Use the newer RUNTIME_PM_OPS() and SYSTEM_SLEEP_PM_OPS() macros
instead of SET_RUNTIME_PM_OPS() and SET_SYSTEM_SLEEP_PM_OPS() together
with pm_ptr()
This optimizes slightly when CONFIG_PM is disabled, too.
Cc: Shengjiu Wang <shengjiu.wang@gmail.com>
Cc: Xiubo Li <Xiubo.Lee@gmail.com>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Nicolin Chen <nicoleotsuka@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://patch.msgid.link/20250317095603.20073-60-tiwai@suse.de
Signed-off-by: Mark Brown <broonie@kernel.org>
When decimation filter bypass mode is enabled, PDM data can be
written into FIFO directly without any processing.
The interface of this mode is DSD big endian format, when
user needs this format, then this mode is enabled.
This mode is only for the i.MX943 platform currently.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://patch.msgid.link/20250206030306.2618620-1-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
If 'micfil->quality' received from micfil_quality_set() somehow ends
up with an unpredictable value, switch() operator will fail to
initialize local variable qsel before regmap_update_bits() tries
to utilize it.
While it is unlikely, play it safe and enable a default case that
returns -EINVAL error.
Found by Linux Verification Center (linuxtesting.org) with static
analysis tool SVACE.
Fixes: bea1d61d58 ("ASoC: fsl_micfil: rework quality setting")
Cc: stable@vger.kernel.org
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Link: https://patch.msgid.link/20250116142436.22389-1-n.zhandarovich@fintech.ru
Signed-off-by: Mark Brown <broonie@kernel.org>
On i.MX943, the FIFO data address is changed to 0x20~0x3c,
compared to previous version, there is a minus 4 offset,
so add a new regmap configuration for it.
And the bit width of CICOSR is changed to 5 bits, from bit
16th to 20th in REG_MICFIL_CTRL2 register, so update its
definition header file.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://patch.msgid.link/20250114102720.3664667-2-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
fsl_asoc_constrain_rates() is a common function to constrain rates.
Let's switch to this function.
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Link: https://patch.msgid.link/20241126115440.3929061-3-chancel.liu@nxp.com
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
On some platforms, for example i.MX93, there is only one
audio PLL source, so some sample rate can't be supported.
If the PLL source is used for 8kHz series rates, then 11kHz
series rates can't be supported.
So add constraints according to the frequency of available
clock sources, then alsa-lib will help to convert the
unsupported rate for the driver.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://patch.msgid.link/1728884313-6778-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
On i.MX8MM the register of volume control has positive and negative
values. It is different from other platforms like i.MX8MP and i.MX93
which only have positive values. Add a volume_sx flag to use SX_TLV
volume control for this kind of platform. Use common TLV volume control
for other platforms.
Fixes: cdfa92eb90 ("ASoC: fsl_micfil: Correct the number of steps on SX controls")
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Link: https://patch.msgid.link/20241017071507.2577786-1-chancel.liu@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Merge series from Shengjiu Wang <shengjiu.wang@nxp.com>:
Fix the usage of regmap_write_bits().
Move mclk clock enablement to late stage.
Enable the micfil error interrupt.
Several of the NXP drivers use regmaps with a rbtree register cache. Since
the maple tree cache is using a generally more modern data structure which
makes implementation choices more suitable for modern systems let's convert
these drivers to it. This should have no practical impact.
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://patch.msgid.link/20240924-asoc-imx-maple-v1-1-8b993901f71e@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Previously the mclk is enabled in probe() stage, which
is not necessary. Move mclk enablement to hw_params()
and mclk disablement to hw_free() will be more efficient.
'mclk_flag' is used for this case.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://patch.msgid.link/1727424031-19551-3-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The last parameter 1 means BIT(0), which should be the
correct BIT(X).
Fixes: 47a70e6fc9 ("ASoC: Add MICFIL SoC Digital Audio Interface driver.")
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Link: https://patch.msgid.link/1727424031-19551-2-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
After commit 0edb555a65 ("platform: Make platform_driver::remove()
return void") .remove() is (again) the right callback to implement for
platform drivers.
Convert all drivers below sound/soc to use .remove(), with the eventual
goal to drop struct platform_driver::remove_new(). As .remove() and
.remove_new() have the same prototypes, conversion is done by just
changing the structure member name in the driver initializer.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Link: https://patch.msgid.link/20240909151230.909818-2-u.kleine-koenig@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>
On i.MX9x platforms, the REG_MICFIL_FSYNC_CTRL, REG_MICFIL_VERID,
REG_MICFIL_PARAM are added, but they are not existed on i.MX8x
platforms.
Use the existed micfil->soc->use_verid to distinguish the access
permission for these platforms.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://patch.msgid.link/1721897694-6088-3-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
On the i.MX9x platforms, the mask of FIFO watermark
is 0x1F, on i.MX8x platforms, the mask of FIFO watermark
is 0X7. So use the mask 0x1F for all platforms to make them
compatible.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://patch.msgid.link/1721897694-6088-2-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
We need this in order to easily reuse register definitions
and some functions with Sound Open Firmware driver.
According to Documentation/process/license-rules.rst:
"Dual BSD/GPL" The module is dual licensed under a GPL v2
variant or BSD license choice. The exact
variant of the BSD license can only be
determined via the license information
in the corresponding source files.
so use "Dual BSD/GPL" for license string.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Link: https://lore.kernel.org/r/20230803072638.640789-1-daniel.baluta@oss.nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Use SET_SYSTEM_SLEEP_PM_OPS to simplify suspend and resume function.
fsl_micfil_suspend() and fsl_micfil_resume() can be deleted.
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20230802052117.1293029-4-chancel.liu@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
fsl_micfil_use_verid() can help to parse the version info in VERID and
PARAM registers. Since the two registers are added only on i.MX93
platform, a member flag called use_verid is introduced to soc data
structure which indicates acquiring MICFIL version.
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20230802052117.1293029-3-chancel.liu@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
MICFIL IP is upgraded on i.MX93 platform. These new registers and new
bit definition are added to complete the register list.
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20230802052117.1293029-2-chancel.liu@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
`strncpy` is deprecated for use on NUL-terminated destination strings [1].
A suitable replacement is `strscpy` [2] due to the fact that it
guarantees NUL-termination on its destination buffer argument which is
_not_ always the case for `strncpy`!
In this case, though, there was great care taken to ensure that the
destination buffer would be NUL-terminated through the use of `len - 1`
ensuring that the previously zero-initialized buffer would not overwrite
the last NUL byte. This means that there's no bug here.
However, `strscpy` will add a mandatory NUL byte to the destination
buffer as promised by the following `strscpy` implementation [3]:
| /* Hit buffer length without finding a NUL; force NUL-termination. */
| if (res)
| dest[res-1] = '\0';
This means we can lose the `- 1` which clears up whats happening here.
All the while, we get one step closer to eliminating the ambiguous
`strncpy` api in favor of its less ambiguous replacement like `strscpy`,
`strscpy_pad`, `strtomem` and `strtomem_pad` amongst others.
[1]: www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings
[2]: manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html
[3]: https://elixir.bootlin.com/linux/v6.3/source/lib/string.c#L183
Link: https://github.com/KSPP/linux/issues/90
Signed-off-by: Justin Stitt <justinstitt@google.com>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20230727-sound-soc-fsl-v1-1-4fc0ed7e0366@google.com
Signed-off-by: Mark Brown <broonie@kernel.org>
There is error message when defer probe happens:
fsl-micfil-dai 30ca0000.micfil: Unbalanced pm_runtime_enable!
Fix the error handler with pm_runtime_enable and add
fsl_micfil_remove() for pm_runtime_disable.
Fixes: 47a70e6fc9 ("ASoC: Add MICFIL SoC Digital Audio Interface driver.")
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com
Link: https://lore.kernel.org/r/1683540996-6136-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org
The parameter "max" of SOC_SINGLE_SX_TLV() means the number of steps
rather than maximum value. This patch corrects the minimum value to -8
and the number of steps to 15.
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20230104025754.3019235-1-chancel.liu@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
On i.MX93 platform MICFIL uses eDMA. The maxburst should be set to the
number of channels in eDMA multiple FIFO mode.
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20221028082750.991822-4-chancel.liu@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add compatible string and specific soc data to support MICFIL on i.MX93
platform.
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20221028082750.991822-3-chancel.liu@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The Hardware Voice Activity Detector (HWVAD) is a block
responsible for detect voice activity in a channel selected
by the user. It can be configured in Envelope-based or
Energy-based mode.
There are additional two interrupts for HWVAD, one is event
interrupt, another is error interrupt.
Enable hwvad in parallel with recording.
when voice activity detected, HWVAD will be disabled and
alsa control notification is triggerred.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1663925494-9941-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
i.MX8MQ/MN/MM/MP platforms typically have 2 AUDIO PLLs being
configured to handle 8kHz and 11kHz series audio rates.
The patch implements the functionality to select at runtime
the appropriate AUDIO PLL as function of audio file rate.
Signed-off-by: Viorel Suman <viorel.suman@nxp.com>
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1656667961-1799-4-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Need to add legacy_dai_naming flag otherwise there
will be issue when registerring component, that cause
the probe failure.
Fixes: 1e63fcc74a ("ASoC: fsl: Migrate to new style legacy DAI naming flag")
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1656899417-4775-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Previous default settings resulted in loose dynamic
range and low sound level.
New default configuration changes:
- outgain = 2
- quality mode = VLOW0
- dc remover = bypass
Signed-off-by: Irina Patru <ioana-irina.patru@nxp.com>
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1656405589-29850-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
There may be failure when start 1 channel recording after
8 channels recording. The reason is that the CHnF
flags are not cleared successfully by software reset.
This issue is triggerred by the change of clearing
software reset bit.
CHnF flags are write 1 clear bits. Clear them by force
write.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1651925654-32060-2-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
SRES is self-cleared bit, but REG_MICFIL_CTRL1 is defined as
non volatile register, it still remain in regmap cache after set,
then every update of REG_MICFIL_CTRL1, software reset happens.
to avoid this, clear it explicitly.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1651925654-32060-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
On i.MX8Plus there are two updates for micfil module.
One is that the output format is S32_LE, only the 24 more
significative bits have information, the other bits are always
zero. Add 'formats' variable in soc data to distinguish the
format on different platform.
Another is that the fifo depth is 32 entries.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1652087663-1908-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The micfil driver prints out the IRQ numbers for each interrupt at error
level. This information is useful for debugging at best, remove it.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-22-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
fsl_set_clock_params() is used only once and easily be folded into its
caller, do so.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-21-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
The "fsl,shared-interrupt" property is undocumented and unnecessary.
Just pass IRQF_SHARED unconditionally.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-20-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
FSL_MICFIL_RATES and FSL_MICFIL_FORMATS is only used once. Drop
the unnecesary indirection and use SNDRV_PCM_RATE_8000_48000 and
SNDRV_PCM_FMTBIT_S16_LE directly.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-19-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
The micfil driver doesn't use anything from imx-pcm.h. Drop its
inclusion.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-18-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
For the quality setting the quality setting register values are directly
exposed to the kcontrol and thus to userspace. This is unfortunate
because the register settings contains invalid bit combinations marked
as "N/A". For userspace it doesn't make much sense to be able to set
these just to see that the driver responds with "Please make sure you
select a valid quality." in the kernel log.
Work around this by adding get/set functions for the quality setting.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-17-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
The reference manual has this for calculating the micfil internal clock
divider:
MICFIL Clock rate
clkdiv = -----------------
8 * OSR * outrate
(with OSR == Oversampling Rate, outrate == output sample rate)
The driver first sets the MICFIL Clock rate to (outrate * 1024) and then
calculates back the clkdiv value from the above calculation.
Simplify this by using a fixed clkdiv value of 8 and set the MICFIL
Clock rate to (outrate * clkdiv * OSR * 8).
While at it drop disabling the clock before setting its rate. The MICFIL
module is disabled when the rate is changed and it is also resetted
before it is started again, so I doubt it's necessary to disable the
clock.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-16-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
get_pdm_clk() calculates the PDM clock based on the quality setting,
but really the PDM clock is independent of the quality, it's always
rate * 4 * micfil->osr. Just drop the function and do the calculation
in the caller.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-15-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
The OSR (OverSampling Rate) setting is set once to the default value
and never changed throughout the driver. Nevertheless the value is
read back from the register for further calculations. Just use the
default value because we know what we have written.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-14-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
The micfil hardware provides the microphone data on multiple successive
FIFO registers, one register per stereo pair. Also to work properly the
SDMA_DONE0_CONFIG_DONE_SEL bit in the SDMA engines SDMA_DONE0_CONFIG
register must be set. This patch provides the necessary information to
the SDMA engine driver.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-13-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
Failed register accesses are really not expected in memory mapped
registers. When it fails then the register access itself is likely not
the reason, so no need to have extra error messages for each regmap
access. Just drop the error messages. This also fixes some places where
a return value is concatenated using 'ret |=' and then returned as
error value.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-8-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
Instead regmap_update_bits() use the simpler variants
regmap_[set|clear]_bits() where appropriate.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-7-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
Use GENMASK along with FIELD_PREP and FIELD_GET to access bitfields in
registers to straighten register access and to drop a lot of defines.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-6-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
