| View previous topic :: View next topic |
| Author |
Message |
sublogic
Guru
Joined: 21 Mar 2022
Posts: 309
Location: Pennsylvania, USA
|
 Posted: Sat Apr 08, 2023 5:44 pm Post subject: simplefb borked in 6.3-rc5 kernel |
 |
|
Normally I would not run an -rc kernel, but 6.3-rc5 has a fix for a regression I reported. (You saw it here first!)
So, they fixed my problem, but the early boot screen looks like this. Compare with the kernel from Gentoo: it used to be like so.
The images are of a 16 year old laptop, but I see the same issue ~2019 vintage Dell tower. I bisected it and reported it upstream, we'll see what they do. My two computers use a legacy BIOS but the breaking patch says: | Code: | Tested on x86-64 with EFI output and x86 with various VESA color
modes. Also tested on ppc64 with OF output. |
I wonder if it's a BIOS vs EFI issue.
Anyone interested in testing? Run dmesg and check that you are using the simple-framebuffer. If yes, you can test. You don't have to do a full install, just boot the bzImage and let it run to a rescueshell like I did, or let it panic for lack of a root partition. All you have to do is look at the screen. Something like this: | Code: | $ wget https://git.kernel.org/torvalds/t/linux-6.3-rc5.tar.gz
$ tar xzvf linux-6.3-rc5.tar.gz
$ : if -rc6 comes out, use that instead
$ cd linux-6.3-rc5
$ zcat /proc/config.gz > .config
$ make olddefconfig
$ make bzImage
$ : customize the last few lines for your installation
$ : this works on mine
$ sudo mount /boot
$ sudo cp arch/x66/boot/bzImage /boot
$ sudo reboot |
I use grub, I can edit a menu entry to use the bzImage and pass a debug option. If you don't have an initramfs, just clobber the root= option. If you have a root configured in the kernel, take it out from menuconfig before compiling.
Uh, backups recommended in case the kernel under test decides to shred your disks (highly unlikely in this case).
If you get your name on a Tested-by: patch header, you get a nice email from Greg KH asking how to classify you in the kernel contributor statistics  . |
|
| Back to top |
|
 |
Goverp
Advocate
Joined: 07 Mar 2007
Posts: 2204
|
| Posted: Sat Apr 08, 2023 6:39 pm Post subject: |
|
|
I presume you've spotted it, but in case not, the "format" in the second "simple-framebuffer" line following the takeover has changed between "x8r8g8b8" and "r8g8b8", which presumably explains the corrupted display.
_________________
Greybeard |
|
| Back to top |
|
|
sublogic
Guru
Joined: 21 Mar 2022
Posts: 309
Location: Pennsylvania, USA
|
| Posted: Sat Apr 08, 2023 6:56 pm Post subject: |
|
|
Yes, and the "Genkernel" line, which has the same text on the two displays, is physically 25% shorter on the corrupted display.
I'm hoping the developers will pick up some clues from these dmesg lines. |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Kernel & Hardware
