On Fri, Jun 17, 2016 at 03:41:38PM -0400, Vince Weaver wrote:
> FWIW on the trinity machine (F15hM13h) with your patch applied I get the
> following:
>
> vince@a10:~$ sensors
> k10temp-pci-00c3
> Adapter: PCI adapter
> temp1: +0.0°C (high = +70.0°C)
>(crit = +70.0
On Fri, Jun 17, 2016 at 02:58:41PM -0400, Vince Weaver wrote:
> It's a dual-socket server machine.
> AMD Opteron(tm) Processor 6376
> 2 packages, 16 cores each.
Right, so this is 2 MCMs. You have 2 physical nodes and each physical is
separated into 2 logical nodes additionally.
So provided the wa
On Fri, 17 Jun 2016, Vince Weaver wrote:
> I'll try applying the patch. My code that probes the msrs/pci directly
> from userspace seems to think it has support.
FWIW on the trinity machine (F15hM13h) with your patch applied I get the
following:
vince@a10:~$ sensors
k10temp-pci-00c3
Adapter:
On Fri, 17 Jun 2016, Borislav Petkov wrote:
> On Fri, Jun 17, 2016 at 01:38:44PM -0400, Vince Weaver wrote:
> > F15hM02h
> >
> > fam15h_power-pci-00c4
> > Adapter: PCI adapter
> > power1: 105.64 W (crit = 115.17 W)
> >
> > fam15h_power-pci-00d4
> > Adapter: PCI adapter
> > power1: 105
On Fri, Jun 17, 2016 at 01:38:44PM -0400, Vince Weaver wrote:
> F15hM02h
>
> fam15h_power-pci-00c4
> Adapter: PCI adapter
> power1: 105.64 W (crit = 115.17 W)
>
> fam15h_power-pci-00d4
> Adapter: PCI adapter
> power1: 105.59 W (crit = 115.17 W)
Is that a dual-socket machine?
Because
On Fri, 17 Jun 2016, Borislav Petkov wrote:
> Piledriver? What f/m/s is that?
>
> Because I have a PD here - AMD FX(tm)-8350 Eight-Core Processor -
> F15hM02 and sensors gives only this below. But my PD doesn't have
> X86_FEATURE_ACC_POWER.
F15hM02h
fam15h_power-pci-00c4
Adapter: PCI adapter
po
On Fri, Jun 17, 2016 at 11:56:31AM -0400, Vince Weaver wrote:
> Has anyone actually verified that the fam15h excavator machines are
> producing "stable" results?
Not that I know of. Rui, any ideas?
> I know on my fam15h piledriver machines the results returned by the
> drivers/hwmon/fam15h_power.
On Fri, 17 Jun 2016, Borislav Petkov wrote:
> On Fri, Jun 17, 2016 at 06:07:47PM +0800, Huang Rui wrote:
> > Agree with you. If the some chips are not stable, we can add a check
> > to ignore them with family and model id.
>
> So if family 0x16 is not "stable" as you say, we probably should keep
On Fri, Jun 17, 2016 at 06:07:47PM +0800, Huang Rui wrote:
> Agree with you. If the some chips are not stable, we can add a check
> to ignore them with family and model id.
So if family 0x16 is not "stable" as you say, we probably should keep
the cpu_match array too.
Actually, you could merge the
On Thu, Jun 16, 2016 at 11:12:18PM +0200, Borislav Petkov wrote:
> On Thu, Jun 16, 2016 at 05:00:04PM -0400, Vince Weaver wrote:
> >
> > According to the BKDG the AMD Family16h Model30h "Jaguar Mullins"
> > also supports the accumulated power interface. I've tested on
> > hardware I have and with
On Thu, Jun 16, 2016 at 05:00:04PM -0400, Vince Weaver wrote:
>
> According to the BKDG the AMD Family16h Model30h "Jaguar Mullins"
> also supports the accumulated power interface. I've tested on
> hardware I have and with this patch I indeed get power readings using
> perf.
>
> Signed-off-by:
According to the BKDG the AMD Family16h Model30h "Jaguar Mullins"
also supports the accumulated power interface. I've tested on
hardware I have and with this patch I indeed get power readings using
perf.
Signed-off-by: Vince Weaver
diff --git a/arch/x86/events/amd/power.c b/arch/x86/events/am
12 matches
Mail list logo