GCC regression "native build" failure reason changed

[Geoff Keating]

Hans-Peter Nilsson <hp@bitrange.com> writes:

> Not sure if you noticed, but the current badly regressed state
> regarding "native" build from head has changed reason.  It used
> to complain about multiple symbol definitions in libstdc++ but
> now it says:
> 
> checking for exception model to use... configure: error: unable to detect exception model
> make[1]: *** [configure-target-libstdc++-v3] Error 1
> 
> IIRC that *could* be a binutils problem.  (I had a bug with that
> symptom.)

This is the generic "your C++ compiler doesn't work" result.

It started with
<http://gcc.gnu.org/ml/gcc-regression/2003-03/msg00034.html>.  I've
listed in the CC: all the authors whose patch could possibly have
caused the problem.

-- 
- Geoffrey Keating <geoffk@geoffk.org>