19317ad1d9
The original Interpreter implementation had a hard dependency on ORC and grew organically with the addition of out-of-process JIT support. This tightly coupled the Interpreter to a specific execution engine and leaked ORC-specific assumptions (runtime layout, symbol lookup, exception model) into higher layers. The WebAssembly integration demonstrated that incremental execution can be implemented without ORC, exposing the need for a cleaner abstraction boundary. This change introduces an IncrementalExecutor interface and moves ORC-based execution behind a concrete implementation. The Interpreter now depends only on the abstract executor, improving layering and encapsulation. In addition, the Interpreter can be configured with user-provided incremental executor implementations, enabling ORC-independent execution, easier testing, and future extensions without modifying the core Interpreter.
122 lines
4.1 KiB
C++
122 lines
4.1 KiB
C++
//===--- OrcIncrementalExecutor.cpp - Orc Incremental Execution -*- 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
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
// This file implements an Orc-based incremental code execution.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "OrcIncrementalExecutor.h"
|
|
#include "clang/Interpreter/PartialTranslationUnit.h"
|
|
|
|
#include "llvm/ExecutionEngine/Orc/EPCDynamicLibrarySearchGenerator.h"
|
|
#include "llvm/ExecutionEngine/Orc/ExecutionUtils.h"
|
|
#include "llvm/ExecutionEngine/Orc/JITTargetMachineBuilder.h"
|
|
#include "llvm/ExecutionEngine/Orc/LLJIT.h"
|
|
#include "llvm/ExecutionEngine/Orc/Shared/OrcRTBridge.h"
|
|
#include "llvm/ExecutionEngine/Orc/Shared/SimpleRemoteEPCUtils.h"
|
|
#include "llvm/ExecutionEngine/Orc/TargetProcess/JITLoaderGDB.h"
|
|
#include "llvm/Support/Error.h"
|
|
#include "llvm/Support/raw_ostream.h"
|
|
|
|
#ifdef LLVM_ON_UNIX
|
|
#include <netdb.h>
|
|
#include <netinet/in.h>
|
|
#include <sys/socket.h>
|
|
#include <unistd.h>
|
|
#endif // LLVM_ON_UNIX
|
|
|
|
// Force linking some of the runtimes that helps attaching to a debugger.
|
|
LLVM_ATTRIBUTE_USED void linkComponents() {
|
|
llvm::errs() << (void *)&llvm_orc_registerJITLoaderGDBAllocAction;
|
|
}
|
|
|
|
namespace clang {
|
|
OrcIncrementalExecutor::OrcIncrementalExecutor(
|
|
llvm::orc::ThreadSafeContext &TSC)
|
|
: TSCtx(TSC) {}
|
|
|
|
OrcIncrementalExecutor::OrcIncrementalExecutor(
|
|
llvm::orc::ThreadSafeContext &TSC, llvm::orc::LLJITBuilder &JITBuilder,
|
|
llvm::Error &Err)
|
|
: TSCtx(TSC) {
|
|
using namespace llvm::orc;
|
|
llvm::ErrorAsOutParameter EAO(&Err);
|
|
|
|
if (auto JitOrErr = JITBuilder.create())
|
|
Jit = std::move(*JitOrErr);
|
|
else {
|
|
Err = JitOrErr.takeError();
|
|
return;
|
|
}
|
|
}
|
|
|
|
OrcIncrementalExecutor::~OrcIncrementalExecutor() {}
|
|
|
|
llvm::Error OrcIncrementalExecutor::addModule(PartialTranslationUnit &PTU) {
|
|
llvm::orc::ResourceTrackerSP RT =
|
|
Jit->getMainJITDylib().createResourceTracker();
|
|
ResourceTrackers[&PTU] = RT;
|
|
|
|
return Jit->addIRModule(RT, {std::move(PTU.TheModule), TSCtx});
|
|
}
|
|
|
|
llvm::Error OrcIncrementalExecutor::removeModule(PartialTranslationUnit &PTU) {
|
|
|
|
llvm::orc::ResourceTrackerSP RT = std::move(ResourceTrackers[&PTU]);
|
|
if (!RT)
|
|
return llvm::Error::success();
|
|
|
|
ResourceTrackers.erase(&PTU);
|
|
if (llvm::Error Err = RT->remove())
|
|
return Err;
|
|
return llvm::Error::success();
|
|
}
|
|
|
|
// Clean up the JIT instance.
|
|
llvm::Error OrcIncrementalExecutor::cleanUp() {
|
|
// This calls the global dtors of registered modules.
|
|
return Jit->deinitialize(Jit->getMainJITDylib());
|
|
}
|
|
|
|
llvm::Error OrcIncrementalExecutor::runCtors() const {
|
|
return Jit->initialize(Jit->getMainJITDylib());
|
|
}
|
|
|
|
llvm::Expected<llvm::orc::ExecutorAddr>
|
|
OrcIncrementalExecutor::getSymbolAddress(llvm::StringRef Name,
|
|
SymbolNameKind NameKind) const {
|
|
using namespace llvm::orc;
|
|
auto SO = makeJITDylibSearchOrder({&Jit->getMainJITDylib(),
|
|
Jit->getPlatformJITDylib().get(),
|
|
Jit->getProcessSymbolsJITDylib().get()});
|
|
|
|
ExecutionSession &ES = Jit->getExecutionSession();
|
|
|
|
auto SymOrErr = ES.lookup(SO, (NameKind == SymbolNameKind::LinkerName)
|
|
? ES.intern(Name)
|
|
: Jit->mangleAndIntern(Name));
|
|
if (auto Err = SymOrErr.takeError())
|
|
return std::move(Err);
|
|
return SymOrErr->getAddress();
|
|
}
|
|
|
|
llvm::Error OrcIncrementalExecutor::LoadDynamicLibrary(const char *name) {
|
|
// FIXME: Eventually we should put each library in its own JITDylib and
|
|
// turn off process symbols by default.
|
|
llvm::orc::ExecutionSession &ES = Jit->getExecutionSession();
|
|
auto DLSGOrErr = llvm::orc::EPCDynamicLibrarySearchGenerator::Load(ES, name);
|
|
if (!DLSGOrErr)
|
|
return DLSGOrErr.takeError();
|
|
|
|
Jit->getProcessSymbolsJITDylib()->addGenerator(std::move(*DLSGOrErr));
|
|
|
|
return llvm::Error::success();
|
|
}
|
|
|
|
} // namespace clang
|