From: Daniel Arndt <arndtd@ornl.gov>
Date: Mon, 3 Apr 2023 20:43:56 +0000 (-0400)
Subject: Improve LinearAlgebra::parallel::Vector documentation
X-Git-Tag: v9.5.0-rc1~378^2
X-Git-Url: https://gitweb.dealii.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=refs%2Fpull%2F15014%2Fhead;p=dealii.git

Improve LinearAlgebra::parallel::Vector documentation
---

diff --git a/include/deal.II/lac/la_parallel_vector.h b/include/deal.II/lac/la_parallel_vector.h
index cb2d74d36c..afd9a24bcd 100644
--- a/include/deal.II/lac/la_parallel_vector.h
+++ b/include/deal.II/lac/la_parallel_vector.h
@@ -188,14 +188,13 @@ namespace LinearAlgebra
      * <h4>GPU support</h4>
      *
      * This vector class supports two different memory spaces: Host and Default.
-     * By default, the memory space is Host and all the data is allocated on the
-     * CPU. When the memory space is Default, all the data is allocated on
-     * Kokkos' default memory space. That means that if Kokkos was configured
-     * with a GPU backend, the data is allocated on a GPU. The operations on the
-     * vector are performed on the chosen memory space. From the host, there are
-     * two methods to access the elements of the Vector when using the Default
-     * memory space:
-     * <ul>
+     * If the MemorySpace template argument is not specified, the memory space
+     * is Host and all the data is allocated on the CPU. When the memory space
+     * is Default, all the data is allocated on Kokkos' default memory space.
+     * That means that if Kokkos was configured with a GPU backend, the data is
+     * allocated on a GPU. The operations on the vector are performed on the
+     * chosen memory space. From the host, there are two methods to access the
+     * elements of the Vector when using the Default memory space: <ul>
      * <li> use get_values():
      * @code
      * Vector<double, MemorySpace::Default> vector(local_range, comm);