[OpenMP] Introduce the OpenMPOpt transformation pass

The OpenMPOpt pass is a CGSCC pass in which OpenMP specific
optimizations can reside.

The OpenMPOpt pass uses the OpenMPKinds.def file to identify runtime
calls and their uses. This allows targeted transformations and eases
their implementation.

This initial patch deduplicates `__kmpc_global_thread_num` and
`omp_get_thread_num` calls. We can also identify arguments that are
equivalent to such a call result and use it instead. Later we can
determine "gtid" arguments based on the use in kernel functions etc.

Reviewed By: JonChesterfield

Differential Revision: https://reviews.llvm.org/D69930

llvm/include/llvm/Frontend/OpenMP/OMPKinds.def

@@ -177,6 +177,8 @@ __OMP_RTL(__kmpc_end_serialized_parallel, false, Void, IdentPtr, Int32)
 
 __OMP_RTL(omp_get_thread_num, false, Int32, )
 
+__OMP_RTL(__last, false, Void, )
+
 #undef __OMP_RTL
 #undef OMP_RTL
 

llvm/include/llvm/InitializePasses.h

@@ -71,6 +71,7 @@ void initializeAggressiveInstCombinerLegacyPassPass(PassRegistry&);
 void initializeAliasSetPrinterPass(PassRegistry&);
 void initializeAlignmentFromAssumptionsPass(PassRegistry&);
 void initializeAlwaysInlinerLegacyPassPass(PassRegistry&);
+void initializeOpenMPOptLegacyPassPass(PassRegistry &);
 void initializeArgPromotionPass(PassRegistry&);
 void initializeAssumptionCacheTrackerPass(PassRegistry&);
 void initializeAtomicExpandPass(PassRegistry&);

llvm/include/llvm/LinkAllPasses.h

@@ -71,6 +71,7 @@ namespace {
       (void) llvm::createAggressiveDCEPass();
       (void) llvm::createAggressiveInstCombinerPass();
       (void) llvm::createBitTrackingDCEPass();
+      (void) llvm::createOpenMPOptLegacyPass();
       (void) llvm::createArgumentPromotionPass();
       (void) llvm::createAlignmentFromAssumptionsPass();
       (void) llvm::createBasicAAWrapperPass();

llvm/include/llvm/Transforms/IPO.h

@@ -152,6 +152,10 @@ ModulePass *createDeadArgHackingPass();
 ///
 Pass *createArgumentPromotionPass(unsigned maxElements = 3);
 
+//===----------------------------------------------------------------------===//
+/// createOpenMPOptLegacyPass - OpenMP specific optimizations.
+Pass *createOpenMPOptLegacyPass();
+
 //===----------------------------------------------------------------------===//
 /// createIPConstantPropagationPass - This pass propagates constants from call
 /// sites into the bodies of functions.

llvm/include/llvm/Transforms/IPO/OpenMPOpt.h

@@ -0,0 +1,54 @@
+//===- IPO/OpenMPOpt.h - Collection of OpenMP optimizations -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_TRANSFORMS_IPO_OPENMP_OPT_H
+#define LLVM_TRANSFORMS_IPO_OPENMP_OPT_H
+
+#include "llvm/Analysis/CGSCCPassManager.h"
+#include "llvm/Analysis/LazyCallGraph.h"
+#include "llvm/IR/PassManager.h"
+
+namespace llvm {
+
+namespace omp {
+
+/// Helper to remember if the module contains OpenMP (runtime calls), to be used
+/// foremost with containsOpenMP.
+struct OpenMPInModule {
+  OpenMPInModule &operator=(bool Found) {
+    if (Found)
+      Value = OpenMPInModule::OpenMP::FOUND;
+    else
+      Value = OpenMPInModule::OpenMP::NOT_FOUND;
+    return *this;
+  }
+  bool isKnown() { return Value != OpenMP::UNKNOWN; }
+  operator bool() { return Value != OpenMP::NOT_FOUND; }
+
+private:
+  enum class OpenMP { FOUND, NOT_FOUND, UNKNOWN } Value = OpenMP::UNKNOWN;
+};
+
+/// Helper to determine if \p M contains OpenMP (runtime calls).
+bool containsOpenMP(Module &M, OpenMPInModule &OMPInModule);
+
+} // namespace omp
+
+/// OpenMP optimizations pass.
+class OpenMPOptPass : public PassInfoMixin<OpenMPOptPass> {
+  /// Helper to remember if the module contains OpenMP (runtime calls).
+  omp::OpenMPInModule OMPInModule;
+
+public:
+  PreservedAnalyses run(LazyCallGraph::SCC &C, CGSCCAnalysisManager &AM,
+                        LazyCallGraph &CG, CGSCCUpdateResult &UR);
+};
+
+} // end namespace llvm
+
+#endif // LLVM_TRANSFORMS_IPO_OPENMP_OPT_H

llvm/lib/LTO/LTOCodeGenerator.cpp

@@ -134,6 +134,7 @@ void LTOCodeGenerator::initializeLTOPasses() {
   initializeSimpleInlinerPass(R);
   initializePruneEHPass(R);
   initializeGlobalDCELegacyPassPass(R);
+  initializeOpenMPOptLegacyPassPass(R);
   initializeArgPromotionPass(R);
   initializeJumpThreadingPass(R);
   initializeSROALegacyPassPass(R);

llvm/lib/Passes/PassBuilder.cpp

@@ -87,6 +87,7 @@
 #include "llvm/Transforms/IPO/Internalize.h"
 #include "llvm/Transforms/IPO/LowerTypeTests.h"
 #include "llvm/Transforms/IPO/MergeFunctions.h"
+#include "llvm/Transforms/IPO/OpenMPOpt.h"
 #include "llvm/Transforms/IPO/PartialInlining.h"
 #include "llvm/Transforms/IPO/SCCP.h"
 #include "llvm/Transforms/IPO/SampleProfile.h"
@@ -837,6 +838,11 @@ PassBuilder::buildModuleSimplificationPipeline(OptimizationLevel Level,
   if (Level == OptimizationLevel::O3)
     MainCGPipeline.addPass(ArgumentPromotionPass());
 
+  // Try to perform OpenMP specific optimizations. This is a (quick!) no-op if
+  // there are no OpenMP runtime calls present in the module.
+  if (Level == OptimizationLevel::O2 || Level == OptimizationLevel::O3)
+    MainCGPipeline.addPass(OpenMPOptPass());
+
   // Lastly, add the core function simplification pipeline nested inside the
   // CGSCC walk.
   MainCGPipeline.addPass(createCGSCCToFunctionPassAdaptor(

llvm/lib/Passes/PassRegistry.def

@@ -109,6 +109,7 @@ CGSCC_PASS("argpromotion", ArgumentPromotionPass())
 CGSCC_PASS("invalidate<all>", InvalidateAllAnalysesPass())
 CGSCC_PASS("function-attrs", PostOrderFunctionAttrsPass())
 CGSCC_PASS("inline", InlinerPass())
+CGSCC_PASS("openmpopt", OpenMPOptPass())
 CGSCC_PASS("no-op-cgscc", NoOpCGSCCPass())
 #undef CGSCC_PASS
 

llvm/lib/Transforms/IPO/CMakeLists.txt

@@ -26,6 +26,7 @@ add_llvm_component_library(LLVMipo
   LoopExtractor.cpp
   LowerTypeTests.cpp
   MergeFunctions.cpp
+  OpenMPOpt.cpp
   PartialInlining.cpp
   PassManagerBuilder.cpp
   PruneEH.cpp

llvm/lib/Transforms/IPO/IPO.cpp

@@ -23,6 +23,7 @@
 using namespace llvm;
 
 void llvm::initializeIPO(PassRegistry &Registry) {
+  initializeOpenMPOptLegacyPassPass(Registry);
   initializeArgPromotionPass(Registry);
   initializeCalledValuePropagationLegacyPassPass(Registry);
   initializeConstantMergeLegacyPassPass(Registry);

llvm/lib/Transforms/IPO/LLVMBuild.txt

@@ -19,4 +19,4 @@ type = Library
 name = IPO
 parent = Transforms
 library_name = ipo
-required_libraries = AggressiveInstCombine Analysis BitReader BitWriter Core InstCombine IRReader Linker Object ProfileData Scalar Support TransformUtils Vectorize Instrumentation
+required_libraries = AggressiveInstCombine Analysis BitReader BitWriter Core FrontendOpenMP InstCombine IRReader Linker Object ProfileData Scalar Support TransformUtils Vectorize Instrumentation