Applied patch 3 by Ondřej Bílka with some manual corrections,

fixing typos:
  https://sympa.inria.fr/sympa/arc/mpfr/2013-07/msg00016.html

git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@8621 280ebfd0-de03-0410-8827-d642c229c3f4

2 files changed, 2 insertions, 2 deletions

diff --git a/tune/bidimensional_sample.c b/tune/bidimensional_sample.c
index 47aec1f2f..db9c28997 100644
--- a/tune/bidimensional_sample.c
+++ b/tune/bidimensional_sample.c
@@ -32,7 +32,7 @@ http://www.gnu.org/licenses/ or write to the Free Software Foundation, Inc.,
 
 /* Let f be a function for which we have several implementations f1, f2... */
 /* We wish to have a quick overview of which implementation is the best    */
-/* in function of the point x where f(x) is computed and of the prectision */
+/* in function of the point x where f(x) is computed and of the precision */
 /* prec requested by the user.                                             */
 /* This is performed by drawing a 2D graphic with color indicating which   */
 /* method is the best.                                                     */
diff --git a/tune/tuneup.c b/tune/tuneup.c
index 181ec7336..058d3c4b1 100644
--- a/tune/tuneup.c
+++ b/tune/tuneup.c
@@ -562,7 +562,7 @@ tune_simple_func (mpfr_prec_t *threshold,
    It assumes that for (x,p) close to zero, algo1 is used
    and algo2 is used when (x,p) is far from zero.
    If algo2 is better for low prec, and algo1 better for high prec,
-   the behaviour of this function is undefined.
+   the behavior of this function is undefined.
    This tuning function tries couples (x,p) of the form (ell*dirx, ell*dirp)
    until it finds a point on the boundary. It returns ell.
  */