| 28 Nov 2024 |
 SomeoneSerge (back on matrix) | My latest suspicion was broken TURN | 14:59:21 |
| SomeoneSerge (back on matrix) | Still haven't tried the official instance | 15:00:12 |
 sielicki | is the matrix video/voice stuff any good? | 15:00:50 |
| SomeoneSerge (back on matrix) | Latest I tried it only worked for me when in the same vpn, probably needed some extra TURN configuration too:) | 15:02:34 |
| Jitsi widget added by SomeoneSerge (back on matrix) | 15:02:42 |
| SomeoneSerge (back on matrix) | Let's try | 15:02:57 |
| SomeoneSerge (back on matrix) | https://pad.lassul.us/DwBgo1SZSzSVL29wsnK0fg# | 15:07:54 |
 @palasso:matrix.org | Last time I tried it, it wasn't good. Also letting you know that it notifies everyone in the room 😃 | 15:09:36 |
| sielicki | https://dev-discuss.pytorch.org/t/pytorch-deprecation-of-conda-nightly-builds/2590 | 15:27:04 |
| SomeoneSerge (back on matrix) | pkgs/development/cuda-modules/cuda/manifests/feature_12.4.0.json | 15:51:10 |
 connor (he/him) | https://github.com/ConnorBaker/cuda-packages/tree/main/scripts/cuda-redist | 16:00:55 |
| connor (he/him) | Version (I think) used to generate manifests which are currently in Nixpkgs: https://github.com/ConnorBaker/cuda-redist-find-features | 16:01:33 |
| SomeoneSerge (back on matrix) | Yeah idk how to keep things from going overtime | 16:23:28 |
| SomeoneSerge (back on matrix) | I wonder if we should consider starting like monthly office hours though | 16:24:05 |
| sielicki | i didn't mind at all, was fun to chat | 16:32:22 |
| connor (he/him) | SomeoneSerge (utc+3): I'm reaaaaally interested in extending the *Platform attributes so they include CUDA support and capabilities; do you have a link to Ross' PR changing the platform matcher functionality handy? | 16:51:43 |
| SomeoneSerge (back on matrix) | https://github.com/NixOS/nixpkgs/pull/348192 | 16:53:32 |
| sielicki | 
Download Screenshot from 2024-11-28 09-40-11.png | 17:42:24 |
| sielicki | lol | 17:42:25 |
| sielicki | * lol, compiling torch be like | 17:42:43 |
| SomeoneSerge (back on matrix) | In reply to @ss:someonex.net
Let's try I'm back. connor (he/him) (UTC-7): I added the "People" section to the pad. I'm sure I missed a lot but that's to be expected. If you can think of any names who aren't on any teams but are active? | 18:30:01 |
 nim65s | Only 35G ? that's fine ! | 18:36:50 |
| nim65s changed their display name from Guilhem to nim65s. | 18:36:58 |
 Gaétan Lepage | In reply to @sielicki:matrix.org
lol, compiling torch be like I feel you... | 20:43:45 |
| connor (he/him) | In reply to @ss:someonex.net
I'm back. connor (he/him) (UTC-7): I added the "People" section to the pad. I'm sure I missed a lot but that's to be expected. If you can think of any names who aren't on any teams but are active? you got everyone I can think of | 21:10:43 |
 hexa | In reply to @sielicki:matrix.org
sent an image. red is not the load I'd like to see 😄 or do you have an odd theme? | 21:27:02 |
| 29 Nov 2024 |
|  shekhinah changed their profile picture. | 10:05:12 |
| shekhinah changed their profile picture. | 10:08:49 |
| 30 Nov 2024 |
| sielicki | pkgs/os-specific/linux/nvidia-x11/open.nix presently does not ship Module.symvers in $dev, which is necessary to package other out-of-tree modules that want to consume symbols from it.
creating the output and installing it is simple enough, but it's not clear to me how that ought to be consumed. If another kmod derivation takes nvidia_x11 as an input, it's just always going to point to nvidiaPackages.stable.
is it correct that I'd need to essentially put it as an attribute nested beneath nvidia_x11? ie: adjacent to nvidia_x11.open?
| 10:31:25 |
| sielicki | * pkgs/os-specific/linux/nvidia-x11/open.nix presently does not ship Module.symvers in $dev, which is necessary to package other out-of-tree modules that want to consume symbols from it.
creating the output and installing it is simple enough, but it's not clear to me how that ought to be consumed. If another kmod derivation takes nvidia_x11 as an input, it's just always going to point to nvidiaPackages.stable.
is it correct that I'd need to essentially put any consuming kmod as an attribute nested beneath nvidia_x11? ie: adjacent to nvidia_x11.open?
| 10:39:18 |
