On Sun, 2006-09-24 at 23:25 -0700, Jeremy Fitzhardinge wrote:
> Rusty Russell wrote:
> > I don't think so. There's *never* address subtraction, there's
> > sometimes 32 bit wrap (glibc uses this to effect subtraction, sure).
> > But there's no wrap here.
> >
> Hm, I guess, so long as you assume the kernel data segment is always
> below the kernel heap.
Agreed, we should BUG_ON() in case anyone ever changes this... I will
create a patch for this...
> > To test, I changed the following:
> >
> > --- smpboot.c.~8~ 2006-09-25 15:51:50.000000000 +1000
> > +++ smpboot.c 2006-09-25 16:00:36.000000000 +1000
> > @@ -926,8 +926,9 @@
> > unsigned long per_cpu_off)
> > {
> > unsigned limit, flags;
> > + extern char __per_cpu_end[];
> >
> > - limit = (1 << 20);
> > + limit = PAGE_ALIGN((long)__per_cpu_end) >> PAGE_SHIFT;
> >
> limit is a size, rather than the end address, so this isn't quite right.
I think it's OK. For every "%gs:var", var will be less than
__per_cpu_end.
Thanks!
Rusty.
--
Help! Save Australia from the worst of the DMCA: http://linux.org.au/law
