Bruce Evans 8b438ea8dd Use double precision to simplify and optimize arg reduction for small ... and medium size args too: instead of conditionally subtracting a float 17+24, 17+17+24 or 17+17+17+24 bit approximation to pi/2, always subtract a double 33+53 bit one. The float version is now closer to the double version than to old versions of itself -- it uses the same 33+53 bit approximation as the simplest cases in the double version, and where the float version had to switch to the slow general case at |x| == 2^7*pi/2, it now switches at |x| == 2^19*pi/2 the same as the double version. This speeds up arg reduction by a factor of 2 for |x| between 3*pi/4 and 2^7*pi/4, and by a factor of 7 for |x| between 2^7*pi/4 and 2^19*pi/4.		2005-10-29 16:34:50 +00:00
..
alpha	Replace fegetmask() and fesetmask() with feenableexcept(),	2005-03-16 19:03:46 +00:00
amd64	Add a missing ldexpf() alias for amd64.	2005-09-12 20:54:00 +00:00
arm	Replace fegetmask() and fesetmask() with feenableexcept(),	2005-03-16 19:03:46 +00:00
bsdsrc	Fixed aliasing bugs in TRUNC() by using the fdlibm macros for access	2005-09-19 11:28:19 +00:00
i387	Prevent these functions from using stack outside of their frame.	2005-05-06 15:44:20 +00:00
ia64	Replace fegetmask() and fesetmask() with feenableexcept(),	2005-03-16 19:03:46 +00:00
man	Markup nit.	2005-06-16 21:56:03 +00:00
powerpc	Replace fegetmask() and fesetmask() with feenableexcept(),	2005-03-16 19:03:46 +00:00
sparc64	Replace fegetmask() and fesetmask() with feenableexcept(),	2005-03-16 19:03:46 +00:00
src	Use double precision to simplify and optimize arg reduction for small	2005-10-29 16:34:50 +00:00
Makefile	Bump the shared library version number of all libraries that have not	2005-07-22 17:19:05 +00:00