From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:46060) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SrV4b-0004S8-P5 for qemu-devel@nongnu.org; Wed, 18 Jul 2012 10:17:54 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1SrV4a-0000No-4p for qemu-devel@nongnu.org; Wed, 18 Jul 2012 10:17:53 -0400 Received: from mx1.redhat.com ([209.132.183.28]:51459) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SrV4Z-0000Lo-Sn for qemu-devel@nongnu.org; Wed, 18 Jul 2012 10:17:52 -0400 Message-ID: <5006C58C.5060500@redhat.com> Date: Wed, 18 Jul 2012 08:17:48 -0600 From: Eric Blake MIME-Version: 1.0 References: <20120718110226.23a50a7c@doriath.home> In-Reply-To: <20120718110226.23a50a7c@doriath.home> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="------------enig18DB9BE1FABE30F5269D793E" Subject: Re: [Qemu-devel] QMP event for suspend to disk List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Luiz Capitulino Cc: qemu-devel , Anthony Liguori This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig18DB9BE1FABE30F5269D793E Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable On 07/18/2012 08:02 AM, Luiz Capitulino wrote: > Hi, >=20 > QEMU can now distinguish between S4 and power off for the guest OSes th= at > suspends to disk through S4. This means that we can have a QMP event fo= r S4 > (so that QMP clients can distinguish between S3 and S4). >=20 > However, as we already emit the SHUTDOWN event for this, it turns out t= hat > there are three different ways of supporting the new event. We could: >=20 > 1. not emit the SHUTDOWN event and only emit the SUSPEND_DISK event >=20 > cons: breaks compatibility, as we'll stop sending an event we send = today. At least the new 'query-events' monitor command will allow libvirt to know which events to listen for; so this is workable. >=20 > 2. emit both, that is, first SUSPEND_DISK and then SHUTDOWN >=20 > cons: emitting two very similar events. This has the most back-compat, but is noisier in that it generates more events. >=20 > 3. extend the SHUTDOWN event to say whether it's a suspend to disk or = not, > like this: >=20 > { "event": "SHUTDOWN", "data": { "suspend-to-disk": true }, ... = } >=20 > cons: we already have the SUSPEND event, so from an API perspective= > it would make sense to have SUSPEND_DISK. >=20 > I prefer item 3, because it seems to be cleanest one. I'm also liking 3, then 1, then 2; but think any of them will work from libvirt's POV. --=20 Eric Blake eblake@redhat.com +1-919-301-3266 Libvirt virtualization library http://libvirt.org --------------enig18DB9BE1FABE30F5269D793E Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) Comment: Public key at http://people.redhat.com/eblake/eblake.gpg Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBCAAGBQJQBsWMAAoJEKeha0olJ0Nq+4YIAJZJ0vabTui1tAzVMSy7ZalD MfN4ke9WtYjsCSsOvndTsRJLA7FFkIZm5zmlu9nJ5gDC+p/7gQZuS79ope5EKAGr lAY69BUXuauhUjyE+BfL8gdfkpNpCOMYUaFssV69ilMLyFS0oW/DV+meVyXOnnGr VJumh3C1fAOWyp2bCwtlTxHAUT4XuPKAXqPYfKbwEz39qp6kZdCOrDcmcPDKF8hx nS0qE2AxFeaS9C2x2lob3HkmeGKhBa/YE0Yb3zGMck84VrwsHz53oJa4N3I8PG+W oR32f7KvDRBgqWXV6jSL1dNFK0ZF8QComTp2OFvRACTw8VCrJFnJkYzD5aqwhPw= =xGYC -----END PGP SIGNATURE----- --------------enig18DB9BE1FABE30F5269D793E--