From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gleb Natapov Subject: Re: cpufreq and QEMU guests Date: Mon, 16 Sep 2013 18:32:39 +0300 Message-ID: <20130916153239.GD906@redhat.com> References: <20130916121545.GH5105@irqsave.net> <8668D877-8B37-48E3-97B8-CE36DB884E54@suse.de> <20130916150544.GJ5105@irqsave.net> Mime-Version: 1.0 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <20130916150544.GJ5105@irqsave.net> Sender: cpufreq-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="utf-8" To: =?utf-8?Q?Beno=C3=AEt?= Canet Cc: Alexander Graf , "cpufreq@vger.kernel.org" , "qemu-devel@nongnu.org" , "rjw@sisk.pl" , "viresh.kumar@linaro.org" , "peter.maydell@linaro.org" , "pbonzini@redhat.com" On Mon, Sep 16, 2013 at 05:05:45PM +0200, Beno=C3=AEt Canet wrote: > Le Monday 16 Sep 2013 =C3=A0 09:39:10 (-0500), Alexander Graf a =C3=A9= crit : > >=20 > >=20 > > Am 16.09.2013 um 07:15 schrieb Beno=C3=AEt Canet : > >=20 > > >=20 > > > Hello, > > >=20 > > > I know a cloud provider worried about the fact that the /proc/cpu= info of his > > > guests give a bogus frequency to his customer. > > >=20 > > > QEMU and the guests kernel currently have no way to reflect the h= ost frequency > > > changes to the guests. > > >=20 > > > The customer compute intensive application then read this informa= tion and take > > > wrong decisions. > >=20 > > Why do they care about the frequency? Is it for scheduling workload= s? The only other case I can think of would be the TSC and that should = be fixed frequency these days. > >=20 > > If it's scheduling, you could maybe expose the unavailable compute = time as steal time to the guest. Exposibg frequency in a virtual enviro= nment feels backwards. >=20 > The final customer have a compute intensive workload. > At startup the code retrieve the cpu cache topology, the cpu model, a= nd various > informations including the guest cpu frequency before starting the co= mpute job. > The QEMU instance typicaly use -cpu host. >=20 > The code inspects the cpu frequency has seen by the guests to choose = the number > of vms to instanciate to compute the given task. I am not sure I understand. They look at guest cpu frequency to estimat= e guest's performance? > They even destroy and recreate some vms that would be underperforming= to > mitigate the high inter vm communication costs. >=20 > Do you think the steal time trick would work for this ? >=20 -- Gleb.