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, 2 insertions, 2 deletions

diff --git a/src/internal/libm.h b/src/internal/libm.h
index b5bd26b8..7533f6ba 100644
--- a/src/internal/libm.h
+++ b/src/internal/libm.h
@@ -236,13 +236,13 @@ hidden int    __rem_pio2(double,double*);
 hidden double __sin(double,double,int);
 hidden double __cos(double,double);
 hidden double __tan(double,double,int);
-hidden double __expo2(double);
+hidden double __expo2(double,double);
 
 hidden int    __rem_pio2f(float,double*);
 hidden float  __sindf(double);
 hidden float  __cosdf(double);
 hidden float  __tandf(double,int);
-hidden float  __expo2f(float);
+hidden float  __expo2f(float,float);
 
 hidden int __rem_pio2l(long double, long double *);
 hidden long double __sinl(long double, long double, int);