diff --git a/bolt/lib/Profile/StaleProfileMatching.cpp b/bolt/lib/Profile/StaleProfileMatching.cpp --- a/bolt/lib/Profile/StaleProfileMatching.cpp +++ b/bolt/lib/Profile/StaleProfileMatching.cpp @@ -27,17 +27,18 @@ #include "bolt/Core/HashUtilities.h" #include "bolt/Profile/YAMLProfileReader.h" +#include "llvm/ADT/Bitfields.h" #include "llvm/ADT/Hashing.h" #include "llvm/Support/CommandLine.h" #include "llvm/Transforms/Utils/SampleProfileInference.h" #include +using namespace llvm; + #undef DEBUG_TYPE #define DEBUG_TYPE "bolt-prof" -using namespace llvm; - namespace opts { extern cl::OptionCategory BoltOptCategory; @@ -141,49 +142,29 @@ /// components are of smaller size (e.g., uint16_t or uint8_t). struct BlendedBlockHash { private: - static uint64_t combineHashes(uint16_t Hash1, uint16_t Hash2, uint16_t Hash3, - uint16_t Hash4) { - uint64_t Hash = 0; - - Hash |= uint64_t(Hash4); - Hash <<= 16; - - Hash |= uint64_t(Hash3); - Hash <<= 16; - - Hash |= uint64_t(Hash2); - Hash <<= 16; - - Hash |= uint64_t(Hash1); - - return Hash; - } - - static void parseHashes(uint64_t Hash, uint16_t &Hash1, uint16_t &Hash2, - uint16_t &Hash3, uint16_t &Hash4) { - Hash1 = Hash & 0xffff; - Hash >>= 16; - - Hash2 = Hash & 0xffff; - Hash >>= 16; - - Hash3 = Hash & 0xffff; - Hash >>= 16; - - Hash4 = Hash & 0xffff; - Hash >>= 16; - } + using ValueOffset = Bitfield::Element; + using ValueOpcode = Bitfield::Element; + using ValueInstr = Bitfield::Element; + using ValueNeighbor = Bitfield::Element; public: explicit BlendedBlockHash() {} - explicit BlendedBlockHash(uint64_t CombinedHash) { - parseHashes(CombinedHash, Offset, OpcodeHash, InstrHash, NeighborHash); + explicit BlendedBlockHash(uint64_t Hash) { + Offset = Bitfield::get(Hash); + OpcodeHash = Bitfield::get(Hash); + InstrHash = Bitfield::get(Hash); + NeighborHash = Bitfield::get(Hash); } /// Combine the blended hash into uint64_t. uint64_t combine() const { - return combineHashes(Offset, OpcodeHash, InstrHash, NeighborHash); + uint64_t Hash = 0; + Bitfield::set(Hash, Offset); + Bitfield::set(Hash, OpcodeHash); + Bitfield::set(Hash, InstrHash); + Bitfield::set(Hash, NeighborHash); + return Hash; } /// Compute a distance between two given blended hashes. The smaller the @@ -236,9 +217,8 @@ /// Find the most similar block for a given hash. const FlowBlock *matchBlock(BlendedBlockHash BlendedHash) const { auto BlockIt = OpHashToBlocks.find(BlendedHash.OpcodeHash); - if (BlockIt == OpHashToBlocks.end()) { + if (BlockIt == OpHashToBlocks.end()) return nullptr; - } FlowBlock *BestBlock = nullptr; uint64_t BestDist = std::numeric_limits::max(); for (auto It : BlockIt->second) { @@ -306,6 +286,7 @@ BB->setHash(BlendedHashes[I].combine()); } } + /// Create a wrapper flow function to use with the profile inference algorithm, /// and initialize its jumps and metadata. FlowFunction @@ -314,7 +295,7 @@ // Add a special "dummy" source so that there is always a unique entry point. // Because of the extra source, for all other blocks in FlowFunction it holds - // that Block.Index == BB->getLayoutIndex() + 1 + // that Block.Index == BB->getIndex() + 1 FlowBlock EntryBlock; EntryBlock.Index = 0; Func.Blocks.push_back(EntryBlock); @@ -325,7 +306,7 @@ FlowBlock &Block = Func.Blocks.back(); Block.Index = Func.Blocks.size() - 1; (void)BB; - assert(Block.Index == BB->getLayoutIndex() + 1 && + assert(Block.Index == BB->getIndex() + 1 && "incorrectly assigned basic block index"); } @@ -341,8 +322,8 @@ Func.Jumps.emplace_back(); FlowJump &Jump = Func.Jumps.back(); - Jump.Source = SrcBB->getLayoutIndex() + 1; - Jump.Target = DstBB->getLayoutIndex() + 1; + Jump.Source = SrcBB->getIndex() + 1; + Jump.Target = DstBB->getIndex() + 1; InDegree[Jump.Target]++; UniqueSuccs.insert(DstBB); } @@ -354,8 +335,8 @@ Func.Jumps.emplace_back(); FlowJump &Jump = Func.Jumps.back(); - Jump.Source = SrcBB->getLayoutIndex() + 1; - Jump.Target = DstBB->getLayoutIndex() + 1; + Jump.Source = SrcBB->getIndex() + 1; + Jump.Target = DstBB->getIndex() + 1; InDegree[Jump.Target]++; UniqueSuccs.insert(DstBB); } @@ -691,31 +672,33 @@ bool YAMLProfileReader::inferStaleProfile( BinaryFunction &BF, const yaml::bolt::BinaryFunctionProfile &YamlBF) { - // Make sure that block indices and hashes are up to date - BF.getLayout().updateLayoutIndices(); + LLVM_DEBUG(dbgs() << "BOLT-INFO: applying profile inference for " + << "\"" << BF.getPrintName() << "\"\n"); + + // Make sure that block hashes are up to date. BF.computeBlockHashes(); const BinaryFunction::BasicBlockOrderType BlockOrder( BF.getLayout().block_begin(), BF.getLayout().block_end()); - // Create a wrapper flow function to use with the profile inference algorithm + // Create a wrapper flow function to use with the profile inference algorithm. FlowFunction Func = createFlowFunction(BlockOrder); - // Match as many block/jump counts from the stale profile as possible + // Match as many block/jump counts from the stale profile as possible. matchWeightsByHashes(BlockOrder, YamlBF, Func); // Adjust the flow function by marking unreachable blocks Unlikely so that - // they don't get any counts assigned + // they don't get any counts assigned. preprocessUnreachableBlocks(Func); - // Check if profile inference can be applied for the instance + // Check if profile inference can be applied for the instance. if (!canApplyInference(Func)) return false; - // Apply the profile inference algorithm + // Apply the profile inference algorithm. applyInference(Func); - // Collect inferred counts and update function annotations + // Collect inferred counts and update function annotations. assignProfile(BF, BlockOrder, Func); // As of now, we always mark the binary function having "correct" profile. diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp b/bolt/lib/Profile/YAMLProfileReader.cpp --- a/bolt/lib/Profile/YAMLProfileReader.cpp +++ b/bolt/lib/Profile/YAMLProfileReader.cpp @@ -249,9 +249,6 @@ << " edges in profile did not match function " << BF << '\n'; if (!ProfileMatched && opts::InferStaleProfile) { - if (opts::Verbosity >= 1) - outs() << "BOLT-INFO: applying profile inference for " - << "\"" << BF.getPrintName() << "\"\n"; if (inferStaleProfile(BF, YamlBF)) { ProfileMatched = true; BF.markProfiled(YamlBP.Header.Flags);