From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: [RFC][PATCH] qemu-kvm: Introduce writeback scope for cpu_synchronize_state Date: Tue, 17 Nov 2009 17:50:12 +0100 Message-ID: <4B02D444.6080402@web.de> References: <4B018542.3020602@siemens.com> <4B01A487.3020808@redhat.com> <4B01C2B0.3000205@web.de> <4B02592C.6060004@redhat.com> <4B025B50.4070505@web.de> <4B0260D7.1060107@redhat.com> <4B026A03.4080600@web.de> <4B0298F0.3080007@redhat.com> <4B029FA8.5080205@web.de> <4B02A4FD.4010802@redhat.com> <4B02AF58.4010407@web.de> <4B02B252.5080207@redhat.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enigAF7F562A33036648B88C6CCB" Cc: Marcelo Tosatti , kvm , Gleb Natapov To: Avi Kivity Return-path: Received: from fmmailgate03.web.de ([217.72.192.234]:41382 "EHLO fmmailgate03.web.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752955AbZKQQuJ (ORCPT ); Tue, 17 Nov 2009 11:50:09 -0500 In-Reply-To: <4B02B252.5080207@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigAF7F562A33036648B88C6CCB Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Avi Kivity wrote: > On 11/17/2009 04:12 PM, Jan Kiszka wrote: >>>> The alternative would be a complex get&lock/put&unlock + a queue for= >>>> async events during the lock + an option to ignore what was queued w= hen >>>> doing a true reset. Back to square #1: we would still need the propo= sed >>>> high-level interface to communicate the difference between replay an= d >>>> drop queue. >>>> =20 >>>> =20 >>> There's no need for get+lock / put+unlock; a normal get/put with the >>> =20 >> You need to track when to queue and when to apply directly. Call it lo= ck >> or call it something else. >> =20 >=20 > You always queue. When starting vcpu_run() or reading state to > userspace you flush the queue. Now I finally got your idea. >=20 > The hardware equivalent is posting APIC messages, and the core executin= g > them. >=20 >>> addition that get flushes the queue suffices. To make sure queued >>> events don't affect set you need to stop the entire VM before setting= >>> state, but you need to do that anyway for non-rmw writes. >>> >>> =20 >> Well, sounds good, but it will be a non-trivial change in the interfac= e >> semantics. At bare minimum, we would need a new mp_state interface. If= >> we would count mp_state to our new event structure (hmm...), then we >> could confine the semantical changes to that new IOCTL pair. But how t= o >> deal with existing KVM kernels with their mp_state interface? It's a b= it >> like the vcpu state thing: we are already down a specific road, and it= 's >> hard to turn around. >> =20 >=20 > I think we're not on the same page here. As I see it, no interface > change is needed at all. >=20 > It's true that existing kernels don't handle this properly, which is wh= y > I said I'm willing to treat it as a bug (and thus the -stable treatment= > etc.). I admit it's a stretch since this is not going to be trivial > (though I think less complex that you believe). >=20 > Putting mp_state into the events structure is reasonable regardless of > this issue (and doable since we haven't pushed it to 2.6.33 yet). But = I > want to understand why you think it's needed. >=20 That wouldn't be required anymore with the "always queue" policy. But what would you queue at all? Only mp_state, nmi_pending and sipi_vector? Or also all the relevant PIC and LAPIC states that might be changed asynchronously? Jan --------------enigAF7F562A33036648B88C6CCB Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org iEYEARECAAYFAksC1EUACgkQitSsb3rl5xRi2ACgiCcURvrdMsPHKstR9A4ClYN6 ihIAnR2gYfA3eD/WAbi2IgdXpIbV+Nfr =ibTc -----END PGP SIGNATURE----- --------------enigAF7F562A33036648B88C6CCB--