master
aarch64: add single instruction math functions
arm fabs and sqrt: support single-precision-only fpu variants
math: move x87-family fmod functions to C with inline asm
add m68k sqrtl using native instruction
mips: add single-instruction math functions
add SPE FPU support to powerpc-sf
powerpc64: add single-instruction math functions
riscv32: add fenv and math
add riscv64 architecture support
s390x: add single instruction fma and fmaf
math/x32: correct lrintl.s for 32-bit long
math: add x86_64 remquol
math: use double_t for temporaries to avoid stores on i386
math: long double trigonometric cleanup (cosl, sinl, sincosl, tanl)
math: fix sinh overflows in non-nearest rounding
math: cosmetic cleanup (use explicit union instead of fshape and dshape)
add support for m68k 80-bit long double variant
math: fix 128bit long double inverse trigonometric functions
apply hidden visibility to various remaining internal interfaces
math: add double precision error handling functions
math: add single precision error handling functions
math: add __math_invalidl
prune math code on archs with binary64 long double
math: fix __rem_pio2 in non-nearest rounding modes
math: clean up __rem_pio2
remove a fixme comment
always provide __fpclassifyl and __signbitl definitions
math: tan cleanups
math: use sqrtl if FLT_EVAL_METHOD==2 in acosh and acoshf
math: fix acoshf for negative inputs
math: fix ld80 acoshl(x) for x < 0
math: long double inverse trigonometric cleanup (acosl, asinl, atanl, atan2l)
math: fix asin, atan, log1p, tanh to raise underflow on subnormal
math: use 0x1p-120f and 0x1p120f for tiny and huge values
math: rewrite inverse hyperbolic functions to be simpler/smaller
math: add dummy implementations of 128 bit long double functions
math: clean up atan2.c
math: fix atanh (overflow and underflow issues)
fix unintended global symbols in atanl.c
math: cbrt cleanup and long double fix
math: use the rounding idiom consistently
math: rewrite rounding functions (ceil, floor, trunc, round, rint)
long double cleanup, initial commit
math: sin cos cleanup
math: new exp and exp2
reduce spurious inclusion of libc.h
math: new exp2f and expf
math: add ld128 exp2l based on the freebsd implementation
math: remove STRICT_ASSIGN macro
math: fix expm1f overflow threshold
remove libm.h includes when math.h and float.h are enough
math: add (obsolete) bsd drem and finite functions
math: fix fma(x,y,0) when x*y rounds to -0
math: fix fmaf not to depend on FE_TOWARDZERO
math: fix rare underflow issue in fma
math: rewrite remainder functions (remainder, remquo, fmod, modf)
math: fix fmodl for IEEE binary128
first commit of the new libm!
math: long double fix (use ldshape union)
math: explicitly promote expressions to excess-precision types
math: rewrite hypot
math: ilogb cleanup
fix threshold constants in j0f, y0f, j1f, y1f
math: bessel cleanup (jn.c and jnf.c)
move lgamma-related internal declarations to libm.h
make lrint and llrint functions work without fenv support
math: use '#pragma STDC FENV_ACCESS ON' when fenv is accessed
simplify lround and llround functions
math: new log
math: extensive log*.c cleanup
math: new log2
math: new log2f
math: fix logb(-0.0) in downward rounding mode
math: include missing float.h in logbl
math: fix undefined shift in logf
math: new logf
math: optimize lrint on 32bit targets
implement nan, nanf, nanl
math: new pow
math: new powf
remove incorrect comment regarding powl exceptional cases
fix rint.c and rintf.c when FLT_EVAL_METHOD!=0
math: define _GNU_SOURCE when implementing non-standard math functions
fix scalbn when result is in the subnormal range
add significand[fl] math functions
math: fix aliasing violation in long double wrappers
math: new software sqrt
math: new software sqrtf
sqrtl: fix invalid use of a non-constant-expression as static initializer
math: lgamma cleanup (simpler sin(pi*x) for the negative case)
math: add a non-dummy tgamma implementation