search

hexdump0815 / imagebuilder

velvet os - simple script framework to build ubuntu 22.04 lts jammy (in older versions also 20.04 lts focal) and debian 12 bookworm (in older versions also 11 bullseye) bootable usb / sd card images for some arm and intel devices - lots of prebuilt images as well
GNU General Public License v3.0
298 stars 45 forks source link

Add ucm & libinput config for trogdor-coachz #201

Open pulakctl opened 7 months ago

pulakctl commented 7 months ago

As discussed in #44, this will add the ucm and libinput quirks to extra-files for trogdor.

Libinput changes:

Alsa UCM:

hexdump0815 commented 7 months ago

thanks a lot for this pr - i'll have a look at it during the next days

hexdump0815 commented 6 months ago

i just wanted to note that i did not forget about this pr, i'm just having very little time right now, but i'll come back to it (and other stuff) for sure when more time will be available

jenneron commented 5 months ago

UCM configuration is added to not conflict with upstream alsa-lib tree but I have not enabled it in conf.d as this commit will cause a conflict with upstream. I'm not sure how to best handle this, so let me know what would the best approach.

This part doesn't really explain why UCM configs in this PR conflict with upstream UCM, it just points to some files and commits. It is okay if you don't know it yet, but the first part of handling this would be to figure out why there is a conflict in the first place.

Is it because there are different devices using the same sound card name, but having different audio configuration? does strace alsaucm reload output suggests a better path for ucm configs, so they don't conflict? e.g. something more board-specific

Also, it is better to submit these changes to https://github.com/alsa-project/alsa-ucm-conf to make this work available for other projects and reduce amount of configs (thus reducing maintenance burden) here

jenneron commented 5 months ago

Is it because there are different devices using the same sound card name, but having different audio configuration? does strace alsaucm reload output suggests a better path for ucm configs, so they don't conflict? e.g. something more board-specific

if no, we can make different ucm configuration depending on device-tree compatible like here https://gitlab.com/postmarketOS/pmaports/-/blob/1f86677848a69f26ba60ec46b2a58b7283e7dead/device/community/alsa-ucm-conf-google-kukui/mt8183_mt6358_HiFi.conf#L36-40. You can figure out your compatible strings with cat /proc/device-tree/compatible or looking into kernel source. So, for trogdor-coachz we need to care about these compatible strings:

pulakctl commented 5 months ago

Also, it is better to submit these changes to https://github.com/alsa-project/alsa-ucm-conf to make this work available for other projects and reduce amount of configs (thus reducing maintenance burden) here

Sorry, I forgot to mention that I've sent a PR upstream, if I get any updates on that I will update this PR accordingly.

This part doesn't really explain why UCM configs in this PR conflict with upstream UCM, it just points to some files and commits.

Regarding the conflict, the commit here symlinks /usr/share/alsa/ucm2/conf.d/SC7180 to /usr/share/alsa/ucm2/SC7180, which has a config for sc7180-rt5682-max98357a-1mic. Meanwhile upstream has /usr/share/alsa/ucm2/conf.d/SC7180 as a directory, with a symlink inside it for sc7180-rt5682-max98357a-1mic. On top of that the upstream HiFi.conf for the card differs from the one in imagebuilder. Wouldn't this override the upstream config?

To enable my changes with current imagebuilder setup, we wouldn't really need to do device detection in UCM imo, we can just move the ucm config to be inline with imagebuilder's setup (/usr/share/alsa/ucm2/SC7180 would have ucm for both sc7180-rt5682-max98357a-1mic & sc7180-adau7002-max98357a.

pulakctl commented 5 months ago

Update: the alsa-ucm PR has been merged upstream