llvm-project/llvm/lib/Target/Sparc/SparcTargetMachine.cpp

195 lines
6.8 KiB
C++
Raw Normal View History

//===-- SparcTargetMachine.cpp - Define TargetMachine for Sparc -----------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
//
//===----------------------------------------------------------------------===//
#include "SparcTargetMachine.h"
#include "SparcTargetObjectFile.h"
#include "Sparc.h"
#include "LeonPasses.h"
#include "llvm/CodeGen/Passes.h"
#include "llvm/CodeGen/TargetPassConfig.h"
#include "llvm/IR/LegacyPassManager.h"
#include "llvm/Support/TargetRegistry.h"
using namespace llvm;
extern "C" void LLVMInitializeSparcTarget() {
// Register the target.
RegisterTargetMachine<SparcV8TargetMachine> X(getTheSparcTarget());
RegisterTargetMachine<SparcV9TargetMachine> Y(getTheSparcV9Target());
RegisterTargetMachine<SparcelTargetMachine> Z(getTheSparcelTarget());
}
static std::string computeDataLayout(const Triple &T, bool is64Bit) {
// Sparc is typically big endian, but some are little.
std::string Ret = T.getArch() == Triple::sparcel ? "e" : "E";
Ret += "-m:e";
// Some ABIs have 32bit pointers.
if (!is64Bit)
Ret += "-p:32:32";
// Alignments for 64 bit integers.
Ret += "-i64:64";
// On SparcV9 128 floats are aligned to 128 bits, on others only to 64.
// On SparcV9 registers can hold 64 or 32 bits, on others only 32.
if (is64Bit)
Ret += "-n32:64";
else
Ret += "-f128:64-n32";
if (is64Bit)
Ret += "-S128";
else
Ret += "-S64";
return Ret;
}
static Reloc::Model getEffectiveRelocModel(Optional<Reloc::Model> RM) {
if (!RM.hasValue())
return Reloc::Static;
return *RM;
}
/// Create an ILP32 architecture model
SparcTargetMachine::SparcTargetMachine(const Target &T, const Triple &TT,
StringRef CPU, StringRef FS,
const TargetOptions &Options,
Optional<Reloc::Model> RM,
CodeModel::Model CM,
CodeGenOpt::Level OL, bool is64bit)
: LLVMTargetMachine(T, computeDataLayout(TT, is64bit), TT, CPU, FS, Options,
getEffectiveRelocModel(RM), CM, OL),
TLOF(make_unique<SparcELFTargetObjectFile>()),
Subtarget(TT, CPU, FS, *this, is64bit), is64Bit(is64bit) {
initAsmInfo();
}
SparcTargetMachine::~SparcTargetMachine() {}
const SparcSubtarget *
SparcTargetMachine::getSubtargetImpl(const Function &F) const {
Attribute CPUAttr = F.getFnAttribute("target-cpu");
Attribute FSAttr = F.getFnAttribute("target-features");
std::string CPU = !CPUAttr.hasAttribute(Attribute::None)
? CPUAttr.getValueAsString().str()
: TargetCPU;
std::string FS = !FSAttr.hasAttribute(Attribute::None)
? FSAttr.getValueAsString().str()
: TargetFS;
// FIXME: This is related to the code below to reset the target options,
// we need to know whether or not the soft float flag is set on the
// function, so we can enable it as a subtarget feature.
bool softFloat =
F.hasFnAttribute("use-soft-float") &&
F.getFnAttribute("use-soft-float").getValueAsString() == "true";
if (softFloat)
FS += FS.empty() ? "+soft-float" : ",+soft-float";
auto &I = SubtargetMap[CPU + FS];
if (!I) {
// This needs to be done before we create a new subtarget since any
// creation will depend on the TM and the code generation flags on the
// function that reside in TargetOptions.
resetTargetOptions(F);
I = llvm::make_unique<SparcSubtarget>(TargetTriple, CPU, FS, *this,
this->is64Bit);
}
return I.get();
}
namespace {
/// Sparc Code Generator Pass Configuration Options.
class SparcPassConfig : public TargetPassConfig {
public:
SparcPassConfig(SparcTargetMachine *TM, PassManagerBase &PM)
: TargetPassConfig(TM, PM) {}
SparcTargetMachine &getSparcTargetMachine() const {
return getTM<SparcTargetMachine>();
}
Erase fence insertion from SelectionDAGBuilder.cpp (NFC) Summary: Backends can use setInsertFencesForAtomic to signal to the middle-end that montonic is the only memory ordering they can accept for stores/loads/rmws/cmpxchg. The code lowering those accesses with a stronger ordering to fences + monotonic accesses is currently living in SelectionDAGBuilder.cpp. In this patch I propose moving this logic out of it for several reasons: - There is lots of redundancy to avoid: extremely similar logic already exists in AtomicExpand. - The current code in SelectionDAGBuilder does not use any target-hooks, it does the same transformation for every backend that requires it - As a result it is plain *unsound*, as it was apparently designed for ARM. It happens to mostly work for the other targets because they are extremely conservative, but Power for example had to switch to AtomicExpand to be able to use lwsync safely (see r218331). - Because it produces IR-level fences, it cannot be made sound ! This is noted in the C++11 standard (section 29.3, page 1140): ``` Fences cannot, in general, be used to restore sequential consistency for atomic operations with weaker ordering semantics. ``` It can also be seen by the following example (called IRIW in the litterature): ``` atomic<int> x = y = 0; int r1, r2, r3, r4; Thread 0: x.store(1); Thread 1: y.store(1); Thread 2: r1 = x.load(); r2 = y.load(); Thread 3: r3 = y.load(); r4 = x.load(); ``` r1 = r3 = 1 and r2 = r4 = 0 is impossible as long as the accesses are all seq_cst. But if they are lowered to monotonic accesses, no amount of fences can prevent it.. This patch does three things (I could cut it into parts, but then some of them would not be tested/testable, please tell me if you would prefer that): - it provides a default implementation for emitLeadingFence/emitTrailingFence in terms of IR-level fences, that mimic the original logic of SelectionDAGBuilder. As we saw above, this is unsound, but the best that can be done without knowing the targets well (and there is a comment warning about this risk). - it then switches Mips/Sparc/XCore to use AtomicExpand, relying on this default implementation (that exactly replicates the logic of SelectionDAGBuilder, so no functional change) - it finally erase this logic from SelectionDAGBuilder as it is dead-code. Ideally, each target would define its own override for emitLeading/TrailingFence using target-specific fences, but I do not know the Sparc/Mips/XCore memory model well enough to do this, and they appear to be dealing fine with the ARM-inspired default expansion for now (probably because they are overly conservative, as Power was). If anyone wants to compile fences more agressively on these platforms, the long comment should make it clear why he should first override emitLeading/TrailingFence. Test Plan: make check-all, no functional change Reviewers: jfb, t.p.northover Subscribers: aemerson, llvm-commits Differential Revision: http://reviews.llvm.org/D5474 llvm-svn: 219957
2014-10-17 04:34:57 +08:00
void addIRPasses() override;
bool addInstSelector() override;
void addPreEmitPass() override;
};
} // namespace
TargetPassConfig *SparcTargetMachine::createPassConfig(PassManagerBase &PM) {
return new SparcPassConfig(this, PM);
}
Erase fence insertion from SelectionDAGBuilder.cpp (NFC) Summary: Backends can use setInsertFencesForAtomic to signal to the middle-end that montonic is the only memory ordering they can accept for stores/loads/rmws/cmpxchg. The code lowering those accesses with a stronger ordering to fences + monotonic accesses is currently living in SelectionDAGBuilder.cpp. In this patch I propose moving this logic out of it for several reasons: - There is lots of redundancy to avoid: extremely similar logic already exists in AtomicExpand. - The current code in SelectionDAGBuilder does not use any target-hooks, it does the same transformation for every backend that requires it - As a result it is plain *unsound*, as it was apparently designed for ARM. It happens to mostly work for the other targets because they are extremely conservative, but Power for example had to switch to AtomicExpand to be able to use lwsync safely (see r218331). - Because it produces IR-level fences, it cannot be made sound ! This is noted in the C++11 standard (section 29.3, page 1140): ``` Fences cannot, in general, be used to restore sequential consistency for atomic operations with weaker ordering semantics. ``` It can also be seen by the following example (called IRIW in the litterature): ``` atomic<int> x = y = 0; int r1, r2, r3, r4; Thread 0: x.store(1); Thread 1: y.store(1); Thread 2: r1 = x.load(); r2 = y.load(); Thread 3: r3 = y.load(); r4 = x.load(); ``` r1 = r3 = 1 and r2 = r4 = 0 is impossible as long as the accesses are all seq_cst. But if they are lowered to monotonic accesses, no amount of fences can prevent it.. This patch does three things (I could cut it into parts, but then some of them would not be tested/testable, please tell me if you would prefer that): - it provides a default implementation for emitLeadingFence/emitTrailingFence in terms of IR-level fences, that mimic the original logic of SelectionDAGBuilder. As we saw above, this is unsound, but the best that can be done without knowing the targets well (and there is a comment warning about this risk). - it then switches Mips/Sparc/XCore to use AtomicExpand, relying on this default implementation (that exactly replicates the logic of SelectionDAGBuilder, so no functional change) - it finally erase this logic from SelectionDAGBuilder as it is dead-code. Ideally, each target would define its own override for emitLeading/TrailingFence using target-specific fences, but I do not know the Sparc/Mips/XCore memory model well enough to do this, and they appear to be dealing fine with the ARM-inspired default expansion for now (probably because they are overly conservative, as Power was). If anyone wants to compile fences more agressively on these platforms, the long comment should make it clear why he should first override emitLeading/TrailingFence. Test Plan: make check-all, no functional change Reviewers: jfb, t.p.northover Subscribers: aemerson, llvm-commits Differential Revision: http://reviews.llvm.org/D5474 llvm-svn: 219957
2014-10-17 04:34:57 +08:00
void SparcPassConfig::addIRPasses() {
addPass(createAtomicExpandPass(&getSparcTargetMachine()));
TargetPassConfig::addIRPasses();
}
bool SparcPassConfig::addInstSelector() {
addPass(createSparcISelDag(getSparcTargetMachine()));
return false;
}
void SparcPassConfig::addPreEmitPass(){
addPass(createSparcDelaySlotFillerPass(getSparcTargetMachine()));
if (this->getSparcTargetMachine().getSubtargetImpl()->insertNOPLoad())
{
addPass(new InsertNOPLoad(getSparcTargetMachine()));
}
if (this->getSparcTargetMachine().getSubtargetImpl()->fixFSMULD())
{
addPass(new FixFSMULD(getSparcTargetMachine()));
}
if (this->getSparcTargetMachine().getSubtargetImpl()->replaceFMULS())
{
addPass(new ReplaceFMULS(getSparcTargetMachine()));
}
if (this->getSparcTargetMachine().getSubtargetImpl()->fixAllFDIVSQRT())
{
addPass(new FixAllFDIVSQRT(getSparcTargetMachine()));
}
}
void SparcV8TargetMachine::anchor() { }
SparcV8TargetMachine::SparcV8TargetMachine(const Target &T, const Triple &TT,
StringRef CPU, StringRef FS,
const TargetOptions &Options,
Optional<Reloc::Model> RM,
CodeModel::Model CM,
CodeGenOpt::Level OL)
: SparcTargetMachine(T, TT, CPU, FS, Options, RM, CM, OL, false) {}
void SparcV9TargetMachine::anchor() { }
SparcV9TargetMachine::SparcV9TargetMachine(const Target &T, const Triple &TT,
StringRef CPU, StringRef FS,
const TargetOptions &Options,
Optional<Reloc::Model> RM,
CodeModel::Model CM,
CodeGenOpt::Level OL)
: SparcTargetMachine(T, TT, CPU, FS, Options, RM, CM, OL, true) {}
void SparcelTargetMachine::anchor() {}
SparcelTargetMachine::SparcelTargetMachine(const Target &T, const Triple &TT,
StringRef CPU, StringRef FS,
const TargetOptions &Options,
Optional<Reloc::Model> RM,
CodeModel::Model CM,
CodeGenOpt::Level OL)
: SparcTargetMachine(T, TT, CPU, FS, Options, RM, CM, OL, false) {}