To prevent rmmod and similar behave unexpectedly when invoked on
snd_soc_avs module while the AudioDSP firmware tracing is ongoing,
increase the module refcount until the tracing is stopped.
Signed-off-by: Cezary Rojewski <[email protected]>
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
In order to instantiate modules on the firmware side, the driver sends
payload with module configuration. In some case size of this information
is not known before hand, so driver allocates temporary memory during
module creation and frees it after use. Optimize the flow a bit, by
preallocating maximum buffer. This removes the time spend on allocating
memory, as well as potential OOM errors during module initialization.
Handlers for modules, where configuration data fits on stack, are left
as is.
Reviewed-by: Cezary Rojewski <[email protected]>
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
Instead of using PAGE_SIZE as base of definitions in headers, use
generic size defines. While x86 platforms use 4096 as page size, there
are platforms which use different page sizes. Two of changed defines are
for memory windows on DSP side, which have fixed size independent of
host side page size. Another one is for CLDMA buffer which also doesn't
need to change with page size.
Reviewed-by: Cezary Rojewski <[email protected]>
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
There are examples in documentation for codec to codec connection.
However they show method before recent series of patches which renamed
the fields. Update documentation accordingly.
Fixes: 7ddc7f91be ("ASoC: soc.h: clarify Codec2Codec params")
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
Merge series from [email protected]:
This patch series create a TDM source select property and use it to
decide which TDM data source is connected.
Following by the below patch disccuion
https://lore.kernel.org/lkml/[email protected]/#t
It may not be a good choice to add the user controlable mixer control
item. Since it follows the board design, make it as a device property.
Add two properties, the "awinic,audio-channel" property and the
"awinic,sync-flag". The "awinic,audio-channel" is used to make
different PA load different configurations, the "awinic,sync-flag"
is used to synchronize the phases of multiple PA. These two properties
will be read by the corresponding driver, allowing multi-PA to
achieve better playback effect.
Signed-off-by: Weidong Wang <[email protected]>
Acked-by: Rob Herring <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
The commit 1da681e528 ("ASoC: soc-pcm.c: Clear DAIs parameters after
stream_active is updated") tries to make sure DAI parameters can be
cleared properly through moving the cleanup to the place where stream
active status is updated. However, it will cause the cleanup only
happening in soc_pcm_close().
Suppose a case: aplay -Dhw:0 44100.wav 48000.wav. The case calls
soc_pcm_open()->soc_pcm_hw_params()->soc_pcm_hw_free()->
soc_pcm_hw_params()->soc_pcm_hw_free()->soc_pcm_close() in order. The
parameters would be remained in the system even if the playback of
44100.wav is finished.
The case requires us clearing parameters in phase of soc_pcm_hw_free().
However, moving the DAI parameters cleanup back to soc_pcm_hw_free()
has the risk that DAIs parameters never be cleared if there're more
than one stream, see commit 1da681e528 ("ASoC: soc-pcm.c: Clear DAIs
parameters after stream_active is updated") for more details.
To meet all these requirements, in addition to do DAI parameters
cleanup in soc_pcm_hw_free(), also check it in soc_pcm_close() to make
sure DAI parameters cleared if the DAI becomes inactive.
Fixes: 1da681e528 ("ASoC: soc-pcm.c: Clear DAIs parameters after stream_active is updated")
Signed-off-by: Chancel Liu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
Merge series from Kuninori Morimoto <[email protected]>:
ASoC is using 2 type of prefix (asoc_xxx() vs snd_soc_xxx()), but there
is no particular reason about that [1].
To reduce confusing, standarding these to snd_soc_xxx() is sensible.
This patch adds asoc_xxx() macro to keep compatible for a while.
It will be removed if all drivers were switched to new style.
Link: https://lore.kernel.org/r/[email protected] [1]
As the pll_id and pll_id can be zero (WM8960_SYSCLK_AUTO)
with the commit 2bbc2df46e ("ASoC: wm8960: Make automatic the
default clocking mode")
Then the machine driver will skip to call set_sysclk() and set_pll()
for codec, when the sysclk rate is different with what wm8960 read
at probe, the output sound frequency is wrong.
So change the fll_id and pll_id initial value, still keep machine
driver's behavior same as before.
Signed-off-by: Shengjiu Wang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
Just as unevaluatedProperties or additionalProperties are required at
the top level of schemas, they should (and will) also be required for
child node schemas. That ensures only documented properties are
present for any node.
Add unevaluatedProperties or additionalProperties as appropriate.
Signed-off-by: Rob Herring <[email protected]>
Acked-by: Herve Codina <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
sound/soc/sh/dma-sh7760.c: In function ‘camelot_prepare’:
./include/linux/kern_levels.h:5:25: warning: format ‘%lx’ expects argument of type ‘long unsigned int’, but argument 2 has type ‘unsigned int’ [-Wformat=]
5 | #define KERN_SOH "\001" /* ASCII Start Of Header */
sound/soc/sh/dma-sh7760.c:198:9: note: in expansion of macro ‘pr_debug’
198 | pr_debug("PCM data: addr 0x%08lx len %d\n",
| ^~~~~~~~
Fix this by using "%pad" and taking the address to format the DMA
address. While at it, use "%zu" to format size_t.
Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Signed-off-by: Geert Uytterhoeven <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mark Brown <[email protected]>
"io-channel-names" is expected to have few values, so there is no real
point to allocate audio_iio_aux_chan structure with a dedicate memory
allocation.
Using a flexible array for struct audio_iio_aux->chans avoids the
overhead of an additional, managed, memory allocation.
This also saves an indirection when the array is accessed.
Finally, __counted_by() can be used for run-time bounds checking if
configured and supported by the compiler.
Signed-off-by: Christophe JAILLET <[email protected]>
Link: https://lore.kernel.org/r/1c0090aaf49504eaeaff5e7dd119fd37173290b5.1695540940.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Mark Brown <[email protected]>
Cezary Rojewski