Files
FEX-Emu--FEX/External/FEXCore/Source/Interface/IR/Passes.h
T
lioncash 5f9052a675 CoreState: Use a union for representing xmm data
In the event that the host doesn't support the requirements for running
AVX-enabled applications (SVE2 with at least 256-bit wide vectors) but
still wanted to run regular SSE-enabled applications, they would be
taking a performance hit due to a load/store pessimization (necessary in
order for 256-bit loads/stores to work)

However, we can add an alternate view into the xmm data that would allow
those hosts to use the previous optimization, while still supporting
AVX-capable hosts.
2022-07-26 16:56:54 -04:00

35 lines
1.5 KiB
C++

#pragma once
#include <memory>
namespace FEXCore::Utils {
class IntrusivePooledAllocator;
}
namespace FEXCore::IR {
class Pass;
class RegisterAllocationPass;
class RegisterAllocationData;
std::unique_ptr<FEXCore::IR::Pass> CreateConstProp(bool InlineConstants, bool SupportsTSOImm9);
std::unique_ptr<FEXCore::IR::Pass> CreateContextLoadStoreElimination(bool SupportsAVX);
std::unique_ptr<FEXCore::IR::Pass> CreateSyscallOptimization();
std::unique_ptr<FEXCore::IR::Pass> CreateDeadFlagCalculationEliminination();
std::unique_ptr<FEXCore::IR::Pass> CreateDeadStoreElimination(bool SupportsAVX);
std::unique_ptr<FEXCore::IR::Pass> CreatePassDeadCodeElimination();
std::unique_ptr<FEXCore::IR::Pass> CreateIRCompaction(FEXCore::Utils::IntrusivePooledAllocator &Allocator);
std::unique_ptr<FEXCore::IR::RegisterAllocationPass> CreateRegisterAllocationPass(FEXCore::IR::Pass* CompactionPass,
bool OptimizeSRA,
bool SupportsAVX);
std::unique_ptr<FEXCore::IR::Pass> CreateStaticRegisterAllocationPass(bool SupportsAVX);
std::unique_ptr<FEXCore::IR::Pass> CreateLongDivideEliminationPass();
namespace Validation {
std::unique_ptr<FEXCore::IR::Pass> CreateIRValidation();
std::unique_ptr<FEXCore::IR::Pass> CreateRAValidation();
std::unique_ptr<FEXCore::IR::Pass> CreatePhiValidation();
std::unique_ptr<FEXCore::IR::Pass> CreateValueDominanceValidation();
}
}