Re: [Xen-devel] [PATCH] xl: remove stale comment

[George Dunlap <dunlapg@xxxxxxxxx>]

On Thu, Feb 2, 2017 at 4:07 PM, Ian Jackson <ian.jackson@xxxxxxxxxxxxx> wrote:
> Wei Liu writes ("Re: [PATCH] xl: remove stale comment"):
>> On Thu, Feb 02, 2017 at 03:56:03PM +0000, Ian Jackson wrote:
>> > Wei Liu writes ("[PATCH] xl: remove stale comment"):
>> > >          case LIBXL_EVENT_TYPE_DISK_EJECT:
>> > > -            /* XXX what is this for? */
>> > >              libxl_cdrom_insert(ctx, domid, &event->u.disk_eject.disk, 
>> > > NULL);
>> > >              break;
>> >
>> > ... why does xl respond to ejection of removeable media by calling
>> > libxl_cdrom_insert with an empty medium ?
>>
>> Oh, so maybe early days there is only one type of removable medium --
>> the cdrom -- but things have changed now?
>
> No, that's not my point.
>
>> Anyway, I don't know much about the history, just curious why this
>> comment is here.
>
> The comment is there because I discovered when making libxl eventy
> that xl responded to disk eject events by re-ejecting the disk and
> didn't undrstand why.  I still don't understand why.
>
> Ie, why does xl have this code at all ?  Why is its handling of
> LIBXL_EVENT_TYPE_DISK_EJECT not to simply ignore it ?  Why does it
> even bother listening for this event ?

Is it possible that this is a disk eject *request* event from the guest?

I can't seem to figure out where these events come from.

  -George

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel