On Tue, Jan 22, 2013 at 03:30:06PM +0100, Richard Biener wrote:
> Marek Polacek <[email protected]> wrote:
>
> >We ICEd on attached testcase because we didn't properly detected the
> >latch edge. Furthermore, I think the code for re-computing latches
> >is somehow broken at the moment. In fix_loop_structure we have
> > /* If there was no latch, schedule the loop for removal. */
> > if (!first_latch)
> > loop->header = NULL;
>
> Are you sure this works when we hit this, thus
> Remove the loop? Otherwise you are right, the current place looks bogus.
I guess it should work--in the very next moment after zapping the
header we'd call delete_loop on that loop. But...
> Ok if you can convince yourself that this is
> The case.
... I will prepare another fix which Zdenek likes more ;). Thanks,
Marek
