From d9830eb79fdc42368d370abeab9a3b56c08e3963 Mon Sep 17 00:00:00 2001 From: Piotr Padlewski Date: Mon, 26 Sep 2016 20:37:32 +0000 Subject: [thinlto] Basic thinlto fdo heuristic Summary: This patch improves thinlto importer by importing 3x larger functions that are called from hot block. I compared performance with the trunk on spec, and there were about 2% on povray and 3.33% on milc. These results seems to be consistant and match the results Teresa got with her simple heuristic. Some benchmarks got slower but I think they are just noisy (mcf, xalancbmki, omnetpp)- running the benchmarks again with more iterations to confirm. Geomean of all benchmarks including the noisy ones were about +0.02%. I see much better improvement on google branch with Easwaran patch for pgo callsite inlining (the inliner actually inline those big functions) Over all I see +0.5% improvement, and I get +8.65% on povray. So I guess we will see much bigger change when Easwaran patch will land (it depends on new pass manager), but it is still worth putting this to trunk before it. Implementation details changes: - Removed CallsiteCount. - ProfileCount got replaced by Hotness - hot-import-multiplier is set to 3.0 for now, didn't have time to tune it up, but I see that we get most of the interesting functions with 3, so there is no much performance difference with higher, and binary size doesn't grow as much as with 10.0. Reviewers: eraman, mehdi_amini, tejohnson Subscribers: mehdi_amini, llvm-commits Differential Revision: https://reviews.llvm.org/D24638 llvm-svn: 282437 --- llvm/lib/Bitcode/Reader/BitcodeReader.cpp | 57 ++++++++++++++++++++----------- 1 file changed, 38 insertions(+), 19 deletions(-) (limited to 'llvm/lib/Bitcode/Reader/BitcodeReader.cpp') diff --git a/llvm/lib/Bitcode/Reader/BitcodeReader.cpp b/llvm/lib/Bitcode/Reader/BitcodeReader.cpp index 49bf813..101e8eb 100644 --- a/llvm/lib/Bitcode/Reader/BitcodeReader.cpp +++ b/llvm/lib/Bitcode/Reader/BitcodeReader.cpp @@ -651,6 +651,9 @@ private: std::pair getGUIDFromValueId(unsigned ValueId); + std::pair + readCallGraphEdge(const SmallVector &Record, unsigned int &I, + bool IsOldProfileFormat, bool HasProfile); }; } // end anonymous namespace @@ -6218,8 +6221,10 @@ std::error_code ModuleSummaryIndexBitcodeReader::parseEntireSummary() { return error("Invalid Summary Block: version expected"); } const uint64_t Version = Record[0]; - if (Version != 1) - return error("Invalid summary version " + Twine(Version) + ", 1 expected"); + const bool IsOldProfileFormat = Version == 1; + if (!IsOldProfileFormat && Version != 2) + return error("Invalid summary version " + Twine(Version) + + ", 1 or 2 expected"); Record.clear(); // Keep around the last seen summary to be used when we see an optional @@ -6264,10 +6269,10 @@ std::error_code ModuleSummaryIndexBitcodeReader::parseEntireSummary() { default: // Default behavior: ignore. break; // FS_PERMODULE: [valueid, flags, instcount, numrefs, numrefs x valueid, - // n x (valueid, callsitecount)] + // n x (valueid)] // FS_PERMODULE_PROFILE: [valueid, flags, instcount, numrefs, // numrefs x valueid, - // n x (valueid, callsitecount, profilecount)] + // n x (valueid, hotness)] case bitc::FS_PERMODULE: case bitc::FS_PERMODULE_PROFILE: { unsigned ValueID = Record[0]; @@ -6296,12 +6301,11 @@ std::error_code ModuleSummaryIndexBitcodeReader::parseEntireSummary() { bool HasProfile = (BitCode == bitc::FS_PERMODULE_PROFILE); for (unsigned I = CallGraphEdgeStartIndex, E = Record.size(); I != E; ++I) { - unsigned CalleeValueId = Record[I]; - unsigned CallsiteCount = Record[++I]; - uint64_t ProfileCount = HasProfile ? Record[++I] : 0; - GlobalValue::GUID CalleeGUID = getGUIDFromValueId(CalleeValueId).first; - FS->addCallGraphEdge(CalleeGUID, - CalleeInfo(CallsiteCount, ProfileCount)); + CalleeInfo::HotnessType Hotness; + GlobalValue::GUID CalleeGUID; + std::tie(CalleeGUID, Hotness) = + readCallGraphEdge(Record, I, IsOldProfileFormat, HasProfile); + FS->addCallGraphEdge(CalleeGUID, CalleeInfo(Hotness)); } auto GUID = getGUIDFromValueId(ValueID); FS->setOriginalName(GUID.second); @@ -6356,10 +6360,9 @@ std::error_code ModuleSummaryIndexBitcodeReader::parseEntireSummary() { break; } // FS_COMBINED: [valueid, modid, flags, instcount, numrefs, - // numrefs x valueid, n x (valueid, callsitecount)] + // numrefs x valueid, n x (valueid)] // FS_COMBINED_PROFILE: [valueid, modid, flags, instcount, numrefs, - // numrefs x valueid, - // n x (valueid, callsitecount, profilecount)] + // numrefs x valueid, n x (valueid, hotness)] case bitc::FS_COMBINED: case bitc::FS_COMBINED_PROFILE: { unsigned ValueID = Record[0]; @@ -6385,12 +6388,11 @@ std::error_code ModuleSummaryIndexBitcodeReader::parseEntireSummary() { bool HasProfile = (BitCode == bitc::FS_COMBINED_PROFILE); for (unsigned I = CallGraphEdgeStartIndex, E = Record.size(); I != E; ++I) { - unsigned CalleeValueId = Record[I]; - unsigned CallsiteCount = Record[++I]; - uint64_t ProfileCount = HasProfile ? Record[++I] : 0; - GlobalValue::GUID CalleeGUID = getGUIDFromValueId(CalleeValueId).first; - FS->addCallGraphEdge(CalleeGUID, - CalleeInfo(CallsiteCount, ProfileCount)); + CalleeInfo::HotnessType Hotness; + GlobalValue::GUID CalleeGUID; + std::tie(CalleeGUID, Hotness) = + readCallGraphEdge(Record, I, IsOldProfileFormat, HasProfile); + FS->addCallGraphEdge(CalleeGUID, CalleeInfo(Hotness)); } GlobalValue::GUID GUID = getGUIDFromValueId(ValueID).first; TheIndex->addGlobalValueSummary(GUID, std::move(FS)); @@ -6456,6 +6458,23 @@ std::error_code ModuleSummaryIndexBitcodeReader::parseEntireSummary() { llvm_unreachable("Exit infinite loop"); } +std::pair +ModuleSummaryIndexBitcodeReader::readCallGraphEdge( + const SmallVector &Record, unsigned int &I, + const bool IsOldProfileFormat, const bool HasProfile) { + + auto Hotness = CalleeInfo::HotnessType::Unknown; + unsigned CalleeValueId = Record[I]; + GlobalValue::GUID CalleeGUID = getGUIDFromValueId(CalleeValueId).first; + if (IsOldProfileFormat) { + I += 1; // Skip old callsitecount field + if (HasProfile) + I += 1; // Skip old profilecount field + } else if (HasProfile) + Hotness = static_cast(Record[++I]); + return {CalleeGUID, Hotness}; +} + // Parse the module string table block into the Index. // This populates the ModulePathStringTable map in the index. std::error_code ModuleSummaryIndexBitcodeReader::parseModuleStringTable() { -- cgit v1.1