mail archive of the barebox mailing list
 help / color / mirror / Atom feed
* [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply
@ 2019-12-03 13:48 Stefan Riedmueller
  2019-12-05  8:07 ` Sascha Hauer
  0 siblings, 1 reply; 6+ messages in thread
From: Stefan Riedmueller @ 2019-12-03 13:48 UTC (permalink / raw)
  To: barebox

There is no driver for the eMMC's vmmc-supply regulator in the barebox.
Use a dummy regulator instead by simply deleting the vmmc-supply
property.

Signed-off-by: Stefan Riedmueller <s.riedmueller@phytec.de>
---
 arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
index 69f252b42382..974e271f453d 100644
--- a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
+++ b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
@@ -112,6 +112,7 @@
 };
 
 &usdhc4 {
+	/delete-property/ vmmc-supply;
 	partitions {
 		compatible = "fixed-partitions";
 		#address-cells = <1>;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply
  2019-12-03 13:48 [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply Stefan Riedmueller
@ 2019-12-05  8:07 ` Sascha Hauer
  2019-12-05  8:37   ` Ahmad Fatoum
  0 siblings, 1 reply; 6+ messages in thread
From: Sascha Hauer @ 2019-12-05  8:07 UTC (permalink / raw)
  To: Stefan Riedmueller; +Cc: barebox

Hi Stefan,

On Tue, Dec 03, 2019 at 02:48:41PM +0100, Stefan Riedmueller wrote:
> There is no driver for the eMMC's vmmc-supply regulator in the barebox.
> Use a dummy regulator instead by simply deleting the vmmc-supply
> property.
> 
> Signed-off-by: Stefan Riedmueller <s.riedmueller@phytec.de>
> ---
>  arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
> index 69f252b42382..974e271f453d 100644
> --- a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
> +++ b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
> @@ -112,6 +112,7 @@
>  };
>  
>  &usdhc4 {
> +	/delete-property/ vmmc-supply;

I don't like this approach very much. It's fine for barebox, but leads
to problems once you start Linux with the barebox device tree. I'd
rather go with the unmodified device trees (apart from the things
barebox changes to the device trees in order to start Linux).

One idea that comes to my mind is: Would it be possible to provide a
call like

int regulator_provide_dummy(struct device_node *np);

Boards could call this to let barebox provide a dummy regulator for the
given node (the "dlg,da9062" node in your case) and its child nodes.
In of_regulator_get() we then see if the desired regulator is a child
node of a node registered as dummy provider, then register a dummy
regulator and return it.

Sascha

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply
  2019-12-05  8:07 ` Sascha Hauer
@ 2019-12-05  8:37   ` Ahmad Fatoum
  2019-12-05 13:08     ` Stefan Riedmüller
  0 siblings, 1 reply; 6+ messages in thread
From: Ahmad Fatoum @ 2019-12-05  8:37 UTC (permalink / raw)
  To: barebox

Hello,

On 12/5/19 9:07 AM, Sascha Hauer wrote:
> On Tue, Dec 03, 2019 at 02:48:41PM +0100, Stefan Riedmueller wrote:
>> There is no driver for the eMMC's vmmc-supply regulator in the barebox.
>> Use a dummy regulator instead by simply deleting the vmmc-supply
>> property.
>>
>> Signed-off-by: Stefan Riedmueller <s.riedmueller@phytec.de>
>> ---
>>  arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
>> index 69f252b42382..974e271f453d 100644
>> --- a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
>> +++ b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
>> @@ -112,6 +112,7 @@
>>  };
>>  
>>  &usdhc4 {
>> +	/delete-property/ vmmc-supply;
> 
> I don't like this approach very much. It's fine for barebox, but leads
> to problems once you start Linux with the barebox device tree. I'd
> rather go with the unmodified device trees (apart from the things
> barebox changes to the device trees in order to start Linux).

How about a barebox,regulator-always-on property? If the regulator is unnecessary,
it basically means you can assume it's always on when barebox runs.

If regulator_get is called with a property that points to such a node, a dummy
regulator is returned with which enable is a no-op. Additionally, do the same
for regulator-always-on, so the in-barebox device tree only needs to mark
non regulator-always-on that are expected to be always on when it runs due to
strapping/fuses as barebox,regulator-always-on.

Thoughts?
Ahmad

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply
  2019-12-05  8:37   ` Ahmad Fatoum
@ 2019-12-05 13:08     ` Stefan Riedmüller
  2019-12-05 13:31       ` Ahmad Fatoum
  0 siblings, 1 reply; 6+ messages in thread
From: Stefan Riedmüller @ 2019-12-05 13:08 UTC (permalink / raw)
  To: barebox

Hi,

On 05.12.19 09:37, Ahmad Fatoum wrote:
> Hello,
> 
> On 12/5/19 9:07 AM, Sascha Hauer wrote:
>> On Tue, Dec 03, 2019 at 02:48:41PM +0100, Stefan Riedmueller wrote:
>>> There is no driver for the eMMC's vmmc-supply regulator in the barebox.
>>> Use a dummy regulator instead by simply deleting the vmmc-supply
>>> property.
>>>
>>> Signed-off-by: Stefan Riedmueller <s.riedmueller@phytec.de>
>>> ---
>>>   arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi | 1 +
>>>   1 file changed, 1 insertion(+)
>>>
>>> diff --git a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
>>> index 69f252b42382..974e271f453d 100644
>>> --- a/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
>>> +++ b/arch/arm/dts/imx6qdl-phytec-phycore-som.dtsi
>>> @@ -112,6 +112,7 @@
>>>   };
>>>   
>>>   &usdhc4 {
>>> +	/delete-property/ vmmc-supply;
>>
>> I don't like this approach very much. It's fine for barebox, but leads
>> to problems once you start Linux with the barebox device tree. I'd
>> rather go with the unmodified device trees (apart from the things
>> barebox changes to the device trees in order to start Linux).
> 
> How about a barebox,regulator-always-on property? If the regulator is unnecessary,
> it basically means you can assume it's always on when barebox runs.
> 
> If regulator_get is called with a property that points to such a node, a dummy
> regulator is returned with which enable is a no-op. Additionally, do the same
> for regulator-always-on, so the in-barebox device tree only needs to mark
> non regulator-always-on that are expected to be always on when it runs due to
> strapping/fuses as barebox,regulator-always-on.

What about regulators that are marked as regulator-always-on but need 
voltage adjustments? Or isn't this a real use case for the bootloader?

Regards,
Stefan

> 
> Thoughts?
> Ahmad
> 

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply
  2019-12-05 13:08     ` Stefan Riedmüller
@ 2019-12-05 13:31       ` Ahmad Fatoum
  2019-12-06 14:23         ` Sascha Hauer
  0 siblings, 1 reply; 6+ messages in thread
From: Ahmad Fatoum @ 2019-12-05 13:31 UTC (permalink / raw)
  To: barebox

Hi,

On 12/5/19 2:08 PM, Stefan Riedmüller wrote:
> What about regulators that are marked as regulator-always-on but need voltage adjustments? Or isn't this a real use case for the bootloader?

If someone absolutely need the voltage adjustments, they'll have to implement
a proper regulator driver to realize it[1]. The [barebox,]regulator-always-on
handling I suggest, is only a fallback if there is no regulator found.


Cheers
Ahmad

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | https://www.pengutronix.de/ |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

^ permalink raw reply	[flat|nested] 6+ messages in thread
* Re: [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply
  2019-12-05 13:31       ` Ahmad Fatoum
@ 2019-12-06 14:23         ` Sascha Hauer
  0 siblings, 0 replies; 6+ messages in thread
From: Sascha Hauer @ 2019-12-06 14:23 UTC (permalink / raw)
  To: Ahmad Fatoum; +Cc: barebox

On Thu, Dec 05, 2019 at 02:31:53PM +0100, Ahmad Fatoum wrote:
> Hi,
> 
> On 12/5/19 2:08 PM, Stefan Riedmüller wrote:
> > What about regulators that are marked as regulator-always-on but need voltage adjustments? Or isn't this a real use case for the bootloader?
> 
> If someone absolutely need the voltage adjustments, they'll have to implement
> a proper regulator driver to realize it[1]. The [barebox,]regulator-always-on
> handling I suggest, is only a fallback if there is no regulator found.

What if the driver is present but not yet probed? We would return a
dummy regulator and no longer -EPROBE_DEFER.

Sascha


-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

^ permalink raw reply	[flat|nested] 6+ messages in thread
end of thread, other threads:[~2019-12-06 14:23 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-12-03 13:48 [PATCH] ARM: dts: imx6qdl: phycore: Remove emmc vmmc-supply Stefan Riedmueller
2019-12-05  8:07 ` Sascha Hauer
2019-12-05  8:37   ` Ahmad Fatoum
2019-12-05 13:08     ` Stefan Riedmüller
2019-12-05 13:31       ` Ahmad Fatoum
2019-12-06 14:23         ` Sascha Hauer

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox