On Thu, 2006-05-18 at 09:08 +0200, Gerd Hoffmann wrote:
> > Since the vsyscall page has been at a fixed address, there might be
> > real programs that hardcoded it etc.
>
> Exploits *grin*?
LOL.
> > @@ -147,6 +152,13 @@ void __set_fixmap (enum fixed_addresses
> > return;
> > }
> > set_pte_pfn(address, phys >> PAGE_SHIFT, flags);
> > + nr_fixmaps++;
> > +}
> > +
> > +void set_fixaddr_top(unsigned long top)
> > +{
> > + BUG_ON(nr_fixmaps > 0);
> > + __FIXADDR_TOP = top - PAGE_SIZE;
> > }
>
> Is this intentional?
>
> That are the bits which make the fixmap runtime-movable (by calling
> set_fixaddr_top early enougth at boot time) to make room for the
> hypervisor address space hole. Not used (yet).
>
> We surely want to have that in the long run. It's not mentioned in the
> patch description though. It also used to be a separate patch on top of
> the move-vsyscall-out-of-fixmap patch.
Yes, this patch is showing its origins here. I'll add it to the
description, though. For me, this is the point of the patch. It's just
that Ingo tricked me into pushing exec-shield 8)
Thanks,
Rusty.
--
ccontrol: http://ccontrol.ozlabs.org
