CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
240 lines
8.2 KiB
C++
240 lines
8.2 KiB
C++
//=== lib/CodeGen/GlobalISel/AArch64O0PreLegalizerCombiner.cpp ------------===//
|
|
//
|
|
// 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 pass does combining of machine instructions at the generic MI level,
|
|
// before the legalizer.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "AArch64.h"
|
|
#include "AArch64GlobalISelUtils.h"
|
|
#include "AArch64TargetMachine.h"
|
|
#include "llvm/CodeGen/GlobalISel/Combiner.h"
|
|
#include "llvm/CodeGen/GlobalISel/CombinerHelper.h"
|
|
#include "llvm/CodeGen/GlobalISel/CombinerInfo.h"
|
|
#include "llvm/CodeGen/GlobalISel/GIMatchTableExecutorImpl.h"
|
|
#include "llvm/CodeGen/GlobalISel/GISelValueTracking.h"
|
|
#include "llvm/CodeGen/GlobalISel/MIPatternMatch.h"
|
|
#include "llvm/CodeGen/GlobalISel/MachineIRBuilder.h"
|
|
#include "llvm/CodeGen/LibcallLoweringInfo.h"
|
|
#include "llvm/CodeGen/MachineDominators.h"
|
|
#include "llvm/CodeGen/MachineFunction.h"
|
|
#include "llvm/CodeGen/MachineFunctionAnalysisManager.h"
|
|
#include "llvm/CodeGen/MachineFunctionPass.h"
|
|
#include "llvm/CodeGen/MachinePassManager.h"
|
|
#include "llvm/CodeGen/TargetPassConfig.h"
|
|
#include "llvm/IR/Instructions.h"
|
|
#include <memory>
|
|
|
|
#define GET_GICOMBINER_DEPS
|
|
#include "AArch64GenO0PreLegalizeGICombiner.inc"
|
|
#undef GET_GICOMBINER_DEPS
|
|
|
|
#define DEBUG_TYPE "aarch64-O0-prelegalizer-combiner"
|
|
|
|
using namespace llvm;
|
|
using namespace MIPatternMatch;
|
|
|
|
#define GET_GICOMBINER_TYPES
|
|
#include "AArch64GenO0PreLegalizeGICombiner.inc"
|
|
#undef GET_GICOMBINER_TYPES
|
|
|
|
namespace {
|
|
|
|
class AArch64O0PreLegalizerCombinerImpl : public Combiner {
|
|
protected:
|
|
const CombinerHelper Helper;
|
|
const AArch64O0PreLegalizerCombinerImplRuleConfig &RuleConfig;
|
|
const AArch64Subtarget &STI;
|
|
const LibcallLoweringInfo &Libcalls;
|
|
|
|
public:
|
|
AArch64O0PreLegalizerCombinerImpl(
|
|
MachineFunction &MF, CombinerInfo &CInfo, GISelCSEInfo *CSEInfo,
|
|
const AArch64O0PreLegalizerCombinerImplRuleConfig &RuleConfig,
|
|
const AArch64Subtarget &STI, const LibcallLoweringInfo &Libcalls);
|
|
|
|
static const char *getName() { return "AArch64O0PreLegalizerCombiner"; }
|
|
|
|
bool tryCombineAll(MachineInstr &I) const override;
|
|
|
|
bool tryCombineAllImpl(MachineInstr &I) const;
|
|
|
|
private:
|
|
#define GET_GICOMBINER_CLASS_MEMBERS
|
|
#include "AArch64GenO0PreLegalizeGICombiner.inc"
|
|
#undef GET_GICOMBINER_CLASS_MEMBERS
|
|
};
|
|
|
|
#define GET_GICOMBINER_IMPL
|
|
#include "AArch64GenO0PreLegalizeGICombiner.inc"
|
|
#undef GET_GICOMBINER_IMPL
|
|
|
|
AArch64O0PreLegalizerCombinerImpl::AArch64O0PreLegalizerCombinerImpl(
|
|
MachineFunction &MF, CombinerInfo &CInfo, GISelCSEInfo *CSEInfo,
|
|
const AArch64O0PreLegalizerCombinerImplRuleConfig &RuleConfig,
|
|
const AArch64Subtarget &STI, const LibcallLoweringInfo &Libcalls)
|
|
: Combiner(MF, CInfo, /*VT=*/nullptr, CSEInfo),
|
|
Helper(Observer, B, /*IsPreLegalize*/ true, /*VT=*/nullptr),
|
|
RuleConfig(RuleConfig), STI(STI), Libcalls(Libcalls),
|
|
#define GET_GICOMBINER_CONSTRUCTOR_INITS
|
|
#include "AArch64GenO0PreLegalizeGICombiner.inc"
|
|
#undef GET_GICOMBINER_CONSTRUCTOR_INITS
|
|
{
|
|
}
|
|
|
|
bool AArch64O0PreLegalizerCombinerImpl::tryCombineAll(MachineInstr &MI) const {
|
|
if (tryCombineAllImpl(MI))
|
|
return true;
|
|
|
|
unsigned Opc = MI.getOpcode();
|
|
switch (Opc) {
|
|
case TargetOpcode::G_MEMCPY_INLINE:
|
|
return Helper.tryEmitMemcpyInline(MI);
|
|
case TargetOpcode::G_MEMCPY:
|
|
case TargetOpcode::G_MEMMOVE:
|
|
case TargetOpcode::G_MEMSET: {
|
|
// At -O0 set a maxlen of 32 to inline;
|
|
unsigned MaxLen = 32;
|
|
// Try to inline memcpy type calls if optimizations are enabled.
|
|
if (Helper.tryCombineMemCpyFamily(MI, MaxLen))
|
|
return true;
|
|
if (Opc == TargetOpcode::G_MEMSET)
|
|
return llvm::AArch64GISelUtils::tryEmitBZero(MI, B, Libcalls,
|
|
CInfo.EnableMinSize);
|
|
return false;
|
|
}
|
|
}
|
|
|
|
return false;
|
|
}
|
|
|
|
bool runCombiner(
|
|
MachineFunction &MF, const LibcallLoweringInfo &Libcalls,
|
|
const AArch64O0PreLegalizerCombinerImplRuleConfig &RuleConfig) {
|
|
const Function &F = MF.getFunction();
|
|
const AArch64Subtarget &ST = MF.getSubtarget<AArch64Subtarget>();
|
|
|
|
CombinerInfo CInfo(/*AllowIllegalOps=*/true, /*ShouldLegalizeIllegal=*/false,
|
|
/*LegalizerInfo=*/nullptr, /*EnableOpt=*/false,
|
|
F.hasOptSize(), F.hasMinSize());
|
|
// Disable fixed-point iteration in the Combiner. This improves compile-time
|
|
// at the cost of possibly missing optimizations. See PR#94291 for details.
|
|
CInfo.MaxIterations = 1;
|
|
|
|
AArch64O0PreLegalizerCombinerImpl Impl(MF, CInfo,
|
|
/*CSEInfo*/ nullptr, RuleConfig, ST,
|
|
Libcalls);
|
|
return Impl.combineMachineInstrs();
|
|
}
|
|
|
|
// Pass boilerplate
|
|
// ================
|
|
|
|
class AArch64O0PreLegalizerCombinerLegacy : public MachineFunctionPass {
|
|
public:
|
|
static char ID;
|
|
|
|
AArch64O0PreLegalizerCombinerLegacy();
|
|
|
|
StringRef getPassName() const override {
|
|
return "AArch64O0PreLegalizerCombiner";
|
|
}
|
|
|
|
bool runOnMachineFunction(MachineFunction &MF) override;
|
|
|
|
void getAnalysisUsage(AnalysisUsage &AU) const override;
|
|
|
|
private:
|
|
AArch64O0PreLegalizerCombinerImplRuleConfig RuleConfig;
|
|
};
|
|
} // end anonymous namespace
|
|
|
|
void AArch64O0PreLegalizerCombinerLegacy::getAnalysisUsage(
|
|
AnalysisUsage &AU) const {
|
|
AU.setPreservesCFG();
|
|
getSelectionDAGFallbackAnalysisUsage(AU);
|
|
AU.addRequired<LibcallLoweringInfoWrapper>();
|
|
MachineFunctionPass::getAnalysisUsage(AU);
|
|
}
|
|
|
|
AArch64O0PreLegalizerCombinerLegacy::AArch64O0PreLegalizerCombinerLegacy()
|
|
: MachineFunctionPass(ID) {
|
|
if (!RuleConfig.parseCommandLineOption())
|
|
report_fatal_error("Invalid rule identifier");
|
|
}
|
|
|
|
bool AArch64O0PreLegalizerCombinerLegacy::runOnMachineFunction(
|
|
MachineFunction &MF) {
|
|
if (MF.getProperties().hasFailedISel())
|
|
return false;
|
|
|
|
const Function &F = MF.getFunction();
|
|
|
|
const AArch64Subtarget &ST = MF.getSubtarget<AArch64Subtarget>();
|
|
const LibcallLoweringInfo &Libcalls =
|
|
getAnalysis<LibcallLoweringInfoWrapper>().getLibcallLowering(
|
|
*F.getParent(), ST);
|
|
|
|
return runCombiner(MF, Libcalls, RuleConfig);
|
|
}
|
|
|
|
char AArch64O0PreLegalizerCombinerLegacy::ID = 0;
|
|
INITIALIZE_PASS_BEGIN(AArch64O0PreLegalizerCombinerLegacy, DEBUG_TYPE,
|
|
"Combine AArch64 machine instrs before legalization",
|
|
false, false)
|
|
INITIALIZE_PASS_DEPENDENCY(GISelCSEAnalysisWrapperPass)
|
|
INITIALIZE_PASS_DEPENDENCY(LibcallLoweringInfoWrapper)
|
|
INITIALIZE_PASS_END(AArch64O0PreLegalizerCombinerLegacy, DEBUG_TYPE,
|
|
"Combine AArch64 machine instrs before legalization", false,
|
|
false)
|
|
|
|
AArch64O0PreLegalizerCombinerPass::AArch64O0PreLegalizerCombinerPass()
|
|
: RuleConfig(
|
|
std::make_unique<AArch64O0PreLegalizerCombinerImplRuleConfig>()) {
|
|
if (!RuleConfig->parseCommandLineOption())
|
|
report_fatal_error("Invalid rule identifier");
|
|
}
|
|
|
|
AArch64O0PreLegalizerCombinerPass::AArch64O0PreLegalizerCombinerPass(
|
|
AArch64O0PreLegalizerCombinerPass &&) = default;
|
|
|
|
AArch64O0PreLegalizerCombinerPass::~AArch64O0PreLegalizerCombinerPass() =
|
|
default;
|
|
|
|
PreservedAnalyses
|
|
AArch64O0PreLegalizerCombinerPass::run(MachineFunction &MF,
|
|
MachineFunctionAnalysisManager &MFAM) {
|
|
if (MF.getProperties().hasFailedISel())
|
|
return PreservedAnalyses::all();
|
|
|
|
const AArch64Subtarget &ST = MF.getSubtarget<AArch64Subtarget>();
|
|
auto &MAMProxy =
|
|
MFAM.getResult<ModuleAnalysisManagerMachineFunctionProxy>(MF);
|
|
const LibcallLoweringModuleAnalysisResult *LibcallResult =
|
|
MAMProxy.getCachedResult<LibcallLoweringModuleAnalysis>(
|
|
*MF.getFunction().getParent());
|
|
if (!LibcallResult)
|
|
reportFatalUsageError("LibcallLoweringModuleAnalysis result not available");
|
|
|
|
const LibcallLoweringInfo &Libcalls = LibcallResult->getLibcallLowering(ST);
|
|
|
|
if (!runCombiner(MF, Libcalls, *RuleConfig))
|
|
return PreservedAnalyses::all();
|
|
|
|
PreservedAnalyses PA = getMachineFunctionPassPreservedAnalyses();
|
|
PA.preserveSet<CFGAnalyses>();
|
|
return PA;
|
|
}
|
|
|
|
namespace llvm {
|
|
FunctionPass *createAArch64O0PreLegalizerCombiner() {
|
|
return new AArch64O0PreLegalizerCombinerLegacy();
|
|
}
|
|
} // end namespace llvm
|