Merge branch 'asoc-5.7' into asoc-next

Merge branch 'asoc-5.6' into asoc-linus

Merge series "ASoC: Intel: add SoundWire machine driver" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

To handle multiple hardware combinations, this patchset suggests a
single machine driver which will create and initialize dailinks
dynamically. This allows us to support new configurations easily, as
shown with the TigerLake rt5682 example.

Each configuration updates the card component string, and UCM can test
for the presence of components to configure them as needed.

Since we use a single the machine driver name, all previous ACPI
tables need to be updated. That should have no impact since the
machine drivers listed at the time were not upstreamed and are no
longer maintained.

Naveen Manohar (2):
  ASoC: Intel: common: add match table for TGL RT5682 SoundWire driver
  ASoC: Intel: sof_sdw: Add Volteer support with RT5682 SNDW helper
    function

Pierre-Louis Bossart (1):
  ASoC: Intel: boards: add sof_sdw machine driver

Rander Wang (1):
  ASoC: Intel: soc-acpi: update topology and driver name for SoundWire
    platforms

 sound/soc/intel/boards/Kconfig                |  24 +
 sound/soc/intel/boards/Makefile               |   8 +-
 sound/soc/intel/boards/sof_sdw.c              | 962 ++++++++++++++++++
 sound/soc/intel/boards/sof_sdw_common.h       | 114 +++
 sound/soc/intel/boards/sof_sdw_dmic.c         |  42 +
 sound/soc/intel/boards/sof_sdw_hdmi.c         |  97 ++
 sound/soc/intel/boards/sof_sdw_rt1308.c       | 151 +++
 sound/soc/intel/boards/sof_sdw_rt5682.c       | 126 +++
 sound/soc/intel/boards/sof_sdw_rt700.c        | 125 +++
 sound/soc/intel/boards/sof_sdw_rt711.c        | 156 +++
 sound/soc/intel/boards/sof_sdw_rt715.c        |  42 +
 .../intel/common/soc-acpi-intel-cml-match.c   |  24 +-
 .../intel/common/soc-acpi-intel-icl-match.c   |   6 +-
 .../intel/common/soc-acpi-intel-tgl-match.c   |  30 +-
 14 files changed, 1896 insertions(+), 11 deletions(-)
 create mode 100644 sound/soc/intel/boards/sof_sdw.c
 create mode 100644 sound/soc/intel/boards/sof_sdw_common.h
 create mode 100644 sound/soc/intel/boards/sof_sdw_dmic.c
 create mode 100644 sound/soc/intel/boards/sof_sdw_hdmi.c
 create mode 100644 sound/soc/intel/boards/sof_sdw_rt1308.c
 create mode 100644 sound/soc/intel/boards/sof_sdw_rt5682.c
 create mode 100644 sound/soc/intel/boards/sof_sdw_rt700.c
 create mode 100644 sound/soc/intel/boards/sof_sdw_rt711.c
 create mode 100644 sound/soc/intel/boards/sof_sdw_rt715.c

--
2.20.1

ASoC: amd: acp3x-pcm-dma: clean up two indentation issues

There are a couple of statements that are not indented correctly,
add in the missing tab and break the lines to address a checkpatch
warning.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/20200327141429.269191-1-colin.king@canonical.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tlv320adcx140: Remove undocumented property

Remove undocumented and unneeded ti,use-internal-reg from the example as
it was an artifact from initial development.  The code does not query
for this property and as the document indicates if areg-supply is
undefined then the internal regulator is used.

Fixes: 302c0b7490cd ("dt-bindings: sound: Add TLV320ADCx140 dt
bindings")
Signed-off-by: Dan Murphy <dmurphy@ti.com>
CC: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200327162432.17067-1-dmurphy@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_sdw: Add Volteer support with RT5682 SNDW helper function

Add support for Google Volteer device. As per new unified soundwire machine
driver, add rt5682-sdw helper function, which configures codec to Link0.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Naveen Manohar <naveen.m@intel.com>
Link: https://lore.kernel.org/r/20200325220746.29601-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: common: add match table for TGL RT5682 SoundWire driver

RT5682 is in SoundWire mode on link0.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Naveen Manohar <naveen.m@intel.com>
Link: https://lore.kernel.org/r/20200325220746.29601-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: boards: add sof_sdw machine driver

This machine driver provides support for different configurations:

RT700, RT711, RT1308 (1x and 2x, I2S or SoundWire mode), and RT715
CometLake, Icelake, TigerLake.
PDM digital microphones
HDMI

To avoid introducing one driver per configuration, this common machine
driver relies on platform-specific information, tables and quirks to
dynamically create the relevant dailinks.

Unlike a lot of machine drivers, we use different DAI links for
SoundWire capture and playback since the Cadence PDIs can do capture
OR playback, not both simultaneously.

For each configuration, the card component string is updated so that UCM
can select the relevant parts.

Signed-off-by: Rander Wang <rander.wang@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20200325220746.29601-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: soc-acpi: update topology and driver name for SoundWire platforms

Update topology and reflect change to unified machine driver for SoundWire.

Signed-off-by: Rander Wang <rander.wang@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325220746.29601-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge branch 'for-5.6' of https://git./linux/kernel/git/broonie/sound into asoc-5.7

Merge series "ASoC: remove rtd->cpu/codec_dai{s}" from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:

Hi Mark

Now, CPU/Codec DAI(s) were replaced by rtd->dais.
Thus, We don't need rtd->cpu/codec_dai{s} anymore.
This pathset replaces it by new macro.

Kuninori Morimoto (36):
  ASoC: soc-core: add asoc_rtd_to_cpu/codec() macro
  ASoC: amd: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: atmel: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: au1x: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: bcm: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: cirrus: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: dwc: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: fsl: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: generic: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: img: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: intel: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: kirkwood: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: mediatek: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: meson: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: mxs: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: pxa: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: qcom: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: rockchip: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: samsung: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: sh: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: sof: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: sprd: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: stm: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: sunxi: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: tegra: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: ti: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: txx9: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: uniphier: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: ux500: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: xtensa: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: arm: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: codecs: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: soc: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer
  ASoC: soc-core: set rtd->num_cpu/codec at soc_new_pcm_runtime()
  ASoC: soc-core: tidyup soc_new_pcm_runtime() rtd setups
  ASoC: soc-core: remove cpu_dai/codec_dai/cpu_dais/codec_dais

 include/sound/soc.h                           | 30 +++++++------
 sound/arm/pxa2xx-pcm-lib.c                    |  8 ++--
 sound/soc/amd/acp-da7219-max98357a.c          |  2 +-
 sound/soc/amd/acp-rt5645.c                    |  4 +-
 sound/soc/amd/acp3x-rt5682-max9836.c          |  6 +--
 sound/soc/atmel/atmel-pcm-dma.c               |  4 +-
 sound/soc/atmel/atmel-pcm-pdc.c               |  2 +-
 sound/soc/atmel/atmel_wm8904.c                |  2 +-
 sound/soc/atmel/mikroe-proto.c                |  2 +-
 sound/soc/atmel/sam9g20_wm8731.c              |  2 +-
 sound/soc/atmel/sam9x5_wm8731.c               |  2 +-
 sound/soc/au1x/db1200.c                       |  2 +-
 sound/soc/au1x/dbdma2.c                       |  2 +-
 sound/soc/au1x/dma.c                          |  2 +-
 sound/soc/au1x/psc-ac97.c                     |  2 +-
 sound/soc/bcm/bcm63xx-pcm-whistler.c          | 16 +++----
 sound/soc/bcm/cygnus-pcm.c                    | 22 +++++-----
 sound/soc/cirrus/edb93xx.c                    |  4 +-
 sound/soc/cirrus/snappercl15.c                |  4 +-
 sound/soc/codecs/cs47l15.c                    |  4 +-
 sound/soc/codecs/cs47l24.c                    |  6 +--
 sound/soc/codecs/cs47l35.c                    |  6 +--
 sound/soc/codecs/cs47l85.c                    |  6 +--
 sound/soc/codecs/cs47l90.c                    |  6 +--
 sound/soc/codecs/cs47l92.c                    |  4 +-
 sound/soc/codecs/wm5110.c                     |  6 +--
 sound/soc/codecs/wm_adsp.c                    | 10 ++---
 sound/soc/dwc/dwc-pcm.c                       |  2 +-
 sound/soc/fsl/eukrea-tlv320.c                 |  4 +-
 sound/soc/fsl/fsl-asoc-card.c                 | 10 ++---
 sound/soc/fsl/fsl_asrc_dma.c                  |  6 +--
 sound/soc/fsl/fsl_spdif.c                     | 10 ++---
 sound/soc/fsl/fsl_ssi.c                       |  8 ++--
 sound/soc/fsl/imx-audmix.c                    |  8 ++--
 sound/soc/fsl/imx-mc13783.c                   |  4 +-
 sound/soc/fsl/imx-sgtl5000.c                  |  2 +-
 sound/soc/fsl/mpc5200_dma.c                   | 10 ++---
 sound/soc/fsl/mpc5200_psc_i2s.c               |  2 +-
 sound/soc/fsl/mpc8610_hpcd.c                  |  4 +-
 sound/soc/fsl/mx27vis-aic32x4.c               |  4 +-
 sound/soc/fsl/p1022_ds.c                      |  4 +-
 sound/soc/fsl/p1022_rdk.c                     |  4 +-
 sound/soc/fsl/wm1133-ev1.c                    |  6 +--
 sound/soc/generic/simple-card-utils.c         | 12 +++---
 sound/soc/img/img-i2s-in.c                    |  2 +-
 sound/soc/img/img-i2s-out.c                   |  2 +-
 sound/soc/intel/atom/sst-mfld-platform-pcm.c  |  6 +--
 sound/soc/intel/boards/bdw-rt5650.c           |  6 +--
 sound/soc/intel/boards/bdw-rt5677.c           |  6 +--
 sound/soc/intel/boards/broadwell.c            |  4 +-
 sound/soc/intel/boards/bxt_da7219_max98357a.c |  8 ++--
 sound/soc/intel/boards/bxt_rt298.c            |  8 ++--
 sound/soc/intel/boards/byt-max98090.c         |  2 +-
 sound/soc/intel/boards/byt-rt5640.c           |  4 +-
 sound/soc/intel/boards/bytcht_cx2072x.c       | 10 ++---
 sound/soc/intel/boards/bytcht_da7213.c        |  8 ++--
 sound/soc/intel/boards/bytcht_es8316.c        |  8 ++--
 sound/soc/intel/boards/bytcht_nocodec.c       |  4 +-
 sound/soc/intel/boards/bytcr_rt5640.c         |  8 ++--
 sound/soc/intel/boards/bytcr_rt5651.c         |  8 ++--
 sound/soc/intel/boards/cht_bsw_max98090_ti.c  |  6 +--
 sound/soc/intel/boards/cht_bsw_nau8824.c      |  4 +-
 sound/soc/intel/boards/cht_bsw_rt5645.c       | 14 +++----
 sound/soc/intel/boards/cht_bsw_rt5672.c       |  8 ++--
 sound/soc/intel/boards/cml_rt1011_rt5682.c    |  6 +--
 sound/soc/intel/boards/glk_rt5682_max98357a.c | 10 ++---
 sound/soc/intel/boards/haswell.c              |  2 +-
 sound/soc/intel/boards/kbl_da7219_max98357a.c |  8 ++--
 sound/soc/intel/boards/kbl_da7219_max98927.c  |  6 +--
 sound/soc/intel/boards/kbl_rt5660.c           |  6 +--
 sound/soc/intel/boards/kbl_rt5663_max98927.c  |  8 ++--
 .../intel/boards/kbl_rt5663_rt5514_max98927.c |  8 ++--
 .../soc/intel/boards/skl_nau88l25_max98357a.c | 12 +++---
 sound/soc/intel/boards/skl_nau88l25_ssm4567.c | 16 +++----
 sound/soc/intel/boards/skl_rt286.c            |  8 ++--
 sound/soc/intel/boards/sof_da7219_max98373.c  |  8 ++--
 sound/soc/intel/boards/sof_pcm512x.c          |  8 ++--
 sound/soc/intel/boards/sof_rt5682.c           |  6 +--
 sound/soc/intel/haswell/sst-haswell-pcm.c     | 26 ++++++------
 sound/soc/intel/skylake/skl-pcm.c             | 10 ++---
 sound/soc/kirkwood/armada-370-db.c            |  2 +-
 sound/soc/kirkwood/kirkwood-dma.c             |  2 +-
 sound/soc/mediatek/common/mtk-afe-fe-dai.c    | 10 ++---
 .../mediatek/common/mtk-afe-platform-driver.c |  2 +-
 sound/soc/mediatek/mt2701/mt2701-afe-pcm.c    |  2 +-
 sound/soc/mediatek/mt2701/mt2701-cs42448.c    |  4 +-
 sound/soc/mediatek/mt2701/mt2701-wm8960.c     |  4 +-
 sound/soc/mediatek/mt6797/mt6797-afe-pcm.c    |  2 +-
 sound/soc/mediatek/mt8173/mt8173-afe-pcm.c    |  2 +-
 sound/soc/mediatek/mt8173/mt8173-max98090.c   |  4 +-
 .../mediatek/mt8173/mt8173-rt5650-rt5514.c    |  2 +-
 .../mediatek/mt8173/mt8173-rt5650-rt5676.c    |  4 +-
 sound/soc/mediatek/mt8173/mt8173-rt5650.c     |  6 +--
 sound/soc/mediatek/mt8183/mt8183-afe-pcm.c    |  2 +-
 .../mediatek/mt8183/mt8183-da7219-max98357.c  |  4 +-
 .../mt8183/mt8183-mt6358-ts3a227-max98357.c   |  2 +-
 sound/soc/meson/aiu-fifo.c                    |  2 +-
 sound/soc/meson/axg-card.c                    |  8 ++--
 sound/soc/meson/axg-fifo.c                    |  2 +-
 sound/soc/meson/meson-card-utils.c            |  2 +-
 sound/soc/mxs/mxs-sgtl5000.c                  |  4 +-
 sound/soc/pxa/brownstone.c                    |  4 +-
 sound/soc/pxa/corgi.c                         |  4 +-
 sound/soc/pxa/hx4700.c                        |  4 +-
 sound/soc/pxa/imote2.c                        |  4 +-
 sound/soc/pxa/magician.c                      |  8 ++--
 sound/soc/pxa/mioa701_wm9713.c                |  4 +-
 sound/soc/pxa/mmp-pcm.c                       |  2 +-
 sound/soc/pxa/mmp-sspa.c                      |  2 +-
 sound/soc/pxa/poodle.c                        |  4 +-
 sound/soc/pxa/pxa2xx-i2s.c                    |  2 +-
 sound/soc/pxa/spitz.c                         |  4 +-
 sound/soc/pxa/ttc-dkb.c                       |  2 +-
 sound/soc/pxa/z2.c                            |  4 +-
 sound/soc/pxa/zylonite.c                      |  6 +--
 sound/soc/qcom/apq8016_sbc.c                  |  2 +-
 sound/soc/qcom/apq8096.c                      |  6 +--
 sound/soc/qcom/lpass-platform.c               |  2 +-
 sound/soc/qcom/qdsp6/q6asm-dai.c              |  4 +-
 sound/soc/qcom/qdsp6/q6routing.c              |  2 +-
 sound/soc/qcom/sdm845.c                       | 22 +++++-----
 sound/soc/qcom/storm.c                        |  2 +-
 sound/soc/rockchip/rk3288_hdmi_analog.c       |  4 +-
 sound/soc/rockchip/rk3399_gru_sound.c         | 16 +++----
 sound/soc/rockchip/rockchip_max98090.c        |  6 +--
 sound/soc/rockchip/rockchip_rt5645.c          |  6 +--
 sound/soc/samsung/arndale.c                   |  6 +--
 sound/soc/samsung/bells.c                     | 16 +++----
 sound/soc/samsung/h1940_uda1380.c             |  2 +-
 sound/soc/samsung/i2s.c                       |  2 +-
 sound/soc/samsung/jive_wm8750.c               |  4 +-
 sound/soc/samsung/littlemill.c                | 14 +++----
 sound/soc/samsung/lowland.c                   |  4 +-
 sound/soc/samsung/neo1973_wm8753.c            | 10 ++---
 sound/soc/samsung/odroid.c                    |  2 +-
 sound/soc/samsung/pcm.c                       |  4 +-
 sound/soc/samsung/rx1950_uda1380.c            |  2 +-
 sound/soc/samsung/s3c-i2s-v2.c                |  2 +-
 sound/soc/samsung/s3c24xx_simtec.c            |  4 +-
 sound/soc/samsung/s3c24xx_uda134x.c           |  6 +--
 sound/soc/samsung/smartq_wm8987.c             |  4 +-
 sound/soc/samsung/smdk_spdif.c                |  2 +-
 sound/soc/samsung/smdk_wm8580.c               |  2 +-
 sound/soc/samsung/smdk_wm8994.c               |  2 +-
 sound/soc/samsung/smdk_wm8994pcm.c            |  4 +-
 sound/soc/samsung/snow.c                      |  4 +-
 sound/soc/samsung/spdif.c                     |  8 ++--
 sound/soc/samsung/speyside.c                  |  8 ++--
 sound/soc/samsung/tm2_wm5110.c                | 16 +++----
 sound/soc/samsung/tobermory.c                 |  8 ++--
 sound/soc/sh/dma-sh7760.c                     | 16 +++----
 sound/soc/sh/fsi.c                            |  2 +-
 sound/soc/sh/migor.c                          |  6 +--
 sound/soc/sh/rcar/core.c                      |  2 +-
 sound/soc/soc-compress.c                      | 36 ++++++++--------
 sound/soc/soc-core.c                          | 42 +++++++------------
 sound/soc/soc-dapm.c                          |  4 +-
 sound/soc/soc-generic-dmaengine-pcm.c         |  6 +--
 sound/soc/soc-pcm.c                           | 30 ++++++-------
 sound/soc/sof/intel/hda-dai.c                 |  6 +--
 sound/soc/sof/intel/hda-dsp.c                 |  2 +-
 sound/soc/sprd/sprd-pcm-compress.c            |  4 +-
 sound/soc/sprd/sprd-pcm-dma.c                 |  2 +-
 sound/soc/stm/stm32_adfsdm.c                  | 12 +++---
 sound/soc/stm/stm32_sai_sub.c                 |  2 +-
 sound/soc/sunxi/sun4i-spdif.c                 |  2 +-
 sound/soc/tegra/tegra_alc5632.c               |  2 +-
 sound/soc/tegra/tegra_max98090.c              |  2 +-
 sound/soc/tegra/tegra_rt5640.c                |  2 +-
 sound/soc/tegra/tegra_rt5677.c                |  2 +-
 sound/soc/tegra/tegra_sgtl5000.c              |  2 +-
 sound/soc/tegra/tegra_wm8753.c                |  2 +-
 sound/soc/tegra/tegra_wm8903.c                |  6 +--
 sound/soc/tegra/trimslice.c                   |  2 +-
 sound/soc/ti/ams-delta.c                      |  4 +-
 sound/soc/ti/davinci-evm.c                    |  4 +-
 sound/soc/ti/davinci-vcif.c                   |  4 +-
 sound/soc/ti/n810.c                           |  2 +-
 sound/soc/ti/omap-abe-twl6040.c               |  6 +--
 sound/soc/ti/omap-mcbsp-st.c                  |  2 +-
 sound/soc/ti/omap-mcbsp.c                     |  4 +-
 sound/soc/ti/omap-mcpdm.c                     |  2 +-
 sound/soc/ti/omap3pandora.c                   |  4 +-
 sound/soc/ti/osk5912.c                        |  2 +-
 sound/soc/ti/rx51.c                           |  2 +-
 sound/soc/txx9/txx9aclc.c                     |  2 +-
 sound/soc/uniphier/aio-compress.c             | 22 +++++-----
 sound/soc/uniphier/aio-dma.c                  |  6 +--
 sound/soc/ux500/mop500_ab8500.c               |  6 +--
 sound/soc/ux500/ux500_pcm.c                   |  8 ++--
 sound/soc/xtensa/xtfpga-i2s.c                 |  2 +-
 191 files changed, 573 insertions(+), 577 deletions(-)

--
2.17.1

Merge series "ASoC: SOF: Intel: add SoundWire support" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

This patchset provides the support for SoundWire support on Intel
CometLake, IcelLake and TigerLake RVP platforms and form-factor
devices to be released 'soon'.

The bulk of the code is about detecting a valid SoundWire
configuration from ACPI, and implementing the interfaces suggested in
'[PATCH 0/8] soundwire: remove platform devices, add SOF interfaces'
for interrupts, PCI wakes and clock-stop configurations.

Since that SoundWire series will not be in 5.7, the build support for
SOF w/ SoundWire is not provided for now, and fall-back functions will
be used. This code is tested on a daily basis in the SOF tree and is
not expected to change in significant ways.

Changes since v2:
Corrected error in ACPI table (thanks Amadeusz)
Added patch 11 to add reset cycle required on some SoundWire platforms

Bard Liao (1):
  ASoC: SOF: Intel: hda: merge IPC, stream and SoundWire interrupt
    handlers

Pierre-Louis Bossart (8):
  ASoC: soc-acpi: expand description of _ADR-based devices
  ASoC: SOF: Intel: add SoundWire configuration interface
  ASoC: SOF: IPC: dai-intel: move ALH declarations in header file
  ASoC: SOF: Intel: hda: add SoundWire stream config/free callbacks
  ASoC: SOF: Intel: hda: initial SoundWire machine driver autodetect
  ASoC: SOF: Intel: hda: disable SoundWire interrupts on suspend
  ASoC: SOF: Intel: hda: add parameter to control SoundWire clock stop
    quirks
  ASoC: SOF: Intel: hda-ctrl: add reset cycle before parsing
    capabilities

Rander Wang (2):
  ASoC: SOF: Intel: hda: add WAKEEN interrupt support for SoundWire
  Asoc: SOF: Intel: hda: check SoundWire wakeen interrupt in irq thread

 include/sound/soc-acpi.h                      |  39 +-
 include/sound/sof/dai-intel.h                 |  18 +-
 .../intel/common/soc-acpi-intel-cml-match.c   |  87 +++-
 .../intel/common/soc-acpi-intel-icl-match.c   |  97 ++++-
 .../intel/common/soc-acpi-intel-tgl-match.c   |  49 ++-
 sound/soc/sof/intel/hda-ctrl.c                |  25 +-
 sound/soc/sof/intel/hda-dsp.c                 |   2 +
 sound/soc/sof/intel/hda-loader.c              |  31 ++
 sound/soc/sof/intel/hda.c                     | 400 ++++++++++++++++++
 sound/soc/sof/intel/hda.h                     |  66 +++
 10 files changed, 750 insertions(+), 64 deletions(-)

--
2.20.1

ASoC: rt5682: move DAI clock registry to I2S mode

The SoundWire mode doesn't need the DAI clocks.
Therefore, the DAI clock registry moves to I2S mode case.

Signed-off-by: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20200327073849.18291-1-shumingf@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pxa: magician: convert to use i2c_new_client_device()

Move away from the deprecated API and return the shiny new ERRPTR where
useful.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20200326211010.13471-2-wsa+renesas@sang-engineering.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda-ctrl: add reset cycle before parsing capabilities

Without this cycle, HDaudio capability parsing fails on some devices.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-12-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Asoc: SOF: Intel: hda: check SoundWire wakeen interrupt in irq thread

If pci device is in D0, wakeen interrupt will be
aggregated at cAVS level as interrupt. This commit
check the wakeen status and process it in irq thread

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-11-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: add WAKEEN interrupt support for SoundWire

When a SoundWire link is in clock stop state, a Slave device may wake
up the Master for some events such as jack detection. The WAKEEN
interrupt will be triggered and processed by the audio pci device.

If audio device is in D3, the interrupt will be routed to PME, or
aggregated at cAVS level as interrupt when audio device is in D0. This
patch only supports D3 case, where the audio pci device will be
resumed by a PME event and the WAKEEN interrupt will be processed
after audio pci device is powered up and ROM is initialized
successfully.

The WAKEEN handling is only enabled after the first boot due to
dependencies on a shim_lock mutex being initialized.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-10-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: add parameter to control SoundWire clock stop quirks

Add module parameter so that the different modes can be quickly tested.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-9-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: merge IPC, stream and SoundWire interrupt handlers

We have a single irq handler for SOF interrupts. We can further merge
SoundWire ones to completely remove MSI interrupts handling issues
leading to timeouts.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-8-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: disable SoundWire interrupts on suspend

Doing this avoid conflicts and errors reported on the bus.

The interrupts are only re-enabled on resume after the firmware is
downloaded, so the behavior is not fully symmetric

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-7-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: initial SoundWire machine driver autodetect

For now we have a limited number of machine driver configurations, and
we can detect them based on the link configuration returned after
checking hardware and firmware (BIOS) configurations.

The link configuration is checked with a link_mask as well as a list
of _ADR descriptors for each link.

There is a chance that in extreme cases where the BIOS contains too
much information we would need to detect which Slave devices actually
report as 'attached'. This would be more accurate than static
table-based solutions, but it also introduces timing dependencies
since we don't know when those devices might become attached, so will
only be only be looked at if we see limitations with static methods
and the usual quirks based e.g. on DMI information.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-6-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: add SoundWire stream config/free callbacks

These callbacks are invoked when a matching hw_params/hw_free() DAI
operation takes place, and will result in IPC operations with the SOF
firmware.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: IPC: dai-intel: move ALH declarations in header file

ALH was inserted in the wrong place during integration, add after DMIC
to mirror the file used by SOF firmware.

No functional change, just text move in the same file to better track
changes, if any.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: add SoundWire configuration interface

Now that the SoundWire core supports the multi-step initialization,
call the relevant APIs.

The actual hardware enablement can be done in two places, ideally we'd
want to startup the SoundWire IP as soon as possible (while still
taking power rail dependencies into account)

However when suspend/resume is implemented, the DSP device will be
resumed first, and only when the DSP firmware is downloaded/booted
would the SoundWire child devices be resumed, so there are only
marginal benefits in starting the IP earlier for the first probe.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-acpi: expand description of _ADR-based devices

For SoundWire, we need to know if endpoints needs to be 'aggregated'
(MIPI parlance, meaning logically grouped), e.g. when two speaker
amplifiers need to be handled as a single logical output.

We don't necessarily have the information at the firmware (BIOS)
level, so add a notion of endpoints and specify if a device/endpoint
is part of a group, with a position.

This may be expanded in future solutions, for now only provide a group
and position information.

Since we modify the header file, change all existing upstream tables
as well to avoid breaking compilation/bisect.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87wo7bhci3.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: arm: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87y2rrhcia.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: xtensa: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87zhc7hcih.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: ux500: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/871rpjir34.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: uniphier: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87369zir3b.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: txx9: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/874kufir3i.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: ti: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/875zevir3p.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tegra: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/877dzbir3w.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sunxi: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/878sjrir43.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: stm: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87a747ir4b.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sprd: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87blonir4j.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sof: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87d093ir4q.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sh: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87eetjir4x.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: samsung: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87ftdzir57.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rockchip: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87h7yfir5i.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87imivir5q.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pxa: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87k13bir62.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mxs: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87lfnrir6c.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: meson: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87mu87ir6j.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87o8snir6s.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: kirkwood: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87pnd3ir71.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: intel: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87r1xjir7a.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: img: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87sghzir7m.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: generic: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87tv2fir7y.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87v9mvir89.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dwc: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87wo7bir8j.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cirrus: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87y2rrir8s.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: bcm: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87zhc7ir94.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: au1x: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/871rpjk5tw.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: atmel: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87369zk5ul.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: amd: use asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for DAI pointer

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/874kufk5uu.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-core: add asoc_rtd_to_cpu/codec() macro

Now, snd_soc_pcm_runtime supports multi cpu_dai/codec_dai.
It still has cpu_dai/codec_dai for single DAI,
and has cpu_dais/codec_dais for multi DAIs.

dais  = [][][][][][][][][][][][][][][][][][]
^cpu_dais         ^codec_dais
|--- num_cpus ---|--- num_codecs --|

/* for multi DAIs */
rtd->cpu_dais   = &rtd->dais[0];
rtd->codec_dais = &rtd->dais[dai_link->num_cpus];

/* for single DAI */
rtd->cpu_dai    = rtd->cpu_dais[0];
rtd->codec_dai = rtd->codec_dais[0];

But, these can be replaced by dais.
This patch adds asoc_rtd_to_cpu() / asoc_rtd_to_codec() macro for it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/875zevk5va.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: rt1308-sdw: configure amplifier with set_tdm_slot()" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

When two (or more) amplifiers are on the same link, the integrator may:
a) assign dedicated slots for each of the amplifiers.
b) provide the same configuration to all amplifiers, and rely on
additional controls/processing in the amplifier to generate different
outputs.

case a) was the initial direction for SoundWire and is required for
amplifiers with limited capabilities, but to deal with orientation or
'posture' changes it's easier to implement case b) when the amplifier
can deal with multiple channels.

This patchset suggest the use of the set_tdm_slot() API to define
which of the channels will be consumed by what amplifiers. This maps
well with SoundWire's 'ChannelEnable' registers. The notion of
slot_width is however irrelevant here and ignored, and SoundWire ports
are typically single direction, so only one of the two masks shall be
used.

Pierre-Louis Bossart (2):
  ASoC: rt1308-sdw: add set_tdm_slot() support
  ASoC: rt1308-sdw: use slot and rx_mask to configure stream

 sound/soc/codecs/rt1308-sdw.c | 38 +++++++++++++++++++++++++++++++----
 sound/soc/codecs/rt1308-sdw.h |  2 ++
 2 files changed, 36 insertions(+), 4 deletions(-)

--
2.20.1

Merge series "ASoC: Intel: machine driver updates" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

One correction for GeminiLake and 2 additional machine drivers for
Chromebooks.

Curtis Malainey (1):
  ASoC: Intel: Make glk+rt5682 echo ref dynamic

Sathyanarayana Nujella (2):
  ASoC: Intel: sof_rt5682: Add support for tgl-max98373-rt5682
  ASoC: Intel: common: Add mach table for tgl-max98373-rt5682

Yong Zhi (1):
  ASoC: intel: sof_da7219_max98373: Add speaker switch

 sound/soc/intel/boards/Kconfig                |  1 +
 sound/soc/intel/boards/Makefile               |  2 +-
 sound/soc/intel/boards/glk_rt5682_max98357a.c |  1 +
 sound/soc/intel/boards/sof_da7219_max98373.c  | 67 +++++++++-------
 sound/soc/intel/boards/sof_maxim_common.c     | 80 +++++++++++++++++++
 sound/soc/intel/boards/sof_maxim_common.h     | 24 ++++++
 sound/soc/intel/boards/sof_rt5682.c           | 21 +++++
 .../intel/common/soc-acpi-intel-tgl-match.c   | 13 +++
 8 files changed, 178 insertions(+), 31 deletions(-)
 create mode 100644 sound/soc/intel/boards/sof_maxim_common.c
 create mode 100644 sound/soc/intel/boards/sof_maxim_common.h

--
2.20.1

Merge series "ASoC: SOF: cleanups and improvements" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

Couple of small patches to improve error handling, inits, logs.

Hope we can have this for 5.7?

Guennadi Liakhovetski (2):
  ASoC: SOF: (cosmetic) use for_each_pcm_streams() in sof_dai_load()
  ASoC: SOF: fix uninitialised "work" with VirtIO

Kai Vehmanen (2):
  ASoC: SOF: Intel: hda: do not leave clock gating off upon error
  ASoC: SOF: Intel: hda: call codec wake at chip init

Ranjani Sridharan (1):
  ASoC: SOF: Intel: hda: Improve DSP state logging

 sound/soc/sof/intel/hda-ctrl.c | 15 +++++++++---
 sound/soc/sof/intel/hda-dsp.c  | 43 ++++++++++++++++++++++++++++++++--
 sound/soc/sof/pcm.c            |  4 +---
 sound/soc/sof/sof-audio.h      |  3 +++
 sound/soc/sof/topology.c       | 17 ++++++++++----
 5 files changed, 69 insertions(+), 13 deletions(-)

--
2.20.1

ASoC: pxa: Enable AC'97 bus support for PXA machines

The AC'97 based PXA machines currently don't build reliably as they don't
ensure that an AC'97 bus is built, causing at least eseries_pxa_defconfig
to fail to build. Add selects to fix this.

Reported-by: KernelCI <bot@kernelci.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20200326180116.21375-1-broonie@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pxa: Select regmap from AC'97 machines

regmap needs to be selected by users which for machine drivers that select
AC'97 CODEC drivers means that we need to also select regmap to ensure that
the CODEC driver will build if nothing else enables regmap as is likely for
such systems.

Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20200326151053.40806-1-broonie@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt1308-sdw: use slot and rx_mask to configure stream

If the DAI was configured with a set_tdm_slots() call, use the information.

A platform or machine driver may configure each amplifier to extract
different bitSlots from the frame, or extract the same data and use
processing to generate the relevant output. The latter case is easier
to handle in case of orientation changes.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325212905.28145-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt1308-sdw: add set_tdm_slot() support

Add ability to select which of the channels is used, or both, in case
two RT1308 amplifiers are located on the same link.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325212905.28145-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: call codec wake at chip init

Further align HDA init sequence to the legacy non-DSP HDA driver by
calling snd_hdac_set_codec_wakeup() during the chip init sequence.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-6-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: do not leave clock gating off upon error

The misc clock gating (MISCBDCGE) is disabled for controller reset and
reenabled once reset is complete.

Fix the case when error happens during reset, and clock gating is
left disabled. The clock gating should be reenabled also in this case.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: fix uninitialised "work" with VirtIO

In the VirtIO case the sof_pcm_open() function isn't called on the
host during guest streaming, which then leaves "work" structures
uninitialised. However it is then used to handle position update
messages from the DSP. Move their initialisation to immediately after
allocation of the containing structure.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: (cosmetic) use for_each_pcm_streams() in sof_dai_load()

Use for_each_pcm_streams() to enumerate streams in sof_dai_load()
instead of doing that manually.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: Improve DSP state logging

Improve the DSP power state logs with the state names
instead of values.

Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: common: Add mach table for tgl-max98373-rt5682

Update tgl mach table with: Maxim98373 Amp and ALC5682 hp codec.
Both of the codecs are on I2S bus.

Signed-off-by: Jairaj Arava <jairaj.arava@intel.com>
Signed-off-by: Sathyanarayana Nujella <sathyanarayana.nujella@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_rt5682: Add support for tgl-max98373-rt5682

This patch does the below:
1. Adds the driver data and updates quirk info for TGL
        with Max98373 speaker amp and ALC5682 headset codec.
2. Added max98373 speaker related code to common file for re-use.

Signed-off-by: Jairaj Arava <jairaj.arava@intel.com>
Signed-off-by: Sathyanarayana Nujella <sathyanarayana.nujella@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: intel: sof_da7219_max98373: Add speaker switch

Add "Spk Switch" and associated widget, route to max98360a
speaker amp for power saving, also remove the speaker_amp_init()
callback with complete separated tables for max98373 and max98360a.

Signed-off-by: Bhat, Uday M <uday.m.bhat@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Yong Zhi <yong.zhi@intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: Make glk+rt5682 echo ref dynamic

Without the dynamic flag to allow runtime routing, the card cannot
probe on chromebooks because SOF is constantly waiting for the link.
Adding flag back to allow upstream kernels to work on rt5682 based
chromebooks since SOF can now ignore the hard coded front end.

Signed-off-by: Curtis Malainey <cujomalainey@chromium.org>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mchp-i2s-mcc: make signed 1 bit bitfields unsigned

The signed 1 bit bitfields should be unsigned, so make them unsigned.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20200325132913.110115-1-colin.king@canonical.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: bdw-rt5650: Revert SSP0 link to use dummy components

Recent series of patches targeting broadwell boards, while enabling
SOF, changed behavior for non-SOF solutions. In essence replacing
platform 'dummy' with actual 'platform' causes redundant stream
initialization to occur during audio start. hw_params for haswell-pcm
destroys initial stream right after its creation - only to recreate it
again from proceed from there.

While harmless so far, this flow isn't right and should be corrected.
The actual need for dummy components for SSP0 link is questionable but
that issue is subject for another series.

Fixes: a40acc6bfceb ("ASoC: Intel: bdw-rt5650: change cpu_dai and platform components for SOF")
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325131611.545-4-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: bdw-rt5677: Revert SSP0 link to use dummy components

Recent series of patches targeting broadwell boards, while enabling
SOF, changed behavior for non-SOF solutions. In essence replacing
platform 'dummy' with actual 'platform' causes redundant stream
initialization to occur during audio start. hw_params for haswell-pcm
destroys initial stream right after its creation - only to recreate it
again from proceed from there.

While harmless so far, this flow isn't right and should be corrected.
The actual need for dummy components for SSP0 link is questionable but
that issue is subject for another series.

Fixes: 4865bde187b2 ("ASoC: Intel: bdw-rt5677: change cpu_dai and platform components for SOF")
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325131611.545-3-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: broadwell: Revert back SSP0 link to use dummy components

Recent series of patches targeting broadwell boards, while enabling
SOF, changed behavior for non-SOF solutions. In essence replacing
platform 'dummy' with actual 'platform' causes redundant stream
initialization to occur during audio start. hw_params for haswell-pcm
destroys initial stream right after its creation - only to recreate it
again from proceed from there.

While harmless so far, this flow isn't correct and should be corrected.
The actual need for dummy components for SSP0 link is questionable but
that issue is subject for another series.

Link to first message in conversation:
https://lkml.org/lkml/2020/3/18/54

Fixes: 64df6afa0dab ("ASoC: Intel: broadwell: change cpu_dai and platform components for SOF")
Reported-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325131611.545-2-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Convert jz4740-i2s doc to YAML

Convert the textual binding documentation for the AIC (AC97/I2S
Controller) of Ingenic SoCs to a YAML schema, and add the new compatible
strings in the process.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200306222931.39664-1-paul@crapouillou.net
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: jz4740-i2s: Add support for the JZ4770

Before the JZ4770, the playback and capture sampling rates had to match.
The JZ4770 supports independent sampling rates for both.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Link: https://lore.kernel.org/r/20200306222931.39664-6-paul@crapouillou.net
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: jz4740-i2s: Add support for the JZ4760

The change of offset for the {rx,tx}_threshold fields in the conf
register predates the JZ4780, and was first introduced in the JZ4760.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Link: https://lore.kernel.org/r/20200306222931.39664-5-paul@crapouillou.net
Signed-off-by: Mark Brown <broonie@kernel.org>

dt-bindings: sound: rockchip-i2s: add #sound-dai-cells property

'#sound-dai-cells' is required to properly interpret
the list of DAI specified in the 'sound-dai' property,
so add them to 'rockchip-i2s.yaml'

Signed-off-by: Johan Jonker <jbx6244@gmail.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200324094149.6904-2-jbx6244@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

dt-bindings: sound: convert rockchip i2s bindings to yaml

Current dts files with 'i2s' nodes are manually verified.
In order to automate this process rockchip-i2s.txt
has to be converted to yaml.

Signed-off-by: Johan Jonker <jbx6244@gmail.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200324094149.6904-1-jbx6244@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8974: remove unused variables

sound/soc/codecs/wm8974.c:200:38: warning:
 wm8974_aux_boost_controls defined but not used [-Wunused-const-variable=]
sound/soc/codecs/wm8974.c:204:38: warning:
 wm8974_mic_boost_controls defined but not used [-Wunused-const-variable=]

commit 8a123ee2a46d ("ASoC: WM8974 DAPM cleanups")
left behind this, remove them.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20200324070615.16248-1-yuehaibing@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tas2562: Fixed incorrect amp_level setting.

According to the tas2562 datasheet,the bits[5:1] represents the amp_level value.
So to set the amp_level value correctly,the shift value should be set to 1.

Signed-off-by: Jonghwan Choi <charlie.jh@kakaocorp.com>
Acked-by: Dan Murphy <dmurphy@ti.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20200319140043.GA6688@jhbirdchoi-MS-7B79
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "Support built-in Mic on Tegra boards that use WM8903" from Dmitry Osipenko <digetx@gmail.com>:

Hello,

This small series adds audio route for built-in microphone on NVIDIA Tegra
boards that use WM8903 CODEC. In particular this is needed in order to unmute
internal microphone on Acer A500 tablet device. I'm planning to send out the
device tree for the A500 for 5.8, so will be nice to get the microphone
sorted out. Please review and apply, thanks in advance.

Dmitry Osipenko (2):
  dt-bindings: sound: tegra-wm8903: Document built-in microphone audio
    source
  ASoC: tegra: tegra_wm8903: Support DAPM events for built-in microphone

 .../sound/nvidia,tegra-audio-wm8903.txt        |  1 +
 sound/soc/tegra/tegra_wm8903.c                 | 18 ++++++++++++++++++
 2 files changed, 19 insertions(+)

--
2.25.1

ASoC: rt5682: Add the descriptions for the DMIC clock rate and delay settings

The patch adds the descriptions for the DMIC clock rate and delay settings.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200323082547.7898-3-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5682: Add a property for DMIC delay

The patch adds a property for DMIC delay (ms) to avoid pop noise and
changes the default delay setting.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200323082547.7898-2-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5682: Add a property for DMIC clock rate

The patch adds a property for DMIC clock rate (hz) and changes the
default to the common optimize DMIC clock rate.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200323082547.7898-1-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: MT6660: make spdxcheck.py happy

The SPDX-License-Identifier shall not be suffixed with anything further.
This makes ./scripts/spdxcheck.py complain:

  sound/soc/codecs/mt6660.c: 1:36 Invalid token: //

Clean up SPDX-License-Identifier line to make spdxcheck.py happy.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Link: https://lore.kernel.org/r/20200321114022.8545-1-lukas.bulwahn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tegra-wm8903: Document built-in microphone audio source

The internal microphone source is needed in order to be able to describe
the hardware audio routing for devices that have the built-in microphone
in addition to the external Mic Jack.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://lore.kernel.org/r/20200320205504.30466-2-digetx@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tegra: tegra_wm8903: Support DAPM events for built-in microphone

The enable-GPIO needs to be toggled on a DAPM event in order to turn
microphone ON/OFF, otherwise microphone won't work.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://lore.kernel.org/r/20200320205504.30466-3-digetx@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Linux 5.6-rc7

Merge tag 'for-5.6-rc6-tag' of git://git./linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "Two fixes.

  The first is a regression: when dropping some incompat bits the
  conditions were reversed. The other is a fix for rename whiteout
  potentially leaving stack memory linked to a list"

* tag 'for-5.6-rc6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: fix removal of raid[56|1c34} incompat flags after removing block group
  btrfs: fix log context list corruption after rename whiteout error

Merge branch 'akpm' (patches from Andrew)

Merge misc fixes from Andrew Morton:
 "10 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  x86/mm: split vmalloc_sync_all()
  mm, slub: prevent kmalloc_node crashes and memory leaks
  mm/mmu_notifier: silence PROVE_RCU_LIST warnings
  epoll: fix possible lost wakeup on epoll_ctl() path
  mm: do not allow MADV_PAGEOUT for CoW pages
  mm, memcg: throttle allocators based on ancestral memory.high
  mm, memcg: fix corruption on 64-bit divisor in memory.high throttling
  page-flags: fix a crash at SetPageError(THP_SWAP)
  mm/hotplug: fix hot remove failure in SPARSEMEM|!VMEMMAP case
  memcg: fix NULL pointer dereference in __mem_cgroup_usage_unregister_event

x86/mm: split vmalloc_sync_all()

Commit 3f8fd02b1bf1 ("mm/vmalloc: Sync unmappings in
__purge_vmap_area_lazy()") introduced a call to vmalloc_sync_all() in
the vunmap() code-path.  While this change was necessary to maintain
correctness on x86-32-pae kernels, it also adds additional cycles for
architectures that don't need it.

Specifically on x86-64 with CONFIG_VMAP_STACK=y some people reported
severe performance regressions in micro-benchmarks because it now also
calls the x86-64 implementation of vmalloc_sync_all() on vunmap().  But
the vmalloc_sync_all() implementation on x86-64 is only needed for newly
created mappings.

To avoid the unnecessary work on x86-64 and to gain the performance
back, split up vmalloc_sync_all() into two functions:

* vmalloc_sync_mappings(), and
* vmalloc_sync_unmappings()

Most call-sites to vmalloc_sync_all() only care about new mappings being
synchronized.  The only exception is the new call-site added in the
above mentioned commit.

Shile Zhang directed us to a report of an 80% regression in reaim
throughput.

Fixes: 3f8fd02b1bf1 ("mm/vmalloc: Sync unmappings in __purge_vmap_area_lazy()")
Reported-by: kernel test robot <oliver.sang@intel.com>
Reported-by: Shile Zhang <shile.zhang@linux.alibaba.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Tested-by: Borislav Petkov <bp@suse.de>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> [GHES]
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20191009124418.8286-1-joro@8bytes.org
Link: https://lists.01.org/hyperkitty/list/lkp@lists.01.org/thread/4D3JPPHBNOSPFK2KEPC6KGKS6J25AIDB/
Link: http://lkml.kernel.org/r/20191113095530.228959-1-shile.zhang@linux.alibaba.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm, slub: prevent kmalloc_node crashes and memory leaks

Sachin reports [1] a crash in SLUB __slab_alloc():

  BUG: Kernel NULL pointer dereference on read at 0x000073b0
  Faulting instruction address: 0xc0000000003d55f4
  Oops: Kernel access of bad area, sig: 11 [#1]
  LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries
  Modules linked in:
  CPU: 19 PID: 1 Comm: systemd Not tainted 5.6.0-rc2-next-20200218-autotest #1
  NIP:  c0000000003d55f4 LR: c0000000003d5b94 CTR: 0000000000000000
  REGS: c0000008b37836d0 TRAP: 0300   Not tainted  (5.6.0-rc2-next-20200218-autotest)
  MSR:  8000000000009033 <SF,EE,ME,IR,DR,RI,LE>  CR: 24004844  XER: 00000000
  CFAR: c00000000000dec4 DAR: 00000000000073b0 DSISR: 40000000 IRQMASK: 1
  GPR00: c0000000003d5b94 c0000008b3783960 c00000000155d400 c0000008b301f500
  GPR04: 0000000000000dc0 0000000000000002 c0000000003443d8 c0000008bb398620
  GPR08: 00000008ba2f0000 0000000000000001 0000000000000000 0000000000000000
  GPR12: 0000000024004844 c00000001ec52a00 0000000000000000 0000000000000000
  GPR16: c0000008a1b20048 c000000001595898 c000000001750c18 0000000000000002
  GPR20: c000000001750c28 c000000001624470 0000000fffffffe0 5deadbeef0000122
  GPR24: 0000000000000001 0000000000000dc0 0000000000000002 c0000000003443d8
  GPR28: c0000008b301f500 c0000008bb398620 0000000000000000 c00c000002287180
  NIP ___slab_alloc+0x1f4/0x760
  LR __slab_alloc+0x34/0x60
  Call Trace:
    ___slab_alloc+0x334/0x760 (unreliable)
    __slab_alloc+0x34/0x60
    __kmalloc_node+0x110/0x490
    kvmalloc_node+0x58/0x110
    mem_cgroup_css_online+0x108/0x270
    online_css+0x48/0xd0
    cgroup_apply_control_enable+0x2ec/0x4d0
    cgroup_mkdir+0x228/0x5f0
    kernfs_iop_mkdir+0x90/0xf0
    vfs_mkdir+0x110/0x230
    do_mkdirat+0xb0/0x1a0
    system_call+0x5c/0x68

This is a PowerPC platform with following NUMA topology:

  available: 2 nodes (0-1)
  node 0 cpus:
  node 0 size: 0 MB
  node 0 free: 0 MB
  node 1 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
  node 1 size: 35247 MB
  node 1 free: 30907 MB
  node distances:
  node   0   1
    0:  10  40
    1:  40  10

  possible numa nodes: 0-31

This only happens with a mmotm patch "mm/memcontrol.c: allocate
shrinker_map on appropriate NUMA node" [2] which effectively calls
kmalloc_node for each possible node.  SLUB however only allocates
kmem_cache_node on online N_NORMAL_MEMORY nodes, and relies on
node_to_mem_node to return such valid node for other nodes since commit
a561ce00b09e ("slub: fall back to node_to_mem_node() node if allocating
on memoryless node").  This is however not true in this configuration
where the _node_numa_mem_ array is not initialized for nodes 0 and 2-31,
thus it contains zeroes and get_partial() ends up accessing
non-allocated kmem_cache_node.

A related issue was reported by Bharata (originally by Ramachandran) [3]
where a similar PowerPC configuration, but with mainline kernel without
patch [2] ends up allocating large amounts of pages by kmalloc-1k
kmalloc-512.  This seems to have the same underlying issue with
node_to_mem_node() not behaving as expected, and might probably also
lead to an infinite loop with CONFIG_SLUB_CPU_PARTIAL [4].

This patch should fix both issues by not relying on node_to_mem_node()
anymore and instead simply falling back to NUMA_NO_NODE, when
kmalloc_node(node) is attempted for a node that's not online, or has no
usable memory.  The "usable memory" condition is also changed from
node_present_pages() to N_NORMAL_MEMORY node state, as that is exactly
the condition that SLUB uses to allocate kmem_cache_node structures.
The check in get_partial() is removed completely, as the checks in
___slab_alloc() are now sufficient to prevent get_partial() being
reached with an invalid node.

[1] https://lore.kernel.org/linux-next/3381CD91-AB3D-4773-BA04-E7A072A63968@linux.vnet.ibm.com/
[2] https://lore.kernel.org/linux-mm/fff0e636-4c36-ed10-281c-8cdb0687c839@virtuozzo.com/
[3] https://lore.kernel.org/linux-mm/20200317092624.GB22538@in.ibm.com/
[4] https://lore.kernel.org/linux-mm/088b5996-faae-8a56-ef9c-5b567125ae54@suse.cz/

Fixes: a561ce00b09e ("slub: fall back to node_to_mem_node() node if allocating on memoryless node")
Reported-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Reported-by: PUVICHAKRAVARTHY RAMACHANDRAN <puvichakravarthy@in.ibm.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Tested-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Tested-by: Bharata B Rao <bharata@linux.ibm.com>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christopher Lameter <cl@linux.com>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Kirill Tkhai <ktkhai@virtuozzo.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Nathan Lynch <nathanl@linux.ibm.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200320115533.9604-1-vbabka@suse.cz
Debugged-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/mmu_notifier: silence PROVE_RCU_LIST warnings

It is safe to traverse mm->notifier_subscriptions->list either under
SRCU read lock or mm->notifier_subscriptions->lock using
hlist_for_each_entry_rcu().  Silence the PROVE_RCU_LIST false positives,
for example,

  WARNING: suspicious RCU usage
  -----------------------------
  mm/mmu_notifier.c:484 RCU-list traversed in non-reader section!!

  other info that might help us debug this:

  rcu_scheduler_active = 2, debug_locks = 1
  3 locks held by libvirtd/802:
   #0: ffff9321e3f58148 (&mm->mmap_sem#2){++++}, at: do_mprotect_pkey+0xe1/0x3e0
   #1: ffffffff91ae6160 (mmu_notifier_invalidate_range_start){+.+.}, at: change_p4d_range+0x5fa/0x800
   #2: ffffffff91ae6e08 (srcu){....}, at: __mmu_notifier_invalidate_range_start+0x178/0x460

  stack backtrace:
  CPU: 7 PID: 802 Comm: libvirtd Tainted: G          I       5.6.0-rc6-next-20200317+ #2
  Hardware name: HP ProLiant BL460c Gen8, BIOS I31 11/02/2014
  Call Trace:
    dump_stack+0xa4/0xfe
    lockdep_rcu_suspicious+0xeb/0xf5
    __mmu_notifier_invalidate_range_start+0x3ff/0x460
    change_p4d_range+0x746/0x800
    change_protection+0x1df/0x300
    mprotect_fixup+0x245/0x3e0
    do_mprotect_pkey+0x23b/0x3e0
    __x64_sys_mprotect+0x51/0x70
    do_syscall_64+0x91/0xae8
    entry_SYSCALL_64_after_hwframe+0x49/0xb3

Signed-off-by: Qian Cai <cai@lca.pw>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Link: http://lkml.kernel.org/r/20200317175640.2047-1-cai@lca.pw
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

epoll: fix possible lost wakeup on epoll_ctl() path

This fixes possible lost wakeup introduced by commit a218cc491420.
Originally modifications to ep->wq were serialized by ep->wq.lock, but
in commit a218cc491420 ("epoll: use rwlock in order to reduce
ep_poll_callback() contention") a new rw lock was introduced in order to
relax fd event path, i.e. callers of ep_poll_callback() function.

After the change ep_modify and ep_insert (both are called on epoll_ctl()
path) were switched to ep->lock, but ep_poll (epoll_wait) was using
ep->wq.lock on wqueue list modification.

The bug doesn't lead to any wqueue list corruptions, because wake up
path and list modifications were serialized by ep->wq.lock internally,
but actual waitqueue_active() check prior wake_up() call can be
reordered with modifications of ep ready list, thus wake up can be lost.

And yes, can be healed by explicit smp_mb():

  list_add_tail(&epi->rdlink, &ep->rdllist);
  smp_mb();
  if (waitqueue_active(&ep->wq))
wake_up(&ep->wp);

But let's make it simple, thus current patch replaces ep->wq.lock with
the ep->lock for wqueue modifications, thus wake up path always observes
activeness of the wqueue correcty.

Fixes: a218cc491420 ("epoll: use rwlock in order to reduce ep_poll_callback() contention")
Reported-by: Max Neunhoeffer <max@arangodb.com>
Signed-off-by: Roman Penyaev <rpenyaev@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Tested-by: Max Neunhoeffer <max@arangodb.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Christopher Kohlhoff <chris.kohlhoff@clearpool.io>
Cc: Davidlohr Bueso <dbueso@suse.de>
Cc: Jason Baron <jbaron@akamai.com>
Cc: Jes Sorensen <jes.sorensen@gmail.com>
Cc: <stable@vger.kernel.org> [5.1+]
Link: http://lkml.kernel.org/r/20200214170211.561524-1-rpenyaev@suse.de
References: https://bugzilla.kernel.org/show_bug.cgi?id=205933
Bisected-by: Max Neunhoeffer <max@arangodb.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: do not allow MADV_PAGEOUT for CoW pages

Jann has brought up a very interesting point [1].  While shared pages
are excluded from MADV_PAGEOUT normally, CoW pages can be easily
reclaimed that way.  This can lead to all sorts of hard to debug
problems.  E.g.  performance problems outlined by Daniel [2].

There are runtime environments where there is a substantial memory
shared among security domains via CoW memory and a easy to reclaim way
of that memory, which MADV_{COLD,PAGEOUT} offers, can lead to either
performance degradation in for the parent process which might be more
privileged or even open side channel attacks.

The feasibility of the latter is not really clear to me TBH but there is
no real reason for exposure at this stage.  It seems there is no real
use case to depend on reclaiming CoW memory via madvise at this stage so
it is much easier to simply disallow it and this is what this patch
does.  Put it simply MADV_{PAGEOUT,COLD} can operate only on the
exclusively owned memory which is a straightforward semantic.

[1] http://lkml.kernel.org/r/CAG48ez0G3JkMq61gUmyQAaCq=_TwHbi1XKzWRooxZkv08PQKuw@mail.gmail.com
[2] http://lkml.kernel.org/r/CAKOZueua_v8jHCpmEtTB6f3i9e2YnmX4mqdYVWhV4E=Z-n+zRQ@mail.gmail.com

Fixes: 9c276cc65a58 ("mm: introduce MADV_COLD")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Daniel Colascione <dancol@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: "Joel Fernandes (Google)" <joel@joelfernandes.org>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200312082248.GS23944@dhcp22.suse.cz
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm, memcg: throttle allocators based on ancestral memory.high

Prior to this commit, we only directly check the affected cgroup's
memory.high against its usage.  However, it's possible that we are being
reclaimed as a result of hitting an ancestor memory.high and should be
penalised based on that, instead.

This patch changes memory.high overage throttling to use the largest
overage in its ancestors when considering how many penalty jiffies to
charge.  This makes sure that we penalise poorly behaving cgroups in the
same way regardless of at what level of the hierarchy memory.high was
breached.

Fixes: 0e4b01df8659 ("mm, memcg: throttle allocators when failing reclaim over memory.high")
Reported-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Chris Down <chris@chrisdown.name>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Nathan Chancellor <natechancellor@gmail.com>
Cc: Roman Gushchin <guro@fb.com>
Cc: <stable@vger.kernel.org> [5.4.x+]
Link: http://lkml.kernel.org/r/8cd132f84bd7e16cdb8fde3378cdbf05ba00d387.1584036142.git.chris@chrisdown.name
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>