[3/3] i915: Don't provide ACPI backlight interface if firmware expects Windows 8
diff mbox

Message ID 1370818899-8595-4-git-send-email-matthew.garrett@nebula.com
State Superseded, archived
Headers show

Commit Message

Matthew Garrett June 9, 2013, 11:01 p.m. UTC
Windows 8 leaves backlight control up to individual graphics drivers rather
than making ACPI calls itself. There's plenty of evidence to suggest that
the Intel driver for Windows doesn't use the ACPI interface, including the
fact that it's broken on a bunch of machines when the OS claims to support
Windows 8. The simplest thing to do appears to be to disable the ACPI
backlight interface on these systems.

Signed-off-by: Matthew Garrett <matthew.garrett@nebula.com>
---
 drivers/gpu/drm/i915/i915_dma.c | 3 +++
 1 file changed, 3 insertions(+)

Comments

Daniel Vetter June 10, 2013, 7:40 a.m. UTC | #1
On Sun, Jun 09, 2013 at 07:01:39PM -0400, Matthew Garrett wrote:
> Windows 8 leaves backlight control up to individual graphics drivers rather
> than making ACPI calls itself. There's plenty of evidence to suggest that
> the Intel driver for Windows doesn't use the ACPI interface, including the
> fact that it's broken on a bunch of machines when the OS claims to support
> Windows 8. The simplest thing to do appears to be to disable the ACPI
> backlight interface on these systems.
> 
> Signed-off-by: Matthew Garrett <matthew.garrett@nebula.com>

Make sense and I guess it's easier to merge this all through the acpi
tree. So

Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>

> ---
>  drivers/gpu/drm/i915/i915_dma.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
> index 3b315ba..23b6292 100644
> --- a/drivers/gpu/drm/i915/i915_dma.c
> +++ b/drivers/gpu/drm/i915/i915_dma.c
> @@ -1661,6 +1661,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
>  		/* Must be done after probing outputs */
>  		intel_opregion_init(dev);
>  		acpi_video_register();
> +		/* Don't use ACPI backlight functions on Windows 8 platforms */
> +		if (acpi_osi_version() >= ACPI_OSI_WIN_8)
> +			acpi_video_backlight_unregister();
>  	}
>  
>  	if (IS_GEN5(dev))
> -- 
> 1.8.2.1
>
joeyli June 10, 2013, 9:22 a.m. UTC | #2
? ??2013-06-09 ? 19:01 -0400?Matthew Garrett ???
> Windows 8 leaves backlight control up to individual graphics drivers rather
> than making ACPI calls itself. There's plenty of evidence to suggest that
> the Intel driver for Windows doesn't use the ACPI interface, including the
> fact that it's broken on a bunch of machines when the OS claims to support
> Windows 8. The simplest thing to do appears to be to disable the ACPI
> backlight interface on these systems.
> 
> Signed-off-by: Matthew Garrett <matthew.garrett@nebula.com>
> ---
>  drivers/gpu/drm/i915/i915_dma.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
> index 3b315ba..23b6292 100644
> --- a/drivers/gpu/drm/i915/i915_dma.c
> +++ b/drivers/gpu/drm/i915/i915_dma.c
> @@ -1661,6 +1661,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
>  		/* Must be done after probing outputs */
>  		intel_opregion_init(dev);
>  		acpi_video_register();
> +		/* Don't use ACPI backlight functions on Windows 8 platforms */
> +		if (acpi_osi_version() >= ACPI_OSI_WIN_8)
> +			acpi_video_backlight_unregister();
>  	}
>  
>  	if (IS_GEN5(dev))

This patch set works to me on Acer Aspire V3 notebook for unregister the
backlight interface of acpi video driver when i915 loaded. Acer Aspire
V3 has the Windows8 support in DSDT.

Tested-by: Lee, Chun-Yi <jlee@suse.com>


Thanks a lot!
Joey Lee

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Aaron Lu June 14, 2013, 6:47 a.m. UTC | #3
On 06/10/2013 07:01 AM, Matthew Garrett wrote:
> Windows 8 leaves backlight control up to individual graphics drivers rather
> than making ACPI calls itself. There's plenty of evidence to suggest that
> the Intel driver for Windows doesn't use the ACPI interface, including the
> fact that it's broken on a bunch of machines when the OS claims to support
> Windows 8. The simplest thing to do appears to be to disable the ACPI
> backlight interface on these systems.
> 
> Signed-off-by: Matthew Garrett <matthew.garrett@nebula.com>
> ---
>  drivers/gpu/drm/i915/i915_dma.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
> index 3b315ba..23b6292 100644
> --- a/drivers/gpu/drm/i915/i915_dma.c
> +++ b/drivers/gpu/drm/i915/i915_dma.c
> @@ -1661,6 +1661,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
>  		/* Must be done after probing outputs */
>  		intel_opregion_init(dev);
>  		acpi_video_register();
> +		/* Don't use ACPI backlight functions on Windows 8 platforms */
> +		if (acpi_osi_version() >= ACPI_OSI_WIN_8)
> +			acpi_video_backlight_unregister();

What about the platform driver created backlight interface? Oh right,
thinkpad_acpi will not create backlight interface for them since they
are not in DMI table, so acpi_video_backlight_support() will return
true and thinkspad_acpi thinks ACPI video driver is controlling
backlight so it will not create backlight interface for them.

Then we will need to remember not to add any of those systems into
the DMI table of video_detect.c, or thinkpad_acpi module will create
backlight interface for them and according to reporter, that interface
does not work either.

What about a priority based solution? We can introduce a new field named
priority to backlight_device and instead of calling another module's
function like the unregister one here(which cause unnecessary module
dependency), we only need to boost priority for its own interface. This
field will be exported to sysfs, so user can change it during runtime
too. And we can also introduce a new kernel command line as
backlight.force_interface=raw/firmware/platform, to overcome the limited
functionality provided by acpi_backlight=video/vendor, which does not
involve GPU's interface.

And we can place the quirk code in backlight layer instead of individual
backlight functionality provider module. Suppose we have a backlight
manager there, for all win8 systems, we can boost the raw type's
priority on its registration, so no need to add code in
intel/amd/etc./'s GPU driver code.

With priority based solution, all backlight control interfaces stay,
the priority field is an indication given by kernel to user space.

For a complete description on backlight control background and the
proposed solution, please take a look at:
https://gist.github.com/aaronlu/5779828

It would be good to know your opinions, thanks.

-Aaron
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matthew Garrett June 14, 2013, 5:29 p.m. UTC | #4
On Fri, 2013-06-14 at 14:47 +0800, Aaron Lu wrote:

> What about a priority based solution? We can introduce a new field named

> priority to backlight_device and instead of calling another module's

> function like the unregister one here(which cause unnecessary module

> dependency), we only need to boost priority for its own interface. This

> field will be exported to sysfs, so user can change it during runtime

> too. And we can also introduce a new kernel command line as

> backlight.force_interface=raw/firmware/platform, to overcome the limited

> functionality provided by acpi_backlight=video/vendor, which does not

> involve GPU's interface.


How would that work with existing userspace?

> And we can place the quirk code in backlight layer instead of individual

> backlight functionality provider module. Suppose we have a backlight

> manager there, for all win8 systems, we can boost the raw type's

> priority on its registration, so no need to add code in

> intel/amd/etc./'s GPU driver code.


But we'd need to add code to every piece of userspace that currently
uses the backlight, right?

> With priority based solution, all backlight control interfaces stay,

> the priority field is an indication given by kernel to user space.


We shouldn't export interfaces if we don't expect them to work.

-- 
Matthew Garrett | mjg59@srcf.ucam.org
Aaron Lu June 15, 2013, 1:26 a.m. UTC | #5
On 06/15/2013 01:29 AM, Matthew Garrett wrote:
> On Fri, 2013-06-14 at 14:47 +0800, Aaron Lu wrote:
> 
>> What about a priority based solution? We can introduce a new field named
>> priority to backlight_device and instead of calling another module's
>> function like the unregister one here(which cause unnecessary module
>> dependency), we only need to boost priority for its own interface. This
>> field will be exported to sysfs, so user can change it during runtime
>> too. And we can also introduce a new kernel command line as
>> backlight.force_interface=raw/firmware/platform, to overcome the limited
>> functionality provided by acpi_backlight=video/vendor, which does not
>> involve GPU's interface.
> 
> How would that work with existing userspace?

User space tool will need to be updated to use this as stated in the
gist page, I've patches for gsd-backlight-helper and xorg-x11-drv-intel,
for others we can add I think if the priority based solution is deemed
useful.

> 
>> And we can place the quirk code in backlight layer instead of individual
>> backlight functionality provider module. Suppose we have a backlight
>> manager there, for all win8 systems, we can boost the raw type's
>> priority on its registration, so no need to add code in
>> intel/amd/etc./'s GPU driver code.
> 
> But we'd need to add code to every piece of userspace that currently
> uses the backlight, right?

Yes that's the case.

> 
>> With priority based solution, all backlight control interfaces stay,
>> the priority field is an indication given by kernel to user space.
> 
> We shouldn't export interfaces if we don't expect them to work.

It's not easy to decide if they work or not sometimes, e.g. I came
across a system that claims win8 in ACPI table and has an Intel GPU,
while its ACPI video interface also works. With this patch, the working
ACPI video interface is removed, while with the priority based solution,
the GPU's interface priority gets higher, but the ACPI video interface
still stays.

Thanks,
Aaron
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matthew Garrett June 15, 2013, 1:38 a.m. UTC | #6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--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Aaron Lu June 15, 2013, 4:14 a.m. UTC | #7
On 06/15/2013 09:38 AM, Matthew Garrett wrote:
> On Sat, 2013-06-15 at 09:26 +0800, Aaron Lu wrote:
>> It's not easy to decide if they work or not sometimes, e.g. I came
>> across a system that claims win8 in ACPI table and has an Intel GPU,
>> while its ACPI video interface also works. With this patch, the working
>> ACPI video interface is removed, while with the priority based solution,
>> the GPU's interface priority gets higher, but the ACPI video interface
>> still stays.
> 
> Well, Windows 8 will only use the ACPI backlight interface if the GPU
> driver decides to, right? So the logic for deciding whether to remove
> the ACPI backlight control or not should be left up to the GPU. There's

I don't know this. From the document I googled, Microsoft suggests under
win8, backlight should be controlled by the graphics driver for smooth
brightness level change, instead of ACPI or other methods. So it is
possible that OEM will not test the ACPI interface well and thus the
interface is likely broken. I don't see why GPU driver has any better
knowledge on which systems the firmware interface is broken or not.

> no harm in refusing to expose a working method if there's another
> working method, but there is harm in exposing a broken one and expecting
> userspace to know the difference.

BTW, the proposed solution is not meant to solve win8 problems alone, it
should make solving other problems easy and make individual backlight
control interface provider module independent with each other such as
the platform drivers will not need to check if ACPI video driver will
control backlight or not and can always create backlight interface(its
default priority is lower that ACPI video driver's so will not be taken
by user space by default, showing the same behavior of the current code).

The current acpi_backlight=video/vendor kernel command line is pretty
misleading, for laptops that do not have vendor backlight interface,
adding acpi_backlight=vendor actually makes the system using the GPU's
interface. Some laptops are using this switch to work around problems in
ACPI video driver and users think they are using vendor interface.
That's why I think we need a new command line as the
backlight.force_interface=raw/firmware/platform.

Instead of letting individual driver to make decisions on which
backlight interface this system should use(either in platform driver as
we currently did, see acer-wmi and asus-wmi, or GPU driver as this case),
I think we need a better and clear way to handle such things. For
example, suppose an acer laptop: vendor does not support backlight, ACPI
video's backlight interface is broken and GPU's works, to make it work,
user will need to select acer-wmi module while this module does not have
anything to do with the functionality, but simply because it serves as
the backlight manager for acer laptops.

The above information and idea is formed while solving bugs reported
in kernel bugzilla video component, the proposed solutin may not be good
enough, but I hope we can find a better way to handle backlight problems.

Thanks,
Aaron
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matthew Garrett June 15, 2013, 4:19 a.m. UTC | #8
On Sat, Jun 15, 2013 at 12:14:42PM +0800, Aaron Lu wrote:
> On 06/15/2013 09:38 AM, Matthew Garrett wrote:
> > Well, Windows 8 will only use the ACPI backlight interface if the GPU
> > driver decides to, right? So the logic for deciding whether to remove
> > the ACPI backlight control or not should be left up to the GPU. There's
> 
> I don't know this. From the document I googled, Microsoft suggests under
> win8, backlight should be controlled by the graphics driver for smooth
> brightness level change, instead of ACPI or other methods. So it is
> possible that OEM will not test the ACPI interface well and thus the
> interface is likely broken. I don't see why GPU driver has any better
> knowledge on which systems the firmware interface is broken or not.

The vendor will presumably have tested that backlight control works - if 
the GPU driver uses the ACPI interface and backlight control is broken, 
then the vendor would fix it.

> > no harm in refusing to expose a working method if there's another
> > working method, but there is harm in exposing a broken one and expecting
> > userspace to know the difference.
> 
> BTW, the proposed solution is not meant to solve win8 problems alone, it
> should make solving other problems easy and make individual backlight
> control interface provider module independent with each other such as
> the platform drivers will not need to check if ACPI video driver will
> control backlight or not and can always create backlight interface(its
> default priority is lower that ACPI video driver's so will not be taken
> by user space by default, showing the same behavior of the current code).

Sure, but it still requires the replacement of existing userspace. We 
need to fix the problem with existing userspace, too.

> The current acpi_backlight=video/vendor kernel command line is pretty
> misleading, for laptops that do not have vendor backlight interface,
> adding acpi_backlight=vendor actually makes the system using the GPU's
> interface. Some laptops are using this switch to work around problems in
> ACPI video driver and users think they are using vendor interface.
> That's why I think we need a new command line as the
> backlight.force_interface=raw/firmware/platform.

No, I think we need to fix the bugs that currently require users to pass 
options.

> Instead of letting individual driver to make decisions on which
> backlight interface this system should use(either in platform driver as
> we currently did, see acer-wmi and asus-wmi, or GPU driver as this case),
> I think we need a better and clear way to handle such things. For
> example, suppose an acer laptop: vendor does not support backlight, ACPI
> video's backlight interface is broken and GPU's works, to make it work,
> user will need to select acer-wmi module while this module does not have
> anything to do with the functionality, but simply because it serves as
> the backlight manager for acer laptops.

How do these machines work under Windows? Until we know the answer to 
that, we don't know what the correct way to handle the problem is under 
Linux.
Aaron Lu June 15, 2013, 12:29 p.m. UTC | #9
On 06/15/2013 12:19 PM, Matthew Garrett wrote:
> On Sat, Jun 15, 2013 at 12:14:42PM +0800, Aaron Lu wrote:
>> On 06/15/2013 09:38 AM, Matthew Garrett wrote:
>>> Well, Windows 8 will only use the ACPI backlight interface if the GPU
>>> driver decides to, right? So the logic for deciding whether to remove
>>> the ACPI backlight control or not should be left up to the GPU. There's
>>
>> I don't know this. From the document I googled, Microsoft suggests under
>> win8, backlight should be controlled by the graphics driver for smooth
>> brightness level change, instead of ACPI or other methods. So it is
>> possible that OEM will not test the ACPI interface well and thus the
>> interface is likely broken. I don't see why GPU driver has any better
>> knowledge on which systems the firmware interface is broken or not.
> 
> The vendor will presumably have tested that backlight control works - if 
> the GPU driver uses the ACPI interface and backlight control is broken, 
> then the vendor would fix it.

I don't think GPU driver uses ACPI interface, that's why for some
systems, ACPI interface is broken while the GPU's is not.

> 
>>> no harm in refusing to expose a working method if there's another
>>> working method, but there is harm in exposing a broken one and expecting
>>> userspace to know the difference.
>>
>> BTW, the proposed solution is not meant to solve win8 problems alone, it
>> should make solving other problems easy and make individual backlight
>> control interface provider module independent with each other such as
>> the platform drivers will not need to check if ACPI video driver will
>> control backlight or not and can always create backlight interface(its
>> default priority is lower that ACPI video driver's so will not be taken
>> by user space by default, showing the same behavior of the current code).
> 
> Sure, but it still requires the replacement of existing userspace. We 
> need to fix the problem with existing userspace, too.

Yes. The problem is two way. First, some of the backlight interface
privoder module provides a broken interface; Two, the userspace picked a
broken interface while there are usable ones. For example, in the win8
thinkpad case, the ACPI video driver provides broken interface and the
GPU driver provides usable interface, but userspace choose ACPI video's
interface. To make things complicated, if we make the broken interface
disappear in ACPI video module, then the platform driver thinkpad_acpi
will start to provide another broken interface. I have to say, solve it
in the GPU code is a clever way, it's just a little weird to me for a
broken interface to be blacklisted, we have to do it in another module,
not the broken module itself.

> 
>> The current acpi_backlight=video/vendor kernel command line is pretty
>> misleading, for laptops that do not have vendor backlight interface,
>> adding acpi_backlight=vendor actually makes the system using the GPU's
>> interface. Some laptops are using this switch to work around problems in
>> ACPI video driver and users think they are using vendor interface.
>> That's why I think we need a new command line as the
>> backlight.force_interface=raw/firmware/platform.
> 
> No, I think we need to fix the bugs that currently require users to pass 
> options.

For ACPI video driver, since it relies on firmware to do the right thing,
if the functionality is broken, then it is, there is hardly anything we
can do...(not always, we can quirk in some cases, but there are cases we
can do nothing). In this case, we can:
1 Ask user to add acpi_backlight=vendor, so that ACPI video driver will
  not create backlight interface and userspace will not pick it;
2 Add that model in DMI table in video_detect.c, eliminating the need for
  that command line;
3 Add that model in DMI table in another module, let that module
  unregister backlight interface provided by ACPI video as is done in
  acer-wmi, asus-wmi, and i915 in this case;
3 Use the backlight section in xorg.conf to make X uses another
  backlight interface. This may not work for distros that use
  gsd-backlight-helper, which always prefers firmware.

Which one do you prefer?

> 
>> Instead of letting individual driver to make decisions on which
>> backlight interface this system should use(either in platform driver as
>> we currently did, see acer-wmi and asus-wmi, or GPU driver as this case),
>> I think we need a better and clear way to handle such things. For
>> example, suppose an acer laptop: vendor does not support backlight, ACPI
>> video's backlight interface is broken and GPU's works, to make it work,
>> user will need to select acer-wmi module while this module does not have
>> anything to do with the functionality, but simply because it serves as
>> the backlight manager for acer laptops.
> 
> How do these machines work under Windows? Until we know the answer to 
> that, we don't know what the correct way to handle the problem is under 
> Linux.

Do you mean we need to understand Windows behavior so that we can
decide where to place the code to do the backlight management thing?
So for win8 case, we know MS will use GPU driver to do backlight
control, all this means to me is, ACPI video's and platform driver's
interface are more likely broken on those systems, but it doesn't
qualify why Linux' GPU driver should do the decision, it's not that the
GPU driver can poke some GPU register and then decide oh, this model
does not have working ACPI backlight interface. As this patch shows, we
make the decision by OSI, which is not specific to GPU driver.

BTW, I don't think any module knows something better than another
module, all quirk logic and DMI entry we currently have in Linux are
got by user's feedback(bug reports), it doesn't seem to me some module
is natrually the one that should make this decision. Or do I miss
something?

Thanks,
Aaron
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matthew Garrett June 15, 2013, 3:16 p.m. UTC | #10
On Sat, Jun 15, 2013 at 08:29:15PM +0800, Aaron Lu wrote:
> On 06/15/2013 12:19 PM, Matthew Garrett wrote:
> > The vendor will presumably have tested that backlight control works - if 
> > the GPU driver uses the ACPI interface and backlight control is broken, 
> > then the vendor would fix it.
> 
> I don't think GPU driver uses ACPI interface, that's why for some
> systems, ACPI interface is broken while the GPU's is not.

On systems where the ACPI interface is broken, the GPU driver clearly 
doesn't use the ACPI interface. As yet, we don't know if that's always 
true, though.

> > Sure, but it still requires the replacement of existing userspace. We 
> > need to fix the problem with existing userspace, too.
> 
> Yes. The problem is two way. First, some of the backlight interface
> privoder module provides a broken interface; Two, the userspace picked a
> broken interface while there are usable ones. For example, in the win8
> thinkpad case, the ACPI video driver provides broken interface and the
> GPU driver provides usable interface, but userspace choose ACPI video's
> interface. To make things complicated, if we make the broken interface
> disappear in ACPI video module, then the platform driver thinkpad_acpi
> will start to provide another broken interface. I have to say, solve it
> in the GPU code is a clever way, it's just a little weird to me for a
> broken interface to be blacklisted, we have to do it in another module,
> not the broken module itself.

The ACPI driver has no way of making this kind of policy decision. The 
GPU driver does.

> > No, I think we need to fix the bugs that currently require users to pass 
> > options.
> 
> For ACPI video driver, since it relies on firmware to do the right thing,
> if the functionality is broken, then it is, there is hardly anything we
> can do...(not always, we can quirk in some cases, but there are cases we
> can do nothing). In this case, we can:
> 1 Ask user to add acpi_backlight=vendor, so that ACPI video driver will
>   not create backlight interface and userspace will not pick it;
> 2 Add that model in DMI table in video_detect.c, eliminating the need for
>   that command line;
> 3 Add that model in DMI table in another module, let that module
>   unregister backlight interface provided by ACPI video as is done in
>   acer-wmi, asus-wmi, and i915 in this case;
> 3 Use the backlight section in xorg.conf to make X uses another
>   backlight interface. This may not work for distros that use
>   gsd-backlight-helper, which always prefers firmware.
> 
> Which one do you prefer?

I'd prefer to figure out how it works in Windows and implement it the 
same way.

> > How do these machines work under Windows? Until we know the answer to 
> > that, we don't know what the correct way to handle the problem is under 
> > Linux.
> 
> Do you mean we need to understand Windows behavior so that we can
> decide where to place the code to do the backlight management thing?
> So for win8 case, we know MS will use GPU driver to do backlight
> control, all this means to me is, ACPI video's and platform driver's
> interface are more likely broken on those systems, but it doesn't
> qualify why Linux' GPU driver should do the decision, it's not that the
> GPU driver can poke some GPU register and then decide oh, this model
> does not have working ACPI backlight interface. As this patch shows, we
> make the decision by OSI, which is not specific to GPU driver.

Sure it is - for all we know there's a value in the opregion space that 
tells the Intel GPU driver which interface we should be using. We don't 
know because Intel haven't released a version of the opregion driver 
newer than 2007. It may be that all Windows 8 GPU drivers use the GPU 
backlight control registers directly and never use any firmware 
interface, but right now we simply don't know that. All we can do is 
look at the existing cases we have and say that it appears that 
Intel graphics systems don't use the ACPI interface. Anything more than 
that requires more evidence.

> BTW, I don't think any module knows something better than another
> module, all quirk logic and DMI entry we currently have in Linux are
> got by user's feedback(bug reports), it doesn't seem to me some module
> is natrually the one that should make this decision. Or do I miss
> something?

The GPU driver makes the policy decision under Windows 8, so it's the 
natural place to make this decision on Windows 8 systems.
Daniel Vetter June 15, 2013, 6:29 p.m. UTC | #11
Hi all,

So to me it looks like the discussion is going in circles a bit, hence let
me drop my maintainer-opinion here:

1. Matthew's patch series here looks reasonable, and if it fixes a bunch
of systems (which it seems to) it has my Ack and imo should go in. If acpi
maintainers can smash their Ack onto the acpi parts I'd very much like to
merge this into drm-intel-next, that should give us the most coverage for
intel systems.

Len, Rafael, are you ok with the acpi part of this and merging it through
drm-intel-next?

2. Imo the current amount of quirking we expose to users (we have kernel
options to disable the acpi interface, blacklist platform modules, all
backlights can be tested through sysfs and on top of that xf86-video-intel
has an xorg.conf to select the backlight used by the driver). I haven't
spotted a compelling reason in this thread to add another one, what we
have seems to be good enough to debug backligh issues.

3. Also, adding yet another backlight quirk mechanism isn't gonna
magically fix broken systems.

We _really_ should strive to make this just work and not offer the angry
user another roll of duct-tape for free.

4. The currently established priority selection for backlights of platform
> firmware > raw seems to be good enough. Note that the explicit list in
xf86-vidoe-intel is only used as a fallback for really old kernels which
do not expose this information. We could probably rip it out.

5. We've recently looked at opregion again and couldn't spot a hint.
Unfortnately it looks like both noodling better information out of Intel
and trying to publish an updated opregion spec seem to be losing games :(
We'll keep on trying though.

Aside at the end: If the gnome tool indeed has its own backlight code and
doesn't just use that as a fallback if the xrandr backligh property isn't
available, then that's just a serious bug in gnome and should be fixed
asap. But imo that's not something we should try to (nor do I see any way
how to) work around in the kernel.

Cheers, Daniel
Matthew Garrett June 15, 2013, 6:44 p.m. UTC | #12
On Sat, Jun 15, 2013 at 08:29:42PM +0200, Daniel Vetter wrote:

> Aside at the end: If the gnome tool indeed has its own backlight code and
> doesn't just use that as a fallback if the xrandr backligh property isn't
> available, then that's just a serious bug in gnome and should be fixed
> asap. But imo that's not something we should try to (nor do I see any way
> how to) work around in the kernel.

It's only used if there's no backlight property on the display.
Rafael Wysocki June 15, 2013, 8:27 p.m. UTC | #13
On Saturday, June 15, 2013 08:29:42 PM Daniel Vetter wrote:
> Hi all,
> 
> So to me it looks like the discussion is going in circles a bit, hence let
> me drop my maintainer-opinion here:
> 
> 1. Matthew's patch series here looks reasonable, and if it fixes a bunch
> of systems (which it seems to) it has my Ack and imo should go in. If acpi
> maintainers can smash their Ack onto the acpi parts I'd very much like to
> merge this into drm-intel-next, that should give us the most coverage for
> intel systems.
> 
> Len, Rafael, are you ok with the acpi part of this and merging it through
> drm-intel-next?

It has to go through the ACPI tree because of the ACPICA patch that needs to
be synchronized with the ACPICA upstream.  Sorry.

That said, I'm going to take this patchset.

> 2. Imo the current amount of quirking we expose to users (we have kernel
> options to disable the acpi interface, blacklist platform modules, all
> backlights can be tested through sysfs and on top of that xf86-video-intel
> has an xorg.conf to select the backlight used by the driver). I haven't
> spotted a compelling reason in this thread to add another one, what we
> have seems to be good enough to debug backligh issues.
> 
> 3. Also, adding yet another backlight quirk mechanism isn't gonna
> magically fix broken systems.
> 
> We _really_ should strive to make this just work and not offer the angry
> user another roll of duct-tape for free.
> 
> 4. The currently established priority selection for backlights of platform
> > firmware > raw seems to be good enough. Note that the explicit list in
> xf86-vidoe-intel is only used as a fallback for really old kernels which
> do not expose this information. We could probably rip it out.
> 
> 5. We've recently looked at opregion again and couldn't spot a hint.
> Unfortnately it looks like both noodling better information out of Intel
> and trying to publish an updated opregion spec seem to be losing games :(
> We'll keep on trying though.
> 
> Aside at the end: If the gnome tool indeed has its own backlight code and
> doesn't just use that as a fallback if the xrandr backligh property isn't
> available, then that's just a serious bug in gnome and should be fixed
> asap. But imo that's not something we should try to (nor do I see any way
> how to) work around in the kernel.

Agreed.

Thanks,
Rafael
Daniel Vetter June 15, 2013, 8:35 p.m. UTC | #14
On Sat, Jun 15, 2013 at 10:27:06PM +0200, Rafael J. Wysocki wrote:
> On Saturday, June 15, 2013 08:29:42 PM Daniel Vetter wrote:
> > Hi all,
> > 
> > So to me it looks like the discussion is going in circles a bit, hence let
> > me drop my maintainer-opinion here:
> > 
> > 1. Matthew's patch series here looks reasonable, and if it fixes a bunch
> > of systems (which it seems to) it has my Ack and imo should go in. If acpi
> > maintainers can smash their Ack onto the acpi parts I'd very much like to
> > merge this into drm-intel-next, that should give us the most coverage for
> > intel systems.
> > 
> > Len, Rafael, are you ok with the acpi part of this and merging it through
> > drm-intel-next?
> 
> It has to go through the ACPI tree because of the ACPICA patch that needs to
> be synchronized with the ACPICA upstream.  Sorry.

No problem, since we're pretty close to the merge window that would have
at most resulted in a few more weeks of testing in i915 trees anyway. -rc
kernels should still give us plenty of time to catch fallout, if there is
any.

> That said, I'm going to take this patchset.

Thanks, Daniel
Rafael Wysocki July 5, 2013, 12:20 p.m. UTC | #15
On Sunday, June 09, 2013 07:01:39 PM Matthew Garrett wrote:
> Windows 8 leaves backlight control up to individual graphics drivers rather
> than making ACPI calls itself. There's plenty of evidence to suggest that
> the Intel driver for Windows doesn't use the ACPI interface, including the
> fact that it's broken on a bunch of machines when the OS claims to support
> Windows 8. The simplest thing to do appears to be to disable the ACPI
> backlight interface on these systems.
> 
> Signed-off-by: Matthew Garrett <matthew.garrett@nebula.com>
> ---
>  drivers/gpu/drm/i915/i915_dma.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
> index 3b315ba..23b6292 100644
> --- a/drivers/gpu/drm/i915/i915_dma.c
> +++ b/drivers/gpu/drm/i915/i915_dma.c
> @@ -1661,6 +1661,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
>  		/* Must be done after probing outputs */
>  		intel_opregion_init(dev);
>  		acpi_video_register();
> +		/* Don't use ACPI backlight functions on Windows 8 platforms */
> +		if (acpi_osi_version() >= ACPI_OSI_WIN_8)
> +			acpi_video_backlight_unregister();
>  	}
>  
>  	if (IS_GEN5(dev))
> 

Well, this causes build failures to happen when the ACPI video driver is
modular and the graphics driver is not.

I'm not sure how to resolve that, so suggestions are welcome.

Thanks,
Rafael
Rafael Wysocki July 5, 2013, 8 p.m. UTC | #16
On Friday, July 05, 2013 02:20:14 PM Rafael J. Wysocki wrote:
> On Sunday, June 09, 2013 07:01:39 PM Matthew Garrett wrote:
> > Windows 8 leaves backlight control up to individual graphics drivers rather
> > than making ACPI calls itself. There's plenty of evidence to suggest that
> > the Intel driver for Windows doesn't use the ACPI interface, including the
> > fact that it's broken on a bunch of machines when the OS claims to support
> > Windows 8. The simplest thing to do appears to be to disable the ACPI
> > backlight interface on these systems.
> > 
> > Signed-off-by: Matthew Garrett <matthew.garrett@nebula.com>
> > ---
> >  drivers/gpu/drm/i915/i915_dma.c | 3 +++
> >  1 file changed, 3 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
> > index 3b315ba..23b6292 100644
> > --- a/drivers/gpu/drm/i915/i915_dma.c
> > +++ b/drivers/gpu/drm/i915/i915_dma.c
> > @@ -1661,6 +1661,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
> >  		/* Must be done after probing outputs */
> >  		intel_opregion_init(dev);
> >  		acpi_video_register();
> > +		/* Don't use ACPI backlight functions on Windows 8 platforms */
> > +		if (acpi_osi_version() >= ACPI_OSI_WIN_8)
> > +			acpi_video_backlight_unregister();
> >  	}
> >  
> >  	if (IS_GEN5(dev))
> > 
> 
> Well, this causes build failures to happen when the ACPI video driver is
> modular and the graphics driver is not.
> 
> I'm not sure how to resolve that, so suggestions are welcome.

Actually, that happened with the radeon patch.

That said, ACPI_OSI_WIN_8 doesn't make much sense for !CONFIG_ACPI, for
example.

What about making acpi_video_register() do the quirk instead?  We could add an
argument to it indicating whether or not quirks should be applied.

Thanks,
Rafael

Patch
diff mbox

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 3b315ba..23b6292 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1661,6 +1661,9 @@  int i915_driver_load(struct drm_device *dev, unsigned long flags)
 		/* Must be done after probing outputs */
 		intel_opregion_init(dev);
 		acpi_video_register();
+		/* Don't use ACPI backlight functions on Windows 8 platforms */
+		if (acpi_osi_version() >= ACPI_OSI_WIN_8)
+			acpi_video_backlight_unregister();
 	}
 
 	if (IS_GEN5(dev))