Received-SPF: pass (google.com: domain of linux-kernel+bounces-145081-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Sender: Guenter Roeck <groeck7@gmail.com>
Date: Mon, 15 Apr 2024 05:42:09 -0700
From: Guenter Roeck <linux@roeck-us.net>
To: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Cc: "Zhang, Rui" <rui.zhang@intel.com>,
	"Wysocki, Rafael J" <rafael.j.wysocki@intel.com>,
	"jdelvare@suse.com" <jdelvare@suse.com>,
	"srinivas.pandruvada@linux.intel.com" <srinivas.pandruvada@linux.intel.com>,
	"lukasz.luba@arm.com" <lukasz.luba@arm.com>,
	"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
	"linux-hwmon@vger.kernel.org" <linux-hwmon@vger.kernel.org>,
	"daniel.lezcano@linaro.org" <daniel.lezcano@linaro.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"Neri, Ricardo" <ricardo.neri@intel.com>
Subject: Re: [PATCH 3/3] hwmon: (coretemp) Use a model-specific bitmask to
 read registers
Message-ID: <d807bb47-7501-4423-9a8f-d51b5349c96b@roeck-us.net>
References: <20240406010416.4821-1-ricardo.neri-calderon@linux.intel.com>
 <20240406010416.4821-4-ricardo.neri-calderon@linux.intel.com>
 <6411e721a1cd559f59aecd8ac57a941cdaa13ff1.camel@intel.com>
 <20240415011946.GA12551@ranerica-svr.sc.intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20240415011946.GA12551@ranerica-svr.sc.intel.com>

On Sun, Apr 14, 2024 at 06:19:46PM -0700, Ricardo Neri wrote:
> On Sun, Apr 07, 2024 at 08:24:40AM +0000, Zhang, Rui wrote:
> > On Fri, 2024-04-05 at 18:04 -0700, Ricardo Neri wrote:
> > > The Intel Software Development manual defines states the temperature
> > 
> > I failed to parse this, is the above "states" redundant?
> 
> Sorry Rui! I missed this repy.
> 
> Ah, the commit message is wrong. I will do s/defines//
> 
> > 
> > [...]
> > 
> > > digital readout as the bits [22:16] of the
> > > IA32_[PACKAGE]_THERM_STATUS
> > > registers. In recent processor, however, the range is [23:16]. Use a
> > > model-specific bitmask to extract the temperature readout correctly.
> > > 
> > > diff --git a/drivers/hwmon/coretemp.c b/drivers/hwmon/coretemp.c
> > > index 616bd1a5b864..5632e1b1dfb1 100644
> > > --- a/drivers/hwmon/coretemp.c
> > > +++ b/drivers/hwmon/coretemp.c
> > > @@ -17,6 +17,7 @@
> > > ?#include <linux/sysfs.h>
> > > ?#include <linux/hwmon-sysfs.h>
> > > ?#include <linux/err.h>
> > > +#include <linux/intel_tcc.h>
> > > ?#include <linux/mutex.h>
> > > ?#include <linux/list.h>
> > > ?#include <linux/platform_device.h>
> > > @@ -404,6 +405,8 @@ static ssize_t show_temp(struct device *dev,
> > > ????????tjmax = get_tjmax(tdata, dev);
> > > ????????/* Check whether the time interval has elapsed */
> > > ????????if (time_after(jiffies, tdata->last_updated + HZ)) {
> > > +???????????????u32 mask =
> > > intel_tcc_get_temp_mask(is_pkg_temp_data(tdata));
> > > +
> > > ????????????????rdmsr_on_cpu(tdata->cpu, tdata->status_reg, &eax,
> > > &edx);
> > > ????????????????/*
> > > ???????????????? * Ignore the valid bit. In all observed cases the
> > > register
> > > @@ -411,7 +414,7 @@ static ssize_t show_temp(struct device *dev,
> > > ???????????????? * Return it instead of reporting an error which
> > > doesn't
> > > ???????????????? * really help at all.
> > > ???????????????? */
> > > -???????????????tdata->temp = tjmax - ((eax >> 16) & 0x7f) * 1000;
> > > +???????????????tdata->temp = tjmax - ((eax >> 16) & mask) * 1000;
> > > ????????????????tdata->last_updated = jiffies;
> > > ????????}
> > > 
> > Besides this one, we can also convert to use intel_tcc_get_tjmax() in
> > get_tjmax().
> 
> I thought about this, but realized that the bitmask of TjMax is always
> [23:16]; no need for a model check. If anything, intel_tcc_get_tjmax()
> would remove some duplicated code. But coretemp.c would need to depend
> on INTEL_TCC, which seems to be a non-starter.
> 

Calling intel_tcc_get_temp_mask() in practice already introduces that
dependency because it returns a fixed mask if INTEL_TCC is not enabled.
If that doesn't matter, the dynamic mask is unnecessary to start with.

Guenter