| 19 Feb 2024 |
 raitobezarius | isn't this patch something K900 worked on? | 00:21:49 |
 @elvishjerricco:matrix.org | In reply to @raitobezarius:matrix.org
isn't this patch something K900 worked on? that would not surprise me but I have not kept up with that story at all | 03:06:18 |
 K900 | That specific patch is @pedrohlc ⚡️ based on the same idea | 05:48:14 |
|  O33 joined the room. | 12:44:19 |
 pedrohlc ⚡️ | Yeah, K900 brought the idea of “we should make libgbm a loader-only lib”, meanwhile as I couldn't do that at in my repo, I made that patch that “changes the default-fallback backend to something different” | 12:57:40 |
| K900 | I do still want to fix that | 12:59:44 |
| K900 | Upstream | 12:59:48 |
| K900 | But need spoons | 12:59:55 |
| pedrohlc ⚡️ | I brought other ideas to the table before, like putting mesa's hash as the GBM backend's name, this would make it compatible with NixOS and non-NixOS scenarios, and support all different builds -- but hell that's yet another hack | 13:00:02 |
| K900 | I mean technically we don't even have to do that | 13:00:41 |
| K900 | We can just load the libgbm impurely always | 13:00:50 |
