platform/kernel/linux-rpi.git
3 years agoMerge series "ASoC: wm8962: Conversion to json-schema and fix" from Geert Uytterhoeve...
Mark Brown [Mon, 25 Oct 2021 16:37:51 +0000 (17:37 +0100)]
Merge series "ASoC: wm8962: Conversion to json-schema and fix" from Geert Uytterhoeven <geert+renesas@glider.be>:

Hi all,

This patch series converts the Wolfson WM8962 Device Tree binding
documentation to json-schema, after fixing an issue in the imx8mn-beacon
DTS file.

Thanks for your comments!

Geert Uytterhoeven (2):
  arm64: dts: imx: imx8mn-beacon: Drop undocumented clock-names
    reference
  ASoC: dt-bindings: wlf,wm8962: Convert to json-schema

 .../devicetree/bindings/sound/wlf,wm8962.yaml | 118 ++++++++++++++++++
 .../devicetree/bindings/sound/wm8962.txt      |  43 -------
 .../freescale/imx8mn-beacon-baseboard.dtsi    |   1 -
 3 files changed, 118 insertions(+), 44 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/sound/wlf,wm8962.yaml
 delete mode 100644 Documentation/devicetree/bindings/sound/wm8962.txt

--
2.25.1

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
    -- Linus Torvalds

3 years agoASoC: dt-bindings: wlf,wm8962: Convert to json-schema
Geert Uytterhoeven [Mon, 18 Oct 2021 13:59:03 +0000 (15:59 +0200)]
ASoC: dt-bindings: wlf,wm8962: Convert to json-schema

Convert the Wolfson WM8962 Ultra-Low Power Stereo CODEC Device Tree
binding documentation to json-schema.

Add missing *-supply and port properties.
Update the example.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/b0868d2f62fd57499c79d96298e99e5f9e4fbc76.1634565154.git.geert+renesas@glider.be
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: cs42l42: Prevent NULL pointer deref in interrupt handler
Richard Fitzgerald [Mon, 25 Oct 2021 11:22:58 +0000 (12:22 +0100)]
ASoC: cs42l42: Prevent NULL pointer deref in interrupt handler

The interrupt handling code was getting the struct device* from a
struct snd_soc_component* stored in struct cs42l42_private. If the
interrupt was asserted before ASoC calls component_probe() the
snd_soc_component* will be NULL.

The stored snd_soc_component* is not actually used for anything other
than indirectly getting the struct device*. Remove it, and store the
struct device* in struct cs42l42_private.

Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20211025112258.9282-1-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: wm8962: Convert to devm_clk_get_optional()
Geert Uytterhoeven [Mon, 18 Oct 2021 14:01:11 +0000 (16:01 +0200)]
ASoC: wm8962: Convert to devm_clk_get_optional()

Use the existing devm_clk_get_optional() helper instead of building a
similar construct on top of devm_clk_get() that fails to handle all
errors but -EPROBE_DEFER.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/c2a8a1a628804a4439732d02847e25c227083690.1634565564.git.geert+renesas@glider.be
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: fix unmet dependency on GPIOLIB for SND_SOC_MAX98357A
Julian Braha [Mon, 25 Oct 2021 01:06:15 +0000 (21:06 -0400)]
ASoC: fix unmet dependency on GPIOLIB for SND_SOC_MAX98357A

When SND_SOC_INTEL_KBL_DA7219_MAX98357A_MACH is selected,
and GPIOLIB is not selected, Kbuild gives the
following warnings:

WARNING: unmet direct dependencies detected for SND_SOC_MAX98357A
  Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && GPIOLIB [=n]
  Selected by [y]:
  - SND_SOC_INTEL_DA7219_MAX98357A_GENERIC [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_INTEL_MACH [=y]

WARNING: unmet direct dependencies detected for SND_SOC_DMIC
  Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && GPIOLIB [=n]
  Selected by [y]:
  - SND_SOC_INTEL_DA7219_MAX98357A_GENERIC [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_INTEL_MACH [=y]

WARNING: unmet direct dependencies detected for SND_SOC_INTEL_DA7219_MAX98357A_GENERIC
  Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_INTEL_MACH [=y] && GPIOLIB [=n]
  Selected by [y]:
  - SND_SOC_INTEL_KBL_DA7219_MAX98357A_MACH [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_INTEL_MACH [=y] && SND_SOC_INTEL_KBL [=y] && I2C [=y] && ACPI [=y] && (MFD_INTEL_LPSS [=y] || COMPILE_TEST [=n])

This is because SND_SOC_DMIC and SND_SOC_MAX98357A are
selected by SND_SOC_INTEL_DA7219_MAX98357A_GENERIC, which
is also selected by SND_SOC_INTEL_KBL_DA7219_MAX98357A_MACH.
However, the selectors do not depend on or select GPIOLIB,
despite SND_SOC_DMIC and SND_SOC_MAX98357A depending on GPIOLIB.

These unmet dependency bugs were detected by Kismet,
a static analysis tool for Kconfig. Please advise if this
is not the appropriate solution.

Signed-off-by: Julian Braha <julianbraha@gmail.com>
Link: https://lore.kernel.org/r/20211025010615.10070-1-julianbraha@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: cs35l41: Make cs35l41_remove() return void
Uwe Kleine-König [Wed, 20 Oct 2021 13:24:16 +0000 (15:24 +0200)]
ASoC: cs35l41: Make cs35l41_remove() return void

Up to now cs35l41_remove() returns zero unconditionally. Make it
return void instead which makes it easier to see in the callers that
there is no error to handle.

Also the return value of i2c, platform and spi remove callbacks is
ignored anyway.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20211020132416.30288-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoc: wm8900: Drop empty spi_driver remove callback
Uwe Kleine-König [Wed, 20 Oct 2021 12:57:26 +0000 (14:57 +0200)]
ASoc: wm8900: Drop empty spi_driver remove callback

A driver with a remove callback that just returns 0 behaves identically
to a driver with no remove callback at all. So simplify accordingly.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20211020125726.22946-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: tegra: Set default card name for Trimslice
Dmitry Osipenko [Sun, 24 Oct 2021 19:28:53 +0000 (22:28 +0300)]
ASoC: tegra: Set default card name for Trimslice

The default card name for Trimslice device should be "tegra-trimslice".
It got lost by accident during unification of machine sound drivers,
fix it.

Cc: <stable@vger.kernel.org>
Fixes: cc8f70f56039 ("ASoC: tegra: Unify ASoC machine drivers")
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Link: https://lore.kernel.org/r/20211024192853.21957-2-digetx@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: tegra: Restore AC97 support
Dmitry Osipenko [Sun, 24 Oct 2021 19:28:52 +0000 (22:28 +0300)]
ASoC: tegra: Restore AC97 support

The device-tree of AC97 codecs need to be parsed differently from I2S
codecs, plus codec device may need to be created. This was missed by the
patch that unified machine drivers into a single driver, fix it. It should
restore audio on Toradex Colibri board.

Cc: <stable@vger.kernel.org>
Fixes: cc8f70f56039 ("ASoC: tegra: Unify ASoC machine drivers")
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Link: https://lore.kernel.org/r/20211024192853.21957-1-digetx@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoc: wm8731: Drop empty spi_driver remove callback
Uwe Kleine-König [Wed, 20 Oct 2021 12:58:03 +0000 (14:58 +0200)]
ASoc: wm8731: Drop empty spi_driver remove callback

A driver with a remove callback that just returns 0 behaves identically
to a driver with no remove callback at all. So simplify accordingly.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20211020125803.23117-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: doc: update codec example code
John Keeping [Sun, 24 Oct 2021 15:17:31 +0000 (16:17 +0100)]
ASoC: doc: update codec example code

As the codec API has evolved the documentation has not kept up and still
uses old fields that have been removed.  Update the examples to
represent the current API.

Signed-off-by: John Keeping <john@metanate.com>
Link: https://lore.kernel.org/r/20211024151731.360638-1-john@metanate.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: acp: Add acp_machine struct for renoir platform.
Ajit Kumar Pandey [Mon, 25 Oct 2021 07:48:00 +0000 (13:18 +0530)]
ASoC: amd: acp: Add acp_machine struct for renoir platform.

Add acpi_mach struct for renoir platform to select machine driver
based on codec and amp ACPI id.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211025074808.471333-1-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoMerge series "ASoC: Add common modules support for ACP hw block" from Ajit Kumar...
Mark Brown [Sat, 23 Oct 2021 16:30:02 +0000 (17:30 +0100)]
Merge series "ASoC: Add common modules support for ACP hw block" from Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>:

changes since v3:
- rebase and fixes merge conflict.
- Fixed kernel autobot warning.

Ajit Kumar Pandey (8):
  ASoC: amd: Add common framework to support I2S on ACP SOC
  ASoC: amd: acp: Add I2S support on Renoir platform
  ASoC: amd: acp: Add callback for machine driver on ACP
  ASoC: amd: acp: Add generic machine driver support for ACP cards
  ASoC: amd: acp: Add legacy sound card support for Chrome audio
  ASoC: amd: acp: Add SOF audio support on Chrome board
  ASoC: amd: acp: Add support for Maxim amplifier codec
  ASoC: amd: acp: Add support for RT5682-VS codec

 sound/soc/amd/Kconfig                |   2 +
 sound/soc/amd/Makefile               |   1 +
 sound/soc/amd/acp/Kconfig            |  51 +++
 sound/soc/amd/acp/Makefile           |  26 ++
 sound/soc/amd/acp/acp-i2s.c          | 340 +++++++++++++++
 sound/soc/amd/acp/acp-legacy-mach.c  | 104 +++++
 sound/soc/amd/acp/acp-mach-common.c  | 600 +++++++++++++++++++++++++++
 sound/soc/amd/acp/acp-mach.h         |  57 +++
 sound/soc/amd/acp/acp-platform.c     | 315 ++++++++++++++
 sound/soc/amd/acp/acp-renoir.c       | 144 +++++++
 sound/soc/amd/acp/acp-sof-mach.c     | 131 ++++++
 sound/soc/amd/acp/amd.h              | 146 +++++++
 sound/soc/amd/acp/chip_offset_byte.h |  76 ++++
 13 files changed, 1993 insertions(+)
 create mode 100644 sound/soc/amd/acp/Kconfig
 create mode 100644 sound/soc/amd/acp/Makefile
 create mode 100644 sound/soc/amd/acp/acp-i2s.c
 create mode 100644 sound/soc/amd/acp/acp-legacy-mach.c
 create mode 100644 sound/soc/amd/acp/acp-mach-common.c
 create mode 100644 sound/soc/amd/acp/acp-mach.h
 create mode 100644 sound/soc/amd/acp/acp-platform.c
 create mode 100644 sound/soc/amd/acp/acp-renoir.c
 create mode 100644 sound/soc/amd/acp/acp-sof-mach.c
 create mode 100644 sound/soc/amd/acp/amd.h
 create mode 100644 sound/soc/amd/acp/chip_offset_byte.h

--
2.25.1

3 years agoMerge series "Add Yellow Carp platform ASoC driver" from Vijendar Mukunda <Vijendar...
Mark Brown [Sat, 23 Oct 2021 16:29:20 +0000 (17:29 +0100)]
Merge series "Add Yellow Carp platform ASoC driver" from Vijendar Mukunda <Vijendar.Mukunda@amd.com>:

Yellow Carp platform is new APU series based on acp6.x design.
This patch set adds an ASoC driver for the ACP (Audio CoProcessor)
block on AMD Yellow Carp APU with DMIC endpoint support.

changes since v1:
  - fixed SPDX commenting style in acp6x pci driver.
  - fixed Copyright commenting style in acp6x pci driver,
    pdm dma driver and machine driver files.
  - fixed local variable check logic in acp6x_power_on() function.
  - fixed Kconfig indentation issue
  - removed irq_flag local variable in irq handler.
  - add dmi info check in machine driver probe

Vijendar Mukunda (13):
  ASoC: amd: add Yellow Carp ACP6x IP register header
  ASoC: amd: add Yellow Carp ACP PCI driver
  ASoC: amd: add acp6x init/de-init functions
  ASoC: amd: add platform devices for acp6x pdm driver and dmic driver
  ASoC: amd: add acp6x pdm platform driver
  ASoC: amd: add acp6x irq handler
  ASoC: amd: add acp6x pdm driver dma ops
  ASoC: amd: add acp6x pci driver pm ops
  ASoC: amd: add acp6x pdm driver pm ops
  ASoC: amd: enable Yellow carp acp6x drivers build
  ASoC: amd: create platform device for acp6x machine driver
  ASoC: amd: add YC machine driver using dmic
  ASoC: amd: enable Yellow Carp platform machine driver build

 sound/soc/amd/Kconfig                     |  21 +
 sound/soc/amd/Makefile                    |   1 +
 sound/soc/amd/yc/Makefile                 |   9 +
 sound/soc/amd/yc/acp6x-mach.c             | 194 ++++++++++
 sound/soc/amd/yc/acp6x-pdm-dma.c          | 448 ++++++++++++++++++++++
 sound/soc/amd/yc/acp6x.h                  | 107 ++++++
 sound/soc/amd/yc/acp6x_chip_offset_byte.h | 444 +++++++++++++++++++++
 sound/soc/amd/yc/pci-acp6x.c              | 338 ++++++++++++++++
 8 files changed, 1562 insertions(+)
 create mode 100644 sound/soc/amd/yc/Makefile
 create mode 100644 sound/soc/amd/yc/acp6x-mach.c
 create mode 100644 sound/soc/amd/yc/acp6x-pdm-dma.c
 create mode 100644 sound/soc/amd/yc/acp6x.h
 create mode 100644 sound/soc/amd/yc/acp6x_chip_offset_byte.h
 create mode 100644 sound/soc/amd/yc/pci-acp6x.c

--
2.25.1

3 years agoMerge series "ASoC: meson: axg: fix TDM channel order sync" from Jerome Brunet <jbrun...
Mark Brown [Sat, 23 Oct 2021 13:32:19 +0000 (14:32 +0100)]
Merge series "ASoC: meson: axg: fix TDM channel order sync" from Jerome Brunet <jbrunet@baylibre.com>:

On the Amlogic AXG series, the TODDR FIFO may get out of sync with the TDM
decoder if the decoder is started before the FIFO. The channel appears
shifted in memory in an unpredictable way.

To fix this, the trick is to start the FIFO before the TDM decoder. This
way the FIFO is already waiting when the 1st channel is produced and it is
correctly placed in memory.

Jerome Brunet (2):
  ASoC: meson: axg-card: make links nonatomic
  ASoC: meson: axg-tdm-interface: manage formatters in trigger

 sound/soc/meson/axg-card.c          |  1 +
 sound/soc/meson/axg-tdm-interface.c | 26 +++++++++++++++++++++-----
 2 files changed, 22 insertions(+), 5 deletions(-)

--
2.33.0

3 years agoASoC: rt5682s: Downsizing the DAC volume scale
Derek Fang [Thu, 21 Oct 2021 12:03:03 +0000 (20:03 +0800)]
ASoC: rt5682s: Downsizing the DAC volume scale

Use 0.75db/step of DAC volume instead of 1.5 to get
a more smooth volume curve.

Signed-off-by: Derek Fang <derek.fang@realtek.com>
Link: https://lore.kernel.org/r/20211021120303.4601-1-derek.fang@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: meson: axg-tdm-interface: manage formatters in trigger
Jerome Brunet [Wed, 20 Oct 2021 11:42:17 +0000 (13:42 +0200)]
ASoC: meson: axg-tdm-interface: manage formatters in trigger

So far, the formatters have been reset/enabled using the .prepare()
callback. This was done in this callback because walking the formatters use
a mutex so it could not be done in .trigger(), which is atomic by default.

It turns out there is a problem on capture path of the AXG series.
The FIFO may get out of sync with the TDM decoder if the IP are not enabled
in a specific order. The FIFO must be enabled before the formatter starts
producing data. IOW, we must deal with FE before the BE. The .prepare()
callback is called on the BEs before the FE so it is not OK for the AXG.

The .trigger() callback order can be configured, and it deals with the FE
before the BEs by default. To solve our problem, we just need to start and
stop the formatters from the .trigger() callback. It is OK do so now that
the links have been made 'nonatomic' in the card driver.

Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20211020114217.133153-3-jbrunet@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: meson: axg-card: make links nonatomic
Jerome Brunet [Wed, 20 Oct 2021 11:42:16 +0000 (13:42 +0200)]
ASoC: meson: axg-card: make links nonatomic

Non atomic operations need to be performed in the trigger callback
of the TDM interfaces. Those are BEs but what matters is the nonatomic
flag of the FE in the DPCM context. Just set nonatomic for everything so,
at least, it is clear.

Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20211020114217.133153-2-jbrunet@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: cs42l42: Remove unused runtime_suspend/runtime_resume callbacks
Richard Fitzgerald [Mon, 18 Oct 2021 16:44:31 +0000 (17:44 +0100)]
ASoC: cs42l42: Remove unused runtime_suspend/runtime_resume callbacks

The driver has runtime_suspend and runtime_resume callbacks, but
pm_runtime is never enabled so these functions won't be called. They
could not be used anyway because the runtime_suspend would cause jack
detect to stop working.

These functions are unused - delete them.

Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20211018164431.5871-1-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: fix unmet dependency on GPIOLIB
Julian Braha [Sun, 10 Oct 2021 21:56:27 +0000 (17:56 -0400)]
ASoC: fix unmet dependency on GPIOLIB

When SND_SOC_SC7180 or SND_SOC_STORM is selected,
and GPIOLIB is not selected, Kbuild gives the following
warning:

WARNING: unmet direct dependencies detected for SND_SOC_MAX98357A
  Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && GPIOLIB [=n]
  Selected by [y]:
  - SND_SOC_STORM [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_QCOM [=y]
  - SND_SOC_SC7180 [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_QCOM [=y] && I2C [=y]

This is because SND_SOC_MAX98357A is selected
by SND_SOC_STORM and SND_SOC_SC7180, but
these config options do not select or depend on
GPIOLIB, despite SND_SOC_MAX98357A depending on
GPIOLIB.

These unmet dependency bugs were detected by Kismet,
a static analysis tool for Kconfig. Please advise if this
is not the appropriate solution.

Signed-off-by: Julian Braha <julianbraha@gmail.com>
Link: https://lore.kernel.org/r/20211010215627.17869-1-julianbraha@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoMerge branch 'asoc-5.15' into asoc-5.16
Mark Brown [Thu, 21 Oct 2021 13:41:32 +0000 (14:41 +0100)]
Merge branch 'asoc-5.15' into asoc-5.16

3 years agoMerge series "ASoC: qcom: sm8250: add support for TX and RX Macro dais" from Srinivas...
Mark Brown [Wed, 20 Oct 2021 11:11:40 +0000 (12:11 +0100)]
Merge series "ASoC: qcom: sm8250: add support for TX and RX Macro dais" from Srinivas Kandagatla <srinivas.kandagatla@linaro.org>:

This patchset adds support for testing WCD938X connected via TX and RX Macros
on SM8250 MTP.

Srinivas Kandagatla (2):
  ASoC: qcom: sm8250: add support for TX and RX Macro dais
  ASoC: qcom: sm8250: Add Jack support

 sound/soc/qcom/sm8250.c | 79 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 79 insertions(+)

--
2.21.0

3 years agoASoC: amd: acp: Add support for RT5682-VS codec
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:38 +0000 (12:39 +0530)]
ASoC: amd: acp: Add support for RT5682-VS codec

In newer variants primary codec is rt5682vs. Add support for newer
codec variants in generic machine driver module and define driver
data to register SOF sound card.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-9-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: acp: Add support for Maxim amplifier codec
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:37 +0000 (12:39 +0530)]
ASoC: amd: acp: Add support for Maxim amplifier codec

In newer chrome boards we have max98360a as an amplifier codec.
Add support for max98360a in generic machine driver and configure
driver data to enable SOF sound card support on newer boards .

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-8-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: acp: Add SOF audio support on Chrome board
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:36 +0000 (12:39 +0530)]
ASoC: amd: acp: Add SOF audio support on Chrome board

Chrome board has RT5682 codec and RT1019 amp connected to I2S SP
controller on ACP hw. Also it support DMIC capture endpoints with
inbuilt pdm controller on ACP hw block. Add driver module to create
backend dai links for sof dsp core. We pass driver data with audio
end points configuration to register sound cards and create device
nodes for all audio endpoints.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-7-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: acp: Add legacy sound card support for Chrome audio
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:35 +0000 (12:39 +0530)]
ASoC: amd: acp: Add legacy sound card support for Chrome audio

Renoir based Chrome board has RT5682 as primary headset codec and
RT1019 amp device connected to I2SSP ACP i2s controller. Add driver
to register legacy sound card devices on Chrome board.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-6-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: acp: Add generic machine driver support for ACP cards
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:34 +0000 (12:39 +0530)]
ASoC: amd: acp: Add generic machine driver support for ACP cards

We have machines with different audio endpoints configurations
across various distributions. We need to support multiple sound
cards for different combinations of I2S instance and codecs hw.
Now we also need to support SOF-DSP endpoints based sound cards.
All such card combinations slightly differs in terms of machine
ops callback. This patch adds ACP generic machine driver module
that exposes method to create ACP cards dai links and define new
ops for audio endpoints configurations. Initially we have added
dailink support for RT5682 and RT1019 codec connection with ACP
I2S_SP instance. We will add newer codecs in this module to use
this for all AMD's ACP block sound cards supports in future.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-5-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: acp: Add callback for machine driver on ACP
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:33 +0000 (12:39 +0530)]
ASoC: amd: acp: Add callback for machine driver on ACP

Add method to select and register machine driver for acp platform
based on ACPI ID.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-4-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: acp: Add I2S support on Renoir platform
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:32 +0000 (12:39 +0530)]
ASoC: amd: acp: Add I2S support on Renoir platform

Add I2S dai driver for Renoir platform and register with common
acp framework to support non dsp I2S use case on Renoir.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-3-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: Add common framework to support I2S on ACP SOC
Ajit Kumar Pandey [Tue, 19 Oct 2021 07:09:31 +0000 (12:39 +0530)]
ASoC: amd: Add common framework to support I2S on ACP SOC

We are using legacy way of exposing dais and DMA configuration that
requires separate driver modules for various ACP SOC with almost
similar hw configuration. Moreover the legacy approach requires
separate I2S and DMA module platform devices registration and need
machine specific quirk to control various I2S endpoints. Add generic
dai driver and platform driver for I2S controller on ACP hw block.
This common framework can be used by various ACP platform devices
that shares common specs.

Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@amd.com>
Link: https://lore.kernel.org/r/20211019070938.5076-2-AjitKumar.Pandey@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: rockchip: Use generic dmaengine code
Sugar Zhang [Tue, 28 Sep 2021 01:35:57 +0000 (09:35 +0800)]
ASoC: rockchip: Use generic dmaengine code

This reverts commit 75b31192fe6ad20b42276b20ee3bdf1493216d63.

The original purpose of customized pcm was to config prealloc buffer size
flexibly. but, we can do the same thing by soc-generic-dmaengine-pcm.

And the generic one can generated the better config by querying DMA
capabilities from dmaengine driver rather than the Hard-Coded one.

e.g.

the customized one:

  static const struct snd_pcm_hardware snd_rockchip_hardware = {
         .info                   = SNDRV_PCM_INFO_MMAP |
                                   SNDRV_PCM_INFO_MMAP_VALID |
                                   SNDRV_PCM_INFO_PAUSE |
                                   SNDRV_PCM_INFO_RESUME |
                                   SNDRV_PCM_INFO_INTERLEAVED,
  ...

the generic one:

  ret = dma_get_slave_caps(chan, &dma_caps);
  if (ret == 0) {
          if (dma_caps.cmd_pause && dma_caps.cmd_resume)
                  hw.info |= SNDRV_PCM_INFO_PAUSE | SNDRV_PCM_INFO_RESUME;
          if (dma_caps.residue_granularity <= DMA_RESIDUE_GRANULARITY_SEGMENT)
                  hw.info |= SNDRV_PCM_INFO_BATCH;
  ...

So, let's revert back to use the generic dmaengine pcm.

Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Reviewed-by: John Keeping <john@metanate.com>
Link: https://lore.kernel.org/r/1632792957-80428-1-git-send-email-sugar.zhang@rock-chips.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoMerge branch 'asoc-5.15' into asoc-5.16
Mark Brown [Mon, 18 Oct 2021 20:15:13 +0000 (21:15 +0100)]
Merge branch 'asoc-5.15' into asoc-5.16

3 years agoASoC: qcom: sm8250: Add Jack support
Srinivas Kandagatla [Wed, 6 Oct 2021 17:27:45 +0000 (18:27 +0100)]
ASoC: qcom: sm8250: Add Jack support

WCD938X on SM8250 MTP is connected via TX macro which has MBHC support,
So add this jack support in the soundcard driver too.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20211006172745.22103-3-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: qcom: sm8250: add support for TX and RX Macro dais
Srinivas Kandagatla [Wed, 6 Oct 2021 17:27:44 +0000 (18:27 +0100)]
ASoC: qcom: sm8250: add support for TX and RX Macro dais

On SM8250 MTP boards WCD938x codec is connected via TX and RX Macros,
so add support for this dais in the soundcard driver.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20211006172745.22103-2-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: enable Yellow Carp platform machine driver build
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:44 +0000 (16:50 +0530)]
ASoC: amd: enable Yellow Carp platform machine driver build

This patch enables Yellow Carp platform machine driver build.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-14-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add YC machine driver using dmic
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:43 +0000 (16:50 +0530)]
ASoC: amd: add YC machine driver using dmic

Add Yellow Carp platform machine driver using dmic.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-13-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: create platform device for acp6x machine driver
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:42 +0000 (16:50 +0530)]
ASoC: amd: create platform device for acp6x machine driver

Create platform device for acp6x machine driver.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-12-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: enable Yellow carp acp6x drivers build
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:41 +0000 (16:50 +0530)]
ASoC: amd: enable Yellow carp acp6x drivers build

Yellow Carp ACP6x drivers can be built by selecting necessary
kernel config option.
The patch enables build support of the same.

Signed-off-by: Vijendar Mukunda<Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-11-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add acp6x pdm driver pm ops
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:40 +0000 (16:50 +0530)]
ASoC: amd: add acp6x pdm driver pm ops

Add acp6x pdm driver pm ops.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-10-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add acp6x pci driver pm ops
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:39 +0000 (16:50 +0530)]
ASoC: amd: add acp6x pci driver pm ops

Add acp6x pci driver pm ops.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-9-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add acp6x pdm driver dma ops
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:38 +0000 (16:50 +0530)]
ASoC: amd: add acp6x pdm driver dma ops

This patch adds PDM driver DMA operations.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-8-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add acp6x irq handler
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:37 +0000 (16:50 +0530)]
ASoC: amd: add acp6x irq handler

Add ACP6x irq handler for handling irq events for ACP IP.
Add pdm irq events handling.
Whenever audio data equal to the PDM watermark level are consumed,
interrupt is generated. Acknowledge the interrupt.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-7-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add acp6x pdm platform driver
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:36 +0000 (16:50 +0530)]
ASoC: amd: add acp6x pdm platform driver

PDM platform driver binds to the platform device created by
ACP6x PCI device. PDM driver registers ALSA DMA and CPU DAI
components with ASoC framework.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-6-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add platform devices for acp6x pdm driver and dmic driver
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:35 +0000 (16:50 +0530)]
ASoC: amd: add platform devices for acp6x pdm driver and dmic driver

ACP6.x IP has PDM decoder block.
Create a platform device for it, so that the PDM platform driver
can be bound to this device.
Pass PCI resources like MMIO to this platform device.

Create a platform device for generic dmic codec driver.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-5-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add acp6x init/de-init functions
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:34 +0000 (16:50 +0530)]
ASoC: amd: add acp6x init/de-init functions

Add Yellow Carp platform ACP6x PCI driver init/deinit functions.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-4-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add Yellow Carp ACP PCI driver
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:33 +0000 (16:50 +0530)]
ASoC: amd: add Yellow Carp ACP PCI driver

ACP is a PCI audio device.
This patch adds PCI driver to bind to this device and get
PCI resources.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-3-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: amd: add Yellow Carp ACP6x IP register header
Vijendar Mukunda [Mon, 18 Oct 2021 11:20:32 +0000 (16:50 +0530)]
ASoC: amd: add Yellow Carp ACP6x IP register header

Add register header for ACP6x IP in Yellow Carp platform.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Link: https://lore.kernel.org/r/20211018112044.1705805-2-Vijendar.Mukunda@amd.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoMerge series "ASoC: cleanup / tidyup soc-pcm/core/component" from Kuninori Morimoto...
Mark Brown [Mon, 18 Oct 2021 20:10:09 +0000 (21:10 +0100)]
Merge series "ASoC: cleanup / tidyup soc-pcm/core/component" from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:

Hi Mark

These are not a big deal, but cleanup / tidyup patch for ALSA SoC.

Kuninori Morimoto (5):
  ASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step1
  ASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step2
  ASoC: soc-component: add snd_soc_component_is_codec()
  ASoC: soc-core: tidyup empty function
  ASoC: soc-core: accept zero format at snd_soc_runtime_set_dai_fmt()

 include/sound/soc-component.h         |  5 +++
 sound/soc/generic/simple-card-utils.c |  4 +--
 sound/soc/meson/meson-codec-glue.c    |  3 --
 sound/soc/soc-core.c                  | 51 ++++++++-------------------
 sound/soc/soc-pcm.c                   | 13 +++----
 5 files changed, 26 insertions(+), 50 deletions(-)

--
2.25.1

3 years agoASoC: dt-bindings: mediatek: rename reset controller headers in DT example
Tzung-Bi Shih [Mon, 18 Oct 2021 10:16:08 +0000 (18:16 +0800)]
ASoC: dt-bindings: mediatek: rename reset controller headers in DT example

Commit f07c776f6d7e ("arm64: dts: mediatek: Move reset controller
constants into common location") moves the reset controller headers.
However, it forgot to rename the DT example in mt8192-afe-pcm.yaml.

Renames the DT example to pass dt_binding_check.

Fixes: f07c776f6d7e ("arm64: dts: mediatek: Move reset controller constants into common location")
Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
Link: https://lore.kernel.org/r/20211018101608.3818840-1-tzungbi@google.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: rockchip: i2s-tdm: Fix refcount test
Nicolas Frattaroli [Fri, 15 Oct 2021 21:07:29 +0000 (23:07 +0200)]
ASoC: rockchip: i2s-tdm: Fix refcount test

During development of V5 of the i2s-tdm patch series, I replaced
the atomic refcount with a regular integer, as it was only ever
accessed within a spinlock.

Foolishly, I got the semantics of atomic_dec_and_test wrong, which
resulted in a test for 0 actually becoming a test for >0.

The result was that setting the audio frequency broke; switching
from 44100 Hz audio playback to 96000 Hz audio playback would
garble the sound most unpleasantly.

Fix this by checking for --refcount == 0, which is what it should
have been all along.

Fixes: 081068fd6414 ("ASoC: rockchip: add support for i2s-tdm controller")
Signed-off-by: Nicolas Frattaroli <frattaroli.nicolas@gmail.com>
Link: https://lore.kernel.org/r/20211015210730.308946-1-frattaroli.nicolas@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: meson: implement driver_name for snd_soc_card in meson-card-utils
Christian Hewitt [Sun, 17 Oct 2021 16:00:28 +0000 (16:00 +0000)]
ASoC: meson: implement driver_name for snd_soc_card in meson-card-utils

Implement driver_name to provide an alternative to card_name for userspace
configuration of Amlogic audio cards.

Suggested-by: Matthias Reichl <hias@horus.com>
Signed-off-by: Christian Hewitt <christianshewitt@gmail.com>
Acked-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20211017160028.23318-1-christianshewitt@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: tlv320aic32x4: Make aic32x4_remove() return void
Uwe Kleine-König [Fri, 15 Oct 2021 07:11:13 +0000 (09:11 +0200)]
ASoC: tlv320aic32x4: Make aic32x4_remove() return void

Up to now aic32x4_remove() returns zero unconditionally. Make it return
void instead which makes it easier to see in the callers that there is
no error to handle.

Also the return value of i2c and spi remove callbacks is ignored anyway.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20211015071113.2795767-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: soc-core: accept zero format at snd_soc_runtime_set_dai_fmt()
Kuninori Morimoto [Mon, 18 Oct 2021 02:05:44 +0000 (11:05 +0900)]
ASoC: soc-core: accept zero format at snd_soc_runtime_set_dai_fmt()

Do nothing if format was zero at snd_soc_runtime_set_dai_fmt().
soc-core.c can be more simple code by this patch.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87ee8jt7d3.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: soc-core: tidyup empty function
Kuninori Morimoto [Mon, 18 Oct 2021 02:05:34 +0000 (11:05 +0900)]
ASoC: soc-core: tidyup empty function

This patch makes empty function to 1 line.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87fsszt7dd.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: soc-component: add snd_soc_component_is_codec()
Kuninori Morimoto [Mon, 18 Oct 2021 02:05:24 +0000 (11:05 +0900)]
ASoC: soc-component: add snd_soc_component_is_codec()

Checking .non_legacy_dai_naming is not readable.
Let's add new snd_soc_component_is_codec().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87h7dft7dn.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step2
Kuninori Morimoto [Mon, 18 Oct 2021 02:05:05 +0000 (11:05 +0900)]
ASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step2

DAI active count is not exchanged during for_each_rtd_dais()
loops. We don't need to keep snd_soc_dai_stream_active() as
"active" on soc_pcm_hw_clean(). This patch avoid verbose code.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87ilxvt7e6.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step1
Kuninori Morimoto [Mon, 18 Oct 2021 02:04:52 +0000 (11:04 +0900)]
ASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step1

soc_pcm_hw_clean() is using "continue" during for_each_rtd_dais(),
but it is very verbose. This patch cleanup it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87k0ibt7ej.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoMerge tag 'v5.15-rc6' into asoc-5.16
Mark Brown [Mon, 18 Oct 2021 12:50:42 +0000 (13:50 +0100)]
Merge tag 'v5.15-rc6' into asoc-5.16

Linux 5.15-rc6

3 years agoLinux 5.15-rc6
Linus Torvalds [Mon, 18 Oct 2021 06:00:13 +0000 (20:00 -1000)]
Linux 5.15-rc6

3 years agoMerge tag 'libata-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal...
Linus Torvalds [Mon, 18 Oct 2021 05:39:22 +0000 (19:39 -1000)]
Merge tag 'libata-5.15-rc6' of git://git./linux/kernel/git/dlemoal/libata

Pull libata fixes from Damien Le Moal:
 "Two fixes for this cycle:

   - Fix a null pointer dereference in ahci-platform driver (from Hai)

   - Fix uninitialized variables in pata_legacy driver (from Dan)"

* tag 'libata-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/libata:
  ata: ahci_platform: fix null-ptr-deref in ahci_platform_enable_regulators()
  pata_legacy: fix a couple uninitialized variable bugs

3 years agoMerge tag 'block-5.15-2021-10-17' of git://git.kernel.dk/linux-block
Linus Torvalds [Mon, 18 Oct 2021 05:25:20 +0000 (19:25 -1000)]
Merge tag 'block-5.15-2021-10-17' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "Bigger than usual for this point in time, the majority is fixing some
  issues around BDI lifetimes with the move from the request_queue to
  the disk in this release. In detail:

   - Series on draining fs IO for del_gendisk() (Christoph)

   - NVMe pull request via Christoph:
        - fix the abort command id (Keith Busch)
        - nvme: fix per-namespace chardev deletion (Adam Manzanares)

   - brd locking scope fix (Tetsuo)

   - BFQ fix (Paolo)"

* tag 'block-5.15-2021-10-17' of git://git.kernel.dk/linux-block:
  block, bfq: reset last_bfqq_created on group change
  block: warn when putting the final reference on a registered disk
  brd: reduce the brd_devices_mutex scope
  kyber: avoid q->disk dereferences in trace points
  block: keep q_usage_counter in atomic mode after del_gendisk
  block: drain file system I/O on del_gendisk
  block: split bio_queue_enter from blk_queue_enter
  block: factor out a blk_try_enter_queue helper
  block: call submit_bio_checks under q_usage_counter
  nvme: fix per-namespace chardev deletion
  block/rnbd-clt-sysfs: fix a couple uninitialized variable bugs
  nvme-pci: Fix abort command id

3 years agoMerge tag 'io_uring-5.15-2021-10-17' of git://git.kernel.dk/linux-block
Linus Torvalds [Mon, 18 Oct 2021 05:20:13 +0000 (19:20 -1000)]
Merge tag 'io_uring-5.15-2021-10-17' of git://git.kernel.dk/linux-block

Pull io_uring fix from Jens Axboe:
 "Just a single fix for a wrong condition for grabbing a lock, a
  regression in this merge window"

* tag 'io_uring-5.15-2021-10-17' of git://git.kernel.dk/linux-block:
  io_uring: fix wrong condition to grab uring lock

3 years agoMerge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
Linus Torvalds [Mon, 18 Oct 2021 04:17:19 +0000 (18:17 -1000)]
Merge tag 'for_linus' of git://git./linux/kernel/git/mst/vhost

Pull virtio fixes from Michael Tsirkin:
 "Fixes up some issues in rc5"

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  vhost-vdpa: Fix the wrong input in config_cb
  VDUSE: fix documentation underline warning
  Revert "virtio-blk: Add validation for block size in config space"
  vhost_vdpa: unset vq irq before freeing irq
  virtio: write back F_VERSION_1 before validate

3 years agoMerge tag 'powerpc-5.15-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc...
Linus Torvalds [Mon, 18 Oct 2021 04:01:32 +0000 (18:01 -1000)]
Merge tag 'powerpc-5.15-4' of git://git./linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:

 - Fix a bug where guests on P9 with interrupts passed through could get
   stuck in synchronize_irq().

 - Fix a bug in KVM on P8 where secondary threads entering a guest would
   write outside their allocated stack.

 - Fix a bug in KVM on P8 where secondary threads could confuse the host
   offline code and cause the guest or host to crash.

Thanks to Cédric Le Goater.

* tag 'powerpc-5.15-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  KVM: PPC: Book3S HV: Make idle_kvm_start_guest() return 0 if it went to guest
  KVM: PPC: Book3S HV: Fix stack handling in idle_kvm_start_guest()
  powerpc/xive: Discard disabled interrupts in get_irqchip_state()

3 years agoMerge tag 'objtool_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 18 Oct 2021 03:41:39 +0000 (17:41 -1000)]
Merge tag 'objtool_urgent_for_v5.15_rc6' of git://git./linux/kernel/git/tip/tip

Pull objtool fixes from Borislav Petkov:

 - Update section headers before the respective relocations to not
   trigger a safety check in elftoolchain's implementation of libelf

 - Do not add garbage data to the .rela.orc_unwind_ip section

* tag 'objtool_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  objtool: Update section header before relocations
  objtool: Check for gelf_update_rel[a] failures

3 years agoMerge tag 'edac_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 18 Oct 2021 03:36:39 +0000 (17:36 -1000)]
Merge tag 'edac_urgent_for_v5.15_rc6' of git://git./linux/kernel/git/ras/ras

Pull EDAC fix from Borislav Petkov:

 - Log the "correct" uncorrectable error count in the armada_xp driver

* tag 'edac_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
  EDAC/armada-xp: Fix output of uncorrectable error counter

3 years agoMerge tag 'perf_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 18 Oct 2021 03:34:18 +0000 (17:34 -1000)]
Merge tag 'perf_urgent_for_v5.15_rc6' of git://git./linux/kernel/git/tip/tip

Pull perf fix from Borislav Petkov:

 - Add Sapphire Rapids to the list of CPUs supporting the SMI count MSR

* tag 'perf_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86/msr: Add Sapphire Rapids CPU support

3 years agoMerge tag 'efi-urgent-for-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Mon, 18 Oct 2021 03:30:49 +0000 (17:30 -1000)]
Merge tag 'efi-urgent-for-v5.15' of git://git./linux/kernel/git/tip/tip

Pull EFI fixes from Borislav Petkov:
 "Forwarded from Ard Biesheuvel through the tip tree. Ard will send
  stuff directly in the near future.

  Low priority fixes but fixes nonetheless:

   - update stub diagnostic print that is no longer accurate

   - avoid statically allocated buffer for CPER error record decoding

   - avoid sleeping on the efi_runtime semaphore when calling the
     ResetSystem EFI runtime service"

* tag 'efi-urgent-for-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi: Change down_interruptible() in virt_efi_reset_system() to down_trylock()
  efi/cper: use stack buffer for error record decoding
  efi/libstub: Simplify "Exiting bootservices" message

3 years agoMerge tag 'x86_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 18 Oct 2021 03:27:22 +0000 (17:27 -1000)]
Merge tag 'x86_urgent_for_v5.15_rc6' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Borislav Petkov:

 - Do not enable AMD memory encryption in Kconfig by default due to
   shortcomings of some platforms, leading to boot failures.

 - Mask out invalid bits in the MXCSR for 32-bit kernels again because
   Thomas and I don't know how to mask out bits properly. Third time's
   the charm.

* tag 'x86_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/fpu: Mask out the invalid MXCSR bits properly
  x86/Kconfig: Do not enable AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT automatically

3 years agoMerge tag 'driver-core-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Mon, 18 Oct 2021 03:17:28 +0000 (17:17 -1000)]
Merge tag 'driver-core-5.15-rc6' of git://git./linux/kernel/git/gregkh/driver-core

Pull driver core fixes from Greg KH:
 "Here are some small driver core fixes for 5.15-rc6, all of which have
  been in linux-next for a while with no reported issues.

  They include:

   - kernfs negative dentry bugfix

   - simple pm bus fixes to resolve reported issues"

* tag 'driver-core-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  drivers: bus: Delete CONFIG_SIMPLE_PM_BUS
  drivers: bus: simple-pm-bus: Add support for probing simple bus only devices
  driver core: Reject pointless SYNC_STATE_ONLY device links
  kernfs: don't create a negative dentry if inactive node exists

3 years agoMerge tag 'char-misc-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregk...
Linus Torvalds [Mon, 18 Oct 2021 03:14:00 +0000 (17:14 -1000)]
Merge tag 'char-misc-5.15-rc6' of git://git./linux/kernel/git/gregkh/char-misc

Pull char/misc driver fixes from Greg KH:
 "Here are some small char/misc driver fixes for 5.15-rc6 for reported
  issues that include:

   - habanalabs driver fixes

   - mei driver fixes and new ids

   - fpga new device ids

   - MAINTAINER file updates for fpga subsystem

   - spi module id table additions and fixes

   - fastrpc locking fixes

   - nvmem driver fix

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'char-misc-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  eeprom: 93xx46: fix MODULE_DEVICE_TABLE
  nvmem: Fix shift-out-of-bound (UBSAN) with byte size cells
  mei: hbm: drop hbm responses on early shutdown
  mei: me: add Ice Lake-N device id.
  eeprom: 93xx46: Add SPI device ID table
  eeprom: at25: Add SPI ID table
  misc: HI6421V600_IRQ should depend on HAS_IOMEM
  misc: fastrpc: Add missing lock before accessing find_vma()
  cb710: avoid NULL pointer subtraction
  misc: gehc: Add SPI ID table
  MAINTAINERS: Drop outdated FPGA Manager website
  MAINTAINERS: Add Hao and Yilun as maintainers
  habanalabs: fix resetting args in wait for CS IOCTL
  fpga: ice40-spi: Add SPI device ID table

3 years agoMerge tag 'staging-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Mon, 18 Oct 2021 03:10:00 +0000 (17:10 -1000)]
Merge tag 'staging-5.15-rc6' of git://git./linux/kernel/git/gregkh/staging

Pull staging and IIO driver fixes from Greg KH:
 "Here are a number of small IIO and staging driver fixes for 5.15-rc6.

  They include:

   - vc04_services bugfix for reported problem

   - r8188eu array underflow fix

   - iio driver fixes for a lot of tiny reported issues.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'staging-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: r8188eu: prevent array underflow in rtw_hal_update_ra_mask()
  staging: vc04_services: shut up out-of-range warning
  iio: light: opt3001: Fixed timeout error when 0 lux
  iio: adis16480: fix devices that do not support sleep mode
  iio: mtk-auxadc: fix case IIO_CHAN_INFO_PROCESSED
  iio: adis16475: fix deadlock on frequency set
  iio: ssp_sensors: add more range checking in ssp_parse_dataframe()
  iio: ssp_sensors: fix error code in ssp_print_mcu_debug()
  iio: adc: ad7793: Fix IRQ flag
  iio: adc: ad7780: Fix IRQ flag
  iio: adc: ad7192: Add IRQ flag
  iio: adc: aspeed: set driver data when adc probe.
  iio: adc: rzg2l_adc: add missing clk_disable_unprepare() in rzg2l_adc_pm_runtime_resume()
  iio: adc: max1027: Fix the number of max1X31 channels
  iio: adc: max1027: Fix wrong shift with 12-bit devices
  iio: adc128s052: Fix the error handling path of 'adc128_probe()'
  iio: adc: rzg2l_adc: Fix -EBUSY timeout error return
  iio: accel: fxls8962af: return IRQ_HANDLED when fifo is flushed
  iio: dac: ti-dac5571: fix an error code in probe()

3 years agoMerge tag 'tty-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Linus Torvalds [Mon, 18 Oct 2021 03:06:31 +0000 (17:06 -1000)]
Merge tag 'tty-5.15-rc6' of git://git./linux/kernel/git/gregkh/tty

Pull serial driver fix from Greg KH:
 "Here is a single 8250 Kconfig fix for 5.15-rc6 that resolves a
  regression that showed up in 5.15-rc1. It has been in linux-next for a
  while with no reported issues"

* tag 'tty-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  serial: 8250: allow disabling of Freescale 16550 compile test

3 years agoMerge tag 'usb-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Mon, 18 Oct 2021 03:02:00 +0000 (17:02 -1000)]
Merge tag 'usb-5.15-rc6' of git://git./linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are some small USB fixes that resolve a number of tiny issues.
  They include:

   - new USB serial driver ids

   - xhci driver fixes for a bunch of issues

   - musb error path fixes.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'usb-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: musb: dsps: Fix the probe error path
  xhci: Enable trust tx length quirk for Fresco FL11 USB controller
  xhci: Fix command ring pointer corruption while aborting a command
  USB: xhci: dbc: fix tty registration race
  xhci: add quirk for host controllers that don't update endpoint DCS
  xhci: guard accesses to ep_state in xhci_endpoint_reset()
  USB: serial: qcserial: add EM9191 QDL support
  USB: serial: option: add Quectel EC200S-CN module support
  USB: serial: option: add prod. id for Quectel EG91
  USB: serial: option: add Telit LE910Cx composition 0x1204

3 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Mon, 18 Oct 2021 02:57:06 +0000 (16:57 -1000)]
Merge branch 'for-linus' of git://git./linux/kernel/git/dtor/input

Pull input fixes from Dmitry Torokhov:

 - a new product ID for the xpad joystick driver

 - fixes to resistive-adc-touch and snvs_pwrkey drivers

 - a change to touchscreen helpers to make clang happier

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: touchscreen - avoid bitwise vs logical OR warning
  Input: xpad - add support for another USB ID of Nacon GC-100
  Input: resistive-adc-touch - fix division by zero error on z1 == 0
  Input: snvs_pwrkey - add clk handling

3 years agoblock, bfq: reset last_bfqq_created on group change
Paolo Valente [Fri, 15 Oct 2021 14:43:36 +0000 (16:43 +0200)]
block, bfq: reset last_bfqq_created on group change

Since commit 430a67f9d616 ("block, bfq: merge bursts of newly-created
queues"), BFQ maintains a per-group pointer to the last bfq_queue
created. If such a queue, say bfqq, happens to move to a different
group, then bfqq is no more a valid last bfq_queue created for its
previous group. That pointer must then be cleared. Not resetting such
a pointer may also cause UAF, if bfqq happens to also be freed after
being moved to a different group. This commit performs this missing
reset. As such it fixes commit 430a67f9d616 ("block, bfq: merge bursts
of newly-created queues").

Such a missing reset is most likely the cause of the crash reported in [1].
With some analysis, we found that this crash was due to the
above UAF. And such UAF did go away with this commit applied [1].

Anyway, before this commit, that crash happened to be triggered in
conjunction with commit 2d52c58b9c9b ("block, bfq: honor already-setup
queue merges"). The latter was then reverted by commit ebc69e897e17
("Revert "block, bfq: honor already-setup queue merges""). Yet commit
2d52c58b9c9b ("block, bfq: honor already-setup queue merges") contains
no error related with the above UAF, and can then be restored.

[1] https://bugzilla.kernel.org/show_bug.cgi?id=214503

Fixes: 430a67f9d616 ("block, bfq: merge bursts of newly-created queues")
Tested-by: Grzegorz Kowal <custos.mentis@gmail.com>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Link: https://lore.kernel.org/r/20211015144336.45894-2-paolo.valente@linaro.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agoblock: warn when putting the final reference on a registered disk
Christoph Hellwig [Thu, 14 Oct 2021 13:02:31 +0000 (15:02 +0200)]
block: warn when putting the final reference on a registered disk

Warn when the last reference on a live disk is put without calling
del_gendisk first.  There are some BDI related bug reports that look
like a case of this, so make sure we have the proper instrumentation
to catch it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211014130231.1468538-1-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agobrd: reduce the brd_devices_mutex scope
Tetsuo Handa [Tue, 7 Sep 2021 10:18:00 +0000 (19:18 +0900)]
brd: reduce the brd_devices_mutex scope

As with commit 8b52d8be86d72308 ("loop: reorder loop_exit"),
unregister_blkdev() needs to be called first in order to avoid calling
brd_alloc() from brd_probe() after brd_del_one() from brd_exit(). Then,
we can avoid holding global mutex during add_disk()/del_gendisk() as with
commit 1c500ad706383f1a ("loop: reduce the loop_ctl_mutex scope").

Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/e205f13d-18ff-a49c-0988-7de6ea5ff823@i-love.sakura.ne.jp
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agoASoC: dt-bindings: rockchip: i2s-tdm: Drop rockchip,cru property
Nicolas Frattaroli [Sat, 16 Oct 2021 10:53:51 +0000 (12:53 +0200)]
ASoC: dt-bindings: rockchip: i2s-tdm: Drop rockchip,cru property

This property was only needed for a driver hack, which we can
remove. Since the bindings were not in any kernel release yet, we
are able to just drop the property instead of silently accepting
and ignoring it.

Signed-off-by: Nicolas Frattaroli <frattaroli.nicolas@gmail.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Message-Id: <20211016105354.116513-3-frattaroli.nicolas@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoASoC: rockchip: i2s-tdm: Strip out direct CRU use
Nicolas Frattaroli [Sat, 16 Oct 2021 10:53:50 +0000 (12:53 +0200)]
ASoC: rockchip: i2s-tdm: Strip out direct CRU use

In cases where both rx and tx lrck are synced to the same source,
the resets for rx and tx need to be triggered simultaneously,
according to the downstream driver.

As there is no reset API to atomically bulk (de)assert two resets
at once, what the driver did was implement half a reset controller
specific to Rockchip, which tried to write the registers for the
resets within one write ideally or several writes within an irqsave
section.

This of course violates abstractions quite badly. The driver should
not write to the CRU's registers directly.

In practice, for the cases I tested the driver with, which is audio
playback, replacing the synchronised asserts with just individual
ones does not seem to make any difference.

If it turns out that this breaks something in the future, it should
be fixed through the specification and implementation of an atomic
bulk reset API, not with a CRU hack.

Signed-off-by: Nicolas Frattaroli <frattaroli.nicolas@gmail.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Message-Id: <20211016105354.116513-2-frattaroli.nicolas@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
3 years agoMerge tag 'perf-tools-fixes-for-v5.15-2021-10-16' of git://git.kernel.org/pub/scm...
Linus Torvalds [Sat, 16 Oct 2021 18:11:07 +0000 (11:11 -0700)]
Merge tag 'perf-tools-fixes-for-v5.15-2021-10-16' of git://git./linux/kernel/git/acme/linux

Pull perf tools fixes from Arnaldo Carvalho de Melo:

 - Fix 'perf test evsel' build error on !x86 architectures

 - Fix libperf's test_stat_cpu mixup of CPU numbers and CPU indexes

 - Output offsets for decompressed records, not just useless zeros

* tag 'perf-tools-fixes-for-v5.15-2021-10-16' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  libperf tests: Fix test_stat_cpu
  libperf test evsel: Fix build error on !x86 architectures
  perf report: Output non-zero offset for decompressed records

3 years agoMerge tag 'fixes-2021-10-16' of git://git.kernel.org/pub/scm/linux/kernel/git/rppt...
Linus Torvalds [Sat, 16 Oct 2021 17:57:13 +0000 (10:57 -0700)]
Merge tag 'fixes-2021-10-16' of git://git./linux/kernel/git/rppt/memblock

Pull memblock fix from Mike Rapoport:
 "Fix handling of NOMAP regions with kmemleak.

  NOMAP regions don't have linear map entries so an attempt to scan
  these areas in kmemleak would fault.

  Prevent such faults by excluding NOMAP regions from kmemleak"

* tag 'fixes-2021-10-16' of git://git.kernel.org/pub/scm/linux/kernel/git/rppt/memblock:
  memblock: exclude NOMAP regions from kmemleak

3 years agoMerge tag 'trace-v5.15-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt...
Linus Torvalds [Sat, 16 Oct 2021 17:51:41 +0000 (10:51 -0700)]
Merge tag 'trace-v5.15-rc3' of git://git./linux/kernel/git/rostedt/linux-trace

Tracing fixes for 5.15:

 - Fix defined but not use warning/error for osnoise function

 - Fix memory leak in event probe

 - Fix memblock leak in bootconfig

 - Fix the API of event probes to be like kprobes

 - Added test to check removal of event probe API

 - Fix recordmcount.pl for nds32 failed build

* tag 'trace-v5.15-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  nds32/ftrace: Fix Error: invalid operands (*UND* and *UND* sections) for `^'
  selftests/ftrace: Update test for more eprobe removal process
  tracing: Fix event probe removal from dynamic events
  tracing: Fix missing * in comment block
  bootconfig: init: Fix memblock leak in xbc_make_cmdline()
  tracing: Fix memory leak in eprobe_register()
  tracing: Fix missing osnoise tracer on max_latency

3 years agoMerge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 16 Oct 2021 17:22:08 +0000 (10:22 -0700)]
Merge tag 'clk-fixes-for-linus' of git://git./linux/kernel/git/clk/linux

Pull clk driver fixes from Stephen Boyd:
 "Clk driver fixes for critical issues found in the past few weeks:

   - Select gdsc config so qcom sm6350 driver probes

   - Fix a register offset in qcom gcc-sm6115 so the correct clk is
     controlled

   - Fix inverted logic in Renesas RZ/G2L .is_enabled()

   - Mark some more clks critical in Renesas clk driver

   - Remove a duplicate clk in the agilex driver"

* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
  clk: qcom: add select QCOM_GDSC for SM6350
  clk: qcom: gcc-sm6115: Fix offset for hlos1_vote_turing_mmu_tbu0_gdsc
  clk: socfpga: agilex: fix duplicate s2f_user0_clk
  clk: renesas: rzg2l: Fix clk status function
  clk: renesas: r9a07g044: Mark IA55_CLK and DMAC_ACLK critical

3 years agoMerge tag 'for-5.15/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device...
Linus Torvalds [Sat, 16 Oct 2021 17:12:21 +0000 (10:12 -0700)]
Merge tag 'for-5.15/dm-fixes' of git://git./linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mike Snitzer:

 - Fix DM verity target to skip redundant processing on I/O errors.

 - Fix request-based DM so that it doesn't queue request to blk-mq when
   DM device is suspended.

 - Fix DM core mempool NULL pointer race when completing IO.

 - Make DM clone target's 'descs' array static.

* tag 'for-5.15/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm: fix mempool NULL pointer race when completing IO
  dm rq: don't queue request to blk-mq during DM suspend
  dm clone: make array 'descs' static
  dm verity: skip redundant verity_handle_err() on I/O errors

3 years agoMerge tag 's390-5.15-6' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Linus Torvalds [Sat, 16 Oct 2021 16:14:55 +0000 (09:14 -0700)]
Merge tag 's390-5.15-6' of git://git./linux/kernel/git/s390/linux

Pull s390 fixes from Vasily Gorbik:

 - Maintainers and reviewers changes:

    * Cornelia decided to free up her time and step down from vfio-ccw
      maintainer and s390 kvm reviewer duties

    * Add Alexander Gordeev as s390 arch code reviewer

 - Fix broken strrchr implementation

* tag 's390-5.15-6' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390: add Alexander Gordeev as reviewer
  s390: fix strrchr() implementation
  vfio-ccw: step down as maintainer
  KVM: s390: remove myself as reviewer

3 years agoMerge tag 'csky-for-linus-5.15-rc6' of git://github.com/c-sky/csky-linux
Linus Torvalds [Sat, 16 Oct 2021 16:12:18 +0000 (09:12 -0700)]
Merge tag 'csky-for-linus-5.15-rc6' of git://github.com/c-sky/csky-linux

Pull csky fixes from Guo Ren:
 "Only 5 fixups:

   - Make HAVE_TCM depend on !COMPILE_TEST

   - bitops: Remove duplicate __clear_bit define

   - Select ARCH_WANT_FRAME_POINTERS only if compiler supports it

   - Fixup regs.sr broken in ptrace

   - don't let sigreturn play with priveleged bits of status register"

* tag 'csky-for-linus-5.15-rc6' of git://github.com/c-sky/csky-linux:
  csky: Make HAVE_TCM depend on !COMPILE_TEST
  csky: bitops: Remove duplicate __clear_bit define
  csky: Select ARCH_WANT_FRAME_POINTERS only if compiler supports it
  csky: Fixup regs.sr broken in ptrace
  csky: don't let sigreturn play with priveleged bits of status register

3 years agoMerge tag 'arc-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc
Linus Torvalds [Sat, 16 Oct 2021 16:09:27 +0000 (09:09 -0700)]
Merge tag 'arc-5.15-rc6' of git://git./linux/kernel/git/vgupta/arc

Pull ARC fix from Vineet Gupta:
 "Small fixlet for ARC"

* tag 'arc-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
  ARC: fix potential build snafu

3 years agoMerge tag 'arm-soc-fixes-5.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 16 Oct 2021 16:05:58 +0000 (09:05 -0700)]
Merge tag 'arm-soc-fixes-5.15-2' of git://git./linux/kernel/git/soc/soc

Pull ARM SoC fixes from Arnd Bergmann:
 "A small number fixes this time, mostly touching actual code:

   - Add platform device for i.MX System Reset Controller (SRC) to
     fix a regression caused by fw_devlink change

   - A fixup for a boot regression caused by my own rework for the
     Qualcomm SCM driver

   - Multiple bugfixes for the Arm FFA and optee firmware drivers,
     addressing problems when they are built as a loadable module

   - Four dts bugfixes for the Broadcom SoC used in Raspberry pi,
     addressing VEC (video encoder), MDIO bus controller
     #address-cells/#size-cells, SDIO voltage and PCIe host bridge
     dtc warnings"

* tag 'arm-soc-fixes-5.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
  ARM: imx: register reset controller from a platform driver
  iommu/arm: fix ARM_SMMU_QCOM compilation
  ARM: dts: bcm2711-rpi-4-b: Fix usb's unit address
  ARM: dts: bcm2711-rpi-4-b: Fix pcie0's unit address formatting
  tee: optee: Fix missing devices unregister during optee_remove
  ARM: dts: bcm2711-rpi-4-b: fix sd_io_1v8_reg regulator states
  ARM: dts: bcm2711: fix MDIO #address- and #size-cells
  ARM: dts: bcm283x: Fix VEC address for BCM2711
  firmware: arm_ffa: Fix __ffa_devices_unregister
  firmware: arm_ffa: Add missing remove callback to ffa_bus_type

3 years agoMerge tag 'pci-v5.15-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaa...
Linus Torvalds [Sat, 16 Oct 2021 16:00:46 +0000 (09:00 -0700)]
Merge tag 'pci-v5.15-fixes-2' of git://git./linux/kernel/git/helgaas/pci

Pull pci fix from Bjorn Helgaas:

 - Don't save msi_populate_sysfs() error code as dev->msi_irq_groups so
   we don't dereference the error code as a pointer (Wang Hai)

* tag 'pci-v5.15-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
  PCI/MSI: Handle msi_populate_sysfs() errors correctly

3 years agoMerge tag 'acpi-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Sat, 16 Oct 2021 15:45:46 +0000 (08:45 -0700)]
Merge tag 'acpi-5.15-rc6' of git://git./linux/kernel/git/rafael/linux-pm

Pull ACPI fix from Rafael Wysocki:
 "Add a missing device ID to a quirk list in the suspend-to-idle support
  code"

* tag 'acpi-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: PM: Include alternate AMDI0005 id in special behaviour

3 years agox86/fpu: Mask out the invalid MXCSR bits properly
Borislav Petkov [Fri, 15 Oct 2021 10:46:25 +0000 (12:46 +0200)]
x86/fpu: Mask out the invalid MXCSR bits properly

This is a fix for the fix (yeah, /facepalm).

The correct mask to use is not the negation of the MXCSR_MASK but the
actual mask which contains the supported bits in the MXCSR register.

Reported and debugged by Ville Syrjälä <ville.syrjala@linux.intel.com>

Fixes: d298b03506d3 ("x86/fpu: Restore the masking out of reserved MXCSR bits")
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Tested-by: Ser Olmy <ser.olmy@protonmail.com>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/YWgYIYXLriayyezv@intel.com
3 years agoInput: touchscreen - avoid bitwise vs logical OR warning
Nathan Chancellor [Fri, 15 Oct 2021 20:13:06 +0000 (13:13 -0700)]
Input: touchscreen - avoid bitwise vs logical OR warning

A new warning in clang points out a few places in this driver where a
bitwise OR is being used with boolean types:

drivers/input/touchscreen.c:81:17: warning: use of bitwise '|' with boolean operands [-Wbitwise-instead-of-logical]
        data_present = touchscreen_get_prop_u32(dev, "touchscreen-min-x",
                       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

This use of a bitwise OR is intentional, as bitwise operations do not
short circuit, which allows all the calls to touchscreen_get_prop_u32()
to happen so that the last parameter is initialized while coalescing the
results of the calls to make a decision after they are all evaluated.

To make this clearer to the compiler, use the '|=' operator to assign
the result of each touchscreen_get_prop_u32() call to data_present,
which keeps the meaning of the code the same but makes it obvious that
every one of these calls is expected to happen.

Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reported-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Link: https://lore.kernel.org/r/20211014205757.3474635-1-nathan@kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
3 years agoInput: xpad - add support for another USB ID of Nacon GC-100
Michael Cullen [Fri, 15 Oct 2021 20:17:50 +0000 (13:17 -0700)]
Input: xpad - add support for another USB ID of Nacon GC-100

The Nacon GX100XF is already mapped, but it seems there is a Nacon
GC-100 (identified as NC5136Wht PCGC-100WHITE though I believe other
colours exist) with a different USB ID when in XInput mode.

Signed-off-by: Michael Cullen <michael@michaelcullen.name>
Link: https://lore.kernel.org/r/20211015192051.5196-1-michael@michaelcullen.name
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
3 years agoInput: resistive-adc-touch - fix division by zero error on z1 == 0
Oleksij Rempel [Sat, 16 Oct 2021 04:12:06 +0000 (21:12 -0700)]
Input: resistive-adc-touch - fix division by zero error on z1 == 0

For proper pressure calculation we need at least x and z1 to be non
zero. Even worse, in case z1 we may run in to division by zero
error.

Fixes: 60b7db914ddd ("Input: resistive-adc-touch - rework mapping of channels")
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://lore.kernel.org/r/20211007095727.29579-1-o.rempel@pengutronix.de
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
3 years agoInput: snvs_pwrkey - add clk handling
Uwe Kleine-König [Sat, 16 Oct 2021 04:19:33 +0000 (21:19 -0700)]
Input: snvs_pwrkey - add clk handling

On i.MX7S and i.MX8M* (but not i.MX6*) the pwrkey device has an
associated clock. Accessing the registers requires that this clock is
enabled. Binding the driver on at least i.MX7S and i.MX8MP while not
having the clock enabled results in a complete hang of the machine.
(This usually only happens if snvs_pwrkey is built as a module and the
rtc-snvs driver isn't already bound because at bootup the required clk
is on and only gets disabled when the clk framework disables unused clks
late during boot.)

This completes the fix in commit 135be16d3505 ("ARM: dts: imx7s: add
snvs clock to pwrkey").

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20211013062848.2667192-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
3 years agokyber: avoid q->disk dereferences in trace points
Christoph Hellwig [Tue, 12 Oct 2021 09:33:01 +0000 (11:33 +0200)]
kyber: avoid q->disk dereferences in trace points

q->disk becomes invalid after the gendisk is removed.  Work around this
by caching the dev_t for the tracepoints.  The real fix would be to
properly tear down the I/O schedulers with the gendisk, but that is
a much more invasive change.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211012093301.GA27795@lst.de
Tested-by: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agoblock: keep q_usage_counter in atomic mode after del_gendisk
Christoph Hellwig [Wed, 29 Sep 2021 07:12:41 +0000 (09:12 +0200)]
block: keep q_usage_counter in atomic mode after del_gendisk

Don't switch back to percpu mode to avoid the double RCU grace period
when tearing down SCSI devices.  After removing the disk only passthrough
commands can be send anyway.

Suggested-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Darrick J. Wong <djwong@kernel.org>
Link: https://lore.kernel.org/r/20210929071241.934472-6-hch@lst.de
Tested-by: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agoblock: drain file system I/O on del_gendisk
Christoph Hellwig [Wed, 29 Sep 2021 07:12:40 +0000 (09:12 +0200)]
block: drain file system I/O on del_gendisk

Instead of delaying draining of file system I/O related items like the
blk-qos queues, the integrity read workqueue and timeouts only when the
request_queue is removed, do that when del_gendisk is called.  This is
important for SCSI where the upper level drivers that control the gendisk
are separate entities, and the disk can be freed much earlier than the
request_queue, or can even be unbound without tearing down the queue.

Fixes: edb0872f44ec ("block: move the bdi from the request_queue to the gendisk")
Reported-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Darrick J. Wong <djwong@kernel.org>
Link: https://lore.kernel.org/r/20210929071241.934472-5-hch@lst.de
Tested-by: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agoblock: split bio_queue_enter from blk_queue_enter
Christoph Hellwig [Wed, 29 Sep 2021 07:12:39 +0000 (09:12 +0200)]
block: split bio_queue_enter from blk_queue_enter

To prepare for fixing a gendisk shutdown race, open code the
blk_queue_enter logic in bio_queue_enter.  This also removes the
pointless flags translation.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Darrick J. Wong <djwong@kernel.org>
Link: https://lore.kernel.org/r/20210929071241.934472-4-hch@lst.de
Tested-by: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>