!UNVBThoJtlIiVwiDjU:nixos.org

Staging

286 Members
Staging merges | Running staging cycles: https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+head%3Astaging-next+head%3Astaging-next-25.05 | Review Reports: https://malob.github.io/nix-review-tools-reports/99 Servers

Load older messages


SenderMessageTime
15 Sep 2025
@emilazy:matrix.orgemilyif so pls let me do it unless it's urgent13:48:46
@emilazy:matrix.orgemilycan do in a few hours13:48:52
@k900:0upti.meK900I'll try to run it13:50:23
@k900:0upti.meK900 @emily 13:53:13
@k900:0upti.meK900image.png
Download image.png
13:53:13
@emilazy:matrix.orgemilywelp13:54:29
@emilazy:matrix.orgemilyI guess nobody touched LLVM13:54:34
@emilazy:matrix.orgemilycheers13:54:36
@hexa:lossy.networkhexahttps://github.com/libexpat/libexpat/pull/103415:07:13
@hexa:lossy.networkhexaanyone fancy another round of xml parser changes for staging?15:07:29
@hexa:lossy.networkhexarelease is pending https://github.com/libexpat/libexpat/issues/103015:07:53
@connorbaker:matrix.orgconnor (he/him) (UTC-7)For staging rebuilds, how much time would you all say is spent waiting for x86_64-linux to churn through everything? For instance, if the build capacity for the platform were tripled or done on much larger machines, what would that mean for iteration?15:54:15
@emilazy:matrix.orgemilyit would mean very little, since it's not the bottleneck15:54:46
@emilazy:matrix.orgemilymore/beefier Macs would probably help, though even there the queue runner is preventing full utilization I believe15:55:13
@emilazy:matrix.orgemily switching off x86_64-darwin in a few months will help 15:55:24
@connorbaker:matrix.orgconnor (he/him) (UTC-7)Any idea if there are plans to promote aarch64-darwin to a higher tier of support?15:56:26
@emilazy:matrix.orgemilyit is de facto pretty high tier15:57:54
@emilazy:matrix.orgemily blocks nixpkgs-unstable etc. 15:58:03
@emilazy:matrix.orgemily aarch64-linux is nominally more blocking and also easier to support, but probably has significantly fewer users 15:58:35
@emilazy:matrix.orgemilyif you mean the list of tier assignments in the RFC, that's explicitly listed as only being an informative guide at the time of writing rather than a fixed reality15:58:59
@emilazy:matrix.orgemilybut I would also say that the tiers RFC was never really fully implemented15:59:13
@emilazy:matrix.orgemily https://nixos.org/manual/nixpkgs/stable/#chap-platform-support continues to be the most accurate guide to platform support :) 15:59:37
@k900:0upti.meK900 @emily are we actually ready on GCC 15? 16:51:14
@emilazy:matrix.orgemilyI mean, nnno16:51:38
@emilazy:matrix.orgemilyprobably we never will be though16:51:40
@emilazy:matrix.orgemilyI was going to do some builds but if we're looking at next two days probably next cycle is the play16:51:54
@emilazy:matrix.orgemilyI am CMake 4ing though16:52:52
@emilazy:matrix.orgemilyFWIW https://github.com/NixOS/nixpkgs/pull/442977 and https://github.com/NixOS/nixpkgs/pull/443007 are CMake 4 and should be ready16:53:11
@emilazy:matrix.orgemilyoh wait no TBB is LLVM 2116:53:21
@emilazy:matrix.orgemilycan't keep these things straight16:53:28

Show newer messages


Back to Room ListRoom Version: 6