kernel/linux.git/include/sound/soc-component.h, branch v6.1.168

ASoC: soc-component: Remove non_legacy_dai_naming flag

2022-06-27T12:17:43+00:00

Now all the users are moved over to the new legacy_dai_naming flag, remove the now unused old flag. Signed-off-by: Charles Keepax Link: https://lore.kernel.org/r/20220623125250.2355471-97-ckeepax@opensource.cirrus.com Signed-off-by: Mark Brown

ASoC: soc-component: Add legacy_dai_naming flag

2022-06-27T12:16:14+00:00

Historically, the legacy DAI naming scheme was applied to platform drivers and the newer scheme to CODEC drivers. During componentisation the core lost the knowledge of if a driver was a CODEC or platform, they were all now components. To continue to support the legacy naming on older platform drivers a flag was added to the snd_soc_component_driver structure, non_legacy_dai_naming, to indicate to use the new scheme and this was applied to all CODECs as part of the migration. However, a slight issue appears to be developing with respect to this flag being opt in for the non-legacy scheme, which presumably we want to be the primary scheme used. Many codec drivers appear to forget to include this flag: grep -l -r "snd_soc_component_driver" sound/soc/codecs/*.c | xargs grep -L "non_legacy_dai_naming" | wc 48 48 556 It would seem more sensible to change the flag to legacy_dai_naming making the new scheme opt out. As a first step this patch adds a new flag for this so that the users can be updated. Signed-off-by: Charles Keepax Link: https://lore.kernel.org/r/20220623125250.2355471-2-ckeepax@opensource.cirrus.com Signed-off-by: Mark Brown

ASoC: simple-card-utils: Move snd_soc_component_is_codec to be local

2022-06-06T11:34:22+00:00

The helper function snd_soc_component_is_codec is based off the presence of the non_legacy_dai_naming flag. This isn't super robust as CPU side components may also specify this flag, and indeed the kernel already contains a couple that do. After componentisation there isn't really a totally robust solution to identifying what is a CODEC driver, without introducing a flag specifically for that purpose, and really the desirable direction to move in is that the distinction doesn't matter. This patch does two things to try to mitigate these problems. Firstly, now that all the other users of the helper function have been removed, it makes the helper function local to the driver rather, than being part of the core. This should help to discourage any new code from being created that depends on the CODEC driver distinction. Secondly, it updates the helper function itself to use the endianness flag rather than the non_legacy_dai_naming flag. The endianness flag is definitely invalid on a CPU side component, so it a more reliable indicator that the device is definitely a CODEC. The vast majority of buses require the CODEC to set the endianness flag, so the number of corner cases should be fairly minimal. It is worth noting that CODECs sending audio over SPI, or built into the CPU CODECs are potential corner cases, however the hope is that in most cases those types of devices do not consitute a simple audio card. Signed-off-by: Charles Keepax Link: https://lore.kernel.org/r/20220519154318.2153729-57-ckeepax@opensource.cirrus.com Signed-off-by: Mark Brown

ASoC: soc-component: Add comment for the endianness flag

2022-05-09T11:46:06+00:00

Add a comment to make the purpose of the endianness flag on the snd_soc_component structure more clear. Signed-off-by: Charles Keepax Link: https://lore.kernel.org/r/20220504170905.332415-2-ckeepax@opensource.cirrus.com Signed-off-by: Mark Brown

ASoC: soc-core: add debugfs_prefix member to snd_soc_component_driver

2022-04-05T18:14:20+00:00

Allow the component debugfs_prefix to be set from snd_soc_component_driver. First use case is avoiding a duplicate debugfs entry error in case a device has multiple components which have the same name therefore. Note that we don't set component->debugfs_prefix if it's set already. That's needed because partially component->debugfs_prefix is set before calling snd_soc_component_initialize(). Signed-off-by: Heiner Kallweit Link: https://lore.kernel.org/r/d18bff6a-1df1-5f95-0cf8-10dbaa62d7be@gmail.com Signed-off-by: Mark Brown

ASoC: soc-component: add snd_soc_pcm_component_delay()

2021-11-29T12:19:41+00:00

Current soc-pcm.c :: soc_pcm_pointer() is assuming that component driver might update runtime->delay silently in snd_soc_pcm_component_pointer() (= A). static snd_pcm_uframes_t soc_pcm_pointer(...) { ... /* clearing the previous total delay */ => runtime->delay = 0; (A) offset = snd_soc_pcm_component_pointer(substream); /* base delay if assigned in pointer callback */ => delay = runtime->delay; ... } 1) The behavior that ".pointer callback secretly updates runtime->delay" is strange and confusable. 2) Current snd_soc_pcm_component_pointer() uses 1st found component's .pointer callback only, thus it is no problem for now. But runtime->delay might be overwrote if it adjusted to multiple components in the future. 3) Component delay is updated at .pointer callback timing (secretly). But some components which doesn't have .pointer callback might want to increase runtime->delay for some reasons. We already have .delay function for DAI, but not have for Component. This patch adds new snd_soc_pcm_component_delay() for it. Signed-off-by: Kuninori Morimoto Link: https://lore.kernel.org/r/874k8cy25t.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown

ASoC: soc-component: add snd_soc_component_is_codec()

2021-10-18T12:56:37+00:00

Checking .non_legacy_dai_naming is not readable. Let's add new snd_soc_component_is_codec(). Signed-off-by: Kuninori Morimoto Link: https://lore.kernel.org/r/87h7dft7dn.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown

ASoC: soc-component: Remove conditional definition of debugfs data members

2021-10-01T18:34:58+00:00

This simplification allows the use of the standard kernel pattern of static inline dummy functions for debugfs code. Most systems will only have a small number of snd_soc_components so the memory impact is minimal. Signed-off-by: Simon Trimmer Suggested-by: Mark Brown Link: https://lore.kernel.org/r/20210930142116.528878-1-simont@opensource.cirrus.com Signed-off-by: Mark Brown

ASoC: compress/component: Use module_get_when_open/put_when_close for cstream

2021-09-20T12:30:18+00:00

Currently the try_module_get() and module_put() is not possible for compressed streams if the module_get_upon_open is set to 1 which means that\ the components are not protected in a same way as components when normal audio is used. SOF is setting module_get_upon_open to 1 for component drivers which works correctly for audio stream but when compressed stream is used then the module is not protected. Convert the compress open and free operation to mimic the steps of it's pcm counterpart to fix this issue. Signed-off-by: Peter Ujfalusi Link: https://lore.kernel.org/r/20210901095255.3617-3-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown

ASoC: soc-component: Convert the mark_module to void*

2021-09-20T12:30:17+00:00

The mark_module of the snd_soc_component is strict snd_pcm_substream type which prevents it to be used by compressed streams. Change the type to void* along with the snd_soc_component_module_get() and snd_soc_component_module_put() to allow the same mark to be used by compressed when it's module_get_upon_open is set to 1. Signed-off-by: Peter Ujfalusi Link: https://lore.kernel.org/r/20210901095255.3617-2-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown