llvm/tools/llc/llc.cpp

358 lines
12 KiB
C++
Raw Normal View History

//===-- llc.cpp - Implement the LLVM Native Code Generator ----------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This is the llc code generator driver. It provides a convenient
// command-line interface for generating native assembly-language code
// or C code, given LLVM bitcode.
//
//===----------------------------------------------------------------------===//
#include "llvm/ADT/Triple.h"
#include "llvm/Analysis/TargetLibraryInfo.h"
#include "llvm/CodeGen/CommandFlags.h"
#include "llvm/CodeGen/LinkAllAsmWriterComponents.h"
#include "llvm/CodeGen/LinkAllCodegenComponents.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/IRPrintingPasses.h"
#include "llvm/IR/LLVMContext.h"
#include "llvm/IR/Module.h"
#include "llvm/IRReader/IRReader.h"
#include "llvm/MC/SubtargetFeature.h"
#include "llvm/Pass.h"
#include "llvm/PassManager.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/FormattedStream.h"
#include "llvm/Support/Host.h"
#include "llvm/Support/ManagedStatic.h"
#include "llvm/Support/PluginLoader.h"
#include "llvm/Support/PrettyStackTrace.h"
#include "llvm/Support/Signals.h"
#include "llvm/Support/SourceMgr.h"
#include "llvm/Support/TargetRegistry.h"
#include "llvm/Support/TargetSelect.h"
#include "llvm/Support/ToolOutputFile.h"
#include "llvm/Target/TargetMachine.h"
#include "llvm/Target/TargetSubtargetInfo.h"
#include <memory>
using namespace llvm;
// General options for llc. Other pass-specific options are specified
// within the corresponding llc passes, and target-specific options
// and back-end code generation options are specified with the target machine.
//
static cl::opt<std::string>
InputFilename(cl::Positional, cl::desc("<input bitcode>"), cl::init("-"));
static cl::opt<std::string>
OutputFilename("o", cl::desc("Output filename"), cl::value_desc("filename"));
static cl::opt<unsigned>
TimeCompilations("time-compilations", cl::Hidden, cl::init(1u),
cl::value_desc("N"),
cl::desc("Repeat compilation N times for timing"));
static cl::opt<bool>
NoIntegratedAssembler("no-integrated-as", cl::Hidden,
cl::desc("Disable integrated assembler"));
// Determine optimization level.
static cl::opt<char>
OptLevel("O",
cl::desc("Optimization level. [-O0, -O1, -O2, or -O3] "
"(default = '-O2')"),
cl::Prefix,
cl::ZeroOrMore,
cl::init(' '));
static cl::opt<std::string>
TargetTriple("mtriple", cl::desc("Override target triple for module"));
static cl::opt<bool> NoVerify("disable-verify", cl::Hidden,
cl::desc("Do not verify input module"));
static cl::opt<bool> DisableSimplifyLibCalls("disable-simplify-libcalls",
cl::desc("Disable simplify-libcalls"));
static cl::opt<bool> ShowMCEncoding("show-mc-encoding", cl::Hidden,
cl::desc("Show encoding in .s output"));
static cl::opt<bool> EnableDwarfDirectory(
"enable-dwarf-directory", cl::Hidden,
cl::desc("Use .file directives with an explicit directory."));
static cl::opt<bool> AsmVerbose("asm-verbose",
cl::desc("Add comments to directives."),
cl::init(true));
static int compileModule(char **, LLVMContext &);
static std::unique_ptr<tool_output_file>
GetOutputStream(const char *TargetName, Triple::OSType OS,
const char *ProgName) {
// If we don't yet have an output filename, make one.
if (OutputFilename.empty()) {
if (InputFilename == "-")
OutputFilename = "-";
else {
// If InputFilename ends in .bc or .ll, remove it.
StringRef IFN = InputFilename;
if (IFN.endswith(".bc") || IFN.endswith(".ll"))
OutputFilename = IFN.drop_back(3);
else
OutputFilename = IFN;
switch (FileType) {
case TargetMachine::CGFT_AssemblyFile:
if (TargetName[0] == 'c') {
if (TargetName[1] == 0)
OutputFilename += ".cbe.c";
else if (TargetName[1] == 'p' && TargetName[2] == 'p')
OutputFilename += ".cpp";
else
OutputFilename += ".s";
} else
OutputFilename += ".s";
break;
case TargetMachine::CGFT_ObjectFile:
if (OS == Triple::Win32)
OutputFilename += ".obj";
else
OutputFilename += ".o";
break;
case TargetMachine::CGFT_Null:
OutputFilename += ".null";
break;
}
}
}
// Decide if we need "binary" output.
bool Binary = false;
switch (FileType) {
case TargetMachine::CGFT_AssemblyFile:
break;
case TargetMachine::CGFT_ObjectFile:
case TargetMachine::CGFT_Null:
Binary = true;
break;
}
// Open the file.
std::error_code EC;
sys::fs::OpenFlags OpenFlags = sys::fs::F_None;
if (!Binary)
OpenFlags |= sys::fs::F_Text;
auto FDOut = llvm::make_unique<tool_output_file>(OutputFilename, EC,
OpenFlags);
if (EC) {
errs() << EC.message() << '\n';
return nullptr;
}
return FDOut;
}
// main - Entry point for the llc compiler.
//
int main(int argc, char **argv) {
sys::PrintStackTraceOnErrorSignal();
PrettyStackTraceProgram X(argc, argv);
// Enable debug stream buffering.
EnableDebugBuffering = true;
LLVMContext &Context = getGlobalContext();
llvm_shutdown_obj Y; // Call llvm_shutdown() on exit.
// Initialize targets first, so that --version shows registered targets.
InitializeAllTargets();
InitializeAllTargetMCs();
InitializeAllAsmPrinters();
InitializeAllAsmParsers();
// Initialize codegen and IR passes used by llc so that the -print-after,
// -print-before, and -stop-after options work.
PassRegistry *Registry = PassRegistry::getPassRegistry();
initializeCore(*Registry);
initializeCodeGen(*Registry);
initializeLoopStrengthReducePass(*Registry);
initializeLowerIntrinsicsPass(*Registry);
initializeUnreachableBlockElimPass(*Registry);
// Register the target printer for --version.
cl::AddExtraVersionPrinter(TargetRegistry::printRegisteredTargetsForVersion);
cl::ParseCommandLineOptions(argc, argv, "llvm system compiler\n");
// Compile the module TimeCompilations times to give better compile time
// metrics.
for (unsigned I = TimeCompilations; I; --I)
if (int RetVal = compileModule(argv, Context))
return RetVal;
return 0;
}
static int compileModule(char **argv, LLVMContext &Context) {
// Load the module to be compiled...
SMDiagnostic Err;
std::unique_ptr<Module> M;
Triple TheTriple;
bool SkipModule = MCPU == "help" ||
(!MAttrs.empty() && MAttrs.front() == "help");
// If user asked for the 'native' CPU, autodetect here. If autodection fails,
// this will set the CPU to an empty string which tells the target to
// pick a basic default.
if (MCPU == "native")
MCPU = sys::getHostCPUName();
// If user just wants to list available options, skip module loading
if (!SkipModule) {
M = parseIRFile(InputFilename, Err, Context);
if (!M) {
Err.print(argv[0], errs());
return 1;
}
// If we are supposed to override the target triple, do so now.
if (!TargetTriple.empty())
M->setTargetTriple(Triple::normalize(TargetTriple));
TheTriple = Triple(M->getTargetTriple());
} else {
TheTriple = Triple(Triple::normalize(TargetTriple));
}
if (TheTriple.getTriple().empty())
TheTriple.setTriple(sys::getDefaultTargetTriple());
// Get the target specific parser.
std::string Error;
const Target *TheTarget = TargetRegistry::lookupTarget(MArch, TheTriple,
Error);
if (!TheTarget) {
errs() << argv[0] << ": " << Error;
return 1;
}
// Package up features to be passed to target/subtarget
std::string FeaturesStr;
if (MAttrs.size()) {
SubtargetFeatures Features;
for (unsigned i = 0; i != MAttrs.size(); ++i)
Features.AddFeature(MAttrs[i]);
FeaturesStr = Features.getString();
}
CodeGenOpt::Level OLvl = CodeGenOpt::Default;
switch (OptLevel) {
default:
errs() << argv[0] << ": invalid optimization level.\n";
return 1;
case ' ': break;
case '0': OLvl = CodeGenOpt::None; break;
case '1': OLvl = CodeGenOpt::Less; break;
case '2': OLvl = CodeGenOpt::Default; break;
case '3': OLvl = CodeGenOpt::Aggressive; break;
}
TargetOptions Options = InitTargetOptionsFromCodeGenFlags();
Options.DisableIntegratedAS = NoIntegratedAssembler;
Options.MCOptions.ShowMCEncoding = ShowMCEncoding;
Options.MCOptions.MCUseDwarfDirectory = EnableDwarfDirectory;
Options.MCOptions.AsmVerbose = AsmVerbose;
std::unique_ptr<TargetMachine> Target(
TheTarget->createTargetMachine(TheTriple.getTriple(), MCPU, FeaturesStr,
Options, RelocModel, CMModel, OLvl));
assert(Target && "Could not allocate target machine!");
// If we don't have a module then just exit now. We do this down
// here since the CPU/Feature help is underneath the target machine
// creation.
if (SkipModule)
return 0;
assert(M && "Should have exited if we didn't have a module!");
if (GenerateSoftFloatCalls)
FloatABIForCalls = FloatABI::Soft;
// Figure out where we are going to send the output.
std::unique_ptr<tool_output_file> Out =
GetOutputStream(TheTarget->getName(), TheTriple.getOS(), argv[0]);
if (!Out) return 1;
// Build up all of the passes that we want to do to the module.
PassManager PM;
// Add an appropriate TargetLibraryInfo pass for the module's triple.
[PM] Rework how the TargetLibraryInfo pass integrates with the new pass manager to support the actual uses of it. =] When I ported instcombine to the new pass manager I discover that it didn't work because TLI wasn't available in the right places. This is a somewhat surprising and/or subtle aspect of the new pass manager design that came up before but I think is useful to be reminded of: While the new pass manager *allows* a function pass to query a module analysis, it requires that the module analysis is already run and cached prior to the function pass manager starting up, possibly with a 'require<foo>' style utility in the pass pipeline. This is an intentional hurdle because using a module analysis from a function pass *requires* that the module analysis is run prior to entering the function pass manager. Otherwise the other functions in the module could be in who-knows-what state, etc. A somewhat surprising consequence of this design decision (at least to me) is that you have to design a function pass that leverages a module analysis to do so as an optional feature. Even if that means your function pass does no work in the absence of the module analysis, you have to handle that possibility and remain conservatively correct. This is a natural consequence of things being able to invalidate the module analysis and us being unable to re-run it. And it's a generally good thing because it lets us reorder passes arbitrarily without breaking correctness, etc. This ends up causing problems in one case. What if we have a module analysis that is *definitionally* impossible to invalidate. In the places this might come up, the analysis is usually also definitionally trivial to run even while other transformation passes run on the module, regardless of the state of anything. And so, it follows that it is natural to have a hard requirement on such analyses from a function pass. It turns out, that TargetLibraryInfo is just such an analysis, and InstCombine has a hard requirement on it. The approach I've taken here is to produce an analysis that models this flexibility by making it both a module and a function analysis. This exposes the fact that it is in fact safe to compute at any point. We can even make it a valid CGSCC analysis at some point if that is useful. However, we don't want to have a copy of the actual target library info state for each function! This state is specific to the triple. The somewhat direct and blunt approach here is to turn TLI into a pimpl, with the state and mutators in the implementation class and the query routines primarily in the wrapper. Then the analysis can lazily construct and cache the implementations, keyed on the triple, and on-demand produce wrappers of them for each function. One minor annoyance is that we will end up with a wrapper for each function in the module. While this is a bit wasteful (one pointer per function) it seems tolerable. And it has the advantage of ensuring that we pay the absolute minimum synchronization cost to access this information should we end up with a nice parallel function pass manager in the future. We could look into trying to mark when analysis results are especially cheap to recompute and more eagerly GC-ing the cached results, or we could look at supporting a variant of analyses whose results are specifically *not* cached and expected to just be used and discarded by the consumer. Either way, these seem like incremental enhancements that should happen when we start profiling the memory and CPU usage of the new pass manager and not before. The other minor annoyance is that if we end up using the TLI in both a module pass and a function pass, those will be produced by two separate analyses, and thus will point to separate copies of the implementation state. While a minor issue, I dislike this and would like to find a way to cleanly allow a single analysis instance to be used across multiple IR unit managers. But I don't have a good solution to this today, and I don't want to hold up all of the work waiting to come up with one. This too seems like a reasonable thing to incrementally improve later. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@226981 91177308-0d34-0410-b5e6-96231b3b80d8
2015-01-24 02:06:09 +00:00
TargetLibraryInfoImpl TLII(Triple(M->getTargetTriple()));
// The -disable-simplify-libcalls flag actually disables all builtin optzns.
if (DisableSimplifyLibCalls)
[PM] Rework how the TargetLibraryInfo pass integrates with the new pass manager to support the actual uses of it. =] When I ported instcombine to the new pass manager I discover that it didn't work because TLI wasn't available in the right places. This is a somewhat surprising and/or subtle aspect of the new pass manager design that came up before but I think is useful to be reminded of: While the new pass manager *allows* a function pass to query a module analysis, it requires that the module analysis is already run and cached prior to the function pass manager starting up, possibly with a 'require<foo>' style utility in the pass pipeline. This is an intentional hurdle because using a module analysis from a function pass *requires* that the module analysis is run prior to entering the function pass manager. Otherwise the other functions in the module could be in who-knows-what state, etc. A somewhat surprising consequence of this design decision (at least to me) is that you have to design a function pass that leverages a module analysis to do so as an optional feature. Even if that means your function pass does no work in the absence of the module analysis, you have to handle that possibility and remain conservatively correct. This is a natural consequence of things being able to invalidate the module analysis and us being unable to re-run it. And it's a generally good thing because it lets us reorder passes arbitrarily without breaking correctness, etc. This ends up causing problems in one case. What if we have a module analysis that is *definitionally* impossible to invalidate. In the places this might come up, the analysis is usually also definitionally trivial to run even while other transformation passes run on the module, regardless of the state of anything. And so, it follows that it is natural to have a hard requirement on such analyses from a function pass. It turns out, that TargetLibraryInfo is just such an analysis, and InstCombine has a hard requirement on it. The approach I've taken here is to produce an analysis that models this flexibility by making it both a module and a function analysis. This exposes the fact that it is in fact safe to compute at any point. We can even make it a valid CGSCC analysis at some point if that is useful. However, we don't want to have a copy of the actual target library info state for each function! This state is specific to the triple. The somewhat direct and blunt approach here is to turn TLI into a pimpl, with the state and mutators in the implementation class and the query routines primarily in the wrapper. Then the analysis can lazily construct and cache the implementations, keyed on the triple, and on-demand produce wrappers of them for each function. One minor annoyance is that we will end up with a wrapper for each function in the module. While this is a bit wasteful (one pointer per function) it seems tolerable. And it has the advantage of ensuring that we pay the absolute minimum synchronization cost to access this information should we end up with a nice parallel function pass manager in the future. We could look into trying to mark when analysis results are especially cheap to recompute and more eagerly GC-ing the cached results, or we could look at supporting a variant of analyses whose results are specifically *not* cached and expected to just be used and discarded by the consumer. Either way, these seem like incremental enhancements that should happen when we start profiling the memory and CPU usage of the new pass manager and not before. The other minor annoyance is that if we end up using the TLI in both a module pass and a function pass, those will be produced by two separate analyses, and thus will point to separate copies of the implementation state. While a minor issue, I dislike this and would like to find a way to cleanly allow a single analysis instance to be used across multiple IR unit managers. But I don't have a good solution to this today, and I don't want to hold up all of the work waiting to come up with one. This too seems like a reasonable thing to incrementally improve later. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@226981 91177308-0d34-0410-b5e6-96231b3b80d8
2015-01-24 02:06:09 +00:00
TLII.disableAllFunctions();
PM.add(new TargetLibraryInfoWrapperPass(TLII));
// Add the target data from the target machine, if it exists, or the module.
if (const DataLayout *DL = Target->getDataLayout())
M->setDataLayout(DL);
PM.add(new DataLayoutPass());
if (RelaxAll.getNumOccurrences() > 0 &&
FileType != TargetMachine::CGFT_ObjectFile)
errs() << argv[0]
<< ": warning: ignoring -mc-relax-all because filetype != obj";
{
formatted_raw_ostream FOS(Out->os());
AnalysisID StartAfterID = nullptr;
AnalysisID StopAfterID = nullptr;
const PassRegistry *PR = PassRegistry::getPassRegistry();
if (!StartAfter.empty()) {
const PassInfo *PI = PR->getPassInfo(StartAfter);
if (!PI) {
errs() << argv[0] << ": start-after pass is not registered.\n";
return 1;
}
StartAfterID = PI->getTypeInfo();
}
if (!StopAfter.empty()) {
const PassInfo *PI = PR->getPassInfo(StopAfter);
if (!PI) {
errs() << argv[0] << ": stop-after pass is not registered.\n";
return 1;
}
StopAfterID = PI->getTypeInfo();
}
// Ask the target to add backend passes as necessary.
if (Target->addPassesToEmitFile(PM, FOS, FileType, NoVerify,
StartAfterID, StopAfterID)) {
errs() << argv[0] << ": target does not support generation of this"
<< " file type!\n";
return 1;
}
// Before executing passes, print the final values of the LLVM options.
cl::PrintOptionValues();
PM.run(*M);
}
// Declare success.
Out->keep();
return 0;
}