If i915 does not support given platform but the hardware i.e.: HDAudio
codec is still there, the codec-probing procedure will succeed for such
device but the follow up initialization will always end up with -ENODEV.
While bus could filter out address '2' which Intel's HDMI/DP codecs
always enumerate on, more robust approach is to check for i915 presence
before registering display codecs.
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Acked-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20240226124432.1203798-3-cezary.rojewski@intel.com
Per discussion on the alsa-devel mailing list [1], the legacy PIN to PCM
device mapping is obsolete nowadays. The maximum number of the simultaneously
usable PCM devices is equal to the HDMI codec converters.
Remove the extra PCM devices (beyond the detected converters) and force
the use of the dynamic PCM device allocation. The legacy code is removed.
I believe that all HDMI codecs have the jack sensing feature. Move the check
to the codec probe function and print a warning, if a codec without this
feature is detected.
[1] https://lore.kernel.org/alsa-devel/2f37e0b2-1e82-8c0b-2bbd-1e5038d6ecc6@perex.cz/
Cc: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Link: https://lore.kernel.org/r/20220922084017.25925-1-perex@perex.cz
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Add generic ASoC equivalent of ALSA HD-Audio codec. This codec is
designed to follow HDA_DEV_LEGACY convention. Driver wrapps existing
hda_codec.c handlers to prevent code duplication within the newly added
code. Number of DAIs created is dependent on capabilities exposed by the
codec itself. Because of this, single solution can be applied to support
every single HD-Audio codec type.
At the same time, through the ASoC topology, platform drivers may limit
the number of endpoints available to the userspace as codec driver
exposes BE DAIs only.
Both hda_codec_probe() and hda_codec_remove() declare their expectations
on device's usage_count and suspended-status. This is to catch any
unexpected behavior as PM-related code for HD-Audio has been changing
quite a bit throughout the years.
In order for codec DAI list to reflect its actual PCM capabilities, PCMs
need to be built and that can only happen once codec device is
constructed. To do that, a valid component->card->snd_card pointer is
needed. Said pointer will be provided by the framework once all card
components are accounted for and their probing can begin. Usage of
"binder" BE DAI solves the problem - codec can be listed as one of
HD-Audio card components without declaring any actual BE DAIs
statically.
Relation with hdac_hda:
Addition of parallel solution is motivated by behavioral differences
between hdac_hda.c and its legacy equivalent found in sound/pci/hda
e.g.: lack of dynamic, based on codec capabilities, resource allocation
and high cost of removing such differences on actively used targets.
Major goal of codec driver presented here is to follow HD-Audio legacy
behavior in 1:1 fashion by becoming a wrapper. Doing so increases code
coverage of the legacy code and reduces the maintenance cost for both
solutions.
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220511162403.3987658-3-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
