C99 complex numbers in cygwin?

Billinghurst, David (CALCRTS) david.billinghurst@comalco.riotinto.com.au
Sun Jul 18 20:15:00 GMT 2004


> From: dgun@umpire.com 
>> From: "Billinghurst, David (CALCRTS)"
>> > gcc uses the complex math functions from the system libraries,
>> > (excluding builtins).  They aren't in newlib, so cygwin 
>> doesn't have them.
>> > 
>> > I, too, would like them as they are required by gfortran.
>> which will be 
>> > (is) the fortran compiler in gcc-3.5.  
>> 
>> UPDATE: This just in...
>> 
>> http://gcc.gnu.org/onlinedocs/gcc-3.4.1/gcc/Other-Builtins.htm
>l#Other%20Builtins
>
>"The ISO C99 functions ... cabsf, cabsl, cabs, cacosf, cacoshf, cacoshl,
>cacosh, cacosl, cacos, cargf, cargl, carg, casinf, casinhf, casinhl,
>casinh, casinl, casin, catanf, catanhf, catanhl, catanh, catanl, catan,
>cbrtf, cbrtl, cbrt, ccosf, ccoshf, ccoshl, ccosh, ccosl, ccos, cexpf,
>cexpl, cexp, cimagf, cimagl, cimag, conjf, conjl, conj,..., cpowf,
>cpowl, cpow, cprojf, cprojl, cproj, crealf, creall, creal, csinf, csinhf,
>csinhl, csinh, csinl, csin, csqrtf, csqrtl, csqrt, ctanf, ctanhf, ctanhl,
>ctanh, ctanl, ctan ... are handled as built-in functions except in strict
>ISO C90 mode (-ansi or -std=c89)."
>
>Sounds like I just need to wait for gcc 3.4.  Is there a build of it
>available for cygwin yet?

3.4 is not enough.  You still need library support for most of the math 
functions.

I got working float and double functions from Stephen Moshier's code
http://www.moshier.net in a few hours coding last night.  Seems to work
with gcc CVS - accurate to one or two bits.  Need some polishing though.

 



More information about the Newlib mailing list