mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH] hwmon: applesmc: restore keyboard backlight on resume
@ 2009-08-22 21:33 Henrik Rydberg
  2009-08-24 21:07 ` Andrew Morton
  0 siblings, 1 reply; 4+ messages in thread
From: Henrik Rydberg @ 2009-08-22 21:33 UTC (permalink / raw)
  To: Andrew Morton; +Cc: nicolas, lm-sensors, linux-kernel, Henrik Rydberg

On resume from suspend, the driver currently resets the logical
state as if it was brought up from halt. This patch uses the
dev_pm_ops.resume method to synchronize the hardware with the
memorized logical state, bringing back the backlight to the level
prior to suspend.

Signed-off-by: Henrik Rydberg <rydberg@euromail.se>
---
 drivers/hwmon/applesmc.c |   27 +++++++++++++++------------
 1 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/drivers/hwmon/applesmc.c b/drivers/hwmon/applesmc.c
index 678e34b..0b781c8 100644
--- a/drivers/hwmon/applesmc.c
+++ b/drivers/hwmon/applesmc.c
@@ -178,6 +178,8 @@ static const int debug;
 static struct platform_device *pdev;
 static s16 rest_x;
 static s16 rest_y;
+static u8 backlight_state[2];
+
 static struct device *hwmon_dev;
 static struct input_polled_dev *applesmc_idev;
 
@@ -497,17 +499,25 @@ static int applesmc_probe(struct platform_device *dev)
 	return 0;
 }
 
-static int applesmc_resume(struct platform_device *dev)
+/* Synchronize device with memorized backlight state */
+static int applesmc_pm_resume(struct device *dev)
 {
-	return applesmc_device_init();
+	mutex_lock(&applesmc_lock);
+	applesmc_write_key(BACKLIGHT_KEY, backlight_state, 2);
+	mutex_unlock(&applesmc_lock);
+	return 0;
 }
 
+static struct dev_pm_ops applesmc_pm_ops = {
+	.resume = applesmc_pm_resume,
+};
+
 static struct platform_driver applesmc_driver = {
 	.probe = applesmc_probe,
-	.resume = applesmc_resume,
 	.driver	= {
 		.name = "applesmc",
 		.owner = THIS_MODULE,
+		.pm = &applesmc_pm_ops,
 	},
 };
 
@@ -804,17 +814,10 @@ static ssize_t applesmc_calibrate_store(struct device *dev,
 	return count;
 }
 
-/* Store the next backlight value to be written by the work */
-static unsigned int backlight_value;
-
 static void applesmc_backlight_set(struct work_struct *work)
 {
-	u8 buffer[2];
-
 	mutex_lock(&applesmc_lock);
-	buffer[0] = backlight_value;
-	buffer[1] = 0x00;
-	applesmc_write_key(BACKLIGHT_KEY, buffer, 2);
+	applesmc_write_key(BACKLIGHT_KEY, backlight_state, 2);
 	mutex_unlock(&applesmc_lock);
 }
 static DECLARE_WORK(backlight_work, &applesmc_backlight_set);
@@ -824,7 +827,7 @@ static void applesmc_brightness_set(struct led_classdev *led_cdev,
 {
 	int ret;
 
-	backlight_value = value;
+	backlight_state[0] = value;
 	ret = queue_work(applesmc_led_wq, &backlight_work);
 
 	if (debug && (!ret))
-- 
1.6.3.3


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] hwmon: applesmc: restore keyboard backlight on resume
  2009-08-22 21:33 [PATCH] hwmon: applesmc: restore keyboard backlight on resume Henrik Rydberg
@ 2009-08-24 21:07 ` Andrew Morton
  2009-08-24 21:49   ` Henrik Rydberg
  2009-08-24 23:37   ` Henrik Rydberg
  0 siblings, 2 replies; 4+ messages in thread
From: Andrew Morton @ 2009-08-24 21:07 UTC (permalink / raw)
  To: Henrik Rydberg; +Cc: nicolas, lm-sensors, linux-kernel, rydberg

On Sat, 22 Aug 2009 23:33:52 +0200
"Henrik Rydberg" <rydberg@euromail.se> wrote:

> On resume from suspend, the driver currently resets the logical
> state as if it was brought up from halt. This patch uses the
> dev_pm_ops.resume method to synchronize the hardware with the
> memorized logical state, bringing back the backlight to the level
> prior to suspend.

What are the runtime effects of this change?

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] hwmon: applesmc: restore keyboard backlight on resume
  2009-08-24 21:07 ` Andrew Morton
@ 2009-08-24 21:49   ` Henrik Rydberg
  2009-08-24 23:37   ` Henrik Rydberg
  1 sibling, 0 replies; 4+ messages in thread
From: Henrik Rydberg @ 2009-08-24 21:49 UTC (permalink / raw)
  To: Andrew Morton; +Cc: nicolas, lm-sensors, linux-kernel

Andrew Morton wrote:
> On Sat, 22 Aug 2009 23:33:52 +0200
> "Henrik Rydberg" <rydberg@euromail.se> wrote:
> 
>> On resume from suspend, the driver currently resets the logical
>> state as if it was brought up from halt. This patch uses the
>> dev_pm_ops.resume method to synchronize the hardware with the
>> memorized logical state, bringing back the backlight to the level
>> prior to suspend.
> 
> What are the runtime effects of this change?

The patch has zero impact on a running system, only suspend/resume
behavior is affected.

Precondition: Suspend computer with keyboard backlight brightness
greater than zero.

Before patch: On resume, the backlight stays off.

After patch: On resume, the backlight is set to the brightness prior to
suspend.

There is one side-effect of the patch: the accelerometer no longer
re-calibrates on resume, but keeps the calibration from before suspend.
I consider this a good change, but it is after all a change.

Henrik


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] hwmon: applesmc: restore keyboard backlight on resume
  2009-08-24 21:07 ` Andrew Morton
  2009-08-24 21:49   ` Henrik Rydberg
@ 2009-08-24 23:37   ` Henrik Rydberg
  1 sibling, 0 replies; 4+ messages in thread
From: Henrik Rydberg @ 2009-08-24 23:37 UTC (permalink / raw)
  To: Andrew Morton; +Cc: Henrik Rydberg, nicolas, lm-sensors, linux-kernel

Andrew Morton wrote:
> On Sat, 22 Aug 2009 23:33:52 +0200
> "Henrik Rydberg" <rydberg@euromail.se> wrote:
> 
>> On resume from suspend, the driver currently resets the logical
>> state as if it was brought up from halt. This patch uses the
>> dev_pm_ops.resume method to synchronize the hardware with the
>> memorized logical state, bringing back the backlight to the level
>> prior to suspend.
> 
> What are the runtime effects of this change?
> 

My apologies, the patch has two flaws.

1) For macbook models without keyboard backlight, the resume function
will fail internally, presumably resulting in a small unnecessary delay.

2) The accelerometer does not reset properly after hibernation.

Please drop this patch from -mm, I am sending a corrected patch
with a slightly different title. Sorry about this.

Henrik

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2009-08-24 23:48 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-08-22 21:33 [PATCH] hwmon: applesmc: restore keyboard backlight on resume Henrik Rydberg
2009-08-24 21:07 ` Andrew Morton
2009-08-24 21:49   ` Henrik Rydberg
2009-08-24 23:37   ` Henrik Rydberg

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

all inboxes | Powered by JetHome®