On Sat, Jun 10, 2017 at 09:59:19 -0700, Richard Henderson wrote:
> On 06/10/2017 01:51 AM, Alex Bennée wrote:
> >I would agree with Emilio that we revert but I can't quite shake the
> >feeling we are missing an underlying problem. Would just skipping the
> >htable lookup (but keeping the tb_jmp_cache) be an OK fix for now?
>
> I agree. It seems like there's some real problem that this is uncovering.
>
> Dropping the htable lookup is certainly ok by me. If that's enough to
> un-stick your regression testing for aarch64 guest.
Unfortunately not checking the htable isn't enough to fix it for me.
I spent a few hours on Friday trying (and failing) to understand
what's going on. Regarding interrupts, I see *more* TB_EXIT_INTERRUPT
exits than without the patch, so my current best guess is that this is
a livelock kind of thing.
I'll keep digging, but regardless I think we should revert
e75449a346 ("target/aarch64: optimize indirect branches") ASAP, and
then reenable the feature ("revert the revert") once we get to the
bottom of this.
E.
