!RROtHmAaQIkiJzJZZE:nixos.org

NixOS Infrastructure

391 Members
Next Infra call: 2024-07-11, 18:00 CEST (UTC+2) | Infra operational issues backlog: https://github.com/orgs/NixOS/projects/52 | See #infra-alerts:nixos.org for real time alerts from Prometheus.121 Servers

Load older messages


SenderMessageTime
28 Feb 2025
@hexa:lossy.networkhexawhat kind of package are we talking here?22:06:04
@sandro:supersandro.deSandro ^ lun 22:55:06
@lt1379:matrix.orgLuncomposable_kernel, part of ROCm. not tests, all ridiculously templated cpp files for multiple gfx arches23:36:00
@tom:dragar.deTomhttps://github.com/ROCm/composable_kernel23:42:13
@lt1379:matrix.orgLunIf that large a single build is likely to cause problems for hydra I can try to split it up. It's likely to be a significant hassle and make the package harder to keep up to date but should be possible.23:43:16
1 Mar 2025
@sandro:supersandro.deSandroso like boost but even worse? 😂01:07:53
@hexa:lossy.networkhexawondering if it wouldn't exceed the output limit01:08:33
@hexa:lossy.networkhexaof around 3.7GB right now01:08:37
@hexa:lossy.networkhexalots of rocm things have been split up for that reason01:08:53
@lt1379:matrix.orgLunOutput size is ~2.5G01:14:06
@hexa:lossy.networkhexalet's give it a try then01:15:40
@hexa:lossy.networkhexasend a PR, and mention it here01:15:58
@lt1379:matrix.orgLunIt's part of https://github.com/NixOS/nixpkgs/pull/36769501:28:21
@hexa:lossy.networkhexathanks for taking care of that ecosystem 🙂 01:29:56
@hexa:lossy.networkhexapaused unstable-small due to https://github.com/NixOS/nixpkgs/pull/38587903:52:38
@hexa:lossy.networkhexa* cancelled unstable-small due to https://github.com/NixOS/nixpkgs/pull/38587903:52:52
@hexa:lossy.networkhexaok, revert has been merged, evaling unstable-smol05:02:43
@hexa:lossy.networkhexatrying on a 40C machine and I'm at 8% after 2h40m 😄 05:16:25
@hexa:lossy.networkhexawe05:16:32
@hexa:lossy.networkhexa* we'll have to see how this plays out on hydra05:16:39
@hexa:lossy.networkhexawe have epyc 9454p (48C/96T) and schedule 4 big-parallel jobs with 24 "cores"05:18:04
@hexa:lossy.networkhexa* we have epyc 9454p (48C/96T) and schedule up to 4 big-parallel jobs with 24 "cores"05:18:15
@hexa:lossy.networkhexa
  nix.settings = {
    cores = 24;
    max-jobs = 4;
    system-features = [ "big-parallel" ];
  };
05:18:29
@hexa:lossy.networkhexaRedacted or Malformed Event05:19:41
@hexa:lossy.networkhexa

[40 jobs | 9% 389/4212 @ 0.0/s | 2:46:42 | ETA 27:18:26 ]

05:26:06
@hexa:lossy.networkhexaI mean, these are Skylake Cores, so I'm not sure what that means 😄 05:28:22
@hexa:lossy.networkhexahttps://www.intel.com/content/www/us/en/products/sku/123550/intel-xeon-silver-4114-processor-13-75m-cache-2-20-ghz/specifications.html05:28:41
@hexa:lossy.networkhexaI sure hope that package won't surprise me in python-updates one day05:29:14
@vcunat:matrix.orgVladimír Čunát If it's 4h with 128 threads, I don't think it will fit into the 10h default limit in our current infra config. (--cores 24 for big-parallel mentioned here; I didn't verify now) 06:57:47
@vcunat:matrix.orgVladimír ČunátBut that's easy to override per job.06:58:34

Show newer messages


Back to Room ListRoom Version: 6