From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PART1 RFC v2 07/10] svm: Add VMEXIT handlers for AVIC Date: Wed, 9 Mar 2016 11:56:58 +0100 Message-ID: <56E0017A.2020201@redhat.com> References: <1457124368-2025-1-git-send-email-Suravee.Suthikulpanit@amd.com> <1457124368-2025-8-git-send-email-Suravee.Suthikulpanit@amd.com> <56DDA50B.7090609@redhat.com> <20160308220541.GA29455@potion.brq.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Suravee Suthikulpanit , joro@8bytes.org, bp@alien8.de, gleb@kernel.org, alex.williamson@redhat.com, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, wei@redhat.com, sherry.hurwitz@amd.com To: =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= Return-path: Received: from mx1.redhat.com ([209.132.183.28]:47129 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752214AbcCIK5D (ORCPT ); Wed, 9 Mar 2016 05:57:03 -0500 In-Reply-To: <20160308220541.GA29455@potion.brq.redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 08/03/2016 23:05, Radim Kr=C4=8Dm=C3=A1=C5=99 wrote: >>> >> + case AVIC_INCMP_IPI_ERR_INV_TARGET: >>> >> + pr_err("%s: Invalid IPI target (icr=3D%#08x:%08x, idx=3D%u)\n= ", >>> >> + __func__, icrh, icrl, index); >>> >> + BUG(); >>> >> + break; >>> >> + case AVIC_INCMP_IPI_ERR_INV_BK_PAGE: >>> >> + pr_err("%s: Invalid bk page (icr=3D%#08x:%08x, idx=3D%u)\n", >>> >> + __func__, icrh, icrl, index); >>> >> + BUG(); >>> >> + break; >> >=20 >> > Please use WARN(1, "%s: Invalid bk page (icr=3D%#08x:%08x, idx=3D%= u)\n", >> > __func__, icrh, icrl, index) (and likewise for invalid target) ins= tead >> > of BUG(). > I think that if we hit one of these, then WARNs would just flood the > log. I'd prefer WARN_ONCE on AVIC_INCMP_IPI_ERR_INV_BK_PAGE. > (Btw. aren't icr and idx are pointless on this error? and the functi= on > name should be printed by WARN.) Agreed. > Invalid target is triggerable by the guest (by sending IPI to a > non-existent LAPIC), so warning log level seems too severe. > pr_info_ratelimited() or nothing would be better. Definitely should be nothing. Paolo