|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-devel] [Minios-devel] [PATCH MINI-OS v2] xenbus: notify the other end when necessary
Wei Liu, le Mon 26 Oct 2015 13:55:31 +0000, a écrit :
> On Mon, Oct 26, 2015 at 01:52:47PM +0100, Samuel Thibault wrote:
> > Wei Liu, le Mon 26 Oct 2015 12:47:56 +0000, a écrit :
> > > - if (xenstore_buf->rsp_prod - xenstore_buf->rsp_cons <
> > > sizeof(msg))
> > > + if (xenstore_buf->rsp_prod - xenstore_buf->rsp_cons <
> > > sizeof(msg)) {
> > > + notify_remote_via_evtchn(start_info.store_evtchn);
> > > break;
> > > + }
> >
> > > if (xenstore_buf->rsp_prod - xenstore_buf->rsp_cons <
> > > - sizeof(msg) + msg.len)
> > > + sizeof(msg) + msg.len) {
> > > + notify_remote_via_evtchn(start_info.store_evtchn);
> >
> > Actually thinking... In principle we should not need these two
> > notifications: we have already notified last time we consumed a message.
> > Notifying again shouldn't be bringing anything new. Do you actually see
> > a difference with these?
> >
>
> Yes. The ring still gets stalled somehow without those two
> notifications.
Ok... This is still very worrying. The more I'm thinking about it,
the more I believe we shouldn't *have* to do this. Which version of
the xenstore are you testing with, exactly? I guess that while we
investigate this oddness we can apply this fix to stable trees, so as to
be safer than sorry.
Samuel
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel
|
![]() |
Lists.xenproject.org is hosted with RackSpace, monitoring our |