diff --git a/example/01_gemm/run_gemm_example_streamk_v2.inc b/example/01_gemm/run_gemm_example_streamk_v2.inc
index 0a07015c08..6679f95157 100644
--- a/example/01_gemm/run_gemm_example_streamk_v2.inc
+++ b/example/01_gemm/run_gemm_example_streamk_v2.inc
@@ -116,27 +116,38 @@ bool run_gemm(const ProblemType& problem_size, const ExecutionConfig& config)
         };
 
     auto f_get_default_stride =
-        [](std::size_t row, std::size_t col, std::size_t stride, auto layout) {
+        [](std::size_t row, std::size_t col, ck::index_t stride, auto layout) {
             if(stride == -1)
             {
-                // give a chance if stride is zero, return a default packed stride
+                // give a chance if stride is -1, return a default packed stride
                 if constexpr(std::is_same_v<decltype(layout), ck::tensor_layout::gemm::RowMajor>)
                 {
-                    return col;
+                    return static_cast<std::size_t>(col);
                 }
                 else
                 {
-                    return row;
+                    return static_cast<std::size_t>(row);
                 }
             }
             else
-                return stride;
+                return static_cast<std::size_t>(stride);
         };
 
+    auto f_get_default_streamk_policy = [](ck::index_t streamk_sel) {
+        if(streamk_sel == -1)
+        {
+            return static_cast<std::size_t>(4);
+        }
+        else
+            return static_cast<std::size_t>(streamk_sel);
+    };
+
     StrideA = f_get_default_stride(M, K, StrideA, ALayout{});
     StrideB = f_get_default_stride(K, N, StrideB, BLayout{});
     StrideC = f_get_default_stride(M, N, StrideC, CLayout{});
 
+    Streamk_sel = f_get_default_streamk_policy(Streamk_sel);
+
     Tensor<ADataType> a_m_k(f_host_tensor_descriptor(M, K, StrideA, ALayout{}));
     Tensor<BDataType> b_k_n(f_host_tensor_descriptor(K, N, StrideB, BLayout{}));
 
diff --git a/include/ck/tensor_operation/gpu/grid/block_to_ctile_map.hpp b/include/ck/tensor_operation/gpu/grid/block_to_ctile_map.hpp
index 828a0d1e9f..56c37b1b72 100644
--- a/include/ck/tensor_operation/gpu/grid/block_to_ctile_map.hpp
+++ b/include/ck/tensor_operation/gpu/grid/block_to_ctile_map.hpp
@@ -1517,30 +1517,6 @@ struct BlockToCTileMap_GemmStreamK_v2
             equiv_tiles_big       = MDiv(upper_big / k_iters_per_tile.get());
             equiv_tiles_little    = MDiv(upper_little / k_iters_per_tile.get());
         }
-
-        if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING)))
-        {
-            printf("streamk_sel=%0d,grid_size=%0d, num_tiles:%d, dp_tiles:%d, sk_tiles:%u, "
-                   "sk_num_blocks:%d,dp_num_blocks:%d,sk_num_big_blocks:%d, "
-                   "sk_total_iters:%d, dp_start_block_idx:%d,  "
-                   "k_iters_per_tile:%d, k_iters_per_big_block:%d, reduction_start_block_idx:%u, "
-                   " workspace(acc float):%u\n",
-                   streamk_sel,
-                   grid_size,
-                   num_tiles,
-                   dp_tiles,
-                   get_sk_tiles(),
-                   sk_num_blocks,
-                   dp_num_blocks,
-                   sk_num_big_blocks,
-                   sk_total_iters,
-                   dp_start_block_idx,
-
-                   k_iters_per_tile.get(),
-                   k_iters_per_big_block,
-                   reduction_start_block_idx,
-                   get_workspace_size(sizeof(float)));
-        }
     }
 
     __host__ __device__ static constexpr index_t CalculateGridSize(index_t M, index_t N)
diff --git a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_streamk_v3.hpp b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_streamk_v3.hpp
index c5e7feb6dc..ff10215353 100644
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_streamk_v3.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_streamk_v3.hpp
@@ -1315,7 +1315,7 @@ struct GridwiseGemm_xdl_cshuffle_streamk_v3
 
                 num_k_block_main_loop = __builtin_amdgcn_readfirstlane(
                     (a_grid_desc_ak0_m_ak1.GetLength(I0) * a_grid_desc_ak0_m_ak1.GetLength(I2)) /
-                    KPerBlock); :AK0*KPadded/KPerBlock
+                    KPerBlock);
 
                 blockwise_gemm_pipeline.template Run<HasMainKBlockLoop, TailNum>(
                     a_grid_desc_ak0_m_ak1,