thermal: tango: add resume support

From: Marc Gonzalez <[email protected]>

When this platform is suspended, firmware powers the entire SoC down, except a few hardware blocks waiting for wakeup events. And there is no context to save for this particular block.

Therefore, there is nothing useful for the driver to do on suspend; so we define a NULL suspend hook. On resume, the driver initializes the block exactly as is done in the probe callback.

Reviewed-by Kevin Hilman <[email protected]>

Signed-off-by Marc Gonzalez <[email protected]>

---
Add Kevin's Reviewed-by tag. Eduardo/Zhang, can you pick this patch up for 4.8?

---
 drivers/thermal/tango_thermal.c | 27 +++++++++++++++++++++++++--
 1 file changed, 25 insertions(+), 2 deletions(-)

drivers/thermal/tango_thermal.c +8 -2

index 70e0d9f406e9..d571ce2f546d 100644
@@ -64,6 +64,12 @@ 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;

@@ -79,14 +85,30 @@ static int tango_thermal_probe(struct platform_device *pdev)
 	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);

Quoted 8 lines

drivers/thermal/tango_thermal.c +8 -0

+
+     return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm, NULL, tango_thermal_resume);
+
+#define DEV_PM_OPS   &tango_thermal_pm
+#else

drivers/thermal/tango_thermal.c static int tango_thermal_probe(struct platform_device *pdev) · line 110

Quoted 2 lines

drivers/thermal/tango_thermal.c +2 -0

+#define DEV_PM_OPS   NULL
+#endif

In my experience it's often not useful to #ifdef the struct pm_ops. These days you almost certainly want PM enabled, and the conditional doesn't save you all that much in the first place, because it's not unlikely for this to fit into some of the space that would be padded out anyway.

As a side-note, I've noticed that this driver has the following dependencies:

        depends on ARCH_TANGO || COMPILE_TEST

which, last I checked, is probably going to fail on some architectures because you need at least another one on HAS_IOMEM (for readl() and writel()). That's a pre-existing problem, of course, so should be fixed in a separate patch.

Quoted 8 lines · CEST Thierry Reding

Quoted 8 lines

drivers/thermal/tango_thermal.c +8 -0

+
+     return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm, NULL, tango_thermal_resume);
+
+#define DEV_PM_OPS   &tango_thermal_pm
+#else

Quoted 8 lines

drivers/thermal/tango_thermal.c static int tango_thermal_probe(struct platform_device *pdev) · line 110

Quoted 2 lines

drivers/thermal/tango_thermal.c +2 -0

+#define DEV_PM_OPS   NULL
+#endif

In my experience it's often not useful to #ifdef the struct pm_ops. These days you almost certainly want PM enabled, and the conditional doesn't save you all that much in the first place, because it's not unlikely for this to fit into some of the space that would be padded out anyway.

This will also generate a warning when CONFIG_PM_SLEEP is not set. Better write this as

#define DEV_PM_OPS (IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm : NULL)

so the compiler can drop the variable definition when it's not needed.

Quoted 5 lines

As a side-note, I've noticed that this driver has the following dependencies:

        depends on ARCH_TANGO || COMPILE_TEST

Quoted 4 lines

which, last I checked, is probably going to fail on some architectures because you need at least another one on HAS_IOMEM (for readl() and writel()). That's a pre-existing problem, of course, so should be fixed in a separate patch.

No need, we just merged a patch to no longer allow COMPILE_TEST on arch/um/, so we can safely rely on MMIO to be available for COMPILE_TEST.

Quoted 24 lines · Arnd Bergmann

Quoted 8 lines · CEST Thierry Reding

Quoted 8 lines

drivers/thermal/tango_thermal.c +8 -0

+
+     return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm, NULL, tango_thermal_resume);
+
+#define DEV_PM_OPS   &tango_thermal_pm
+#else

Quoted 8 lines

drivers/thermal/tango_thermal.c static int tango_thermal_probe(struct platform_device *pdev) · line 110

Quoted 2 lines

drivers/thermal/tango_thermal.c +2 -0

+#define DEV_PM_OPS   NULL
+#endif

In my experience it's often not useful to #ifdef the struct pm_ops. These days you almost certainly want PM enabled, and the conditional doesn't save you all that much in the first place, because it's not unlikely for this to fit into some of the space that would be padded out anyway.

This will also generate a warning when CONFIG_PM_SLEEP is not set. Better write this as

#define DEV_PM_OPS (IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm : NULL)

Quoted 2 lines

so the compiler can drop the variable definition when it's not needed.

My suggestion was to define tango_thermal_pm unconditionally to avoid any of these tricks. For any real use-case in which the 92 bytes for the struct dev_pm_ops would matter you most likely want PM_SLEEP anyway, so I don't really see why we would even want to make it optional.

Quoted 10 lines

Quoted 9 lines

As a side-note, I've noticed that this driver has the following dependencies:

        depends on ARCH_TANGO || COMPILE_TEST

which, last I checked, is probably going to fail on some architectures because you need at least another one on HAS_IOMEM (for readl() and writel()). That's a pre-existing problem, of course, so should be fixed in a separate patch.

Quoted 2 lines

No need, we just merged a patch to no longer allow COMPILE_TEST on arch/um/, so we can safely rely on MMIO to be available for COMPILE_TEST.

I thought at least S390 didn't have readl() and writel() either, at least when PCI wasn't enabled, or some such.

Quoted 25 lines · CEST Thierry Reding

Quoted 24 lines · Arnd Bergmann

Quoted 8 lines · CEST Thierry Reding

Quoted 8 lines

drivers/thermal/tango_thermal.c +8 -0

+
+     return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm, NULL, tango_thermal_resume);
+
+#define DEV_PM_OPS   &tango_thermal_pm
+#else

Quoted 8 lines

drivers/thermal/tango_thermal.c static int tango_thermal_probe(struct platform_device *pdev) · line 110

Quoted 2 lines

drivers/thermal/tango_thermal.c +2 -0

+#define DEV_PM_OPS   NULL
+#endif

In my experience it's often not useful to #ifdef the struct pm_ops. These days you almost certainly want PM enabled, and the conditional doesn't save you all that much in the first place, because it's not unlikely for this to fit into some of the space that would be padded out anyway.

This will also generate a warning when CONFIG_PM_SLEEP is not set. Better write this as

#define DEV_PM_OPS (IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm : NULL)

Quoted 7 lines

Quoted 2 lines

so the compiler can drop the variable definition when it's not needed.

My suggestion was to define tango_thermal_pm unconditionally to avoid any of these tricks. For any real use-case in which the 92 bytes for the struct dev_pm_ops would matter you most likely want PM_SLEEP anyway, so I don't really see why we would even want to make it optional.

Sure, leaving it unconditional works too.

Quoted 10 lines

Quoted 10 lines

Quoted 9 lines

As a side-note, I've noticed that this driver has the following dependencies:

        depends on ARCH_TANGO || COMPILE_TEST

which, last I checked, is probably going to fail on some architectures because you need at least another one on HAS_IOMEM (for readl() and writel()). That's a pre-existing problem, of course, so should be fixed in a separate patch.

Quoted 5 lines

Quoted 2 lines

No need, we just merged a patch to no longer allow COMPILE_TEST on arch/um/, so we can safely rely on MMIO to be available for COMPILE_TEST.

I thought at least S390 didn't have readl() and writel() either, at least when PCI wasn't enabled, or some such.

Yes, but they've never complained about COMPILE_TEST breakage because of that. Tile is in the same boat too in some configurations.

From: Marc Gonzalez <[email protected]>

When this platform is suspended, firmware powers the entire SoC down, except a few hardware blocks waiting for wakeup events. And there is no context to save for this particular block.

Therefore, there is nothing useful for the driver to do on suspend; so we define a NULL suspend hook. On resume, the driver initializes the block exactly as is done in the probe callback.

Signed-off-by Marc Gonzalez <[email protected]>

---
Changes in v3
Drop the DEV_PM_OPS macro, which caused some confusion.
Note: tango_thermal_pm struct only exists when CONFIG_PM_SLEEP is defined @Kevin, can I add your Reviewed-by tag back?

---
 drivers/thermal/tango_thermal.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

From: Marc Gonzalez <[email protected]>

When this platform is suspended, firmware powers the entire SoC down, except a few hardware blocks waiting for wakeup events. And there is no context to save for this particular block.

Therefore, there is nothing useful for the driver to do on suspend; so we define a NULL suspend hook. On resume, the driver initializes the block exactly as is done in the probe callback.

Signed-off-by Marc Gonzalez <[email protected]>

---
Changes in v3
Drop the DEV_PM_OPS macro, which caused some confusion.
Note: tango_thermal_pm struct only exists when CONFIG_PM_SLEEP is defined @Kevin, can I add your Reviewed-by tag back?

---
 drivers/thermal/tango_thermal.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

drivers/thermal/tango_thermal.c +8 -2

index 70e0d9f406e9..957f8937e126 100644
@@ -64,6 +64,12 @@ 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;

@@ -79,14 +85,24 @@ static int tango_thermal_probe(struct platform_device *pdev)
 	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);
 }

drivers/thermal/tango_thermal.c static int tango_thermal_probe(struct platform_device *pdev) · line 104

Quoted 9 lines · Mason

drivers/thermal/tango_thermal.c +9 -0

+#ifdef CONFIG_PM_SLEEP
+static int tango_thermal_resume(struct device *dev)
+{
+	tango_thermal_init(dev_get_drvdata(dev));
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm, NULL, tango_thermal_resume);
+#endif

#else
#define tango_thermal_resume NULL
#endif

And then move the SIMPLE_DEV_PM_OPS here...

Quoted 6 lines

drivers/thermal/tango_thermal.c +1 -0

+
 static const struct of_device_id tango_sensor_ids[] = {
 	{
 		.compatible = "sigma,smp8758-thermal",

@@ -99,6 +115,9 @@ static struct platform_driver tango_thermal_driver = {
 	.driver	= {

drivers/thermal/tango_thermal.c static struct platform_driver tango_thermal_driver = { · line 120

Quoted 5 lines

drivers/thermal/tango_thermal.c +3 -0

 		.name		= "tango-thermal",
 		.of_match_table	= tango_sensor_ids,
+#ifdef CONFIG_PM_SLEEP
+		.pm		= &tango_thermal_pm,
+#endif

... which allows you to get rid of the ugly ifdef here. (c.f. CodingStyle, Chapter 20,)

Quoted 2 lines · Kevin Hilman

Mason wrote:

Quoted 5 lines

Quoted 5 lines

drivers/thermal/tango_thermal.c +5 -0

+#ifdef CONFIG_PM_SLEEP
+static int tango_thermal_resume(struct device *dev)
+{
+    tango_thermal_init(dev_get_drvdata(dev));
+    return 0;

Quoted 8 lines

drivers/thermal/tango_thermal.c static int tango_thermal_probe(struct platform_device *pdev) · line 104

Quoted 4 lines

drivers/thermal/tango_thermal.c +4 -0

+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm, NULL, tango_thermal_resume);
+#endif

#else
#define tango_thermal_resume NULL
#endif

Quoted 1 line

And then move the SIMPLE_DEV_PM_OPS here...

Moving the SIMPLE_DEV_PM_OPS macro outside the CONFIG_PM_SLEEP guard would unconditionally define a struct dev_pm_ops, which just wastes space when CONFIG_PM_SLEEP is undefined (if I'm not mistaken).

That's why I put SIMPLE_DEV_PM_OPS inside the CONFIG_PM_SLEEP guard.

Quoted 6 lines

Quoted 6 lines

drivers/thermal/tango_thermal.c +1 -0

+
 static const struct of_device_id tango_sensor_ids[] = {
     {
             .compatible = "sigma,smp8758-thermal",

@@ -99,6 +115,9 @@ static struct platform_driver tango_thermal_driver = {
     .driver = {

Quoted 8 lines

drivers/thermal/tango_thermal.c static struct platform_driver tango_thermal_driver = { · line 120

Quoted 5 lines

drivers/thermal/tango_thermal.c +3 -0

             .name           = "tango-thermal",
             .of_match_table = tango_sensor_ids,
+#ifdef CONFIG_PM_SLEEP
+            .pm             = &tango_thermal_pm,
+#endif

... which allows you to get rid of the ugly ifdef here. (c.f. CodingStyle, Chapter 20,)

The previous solution (v2) avoided duplicating the ifdef block, but Arnd and Thierry objected to the code. Later, they agreed that it should work; but if they didn't catch the code's intent at a glance, maybe there is a problem with it anyway.

What do you think?

I copied the DEV_PM_OPS "trick" from drivers/thermal/ti-soc-thermal/ti-bandgap.c (which was done by Eduardo, a thermal maintainer, so maybe he prefers that solution after all? commit 8feaf0ce1a043)

Quoted 4 lines · CEST Mason

Moving the SIMPLE_DEV_PM_OPS macro outside the CONFIG_PM_SLEEP guard would unconditionally define a struct dev_pm_ops, which just wastes space when CONFIG_PM_SLEEP is undefined (if I'm not mistaken).

Quoted 1 line

That's why I put SIMPLE_DEV_PM_OPS inside the CONFIG_PM_SLEEP guard.

If you want to avoid the extra few bytes, just use the trick I suggested:

.pm = IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm : NULL,

You should basically never have that #ifdef inside of the platform_driver definition.

Quoted 6 lines · Arnd Bergmann

Quoted 4 lines · CEST Mason

Moving the SIMPLE_DEV_PM_OPS macro outside the CONFIG_PM_SLEEP guard would unconditionally define a struct dev_pm_ops, which just wastes space when CONFIG_PM_SLEEP is undefined (if I'm not mistaken).

Quoted 6 lines

Quoted 1 line

That's why I put SIMPLE_DEV_PM_OPS inside the CONFIG_PM_SLEEP guard.

If you want to avoid the extra few bytes, just use the trick I suggested:

.pm = IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm : NULL,

This would achieve the same result as the solution I proposed in my v2 patch, right?

So you're saying you prefer the IS_ENABLED macro over using #ifdef ... #else define stuff as NULL #endif

Did I get that right?

Eduardo, Zhang, what do thermal maintainers prefer?

Quoted 2 lines

You should basically never have that #ifdef inside of the platform_driver definition.

Except when the fields don't exist, like the bug I introduced in struct smp_operations (which you fixed).

Quoted 23 lines · CEST Mason

Quoted 10 lines · Arnd Bergmann

Quoted 7 lines · CEST Mason

Moving the SIMPLE_DEV_PM_OPS macro outside the CONFIG_PM_SLEEP guard
would unconditionally define a struct dev_pm_ops, which just wastes
space when CONFIG_PM_SLEEP is undefined (if I'm not mistaken).

Quoted 7 lines

Quoted 2 lines

That's why I put SIMPLE_DEV_PM_OPS inside the CONFIG_PM_SLEEP guard.

If you want to avoid the extra few bytes, just use the trick I suggested:

??????.pm = IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm :
NULL,

This would achieve the same result as the solution I proposed in my v2 patch, right?

Quoted 4 lines

So you're saying you prefer the IS_ENABLED macro over using #ifdef ... #else define stuff as NULL #endif

Did I get that right?

Yes, but I'd also prefer not to hide the operations structure at all and just rely on the __maybe_unused (ideally) or #ifdef (not as good, but commonly used) to leave out the functions.

Quoted 8 lines

Eduardo, Zhang, what do thermal maintainers prefer?

Quoted 3 lines

You should basically never have that #ifdef inside of the platform_driver definition.

Except when the fields don't exist, like the bug I introduced in struct smp_operations (which you fixed).

Right.

Arnd

Quoted 28 lines · Arnd Bergmann

Quoted 27 lines · CEST Mason

Quoted 10 lines · Arnd Bergmann

Quoted 7 lines · CEST Mason

Moving the SIMPLE_DEV_PM_OPS macro outside the CONFIG_PM_SLEEP guard
would unconditionally define a struct dev_pm_ops, which just wastes
space when CONFIG_PM_SLEEP is undefined (if I'm not mistaken).

Quoted 7 lines

Quoted 2 lines

That's why I put SIMPLE_DEV_PM_OPS inside the CONFIG_PM_SLEEP guard.

If you want to avoid the extra few bytes, just use the trick I suggested:

.pm = IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm :
NULL,

This would achieve the same result as the solution I proposed in my v2 patch, right?

So you're saying you prefer the IS_ENABLED macro over using #ifdef ... #else define stuff as NULL #endif

Did I get that right?

Quoted 5 lines

Yes, but I'd also prefer not to hide the operations structure at all and just rely on the __maybe_unused (ideally) or #ifdef (not as good, but commonly used) to leave out the functions.

IMO, the typical way is to use #ifdef for the pm callbacks, and leave SIMPLE_DEV_PM_OPS outside the #ifdef.
For example, drivers/ata/ahci_imx.c.

Quoted 44 lines · CEST Zhang Rui

Quoted 42 lines · Arnd Bergmann

Quoted 28 lines · CEST Mason

Quoted 18 lines · Arnd Bergmann

Quoted 11 lines · CEST Mason

Moving the SIMPLE_DEV_PM_OPS macro outside the CONFIG_PM_SLEEP
guard
would unconditionally define a struct dev_pm_ops, which just
wastes
space when CONFIG_PM_SLEEP is undefined (if I'm not mistaken).

Quoted 3 lines

That's why I put SIMPLE_DEV_PM_OPS inside the
CONFIG_PM_SLEEP
guard.

Quoted 6 lines

If you want to avoid the extra few bytes, just use the trick I
suggested:

??????.pm = IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm :
NULL,

Quoted 7 lines

This would achieve the same result as the solution I proposed in my v2 patch, right?

So you're saying you prefer the IS_ENABLED macro over using #ifdef ... #else define stuff as NULL #endif

Did I get that right?

Yes, but I'd also prefer not to hide the operations structure at all and just rely on the __maybe_unused (ideally) or #ifdef (not as good, but commonly used) to leave out the functions.

Quoted 5 lines

IMO, the typical way is to use #ifdef for the pm callbacks, and leave
SIMPLE_DEV_PM_OPS outside the #ifdef.
For example, drivers/ata/ahci_imx.c.

Lots of drivers do it like that, the main downside I see is that a lot of them also get it wrong and use incorrect #ifdef guards, either checking the wrong Kconfig symbol,

or hiding the wrong
subset of functions.

Quoted 50 lines · Arnd Bergmann

Quoted 49 lines · CEST Zhang Rui

Quoted 33 lines · Arnd Bergmann

Quoted 31 lines · CEST Mason

Quoted 18 lines · Arnd Bergmann

Quoted 11 lines · CEST Mason

Moving the SIMPLE_DEV_PM_OPS macro outside the CONFIG_PM_SLEEP
guard
would unconditionally define a struct dev_pm_ops, which just
wastes
space when CONFIG_PM_SLEEP is undefined (if I'm not mistaken).

Quoted 3 lines

That's why I put SIMPLE_DEV_PM_OPS inside the
CONFIG_PM_SLEEP
guard.

Quoted 6 lines

If you want to avoid the extra few bytes, just use the trick I
suggested:

.pm = IS_ENABLED(CONFIG_PM_SLEEP) ? &tango_thermal_pm :
NULL,

This would achieve the same result as the solution I proposed in my v2 patch, right?

Quoted 9 lines

Quoted 4 lines

So you're saying you prefer the IS_ENABLED macro over using #ifdef ... #else define stuff as NULL #endif

Did I get that right?

Yes, but I'd also prefer not to hide the operations structure at all and just rely on the __maybe_unused (ideally) or #ifdef (not as good, but commonly used) to leave out the functions.

IMO, the typical way is to use #ifdef for the pm callbacks, and leave
SIMPLE_DEV_PM_OPS outside the #ifdef.
For example, drivers/ata/ahci_imx.c.

Quoted 3 lines

Lots of drivers do it like that, the main downside I see is that a lot of them also get it wrong and use incorrect #ifdef guards, either checking the wrong Kconfig symbol,

This also happens when IS_ENABLED macro is used.

Quoted 3 lines

or hiding the wrong
subset of functions.

This also sounds a driver bug to me, and the driver should get fixed. For us, it's not a problem if we do it right here, right? :)

I think the ideal is to have only one set of conditionals in each driver, so at least you don't get a mismatch between them.

SIMPLE_DEV_PM_OPS uses conditional evaluation (but does not show the reference to the compiler). Annotating the functions as __maybe_unused lets the compiler decide for itself if they should be dropped or not, which means we use only the conditional inside of SIMPLE_DEV_PM_OPS.

Ideally, SIMPLE_DEV_PM_OPS itself should have used an IS_ENABLED() check instead of the #ifdef, that would have made it possible to just leave the function always defined with no __maybe_unused, but still have it dropped from the object code without a warning when there is no runtime reference.

When this platform is suspended, firmware powers the entire SoC down, except a few hardware blocks waiting for wakeup events. There is no context to save for this particular block.

Therefore, there is nothing useful for the driver to do on suspend; so we define a NULL suspend hook. On resume, the driver initializes the block exactly as is done in the probe callback.

Signed-off-by Marc Gonzalez <[email protected]>

---
Earlier versions of this patch were reviewed by Kevin and Thierry. The current form is preferred by Arnd.

---
 drivers/thermal/tango_thermal.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

drivers/thermal/tango_thermal.c +17 -2

index 70e0d9f406e9..201304aeafeb 100644
@@ -64,6 +64,12 @@ 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;

@@ -79,14 +85,22 @@ static int tango_thermal_probe(struct platform_device *pdev)
 	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 __maybe_unused tango_thermal_resume(struct device *dev)
+{
+	tango_thermal_init(dev_get_drvdata(dev));
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(tango_thermal_pm, NULL, tango_thermal_resume);
+
 static const struct of_device_id tango_sensor_ids[] = {
 	{
 		.compatible = "sigma,smp8758-thermal",

@@ -99,6 +113,7 @@ static struct platform_driver tango_thermal_driver = {
 	.driver	= {
 		.name		= "tango-thermal",
 		.of_match_table	= tango_sensor_ids,
+		.pm		= &tango_thermal_pm,
 	},
 };