Hi,
I was wrong... Described problem with vhost-user is not possible.
It cannot enter tcp_chr_disconnect twice because of qemu global mutex (BQL).
Thus chardev event handler (main loop) and vhost_user_blk_set_status (VCPU) are
working under BQL.
30.10.2018, 20:12, "Marc-André Lureau" :
> Hi
>
> O
Hi
On Mon, Oct 29, 2018 at 5:35 PM Yury Kotov wrote:
>
>
>
> 29.10.2018, 09:46, "Marc-André Lureau" :
> > Hi
> >
> > On Mon, Oct 22, 2018 at 5:24 PM Yury Kotov
> > wrote:
> >
> >> Hi,
> >>
> >> I examined vhost-user devices and found some chardev using strangeness.
> >>
> >> Is it ok, that v
29.10.2018, 09:46, "Marc-André Lureau" :
> Hi
>
> On Mon, Oct 22, 2018 at 5:24 PM Yury Kotov wrote:
>
>> Hi,
>>
>> I examined vhost-user devices and found some chardev using strangeness.
>>
>> Is it ok, that vhost-user's set_status do sync chardev io ops from KVM
>> thread?
>> It seems tha
Hi
On Mon, Oct 22, 2018 at 5:24 PM Yury Kotov wrote:
>
> Hi,
>
> I examined vhost-user devices and found some chardev using strangeness.
>
> Is it ok, that vhost-user's set_status do sync chardev io ops from KVM thread?
> It seems that chardev doesn't support working with different threads.
>
> F
Hi,
I examined vhost-user devices and found some chardev using strangeness.
Is it ok, that vhost-user's set_status do sync chardev io ops from KVM thread?
It seems that chardev doesn't support working with different threads.
For example, I think such race is possible (two simultaneous events):
1
