11 Dec, 2015
1 commit
-
TMP102 works based on conversions done periodically. However, as per
the TMP102 data sheet[1] the first conversion is triggered immediately
after we program the configuration register. The temperature data
registers do not reflect proper data until the first conversion is
complete (in our case HZ/4).The driver currently sets the last_update to be jiffies - HZ, just
after the configuration is complete. When TMP102 driver registers
with the thermal framework, it immediately tries to read the sensor
temperature data. This takes place even before the conversion on the
TMP102 is complete and results in an invalid temperature read.Depending on the value read, this may cause thermal framework to
assume that a critical temperature event has occurred and attempts to
shutdown the system.Instead of causing an invalid mid-conversion value to be read
erroneously, we mark the last_update to be in-line with the current
jiffies. This allows the tmp102_update_device function to skip update
until the required conversion time is complete. Further, we ensure to
return -EAGAIN result instead of returning spurious temperature (such
as 0C) values to the caller to prevent any wrong decisions made with
such values. NOTE: this allows the read functions not to be blocking
and allows the callers to make the decision if they would like to
block or try again later. At least the current user(thermal) seems to
handle this by retrying later.A simpler alternative approach could be to sleep in the probe for the
duration required, but that will result in latency that is undesirable
and delay boot sequence un-necessarily.[1] http://www.ti.com/lit/ds/symlink/tmp102.pdf
Cc: Eduardo Valentin
Reported-by: Aparna Balasubramanian
Reported-by: Elvita Lobo
Reported-by: Yan Liu
Signed-off-by: Nishanth Menon
Signed-off-by: Guenter Roeck
03 Aug, 2015
1 commit
-
The thermal code uses int, long and unsigned long for temperatures
in different places.Using an unsigned type limits the thermal framework to positive
temperatures without need. Also several drivers currently will report
temperatures near UINT_MAX for temperatures below 0°C. This will probably
immediately shut the machine down due to overtemperature if started below
0°C.'long' is 64bit on several architectures. This is not needed since INT_MAX °mC
is above the melting point of all known materials.Consistently use a plain 'int' for temperatures throughout the thermal code and
the drivers. This only changes the places in the drivers where the temperature
is passed around as pointer, when drivers internally use another type this is
not changed.Signed-off-by: Sascha Hauer
Acked-by: Geert Uytterhoeven
Reviewed-by: Jean Delvare
Reviewed-by: Lukasz Majewski
Reviewed-by: Darren Hart
Reviewed-by: Heiko Stuebner
Reviewed-by: Peter Feuerer
Cc: Punit Agrawal
Cc: Zhang Rui
Cc: Eduardo Valentin
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: Jean Delvare
Cc: Peter Feuerer
Cc: Heiko Stuebner
Cc: Lukasz Majewski
Cc: Stephen Warren
Cc: Thierry Reding
Cc: linux-acpi@vger.kernel.org
Cc: platform-driver-x86@vger.kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-omap@vger.kernel.org
Cc: linux-samsung-soc@vger.kernel.org
Cc: Guenter Roeck
Cc: Rafael J. Wysocki
Cc: Maxime Ripard
Cc: Darren Hart
Cc: lm-sensors@lm-sensors.org
Signed-off-by: Zhang Rui
04 Feb, 2015
1 commit
-
Setting a dev_pm_ops suspend/resume pair but not a set of
hibernation functions means those pm functions will not be
called upon hibernation.
Fix this by using SIMPLE_DEV_PM_OPS, which appropriately
assigns the suspend and hibernation handlers and move
mp102_suspend/tmp102_resume under CONFIG_PM_SLEEP to avoid
build warnings.Signed-off-by: Grygorii Strashko
[groeck: Declare tmp102_dev_pm_ops as static variable]
Signed-off-by: Guenter Roeck
20 Nov, 2014
1 commit
-
Different drivers request API extensions in of-thermal. For this reason,
additional callbacks are required to fit the new drivers needs.The current API implementation expects the registering sensor driver
to provide a get_temp and get_trend callbacks as function parameters.
As the amount of callbacks is growing, this patch changes the existing
implementation to use a .ops field to hold all the of thermal callbacks
to sensor drivers.This patch also changes the existing of-thermal users to fit the new
API design. No functional change is introduced in this patch.Cc: Alexandre Courbot
Cc: devicetree@vger.kernel.org
Cc: Grant Likely
Cc: Guenter Roeck
Cc: Jean Delvare
Cc: linux-kernel@vger.kernel.org
Cc: linux-pm@vger.kernel.org
Cc: linux-tegra@vger.kernel.org
Cc: lm-sensors@lm-sensors.org
Cc: Rob Herring
Cc: Stephen Warren
Cc: Thierry Reding
Cc: Zhang Rui
Acked-by: Guenter Roeck
Tested-by: Mikko Perttunen
Reviewed-by: Mikko Perttunen
Reviewed-by: Alexandre Courbot
Reviewed-by: Lukasz Majewski
Signed-off-by: Eduardo Valentin
22 May, 2014
2 commits
-
Simplify code, reduce code size, and attach sysfs attributes to hwmon device.
Reviewed-by: Jean Delvare
Signed-off-by: Guenter Roeck -
The pointer to client->dev is used several times in the probe function.
Simplify code by introducing a separate variable for it.Reviewed-by: Jean Delvare
Signed-off-by: Guenter Roeck
04 Dec, 2013
1 commit
-
This patch adds to tmp102 temperature sensor the possibility
to expose itself as thermal zone device, registered on the
thermal framework.The thermal zone is built only if a device tree node
describing a thermal zone for this sensor is present
inside the tmp102 DT node. Otherwise, the driver behavior
will be the same.Note: This patch has also been reviewed by Jean D. He has
requested to perform a wider inspection of possible
users of thermal and hwmon interaction API. On the other
hand, the change on this patch is acceptable on first
step of overall code change.Cc: Jean Delvare
Cc: lm-sensors@lm-sensors.org
Cc: linux-kernel@vger.kernel.org
Acked-by: Guenter Roeck
Signed-off-by: Eduardo Valentin
08 Apr, 2013
1 commit
-
Cc: Corentin Labbe
Cc: Mark M. Hoffman
Cc: Fenghua Yu
Cc: Juerg Haefliger
Cc: Andreas Herrmann
Cc: Rudolf Marek
Cc: Jim Cromie
Cc: Roger Lucas
Cc: Marc Hulsman
Signed-off-by: Guenter Roeck
26 Jan, 2013
1 commit
-
SENSORS_LIMIT and the generic clamp_val have the same functionality,
and clamp_val is more efficient.This patch reduces text size by 9052 bytes and bss size by 11624 bytes
for x86_64 builds.Signed-off-by: Guenter Roeck
Acked-by: George Joseph
Acked-by: Jean Delvare
29 Nov, 2012
3 commits
-
CONFIG_HOTPLUG is going away as an option so __devexit is no
longer needed.Signed-off-by: Bill Pemberton
Cc: Hans de Goede
Cc: Jean Delvare
Cc: Alistair John Strachan
Cc: Fenghua Yu
Cc: Juerg Haefliger
Cc: Andreas Herrmann
Cc: Clemens Ladisch
Cc: Rudolf Marek
Cc: Jim Cromie
Cc: "Mark M. Hoffman"
Cc: Roger Lucas
Acked-by: Guenter Roeck
Acked-by: Mark Brown
Signed-off-by: Greg Kroah-Hartman -
CONFIG_HOTPLUG is going away as an option so __devinit is no longer
needed.Signed-off-by: Bill Pemberton
Cc: Hans de Goede
Cc: Jean Delvare
Cc: Alistair John Strachan
Cc: Fenghua Yu
Cc: Juerg Haefliger
Cc: Andreas Herrmann
Cc: Clemens Ladisch
Cc: Rudolf Marek
Cc: Jim Cromie
Cc: "Mark M. Hoffman"
Cc: Roger Lucas
Acked-by: Guenter Roeck
Acked-by: Mark Brown
Signed-off-by: Greg Kroah-Hartman -
CONFIG_HOTPLUG is going away as an option so __devexit_p is no longer
needed.Signed-off-by: Bill Pemberton
Cc: Hans de Goede
Cc: Jean Delvare
Cc: Alistair John Strachan
Cc: Fenghua Yu
Cc: Juerg Haefliger
Cc: Andreas Herrmann
Cc: Clemens Ladisch
Cc: Rudolf Marek
Cc: Jim Cromie
Cc: "Mark M. Hoffman"
Cc: Roger Lucas
Acked-by: Guenter Roeck
Acked-by: Mark Brown
Signed-off-by: Greg Kroah-Hartman
10 Oct, 2012
1 commit
-
Many hwmon drivers use jiffies but omit the inclusion of the header
file. Fix that, and also fix one driver which was including the header
file but didn't need it.Signed-off-by: Jean Delvare
Acked-by: Guenter Roeck
Acked-by: Luca Tettamanti
Cc: Marc Hulsman
Cc: Rudolf Marek
22 Jul, 2012
1 commit
-
Convert to use devm_ functions to reduce code size and simplify the code.
Signed-off-by: Guenter Roeck
Acked-by: Jean Delvare
19 Mar, 2012
1 commit
-
This patch converts the drivers in drivers/hwmon/* to use the
module_i2c_driver() macro which makes the code smaller and a bit simpler.Signed-off-by: Axel Lin
Cc: Corentin Labbe
Cc: Dirk Eibach
Cc: "Mark M. Hoffman"
Cc: Steve Glendinning
Cc: Riku Voipio
Cc: Guillaume Ligneul
Cc: David George
Cc: "Hans J. Koch"
Cc: Marc Hulsman
Cc: Rudolf Marek
Signed-off-by: Guenter Roeck
06 Jan, 2012
1 commit
-
replaced strict_strtol with kstrtol and
replaced strict_strtuol with kstrtuolThis satisfies checkpatch -f
Compile tested only: no warnings or errors givenSigned-off-by: Frans Meulenbroeks
Signed-off-by: Guenter Roeck
04 Nov, 2011
1 commit
-
Make use of the new i2c_smbus_{read,write}_word_swapped functions.
This makes the driver code more compact and readable. It also ensures
proper error handling.Signed-off-by: Jean Delvare
Acked-by: Jonathan Cameron
Acked-by: Guenter Roeck
Cc: Dirk Eibach
Cc: "Mark M. Hoffman"
Cc: Guillaume Ligneul
31 Mar, 2011
1 commit
-
Fixes generated by 'codespell' and manually reviewed.
Signed-off-by: Lucas De Marchi
03 Jun, 2010
1 commit
-
I2C drivers can use the clientdata-pointer to point to private data. As I2C
devices are not really unregistered, but merely detached from their driver, it
used to be the drivers obligation to clear this pointer during remove() or a
failed probe(). As a couple of drivers forgot to do this, it was agreed that it
was cleaner if the i2c-core does this clearance when appropriate, as there is
no guarantee for the lifetime of the clientdata-pointer after remove() anyhow.
This feature was added to the core with commit
e4a7b9b04de15f6b63da5ccdd373ffa3057a3681 to fix the faulty drivers.As there is no need anymore to clear the clientdata-pointer, remove all current
occurrences in the drivers to simplify the code and prevent confusion.Signed-off-by: Wolfram Sang
Acked-by: Mark Brown
Acked-by: Greg Kroah-Hartman
Acked-by: Richard Purdie
Acked-by: Dmitry Torokhov
Signed-off-by: Jean Delvare
28 May, 2010
4 commits
-
Only stop the chip at driver exit if it was stopped when driver was
loaded. Leave it running otherwise.Also restore the device configuration if probe failed, to not leave
the system in a dangling state.Signed-off-by: Jean Delvare
Cc: Steven King -
Suspend and resume functions shouldn't overwrite the configuration
register. They should only alter the one bit they have to touch.Also don't assume that register reads and writes always succeed.
Handle errors properly, shall they happen.Signed-off-by: Jean Delvare
Cc: Steven King -
Fixes from my driver review:
http://lists.lm-sensors.org/pipermail/lm-sensors/2010-March/028051.htmlOnly the small changes are in there, more important changes will come
later separately as time permits.* Drop the remnants of the now gone detect function
* The TMP102 has no known compatible chip
* Include the right header files
* Clarify why byte swapping of register values is needed
* Strip resolution info bit from temperature register value
* Set cache lifetime to 1/3 second
* Don't arbitrarily reject limit values; clamp as needed
* Make limit writing unconditional
* Don't check for transaction types the driver doesn't use
* Properly check for error when setting configuration
* Report error on failed probe
* Make the driver load automatically where needed
* Various other minor fixesSigned-off-by: Jean Delvare
Cc: Steven King -
Driver for the TI TMP102.
The TI TMP102 is similar to the LM75. It differs from the LM75 by
having a 16-bit conf register and the temp registers have a minimum
resolution of 12 bits; the extended conf register can select 13-bit
resolution (which this driver does) and also change the update rate
(which this driver currently doesn't use).[JD: Fix tmp102_exit tag, must be __exit, not __init.]
Signed-off-by: Steven King
Signed-off-by: Jean Delvare
