From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754086Ab3APOfa (ORCPT ); Wed, 16 Jan 2013 09:35:30 -0500 Received: from youngberry.canonical.com ([91.189.89.112]:35067 "EHLO youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751914Ab3APOf3 (ORCPT ); Wed, 16 Jan 2013 09:35:29 -0500 Message-ID: <50F6BA8C.1070709@canonical.com> Date: Wed, 16 Jan 2013 15:34:52 +0100 From: Stefan Bader User-Agent: Mozilla/5.0 (X11; Linux i686; rv:17.0) Gecko/20130105 Thunderbird/17.0.2 MIME-Version: 1.0 To: Jan Beulich CC: Konrad Rzeszutek Wilk , Borislav Petkov , Andre Przywara , xen-devel , Matthew Garrett , "Rafael J. Wysocki" , Linux Kernel Mailing List Subject: Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0 References: <50F42B3E.7090602@canonical.com> <20130114163445.GA4867@liondog.tnic> <20130115175305.GA12449@phenom.dumpdata.com> <50F68E4902000078000B61AC@nat28.tlf.novell.com> In-Reply-To: <50F68E4902000078000B61AC@nat28.tlf.novell.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/16/2013 11:26 AM, Jan Beulich wrote: >>>> On 15.01.13 at 18:53, Konrad Rzeszutek Wilk wrote: >> On Mon, Jan 14, 2013 at 05:34:45PM +0100, Borislav Petkov wrote: >>> On Mon, Jan 14, 2013 at 04:58:54PM +0100, Stefan Bader wrote: >>>> --- a/drivers/acpi/processor_perflib.c >>>> +++ b/drivers/acpi/processor_perflib.c >>>> @@ -340,6 +340,9 @@ static void amd_fixup_frequency(struct acpi_processor_px *px >>>> if ((boot_cpu_data.x86 == 0x10 && boot_cpu_data.x86_model < 10) >>>> || boot_cpu_data.x86 == 0x11) { >>>> rdmsr(MSR_AMD_PSTATE_DEF_BASE + index, lo, hi); >>>> + /* Bit 63 indicates whether contents are valid */ >>>> + if (!(hi & 0x8000000)) >>>> + return; >>> >>> I don't think that's the right change - this is fixing baremetal so that >>> it works on xen. And besides, this code was in powernow-k8 before so I'm >>> wondering why did it work then. >> >> Powernow-k8 only populated the cpufreq policy information. This library >> (processor_perflib) is the generic library used for ACPI P-states parsing. >> This specific function (acpi_processor_get_performance_states) is just >> used to fetch and parse the P-states. >> >> Xen-acpi-processor (which we use to upload the P and C-states to the >> hypervisor) ends up calling this library to parse the P-states >> and this unfortunate quirk clamps the P-states based on the MSRS. >> >> It is odd that this CPU specific quirk got added in this generic >> library. Is there no ACPI quirk system similar to how DMI quirks >> are handled? >> >> Anyhow, I think this patch makes sense - it makes sure that the >> MSR value is sane. >> >> Reviewed-by: Konrad Rzeszutek Wilk > > Did someone actually _test_ that patch? I ask because the mask > used (0x8000000) doesn't check bit 63 as the comment says, but > bit 59 instead... > > Jan > Not this version which I did at the end of a day to have a cleaned up version to discuss. And obviously I managed to get the number of zeros wrong. :( The comment is right and it should be 0x80000000