| 9 Oct 2025 |
connor (he/him) | Okay I think the CUDA 13 PR is ready: https://github.com/NixOS/nixpkgs/pull/437723
And by that I mean I'm exhausted and don't really want to think about it any more | 22:47:32 |
connor (he/him) | TL;DR: expect basically nothing in-tree to work with CUDA 13. If it does, rejoice! | 22:48:06 |
connor (he/him) | I'm currently running nixpkgs-review on x86_64-linux | 22:55:29 |
connor (he/him) | kill me cudaPackages_13.saxpy doesn't build
cuda13.0-saxpy> CMake Error in CMakeLists.txt:
cuda13.0-saxpy> Imported target "CUDA::cublas" includes non-existent path
cuda13.0-saxpy>
cuda13.0-saxpy> "/nix/store/96n5czdjq66csa28ml9s1kwa13xnsbdp-cuda13.0-cuda_nvcc-13.0.88/include/cccl"
cuda13.0-saxpy>
cuda13.0-saxpy> in its INTERFACE_INCLUDE_DIRECTORIES. Possible reasons include:
cuda13.0-saxpy>
cuda13.0-saxpy> * The path was deleted, renamed, or moved to another location.
cuda13.0-saxpy>
cuda13.0-saxpy> * An install or uninstall procedure did not complete successfully.
cuda13.0-saxpy>
cuda13.0-saxpy> * The installation package was faulty and references files it does not
cuda13.0-saxpy> provide.
| 23:04:06 |
connor (he/him) | nvcc.profile gets patched from SYSTEM_INCLUDES += "-isystem" "$(TOP)/$(_TARGET_DIR_)/include/cccl" $(_SPACE_) to SYSTEM_INCLUDES += "-isystem" "/nix/store/96n5czdjq66csa28ml9s1kwa13xnsbdp-cuda13.0-cuda_nvcc-13.0.88/include/cccl" $(_SPACE_) 🥴 | 23:32:04 |
connor (he/him) | https://github.com/NixOS/nixpkgs/pull/437723/commits/ffead29ec174980fbcc2ac610195f64328856705 | 23:42:31 |
| 10 Oct 2025 |
connor (he/him) | cuda-legacy is going to be such a pain in the ass if the roughly nine hours I just spent trying to build PyTorch against CUDA 11.4 is any indication | 23:25:40 |
connor (he/him) | (I was not successful; will resume trying with PyTorch 2.6 instead of 2.7 later) | 23:26:30 |
| 11 Oct 2025 |
Tristan Ross | Hey, connor (he/him) (UTC-7) & SomeoneSerge (back on matrix). Either of you wanna collab on getting Tenstorrent support into nixpkgs? I'm the only one working on it but I think since this is in a realm of AI, ML, and GPU-like computing, it would make sense to involve people already touching that stuff. | 02:29:45 |
connor (he/him) | I’d love to but I don’t have time :( | 15:37:38 |
Gaétan Lepage | FYI: I'm working on bumping onnx[runtime] in https://github.com/NixOS/nixpkgs/pull/450587
However, the build fails... More investigation needed. | 16:20:35 |
SomeoneSerge (back on matrix) | In reply to @rosscomputerguy:matrix.org Hey, connor (he/him) (UTC-7) & SomeoneSerge (back on matrix). Either of you wanna collab on getting Tenstorrent support into nixpkgs? I'm the only one working on it but I think since this is in a realm of AI, ML, and GPU-like computing, it would make sense to involve people already touching that stuff. YES! /looks at the calendar, lowers the volume/ yes, though very much part time god hiw do i learn to say no | 18:17:28 |
SomeoneSerge (back on matrix) | * YES! /looks at the calendar, lowers the volume/ yes, though very much part time god how do i learn to say no | 18:17:37 |
| 12 Oct 2025 |
Tristan Ross | Heh, it's not too much. Bulk of the reviewing is https://github.com/NixOS/nixpkgs/pull/444813 | 17:19:19 |
| 13 Oct 2025 |
Collin Arnett | Hello! Have ya'll run into this problem with the nvidia-runtime-container?
https://github.com/llm-d/llm-d/issues/117#issuecomment-2992256350
apparently there is a patch for it here https://github.com/NVIDIA/k8s-device-plugin/pull/1183/files | 15:41:41 |
Collin Arnett | * Hello! Have ya'll run into this problem with the nvidia-container-toolkit?
https://github.com/llm-d/llm-d/issues/117#issuecomment-2992256350
apparently there is a patch for it here https://github.com/NVIDIA/k8s-device-plugin/pull/1183/files | 15:52:49 |
SomeoneSerge (back on matrix) | connor (he/him) (UTC-7): 8am instead of 7, rsvp? | 16:01:29 |
connor (he/him) | Yes, 8am Pacific | 17:22:55 |
| @gmacon:matrix.org left the room. | 17:53:23 |
| 15 Oct 2025 |
danielrf | connor (he/him) (UTC-7): Hey, just fyi. This looks very similar to an issue we had fixed in jetpack-nixos: https://github.com/NixOS/nixpkgs/issues/451912 I can't recall if our fix was generic enough to also be applicable to the nixpkgs' nvidia-container-toolkit | 03:56:45 |
connor (he/him) | pain | 07:03:11 |
connor (he/him) | Yeah Jared had written a udevadm settle for some devices; when I refactored to use upstream’s container toolkit stuff I commented it out hoping waiting on the modprobe nvgpu service was enough | 07:05:36 |
connor (he/him) | Relevant PRs:
- https://github.com/anduril/jetpack-nixos/pull/317
- https://github.com/anduril/jetpack-nixos/pull/331
| 07:06:36 |
Gaétan Lepage | connor (he/him) (UTC-7) SomeoneSerge (back on matrix) Torch 2.9.0 was just released.
I'm working on the bump.
It requires libnvshmem_host.so.3. I never heard of OpenSHMEM before.
Do we have it already in nixpkgs? | 18:50:04 |
apyh | oh i have this in a fork, sec | 19:43:52 |
| Ari Lotter joined the room. | 19:44:18 |
Ari Lotter | (still me sorry, diff devices w bad key management) | 19:44:26 |
Ari Lotter | https://github.com/PsycheFoundation/psyche/blob/main/nix/nvshmem.nix | 19:44:30 |
Ari Lotter | we don't have nvshmem in nixpkgs | 19:44:35 |
Ari Lotter | i'm using this with torch 2.9.0 :) | 19:45:38 |