On Tue, 17 Oct 2023 at 10:38, Albert Esteve <aest...@redhat.com> wrote:
> On Tue, Oct 17, 2023 at 12:57 PM Stefan Hajnoczi <stefa...@gmail.com> wrote:
>>
>> On Tue, 17 Oct 2023 at 04:26, Albert Esteve <aest...@redhat.com> wrote:

Thanks for considering my feedback!

> There is not. In a first thought, I assumed that the backends will be in 
> charge
> of cleaning their entries from the shared hash table when they are destroyed
> (prematurely or no). I will look into occurrences of vhost_dev getting 
> destroyed
> that may need explicit handling of the leftover entries.

QEMU supports hot (un)plug of vhost-user devices. vhost-user backends
may also crash.

QEMU cannot rely solely on the backend to take any cleanup action
because a backend may be buggy or misbehave.

Stefan

Reply via email to