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

[PATCH v2 0/2] ASoC: qcom: common: Parse auxiliary devices from device tree
by Stephan Gerhold 08 Sep '20

08 Sep '20

In some cases we need to probe additional audio components that do not appear as part of the DAI links specified in the device tree. Examples for this are auxiliary devices such as analog amplifiers or codecs. The ASoC core provides a way to probe these components by adding them to snd_soc_card->aux_dev. This patch set allows specifying them in the device tree through a new "aux-devs" property. v1: https://lore.kernel.org/linux-arm-msm/20200819091533.2334-1-stephan@gerhold… Changes in v2: - Fix value type in device tree bindings: aux-devs should be array of phandles without any arguments, so change <phandles with arguments> -> <array of phandles> Stephan Gerhold (2): ASoC: dt-bindings: qcom: Document "aux-devs" property ASoC: qcom: common: Parse auxiliary devices from device tree .../devicetree/bindings/sound/qcom,apq8016-sbc.txt | 7 +++++++ Documentation/devicetree/bindings/sound/qcom,apq8096.txt | 8 ++++++++ Documentation/devicetree/bindings/sound/qcom,sdm845.txt | 8 ++++++++ sound/soc/qcom/common.c | 4 ++++ 4 files changed, 27 insertions(+) -- 2.28.0

2 3

[PATCH v2 0/4] soundwire: qcom: add support for mmio soundwire master
by Jonathan Marek 08 Sep '20

08 Sep '20

This adds initial support for soundwire device on sm8250. Tested with the "wsa" sdw device, which is simpler than the others. v2 addresses some feedback, but I kept this series as simple as possible. In particular, I didn't implement CMD_NACKED from FIFO_STATUS, because the downstream driver doesn't define this bit, so I can't implement it. Soundwire works without it and It shouldn't be difficult to implement later. Jonathan Marek (4): soundwire: qcom: fix abh/ahb typo soundwire: qcom: avoid dependency on CONFIG_SLIMBUS soundwire: qcom: add support for mmio soundwire master devices soundwire: qcom: add v1.5.1 compatible .../bindings/soundwire/qcom,sdw.txt | 1 + drivers/soundwire/Kconfig | 2 +- drivers/soundwire/qcom.c | 38 +++++++++++++++++-- 3 files changed, 36 insertions(+), 5 deletions(-) -- 2.26.1

4 9

[PATCH] ASoC: tlv320adcx140: Add support for master mode
by Camel Guo 08 Sep '20

08 Sep '20

From: Camel Guo <camelg(a)axis.com> Add setup of bclk-to-ws ratio and sample rate when in master mode, as well as MCLK input pin setup. Limitations: - MCLK is assumed to be connected to GPIO1 (other possible choices would be GPI0..3 if they weren't used for analog positive inputs). - PLL is always enabled, and clock generation is in automatic mode. Signed-off-by: Camel Guo <camelg(a)axis.com> --- sound/soc/codecs/tlv320adcx140.c | 134 ++++++++++++++++++++++++++++++- sound/soc/codecs/tlv320adcx140.h | 35 ++++++++ 2 files changed, 165 insertions(+), 4 deletions(-) diff --git a/sound/soc/codecs/tlv320adcx140.c b/sound/soc/codecs/tlv320adcx140.c index 13ece039e9fe..f143b6b62f84 100644 --- a/sound/soc/codecs/tlv320adcx140.c +++ b/sound/soc/codecs/tlv320adcx140.c @@ -35,6 +35,7 @@ struct adcx140_priv { unsigned int dai_fmt; unsigned int tdm_delay; unsigned int slot_width; + int master; }; static const char * const gpo_config_names[] = { @@ -614,12 +615,85 @@ static int adcx140_reset(struct adcx140_priv *adcx140) return ret; } +static int adcx140_fs_bclk_ratio(unsigned int bclk_ratio) +{ + switch (bclk_ratio) { + case 16: + return ADCX140_RATIO_16; + case 24: + return ADCX140_RATIO_24; + case 32: + return ADCX140_RATIO_32; + case 48: + return ADCX140_RATIO_48; + case 64: + return ADCX140_RATIO_64; + case 96: + return ADCX140_RATIO_96; + case 128: + return ADCX140_RATIO_128; + case 192: + return ADCX140_RATIO_192; + case 256: + return ADCX140_RATIO_256; + case 384: + return ADCX140_RATIO_384; + case 512: + return ADCX140_RATIO_512; + case 1024: + return ADCX140_RATIO_1024; + case 2048: + return ADCX140_RATIO_2048; + default: + break; + } + return -EINVAL; +} + +static int adcx140_fs_rate(unsigned int rate) +{ + switch (rate) { + case 7350: + case 8000: + return ADCX140_8_OR_7_35KHZ; + case 14700: + case 16000: + return ADCX140_16_OR_14_7KHZ; + case 22050: + case 24000: + return ADCX140_24_OR_22_05KHZ; + case 29400: + case 32000: + return ADCX140_32_OR_29_4KHZ; + case 44100: + case 48000: + return ADCX140_48_OR_44_1KHZ; + case 88200: + case 96000: + return ADCX140_96_OR_88_2KHZ; + case 176400: + case 192000: + return ADCX140_192_OR_176_4KHZ; + case 352800: + case 384000: + return ADCX140_384_OR_352_8KHZ; + case 705600: + case 768000: + return ADCX140_768_OR_705_6KHZ; + default: + break; + } + return -EINVAL; +} + static int adcx140_hw_params(struct snd_pcm_substream *substream, struct snd_pcm_hw_params *params, struct snd_soc_dai *dai) { struct snd_soc_component *component = dai->component; + struct adcx140_priv *adcx140 = snd_soc_component_get_drvdata(component); u8 data = 0; + u8 mst_cfg0 = 0; switch (params_width(params)) { case 16: @@ -640,6 +714,60 @@ static int adcx140_hw_params(struct snd_pcm_substream *substream, return -EINVAL; } + if (adcx140->master) { + u8 mst_cfg1 = 0; + u8 gpio_cfg0 = 0; + unsigned int bclk_ratio; + int ret; + + mst_cfg0 = ADCX140_MST_SLV_CFG_BIT; + if (params_rate(params) % 1000) + mst_cfg0 |= ADCX140_FS_MODE_BIT; /* 44.1 kHz et al */ + + ret = adcx140_fs_rate(params_rate(params)); + if (ret < 0) { + dev_err(component->dev, "%s: Unsupported rate %d\n", + __func__, params_rate(params)); + return ret; + } + mst_cfg1 |= ret; + + /* In slave mode when using automatic clock configuration, + * the codec figures out the BCLK to FSYNC ratio itself. But + * here in master mode, we need to tell it. + */ + + bclk_ratio = snd_soc_params_to_frame_size(params); + ret = adcx140_fs_bclk_ratio(bclk_ratio); + if (ret < 0) { + dev_err(component->dev, "%s: Unsupported bclk_ratio %d\n", + __func__, bclk_ratio); + return ret; + } + mst_cfg1 |= ret; + + /* The codec can receive MCLK on any input: GPIO, GPI0..3. + * While in future it would be nice to have these selectable + * via clk_id for instance, for now we fix GPIO1 as the MCLK + * source. + */ + gpio_cfg0 |= ADCX140_GPIO1_MCLK; + + snd_soc_component_update_bits(component, ADCX140_MST_CFG1, + ADCX140_FS_RATE_MSK | + ADCX140_RATIO_MSK, + mst_cfg1); + + snd_soc_component_update_bits(component, ADCX140_GPIO_CFG0, + ADCX140_GPIO1_CFG_MSK, + gpio_cfg0); + } + + snd_soc_component_update_bits(component, ADCX140_MST_CFG0, + ADCX140_FS_MODE_BIT | + ADCX140_MST_SLV_CFG_BIT, + mst_cfg0); + snd_soc_component_update_bits(component, ADCX140_ASI_CFG0, ADCX140_WORD_LEN_MSK, data); @@ -652,16 +780,16 @@ static int adcx140_set_dai_fmt(struct snd_soc_dai *codec_dai, struct snd_soc_component *component = codec_dai->component; struct adcx140_priv *adcx140 = snd_soc_component_get_drvdata(component); u8 iface_reg1 = 0; - u8 iface_reg2 = 0; int offset = 0; int width = adcx140->slot_width; /* set master/slave audio interface */ switch (fmt & SND_SOC_DAIFMT_MASTER_MASK) { case SND_SOC_DAIFMT_CBM_CFM: - iface_reg2 |= ADCX140_BCLK_FSYNC_MASTER; + adcx140->master = 1; break; case SND_SOC_DAIFMT_CBS_CFS: + adcx140->master = 0; break; case SND_SOC_DAIFMT_CBS_CFM: case SND_SOC_DAIFMT_CBM_CFS: @@ -714,8 +842,6 @@ static int adcx140_set_dai_fmt(struct snd_soc_dai *codec_dai, ADCX140_BCLKINV_BIT | ADCX140_ASI_FORMAT_MSK, iface_reg1); - snd_soc_component_update_bits(component, ADCX140_MST_CFG0, - ADCX140_BCLK_FSYNC_MASTER, iface_reg2); /* Configure data offset */ snd_soc_component_update_bits(component, ADCX140_ASI_CFG1, diff --git a/sound/soc/codecs/tlv320adcx140.h b/sound/soc/codecs/tlv320adcx140.h index eedbc1d7221f..bac34800b189 100644 --- a/sound/soc/codecs/tlv320adcx140.h +++ b/sound/soc/codecs/tlv320adcx140.h @@ -145,4 +145,39 @@ #define ADCX140_GPO_CFG_MAX 4 #define ADCX140_GPO_DRV_MAX 5 +/* MST_CFG0 */ +#define ADCX140_MST_SLV_CFG_BIT BIT(7) +#define ADCX140_FS_MODE_BIT BIT(3) + +/* MST_CFG1 */ +#define ADCX140_8_OR_7_35KHZ (0 << 4) +#define ADCX140_16_OR_14_7KHZ (1 << 4) +#define ADCX140_24_OR_22_05KHZ (2 << 4) +#define ADCX140_32_OR_29_4KHZ (3 << 4) +#define ADCX140_48_OR_44_1KHZ (4 << 4) +#define ADCX140_96_OR_88_2KHZ (5 << 4) +#define ADCX140_192_OR_176_4KHZ (6 << 4) +#define ADCX140_384_OR_352_8KHZ (7 << 4) +#define ADCX140_768_OR_705_6KHZ (8 << 4) +#define ADCX140_FS_RATE_MSK GENMASK(7, 4) + +#define ADCX140_RATIO_16 0 +#define ADCX140_RATIO_24 1 +#define ADCX140_RATIO_32 2 +#define ADCX140_RATIO_48 3 +#define ADCX140_RATIO_64 4 +#define ADCX140_RATIO_96 5 +#define ADCX140_RATIO_128 6 +#define ADCX140_RATIO_192 7 +#define ADCX140_RATIO_256 8 +#define ADCX140_RATIO_384 9 +#define ADCX140_RATIO_512 10 +#define ADCX140_RATIO_1024 11 +#define ADCX140_RATIO_2048 12 +#define ADCX140_RATIO_MSK GENMASK(3, 0) + +/* GPIO_CFG0 */ +#define ADCX140_GPIO1_MCLK (10 << 4) +#define ADCX140_GPIO1_CFG_MSK GENMASK(7, 4) + #endif /* _TLV320ADCX140_ */ -- 2.20.1

2 1

[PATCH v2 0/4] soundwire: Remove sdw stream operations from Intel
by Bard Liao 08 Sep '20

08 Sep '20

Sdw stream operation APIs can be called once per stream. Move these operations to dailink ops. The linked series is "ASoC: Add sdw stream operations to dailink ops". Pierre-Louis Bossart (4): soundwire: stream: fix NULL/IS_ERR confusion soundwire: intel: fix NULL/ERR_PTR confusion soundwire: intel: remove .trigger operation soundwire: intel: remove stream handling from .prepare and .hw_free drivers/soundwire/intel.c | 60 +++++--------------------------------- drivers/soundwire/stream.c | 2 +- 2 files changed, 9 insertions(+), 53 deletions(-) -- 2.17.1

3 6

[PATCH v2] Revert "ALSA: hda: Add support for Loongson 7A1000 controller"
by Tiezhu Yang 08 Sep '20

08 Sep '20

This reverts commit 61eee4a7fc40 ("ALSA: hda: Add support for Loongson 7A1000 controller") to fix the following error on the Loongson LS7A platform: rcu: INFO: rcu_preempt self-detected stall on CPU <SNIP> NMI backtrace for cpu 0 CPU: 0 PID: 68 Comm: kworker/0:2 Not tainted 5.8.0+ #3 Hardware name: , BIOS Workqueue: events azx_probe_work [snd_hda_intel] <SNIP> Call Trace: [<ffffffff80211a64>] show_stack+0x9c/0x130 [<ffffffff8065a740>] dump_stack+0xb0/0xf0 [<ffffffff80665774>] nmi_cpu_backtrace+0x134/0x140 [<ffffffff80665910>] nmi_trigger_cpumask_backtrace+0x190/0x200 [<ffffffff802b1abc>] rcu_dump_cpu_stacks+0x12c/0x190 [<ffffffff802b08cc>] rcu_sched_clock_irq+0xa2c/0xfc8 [<ffffffff802b91d4>] update_process_times+0x2c/0xb8 [<ffffffff802cad80>] tick_sched_timer+0x40/0xb8 [<ffffffff802ba5f0>] __hrtimer_run_queues+0x118/0x1d0 [<ffffffff802bab74>] hrtimer_interrupt+0x12c/0x2d8 [<ffffffff8021547c>] c0_compare_interrupt+0x74/0xa0 [<ffffffff80296bd0>] __handle_irq_event_percpu+0xa8/0x198 [<ffffffff80296cf0>] handle_irq_event_percpu+0x30/0x90 [<ffffffff8029d958>] handle_percpu_irq+0x88/0xb8 [<ffffffff80296124>] generic_handle_irq+0x44/0x60 [<ffffffff80b3cfd0>] do_IRQ+0x18/0x28 [<ffffffff8067ace4>] plat_irq_dispatch+0x64/0x100 [<ffffffff80209a20>] handle_int+0x140/0x14c [<ffffffff802402e8>] irq_exit+0xf8/0x100 Because AZX_DRIVER_GENERIC can not work well for Loongson LS7A HDA controller, it needs some workarounds which are not merged into the upstream kernel at this time, so it should revert this patch now. Fixes: 61eee4a7fc40 ("ALSA: hda: Add support for Loongson 7A1000 controller") Cc: <stable(a)vger.kernel.org> # 5.9-rc1+ Signed-off-by: Tiezhu Yang <yangtiezhu(a)loongson.cn> --- v2: update commit message sound/pci/hda/hda_intel.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index e34a4d5..0f86e37 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2745,8 +2745,6 @@ static const struct pci_device_id azx_ids[] = { .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_HDMI }, /* Zhaoxin */ { PCI_DEVICE(0x1d17, 0x3288), .driver_data = AZX_DRIVER_ZHAOXIN }, - /* Loongson */ - { PCI_DEVICE(0x0014, 0x7a07), .driver_data = AZX_DRIVER_GENERIC }, { 0, } }; MODULE_DEVICE_TABLE(pci, azx_ids); -- 2.1.0

4 4

Additional features for ALSA Loopback Device?
by info＠definiteaudio.de 08 Sep '20

08 Sep '20

Hello, I'm new here and I urgently need some functionality enhancements for the ALSA loopback device. Probably this is not the right place to drop my request but my thought was, that maybe here I can find a developer that is able and willing to realize my requests. For sure my company is willing to pay for it! Thank you for reading and I really appreciate your help. Here are the requested additional features: 1. Display a custom name for the card instead of "Loopback PCM" (should be entered as parameter in modprobe). 2. Increase max. sample rate to 384k (currently max. is 192k) 3. Currently two soundcards are installed that are interconnected vice versa. We need to have one card only where the output is interconnected to the input. Hope this can be realized. Best Dr. Carl-Werner Oehlrich definiteAudio GmbH mail: info(a)definiteaudio.de <mailto:info@definiteaudio.de>

1 0

[PATCH 0/9] ASoC: sun8i-codec driver cleanup
by Samuel Holland 07 Sep '20

07 Sep '20

Now that the fixes series is merged, here is a series of small cleanups to the sun8i-codec driver. These help shorten the patch stack for the next series, which will add support for the other two DAIs in this codec: AIF2 and AIF3. Samuel Holland (9): ASoC: sun8i-codec: Remove extraneous widgets ASoC: sun8i-codec: Fix AIF1 MODCLK widget name ASoC: sun8i-codec: Fix AIF1_ADCDAT_CTRL field names ASoC: sun8i-codec: Fix AIF1_MXR_SRC field names ASoC: sun8i-codec: Fix ADC_DIG_CTRL field name ASoC: sun8i-codec: Fix field bit number indentation ASoC: sun8i-codec: Sort masks in a consistent order ASoC: sun8i-codec: Attach the bus clock to the regmap ASoC: sun8i-codec: Manage module clock via DAPM sound/soc/sunxi/sun8i-codec.c | 104 ++++++++++++---------------------- 1 file changed, 37 insertions(+), 67 deletions(-) -- 2.26.2

3 12

[PATCH 00/16] ASoC: SOF: component UUID support for 5.10
by Kai Vehmanen 07 Sep '20

07 Sep '20

This series adds support for UUID based component identification in SOF. UUIDs provide a more scalable alternative to the old component type based approach to identify which DSP components should be loaded. More detailed description of UUID usage in SOF is available in: https://thesofproject.github.io/latest/developer_guides/uuid/ UUID support is an incremental update to the SOF IPC interface. Driver remains compatible with pre-UUID (ABI <3.17) firmware versions. Keyon Jie (16): ASoC: SOF: tokens: add token for component UUID ASoC: SOF: add comp_ext to struct snd_sof_widget ASoC: SOF: topology: create component extended tokens ASoC: SOF: topology: parse comp_ext_tokens for all widgets ASoC: SOF: use the sof_ipc_comp reserved bytes for extended data ASoC: SOF: topology: add helper for setting up IPC component ASoC: SOF: append extended data to sof_ipc_comp_dai ASoC: SOF: append extended data to sof_ipc_comp_mixer ASoC: SOF: append extended data to sof_ipc_comp_volume ASoC: SOF: append extended data to sof_ipc_comp_host ASoC: SOF: append extended data to sof_ipc_comp_src ASoC: SOF: append extended data to sof_ipc_comp_asrc ASoC: SOF: append extended data to sof_ipc_comp_tone ASoC: SOF: append extended data to sof_ipc_comp_process ASoC: SOF: append extended data to sof_ipc_comp_mux ASoC: SOF: topology: make process type optional include/sound/sof/topology.h | 12 +- include/uapi/sound/sof/tokens.h | 1 + sound/soc/sof/sof-audio.c | 23 +++- sound/soc/sof/sof-audio.h | 3 + sound/soc/sof/topology.c | 208 ++++++++++++++++++++------------ 5 files changed, 161 insertions(+), 86 deletions(-) -- 2.27.0

2 17

[PATCH] ASoC: Intel: KeemBay: Fix warning potential ! vs ~ typo
by Michael Sit Wei Hong 07 Sep '20

07 Sep '20

To set platform in slave mode setting the MASTER_MODE bit is not needed. Removing !MASTER_MODE conditional to avoid potential errors and warning. Signed-off-by: Michael Sit Wei Hong <michael.wei.hong.sit(a)intel.com> Reported-by: Dan Carpenter <dan.carpenter(a)oracle.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> --- sound/soc/intel/keembay/kmb_platform.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/intel/keembay/kmb_platform.c b/sound/soc/intel/keembay/kmb_platform.c index ca25a6e40cc9..f54b710ee1c2 100644 --- a/sound/soc/intel/keembay/kmb_platform.c +++ b/sound/soc/intel/keembay/kmb_platform.c @@ -515,7 +515,7 @@ static int kmb_dai_hw_params(struct snd_pcm_substream *substream, write_val = ((config->chan_nr / 2) << TDM_CHANNEL_CONFIG_BIT) | (config->data_width << DATA_WIDTH_CONFIG_BIT) | - !MASTER_MODE | TDM_OPERATION; + TDM_OPERATION; writel(write_val, kmb_i2s->pss_base + I2S_GEN_CFG_0); break; -- 2.17.1

2 1

[PATCH v2] ASoC: core: Do not cleanup uninitialized dais on soc_pcm_open failure
by Cezary Rojewski 07 Sep '20

07 Sep '20

Introduce for_each_rtd_dais_rollback macro which behaves exactly like for_each_codec_dais_rollback and its cpu_dais equivalent but for all dais instead. Use newly added macro to fix soc_pcm_open error path and prevent uninitialized dais from being cleaned-up. Signed-off-by: Cezary Rojewski <cezary.rojewski(a)intel.com> Fixes: 5d9fa03e6c35 ("ASoC: soc-pcm: tidyup soc_pcm_open() order") Acked-by: Liam Girdwood <liam.r.girdwood(a)linux.intel.com> Acked-by: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> --- Changes in v2: - no functional changes - appended missing Fixes tag, plus Acked-by from review include/sound/soc.h | 2 ++ sound/soc/soc-pcm.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/include/sound/soc.h b/include/sound/soc.h index b94ebff12350..09a7d8409ade 100644 --- a/include/sound/soc.h +++ b/include/sound/soc.h @@ -1193,6 +1193,8 @@ struct snd_soc_pcm_runtime { ((i) < (rtd)->num_cpus + (rtd)->num_codecs) && \ ((dai) = (rtd)->dais[i]); \ (i)++) +#define for_each_rtd_dais_rollback(rtd, i, dai) \ + for (; (--(i) >= 0) && ((dai) = (rtd)->dais[i]);) void snd_soc_close_delayed_work(struct snd_soc_pcm_runtime *rtd); diff --git a/sound/soc/soc-pcm.c b/sound/soc/soc-pcm.c index 00ac1cbf6f88..4c9d4cd8cf0b 100644 --- a/sound/soc/soc-pcm.c +++ b/sound/soc/soc-pcm.c @@ -812,7 +812,7 @@ static int soc_pcm_open(struct snd_pcm_substream *substream) return 0; config_err: - for_each_rtd_dais(rtd, i, dai) + for_each_rtd_dais_rollback(rtd, i, dai) snd_soc_dai_shutdown(dai, substream); snd_soc_link_shutdown(substream); -- 2.17.1

2 1