!djTaTBQyWEPRQxrPTb:nixos.org

Nixpkgs Architecture Team

227 Members
https://github.com/nixpkgs-architecture, weekly public meetings on Wednesday 15:00-16:00 UTC at https://meet.jit.si/nixpkgs-architecture52 Servers

Load older messages


SenderMessageTime
20 Dec 2022
@k900:0upti.meK900
In reply to @toddgamblin:matrix.org
Are there docs somewhere about how nixpkgs handles CI as contributions come in on github?
#ofborg:nixos.org
17:46:24
@piegames:matrix.org@piegames:matrix.orgMay I request some feedback for https://github.com/NixOS/rfcs/pull/127 ?21:52:02
23 Dec 2022
@hariamoor:matrix.orgHari joined the room.13:40:08
24 Dec 2022
@infinisil:matrix.orginfinisil piegames: Left some feeback 03:26:08
@infinisil:matrix.orginfinisil * piegames: Left some feedback 04:11:57
@piegames:matrix.org@piegames:matrix.orgThanks09:00:03
27 Dec 2022
@zzantares:matrix.org@zzantares:matrix.org joined the room.18:02:11
@zzantares:matrix.org@zzantares:matrix.org changed their display name from Julio César to zzantares.18:16:52
29 Dec 2022
@growpotkin1:matrix.orggrowpotkin ( Alex Ameen )

I made this benchmark for the folks at dream2nix, and it likely isn't "big news" to anyone; but for trivial builds derivation can be ~2x faster than stdenv.mkDerivation.

I ran 1,000,000 trivial builds that just touch "$out";. These use preferLocalBuild = true; and allowSubstitutes = false; for all drvs.
https://github.com/aameen-tulip/benchmark-mkderivation

I know "touch isn't really fair"; but it's just meant to try and find out exactly how much runtime is spent processing setup.sh ( ~0.01s/build from what I can tell if you only run installPhase ).

17:47:19
2 Jan 2023
@nbp:mozilla.orgnbp changed their display name from nbp to nbp | {pto until 2023-01-11}.09:46:25
@infinisil:matrix.orginfinisil growpotkin ( Alex Ameen ): Reminder to comment your findings regarding readDir speed in https://github.com/nixpkgs-architecture/simple-package-paths/issues/17 :) 14:14:44
@infinisil:matrix.orginfinisil @room: The next meeting will take place in about 15 minutes, we'll continue discussing https://github.com/nixpkgs-architecture/simple-package-paths - meeting link - live stream - meeting notes 15:16:23
@infinisil:matrix.orginfinisil John Ericson: Joining? 15:33:30
@infinisil:matrix.orginfinisilWhoops, laptop wasn't connected to power16:23:26
@infinisil:matrix.orginfinisil Good timing though, I'll see you next week 16:24:09
3 Jan 2023
@guydoodmanbro:matrix.orgErnest joined the room.21:45:41
@guydoodmanbro:matrix.orgErnest changed their display name from guydoodmanbro to Ernest.21:46:33
4 Jan 2023
@rewina:matrix.orgrewine joined the room.02:16:03
@enemycube:matrix.orgenemycube joined the room.14:27:16
7 Jan 2023
@aktaboot:tchncs.deaktaboot changed their profile picture.20:27:23
9 Jan 2023
@infinisil:matrix.orginfinisil @room: The next meeting will take place in ~8 minutes, we'll continue discussion issues in https://github.com/nixpkgs-architecture/simple-package-paths - meeting link - live stream - meeting notes 15:22:40
@infinisil:matrix.orginfinisil John Ericson: growpotkin ( Alex Ameen ): Joining? 15:31:54
@growpotkin1:matrix.orggrowpotkin ( Alex Ameen )yes15:32:02
@Ericson2314:matrix.orgJohn Ericsonsorry! will be there in < 3 min15:32:16
@infinisil:matrix.orginfinisilLooks like the jitsi recording messed up too, there's no audio and constant video in the last ~10 minutes17:11:21
@infinisil:matrix.orginfinisil Robert Hensing (roberth): Regarding that reference requirement, I think the arguments here are pretty good for keeping it: https://github.com/nixpkgs-architecture/simple-package-paths/blob/master/README.md#relax-design-to-try-to-attack-issues-like-package-variants-up-front 17:24:56
@roberthensing:matrix.orgRobert Hensing (roberth)are you sure? I just added the argumentation here https://github.com/nixpkgs-architecture/simple-package-paths/pull/27/files17:26:08
@infinisil:matrix.orginfinisilAnd especially since the new convention won't be applicable to all packages, I don't think it makes sense to encourage the new convention to be applied manually17:26:10
@infinisil:matrix.orginfinisil Robert Hensing (roberth): Ah nice I'll read through that 17:26:34
@roberthensing:matrix.orgRobert Hensing (roberth)
In reply to @infinisil:matrix.org
And especially since the new convention won't be applicable to all packages, I don't think it makes sense to encourage the new convention to be applied manually
tell everyone not to add to pkgs/unit?
17:27:06

Show newer messages


Back to Room ListRoom Version: 9