linux_5_12: init at 5.12.2

[xaverdh]

Motivation for this change

Update to add the latest release

Things done

My system is happily running the kernel from this pr.

• Tested using sandboxing (nix.useSandbox on NixOS, or option sandbox in nix.conf on non-NixOS linux)
• Built on platform(s)
  • NixOS
  • macOS
  • other Linux distributions
• Tested via one or more NixOS test(s) if existing and applicable for the change (look inside nixos/tests)
• Tested compilation of all pkgs that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review wip"
• Tested execution of all binary files (usually in ./result/bin/)
• Determined the impact on package closure size (by running nix path-info -S before and after)
• Ensured that relevant documentation is up to date
• Fits CONTRIBUTING.md.

[xaverdh]

Should I also update linux_latest to point to the new version?

Side remark: Is there a script that automates this procedure (I couldn't find one)? would be cool to have one

[xaverdh]

cc @NeQuissimus (although he is probably still not available)
cc @TredwellGit @Atemu (might be interested, judging from the pr updating the stable kernels)

[moni-dz]

about zfs making ofborg fail eval: #121113

[xaverdh]

ah I missed #121262 somehow

But I think we can merge this anyway (independent of any zfs issues)? Not so sure about bumping linux_latest then though..

[davidak]

this pr seem better since it has some patches and a later version

it should be set as linux_latest (that's what it's for i guess), but add changelog for zfs problem.

[TredwellGit]

There should be something like linuxPackages_latest_zfs and linuxPackages_latest_zfsUnstable that linuxPackages_latest should be redefined to if ZFS is used.

[TredwellGit]

Provide rationale if you disagree.

[moni-dz]

@ofborg eval

[davidak]

Change looks good

Builds, works

I'm using the kernel the last hours. It seem more stable than previous versions.

[hmenke]

@TredwellGit zfs is a kernel module like any other. Or do you also want to have linuxPackages_latest_bbswitch_exfat_nvidia_x11_beta_virtualbox_wireguard_zfsUnstable?

[xaverdh]

Ok, this just adds the new kernel to nixpkgs currently. Can we merge it, so people can start using the new kernel?
I will open a separate pr, which points linux_unstable linux_latest at the new version, and adds release notes.

[xaverdh]

@hmenke I think not all kernel modules are equal in that zfs is important to many. But in the long term we might want to think this through a bit. Perhaps we can come up with some kind of kernelWithModules [...] infrastructure, which points to the latest kernel that supports all desired features or something like that.

[TredwellGit]

I am trying to suggest something that easily works now to keep ZFS working with linuxPackages_latest.

Do you mean linuxPackages_testing instead of linux_unstable?

[xaverdh]

I am trying to suggest something that easily works now to keep ZFS working with linuxPackages_latest.

Do you mean linuxPackages_testing instead of linux_unstable?

I meant linuxPackages_latest actually, sorry about that. I guess introducing linuxPackages_latest_zfs doesn't do any harm. At least if people are willing to keep an eye on it, so it doesn't lag behind. It would be really good to have an opinion by @NeQuissimus on that.

[xaverdh]

There should be something like linuxPackages_latest_zfs and linuxPackages_latest_zfsUnstable that linuxPackages_latest should be redefined to if ZFS is used.

oh by refined do you mean automatically, so that the meaning of linuxPackages_latest changes depending on if zfs was used? that would feel like a little too much magic to me..

[Atemu]

Can be merged as-is.

Discussion about ZFS' interaction with alinux_latest bump should be moved to #122598 or a completely separate issue IMO.