Not really. The register allocator is shooting itself in the foot.
4.5/4.6 may be better but I think the work around should be applied
to resolve the problem.
Dave
> Any update on this ?
>
> I'm still running into the issue with gcc 4.4.4 and linux 2.6.37-rc1.
>
> Regards,
> Guy
>
> On Sun, 7 Feb 2010 22:20:34 -0500 (EST)
> "John David Anglin" <dave@xxxxxxxxxxxxxxxxxx> wrote:
>
> > > On Mon, Dec 07, 2009 at 10:01:14PM -0500, John David Anglin wrote:
> > > > - volatile unsigned int *a;
> > > > + volatile register unsigned int *a asm ("r20") =
> > > > __ldcw_align(x);
> > >
> > > I'm reticent about moving the assignment around the mb(), should I
> > > edit this to return the assignment and apply, or apply as is?
> >
> > I don't believe the __ldcw_align operation involves a memory read or
> > write. It's just adjusting a pointer. Thus moving it past the mb()
> > shouldn't matter. That said, I been wrong before...
> >
> > Dave
>
--
J. David Anglin dave.anglin@xxxxxxxxxxxxxx
National Research Council of Canada (613) 990-0752 (FAX: 952-6602)
--
To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
