summaryrefslogtreecommitdiff
path: root/lib/libc/softfloat/bits64/softfloat.c
AgeCommit message (Collapse)Author
2016-03-29Avoid warnings (signed/unsigned comparision and unused variable)martin
2013-11-22Fix a cast from the lint cleanup that made small exponents (i.e. values < 1)martin
sign extend wrong and overflow, causing an underflow in all 128 bit sqrt calculations.
2013-01-10Add hooks to softfloat to support per-lwp exception state.matt
2012-03-24Fix a bug introduced by lint cleanup.matt
2012-03-21cast to appropriate types.christos
2012-03-20Remove initialized but unused variablemartin
2011-07-10Fix so that float128_to_uint64_round_to_zero is defined for sparc64.matt
(isn't tha a namespace violation?)
2011-07-09Add SOFTFLOAT_NEED_FIXUNS condition around the float128 fixunsmatt
2011-07-04Add __floatunsidf __floatunsisf __floatunsitf routines.matt
XXX i think they are correct but not sure.
2007-11-08When creating unsigned int results, do not distinguish positive versusmartin
negative overflow.
2007-11-08When converting long double values to integer types, explicitly use themartin
"round to zero" variants of the softfloat conversion functions. Add a variant to convert long double to unsigned long - the "to long" variant checked for overflows that do not apply to unsigned results. This fixes the regress/lib/libc/convfp tests for sparc64.
2006-05-11add128() wants bits64 * as it's 5th argument.mrg
2003-07-26netbsd.org->NetBSD.orgsalo
2002-05-21Import the 32-bit version of SoftFloat 2a and separate it and the 64-bitbjh21
version into different directories. Which version a given port uses is controlled by the SOFTFLOAT_BITS make variable. This is set to 64 (which uses the same code we had before) by default. 32-bit platforms that don't need extended precision support might get better performance by using 32. Set the ARM port to use the 32-bit version of SoftFloat, since this is more than a factor of two faster than the 64-bit version. This should get the floating-point performance back to what it was in 1.5.