OMP_ATOMIC expand/gimplify changes

[Diego Novillo]

Zdenek Dvorak wrote:

>> I don't think this is terribly important for now, but we should think
>> about a better expansion strategy for the future.  Any ideas?  For
>> now, we could just add a FIXME here for future reference.
> 
> I thought about this too; unfortunately, as expansion needs to work
> also without SSA form, this is somewhat complicated.

That's fine, actually.  If we are not in SSA form, we'll likely not have
done heavy optimization and it'll be unlikely that new statements have
been inserted in the middle.  So, we really only want the smart
expansion when we are in SSA form.

> We use this function to set up the virtual operands of OMP_LOAD, as it
> has the address of the memory reference as an operand, not the memory
> reference itself (the same as OMP_ATOMIC).

Ah, right, thanks.