1>{}, c_warp_y_lengths),
                         c_warp_tensor.get_thread_buffer());
+                    __builtin_amdgcn_sched_barrier(0x7F6);
                 });
             });
         });