Merge remote-tracking branch 'origin/develop' into myamlak/cgemm

include/ck/tensor_operation/gpu/grid/gridwise_gemm_reduce_xdl_cshuffle_v1.hpp

@@ -306,7 +306,7 @@ struct GridwiseGemmReduce_k0mk1_k0nk1_mn_xdl_cshuffle_v1
     __host__ __device__ static constexpr auto
     MakeDefaultBlock2CTileMap(const CGridDesc_M_N& c_grid_desc_m_n)
     {
-        return BlockToCTileMap_M00_N00_M01_N01<MPerBlock, NPerBlock, CGridDesc_M_N>(
+        return BlockToCTileMap_M00_N0_M01Adapt<MPerBlock, NPerBlock, CGridDesc_M_N>(
             c_grid_desc_m_n);
     }
 

include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v1.hpp

@@ -259,7 +259,7 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdl_cshuffle_v1
     __host__ __device__ static constexpr auto
     MakeDefaultBlock2CTileMap(const CGridDesc_M_N& c_grid_desc_m_n)
     {
-        return BlockToCTileMap_M00_N00_M01_N01<MPerBlock, NPerBlock, CGridDesc_M_N>(
+        return BlockToCTileMap_M00_N0_M01Adapt<MPerBlock, NPerBlock, CGridDesc_M_N>(
             c_grid_desc_m_n);
     }
 

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 =