Skip to contentSkip to conversation

drm/i915: Fix eDP blank screen after S3 resume on HP desktops

PATCH Superseded Takashi Iwai 2 participants 8 messages Jun 20, 2012 – Jun 21, 2012

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/i915: Fix eDP blank screen after S3 resume on HP desktops

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Tested with 3.5-rc3 kernel.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

---
 drivers/gpu/drm/i915/i915_drv.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/i915/i915_drv.c +3 -0
index 9fe9ebe..2ba1350 100644
@@ -658,6 +658,9 @@ static int i915_drm_thaw(struct drm_device *dev)
 
 	intel_opregion_init(dev);
 
+	if (dev_priv->backlight)
+		backlight_update_status(dev_priv->backlight);
+
 	dev_priv->modeset_on_lid = 0;
 
 	console_lock();
Signature · 1 line
1.7.10.4
#1 lore raw
Quoted 8 lines · Takashi Iwai

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Quoted 5 lines

Tested with 3.5-rc3 kernel.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

This patch looks very fishy as-is, simply because I don't like adding random calls to random functions at pretty much random places without any hint why it works. Just a few weeks ago we've had tons of fun with writes that don't stick when the ring isn't yet set up. Can you please check with a bunch of printks what exactly happens to the value of BLC_PWM_CPU_CTL? I.e. whether the write doesn't stick or whether it gets reset somewhere between the register restore functions and the new place.

I suspect the opregion_init call could allow the bios to frob with these registers. If that's the case, the patch is missing a comment that to that effect.

Essentially I want to know why this place here works and why not move the call a few lines up or down.
-Daniel

Quoted 25 lines
---
 drivers/gpu/drm/i915/i915_drv.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/i915/i915_drv.c +3 -0
index 9fe9ebe..2ba1350 100644
@@ -658,6 +658,9 @@ static int i915_drm_thaw(struct drm_device *dev)
 
 	intel_opregion_init(dev);
 
+	if (dev_priv->backlight)
+		backlight_update_status(dev_priv->backlight);
+
 	dev_priv->modeset_on_lid = 0;
 
 	console_lock();
Signature · 6 lines
1.7.10.4

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter
Mail: [email protected]
Mobile: +41 (0)79 365 57 48
#2 lore raw

At Wed, 20 Jun 2012 10:05:12 +0200,

Quoted 12 lines · Daniel Vetter
Quoted 10 lines · Takashi Iwai

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Tested with 3.5-rc3 kernel.

Quoted 7 lines
Quoted 3 lines

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

This patch looks very fishy as-is, simply because I don't like adding random calls to random functions at pretty much random places without any hint why it works.

It's difficult to know exactly why it works. We can only guess, but nothing more than that, just like between husband and wife :)

Quoted 5 lines

Just a few weeks ago we've had tons of fun with writes
that don't stick when the ring isn't yet set up. Can you please check with a bunch of printks what exactly happens to the value of BLC_PWM_CPU_CTL? I.e. whether the write doesn't stick or whether it gets reset somewhere between the register restore functions and the new place.

The register value is already set, and not touched after that. I've already checked.

Quoted 3 lines

I suspect the opregion_init call could allow the bios to frob with these registers. If that's the case, the patch is missing a comment that to that effect.

Right. That's also my suspect.

Quoted 2 lines

Essentially I want to know why this place here works and why not move the call a few lines up or down.

A few lines down works definitely. Even just writing to sysfs works. A few lines up, well, I need to check where is the border line.

Takashi

Quoted 33 lines

-Daniel

Quoted 25 lines
---
 drivers/gpu/drm/i915/i915_drv.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/i915/i915_drv.c +3 -0
index 9fe9ebe..2ba1350 100644
@@ -658,6 +658,9 @@ static int i915_drm_thaw(struct drm_device *dev)
 
 	intel_opregion_init(dev);
 
+	if (dev_priv->backlight)
+		backlight_update_status(dev_priv->backlight);
+
 	dev_priv->modeset_on_lid = 0;
 
 	console_lock();
Signature · 6 lines
1.7.10.4

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter
Mail: [email protected]
Mobile: +41 (0)79 365 57 48
#3 lore raw
Quoted 16 lines · Takashi Iwai

At Wed, 20 Jun 2012 10:05:12 +0200,

Quoted 14 lines · Daniel Vetter
Quoted 12 lines · Takashi Iwai

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Tested with 3.5-rc3 kernel.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

This patch looks very fishy as-is, simply because I don't like adding random calls to random functions at pretty much random places without any hint why it works.

It's difficult to know exactly why it works. We can only guess, but nothing more than that, just like between husband and wife :)

Yeah I know, but I've been burnt a bit recently with too much cargo-culting. Hence why I'm so dense about this stuff ;-)

Quoted 8 lines
Quoted 5 lines

Just a few weeks ago we've had tons of fun with writes
that don't stick when the ring isn't yet set up. Can you please check with a bunch of printks what exactly happens to the value of BLC_PWM_CPU_CTL? I.e. whether the write doesn't stick or whether it gets reset somewhere between the register restore functions and the new place.

The register value is already set, and not touched after that. I've already checked.

So immediately before the backlight_update_status call we have the _correct_ value in BLC_PWM_CPU_CTL? That would be indeed puzzling ...

Quoted 11 lines
Quoted 3 lines

I suspect the opregion_init call could allow the bios to frob with these registers. If that's the case, the patch is missing a comment that to that effect.

Right. That's also my suspect.

Quoted 2 lines

Essentially I want to know why this place here works and why not move the call a few lines up or down.

A few lines down works definitely. Even just writing to sysfs works. A few lines up, well, I need to check where is the border line.

Thanks, that would be interesting. Especially if it's opregion that we need to have set up before restoring the backlight properly. -Daniel

Quoted 37 lines

Takashi

Quoted 33 lines

-Daniel

Quoted 25 lines
---
 drivers/gpu/drm/i915/i915_drv.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/i915/i915_drv.c +3 -0
index 9fe9ebe..2ba1350 100644
@@ -658,6 +658,9 @@ static int i915_drm_thaw(struct drm_device *dev)
 
 	intel_opregion_init(dev);
 
+	if (dev_priv->backlight)
+		backlight_update_status(dev_priv->backlight);
+
 	dev_priv->modeset_on_lid = 0;
 
 	console_lock();
Signature · 6 lines
1.7.10.4

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter
Mail: [email protected]
Mobile: +41 (0)79 365 57 48
Signature · 3 lines
Daniel Vetter
Mail: [email protected]
Mobile: +41 (0)79 365 57 48
#4 lore raw

At Wed, 20 Jun 2012 11:36:51 +0200,

Quoted 36 lines · Daniel Vetter
Quoted 16 lines · Takashi Iwai

At Wed, 20 Jun 2012 10:05:12 +0200,

Quoted 14 lines · Daniel Vetter
Quoted 12 lines · Takashi Iwai

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Tested with 3.5-rc3 kernel.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

This patch looks very fishy as-is, simply because I don't like adding random calls to random functions at pretty much random places without any hint why it works.

It's difficult to know exactly why it works. We can only guess, but nothing more than that, just like between husband and wife :)

Yeah I know, but I've been burnt a bit recently with too much cargo-culting. Hence why I'm so dense about this stuff ;-)

Quoted 6 lines
Quoted 5 lines

Just a few weeks ago we've had tons of fun with writes
that don't stick when the ring isn't yet set up. Can you please check with a bunch of printks what exactly happens to the value of BLC_PWM_CPU_CTL? I.e. whether the write doesn't stick or whether it gets reset somewhere between the register restore functions and the new place.

Quoted 5 lines
Quoted 2 lines

The register value is already set, and not touched after that. I've already checked.

So immediately before the backlight_update_status call we have the _correct_ value in BLC_PWM_CPU_CTL? That would be indeed puzzling ...

Yes.

Quoted 6 lines
Quoted 6 lines
Quoted 3 lines

I suspect the opregion_init call could allow the bios to frob with these registers. If that's the case, the patch is missing a comment that to that effect.

Right. That's also my suspect.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Essentially I want to know why this place here works and why not move the call a few lines up or down.

A few lines down works definitely. Even just writing to sysfs works. A few lines up, well, I need to check where is the border line.

Thanks, that would be interesting. Especially if it's opregion that we need to have set up before restoring the backlight properly.

This is getting really puzzling. It's not about the opregion. Just some order of the register restoration.

Essentially the patch below fixes the problem. But I have absolutely no idea why.

Takashi

---

drivers/gpu/drm/i915/i915_suspend.c +1 -1
index 0ede02a..0f89dd0 100644
@@ -740,8 +740,8 @@ static void i915_restore_display(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		I915_WRITE(BLC_PWM_PCH_CTL1, dev_priv->saveBLC_PWM_CTL);
 		I915_WRITE(BLC_PWM_PCH_CTL2, dev_priv->saveBLC_PWM_CTL2);
-		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(BLC_PWM_CPU_CTL2, dev_priv->saveBLC_CPU_PWM_CTL2);
+		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(PCH_PP_ON_DELAYS, dev_priv->savePP_ON_DELAYS);
 		I915_WRITE(PCH_PP_OFF_DELAYS, dev_priv->savePP_OFF_DELAYS);
 		I915_WRITE(PCH_PP_DIVISOR, dev_priv->savePP_DIVISOR);
#5 lore raw

At Wed, 20 Jun 2012 11:39:51 +0200,

Quoted 59 lines · Takashi Iwai

At Wed, 20 Jun 2012 11:36:51 +0200,

Quoted 36 lines · Daniel Vetter
Quoted 16 lines · Takashi Iwai

At Wed, 20 Jun 2012 10:05:12 +0200,

Quoted 14 lines · Daniel Vetter
Quoted 12 lines · Takashi Iwai

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Tested with 3.5-rc3 kernel.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

This patch looks very fishy as-is, simply because I don't like adding random calls to random functions at pretty much random places without any hint why it works.

It's difficult to know exactly why it works. We can only guess, but nothing more than that, just like between husband and wife :)

Yeah I know, but I've been burnt a bit recently with too much cargo-culting. Hence why I'm so dense about this stuff ;-)

Quoted 6 lines
Quoted 5 lines

Just a few weeks ago we've had tons of fun with writes
that don't stick when the ring isn't yet set up. Can you please check with a bunch of printks what exactly happens to the value of BLC_PWM_CPU_CTL? I.e. whether the write doesn't stick or whether it gets reset somewhere between the register restore functions and the new place.

Quoted 5 lines
Quoted 2 lines

The register value is already set, and not touched after that. I've already checked.

So immediately before the backlight_update_status call we have the _correct_ value in BLC_PWM_CPU_CTL? That would be indeed puzzling ...

Yes.

Quoted 12 lines
Quoted 11 lines
Quoted 3 lines

I suspect the opregion_init call could allow the bios to frob with these registers. If that's the case, the patch is missing a comment that to that effect.

Right. That's also my suspect.

Quoted 2 lines

Essentially I want to know why this place here works and why not move the call a few lines up or down.

A few lines down works definitely. Even just writing to sysfs works. A few lines up, well, I need to check where is the border line.

Quoted 8 lines
Quoted 2 lines

Thanks, that would be interesting. Especially if it's opregion that we need to have set up before restoring the backlight properly.

This is getting really puzzling. It's not about the opregion. Just some order of the register restoration.

Essentially the patch below fixes the problem. But I have absolutely no idea why.

Daniel, do you have any clue about this?

Is it a strict order of these registers?

Takashi

Quoted 18 lines

Takashi

---

drivers/gpu/drm/i915/i915_suspend.c +1 -1
index 0ede02a..0f89dd0 100644
@@ -740,8 +740,8 @@ static void i915_restore_display(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		I915_WRITE(BLC_PWM_PCH_CTL1, dev_priv->saveBLC_PWM_CTL);
 		I915_WRITE(BLC_PWM_PCH_CTL2, dev_priv->saveBLC_PWM_CTL2);
-		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(BLC_PWM_CPU_CTL2, dev_priv->saveBLC_CPU_PWM_CTL2);
+		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(PCH_PP_ON_DELAYS, dev_priv->savePP_ON_DELAYS);
 		I915_WRITE(PCH_PP_OFF_DELAYS, dev_priv->savePP_OFF_DELAYS);
 		I915_WRITE(PCH_PP_DIVISOR, dev_priv->savePP_DIVISOR);
#6 lore raw
Quoted 67 lines · Takashi Iwai

At Wed, 20 Jun 2012 11:39:51 +0200,

Quoted 65 lines · Takashi Iwai

At Wed, 20 Jun 2012 11:36:51 +0200,

Quoted 36 lines · Daniel Vetter
Quoted 16 lines · Takashi Iwai

At Wed, 20 Jun 2012 10:05:12 +0200,

Quoted 14 lines · Daniel Vetter
Quoted 12 lines · Takashi Iwai

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Tested with 3.5-rc3 kernel.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

This patch looks very fishy as-is, simply because I don't like adding random calls to random functions at pretty much random places without any hint why it works.

It's difficult to know exactly why it works. We can only guess, but nothing more than that, just like between husband and wife :)

Yeah I know, but I've been burnt a bit recently with too much cargo-culting. Hence why I'm so dense about this stuff ;-)

Quoted 6 lines
Quoted 5 lines

Just a few weeks ago we've had tons of fun with writes
that don't stick when the ring isn't yet set up. Can you please check with a bunch of printks what exactly happens to the value of BLC_PWM_CPU_CTL? I.e. whether the write doesn't stick or whether it gets reset somewhere between the register restore functions and the new place.

Quoted 5 lines
Quoted 2 lines

The register value is already set, and not touched after that. I've already checked.

So immediately before the backlight_update_status call we have the _correct_ value in BLC_PWM_CPU_CTL? That would be indeed puzzling ...

Yes.

Quoted 6 lines
Quoted 6 lines
Quoted 3 lines

I suspect the opregion_init call could allow the bios to frob with these registers. If that's the case, the patch is missing a comment that to that effect.

Right. That's also my suspect.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Essentially I want to know why this place here works and why not move the call a few lines up or down.

A few lines down works definitely. Even just writing to sysfs works. A few lines up, well, I need to check where is the border line.

Thanks, that would be interesting. Especially if it's opregion that we need to have set up before restoring the backlight properly.

This is getting really puzzling. It's not about the opregion. Just some order of the register restoration.

Quoted 6 lines
Quoted 2 lines

Essentially the patch below fixes the problem. But I have absolutely no idea why.

Daniel, do you have any clue about this?

Is it a strict order of these registers?

Could very well be that the write does not get forwarded internally when the BLC stuff isn't enabled yet. Or the other way round. E.g. the new pipe select stuff for the blc clock I've implemented just recently only works while the backlight is off.

I'm much happier with this diff than the older one to just hit the regs harder, so can you please bake this into a proper patch? And please add a 1-line comment to the code that for some unknown reason we need to write the _CTL reg after the _CTL2.

Also, I loathe this save/restore register dance. It duplicates code we already have, but with the twist that we might accidentally get the ordering (or some timing constraint) slightly wrong. Resulting in broken s/r, at least on some machines. Imo we should try to wean off these in the long term.

Thanks, Daniel

Quoted 23 lines

Takashi

Quoted 18 lines

Takashi

---

drivers/gpu/drm/i915/i915_suspend.c +1 -1
index 0ede02a..0f89dd0 100644
@@ -740,8 +740,8 @@ static void i915_restore_display(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		I915_WRITE(BLC_PWM_PCH_CTL1, dev_priv->saveBLC_PWM_CTL);
 		I915_WRITE(BLC_PWM_PCH_CTL2, dev_priv->saveBLC_PWM_CTL2);
-		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(BLC_PWM_CPU_CTL2, dev_priv->saveBLC_CPU_PWM_CTL2);
+		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(PCH_PP_ON_DELAYS, dev_priv->savePP_ON_DELAYS);
 		I915_WRITE(PCH_PP_OFF_DELAYS, dev_priv->savePP_OFF_DELAYS);
 		I915_WRITE(PCH_PP_DIVISOR, dev_priv->savePP_DIVISOR);
Signature · 3 lines
Daniel Vetter
Mail: [email protected]
Mobile: +41 (0)79 365 57 48
#7 lore raw

At Thu, 21 Jun 2012 14:33:07 +0200,

Quoted 72 lines · Daniel Vetter
Quoted 70 lines · Takashi Iwai

At Wed, 20 Jun 2012 11:39:51 +0200,

Quoted 67 lines · Takashi Iwai

At Wed, 20 Jun 2012 11:36:51 +0200,

Quoted 36 lines · Daniel Vetter
Quoted 16 lines · Takashi Iwai

At Wed, 20 Jun 2012 10:05:12 +0200,

Quoted 14 lines · Daniel Vetter
Quoted 12 lines · Takashi Iwai

This patch fixes the problem on some HP desktop machines with eDP which give blank screens after S3 resume.

The problem looks like a timing issue. Although BLC_PWM_CPU_CTL register is already restored at the beginning of resume, it doesn't seem to take effect. Simply re-issuing the register write restores the backlight gracefully.

Tested with 3.5-rc3 kernel.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=49233

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

This patch looks very fishy as-is, simply because I don't like adding random calls to random functions at pretty much random places without any hint why it works.

It's difficult to know exactly why it works. We can only guess, but nothing more than that, just like between husband and wife :)

Yeah I know, but I've been burnt a bit recently with too much cargo-culting. Hence why I'm so dense about this stuff ;-)

Quoted 6 lines
Quoted 5 lines

Just a few weeks ago we've had tons of fun with writes
that don't stick when the ring isn't yet set up. Can you please check with a bunch of printks what exactly happens to the value of BLC_PWM_CPU_CTL? I.e. whether the write doesn't stick or whether it gets reset somewhere between the register restore functions and the new place.

Quoted 5 lines
Quoted 2 lines

The register value is already set, and not touched after that. I've already checked.

So immediately before the backlight_update_status call we have the _correct_ value in BLC_PWM_CPU_CTL? That would be indeed puzzling ...

Yes.

Quoted 6 lines
Quoted 6 lines
Quoted 3 lines

I suspect the opregion_init call could allow the bios to frob with these registers. If that's the case, the patch is missing a comment that to that effect.

Right. That's also my suspect.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Essentially I want to know why this place here works and why not move the call a few lines up or down.

A few lines down works definitely. Even just writing to sysfs works. A few lines up, well, I need to check where is the border line.

Thanks, that would be interesting. Especially if it's opregion that we need to have set up before restoring the backlight properly.

This is getting really puzzling. It's not about the opregion. Just some order of the register restoration.

Essentially the patch below fixes the problem. But I have absolutely no idea why.

Quoted 8 lines
Quoted 3 lines

Daniel, do you have any clue about this?

Is it a strict order of these registers?

Could very well be that the write does not get forwarded internally when the BLC stuff isn't enabled yet. Or the other way round. E.g. the new pipe select stuff for the blc clock I've implemented just recently only works while the backlight is off.

Sounds possible.

Quoted 4 lines

I'm much happier with this diff than the older one to just hit the regs harder, so can you please bake this into a proper patch? And please add a 1-line comment to the code that for some unknown reason we need to write the _CTL reg after the _CTL2.

Sure, will be sent soon later.

Quoted 5 lines

Also, I loathe this save/restore register dance. It duplicates code we already have, but with the twist that we might accidentally get the ordering (or some timing constraint) slightly wrong. Resulting in broken s/r, at least on some machines. Imo we should try to wean off these in the long term.

OTOH, these allow shorter initialization than the normal full init. But I agree with the point of code maintenance mess by such.

thanks,

Takashi

Quoted 32 lines

Thanks, Daniel

Quoted 23 lines

Takashi

Quoted 18 lines

Takashi

---

drivers/gpu/drm/i915/i915_suspend.c +1 -1
index 0ede02a..0f89dd0 100644
@@ -740,8 +740,8 @@ static void i915_restore_display(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		I915_WRITE(BLC_PWM_PCH_CTL1, dev_priv->saveBLC_PWM_CTL);
 		I915_WRITE(BLC_PWM_PCH_CTL2, dev_priv->saveBLC_PWM_CTL2);
-		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(BLC_PWM_CPU_CTL2, dev_priv->saveBLC_CPU_PWM_CTL2);
+		I915_WRITE(BLC_PWM_CPU_CTL, dev_priv->saveBLC_CPU_PWM_CTL);
 		I915_WRITE(PCH_PP_ON_DELAYS, dev_priv->savePP_ON_DELAYS);
 		I915_WRITE(PCH_PP_OFF_DELAYS, dev_priv->savePP_OFF_DELAYS);
 		I915_WRITE(PCH_PP_DIVISOR, dev_priv->savePP_DIVISOR);
Signature · 3 lines
Daniel Vetter
Mail: [email protected]
Mobile: +41 (0)79 365 57 48