On 06/13, Andy Lutomirski wrote: > > On Mon, Jun 13, 2016 at 6:50 AM, Oleg Nesterov <o...@redhat.com> wrote: > > > > Well, I can't resist but to me SA_IA32_ABI/SA_X32_ABI looks like a hack > > too. We actually shift TIF_*32 into k_sigaction->flags, and the fact > > that we do this per-signal looks, well, interesting ;) > > Is anything actually wrong with this, though?
No, I think the patch is correct... It adds the user-visible change, but I do not think it can break something. Well, perhaps we should also remove SA_IA32_ABI|SA_X32_ABI from oact->sa.sa_flags in do_sigaction() to ensure that these non-uapi flags won't leak to user-space, but probably we do not really care. Oleg.