Don't populate the arrays mul and div on the stack but instead make them
static const. Makes the object code smaller by 4 bytes.
Before:
text data bss dec hex filename
16226 4984 64 21274 531a ./sound/soc/uniphier/aio-cpu.o
After:
text data bss dec hex filename
16062 5144 64 21270 5316 ./sound/soc/uniphier/aio-cpu.o
(gcc version 10.2.0)
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Link: https://lore.kernel.org/r/20210818151746.38520-1-colin.king@canonical.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Use "cfg-lineout:2" in the components string on boards with a lineout
instead of "cfg-lineout:1", this better mirrors the speaker part of
the components string where we use "cfg-spk:1" for devices with a single
speaker and "cfg-spk:2" for stereo speakers.
The lineout is stereo by default, so using ":2" makes more sense, this
way we keep ":1" reserved in case we ever encounter a device with
a mono lineout.
We can make this change without breaking userspace because no kernel
has shipped with "cfg-lineout:1" in the component-string yet; and there
also are no userspace bits (UCM profiles) checking for this yet.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210816114722.107363-1-hdegoede@redhat.com
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch cleanups below cppcheck warning.
sound/soc/soc-ac97.c:41:15: style: struct member 'snd_ac97_gpio_priv::gpios_set' is never used. [unusedStructMember]
unsigned int gpios_set;
^
sound/soc/soc-ac97.c:42:28: style: struct member 'snd_ac97_gpio_priv::component' is never used. [unusedStructMember]
struct snd_soc_component *component;
^
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87r1euyolk.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch cleanups below cppcheck warning.
sound/soc/soc-jack.c:21:6: style: struct member 'jack_gpio_tbl::count' is never used. [unusedStructMember]
int count;
^
sound/soc/soc-jack.c:23:28: style: struct member 'jack_gpio_tbl::gpios' is never used. [unusedStructMember]
struct snd_soc_jack_gpio *gpios;
^
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87tujqyols.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch cleanups below cppcheck warning.
sound/soc/soc-dai.c:454:7: style: The scope of the variable 'supported_cpu' can be reduced. [variableScope]
bool supported_cpu;
^
sound/soc/soc-dai.c:455:7: style: The scope of the variable 'supported_codec' can be reduced. [variableScope]
bool supported_codec;
^
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87y292yom6.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
This is an update on an earlier contribution from Heikki Krogerus
The function device_add_properties() is going to be removed.
Replacing it with software node API equivalents.
Thanks for Hans de Goede and Andy Shevchenko for their comments,
suggestions and Reviewed-by tags on GitHub. The review thread can be
found at https://github.com/thesofproject/linux/pull/3041)
v3 changes:
Fixed nit-picks from Andy: label, return value, missing commas/periods.
Added Andy's Reviewed-by tag
v2 changes: feedback from Andy and Hans
Better error handling
Codec reference is kept until the .remove callback
Remove bus search to find device
v1 changes from Heikki's patches:
Avoid the use of devm_ routines for Baytrail machine drivers.
Heikki Krogerus (1):
ASoC: Intel: boards: use software node API in Atom boards
Pierre-Louis Bossart (7):
ASoC: Intel: boards: harden codec property handling
ASoC: Intel: boards: handle errors with acpi_dev_get_first_match_dev()
ASoC: Intel: boards: get codec device with ACPI instead of bus search
ASoC: Intel: sof_sdw: pass card information to init/exit functions
ASoC: Intel: sof_sdw_rt711*: keep codec device reference until remove
ASoC: Intel: use software node API in SoundWire machines
ASoC: Intel: remove device_properties for Atom boards
sound/soc/intel/boards/bytcht_es8316.c | 31 ++++++++--
sound/soc/intel/boards/bytcr_rt5640.c | 60 +++++++++++++++-----
sound/soc/intel/boards/bytcr_rt5651.c | 63 ++++++++++++++-------
sound/soc/intel/boards/sof_sdw.c | 20 ++++---
sound/soc/intel/boards/sof_sdw_common.h | 37 +++++++-----
sound/soc/intel/boards/sof_sdw_max98373.c | 3 +-
sound/soc/intel/boards/sof_sdw_rt1308.c | 3 +-
sound/soc/intel/boards/sof_sdw_rt1316.c | 3 +-
sound/soc/intel/boards/sof_sdw_rt5682.c | 3 +-
sound/soc/intel/boards/sof_sdw_rt700.c | 3 +-
sound/soc/intel/boards/sof_sdw_rt711.c | 51 +++++++++--------
sound/soc/intel/boards/sof_sdw_rt711_sdca.c | 52 +++++++++--------
sound/soc/intel/boards/sof_sdw_rt715.c | 3 +-
sound/soc/intel/boards/sof_sdw_rt715_sdca.c | 3 +-
14 files changed, 223 insertions(+), 112 deletions(-)
--
2.25.1
This patch series aims to add ASoC support on RZ/G2L SoC's.
It is based on the work done by Chris Brandt for RZ/A ASoC driver.
v4->v5
* Moved validation of sample bits in hw_params
* Removed validation of frame bits as it is redundant
* split the rz_ssi_start_stop function into rz_ssi_start and rz_ssi_stop.
* remove the spin_lock around rz_ssi_stream_init.
* Updated dmas description and removed fixes as it is an enhancement
now.
* updated ssi_start functions with setting fifo thresholds
and ssi_stop function with cancel all dma txn.
v3->v4:
* Updated the subject line as per style for the subsystem.
* Removed select SND_SIMPLE_CARD from Kconfig
* Added C++ comments for copyright and driver description.
* Moved validation of channels in hw_params
* removed asm issue reported by bot as well as Mark
* replaced master/slave macros with provider/consumer macros
* Improved locking and added more null pointer checks.
v2->v3:
* Fixed the dependency on KCONFIG
* Merged the binding patch with dma feature added
* Updated dt binding example with encoded #dma-cells value.
* Improved Error handling in probe function
* Removed the passing legacy channel configuration parameters from
dmaengine_slave_config function
* started using dma_request_chan instead of deprecated
dma_request_slave_channel
* Removed SoC dtsi and config patches from this series. Will send it later.
v1->v2:
* Rebased to latest rc kernel
Biju Das (3):
ASoC: sh: Add RZ/G2L SSIF-2 driver
ASoC: dt-bindings: renesas,rz-ssi: Update slave dma channel
configuration parameters
ASoC: sh: rz-ssi: Add SSI DMAC support
.../bindings/sound/renesas,rz-ssi.yaml | 22 +-
sound/soc/sh/Kconfig | 6 +
sound/soc/sh/Makefile | 4 +
sound/soc/sh/rz-ssi.c | 1063 +++++++++++++++++
4 files changed, 1093 insertions(+), 2 deletions(-)
create mode 100644 sound/soc/sh/rz-ssi.c
--
2.17.1
This patchset provides an optimization that result in significant power
savings on Intel HDAudio platforms using SOF (Sound Open Firmware).
We previously prevented the Intel DSP from enabling the DMI_L1
capability to work-around issues with pause on capture streams. It
turns out that this also prevented the platform from entering high C
states in full-duplex usages such as videoconferencing - a rather
basic use case since the start of the pandemic.
The support for pause_push/release was already a bit controversial for
Intel platforms, in theory platforms should only enable PAUSE if they
can resume on the same sample, which is not the case on any Intel
platform.
With this patchset, when the user enables DMI L1 via a kernel
parameter, the PAUSE support is disabled for capture streams. A kernel
parameter is far from ideal but it's a placeholder until we have an
API to negotiate capabilities between applications and driver, and
it's far less confusing than a Kconfig option.
Changes since v1:
Removal of SPIB support since it may conflict with Takashi's memalloc
changes. These SPIB changes will be provided after rebase.
Addition of one cleanup for cppcheck warning
Move all changes to intel/ directory, no changes in shared code
Flipped the logic: the selection of DMI L1 disables PAUSE
Pierre-Louis Bossart (4):
ASoC: SOF: Intel: Kconfig: clarify DMI L1 option description
ASoC: SOF: Intel: hda-stream: remove always true condition
ASoC: SOF: Intel: simplify logic for DMI_L1 handling
ASoC: SOF: Intel: make DMI L1 selection more robust
sound/soc/sof/intel/Kconfig | 10 ----------
sound/soc/sof/intel/hda-pcm.c | 16 ++++++++++++++--
sound/soc/sof/intel/hda-stream.c | 11 +++++------
3 files changed, 19 insertions(+), 18 deletions(-)
--
2.25.1
In current ACPI-based devices, the DSDT does not include any of the
properties required by the codec driver. This is not an ACPI
limitation proper since the _DSD method could be used, as done for
Camera and SoundWire in newer platforms. For legacy devices, there is
unfortunately no other option than using a work-around: we add
properties to the codec device from the machine driver.
To avoid any issues with the codec driver being unbound, we need to
keep a reference to the codec device until the card is removed.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Co-developed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210813151116.23931-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Exposing the DMI L1 configuration as a kernel Kconfig option was in
hindsight a really bad idea. It led to several errors reported by
distributions which selected it by mistake.
The Kconfig is now replaced with a kernel parameter. Since DMI L1
entry is incompatible with pause on a capture stream, the latter is
disabled when the kernel parameter is set.
Experimental results show an increased residency in higher C states
and a significant decrease of system power consumption for "work from
home" usages such as VoIP calls.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@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/20210812231940.172547-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
When I booted up on a board that had a slightly different codec
stuffed on it, I got this message at bootup:
rt5682 9-001a: Device with ID register 6749 is not rt5682
That's normal/expected, but what wasn't normal was the splat that I
got after:
WARNING: CPU: 7 PID: 176 at drivers/regulator/core.c:2151 _regulator_put+0x150/0x158
pc : _regulator_put+0x150/0x158
...
Call trace:
_regulator_put+0x150/0x158
regulator_bulk_free+0x48/0x70
devm_regulator_bulk_release+0x20/0x2c
release_nodes+0x1cc/0x244
devres_release_all+0x44/0x60
really_probe+0x17c/0x378
...
This is because the error paths don't turn off the regulator. Let's
fix that.
Fixes: 0ddce71c21 ("ASoC: rt5682: add rt5682 codec driver")
Fixes: 87b42abae9 ("ASoC: rt5682: Implement remove callback")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Link: https://lore.kernel.org/r/20210811081751.v2.1.I4a1d9aa5d99e05aeee15c2768db600158d76cab8@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
