linux_latest-libre: 19812 -> 19835#418293
Conversation
This comment was marked as outdated.
This comment was marked as outdated.
You should be able to test this via the kernel-generic test at least fwiw. But then... how long is this broken now? |
|
I'd prefer to just merge a fix considering the update script was sufficient. I just upgraded it since I took a look at the contents. |
alyssais
left a comment
There was a problem hiding this comment.
We might as well get this merged in the meantime even if the outcome of the discussion is to drop it. FWIW: I added linux-libre to Nixpkgs many years ago, and I have no objections to dropping it.
|
fwiw for me, the blocking part was that noone else was interested enough to verify that it boots (which is what I did yesterday, but got distracted by something else for GPN-reasons, so I didn't give feedback). |
|
Alyssa: the ❌ is for splitting the commit and nothing else, correct? |
f4a9a3d to
b0ec9fb
Compare
|
I've addressed the commit-splitting feedback, ran the update script which yielded a new revision, and ran the |
This comment was marked as outdated.
This comment was marked as outdated.
b0ec9fb to
4cb34ef
Compare
|
|
|
Successfully created backport PR for |
Also, update the phases to run their hooks, and the hash to be SRI, and the update script to support that.
This fixes the build. I don't have great passion for Linux Libre; I just wanted to see fewer of these build failures being "expected". Though I did not attempt to boot the kernel on a real machine, I did run the
nixosTests.kernel-generic.linux_libretest and the machine does in fact boot and do the needful.Things done
nix.conf? (See Nix manual)sandbox = relaxedsandbox = truenix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)