Fix PR #193175

[roberth]

Description of changes

Unblock channel.

Fixes error introduced as #193175 and cleans up the config mess.

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• For non-Linux: Is sandbox = true set in nix.conf? (See Nix manual)
• Tested, as applicable:
  • NixOS test(s) (look inside nixos/tests)
  • and/or package tests
  • or, for functions and "core" functionality, tests in lib/tests or pkgs/test
  • made sure NixOS tests are linked to the relevant packages
• Tested compilation of all packages that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage
• Tested basic functionality of all binary files (usually in ./result/bin/)
• 22.11 Release Notes (or backporting 22.05 Release notes)
  • (Package updates) Added a release notes entry if the change is major or breaking
  • (Module updates) Added a release notes entry if the change is significant
  • (Module addition) Added a release notes entry if adding a new NixOS module
  • (Release notes changes) Ran nixos/doc/manual/md-to-db.sh to update generated release notes
• Fits CONTRIBUTING.md.

[roberth]

@ofborg test boot.biosCdrom

[roberth]

tests.boot.biosCdrom on aarch64-linux Skipped — No attempt

Expected because

// optionalAttrs (pkgs.stdenv.hostPlatform.system == "x86_64-linux") {
    biosCdrom = makeBootTest "bios-cdrom" {

[alyssais]

Ugh, sorry about that. OfBorg was happy.

[roberth]

Ugh, sorry about that. OfBorg was happy.

Happens to all of us.

[alyssais]

So, do I understand correctly that config here was just the config value passed in by the module system (the one containing the final values of all NixOS options), and that it ended up just shadowing itself? And wasn't actually used in at least some of the functions it was passed into? Trying to follow what this did before makes my head hurt.

[unbel13ver]

Sorry for introducing this bug! I actually tested that with
$ nix-shell -p nixos-generators --run "nixos-generate --format iso --configuration ./iso.nix -o result" -I nixpkgs=<patched nixpkgs>
where iso.nix is:

{ pkgs, modulesPath, lib, ... }: {
  imports = [
    "${modulesPath}/installer/cd-dvd/installation-cd-base.nix"
  ];
  boot.loader.grub.forcei686 = true;
  # use the latest Linux kernel
  boot.kernelPackages = pkgs.linuxPackages_latest;
  boot.supportedFilesystems = lib.mkForce ["ext4fs"];
}

[roberth]

So, do I understand correctly that config here was just the config value passed in by the module system (the one containing the final values of all NixOS options), and that it ended up just shadowing itself? And wasn't actually used in at least some of the functions it was passed into? Trying to follow what this did before makes my head hurt.

Correct. I've double-checked that config is only brought into scope 1. as the (top level) module's module argument, 2. in the two functions that always received the config module argument; and nowhere else.

[alyssais]

Thank you for the additional refactor to make this less confusing!

[roberth]

Sorry for introducing this bug! I actually tested that with $ nix-shell -p nixos-generators --run "nixos-generate --format iso --configuration ./iso.nix -o result" -I nixpkgs=<patched nixpkgs> where iso.nix is:

{ pkgs, modulesPath, lib, ... }: {
  imports = [
    "${modulesPath}/installer/cd-dvd/installation-cd-base.nix"
  ];
  boot.loader.grub.forcei686 = true;
  # use the latest Linux kernel
  boot.kernelPackages = pkgs.linuxPackages_latest;
  boot.supportedFilesystems = lib.mkForce ["ext4fs"];
}

I'm surprised that that worked. Maybe something else was also going on.
At least this was caught early. Very early in fact, because the error was discovered by Nix's static scope checking. 🤓