blob: e6f8836aa367f54f1c4bc524e70506ee4ac04c5a (
plain) (
blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
|
# Global configuration for the SSH client.
{ config, lib, pkgs, ... }:
with lib;
let
cfg = config.programs.turbovnc;
in
{
options = {
programs.turbovnc = {
ensureHeadlessSoftwareOpenGL = mkOption {
type = types.bool;
default = false;
description = ''
Whether to set up NixOS such that TurboVNC's built-in software OpenGL
implementation works.
This will enable <option>hardware.opengl.enable</option> so that OpenGL
programs can find Mesa's llvmpipe drivers.
Setting this option to <code>false</code> does not mean that software
OpenGL won't work; it may still work depending on your system
configuration.
This option is also intended to generate warnings if you are using some
configuration that's incompatible with using headless software OpenGL
in TurboVNC.
'';
};
};
};
config = mkIf cfg.ensureHeadlessSoftwareOpenGL {
# TurboVNC has builtin support for Mesa llvmpipe's `swrast`
# software rendering to implemnt GLX (OpenGL on Xorg).
# However, just building TurboVNC with support for that is not enough
# (it only takes care of the X server side part of OpenGL);
# the indiviudual applications (e.g. `glxgears`) also need to directly load
# the OpenGL libs.
# Thus, this creates `/run/opengl-driver` populated by Mesa so that the applications
# can find the llvmpipe `swrast.so` software rendering DRI lib via `libglvnd`.
# This comment exists to explain why `hardware.` is involved,
# even though 100% software rendering is used.
hardware.opengl.enable = true;
};
}
|