!eWOErHSaiddIbsUNsJ:nixos.org

NixOS CUDA

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

Load older messages


SenderMessageTime
29 Sep 2022
@FRidh:matrix.orgFRidhRight it's only a convenience for compilation, it's not like you somehow can make it a symlink.11:29:42
@FRidh:matrix.orgFRidhI misunderstood what is possible then.11:30:07
30 Sep 2022
@aidalgol:matrix.orgaidalgol
In reply to@aidalgol:matrix.org
I've tried setting hardware.nvidia.nvidiaPersistenced = true; in my NixOS config, and that seems to have resolved it so far.
Nope, still happening sporadically.
06:03:22
@0b11:matrix.org@0b11:matrix.org removed their display name 0x53.17:33:24
@0b11:matrix.org@0b11:matrix.org left the room.17:33:30
3 Oct 2022
@ss:someonex.netSomeoneSerge (back on matrix)Btw if anyone uses blender: opensubdiv (and thus blender) has been broken in nixos-22.05 for almost a month now, but got fixed in master and unstable recently. There's hopefully some commit that that hasn't reached the release channel but maybe could be backported...10:15:01
@spacesbot:nixos.devspacesbot - keeps a log of public NixOS channels joined the room.14:25:38
4 Oct 2022
@justbrowsing:matrix.orgKevin Mittman (EOY sleep) CUDA 11.8.0 is live. Notably linux-aarch64 tarballs for Jetson are available 17:58:28
5 Oct 2022
@ss:someonex.netSomeoneSerge (back on matrix)Nix usability study mentioning CUDA infra: https://discourse.nixos.org/t/usability-study-session-6/2143816:06:28
6 Oct 2022
@ss:someonex.netSomeoneSerge (back on matrix)
In reply to @justbrowsing:matrix.org
CUDA 11.8.0 is live. Notably linux-aarch64 tarballs for Jetson are available

Missing jetson pieces so far:

  • cudnn (needed e.g. for pytorch)
  • cuda_nvprof (needed e.g. for faiss)
20:19:44
7 Oct 2022
@justbrowsing:matrix.orgKevin Mittman (EOY sleep)
In reply to @ss:someonex.net

Missing jetson pieces so far:

  • cudnn (needed e.g. for pytorch)
  • cuda_nvprof (needed e.g. for faiss)
  • cuDNN, TensorRT, NCCL, etc. are separate product releases from CUDA SDK.
    • Still working on getting sign-off to ship Jetson tarballs in a future release of those products
20:42:25
@justbrowsing:matrix.orgKevin Mittman (EOY sleep)
  • Investigatingcuda_nvprof but seems support on linux-aarch64 platforms was dropped sometime back
    • Trying to find any public release notes that mention this
21:01:00
@ss:someonex.netSomeoneSerge (back on matrix)

Investigating cuda_nvprof

Actually, I found today that cuda_profiler.h isn't in cuda_nvprof anymore anyway

21:02:10
@justbrowsing:matrix.orgKevin Mittman (EOY sleep) Yes, cuda_profiler_api was split from cuda_nvprof in 11.8 21:03:46
@ss:someonex.netSomeoneSerge (back on matrix)Oh, I see. Thanks!21:04:05
8 Oct 2022
@bitmapper:amber.telbitmapper joined the room.07:51:46
9 Oct 2022
@mt-caret:matrix.orgmt-caret joined the room.01:57:57
14 Oct 2022
@tpw_rules:matrix.orgtpw_rulesanybody ever seen glxinfo straight up hang for liek 45 seconds at 100% cpu on nvidia drivers?21:16:32
15 Oct 2022
@tinybronca:sibnsk.netunderpantsgnome! changed their display name from underpantsgnome to underpantsgnome!.00:33:33
@aidalgol:matrix.orgaidalgol The worst I have ever experienced is one of the demo applications, like glxgears running at a horribly low framerate when the drivers are not properly installed. 19:19:02
@aidalgol:matrix.orgaidalgolBut then again, I think that was on Debian.19:19:14
@aidalgol:matrix.orgaidalgolAre you by any chance, running a version other than the "stable" driver version?19:19:42
16 Oct 2022
@ss:someonex.netSomeoneSerge (back on matrix)Never happened to me...02:31:06
@tpw_rules:matrix.orgtpw_rulesturns out it was because nvidia-persistenced wasn't running02:31:14
@ss:someonex.netSomeoneSerge (back on matrix)That sounds quite obscure02:31:47
@tpw_rules:matrix.orgtpw_ruleshttps://github.com/NixOS/nixpkgs/pull/195608 this may be of interest to some of you. would appreciate a review02:31:47
@lucernae:matrix.orgRizky Maulana Nugraha joined the room.23:23:13
22 Oct 2022
@tpw_rules:matrix.orgtpw_rulesanybody thought about how to have several versions of cutensor?01:33:16
@tpw_rules:matrix.orgtpw_rulesfor now there's only two old ones and they are auto-selected by what cuda toolkit. anyone remember why?01:33:28
@tpw_rules:matrix.orgtpw_rules FRidh: ^ 01:36:48

Show newer messages


Back to Room ListRoom Version: 9