packaging: Build without symbolic interposition on GCC

This turns out to be a big problem for performance of Bison
generated code, that for whatever reason cannot be made internal
to the shared library. This causes GCC to make a bunch of function
calls go through PLT. Ideally these hot functions (like move/copy ctor) could become
inline in upstream Bison. That will make sure that GCC can do interprocedular
optimizations without -fno-semantic-interposition [^]. Considering that
LLVM already does inlining and whatnot is a good motivation for this change.
I don't know of any case where Nix relies on LD_PRELOAD tricks for the shared
libraries in production use-cases.

[^]: https://maskray.me/blog/2021-05-09-fno-semantic-interposition

packaging/components.nix

@@ -164,6 +164,24 @@ let
   };
 
   mesonLibraryLayer = finalAttrs: prevAttrs: {
+    preConfigure =
+      let
+        interpositionFlags = [
+          "-fno-semantic-interposition"
+          "-Wl,-Bsymbolic-functions"
+        ];
+      in
+      # NOTE: By default GCC disables interprocedular optimizations (in particular inlining) for
+      # position-independent code and thus shared libraries.
+      # Since LD_PRELOAD tricks aren't worth losing out on optimizations, we disable it for good.
+      # This is not the case for Clang, where inlining is done by default even without -fno-semantic-interposition.
+      # https://reviews.llvm.org/D102453
+      # https://fedoraproject.org/wiki/Changes/PythonNoSemanticInterpositionSpeedup
+      prevAttrs.preConfigure or ""
+      + lib.optionalString stdenv.cc.isGNU ''
+        export CFLAGS="''${CFLAGS:-} ${toString interpositionFlags}"
+        export CXXFLAGS="''${CXXFLAGS:-} ${toString interpositionFlags}"
+      '';
     outputs = prevAttrs.outputs or [ "out" ] ++ [ "dev" ];
   };