Bug 1940163 - Console output not shown on display unless plymouth installed, even when booted with `console=tty0` (aarch64)
Summary: Console output not shown on display unless plymouth installed, even when boot...
Keywords:
Status: CLOSED EOL
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 34
Hardware: aarch64
OS: Unspecified
unspecified
low
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard: openqa https://fedoraproject.org/wiki...
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2021-03-17 17:41 UTC by Adam Williamson
Modified: 2022-06-08 06:24 UTC (History)
21 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2022-06-08 06:24:33 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)

Description Adam Williamson 2021-03-17 17:41:53 UTC
On a Fedora aarch64 system with a physical display (or a VM with an emulated one), if plymouth is not installed, then even when booting with parameter `console=tty0`, no boot messages are shown on the monitor. This includes the decryption prompt for any encrypted partition; the prompt is shown on the serial console instead.

This has been the case for a very long time, it seems, but we only recently noticed because we stopped installing plymouth in non-graphical installs by default (see https://bugzilla.redhat.com/show_bug.cgi?id=1933378 for more on that).

I'm proposing this as a blocker as a violation of Basic criterion "A system installed without a graphical package set must boot to a state where it is possible to log in through at least one of the default virtual consoles" footnote "In all of the above cases, if any system partitions were encrypted as part of the installation, the boot process must prompt for the passphrase(s) and correctly unlock the partition(s) when provided with the correct passphrase(s)".

Comment 1 Adam Williamson 2021-03-17 17:43:23 UTC
Alper Nebi Yasak (added to cc) proposed a patch series for this last year, though it seems to have not received a single reply:
https://lore.kernel.org/lkml/44156595-0eee-58da-4376-fd25b634d21b@gmail.com/T/

Comment 2 Alper Nebi Yasak 2021-03-17 18:12:34 UTC
(As I wrote on the other thread, there is a v2 of my patch [1] with some replies.)

[1] https://lore.kernel.org/linux-serial/20200430161438.17640-1-alpernebiyasak@gmail.com/T/

Comment 3 Alper Nebi Yasak 2021-03-17 18:32:20 UTC
> if plymouth is not installed, then even when booting with parameter `console=tty0`, no boot messages are shown on the monitor.

Just to make sure, are you putting tty0 as the last console= parameter on the kernel cmdline? What does /proc/consoles look like? IIRC, it should have tty0 as the first entry and with (C) if it's the preferred console.

Comment 4 Adam Williamson 2021-03-17 19:09:36 UTC
"Just to make sure, are you putting tty0 as the last console= parameter on the kernel cmdline?"

Yes, there aren't any others.

"What does /proc/consoles look like?"

Don't know offhand, but I'll get a test to check.

Comment 5 Adam Williamson 2021-03-17 20:31:20 UTC
This is what /proc/consoles looks like:

tty0                 -WU (EC p  )    4:1
ttyAMA0              -W- (E  p a)  204:64

Comment 6 Adam Williamson 2021-03-17 23:07:41 UTC
+1 / -5 in https://pagure.io/fedora-qa/blocker-review/issue/312 , marking rejected. Adding commonbugs flag.

Comment 7 Alper Nebi Yasak 2021-03-18 14:56:24 UTC
> This is what /proc/consoles looks like:
> 
> tty0                 -WU (EC p  )    4:1
> ttyAMA0              -W- (E  p a)  204:64

Then, my patch would not really help here as this is the intended result after that patch (but without the need to pass console=tty0 explicitly). I don't know why boot messages wouldn't show up on tty0 in this case.

Comment 8 Hans de Goede 2021-03-18 15:10:36 UTC
I wonder if this is a systemd issue, where systemd somehow prefers the serial-console for asking for the diskcrypt passwd.

Adam, if you drop the "quiet" from the kernel commandline then where do the kernel messages get logged ?

It also might be interesting to try and disable fbcon deferral, by adding: "fbcon=nodefer" to the kernel commandline. I would not expect that to change much, but maybe systemd detects the console is a dummy-console and that causes it to use the serial-console instead ?

Comment 9 Adam Williamson 2021-04-06 21:12:37 UTC
https://pagure.io/fedora-comps/pull-request/642 is a PR to put plymouth back in core to avoid this.

Comment 10 Adam Williamson 2021-04-08 20:17:17 UTC
As we missed the freeze, I'm proposing this as a Final FE. I do think it would make sense to put plymouth back for final, it resolves this and is more in line with how we shipped all previous releases so it should be a safe change.

Comment 11 Adam Williamson 2021-04-12 16:09:56 UTC
+4 in https://pagure.io/fedora-qa/blocker-review/issue/345 , marking accepted.

Comment 12 Adam Williamson 2021-04-13 00:50:02 UTC
setting POST per #c9.

Comment 13 Adam Williamson 2021-04-14 18:53:24 UTC
OK, so putting plymouth back in seems to have worked, per recent openQA results. I'm dropping priority to low and clearing F34 FE status. I won't close the bug as the bug, as described, still exists - if you *don't* have plymouth, you don't get kernel output. But we've worked around it for most practical cases.

Comment 14 George R. Goffe 2021-04-26 15:59:48 UTC
Adam,

This sounds a lot like the question I posed to the test mailing list. In my case, plymouth is installed.

tty broken after "latest" upgrade

The problem happens with or without "nomodeset" in boot cmdline. 

fc35-bash 5.1 ~# cat /proc/consoles
tty0                 -WU (EC p  )    4:1

I'll try console=tty0 and report.

George...

Comment 15 Adam Williamson 2021-04-26 16:02:37 UTC
Unless you're running on aarch64, no, this is not the same bug.

Comment 16 George R. Goffe 2021-04-26 20:45:52 UTC
Adam,

Not running aarch64... x86_64.

A lot of similarities in perceived behavior though.

Regards,

George...

Comment 17 Adam Williamson 2021-04-26 21:05:00 UTC
It's not the same bug. Perception is not always the same thing as reality...

Comment 18 Ben Cotton 2022-05-12 16:52:46 UTC
This message is a reminder that Fedora Linux 34 is nearing its end of life.
Fedora will stop maintaining and issuing updates for Fedora Linux 34 on 2022-06-07.
It is Fedora's policy to close all bug reports from releases that are no longer
maintained. At that time this bug will be closed as EOL if it remains open with a
'version' of '34'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, change the 'version' 
to a later Fedora Linux version.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora Linux 34 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora Linux, you are encouraged to change the 'version' to a later version
prior to this bug being closed.

Comment 19 Ben Cotton 2022-06-08 06:24:33 UTC
Fedora Linux 34 entered end-of-life (EOL) status on 2022-06-07.

Fedora Linux 34 is no longer maintained, which means that it
will not receive any further security or bug fix updates. As a result we
are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release.

Thank you for reporting this bug and we are sorry it could not be fixed.


Note You need to log in before you can comment on or make changes to this bug.