January 2023 - Alsa-devel - mailman.alsa-project.org

[PATCH] ASoC: mediatek: mt8186: add AFE_DAC_CON0 to volatile register list
by Chunxu Li 23 Jan '23

23 Jan '23

Mark AFE_DAC_CON0 as volatile since DSP firmware will access this register too. Signed-off-by: Chunxu Li <chunxu.li(a)mediatek.com> --- sound/soc/mediatek/mt8186/mt8186-afe-pcm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c b/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c index d7e94e6a19c7..41172a82103e 100644 --- a/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c +++ b/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c @@ -2395,6 +2395,7 @@ static bool mt8186_is_volatile_reg(struct device *dev, unsigned int reg) case AUDIO_TOP_CON1: /* reg bit controlled by CCF */ case AUDIO_TOP_CON2: case AUDIO_TOP_CON3: + case AFE_DAC_CON0: case AFE_DL1_CUR_MSB: case AFE_DL1_CUR: case AFE_DL1_END: -- 2.25.1

2 1

[PATCH] ASoC: qdsp6: audioreach: Add ADSP ready check
by Srinivasa Rao Mandadapu 23 Jan '23

23 Jan '23

Check for SPF readiness in prm driver probe to avoid race conditions during ADSP pil loading. This patch is to avoid, sending requests to ADSP before it's power domains are up and ready. Signed-off-by: Srinivasa Rao Mandadapu <quic_srivasam(a)quicinc.com> Tested-by: Ratna Deepthi Kudaravalli <rkudarav(a)qti.qualcomm.com> --- sound/soc/qcom/qdsp6/q6apm.c | 13 +++++++++++++ sound/soc/qcom/qdsp6/q6apm.h | 2 ++ sound/soc/qcom/qdsp6/q6prm.c | 4 ++++ 3 files changed, 19 insertions(+) diff --git a/sound/soc/qcom/qdsp6/q6apm.c b/sound/soc/qcom/qdsp6/q6apm.c index 5beb898..8a7dfd2 100644 --- a/sound/soc/qcom/qdsp6/q6apm.c +++ b/sound/soc/qcom/qdsp6/q6apm.c @@ -27,6 +27,8 @@ struct apm_graph_mgmt_cmd { #define APM_GRAPH_MGMT_PSIZE(p, n) ALIGN(struct_size(p, sub_graph_id_list, n), 8) +struct q6apm *g_apm; + int q6apm_send_cmd_sync(struct q6apm *apm, struct gpr_pkt *pkt, uint32_t rsp_opcode) { gpr_device_t *gdev = apm->gdev; @@ -143,6 +145,15 @@ static void q6apm_put_audioreach_graph(struct kref *ref) kfree(graph); } +bool q6apm_is_adsp_ready(void) +{ + if (g_apm && g_apm->state) + return true; + + return false; +} +EXPORT_SYMBOL_GPL(q6apm_is_adsp_ready); + static int q6apm_get_apm_state(struct q6apm *apm) { struct gpr_pkt *pkt; @@ -658,6 +669,8 @@ static int apm_probe(gpr_device_t *gdev) idr_init(&apm->modules_idr); + g_apm = apm; + q6apm_get_apm_state(apm); ret = devm_snd_soc_register_component(dev, &q6apm_audio_component, NULL, 0); diff --git a/sound/soc/qcom/qdsp6/q6apm.h b/sound/soc/qcom/qdsp6/q6apm.h index 273f978..7005be9 100644 --- a/sound/soc/qcom/qdsp6/q6apm.h +++ b/sound/soc/qcom/qdsp6/q6apm.h @@ -145,4 +145,6 @@ struct audioreach_module *q6apm_find_module_by_mid(struct q6apm_graph *graph, void q6apm_set_fe_dai_ops(struct snd_soc_dai_driver *dai_drv); int q6apm_graph_get_rx_shmem_module_iid(struct q6apm_graph *graph); +bool q6apm_is_adsp_ready(void); + #endif /* __APM_GRAPH_ */ diff --git a/sound/soc/qcom/qdsp6/q6prm.c b/sound/soc/qcom/qdsp6/q6prm.c index 8aa1a21..3aa63aa 100644 --- a/sound/soc/qcom/qdsp6/q6prm.c +++ b/sound/soc/qcom/qdsp6/q6prm.c @@ -12,6 +12,7 @@ #include <linux/soc/qcom/apr.h> #include <dt-bindings/soc/qcom,gpr.h> #include <dt-bindings/sound/qcom,q6dsp-lpass-ports.h> +#include "q6apm.h" #include "q6prm.h" #include "audioreach.h" @@ -226,6 +227,9 @@ static int prm_probe(gpr_device_t *gdev) init_waitqueue_head(&cc->wait); dev_set_drvdata(dev, cc); + if (!q6apm_is_adsp_ready()) + return -EPROBE_DEFER; + return devm_of_platform_populate(dev); } -- 2.7.4

2 1

[PATCH] ASoC: Intel: avs: Simplify probe-component implementation
by Cezary Rojewski 23 Jan '23

23 Jan '23

There is no need for the probe-component to be part of the PCM component list as it does not make use of ASoC-topology and does not participate in creating any PCM streams. To achieve that, remove probe() and remove() functions. Fixes: ed914a2a45a4 ("ASoC: Intel: avs: Data probing soc-component") Signed-off-by: Cezary Rojewski <cezary.rojewski(a)intel.com> --- This change should be part of the initial series [1] from the get-go, differences between the internal tree and the external one caused me to miss the change when upstreaming. And thus the "fixes" tag. [1]: https://lore.kernel.org/all/20221202152841.672536-1-cezary.rojewski@intel.c… sound/soc/intel/avs/probes.c | 23 ----------------------- 1 file changed, 23 deletions(-) diff --git a/sound/soc/intel/avs/probes.c b/sound/soc/intel/avs/probes.c index 29d63f2a9616..b04f2d1a4c07 100644 --- a/sound/soc/intel/avs/probes.c +++ b/sound/soc/intel/avs/probes.c @@ -277,31 +277,8 @@ static struct snd_soc_dai_driver probe_cpu_dais[] = { }, }; -static int avs_probe_component_probe(struct snd_soc_component *component) -{ - struct avs_soc_component *acomp = to_avs_soc_component(component); - struct avs_dev *adev = to_avs_dev(component->dev); - - mutex_lock(&adev->comp_list_mutex); - list_add_tail(&acomp->node, &adev->comp_list); - mutex_unlock(&adev->comp_list_mutex); - return 0; -} - -static void avs_probe_component_remove(struct snd_soc_component *component) -{ - struct avs_soc_component *acomp = to_avs_soc_component(component); - struct avs_dev *adev = to_avs_dev(component->dev); - - mutex_lock(&adev->comp_list_mutex); - list_del(&acomp->node); - mutex_unlock(&adev->comp_list_mutex); -} - static const struct snd_soc_component_driver avs_probe_component_driver = { .name = "avs-probe-compr", - .probe = avs_probe_component_probe, - .remove = avs_probe_component_remove, .compress_ops = &avs_probe_compress_ops, .module_get_upon_open = 1, /* increment refcount when a stream is opened */ }; -- 2.25.1

2 1

[PATCH v2 0/8] ASoC: cs42l42: Add Soundwire support
by Stefan Binding 23 Jan '23

23 Jan '23

The CS42L42 has a Soundwire interface for control and audio. This chain of patches adds support for this. Patches #1 .. #5 split out various changes to the existing code that are needed for adding Soundwire. These are mostly around clocking and supporting the separate probe and enumeration stages in Soundwire. Patches #6 .. #8 actually adds the Soundwire handling. Changes since v1: - fixes for various review comments from v1 - add support for wakeup from clock stop using hardware interrupts - use port_prep callback to prepare/deprepare codec Richard Fitzgerald (6): ASoC: cs42l42: Add SOFT_RESET_REBOOT register ASoC: cs42l42: Ensure MCLKint is a multiple of the sample rate ASoC: cs42l42: Separate ASP config from PLL config ASoC: cs42l42: Export some functions for Soundwire ASoC: cs42l42: Add Soundwire support ASoC: cs42l42: Don't set idle_bias_on Stefan Binding (2): soundwire: stream: Add specific prep/deprep commands to port_prep callback ASoC: cs42l42: Wait for debounce interval after resume drivers/soundwire/stream.c | 4 +- include/linux/soundwire/sdw.h | 8 +- include/sound/cs42l42.h | 5 + sound/soc/codecs/Kconfig | 8 + sound/soc/codecs/Makefile | 2 + sound/soc/codecs/cs42l42-sdw.c | 603 +++++++++++++++++++++++++++++++++ sound/soc/codecs/cs42l42.c | 127 ++++--- sound/soc/codecs/cs42l42.h | 9 +- 8 files changed, 716 insertions(+), 50 deletions(-) create mode 100644 sound/soc/codecs/cs42l42-sdw.c -- 2.34.1

3 18

Question about DPCM private_data of soc-compress.c
by Kuninori Morimoto 23 Jan '23

23 Jan '23

Hi ALSA ML I'm not familiar with soc-compress.c, but I think dpcm_capture case don't need "else" ? I think it is same as soc-pcm.c :: soc_new_pcm(). I'm happy to post patch if someone acknowledge this question. --- soc-compress.c -------- int snd_soc_new_compress(...) { ... if (rtd->dai_link->dynamic) { ... if (rtd->dai_link->dpcm_playback) be_pcm->streams[SNDRV_PCM_STREAM_PLAYBACK].substream->private_data = rtd; - else if (rtd->dai_link->dpcm_capture) + if (rtd->dai_link->dpcm_capture) be_pcm->streams[SNDRV_PCM_STREAM_CAPTURE].substream->private_data = rtd; ... } ... } --- soc-pcm.c --------------- int soc_new_pcm(...) { ... if (rtd->dai_link->no_pcm || rtd->dai_link->params) { if (playback) pcm->streams[SNDRV_PCM_STREAM_PLAYBACK].substream->private_data = rtd; => if (capture) pcm->streams[SNDRV_PCM_STREAM_CAPTURE].substream->private_data = rtd; goto out; } ... } Thank you for your help !! Best regards --- Kuninori Morimoto

3 2

[PATCH v2 0/3] Add the Renesas IDT821034 codec support
by Herve Codina 23 Jan '23

23 Jan '23

Hi, The Renesas IDT821034 codec is four channel PCM codec with on-chip filters and programmable gain setting. It also provides SLIC (Subscriber Line Interface Circuit) signals as GPIOs. Based on previous iteration: https://lore.kernel.org/all/20230111134905.248305-1-herve.codina@bootlin.co… the bigger change is the codec driver rework in order to remove the regmap virtual registers layer. Best regards, Herve Codina Changes v1 -> v2: - All patches Reformat commit log - Patch 1 Remove '$ref: /schemas/gpio/gpio.yaml#' Use 'unevaluatedProperties: false' Update the node name and remove the sound node in the example - Patch 2 Change the file header comment format Rework in order to remove the regmap virtual registers Herve Codina (3): dt-bindings: sound: Add Renesas IDT821034 codec ASoC: codecs: Add support for the Renesas IDT821034 codec MAINTAINERS: add the Renesas IDT821034 codec entry .../bindings/sound/renesas,idt821034.yaml | 75 ++ MAINTAINERS | 7 + sound/soc/codecs/Kconfig | 12 + sound/soc/codecs/Makefile | 2 + sound/soc/codecs/idt821034.c | 1200 +++++++++++++++++ 5 files changed, 1296 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/renesas,idt821034.yaml create mode 100644 sound/soc/codecs/idt821034.c -- 2.39.0

4 14

[PATCH] ASoC: amd: ps: Make acp63_fill_platform_dev_info static
by Jiapeng Chong 23 Jan '23

23 Jan '23

This symbol is not used outside of pci-ps.c, so marks it static. sound/soc/amd/ps/pci-ps.c:170:6: warning: no previous prototype for ‘acp63_fill_platform_dev_info’. Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=3640 Reported-by: Abaci Robot <abaci(a)linux.alibaba.com> Signed-off-by: Jiapeng Chong <jiapeng.chong(a)linux.alibaba.com> --- sound/soc/amd/ps/pci-ps.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/soc/amd/ps/pci-ps.c b/sound/soc/amd/ps/pci-ps.c index 401cfd0036be..570736e1b5e8 100644 --- a/sound/soc/amd/ps/pci-ps.c +++ b/sound/soc/amd/ps/pci-ps.c @@ -167,10 +167,10 @@ static void get_acp63_device_config(u32 config, struct pci_dev *pci, } } -void acp63_fill_platform_dev_info(struct platform_device_info *pdevinfo, struct device *parent, - struct fwnode_handle *fw_node, char *name, unsigned int id, - const struct resource *res, unsigned int num_res, - const void *data, size_t size_data) +static void acp63_fill_platform_dev_info(struct platform_device_info *pdevinfo, + struct device *parent, struct fwnode_handle *fw_node, + char *name, unsigned int id, const struct resource *res, + unsigned int num_res, const void *data, size_t size_data) { pdevinfo->name = name; pdevinfo->id = id; -- 2.20.1.7.g153144c

2 1

[PATCH v2] ALSA: hda: Do not unset preset when cleaning up codec
by Cezary Rojewski 23 Jan '23

23 Jan '23

Several functions that take part in codec's initialization and removal are re-used by ASoC codec drivers implementations. Drivers mimic the behavior of hda_codec_driver_probe/remove() found in sound/pci/hda/hda_bind.c with their component->probe/remove() instead. One of the reasons for that is the expectation of snd_hda_codec_device_new() to receive a valid pointer to an instance of struct snd_card. This expectation can be met only once sound card components probing commences. As ASoC sound card may be unbound without codec device being actually removed from the system, unsetting ->preset in snd_hda_codec_cleanup_for_unbind() interferes with module unload -> load scenario causing null-ptr-deref. Preset is assigned only once, during device/driver matching whereas ASoC codec driver's module reloading may occur several times throughout the lifetime of an audio stack. Suggested-by: Takashi Iwai <tiwai(a)suse.com> Signed-off-by: Cezary Rojewski <cezary.rojewski(a)intel.com> --- Changes in v2: - relocated the operation to routines found in hda_bind.c rather than just removing it from the cleanup function This is a continuation of a discussion that begun in the middle of 2022 [1] and was part of a larger series addressing several HDAudio topics. Single rmmod on ASoC's codec driver module is enough to cause a panic. Given our results, no regression shows up with modprobe/rmmod on snd_hda_intel side with this patch applied. [1]: https://lore.kernel.org/alsa-devel/20220706120230.427296-2-cezary.rojewski@… sound/pci/hda/hda_bind.c | 2 ++ sound/pci/hda/hda_codec.c | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/sound/pci/hda/hda_bind.c b/sound/pci/hda/hda_bind.c index 1a868dd9dc4b..890c2f7c33fc 100644 --- a/sound/pci/hda/hda_bind.c +++ b/sound/pci/hda/hda_bind.c @@ -144,6 +144,7 @@ static int hda_codec_driver_probe(struct device *dev) error: snd_hda_codec_cleanup_for_unbind(codec); + codec->preset = NULL; return err; } @@ -166,6 +167,7 @@ static int hda_codec_driver_remove(struct device *dev) if (codec->patch_ops.free) codec->patch_ops.free(codec); snd_hda_codec_cleanup_for_unbind(codec); + codec->preset = NULL; module_put(dev->driver->owner); return 0; } diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index edd653ece70d..ac1cc7c5290e 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -795,7 +795,6 @@ void snd_hda_codec_cleanup_for_unbind(struct hda_codec *codec) snd_array_free(&codec->cvt_setups); snd_array_free(&codec->spdif_out); snd_array_free(&codec->verbs); - codec->preset = NULL; codec->follower_dig_outs = NULL; codec->spdif_status_reset = 0; snd_array_free(&codec->mixers); -- 2.25.1

2 1

[PATCH 00/25] ASoC: use helper function and cleanup
by Kuninori Morimoto 23 Jan '23

23 Jan '23

Hi Mark I'm posting about DT schema fixup, but it seems take longer time. This patch-set is 100% independent from it, so I will post it. struct snd_soc_dai need to have info for playback/capture, but it is using "playback/capture_xxx" or "tx/tx_xxx" or array. This kind of random definition is very difficult to read. This patch-set add helper functions and each driver use it. And cleanup the definition. Kuninori Morimoto (25): ASoC: soc-dai.h: add missing snd_soc_dai_set_widget() ASoC: soc-dai.h: add snd_soc_dai_dma_data_set/get() for low level ASoC: soc-dai.h: add snd_soc_dai_tdm_mask_set/get() helper ASoC: soc-dai.h: add snd_soc_dai_get/set_widget_playback/capture() helper ASoC: soc.h: add snd_soc_card_is_instantiated() helper ASoC: hdmi-codec: use helper function ASoC: hda: use helper function ASoC: max: use helper function ASoC: rt: use helper function ASoC: cirrus: use helper function ASoC: spear: use helper function ASoC: sdw-mockup: use helper function ASoC: intel: use helper function ASoC: mediatek: use helper function ASoC: meson: use helper function ASoC: rockchip: use helper function ASoC: sof: use helper function ASoC: tegra: use helper function ASoC: ti: use helper function ASoC: soc-core.c: use helper function ASoC: soc-dapm.c: use helper function ASoC: soc-dai.c: use helper function ASoC: soc-pcm.c: use helper function ASoC: soc-topology.c: use helper function ASoC: soc-dai.h: cleanup Playback/Capture data for snd_soc_dai include/sound/soc-dai.h | 90 ++++++++------- include/sound/soc.h | 6 + sound/soc/cirrus/ep93xx-ac97.c | 4 +- sound/soc/cirrus/ep93xx-i2s.c | 4 +- sound/soc/codecs/hda.c | 7 +- sound/soc/codecs/hdmi-codec.c | 11 +- sound/soc/codecs/max98373-sdw.c | 5 +- sound/soc/codecs/rt1308-sdw.c | 5 +- sound/soc/codecs/rt1316-sdw.c | 5 +- sound/soc/codecs/rt1318-sdw.c | 5 +- sound/soc/codecs/rt5645.c | 2 +- sound/soc/codecs/rt5665.c | 4 +- sound/soc/codecs/rt5668.c | 4 +- sound/soc/codecs/rt5682-sdw.c | 5 +- sound/soc/codecs/rt5682.c | 4 +- sound/soc/codecs/rt5682s.c | 4 +- sound/soc/codecs/rt700.c | 7 +- sound/soc/codecs/rt711-sdca.c | 7 +- sound/soc/codecs/rt711.c | 7 +- sound/soc/codecs/rt715-sdca.c | 5 +- sound/soc/codecs/rt715.c | 5 +- sound/soc/codecs/sdw-mockup.c | 5 +- sound/soc/intel/atom/sst-atom-controls.c | 9 +- sound/soc/intel/avs/boards/nau8825.c | 5 +- sound/soc/intel/avs/pcm.c | 15 +-- sound/soc/intel/skylake/skl-topology.c | 19 +--- .../soc/mediatek/common/mtk-dsp-sof-common.c | 18 ++- sound/soc/mediatek/mt6797/mt6797-dai-pcm.c | 8 +- sound/soc/mediatek/mt8183/mt8183-dai-pcm.c | 7 +- sound/soc/mediatek/mt8186/mt8186-dai-pcm.c | 7 +- sound/soc/mediatek/mt8192/mt8192-dai-pcm.c | 8 +- sound/soc/mediatek/mt8195/mt8195-dai-pcm.c | 7 +- sound/soc/meson/aiu-fifo-i2s.c | 4 +- sound/soc/meson/aiu-fifo-spdif.c | 2 +- sound/soc/meson/aiu-fifo.c | 21 ++-- sound/soc/meson/axg-tdm-interface.c | 34 +++--- sound/soc/meson/axg-tdmin.c | 2 +- sound/soc/meson/axg-tdmout.c | 2 +- sound/soc/meson/meson-codec-glue.c | 8 +- sound/soc/rockchip/rockchip_i2s_tdm.c | 4 +- sound/soc/rockchip/rockchip_pdm.c | 2 +- sound/soc/rockchip/rockchip_spdif.c | 2 +- sound/soc/soc-core.c | 8 +- sound/soc/soc-dai.c | 13 ++- sound/soc/soc-dapm.c | 26 ++--- sound/soc/soc-pcm.c | 12 +- sound/soc/soc-topology.c | 4 +- sound/soc/sof/topology.c | 104 +++++++----------- sound/soc/spear/spdif_out.c | 3 +- sound/soc/tegra/tegra20_ac97.c | 4 +- sound/soc/tegra/tegra20_i2s.c | 4 +- sound/soc/tegra/tegra20_spdif.c | 3 +- sound/soc/tegra/tegra210_admaif.c | 4 +- sound/soc/tegra/tegra30_i2s.c | 4 +- sound/soc/ti/davinci-i2s.c | 5 +- sound/soc/ti/davinci-mcasp.c | 5 +- sound/soc/ti/davinci-vcif.c | 5 +- 57 files changed, 274 insertions(+), 315 deletions(-) -- 2.25.1

3 34

[PATCH v2 0/2] Update lpass dt-bindings
by Srinivasa Rao Mandadapu 22 Jan '23

22 Jan '23

Update va-macro driver dt-bindings and add missing properties in sc7280 machine driver dt-bindings. Changes since v1: -- Remove sound-dai cells property patch. -- Update example in clock name change patch. -- Update commit message and add maxItems in platform proerty patch. Srinivasa Rao Mandadapu (2): ASoC: qcom: dt-bindings: lpass-va-macro: Update clock name ASoC: dt-bindings: google,sc7280-herobrine: Add platform property .../devicetree/bindings/sound/google,sc7280-herobrine.yaml | 12 ++++++++++++ .../devicetree/bindings/sound/qcom,lpass-va-macro.yaml | 4 ++-- 2 files changed, 14 insertions(+), 2 deletions(-) -- 2.7.4

2 4