Which Linux distro are you using? Fedora 41 Workstation
Which kernel are you using? 6.12.10-200.fc41.x86_64
Which BIOS version are you using? 3.05
Which Framework Laptop 13 model are you using? AMD Ryzen™ 7040 Series
I am noticing recently that there is a strange input delay when typing on the keyboard.
Typing feels responsive, however every few seconds there will be an input lag, it appears to happen at regular intervals and severely affects the typing experience.
I don’t use any kernel flags, but do use powertop defaults and tuned-ppd as AMD and Fedora prescribe.
I am not sure if this is expected from the Framework, or Gnome 47.
Has anyone else experience this lag? Or could it be something to do with how power management is handled by powertop? Should certain acpi settings be set to “bad”?
Hi. I tend to leave the default ones that come as “Bad”, and leave them as “Bad”.
You can experiment with turning “Bad” to “Good”.
I suspect leaving the Keyboard module as “Bad” might fix your problems.
It also causes other unwanted affects, such as unwanted repeated keys.
Varies by distro and bootloader. You’re going to have to try to look that up for your particular system.
It seems like the issue was fixed for me on newer kernels though, so maybe check if your distro has an update for 6.12.11 or 6.13 first.
I’m also experiencing this issue running kernel 6.13.1-arch1-1 on Arch Linux. I added the option you suggested to my GRUB defaults and rebuilt the GRUB config. This unfortunately did not fix the issue for me
I also ran wlr-randr and it seems to suggest that adaptive sync is already disabled
That just sounds like it’s a different issue then, lol. I can’t reproduce it, so it’s hard for me to help you, but you can try going through some other recent issues.
Okay, thanks. You might be right, since that Gitlab issue you linked to before says the problem also appeared in 6.11, but it only showed up for me in 6.12, if I remember correctly. I’ll try going through the issues like you said
Update: Found it. For anyone who might have the same problem as me, the workaround proposed in this issue, setting amdgpu.dcdebugmask=0x10 in the kernel cmdline seems to fix the problem for me.