bumblebee: add xf86inputmouse to x11 modules

[abmantis]

Description of changes

Without this the X11 server would fail to start with the error:

(EE) Failed to load module "mouse" (module does not exist, 0)

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• For non-Linux: Is sandboxing enabled in nix.conf? (See Nix manual)
  • sandbox = relaxed
  • sandbox = true
• Tested, as applicable:
  • NixOS test(s) (look inside nixos/tests)
  • and/or package tests
  • or, for functions and "core" functionality, tests in lib/tests or pkgs/test
  • made sure NixOS tests are linked to the relevant packages
• Tested compilation of all packages that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage
• Tested basic functionality of all binary files (usually in ./result/bin/)
• 24.05 Release Notes (or backporting 23.05 and 23.11 Release notes)
  • (Package updates) Added a release notes entry if the change is major or breaking
  • (Module updates) Added a release notes entry if the change is significant
  • (Module addition) Added a release notes entry if adding a new NixOS module
• Fits CONTRIBUTING.md.

---

Add a 👍 reaction to pull requests you find important.

[puzzlewolf]

Hi, thanks for your PR :)

Can you please change the commit message and PR title to something that matches the commit conventions?

I don't have an Nvidia GPU, so I can't review the technical part of the PR, I hope someone (@abbradar?) comes along.

[abmantis]

@puzzlewolf anything else I could do to help this move forward?

[puzzlewolf]

Pinging @abbradar, you're the maintainer of the bumblebee package, it would be great if you could review this PR :)

If they don't turn up, @abmantis, you can find some other people (not too many :D) who recently worked on this or other bumblebee related packages (made a PR or reviewed or so) and kindly ask them for a review here. Also, you can post in PRs ready for review _.

[abmantis]

Hey @Mic92 , do you have a chance to try this?
I saw your change in the hardware repo and was hopping you could test this out.

[Mic92]

Hey @Mic92 , do you have a chance to try this? I saw your change in the hardware repo and was hopping you could test this out.

The only thing, I can test here, is if it evaluates. I don't have a dual-gpu setup with Nvidia. Only some big enterprise Nvidia GPU in a headless server.

[nixos-discourse]

This pull request has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/prs-ready-for-review/3032/3644

[anonimo0-0]

I have tested this, and it works. 👍

[abmantis]

@abbradar any chance to try this?

[t-gebauer]

I just tested this by rebasing your commit onto my current nixos-version --revision (42f4c8c).

I am using bumblebee and intel-virtual-output to use external monitors on a Thinkpad P52 with a Quadro P1000 GPU.
For me, this already works fine without your fix. I have no complains. (As long as I don't do anything weird, like trying to sleep or hibernate...)
Although, I can see the same error message in the log, every time I do something with the nvidia card: intel-virtual-output, or optirun glxgears.
But that has never bothered me.

Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (EE) Failed to load module "mouse" (module does not exist, 0)
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (WW) NVIDIA: No DRM device: Direct render devices found but none could be
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (WW) NVIDIA:     used.
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (WW) NVIDIA(0): Option "NoLogo" is not used
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (WW) Warning, couldn't open module mouse
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (EE) Failed to load module "mouse" (module does not exist, 0)
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (WW) Warning, couldn't open module libinput
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (EE) Failed to load module "libinput" (module does not exist, 0)
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (WW) Warning, couldn't open module libinput
Jun 15 17:56:15 A014 bumblebeed[1212]: [XORG] (EE) Failed to load module "libinput" (module does not exist, 0)

Now, with your change, the mentioned EE) Failed to load module "mouse" is indeed gone:

Jun 15 18:03:22 A014 bumblebeed[1201]: [XORG] (WW) Open ACPI failed (/var/run/acpid.socket) (No such file or directory)
Jun 15 18:03:22 A014 bumblebeed[1201]: [XORG] (WW) NVIDIA: No DRM device: Direct render devices found but none could be
Jun 15 18:03:22 A014 bumblebeed[1201]: [XORG] (WW) NVIDIA:     used.
Jun 15 18:03:22 A014 bumblebeed[1201]: [XORG] (WW) NVIDIA(0): Option "NoLogo" is not used
Jun 15 18:03:22 A014 bumblebeed[1201]: [XORG] (WW) Warning, couldn't open module libinput
Jun 15 18:03:22 A014 bumblebeed[1201]: [XORG] (EE) Failed to load module "libinput" (module does not exist, 0)

BUT now the mouse cursor on external monitors is lagging/jumping around. Only when moving. Hardly noticeable when moving slowly. But very noticeably when moving fast.
This sounds very much like multiple actors are calculating the mouse position differently.
And indeed, if I revert my non-standard mouse acceleration settings from flat to the default adaptive profile, the lag is gone:

#xinput set-prop 12 "libinput Accel Profile Enabled" 0 1 0 # flat
xinput set-prop 12 "libinput Accel Profile Enabled" 1 0 0 # adaptive

But I like the flat profile and don't yet know how to enable it in bumblebees xserver.

So, it appears, that in my case it was positive, that the mouse module did not load.

Now, I wonder if the mouse lag is also related to the other error in my logs:
(EE) Failed to load module "libinput" (module does not exist, 0)
Would the mouse lag be fixed if the libinput module would load?

Or, in my case, maybe I should try to stop x11 from even trying to load both the mouse and libinput modules? They do not appear to be necessary at all.