@janne@social.treehouse.systems

Janne Grunau @janne@social.treehouse.systems · May 08, 2026

Replying to @meluzzy@woof.tech

@meluzzy@woof.tech there should be only variants ending in “+16K” Can you deinstall “kernel-core” (you might need to specify the versions), that removes dependent kernel packages as well which is intended. Make sure no packages starting with “kernel-16k” will be removed.

1

0

Janne Grunau @janne@social.treehouse.systems · May 08, 2026

Replying to @meluzzy@woof.tech

@meluzzy@woof.tech there is something wrong. you shouldn’t have kernel-uki-dtbloader to begin with. What large (starting at 10MB) files do you have in boot? Be careful not to remove any kernel-16k packages

0

3

0

Janne Grunau @janne@social.treehouse.systems · May 08, 2026

Replying to @Aissen@social.treehouse.systems

@Aissen@social.treehouse.systems that’s me failing to update the documentation/script after the f44 changes

0

1

0

Janne Grunau @janne@social.treehouse.systems · May 07, 2026

Replying to @Aissen@social.treehouse.systems

@Aissen@social.treehouse.systems @adamw@fosstodon.org is Calamares still present in the Asahi F44 images? That would be unintentional. I thought I dropped it in favor of plasma-setup

0

3

0

Janne Grunau @janne@social.treehouse.systems · Apr 30, 2026

Replying to @rysiek@mstdn.social

@rysiek there is a LTP patch with saner reproducer https://lore.kernel.org/all/20260430-cve-2026-31431-v1-1-7fdc16c25785@suse.com/

2

0

3

0

Janne Grunau @janne@social.treehouse.systems · Apr 30, 2026

Replying to @elly@donotsta.re

@elly while not immediately required watchdog should be high on the list of what to support unless it is or can be disabled in the boot loader. very frustrating to have a device which boots but resets after X seconds.

8

0

Janne Grunau @janne@social.treehouse.systems · Apr 18, 2026

someone hasn’t thought applying display scaling in macos recovery through.

This the result with an 13-inch FullHD display

16

3

2

0

Janne Grunau @janne@social.treehouse.systems · Feb 21, 2026

Replying to @trini@floss.social

@trini u-boot env has to modified anyway to set the boot command to booti/bootefi with the actual kernel image and initramfs addresses and sizes. iirc there wasn’t even enough space for that. Bootargs are already modified in the dtb so it should be unnecessary to add them to the u-boot env. The alternate bootcmd can then added to the default env for apple devices since it has a known maximal size. That’s better than my quick workaround and the alternate solutions I was thinking of. thanks

1

0

Janne Grunau @janne@social.treehouse.systems · Feb 21, 2026

Replying to @trini@floss.social

@trini there is a script (https://github.com/AsahiLinux/m1n1/blob/main/proxyclient/tools/linux.py) for m1n1's proxy mode which uploads kernel, initramfs and optionally u-boot and boots them. When using u-boot the script modifies the embedded default environment to boot kernel and initramfs from the known memory addresses. See the script source starting at line 84. This worked with the distro boot scripts since it replaced the complex boot script as that is not necessary. This broke with bootstd as the default environment is tiny and there is not much which can be dropped. I probably got overboard with the 8K padding but one intended use case is https://gitlab.freedesktop.org/gfx-ci/boot2container and I wasn't sure how long the boot arguments for that can get. I started looking how to solve this for upstream u-boot. One option I thought of is using ... char default_environment[CONFIG_ENV_DEFAULT_ENV_EMBEDDED_SIZE] = { with CONFIG_ENV_DEFAULT_ENV_EMBEDDED_SIZE unset by default. The other one is to let the build system append a configurable sized padding at the end.

1

0

Janne Grunau @janne@social.treehouse.systems · Feb 20, 2026

and so it begins:

for https://gitlab.freedesktop.org/gfx-ci/ci-tron/-/merge_requests/1078

34

6

12

0

Janne Grunau @janne@social.treehouse.systems · Feb 14, 2026

testing absurd things:

8

1

0

Janne Grunau @janne@social.treehouse.systems · Feb 13, 2026

Is my thunderbolt/usb4 cable broken?

2

0

1

0

Janne Grunau @janne@social.treehouse.systems · Jan 31, 2026

Replying to @videolan@floss.social

@videolan congratulations enjoy #FOSDEM I was planning to attend but DB decided on Thursday not to run trains between Berlin and Cologne both on Friday and Sunday

0

Janne Grunau @janne@social.treehouse.systems · Jan 20, 2026

Replying to @jstultz@hachyderm.io

@jstultz there is scripts/dtc/dt_to_config but it does not generate a config but prints possible missing CONFIG_s by mapping compatible strings via Makefiles to CONFIG symbols

1

2

1

0

Janne Grunau @janne@social.treehouse.systems · Jan 07, 2026

Replying to @sven@social.treehouse.systems

@sven still wondering why it even helps if we might just chasing the “ghost” of initializing the usb controllers in device mode in our m1n1 test setup. that needs of work as well

5

1

0

Janne Grunau @janne@social.treehouse.systems · Jun 29, 2025

Replying to @ariadne@social.treehouse.systems

@ariadne@social.treehouse.systems display part works on Apple silicon devices like a charm. No tearing (compared xorg-xserver / modesetting) and holding the refresh rate. Compatibility with x11 desktop environments in fedora 42 is janky (or I’m holding it wrong). starting things through xterm works partially. reported resolutions and display size through xrandr is off by factor 2 for the supported resolution and screen size seems to be calculated from that assuming 96 dpi. probably an xwayland thing

0

Janne Grunau @janne@social.treehouse.systems · Apr 02, 2025

Replying to @chandlerc@hachyderm.io

@chandlerc @AsahiLinux github does not charge fees on sponsorships from personal accounts and it is forwarded without further transaction fees to Open Collective (the 10% host fees still applies). So it includes a small additional sponsorship from Github (and Github probably pays less transaction fees to begin with). It breaks the vanity features on Open Collective side (just one large contribution from github sponsors), possibly breaks the monthly recurring stats and does not support payments via PayPal. Overall I’d say Github Sponsors is the preferred option if viable.

5

0

1

0

Janne Grunau

Posts

Remote instance

Media