Paolo Bonzini <pbonz...@redhat.com> writes:
> On 30/10/2018 13:21, Alex Bennée wrote:
>>
>> Paolo Bonzini <pbonz...@redhat.com> writes:
>>
>>> This is a race that can happen when migrating TCG guests under load.
>>> It was introduced by the change to run vCPUs outside the big QEMU
>>> lock.
>>
>> Did this ever get re-spun?
>
> No, the issue is that *without MTTCG* all work items must come from the
> same list, otherwise you get a deadlock. However the cleanup part of
> the series was applied.
I say there was:
commit a8efa60633575a2ee4dbf807a71cb44d44b0e0f8
Author: Paolo Bonzini <pbonz...@redhat.com>
Date: Wed Nov 14 12:36:57 2018 +0100
cpus: run work items for all vCPUs if single-threaded
Was this an effort to fix this problem?
--
Alex Bennée
