Re: overridden optional reload may cause a needed reload to bedropped

[Bernd Schmidt <bernds at redhat dot co dot uk>]

On 26 Oct 2000, Alexandre Oliva wrote:

> On Oct 26, 2000, Bernd Schmidt <bernds@redhat.com> wrote:
> 
> > Apparently we lose the contents of reload_override_in at some point;
> 
> We don't.  It's just that we decide to not reload it because the
> reload is optional.

Hmmm.
The point of optional reloads is that they get done exactly in those cases
where we can inherit them.  In this case, we can't use the inherited reg
as the reload register itself, but we can use it as a better input (that's
what reload_override_in means).  It may be that the inheritance code can't
handle the combination of optional reload with reload_override_in, although
I'd be surprised if that was really the problem - the usual feeling of "it
should have shown up long ago".

> Or at least that was my understanding of the
> problem.  I may point you to the testcase, if you want to play with
> it.

Please do.


Bernd