Update BatchNormKernels.cpp

Author: xytintel

src/ATen/native/xpu/sycl/BatchNormKernels.cpp

@@ -1287,7 +1287,7 @@ struct BatchNormTransformInputVectorizedKernelFunctor {
     } else {
       invstd =
           static_cast<stat_accscalar_t>(1) /
-          device_sqrt(
+          std::sqrt(
               static_cast<stat_accscalar_t>(var_or_invstd_[plane]) + epsilon_);
     }
 
@@ -1456,27 +1456,13 @@ void batch_norm_elemt_template(
   nwg_y = std::min<int>(nwg_y, syclMaxWorkItemsPerTile() / (tf * tb));
   sycl::range<2> global_range(nwg_y * tb, nwg_x * tf);
 
-  auto output_ptr = (char*)output_reshaped.data_ptr();
-  if (output_reshaped.is_contiguous() &&
-      memory::can_vectorize_up_to<input_scalar_t>(output_ptr) >= 4 &&
-      sizeof(input_scalar_t) < sizeof(float)) {
-    auto kfn = BatchNormTransformInputVectorizedKernelFunctor<
-        4,
-        input_scalar_t,
-        stat_scalar_t,
-        stat_accscalar_t,
-        true,
-        index_t>(input, output, mean, invstd, weight, bias, dummy_epsilon);
-    sycl_kernel_submit(global_range, local_range, queue, kfn);
-  } else {
-    auto kfn = BatchNormTransformInputKernelFunctor<
-        input_scalar_t,
-        stat_scalar_t,
-        stat_accscalar_t,
-        true,
-        index_t>(input, output, mean, invstd, weight, bias, dummy_epsilon);
-    sycl_kernel_submit(global_range, local_range, queue, kfn);
-  }
+  auto kfn = BatchNormTransformInputKernelFunctor<
+      input_scalar_t,
+      stat_scalar_t,
+      stat_accscalar_t,
+      true,
+      index_t>(input, output, mean, invstd, weight, bias, dummy_epsilon);
+  sycl_kernel_submit(global_range, local_range, queue, kfn);
 }
 
 template <