| 29 Oct 2025 |
connor (he/him) | Is that good? | 04:40:42 |
Gaétan Lepage | Yes and no, the "+125" is mostly the new stuff introduced by the CUDA 13 PR. | 08:48:19 |
connor (he/him) | This may be worth considering: https://github.com/garnix-io/yensid | 14:56:27 |
SomeoneSerge (back on matrix) | Neat, literally haproxy | 16:00:07 |
Gaétan Lepage | FYI, the update to the latest torch 2.9.0 was just merged!
Changelog: https://github.com/pytorch/pytorch/releases/tag/v2.9.0
We also just added torchcodec a media decoder/encoder working along with PyTorch | 23:19:33 |
| 30 Oct 2025 |
Gaétan Lepage | connor (he/him) (UTC-7) any idea what, in the CUDA 13 PR, could be responsible for gcc-wrapper leaking into the firefox derivation? | 21:42:34 |
Gaétan Lepage | ❮ nix-build --arg config '{ allowUnfree = true; cudaSupport = true; }' -A firefox
this derivation will be built:
/nix/store/7s1dyndj7r33wck6dywjf6kqxah7ngm2-firefox-144.0.2.drv
building '/nix/store/7s1dyndj7r33wck6dywjf6kqxah7ngm2-firefox-144.0.2.drv'...
structuredAttrs is enabled
error: output '/nix/store/b3cyrj28jf03mrf15jmy9whhfrgy2bxw-firefox-144.0.2' is not allowed to refer to the following paths:
/nix/store/x8mydcgbry214s802nzvy7fdljx404ym-gcc-wrapper-14.3.0
| 21:42:54 |
connor (he/him) | Check onnxruntime | 22:04:03 |
Gaétan Lepage | Before: | 22:11:06 |
Gaétan Lepage |  Download clipboard.png | 22:11:08 |
Gaétan Lepage | After: | 22:11:19 |
Gaétan Lepage |  Download clipboard.png | 22:11:21 |
Gaétan Lepage | Probably nccl's fault... | 22:18:18 |
| 31 Oct 2025 |
| connor (he/him) changed their profile picture. | 03:16:13 |
| connor (he/him) changed their display name from connor (he/him) (UTC-7) to connor (burnt/out) (UTC-7). | 03:16:45 |
connor (he/him) | gonna slow down and take a step back for a bit | 03:18:20 |
Daniel Fahey | You said upfront "The addition of CUDA 13 does not mean packages will suddenly work with CUDA 13. Expect breakages." I know I'm just a random bloke from GitHub an fairly new but I've had really bad burnout in the past, I'd suggest still doing a little bit of triaging and technical support here and there for the CUDA Team in strict time blocks, so you can at least see the fruits of your labour (given to the world, for free) as the breakages all get sorted out in the coming weeks in our collective efforts.
From my perspective I'm just excited about the prospect of using CUDA 13 with Nixpkgs, I've basicaly used nixos-unstable, sometimes master, since starting to use Nix, and still have no idea how the release cycle is supposed to work 🙃.
I reckon for the next big CUDA update, do something like how the haskell-updates branch gets merged into staging first.
| 10:26:33 |
Daniel Fahey | * You said upfront "The addition of CUDA 13 does not mean packages will suddenly work with CUDA 13. Expect breakages." I know I'm just a random bloke from GitHub and fairly new but I've had really bad burnout in the past, I'd suggest still doing a little bit of triaging and technical support here and there for the CUDA Team in strict time blocks, so you can at least see the fruits of your labour (given to the world, for free) as the breakages all get sorted out in the coming weeks in our collective efforts.
From my perspective I'm just excited about the prospect of using CUDA 13 with Nixpkgs, I've basicaly used nixos-unstable, sometimes master, since starting to use Nix, and still have no idea how the release cycle is supposed to work 🙃.
I reckon for the next big CUDA update, do something like how the haskell-updates branch gets merged into staging first.
| 10:26:44 |
connor (he/him) | I mean I’ll still be around, just not doing as much. I’ll still be in the team weeklies, etc. | 14:35:31 |
connor (he/him) | CUDA 13 isn’t the default because the stuff we have in tree is too old or doesn’t support it; the expect breakages was in reference to trying to use CUDA 13 as the default. | 14:36:20 |
connor (he/him) | Haskell stuff goes into staging (at least partly) because of the sheer number of packages, to allow Hydra to churn through them. None of our stuff is built upstream, so there’s not really a point. | 14:37:36 |
Robbie Buxton | I think also a fair amount of stuff upstream doesn’t even build with cuda 13 yet either | 14:37:54 |
connor (he/him) | Yeah NVIDIA does not care outside of projects they dedicate engineering hours to supporting, and changing the default version of OpenCV or other large projects to a commit from master adding support would be dead on arrival, and trying to special case it just for when CUDA is configured would be difficult. | 14:39:51 |
| Daniel Fahey set a profile picture. | 14:56:01 |
Daniel Fahey | This is quite a convincing argument to revert the 99 commits https://github.com/NixOS/nixpkgs/pull/437723#issuecomment-3472997390
Maybe there could be a cuda-refactor branch that is continually built and tested by e.g. https://hydra.nixos-cuda.org/jobset/nixpkgs/cuda-refactor while it gets the attention it deserves?
| 15:50:50 |
Daniel Fahey | (all a bit over my current pay grade with my limited Nixpkgs experience though, lol) just really want to express my gratitude to the CUDA Team | 15:52:21 |
Robbie Buxton | My understanding (which may be incorrect) is that CUDA 13 is opt in so will only break if you try and use it instead of the default? | 16:03:44 |
connor (he/him) | Gaétan LepageSomeoneSerge (back on matrix) are you okay with merging:
- https://github.com/NixOS/nixpkgs/pull/457338
- https://github.com/NixOS/nixpkgs/pull/457220
I’d like there to be consensus as a team for those reverts to go through. Serge, I know you’re in favor of the config.cudaSupport one, but I’d like to issue the statement/decision as a team.
| 19:40:25 |
connor (he/him) | Correct | 19:46:10 |
connor (he/him) | We don’t have anywhere near the capacity (hardware or labor) to do that on a regular cadence, but that would be nice | 19:47:00 |