- 25 Oct, 2018 10 commits
-
-
Geert Uytterhoeven authored
The workqueue used for monitoring the hardware may run while the device is already suspended. Fix this by using the freezable system workqueue instead, cfr. commit 51e20d0e ("thermal: Prevent polling from happening during system suspend"). Fixes: 608567aa ("thermal: da9062/61: Thermal junction temperature monitoring driver") Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Acked-by:
Steve Twiss <stwiss.opensource@diasemi.com> Signed-off-by:
Eduardo Valentin <edubezval@gmail.com>
-
Geert Uytterhoeven authored
When testing bind/unbind on r8a7791/koelsch: WARNING: CPU: 1 PID: 697 at lib/debugobjects.c:329 debug_print_object+0x8c/0xb4 ODEBUG: free active (active state 0) object type: timer_list hint: delayed_work_timer_fn+0x0/0x10 This happens if the workqueue runs after the device has been unbound. Fix this by cancelling any queued work during remove. Fixes: e0a5172e ("thermal: rcar: add interrupt support") Signed-off-by:Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Signed-off-by:
-
Geert Uytterhoeven authored
On r8a7791/koelsch, sometimes the following message is printed during system suspend: rcar_thermal e61f0000.thermal: thermal sensor was broken This happens if the workqueue runs while the device is already suspended. Fix this by using the freezable system workqueue instead, cfr. commit 51e20d0e ("thermal: Prevent polling from happening during system suspend"). Fixes: e0a5172e ("thermal: rcar: add interrupt support") Signed-off-by: -
Sergei Shtylyov authored
Add the R-Car V3H (R8A77980) SoC support to the R-Car gen3 thermal driver. Signed-off-by:
Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Reviewed-by:
Simon Horman <horms+renesas@verge.net.au> Reviewed-by:
-
Sergei Shtylyov authored
Document the R-Car V3H (R8A77980) SoC in the Renesas R-Car gen3 thermal bindings. Signed-off-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
David HERNANDEZ SANCHEZ authored
Add support for DTS thermal sensor that can be found on some STM32 platforms. This driver is based on OF and works in interrupt mode. It offers two temperature trip points: passive and critical. The first is intended for passive cooling notification while the second is used for over-temperature reset. Signed-off-by:
David Hernandez Sanchez <david.hernandezsanchez@st.com> Signed-off-by:
-
David HERNANDEZ SANCHEZ authored
Add thermal binding documentation for STM32 DTS sensor Signed-off-by:
-
Sergei Shtylyov authored
Add the R-Car V3M (R8A77970) SoC support to the R-Car gen2 thermal driver. The hardware is the same as in the R-Car D3 (R8A77995) plus the CIVM status register (we don't use). Signed-off-by:
-
Sergei Shtylyov authored
Document the R-Car V3M (R8A77970) SoC in the Renesas R-Car gen2 thermal bindings. The hardware is the same as in the R-Car D3 (R8A77995) plus an extra status register. Signed-off-by:
-
Sergei Shtylyov authored
The driver on R8A77995 requests the same IRQ twice since platform_get_resource() is always called for the 1st IRQ resource. Fixes: 1969d9dc ("thermal: rcar_thermal: add r8a77995 support") Signed-off-by:
Daniel Lezcano <daniel.lezcano@linaro.org> Signed-off-by:
-
- 23 Oct, 2018 25 commits
-
-
Biju Das authored
Add thermal sensor support for r8a7744 SoC. The Renesas RZ/G1N (r8a7744) thermal sensor module is identical to the R-Car Gen2 family. No driver change is needed due to the fallback compatible value "renesas,rcar-gen2-thermal". Signed-off-by:
Biju Das <biju.das@bp.renesas.com> Reviewed-by:
Chris Paterson <Chris.Paterson2@renesas.com> Reviewed-by:
-
Daniel Lezcano authored
The code is ready to support multiple sensors on the hi3660. The DT defines a thermal zone per cluster. Add the little cluster sensor and let it bind with the thermal zone. Signed-off-by:
-
Daniel Lezcano authored
Add the sensor channels id for the little, g3d and modem. Signed-off-by:
-
Daniel Lezcano authored
The irq field in the data structure is pointless as the scope of its usage is just to request the interrupt. It can be replaced by a local variable. Use the 'ret' variable to get the interrupt number. Signed-off-by:
-
Daniel Lezcano authored
As we have the interrupt names defines, replace platform_get_irq() by platform_get_irq_byname(), so no confusion can be made when getting the interrupt with the sensor id. Signed-off-by:
-
Daniel Lezcano authored
Change the macro name in order to give a better indication of the sensor location. Signed-off-by:
-
Daniel Lezcano authored
Change the code as it is dealing with several sensors. For git-bisect compatibility (compilation and booting), assume the DT is not yet changed and we have a single interrupt. Next changes will support multiple interrupt sorted by their name. Signed-off-by:
-
Daniel Lezcano authored
Convert the 'sensor' field to a pointer and propagate the change in the file. Havintg a pointer, gives us the opportunity to define multiple sensors. Signed-off-by:
-
Daniel Lezcano authored
The hi6220 and the hi3660 probe functions are doing almost the same operations, they can share 90% of their code. Factor out the probe functions by moving the common code in the common probe function. Signed-off-by:
-
Daniel Lezcano authored
Store the sensor pointer in the thermal zone private data and use it in the callback functions. That allows to continue the conversion to sensor oriented code where the pointers are the sensors. Signed-off-by:
-
Daniel Lezcano authored
In order to support multiple sensors, we have to change the code to deal with sensors and not the hisi thermal structure. Add a back pointer to the hisi thermal structure (containerof is not a good option because later we convert the sensor field to a pointer). Change the functions parameters to take a sensor instead of this hisi thermal 'data' structure. Signed-off-by:
-
Daniel Lezcano authored
Group the temperature sensor specific ops into a single structure and assign it to hisi thermal data structure. Change the platform data pointer to reference the specific sensor ops instead of the probe functions. Moving out those allow to split the code to self-encapsulate the sensor object. Signed-off-by:
-
Dan Carpenter authored
The platform_get_resource() function doesn't return error pointers, it returns NULL on error. Fixes: 3d4e5184 ("thermal: armada: convert driver to syscon register accesses") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by:
Miquel Raynal <miquel.raynal@bootlin.com> Signed-off-by:
-
Anson Huang authored
During probe phase, the error path can be handled in one place and use goto method to save many duplicated code. Signed-off-by:
Anson Huang <Anson.Huang@nxp.com> Reviewed-by:
-
Anson Huang authored
Remove the duplicated "from" to improve the error message. Signed-off-by:
-
Amit Kucheria authored
c(1) + x(1) was actually meant to be c(1) * x(1). Signed-off-by:
Amit Kucheria <amit.kucheria@linaro.org> Reviewed-by:
Matthias Kaehlcke <mka@chromium.org> Acked-by:
Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by:
-
Amit Kucheria authored
The SROT registers are initialised by the secure firmware at boot. We don't have write access to the registers. Check if the block is enabled before continuing. Signed-off-by:
-
Amit Kucheria authored
Registers have moved around across TSENS generations. For example, the CTRL register was at offset 0x0 in the SROT region on msm8916 but is at offset 0x4 in newer v2 based TSENS HW blocks. Allow passing offsets of important registers so that we can continue to use common functions. Signed-off-by:
-
Amit Kucheria authored
On platforms whose device trees specify two address spaces for TSENS, the second one points to the SROT registers. Initialise the SROT map on those platforms. Signed-off-by:
-
Amit Kucheria authored
The TSENS driver currently only uses a limited set of registers from the TM address space. So it was ok to map just that set of registers and call it "map". We'd now like to map a second set: SROT registers to introduce new functionality. Rename the "map" field to a more appropriate "tm_map". The 8960 doesn't have a clear split between TM and SROT registers. To avoid complicating the data structure, it will switchover to using tm_map for its maps. There is no functional change with this patch. Signed-off-by:
-
Amit Kucheria authored
hw_id is dynamically allocated but not used anywhere. Get rid of dead code. Signed-off-by:
-
Amit Kucheria authored
The TSENS drivers use a GPL-2.0 license. Replace with equivalent SPDX tags and delete the full license text. Signed-off-by:
-
Amit Kucheria authored
We've already converted over the devicetree of platforms using v2 version of the TSENS IP to use two address spaces. Now prepare to convert over the 8916 and 8974 platforms to use separate SROT and TM address spaces. This patch will work with device trees with one or two address spaces because we set the tm_offset in commit 5b128398 ("thermal: tsens: Add support to split up register address space into two"). Signed-off-by:
-
Anson Huang authored
Add i.mx8mq specific compatible string. Signed-off-by:
-
Rob Herring authored
In preparation to remove the node name pointer from struct device_node, convert printf users to use the %pOFn format specifier. Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: Daniel Lezcano <daniel.lezcano@linaro.org> Cc: linux-pm@vger.kernel.org Signed-off-by:
-
- 24 Aug, 2018 5 commits
-
-
Fabrizio Castro authored
Document RZ/G2M (R8A774A1) SoC bindings. Signed-off-by:
Fabrizio Castro <fabrizio.castro@bp.renesas.com> Reviewed-by:
-
Fabrizio Castro authored
Add r8a774a1 specific compatible string. Signed-off-by:
-
Viresh Kumar authored
A cooling map entry may now contain a list of phandles and their arguments representing multiple devices which share the trip point. This patch updates the thermal OF core to parse them properly. The trip point and contribution value is shared by multiple cooling devices now and so a new structure is created, struct __thermal_cooling_bind_param, which represents a cooling device and its min/max states and the existing struct __thermal_bind_params now contains an array of this new cooling device structure. Tested on Hikey960. Signed-off-by:
Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by:
-
Matthias Kaehlcke authored
There are three thermal stages defined in the PMIC: stage 1: warning stage 2: system should shut down stage 3: emergency shut down By default the PMIC assumes that the OS isn't doing anything and thus at stage 2 it does a partial PMIC shutdown and at stage 3 it kills all power. When switching between thermal stages the PMIC generates an interrupt which is handled by the driver. The partial PMIC shutdown at stage 2 can be disabled by software, which allows the OS to initiate a shutdown at stage 2 with a thermal zone configured accordingly. If a critical trip point is configured in the thermal zone the driver adjusts the stage 1-3 temperature thresholds to (closely) match the critical temperature with a stage 2 threshold (125/130/135/140 °C). If a suitable match is found the partial shutdown at stage 2 is disabled. If for some reason the system doesn't shutdown at stage 2 the emergency shutdown at stage 3 kicks in. The partial shutdown at stage 2 remains enabled in these cases: - no critical trip point defined - the temperature of the critical trip point is < 125°C - the temperature of the critical trip point is > 140°C and no ADC channel is configured (thus the OS is not notified when the critical temperature is reached) Suggested-by:
Douglas Anderson <dianders@chromium.org> Signed-off-by:
-
Matthias Kaehlcke authored
The current example for a thermal zone isn't very useful as reference since it would result in a hardware shutdown at 145°C, instead of allowing the system to try to shutdown gracefully. Without an ADC channel a maximum of two trip points is useful in practice for this sensor, with temperatures corresponding to the stage 1 and stage 2 'hardware trip points'. A critical trip point at stage 2 may allow the system to shutdown before a hardware shutdown at stage 3 kicks in. It should be noted though that by default the chip performs a 'partial shutdown' when the temperature reaches stage 2, which may prevent an orderly shutdown. The 'partial shutdown' can be disabled by software. Signed-off-by:
-
