pcsx2: 1.7.5587 -> 1.7.5779

[SuperSamus]

Description of changes

Diff: PCSX2/pcsx2@v1.7.5587...v1.7.5779

Depends on #305508 (which is the base of this branch). Merged.

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• For non-Linux: Is sandboxing enabled in nix.conf? (See Nix manual)
  • sandbox = relaxed
  • sandbox = true
• 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/)
• 24.05 Release Notes (or backporting 23.05 and 23.11 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
• Fits CONTRIBUTING.md.

---

Add a 👍 reaction to pull requests you find important.

[nixos-discourse]

This pull request has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/prs-ready-for-review/3032/3849

[SuperSamus]

Waiting to know whether it's better to patch shaderc inside the package or in all-packages. (#309064)

[hrdinka]

@SuperSamus thanks a lot for continuously updating this package! As always I am very happy about your contribution.

Regarding the shaderc patch: IMHO it should be overriden in all-packages.nix. There are many other packages doing it this way as well. The pcsx2 derivation should be agnostic of its input and not try to override them, it cannot know what exact derivations are passed as input. It should basically just ask for a shaderc package and be happy with whatever it gets. The caller should make sure to pass the right one. If a package needs a special version of an input, it is always done via callPackage in all-packages.nix and this here is not different.

Further a fetch* variant should be preferred and used whenever possible like in the duckstation PR to not bloat nixpkgs unnecessarily.

I would however not be happy about moving it to the bottom in all-packages.nix like it was proposed in the duckstation PR. I appreciate its good intentions but in the end this will mess up the ordering in all-packages.nix even more since I expect only few people to notice and adhere to it. In the light of RCF140, eventually all default package calls will be removed from all-packages.nix anyway.

[SuperSamus]

Rebased, done what was done with DuckStation, and backported upstream patches to fix startup crashes

(Current master requires some stuff that nixpkgs doesn't have yet, like an updated SDL2 and Shaderc. One step at a time, this PR has been open long enough already.)

[AndersonTorres]

Some minor cleanup.

[AndersonTorres]

Besides, LGTM

[AndersonTorres]

P.S.: we should have a team for emulators.

[SuperSamus]

Anything left to do before this can be merged?

#287169

[hrdinka]

Tested locally. Thanks to everyone and thanks @SuperSamus for maintaining pcsx2 🙂

I had to dismiss the outdated review which blocked the merge.