00:22 <ukky> farkuhar: I asked for VID & DID hoping to find special treatment in the kernel under 'Device Drivers->HID Support->HID bus support->Special HID drivers'. There are none for your keyboard.

00:52 <farkuhar> ukky: thanks for investigating. It's not a dealbreaker that I have to toggle NumLock before trying to log in. The benefit of extra desk space (due to the tenkeyless layout) is worth the hassle. If I had a bigger desk I would have just used the full-size Model M that I got from the same company in 2004.

00:53 <farkuhar> Besides, I'll probably upgrade the motherboard before this keyboard dies on me. And now I know to avoid AsRock or ASUS.

00:57 <ukky> One other test to try: Boot install ISO and see how your keyboard gets initialized (need multiple reboots/tests). Reason: CRUX ISO does not use udev and uses mdev instead.

01:20 <ukky> farkuhar: Maybe you have these BIOS options: https://ubuntuforums.org/showthread.php?t=2228383&p=13050659#post13050659

01:25 <farkuhar> ukky: they do sound familiar, from one of my previous motherboards, though ... I don't think AsRock gives the user any control over those options.

01:28 <farkuhar> thanks anyway for taking the trouble to do the research ... something I should have done before selecting the AsRock motherboard back in 2020.

01:29 <jaeger> Interesting issue. I've had quite a few asrock motherboards over the years but never ran into that with my keyboards

01:45 <ukky> That keyboard is custom and might have skipped USB compliance test(s).

01:51 <jaeger> true

01:53 <ukky> Here guy also has trouble with Unicomp Mini M keyboard: https://www.reddit.com/r/modelm/comments/o7m5so/comment/h323rvi/ and even lays out idea to replace microcontroller and write his own firmware

05:00 <cruxbot> [contrib.git/3.7]: highlight: 4.7 -> 4.8

05:00 <cruxbot> [contrib.git/3.7]: python3-filelock: 3.12.2 -> 3.12.3

05:00 <cruxbot> [contrib.git/3.7]: python3-twisted: 22.10.0 -> 23.8.0

05:00 <cruxbot> [contrib.git/3.7]: libavif: 0.11.1 -> 1.0.0

06:21 <cruxbot> [contrib.git/3.7]: clamav: updated to version 1.2.0

07:25 <cruxbot> [opt.git/3.7]: qemu-all: add man page

07:25 <cruxbot> [opt.git/3.7]: qemu: add man page

08:49 <cruxbot> [contrib.git/3.7]: transmission-gtk: update to 4.0.4

08:49 <cruxbot> [contrib.git/3.7]: pmwiki: update to 2.3.25

08:50 <cruxbot> [opt.git/3.7]: transmission-qt: update to 4.0.4

08:50 <cruxbot> [opt.git/3.7]: transmission: update to 4.0.4

09:53 <lavaball> i don't need config_x86_umip, right?

11:01 <lavaball> how do i find out if i have config_acpi_dptf support? do i have intel. i do have atom. nothing in dmesg though.

11:14 <SiFuh> ls /sys/bus/platform/devices/INT*/ |grep dptf

11:15 <lavaball> apparently not.

11:15 <lavaball> thanks a bunch.

11:16 <SiFuh> config_x86_umip is for X86 both 32 and 64. Intel and AMD. It was originally called config_intel_x86_umip but to stop confusion they removed intel from the name.

11:16 <SiFuh> You can use it for some games and some vitrual machines. It is generally safe to disable it.

11:25 <lavaball> thanks a bunch.

11:30 <SiFuh> According to line 138 of /usr/src/linux-6.1.49/Documentation/x86/cpuinfo.rst you can disable config_x86_umip without recompiling your kernel at the kernel cmdline with clearcpuid=514. Could be useful with a crashing 32 bit program to see if it is responsible.

12:22 <lavaball> 32bit?

12:23 <lavaball> where we are going we don't need 32bit!

12:25 <SiFuh> VirtualBox, Wine32 and virtual machines, 32bit compatability, 32 bit games.

18:15 <cruxbot> [contrib.git/3.7]: objfw: initial commit, version 1.0

19:30 <cruxbot> [contrib.git/3.7]: easyeffects: 7.0.7 -> 7.0.8

19:30 <cruxbot> [contrib.git/3.7]: pipewire: 0.3.78 -> 0.3.79

19:30 <cruxbot> [contrib.git/3.7]: python3-pyproject-api: 1.5.4 -> 1.6.1

19:38 <cruxbot> [compat-32.git/3.7]: pipewire-32: 0.3.78 -> 0.3.79