Arch Linux

WanderingSoul

Member

Registered: 2021-05-19

Posts: 6

[SOLVED]Distorted kernel ouput on screen during bootup

Hi,

My screen shows some distorted garbage right after grub finished loading kernel. You can see a snapshot here or a short recording here.

It shows that output for few seconds and then turns normal( systemd outputs show fine). I ignored it before since it didn't affect anything else but now that I have encrypted my root partition, I need to enter passphrase at boot time and I can't see prompt. I just type in passphrase when flashing across screen pauses and hit enter and the boot continues. But annoying.

I tried to pass screen resolution as 'video' kernel parameter but that didn't affect it. Journal doesn't show anything helpful ( at least for me). What's more, something similar happens during poweroff. But this time characters are visible, just not correctly placed. During boot nothing can be read for few seconds until it goes normal.

Any help?

Last edited by WanderingSoul (2022-06-13 19:41:06)

V1del

Forum Moderator

Registered: 2012-10-16

Posts: 25,273

Re: [SOLVED]Distorted kernel ouput on screen during bootup

What hardware and even if the journal isn't helpful to you it could be helpful to others.

Based on this extremely minimal set of information this just sounds like early KMS being off and so you'd want to do: https://wiki.archlinux.org/title/Kernel … _KMS_start

Last edited by V1del (2022-01-02 13:17:24)

WanderingSoul

Member

Registered: 2021-05-19

Posts: 6

Re: [SOLVED]Distorted kernel ouput on screen during bootup

Oh that did it. After including i915 module in /etc/mkinitcpio.conf and regenerating initramfs, I can see prompt.

Though it starts off same, like it used to, for split second. I see that flash once and it goes normal. Unavoidable?
And poweroff/reboot output is still terse, not affected by the change.

The output of

journalctl -b | grep -i -e 'vga' -e 'console' -e 'i915'

is this.

Jan 02 21:33:54 User kernel: Console: colour dummy device 80x25
Jan 02 21:33:54 User kernel: printk: console [tty0] enabled
Jan 02 21:33:54 User kernel: fbcon: Taking over console
Jan 02 21:33:54 User kernel: pci 0000:00:02.0: vgaarb: setting as boot VGA device
Jan 02 21:33:54 User kernel: pci 0000:00:02.0: vgaarb: VGA device added: decodes=io+mem,owns=io+mem,locks=none
Jan 02 21:33:54 User kernel: pci 0000:00:02.0: vgaarb: bridge control possible
Jan 02 21:33:54 User kernel: vgaarb: loaded
Jan 02 21:33:54 User kernel: Console: switching to colour frame buffer device 170x48
Jan 02 21:33:54 User kernel: fb0: EFI VGA frame buffer device
Jan 02 21:33:54 User kernel: fb0: switching to inteldrmfb from EFI VGA
Jan 02 21:33:54 User kernel: Console: switching to colour dummy device 80x25
Jan 02 21:33:54 User kernel: i915 0000:00:02.0: vgaarb: deactivate vga console
Jan 02 21:33:54 User kernel: i915 0000:00:02.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem
Jan 02 21:33:54 User kernel: [drm] Initialized i915 1.6.0 20200917 for 0000:00:02.0 on minor 0
Jan 02 21:33:54 User kernel: fbcon: i915drmfb (fb0) is primary device
Jan 02 21:33:54 User kernel: Console: switching to colour frame buffer device 170x48
Jan 02 21:33:54 User kernel: i915 0000:00:02.0: [drm] fb0: i915drmfb frame buffer device
Jan 02 21:33:54 User systemd[1]: Started Dispatch Password Requests to Console Directory Watch.
Jan 02 21:34:01 User kernel: snd_hda_intel 0000:00:03.0: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
Jan 02 22:44:00 User kernel: printk: Suspending console(s) (use no_console_suspend to debug)
Jan 02 22:45:01 User kernel: printk: Suspending console(s) (use no_console_suspend to debug)
Jan 03 08:13:36 User kernel: printk: Suspending console(s) (use no_console_suspend to debug)
Jan 03 08:45:22 User kernel: printk: Suspending console(s) (use no_console_suspend to debug)

The output of

journalctl -b -p 4

is this.

Jan 03 09:59:02 User kernel: DMAR: [Firmware Bug]: No firmware reserved region can cover this RMRR [0x00000000af000000-0x00000000bf1fffff], contact BIOS vendor for fixes
Jan 03 09:59:02 User kernel: DMAR: [Firmware Bug]: Your BIOS is broken; bad RMRR [0x00000000af000000-0x00000000bf1fffff]
                               BIOS vendor: Hewlett-Packard; Ver: L71 Ver. 01.49; Product Version: A3009DD10203
Jan 03 09:59:02 User kernel: MDS CPU bug present and SMT on, data leak possible. See https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/mds.html for more details.
Jan 03 09:59:02 User kernel:  #2 #3
Jan 03 09:59:02 User kernel: ENERGY_PERF_BIAS: Set to 'normal', was 'performance'
Jan 03 09:59:02 User kernel: pci 0000:00:14.0: xHCI BIOS handoff failed (BIOS bug ?) 00010801
Jan 03 09:59:02 User kernel: hpet_acpi_add: no address or irqs in _CRS
Jan 03 09:59:02 User kernel: usb: port power management may be unreliable
Jan 03 09:59:08 User kernel: vboxdrv: loading out-of-tree module taints kernel.
Jan 03 09:59:08 User kernel: VBoxNetAdp: Successfully started.
Jan 03 09:59:08 User systemd-journald[377]: File /var/log/journal/845a9ca94e69434f83fea237d9c7a2e3/system.journal corrupted or uncleanly shut down, renaming and replacing.
Jan 03 09:59:08 User kernel: VBoxNetFlt: Successfully started.
Jan 03 09:59:08 User kernel: platform regulatory.0: Direct firmware load for regulatory.db failed with error -2
Jan 03 09:59:08 User kernel: e1000e 0000:00:19.0 0000:00:19.0 (uninitialized): Failed to disable ULP
Jan 03 09:59:08 User kernel: e1000e: probe of 0000:00:19.0 failed with error -3
Jan 03 09:59:08 User kernel: kauditd_printk_skb: 2 callbacks suppressed
Jan 03 09:59:08 User kernel: at24 9-0050: supply vcc not found, using dummy regulator
Jan 03 09:59:09 User lvm[524]: PV /dev/sda2 online, VG root is complete.
Jan 03 09:59:09 User lvm[524]: VG root finished
Jan 03 09:59:10 User lvm[538]: PV /dev/sda4 online, VG pool is complete.
Jan 03 09:59:10 User lvm[538]: VG pool finished
Jan 03 09:59:13 User kernel: kauditd_printk_skb: 4 callbacks suppressed
Jan 03 09:59:18 User kernel: kauditd_printk_skb: 8 callbacks suppressed
Jan 03 09:59:23 User kernel: kauditd_printk_skb: 230 callbacks suppressed
Jan 03 09:59:32 User lightdm[776]: gkr-pam: unable to locate daemon control file
Jan 03 09:59:32 User systemd-journald[377]: File /var/log/journal/845a9ca94e69434f83fea237d9c7a2e3/user-1000.journal corrupted or uncleanly shut down, renaming and replacing.
Jan 03 09:59:37 User dbus-daemon[854]: writing oom_score_adj error: Permission denied
Jan 03 09:59:42 User kernel: kauditd_printk_skb: 7 callbacks suppressed
Jan 03 10:00:01 User kernel: smpboot: Scheduler frequency invariance went wobbly, disabling!
Jan 03 10:00:01 User kernel: acpi PNP0501:00: Still not present
Jan 03 10:00:01 User kernel: done.
Jan 03 10:43:10 User kernel: usb 3-3: Disable of device-initiated U1 failed.
Jan 03 10:43:10 User kernel: usb 3-3: Disable of device-initiated U2 failed.
Jan 03 10:43:10 User kernel: acpi PNP0501:00: Still not present
Jan 03 10:43:10 User kernel: done.
Jan 03 10:43:11 User kernel: acpi PNP0501:00: Still not present
Jan 03 10:43:11 User kernel: acpi PNP0501:00: Still not present
Jan 03 10:46:04 User pulseaudio[5922]: stat('/etc/pulse/default.pa.d'): No such file or directory
Jan 03 11:00:42 User kernel: kauditd_printk_skb: 8 callbacks suppressed

Whole journal is too big to big to put here.

My machine is HP elitebook 840 G1, has intel graphics card.

V1del

Forum Moderator

Registered: 2012-10-16

Posts: 25,273

Re: [SOLVED]Distorted kernel ouput on screen during bootup

It's likely happening too early to be part of the journal anyway FWIW, one thing you can try to get rid of modeset switches is check whether your UEFI is configured for CSM compat mode and disable that. Which boot loader? You might not generally be able to get rid of that depending on a few factors, another param you could try is i915.fastboot=1 on your kernel params, but that has afaik had some issues historically.

WanderingSoul

Member

Registered: 2021-05-19

Posts: 6

Re: [SOLVED]Distorted kernel ouput on screen during bootup

Boot loader is grub. I'm guessing I check for UEFI CSM compact mode in BIOS settings?

V1del

Forum Moderator

Registered: 2012-10-16

Posts: 25,273

Re: [SOLVED]Distorted kernel ouput on screen during bootup

Yes you'd check that there. For GRUB you could additionally look at https://wiki.archlinux.org/title/GRUB/T … resolution but afaik the defaults should work well if the EFI sets the correct resolution in the first place which can often only happen with CSM disabled.

WanderingSoul

Member

Registered: 2021-05-19

Posts: 6

Re: [SOLVED]Distorted kernel ouput on screen during bootup

UEFI was on "UEFI native ( without CSM)" in BIOS already. Tried i915.fastboot=1 param too. I saw something about including intel_agp module in KMS archwiki page and tried that as well. Grub resolution setting too ( grub was displaying just fine already though).

None of them changed anything. I guess I should just tolerate it for now. And it isn't so bad now except for minor annoyance.

Head_on_a_Stick

Member

From: The Wirral

Registered: 2014-02-20

Posts: 9,003

Website

Re: [SOLVED]Distorted kernel ouput on screen during bootup

It could be a corruption of the firmware ("BIOS") splash image so try adding the bgrt_disable=1 kernel command line parameter to disable that.

---

Jin, Jîyan, Azadî

WanderingSoul

Member

Registered: 2021-05-19

Posts: 6

Re: [SOLVED]Distorted kernel ouput on screen during bootup

I used this archwiki page https://wiki.archlinux.org/title/Kernel … _KMS_start as suggested and preloaded i915 module by adding i915 in modules in /etc/mkinitcpio.conf(MODULES=(i915)). Now I have my prompt to enter password and proceed from there. There is that  distortion for a fraction of a second right after grub output, but that is harmless and bearable for now. So I'm marking this issue SOLVED.

WanderingSoul

Member

Registered: 2021-05-19

Posts: 6

Re: [SOLVED]Distorted kernel ouput on screen during bootup

It seems that the problem lies with grub(or my configuration of it). Today I switched to systemd-boot (for unrelated reasons) and it ended up resolving this issue completely. I even removed i915 from initramfs and the distortion is gone without a trace.