Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Bump rockchip64 edge kernel to 6.10 #6980

Merged
merged 3 commits into from
Jul 22, 2024

Conversation

paolosabatino
Copy link
Contributor

@paolosabatino paolosabatino commented Jul 21, 2024

Description

Bump rockchip64 edge kernel to v6.10.
Includes the rk3328 gpio2 bank fix from #6974

Minor changes have been needed to bring patches from 6.9 to 6.10 (notably, very small rework of the overlay compilation).

@brentr rk3308 now has a mainline audio codec driver. It breaks several patches that previously introduced a driver and manipulated it to "sync with BSP" and convert to "component". This commit contains all the modifications I did to allow the kernel to compile cleanly, but the result may not provide working audio or may not provide the same functionalities as before, hence rk3308 audio section requires some more separate (ie: another pull request) analisys and perhaps some rework.

GitHub issue reference:
Jira reference number AR-2428

How Has This Been Tested?

  • Compile deb packages
  • Test on a rk3328 live system
  • Test on a rk3399 live system

Checklist:

  • My code follows the style guidelines of this project
  • I have performed a self-review of my own code
  • I have commented my code, particularly in hard-to-understand areas
  • My changes generate no new warnings

@paolosabatino paolosabatino requested a review from a team July 21, 2024 13:37
@github-actions github-actions bot added Hardware Hardware related like kernel, U-Boot, ... Patches Patches related to kernel, U-Boot, ... labels Jul 21, 2024
@paolosabatino paolosabatino added the Ready to merge Reviewed, tested and ready for merge label Jul 22, 2024
* the existing patches for rk3308 audio codec have been disabled:
  the rk3308 driver has been mainlined and thus the patches
  don't apply anymore
* add the missing i2s_8ch_0 and _1 i2s nodes
* adjust "acodec" with "codec" device tree labels, to match
  mainline kernel nomenclature
@igorpecovnik igorpecovnik merged commit 1f7289d into armbian:main Jul 22, 2024
@brentr
Copy link
Collaborator

brentr commented Nov 25, 2024

@paolosabatino
I was on vacation when this hit. Missed this when I returned.
Audio on the RK3308 is, indeed, broken by these changes.
The device isn't even recognized. Does no one else have a RockPI-S to test?

What's worse is that both current and edge have broken audio.
There's just the:
board-rockpis-0001-arm64-dts.patch.disabled
in the current 6.6 kernel.

I confess, I don't know a lot about the audio subsystem internals.
Any pointers appreciated.

I thinking, at this point, that my priority has to be to get it working again in the "current" kernel.
Will 6.6 be "current" for a while?

However, I also have reports from users that the 6.9 kernel works!
It looks like 6.9 is still in the tree and buildable. Testing it now.

So, if this works, can we make 6.9 "current" for the RockPI-S?
(if so, how?)

@brentr
Copy link
Collaborator

brentr commented Nov 25, 2024

@igorpecovnik @paolosabatino @rpardini
The 6.9 kernel still in supported in the Armbian build repo works great on the RockPI-S
It supports audio. The mainlined driver does not work. Maybe that can be fixed, but, in the short term,
could we not make 6.9 the new "current" kernel for the RockPI-S?
If this is acceptable, how best to configure it?

(For my test, I changed the current kernel in rockchip64_common.inc)

@amazingfate
Copy link
Contributor

If this is an issue introduced by 6.10, it should be possible to find the commit causing it, by git revert the commits of driver one by one. Then a patch reverting the root cause on 6.10 can fix the issue.

@paolosabatino paolosabatino deleted the rockchip64-kernel-6.10 branch November 25, 2024 11:48
@paolosabatino
Copy link
Contributor Author

paolosabatino commented Nov 25, 2024

Hello, I agree with @amazingfate , it is much better to find the cause and patch the thing rather than pinning the kernel on an older version because of a single regression, even though edge kernel for rockchip64 recently moved to 6.12.

I spotted also some other regression against rockpi-s in the past with both kernel 6.6 and 6.11 regarding the broken support for realtek 8821cu devices.

I can do a round of checks with this newer kernel, I (finally) have a couple of days with some spare time I can invest into.

edit: I guess it is a good idea to fix first in kernel 6.6: it will probably be the "current" up to release of the kernel 6.13. When 6.13 will be relesed, the 6.12 (which is supposedly the next LTS version) will become the new "current" and 6.13 will become "edge". I guess it will happen when there is general consesus that 6.12 is good enough to be current, so timing is not strict, but it will happen sooner or later.

@brentr
Copy link
Collaborator

brentr commented Nov 25, 2024

For what it's worth, I agree with @paolosabatino post above.
the "edge" kernels are often broken.
The only urgent issue here is fixing the regression in the "current" 6.6 kernel.

@paolosabatino
Copy link
Contributor Author

I brought up a general clean-up pull request: #7519 - I reworked the disabled patches and now, in my case, analog sound is back

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Hardware Hardware related like kernel, U-Boot, ... Patches Patches related to kernel, U-Boot, ... Ready to merge Reviewed, tested and ready for merge
Development

Successfully merging this pull request may close these issues.

5 participants