[pypy-dev] A bug in cc?

[Elmo Mäntynen]

Armin Rigo wrote:
> Hi Elmo,
>
> On Sun, Jul 23, 2006 at 09:43:05PM +0300, Elmo M?ntynen wrote:
>   
>> Now I got an internal compiler error from cc:
>>     
>
> Couldn't reproduce it...  I ran
>
>   python translate.py --backend=c --profopt='-c "from richards import
>   *;main(iterations=1)"' --text --batch targetpypystandalone.py
>
> but everything went fine.  It's a gcc 3.4.5 (Gentoo 3.4.5-r1,
> ssp-3.4.5-1.0, pie-8.7.9).  Either it's a bug in a specific gcc version,
> or else maybe (just guessing) the floating-point exception is when gcc
> computes something with the profile data, maybe timings, in which case
> the bug could be hard to reproduce on a different machine.
>
> In any case, if the bug is still there in the latest gcc, yes, I'd
> consider reporting it.  PyPy is good to push many limits of its backends
> -- e.g. it gave quite a few LLVM bug reports and I wouldn't be surpized
> if mono was next :-)
>
>
> A bientot,
>
> Armin
>   
I have 4.0.4 (debian 4.0.3-4). How should we go about testing on the
latest gcc, or are you doing it right now. I can definitely reproduce
the bug with the that certain file(s). Which of the .gcda and .gcno (or
the rest) files should I send?

Elmo