[PATCH v2 0/6] platform/chrome: cros_ec: miscellaneous cleanups

[PATCH v2 0/6] platform/chrome: cros_ec: miscellaneous cleanups

[Tzung-Bi Shih]

The 1st patch fixes unhandled undos in error handling path.

The rest of patches cleans drivers/platform/chrome/cros_ec.c.

Changes from v1[1]:
- Use imperative mood in commit messages.
- Use IS_ERR_OR_NULL() in 1st patch.

[1]: https://lore.kernel.org/lkml/20220125101527.1812887-1-tzungbi@google.com/T/#u

Tzung-Bi Shih (6):
  platform/chrome: cros_ec: fix error handling in cros_ec_register()
  platform/chrome: cros_ec: remove unused variable `was_wake_device`
  platform/chrome: cros_ec: determine `wake_enabled` in
    cros_ec_suspend()
  platform/chrome: cros_ec: don't initialize `err` in cros_ec_register()
  platform/chrome: cros_ec: sort header inclusion alphabetically
  platform/chrome: cros_ec: append newline to all logs

 drivers/platform/chrome/cros_ec.c           | 38 ++++++++++++---------
 include/linux/platform_data/cros_ec_proto.h |  3 --
 2 files changed, 21 insertions(+), 20 deletions(-)

-- 
2.35.0.263.gb82422642f-goog

[PATCH v2 1/6] platform/chrome: cros_ec: fix error handling in cros_ec_register()

[Tzung-Bi Shih]

Fix cros_ec_register() doesn't unregister platform devices if
blocking_notifier_chain_register() fails.

Also use the single exit path to handle the platform device
unregistration.

This fix depends on the fact that all the callers of cros_ec_register()
allocate zeroed memory.

Fixes: 42cd0ab476e2 ("platform/chrome: cros_ec: Query EC protocol version if EC transitions between RO/RW")
Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
---
 drivers/platform/chrome/cros_ec.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/platform/chrome/cros_ec.c b/drivers/platform/chrome/cros_ec.c
index fc5aa1525d13..7ce667ff08e0 100644
--- a/drivers/platform/chrome/cros_ec.c
+++ b/drivers/platform/chrome/cros_ec.c
@@ -245,18 +245,16 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
 		if (IS_ERR(ec_dev->pd)) {
 			dev_err(ec_dev->dev,
 				"Failed to create CrOS PD platform device\n");
-			platform_device_unregister(ec_dev->ec);
-			return PTR_ERR(ec_dev->pd);
+			err = PTR_ERR(ec_dev->pd);
+			goto exit;
 		}
 	}
 
 	if (IS_ENABLED(CONFIG_OF) && dev->of_node) {
 		err = devm_of_platform_populate(dev);
 		if (err) {
-			platform_device_unregister(ec_dev->pd);
-			platform_device_unregister(ec_dev->ec);
 			dev_err(dev, "Failed to register sub-devices\n");
-			return err;
+			goto exit;
 		}
 	}
 
@@ -278,7 +276,7 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
 		err = blocking_notifier_chain_register(&ec_dev->event_notifier,
 						      &ec_dev->notifier_ready);
 		if (err)
-			return err;
+			goto exit;
 	}
 
 	dev_info(dev, "Chrome EC device registered\n");
@@ -291,6 +289,12 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
 		cros_ec_irq_thread(0, ec_dev);
 
 	return 0;
+exit:
+	if (!IS_ERR_OR_NULL(ec_dev->pd))
+		platform_device_unregister(ec_dev->pd);
+	if (!IS_ERR_OR_NULL(ec_dev->ec))
+		platform_device_unregister(ec_dev->ec);
+	return err;
 }
 EXPORT_SYMBOL(cros_ec_register);
 
-- 
2.35.0.263.gb82422642f-goog

[PATCH v2 2/6] platform/chrome: cros_ec: remove unused variable `was_wake_device`

[Tzung-Bi Shih]

`was_wake_device` is unused.  Remove it.

Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
---
 drivers/platform/chrome/cros_ec.c           | 1 -
 include/linux/platform_data/cros_ec_proto.h | 3 ---
 2 files changed, 4 deletions(-)

diff --git a/drivers/platform/chrome/cros_ec.c b/drivers/platform/chrome/cros_ec.c
index 7ce667ff08e0..303af630c03e 100644
--- a/drivers/platform/chrome/cros_ec.c
+++ b/drivers/platform/chrome/cros_ec.c
@@ -344,7 +344,6 @@ int cros_ec_suspend(struct cros_ec_device *ec_dev)
 		ec_dev->wake_enabled = !enable_irq_wake(ec_dev->irq);
 
 	disable_irq(ec_dev->irq);
-	ec_dev->was_wake_device = ec_dev->wake_enabled;
 	ec_dev->suspended = true;
 
 	return 0;
diff --git a/include/linux/platform_data/cros_ec_proto.h b/include/linux/platform_data/cros_ec_proto.h
index df3c78c92ca2..c65971ec90ea 100644
--- a/include/linux/platform_data/cros_ec_proto.h
+++ b/include/linux/platform_data/cros_ec_proto.h
@@ -76,8 +76,6 @@ struct cros_ec_command {
  * struct cros_ec_device - Information about a ChromeOS EC device.
  * @phys_name: Name of physical comms layer (e.g. 'i2c-4').
  * @dev: Device pointer for physical comms device
- * @was_wake_device: True if this device was set to wake the system from
- *                   sleep at the last suspend.
  * @cros_class: The class structure for this device.
  * @cmd_readmem: Direct read of the EC memory-mapped region, if supported.
  *     @offset: Is within EC_LPC_ADDR_MEMMAP region.
@@ -137,7 +135,6 @@ struct cros_ec_device {
 	/* These are used by other drivers that want to talk to the EC */
 	const char *phys_name;
 	struct device *dev;
-	bool was_wake_device;
 	struct class *cros_class;
 	int (*cmd_readmem)(struct cros_ec_device *ec, unsigned int offset,
 			   unsigned int bytes, void *dest);
-- 
2.35.0.263.gb82422642f-goog

[PATCH v2 3/6] platform/chrome: cros_ec: determine `wake_enabled` in cros_ec_suspend()

[Tzung-Bi Shih]

`wake_enabled` indicates cros_ec_resume() needs to call
disable_irq_wake() to undo enable_irq_wake() in cros_ec_suspend().

Determine `wake_enabled` in cros_ec_suspend() instead of
reset-after-used in cros_ec_resume().

Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
---
 drivers/platform/chrome/cros_ec.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/platform/chrome/cros_ec.c b/drivers/platform/chrome/cros_ec.c
index 303af630c03e..a97a47ae0472 100644
--- a/drivers/platform/chrome/cros_ec.c
+++ b/drivers/platform/chrome/cros_ec.c
@@ -342,6 +342,8 @@ int cros_ec_suspend(struct cros_ec_device *ec_dev)
 
 	if (device_may_wakeup(dev))
 		ec_dev->wake_enabled = !enable_irq_wake(ec_dev->irq);
+	else
+		ec_dev->wake_enabled = false;
 
 	disable_irq(ec_dev->irq);
 	ec_dev->suspended = true;
@@ -383,10 +385,9 @@ int cros_ec_resume(struct cros_ec_device *ec_dev)
 		dev_dbg(ec_dev->dev, "Error %d sending resume event to ec",
 			ret);
 
-	if (ec_dev->wake_enabled) {
+	if (ec_dev->wake_enabled)
 		disable_irq_wake(ec_dev->irq);
-		ec_dev->wake_enabled = 0;
-	}
+
 	/*
 	 * Let the mfd devices know about events that occur during
 	 * suspend. This way the clients know what to do with them.
-- 
2.35.0.263.gb82422642f-goog

[PATCH v2 4/6] platform/chrome: cros_ec: don't initialize `err` in cros_ec_register()

[Tzung-Bi Shih]

`err` in cros_ec_register() doesn't need to be initialized.

Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
---
 drivers/platform/chrome/cros_ec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/platform/chrome/cros_ec.c b/drivers/platform/chrome/cros_ec.c
index a97a47ae0472..c2105621b368 100644
--- a/drivers/platform/chrome/cros_ec.c
+++ b/drivers/platform/chrome/cros_ec.c
@@ -182,7 +182,7 @@ static int cros_ec_ready_event(struct notifier_block *nb,
 int cros_ec_register(struct cros_ec_device *ec_dev)
 {
 	struct device *dev = ec_dev->dev;
-	int err = 0;
+	int err;
 
 	BLOCKING_INIT_NOTIFIER_HEAD(&ec_dev->event_notifier);
 
-- 
2.35.0.263.gb82422642f-goog

[PATCH v2 5/6] platform/chrome: cros_ec: sort header inclusion alphabetically

[Tzung-Bi Shih]

Sort header inclusion alphabetically.

Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
---
 drivers/platform/chrome/cros_ec.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/platform/chrome/cros_ec.c b/drivers/platform/chrome/cros_ec.c
index c2105621b368..008dc2fc5066 100644
--- a/drivers/platform/chrome/cros_ec.c
+++ b/drivers/platform/chrome/cros_ec.c
@@ -9,12 +9,12 @@
  * battery charging and regulator control, firmware update.
  */
 
-#include <linux/of_platform.h>
 #include <linux/interrupt.h>
-#include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/of_platform.h>
 #include <linux/platform_data/cros_ec_commands.h>
 #include <linux/platform_data/cros_ec_proto.h>
+#include <linux/slab.h>
 #include <linux/suspend.h>
 
 #include "cros_ec.h"
-- 
2.35.0.263.gb82422642f-goog

[PATCH v2 6/6] platform/chrome: cros_ec: append newline to all logs

[Tzung-Bi Shih]

To be consistent, append newline ("\n") to all logs.

Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
---
 drivers/platform/chrome/cros_ec.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/platform/chrome/cros_ec.c b/drivers/platform/chrome/cros_ec.c
index 008dc2fc5066..426314e4b7f6 100644
--- a/drivers/platform/chrome/cros_ec.c
+++ b/drivers/platform/chrome/cros_ec.c
@@ -213,7 +213,7 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
 						IRQF_TRIGGER_LOW | IRQF_ONESHOT,
 						"chromeos-ec", ec_dev);
 		if (err) {
-			dev_err(dev, "Failed to request IRQ %d: %d",
+			dev_err(dev, "Failed to request IRQ %d: %d\n",
 				ec_dev->irq, err);
 			return err;
 		}
@@ -264,7 +264,7 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
 	 */
 	err = cros_ec_sleep_event(ec_dev, 0);
 	if (err < 0)
-		dev_dbg(ec_dev->dev, "Error %d clearing sleep event to ec",
+		dev_dbg(ec_dev->dev, "Error %d clearing sleep event to ec\n",
 			err);
 
 	if (ec_dev->mkbp_event_supported) {
@@ -337,7 +337,7 @@ int cros_ec_suspend(struct cros_ec_device *ec_dev)
 
 	ret = cros_ec_sleep_event(ec_dev, sleep_event);
 	if (ret < 0)
-		dev_dbg(ec_dev->dev, "Error %d sending suspend event to ec",
+		dev_dbg(ec_dev->dev, "Error %d sending suspend event to ec\n",
 			ret);
 
 	if (device_may_wakeup(dev))
@@ -382,7 +382,7 @@ int cros_ec_resume(struct cros_ec_device *ec_dev)
 
 	ret = cros_ec_sleep_event(ec_dev, sleep_event);
 	if (ret < 0)
-		dev_dbg(ec_dev->dev, "Error %d sending resume event to ec",
+		dev_dbg(ec_dev->dev, "Error %d sending resume event to ec\n",
 			ret);
 
 	if (ec_dev->wake_enabled)
-- 
2.35.0.263.gb82422642f-goog

Re: [PATCH v2 1/6] platform/chrome: cros_ec: fix error handling in cros_ec_register()

[Prashant Malani]

Hi,

On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
>
> Fix cros_ec_register() doesn't unregister platform devices if
> blocking_notifier_chain_register() fails.

This isn't grammatically correct. Instead:
Fix cros_ec_register() to unregister platform devices if blocking....

>
> Also use the single exit path to handle the platform device
> unregistration.
>
> This fix depends on the fact that all the callers of cros_ec_register()
> allocate zeroed memory.

Is that a fair assumption? What happens if a future driver calls
cros_ec_register()
without zeroed memory?

>
> Fixes: 42cd0ab476e2 ("platform/chrome: cros_ec: Query EC protocol version if EC transitions between RO/RW")
> Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
> ---
What has changed since v2? Please add a change log of the versions here.

>  drivers/platform/chrome/cros_ec.c | 16 ++++++++++------
>  1 file changed, 10 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/platform/chrome/cros_ec.c b/drivers/platform/chrome/cros_ec.c
> index fc5aa1525d13..7ce667ff08e0 100644
> --- a/drivers/platform/chrome/cros_ec.c
> +++ b/drivers/platform/chrome/cros_ec.c
> @@ -245,18 +245,16 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
>                 if (IS_ERR(ec_dev->pd)) {
>                         dev_err(ec_dev->dev,
>                                 "Failed to create CrOS PD platform device\n");
> -                       platform_device_unregister(ec_dev->ec);
> -                       return PTR_ERR(ec_dev->pd);
> +                       err = PTR_ERR(ec_dev->pd);
> +                       goto exit;
>                 }
>         }
>
>         if (IS_ENABLED(CONFIG_OF) && dev->of_node) {
>                 err = devm_of_platform_populate(dev);
>                 if (err) {
> -                       platform_device_unregister(ec_dev->pd);
> -                       platform_device_unregister(ec_dev->ec);
>                         dev_err(dev, "Failed to register sub-devices\n");
> -                       return err;
> +                       goto exit;
>                 }
>         }
>
> @@ -278,7 +276,7 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
>                 err = blocking_notifier_chain_register(&ec_dev->event_notifier,
>                                                       &ec_dev->notifier_ready);
>                 if (err)
> -                       return err;
> +                       goto exit;
>         }
>
>         dev_info(dev, "Chrome EC device registered\n");
> @@ -291,6 +289,12 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
>                 cros_ec_irq_thread(0, ec_dev);
>
>         return 0;
> +exit:
> +       if (!IS_ERR_OR_NULL(ec_dev->pd))
> +               platform_device_unregister(ec_dev->pd);
> +       if (!IS_ERR_OR_NULL(ec_dev->ec))
> +               platform_device_unregister(ec_dev->ec);
> +       return err;
You don't need these "if" checks. They are already performed by
platform_device_unregister(), or rather, the functions that it calls [1][2].

[1] https://elixir.bootlin.com/linux/latest/source/drivers/base/platform.c#L756
[2] https://elixir.bootlin.com/linux/latest/source/drivers/base/platform.c#L553

Re: [PATCH v2 2/6] platform/chrome: cros_ec: remove unused variable `was_wake_device`

[Prashant Malani]

Hi,

On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
>
> `was_wake_device` is unused.  Remove it.
nit: This commit message doesn't add anything to the patch (it is
repeating the subject), so it
should be fine to drop it.

>
> Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
Reviewed-by: Prashant Malani <pmalani@chromium.org>
> ---
In the future, please add a change log for versions here. Quoting the
patch submission guide [1]

"One good use for the additional comments after the --- marker
is...Other comments relevant
only to the moment or the maintainer, not suitable for the permanent
changelog, should also
go here. A good example of such comments might be patch changelogs
which describe what
has changed between the v1 and v2 version of the patch."

[1] https://www.kernel.org/doc/html/latest/process/submitting-patches.html#the-canonical-patch-format

Re: [PATCH v2 3/6] platform/chrome: cros_ec: determine `wake_enabled` in cros_ec_suspend()

[Prashant Malani]

Hi,

On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
>
> `wake_enabled` indicates cros_ec_resume() needs to call
> disable_irq_wake() to undo enable_irq_wake() in cros_ec_suspend().
>
> Determine `wake_enabled` in cros_ec_suspend() instead of
> reset-after-used in cros_ec_resume().
It sounds like we can accomplish the same thing as this patch by either:
- Initializing ec_dev->wake_enabled = false during cros_ec_register()
or
- Setting ec_dev->wake_enabled = false just before the
device_may_wakeup(dev) check.

Both of these options accomplish the same thing as this patch, with a
much smaller diff.

Re: [PATCH v2 4/6] platform/chrome: cros_ec: don't initialize `err` in cros_ec_register()

[Prashant Malani]

Hi,

On Tue, Feb 8, 2022 at 8:51 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
>
> `err` in cros_ec_register() doesn't need to be initialized.

What is the issue with it being initialized to 0? This doesn't strike
me as necessary.

Re: [PATCH v2 1/6] platform/chrome: cros_ec: fix error handling in cros_ec_register()

[Tzung-Bi Shih]

On Tue, Feb 08, 2022 at 09:38:55PM -0800, Prashant Malani wrote:
> On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> >
> > Fix cros_ec_register() doesn't unregister platform devices if
> > blocking_notifier_chain_register() fails.
> 
> This isn't grammatically correct. Instead:
> Fix cros_ec_register() to unregister platform devices if blocking....

Thanks, will fix in next version.

> > Also use the single exit path to handle the platform device
> > unregistration.
> >
> > This fix depends on the fact that all the callers of cros_ec_register()
> > allocate zeroed memory.
> 
> Is that a fair assumption? What happens if a future driver calls
> cros_ec_register()
> without zeroed memory?

Suppose ec_dev->pd is garbage data, platform_device_unregister() will operate
on invalid memory address.

Let's use another guard condition in next version instead of relying on
zeroed memory.

> > ---
> What has changed since v2? Please add a change log of the versions here.

They are in the cover letter actually.  Will copy them to each patch.

> > @@ -291,6 +289,12 @@ int cros_ec_register(struct cros_ec_device *ec_dev)
> >                 cros_ec_irq_thread(0, ec_dev);
> >
> >         return 0;
> > +exit:
> > +       if (!IS_ERR_OR_NULL(ec_dev->pd))
> > +               platform_device_unregister(ec_dev->pd);
> > +       if (!IS_ERR_OR_NULL(ec_dev->ec))
> > +               platform_device_unregister(ec_dev->ec);
> > +       return err;
> You don't need these "if" checks. They are already performed by
> platform_device_unregister(), or rather, the functions that it calls [1][2].

Ack, will remove the check in next version.

Re: [PATCH v2 2/6] platform/chrome: cros_ec: remove unused variable `was_wake_device`

[Tzung-Bi Shih]

On Tue, Feb 08, 2022 at 09:58:41PM -0800, Prashant Malani wrote:
> On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> >
> > `was_wake_device` is unused.  Remove it.
> nit: This commit message doesn't add anything to the patch (it is
> repeating the subject), so it
> should be fine to drop it.

Ack.

> > ---
> In the future, please add a change log for versions here. Quoting the
> patch submission guide [1]
> 
> "One good use for the additional comments after the --- marker
> is...Other comments relevant
> only to the moment or the maintainer, not suitable for the permanent
> changelog, should also
> go here. A good example of such comments might be patch changelogs
> which describe what
> has changed between the v1 and v2 version of the patch."
> 
> [1] https://www.kernel.org/doc/html/latest/process/submitting-patches.html#the-canonical-patch-format

They are in the cover letter.  Will copy them to each patch.

Re: [PATCH v2 3/6] platform/chrome: cros_ec: determine `wake_enabled` in cros_ec_suspend()

[Tzung-Bi Shih]

On Tue, Feb 08, 2022 at 10:05:58PM -0800, Prashant Malani wrote:
> On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> >
> > `wake_enabled` indicates cros_ec_resume() needs to call
> > disable_irq_wake() to undo enable_irq_wake() in cros_ec_suspend().
> >
> > Determine `wake_enabled` in cros_ec_suspend() instead of
> > reset-after-used in cros_ec_resume().
> It sounds like we can accomplish the same thing as this patch by either:
> - Initializing ec_dev->wake_enabled = false during cros_ec_register()

This doesn't sound like a good idea.  Value of device_may_wakeup(dev) changes
during runtime.  It should check in cros_ec_suspend() instead of in
cros_ec_register().

> or
> - Setting ec_dev->wake_enabled = false just before the
> device_may_wakeup(dev) check.

Did you mean:

ec_dev->wake_enabled = false;
if (device_may_wakeup(dev))
        ec_dev->wake_enabled = !enable_irq_wake(ec_dev->irq);

If so, this way is suboptimal to me.  ec_dev->wake_enabled can be written
once; however, it is written twice if device_may_wakeup(dev) is true.

Re: [PATCH v2 4/6] platform/chrome: cros_ec: don't initialize `err` in cros_ec_register()

[Tzung-Bi Shih]

On Tue, Feb 08, 2022 at 10:08:59PM -0800, Prashant Malani wrote:
> On Tue, Feb 8, 2022 at 8:51 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> >
> > `err` in cros_ec_register() doesn't need to be initialized.
> 
> What is the issue with it being initialized to 0? This doesn't strike
> me as necessary.

Although it might be optimized out, the initialization is redundant.  `err` is
going to be overridden soon at [1].

[1]: https://elixir.bootlin.com/linux/v5.17-rc3/source/drivers/platform/chrome/cros_ec.c#L203

The patch makes the code concise and reduces the code size (hopefully!).  It's
fine to drop the change if the code wouldn't bother you.

Re: [PATCH v2 3/6] platform/chrome: cros_ec: determine `wake_enabled` in cros_ec_suspend()

[Prashant Malani]

On Wed, Feb 9, 2022 at 1:36 AM Tzung-Bi Shih <tzungbi@google.com> wrote:
>
> On Tue, Feb 08, 2022 at 10:05:58PM -0800, Prashant Malani wrote:
> > On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> > >
> > > `wake_enabled` indicates cros_ec_resume() needs to call
> > > disable_irq_wake() to undo enable_irq_wake() in cros_ec_suspend().
> > >
> > > Determine `wake_enabled` in cros_ec_suspend() instead of
> > > reset-after-used in cros_ec_resume().
> > It sounds like we can accomplish the same thing as this patch by either:
> > - Initializing ec_dev->wake_enabled = false during cros_ec_register()
>
> This doesn't sound like a good idea.  Value of device_may_wakeup(dev) changes
> during runtime.  It should check in cros_ec_suspend() instead of in
> cros_ec_register().

Hmm, I'm pretty sure it shouldn't change for Cros EC. In any case, I'm
not suggesting
moving the check away from cros_ec_suspend(), just the initialization
of the flag.

>
> > or
> > - Setting ec_dev->wake_enabled = false just before the
> > device_may_wakeup(dev) check.
>
> Did you mean:
>
> ec_dev->wake_enabled = false;
> if (device_may_wakeup(dev))
>         ec_dev->wake_enabled = !enable_irq_wake(ec_dev->irq);
>
> If so, this way is suboptimal to me.  ec_dev->wake_enabled can be written
> once; however, it is written twice if device_may_wakeup(dev) is true.

I don't think it is a concern on any modern computer hardware, and
seems more readable to me, but
it's up to you.

Re: [PATCH v2 3/6] platform/chrome: cros_ec: determine `wake_enabled` in cros_ec_suspend()

[Tzung-Bi Shih]

On Tue, Feb 15, 2022 at 05:07:43PM -0800, Prashant Malani wrote:
> On Wed, Feb 9, 2022 at 1:36 AM Tzung-Bi Shih <tzungbi@google.com> wrote:
> >
> > On Tue, Feb 08, 2022 at 10:05:58PM -0800, Prashant Malani wrote:
> > > On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> > > >
> > > > `wake_enabled` indicates cros_ec_resume() needs to call
> > > > disable_irq_wake() to undo enable_irq_wake() in cros_ec_suspend().
> > > >
> > > > Determine `wake_enabled` in cros_ec_suspend() instead of
> > > > reset-after-used in cros_ec_resume().
> > > It sounds like we can accomplish the same thing as this patch by either:
> > > - Initializing ec_dev->wake_enabled = false during cros_ec_register()
> >
> > This doesn't sound like a good idea.  Value of device_may_wakeup(dev) changes
> > during runtime.  It should check in cros_ec_suspend() instead of in
> > cros_ec_register().
> 
> Hmm, I'm pretty sure it shouldn't change for Cros EC. In any case, I'm
> not suggesting
> moving the check away from cros_ec_suspend(), just the initialization
> of the flag.

Got it.  I misunderstood.  Let's go this direction.  Will fix in the next
version.

Re: [PATCH v2 3/6] platform/chrome: cros_ec: determine `wake_enabled` in cros_ec_suspend()

[Prashant Malani]

On Tue, Feb 15, 2022 at 8:13 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
>
> On Tue, Feb 15, 2022 at 05:07:43PM -0800, Prashant Malani wrote:
> > On Wed, Feb 9, 2022 at 1:36 AM Tzung-Bi Shih <tzungbi@google.com> wrote:
> > >
> > > On Tue, Feb 08, 2022 at 10:05:58PM -0800, Prashant Malani wrote:
> > > > On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> > > > >
> > > > > `wake_enabled` indicates cros_ec_resume() needs to call
> > > > > disable_irq_wake() to undo enable_irq_wake() in cros_ec_suspend().
> > > > >
> > > > > Determine `wake_enabled` in cros_ec_suspend() instead of
> > > > > reset-after-used in cros_ec_resume().
> > > > It sounds like we can accomplish the same thing as this patch by either:
> > > > - Initializing ec_dev->wake_enabled = false during cros_ec_register()
> > >
> > > This doesn't sound like a good idea.  Value of device_may_wakeup(dev) changes
> > > during runtime.  It should check in cros_ec_suspend() instead of in
> > > cros_ec_register().
Actually, circling back to this (apologies), can you provide an
example of this changing during runtime?
I'm not aware of one, but I could be omitting some use cases that you
might be familiar with.

> >
> > Hmm, I'm pretty sure it shouldn't change for Cros EC. In any case, I'm
> > not suggesting
> > moving the check away from cros_ec_suspend(), just the initialization
> > of the flag.
>
> Got it.  I misunderstood.  Let's go this direction.  Will fix in the next
> version.

Re: [PATCH v2 3/6] platform/chrome: cros_ec: determine `wake_enabled` in cros_ec_suspend()

[Tzung-Bi Shih]

On Tue, Feb 15, 2022 at 09:55:13PM -0800, Prashant Malani wrote:
> On Tue, Feb 15, 2022 at 8:13 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> >
> > On Tue, Feb 15, 2022 at 05:07:43PM -0800, Prashant Malani wrote:
> > > On Wed, Feb 9, 2022 at 1:36 AM Tzung-Bi Shih <tzungbi@google.com> wrote:
> > > >
> > > > On Tue, Feb 08, 2022 at 10:05:58PM -0800, Prashant Malani wrote:
> > > > > On Tue, Feb 8, 2022 at 8:50 PM Tzung-Bi Shih <tzungbi@google.com> wrote:
> > > > > >
> > > > > > `wake_enabled` indicates cros_ec_resume() needs to call
> > > > > > disable_irq_wake() to undo enable_irq_wake() in cros_ec_suspend().
> > > > > >
> > > > > > Determine `wake_enabled` in cros_ec_suspend() instead of
> > > > > > reset-after-used in cros_ec_resume().
> > > > > It sounds like we can accomplish the same thing as this patch by either:
> > > > > - Initializing ec_dev->wake_enabled = false during cros_ec_register()
> > > >
> > > > This doesn't sound like a good idea.  Value of device_may_wakeup(dev) changes
> > > > during runtime.  It should check in cros_ec_suspend() instead of in
> > > > cros_ec_register().
> Actually, circling back to this (apologies), can you provide an
> example of this changing during runtime?
> I'm not aware of one, but I could be omitting some use cases that you
> might be familiar with.

If someone `echo disabled > /sys/devices/.../spi0/spi0.0/power/wakeup`, the
value of device_may_wakeup(dev) would change.