From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.linutronix.de (146.0.238.70:993) by crypto-ml.lab.linutronix.de with IMAP4-SSL for ; 31 Jul 2018 08:22:17 -0000 Received: from mx3-rdu2.redhat.com ([66.187.233.73] helo=mx1.redhat.com) by Galois.linutronix.de with esmtps (TLS1.2:DHE_RSA_AES_256_CBC_SHA256:256) (Exim 4.80) (envelope-from ) id 1fkPv3-0007al-Cr for speck@linutronix.de; Tue, 31 Jul 2018 10:22:13 +0200 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 5C3F940241C9 for ; Tue, 31 Jul 2018 08:22:07 +0000 (UTC) Received: from [10.36.117.153] (ovpn-117-153.ams2.redhat.com [10.36.117.153]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 8BB9E215670D for ; Tue, 31 Jul 2018 08:22:06 +0000 (UTC) Subject: [MODERATED] Re: [PATCH v2 2/4] L1TF KVM ARCH_CAPABILITIES #2 References: <20180725143100.16309-1-pbonzini@redhat.com> <20180725143100.16309-3-pbonzini@redhat.com> From: Paolo Bonzini Message-ID: <0483e08d-519b-ab68-11b6-c324ba88e8c0@redhat.com> Date: Tue, 31 Jul 2018 10:22:04 +0200 MIME-Version: 1.0 In-Reply-To: Content-Type: multipart/mixed; boundary="gxwwAPedUHf3tOxfBEtec7KB256vw8oKV"; protected-headers="v1" To: speck@linutronix.de List-ID: This is an OpenPGP/MIME encrypted message (RFC 4880 and 3156) --gxwwAPedUHf3tOxfBEtec7KB256vw8oKV Content-Type: text/plain; charset=windows-1252 Content-Language: en-US Content-Transfer-Encoding: quoted-printable On 30/07/2018 23:27, speck for Thomas Gleixner wrote: > On Wed, 25 Jul 2018, speck for Paolo Bonzini wrote: >=20 >> From: Paolo Bonzini >> Subject: [PATCH v2 2/4] x86: SMT doesn't matter for VMX L1TF if EPT di= sabled or >> mitigation disabled >> >> If EPT is disabled, L1TF cannot be exploited even across threads on th= e >> same core, and SMT is irrelevant. >=20 > Ack. >=20 >> If mitigation is completely disabled, L1TF can be exploited even withi= n a >> single thread, so SMT is again irrelevant. >=20 > I'm not sure about this one. You might decide that the risk of not flus= hing > is acceptable if SMT is off, but then if SMT is enabled the whole thing= > might be less acceptable. So keeping that information in the sysfs file= > makes some sense. No strong opionion though. Fair enough, I'll change it to still do "VMX: SMT disabled, L1D vulnerable". What do you think actually about switching the order to "VMX: xxx, SMT yyy"? Paolo --gxwwAPedUHf3tOxfBEtec7KB256vw8oKV--