Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-22 Thread Rafael J. Wysocki
On Monday, January 21, 2013 04:31:05 PM Borislav Petkov wrote: > On Mon, Jan 21, 2013 at 03:03:43PM +, Stefan Bader wrote: > > From 9870926d4a847e36c0f61921762fd50f1c92f75d Mon Sep 17 00:00:00 2001 > > From: Stefan Bader > > Date: Mon, 14 Jan 2013 16:17:00 +0100 > > Subject: [PATCH] ACPI: Chec

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Boris Ostrovsky
On 01/18/2013 02:00 PM, Konrad Rzeszutek Wilk wrote: So something like this in the hypervisor maybe (not even tested): diff --git a/xen/arch/x86/acpi/cpufreq/powernow.c b/xen/arch/x86/acpi/cpufreq/powernow.c index a9b7792..54e7808 100644 --- a/xen/arch/x86/acpi/cpufreq/powernow.c +++ b/xen/a

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Borislav Petkov
On Mon, Jan 21, 2013 at 03:03:43PM +, Stefan Bader wrote: > From 9870926d4a847e36c0f61921762fd50f1c92f75d Mon Sep 17 00:00:00 2001 > From: Stefan Bader > Date: Mon, 14 Jan 2013 16:17:00 +0100 > Subject: [PATCH] ACPI: Check MSR valid bit before using P-state frequencies > > To fix incorrect P-

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Stefan Bader
Ok, so how about this? -Stefan From 9870926d4a847e36c0f61921762fd50f1c92f75d Mon Sep 17 00:00:00 2001 From: Stefan Bader Date: Mon, 14 Jan 2013 16:17:00 +0100 Subject: [PATCH] ACPI: Check MSR valid bit before using P-state frequencies To fix incorrect P-state frequencies which can happen on so

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Stefan Bader
On 01/21/2013 12:42 PM, Borislav Petkov wrote: On Mon, Jan 21, 2013 at 12:22:18PM +, Stefan Bader wrote: So for having the "check for sensible BIOS" in mainline I refreshed the patch (fixed the bit test, and actually tested it this time) and also added some hopefully sensible explanation to

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Borislav Petkov
On Mon, Jan 21, 2013 at 01:53:38PM +0100, Rafael J. Wysocki wrote: > I can, if you ACK it for me. :-) Will do, after Stefan corrects some cosmetic issues :-) Thanks. -- Regards/Gruss, Boris. Sent from a fat crate under my desk. Formatting is fine. -- -- To unsubscribe from this list: send

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Rafael J. Wysocki
On Monday, January 21, 2013 01:42:55 PM Borislav Petkov wrote: > On Mon, Jan 21, 2013 at 12:22:18PM +, Stefan Bader wrote: > > So for having the "check for sensible BIOS" in mainline I refreshed > > the patch (fixed the bit test, and actually tested it this time) and > > also added some hopeful

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Borislav Petkov
On Mon, Jan 21, 2013 at 12:22:18PM +, Stefan Bader wrote: > So for having the "check for sensible BIOS" in mainline I refreshed > the patch (fixed the bit test, and actually tested it this time) and > also added some hopefully sensible explanation to it (attached > below). > > Should I send it

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-21 Thread Stefan Bader
On 01/18/2013 08:03 PM, Borislav Petkov wrote: On Fri, Jan 18, 2013 at 02:00:15PM -0500, Konrad Rzeszutek Wilk wrote: I did not explain myself well. The fix is OK - it just that the hypervisor causes the quirk to not work correctly. Hmm, I wonder if there BIOSes that do the same thing (cause the

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-18 Thread Konrad Rzeszutek Wilk
> > > Right, that information is gathered from the MSRs. I think the Xen would > > need to do this since it can do the MSRs correctly and modify the P-states. > > > > So something like this in the hypervisor maybe (not even tested): > > Yeah, something like that. Basically you can copy the quirk

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-18 Thread Borislav Petkov
On Fri, Jan 18, 2013 at 02:00:15PM -0500, Konrad Rzeszutek Wilk wrote: > I did not explain myself well. The fix is OK - it just that the > hypervisor causes the quirk to not work correctly. Hmm, I wonder if > there BIOSes that do the same thing (cause the MSR to return 0). Per > you estimation of B

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-18 Thread Boris Ostrovsky
On 01/18/2013 02:00 PM, Konrad Rzeszutek Wilk wrote: Right, that information is gathered from the MSRs. I think the Xen would need to do this since it can do the MSRs correctly and modify the P-states. So something like this in the hypervisor maybe (not even tested): Is there any harm in allo

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-18 Thread Konrad Rzeszutek Wilk
On Tue, Jan 15, 2013 at 07:18:39PM +0100, Borislav Petkov wrote: > On Tue, Jan 15, 2013 at 12:53:05PM -0500, Konrad Rzeszutek Wilk wrote: > > > 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 > >

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-16 Thread Stefan Bader
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_perf

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-16 Thread Jan Beulich
>>> 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 @

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-15 Thread Borislav Petkov
On Tue, Jan 15, 2013 at 12:53:05PM -0500, Konrad Rzeszutek Wilk wrote: > > 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 cp

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-15 Thread Matt Wilson
On Tue, Jan 15, 2013 at 02:04:30PM +0100, Matt Wilson wrote: > On Mon, Jan 14, 2013 at 04:58:54PM +0100, Stefan Bader wrote: > > Starting with kernel v3.7 the following commit added a quirk > > to obtain the real frequencies of certain AMD systems: > > > > commit f594065faf4f9067c2283a34619fc0714e

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-15 Thread Konrad Rzeszutek Wilk
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: > > Starting with kernel v3.7 the following commit added a quirk > > to obtain the real frequencies of certain AMD systems: > > > > commit f594065faf4f9067c2283a34619fc0

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-15 Thread Matt Wilson
On Mon, Jan 14, 2013 at 04:58:54PM +0100, Stefan Bader wrote: > Starting with kernel v3.7 the following commit added a quirk > to obtain the real frequencies of certain AMD systems: > > commit f594065faf4f9067c2283a34619fc0714e79a98d > Author: Matthew Garrett > Date: Tue Sep 4 08:28:06 2012 +00

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-14 Thread André Przywara
On Mon, 14 Jan 2013 18:08:45 +0100 Stefan Bader wrote: > On 14.01.2013 17:34, Borislav Petkov wrote: > > On Mon, Jan 14, 2013 at 04:58:54PM +0100, Stefan Bader wrote: > >> Starting with kernel v3.7 the following commit added a quirk > >> to obtain the real frequencies of certain AMD systems: > >>

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-14 Thread Stefan Bader
On 14.01.2013 17:34, Borislav Petkov wrote: > On Mon, Jan 14, 2013 at 04:58:54PM +0100, Stefan Bader wrote: >> Starting with kernel v3.7 the following commit added a quirk >> to obtain the real frequencies of certain AMD systems: >> >> commit f594065faf4f9067c2283a34619fc0714e79a98d >> Author: Matt

Re: [Xen-devel] kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-14 Thread Jan Beulich
>>> On 14.01.13 at 17:34, 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

Re: kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-14 Thread Borislav Petkov
On Mon, Jan 14, 2013 at 04:58:54PM +0100, Stefan Bader wrote: > Starting with kernel v3.7 the following commit added a quirk > to obtain the real frequencies of certain AMD systems: > > commit f594065faf4f9067c2283a34619fc0714e79a98d > Author: Matthew Garrett > Date: Tue Sep 4 08:28:06 2012 +00

kernel 3.7+ cpufreq regression on AMD system running as dom0

2013-01-14 Thread Stefan Bader
Starting with kernel v3.7 the following commit added a quirk to obtain the real frequencies of certain AMD systems: commit f594065faf4f9067c2283a34619fc0714e79a98d Author: Matthew Garrett Date: Tue Sep 4 08:28:06 2012 + ACPI: Add fixups for AMD P-state figures When running bare-metal,