haskell.compiler: stop reliance on LLVM 6 & 7; haskell.compiler.ghc884: remove at 8.8.4; llvmPackages_7: remove at 7.1.0

[sternenseemann]

Description of changes

cc @RaitoBezarius

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/)
• 23.11 Release Notes (or backporting 23.05 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.

[RaitoBezarius]

Looks good to me, thank you.

[sternenseemann]

cc @amjoseph-nixpkgs for GHC 8.6.5 changes

[sternenseemann]

Also I'd like someone to look over the alias stuff and suggest something for the release notes. Maybe it's easier to do the LLVM removals separately?

[maralorn]

LGTM. Just curious, why are we keeping LLVM 6?

[sternenseemann]

@ofborg build maude, root5

[sternenseemann]

Can someone test this for maude/root5 on darwin?

diff --git a/pkgs/top-level/all-packages.nix b/pkgs/top-level/all-packages.nix
index 3ac47ca930d5..550200c6e8fa 100644
--- a/pkgs/top-level/all-packages.nix
+++ b/pkgs/top-level/all-packages.nix
@@ -17733,7 +17733,7 @@ with pkgs;
   ### End of CuboCore
 
   maude = callPackage ../development/interpreters/maude {
-    stdenv = if stdenv.cc.isClang then llvmPackages_7.stdenv else stdenv;
+    stdenv = if stdenv.cc.isClang then llvmPackages_6.stdenv else stdenv;
   };
 
   me_cleaner = callPackage ../tools/misc/me_cleaner { };
@@ -40003,7 +40003,7 @@ with pkgs;
 
   root5 = lowPrio (callPackage ../applications/science/misc/root/5.nix {
     inherit (darwin.apple_sdk.frameworks) Cocoa OpenGL;
-    stdenv = if stdenv.cc.isClang then llvmPackages_7.stdenv else stdenv;
+    stdenv = if stdenv.cc.isClang then llvmPackages_6.stdenv else stdenv;
   });
 
   rinetd = callPackage ../servers/rinetd { };

[veprbl]

No objection to root5 change from me.

[ghost]

cc @amjoseph-nixpkgs for GHC 8.6.5 changes

I have no objections.

The contraption I hacked together to use ancient (gentoo?) powerpc64le binaries to bootstrap ghc just isn't sustainable.

We need to invest in being able to cross-compile ghc (i.e. build a build!=(host==target) compiler). Any other strategy for non-x86_64/aarch64 platforms isn't a good use of resources in the long run.

[sternenseemann]

LLVM 6 remains to be removed, but that requires a bigger effort since a lot of stuff now uses LLVM 6 post LLVM-5-removal.