On Mon, Sep 14, 2026 at 12:58:41AM -0400, Daniel Dickman wrote: > > > > On Sep 14, 2026, at 12:38 AM, Theo Buehler <[email protected]> wrote: > > > > On Mon, Sep 14, 2026 at 12:32:44AM -0400, Daniel Dickman wrote: > >> > >> > >>>> On Sep 14, 2026, at 12:07 AM, Theo Buehler <[email protected]> wrote: > >>> > >>> On Sun, Sep 13, 2026 at 10:49:38PM -0400, Daniel Dickman wrote: > >>>> The reproducer did what you said on an IBT amd64 box and I confirm the > >>>> SIGILL. > >>> > >>> Thanks for testing. > >>> > >>>> But the same reproducer did not SIGILL for me on my BT aarch64 laptop. > >>>> Does it need a different reproducer to trigger it on that platform? > >>> > >>> Apparently. There's clearly bad code with an obvious fix. Does it matter? > >> > >> I’m not totally following your thought process here. What is the diff > >> fixing? > > > > Bad code. > > I’m not totally sure. I think we’d need to check if the callers are direct or > indirect. > > If they are all direct then I think this diff may introduce new landing pads > that weren’t there before and it broadens the BTI target set unnecessarily. > > I suggest that should be checked first. What do you think?
I think doing this is a waste of time and you're asking for too much. ruff cranks out a release every week so today's analysis will be wrong tomorrow. > > Anyway that’s why I was checking for a reproducer first. > > Hope that helps…
