math: fix sinh overflows in non-nearest rounding

The final rounding operation should be done with the correct sign
otherwise huge results may incorrectly get rounded to or away from
infinity in upward or downward rounding modes.

This affected sinh and sinhf which set the sign on the result after
a potentially overflowing mul. There may be other non-nearest rounding
issues, but this was a known long standing issue with large ulp error
(depending on how ulp is defined near infinity).

The fix should have no effect on sinh and sinhf performance but may
have a tiny effect on cosh and coshf.

1 files changed, 1 insertions, 1 deletions

diff --git a/src/math/sinh.c b/src/math/sinh.c
index 00022c4e..a01951ae 100644
--- a/src/math/sinh.c
+++ b/src/math/sinh.c
@@ -34,6 +34,6 @@ double sinh(double x)
 
 	/* |x| > log(DBL_MAX) or nan */
 	/* note: the result is stored to handle overflow */
-	t = 2*h*__expo2(absx);
+	t = __expo2(absx, 2*h);
 	return t;
 }