On Wed, 2012-10-10 at 09:23 +1100, Paul Mackerras wrote:
> On Wed, Oct 10, 2012 at 01:20:37AM +1100, Michael Ellerman wrote:
> > We have over 15 routines that implement essentially the same logic
> > in terms of catching faults. Pull the logic out into two helper
> > routines.
> >
> > The pattern becomes:
> >
> > if (start_bus_error_jump() == 0) {
> > <do potentially faulting things>
> >
> > end_bus_error_jump();
> > } else {
> > printf("Faulting thing faulted!\n");
> > }
>
> NAK... think about how setjmp() works. You can't return from the
> function that called setjmp() while there is any possibility that
> longjmp() could be called for the same jmp_buf.
Sigh, yeah. Somehow I'd convinced myself that it worked in this case,
but after a night's sleep I'm not sure what my logic was.
cheers
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
