Sergio Afonso 837bff11cb
[Flang][Lower] Attach target_cpu and target_features attributes to MLIR functions (#78289)
This patch forwards the target CPU and features information from the
Flang frontend to MLIR func.func operation attributes, which are later
used to populate the target_cpu and target_features llvm.func
attributes.

This is achieved in two stages:

1. Introduce the `fir.target_cpu` and `fir.target_features` module
attributes with information from the target machine immediately after
the initial creation of the MLIR module in the lowering bridge.

2. Update the target rewrite flang pass to get this information from the
module and pass it along to all func.func MLIR operations, respectively
as attributes named `target_cpu` and `target_features`. These attributes
will be automatically picked up during Func to LLVM dialect lowering and
used to initialize the corresponding llvm.func named attributes.

The target rewrite and FIR to LLVM lowering passes are updated with the
ability to override these module attributes, and the `CodeGenSpecifics`
optimizer class is augmented to make this information available to
target-specific MLIR transformations.

This completes a full flow by which target CPU and features make it all
the way from compiler options to LLVM IR function attributes.
2024-01-30 13:45:56 +00:00

514 lines
20 KiB
C++

//===- bbc.cpp - Burnside Bridge Compiler -----------------------*- 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
//
//===----------------------------------------------------------------------===//
//
// Coding style: https://mlir.llvm.org/getting_started/DeveloperGuide/
//
//===----------------------------------------------------------------------===//
///
/// This is a tool for translating Fortran sources to the FIR dialect of MLIR.
///
//===----------------------------------------------------------------------===//
#include "flang/Common/Fortran-features.h"
#include "flang/Common/OpenMP-features.h"
#include "flang/Common/Version.h"
#include "flang/Common/default-kinds.h"
#include "flang/Lower/Bridge.h"
#include "flang/Lower/PFTBuilder.h"
#include "flang/Lower/Support/Verifier.h"
#include "flang/Optimizer/Dialect/Support/FIRContext.h"
#include "flang/Optimizer/Dialect/Support/KindMapping.h"
#include "flang/Optimizer/Support/InitFIR.h"
#include "flang/Optimizer/Support/InternalNames.h"
#include "flang/Optimizer/Support/Utils.h"
#include "flang/Optimizer/Transforms/Passes.h"
#include "flang/Parser/characters.h"
#include "flang/Parser/dump-parse-tree.h"
#include "flang/Parser/message.h"
#include "flang/Parser/parse-tree-visitor.h"
#include "flang/Parser/parse-tree.h"
#include "flang/Parser/parsing.h"
#include "flang/Parser/provenance.h"
#include "flang/Parser/unparse.h"
#include "flang/Semantics/expression.h"
#include "flang/Semantics/runtime-type-info.h"
#include "flang/Semantics/semantics.h"
#include "flang/Semantics/unparse-with-symbols.h"
#include "flang/Tools/CrossToolHelpers.h"
#include "flang/Tools/TargetSetup.h"
#include "flang/Version.inc"
#include "mlir/Dialect/OpenMP/OpenMPDialect.h"
#include "mlir/IR/AsmState.h"
#include "mlir/IR/BuiltinOps.h"
#include "mlir/IR/MLIRContext.h"
#include "mlir/Parser/Parser.h"
#include "mlir/Pass/Pass.h"
#include "mlir/Pass/PassManager.h"
#include "mlir/Pass/PassRegistry.h"
#include "mlir/Transforms/GreedyPatternRewriteDriver.h"
#include "mlir/Transforms/Passes.h"
#include "llvm/MC/TargetRegistry.h"
#include "llvm/Passes/OptimizationLevel.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/ErrorOr.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/InitLLVM.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/Path.h"
#include "llvm/Support/SourceMgr.h"
#include "llvm/Support/TargetSelect.h"
#include "llvm/Support/ToolOutputFile.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/TargetParser/Host.h"
#include "llvm/TargetParser/Triple.h"
#include <memory>
//===----------------------------------------------------------------------===//
// Some basic command-line options
//===----------------------------------------------------------------------===//
static llvm::cl::opt<std::string> inputFilename(llvm::cl::Positional,
llvm::cl::Required,
llvm::cl::desc("<input file>"));
static llvm::cl::opt<std::string>
outputFilename("o", llvm::cl::desc("Specify the output filename"),
llvm::cl::value_desc("filename"));
static llvm::cl::list<std::string>
includeDirs("I", llvm::cl::desc("include module search paths"));
static llvm::cl::alias includeAlias("module-directory",
llvm::cl::desc("module search directory"),
llvm::cl::aliasopt(includeDirs));
static llvm::cl::list<std::string>
intrinsicIncludeDirs("J", llvm::cl::desc("intrinsic module search paths"));
static llvm::cl::alias
intrinsicIncludeAlias("intrinsic-module-directory",
llvm::cl::desc("intrinsic module directory"),
llvm::cl::aliasopt(intrinsicIncludeDirs));
static llvm::cl::opt<std::string>
moduleDir("module", llvm::cl::desc("module output directory (default .)"),
llvm::cl::init("."));
static llvm::cl::opt<std::string>
moduleSuffix("module-suffix", llvm::cl::desc("module file suffix override"),
llvm::cl::init(".mod"));
static llvm::cl::opt<bool>
emitFIR("emit-fir",
llvm::cl::desc("Dump the FIR created by lowering and exit"),
llvm::cl::init(false));
static llvm::cl::opt<bool>
emitHLFIR("emit-hlfir",
llvm::cl::desc("Dump the HLFIR created by lowering and exit"),
llvm::cl::init(false));
static llvm::cl::opt<bool> warnStdViolation("Mstandard",
llvm::cl::desc("emit warnings"),
llvm::cl::init(false));
static llvm::cl::opt<bool> warnIsError("Werror",
llvm::cl::desc("warnings are errors"),
llvm::cl::init(false));
static llvm::cl::opt<bool> dumpSymbols("dump-symbols",
llvm::cl::desc("dump the symbol table"),
llvm::cl::init(false));
static llvm::cl::opt<bool> pftDumpTest(
"pft-test",
llvm::cl::desc("parse the input, create a PFT, dump it, and exit"),
llvm::cl::init(false));
static llvm::cl::opt<bool> enableOpenMP("fopenmp",
llvm::cl::desc("enable openmp"),
llvm::cl::init(false));
static llvm::cl::opt<bool>
enableOpenMPDevice("fopenmp-is-target-device",
llvm::cl::desc("enable openmp device compilation"),
llvm::cl::init(false));
static llvm::cl::opt<bool>
enableOpenMPGPU("fopenmp-is-gpu",
llvm::cl::desc("enable openmp GPU target codegen"),
llvm::cl::init(false));
// A simplified subset of the OpenMP RTL Flags from Flang, only the primary
// positive options are available, no negative options e.g. fopen_assume* vs
// fno_open_assume*
static llvm::cl::opt<uint32_t>
setOpenMPVersion("fopenmp-version",
llvm::cl::desc("OpenMP standard version"),
llvm::cl::init(11));
static llvm::cl::opt<uint32_t> setOpenMPTargetDebug(
"fopenmp-target-debug",
llvm::cl::desc("Enable debugging in the OpenMP offloading device RTL"),
llvm::cl::init(0));
static llvm::cl::opt<bool> setOpenMPThreadSubscription(
"fopenmp-assume-threads-oversubscription",
llvm::cl::desc("Assume work-shared loops do not have more "
"iterations than participating threads."),
llvm::cl::init(false));
static llvm::cl::opt<bool> setOpenMPTeamSubscription(
"fopenmp-assume-teams-oversubscription",
llvm::cl::desc("Assume distributed loops do not have more iterations than "
"participating teams."),
llvm::cl::init(false));
static llvm::cl::opt<bool> setOpenMPNoThreadState(
"fopenmp-assume-no-thread-state",
llvm::cl::desc(
"Assume that no thread in a parallel region will modify an ICV."),
llvm::cl::init(false));
static llvm::cl::opt<bool> setOpenMPNoNestedParallelism(
"fopenmp-assume-no-nested-parallelism",
llvm::cl::desc("Assume that no thread in a parallel region will encounter "
"a parallel region."),
llvm::cl::init(false));
static llvm::cl::opt<bool>
setNoGPULib("nogpulib",
llvm::cl::desc("Do not link device library for CUDA/HIP device "
"compilation"),
llvm::cl::init(false));
static llvm::cl::opt<bool> enableOpenACC("fopenacc",
llvm::cl::desc("enable openacc"),
llvm::cl::init(false));
static llvm::cl::opt<bool> enablePolymorphic(
"polymorphic-type",
llvm::cl::desc("enable polymorphic type lowering (experimental)"),
llvm::cl::init(false));
static llvm::cl::opt<bool> enableNoPPCNativeVecElemOrder(
"fno-ppc-native-vector-element-order",
llvm::cl::desc("no PowerPC native vector element order."),
llvm::cl::init(false));
static llvm::cl::opt<bool> useHLFIR("hlfir",
llvm::cl::desc("Lower to high level FIR"),
llvm::cl::init(true));
static llvm::cl::opt<bool> enableCUDA("fcuda",
llvm::cl::desc("enable CUDA Fortran"),
llvm::cl::init(false));
static llvm::cl::opt<bool> fixedForm("ffixed-form",
llvm::cl::desc("enable fixed form"),
llvm::cl::init(false));
static llvm::cl::opt<std::string>
targetTripleOverride("target",
llvm::cl::desc("Override host target triple"),
llvm::cl::init(""));
#define FLANG_EXCLUDE_CODEGEN
#include "flang/Tools/CLOptions.inc"
//===----------------------------------------------------------------------===//
using ProgramName = std::string;
// Print the module with the "module { ... }" wrapper, preventing
// information loss from attribute information appended to the module
static void printModule(mlir::ModuleOp mlirModule, llvm::raw_ostream &out) {
out << mlirModule << '\n';
}
static void registerAllPasses() {
fir::support::registerMLIRPassesForFortranTools();
fir::registerOptTransformPasses();
}
/// Create a target machine that is at least sufficient to get data-layout
/// information required by flang semantics and lowering. Note that it may not
/// contain all the CPU feature information to get optimized assembly generation
/// from LLVM IR. Drivers that needs to generate assembly from LLVM IR should
/// create a target machine according to their specific options.
static std::unique_ptr<llvm::TargetMachine>
createTargetMachine(llvm::StringRef targetTriple, std::string &error) {
std::string triple{targetTriple};
if (triple.empty())
triple = llvm::sys::getDefaultTargetTriple();
const llvm::Target *theTarget =
llvm::TargetRegistry::lookupTarget(triple, error);
if (!theTarget)
return nullptr;
return std::unique_ptr<llvm::TargetMachine>{
theTarget->createTargetMachine(triple, /*CPU=*/"",
/*Features=*/"", llvm::TargetOptions(),
/*Reloc::Model=*/std::nullopt)};
}
//===----------------------------------------------------------------------===//
// Translate Fortran input to FIR, a dialect of MLIR.
//===----------------------------------------------------------------------===//
static mlir::LogicalResult convertFortranSourceToMLIR(
std::string path, Fortran::parser::Options options,
const ProgramName &programPrefix,
Fortran::semantics::SemanticsContext &semanticsContext,
const mlir::PassPipelineCLParser &passPipeline,
const llvm::TargetMachine &targetMachine) {
// prep for prescan and parse
Fortran::parser::Parsing parsing{semanticsContext.allCookedSources()};
parsing.Prescan(path, options);
if (!parsing.messages().empty() && (parsing.messages().AnyFatalError())) {
llvm::errs() << programPrefix << "could not scan " << path << '\n';
parsing.messages().Emit(llvm::errs(), parsing.allCooked());
return mlir::failure();
}
// parse the input Fortran
parsing.Parse(llvm::outs());
parsing.messages().Emit(llvm::errs(), parsing.allCooked());
if (!parsing.consumedWholeFile()) {
parsing.EmitMessage(llvm::errs(), parsing.finalRestingPlace(),
"parser FAIL (final position)",
"error: ", llvm::raw_ostream::RED);
return mlir::failure();
}
if ((!parsing.messages().empty() && (parsing.messages().AnyFatalError())) ||
!parsing.parseTree().has_value()) {
llvm::errs() << programPrefix << "could not parse " << path << '\n';
return mlir::failure();
}
// run semantics
auto &parseTree = *parsing.parseTree();
Fortran::semantics::Semantics semantics(semanticsContext, parseTree);
semantics.Perform();
semantics.EmitMessages(llvm::errs());
if (semantics.AnyFatalError()) {
llvm::errs() << programPrefix << "semantic errors in " << path << '\n';
return mlir::failure();
}
Fortran::semantics::RuntimeDerivedTypeTables tables;
if (!semantics.AnyFatalError()) {
tables =
Fortran::semantics::BuildRuntimeDerivedTypeTables(semanticsContext);
if (!tables.schemata)
llvm::errs() << programPrefix
<< "could not find module file for __fortran_type_info\n";
}
if (dumpSymbols) {
semantics.DumpSymbols(llvm::outs());
return mlir::success();
}
if (pftDumpTest) {
if (auto ast = Fortran::lower::createPFT(parseTree, semanticsContext)) {
Fortran::lower::dumpPFT(llvm::outs(), *ast);
return mlir::success();
}
llvm::errs() << "Pre FIR Tree is NULL.\n";
return mlir::failure();
}
// translate to FIR dialect of MLIR
mlir::DialectRegistry registry;
fir::support::registerNonCodegenDialects(registry);
mlir::MLIRContext ctx(registry);
fir::support::loadNonCodegenDialects(ctx);
auto &defKinds = semanticsContext.defaultKinds();
fir::KindMapping kindMap(
&ctx, llvm::ArrayRef<fir::KindTy>{fir::fromDefaultKinds(defKinds)});
std::string targetTriple = targetMachine.getTargetTriple().normalize();
// Use default lowering options for bbc.
Fortran::lower::LoweringOptions loweringOptions{};
loweringOptions.setPolymorphicTypeImpl(enablePolymorphic);
loweringOptions.setNoPPCNativeVecElemOrder(enableNoPPCNativeVecElemOrder);
loweringOptions.setLowerToHighLevelFIR(useHLFIR || emitHLFIR);
auto burnside = Fortran::lower::LoweringBridge::create(
ctx, semanticsContext, defKinds, semanticsContext.intrinsics(),
semanticsContext.targetCharacteristics(), parsing.allCooked(),
targetTriple, kindMap, loweringOptions, {},
semanticsContext.languageFeatures(), targetMachine);
burnside.lower(parseTree, semanticsContext);
mlir::ModuleOp mlirModule = burnside.getModule();
if (enableOpenMP) {
if (enableOpenMPGPU && !enableOpenMPDevice) {
llvm::errs() << "FATAL: -fopenmp-is-gpu can only be set if "
"-fopenmp-is-target-device is also set";
return mlir::failure();
}
auto offloadModuleOpts =
OffloadModuleOpts(setOpenMPTargetDebug, setOpenMPTeamSubscription,
setOpenMPThreadSubscription, setOpenMPNoThreadState,
setOpenMPNoNestedParallelism, enableOpenMPDevice,
enableOpenMPGPU, setOpenMPVersion, "", setNoGPULib);
setOffloadModuleInterfaceAttributes(mlirModule, offloadModuleOpts);
setOpenMPVersionAttribute(mlirModule, setOpenMPVersion);
}
std::error_code ec;
std::string outputName = outputFilename;
if (!outputName.size())
outputName = llvm::sys::path::stem(inputFilename).str().append(".mlir");
llvm::raw_fd_ostream out(outputName, ec);
if (ec)
return mlir::emitError(mlir::UnknownLoc::get(&ctx),
"could not open output file ")
<< outputName;
// Otherwise run the default passes.
mlir::PassManager pm(mlirModule->getName(),
mlir::OpPassManager::Nesting::Implicit);
if (enableOpenMP)
// WARNING: This pipeline must be run immediately after the lowering to
// ensure that the FIR is correct with respect to OpenMP operations/
// attributes.
fir::createOpenMPFIRPassPipeline(pm, enableOpenMPDevice);
pm.enableVerifier(/*verifyPasses=*/true);
(void)mlir::applyPassManagerCLOptions(pm);
if (passPipeline.hasAnyOccurrences()) {
// run the command-line specified pipeline
hlfir::registerHLFIRPasses();
(void)passPipeline.addToPipeline(pm, [&](const llvm::Twine &msg) {
mlir::emitError(mlir::UnknownLoc::get(&ctx)) << msg;
return mlir::failure();
});
} else if (emitFIR || emitHLFIR) {
// --emit-fir: Build the IR, verify it, and dump the IR if the IR passes
// verification. Use --dump-module-on-failure to dump invalid IR.
pm.addPass(std::make_unique<Fortran::lower::VerifierPass>());
if (mlir::failed(pm.run(mlirModule))) {
llvm::errs() << "FATAL: verification of lowering to FIR failed";
return mlir::failure();
}
if (emitFIR && useHLFIR) {
// lower HLFIR to FIR
fir::createHLFIRToFIRPassPipeline(pm, llvm::OptimizationLevel::O2);
if (mlir::failed(pm.run(mlirModule))) {
llvm::errs() << "FATAL: lowering from HLFIR to FIR failed";
return mlir::failure();
}
}
printModule(mlirModule, out);
return mlir::success();
} else {
// run the default canned pipeline
pm.addPass(std::make_unique<Fortran::lower::VerifierPass>());
// Add O2 optimizer pass pipeline.
fir::createDefaultFIROptimizerPassPipeline(
pm, MLIRToLLVMPassPipelineConfig(llvm::OptimizationLevel::O2));
}
if (mlir::succeeded(pm.run(mlirModule))) {
// Emit MLIR and do not lower to LLVM IR.
printModule(mlirModule, out);
return mlir::success();
}
// Something went wrong. Try to dump the MLIR module.
llvm::errs() << "oops, pass manager reported failure\n";
return mlir::failure();
}
int main(int argc, char **argv) {
[[maybe_unused]] llvm::InitLLVM y(argc, argv);
llvm::InitializeAllTargets();
llvm::InitializeAllTargetMCs();
registerAllPasses();
mlir::registerMLIRContextCLOptions();
mlir::registerAsmPrinterCLOptions();
mlir::registerPassManagerCLOptions();
mlir::PassPipelineCLParser passPipe("", "Compiler passes to run");
llvm::cl::ParseCommandLineOptions(argc, argv, "Burnside Bridge Compiler\n");
ProgramName programPrefix;
programPrefix = argv[0] + ": "s;
if (includeDirs.size() == 0) {
includeDirs.push_back(".");
// Default Fortran modules should be installed in include/flang (a sibling
// to the bin) directory.
intrinsicIncludeDirs.push_back(
llvm::sys::path::parent_path(
llvm::sys::path::parent_path(
llvm::sys::fs::getMainExecutable(argv[0], nullptr)))
.str() +
"/include/flang");
}
Fortran::parser::Options options;
options.predefinitions.emplace_back("__flang__"s, "1"s);
options.predefinitions.emplace_back("__flang_major__"s,
std::string{FLANG_VERSION_MAJOR_STRING});
options.predefinitions.emplace_back("__flang_minor__"s,
std::string{FLANG_VERSION_MINOR_STRING});
options.predefinitions.emplace_back(
"__flang_patchlevel__"s, std::string{FLANG_VERSION_PATCHLEVEL_STRING});
// enable parsing of OpenMP
if (enableOpenMP) {
options.features.Enable(Fortran::common::LanguageFeature::OpenMP);
Fortran::common::setOpenMPMacro(setOpenMPVersion, options.predefinitions);
}
// enable parsing of OpenACC
if (enableOpenACC) {
options.features.Enable(Fortran::common::LanguageFeature::OpenACC);
options.predefinitions.emplace_back("_OPENACC", "202211");
}
// enable parsing of CUDA Fortran
if (enableCUDA) {
options.features.Enable(Fortran::common::LanguageFeature::CUDA);
}
if (fixedForm) {
options.isFixedForm = fixedForm;
}
Fortran::common::IntrinsicTypeDefaultKinds defaultKinds;
Fortran::parser::AllSources allSources;
Fortran::parser::AllCookedSources allCookedSources(allSources);
Fortran::semantics::SemanticsContext semanticsContext{
defaultKinds, options.features, allCookedSources};
semanticsContext.set_moduleDirectory(moduleDir)
.set_moduleFileSuffix(moduleSuffix)
.set_searchDirectories(includeDirs)
.set_intrinsicModuleDirectories(intrinsicIncludeDirs)
.set_warnOnNonstandardUsage(warnStdViolation)
.set_warningsAreErrors(warnIsError);
std::string error;
// Create host target machine.
std::unique_ptr<llvm::TargetMachine> targetMachine =
createTargetMachine(targetTripleOverride, error);
if (!targetMachine) {
llvm::errs() << "failed to create target machine: " << error << "\n";
return mlir::failed(mlir::failure());
}
std::string compilerVersion = Fortran::common::getFlangToolFullVersion("bbc");
std::string compilerOptions = "";
Fortran::tools::setUpTargetCharacteristics(
semanticsContext.targetCharacteristics(), *targetMachine, compilerVersion,
compilerOptions);
return mlir::failed(
convertFortranSourceToMLIR(inputFilename, options, programPrefix,
semanticsContext, passPipe, *targetMachine));
}