Since "abs" truncates any value in (-1..1) to 0, the tests below would
succeed for any value in a much larger than intended range: (-0.5, 1.5).
The corrected (stricter) tests still pass on at least a few targets,
so I've pushed this change:
test-strtod: fix typos: s/abs/fabs/
* tests/test-strtod.c (main): Use fabs, not narrowing-to-int "abs".
diff --git a/tests/test-strtod.c b/tests/test-strtod.c
index 6c63e13..12d3606 100644
--- a/tests/test-strtod.c
+++ b/tests/test-strtod.c
@@ -163,7 +163,7 @@ main ()
/* FIXME - gnulib's version is rather inaccurate. It would be
nice to guarantee an exact result, but for now, we settle for a
1-ulp error. */
- ASSERT (abs (result - 0.5) < DBL_EPSILON);
+ ASSERT (fabs (result - 0.5) < DBL_EPSILON);
ASSERT (ptr == input + 2);
ASSERT (errno == 0);
}
@@ -246,7 +246,7 @@ main ()
/* FIXME - gnulib's version is rather inaccurate. It would be
nice to guarantee an exact result, but for now, we settle for a
1-ulp error. */
- ASSERT (abs (result - 0.5) < DBL_EPSILON);
+ ASSERT (fabs (result - 0.5) < DBL_EPSILON);
ASSERT (ptr == input + 4);
ASSERT (errno == 0);
}
--
1.5.5.50.gab781
