Ryujinx/ARMeilleure/Translation/ArmEmitterContext.cs
LDj3SNuD 5e724cf24e
Add Profiled Persistent Translation Cache. (#769)
* Delete DelegateTypes.cs

* Delete DelegateCache.cs

* Add files via upload

* Update Horizon.cs

* Update Program.cs

* Update MainWindow.cs

* Update Aot.cs

* Update RelocEntry.cs

* Update Translator.cs

* Update MemoryManager.cs

* Update InstEmitMemoryHelper.cs

* Update Delegates.cs

* Nit.

* Nit.

* Nit.

* 10 fewer MSIL bytes for us

* Add comment. Nits.

* Update Translator.cs

* Update Aot.cs

* Nits.

* Opt..

* Opt..

* Opt..

* Opt..

* Allow to change compression level.

* Update MemoryManager.cs

* Update Translator.cs

* Manage corner cases during the save phase. Nits.

* Update Aot.cs

* Translator response tweak for Aot disabled. Nit.

* Nit.

* Nits.

* Create DelegateHelpers.cs

* Update Delegates.cs

* Nit.

* Nit.

* Nits.

* Fix due to #784.

* Fixes due to #757 & #841.

* Fix due to #846.

* Fix due to #847.

* Use MethodInfo for managed method calls.

Use IR methods instead of managed methods about Max/Min (S/U).
Follow-ups & Nits.

* Add missing exception messages.

Reintroduce slow path for Fmov_Vi.
Implement slow path for Fmov_Si.

* Switch to the new folder structure.

Nits.

* Impl. index-based relocation information. Impl. cache file version field.

* Nit.

* Address gdkchan comments.

Mainly:
- fixed cache file corruption issue on exit; - exposed a way to disable AOT on the GUI.

* Address AcK77 comment.

* Address Thealexbarney, jduncanator & emmauss comments.

Header magic, CpuId (FI) & Aot -> Ptc.

* Adaptation to the new application reloading system.

Improvements to the call system of managed methods.
Follow-ups.
Nits.

* Get the same boot times as on master when PTC is disabled.

* Profiled Aot.

* A32 support (#897).

* #975 support (1 of 2).

* #975 support (2 of 2).

* Rebase fix & nits.

* Some fixes and nits (still one bug left).

* One fix & nits.

* Tests fix (by gdk) & nits.

* Support translations not only in high quality and rejit.

Nits.

* Added possibility to skip translations and continue execution, using `ESC` key.

* Update SettingsWindow.cs

* Update GLRenderer.cs

* Update Ptc.cs

* Disabled Profiled PTC by default as requested in the past by gdk.

* Fix rejit bug. Increased number of parallel translations. Add stack unwinding stuffs support (1 of 2).

Nits.

* Add stack unwinding stuffs support (2 of 2). Tuned number of parallel translations.

* Restored the ability to assemble jumps with 8-bit offset when Profiled PTC is disabled or during profiling.

Modifications due to rebase.
Nits.

* Limited profiling of the functions to be translated to the addresses belonging to the range of static objects only.

* Nits.

* Nits.

* Update Delegates.cs

* Nit.

* Update InstEmitSimdArithmetic.cs

* Address riperiperi comments.

* Fixed the issue of unjustifiably longer boot times at the second boot than at the first boot, measured at the same time or reference point and with the same number of translated functions.

* Implemented a simple redundant load/save mechanism.

Halved the value of Decoder.MaxInstsPerFunction more appropriate for the current performance of the Translator.
Replaced by Logger.PrintError to Logger.PrintDebug in TexturePool.cs about the supposed invalid texture format to avoid the spawn of the log.
Nits.

* Nit.

Improved Logger.PrintError in TexturePool.cs to avoid log spawn.
Added missing code for FZ handling (in output) for fp max/min instructions (slow paths).

* Add configuration migration for PTC

Co-authored-by: Thog <me@thog.eu>
2020-06-16 20:28:02 +02:00

162 lines
5.6 KiB
C#

using ARMeilleure.Decoders;
using ARMeilleure.Instructions;
using ARMeilleure.IntermediateRepresentation;
using ARMeilleure.Memory;
using ARMeilleure.State;
using System.Collections.Generic;
using static ARMeilleure.IntermediateRepresentation.OperandHelper;
namespace ARMeilleure.Translation
{
class ArmEmitterContext : EmitterContext
{
private readonly Dictionary<ulong, Operand> _labels;
private OpCode _optOpLastCompare;
private OpCode _optOpLastFlagSet;
private Operand _optCmpTempN;
private Operand _optCmpTempM;
private Block _currBlock;
public Block CurrBlock
{
get
{
return _currBlock;
}
set
{
_currBlock = value;
ResetBlockState();
}
}
public OpCode CurrOp { get; set; }
public IMemoryManager Memory { get; }
public Aarch32Mode Mode { get; }
public JumpTable JumpTable { get; }
public long BaseAddress { get; }
public bool HighCq { get; }
public ArmEmitterContext(IMemoryManager memory, JumpTable jumpTable, long baseAddress, bool highCq, Aarch32Mode mode)
{
Memory = memory;
JumpTable = jumpTable;
BaseAddress = baseAddress;
HighCq = highCq;
Mode = mode;
_labels = new Dictionary<ulong, Operand>();
}
public Operand GetLabel(ulong address)
{
if (!_labels.TryGetValue(address, out Operand label))
{
label = Label();
_labels.Add(address, label);
}
return label;
}
public void MarkComparison(Operand n, Operand m)
{
_optOpLastCompare = CurrOp;
_optCmpTempN = Copy(n);
_optCmpTempM = Copy(m);
}
public void MarkFlagSet(PState stateFlag)
{
// Set this only if any of the NZCV flag bits were modified.
// This is used to ensure that when emiting a direct IL branch
// instruction for compare + branch sequences, we're not expecting
// to use comparison values from an old instruction, when in fact
// the flags were already overwritten by another instruction further along.
if (stateFlag >= PState.VFlag)
{
_optOpLastFlagSet = CurrOp;
}
}
private void ResetBlockState()
{
_optOpLastCompare = null;
_optOpLastFlagSet = null;
}
public Operand TryGetComparisonResult(Condition condition)
{
if (_optOpLastCompare == null || _optOpLastCompare != _optOpLastFlagSet)
{
return null;
}
Operand n = _optCmpTempN;
Operand m = _optCmpTempM;
InstName cmpName = _optOpLastCompare.Instruction.Name;
if (cmpName == InstName.Subs)
{
switch (condition)
{
case Condition.Eq: return ICompareEqual (n, m);
case Condition.Ne: return ICompareNotEqual (n, m);
case Condition.GeUn: return ICompareGreaterOrEqualUI(n, m);
case Condition.LtUn: return ICompareLessUI (n, m);
case Condition.GtUn: return ICompareGreaterUI (n, m);
case Condition.LeUn: return ICompareLessOrEqualUI (n, m);
case Condition.Ge: return ICompareGreaterOrEqual (n, m);
case Condition.Lt: return ICompareLess (n, m);
case Condition.Gt: return ICompareGreater (n, m);
case Condition.Le: return ICompareLessOrEqual (n, m);
}
}
else if (cmpName == InstName.Adds && _optOpLastCompare is IOpCodeAluImm op)
{
// There are several limitations that needs to be taken into account for CMN comparisons:
// - The unsigned comparisons are not valid, as they depend on the
// carry flag value, and they will have different values for addition and
// subtraction. For addition, it's carry, and for subtraction, it's borrow.
// So, we need to make sure we're not doing a unsigned compare for the CMN case.
// - We can only do the optimization for the immediate variants,
// because when the second operand value is exactly INT_MIN, we can't
// negate the value as theres no positive counterpart.
// Such invalid values can't be encoded on the immediate encodings.
if (op.RegisterSize == RegisterSize.Int32)
{
m = Const((int)-op.Immediate);
}
else
{
m = Const(-op.Immediate);
}
switch (condition)
{
case Condition.Eq: return ICompareEqual (n, m);
case Condition.Ne: return ICompareNotEqual (n, m);
case Condition.Ge: return ICompareGreaterOrEqual(n, m);
case Condition.Lt: return ICompareLess (n, m);
case Condition.Gt: return ICompareGreater (n, m);
case Condition.Le: return ICompareLessOrEqual (n, m);
}
}
return null;
}
}
}