Re: [PATCH v8 10/13] hwmon: peci: Add cputemp driver

From: Paul Fertser <fercerpav@gmail.com>
To: Iwona Winiarska <iwona.winiarska@intel.com>
Cc: linux-aspeed@lists.ozlabs.org, linux-doc@vger.kernel.org,
	Dave Hansen <dave.hansen@intel.com>,
	Zev Weiss <zweiss@equinix.com>,
	Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>,
	Jonathan Corbet <corbet@lwn.net>,
	openbmc@lists.ozlabs.org,
	Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>,
	Guenter Roeck <linux@roeck-us.net>,
	devicetree@vger.kernel.org, Jean Delvare <jdelvare@suse.com>,
	Arnd Bergmann <arnd@arndb.de>,
	Billy Tsai <billy_tsai@aspeedtech.com>,
	Rob Herring <robh+dt@kernel.org>, Borislav Petkov <bp@alien8.de>,
	Dan Williams <dan.j.williams@intel.com>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	linux-arm-kernel@lists.infradead.org,
	linux-hwmon@vger.kernel.org, Tony Luck <tony.luck@intel.com>,
	Andrew Jeffery <andrew@aj.id.au>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Randy Dunlap <rdunlap@infradead.org>,
	linux-kernel@vger.kernel.org, Olof Johansson <olof@lixom.net>
Subject: Re: [PATCH v8 10/13] hwmon: peci: Add cputemp driver
Date: Mon, 20 Mar 2023 14:45:15 +0300	[thread overview]
Message-ID: <ZBhHS7v+98NK56is@home.paul.comp> (raw)
In-Reply-To: <20220208153639.255278-11-iwona.winiarska@intel.com>

Hello,

We are seeing wrong DTS temperatures on at least "Intel(R) Xeon(R)
Bronze 3204 CPU @ 1.90GHz" and most probably other Skylake Xeon CPUs
are also affected, see inline.

On Tue, Feb 08, 2022 at 04:36:36PM +0100, Iwona Winiarska wrote:
> Add peci-cputemp driver for Digital Thermal Sensor (DTS) thermal
> readings of the processor package and processor cores that are
> accessible via the PECI interface.
...
> +static const struct cpu_info cpu_hsx = {
> +	.reg		= &resolved_cores_reg_hsx,
> +	.min_peci_revision = 0x33,
> +	.thermal_margin_to_millidegree = &dts_eight_dot_eight_to_millidegree,
> +};
> +
> +static const struct cpu_info cpu_icx = {
> +	.reg		= &resolved_cores_reg_icx,
> +	.min_peci_revision = 0x40,
> +	.thermal_margin_to_millidegree = &dts_ten_dot_six_to_millidegree,
> +};
...
> +	{
> +		.name = "peci_cpu.cputemp.skx",
> +		.driver_data = (kernel_ulong_t)&cpu_hsx,
> +	},

With this configuration we get this data:

/sys/bus/peci/devices/0-30/peci_cpu.cputemp.skx.48/hwmon/hwmon15# grep . temp[123]_{label,input}
temp1_label:Die
temp2_label:DTS
temp3_label:Tcontrol
temp1_input:30938
temp2_input:67735
temp3_input:80000

On the host system "sensors" report

Package id 0:  +31.C (high = +80.C, crit = +90.C)

So I conclude Die temperature as retrieved over PECI is correct while
DTS is mis-calculated. The old downstream code in OpenBMC was using
ten_dot_six_to_millidegree() function for conversion, and that was
providing expected results. And indeed if we reverse the calculation
here we get 80000 - ((80000-67735) * 256 / 64) = 30940 which matches
expectations.

-- 
Be free, use free (http://www.gnu.org/philosophy/free-sw.html) software!
mailto:fercerpav@gmail.com