linux-kernel - Re: [patch] events/amd/power add support for fam16h model30h

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20160617181906.GM3912@pd.tnic>
Date:	Fri, 17 Jun 2016 20:19:06 +0200
From:	Borislav Petkov <bp@...en8.de>
To:	Vince Weaver <vincent.weaver@...ne.edu>,
	Huang Rui <ray.huang@....com>
Cc:	linux-kernel@...r.kernel.org,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...hat.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>
Subject: Re: [patch] events/amd/power add support for fam16h model30h

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 I have only the fam15h_power-pci-00c4 adapter. And mine goes
between 20W and 40W in idle.

> The whole system only uses 167W total at idle (according to a
> wall-outlet power meter), so the two packages each drawing 105W seems
> a bit off.

Or is that an MCM box? I.e., a Multi-node CPU? Because:

Documentation/hwmon/fam15h_power:
"On multi-node processors the calculated value is for the entire package
and not for a single node. Thus the driver creates sysfs attributes only
for internal node0 of a multi-node processor."

And I remember Andreas did that should_load_on_this_node() thing.

If so, then 105W vs 167W should fit.

> I also have a F15hM13h system but it doesn't seem to support TDP
> measurement.

That's Trinity. That PCI device ID is not even in the kernel - I'm
guessing no one has even thought of enabling fam15_power on it. I don't
know even whether it supports the different PCI interfaces like TDP, TDP
limit, etc.

IOW, something like that might not really help.

Rui, you could take a look if you feel bored and if you have a TN system
somewhere lying around.

---
diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index eb97a9241d17..b31ef2779aac 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -503,6 +503,7 @@ static int fam15h_power_probe(struct pci_dev *pdev,

 static const struct pci_device_id fam15h_power_id_table[] = {
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_NB_F4) },
+	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_M10H_F4) },
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_M30H_NB_F4) },
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_M60H_NB_F4) },
 	{ PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_15H_M70H_NB_F4) },
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index c58752fe16c4..f7385ed3d5f1 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -521,6 +521,7 @@
 #define PCI_DEVICE_ID_AMD_11H_NB_MISC	0x1303
 #define PCI_DEVICE_ID_AMD_11H_NB_LINK	0x1304
 #define PCI_DEVICE_ID_AMD_15H_M10H_F3	0x1403
+#define PCI_DEVICE_ID_AMD_15H_M10H_F4	0x1404
 #define PCI_DEVICE_ID_AMD_15H_M30H_NB_F3 0x141d
 #define PCI_DEVICE_ID_AMD_15H_M30H_NB_F4 0x141e
 #define PCI_DEVICE_ID_AMD_15H_M60H_NB_F3 0x1573

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.