[DFAPacketizer] Fix large compile-time regression for VLIW targets
D68992 / rL375086 refactored the packetizer and removed a bunch of logic. Unfortunately it creates an Automaton object whenever a DFAPacketizer is required. These objects have no longevity, and in particular on a debug build the population of the Automaton's transition map from the underlying table is very slow (because it is called ~10 times per MachineFunction, in the testcase I'm looking at).
This patch changes Automaton to wrap its underlying constant data in std::shared_ptr, which allows trivial copy construction. The DFAPacketizer creation function now creates a static archetypical Automaton and copies that whenever a new DFAPacketizer is required.
This takes a testcase down from ~20s to ~0.5s in debug mode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@375240 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/include/llvm/Support/Automaton.h b/include/llvm/Support/Automaton.h
index a5a4aef..7c13a69 100644
--- a/include/llvm/Support/Automaton.h
+++ b/include/llvm/Support/Automaton.h
@@ -161,10 +161,13 @@
/// FIXME: This uses a std::map because ActionT can be a pair type including
/// an enum. In particular DenseMapInfo<ActionT> must be defined to use
/// DenseMap here.
- std::map<std::pair<uint64_t, ActionT>, std::pair<uint64_t, unsigned>> M;
+ /// This is a shared_ptr to allow very quick copy-construction of Automata; this
+ /// state is immutable after construction so this is safe.
+ using MapTy = std::map<std::pair<uint64_t, ActionT>, std::pair<uint64_t, unsigned>>;
+ std::shared_ptr<MapTy> M;
/// An optional transcription object. This uses much more state than simply
/// traversing the DFA for acceptance, so is heap allocated.
- std::unique_ptr<internal::NfaTranscriber> Transcriber;
+ std::shared_ptr<internal::NfaTranscriber> Transcriber;
/// The initial DFA state is 1.
uint64_t State = 1;
/// True if we should transcribe and false if not (even if Transcriber is defined).
@@ -187,13 +190,15 @@
ArrayRef<NfaStatePair> TranscriptionTable = {}) {
if (!TranscriptionTable.empty())
Transcriber =
- std::make_unique<internal::NfaTranscriber>(TranscriptionTable);
+ std::make_shared<internal::NfaTranscriber>(TranscriptionTable);
Transcribe = Transcriber != nullptr;
+ M = std::make_shared<MapTy>();
for (const auto &I : Transitions)
// Greedily read and cache the transition table.
- M.emplace(std::make_pair(I.FromDfaState, I.Action),
- std::make_pair(I.ToDfaState, I.InfoIdx));
+ M->emplace(std::make_pair(I.FromDfaState, I.Action),
+ std::make_pair(I.ToDfaState, I.InfoIdx));
}
+ Automaton(const Automaton &) = default;
/// Reset the automaton to its initial state.
void reset() {
@@ -218,8 +223,8 @@
/// If this function returns false, all methods are undefined until reset() is
/// called.
bool add(const ActionT &A) {
- auto I = M.find({State, A});
- if (I == M.end())
+ auto I = M->find({State, A});
+ if (I == M->end())
return false;
if (Transcriber && Transcribe)
Transcriber->transition(I->second.second);
@@ -229,8 +234,8 @@
/// Return true if the automaton can be transitioned based on input symbol A.
bool canAdd(const ActionT &A) {
- auto I = M.find({State, A});
- return I != M.end();
+ auto I = M->find({State, A});
+ return I != M->end();
}
/// Obtain a set of possible paths through the input nondeterministic
diff --git a/utils/TableGen/DFAPacketizerEmitter.cpp b/utils/TableGen/DFAPacketizerEmitter.cpp
index 0d3c6fd..ccb4ef1 100644
--- a/utils/TableGen/DFAPacketizerEmitter.cpp
+++ b/utils/TableGen/DFAPacketizerEmitter.cpp
@@ -538,10 +538,10 @@
OS << "DFAPacketizer *" << SubTargetClassName << "::"
<< "create" << DFAName
<< "DFAPacketizer(const InstrItineraryData *IID) const {\n"
- << " Automaton<uint64_t> A(ArrayRef<" << TargetAndDFAName
+ << " static Automaton<uint64_t> A(ArrayRef<" << TargetAndDFAName
<< "Transition>(" << TargetAndDFAName << "Transitions), "
<< TargetAndDFAName << "TransitionInfo);\n"
- << " return new DFAPacketizer(IID, std::move(A));\n"
+ << " return new DFAPacketizer(IID, A);\n"
<< "\n}\n\n";
}