diff mbox series

[01/18] ARM: dts: exynos: correct fuel gauge interrupt trigger level on GT-I9100

Message ID 20201210212534.216197-1-krzk@kernel.org
State Accepted
Commit 46799802136670e00498f19898f1635fbc85f583
Headers show
Series [01/18] ARM: dts: exynos: correct fuel gauge interrupt trigger level on GT-I9100 | expand

Commit Message

Krzysztof Kozlowski Dec. 10, 2020, 9:25 p.m. UTC
The Maxim fuel gauge datasheets describe the interrupt line as active
low with a requirement of acknowledge from the CPU.  The falling edge
interrupt will mostly work but it's not correct.

Fixes: 8620cc2f99b7 ("ARM: dts: exynos: Add devicetree file for the Galaxy S2")
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
---
 arch/arm/boot/dts/exynos4210-i9100.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Sebastian Reichel Dec. 10, 2020, 10:16 p.m. UTC | #1
(I shortened the Cc)

Hi,

On Thu, Dec 10, 2020 at 10:25:34PM +0100, Krzysztof Kozlowski wrote:
> [...]
> index d956c67d5155..f737de0470de 100644
> --- a/drivers/power/supply/max17040_battery.c
> +++ b/drivers/power/supply/max17040_battery.c
> @@ -367,7 +367,7 @@ static int max17040_enable_alert_irq(struct max17040_chip *chip)
>  
>  	flags = IRQF_TRIGGER_FALLING | IRQF_ONESHOT;
>  	ret = devm_request_threaded_irq(&client->dev, client->irq, NULL,
> -					max17040_thread_handler, flags,
> +					max17040_thread_handler, IRQF_ONESHOT,
>  					chip->battery->desc->name, chip);
>  
>  	return ret;

'flags' is unused after this change.

-- Sebastian
Krzysztof Kozlowski Dec. 11, 2020, 7:47 a.m. UTC | #2
On Thu, Dec 10, 2020 at 10:25:34PM +0100, Krzysztof Kozlowski wrote:
> Interrupt line can be configured on different hardware in different way,
> even inverted.  Therefore driver should not enforce specific trigger
> type - edge falling - but instead rely on Devicetree to configure it.
> 
> The Maxim 14577/77836 datasheets describe the interrupt line as active
> low with a requirement of acknowledge from the CPU therefore the edge
> falling is not correct.
> 
> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
> 
> ---
> 
> This patch should wait till DTS changes are merged, as it relies on
> proper Devicetree.
> ---
>  .../devicetree/bindings/power/supply/max17040_battery.txt       | 2 +-
>  drivers/power/supply/max17040_battery.c                         | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/power/supply/max17040_battery.txt b/Documentation/devicetree/bindings/power/supply/max17040_battery.txt
> index c802f664b508..194eb9fe574d 100644
> --- a/Documentation/devicetree/bindings/power/supply/max17040_battery.txt
> +++ b/Documentation/devicetree/bindings/power/supply/max17040_battery.txt
> @@ -39,7 +39,7 @@ Example:
>  		reg = <0x36>;
>  		maxim,alert-low-soc-level = <10>;
>  		interrupt-parent = <&gpio7>;
> -		interrupts = <2 IRQ_TYPE_EDGE_FALLING>;
> +		interrupts = <2 IRQ_TYPE_LEVEL_LOW>;
>  		wakeup-source;
>  	};
>  
> diff --git a/drivers/power/supply/max17040_battery.c b/drivers/power/supply/max17040_battery.c
> index d956c67d5155..f737de0470de 100644
> --- a/drivers/power/supply/max17040_battery.c
> +++ b/drivers/power/supply/max17040_battery.c
> @@ -367,7 +367,7 @@ static int max17040_enable_alert_irq(struct max17040_chip *chip)
>  
>  	flags = IRQF_TRIGGER_FALLING | IRQF_ONESHOT;

This has to be removed. I will fix it in v2.

Best regards,
Krzysztof


>  	ret = devm_request_threaded_irq(&client->dev, client->irq, NULL,
> -					max17040_thread_handler, flags,
> +					max17040_thread_handler, IRQF_ONESHOT,
>  					chip->battery->desc->name, chip);
>  
>  	return ret;
> -- 
> 2.25.1
>
Krzysztof Kozlowski Dec. 11, 2020, 7:52 a.m. UTC | #3
On Thu, Dec 10, 2020 at 11:16:00PM +0100, Sebastian Reichel wrote:
> (I shortened the Cc)

> 

> Hi,

> 

> On Thu, Dec 10, 2020 at 10:25:34PM +0100, Krzysztof Kozlowski wrote:

> > [...]

> > index d956c67d5155..f737de0470de 100644

> > --- a/drivers/power/supply/max17040_battery.c

> > +++ b/drivers/power/supply/max17040_battery.c

> > @@ -367,7 +367,7 @@ static int max17040_enable_alert_irq(struct max17040_chip *chip)

> >  

> >  	flags = IRQF_TRIGGER_FALLING | IRQF_ONESHOT;

> >  	ret = devm_request_threaded_irq(&client->dev, client->irq, NULL,

> > -					max17040_thread_handler, flags,

> > +					max17040_thread_handler, IRQF_ONESHOT,

> >  					chip->battery->desc->name, chip);

> >  

> >  	return ret;

> 

> 'flags' is unused after this change.


Indeed, thanks. Kbuild also complained now on my github branch. I'll fix
it in v2.
Iskren Chernev Dec. 11, 2020, 3:29 p.m. UTC | #4
On 12/10/20 11:25 PM, Krzysztof Kozlowski wrote:
 > The Maxim fuel gauge datasheets describe the interrupt line as active
 > low with a requirement of acknowledge from the CPU.  The falling edge
 > interrupt will mostly work but it's not correct.
 >
 > Fixes: 45dfa741df86 ("ARM: dts: qcom: msm8974-lge-nexus5: Add fuel 
gauge")
 > Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
 > ---
 >  arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts | 2 +-
 >  1 file changed, 1 insertion(+), 1 deletion(-)
 >
 > diff --git a/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts 
b/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts
 > index e769f638f205..4c6f54aa9f66 100644
 > --- a/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts
 > +++ b/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts
 > @@ -575,7 +575,7 @@ fuelgauge: max17048@36 {
 >              maxim,rcomp = /bits/ 8 <0x4d>;
 >
 >              interrupt-parent = <&msmgpio>;
 > -            interrupts = <9 IRQ_TYPE_EDGE_FALLING>;
 > +            interrupts = <9 IRQ_TYPE_LEVEL_LOW>;
 >
 >              pinctrl-names = "default";
 >              pinctrl-0 = <&fuelgauge_pin>;

According to the datasheet for max17048 the ALRT pin is active low, so that
looks good. The reason it was implemented EDGE_FALLING is mostly due to
fragments taken from downstream, and the fact that it worked :)

Acked-by: Iskren Chernev <iskren.chernev@gmail.com>
Krzysztof Kozlowski Dec. 11, 2020, 3:51 p.m. UTC | #5
On Fri, Dec 11, 2020 at 05:44:26PM +0200, Iskren Chernev wrote:
> 
> On 12/11/20 9:47 AM, Krzysztof Kozlowski wrote:
> > On Thu, Dec 10, 2020 at 10:25:34PM +0100, Krzysztof Kozlowski wrote:
> >> Interrupt line can be configured on different hardware in different way,
> >> even inverted.  Therefore driver should not enforce specific trigger
> >> type - edge falling - but instead rely on Devicetree to configure it.
> >>
> >> The Maxim 14577/77836 datasheets describe the interrupt line as active
> >> low with a requirement of acknowledge from the CPU therefore the edge
> >> falling is not correct.
> >>
> >> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
> >>
> >> ---
> >>
> >> This patch should wait till DTS changes are merged, as it relies on
> >> proper Devicetree.
> >> ---
> >> .../devicetree/bindings/power/supply/max17040_battery.txt       | 2 +-
> >> drivers/power/supply/max17040_battery.c                         | 2 +-
> >>  2 files changed, 2 insertions(+), 2 deletions(-)
> >>
> >> diff --git
> a/Documentation/devicetree/bindings/power/supply/max17040_battery.txt
> b/Documentation/devicetree/bindings/power/supply/max17040_battery.txt
> >> index c802f664b508..194eb9fe574d 100644
> >> --- a/Documentation/devicetree/bindings/power/supply/max17040_battery.txt
> >> +++ b/Documentation/devicetree/bindings/power/supply/max17040_battery.txt
> >> @@ -39,7 +39,7 @@ Example:
> >>          reg = <0x36>;
> >>          maxim,alert-low-soc-level = <10>;
> >>          interrupt-parent = <&gpio7>;
> >> -        interrupts = <2 IRQ_TYPE_EDGE_FALLING>;
> >> +        interrupts = <2 IRQ_TYPE_LEVEL_LOW>;
> >>          wakeup-source;
> >>      };
> >>
> >> diff --git a/drivers/power/supply/max17040_battery.c
> b/drivers/power/supply/max17040_battery.c
> >> index d956c67d5155..f737de0470de 100644
> >> --- a/drivers/power/supply/max17040_battery.c
> >> +++ b/drivers/power/supply/max17040_battery.c
> >> @@ -367,7 +367,7 @@ static int max17040_enable_alert_irq(struct
> max17040_chip *chip)
> >>
> >>      flags = IRQF_TRIGGER_FALLING | IRQF_ONESHOT;
> >
> > This has to be removed. I will fix it in v2.
> >
> > Best regards,
> > Krzysztof
> 
> I removed the IRQF_TRIGGER_FALLING, tweaked the DT as per the DT patch, and
> it worked on the samsung klte.
> 
> I don't understand how the DT irq flag ends up being used by the kernel. It
> is never explicitly read from DT or passed to interrupt API, only i2c->irq,
> which is a pure int.

The core __setup_irq() calls irqd_get_trigger_type() on IRQ data
matching the IRQ.

Best regards,
Krzysztof
Rob Herring (Arm) Dec. 21, 2020, 6:45 p.m. UTC | #6
On Thu, 10 Dec 2020 22:25:29 +0100, Krzysztof Kozlowski wrote:
> Interrupt line can be configured on different hardware in different way,
> even inverted.  Therefore driver should not enforce specific trigger
> type - edge falling - but instead rely on Devicetree to configure it.
> 
> The Maxim 77686 datasheet describes the interrupt line as active low
> with a requirement of acknowledge from the CPU therefore the edge
> falling is not correct.
> 
> The interrupt line is shared between PMIC and RTC driver, so using level
> sensitive interrupt is here especially important to avoid races.  With
> an edge configuration in case if first PMIC signals interrupt followed
> shortly after by the RTC, the interrupt might not be yet cleared/acked
> thus the second one would not be noticed.
> 
> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
> 
> ---
> 
> This patch should wait till DTS changes are merged, as it relies on
> proper Devicetree.
> ---
>  Documentation/devicetree/bindings/clock/maxim,max77686.txt | 4 ++--
>  Documentation/devicetree/bindings/mfd/max77686.txt         | 2 +-
>  Documentation/devicetree/bindings/regulator/max77686.txt   | 2 +-
>  drivers/mfd/max77686.c                                     | 3 +--
>  4 files changed, 5 insertions(+), 6 deletions(-)
> 

Acked-by: Rob Herring <robh@kernel.org>
Krzysztof Kozlowski March 3, 2021, 6:19 p.m. UTC | #7
On Thu, Dec 10, 2020 at 10:25:17PM +0100, Krzysztof Kozlowski wrote:
> The Maxim fuel gauge datasheets describe the interrupt line as active
> low with a requirement of acknowledge from the CPU.  The falling edge
> interrupt will mostly work but it's not correct.
> 
> Fixes: 8620cc2f99b7 ("ARM: dts: exynos: Add devicetree file for the Galaxy S2")
> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
> ---
>  arch/arm/boot/dts/exynos4210-i9100.dts | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 

Applied 1-10 (Exynos and S5P dts patches).

Best regards,
Krzysztof
diff mbox series

Patch

diff --git a/arch/arm/boot/dts/exynos4210-i9100.dts b/arch/arm/boot/dts/exynos4210-i9100.dts
index a0c3bab382ae..e56b64e237d3 100644
--- a/arch/arm/boot/dts/exynos4210-i9100.dts
+++ b/arch/arm/boot/dts/exynos4210-i9100.dts
@@ -136,7 +136,7 @@  battery@36 {
 			compatible = "maxim,max17042";
 
 			interrupt-parent = <&gpx2>;
-			interrupts = <3 IRQ_TYPE_EDGE_FALLING>;
+			interrupts = <3 IRQ_TYPE_LEVEL_LOW>;
 
 			pinctrl-0 = <&max17042_fuel_irq>;
 			pinctrl-names = "default";