LWPCK-2429: Device grouped GEMM uses Async Memcpy (#1695)

* LWPCK-2429: Device grouped GEMM uses Async Memcpy
Resolving merge conflicts

* reverting changes to profile_grouped_gemm

* revert date change

---------
Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>

include/ck/tensor_operation/gpu/device/impl/device_grouped_gemm_multiple_d_dl.hpp

 #pragma once
 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
 
 #pragma once
 
@@ -603,7 +603,7 @@ struct DeviceGroupedGemmMultipleD_Dl : public DeviceGroupedGemm<ALayout,
             }
 
             hipGetErrorString(
-                hipMemcpyWithStream(arg.p_workspace_,
+                hipMemcpyAsync(arg.p_workspace_,
                                arg.gemm_desc_kernel_arg_.data(),
                                arg.gemm_desc_kernel_arg_.size() * sizeof(GemmKernelArg),
                                hipMemcpyHostToDevice,

include/ck/tensor_operation/gpu/device/impl/device_grouped_gemm_multiple_d_splitk_xdl_cshuffle_two_stage.hpp

@@ -761,7 +761,7 @@ struct DeviceGroupedGemmMultipleDSplitKXdlCShuffleTwoStage
             float time{0.f};
 
             hip_check_error(
-                hipMemcpyWithStream(dev_gemm_kargs,
+                hipMemcpyAsync(dev_gemm_kargs,
                                arg.gemm_kernel_args_.data(),
                                arg.gemm_kernel_args_.size() * sizeof(GemmTransKernelArg),
                                hipMemcpyHostToDevice,

include/ck/tensor_operation/gpu/device/impl/device_grouped_gemm_multiple_d_xdl_cshuffle_tile_loop.hpp

@@ -940,7 +940,7 @@ struct DeviceGroupedGemmMultipleDXdlCShuffleTileLoop
                              const void* p_host_kernel_args) const
     {
         arg.p_dev_gemm_args_ = p_dev_kernel_args;
-        hip_check_error(hipMemcpy(p_dev_kernel_args,
+        hip_check_error(hipMemcpyAsync(p_dev_kernel_args,
                                        p_host_kernel_args,
                                        GetDeviceKernelArgSize(&arg),
                                        hipMemcpyHostToDevice));

include/ck/tensor_operation/gpu/device/impl/device_grouped_gemm_xdl.hpp

@@ -557,10 +557,10 @@ struct DeviceGroupedGemm_Xdl : public DeviceGroupedGemm<ALayout,
                 }
             }
 
-            hipGetErrorString(hipMemcpyWithStream(arg.p_workspace_,
+            hipGetErrorString(
+                hipMemcpyAsync(arg.p_workspace_,
                                arg.gemm_desc_kernel_arg_.data(),
-                                                  arg.gemm_desc_kernel_arg_.size() *
-                                                      sizeof(GemmBiasTransKernelArg),
+                               arg.gemm_desc_kernel_arg_.size() * sizeof(GemmBiasTransKernelArg),
                                hipMemcpyHostToDevice,
                                stream_config.stream_id_));
 

include/ck/tensor_operation/gpu/device/impl/device_grouped_gemm_xdl_splitk_cshuffle.hpp

@@ -421,7 +421,7 @@ struct DeviceGroupedGemmXdlSplitKCShuffle : public DeviceGroupedGemmSplitK<ALayo
             }
 
             hip_check_error(
-                hipMemcpyWithStream(arg.p_workspace_,
+                hipMemcpyAsync(arg.p_workspace_,
                                arg.gemm_kernel_args_.data(),
                                arg.gemm_kernel_args_.size() * sizeof(GemmTransKernelArg),
                                hipMemcpyHostToDevice,