c++/9868: [3.3/3.4 regression] ICE in c_expand_expr when qualifying member variable

[Wolfgang Bangerth]

> > Correct, at least close. I had a problem with 9524 before, and still have:
> > it doesn't ICE on mainline for me.
> 
> Ahh, yes. I forgot.  It would be interesting to know in which way our
> compilers differ, that even the frontend behaves differently.  Hmm.

Valid question. I am building straight from CVS, on a SuSE 8 box. Are you 
using additional SuSE patches?

At any rate, 9868 seems to be a superset of the problem in 9524, so should 
9868 be fixed, this would likely cure 9524 as well, and for both of us.

W.

-------------------------------------------------------------------------
Wolfgang Bangerth             email:            bangerth@ticam.utexas.edu
                              www: http://www.ticam.utexas.edu/~bangerth/