Build errors

[David Malcolm]

On Wed, 2013-10-23 at 20:31 -0700, Simon Feltman wrote:
> On Wed, Oct 23, 2013 at 9:07 AM, David Malcolm <dmalcolm@redhat.com> wrote:
> >> <snip>
> >> make: *** [all] Error 2

[...]

> The problem turned out to be debian multiarch shananagins and adding
> "--disable-multilib" to configure fixed everything up.

Thanks for tracking this down and posting the results.

> Perhaps the gccjit wiki could have a note that people running into
> build problems should consult the FAQ. Basically gccjit was my entry
> point into building gcc and it might be helpful for others going down
> a similar path (I'm also happy to add this kind of stuff to the wiki
> if you think it makes sense.)

Good point; sorry about not spelling this out.  I went ahead and added a
note about this to the wiki page:
  http://gcc.gnu.org/wiki/JIT?action=diff&rev2=14&rev1=13

[...]

> >> But I'm now hitting this when running test.py:
> >> $ python3 test.py
> >> .libgccjit.so: internal compiler error: Segmentation fault
> >> ...
> >> 0x7fad426c2485 gcc::jit::context::invoke_code_factory()
> >>         ../../src/gcc/jit/internal-api.c:1345
> >
> > I'm not seeing that (with Python 3.2 fwiw).
> 
> I just tried with with Python 2.7 and it was broken as well. Note that
> the first unittest "test_square" works well, only
> "test_sum_of_squares" is blowing up. I'll try to debug this a bit
> more.
> 
> > What version of the dmalcolm/jit branch are you using?   I added a lot
> > of error-checking last Friday in
> > d4019d9ddb77354c242798596268c7043bb7f3dd.
> 
> The clone was made yesterday (and I updated it again today).

That's unexpected.

You know this, but for the list archives, you can debug the python
bindings using:

  gdb --args python3 test.py

Hopefully you'll be able to get a backtrace of the segfault that way.
If it's no longer *segfaulting*, but instead merely bailing with an
"internal compiler error", try putting a breakpoint on the function
"internal_error".

Dave