Platform: Brightness quirk for samsung laptop driver

From: Jason Stubbs <[email protected]>

On some samsung laptops the brightness regulation works slightly different. All SABI commands except for set_brightness work as expected. The behaviour of set_brightness is as follows:

- Setting a new brightness will only step one level toward the new brightness level. For example, setting a level of 5 when the current level is 2 will result in a brightness level of 3.
- A spurious KEY_BRIGHTNESS_UP or KEY_BRIGHTNESS_DOWN event is also generated along with the change in brightness.
- Neither of the above two issues occur when changing from/to brightness level 0.

This patch adds detection and a non-intrusive workaround for the above issues.

Signed-off-by David Herrmann <[email protected]>

---
Hi

This patch is originally from Jason Stubbs. See here:
http://git.kernel.org/?p=linux/kernel/git/gregkh/patches.git;a=blob;f=platform-samsung_laptop-add-support-for-samsung-nc210-nc110.patch;hb=HEAD

I just wanted to ask whether there is some work going on here? My laptop model needs this fix and I reviewed and tested it and it works fine. It would be nice to see this upstream.

Quoted 22 lines · David Herrmann

From: Jason Stubbs<[email protected]>

On some samsung laptops the brightness regulation works slightly different. All SABI commands except for set_brightness work as expected. The behaviour of set_brightness is as follows:

- Setting a new brightness will only step one level toward the new brightness level. For example, setting a level of 5 when the current level is 2 will result in a brightness level of 3.
- A spurious KEY_BRIGHTNESS_UP or KEY_BRIGHTNESS_DOWN event is also generated along with the change in brightness.
- Neither of the above two issues occur when changing from/to brightness level 0.

This patch adds detection and a non-intrusive workaround for the above issues.

Signed-off-by David Herrmann<[email protected]>

---
Hi

This patch is originally from Jason Stubbs. See here:
http://git.kernel.org/?p=linux/kernel/git/gregkh/patches.git;a=blob;f=platform-samsung_laptop-add-support-for-samsung-nc210-nc110.patch;hb=HEAD

Quoted 3 lines

I just wanted to ask whether there is some work going on here? My laptop model needs this fix and I reviewed and tested it and it works fine. It would be nice to see this upstream.

I have a different patch from Jason that should fix this properly, I'll send it to Matthew later today.

Quoted 23 lines · Greg KH

Quoted 22 lines · David Herrmann

From: Jason Stubbs<[email protected]>

On some samsung laptops the brightness regulation works slightly different. All SABI commands except for set_brightness work as expected. The behaviour of set_brightness is as follows:

- Setting a new brightness will only step one level toward the new brightness level. For example, setting a level of 5 when the current level is 2 will result in a brightness level of 3.
- A spurious KEY_BRIGHTNESS_UP or KEY_BRIGHTNESS_DOWN event is also generated along with the change in brightness.
- Neither of the above two issues occur when changing from/to brightness level 0.

This patch adds detection and a non-intrusive workaround for the above issues.

Signed-off-by David Herrmann<[email protected]>

---
Hi

This patch is originally from Jason Stubbs. See here:
http://git.kernel.org/?p=linux/kernel/git/gregkh/patches.git;a=blob;f=platform-samsung_laptop-add-support-for-samsung-nc210-nc110.patch;hb=HEAD

Quoted 6 lines

Quoted 3 lines

I just wanted to ask whether there is some work going on here? My laptop model needs this fix and I reviewed and tested it and it works fine. It would be nice to see this upstream.

I have a different patch from Jason that should fix this properly, I'll send it to Matthew later today.

If you're talking about the min/max calculations patch, that fixes a separate issue to this patch. Both patches touch set_brightness() though, which is probably where the confusion came from.

The version of the patch in git doesn't apply on top of the min/max calcs patch that you've added to the gregkh-2.6 tree, but I've previously posted a version that does. You can find it at
https://lkml.org/lkml/2011/5/13/150 or I can post it again.

From: Jason Stubbs <[email protected]>

On some Samsung laptops the brightness regulation works slightly different. All SABI commands except for set_brightness work as expected. The behaviour of set_brightness is as follows:

- Setting a new brightness will only step one level toward the new brightness level. For example, setting a level of 5 when the current level is 2 will result in a brightness level of 3.
- A spurious KEY_BRIGHTNESS_UP or KEY_BRIGHTNESS_DOWN event is also generated along with the change in brightness.
- Neither of the above two issues occur when changing from/to brightness level 0.

This patch adds detection and a non-intrusive workaround for the above issues.

Signed-off-by Jason Stubbs <[email protected]>

---

drivers/platform/x86/samsung-laptop.c +40 -0

index ee68c1c..b193bf0 100644
@@ -226,6 +226,7 @@ static struct backlight_device *backlight_device;
  static struct mutex sabi_mutex;
  static struct platform_device *sdev;
  static struct rfkill *rfk;
+static bool has_stepping_quirk;
 
  static int force;
  module_param(force, bool, 0);

@@ -382,6 +383,17 @@ static void set_brightness(u8 user_brightness)
  {
        u8 user_level = user_brightness + sabi_config->min_brightness;
 
+       if (has_stepping_quirk && user_level != 0) {
+               /*
+                * short circuit if the specified level is what's already set
+                * to prevent the screen from flickering needlessly
+                */
+               if (user_brightness == read_brightness())
+                       return;
+
+               sabi_set_command(sabi_config->commands.set_brightness, 0);
+       }
+
        sabi_set_command(sabi_config->commands.set_brightness, user_level);
  }
 

@@ -390,6 +402,34 @@ static int get_brightness(struct backlight_device *bd)
        return (int)read_brightness();
  }
 
+static void check_for_stepping_quirk(void)
+{
+       u8 initial_level = read_brightness();
+       u8 check_level;
+
+       /*
+        * Some laptops exhibit the strange behaviour of stepping toward
+        * (rather than setting) the brightness except when changing to/from
+        * brightness level 0. This behaviour is checked for here and worked
+        * around in set_brightness.
+        */
+
+       if (initial_level <= 2)
+               check_level = initial_level + 2;
+       else
+               check_level = initial_level - 2;
+
+       has_stepping_quirk = false;
+       set_brightness(check_level);
+
+       if (read_brightness() != check_level) {
+               has_stepping_quirk = true;
+               pr_info("enabled workaround for brightness stepping quirk\n");
+       }
+
+       set_brightness(initial_level);
+}
+
  static int update_status(struct backlight_device *bd)
  {
        set_brightness(bd->props.brightness);

Quoted 88 lines · Jason Stubbs

From: Jason Stubbs <[email protected]>

On some Samsung laptops the brightness regulation works slightly different. All SABI commands except for set_brightness work as expected. The behaviour of set_brightness is as follows:

- Setting a new brightness will only step one level toward the new brightness level. For example, setting a level of 5 when the current level is 2 will result in a brightness level of 3.
- A spurious KEY_BRIGHTNESS_UP or KEY_BRIGHTNESS_DOWN event is also generated along with the change in brightness.
- Neither of the above two issues occur when changing from/to brightness level 0.

This patch adds detection and a non-intrusive workaround for the above issues.

Signed-off-by Jason Stubbs <[email protected]>

---

drivers/platform/x86/samsung-laptop.c +40 -0

index ee68c1c..b193bf0 100644
@@ -226,6 +226,7 @@ static struct backlight_device *backlight_device;
  static struct mutex sabi_mutex;
  static struct platform_device *sdev;
  static struct rfkill *rfk;
+static bool has_stepping_quirk;
 
  static int force;
  module_param(force, bool, 0);

@@ -382,6 +383,17 @@ static void set_brightness(u8 user_brightness)
  {
        u8 user_level = user_brightness + sabi_config->min_brightness;
 
+       if (has_stepping_quirk && user_level != 0) {
+               /*
+                * short circuit if the specified level is what's already set
+                * to prevent the screen from flickering needlessly
+                */
+               if (user_brightness == read_brightness())
+                       return;
+
+               sabi_set_command(sabi_config->commands.set_brightness, 0);
+       }
+
        sabi_set_command(sabi_config->commands.set_brightness, user_level);
  }
 

@@ -390,6 +402,34 @@ static int get_brightness(struct backlight_device *bd)
        return (int)read_brightness();
  }
 
+static void check_for_stepping_quirk(void)
+{
+       u8 initial_level = read_brightness();
+       u8 check_level;
+
+       /*
+        * Some laptops exhibit the strange behaviour of stepping toward
+        * (rather than setting) the brightness except when changing to/from
+        * brightness level 0. This behaviour is checked for here and worked
+        * around in set_brightness.
+        */
+
+       if (initial_level <= 2)
+               check_level = initial_level + 2;
+       else
+               check_level = initial_level - 2;
+
+       has_stepping_quirk = false;
+       set_brightness(check_level);
+
+       if (read_brightness() != check_level) {
+               has_stepping_quirk = true;
+               pr_info("enabled workaround for brightness stepping quirk\n");
+       }
+
+       set_brightness(initial_level);
+}
+
  static int update_status(struct backlight_device *bd)
  {
        set_brightness(bd->props.brightness);

drivers/platform/x86/samsung-laptop.c static int __init samsung_init(void) · line 844

Quoted 9 lines

drivers/platform/x86/samsung-laptop.c +3 -0

@@ -796,6 +836,9 @@ static int __init samsung_init(void)
                }
        }
 
+       /* Check for stepping quirk */
+       check_for_stepping_quirk();
+
        /* knock up a platform device to hang stuff off of */
        sdev = platform_device_register_simple("samsung", -1, NULL, 0);
        if (IS_ERR(sdev))

Thank you, Jason.

Tested-by David Herrmann <[email protected]>

And Greg, could you actually "git add" the new samsung patches in your repository? You added them to "series" but didn't submit the actual patch files. That would be great.