Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.kernel > #90699 > unrolled thread

Bug#1124634: linux: VT refresh doesn't work reliably on Surface Go 2 tablet

Back to article view | Back to linux.debian.kernel

Contents

  Bug#1124634: linux: VT refresh doesn't work reliably on Surface Go 2 tablet Andrea Bolognani <eof@kiyuko.org> - 2026-01-04 18:50 +0100
    Bug#1124634: linux: VT refresh doesn't work reliably on Surface Go 2 tablet Andrea Bolognani <eof@kiyuko.org> - 2026-04-06 01:00 +0200

#90699 — Bug#1124634: linux: VT refresh doesn't work reliably on Surface Go 2 tablet

[Multipart message — attachments visible in raw view] — view raw

Source: linux
Version: 6.17.13-1
Severity: normal

I'm running testing on a Surface Go 2 tablet and things overall work
pretty nicely. I had installed stable originally, but later upgraded
primarily to enjoy newer and more refined versions of phosh.

I have, however, encountered one regression: VTs don't work correctly
with the latest kernel, to the point where they're pretty much
unusable.

The behavior I'm seeing is as such: from the Wayland (phosh) session
running on tty7, I can switch to a tty2 using Ctrl+Alt+F2, and the
usual login prompt is displayed correctly. When I type in my
username, however, the screen doesn't get updated.

I can force an update by switching to other VTs and then back. There
doesn't seem to be a precise pattern to how many of these switches
are necessary to trigger a refresh: sometimes it's enough to do

  tty2 -> tty3 -> tty2

but other times additional switches are needed. For each of these
switches, there's a chance that the login prompt will be updated to
show the actual VT in the first line, while other times that won't
happen. Again, if there's a pattern to this, I haven't been able to
figure out what it is. Switching to the Wayland session on tty7 and
then back reliably forces a refresh.

Additionally, the entire screen occasionally gets corrupted while
sitting idle. It's hard to describe, but it looks as if the rows of
pixels all get shifted horizontally by different amounts. The
corruption lasts maybe 20-30 seconds, then things return to normal. I
can take a picture of the screen and attach it to the bug report if
that's useful.

None of this happens when booting the current stable kernel,
linux-image-6.12.57+deb13-amd64 (6.12.57-1), so it appears to be a
kernel regression.

Note that I'm working under the assumption that the issue is limited
to the Surface Go 2 hardware, but in reality I'm not running testing
on any other machine right now so it could be more widespread. I just
expect that it would have been noticed and reported already if that
had been the case.

I'm happy to provide any additional information upon request.


-- System Information:
Debian Release: 13.2
  APT prefers stable-updates
  APT policy: (500, 'stable-updates'), (500, 'stable-security'), (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 6.12.57+deb13-amd64 (SMP w/8 CPU threads; PREEMPT)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

[toc] | [next] | [standalone]

#91925

[Multipart message — attachments visible in raw view] — view raw

Control: fixed linux/6.19.10-1

On Sun, Jan 04, 2026 at 06:39:45PM +0100, Andrea Bolognani wrote:
> I'm running testing on a Surface Go 2 tablet and things overall work
> pretty nicely. I had installed stable originally, but later upgraded
> primarily to enjoy newer and more refined versions of phosh.
> 
> I have, however, encountered one regression: VTs don't work correctly
> with the latest kernel, to the point where they're pretty much
> unusable.

The most recent kernel available from testing seems to work fine.
Marking this as fixed in that version, feel free to narrow it down
further if you think that's useful; the bug can be closed as far as
I'm concerned.

-- 
Andrea Bolognani <eof@kiyuko.org>
Resistance is futile, you will be garbage collected.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.kernel

csiph-web