lib.options: several small performance cleanups

[llakala]

I'm getting fast at this! Most of these are very small savings, but 0.1% of 4 million function calls is still 4 thousand function calls saved. Hash is the same before/after for both a minimal NixOS config and my own. Stats diff:

NIX_SHOW_STATS=1 \
nix-instantiate -I nixpkgs=. --eval -E '
  let nixos = import ./nixos/lib/eval-config.nix {
    modules = [
      ./nixos/modules/profiles/minimal.nix
      { 
        fileSystems."/" = { device = "/dev/sda1"; fsType = "ext4"; }; 
        boot.loader.grub.devices = ["/dev/sda"];
      }
    ];
  }; in nixos.config.system.build.toplevel.drvPath
'

{
  "attrset": {
    "lookups": "-0.11%",
    "merges": "-0.69%",
    "mergeCopies": "-0.09%"
  },
  "list": {
    "concats": null
  },
  "parser": {
    "expressions": null
  },
  "memory": {
    "envs": "-0.25%",
    "list": "-0.1%",
    "sets": "-0.07%",
    "symbols": null,
    "values": false,
    "total": "-0.12%"
  },
  "speed": {
    "primops": "-0.07%",
    "functionCalls": "-0.1%",
    "thunksMade": "-0.1%",
    "thunksAvoided": "-0.22%"
  }
}

Things done

• Built on platform:
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• Tested, as applicable:
  • NixOS tests in nixos/tests.
  • Package tests at passthru.tests.
  • Tests in lib/tests or pkgs/test for functions and "core" functionality.
• Ran nixpkgs-review on this PR. See nixpkgs-review usage.
• Tested basic functionality of all binary files, usually in ./result/bin/.
• Nixpkgs Release Notes
  • Package update: when the change is major or breaking.
• NixOS Release Notes
  • Module addition: when adding a new NixOS module.
  • Module update: when the change is significant.
• Fits CONTRIBUTING.md, pkgs/README.md, maintainers/README.md and other READMEs.

[hsjobeki]

lgtm nice improvements overall. Left some nits

[hsjobeki]

Not sure if i like the inlining for a performance win of what exactly? Can you measure how much we save here?
dry maintainable code vs inlining for performance reasons is a tradeoff in nix unfortunately.

[llakala]

Sure. With the test case of:

NIX_SHOW_STATS=1 NIX_SHOW_STATS_PATH=after.json \
 nix-instantiate -I nixpkgs=. --eval -E '
    let nixos = import ./nixos/lib/eval-config.nix {
      modules = [ ./nixos/modules/profiles/minimal.nix
        { fileSystems."/" = { device = "/dev/sda1"; fsType = "ext4"; };
        boot.loader.grub.devices = ["/dev/sda"]; } ];
      }; in nixos.config.system.build.toplevel.drvPath
    '

Absolute stats diff:

{
  "attrset": {
    "lookups": -3754,
    "merges": -4118,
    "mergeCopies": -2059
  },
  "list": {
    "concats": 0
  },
  "parser": {
    "expressions": 2
  },
  "memory": {
    "envs": -197648,
    "list": 0,
    "sets": -131776,
    "symbols": 0,
    "values": -65872,
    "total": -395296
  },
  "speed": {
    "primops": 0,
    "functionCalls": -2059,
    "thunksMade": -4117,
    "thunksAvoided": -10295
  }
}

And with percentages enabled:

{
  "attrset": {
    "lookups": "-0.11%",
    "merges": "-0.65%",
    "mergeCopies": "-0.04%"
  },
  "list": {
    "concats": null
  },
  "parser": {
    "expressions": "+0%"
  },
  "memory": {
    "envs": "-0.19%",
    "list": null,
    "sets": "-0.06%",
    "symbols": null,
    "values": "-0.04%",
    "total": "-0.08%"
  },
  "speed": {
    "primops": null,
    "functionCalls": "-0.04%",
    "thunksMade": "-0.06%",
    "thunksAvoided": "-0.18%"
  }
}

-0.65% merges does seem nice - everything else is rather minor. Up to you.