[OpenMP][NFC] Add option to print module after OpenMPOpt for debugging

[jhuber6]

Summary

This patch adds an extra option to print the module after running one of
the OpenMPOpt passes if debugging is enabled. This makes it much easier
to inspect the effects of this pass when doing debugging.

[jdoerfert]

LG

llvm/lib/Transforms/IPO/OpenMPOpt.cpp

static cl::opt<bool> DisableOpenMPOptFolding(
     cl::desc("Disable OpenMP optimizations involving folding."), cl::Hidden,
     cl::init(false));
 
 static cl::opt<bool> DisableOpenMPOptStateMachineRewrite(
     "openmp-opt-disable-state-machine-rewrite", cl::ZeroOrMore,
     cl::desc("Disable OpenMP optimizations that replace the state machine."),
     cl::Hidden, cl::init(false));
 
+static cl::opt<bool> PrintModuleAfterOptimizations(
+    "openmp-opt-print-module", cl::ZeroOrMore,
+    cl::desc("Print the current module after OpenMP optimizations."),
+    cl::Hidden, cl::init(false));
+
 STATISTIC(NumOpenMPRuntimeCallsDeduplicated,
           "Number of OpenMP runtime calls deduplicated");
 STATISTIC(NumOpenMPParallelRegionsDeleted,
           "Number of OpenMP parallel regions deleted");
 STATISTIC(NumOpenMPRuntimeFunctionsIdentified,
           "Number of OpenMP runtime functions identified");
 STATISTIC(NumOpenMPRuntimeFunctionUsesIdentified,
           "Number of OpenMP runtime function uses identified");
PreservedAnalyses OpenMPOptPass::run(Module &M, ModuleAnalysisManager &AM) {
   OMPInformationCache InfoCache(M, AG, Allocator, /*CGSCC*/ Functions, Kernels);
 
   unsigned MaxFixpointIterations = (isOpenMPDevice(M)) ? 128 : 32;
   Attributor A(Functions, InfoCache, CGUpdater, nullptr, true, false,
                MaxFixpointIterations, OREGetter, DEBUG_TYPE);
 
   OpenMPOpt OMPOpt(SCC, CGUpdater, OREGetter, InfoCache, A);
   bool Changed = OMPOpt.run(true);
+
+  if (PrintModuleAfterOptimizations)
+    LLVM_DEBUG(dbgs() << TAG << "Module after OpenMPOpt Module Pass:\n" << M);
+
   if (Changed)
     return PreservedAnalyses::none();
 
   return PreservedAnalyses::all();
 }
 
 PreservedAnalyses OpenMPOptCGSCCPass::run(LazyCallGraph::SCC &C,
                                           CGSCCAnalysisManager &AM,
OMPInformationCache InfoCache(*(Functions.back()->getParent()), AG, Allocator,
                                 /*CGSCC*/ Functions, Kernels);
 
   unsigned MaxFixpointIterations = (isOpenMPDevice(M)) ? 128 : 32;
   Attributor A(Functions, InfoCache, CGUpdater, nullptr, false, true,
                MaxFixpointIterations, OREGetter, DEBUG_TYPE);
 
   OpenMPOpt OMPOpt(SCC, CGUpdater, OREGetter, InfoCache, A);
   bool Changed = OMPOpt.run(false);
+
+  if (PrintModuleAfterOptimizations)
+    LLVM_DEBUG(dbgs() << TAG << "Module after OpenMPOpt CGSCC Pass:\n" << M);
+
   if (Changed)
     return PreservedAnalyses::none();
 
   return PreservedAnalyses::all();
 }
 
 namespace {
 
OMPInformationCache InfoCache(*(Functions.back()->getParent()), AG,
                                   Allocator,
                                   /*CGSCC*/ Functions, Kernels);
 
     unsigned MaxFixpointIterations = (isOpenMPDevice(M)) ? 128 : 32;
     Attributor A(Functions, InfoCache, CGUpdater, nullptr, false, true,
                  MaxFixpointIterations, OREGetter, DEBUG_TYPE);
 
     OpenMPOpt OMPOpt(SCC, CGUpdater, OREGetter, InfoCache, A);
-    return OMPOpt.run(false);
+    bool Result = OMPOpt.run(false);
+
+    if (PrintModuleAfterOptimizations)
+      LLVM_DEBUG(dbgs() << TAG << "Module after OpenMPOpt CGSCC Pass:\n" << M);
+
+    return Result;
   }
 
   bool doFinalization(CallGraph &CG) override { return CGUpdater.finalize(); }
 };
 
 } // end anonymous namespace
 
 KernelSet llvm::omp::getDeviceKernels(Module &M) {