Return-Path: Received: from mail.linutronix.de (146.0.238.70:993) by crypto-ml.lab.linutronix.de with IMAP4-SSL for ; 07 Feb 2019 22:45:56 -0000 Received: from smtp.eu.citrix.com ([185.25.65.24]) by Galois.linutronix.de with esmtps (TLS1.2:DHE_RSA_AES_256_CBC_SHA256:256) (Exim 4.80) (envelope-from ) id 1grsQc-0005S5-QH for speck@linutronix.de; Thu, 07 Feb 2019 23:45:55 +0100 Subject: [MODERATED] Re: [PATCH v2 0/8] PERFv2 References: <20190207132709.GA32477@hirez.programming.kicks-ass.net> <20190207143759.GP31598@tassilo.jf.intel.com> <20190207153728.GE32477@hirez.programming.kicks-ass.net> <2875fb7c-1354-c401-a7e5-6cb2cb02b40c@citrix.com> <20190207170059.GD32511@hirez.programming.kicks-ass.net> <20190207220450.GU31598@tassilo.jf.intel.com> From: Andrew Cooper Message-ID: Date: Thu, 7 Feb 2019 22:45:45 +0000 MIME-Version: 1.0 In-Reply-To: <20190207220450.GU31598@tassilo.jf.intel.com> Content-Type: text/plain; charset="windows-1252" Content-Transfer-Encoding: quoted-printable Content-Language: en-GB To: speck@linutronix.de List-ID: On 07/02/2019 22:04, speck for Andi Kleen wrote: >>> This feature and MSR are not architectural, and not expected to exist on >>> future hardware with a silicon fix for the issue (CascadeLake B1 >>> stepping and later, IIRC). > If you really want to migrate you could just make it a noop. After > all the other CPUs have all counters always working. I suppose you could, but which hypervisor is going to want to carry a compatibility workaround forever more just so an errata workaround for a single uarch can be given to the guest kernel for control. > > >>> PMU deliberately >>> isn't available to guests in general, > That's a sad attitude, but luckily less and less true in real deployments. If you say so.=A0 I'll let others judge the irony of that claim, considering the current track record of how easy it is to infer data across a privilege boundary. ~Andrew