resolve libstdc++/5198

[Richard Henderson]

On Thu, Jan 24, 2002 at 08:03:17PM +0100, Roman Zippel wrote:
> add.l is only atomic local to the cpu, so that would be only a problem
> for m68k smp systems. I heard of rumors that such machines would exist,
> but does gcc support them?

As much as we support any other smp system.  That is, the compiler
couldn't care less.  

Anyway, m68k is freaky in saving so much state on faults, I didn't
know if it would save intermediate results when faulting in the 
middle of an indirect memory thing.


r~