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 ; 12 Dec 2018 14:02:38 -0000 Received: from mail-wm1-f65.google.com ([209.85.128.65]) by Galois.linutronix.de with esmtps (TLS1.2:RSA_AES_128_CBC_SHA1:128) (Exim 4.80) (envelope-from ) id 1gX55w-0003cW-Lr for speck@linutronix.de; Wed, 12 Dec 2018 15:02:36 +0100 Received: by mail-wm1-f65.google.com with SMTP id a18so6106036wmj.1 for ; Wed, 12 Dec 2018 06:02:35 -0800 (PST) Received: from ?IPv6:2001:b07:6468:f312:7020:93c3:970b:67cb? ([2001:b07:6468:f312:7020:93c3:970b:67cb]) by smtp.gmail.com with ESMTPSA id o9sm2738903wmh.3.2018.12.12.06.02.29 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 12 Dec 2018 06:02:29 -0800 (PST) Subject: [MODERATED] Re: [PATCH v2 3/8] MDSv2 5 References: <0d6a3fbe4c511152a0f5350e62e9e09ec545f709.1544464266.git.ak@linux.intel.com> <20181211000303.GB16024@tassilo.jf.intel.com> <20181211032503.GB25620@tassilo.jf.intel.com> From: Paolo Bonzini Message-ID: Date: Wed, 12 Dec 2018 15:02:23 +0100 MIME-Version: 1.0 In-Reply-To: <20181211032503.GB25620@tassilo.jf.intel.com> Content-Type: multipart/mixed; boundary="ZqAl4uMclFAubhRwouawEpgqZLCo2MtSS"; protected-headers="v1" To: speck@linutronix.de List-ID: This is an OpenPGP/MIME encrypted message (RFC 4880 and 3156) --ZqAl4uMclFAubhRwouawEpgqZLCo2MtSS Content-Type: text/plain; charset=windows-1252 Content-Language: en-US Content-Transfer-Encoding: quoted-printable On 11/12/18 04:25, speck for Andi Kleen wrote: > Okay. >=20 > We'll have to disable interrupts in that window, but that > should be ok. >=20 > I guess it could be also set in some skb_* functions to catch > the network cases. >=20 > Or maybe setting when calling into unaudited code in interrupts/timers?= >=20 > And the flag would allow a "paranoid" mode which sets > it unconditionally.=20 That sounds a lot like kvm_get_cpu_l1tf_flush_l1d and kvm_clear_cpu_l1tf_flush_l1d. Maybe we can just add another bit in irq_cpustat_t's kvm_cpu_l1tf_flush_l1d, which would be cleared on return to userspace before doing the verw stuff. Paolo --ZqAl4uMclFAubhRwouawEpgqZLCo2MtSS--