[OpenMP][NFC] Add an option to print the module before in OpenMPOpt

[jhuber6]

Summary

Previously there was a debug flag to print the module after
optimizations. Sometimes we wanted to print the module before
optimizations so this is being split into two flags.
-openmp-opt-print-module is now -openmp-opt-print-module-after.

[jdoerfert]

LG

llvm/lib/Transforms/IPO/OpenMPOpt.cpp

static cl::opt<bool> DisableOpenMPOptStateMachineRewrite(
     cl::Hidden, cl::init(false));
 
 static cl::opt<bool> DisableOpenMPOptBarrierElimination(
     "openmp-opt-disable-barrier-elimination", cl::ZeroOrMore,
     cl::desc("Disable OpenMP optimizations that eliminate barriers."),
     cl::Hidden, cl::init(false));
 
 static cl::opt<bool> PrintModuleAfterOptimizations(
-    "openmp-opt-print-module", cl::ZeroOrMore,
+    "openmp-opt-print-module-after", cl::ZeroOrMore,
     cl::desc("Print the current module after OpenMP optimizations."),
     cl::Hidden, cl::init(false));
 
+static cl::opt<bool> PrintModuleBeforeOptimizations(
+    "openmp-opt-print-module-before", cl::ZeroOrMore,
+    cl::desc("Print the current module before OpenMP optimizations."),
+    cl::Hidden, cl::init(false));
+
 static cl::opt<bool> AlwaysInlineDeviceFunctions(
     "openmp-opt-inline-device", cl::ZeroOrMore,
     cl::desc("Inline all applicible functions on the device."), cl::Hidden,
     cl::init(false));
 
 static cl::opt<bool>
     EnableVerboseRemarks("openmp-opt-verbose-remarks", cl::ZeroOrMore,
                          cl::desc("Enables more verbose remarks."), cl::Hidden,
if (!containsOpenMP(M))
     return PreservedAnalyses::all();
   if (DisableOpenMPOptimizations)
     return PreservedAnalyses::all();
 
   FunctionAnalysisManager &FAM =
       AM.getResult<FunctionAnalysisManagerModuleProxy>(M).getManager();
   KernelSet Kernels = getDeviceKernels(M);
 
+  if (PrintModuleBeforeOptimizations)
+    LLVM_DEBUG(dbgs() << TAG << "Module before OpenMPOpt Module Pass:\n" << M);
+
   auto IsCalled = [&](Function &F) {
     if (Kernels.contains(&F))
       return true;
     for (const User *U : F.users())
       if (!isa<BlockAddress>(U))
         return true;
     return false;
   };
for (LazyCallGraph::Node &N : C) {
     SCC.push_back(Fn);
   }
 
   if (SCC.empty())
     return PreservedAnalyses::all();
 
   Module &M = *C.begin()->getFunction().getParent();
 
+  if (PrintModuleBeforeOptimizations)
+    LLVM_DEBUG(dbgs() << TAG << "Module before OpenMPOpt CGSCC Pass:\n" << M);
+
   KernelSet Kernels = getDeviceKernels(M);
 
   FunctionAnalysisManager &FAM =
       AM.getResult<FunctionAnalysisManagerCGSCCProxy>(C, CG).getManager();
 
   AnalysisGetter AG(FAM);
 
   auto OREGetter = [&FAM](Function *F) -> OptimizationRemarkEmitter & {