New Raspberry Pi Kernels & Related Packages

I’ve shut it down and then power it on but I get the same issue :

I can log in (lightdm) and I get a desktop with only the desktop wallpaper and home icons.
If I use the right-click I can run a “open terminal here” but I cannot write in it and the windows doesn’t have controls.

I guess it’s more a xfce issue : it was updated to 4.16

I bet so also. There was a post about it being a disaster.

https://forum.manjaro.org/t/xfce4-4-16-warning/44774

Thanks I’ve add some comments. I hope someone will be able to help.

@Darksky

thanks , unfortunate again unlucky :slight_smile: guess it really up to upstream or we need to wait untill somebody applies patches ( both mentioned on arch arm forum )

https://archlinuxarm.org/forum/viewtopic.php?f=65&t=14754&sid=54382fe31ee2a4f24a771b7668d4ba79&start=20

I have been watching a Work In Progress the RPi people are doing with those patches in their tree. Looks like it is not ready for prime time yet.

https://github.com/raspberrypi/linux/pull/3999

I have pushed linux-rpi4-rc 5.11-rc2 kernel/headers and the 2 new raspberrypi-bootloader packages to the unstable branch when the mirrors sync.

While building the new kernel config I noticed a ton of changes as opposed to the 5.10 kernel. I hope they all turn out to be good. In testing I found that anbox is also working.

Be sure to upgrade the kernel and the 2 new raspberrypi-bootloader packages at the same time. They go hand in hand with each other.

linux-rpi4-rc 5.11.rc2-1
linux-rpi4-rc-headers 5.11.rc2-1
raspberrypi-bootloader 20210106-1
raspberrypi-bootloader-x 20210106-1

Sorry, my update to this kernel blew up and I had to work save my install. :slight_smile:
(I have had it occur a few times now, not sure why. I end up with 0 block files in /boot/overlay some in /boot/* and the kernel modules. Rather ugly. I am not sure if this is a filesystem issue, bad luck with overclocking or something else funky I have going on.)

When you say there are many changes, are you referring to RPi code? Or code in general? I assume you are referring to changes reflected by kernel config changes.

Wow this is the second time that I know of with kernel install issues you have had. Overclocking can cause corruption. I was thinking about some thing common with some of your issues lately and the only thing I can come up with is it seems that they involved installing packages that are compressed. My first step would be testing with this and see if the issue goes away:

over_voltage=5
arm_freq=2000

I was referring to the changes in making the new kernel config. I start out with the RPi defconfig and then in meld make the changes that people here want enabled. The reason is the pi folks sometimes use the upstream modules and will create their own and disable some upstream modules. Then some times they will remove theirs and go back to upstream as they get code pushed upstream. If you do not use this method occasionally you will wind up with a kernel that does not boot or something will not work right. I use the latest config that is close to the -rc when they first push it and in this case it would be the 5.10.4 config as a reference.

Here is meld and look and all of the adding/deletion of modules looking at the side bars. Usually there are just a few but very seldom this many.

I removed all overclocking after this last episode, so let’s see how it goes. I even removed hdmi_enable_4kp60=1 as I recently learned it bumps the core clock from 500 to 550.

With the recent changes to Plasma with mesa vulkan support, I don’t need the overclocking. :slight_smile:

And wow, yes. It would appear there are a great many changes as you said, thanks for the peek.

I do not have much time to test on the pi so let me know how the new kernel does compared to the others if you would.

@Darksky kernel 5.10.5 is available. I won’t test rc kernel for now because I have to fix my xfce desktop and don’t want to add some incertainty.

Thanks. Booting up my distcc farm…

Just updated kernel from unstable branch to 5.11.0-rc2-1-MANJARO-ARM, installed mesa-git on Manjaro kde, overclocked and it is really fast. Perfect youtube, smooth desktop. Thanks people.
One odd thing, I’m loosing audio with kms (works with fkms) on my external usb M-Track 2X2 interface.

1 Like

@Dulbi (And everyone else)

Here you go. The 2 latest raspberrypi-bootloader packages and the linux-rpi4-mainline 5.10.5-1 kernel packages has been pushed to the unstable branch when the mirrors sync.

linux-rpi4-mainline 5.10.5-1
linux-rpi4-mainline-headers 5.10.5-1
raspberrypi-bootloader 20210107-1
raspberrypi-bootloader-x 20210107-1

Seems like I read something about audio and kms on the pi forums. They were doing some type of work around. But his was on kernel 5.10. It may be audio on the headphone jack though.

This is happening only with last rc kernel. was working with previous just fine. Thanks for fast answer Darksky.

I successfully updated to the new rc from mainline, without issue. The kernel upgrade went smoothly. I reviewed dmesg and journalctl for issues but did not see anything out of the ordinary. However I do have these errors but they existed on 5.10.

Jan 07 17:14:56 future kernel: vc4_hdmi fef05700.hdmi: ASoC: error at snd_soc_dai_startup on fef05700.hdmi: -19
Jan 07 17:14:56 future kernel: vc4_hdmi fef05700.hdmi: ASoC: error at snd_soc_dai_startup on fef05700.hdmi: -19
Jan 07 17:14:56 future kernel: vc4_hdmi fef05700.hdmi: ASoC: error at snd_soc_dai_startup on fef05700.hdmi: -19
Jan 07 17:14:56 future kernel: vc4_hdmi fef05700.hdmi: ASoC: error at snd_soc_dai_startup on fef05700.hdmi: -19
Jan 07 17:14:56 future kernel: vc4_hdmi fef05700.hdmi: ASoC: error at snd_soc_dai_startup on fef05700.hdmi: -19

A fix is in the works. It involves pulseaudio too and I do see these pulseaudio complaints:

Jan 07 17:14:56 future pulseaudio[793]: Failed to find a working profile.
Jan 07 17:14:56 future pulseaudio[793]: Failed to load module "module-alsa-card" (argument: "device_id="1" name="platform-fef05700.hdmi" card_name="alsa_card.platform-fef05700.hdmi" namereg_fail=false tsched=no fixed_latency_range=no ignore_dB=no deferred_volume=yes use_ucm=yes avoid_resampling=no card_properties="module-udev-detect.discovered=1""): initialization failed.
Jan 07 17:14:56 future dbus-daemon[491]: [system] Activating via systemd: service name='org.bluez' unit='dbus-org.bluez.service' requested by ':1.31' (uid=1000 pid=793 comm="/usr/bin/pulseaudio --daemonize=no --log-target=jo")
Jan 07 17:14:56 future pulseaudio[793]: module-rescue-stream is obsolete and should no longer be loaded. Please remove it from your configuration.
Jan 07 17:14:57 future pulseaudio[793]: GetManagedObjects() failed: org.freedesktop.systemd1.NoSuchUnit: Unit dbus-org.bluez.service not found.
Jan 07 17:15:00 future systemd[617]: pulseaudio.socket: Succeeded.
Jan 07 17:15:04 future plasmashell[781]: org.kde.plasma.pulseaudio: No object for name "alsa_output.platform-fef00700.hdmi.iec958-stereo.monitor"
Jan 07 17:24:09 future pulseaudio[793]: ALSA woke us up to write new data to the device, but there was actually nothing to write.
Jan 07 17:24:09 future pulseaudio[793]: Most likely this is a bug in the ALSA driver 'vc4'. Please report this issue to the ALSA developers.
Jan 07 17:24:09 future pulseaudio[793]: We were woken up with POLLOUT set -- however a subsequent snd_pcm_avail() returned 0 or another value < min_avail.

But I do have working audio in chromium via HDMI and volume control in Plasma.

Thanks for reporting and your research.

sounds ok for me, thanks !

1 Like

The latest linux-rpi4-mainline / linux-rpi4-rc kernel packages and the latest 2 raspberrypi-bootloader packages has been pushed to the unstable branch when the mirrors sync.

linux-rpi4-mainline 5.10.6-1
linux-rpi4-mainline-headers 5.10.6-1
linux-rpi4-rc 5.11.rc3-1
linux-rpi4-rc-headers 5.11.rc3-1
raspberrypi-bootloader 20210111-1
raspberrypi-bootloader-x 20210111-1