[llvm] [AMDGPU] Move `AMDGPUAttributorPass` to full LTO post link stage (PR #102086)

[Matt Arsenault via llvm-commits]

================
@@ -734,21 +739,15 @@ void AMDGPUTargetMachine::registerPassBuilderCallbacks(PassBuilder &PB) {
         PM.addPass(createCGSCCToFunctionPassAdaptor(std::move(FPM)));
       });
 
-  // FIXME: Why is AMDGPUAttributor not in CGSCC?
-  PB.registerOptimizerLastEPCallback(
-      [this](ModulePassManager &MPM, OptimizationLevel Level) {
-        if (Level != OptimizationLevel::O0) {
-          MPM.addPass(AMDGPUAttributorPass(*this));
-        }
-      });
-
   PB.registerFullLinkTimeOptimizationLastEPCallback(
       [this](ModulePassManager &PM, OptimizationLevel Level) {
         // We want to support the -lto-partitions=N option as "best effort".
         // For that, we need to lower LDS earlier in the pipeline before the
         // module is partitioned for codegen.
         if (EnableLowerModuleLDS)
           PM.addPass(AMDGPULowerModuleLDSPass(*this));
+        if (EnableAMDGPUAttributor)
----------------
arsenm wrote:

This should still skip for -O0.

https://github.com/llvm/llvm-project/pull/102086