1 |
Please don't top-post. |
2 |
|
3 |
On Wednesday, May 20, 2020 1:37:45 PM CEST Victor Ivanov wrote: |
4 |
> When the lbglvnd flag was introduced I remember I solved this issue by: |
5 |
> |
6 |
> # emerge --unmerge eselect-opengl |
7 |
> # emerge -1qv mesa |
8 |
> |
9 |
> After that, a simple update of @world rebuilt everything else on its own. |
10 |
> |
11 |
> Personally, I had been waiting for libglvnd support for _a long time_. |
12 |
> This - and I mean GLVND in general - is something that should have come |
13 |
> to Linux many years ago, along with NVIDIAs PRIME render offloading. |
14 |
> |
15 |
> 10y ago I used to have an Optimus laptop with an Nvidia GPU and it was |
16 |
> an absolute hell to get it running, I remember writing tonnes of scripts |
17 |
> using VirtualGL and a dummy X server running on the Nvidia GPU. This was |
18 |
> before bumblebee. |
19 |
> |
20 |
> Today, I still need this with an external GPU. |
21 |
> |
22 |
> But now it takes 1 environment variable to offload to the other GPU! |
23 |
> GLVND literally made my Linux work experience a million times better. |
24 |
> I'm extatic. |
25 |
> |
26 |
> - V |
27 |
> |
28 |
|
29 |
It is precisely why I had to implement GLVND before it became stable (using |
30 |
external overlays even and manually adding patches) |
31 |
|
32 |
My new laptop doesn't work at all with bumblebee, because the external display |
33 |
ports are connected to the nvidia-chip and not to the intel-chip. |
34 |
If reverse-PRIME would be supported, I would be able to disable the nvidia- |
35 |
chip during 80% of the time and only enable it when playing games. |
36 |
|
37 |
-- |
38 |
Joost |