summaryrefslogtreecommitdiff
path: root/src/math
AgeCommit message (Expand)AuthorFilesLines
2017-06-23s390x: add single-instruction math functionsRich Felker24-0/+360
2017-04-21fix scalbn when result is in the subnormal rangeSzabolcs Nagy3-12/+14
2017-03-21aarch64: add single instruction math functionsSzabolcs Nagy34-24/+226
2017-03-15fix threshold constants in j0f, y0f, j1f, y1fSzabolcs Nagy2-13/+12
2016-10-20math: fix pow signed shift ubSzabolcs Nagy1-2/+2
2016-08-30math: fix 128bit long double inverse trigonometric functionsSzabolcs Nagy1-1/+1
2016-03-04math: fix expf(-NAN) and exp2f(-NAN) to return -NAN instead of 0Szabolcs Nagy2-0/+4
2016-02-19work around regression building for armhf with clang (compiler bug)Rich Felker2-2/+2
2016-02-18improve macro logic for enabling arm math asmRich Felker2-2/+2
2016-01-20replace armhf math asm source files with inline asmRich Felker16-40/+60
2015-11-21math: explicitly promote expressions to excess-precision typesRich Felker3-4/+4
2015-11-10explicitly assemble all arm asm sources as UALRich Felker4-0/+4
2015-10-19declare fpu usage to the assembler in arm hard-float asm filesSzabolcs Nagy4-0/+4
2015-04-23fix regression in x86_64 math asm with old binutilsRich Felker2-6/+6
2015-04-18remove potentially PIC-incompatible relocations from x86_64 and x32 asmRich Felker2-2/+2
2015-04-18remove the last of possible-textrels from i386 asmRich Felker2-1/+5
2015-04-18math: fix pow(+-0,-inf) not to raise divbyzero flagSzabolcs Nagy3-3/+3
2015-03-11add aarch64 portSzabolcs Nagy4-0/+24
2015-03-11math: add dummy implementations of 128 bit long double functionsSzabolcs Nagy16-4/+97
2015-03-11math: add ld128 exp2l based on the freebsd implementationSzabolcs Nagy1-1/+366
2015-02-09math: fix fmodl for IEEE binary128Szabolcs Nagy1-1/+1
2015-02-08math: fix __fpclassifyl(-0.0) for IEEE binary128Szabolcs Nagy1-3/+2
2015-02-08add parenthesis in fma.c to clarify intent and silence warningsSzabolcs Nagy1-1/+1
2014-11-05math: use fnstsw consistently instead of fstsw in x87 asmSzabolcs Nagy11-11/+11
2014-11-05math: fix x86_64 and x32 asm not to use sahf instructionSzabolcs Nagy6-28/+14
2014-10-31math: use the rounding idiom consistentlySzabolcs Nagy13-58/+89
2014-10-31fix rint.c and rintf.c when FLT_EVAL_METHOD!=0Szabolcs Nagy2-4/+22
2014-10-08always provide __fpclassifyl and __signbitl definitionsRich Felker2-1/+9
2014-09-18math: fix exp10 not to raise invalid exception on NaNSzabolcs Nagy3-4/+13
2014-09-08fix exp10l.c to include float.hSzabolcs Nagy1-0/+1
2014-09-08prune math code on archs with binary64 long doubleSzabolcs Nagy2-0/+10
2014-04-11math: fix aliasing violation in long double wrappersSzabolcs Nagy2-2/+10
2014-02-23x32 port (diff against vanilla x86_64)rofl0r18-69/+69
2014-02-23import vanilla x86_64 code as x32rofl0r30-0/+396
2014-01-08math: add drem and dremf weak aliases to i386 remainder asmSzabolcs Nagy2-0/+6
2013-12-12math: define _GNU_SOURCE when implementing non-standard math functionsSzabolcs Nagy6-0/+6
2013-11-24math: clean up __rem_pio2Szabolcs Nagy3-71/+53
2013-11-21math: add (obsolete) bsd drem and finite functionsSzabolcs Nagy4-0/+20
2013-11-21math: lgamma cleanup (simpler sin(pi*x) for the negative case)Szabolcs Nagy4-202/+110
2013-10-28math: extensive log*.c cleanupSzabolcs Nagy14-583/+369
2013-10-07math: fix rare underflow issue in fmaSzabolcs Nagy3-13/+55
2013-10-07math: use sqrtl if FLT_EVAL_METHOD==2 in acosh and acoshfSzabolcs Nagy2-0/+13
2013-10-06math: remove an unused variable from modflSzabolcs Nagy1-1/+0
2013-10-04math: remove code duplication in erfl found by clang analyzerSzabolcs Nagy1-13/+2
2013-10-04math: remove a useless assignment in lgammal found by clang analyzerSzabolcs Nagy1-2/+2
2013-09-13fix x86_64 lrintl asm, againRich Felker1-2/+2
2013-09-06math: remove STRICT_ASSIGN from exp2f (see previous commit)Szabolcs Nagy1-1/+1
2013-09-06math: remove STRICT_ASSIGN macroSzabolcs Nagy10-12/+13
2013-09-05math: support invalid ld80 representations in fpclassifySzabolcs Nagy1-2/+4
2013-09-05math: fix atanh (overflow and underflow issues)Szabolcs Nagy3-14/+37