[NewPassManager] Add tuning option: LoopUnrolling [NFC].

[asbirlea]

Summary

Mirror tuning option from old pass manager in new pass manager.

[chandlerc]

Hmm, how should we test that this works?

[asbirlea]

I think the test in the clang patch that needs this should do?

[chandlerc]

Eh, I'd like long term to have a way of testing this field in LLVM, but I think this is fine for now to unblock the Clang changes and other changes.

Maybe in a follow-up either a unittest that works like Clang does or add a commandline flag so it can be tested with opt?

[asbirlea]

Ack, will do!

include/llvm/Passes/PassBuilder.h

public:
   /// Tuning option to enable/disable loop vectorization. Its default value is
   /// that of the flag: `-vectorize-loops`.
   bool LoopVectorization;
 
   /// Tuning option to enable/disable slp loop vectorization. Its default value
   /// is that of the flag: `vectorize-slp`.
   bool SLPVectorization;
 
+  /// Tuning option to enable/disable loop unrolling. Its default value is true.
+  bool LoopUnrolling;
+
   /// Tuning option to forget all SCEV loops in LoopUnroll. Its default value
   /// is that of the flag: `-forget-scev-loop-unroll`.
   bool ForgetAllSCEVInLoopUnroll;
 
   /// Tuning option to cap the number of calls to retrive clobbering accesses in
   /// MemorySSA, in LICM.
   unsigned LicmMssaOptCap;
 

lib/Passes/PassBuilder.cpp

 static cl::opt<bool>
     EnableCHR("enable-chr-npm", cl::init(true), cl::Hidden,
               cl::desc("Enable control height reduction optimization (CHR)"));
 
 PipelineTuningOptions::PipelineTuningOptions() {
   LoopInterleaving = EnableLoopInterleaving;
   LoopVectorization = EnableLoopVectorization;
   SLPVectorization = RunSLPVectorization;
+  LoopUnrolling = true;
   ForgetAllSCEVInLoopUnroll = ForgetSCEVInLoopUnroll;
   LicmMssaOptCap = SetLicmMssaOptCap;
   LicmMssaNoAccForPromotionCap = SetLicmMssaNoAccForPromotionCap;
 }
 
 extern cl::opt<bool> EnableHotColdSplit;
 extern cl::opt<bool> EnableOrderFileInstrumentation;
 
PassBuilder::buildFunctionSimplificationPipeline(OptimizationLevel Level,
 
   for (auto &C : LateLoopOptimizationsEPCallbacks)
     C(LPM2, Level);
 
   LPM2.addPass(LoopDeletionPass());
   // Do not enable unrolling in PreLinkThinLTO phase during sample PGO
   // because it changes IR to makes profile annotation in back compile
   // inaccurate.
-  if (Phase != ThinLTOPhase::PreLink || !PGOOpt ||
-      PGOOpt->Action != PGOOptions::SampleUse)
+  if ((Phase != ThinLTOPhase::PreLink || !PGOOpt ||
+       PGOOpt->Action != PGOOptions::SampleUse) &&
+      PTO.LoopUnrolling)
     LPM2.addPass(
         LoopFullUnrollPass(Level, false, PTO.ForgetAllSCEVInLoopUnroll));
 
   for (auto &C : LoopOptimizerEndEPCallbacks)
     C(LPM2, Level);
 
   // We provide the opt remark emitter pass for LICM to use. We only need to do
   // this once as it is immutable.
ModulePassManager PassBuilder::buildModuleOptimizationPipeline(
   // FIXME: It would be really good to use a loop-integrated instruction
   // combiner for cleanup here so that the unrolling and LICM can be pipelined
   // across the loop nests.
   // We do UnrollAndJam in a separate LPM to ensure it happens before unroll
   if (EnableUnrollAndJam) {
     OptimizePM.addPass(
         createFunctionToLoopPassAdaptor(LoopUnrollAndJamPass(Level)));
   }
+  if (PTO.LoopUnrolling)
     OptimizePM.addPass(LoopUnrollPass(
         LoopUnrollOptions(Level, false, PTO.ForgetAllSCEVInLoopUnroll)));
   OptimizePM.addPass(WarnMissedTransformationsPass());
   OptimizePM.addPass(InstCombinePass());
   OptimizePM.addPass(RequireAnalysisPass<OptimizationRemarkEmitterAnalysis, Function>());
   OptimizePM.addPass(createFunctionToLoopPassAdaptor(
       LICMPass(PTO.LicmMssaOptCap, PTO.LicmMssaNoAccForPromotionCap),
       DebugLogging));
 
   // Now that we've vectorized and unrolled loops, we may have more refined