From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stefan Bader Subject: Re: nVMX regression v3.13+, bisected Date: Wed, 26 Feb 2014 21:27:17 +0100 Message-ID: <530E4E25.4050508@canonical.com> References: <530E43EC.7000600@canonical.com> <530E4DB9.5050001@redhat.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha512; protocol="application/pgp-signature"; boundary="Wu80RgAAVLI1h4gpiUu9UlbgebX0kjSiO" Cc: Anthoine Bourgeois To: Paolo Bonzini , kvm@vger.kernel.org Return-path: Received: from youngberry.canonical.com ([91.189.89.112]:53282 "EHLO youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752245AbaBZU1V (ORCPT ); Wed, 26 Feb 2014 15:27:21 -0500 In-Reply-To: <530E4DB9.5050001@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --Wu80RgAAVLI1h4gpiUu9UlbgebX0kjSiO Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: quoted-printable On 26.02.2014 21:25, Paolo Bonzini wrote: > Il 26/02/2014 20:43, Stefan Bader ha scritto: >> Hi, >> >> I was looking at a bug report[1] about a regression on nested VMX that= started >> with kernel v3.13 (same issue still existed with v3.14-rc4). The probl= em shows >> up when running a v3.13 kernel in L0 and then trying to launch a L2 (L= 1 was >> either a v3.2 kernel or v3.13, so seemed to have no immediate influenc= e). L2 is >> trying to boot a iso image and hangs before the isolinux boot loader d= isplays >> anything. A preinstalled hd image fails to boot, too. >> >> I bisected this and ended up on the following commit which, when rever= ted made >> the launch work again: >> >> Author: Anthoine Bourgeois >> Date: Wed Nov 13 11:45:37 2013 +0100 >> >> kvm, vmx: Fix lazy FPU on nested guest >> >> If a nested guest does a NM fault but its CR0 doesn't contain the = TS >> flag (because it was already cleared by the guest with L1 aid) the= n we >> have to activate FPU ourselves in L0 and then continue to L2. If T= S flag >> is set then we fallback on the previous behavior, forward the faul= t to >> L1 if it asked for. >> >> Signed-off-by: Anthoine Bourgeois >> Signed-off-by: Paolo Bonzini >> >> The condition to exit to L0 seems to be according to what the descript= ion says. >> Could it be that the handling in L0 is doing something wrong? >=20 > Thanks, I'll look at it tomorrow or Friday. >=20 > Paolo >=20 Great thanks. And maybe it helps if I actually add the link to the bug re= port as I had intended... :-P [1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1278531 --Wu80RgAAVLI1h4gpiUu9UlbgebX0kjSiO Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBCgAGBQJTDk4lAAoJEOhnXe7L7s6jX6wP/2TxWQ4d17wll8hksNU496S0 XikbpVcbVOwqvPg/uPuLsVwAnz/TpCPZXQQz90hl862irbkiHMoH+JDTw/dck9jg eqHNqpuGEuM5/2WvyQw54JqlZmj/ovEolibYhC+NMnFrm7RfMMfmJkEBJ8wQn9cr P1/NurZ55k9xckhocH98yi8TRL3xtotpQRlDSYwhtB2Vvp62WYzd1pnV7KfNEn6f 80Lt7Di/thHLghsj41Q7IpU5GP3C7w6aODwfSXSt/tuao/m0bnjXAgX7aTB7o9fk PgVaQOEKnr+NIxhfAuXxWRbkIjInJBnlQPNgF5Og1KzOyCHnkp897cbpWz+aeJQQ Ve1f0oavfr/o7bn0AzbFw0AX7phKqoALQ4jhkl1iZwm5qnFtshfxj2l8m1Ch+5MF 17gi3mhPcsY+aMSxGltmQNrzGYdY5reur9Wg8bdmwA9x/7D5GUCZj53RzxsfqSvg Nr6FF0hraqywi8nvgrPqLZvIGws405pC6DEMFb0jOZsOH/Cc4N+k+BXqZvEz6t+l anlGwsf2CS14hb3I1GTW+pZNn7L6IiL7+EvTMNNfi6zyW5/yegAOaGz/m8lpVQUT eQbHJlPGUSm2rZ1sw6iCSm4ZZfdt635OJpjPikqZAj12M37xbn3aDKtvQLe71qr6 mgbphXaX/iXFsZd9r0ZK =KrMZ -----END PGP SIGNATURE----- --Wu80RgAAVLI1h4gpiUu9UlbgebX0kjSiO--