Merge commit '3e6c2610' into gemm_norm

include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp

 #pragma once
+
 #include "common_header.hpp"
 #include "multi_index_transform_helper.hpp"
 #include "tensor_descriptor.hpp"
@@ -287,11 +288,11 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
     }
 
     // return block_id to C matrix tile idx (m0, n0) mapping
-    __host__ __device__ static constexpr auto
-    MakeDefaultBlock2CTileMap(const CGridDesc_M_N& c_grid_desc_m_n, index_t M01, index_t N01)
+    __host__ __device__ static constexpr auto MakeDefaultBlock2CTileMap(
+        const CGridDesc_M_N& c_grid_desc_m_n, index_t /* M01 */, index_t /* N01 */)
     {
-        return BlockToCTileMap_M00_N00_M01_N01<MPerBlock, NPerBlock, CGridDesc_M_N>(
-            c_grid_desc_m_n, M01, N01);
+        return BlockToCTileMap_M00_N0_M01Adapt<MPerBlock, NPerBlock, CGridDesc_M_N>(
+            c_grid_desc_m_n);
     }
 
     using CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2 =

include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4.hpp

@@ -265,10 +265,10 @@ struct GridwiseGemm_bk0mk1_bk0nk1_mn_xdlops_v2r4
 
     // return block_id to C matrix tile idx (m0, n0) mapping
     __host__ __device__ static constexpr auto MakeCBlockClusterAdaptor(
-        const CMNGridDesc& c_m_n_grid_desc, index_t M01, index_t N01, index_t KBatch)
+        const CMNGridDesc& c_m_n_grid_desc, index_t /* M01 */, index_t /* N01 */, index_t KBatch)
     {
-        return BlockToCTileMap_KSplit_M00_N00_M01_N01<MPerBlock, NPerBlock, CMNGridDesc>(
-            c_m_n_grid_desc, M01, N01, KBatch);
+        return BlockToCTileMap_KSplit_M00_N0_M01Adapt<MPerBlock, NPerBlock, CMNGridDesc>(
+            c_m_n_grid_desc, 8, KBatch);
     }
 
     using CM0N0M1N1M2M3M4N2GridDesc = decltype(MakeCM0N0M1N1M2M3M4N2GridDescriptor(CMNGridDesc{}));