mbox series

[v4,00/20] Add support for binding ACPI platform profile to multiple drivers

Message ID 20241105153316.378-1-mario.limonciello@amd.com (mailing list archive)
Headers show
Series Add support for binding ACPI platform profile to multiple drivers | expand

Message

Mario Limonciello Nov. 5, 2024, 3:32 p.m. UTC
Currently there are a number of ASUS products on the market that happen to
have ACPI objects for amd-pmf to bind to as well as an ACPI platform
profile provided by asus-wmi.

The ACPI platform profile support created by amd-pmf on these ASUS
products is "Function 9" which is specifically for "BIOS or EC
notification" of power slider position. This feature is actively used
by some designs such as Framework 13 and Framework 16.

On these ASUS designs we keep on quirking more and more of them to turn
off this notification so that asus-wmi can bind.

This however isn't how Windows works.  "Multiple" things are notified for
the power slider position. This series adjusts Linux to behave similarly.

Multiple drivers can now register an ACPI platform profile and will react
to set requests.

To avoid chaos, only positions that are common to both drivers are
accepted when the legacy /sys/firmware/acpi/platform_profile interface
is used.

This series also adds a new concept of a "custom" profile.  This allows
userspace to discover that there are multiple driver handlers that are
configured differently.

This series also allows dropping all of the PMF quirks from amd-pmf.

v4:
 * Drop the list; iterate classes
 * Drop patches that didn't make sense without list
 * Cover alienware-wmi as well (recently merged to platform-x86/for-next)
 * Drop requirement for balanced
 * Rename platform-profile class members to 'profile', 'options', 'name'
 * Drop the name in /sys/class/platform-profile and just use ida value.
   IE platform-profile-0

Mario Limonciello (20):
  ACPI: platform-profile: Add a name member to handlers
  platform/x86/dell: dell-pc: Create platform device
  ACPI: platform_profile: Add device pointer into platform profile
    handler
  ACPI: platform_profile: Add platform handler argument to
    platform_profile_remove()
  ACPI: platform_profile: Move sanity check out of the mutex
  ACPI: platform_profile: Move matching string for new profile out of
    mutex
  ACPI: platform_profile: Use guard(mutex) for register/unregister
  ACPI: platform_profile: Use `scoped_cond_guard`
  ACPI: platform_profile: Create class for ACPI platform profile
  ACPI: platform_profile: Add name attribute to class interface
  ACPI: platform_profile: Add choices attribute for class interface
  ACPI: platform_profile: Add profile attribute for class interface
  ACPI: platform_profile: Notify change events on register and
    unregister
  ACPI: platform_profile: Only show profiles common for all handlers
  ACPI: platform_profile: Add concept of a "custom" profile
  ACPI: platform_profile: Make sure all profile handlers agree on
    profile
  ACPI: platform_profile: Check all profile handler to calculate next
  ACPI: platform_profile: Allow multiple handlers
  platform/x86/amd: pmf: Drop all quirks
  Documentation: Add documentation about class interface for platform
    profiles

 .../userspace-api/sysfs-platform_profile.rst  |  28 ++
 drivers/acpi/platform_profile.c               | 446 ++++++++++++++----
 .../surface/surface_platform_profile.c        |   8 +-
 drivers/platform/x86/acer-wmi.c               |  10 +-
 drivers/platform/x86/amd/pmf/Makefile         |   2 +-
 drivers/platform/x86/amd/pmf/core.c           |   1 -
 drivers/platform/x86/amd/pmf/pmf-quirks.c     |  66 ---
 drivers/platform/x86/amd/pmf/pmf.h            |   3 -
 drivers/platform/x86/amd/pmf/sps.c            |   4 +-
 drivers/platform/x86/asus-wmi.c               |   6 +-
 drivers/platform/x86/dell/alienware-wmi.c     |   8 +-
 drivers/platform/x86/dell/dell-pc.c           |  39 +-
 drivers/platform/x86/hp/hp-wmi.c              |   8 +-
 drivers/platform/x86/ideapad-laptop.c         |   4 +-
 .../platform/x86/inspur_platform_profile.c    |   7 +-
 drivers/platform/x86/thinkpad_acpi.c          |   4 +-
 include/linux/platform_profile.h              |   7 +-
 17 files changed, 456 insertions(+), 195 deletions(-)
 delete mode 100644 drivers/platform/x86/amd/pmf/pmf-quirks.c


base-commit: d68cb6023356af3bd3193983ad4ec03954a0b3e2

Comments

Mark Pearson Nov. 5, 2024, 10:47 p.m. UTC | #1
Hi Mario,

On Tue, Nov 5, 2024, at 10:32 AM, Mario Limonciello wrote:
> Currently there are a number of ASUS products on the market that happen to
> have ACPI objects for amd-pmf to bind to as well as an ACPI platform
> profile provided by asus-wmi.
>
> The ACPI platform profile support created by amd-pmf on these ASUS
> products is "Function 9" which is specifically for "BIOS or EC
> notification" of power slider position. This feature is actively used
> by some designs such as Framework 13 and Framework 16.
>
> On these ASUS designs we keep on quirking more and more of them to turn
> off this notification so that asus-wmi can bind.
>
> This however isn't how Windows works.  "Multiple" things are notified for
> the power slider position. This series adjusts Linux to behave similarly.
>
> Multiple drivers can now register an ACPI platform profile and will react
> to set requests.
>
> To avoid chaos, only positions that are common to both drivers are
> accepted when the legacy /sys/firmware/acpi/platform_profile interface
> is used.
>
> This series also adds a new concept of a "custom" profile.  This allows
> userspace to discover that there are multiple driver handlers that are
> configured differently.
>
> This series also allows dropping all of the PMF quirks from amd-pmf.
>
> v4:
>  * Drop the list; iterate classes
>  * Drop patches that didn't make sense without list
>  * Cover alienware-wmi as well (recently merged to platform-x86/for-next)
>  * Drop requirement for balanced
>  * Rename platform-profile class members to 'profile', 'options', 'name'
>  * Drop the name in /sys/class/platform-profile and just use ida value.
>    IE platform-profile-0
>
> Mario Limonciello (20):
>   ACPI: platform-profile: Add a name member to handlers
>   platform/x86/dell: dell-pc: Create platform device
>   ACPI: platform_profile: Add device pointer into platform profile
>     handler
>   ACPI: platform_profile: Add platform handler argument to
>     platform_profile_remove()
>   ACPI: platform_profile: Move sanity check out of the mutex
>   ACPI: platform_profile: Move matching string for new profile out of
>     mutex
>   ACPI: platform_profile: Use guard(mutex) for register/unregister
>   ACPI: platform_profile: Use `scoped_cond_guard`
>   ACPI: platform_profile: Create class for ACPI platform profile
>   ACPI: platform_profile: Add name attribute to class interface
>   ACPI: platform_profile: Add choices attribute for class interface
>   ACPI: platform_profile: Add profile attribute for class interface
>   ACPI: platform_profile: Notify change events on register and
>     unregister
>   ACPI: platform_profile: Only show profiles common for all handlers
>   ACPI: platform_profile: Add concept of a "custom" profile
>   ACPI: platform_profile: Make sure all profile handlers agree on
>     profile
>   ACPI: platform_profile: Check all profile handler to calculate next
>   ACPI: platform_profile: Allow multiple handlers
>   platform/x86/amd: pmf: Drop all quirks
>   Documentation: Add documentation about class interface for platform
>     profiles
>
>  .../userspace-api/sysfs-platform_profile.rst  |  28 ++
>  drivers/acpi/platform_profile.c               | 446 ++++++++++++++----
>  .../surface/surface_platform_profile.c        |   8 +-
>  drivers/platform/x86/acer-wmi.c               |  10 +-
>  drivers/platform/x86/amd/pmf/Makefile         |   2 +-
>  drivers/platform/x86/amd/pmf/core.c           |   1 -
>  drivers/platform/x86/amd/pmf/pmf-quirks.c     |  66 ---
>  drivers/platform/x86/amd/pmf/pmf.h            |   3 -
>  drivers/platform/x86/amd/pmf/sps.c            |   4 +-
>  drivers/platform/x86/asus-wmi.c               |   6 +-
>  drivers/platform/x86/dell/alienware-wmi.c     |   8 +-
>  drivers/platform/x86/dell/dell-pc.c           |  39 +-
>  drivers/platform/x86/hp/hp-wmi.c              |   8 +-
>  drivers/platform/x86/ideapad-laptop.c         |   4 +-
>  .../platform/x86/inspur_platform_profile.c    |   7 +-
>  drivers/platform/x86/thinkpad_acpi.c          |   4 +-
>  include/linux/platform_profile.h              |   7 +-
>  17 files changed, 456 insertions(+), 195 deletions(-)
>  delete mode 100644 drivers/platform/x86/amd/pmf/pmf-quirks.c
>
>
> base-commit: d68cb6023356af3bd3193983ad4ec03954a0b3e2
> -- 
> 2.43.0

For the series - I tried it out on my T14s G6 AMD, and it all looks to be working nicely and as expected.
The thinkpad-acpi and amd-pmf classes show up. I can tweak them individually and 'custom' shows up under firmware/acpi/platform_profile.
I tried various combo's and didn't see any issues.
If you have any recommendations of things to try let me know.

Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>

Mark
Mario Limonciello Nov. 6, 2024, 1:11 a.m. UTC | #2
On 11/5/2024 16:47, Mark Pearson wrote:
> Hi Mario,
> 
> On Tue, Nov 5, 2024, at 10:32 AM, Mario Limonciello wrote:
>> Currently there are a number of ASUS products on the market that happen to
>> have ACPI objects for amd-pmf to bind to as well as an ACPI platform
>> profile provided by asus-wmi.
>>
>> The ACPI platform profile support created by amd-pmf on these ASUS
>> products is "Function 9" which is specifically for "BIOS or EC
>> notification" of power slider position. This feature is actively used
>> by some designs such as Framework 13 and Framework 16.
>>
>> On these ASUS designs we keep on quirking more and more of them to turn
>> off this notification so that asus-wmi can bind.
>>
>> This however isn't how Windows works.  "Multiple" things are notified for
>> the power slider position. This series adjusts Linux to behave similarly.
>>
>> Multiple drivers can now register an ACPI platform profile and will react
>> to set requests.
>>
>> To avoid chaos, only positions that are common to both drivers are
>> accepted when the legacy /sys/firmware/acpi/platform_profile interface
>> is used.
>>
>> This series also adds a new concept of a "custom" profile.  This allows
>> userspace to discover that there are multiple driver handlers that are
>> configured differently.
>>
>> This series also allows dropping all of the PMF quirks from amd-pmf.
>>
>> v4:
>>   * Drop the list; iterate classes
>>   * Drop patches that didn't make sense without list
>>   * Cover alienware-wmi as well (recently merged to platform-x86/for-next)
>>   * Drop requirement for balanced
>>   * Rename platform-profile class members to 'profile', 'options', 'name'
>>   * Drop the name in /sys/class/platform-profile and just use ida value.
>>     IE platform-profile-0
>>
>> Mario Limonciello (20):
>>    ACPI: platform-profile: Add a name member to handlers
>>    platform/x86/dell: dell-pc: Create platform device
>>    ACPI: platform_profile: Add device pointer into platform profile
>>      handler
>>    ACPI: platform_profile: Add platform handler argument to
>>      platform_profile_remove()
>>    ACPI: platform_profile: Move sanity check out of the mutex
>>    ACPI: platform_profile: Move matching string for new profile out of
>>      mutex
>>    ACPI: platform_profile: Use guard(mutex) for register/unregister
>>    ACPI: platform_profile: Use `scoped_cond_guard`
>>    ACPI: platform_profile: Create class for ACPI platform profile
>>    ACPI: platform_profile: Add name attribute to class interface
>>    ACPI: platform_profile: Add choices attribute for class interface
>>    ACPI: platform_profile: Add profile attribute for class interface
>>    ACPI: platform_profile: Notify change events on register and
>>      unregister
>>    ACPI: platform_profile: Only show profiles common for all handlers
>>    ACPI: platform_profile: Add concept of a "custom" profile
>>    ACPI: platform_profile: Make sure all profile handlers agree on
>>      profile
>>    ACPI: platform_profile: Check all profile handler to calculate next
>>    ACPI: platform_profile: Allow multiple handlers
>>    platform/x86/amd: pmf: Drop all quirks
>>    Documentation: Add documentation about class interface for platform
>>      profiles
>>
>>   .../userspace-api/sysfs-platform_profile.rst  |  28 ++
>>   drivers/acpi/platform_profile.c               | 446 ++++++++++++++----
>>   .../surface/surface_platform_profile.c        |   8 +-
>>   drivers/platform/x86/acer-wmi.c               |  10 +-
>>   drivers/platform/x86/amd/pmf/Makefile         |   2 +-
>>   drivers/platform/x86/amd/pmf/core.c           |   1 -
>>   drivers/platform/x86/amd/pmf/pmf-quirks.c     |  66 ---
>>   drivers/platform/x86/amd/pmf/pmf.h            |   3 -
>>   drivers/platform/x86/amd/pmf/sps.c            |   4 +-
>>   drivers/platform/x86/asus-wmi.c               |   6 +-
>>   drivers/platform/x86/dell/alienware-wmi.c     |   8 +-
>>   drivers/platform/x86/dell/dell-pc.c           |  39 +-
>>   drivers/platform/x86/hp/hp-wmi.c              |   8 +-
>>   drivers/platform/x86/ideapad-laptop.c         |   4 +-
>>   .../platform/x86/inspur_platform_profile.c    |   7 +-
>>   drivers/platform/x86/thinkpad_acpi.c          |   4 +-
>>   include/linux/platform_profile.h              |   7 +-
>>   17 files changed, 456 insertions(+), 195 deletions(-)
>>   delete mode 100644 drivers/platform/x86/amd/pmf/pmf-quirks.c
>>
>>
>> base-commit: d68cb6023356af3bd3193983ad4ec03954a0b3e2
>> -- 
>> 2.43.0
> 
> For the series - I tried it out on my T14s G6 AMD, and it all looks to be working nicely and as expected.
> The thinkpad-acpi and amd-pmf classes show up. I can tweak them individually and 'custom' shows up under firmware/acpi/platform_profile.
> I tried various combo's and didn't see any issues.
> If you have any recommendations of things to try let me know.
> 
> Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
> 
> Mark

Thanks for testing!  The fact that Lenovo and ASUS are both doing this 
with recent laptops makes it all the stronger of a case to do it rather 
than let them race for who gets the coveted platform profile slot.

If you've already tried mixing and matching combinations that's all I 
was going to suggest.

I'll wait for Armin's review and then I'll respin with the tags and 
small things you and Ilpo found.

BTW if I missed any important feedback of yours from v3 please let me 
know.  It was a lot of overhaul to switch to this way of doing things 
and a lot of it didn't make sense anymore.
Armin Wolf Nov. 6, 2024, 9:04 p.m. UTC | #3
Am 06.11.24 um 02:11 schrieb Mario Limonciello:

> On 11/5/2024 16:47, Mark Pearson wrote:
>> Hi Mario,
>>
>> On Tue, Nov 5, 2024, at 10:32 AM, Mario Limonciello wrote:
>>> Currently there are a number of ASUS products on the market that
>>> happen to
>>> have ACPI objects for amd-pmf to bind to as well as an ACPI platform
>>> profile provided by asus-wmi.
>>>
>>> The ACPI platform profile support created by amd-pmf on these ASUS
>>> products is "Function 9" which is specifically for "BIOS or EC
>>> notification" of power slider position. This feature is actively used
>>> by some designs such as Framework 13 and Framework 16.
>>>
>>> On these ASUS designs we keep on quirking more and more of them to turn
>>> off this notification so that asus-wmi can bind.
>>>
>>> This however isn't how Windows works.  "Multiple" things are
>>> notified for
>>> the power slider position. This series adjusts Linux to behave
>>> similarly.
>>>
>>> Multiple drivers can now register an ACPI platform profile and will
>>> react
>>> to set requests.
>>>
>>> To avoid chaos, only positions that are common to both drivers are
>>> accepted when the legacy /sys/firmware/acpi/platform_profile interface
>>> is used.
>>>
>>> This series also adds a new concept of a "custom" profile. This allows
>>> userspace to discover that there are multiple driver handlers that are
>>> configured differently.
>>>
>>> This series also allows dropping all of the PMF quirks from amd-pmf.
>>>
>>> v4:
>>>   * Drop the list; iterate classes
>>>   * Drop patches that didn't make sense without list
>>>   * Cover alienware-wmi as well (recently merged to
>>> platform-x86/for-next)
>>>   * Drop requirement for balanced
>>>   * Rename platform-profile class members to 'profile', 'options',
>>> 'name'
>>>   * Drop the name in /sys/class/platform-profile and just use ida
>>> value.
>>>     IE platform-profile-0
>>>
>>> Mario Limonciello (20):
>>>    ACPI: platform-profile: Add a name member to handlers
>>>    platform/x86/dell: dell-pc: Create platform device
>>>    ACPI: platform_profile: Add device pointer into platform profile
>>>      handler
>>>    ACPI: platform_profile: Add platform handler argument to
>>>      platform_profile_remove()
>>>    ACPI: platform_profile: Move sanity check out of the mutex
>>>    ACPI: platform_profile: Move matching string for new profile out of
>>>      mutex
>>>    ACPI: platform_profile: Use guard(mutex) for register/unregister
>>>    ACPI: platform_profile: Use `scoped_cond_guard`
>>>    ACPI: platform_profile: Create class for ACPI platform profile
>>>    ACPI: platform_profile: Add name attribute to class interface
>>>    ACPI: platform_profile: Add choices attribute for class interface
>>>    ACPI: platform_profile: Add profile attribute for class interface
>>>    ACPI: platform_profile: Notify change events on register and
>>>      unregister
>>>    ACPI: platform_profile: Only show profiles common for all handlers
>>>    ACPI: platform_profile: Add concept of a "custom" profile
>>>    ACPI: platform_profile: Make sure all profile handlers agree on
>>>      profile
>>>    ACPI: platform_profile: Check all profile handler to calculate next
>>>    ACPI: platform_profile: Allow multiple handlers
>>>    platform/x86/amd: pmf: Drop all quirks
>>>    Documentation: Add documentation about class interface for platform
>>>      profiles
>>>
>>>   .../userspace-api/sysfs-platform_profile.rst  |  28 ++
>>>   drivers/acpi/platform_profile.c               | 446
>>> ++++++++++++++----
>>>   .../surface/surface_platform_profile.c        |   8 +-
>>>   drivers/platform/x86/acer-wmi.c               |  10 +-
>>>   drivers/platform/x86/amd/pmf/Makefile         |   2 +-
>>>   drivers/platform/x86/amd/pmf/core.c           |   1 -
>>>   drivers/platform/x86/amd/pmf/pmf-quirks.c     |  66 ---
>>>   drivers/platform/x86/amd/pmf/pmf.h            |   3 -
>>>   drivers/platform/x86/amd/pmf/sps.c            |   4 +-
>>>   drivers/platform/x86/asus-wmi.c               |   6 +-
>>>   drivers/platform/x86/dell/alienware-wmi.c     |   8 +-
>>>   drivers/platform/x86/dell/dell-pc.c           |  39 +-
>>>   drivers/platform/x86/hp/hp-wmi.c              |   8 +-
>>>   drivers/platform/x86/ideapad-laptop.c         |   4 +-
>>>   .../platform/x86/inspur_platform_profile.c    |   7 +-
>>>   drivers/platform/x86/thinkpad_acpi.c          |   4 +-
>>>   include/linux/platform_profile.h              |   7 +-
>>>   17 files changed, 456 insertions(+), 195 deletions(-)
>>>   delete mode 100644 drivers/platform/x86/amd/pmf/pmf-quirks.c
>>>
>>>
>>> base-commit: d68cb6023356af3bd3193983ad4ec03954a0b3e2
>>> --
>>> 2.43.0
>>
>> For the series - I tried it out on my T14s G6 AMD, and it all looks
>> to be working nicely and as expected.
>> The thinkpad-acpi and amd-pmf classes show up. I can tweak them
>> individually and 'custom' shows up under firmware/acpi/platform_profile.
>> I tried various combo's and didn't see any issues.
>> If you have any recommendations of things to try let me know.
>>
>> Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
>>
>> Mark
>
> Thanks for testing!  The fact that Lenovo and ASUS are both doing this
> with recent laptops makes it all the stronger of a case to do it
> rather than let them race for who gets the coveted platform profile slot.
>
> If you've already tried mixing and matching combinations that's all I
> was going to suggest.
>
> I'll wait for Armin's review and then I'll respin with the tags and
> small things you and Ilpo found.
>
Looking forward to the next iteration.

Thanks,
Armin Wolf

> BTW if I missed any important feedback of yours from v3 please let me
> know.  It was a lot of overhaul to switch to this way of doing things
> and a lot of it didn't make sense anymore.
>