Kernel 5.11

[Doug S]

For the second day in a row "daily" (the only one available) is not booting here. It installs OK, and stalls on graphics enabling... Will follow. Other non-mainline kernels boot perfectly.

Hi zika,

the mainline as of now:

06d5d309a3f1 (HEAD -> master, origin/master, origin/HEAD) Merge tag 'kbuild-fixes-v5.12' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
cd278456d4ca Merge tag 'csky-for-linus-5.12-rc1' of git://github.com/c-sky/csky-linux
7d19ea5e9973 Merge tag 'riscv-for-linus-5.12-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux

boots fine on my test computer. However, it is a CLI only server, no desktop GUI. so, proves nothing really.

[zika]

Resolved on my side, also...

:~$ uname -r5.11.0-051100daily20210228-generic

DRMTip and 5.12 still in the oven...

[tome74]

First time posting in Ubuntu Forum, so forgive me if this is the wrong place to post this.

Ubuntu 20.10, with Linux kernel 5.11 (tried 5.11-rc2, 5.11, 5.11.1, 5.11.3). System boots, but display is corrupted. See attached image.

System specs: Ryzen Pro 4750G APU Renoir, 64GB RAM, Integrated Graphics.

Is this is a known issue ?

Problem doesn't occur with kernel 5.10.20.

Thanks.

-Hock

[Doug S]

First time posting in Ubuntu Forum, so forgive me if this is the wrong place to post this.

Hi, and welcome.

Ubuntu 20.10, with Linux kernel 5.11 (tried 5.11-rc2, 5.11, 5.11.1, 5.11.3). System boots, but display is corrupted. See attached image.
System specs: Ryzen Pro 4750G APU Renoir, 64GB RAM, Integrated Graphics.
Is this is a known issue ?
Problem doesn't occur with kernel 5.10.20.

That is a very interesting image. I have no idea if it is a known issue or not (I am a server person, with little to no knowledge of display issues.)
The first thing for you to try is kernel 5.12-rc1. This will tell you if the problem has already been fixed. On Monday, try 5.12-rc2.

Also try mainline kernels 5.10 and 5.11-rc1. The purpose being to narrow down the faulty commit as much as possible. the next step would be to do a kernel bisection to isolate the exact commit. It'll take around 18 kernel compiles. Or, just wait long enough for someone else to figure it out.

EDIT: kernels such as 5.10.20 and 5.11.1 and such are deviations from the main mainline tree. My input towards isolating the issue is via the mainline tree only.

EDIT 2: Oh, I see that this was a major change for the 5.11 cycle. Link.

EDIT 3: See if you can find your issue listed here.

[tome74]

Thanks for your reply. I will try out your suggestions for testing & isolating the issue.

[tome74]

Just tried with kernel 5.12-rc2. Problem with corrupted screen still occurs.

[Doug S]

Just tried with kernel 5.12-rc2. Problem with corrupted screen still occurs.

ya, this is a lot of new code. Did you look at that list of issues link I gave. The first one listed looked like a contender for your issue to me. (i'll come back and edit in a direct link later).

EDIT: this new one, or is that you that entered it? But this is the one I thought of yesterday.

[tome74]

Yes, I entered the new one. Looked thru all the others, but none that has the kind of whole screen corruption that I'm seeing.

[corradoventu]

Kernel 5.11 status in launchpad: https://bugs.launchpad.net/kernel-sr...w/+bug/1917335

[Doug S]

Yes, I entered the new one. Looked thru all the others, but none that has the kind of whole screen corruption that I'm seeing.

I see in that bug report that you bisected the kernel, which led to a patch set to try which you, and someone else, verified as fixing the issue. A huge time commitment, thank you.