Ryujinx/ARMeilleure/Instructions/InstEmitSimdShift32.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

278 lines
10 KiB
C#

using ARMeilleure.Decoders;
using ARMeilleure.IntermediateRepresentation;
using ARMeilleure.Translation;
using System;
using System.Diagnostics;
using System.Reflection;
using static ARMeilleure.Instructions.InstEmitSimdHelper32;
using static ARMeilleure.IntermediateRepresentation.OperandHelper;
namespace ARMeilleure.Instructions
{
static partial class InstEmit32
{
public static void Vqrshrn(ArmEmitterContext context)
{
OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;
EmitRoundShrImmSaturatingNarrowOp(context, op.U ? ShrImmSaturatingNarrowFlags.VectorZxZx : ShrImmSaturatingNarrowFlags.VectorSxSx);
}
public static void Vqrshrun(ArmEmitterContext context)
{
EmitRoundShrImmSaturatingNarrowOp(context, ShrImmSaturatingNarrowFlags.VectorSxZx);
}
public static void Vrshr(ArmEmitterContext context)
{
OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;
int shift = GetImmShr(op);
long roundConst = 1L << (shift - 1);
if (op.U)
{
if (op.Size < 2)
{
EmitVectorUnaryOpZx32(context, (op1) =>
{
op1 = context.Add(op1, Const(op1.Type, roundConst));
return context.ShiftRightUI(op1, Const(shift));
});
}
else if (op.Size == 2)
{
EmitVectorUnaryOpZx32(context, (op1) =>
{
op1 = context.ZeroExtend32(OperandType.I64, op1);
op1 = context.Add(op1, Const(op1.Type, roundConst));
return context.ConvertI64ToI32(context.ShiftRightUI(op1, Const(shift)));
});
}
else /* if (op.Size == 3) */
{
EmitVectorUnaryOpZx32(context, (op1) => EmitShrImm64(context, op1, signed: false, roundConst, shift));
}
}
else
{
if (op.Size < 2)
{
EmitVectorUnaryOpSx32(context, (op1) =>
{
op1 = context.Add(op1, Const(op1.Type, roundConst));
return context.ShiftRightSI(op1, Const(shift));
});
}
else if (op.Size == 2)
{
EmitVectorUnaryOpSx32(context, (op1) =>
{
op1 = context.SignExtend32(OperandType.I64, op1);
op1 = context.Add(op1, Const(op1.Type, roundConst));
return context.ConvertI64ToI32(context.ShiftRightSI(op1, Const(shift)));
});
}
else /* if (op.Size == 3) */
{
EmitVectorUnaryOpZx32(context, (op1) => EmitShrImm64(context, op1, signed: true, roundConst, shift));
}
}
}
public static void Vshl(ArmEmitterContext context)
{
OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;
EmitVectorUnaryOpZx32(context, (op1) => context.ShiftLeft(op1, Const(op.Shift)));
}
public static void Vshl_I(ArmEmitterContext context)
{
OpCode32SimdReg op = (OpCode32SimdReg)context.CurrOp;
if (op.U)
{
EmitVectorBinaryOpZx32(context, (op1, op2) => EmitShlRegOp(context, op2, op1, op.Size, true));
}
else
{
EmitVectorBinaryOpSx32(context, (op1, op2) => EmitShlRegOp(context, op2, op1, op.Size, false));
}
}
public static void Vshr(ArmEmitterContext context)
{
OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;
int shift = GetImmShr(op);
int maxShift = (8 << op.Size) - 1;
if (op.U)
{
EmitVectorUnaryOpZx32(context, (op1) => (shift > maxShift) ? Const(op1.Type, 0) : context.ShiftRightUI(op1, Const(shift)));
}
else
{
EmitVectorUnaryOpSx32(context, (op1) => context.ShiftRightSI(op1, Const(Math.Min(maxShift, shift))));
}
}
public static void Vshrn(ArmEmitterContext context)
{
OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;
int shift = GetImmShr(op);
EmitVectorUnaryNarrowOp32(context, (op1) => context.ShiftRightUI(op1, Const(shift)));
}
private static Operand EmitShlRegOp(ArmEmitterContext context, Operand op, Operand shiftLsB, int size, bool unsigned)
{
if (shiftLsB.Type == OperandType.I64)
{
shiftLsB = context.ConvertI64ToI32(shiftLsB);
}
shiftLsB = context.SignExtend8(OperandType.I32, shiftLsB);
Debug.Assert((uint)size < 4u);
Operand negShiftLsB = context.Negate(shiftLsB);
Operand isPositive = context.ICompareGreaterOrEqual(shiftLsB, Const(0));
Operand shl = context.ShiftLeft(op, shiftLsB);
Operand shr = unsigned ? context.ShiftRightUI(op, negShiftLsB) : context.ShiftRightSI(op, negShiftLsB);
Operand res = context.ConditionalSelect(isPositive, shl, shr);
if (unsigned)
{
Operand isOutOfRange = context.BitwiseOr(
context.ICompareGreaterOrEqual(shiftLsB, Const(8 << size)),
context.ICompareGreaterOrEqual(negShiftLsB, Const(8 << size)));
return context.ConditionalSelect(isOutOfRange, Const(op.Type, 0), res);
}
else
{
Operand isOutOfRange0 = context.ICompareGreaterOrEqual(shiftLsB, Const(8 << size));
Operand isOutOfRangeN = context.ICompareGreaterOrEqual(negShiftLsB, Const(8 << size));
// Also zero if shift is too negative, but value was positive.
isOutOfRange0 = context.BitwiseOr(isOutOfRange0, context.BitwiseAnd(isOutOfRangeN, context.ICompareGreaterOrEqual(op, Const(op.Type, 0))));
Operand min = (op.Type == OperandType.I64) ? Const(-1L) : Const(-1);
return context.ConditionalSelect(isOutOfRange0, Const(op.Type, 0), context.ConditionalSelect(isOutOfRangeN, min, res));
}
}
[Flags]
private enum ShrImmSaturatingNarrowFlags
{
Scalar = 1 << 0,
SignedSrc = 1 << 1,
SignedDst = 1 << 2,
Round = 1 << 3,
ScalarSxSx = Scalar | SignedSrc | SignedDst,
ScalarSxZx = Scalar | SignedSrc,
ScalarZxZx = Scalar,
VectorSxSx = SignedSrc | SignedDst,
VectorSxZx = SignedSrc,
VectorZxZx = 0
}
private static void EmitRoundShrImmSaturatingNarrowOp(ArmEmitterContext context, ShrImmSaturatingNarrowFlags flags)
{
EmitShrImmSaturatingNarrowOp(context, ShrImmSaturatingNarrowFlags.Round | flags);
}
private static void EmitShrImmSaturatingNarrowOp(ArmEmitterContext context, ShrImmSaturatingNarrowFlags flags)
{
OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;
bool scalar = (flags & ShrImmSaturatingNarrowFlags.Scalar) != 0;
bool signedSrc = (flags & ShrImmSaturatingNarrowFlags.SignedSrc) != 0;
bool signedDst = (flags & ShrImmSaturatingNarrowFlags.SignedDst) != 0;
bool round = (flags & ShrImmSaturatingNarrowFlags.Round) != 0;
if (scalar)
{
// TODO: Support scalar operation.
throw new NotImplementedException();
}
int shift = GetImmShr(op);
long roundConst = 1L << (shift - 1);
EmitVectorUnaryNarrowOp32(context, (op1) =>
{
if (op.Size <= 1 || !round)
{
if (round)
{
op1 = context.Add(op1, Const(op1.Type, roundConst));
}
op1 = signedSrc ? context.ShiftRightSI(op1, Const(shift)) : context.ShiftRightUI(op1, Const(shift));
}
else /* if (op.Size == 2 && round) */
{
op1 = EmitShrImm64(context, op1, signedSrc, roundConst, shift); // shift <= 32
}
return EmitSatQ(context, op1, 8 << op.Size, signedDst);
}, signedSrc);
}
private static int GetImmShr(OpCode32SimdShImm op)
{
return (8 << op.Size) - op.Shift; // Shr amount is flipped.
}
// dst64 = (Int(src64, signed) + roundConst) >> shift;
private static Operand EmitShrImm64(
ArmEmitterContext context,
Operand value,
bool signed,
long roundConst,
int shift)
{
MethodInfo info = signed
? typeof(SoftFallback).GetMethod(nameof(SoftFallback.SignedShrImm64))
: typeof(SoftFallback).GetMethod(nameof(SoftFallback.UnsignedShrImm64));
return context.Call(info, value, Const(roundConst), Const(shift));
}
private static Operand EmitSatQ(ArmEmitterContext context, Operand value, int eSize, bool signed)
{
Debug.Assert(eSize <= 32);
long intMin = signed ? -(1L << (eSize - 1)) : 0;
long intMax = signed ? (1L << (eSize - 1)) - 1 : (1L << eSize) - 1;
Operand gt = context.ICompareGreater(value, Const(value.Type, intMax));
Operand lt = context.ICompareLess(value, Const(value.Type, intMin));
value = context.ConditionalSelect(gt, Const(value.Type, intMax), value);
value = context.ConditionalSelect(lt, Const(value.Type, intMin), value);
Operand lblNoSat = Label();
context.BranchIfFalse(lblNoSat, context.BitwiseOr(gt, lt));
// TODO: Set QC (to 1) on FPSCR here.
context.MarkLabel(lblNoSat);
return value;
}
}
}