[Bug c++/23139] -pedantic -ffast-math breaks working code

2005-07-29 Thread thor at math dot tu-berlin dot de
--- Additional Comments From thor at math dot tu-berlin dot de 2005-07-29 21:18 --- Subject: Re: -pedantic -ffast-math breaks working code pinskia at gcc dot gnu dot org wrote: > --- Additional Comments From pinskia at gcc dot gnu dot org 2005-07-29 > 21:06 --- > Can you atta

[Bug c++/23139] -pedantic -ffast-math breaks working code

2005-07-29 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-07-29 21:18 --- Hmm, if passed in 4.0.0 20050225 but fails on the mainline. It also works with the C front-end. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23139

[Bug c++/23139] -pedantic -ffast-math breaks working code

2005-07-29 Thread thor at math dot tu-berlin dot de
--- Additional Comments From thor at math dot tu-berlin dot de 2005-07-29 21:14 --- As per request, here is the preprocessed source code: int main(int argc,char **argv) { double v = (__extension__ 0x1.0p2047); printf("%f\n",v); return 0; } -- http://gcc.gnu.org/bugzilla/sho

[Bug c++/23139] -pedantic -ffast-math breaks working code

2005-07-29 Thread pinskia at gcc dot gnu dot org
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-07-29 21:06 --- Can you attach the preprocessing source as the glibc I have installed defineds HUGE_VAL as __builtin_huge_val(). -- What|Removed |Added