summaryrefslogtreecommitdiffstats
path: root/lib/libm
AgeCommit message (Expand)AuthorFilesLines
2016-03-09We are done providing support for the vax.deraadt2-32/+2
2015-12-18Fix wrong answer if the imaginary part is zero.shadchin3-3/+9
2015-12-18Fix behavior csqrt, should beshadchin3-6/+6
2015-11-10update NAME section to include all documented functions,jmc2-6/+10
2015-11-06From FreeBSD 23397:shadchin1-7/+7
2015-09-18Move s_signgam.c to PURE_SRCS, so that it also gets built on vax.miod1-3/+3
2015-09-14use .Va for global variables, and .Vt where the type is includedschwarze1-9/+9
2015-08-20Avoid left-shifting a negative integer, which is undefined. Replacenaddy1-3/+3
2015-08-12fix a case where ceill() returns 1.0L: in the x86 extended precision formatnaddy1-1/+1
2015-07-19Make exponents of x and y signed and fix esx and esymartynas1-4/+5
2015-07-16Adapt previous to extended-precision, and fix single-precision constants.martynas1-2/+2
2015-07-16Signs of cacosh/cacoshf were not always correct (e.g., -1.1 -1.1i),martynas2-4/+4
2015-01-22Fix logic botch causing warnings with Clang. Reported by dhill, matchesmiod2-12/+12
2015-01-22Possibly uninitialized variable. From Clang via dhill.miod1-1/+1
2015-01-20Missing $OpenBSD$'s.krw13-0/+13
2015-01-15remove .Tn; from Jan Stary <hans at stare dot cz>schwarze5-107/+47
2014-09-12Provide a sparc64 version of sqrtl(3) that simply calls _Qp_sqrt.kettenis2-1/+31
2014-07-21Switch from <sys/endian.h> or <machine/endian.h> to the new,guenther3-8/+6
2014-06-05Extended precision floating-point numbers have an explicit integer part bit.kettenis1-11/+10
2014-06-02Make sure STRICT_ASSIGN handles double as well. From FreeBSD, where thekettenis1-2/+2
2014-04-18It's been a quarter century: we can assume volatile is present with that name.guenther28-322/+322
2014-03-18Retire hp300, mvme68k and mvme88k ports. These ports have no users, keepingmiod60-2702/+2
2014-02-11Make floor round to integral value less than x, when x is tiny.martynas2-6/+8
2013-11-22unsigned char casts for ctype macrosderaadt1-3/+3
2013-11-12Trick the compiler that volatile temporary is used while raisingmartynas3-0/+12
2013-11-12Declare __p1evll and __polevll protos in math_private.h instead ofmartynas10-27/+30
2013-11-12Add parentheses around comparison in operand of ^. Pointed out by theo.martynas2-10/+10
2013-11-12Add explicit braces to avoid ambiguous else. Pointed out by theo.martynas10-18/+27
2013-11-03NaN -> \*(Namartynas7-23/+23
2013-08-14no longer any need to quote macro lines with >9 args;jmc3-9/+9
2013-08-13Switch time_t, ino_t, clock_t, and struct kevent's ident and dataguenther1-1/+1
2013-08-07Document lgamma_r, lgammaf_r, and lgammal_r.millert1-2/+35
2013-08-07simple inaccurate implementation of lrint* functions on vax, just soespie2-2/+58
2013-08-06scalbnf on vax, okay miod@espie1-1/+8
2013-08-05tgamma(+-0) == +-Infkettenis1-1/+3
2013-08-02Fix a couple of corner cases in the implementation of pow(3) to make itkettenis3-9/+15
2013-07-17more library historyschwarze11-40/+85
2013-07-15Fix atan2f() and hypotf() by actually converting both their arguments to themiod2-4/+6
2013-07-15use some generic code on vax, to do round/roundf/scalbln...espie3-83/+5
2013-07-15use atan2f to add atanfespie1-1/+8
2013-07-15bring reference C code that's not compiled in synch with asm codeespie6-6/+50
2013-07-15add some missing asm functions to vax: long double as aliases to double,espie6-10/+51
2013-07-05VAX ELF userland bits. Consists mostly of register prefix additions.miod9-490/+490
2013-07-05... and gain rintf by using the same technique, along with aliases forespie1-1/+24
2013-07-05implement floorf and ceilf: same technique as floor and ceil, just withespie1-2/+34
2013-07-03reorganize NOIEEE_SRCS so that files not compiled on vax becauseespie3-78/+12
2013-07-03with DBL vs LDBL fixed, tedu some source in noieee which is actually theespie30-1777/+25
2013-07-03test LDBL_MANT_DIG == DBL_MANT_DIG instead of hardcoding 53.espie69-171/+171
2013-06-05use fancy .In macro for includes. from Jan Klemkow. ok jmc schwarzetedu55-166/+166
2013-03-28Switch libc and libm to use strong aliases rather than weak aliasesmartynas95-156/+156