summaryrefslogtreecommitdiffstats
path: root/sys/lib/libkern/softfloat-macros.h (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Override SoftFloat's countLeadingZeros32() with a faster, ff1-based, versionmiod2007-12-291-1/+3
| | | | on m88k.
* Unifdef NO_IEEE in the libkern softfloat code, and put an appropriate rulemiod2007-12-291-4/+28
| | | | | | in files.alpha to compensate for NO_IEEE kernels. This will allow the softfloat code to be used by other platforms than alpha.
* IEEE 754 floating point completion code, and implementation of thepvalchev2002-04-281-0/+753
FP_C (Floating Point Control Quadword). From ross@NetBSD. Added a way to disable it with option NO_IEEE, which appears on the ramdisks to save space. This affects only programs compiled with -mieee, and what it essentially does is enabling infinities and NaNs, instead of generating SIGFPE on division by zero, overflow, etc. ok art, deraadt