Re: multiple definition of `atan2l' on the current Cygwin

[Tatsuro MATSUOKA <tmacchant2@yahoo.co.jp>]

> From: Corinna Vinschen >
> To: cygwin> Cc: JonY
> Date: 2016/4/14, Thu 22:52
> Subject: Re: multiple definition of `atan2l' on the current Cygwin
> 
> On Apr 14 19:00, Takashi Yano wrote:
>>  Hi Corinna,
>> 
>>  On Thu, 14 Apr 2016 09:52:05 +0200
>>  Corinna Vinschen wrote:
>>  > If it's that simple, why does my STC (or a simple variation 
> thereof) not
>>  > complain?
>> 
>>  It is because optimizer removes the function call.
>> 
>>  Try:
>> 
>>  $ cat > m.cc <<EOF
>>  #include <cstdio>
>>  #include <cmath>
>> 
>>  int main()
>>  {
>>    long double x, y;
>>    scanf("%Lf %Lf", &x, &y);
>>    printf("%Lf\n", atan2l(x,y));
>>    return 0;
>>  }
>>  EOF
>>  $ g++ m.cc -lm
> 
> Thanks!  So that's why my testcase didn't work :}
> 
> This is actually a problem in libstdc++-v3.  On platforms missing
> certain long double functions libstdc++-v3 is built with stub functions
> using the same name.
> 
> Jon, any chance to rebuild the gcc packages ASAP?
> 
> 
> Thanks,
> Corinna
I have confirmed the mutiple definition error does not occur on gnuplot build with 
gcc-5.3.0-4 today.
Thanks!
Tatsuro


--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple