Fix compilation error

Author: EnricoDeg

example/65_gemm_multiply_multiply/gemm_multiply_multiply_wmma_fp8_ab_scale.cpp

@@ -143,26 +143,26 @@ int main(int argc, char* argv[])
 
             if(std::is_same<decltype(layout), ck::tensor_layout::gemm::RowMajor>::value)
             {
-                return HostTensorDescriptor({row, col}, {stride, 1_uz});
+                return ck::HostTensorDescriptor({row, col}, {stride, 1_uz});
             }
             else
             {
-                return HostTensorDescriptor({row, col}, {1_uz, stride});
+                return ck::HostTensorDescriptor({row, col}, {1_uz, stride});
             }
         };
 
-    Tensor<A0DataType> a0_m_k(f_host_tensor_descriptor(M, K, StrideA, A0Layout{}));
-    Tensor<A1DataType> a1_m_k(f_host_tensor_descriptor((M + Scale_Block_M - 1) / Scale_Block_M,
-                                                       (K + Scale_Block_K - 1) / Scale_Block_K,
-                                                       Scale_Stride_AM,
-                                                       A0Layout{}));
-    Tensor<B0DataType> b0_k_n(f_host_tensor_descriptor(K, N, StrideB, B0Layout{}));
-    Tensor<B1DataType> b1_k_n(f_host_tensor_descriptor((K + Scale_Block_K - 1) / Scale_Block_K,
-                                                       (N + Scale_Block_N - 1) / Scale_Block_N,
-                                                       Scale_Stride_BN,
-                                                       B0Layout{}));
-    Tensor<EDataType> e_m_n_host_result(f_host_tensor_descriptor(M, N, StrideE, ELayout{}));
-    Tensor<EDataType> e_m_n_device_result(f_host_tensor_descriptor(M, N, StrideE, ELayout{}));
+    ck::Tensor<A0DataType> a0_m_k(f_host_tensor_descriptor(M, K, StrideA, A0Layout{}));
+    ck::Tensor<A1DataType> a1_m_k(f_host_tensor_descriptor((M + Scale_Block_M - 1) / Scale_Block_M,
+                                                           (K + Scale_Block_K - 1) / Scale_Block_K,
+                                                           Scale_Stride_AM,
+                                                           A0Layout{}));
+    ck::Tensor<B0DataType> b0_k_n(f_host_tensor_descriptor(K, N, StrideB, B0Layout{}));
+    ck::Tensor<B1DataType> b1_k_n(f_host_tensor_descriptor((K + Scale_Block_K - 1) / Scale_Block_K,
+                                                           (N + Scale_Block_N - 1) / Scale_Block_N,
+                                                           Scale_Stride_BN,
+                                                           B0Layout{}));
+    ck::Tensor<EDataType> e_m_n_host_result(f_host_tensor_descriptor(M, N, StrideE, ELayout{}));
+    ck::Tensor<EDataType> e_m_n_device_result(f_host_tensor_descriptor(M, N, StrideE, ELayout{}));
 
     std::cout << "a0_m_k: " << a0_m_k.mDesc << std::endl;
     std::cout << "a1_m_k: " << a1_m_k.mDesc << std::endl;
@@ -210,11 +210,11 @@ int main(int argc, char* argv[])
         b1_k_n.GenerateTensorValue(GeneratorTensor_3<B1DataType>{0, 1.0});
     }
 
-    DeviceMem a0_device_buf(sizeof(A0DataType) * a0_m_k.mDesc.GetElementSpaceSize());
-    DeviceMem a1_device_buf(sizeof(A1DataType) * a1_m_k.mDesc.GetElementSpaceSize());
-    DeviceMem b0_device_buf(sizeof(B0DataType) * b0_k_n.mDesc.GetElementSpaceSize());
-    DeviceMem b1_device_buf(sizeof(B1DataType) * b1_k_n.mDesc.GetElementSpaceSize());
-    DeviceMem e_device_buf(sizeof(EDataType) * e_m_n_device_result.mDesc.GetElementSpaceSize());
+    ck::DeviceMem a0_device_buf(sizeof(A0DataType) * a0_m_k.mDesc.GetElementSpaceSize());
+    ck::DeviceMem a1_device_buf(sizeof(A1DataType) * a1_m_k.mDesc.GetElementSpaceSize());
+    ck::DeviceMem b0_device_buf(sizeof(B0DataType) * b0_k_n.mDesc.GetElementSpaceSize());
+    ck::DeviceMem b1_device_buf(sizeof(B1DataType) * b1_k_n.mDesc.GetElementSpaceSize());
+    ck::DeviceMem e_device_buf(sizeof(EDataType) * e_m_n_device_result.mDesc.GetElementSpaceSize());
 
     a0_device_buf.ToDevice(a0_m_k.mData.data());
     a1_device_buf.ToDevice(a1_m_k.mData.data());
@@ -269,9 +269,9 @@ int main(int argc, char* argv[])
 
     if(do_verification)
     {
-        Tensor<AccDataType> c_m_n({M, N});
-        Tensor<float> a_m_k({M, K});
-        Tensor<float> b_k_n({K, N});
+        ck::Tensor<AccDataType> c_m_n({M, N});
+        ck::Tensor<float> a_m_k({M, K});
+        ck::Tensor<float> b_k_n({K, N});
 
         for(int m = 0; m < M; m++)
         {