Now we only allocate 44 bytes for the control name keep it on the stack
to avoid a lot of pointless memory allocation.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
ALSA only supports control names up to 44 bytes, so there is no point
allocating a whole page of memory to hold the control name, just limit
the control name to 44 bytes.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The algorithm region information in the firmware doesn't contain a
length field, explicitly pass this to the create_control function rather
than bundling into wm_adsp_alg_region.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
We have wm_adsp_region, wm_adsp_alg_region, and wmfw_region, the
variables for which are all frequently called region, this can get quite
confusing when reviewing the code especially given some functions are
quite long. Consistently use mem for wm_adsp_regions, alg_region for
wm_adsp_alg_region and region for wmfw_region.
Additionally, we use a mix of adsp and dsp for pointers to the wm_adsp
structure standardise this on dsp.
Finally, we use algs to refer to the number of algorithms quite
frequently, change this to the more descriptive n_algs.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The vast majority of the wm_adsp_setup_algs function is case statements
for ADSP1 or ADSP2, this patch splits this out into two separate
functions wm_adsp1_setup_algs and wm_adsp2_setup_algs. The small amount
of shared code between them is factored out into an extra helper
function. This makes the code a lot cleaner.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Whilst the existing code does correctly round to the next 4-byte boundary
it does so rather inefficiently. This patch changes the rounding to be
simpler and more efficient.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Currently DSP controls are persistent (across DSP On/Off) only
if they were set whilst the DSP is off. This change makes the
controls persistent irrespective of when they are set.
Signed-off-by: Nikesh Oswal <Nikesh.Oswal@wolfsonmicro.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The codec field of the snd_soc_widget struct is eventually going to be
removed, use snd_soc_dapm_to_codec(w->dapm) instead.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Many firmwares do not wait for the start bit before they begin
processing audio, whilst this is a bug on the firmware side there are
too many such firmwares in the wild to ignore the situation. This patch
moves the core enable to happen at same time as the start, the firmware
looses the ability to overlap its own startup with the audio path bring
up but we ensure that all firmwares behave.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: JS Park <aitdark.park@samsung.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
We should not free any buffers associated with writing out coefficients
to the DSP until all the async writes have completed. This patch updates
the out of memory path when allocating a new buffer to include a call to
regmap_async_complete.
Reported-by: JS Park <aitdark.park@samsung.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org
Use vmalloc to allocate the buffer for firmware/coefficient download and
rely on the SPI core to split this up into DMA-able chunks. This should
give better performance and means we no longer need to manually split
the download into page size chunks to avoid allocating overly large
continuous memory regions.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Both the snd_soc_codec and snd_soc_platform struct do have a pointer to the
parent card and both handle this pointer in mostly the same way. This patch
moves the card field to the component level which will allow further code
consolidation between platforms and CODECS.
Since there are only a handful of users of the snd_soc_codec struct's card field
(and none of the snd_soc_platform's) these are update in this patch as well,
which allows it to be removed from the snd_soc_codec struct.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
Since MODULE_LICENSE is missing the module load fails,
so add this for module.
Signed-off-by: Praveen Diwakar <praveen.diwakar@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Cc: stable@vger.kernel.org
We have defines for adsp messages best to consistently use them.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
For CODEC controls snd_kcontrol_chip() currently returns a pointer to the
CODEC that registered the control. With the upcoming consolidation of
platform and CODEC controls this will change. Prepare for this by introducing
the snd_soc_kcontrol_codec() helper function that will hide the implementation
details of how the CODEC for a control can be obtained. This will allow us to
change this easily in the future.
The patch also updates all CODEC drivers to use the new helper function.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
The firmware files can be quite large and allocating the whole firmware
a single DMA safe buffer can be problematic if the system is under a
high memory load. Ease the requirements slightly by writing the firmware
out in page sized chunks.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
It can be useful for debugging purposes to see at what point the DSP has
powered down, so add a message to inform us of this.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Move the start of booting the DSP to earlier in the DAPM process, and
move the final starting of the DSP to later in the DAPM process. This
allows us to overlap some of the processing with other components of the
system being brought up.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Move the ADSP2 boot proceedure into a work structure in preparation for
running it asynchronously with the reset of the audio path bring up.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Some devices are getting very close to the limit whilst polling the RAM
start, this patch adds a small delay to this loop to give a longer
startup timeout.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Cc: stable@vger.kernel.org
This will allow a marginal speed improvement when used with a bus that
supports async I/O by reducing the amount of context thrashing between
writes, allowing the bus to be more fully utilised.
Signed-off-by: Mark Brown <broonie@linaro.org>
When locating the memory region relating to a coefficient block written
through a bin file we keep processing the list of regions even after we
have found the region we require. This patch adds a break, so we don't
process redundant list items.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
This patch does:
- Move the sanity check with WARN_ON() in wm_adsp_region_to_reg() and
remove the checks in the callers,
- Fix wrong WARN_ON() usages, replaced with WARN(),
- Fix unreachable or wrong BUG_ON() usages and replace with WARN_ON().
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Pad the ADSP word (3 bytes) to 4 bytes in the kernel and calculate
lengths based on padded ADSP words instead of treating them as bytes
Signed-off-by: Nariman Poushin <nariman@opensource.wolfsonmicro.com>
Signed-off-by: Dimitris Papastamos <dp@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Cc: stable@vger.kernel.org
Do it in a similar fashion as we do for ADSP2.
Signed-off-by: Dimitris Papastamos <dp@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
No need to hold on to the `codec' pointer. We can use the `dsp'
pointer and grab all the information we need from there. This
makes the parameters for the functions a bit more sane and idiomatic.
Signed-off-by: Dimitris Papastamos <dp@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Get rid off the wm_coeff struct and the wm_coeff_add_kcontrol()
function. We are now using the snd_soc_card_kcontrol() function to
get the kcontrol pointers. No need to call into ALSA code to
register the kcontrols.
Signed-off-by: Dimitris Papastamos <dp@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Reported-by: Dennis May <dennis.may@wolfsonmicro.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
Rename `dirty' to `set' as it is a bit more descriptive. A set
control is any control that has been set by the user. We need to ensure
that everytime we boot the DSP we sync out any controls that were
set.
We could at some point start keeping track of the default values of
the controls to suppress some of the device I/O.
Signed-off-by: Dimitris Papastamos <dp@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
There are two return paths which don't kfree(name).
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Add initial support for runtime tuning for the ADSP cores. This
is achieved by exposing the coefficient configuration blocks as
ALSA binary controls.
The current code assumes that no controls on the DSP are volatile.
Signed-off-by: Dimitris Papastamos <dp@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Firmwares may provide some firmware wide configuration regions which can
be configured by the coefficient files using the firmware ID as the
algorithm ID, include these in the algorithm list.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Now that we have regular register mapped controls we should be splitting
the control sets for ADSP1 and ADSP2 as the register maps are not
identical. Do that.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Since ASoC does not yet really have the framework features needed to
support propagating sample rates through the device well yet implement
basic support for the ISRCs equivalent to that we currently have for the
ASRCs. The user can opt for 8kHz or 16kHz as the rate for the DSP blocks
in addition to the main audio rate, these being the primary use cases.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Older .bin files report the global coefficients as absolute address writes
to zero; maintain compatibility with them.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
'file' is malloced in wm_adsp_load_coeff() and should be freed
before leaving from the error handling cases, otherwise it will cause
memory leak.
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Allow the regmap API to use asynchronous I/O where supported to minimise
the delay between transfers, reducing firmware download times.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Maximise robustness for the widest range of firmwares by ensuring the DSP
is in a consistent state when halted.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
This patch correctly releases the firmware if the magic string in the
firmware header does not match.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Normally kmalloc() returns things that are DMA safe so not visible on all
platforms but we do need to explicitly request DMA safe memory.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Normally kmalloc() returns things that are DMA safe so not visible on all
platforms but we do need to explicitly request DMA safe memory.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Sets the ADSP1 clock rate to match the system clock
rate. To support this the codec driver provides
details of register containing the system clock
control bits.
Signed-off-by: Chris Rattray <crattray@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Implement support for a new revision of the coefficeint file format for
ADSP cores.
Since coefficient file format 0 has not been widely deployed and is very
unlikely to ever be used with this driver code support for it has been
removed.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
There are many firmwares available for ADSP devices. Add basic support
for selecting between them, including a couple of feature sets in the
set of available firmware to start off with.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
WMDR coefficient files can specify coefficients in terms of algorithm
specific data regions. Record the start addresses of these regions while
parsing the algorithms and then use them to handle coefficients with
these formats.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
If we run into I/O problems the algorithm count may be crazy, validate it
before we proceed and dump the read data for diagnostic purposes.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
ADSP devices report information on the algorithms loaded on them. Parse
this data and use it to allow coefficients to be configured for specific
algorithms.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Otherwise we won't run correctly on systems that require this for larger
data transfers.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Some ADSP devices can make use of DVFS to optimise power consumption
depending on the operating frequency of the DSP core. Implement
support for this in the generic ADSP code.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Many current Wolfson devices feature DSPs based around an architecture
known as ADSP. Since there is a lot of commonality in the system
integration of these devices a common library will be used to provide
support for them.
This version provides equivalent support for ADSP1 to that currently
included in the WM2200 driver.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Charles Keepax