March 2025 - Alsa-devel - mailman.alsa-project.org

HDSPconf missing setting...
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-tools issue #26 was edited from juanpc2018: $ alsamixer v1.2.2 ![Image](https://github.com/user-attachments/assets/6b2860d4-73a0-4526-9a97-13a7c5d6785a) ![Image](https://github.com/user-attachments/assets/ec8253d0-627a-4dda-a45e-1a2080d20563) HDSPconf 1.4 from alsa-tools-gui ![Image](https://github.com/user-attachments/assets/d2d1dba5-8da6-43ae-a594-cfd89bdd7e1f) ![Image](https://github.com/user-attachments/assets/17d070d3-067e-4d1b-adf9-131f843a634c) i dont know why iis called DDS Sample Offset but it should be a slider similar to HDSPmixer ![Image](https://github.com/user-attachments/assets/3ac4ca5c-df61-4cb3-8e8d-2af8fc44019c) DDS "Sample Offsett", is similar to a PLL with a more aggressive rejection of the external clock signal. IF the external clock coming from s/pdif is better, vs. RME DDS 200pico second jitter clock, worsens the clock. like a blend control: 0% preference for internal DDS clock = 100% preference for external clock, 100% preference for internal DDS clock = 0% preference for external clock. Issue URL : https://github.com/alsa-project/alsa-tools/issues/26 Repository URL: https://github.com/alsa-project/alsa-tools

1 0

HDSPconf missing setting...
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-tools issue #26 was edited from juanpc2018: $ alsamixer 1.2.2. ![Image](https://github.com/user-attachments/assets/6b2860d4-73a0-4526-9a97-13a7c5d6785a) ![Image](https://github.com/user-attachments/assets/ec8253d0-627a-4dda-a45e-1a2080d20563) HDSPconf 1.4 from alsa-tools-gui ![Image](https://github.com/user-attachments/assets/d2d1dba5-8da6-43ae-a594-cfd89bdd7e1f) ![Image](https://github.com/user-attachments/assets/17d070d3-067e-4d1b-adf9-131f843a634c) i dont know why iis called DDS Sample Offset but it should be a slider similar to HDSPmixer ![Image](https://github.com/user-attachments/assets/3ac4ca5c-df61-4cb3-8e8d-2af8fc44019c) DDS "Sample Offsett", is similar to a PLL with a more aggressive rejection of the external clock signal. IF the external clock coming from s/pdif is better, vs. RME DDS 200pico second jitter clock, it worsens the clock. acts like a blend control: 0% preference for internal DDS clock = 100% preference for external clock, 100% preference for internal DDS clock = 0% preference for external clock. Issue URL : https://github.com/alsa-project/alsa-tools/issues/26 Repository URL: https://github.com/alsa-project/alsa-tools

1 0

HDSPconf missing setting...
by GitHub issues - opened 06 Mar '25

06 Mar '25

alsa-project/alsa-tools issue #26 was opened from juanpc2018: $ alsamixer 1.2.2. ![Image](https://github.com/user-attachments/assets/6b2860d4-73a0-4526-9a97-13a7c5d6785a) ![Image](https://github.com/user-attachments/assets/ec8253d0-627a-4dda-a45e-1a2080d20563) HDSPconf 1.4 from alsa-tools-gui ![Image](https://github.com/user-attachments/assets/d2d1dba5-8da6-43ae-a594-cfd89bdd7e1f) ![Image](https://github.com/user-attachments/assets/17d070d3-067e-4d1b-adf9-131f843a634c) i dont know why iis called DDS Sample Offset should be a slider similar to HDSPmixer ![Image](https://github.com/user-attachments/assets/3ac4ca5c-df61-4cb3-8e8d-2af8fc44019c) DDS "Sample Offsett", is similar to a PLL with a more aggressive rejection of the external clock signal. IF the external clock coming from s/pdif is better, vs. RME DDS 200pico second jitter clock, it worsens the clock. acts like a blend control: 0% preference for internal DDS clock = 100% preference for external clock, 100% preference for internal DDS clock = 0% preference for external clock. Issue URL : https://github.com/alsa-project/alsa-tools/issues/26 Repository URL: https://github.com/alsa-project/alsa-tools

1 0

Midi Ports are ambiguously named after some alsa update, should be distinct.
by GitHub issues - opened 06 Mar '25

06 Mar '25

alsa-project/alsa-lib issue #442 was opened from Thrilled-Oaf: Greetings~. I'm on Slackware, which is a pretty languid, laid back distro... so I have to admit that I don't know when this happened... But I recently updated it, cause I bought a new midi keyboard, a launchkey mk4, and it should have a 'DAW' port, both on its USB connection and the classic pinned midi cable thing. But instead, aconnect -i and all that just gives me ``` client 24: 'Launchkey MK4 61' [type=kernel,card=2] 0 'Launchkey MK4 61 Launchkey MK4 ' 1 'Launchkey MK4 61 Launchkey MK4 ' ``` I actually use JACK audio on top of alsa but even without jack, this still occurs. I can sort of pretty print this for JACK but it would be best to have this working properly at the very base... cause if I disable all alsa seq/raw support from JACK, then DAWs like ardour won't even see any midi ports anymore. I also tried being 'clever' and loading `snd-seq-dummy` and setting up some routing that way... didn't work. Then I actually edited the source file for the dummy device to specifically give me a properly named DAW port if the index of the dummy port was like, 3 or something. And while the naming worked, it was ultimately ...not able to help out with this issue. What is basically assumed by many clients, including the configuration utility `components` for the keyboard is that it has a distinguishable second midi port with a DAW suffix, thus...this is what `aconnect -i` should ideally show. ``` client 24: 'Launchkey MK4 61' [type=kernel,card=2] 0 'Launchkey MK4 61 Launchkey MK4 ' 1 'Launchkey MK4 61 Launchkey MK4 DAW' ``` I think I read an email somewhere that alsa might be running into a name length limit, but ...perhaps there's a way to configure this? If not through a configuration file, then maybe somewhere in the source itself...? I am not shy about that but I would need directions...100% lol. I also looked up other issues here and some talk about portaudio and mixxx and stuff but this is definitely a system wide issue for me. And so sad too, it's a text descriptor... a text label of all things... bringing the mighty ALSA to its knees. A measly, peasly, text label for a little bittle MIDI port... I tried udev too but ...that was utterly fruitless due to my lack of knowledge/skill... but this all should signal that I'm willing to put in some work, too. I just don't know what to do anymore. maybe .asoundrc but all the examples are about PCM and my PCM cards work fine, it's just this MIDI ambiguity thing... Kernel: ` 6.12.17 #1 SMP PREEMPT_DYNAMIC Thu Feb 27 15:22:43 CST 2025 x86_64 AMD Ryzen 7 2700 Eight-Core Processor AuthenticAMD GNU/Linux` All the alsa related packages on my system (many are third party but compiled by me) ``` alsa-lib-1.2.13-x86_64-2 alsa-oss-1.1.8-x86_64-3 alsa-plugin-jack-1.2.5-x86_64-1SBo-git alsa-plugins-1.2.12-x86_64-2 alsa-plugins-compat32-1.2.12-x86_64-2compat32 alsa-tools-1.0.28-x86_64-1ponce alsa-utils-1.2.13-x86_64-3 alsamixergui-0.9.0rc2-x86_64-2_SBo bluez-alsa-3.1.0-x86_64-2ponce zita-alsa-pcmi-0.3.2-x86_64-1alien ``` Issue URL : https://github.com/alsa-project/alsa-lib/issues/442 Repository URL: https://github.com/alsa-project/alsa-lib

1 0

RE: [PATCH v4 1/2] ASoC: codecs: add support for ES8389
by Zhang Yi 06 Mar '25

06 Mar '25

> > > set_tdm_slot() > > > We will register multiple codecs inside a single dai_link and differentiate these > > codecs by of_node. And the adc_slot and the dac_slot may be different, they can > > be decided by the user.If we use set_tdm_slot,the adc_slot and the dac_slot will > > be same. > > No, the machine driver should be configuring different TDM slots for > each device - that's the whole point of the API. We are using multiple codecs as a single device.So we can't use set_tdm_slot to configure different slots for multiple codecs under one device. > > > > + ret = clk_prepare_enable(es8389->mclk); > > > > + if (ret) { > > > > + dev_err(codec->dev, "%s, unable to enable mclk\n", __func__); > > > > + return ret; > > > > + } > > > > + } > > > > Making the use of a MCLK depend on the configuration of a TDM slot for > > > the ADC seems *very* unusual, what's going on there? > > > Because we are associating multiple codecs under a single dai_link, we will be > > executing probe and init many times during initialization.But MCLK only needs > > to be used once.So we decided making the use of a MCLK depend on the configuration > > of a TDM slot for the ADC > > No, each device should just get and enable the MCLK itself - the clock > API does reference counting so there's no problem with this, it's normal > for a clok to have multiple consumers. ok,I'll fix it

2 1

regresion: snd_hdsp
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was edited from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like [stutter](https://www.youtube.com/watch?v=9ZaDn5wne8Y&t=609s).[2](https://ww… i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, APIC ACPI Tables v2 & v3 in Bios, Audacity 3.6. ALSA hw:0 also Crash. but Jack Audio works ok. i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. ``` $ sudo apt show alsa-firmware* Package: alsa-firmware-loaders Version: 1.1.7-1ubuntu1 Priority: extra Section: multiverse/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 211 kB Depends: fxload, udev, libasound2 (>= 1.0.24.1), libc6 (>= 2.4) Homepage: https://www.alsa-project.org/ Download-Size: 25,0 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages Description: ALSA software loaders for specific hardware $ sudo apt show alsa-tools-gui Package: alsa-tools-gui Version: 1.1.7-1ubuntu1 Priority: extra Section: universe/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 1.002 kB Depends: libasound2 (>= 1.0.24.1), libc6 (>= 2.29), libfltk1.1 (>= 1.1.6), libglib2.0-0 (>= 2.35.9), libgtk-3-0 (>= 3.0.0), libgtk2.0-0 (>= 2.8.0), libstdc++6 (>= 5.2) Homepage: https://www.alsa-project.org/ Task: ubuntustudio-audio Download-Size: 220 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages Description: GUI based ALSA utilities for specific hardware ``` Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

regresion: snd_hdsp
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was edited from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like [stutter](https://www.youtube.com/watch?v=9ZaDn5wne8Y&t=609s).[2](https://ww… i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, Audacity 3.6. ALSA hw:0 also Crash. but Jack Audio works ok. i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. ``` $ sudo apt show alsa-firmware* Package: alsa-firmware-loaders Version: 1.1.7-1ubuntu1 Priority: extra Section: multiverse/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 211 kB Depends: fxload, udev, libasound2 (>= 1.0.24.1), libc6 (>= 2.4) Homepage: https://www.alsa-project.org/ Download-Size: 25,0 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages Description: ALSA software loaders for specific hardware $ sudo apt show alsa-tools-gui Package: alsa-tools-gui Version: 1.1.7-1ubuntu1 Priority: extra Section: universe/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 1.002 kB Depends: libasound2 (>= 1.0.24.1), libc6 (>= 2.29), libfltk1.1 (>= 1.1.6), libglib2.0-0 (>= 2.35.9), libgtk-3-0 (>= 3.0.0), libgtk2.0-0 (>= 2.8.0), libstdc++6 (>= 5.2) Homepage: https://www.alsa-project.org/ Task: ubuntustudio-audio Download-Size: 220 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages Description: GUI based ALSA utilities for specific hardware ``` Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

regresion: snd_hdsp
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was edited from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like [stutter](https://www.youtube.com/watch?v=9ZaDn5wne8Y&t=609s).[2](https://ww… i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, Audacity 3.6. ALSA hw:0 also Crash. i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. ``` $ sudo apt show alsa-firmware* Package: alsa-firmware-loaders Version: 1.1.7-1ubuntu1 Priority: extra Section: multiverse/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 211 kB Depends: fxload, udev, libasound2 (>= 1.0.24.1), libc6 (>= 2.4) Homepage: https://www.alsa-project.org/ Download-Size: 25,0 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages Description: ALSA software loaders for specific hardware $ sudo apt show alsa-tools-gui Package: alsa-tools-gui Version: 1.1.7-1ubuntu1 Priority: extra Section: universe/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 1.002 kB Depends: libasound2 (>= 1.0.24.1), libc6 (>= 2.29), libfltk1.1 (>= 1.1.6), libglib2.0-0 (>= 2.35.9), libgtk-3-0 (>= 3.0.0), libgtk2.0-0 (>= 2.8.0), libstdc++6 (>= 5.2) Homepage: https://www.alsa-project.org/ Task: ubuntustudio-audio Download-Size: 220 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages Description: GUI based ALSA utilities for specific hardware ``` Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

regresion: snd_hdsp
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was edited from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like [stutter](https://www.youtube.com/watch?v=9ZaDn5wne8Y&t=609s).[2](https://ww… i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, Audacity 3.6. ALSA also crash. i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. ``` $ sudo apt show alsa-firmware* Package: alsa-firmware-loaders Version: 1.1.7-1ubuntu1 Priority: extra Section: multiverse/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 211 kB Depends: fxload, udev, libasound2 (>= 1.0.24.1), libc6 (>= 2.4) Homepage: https://www.alsa-project.org/ Download-Size: 25,0 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages Description: ALSA software loaders for specific hardware $ sudo apt show alsa-tools-gui Package: alsa-tools-gui Version: 1.1.7-1ubuntu1 Priority: extra Section: universe/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 1.002 kB Depends: libasound2 (>= 1.0.24.1), libc6 (>= 2.29), libfltk1.1 (>= 1.1.6), libglib2.0-0 (>= 2.35.9), libgtk-3-0 (>= 3.0.0), libgtk2.0-0 (>= 2.8.0), libstdc++6 (>= 5.2) Homepage: https://www.alsa-project.org/ Task: ubuntustudio-audio Download-Size: 220 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages Description: GUI based ALSA utilities for specific hardware ``` Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

regresion: snd_hdsp
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was edited from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like [stutter](https://www.youtube.com/watch?v=9ZaDn5wne8Y&t=609s).[2](https://ww… i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. ``` $ sudo apt show alsa-firmware* Package: alsa-firmware-loaders Version: 1.1.7-1ubuntu1 Priority: extra Section: multiverse/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 211 kB Depends: fxload, udev, libasound2 (>= 1.0.24.1), libc6 (>= 2.4) Homepage: https://www.alsa-project.org/ Download-Size: 25,0 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages Description: ALSA software loaders for specific hardware $ sudo apt show alsa-tools-gui Package: alsa-tools-gui Version: 1.1.7-1ubuntu1 Priority: extra Section: universe/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 1.002 kB Depends: libasound2 (>= 1.0.24.1), libc6 (>= 2.29), libfltk1.1 (>= 1.1.6), libglib2.0-0 (>= 2.35.9), libgtk-3-0 (>= 3.0.0), libgtk2.0-0 (>= 2.8.0), libstdc++6 (>= 5.2) Homepage: https://www.alsa-project.org/ Task: ubuntustudio-audio Download-Size: 220 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages Description: GUI based ALSA utilities for specific hardware ``` Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

regresion: snd_hdsp
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was edited from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like [stutter](https://www.youtube.com/watch?v=9ZaDn5wne8Y&t=609s) i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. ``` $ sudo apt show alsa-firmware* Package: alsa-firmware-loaders Version: 1.1.7-1ubuntu1 Priority: extra Section: multiverse/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 211 kB Depends: fxload, udev, libasound2 (>= 1.0.24.1), libc6 (>= 2.4) Homepage: https://www.alsa-project.org/ Download-Size: 25,0 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages Description: ALSA software loaders for specific hardware $ sudo apt show alsa-tools-gui Package: alsa-tools-gui Version: 1.1.7-1ubuntu1 Priority: extra Section: universe/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 1.002 kB Depends: libasound2 (>= 1.0.24.1), libc6 (>= 2.29), libfltk1.1 (>= 1.1.6), libglib2.0-0 (>= 2.35.9), libgtk-3-0 (>= 3.0.0), libgtk2.0-0 (>= 2.8.0), libstdc++6 (>= 5.2) Homepage: https://www.alsa-project.org/ Task: ubuntustudio-audio Download-Size: 220 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages Description: GUI based ALSA utilities for specific hardware ``` Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

regresion: snd_hdsp
by GitHub issues - edited 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was edited from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like stutter. i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. ``` $ sudo apt show alsa-firmware* Package: alsa-firmware-loaders Version: 1.1.7-1ubuntu1 Priority: extra Section: multiverse/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 211 kB Depends: fxload, udev, libasound2 (>= 1.0.24.1), libc6 (>= 2.4) Homepage: https://www.alsa-project.org/ Download-Size: 25,0 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages Description: ALSA software loaders for specific hardware $ sudo apt show alsa-tools-gui Package: alsa-tools-gui Version: 1.1.7-1ubuntu1 Priority: extra Section: universe/sound Source: alsa-tools Origin: Ubuntu Maintainer: Ubuntu Developers <ubuntu-devel-discuss(a)lists.ubuntu.com> Original-Maintainer: Debian ALSA Maintainers <pkg-alsa-devel(a)lists.alioth.debian.org> Bugs: https://bugs.launchpad.net/ubuntu/+filebug Installed-Size: 1.002 kB Depends: libasound2 (>= 1.0.24.1), libc6 (>= 2.29), libfltk1.1 (>= 1.1.6), libglib2.0-0 (>= 2.35.9), libgtk-3-0 (>= 3.0.0), libgtk2.0-0 (>= 2.8.0), libstdc++6 (>= 5.2) Homepage: https://www.alsa-project.org/ Task: ubuntustudio-audio Download-Size: 220 kB APT-Manual-Installed: yes APT-Sources: http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages Description: GUI based ALSA utilities for specific hardware ``` Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

regresion: snd_hdsp
by GitHub issues - opened 06 Mar '25

06 Mar '25

alsa-project/alsa-firmware issue #11 was opened from juanpc2018: ![Image](https://github.com/user-attachments/assets/e57192a7-ad1c-44f9-b51d-7806efed9a32) ``` $ pulseaudio W: [pulseaudio] module.c: module-combine is deprecated: Please use module-combine-sink instead of module-combine! W: [pulseaudio] module-combine.c: We will now load module-combine-sink. Please make sure to remove module-combine from your configuration. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hdsp'. Please report this issue to the ALSA developers. E: [alsa-sink-RME Hammerfall HDSP 9632] alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail. ``` when i play any video or sound file, most of the time PulseAudio takes between 7 to 15 seconds to fill the buffer in proper order, audio is played in wrong order, chopped, like stutter. i have tested /etc/pulse/deamon.conf s24le to s24be problem persists. forced sample rate to 48Khz, i remember hdsp9632 worked ok on U/Kubuntu 10.04 & 10.10 Now 20.04.4 LTS has problems. im using Liquorix 6.3 Kernel. Issue URL : https://github.com/alsa-project/alsa-firmware/issues/11 Repository URL: https://github.com/alsa-project/alsa-firmware

1 0

XRUN handling in pcm driver
by noman pouigt 05 Mar '25

05 Mar '25

Hello Maintainers, I have a question regarding XRUN handling. After an XRUN is triggered, userspace generally calls TRIGGER_STOP followed by PREPARE and TRIGGER_START. In the PREPARE stage, we are currently not sending the ALSA ring buffer (physical address, rate, format, etc.) IPC to the DSP but instead doing that in HW_PARAMS which is causing bad audio after XRUN is triggered. I was able to resolve the issue by moving the IPC to PREPARE, which fixes the XRUN problem and results in good audio even after an XRUN. I have also observed a similar approach in Qualcomm drivers, which I assume is for the same reason. Is moving the IPC to PREPARE the correct approach for handling this issue? I would appreciate any insights or guidance on whether this is the recommended way to handle XRUN recovery. Thanks,

2 1

RE: [PATCH v4 1/2] ASoC: codecs: add support for ES8389
by Zhang Yi 05 Mar '25

05 Mar '25

> > + ret = device_property_read_u8(codec->dev, "everest,adc-slot", &es8389->adc_slot); > > + if (ret != 0) { > > + dev_dbg(codec->dev, "adc-slot return %d", ret); > > + es8389->adc_slot = 0x00; > > + } > > + > > + ret = device_property_read_u8(codec->dev, "everest,dac-slot", &es8389->dac_slot); > > + if (ret != 0) { > > + dev_dbg(codec->dev, "dac-slot return %d", ret); > > + es8389->dac_slot = 0x00; > > + } > > set_tdm_slot() > > Please don't ignore review comments, people are generally making them > for a reason and are likely to have the same concerns if issues remain > unaddressed. Having to repeat the same comments can get repetitive and > make people question the value of time spent reviewing. If you disagree > with the review comments that's fine but you need to reply and discuss > your concerns so that the reviewer can understand your decisions. We will register multiple codecs inside a single dai_link and differentiate these codecs by of_node. And the adc_slot and the dac_slot may be different, they can be decided by the user.If we use set_tdm_slot,the adc_slot and the dac_slot will be same. > > + if (!es8389->adc_slot) { > > + es8389->mclk = devm_clk_get(codec->dev, "mclk"); > > + if (IS_ERR(es8389->mclk)) > > + return dev_err_probe(codec->dev, PTR_ERR(es8389->mclk), > > + "ES8389 is unable to get mclk\n"); > > + > > + if (!es8389->mclk) > > + dev_err(codec->dev, "%s, assuming static mclk\n", __func__); > > + > > + ret = clk_prepare_enable(es8389->mclk); > > + if (ret) { > > + dev_err(codec->dev, "%s, unable to enable mclk\n", __func__); > > + return ret; > > + } > > + } > > Making the use of a MCLK depend on the configuration of a TDM slot for > the ADC seems *very* unusual, what's going on there? Because we are associating multiple codecs under a single dai_link, we will be executing probe and init many times during initialization.But MCLK only needs to be used once.So we decided making the use of a MCLK depend on the configuration of a TDM slot for the ADC

2 1

update ALC222 depop optimize
by Kailang 05 Mar '25

05 Mar '25

Hi Takashi, Update ALC222 depop optimize as attach. BR, Kailang

2 8

[PATCH] ALSA: hda: intel: Add Dell ALC3271 to power_save denylist
by Hoku Ishibe 05 Mar '25

05 Mar '25

Dell XPS 13 7390 with the Realtek ALC3271 codec experiences persistent humming noise when the power_save mode is enabled. This issue occurs when the codec enters power saving mode, leading to unwanted noise from the speakers. This patch adds the affected model (PCI ID 0x1028:0x0962) to the power_save denylist to ensure power_save is disabled by default, preventing power-off related noise issues. Steps to Reproduce 1. Boot the system with `snd_hda_intel` loaded. 2. Verify that `power_save` mode is enabled: ```sh cat /sys/module/snd_hda_intel/parameters/power_save ```` output: 10 (default power save timeout) 3. Wait for the power save timeout 4. Observe a persistent humming noise from the speakers 5. Disable `power_save` manually: ```sh echo 0 | sudo tee /sys/module/snd_hda_intel/parameters/power_save ```` 6. Confirm that the noise disappears immediately. This issue has been observed on my system, and this patch successfully eliminates the unwanted noise. If other users experience similar issues, additional reports would be helpful. Signed-off-by: Hoku Ishibe <me(a)hokuishi.be> --- sound/pci/hda/hda_intel.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 7d7f9aac50a9..ff536a0e480c 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2232,6 +2232,8 @@ static const struct snd_pci_quirk power_save_denylist[] = { SND_PCI_QUIRK(0x1631, 0xe017, "Packard Bell NEC IMEDIA 5204", 0), /* KONTRON SinglePC may cause a stall at runtime resume */ SND_PCI_QUIRK(0x1734, 0x1232, "KONTRON SinglePC", 0), + /* Dell ALC3271 */ + SND_PCI_QUIRK(0x1028, 0x0962, "Dell ALC3271", 0), {} }; -- 2.48.1

2 1

RE: [PATCH v4 1/2] ASoC: codecs: add support for ES8389
by Zhang Yi 05 Mar '25

05 Mar '25

> > @@ -0,0 +1,961 @@ > > +// SPDX-License-Identifier: GPL-2.0-only > > +/* > > + * es8389.c -- ES8389/ES8390 ALSA SoC Audio Codec > > + * > > + * Copyright (C) 2025 Everest Semiconductor Co., Ltd > > Please make the entire comment block a C++ one so things look more > consistent. I will fix it > > + if (es8389->dmic == true) { > > + regmap_update_bits(es8389->regmap, ES8389_DMIC_EN, 0xC0, 0xC0); > > + regmap_update_bits(es8389->regmap, ES8389_ADC_MODE, 0x03, 0x03); > > + } else { > > + regmap_update_bits(es8389->regmap, ES8389_DMIC_EN, 0xC0, 0x00); > > + regmap_update_bits(es8389->regmap, ES8389_ADC_MODE, 0x03, 0x00); > > + } > > We also had the DMIC mux, is that useful as a runtime control when we > have firmware data telling us if there's a DMIC? Can both a DMIC and > analog input be present in the same system? > > It does still look like a lot of these settings might be things that > should be user controllable... I'm going to remove es8389->dmic and everest,dmic-enabled and use only DMIC_MUX

1 0

RE: [PATCH v3 2/2] ASoC: dt-bindings: Add Everest ES8389 audio CODEC
by Zhang Yi 05 Mar '25

05 Mar '25

> > > >>> + > >>> + prefix_name: > >> > >> No underscores in node names, missing vendor prefix... but more > >> important: I don't understand the need for this property. Description copies property name so it is not useful. > > And these comments as well. The prefix_name is used to distinguish the dapm and kcontrol of each individual codec. And I will update name of the property > > > >>> + everest,dmic-enabled: > >>> + $ref: /schemas/types.yaml#/definitions/flag > >>> + description: > >>> + The property is a choice between PDM and AMIC > >>> + > >> > >> No supplies? > > > > I will drop the property > > I did not comment about dmic, but meant missing power supplies, which > usually are placed here. I apologize for misunderstanding your comment. But we don't have properties about the power supplies

1 0

RE: [PATCH v3 2/2] ASoC: dt-bindings: Add Everest ES8389 audio CODEC
by Zhang Yi 05 Mar '25

05 Mar '25

> > + > > +properties: > > + compatible: > > + const: everest,es8389 > > Device is really different than es8388? yes, it's different from es8388 > > + "#sound-dai-cells": > > + const: 0 > > + > > + everest,adc-slot: > > + $ref: /schemas/types.yaml#/definitions/uint8 > > + description: | > > + This property is used to set the slots of recording data when multiple > > + codecs are connected in PTDM mode. > > + please set this property to default if you are setting STDM mode. > > + minimum: 0x00 > > Drop, unsigned does not allow -1. > > > + maximum: 0x07 > > + default: 0x00 > > All of these should be rather decimal. I will try to fix the error > > + > > + prefix_name: > > No underscores in node names, missing vendor prefix... but more > important: I don't understand the need for this property. Description copies property name so it is not useful. > > Drop. You maybe wanted dai prefix, but this is already in dai-common. I will update description at v4 PATCH > > + everest,dmic-enabled: > > + $ref: /schemas/types.yaml#/definitions/flag > > + description: > > + The property is a choice between PDM and AMIC > > + > > No supplies? I will drop the property > > +required: > > + - compatible > > + - reg > > + - "#sound-dai-cells" > > + > > +additionalProperties: false > > + > > +examples: > > + - | > > + i2c{ > > Missing space. > > > + #address-cells = <1>; > > + #size-cells = <0>; > > + es8389: es8389@10 { > > Node names should be generic. See also an explanation and list of examples (not exhaustive) in DT specification: > https://devicetree-specification.readthedocs.io/en/latest/chapter2-devicetr… > > > + compatible = "everest,es8389"; > > + reg = <0x10>; > > + everest,adc-slot = [00]; > > Please open other bindings or DTS and take a look how single number is encoded. In general, please rather base your work on latest bindings and DTS accepted by reviewers and use them as learning/template to avoid common mistakes. At least two comments in this review could be avoided if you did this. > > > + everest,dac-slot = [00]; > > + prefix_name = "es8389_0"; > > Drop I will try to fix it

2 1

[PATCH v4 0/2] ASoC: codecs: add support for ES8389
by Zhang Yi 04 Mar '25

04 Mar '25

The driver is for codec ES8389 of everest-semi. Zhang Yi (2): ASoC: codecs: add support for ES8389 ASoC: dt-bindings: Add Everest ES8389 audio CODEC .../bindings/sound/everest,es8389.yaml | 83 ++ sound/soc/codecs/Kconfig | 6 +- sound/soc/codecs/Makefile | 2 + sound/soc/codecs/es8389.c | 961 ++++++++++++++++++ sound/soc/codecs/es8389.h | 139 +++ 5 files changed, 1190 insertions(+), 1 deletion(-) create mode 100644 Documentation/devicetree/bindings/sound/everest,es8389.yaml create mode 100644 sound/soc/codecs/es8389.c create mode 100644 sound/soc/codecs/es8389.h -- 2.17.1

3 4

[PATCH v3 0/2] ASoC: codecs: add support for ES8389
by Zhang Yi 04 Mar '25

04 Mar '25

The driver is for codec ES8389 of everest-semi. change in v3: - Added reg_format at everest,es8389.yaml Zhang Yi (2): ASoC: codecs: add support for ES8389 ASoC: dt-bindings: Add Everest ES8389 audio CODEC .../bindings/sound/everest,es8389.yaml | 82 ++ sound/soc/codecs/Kconfig | 6 +- sound/soc/codecs/Makefile | 2 + sound/soc/codecs/es8389.c | 971 ++++++++++++++++++ sound/soc/codecs/es8389.h | 139 +++ 5 files changed, 1199 insertions(+), 1 deletion(-) create mode 100644 Documentation/devicetree/bindings/sound/everest,es8389.yaml create mode 100644 sound/soc/codecs/es8389.c create mode 100644 sound/soc/codecs/es8389.h -- 2.17.1

2 4

[PATCH v2 0/2] ASoC: codecs: add support for ES8389
by Zhang Yi 04 Mar '25

04 Mar '25

The driver is for codec ES8389 of everest-semi. Zhang Yi (2): ASoC: codecs: add support for ES8389 ASoC: dt-bindings: Add Everest ES8389 audio CODEC .../bindings/sound/everest,es8389.yaml | 78 ++ sound/soc/codecs/Kconfig | 6 +- sound/soc/codecs/Makefile | 2 + sound/soc/codecs/es8389.c | 971 ++++++++++++++++++ sound/soc/codecs/es8389.h | 139 +++ 5 files changed, 1195 insertions(+), 1 deletion(-) create mode 100644 Documentation/devicetree/bindings/sound/everest,es8389.yaml create mode 100644 sound/soc/codecs/es8389.c create mode 100644 sound/soc/codecs/es8389.h -- 2.17.1

2 3

[PATCH v5] ASoC: tas2781: Support dsp firmware Alpha and Beta seaies
by Shenghao Ding 03 Mar '25

03 Mar '25

For calibration, basic version does not contain any calibration addresses, it depends on calibration tool to convey the addresses to the driver. Since Alpha and Beta firmware, all the calibration addresses are saved into the firmware. Signed-off-by: Shenghao Ding <shenghao-ding(a)ti.com> --- v5: - Instead of the lengthy approach, just replace if-continue with else if, has reduced this by 2/3 of LoCs - Use proper error code in fw_parse_fct_param_address() - return directly in tasdevice_dspfw_ready() - drop Unneeded assignment v4: - drop unnecceary "unsigned short size;" and "char *data;" in struct fct_param_address - replace strstr() with strncmp() in fct_param_address_parser() - replace hex_parse_u8() with memcpy() - replace hex_parse_u24() with direct assignment - drop hex_parse_u8() & hex_parse_u24() - Just return directly and use proper error codes and drop "goto out" and "out:" label in fw_parse_fct_param_address - fixed indentation issues in cali_reg_update - fixed the logical splits in tas2781_calib_stop_put() - return -ENOMEM if kmemdup fails in fw_parse_fct_param_address() - remove unnecessary "out:" label v3: - Merge the 4 patches into one patch v2: - | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: | https://urldefense.com/v3/__https://lore.kernel.org/oe-kbuild-all/2025 | 02160812.P7szOygN-lkp(a)intel.com/__;!!G3vK!WriwhtbQGlg6p00bcWutlhb-y9vS | pDrSu5_g9oOzzGOoABdirEARTntFzyX6Ss3UPRDxi_E-IA$ - | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: | https://urldefense.com/v3/__https://lore.kernel.org/oe-kbuild-all/2025 | 02160632.u0zxG7yD-lkp(a)intel.com/__;!!G3vK!VtEdfur3fZklvPVfHFFfFP60Ub2E | JYvmjSUZ65lJ6BsQPJDl1fjYRDqh3c6Q5y6sA70ssXDG5g$ v1: - update the year from 2024 to 2025 in the files. - Add dspbin_type to distinguish different dspbin types, such as TASDEV_BASIC, TASDEV_ALPHA, and TASDEV_BETA. - Add struct bulk_reg_val alp_cali_bckp to store the value of pilot tone threshold only for TASDEV_ALPHA & TASDEV_BETA firmware. - Add the declaration of (*fw_parse_fct_param_address)() to parse the addresses of calibration params. - Add the minimum version of TASDEV_ALPHA and TASDEV_BETA firmware. - Add the definition of struct fct_param_address. - update the year from 2024 to 2025. - release the memory to save calibration addresses in tasdevice_dsp_remove(). - Add TASDEV_ALPHA and TASDEV_BETA firmware checking code. - Add fct_param_address_parser for Alpha and Beta firmware. - Add an extra branch to handle calibration address for TASDEV_ALPHA and TASDEV_BETA firmware. --- include/sound/tas2781-dsp.h | 30 ++++- include/sound/tas2781.h | 11 ++ sound/pci/hda/tas2781_spi_fwlib.c | 8 +- sound/soc/codecs/tas2781-fmwlib.c | 190 ++++++++++++++++++++++---- sound/soc/codecs/tas2781-i2c.c | 215 ++++++++++++++++++++++-------- 5 files changed, 368 insertions(+), 86 deletions(-) diff --git a/include/sound/tas2781-dsp.h b/include/sound/tas2781-dsp.h index 3cda9da14f6d..c3a9efa73d5d 100644 --- a/include/sound/tas2781-dsp.h +++ b/include/sound/tas2781-dsp.h @@ -2,7 +2,7 @@ // // ALSA SoC Texas Instruments TAS2781 Audio Smart Amplifier // -// Copyright (C) 2022 - 2024 Texas Instruments Incorporated +// Copyright (C) 2022 - 2025 Texas Instruments Incorporated // https://www.ti.com // // The TAS2781 driver implements a flexible and configurable @@ -30,8 +30,10 @@ #define PRE_DEVICE_C 0x12 #define PRE_DEVICE_D 0x16 -#define PPC3_VERSION 0x4100 -#define PPC3_VERSION_TAS2781 0x14600 +#define PPC3_VERSION_BASE 0x4100 +#define PPC3_VERSION_TAS2781_BASIC_MIN 0x14600 +#define PPC3_VERSION_TAS2781_ALPHA_MIN 0x4a00 +#define PPC3_VERSION_TAS2781_BETA_MIN 0x19400 #define TASDEVICE_DEVICE_SUM 8 #define TASDEVICE_CONFIG_SUM 64 @@ -106,6 +108,27 @@ struct tasdevice_calibration { struct tasdevice_data dev_data; }; +struct fct_param_address { + /* Thermal data for PG 1.0 device */ + unsigned char thr[3]; + /* Thermal data for PG 2.0 device */ + unsigned char thr2[3]; + /* Pilot tone enable flag, usually the sine wave */ + unsigned char plt_flg[3]; + /* Pilot tone gain for calibration */ + unsigned char sin_gn[3]; + /* Pilot tone gain for calibration */ + unsigned char sin_gn2[3]; + /* high 32-bit of real-time spk impedance */ + unsigned char r0_reg[3]; + /* check spk connection */ + unsigned char tf_reg[3]; + /* check spk resonant frequency */ + unsigned char a1_reg[3]; + /* check spk resonant frequency */ + unsigned char a2_reg[3]; +}; + struct tasdevice_fw { struct tasdevice_dspfw_hdr fw_hdr; unsigned short nr_programs; @@ -114,6 +137,7 @@ struct tasdevice_fw { struct tasdevice_config *configs; unsigned short nr_calibrations; struct tasdevice_calibration *calibrations; + struct fct_param_address fct_par_addr; struct device *dev; }; diff --git a/include/sound/tas2781.h b/include/sound/tas2781.h index 4d9a37544354..eff011444cc8 100644 --- a/include/sound/tas2781.h +++ b/include/sound/tas2781.h @@ -113,6 +113,12 @@ enum audio_device { TAS2781, }; +enum dspbin_type { + TASDEV_BASIC, + TASDEV_ALPHA, + TASDEV_BETA, +}; + enum device_catlog_id { LENOVO = 0, OTHERS @@ -127,6 +133,7 @@ struct bulk_reg_val { struct tasdevice { struct bulk_reg_val *cali_data_backup; + struct bulk_reg_val alp_cali_bckp; struct tasdevice_fw *cali_data_fmw; unsigned int dev_addr; unsigned int err_code; @@ -171,6 +178,7 @@ struct tasdevice_priv { unsigned char dev_name[32]; const char *name_prefix; unsigned char ndev; + unsigned int dspbin_typ; unsigned int magic_num; unsigned int chip_id; unsigned int sysclk; @@ -196,6 +204,9 @@ struct tasdevice_priv { int (*fw_parse_configuration_data)(struct tasdevice_priv *tas_priv, struct tasdevice_fw *tas_fmw, const struct firmware *fmw, int offset); + int (*fw_parse_fct_param_address)(struct tasdevice_priv *tas_priv, + struct tasdevice_fw *tas_fmw, + const struct firmware *fmw, int offset); int (*tasdevice_load_block)(struct tasdevice_priv *tas_priv, struct tasdev_blk *block); diff --git a/sound/pci/hda/tas2781_spi_fwlib.c b/sound/pci/hda/tas2781_spi_fwlib.c index 0e2acbc3c900..19a693d58ee8 100644 --- a/sound/pci/hda/tas2781_spi_fwlib.c +++ b/sound/pci/hda/tas2781_spi_fwlib.c @@ -2,7 +2,7 @@ // // TAS2781 HDA SPI driver // -// Copyright 2024 Texas Instruments, Inc. +// Copyright 2024 - 2025 Texas Instruments, Inc. // // Author: Baojun Xu <baojun.xu(a)ti.com> @@ -370,10 +370,10 @@ static unsigned char map_dev_idx(struct tasdevice_fw *tas_fmw, int i, n = ARRAY_SIZE(non_ppc3_mapping_table); unsigned char dev_idx = 0; - if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781) { + if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781_BASIC_MIN) { p = (struct blktyp_devidx_map *)ppc3_tas2781_mapping_table; n = ARRAY_SIZE(ppc3_tas2781_mapping_table); - } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION) { + } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION_BASE) { p = (struct blktyp_devidx_map *)ppc3_mapping_table; n = ARRAY_SIZE(ppc3_mapping_table); } @@ -1605,7 +1605,7 @@ static int dspfw_default_callback(struct tasdevice_priv *tas_priv, int rc = 0; if (drv_ver == 0x100) { - if (ppcver >= PPC3_VERSION) { + if (ppcver >= PPC3_VERSION_BASE) { tas_priv->fw_parse_variable_header = fw_parse_variable_header_kernel; tas_priv->fw_parse_program_data = diff --git a/sound/soc/codecs/tas2781-fmwlib.c b/sound/soc/codecs/tas2781-fmwlib.c index bc2f68bd9846..13a197468193 100644 --- a/sound/soc/codecs/tas2781-fmwlib.c +++ b/sound/soc/codecs/tas2781-fmwlib.c @@ -390,10 +390,10 @@ static unsigned char map_dev_idx(struct tasdevice_fw *tas_fmw, int i, n = ARRAY_SIZE(non_ppc3_mapping_table); unsigned char dev_idx = 0; - if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781) { + if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781_BASIC_MIN) { p = (struct blktyp_devidx_map *)ppc3_tas2781_mapping_table; n = ARRAY_SIZE(ppc3_tas2781_mapping_table); - } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION) { + } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION_BASE) { p = (struct blktyp_devidx_map *)ppc3_mapping_table; n = ARRAY_SIZE(ppc3_mapping_table); } @@ -560,6 +560,124 @@ static int fw_parse_configuration_data_kernel( return offset; } +static void fct_param_address_parser(struct cali_reg *r, + struct tasdevice_fw *tas_fmw, const unsigned char *data) +{ + struct fct_param_address *p = &tas_fmw->fct_par_addr; + unsigned int i; + + /* + * Calibration parameters locations and data schema in dsp firmware. + * The number of items are flexible, but not more than 20. The dsp tool + * will reseve 20*24-byte space for fct params. In some cases, the + * number of fct param is less than 20, the data will be saved from the + * beginning, the rest part will be stuffed with zero. + * + * fct_param_num (not more than 20) + * for (i = 0; i < fct_param_num; i++) { + * Alias of fct param (20 bytes) + * Book (1 byte) + * Page (1 byte) + * Offset (1 byte) + * CoeffLength (1 byte) = 0x1 + * } + * if (20 - fct_param_num) + * 24*(20 - fct_param_num) pieces of '0' as stuffing + * + * As follow: + * umg_SsmKEGCye = Book, Page, Offset, CoeffLength + * iks_E0 = Book, Page, Offset, CoeffLength + * yep_LsqM0 = Book, Page, Offset, CoeffLength + * oyz_U0_ujx = Book, Page, Offset, CoeffLength + * iks_GC_GMgq = Book, Page, Offset, CoeffLength + * gou_Yao = Book, Page, Offset, CoeffLength + * kgd_Wsc_Qsbp = Book, Page, Offset, CoeffLength + * yec_CqseSsqs = Book, Page, Offset, CoeffLength + * iks_SogkGgog2 = Book, Page, Offset, CoeffLength + * yec_Sae_Y = Book, Page, Offset, CoeffLength + * Re_Int = Book, Page, Offset, CoeffLength + * SigFlag = Book, Page, Offset, CoeffLength + * a1_Int = Book, Page, Offset, CoeffLength + * a2_Int = Book, Page, Offset, CoeffLength + */ + for (i = 0; i < 20; i++) { + const unsigned char *dat = &data[24 * i]; + + /* + * check whether current fct param is empty. + */ + if (dat[23] != 1) + break; + + if (!strncmp(dat, "umg_SsmKEGCye", 20)) + r->pow_reg = TASDEVICE_REG(dat[20], dat[21], dat[22]); + /* high 32-bit of real-time spk impedance */ + else if (!strncmp(dat, "iks_E0", 20)) + r->r0_reg = TASDEVICE_REG(dat[20], dat[21], dat[22]); + /* inverse of real-time spk impedance */ + else if (!strncmp(dat, "yep_LsqM0", 20)) + r->invr0_reg = + TASDEVICE_REG(dat[20], dat[21], dat[22]); + /* low 32-bit of real-time spk impedance */ + else if (!strncmp(dat, "oyz_U0_ujx", 20)) + r->r0_low_reg = + TASDEVICE_REG(dat[20], dat[21], dat[22]); + /* Delta Thermal Limit */ + else if (!strncmp(dat, "iks_GC_GMgq", 20)) + r->tlimit_reg = + TASDEVICE_REG(dat[20], dat[21], dat[22]); + /* Thermal data for PG 1.0 device */ + else if (!strncmp(dat, "gou_Yao", 20)) + memcpy(p->thr, &dat[20], 3); + /* Pilot tone enable flag, usually the sine wave */ + else if (!strncmp(dat, "kgd_Wsc_Qsbp", 20)) + memcpy(p->plt_flg, &dat[20], 3); + /* Pilot tone gain for calibration */ + else if (!strncmp(dat, "yec_CqseSsqs", 20)) + memcpy(p->sin_gn, &dat[20], 3); + /* Pilot tone gain for calibration, useless in PG 2.0 */ + else if (!strncmp(dat, "iks_SogkGgog2", 20)) + memcpy(p->sin_gn2, &dat[20], 3); + /* Thermal data for PG 2.0 device */ + else if (!strncmp(dat, "yec_Sae_Y", 20)) + memcpy(p->thr2, &dat[20], 3); + /* Spk Equivalent Resistance in fixed-point format */ + else if (!strncmp(dat, "Re_Int", 20)) + memcpy(p->r0_reg, &dat[20], 3); + /* Check whether the spk connection is open */ + else if (!strncmp(dat, "SigFlag", 20)) + memcpy(p->tf_reg, &dat[20], 3); + /* check spk resonant frequency */ + else if (!strncmp(dat, "a1_Int", 20)) + memcpy(p->a1_reg, &dat[20], 3); + /* check spk resonant frequency */ + else if (!strncmp(dat, "a2_Int", 20)) + memcpy(p->a2_reg, &dat[20], 3); + } +} + +static int fw_parse_fct_param_address(struct tasdevice_priv *tas_priv, + struct tasdevice_fw *tas_fmw, const struct firmware *fmw, int offset) +{ + struct calidata *cali_data = &tas_priv->cali_data; + struct cali_reg *r = &cali_data->cali_reg_array; + const unsigned char *data = fmw->data; + + if (offset + 520 > fmw->size) { + dev_err(tas_priv->dev, "%s: File Size error\n", __func__); + return -EINVAL; + } + + /* skip reserved part */ + offset += 40; + + fct_param_address_parser(r, tas_fmw, &data[offset]); + + offset += 480; + + return offset; +} + static int fw_parse_variable_header_kernel( struct tasdevice_priv *tas_priv, const struct firmware *fmw, int offset) @@ -1687,13 +1805,29 @@ static int tasdevice_load_block(struct tasdevice_priv *tas_priv, return rc; } +static void dspbin_type_check(struct tasdevice_priv *tas_priv, + unsigned int ppcver) +{ + if (ppcver >= PPC3_VERSION_TAS2781_ALPHA_MIN) { + if (ppcver >= PPC3_VERSION_TAS2781_BETA_MIN) + tas_priv->dspbin_typ = TASDEV_BETA; + else if (ppcver >= PPC3_VERSION_TAS2781_BASIC_MIN) + tas_priv->dspbin_typ = TASDEV_BASIC; + else + tas_priv->dspbin_typ = TASDEV_ALPHA; + } + if (tas_priv->dspbin_typ != TASDEV_BASIC) + tas_priv->fw_parse_fct_param_address = + fw_parse_fct_param_address; +} + static int dspfw_default_callback(struct tasdevice_priv *tas_priv, unsigned int drv_ver, unsigned int ppcver) { int rc = 0; if (drv_ver == 0x100) { - if (ppcver >= PPC3_VERSION) { + if (ppcver >= PPC3_VERSION_BASE) { tas_priv->fw_parse_variable_header = fw_parse_variable_header_kernel; tas_priv->fw_parse_program_data = @@ -1702,6 +1836,7 @@ static int dspfw_default_callback(struct tasdevice_priv *tas_priv, fw_parse_configuration_data_kernel; tas_priv->tasdevice_load_block = tasdevice_load_block_kernel; + dspbin_type_check(tas_priv, ppcver); } else { switch (ppcver) { case 0x00: @@ -1717,7 +1852,7 @@ static int dspfw_default_callback(struct tasdevice_priv *tas_priv, default: dev_err(tas_priv->dev, "%s: PPCVer must be 0x0 or 0x%02x", - __func__, PPC3_VERSION); + __func__, PPC3_VERSION_BASE); dev_err(tas_priv->dev, " Current:0x%02x\n", ppcver); rc = -EINVAL; @@ -1953,28 +2088,25 @@ static int tasdevice_dspfw_ready(const struct firmware *fmw, struct tasdevice_fw_fixed_hdr *fw_fixed_hdr; struct tasdevice_fw *tas_fmw; int offset = 0; - int ret = 0; + int ret; if (!fmw || !fmw->data) { dev_err(tas_priv->dev, "%s: Failed to read firmware %s\n", __func__, tas_priv->coef_binaryname); - ret = -EINVAL; - goto out; + return -EINVAL; } tas_priv->fmw = kzalloc(sizeof(struct tasdevice_fw), GFP_KERNEL); - if (!tas_priv->fmw) { - ret = -ENOMEM; - goto out; - } + if (!tas_priv->fmw) + return -ENOMEM; + tas_fmw = tas_priv->fmw; tas_fmw->dev = tas_priv->dev; offset = fw_parse_header(tas_priv, tas_fmw, fmw, offset); - if (offset == -EINVAL) { - ret = -EINVAL; - goto out; - } + if (offset == -EINVAL) + return -EINVAL; + fw_fixed_hdr = &(tas_fmw->fw_hdr.fixed_hdr); /* Support different versions of firmware */ switch (fw_fixed_hdr->drv_ver) { @@ -2007,28 +2139,32 @@ static int tasdevice_dspfw_ready(const struct firmware *fmw, ret = dspfw_default_callback(tas_priv, fw_fixed_hdr->drv_ver, fw_fixed_hdr->ppcver); if (ret) - goto out; + return ret; break; } offset = tas_priv->fw_parse_variable_header(tas_priv, fmw, offset); - if (offset < 0) { - ret = offset; - goto out; - } + if (offset < 0) + return offset; + offset = tas_priv->fw_parse_program_data(tas_priv, tas_fmw, fmw, offset); - if (offset < 0) { - ret = offset; - goto out; - } + if (offset < 0) + return offset; + offset = tas_priv->fw_parse_configuration_data(tas_priv, tas_fmw, fmw, offset); if (offset < 0) - ret = offset; + return offset; -out: - return ret; + if (tas_priv->fw_parse_fct_param_address) { + offset = tas_priv->fw_parse_fct_param_address(tas_priv, + tas_fmw, fmw, offset); + if (offset < 0) + return offset; + } + + return 0; } int tasdevice_dsp_parser(void *context) diff --git a/sound/soc/codecs/tas2781-i2c.c b/sound/soc/codecs/tas2781-i2c.c index 90c5b2e74d12..b950914b7d48 100644 --- a/sound/soc/codecs/tas2781-i2c.c +++ b/sound/soc/codecs/tas2781-i2c.c @@ -349,13 +349,31 @@ static int calib_data_get(struct tasdevice_priv *tas_priv, int reg, return rc; } +static int partial_cali_data_update(int *reg, int j) +{ + switch (tas2781_cali_start_reg[j].reg) { + case 0: + return reg[0]; + case TAS2781_PRM_PLT_FLAG_REG: + return reg[1]; + case TAS2781_PRM_SINEGAIN_REG: + return reg[2]; + case TAS2781_PRM_SINEGAIN2_REG: + return reg[3]; + default: + return 0; + } +} + static void sngl_calib_start(struct tasdevice_priv *tas_priv, int i, int *reg, unsigned char *dat) { struct tasdevice *tasdev = tas_priv->tasdevice; struct bulk_reg_val *p = tasdev[i].cali_data_backup; + struct bulk_reg_val *t = &tasdev[i].alp_cali_bckp; const int sum = ARRAY_SIZE(tas2781_cali_start_reg); - int j; + unsigned char val[4]; + int j, r; if (p == NULL) return; @@ -370,30 +388,23 @@ static void sngl_calib_start(struct tasdevice_priv *tas_priv, int i, tasdevice_dev_read(tas_priv, i, p[j].reg, (int *)&p[j].val[0]); } else { - switch (tas2781_cali_start_reg[j].reg) { - case 0: { - if (!reg[0]) - continue; - p[j].reg = reg[0]; + if (!tas_priv->dspbin_typ) { + r = partial_cali_data_update(reg, j); + if (r) + p[j].reg = r; } - break; - case TAS2781_PRM_PLT_FLAG_REG: - p[j].reg = reg[1]; - break; - case TAS2781_PRM_SINEGAIN_REG: - p[j].reg = reg[2]; - break; - case TAS2781_PRM_SINEGAIN2_REG: - p[j].reg = reg[3]; - break; - } - tasdevice_dev_bulk_read(tas_priv, i, p[j].reg, - p[j].val, 4); + + if (p[j].reg) + tasdevice_dev_bulk_read(tas_priv, i, p[j].reg, + p[j].val, 4); } } + if (tas_priv->dspbin_typ == TASDEV_ALPHA) + tasdevice_dev_bulk_read(tas_priv, i, t->reg, t->val, 4); + /* Update the setting for calibration */ - for (j = 0; j < sum - 2; j++) { + for (j = 0; j < sum - 4; j++) { if (p[j].val_len == 1) { if (p[j].is_locked) tasdevice_dev_write(tas_priv, i, @@ -401,17 +412,33 @@ static void sngl_calib_start(struct tasdevice_priv *tas_priv, int i, TAS2781_TEST_PAGE_UNLOCK); tasdevice_dev_write(tas_priv, i, p[j].reg, tas2781_cali_start_reg[j].val[0]); - } else { - if (!p[j].reg) - continue; - tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, - (unsigned char *) - tas2781_cali_start_reg[j].val, 4); } } - tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, &dat[1], 4); - tasdevice_dev_bulk_write(tas_priv, i, p[j + 1].reg, &dat[5], 4); + if (tas_priv->dspbin_typ == TASDEV_ALPHA) { + val[0] = 0x00; + val[1] = 0x00; + val[2] = 0x21; + val[3] = 0x8e; + } else { + val[0] = tas2781_cali_start_reg[j].val[0]; + val[1] = tas2781_cali_start_reg[j].val[1]; + val[2] = tas2781_cali_start_reg[j].val[2]; + val[3] = tas2781_cali_start_reg[j].val[3]; + } + tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, val, 4); + tasdevice_dev_bulk_write(tas_priv, i, p[j + 1].reg, + (unsigned char *)tas2781_cali_start_reg[j + 1].val, 4); + tasdevice_dev_bulk_write(tas_priv, i, p[j + 2].reg, &dat[1], 4); + tasdevice_dev_bulk_write(tas_priv, i, p[j + 3].reg, &dat[5], 4); + if (tas_priv->dspbin_typ == TASDEV_ALPHA) { + val[0] = 0x00; + val[1] = 0x00; + val[2] = 0x2a; + val[3] = 0x0b; + + tasdevice_dev_bulk_read(tas_priv, i, t->reg, val, 4); + } } static int tas2781_calib_start_put(struct snd_kcontrol *kcontrol, @@ -452,14 +479,15 @@ static int tas2781_calib_start_put(struct snd_kcontrol *kcontrol, return 1; } -static void tas2781_calib_stop_put(struct tasdevice_priv *tas_priv) +static void tas2781_calib_stop_put(struct tasdevice_priv *priv) { const int sum = ARRAY_SIZE(tas2781_cali_start_reg); int i, j; - for (i = 0; i < tas_priv->ndev; i++) { - struct tasdevice *tasdev = tas_priv->tasdevice; + for (i = 0; i < priv->ndev; i++) { + struct tasdevice *tasdev = priv->tasdevice; struct bulk_reg_val *p = tasdev[i].cali_data_backup; + struct bulk_reg_val *t = &tasdev[i].alp_cali_bckp; if (p == NULL) continue; @@ -467,18 +495,21 @@ static void tas2781_calib_stop_put(struct tasdevice_priv *tas_priv) for (j = 0; j < sum; j++) { if (p[j].val_len == 1) { if (p[j].is_locked) - tasdevice_dev_write(tas_priv, i, + tasdevice_dev_write(priv, i, TAS2781_TEST_UNLOCK_REG, TAS2781_TEST_PAGE_UNLOCK); - tasdevice_dev_write(tas_priv, i, p[j].reg, + tasdevice_dev_write(priv, i, p[j].reg, p[j].val[0]); } else { if (!p[j].reg) continue; - tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, + tasdevice_dev_bulk_write(priv, i, p[j].reg, p[j].val, 4); } } + + if (priv->dspbin_typ == TASDEV_ALPHA) + tasdevice_dev_bulk_write(priv, i, t->reg, t->val, 4); } } @@ -590,16 +621,20 @@ static int tasdev_cali_data_put(struct snd_kcontrol *kcontrol, i += 2; priv->is_user_space_calidata = true; - p->r0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->r0_low_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->invr0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->pow_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->tlimit_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; + if (priv->dspbin_typ == TASDEV_BASIC) { + p->r0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->r0_low_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->invr0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->pow_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->tlimit_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + } else { + i += 15; + } memcpy(dst, &src[i], cali_data->total_sz); return 1; @@ -646,12 +681,19 @@ static int tasdev_tf_data_get(struct snd_kcontrol *kcontrol, struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; - unsigned int reg; + unsigned int reg = TAS2781_RUNTIME_RE_REG_TF; + + if (tas_priv->chip_id == TAS2781) { + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); - if (tas_priv->chip_id == TAS2781) reg = TAS2781_RUNTIME_RE_REG_TF; - else + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->tf_reg[0], p->tf_reg[1], + p->tf_reg[2]); + } else { reg = TAS2563_RUNTIME_RE_REG_TF; + } guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; @@ -666,12 +708,19 @@ static int tasdev_re_data_get(struct snd_kcontrol *kcontrol, struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; - unsigned int reg; + unsigned int reg = TAS2781_RUNTIME_RE_REG; - if (tas_priv->chip_id == TAS2781) - reg = TAS2781_RUNTIME_RE_REG; - else + if (tas_priv->chip_id == TAS2781) { + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); + + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->r0_reg[0], p->r0_reg[1], + p->r0_reg[2]); + } else { reg = TAS2563_RUNTIME_RE_REG; + } + guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; return calib_data_get(tas_priv, reg, &dst[1]); @@ -705,11 +754,16 @@ static int tasdev_XMA1_data_get(struct snd_kcontrol *kcontrol, { struct snd_soc_component *comp = snd_soc_kcontrol_component(kcontrol); struct tasdevice_priv *tas_priv = snd_soc_component_get_drvdata(comp); + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; unsigned int reg = TASDEVICE_XM_A1_REG; + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->a1_reg[0], p->a1_reg[1], p->a1_reg[2]); + guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; return calib_data_get(tas_priv, reg, &dst[1]); @@ -720,11 +774,16 @@ static int tasdev_XMA2_data_get(struct snd_kcontrol *kcontrol, { struct snd_soc_component *comp = snd_soc_kcontrol_component(kcontrol); struct tasdevice_priv *tas_priv = snd_soc_component_get_drvdata(comp); + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; unsigned int reg = TASDEVICE_XM_A2_REG; + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->a2_reg[0], p->a2_reg[1], p->a2_reg[2]); + guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; return calib_data_get(tas_priv, reg, &dst[1]); @@ -1172,10 +1231,51 @@ static int tasdevice_dsp_create_ctrls(struct tasdevice_priv *tas_priv) nr_controls < mix_index ? nr_controls : mix_index); } +static void cali_reg_update(struct bulk_reg_val *p, + struct fct_param_address *t) +{ + const int sum = ARRAY_SIZE(tas2781_cali_start_reg); + int reg, j; + + for (j = 0; j < sum; j++) { + switch (tas2781_cali_start_reg[j].reg) { + case 0: + reg = TASDEVICE_REG(t->thr[0], t->thr[1], t->thr[2]); + break; + case TAS2781_PRM_PLT_FLAG_REG: + reg = TASDEVICE_REG(t->plt_flg[0], t->plt_flg[1], + t->plt_flg[2]); + break; + case TAS2781_PRM_SINEGAIN_REG: + reg = TASDEVICE_REG(t->sin_gn[0], t->sin_gn[1], + t->sin_gn[2]); + break; + case TAS2781_PRM_SINEGAIN2_REG: + reg = TASDEVICE_REG(t->sin_gn[0], t->sin_gn[1], + t->sin_gn[2]); + break; + default: + reg = 0; + break; + } + if (reg) + p[j].reg = reg; + } +} + +static void alpa_cali_update(struct bulk_reg_val *p, + struct fct_param_address *t) +{ + p->is_locked = false; + p->reg = TASDEVICE_REG(t->thr2[0], t->thr2[1], t->thr2[2]); + p->val_len = 4; +} + static int tasdevice_create_cali_ctrls(struct tasdevice_priv *priv) { struct calidata *cali_data = &priv->cali_data; struct tasdevice *tasdev = priv->tasdevice; + struct tasdevice_fw *fmw = priv->fmw; struct soc_bytes_ext *ext_cali_data; struct snd_kcontrol_new *cali_ctrls; unsigned int nctrls; @@ -1191,14 +1291,25 @@ static int tasdevice_create_cali_ctrls(struct tasdevice_priv *priv) } if (priv->chip_id == TAS2781) { + struct fct_param_address *t = &(fmw->fct_par_addr); + cali_ctrls = (struct snd_kcontrol_new *)tas2781_cali_controls; nctrls = ARRAY_SIZE(tas2781_cali_controls); for (i = 0; i < priv->ndev; i++) { - tasdev[i].cali_data_backup = + struct bulk_reg_val *p; + + p = tasdev[i].cali_data_backup = kmemdup(tas2781_cali_start_reg, sizeof(tas2781_cali_start_reg), GFP_KERNEL); if (!tasdev[i].cali_data_backup) return -ENOMEM; + if (priv->dspbin_typ) { + cali_reg_update(p, t); + if (priv->dspbin_typ == TASDEV_ALPHA) { + p = &tasdev[i].alp_cali_bckp; + alpa_cali_update(p, t); + } + } } } else { cali_ctrls = (struct snd_kcontrol_new *)tas2563_cali_controls; -- 2.34.1

1 0

[PATCH v4] ASoC: tas2781: Support dsp firmware Alpha and Beta seaies
by Shenghao Ding 03 Mar '25

03 Mar '25

For calibration, basic version does not contain any calibration addresses, it depends on calibration tool to convey the addresses to the driver. Since Alpha and Beta firmware, all the calibration addresses are saved into the firmware. Signed-off-by: Shenghao Ding <shenghao-ding(a)ti.com> --- v4: - drop unnecceary "unsigned short size;" and "char *data;" in struct fct_param_address - replace strstr() with strncmp() in fct_param_address_parser() - replace hex_parse_u8() with memcpy() - replace hex_parse_u24() with direct assignment - drop hex_parse_u8() & hex_parse_u24() - Just return directly and use proper error codes and drop "goto out" and "out:" label in fw_parse_fct_param_address - fixed indentation issues in cali_reg_update - fixed the logical splits in tas2781_calib_stop_put() - return -ENOMEM if kmemdup fails in fw_parse_fct_param_address() - remove unnecessary "out:" label v3: - Merge the 4 patches into one patch v2: - | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: | https://urldefense.com/v3/__https://lore.kernel.org/oe-kbuild-all/2025 | 02160812.P7szOygN-lkp(a)intel.com/__;!!G3vK!WriwhtbQGlg6p00bcWutlhb-y9vS | pDrSu5_g9oOzzGOoABdirEARTntFzyX6Ss3UPRDxi_E-IA$ - | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: | https://urldefense.com/v3/__https://lore.kernel.org/oe-kbuild-all/2025 | 02160632.u0zxG7yD-lkp(a)intel.com/__;!!G3vK!VtEdfur3fZklvPVfHFFfFP60Ub2E | JYvmjSUZ65lJ6BsQPJDl1fjYRDqh3c6Q5y6sA70ssXDG5g$ v1: - update the year from 2024 to 2025 in the files. - Add dspbin_type to distinguish different dspbin types, such as TASDEV_BASIC, TASDEV_ALPHA, and TASDEV_BETA. - Add struct bulk_reg_val alp_cali_bckp to store the value of pilot tone threshold only for TASDEV_ALPHA & TASDEV_BETA firmware. - Add the declaration of (*fw_parse_fct_param_address)() to parse the addresses of calibration params. - Add the minimum version of TASDEV_ALPHA and TASDEV_BETA firmware. - Add the definition of struct fct_param_address. - update the year from 2024 to 2025. - release the memory to save calibration addresses in tasdevice_dsp_remove(). - Add TASDEV_ALPHA and TASDEV_BETA firmware checking code. - Add fct_param_address_parser for Alpha and Beta firmware. - Add an extra branch to handle calibration address for TASDEV_ALPHA and TASDEV_BETA firmware. --- include/sound/tas2781-dsp.h | 30 ++++- include/sound/tas2781.h | 11 ++ sound/pci/hda/tas2781_spi_fwlib.c | 8 +- sound/soc/codecs/tas2781-fmwlib.c | 214 +++++++++++++++++++++++++---- sound/soc/codecs/tas2781-i2c.c | 216 +++++++++++++++++++++++------- 5 files changed, 395 insertions(+), 84 deletions(-) diff --git a/include/sound/tas2781-dsp.h b/include/sound/tas2781-dsp.h index 3cda9da14f6d..c3a9efa73d5d 100644 --- a/include/sound/tas2781-dsp.h +++ b/include/sound/tas2781-dsp.h @@ -2,7 +2,7 @@ // // ALSA SoC Texas Instruments TAS2781 Audio Smart Amplifier // -// Copyright (C) 2022 - 2024 Texas Instruments Incorporated +// Copyright (C) 2022 - 2025 Texas Instruments Incorporated // https://www.ti.com // // The TAS2781 driver implements a flexible and configurable @@ -30,8 +30,10 @@ #define PRE_DEVICE_C 0x12 #define PRE_DEVICE_D 0x16 -#define PPC3_VERSION 0x4100 -#define PPC3_VERSION_TAS2781 0x14600 +#define PPC3_VERSION_BASE 0x4100 +#define PPC3_VERSION_TAS2781_BASIC_MIN 0x14600 +#define PPC3_VERSION_TAS2781_ALPHA_MIN 0x4a00 +#define PPC3_VERSION_TAS2781_BETA_MIN 0x19400 #define TASDEVICE_DEVICE_SUM 8 #define TASDEVICE_CONFIG_SUM 64 @@ -106,6 +108,27 @@ struct tasdevice_calibration { struct tasdevice_data dev_data; }; +struct fct_param_address { + /* Thermal data for PG 1.0 device */ + unsigned char thr[3]; + /* Thermal data for PG 2.0 device */ + unsigned char thr2[3]; + /* Pilot tone enable flag, usually the sine wave */ + unsigned char plt_flg[3]; + /* Pilot tone gain for calibration */ + unsigned char sin_gn[3]; + /* Pilot tone gain for calibration */ + unsigned char sin_gn2[3]; + /* high 32-bit of real-time spk impedance */ + unsigned char r0_reg[3]; + /* check spk connection */ + unsigned char tf_reg[3]; + /* check spk resonant frequency */ + unsigned char a1_reg[3]; + /* check spk resonant frequency */ + unsigned char a2_reg[3]; +}; + struct tasdevice_fw { struct tasdevice_dspfw_hdr fw_hdr; unsigned short nr_programs; @@ -114,6 +137,7 @@ struct tasdevice_fw { struct tasdevice_config *configs; unsigned short nr_calibrations; struct tasdevice_calibration *calibrations; + struct fct_param_address fct_par_addr; struct device *dev; }; diff --git a/include/sound/tas2781.h b/include/sound/tas2781.h index 4d9a37544354..eff011444cc8 100644 --- a/include/sound/tas2781.h +++ b/include/sound/tas2781.h @@ -113,6 +113,12 @@ enum audio_device { TAS2781, }; +enum dspbin_type { + TASDEV_BASIC, + TASDEV_ALPHA, + TASDEV_BETA, +}; + enum device_catlog_id { LENOVO = 0, OTHERS @@ -127,6 +133,7 @@ struct bulk_reg_val { struct tasdevice { struct bulk_reg_val *cali_data_backup; + struct bulk_reg_val alp_cali_bckp; struct tasdevice_fw *cali_data_fmw; unsigned int dev_addr; unsigned int err_code; @@ -171,6 +178,7 @@ struct tasdevice_priv { unsigned char dev_name[32]; const char *name_prefix; unsigned char ndev; + unsigned int dspbin_typ; unsigned int magic_num; unsigned int chip_id; unsigned int sysclk; @@ -196,6 +204,9 @@ struct tasdevice_priv { int (*fw_parse_configuration_data)(struct tasdevice_priv *tas_priv, struct tasdevice_fw *tas_fmw, const struct firmware *fmw, int offset); + int (*fw_parse_fct_param_address)(struct tasdevice_priv *tas_priv, + struct tasdevice_fw *tas_fmw, + const struct firmware *fmw, int offset); int (*tasdevice_load_block)(struct tasdevice_priv *tas_priv, struct tasdev_blk *block); diff --git a/sound/pci/hda/tas2781_spi_fwlib.c b/sound/pci/hda/tas2781_spi_fwlib.c index 0e2acbc3c900..19a693d58ee8 100644 --- a/sound/pci/hda/tas2781_spi_fwlib.c +++ b/sound/pci/hda/tas2781_spi_fwlib.c @@ -2,7 +2,7 @@ // // TAS2781 HDA SPI driver // -// Copyright 2024 Texas Instruments, Inc. +// Copyright 2024 - 2025 Texas Instruments, Inc. // // Author: Baojun Xu <baojun.xu(a)ti.com> @@ -370,10 +370,10 @@ static unsigned char map_dev_idx(struct tasdevice_fw *tas_fmw, int i, n = ARRAY_SIZE(non_ppc3_mapping_table); unsigned char dev_idx = 0; - if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781) { + if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781_BASIC_MIN) { p = (struct blktyp_devidx_map *)ppc3_tas2781_mapping_table; n = ARRAY_SIZE(ppc3_tas2781_mapping_table); - } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION) { + } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION_BASE) { p = (struct blktyp_devidx_map *)ppc3_mapping_table; n = ARRAY_SIZE(ppc3_mapping_table); } @@ -1605,7 +1605,7 @@ static int dspfw_default_callback(struct tasdevice_priv *tas_priv, int rc = 0; if (drv_ver == 0x100) { - if (ppcver >= PPC3_VERSION) { + if (ppcver >= PPC3_VERSION_BASE) { tas_priv->fw_parse_variable_header = fw_parse_variable_header_kernel; tas_priv->fw_parse_program_data = diff --git a/sound/soc/codecs/tas2781-fmwlib.c b/sound/soc/codecs/tas2781-fmwlib.c index bc2f68bd9846..2007c25c5778 100644 --- a/sound/soc/codecs/tas2781-fmwlib.c +++ b/sound/soc/codecs/tas2781-fmwlib.c @@ -390,10 +390,10 @@ static unsigned char map_dev_idx(struct tasdevice_fw *tas_fmw, int i, n = ARRAY_SIZE(non_ppc3_mapping_table); unsigned char dev_idx = 0; - if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781) { + if (fw_fixed_hdr->ppcver >= PPC3_VERSION_TAS2781_BASIC_MIN) { p = (struct blktyp_devidx_map *)ppc3_tas2781_mapping_table; n = ARRAY_SIZE(ppc3_tas2781_mapping_table); - } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION) { + } else if (fw_fixed_hdr->ppcver >= PPC3_VERSION_BASE) { p = (struct blktyp_devidx_map *)ppc3_mapping_table; n = ARRAY_SIZE(ppc3_mapping_table); } @@ -560,6 +560,152 @@ static int fw_parse_configuration_data_kernel( return offset; } +static void fct_param_address_parser(struct cali_reg *r, + struct tasdevice_fw *tas_fmw, const unsigned char *data) +{ + struct fct_param_address *p = &tas_fmw->fct_par_addr; + unsigned int i; + + /* + * Calibration parameters locations and data schema in dsp firmware. + * The number of items are flexible, but not more than 20. The dsp tool + * will reseve 20*24-byte space for fct params. In some cases, the + * number of fct param is less than 20, the data will be saved from the + * beginning, the rest part will be stuffed with zero. + * + * fct_param_num (not more than 20) + * for (i = 0; i < fct_param_num; i++) { + * Alias of fct param (20 bytes) + * Book (1 byte) + * Page (1 byte) + * Offset (1 byte) + * CoeffLength (1 byte) = 0x1 + * } + * if (20 - fct_param_num) + * 24*(20 - fct_param_num) pieces of '0' as stuffing + * + * As follow: + * umg_SsmKEGCye = Book, Page, Offset, CoeffLength + * iks_E0 = Book, Page, Offset, CoeffLength + * yep_LsqM0 = Book, Page, Offset, CoeffLength + * oyz_U0_ujx = Book, Page, Offset, CoeffLength + * iks_GC_GMgq = Book, Page, Offset, CoeffLength + * gou_Yao = Book, Page, Offset, CoeffLength + * kgd_Wsc_Qsbp = Book, Page, Offset, CoeffLength + * yec_CqseSsqs = Book, Page, Offset, CoeffLength + * iks_SogkGgog2 = Book, Page, Offset, CoeffLength + * yec_Sae_Y = Book, Page, Offset, CoeffLength + * Re_Int = Book, Page, Offset, CoeffLength + * SigFlag = Book, Page, Offset, CoeffLength + * a1_Int = Book, Page, Offset, CoeffLength + * a2_Int = Book, Page, Offset, CoeffLength + */ + for (i = 0; i < 20; i++) { + const unsigned char *dat = &data[24 * i]; + + /* + * check whether current fct param is empty. + */ + if (dat[23] != 1) + break; + + if (!strncmp(dat, "umg_SsmKEGCye", 20)) { + r->pow_reg = TASDEVICE_REG(dat[20], dat[21], dat[22]); + continue; + } + /* high 32-bit of real-time spk impedance */ + if (!strncmp(dat, "iks_E0", 20)) { + r->r0_reg = TASDEVICE_REG(dat[20], dat[21], dat[22]); + continue; + } + /* inverse of real-time spk impedance */ + if (!strncmp(dat, "yep_LsqM0", 20)) { + r->invr0_reg = + TASDEVICE_REG(dat[20], dat[21], dat[22]); + continue; + } + /* low 32-bit of real-time spk impedance */ + if (!strncmp(dat, "oyz_U0_ujx", 20)) { + r->r0_low_reg = + TASDEVICE_REG(dat[20], dat[21], dat[22]); + continue; + } + /* Delta Thermal Limit */ + if (!strncmp(dat, "iks_GC_GMgq", 20)) { + r->tlimit_reg = + TASDEVICE_REG(dat[20], dat[21], dat[22]); + continue; + } + /* Thermal data for PG 1.0 device */ + if (!strncmp(dat, "gou_Yao", 20)) { + memcpy(p->thr, &dat[20], 3); + continue; + } + /* Pilot tone enable flag, usually the sine wave */ + if (!strncmp(dat, "kgd_Wsc_Qsbp", 20)) { + memcpy(p->plt_flg, &dat[20], 3); + continue; + } + /* Pilot tone gain for calibration */ + if (!strncmp(dat, "yec_CqseSsqs", 20)) { + memcpy(p->sin_gn, &dat[20], 3); + continue; + } + /* Pilot tone gain for calibration, useless in PG 2.0 */ + if (!strncmp(dat, "iks_SogkGgog2", 20)) { + memcpy(p->sin_gn2, &dat[20], 3); + continue; + } + /* Thermal data for PG 2.0 device */ + if (!strncmp(dat, "yec_Sae_Y", 20)) { + memcpy(p->thr2, &dat[20], 3); + continue; + } + /* Spk Equivalent Resistance in fixed-point format */ + if (!strncmp(dat, "Re_Int", 20)) { + memcpy(p->r0_reg, &dat[20], 3); + continue; + } + /* Check whether the spk connection is open */ + if (!strncmp(dat, "SigFlag", 20)) { + memcpy(p->tf_reg, &dat[20], 3); + continue; + } + /* check spk resonant frequency */ + if (!strncmp(dat, "a1_Int", 20)) { + memcpy(p->a1_reg, &dat[20], 3); + continue; + } + /* check spk resonant frequency */ + if (!strncmp(dat, "a2_Int", 20)) { + memcpy(p->a2_reg, &dat[20], 3); + continue; + } + } +} + +static int fw_parse_fct_param_address(struct tasdevice_priv *tas_priv, + struct tasdevice_fw *tas_fmw, const struct firmware *fmw, int offset) +{ + struct calidata *cali_data = &tas_priv->cali_data; + struct cali_reg *r = &cali_data->cali_reg_array; + const unsigned char *data = fmw->data; + + if (offset + 520 > fmw->size) { + dev_err(tas_priv->dev, "%s: File Size error\n", __func__); + return -1; + } + + /* skip reserved part */ + offset += 40; + + fct_param_address_parser(r, tas_fmw, &data[offset]); + + offset += 480; + + return offset; +} + static int fw_parse_variable_header_kernel( struct tasdevice_priv *tas_priv, const struct firmware *fmw, int offset) @@ -1687,13 +1833,29 @@ static int tasdevice_load_block(struct tasdevice_priv *tas_priv, return rc; } +static void dspbin_type_check(struct tasdevice_priv *tas_priv, + unsigned int ppcver) +{ + if (ppcver >= PPC3_VERSION_TAS2781_ALPHA_MIN) { + if (ppcver >= PPC3_VERSION_TAS2781_BETA_MIN) + tas_priv->dspbin_typ = TASDEV_BETA; + else if (ppcver >= PPC3_VERSION_TAS2781_BASIC_MIN) + tas_priv->dspbin_typ = TASDEV_BASIC; + else + tas_priv->dspbin_typ = TASDEV_ALPHA; + } + if (tas_priv->dspbin_typ != TASDEV_BASIC) + tas_priv->fw_parse_fct_param_address = + fw_parse_fct_param_address; +} + static int dspfw_default_callback(struct tasdevice_priv *tas_priv, unsigned int drv_ver, unsigned int ppcver) { int rc = 0; if (drv_ver == 0x100) { - if (ppcver >= PPC3_VERSION) { + if (ppcver >= PPC3_VERSION_BASE) { tas_priv->fw_parse_variable_header = fw_parse_variable_header_kernel; tas_priv->fw_parse_program_data = @@ -1702,6 +1864,7 @@ static int dspfw_default_callback(struct tasdevice_priv *tas_priv, fw_parse_configuration_data_kernel; tas_priv->tasdevice_load_block = tasdevice_load_block_kernel; + dspbin_type_check(tas_priv, ppcver); } else { switch (ppcver) { case 0x00: @@ -1717,7 +1880,7 @@ static int dspfw_default_callback(struct tasdevice_priv *tas_priv, default: dev_err(tas_priv->dev, "%s: PPCVer must be 0x0 or 0x%02x", - __func__, PPC3_VERSION); + __func__, PPC3_VERSION_BASE); dev_err(tas_priv->dev, " Current:0x%02x\n", ppcver); rc = -EINVAL; @@ -1958,23 +2121,20 @@ static int tasdevice_dspfw_ready(const struct firmware *fmw, if (!fmw || !fmw->data) { dev_err(tas_priv->dev, "%s: Failed to read firmware %s\n", __func__, tas_priv->coef_binaryname); - ret = -EINVAL; - goto out; + return -EINVAL; } tas_priv->fmw = kzalloc(sizeof(struct tasdevice_fw), GFP_KERNEL); - if (!tas_priv->fmw) { - ret = -ENOMEM; - goto out; - } + if (!tas_priv->fmw) + return -ENOMEM; + tas_fmw = tas_priv->fmw; tas_fmw->dev = tas_priv->dev; offset = fw_parse_header(tas_priv, tas_fmw, fmw, offset); - if (offset == -EINVAL) { - ret = -EINVAL; - goto out; - } + if (offset == -EINVAL) + return -EINVAL; + fw_fixed_hdr = &(tas_fmw->fw_hdr.fixed_hdr); /* Support different versions of firmware */ switch (fw_fixed_hdr->drv_ver) { @@ -2007,27 +2167,31 @@ static int tasdevice_dspfw_ready(const struct firmware *fmw, ret = dspfw_default_callback(tas_priv, fw_fixed_hdr->drv_ver, fw_fixed_hdr->ppcver); if (ret) - goto out; + return ret; break; } offset = tas_priv->fw_parse_variable_header(tas_priv, fmw, offset); - if (offset < 0) { - ret = offset; - goto out; - } + if (offset < 0) + return offset; + offset = tas_priv->fw_parse_program_data(tas_priv, tas_fmw, fmw, offset); - if (offset < 0) { - ret = offset; - goto out; - } + if (offset < 0) + return offset; + offset = tas_priv->fw_parse_configuration_data(tas_priv, tas_fmw, fmw, offset); if (offset < 0) - ret = offset; + return offset; + + if (tas_priv->fw_parse_fct_param_address) { + offset = tas_priv->fw_parse_fct_param_address(tas_priv, + tas_fmw, fmw, offset); + if (offset < 0) + ret = offset; + } -out: return ret; } diff --git a/sound/soc/codecs/tas2781-i2c.c b/sound/soc/codecs/tas2781-i2c.c index 90c5b2e74d12..7d0b119c828a 100644 --- a/sound/soc/codecs/tas2781-i2c.c +++ b/sound/soc/codecs/tas2781-i2c.c @@ -349,13 +349,31 @@ static int calib_data_get(struct tasdevice_priv *tas_priv, int reg, return rc; } +static int partial_cali_data_update(int *reg, int j) +{ + switch (tas2781_cali_start_reg[j].reg) { + case 0: + return reg[0]; + case TAS2781_PRM_PLT_FLAG_REG: + return reg[1]; + case TAS2781_PRM_SINEGAIN_REG: + return reg[2]; + case TAS2781_PRM_SINEGAIN2_REG: + return reg[3]; + default: + return 0; + } +} + static void sngl_calib_start(struct tasdevice_priv *tas_priv, int i, int *reg, unsigned char *dat) { struct tasdevice *tasdev = tas_priv->tasdevice; struct bulk_reg_val *p = tasdev[i].cali_data_backup; + struct bulk_reg_val *t = &tasdev[i].alp_cali_bckp; const int sum = ARRAY_SIZE(tas2781_cali_start_reg); - int j; + unsigned char val[4]; + int j, r; if (p == NULL) return; @@ -370,30 +388,23 @@ static void sngl_calib_start(struct tasdevice_priv *tas_priv, int i, tasdevice_dev_read(tas_priv, i, p[j].reg, (int *)&p[j].val[0]); } else { - switch (tas2781_cali_start_reg[j].reg) { - case 0: { - if (!reg[0]) - continue; - p[j].reg = reg[0]; + if (!tas_priv->dspbin_typ) { + r = partial_cali_data_update(reg, j); + if (r) + p[j].reg = r; } - break; - case TAS2781_PRM_PLT_FLAG_REG: - p[j].reg = reg[1]; - break; - case TAS2781_PRM_SINEGAIN_REG: - p[j].reg = reg[2]; - break; - case TAS2781_PRM_SINEGAIN2_REG: - p[j].reg = reg[3]; - break; - } - tasdevice_dev_bulk_read(tas_priv, i, p[j].reg, - p[j].val, 4); + + if (p[j].reg) + tasdevice_dev_bulk_read(tas_priv, i, p[j].reg, + p[j].val, 4); } } + if (tas_priv->dspbin_typ == TASDEV_ALPHA) + tasdevice_dev_bulk_read(tas_priv, i, t->reg, t->val, 4); + /* Update the setting for calibration */ - for (j = 0; j < sum - 2; j++) { + for (j = 0; j < sum - 4; j++) { if (p[j].val_len == 1) { if (p[j].is_locked) tasdevice_dev_write(tas_priv, i, @@ -401,17 +412,33 @@ static void sngl_calib_start(struct tasdevice_priv *tas_priv, int i, TAS2781_TEST_PAGE_UNLOCK); tasdevice_dev_write(tas_priv, i, p[j].reg, tas2781_cali_start_reg[j].val[0]); - } else { - if (!p[j].reg) - continue; - tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, - (unsigned char *) - tas2781_cali_start_reg[j].val, 4); } } - tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, &dat[1], 4); - tasdevice_dev_bulk_write(tas_priv, i, p[j + 1].reg, &dat[5], 4); + if (tas_priv->dspbin_typ == TASDEV_ALPHA) { + val[0] = 0x00; + val[1] = 0x00; + val[2] = 0x21; + val[3] = 0x8e; + } else { + val[0] = tas2781_cali_start_reg[j].val[0]; + val[1] = tas2781_cali_start_reg[j].val[1]; + val[2] = tas2781_cali_start_reg[j].val[2]; + val[3] = tas2781_cali_start_reg[j].val[3]; + } + tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, val, 4); + tasdevice_dev_bulk_write(tas_priv, i, p[j + 1].reg, + (unsigned char *)tas2781_cali_start_reg[j + 1].val, 4); + tasdevice_dev_bulk_write(tas_priv, i, p[j + 2].reg, &dat[1], 4); + tasdevice_dev_bulk_write(tas_priv, i, p[j + 3].reg, &dat[5], 4); + if (tas_priv->dspbin_typ == TASDEV_ALPHA) { + val[0] = 0x00; + val[1] = 0x00; + val[2] = 0x2a; + val[3] = 0x0b; + + tasdevice_dev_bulk_read(tas_priv, i, t->reg, val, 4); + } } static int tas2781_calib_start_put(struct snd_kcontrol *kcontrol, @@ -452,14 +479,15 @@ static int tas2781_calib_start_put(struct snd_kcontrol *kcontrol, return 1; } -static void tas2781_calib_stop_put(struct tasdevice_priv *tas_priv) +static void tas2781_calib_stop_put(struct tasdevice_priv *priv) { const int sum = ARRAY_SIZE(tas2781_cali_start_reg); int i, j; - for (i = 0; i < tas_priv->ndev; i++) { - struct tasdevice *tasdev = tas_priv->tasdevice; + for (i = 0; i < priv->ndev; i++) { + struct tasdevice *tasdev = priv->tasdevice; struct bulk_reg_val *p = tasdev[i].cali_data_backup; + struct bulk_reg_val *t = &tasdev[i].alp_cali_bckp; if (p == NULL) continue; @@ -467,18 +495,21 @@ static void tas2781_calib_stop_put(struct tasdevice_priv *tas_priv) for (j = 0; j < sum; j++) { if (p[j].val_len == 1) { if (p[j].is_locked) - tasdevice_dev_write(tas_priv, i, + tasdevice_dev_write(priv, i, TAS2781_TEST_UNLOCK_REG, TAS2781_TEST_PAGE_UNLOCK); - tasdevice_dev_write(tas_priv, i, p[j].reg, + tasdevice_dev_write(priv, i, p[j].reg, p[j].val[0]); } else { if (!p[j].reg) continue; - tasdevice_dev_bulk_write(tas_priv, i, p[j].reg, + tasdevice_dev_bulk_write(priv, i, p[j].reg, p[j].val, 4); } } + + if (priv->dspbin_typ == TASDEV_ALPHA) + tasdevice_dev_bulk_write(priv, i, t->reg, t->val, 4); } } @@ -590,16 +621,20 @@ static int tasdev_cali_data_put(struct snd_kcontrol *kcontrol, i += 2; priv->is_user_space_calidata = true; - p->r0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->r0_low_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->invr0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->pow_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; - p->tlimit_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); - i += 3; + if (priv->dspbin_typ == TASDEV_BASIC) { + p->r0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->r0_low_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->invr0_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->pow_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + p->tlimit_reg = TASDEVICE_REG(src[i], src[i + 1], src[i + 2]); + i += 3; + } else { + i += 15; + } memcpy(dst, &src[i], cali_data->total_sz); return 1; @@ -646,12 +681,19 @@ static int tasdev_tf_data_get(struct snd_kcontrol *kcontrol, struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; - unsigned int reg; + unsigned int reg = TAS2781_RUNTIME_RE_REG_TF; + + if (tas_priv->chip_id == TAS2781) { + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); - if (tas_priv->chip_id == TAS2781) reg = TAS2781_RUNTIME_RE_REG_TF; - else + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->tf_reg[0], p->tf_reg[1], + p->tf_reg[2]); + } else { reg = TAS2563_RUNTIME_RE_REG_TF; + } guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; @@ -666,12 +708,19 @@ static int tasdev_re_data_get(struct snd_kcontrol *kcontrol, struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; - unsigned int reg; + unsigned int reg = TAS2781_RUNTIME_RE_REG; - if (tas_priv->chip_id == TAS2781) - reg = TAS2781_RUNTIME_RE_REG; - else + if (tas_priv->chip_id == TAS2781) { + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); + + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->r0_reg[0], p->r0_reg[1], + p->r0_reg[2]); + } else { reg = TAS2563_RUNTIME_RE_REG; + } + guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; return calib_data_get(tas_priv, reg, &dst[1]); @@ -705,11 +754,16 @@ static int tasdev_XMA1_data_get(struct snd_kcontrol *kcontrol, { struct snd_soc_component *comp = snd_soc_kcontrol_component(kcontrol); struct tasdevice_priv *tas_priv = snd_soc_component_get_drvdata(comp); + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; unsigned int reg = TASDEVICE_XM_A1_REG; + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->a1_reg[0], p->a1_reg[1], p->a1_reg[2]); + guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; return calib_data_get(tas_priv, reg, &dst[1]); @@ -720,11 +774,16 @@ static int tasdev_XMA2_data_get(struct snd_kcontrol *kcontrol, { struct snd_soc_component *comp = snd_soc_kcontrol_component(kcontrol); struct tasdevice_priv *tas_priv = snd_soc_component_get_drvdata(comp); + struct tasdevice_fw *tas_fmw = tas_priv->fmw; + struct fct_param_address *p = &(tas_fmw->fct_par_addr); struct soc_bytes_ext *bytes_ext = (struct soc_bytes_ext *) kcontrol->private_value; unsigned char *dst = ucontrol->value.bytes.data; unsigned int reg = TASDEVICE_XM_A2_REG; + if (tas_priv->dspbin_typ) + reg = TASDEVICE_REG(p->a2_reg[0], p->a2_reg[1], p->a2_reg[2]); + guard(mutex)(&tas_priv->codec_lock); dst[0] = bytes_ext->max; return calib_data_get(tas_priv, reg, &dst[1]); @@ -1172,10 +1231,52 @@ static int tasdevice_dsp_create_ctrls(struct tasdevice_priv *tas_priv) nr_controls < mix_index ? nr_controls : mix_index); } +static void cali_reg_update(struct bulk_reg_val *p, + struct fct_param_address *t) +{ + const int sum = ARRAY_SIZE(tas2781_cali_start_reg); + int reg = 0; + int j; + + for (j = 0; j < sum; j++) { + switch (tas2781_cali_start_reg[j].reg) { + case 0: + reg = TASDEVICE_REG(t->thr[0], t->thr[1], t->thr[2]); + break; + case TAS2781_PRM_PLT_FLAG_REG: + reg = TASDEVICE_REG(t->plt_flg[0], t->plt_flg[1], + t->plt_flg[2]); + break; + case TAS2781_PRM_SINEGAIN_REG: + reg = TASDEVICE_REG(t->sin_gn[0], t->sin_gn[1], + t->sin_gn[2]); + break; + case TAS2781_PRM_SINEGAIN2_REG: + reg = TASDEVICE_REG(t->sin_gn[0], t->sin_gn[1], + t->sin_gn[2]); + break; + default: + reg = 0; + break; + } + if (reg) + p[j].reg = reg; + } +} + +static void alpa_cali_update(struct bulk_reg_val *p, + struct fct_param_address *t) +{ + p->is_locked = false; + p->reg = TASDEVICE_REG(t->thr2[0], t->thr2[1], t->thr2[2]); + p->val_len = 4; +} + static int tasdevice_create_cali_ctrls(struct tasdevice_priv *priv) { struct calidata *cali_data = &priv->cali_data; struct tasdevice *tasdev = priv->tasdevice; + struct tasdevice_fw *fmw = priv->fmw; struct soc_bytes_ext *ext_cali_data; struct snd_kcontrol_new *cali_ctrls; unsigned int nctrls; @@ -1191,14 +1292,25 @@ static int tasdevice_create_cali_ctrls(struct tasdevice_priv *priv) } if (priv->chip_id == TAS2781) { + struct fct_param_address *t = &(fmw->fct_par_addr); + cali_ctrls = (struct snd_kcontrol_new *)tas2781_cali_controls; nctrls = ARRAY_SIZE(tas2781_cali_controls); for (i = 0; i < priv->ndev; i++) { - tasdev[i].cali_data_backup = + struct bulk_reg_val *p; + + p = tasdev[i].cali_data_backup = kmemdup(tas2781_cali_start_reg, sizeof(tas2781_cali_start_reg), GFP_KERNEL); if (!tasdev[i].cali_data_backup) return -ENOMEM; + if (priv->dspbin_typ) { + cali_reg_update(p, t); + if (priv->dspbin_typ == TASDEV_ALPHA) { + p = &tasdev[i].alp_cali_bckp; + alpa_cali_update(p, t); + } + } } } else { cali_ctrls = (struct snd_kcontrol_new *)tas2563_cali_controls; -- 2.34.1

2 1