November 2024 - Alsa-devel - mailman.alsa-project.org

ASoC Kernel 6.1: Query regarding dynamic dai link/dapm registration during bootup
by Shalini Manjunatha (Consultant) (QUIC) 08 Nov '24

08 Nov '24

Hi All, On kernel6.1 based Qualcomm target for automotive platform based on Android U, we have a feature early services to launch certain early apps at early init of android bootup cycle. Hence in order to launch early chime sound of audio at early init of android boot up cycle, we need to load all required audio kernel modules at early init stage in order to get sound card registered. We have certain KPI goals for every early apps to get completed within lesser duration so that we can achieve final KPI goal of android boot complete as per the target specification. In this journey, while optimizing the sound card registration through machine driver probe to achieve lower KPI for audio early chime sound, we find that for early chime launch we required only selected pcm front end dai links only. Hence if we keep FE dai link structure in our machine driver with only required items for early chime, overall sound card registrations gets completed within 200msec from the time it starts, which is a very good KPI number to achieve. So we were trying out code change to split overall FE dai link structure into two, one for early chime and another for all other required dai link's, then some how we need to register early chime related dai link during early init so that sound card registration process gets completed quickly, later stage we want to dynamically load all other FE dai links. Could you please help us to suggest on how can we achieve this?. Is there any possible way out to avoid static FE dai link registration for all required ones from machine driver during machine driver probe, instead register only limited list of dai links. Later can we dynamically register the dai links in order to get PCM devices entries created based on use case. Currently we have this API devm_snd_soc_register_card() being called with snd_soc_card parameter which has all possible FE dai links populated at once. Queries: 1) can we register sound card twice during bootup, once with limited dai links and next with all required ones? 2) can we register dai links/dapm dynamically based on use case requriement? 3) If it is possible, could you share any reference from any other vendor who is able to achieve the same?. Regards, Shalini M

3 3

[PATCH RESEND v4 0/2] Add a driver for the Iron Device SMA1307 Amp
by Kiseok Jo 08 Nov '24

08 Nov '24

This adds basic audio support for the Iron Device SMA1307 amplifier Kiseok Jo (2): ASoC: dt-bindings: irondevice,sma1307: Add initial DT ASoC: sma1307: Add driver for Iron Device SMA1307 .../bindings/sound/irondevice,sma1307.yaml | 53 + sound/soc/codecs/Kconfig | 10 + sound/soc/codecs/Makefile | 2 + sound/soc/codecs/sma1307.c | 2052 +++++++++++++++++ sound/soc/codecs/sma1307.h | 444 ++++ 5 files changed, 2561 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/irondevice,sma1307.yaml create mode 100644 sound/soc/codecs/sma1307.c create mode 100644 sound/soc/codecs/sma1307.h -- 2.39.2

3 5

[GIT PULL] ASoC fixes for v6.12-rc6
by Mark Brown 08 Nov '24

08 Nov '24

The following changes since commit 2db63e92186d7201ee1cb2f5af11757c5e5a1020: wcd937x codec fixes (2024-10-29 19:18:48 +0000) are available in the Git repository at: https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git tags/asoc-fix-v6.12-rc6 for you to fetch changes up to 48b86532c10128cf50c854a90c2d5b1410f4012d: ASoC: SOF: sof-client-probes-ipc4: Set param_size extension bits (2024-11-07 15:35:52 +0000) ---------------------------------------------------------------- ASoC: Fixes for v6.12 A moderately large pile of small changes here, split fairly evenly between fixes and ID additions/quirks and all of it driver specific. ---------------------------------------------------------------- Amelie Delaunay (1): ASoC: stm32: spdifrx: fix dma channel release in stm32_spdifrx_remove Jyri Sarha (1): ASoC: SOF: sof-client-probes-ipc4: Set param_size extension bits Kuninori Morimoto (1): MAINTAINERS: Generic Sound Card section Luo Yifan (2): ASoC: stm: Prevent potential division by zero in stm32_sai_mclk_round_rate() ASoC: stm: Prevent potential division by zero in stm32_sai_get_clk_div() Markus Petri (1): ASoC: amd: yc: Support dmic on another model of Lenovo Thinkpad E14 Gen 6 Mingcong Bai (1): ASoC: amd: yc: fix internal mic on Xiaomi Book Pro 14 2022 Shenghao Ding (1): ASoC: tas2781: Add new driver version for tas2563 & tas2781 qfn chip Venkata Prasad Potturu (1): ASoC: SOF: amd: Fix for incorrect DMA ch status register offset MAINTAINERS | 9 +++++++++ sound/soc/amd/yc/acp6x-mach.c | 14 ++++++++++++++ sound/soc/codecs/tas2781-fmwlib.c | 1 + sound/soc/sof/amd/acp.c | 10 +++++++++- sound/soc/sof/sof-client-probes-ipc4.c | 1 + sound/soc/stm/stm32_sai_sub.c | 6 +++--- sound/soc/stm/stm32_spdifrx.c | 2 +- 7 files changed, 38 insertions(+), 5 deletions(-)

2 1

[PATCH RESEND v4 0/2] Add a driver for the Iron Device SMA1307 Amp
by Kiseok Jo 07 Nov '24

07 Nov '24

This adds basic audio support for the Iron Device SMA1307 amplifier Kiseok Jo (2): ASoC: dt-bindings: irondevice,sma1307: Add initial DT ASoC: sma1307: Add driver for Iron Device SMA1307 .../bindings/sound/irondevice,sma1307.yaml | 53 + sound/soc/codecs/Kconfig | 10 + sound/soc/codecs/Makefile | 2 + sound/soc/codecs/sma1307.c | 2052 +++++++++++++++++ sound/soc/codecs/sma1307.h | 444 ++++ 5 files changed, 2561 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/irondevice,sma1307.yaml create mode 100644 sound/soc/codecs/sma1307.c create mode 100644 sound/soc/codecs/sma1307.h -- 2.39.2

4 7

[PATCH v3 1/2] ASoC: dt-bindings: irondevice,sma1307: Add initial DT binding
by Ki-Seok Jo 07 Nov '24

07 Nov '24

This adds the schema binding for the Iron Device SMA1307 Amp Signed-off-by: Kiseok Jo <kiseok.jo(a)irondevice.com> --- .../bindings/sound/irondevice,sma1307.yaml | 54 ++++++++++++++++++++++ 1 file changed, 54 insertions(+) diff --git a/Documentation/devicetree/bindings/sound/irondevice,sma1307.yaml b/Documentation/devicetree/bindings/sound/irondevice,sma1307.yaml new file mode 100644 index 000000000..0bb4ee664 --- /dev/null +++ b/Documentation/devicetree/bindings/sound/irondevice,sma1307.yaml @@ -0,0 +1,54 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) %YAML 1.2 +--- +$id: http://devicetree.org/schemas/sound/irondevice,sma1307.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Iron Device SMA1307 Audio Amplifier + +maintainers: + - Kiseok Jo <kiseok.jo(a)irondevice.com> + +description: + SMA1307 boosted digital speaker amplifier + with feedback-loop. + +allOf: + - $ref: dai-common.yaml# + +properties: + compatible: + enum: + - irondevice,sma1307a + - irondevice,sma1307aq + description: + If a 'q' is added, it indicated the product is AEC-Q100 + qualified for automotive applications. SMA1307A supports + both WLCSP and QFN packages. However, SMA1307AQ only + supports the QFN package. + + reg: + maxItems: 1 + + '#sound-dai-cells': + const: 1 + +required: + - compatible + - reg + - '#sound-dai-cells' + +additionalProperties: false + +examples: + - | + i2c { + #address-cells = <1>; + #size-cells = <0>; + + amplifier@1e { + compatible = "irondevice,sma1307a"; + reg = <0x1e>; + #sound-dai-cells = <1>; + }; + }; -- 2.39.2

3 4

[PATCH v2] ASoC: SOF: amd: Fix for incorrect DMA ch status register offset
by Venkata Prasad Potturu 07 Nov '24

07 Nov '24

DMA ch status register offset change in acp7.0 platform Incorrect DMA channel status register offset check lead to firmware boot failure. [ 14.432497] snd_sof_amd_acp70 0000:c4:00.5: ------------[ DSP dump start ]------------ [ 14.432533] snd_sof_amd_acp70 0000:c4:00.5: Firmware boot failure due to timeout [ 14.432549] snd_sof_amd_acp70 0000:c4:00.5: fw_state: SOF_FW_BOOT_IN_PROGRESS (3) [ 14.432610] snd_sof_amd_acp70 0000:c4:00.5: invalid header size 0x71c41000. FW oops is bogus [ 14.432626] snd_sof_amd_acp70 0000:c4:00.5: unexpected fault 0x71c40000 trace 0x71c40000 [ 14.432642] snd_sof_amd_acp70 0000:c4:00.5: ------------[ DSP dump end ]------------ [ 14.432657] snd_sof_amd_acp70 0000:c4:00.5: error: failed to boot DSP firmware -5 [ 14.432672] snd_sof_amd_acp70 0000:c4:00.5: fw_state change: 3 -> 4 [ 14.433260] dmic-codec dmic-codec: ASoC: Unregistered DAI 'dmic-hifi' [ 14.433319] snd_sof_amd_acp70 0000:c4:00.5: fw_state change: 4 -> 0 [ 14.433358] snd_sof_amd_acp70 0000:c4:00.5: error: sof_probe_work failed err: -5 Update correct register offset for DMA ch status register. Fixes: 490be7ba2a01 ("ASoC: SOF: amd: add support for acp7.0 based platform") Signed-off-by: Venkata Prasad Potturu <venkataprasad.potturu(a)amd.com> --- sound/soc/sof/amd/acp.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/sound/soc/sof/amd/acp.c b/sound/soc/sof/amd/acp.c index de3001f5b9bb..95d4762c9d93 100644 --- a/sound/soc/sof/amd/acp.c +++ b/sound/soc/sof/amd/acp.c @@ -342,11 +342,19 @@ int acp_dma_status(struct acp_dev_data *adata, unsigned char ch) { struct snd_sof_dev *sdev = adata->dev; unsigned int val; + unsigned int acp_dma_ch_sts; int ret = 0; + switch (adata->pci_rev) { + case ACP70_PCI_ID: + acp_dma_ch_sts = ACP70_DMA_CH_STS; + break; + default: + acp_dma_ch_sts = ACP_DMA_CH_STS; + } val = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_DMA_CNTL_0 + ch * sizeof(u32)); if (val & ACP_DMA_CH_RUN) { - ret = snd_sof_dsp_read_poll_timeout(sdev, ACP_DSP_BAR, ACP_DMA_CH_STS, val, !val, + ret = snd_sof_dsp_read_poll_timeout(sdev, ACP_DSP_BAR, acp_dma_ch_sts, val, !val, ACP_REG_POLL_INTERVAL, ACP_DMA_COMPLETE_TIMEOUT_US); if (ret < 0) -- 2.39.2

2 1

[PATCH] ASoC: SOF: amd: Fix for incorrect DMA ch status register offset
by Venkata Prasad Potturu 07 Nov '24

07 Nov '24

DMA ch status register offset change in acp7.0 platform Incorrect DMA channel status register offset check lead to firmware boot failure. [ 14.432497] snd_sof_amd_acp70 0000:c4:00.5: ------------[ DSP dump start ]------------ [ 14.432533] snd_sof_amd_acp70 0000:c4:00.5: Firmware boot failure due to timeout [ 14.432549] snd_sof_amd_acp70 0000:c4:00.5: fw_state: SOF_FW_BOOT_IN_PROGRESS (3) [ 14.432610] snd_sof_amd_acp70 0000:c4:00.5: invalid header size 0x71c41000. FW oops is bogus [ 14.432626] snd_sof_amd_acp70 0000:c4:00.5: unexpected fault 0x71c40000 trace 0x71c40000 [ 14.432642] snd_sof_amd_acp70 0000:c4:00.5: ------------[ DSP dump end ]------------ [ 14.432657] snd_sof_amd_acp70 0000:c4:00.5: error: failed to boot DSP firmware -5 [ 14.432672] snd_sof_amd_acp70 0000:c4:00.5: fw_state change: 3 -> 4 [ 14.433260] dmic-codec dmic-codec: ASoC: Unregistered DAI 'dmic-hifi' [ 14.433319] snd_sof_amd_acp70 0000:c4:00.5: fw_state change: 4 -> 0 [ 14.433358] snd_sof_amd_acp70 0000:c4:00.5: error: sof_probe_work failed err: -5 Update correct register offset for DMA ch status register. Fixes: bff3c2a4062d ("ASoC: SOF: amd: add support for acp7.0 based platform") Signed-off-by: Venkata Prasad Potturu <venkataprasad.potturu(a)amd.com> --- sound/soc/sof/amd/acp.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/sound/soc/sof/amd/acp.c b/sound/soc/sof/amd/acp.c index ce10d128de25..f7814dadf3ba 100644 --- a/sound/soc/sof/amd/acp.c +++ b/sound/soc/sof/amd/acp.c @@ -342,11 +342,19 @@ int acp_dma_status(struct acp_dev_data *adata, unsigned char ch) { struct snd_sof_dev *sdev = adata->dev; unsigned int val; + unsigned int acp_dma_ch_sts; int ret = 0; + switch (adata->pci_rev) { + case ACP70_PCI_ID: + acp_dma_ch_sts = ACP70_DMA_CH_STS; + break; + default: + acp_dma_ch_sts = ACP_DMA_CH_STS; + } val = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_DMA_CNTL_0 + ch * sizeof(u32)); if (val & ACP_DMA_CH_RUN) { - ret = snd_sof_dsp_read_poll_timeout(sdev, ACP_DSP_BAR, ACP_DMA_CH_STS, val, !val, + ret = snd_sof_dsp_read_poll_timeout(sdev, ACP_DSP_BAR, acp_dma_ch_sts, val, !val, ACP_REG_POLL_INTERVAL, ACP_DMA_COMPLETE_TIMEOUT_US); if (ret < 0) -- 2.39.2

3 3

topology: Fix strcat() to uninitialized memory in add_define()
by GitHub pull_request - opened 07 Nov '24

07 Nov '24

alsa-project/alsa-utils pull request #277 was opened from rfvirgil: If *defs == NULL the realloc() will be a simple malloc() and return a pointer to uninitialized memory. Doing a strcat() to this is bad, so in that case strcpy() the string into the newly allocated buffer. This fixes the buffer overflow bug, such as: *** buffer overflow detected ***: terminated Program received signal SIGABRT, Aborted. __pthread_kill_implementation (no_tid=0, signo=6, threadid=<optimized out>) at ./nptl/pthread_kill.c:44 warning: 44 ./nptl/pthread_kill.c: No such file or directory (gdb) bt at ../sysdeps/posix/libc_fatal.c:132 at ./debug/fortify_fail.c:24 dest=dest@entry=0x5555555682c0 "hUUU\005PLATFORM=mtl,PREPROCESS_PLUGINS=nhlt,NHLT_BIN=nhlt-sof-mtl-sdw-cs42l42-l0-max98363-l2-4ch-48k-96k.bin,NUM_DMICS=4,PDM1_MIC_A_ENABLE=1,PDM1_MIC_B_ENABLE=1,DMIC0_ID=3,DMIC1_ID=4,DMIC1_ENABLE=passth"..., src=src@entry=0x7fffffffe0fa "PLATFORM=mtl,PREPROCESS_PLUGINS=nhlt,NHLT_BIN=nhlt-sof-mtl-sdw-cs42l42-l0-max98363-l2-4ch-48k-96k.bin,NUM_DMICS=4,PDM1_MIC_A_ENABLE=1,PDM1_MIC_B_ENABLE=1,DMIC0_ID=3,DMIC1_ID=4,DMIC1_ENABLE=passthrough"..., destlen=<optimized out>) at ./debug/strcat_chk.c:34 __src=0x7fffffffe0fa "PLATFORM=mtl,PREPROCESS_PLUGINS=nhlt,NHLT_BIN=nhlt-sof-mtl-sdw-cs42l42-l0-max98363-l2-4ch-48k-96k.bin,NUM_DMICS=4,PDM1_MIC_A_ENABLE=1,PDM1_MIC_B_ENABLE=1,DMIC0_ID=3,DMIC1_ID=4,DMIC1_ENABLE=passthrough"..., __dest=<optimized out>) at /usr/include/x86_64-linux-gnu/bits/string_fortified.h:130 d=0x7fffffffe0fa "PLATFORM=mtl,PREPROCESS_PLUGINS=nhlt,NHLT_BIN=nhlt-sof-mtl-sdw-cs42l42-l0-max98363-l2-4ch-48k-96k.bin,NUM_DMICS=4,PDM1_MIC_A_ENABLE=1,PDM1_MIC_B_ENABLE=1,DMIC0_ID=3,DMIC1_ID=4,DMIC1_ENABLE=passthrough"..., defs=<synthetic pointer>) at topology.c:400 Request URL : https://github.com/alsa-project/alsa-utils/pull/277 Patch URL : https://github.com/alsa-project/alsa-utils/pull/277.patch Repository URL: https://github.com/alsa-project/alsa-utils

1 0

[BUG] HDA Intel PCH, HDMI , HDA Intel PCH, ALC886 Digital, no IEC958 / HDMI output/passthrough to AVR
by schorpp 07 Nov '24

07 Nov '24

[BUG] HDA Intel PCH, HDMI , HDA Intel PCH, ALC886 Digital, no IEC958 / HDMI output/passthrough to AVR ALSA versions 1.0.25...current Kernel Driver versions 3.12.40...6.x Hardware: DFI Mainboard MB330-CRM https://www.dfi.com/de/product/index/273#specification Reproducible Always Steps to reproduce and full report: https://www-vdr--portal-de.translate.goog/forum/index.php?thread/136519-gel… Software crosscheck with CM106 USB Soundbox iec958: works OK. Where to start in the drivers source code to fix it? y tom

1 0

BridgeCast v2 not working properly despite V2 configuration files
by GitHub issues - edited 06 Nov '24

06 Nov '24

alsa-project/alsa-ucm-conf issue #458 was edited from acerammos: Hi everybody, I (sadly) made the firmware update for the Roland BridgeCast and it came out broken. It was working flawlessly before the update, thanks to you i had all the Output/Input channels available and i was able to assign sound sources on them individually through the system settings gui (Kmix - Plasma). Today, as a now "V2", it is over. The BridgeCast seems to be assigned to the new conf files with its new usb tag, to `BridgeCastV2-Hifi.conf` and not former the `BridgeCast-Hifi.conf`. It seems to be a good start, thanks to Fulmer for the merge and first testing Roland's update. But now, i have a weird mix of "auxiliary channels" instead of proper audio outputs like before : ![Copie d'écran_20241106_154428](https://github.com/user-attachments/assets/4240924b-… > _(Tell me if you need Fr>En translation on that screenshot)_ Note that "chromium" (wich is in fact a client like spotify, playing music) cannot be assigned to a specific output, because BridgeCast is now detected to be "only one channel" so the option is missing. An exemple after also enabling my speakers : ![Copie d'écran_20241106_155745](https://github.com/user-attachments/assets/d01d1a67-… > _Old behaviour allowed me to choose between all the output channels of the BridgeCast, like "Games", "Aux", "Music", Chat" or "System"_ Something else i've (maybe) noticed is the differences for the channel maps, between the `cat /proc/asound/cardX/stream0` channels output and what is written in the `Roland/BridgeCastV2-Hifi.conf`. `cat /proc/asound/cardX/stream0 :` ``` Roland BRIDGE CAST V2 at usb-0000:00:14.0-1, high speed : USB Audio Playback: Status: Running Interface = 1 Altset = 1 Packet Size = 392 Momentary freq = 48000 Hz (0x6.0000) Interface 1 Altset 1 Format: S32_LE Channels: 14 Endpoint: 0x0d (13 OUT) (ASYNC) Rates: 48000 Data packet interval: 125 us Bits: 24 Channel map: FL FR FC LFE RL RR FLC FRC RC SL SR TC -- -- Sync Endpoint: 0x8e (14 IN) Sync EP Interface: 2 Sync EP Altset: 1 Implicit Feedback Mode: Yes Capture: Status: Stop Interface 2 Altset 1 Format: S32_LE Channels: 6 Endpoint: 0x8e (14 IN) (ASYNC) Rates: 48000 Data packet interval: 125 us Bits: 24 Channel map: FL FR FC LFE RL RR Sync Endpoint: 0x0d (13 OUT) Sync EP Interface: 1 Sync EP Altset: 1 Implicit Feedback Mode: Yes ``` `BridgeCastV2-Hifi.conf :` ``` Macro [ { SplitPCM { Name "bc_stereo_out" Direction Playback Channels 2 HWChannels 14 HWChannelPos0 FL HWChannelPos1 FR HWChannelPos2 FL HWChannelPos3 FR HWChannelPos4 FL HWChannelPos5 FR HWChannelPos6 FL HWChannelPos7 FR HWChannelPos8 FL HWChannelPos9 FR HWChannelPos10 FL HWChannelPos11 FR HWChannelPos12 FL HWChannelPos13 FR } } { SplitPCM { Name "bc_stereo_in" Direction Capture Channels 2 HWChannels 6 HWChannelPos0 FL HWChannelPos1 FR HWChannelPos2 FL HWChannelPos3 FR HWChannelPos4 FL HWChannelPos5 FR } } ] ``` I'm completely new to this kind of code/linux driver wizardry so i haven't been able to gather more data, but i'm not afraid to be messing around, so if i can be of any help, screenshot, command outputs, random file testing, i'll gladly help. Regards, and thank you all ! > _Edits : Spelling/Grammar mistakes_ Issue URL : https://github.com/alsa-project/alsa-ucm-conf/issues/458 Repository URL: https://github.com/alsa-project/alsa-ucm-conf

1 0