!ayCRiZriCVtuCUpeLp:nixos.org

Nix Cross Compiling

576 Members
128 Servers

Load older messages


SenderMessageTime
2 Dec 2024
@kjeremy:matrix.orgkjeremy Artturin: it's crazy to me how easily it is to get the evaluator to OOM with small changes on cross. I would expect memory usage to be linear wrt non-cross but it's definitely not 15:26:27
@rick:matrix.ciphernetics.nlRick (Mindavi)I think it instantiates nixpkgs x amount of times15:28:34
@hexa:lossy.networkhexaand don't try python things 🙈16:12:19
@kjeremy:matrix.orgkjeremy hexa: the python thing is killing me with nix-ros-overlay 16:13:09
@hexa:lossy.networkhexabut you surely remember the issue 😛16:13:09
@hexa:lossy.networkhexasowwy16:13:29
@hexa:lossy.networkhexauntangling this mess goes over my head16:13:42
@xin:xiny.lixin joined the room.16:20:21
@xin:xiny.lixin set a profile picture.16:27:15
@aliarokapis:matrix.orgAlexandros Liarokapis
In reply to @kjeremy:matrix.org
hexa: the python thing is killing me with nix-ros-overlay
Which python thing?
16:49:36
@artturin:matrix.orgArtturinhttps://github.com/NixOS/nixpkgs/blob/9763b802cea94c20b3b85ea408bf5bf4dd1fcaf1/pkgs/development/interpreters/python/cpython/default.nix#L13516:51:47
@kjeremy:matrix.orgkjeremyyep. the corresponding issue: https://github.com/NixOS/nixpkgs/issues/33823116:53:47
@artturin:matrix.orgArtturinThe comment in the code is wrong because it's not interacting with splicing in any way16:54:53
@artturin:matrix.orgArtturin * The comment in the code is wrong because it's not interacting with splicing from splice.nix in any way16:55:19
@artturin:matrix.orgArtturinit's kind of reimplementing splicing16:55:27
@artturin:matrix.orgArtturin * it's kind of reimplementing splicing Override doesn't preserve splicing anyways16:56:13
@szlend:matrix.orgszlendif pkgs was already spliced, it would :> 18:48:38
@artturin:matrix.orgArtturinIt wouldn't18:48:57
@szlend:matrix.orgszlend Are you saying this wouldn't work? foo.override { bla = pkgs.__splicedPackages.bla; } 18:50:29
@artturin:matrix.orgArtturin
In reply to @artturin:matrix.org
it's kind of reimplementing splicing
Override doesn't preserve splicing anyways
It doesn't preserve __spliced because those are not a part of the derivation/package, instead they're just tacked on to package attrset
18:51:16
@artturin:matrix.orgArtturin x.override ...) ? __spliced == false 18:51:49
@artturin:matrix.orgArtturin * __splicedPackages.x.override ...) ? __spliced == false 18:51:56
@szlend:matrix.orgszlendoh right, that makes sense18:52:15
@szlend:matrix.orgszlend (pkgs.splicePackages { pkgsBuildBuild = pkgs.pkgsBuildBuild.override { ... }, ... }).foo :') 18:53:43
@szlend:matrix.orgszlend * (pkgs.splicePackages { pkgsBuildBuild = pkgs.pkgsBuildBuild.foo.override { ... }, ... }).foo :') 18:53:56
@artturin:matrix.orgArtturinNah too cumbersome18:54:53
@artturin:matrix.orgArtturinhttps://github.com/NixOS/nixpkgs/blob/cbefb908e709b0579a3aca712614569952fec10e/lib/customisation.nix#L8818:55:04
@artturin:matrix.orgArtturinFunc like this 18:55:30
@artturin:matrix.orgArtturinlib.overrideAttrs18:55:36
@szlend:matrix.orgszlend wait but doesn't .override call this exact function? 18:57:55

Show newer messages


Back to Room ListRoom Version: 6