mbox series

[0/3] ALSA: hda: bus cleanup

Message ID 20190808095715.29713-1-tiwai@suse.de (mailing list archive)
Headers show
Series ALSA: hda: bus cleanup | expand

Message

Takashi Iwai Aug. 8, 2019, 9:57 a.m. UTC
Hi,

this is a few patches to simplify and cleanup the HD-audio bus ops.

The first two patches translate the indirect calls of DMA page
allocation and MMIO accesses with the direct ones, as well as
eliminating the whole bus->io_ops.

The last one is SOF-specific, and fixes/cleans up by calling the
proper hdaudio bus init function, as formerly discussed.


Takashi

===

Takashi Iwai (3):
  ALSA: hda: Remove page allocation redirection
  ALSA: hda: Direct MMIO accesses
  ASoC: SOF: Intel: Initialize hdaudio bus properly

 include/sound/hdaudio.h                | 69 +++++++++++++--------------
 include/sound/hdaudio_ext.h            |  1 -
 sound/hda/Kconfig                      |  3 ++
 sound/hda/ext/hdac_ext_bus.c           | 60 +-----------------------
 sound/hda/hdac_bus.c                   | 36 ++++++++++++--
 sound/hda/hdac_controller.c            | 18 +++----
 sound/hda/hdac_stream.c                |  8 ++--
 sound/pci/hda/Kconfig                  |  1 +
 sound/pci/hda/hda_controller.c         |  6 +--
 sound/pci/hda/hda_controller.h         |  3 +-
 sound/pci/hda/hda_intel.c              | 71 ++--------------------------
 sound/pci/hda/hda_tegra.c              | 84 +--------------------------------
 sound/soc/intel/skylake/skl-messages.c | 15 +-----
 sound/soc/intel/skylake/skl.c          |  7 ++-
 sound/soc/sof/intel/hda-bus.c          | 85 ++++------------------------------
 sound/soc/sof/intel/hda-dsp.c          |  2 +-
 sound/soc/sof/intel/hda.c              |  6 +--
 sound/soc/sof/intel/hda.h              |  3 +-
 18 files changed, 107 insertions(+), 371 deletions(-)

Comments

Pierre-Louis Bossart Aug. 8, 2019, 2:23 p.m. UTC | #1
On 8/8/19 4:57 AM, Takashi Iwai wrote:
> Hi,
> 
> this is a few patches to simplify and cleanup the HD-audio bus ops.
> 
> The first two patches translate the indirect calls of DMA page
> allocation and MMIO accesses with the direct ones, as well as
> eliminating the whole bus->io_ops.
> 
> The last one is SOF-specific, and fixes/cleans up by calling the
> proper hdaudio bus init function, as formerly discussed.

This is a good cleanup, thanks Takashi.

For the series

Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>

One request from me since I am lazy: could we somehow merge branches 
between you / Mark since at the moment we have two sets of conflicts 
(NHLT and this one). We test all the upstream changes on hardware as 
soon as we can, and manual conflict resolution makes it slower/more 
error prone.

> 
> 
> Takashi
> 
> ===
> 
> Takashi Iwai (3):
>    ALSA: hda: Remove page allocation redirection
>    ALSA: hda: Direct MMIO accesses
>    ASoC: SOF: Intel: Initialize hdaudio bus properly
> 
>   include/sound/hdaudio.h                | 69 +++++++++++++--------------
>   include/sound/hdaudio_ext.h            |  1 -
>   sound/hda/Kconfig                      |  3 ++
>   sound/hda/ext/hdac_ext_bus.c           | 60 +-----------------------
>   sound/hda/hdac_bus.c                   | 36 ++++++++++++--
>   sound/hda/hdac_controller.c            | 18 +++----
>   sound/hda/hdac_stream.c                |  8 ++--
>   sound/pci/hda/Kconfig                  |  1 +
>   sound/pci/hda/hda_controller.c         |  6 +--
>   sound/pci/hda/hda_controller.h         |  3 +-
>   sound/pci/hda/hda_intel.c              | 71 ++--------------------------
>   sound/pci/hda/hda_tegra.c              | 84 +--------------------------------
>   sound/soc/intel/skylake/skl-messages.c | 15 +-----
>   sound/soc/intel/skylake/skl.c          |  7 ++-
>   sound/soc/sof/intel/hda-bus.c          | 85 ++++------------------------------
>   sound/soc/sof/intel/hda-dsp.c          |  2 +-
>   sound/soc/sof/intel/hda.c              |  6 +--
>   sound/soc/sof/intel/hda.h              |  3 +-
>   18 files changed, 107 insertions(+), 371 deletions(-)
>
Takashi Iwai Aug. 8, 2019, 2:45 p.m. UTC | #2
On Thu, 08 Aug 2019 16:23:40 +0200,
Pierre-Louis Bossart wrote:
> 
> One request from me since I am lazy: could we somehow merge branches
> between you / Mark since at the moment we have two sets of conflicts
> (NHLT and this one). We test all the upstream changes on hardware as
> soon as we can, and manual conflict resolution makes it slower/more
> error prone.

I don't mind who actually merges.  I'm going to push out my topic
branch once after gathering enough Ack's.  Then Mark can pull it into
his tree (as well as topic/hda-dmic branch).

Or if Mark can give me the branch, I can pull into my for-next
branch, too.  Basically my for-next branch is persistent and not
rebased, so you can work on it once when set.

Mark, just let me know your preference.


thanks,

Takashi
Mark Brown Aug. 8, 2019, 7:03 p.m. UTC | #3
On Thu, Aug 08, 2019 at 04:45:09PM +0200, Takashi Iwai wrote:

> I don't mind who actually merges.  I'm going to push out my topic
> branch once after gathering enough Ack's.  Then Mark can pull it into
> his tree (as well as topic/hda-dmic branch).

I can do that easily enough.

> Or if Mark can give me the branch, I can pull into my for-next
> branch, too.  Basically my for-next branch is persistent and not
> rebased, so you can work on it once when set.

> Mark, just let me know your preference.

Well, I don't have topic branches any more since Linus hates them and
it's always hard to write sensible summaries for the development when
it's half way through so...
Takashi Iwai Aug. 8, 2019, 8:38 p.m. UTC | #4
On Thu, 08 Aug 2019 21:03:43 +0200,
Mark Brown wrote:
> 
> On Thu, Aug 08, 2019 at 04:45:09PM +0200, Takashi Iwai wrote:
> 
> > I don't mind who actually merges.  I'm going to push out my topic
> > branch once after gathering enough Ack's.  Then Mark can pull it into
> > his tree (as well as topic/hda-dmic branch).
> 
> I can do that easily enough.

OK, I pushed out the branch topic/hda-bus-ops-cleanup.
Please merge this one and topic/hda-dmic branch into yours.


thanks,

Takashi
Mark Brown Aug. 8, 2019, 9:33 p.m. UTC | #5
On Thu, Aug 08, 2019 at 10:38:33PM +0200, Takashi Iwai wrote:
> Mark Brown wrote:

> > I can do that easily enough.

> OK, I pushed out the branch topic/hda-bus-ops-cleanup.
> Please merge this one and topic/hda-dmic branch into yours.

There were some conflicts which hopefully I resolved OK, it compiles and
everything so it can't be that bad right?
Mark Brown Aug. 8, 2019, 10:22 p.m. UTC | #6
On Thu, Aug 08, 2019 at 10:33:11PM +0100, Mark Brown wrote:
> On Thu, Aug 08, 2019 at 10:38:33PM +0200, Takashi Iwai wrote:
> > Mark Brown wrote:
> 
> > > I can do that easily enough.
> 
> > OK, I pushed out the branch topic/hda-bus-ops-cleanup.
> > Please merge this one and topic/hda-dmic branch into yours.
> 
> There were some conflicts which hopefully I resolved OK, it compiles and
> everything so it can't be that bad right?

Had to rebuild this since git has a bug with recording merge conflicts,
just pushing out a new version now - c2f16a94a80497e4b28c27f9ca2cd6cd60706fb6.
Takashi Iwai Aug. 9, 2019, 5:58 a.m. UTC | #7
On Fri, 09 Aug 2019 00:22:01 +0200,
Mark Brown wrote:
> 
> On Thu, Aug 08, 2019 at 10:33:11PM +0100, Mark Brown wrote:
> > On Thu, Aug 08, 2019 at 10:38:33PM +0200, Takashi Iwai wrote:
> > > Mark Brown wrote:
> > 
> > > > I can do that easily enough.
> > 
> > > OK, I pushed out the branch topic/hda-bus-ops-cleanup.
> > > Please merge this one and topic/hda-dmic branch into yours.
> > 
> > There were some conflicts which hopefully I resolved OK, it compiles and
> > everything so it can't be that bad right?
> 
> Had to rebuild this since git has a bug with recording merge conflicts,
> just pushing out a new version now - c2f16a94a80497e4b28c27f9ca2cd6cd60706fb6.

Thanks, it looks good to me.

And Stephen resolved the conflicts in today's linux-next, too, so
let's compare later.


Takashi
Mark Brown Aug. 9, 2019, 11:38 a.m. UTC | #8
On Fri, Aug 09, 2019 at 07:58:06AM +0200, Takashi Iwai wrote:
> Mark Brown wrote:

> > Had to rebuild this since git has a bug with recording merge conflicts,
> > just pushing out a new version now - c2f16a94a80497e4b28c27f9ca2cd6cd60706fb6.

> Thanks, it looks good to me.

> And Stephen resolved the conflicts in today's linux-next, too, so
> let's compare later.

My tree didn't get updated in -next because there's some configurations
that don't build.