Compiling the Linux kernel from scratch takes over an hour on this laptop. Given I'm tracking the unstable channel on a rolling distro means doing that several times a week. Ain't nobody got time for that. Or at least I don't...
I don't have resources to locally build an optimized kennel for every update for each of my systems, thus my interests in keeping with a community cache.
I didn't do much here, just swapped the kernel via config and ran some benchmarks. I posted as I was more curious to hear of what engineering trade offs may be at play, and what experience folks have had in daily driving CatchyOS's kernel patch sets.
I wasn't sure if these results were specific to my older hardware, or more generalizable to newer systems.
Indeed perhaps. I'd really like to see something like a Content-addressed Storage model from snix get upstreamed. That could really revolutionize bandwidth requirements for nix package updates and store sizes. Imagine downloading only the binary diffs for updates to packages like large modern browsers, IDEs, election apps, etcs, that mostly share common files across versions and packages. Suddenly, daily driving a rolling distro on the bleeding edge would be no more taxing on networks or disks than your average infrequent Debian update.
I haven't gotten around to testing this yet, but I think they just share the Android app proot scaffolding. I wasn't too satisfied with the proot performance on Android in general using Termux, as with interpreted runtimes like python, the system call overhead really shows itself. That of course doesn't change here with nix-on-droid.
Sounds like their target customers are commercial service providers and industrial device manufacturers using NixOS, but need to support stable deployments for longer than NixOS's current 6 month release cycle. Probably not to applicable for individual desktop users, but they still plan on hosting a public channel and cache for newer releases.
IBM acquired the company behind Fedora Linux. Although I doubt they had any corporate influence in the production of this particular vlog.
I wonder if they would have had better luck with Kubuntu or KDE Neon? Perhaps if they're still on KDE by the time KDE releases their Arch distro variant, they could test that too on their hardware.
Given the amount of trouble shooting demonstrated throughout the video, don't think the marketing was entirely positive...
Appreciate the detailed context, and thank you for your work!
The comment thread sparked a few new questions for me.
That will likely soon stop working because steam-run is no longer a grab bag for literally every library out there.
What recently changed with steam-run?
Is using pkgs.appimageTools.defaultFhsEnvArgs with pkgs.buildFHSUserEnv a suitable alternative?
#!/usr/bin/env nix-shell
{ pkgs ? import <nixpkgs> { } }:
(
let base = pkgs.appimageTools.defaultFhsEnvArgs; in
pkgs.buildFHSUserEnv (base // {
name = "FHS";
targetPkgs = pkgs: (with pkgs; [
/* add additional packages here e.g */
pcre
tzdata
]);
runScript = "bash";
extraOutputsToInstall = [ "dev" ];
})
).env
Appreciate the heads up. Looks like they use merge bots to auto update the package version JSON files for git packages, making for a very large/frequent commit history. Was that what made bisecting imposable?
I also see they pin the nixpkgs input, but do others normally modify that nixpkgs input to follow their global nixpkgs from their own system flake, or does that invalidate the use of Nyx community cache?