fix dynamic_per_group_scaled_quant_perf_drop dure to amd_buffer_coherence_enum (ROCm#1769)

Author: yzhou103

csrc/kernels/quant_kernels.cu

@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: MIT
-// Copyright (C) 2024-2025, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (C) 2024-2026, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "aiter_hip_common.h"
 #include "dispatch_utils.h"
@@ -134,8 +134,7 @@ dynamic_per_group_scaled_quant_kernel(DTYPE_O* __restrict__ out,
     using DTYPE_STORE = typename ck_tile::vector_traits<DTYPE_O>::scalar_type;
     auto* out_ptr     = reinterpret_cast<DTYPE_STORE*>(out);
     auto buffer_o =
-        ck_tile::make_buffer_view<ck_tile::address_space_enum::global,
-                                  ck_tile::amd_buffer_coherence_enum::glc>(out_ptr, oob_o);
+        ck_tile::make_buffer_view<ck_tile::address_space_enum::global>(out_ptr, oob_o);
     buffer_o.init_raw();
 
     auto out_s =