Re: Strange CPP failures

[Neil Booth <neil at daikokuya dot co dot uk>]

Toon Moene wrote:-

> Neil Booth wrote:
> 
> > Sigh.  There's been a bunch of problems recently with GCC compiled
> > with earlier GCCs, but not with existing GCCs.
>        ^^^^^^^           ^^^^^^^^^^^^^^^^^
> 
> I can see how we sometimes have to say that we do not support "early"
> GCC's, but we do not have to resort to denying they exist, I hope ?

I hope not.  But I don't think we've really nailed what's causing
these issues.  They're kind of random, but they seem to affect CPP
rather than anything else.  So my suspicion rests with CPP.

But, it could just be a chance compiler/binutils bug that breaks
one thing in CPP that  shows up in various weird and wonderful
ways.  For example, 3.0.something miscompiles cppexp.c as of
about 6 weeks ago; Jason worked around that by rewriting one
or two lines of code.  The CPP code was valid regardless.

Neil.