[v2,0/7] ALSA: HD-audio display power fixes

Message

Takashi Iwai Dec. 11, 2018, 1:53 p.m. UTC

Hi,

this patch a revised patchset for fixing possible HD-audio display
power unbalance.
Basically the "fix" is done by refactoring the whole relevant code.
It starts from the Intel HD-audio runtime PM refactoring, followed
by the display PM API change, and lots of code cleanups.

Since it changes the display power API function, it hits both legacy
and ASoC drivers.

The patches are kept in topic/hda-pm-refactor branch of my sound git
tree.  This will be an immutable branch once when merged to for-next,
so that it can be merged to ASoC tree if any conflicting change needs
to be applied.

The major change in v2 series is the complete removal of
AZX_DCAPS_I915_POWERWELL flag in patch#3, and other minor fixes
suggested by Pierre.


Takashi

===

Takashi Iwai (7):
  ALSA: hda/intel: Refactoring PM code
  ALSA: hda: Refactor display power management
  ALSA: hda/intel: Drop superfluous AZX_DCAPS_I915_POWERWELL checks
  ALSA: hda/intel: Properly free the display power at error path
  ALSA: hda: Make snd_hdac_display_power() void function
  ASoC: hdac_hdmi: Add missing display power-off at driver removal
  ALSA: hda/hdmi: Always set display_power_control for Intel HSW+ codecs

 include/sound/hda_codec.h      |   1 +
 include/sound/hda_component.h  |  11 +-
 include/sound/hdaudio.h        |   7 +-
 sound/hda/hdac_component.c     |  39 +++---
 sound/hda/hdac_device.c        |  17 ---
 sound/pci/hda/hda_codec.c      |  16 ++-
 sound/pci/hda/hda_controller.c |  11 --
 sound/pci/hda/hda_controller.h |   6 +-
 sound/pci/hda/hda_intel.c      | 220 +++++++++++++--------------------
 sound/pci/hda/patch_hdmi.c     |   8 +-
 sound/soc/codecs/hdac_hdmi.c   |  24 +---
 sound/soc/intel/skylake/skl.c  |  40 ++----
 12 files changed, 148 insertions(+), 252 deletions(-)

Comments

Takashi Iwai Dec. 13, 2018, 8:17 a.m. UTC | #1

On Tue, 11 Dec 2018 14:53:34 +0100,
Takashi Iwai wrote:
> 
> Hi,
> 
> this patch a revised patchset for fixing possible HD-audio display
> power unbalance.
> Basically the "fix" is done by refactoring the whole relevant code.
> It starts from the Intel HD-audio runtime PM refactoring, followed
> by the display PM API change, and lots of code cleanups.
> 
> Since it changes the display power API function, it hits both legacy
> and ASoC drivers.
> 
> The patches are kept in topic/hda-pm-refactor branch of my sound git
> tree.  This will be an immutable branch once when merged to for-next,
> so that it can be merged to ASoC tree if any conflicting change needs
> to be applied.

Now I merged the patchset to for-next branch, and I'll keep the
topic/hda-pm-refactor immutable.

Mark, if you happen to apply Intel ASoC patches that conflict with the
recent changes, feel free to merge that branch into yours.


thanks,

Takashi

Pierre-Louis Bossart Dec. 13, 2018, 2:37 p.m. UTC | #2

On 12/13/18 2:17 AM, Takashi Iwai wrote:
> On Tue, 11 Dec 2018 14:53:34 +0100,
> Takashi Iwai wrote:
>> Hi,
>>
>> this patch a revised patchset for fixing possible HD-audio display
>> power unbalance.
>> Basically the "fix" is done by refactoring the whole relevant code.
>> It starts from the Intel HD-audio runtime PM refactoring, followed
>> by the display PM API change, and lots of code cleanups.
>>
>> Since it changes the display power API function, it hits both legacy
>> and ASoC drivers.
>>
>> The patches are kept in topic/hda-pm-refactor branch of my sound git
>> tree.  This will be an immutable branch once when merged to for-next,
>> so that it can be merged to ASoC tree if any conflicting change needs
>> to be applied.
> Now I merged the patchset to for-next branch, and I'll keep the
> topic/hda-pm-refactor immutable.
>
> Mark, if you happen to apply Intel ASoC patches that conflict with the
> recent changes, feel free to merge that branch into yours.
>
The next batch of SOF patches will depend on these changes, so having 
Takashi's topic/hda-pm-refactor merged into Mark's for-next branch would 
help. Thanks!