All of lore.kernel.org
 help / color / mirror / Atom feed
* [RFC] Adding PM support to thermal driver
@ 2016-06-15 16:14 Mason
  2016-06-15 21:51 ` Kevin Hilman
  0 siblings, 1 reply; 3+ messages in thread
From: Mason @ 2016-06-15 16:14 UTC (permalink / raw)
  To: Kevin Hilman, Eduardo Valentin, Javi Merino
  Cc: linux-pm, Rafael J. Wysocki, Sebastian Frias

Hello,

I need to support suspend-to-RAM in my platform.
This platform powers down on suspend, so the contents of device registers are lost.
Can you tell me if this patch looks OK for adding S2R support?
(I would then make a formal submission.)


Regards.


diff --git a/drivers/thermal/tango_thermal.c b/drivers/thermal/tango_thermal.c
index 70e0d9f406e9..29ef9f2f1b93 100644
--- a/drivers/thermal/tango_thermal.c
+++ b/drivers/thermal/tango_thermal.c
@@ -62,10 +62,16 @@ static int tango_get_temp(void *arg, int *res)
 
 static const struct thermal_zone_of_device_ops ops = {
 	.get_temp	= tango_get_temp,
 };
 
+static void tango_thermal_init(struct tango_thermal_priv *priv)
+{
+	writel(0, priv->base + TEMPSI_CFG);
+	writel(CMD_ON, priv->base + TEMPSI_CMD);
+}
+
 static int tango_thermal_probe(struct platform_device *pdev)
 {
 	struct resource *res;
 	struct tango_thermal_priv *priv;
 	struct thermal_zone_device *tzdev;
@@ -77,18 +83,28 @@ static int tango_thermal_probe(struct platform_device *pdev)
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	priv->base = devm_ioremap_resource(&pdev->dev, res);
 	if (IS_ERR(priv->base))
 		return PTR_ERR(priv->base);
 
+	platform_set_drvdata(pdev, priv);
 	priv->thresh_idx = IDX_MIN;
-	writel(0, priv->base + TEMPSI_CFG);
-	writel(CMD_ON, priv->base + TEMPSI_CMD);
+	tango_thermal_init(priv);
 
 	tzdev = devm_thermal_zone_of_sensor_register(&pdev->dev, 0, priv, &ops);
 	return PTR_ERR_OR_ZERO(tzdev);
 }
 
+static int tango_thermal_resume(struct device *dev)
+{
+	struct tango_thermal_priv *priv = dev_get_drvdata(dev);
+	tango_thermal_init(priv);
+
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm_ops, NULL, tango_thermal_resume);
+
 static const struct of_device_id tango_sensor_ids[] = {
 	{
 		.compatible = "sigma,smp8758-thermal",
 	},
 	{ /* sentinel */ }
@@ -97,10 +113,11 @@ static const struct of_device_id tango_sensor_ids[] = {
 static struct platform_driver tango_thermal_driver = {
 	.probe	= tango_thermal_probe,
 	.driver	= {
 		.name		= "tango-thermal",
 		.of_match_table	= tango_sensor_ids,
+		.pm		= &tango_thermal_pm_ops,
 	},
 };
 
 module_platform_driver(tango_thermal_driver);
 

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [RFC] Adding PM support to thermal driver
  2016-06-15 16:14 [RFC] Adding PM support to thermal driver Mason
@ 2016-06-15 21:51 ` Kevin Hilman
  2016-06-17 11:11   ` Mason
  0 siblings, 1 reply; 3+ messages in thread
From: Kevin Hilman @ 2016-06-15 21:51 UTC (permalink / raw)
  To: Mason
  Cc: Eduardo Valentin, Javi Merino, linux-pm, Rafael J. Wysocki,
	Sebastian Frias

Mason <slash.tmp@free.fr> writes:

> I need to support suspend-to-RAM in my platform.
> This platform powers down on suspend, so the contents of device registers are lost.
> Can you tell me if this patch looks OK for adding S2R support?

The idea looks right for ensuring a clean resume after context loss.

However, I'm a bit surprised to see an empty ->suspend hook though.  Is
there nothing to do to gracefully shutdown this hardware?  A quick
glance suggests maybe CMD_OFF should be sent?

Kevin

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [RFC] Adding PM support to thermal driver
  2016-06-15 21:51 ` Kevin Hilman
@ 2016-06-17 11:11   ` Mason
  0 siblings, 0 replies; 3+ messages in thread
From: Mason @ 2016-06-17 11:11 UTC (permalink / raw)
  To: Kevin Hilman
  Cc: Eduardo Valentin, Javi Merino, linux-pm, Rafael J. Wysocki,
	Sebastian Frias, Thibaud Cornic

On 15/06/2016 23:51, Kevin Hilman wrote:

> Mason writes:
> 
>> I need to support suspend-to-RAM in my platform.
>> This platform powers down on suspend, so the contents of device registers are lost.
>> Can you tell me if this patch looks OK for adding S2R support?
> 
> The idea looks right for ensuring a clean resume after context loss.

Thanks for your quick review.

> However, I'm a bit surprised to see an empty ->suspend hook though.
> Is there nothing to do to gracefully shutdown this hardware?  A quick
> glance suggests maybe CMD_OFF should be sent?

We discussed this internally. When Linux asks to be suspended to RAM,
the firmware cuts the power to the entire platform (except RAM obviously,
and a few blocks listening for wakeup events).

So it seems pointless to write code to power individual HW blocks down
that are about to be powered down globally.

(NB: the thermal block dissipates ~50 microwatts, so powering it down
ASAP saves only a negligible amount of energy.)

Regards.


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2016-06-17 11:12 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-06-15 16:14 [RFC] Adding PM support to thermal driver Mason
2016-06-15 21:51 ` Kevin Hilman
2016-06-17 11:11   ` Mason

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.