!eWOErHSaiddIbsUNsJ:nixos.org

NixOS CUDA

296 Members
CUDA packages maintenance and support in nixpkgs | https://github.com/orgs/NixOS/projects/27/ | https://nixos.org/manual/nixpkgs/unstable/#cuda58 Servers

You have reached the beginning of time (for this room).


SenderMessageTime
4 Nov 2025
@glepage:matrix.orgGaétan Lepage TIL: python3Packages.torchWithRocm is apprently sensitive to config.cudaSupport. 20:11:25
@arilotter:matrix.orgAri Lotterugh i wish we could compile packages with cudaCapabilities individually per-capability and merge them later, it's such a nightmare adding one new capability level and it causing a huge 8-hour recompile..20:40:40
@connorbaker:matrix.orgconnor (burnt/out) (UTC-8)These aliases must die, they make my life so difficult21:45:22
@connorbaker:matrix.orgconnor (burnt/out) (UTC-8)Join the club And it’s not even like we could do a mega-build in an intermediate derivation and then prune unused capabilities according to whatever the user requested because the amount of generated device code is so large linking will fail lmao21:46:17
@connorbaker:matrix.orgconnor (burnt/out) (UTC-8) Gaétan Lepage are any of SomeoneSerge (back on matrix)’s comments on https://github.com/NixOS/nixpkgs/pull/457803 actionable or is it good to merge? 21:48:00
@connorbaker:matrix.orgconnor (burnt/out) (UTC-8)Also, would you mind reviewing https://github.com/NixOS/nixpkgs/pull/458619?21:48:09
@hacker1024:matrix.orghacker1024

This is most likely due to a dependency, but I will also point out that all torch variants are at the moment due to an unconditional version access

https://github.com/NixOS/nixpkgs/blob/b3d51a0365f6695e7dd5cdf3e180604530ed33b4/pkgs/development/python-modules/torch/source/default.nix#L458

21:48:19
@glepage:matrix.orgGaétan Lepage I quadruppled check.
Both commits of my PR are actually necessary to get a nvcc-free onnxruntime.
21:48:42
@glepage:matrix.orgGaétan LepageLet me change one comment to mention the bisection21:48:57
@glepage:matrix.orgGaétan Lepage connor (burnt/out) (UTC-8), I reviewed nccl-tests. Feel free to merge 22:08:11
@arilotter:matrix.orgAri Lotter i'm trying to fix this exact linker error right now 😭 trying to get flash-attn built for cuda capabilities 7.5 thru 12.0a, and i'm so stuck, and every rebuild with an attempted fix takes ~2 hours... any ideas? 😭 22:17:28
@arilotter:matrix.orgAri Lottermaybe we're just screwed :)22:20:25
@sporeray:matrix.orgRobbie Buxton Which flash attention version 22:24:21
@sporeray:matrix.orgRobbie BuxtonV2 or v322:24:27

Show newer messages


Back to Room ListRoom Version: 9