Implement fast DMA texture to texture copy (#7299)

* Implement fast DMA texture to texture copy

* PR feedback

Directory.Packages.props

@@ -39,9 +39,9 @@
     <PackageVersion Include="securifybv.ShellLink" Version="0.1.0" />
     <PackageVersion Include="shaderc.net" Version="0.1.0" />
     <PackageVersion Include="SharpZipLib" Version="1.4.2" />
-    <PackageVersion Include="Silk.NET.Vulkan" Version="2.16.0" />
+    <PackageVersion Include="Silk.NET.Vulkan" Version="2.21.0" />
-    <PackageVersion Include="Silk.NET.Vulkan.Extensions.EXT" Version="2.16.0" />
+    <PackageVersion Include="Silk.NET.Vulkan.Extensions.EXT" Version="2.21.0" />
-    <PackageVersion Include="Silk.NET.Vulkan.Extensions.KHR" Version="2.16.0" />
+    <PackageVersion Include="Silk.NET.Vulkan.Extensions.KHR" Version="2.21.0" />
     <PackageVersion Include="SkiaSharp" Version="2.88.7" />
     <PackageVersion Include="SkiaSharp.NativeAssets.Linux" Version="2.88.7" />
     <PackageVersion Include="SPB" Version="0.0.4-build32" />

src/ARMeilleure/Decoders/OpCodeTable.cs

@@ -746,6 +746,7 @@ namespace ARMeilleure.Decoders
             SetA32("<<<<01101000xxxxxxxxxxxxxx01xxxx", InstName.Pkh,     InstEmit32.Pkh,     OpCode32AluRsImm.Create);
             SetA32("11110101xx01xxxx1111xxxxxxxxxxxx", InstName.Pld,     InstEmit32.Nop,     OpCode32.Create);
             SetA32("11110111xx01xxxx1111xxxxxxx0xxxx", InstName.Pld,     InstEmit32.Nop,     OpCode32.Create);
+            SetA32("<<<<01100010xxxxxxxx11110001xxxx", InstName.Qadd16,  InstEmit32.Qadd16,  OpCode32AluReg.Create);
             SetA32("<<<<011011111111xxxx11110011xxxx", InstName.Rbit,    InstEmit32.Rbit,    OpCode32AluReg.Create);
             SetA32("<<<<011010111111xxxx11110011xxxx", InstName.Rev,     InstEmit32.Rev,     OpCode32AluReg.Create);
             SetA32("<<<<011010111111xxxx11111011xxxx", InstName.Rev16,   InstEmit32.Rev16,   OpCode32AluReg.Create);
@@ -1034,6 +1035,7 @@ namespace ARMeilleure.Decoders
             SetAsimd("111100101x>>>xxxxxxx0101>xx1xxxx", InstName.Vshl,        InstEmit32.Vshl,         OpCode32SimdShImm.Create,       OpCode32SimdShImm.CreateT32);
             SetAsimd("1111001x0xxxxxxxxxxx0100xxx0xxxx", InstName.Vshl,        InstEmit32.Vshl_I,       OpCode32SimdReg.Create,         OpCode32SimdReg.CreateT32);
             SetAsimd("1111001x1x>>>xxxxxxx101000x1xxxx", InstName.Vshll,       InstEmit32.Vshll,        OpCode32SimdShImmLong.Create,   OpCode32SimdShImmLong.CreateT32); // A1 encoding.
+            SetAsimd("111100111x11<<10xxxx001100x0xxxx", InstName.Vshll,       InstEmit32.Vshll2,       OpCode32SimdMovn.Create,        OpCode32SimdMovn.CreateT32); // A2 encoding.
             SetAsimd("1111001x1x>>>xxxxxxx0000>xx1xxxx", InstName.Vshr,        InstEmit32.Vshr,         OpCode32SimdShImm.Create,       OpCode32SimdShImm.CreateT32);
             SetAsimd("111100101x>>>xxxxxxx100000x1xxx0", InstName.Vshrn,       InstEmit32.Vshrn,        OpCode32SimdShImmNarrow.Create, OpCode32SimdShImmNarrow.CreateT32);
             SetAsimd("111100111x>>>xxxxxxx0101>xx1xxxx", InstName.Vsli,        InstEmit32.Vsli_I,       OpCode32SimdShImm.Create,       OpCode32SimdShImm.CreateT32);

src/ARMeilleure/Instructions/InstEmitAlu32.cs

@@ -292,6 +292,16 @@ namespace ARMeilleure.Instructions
             EmitAluStore(context, res);
         }
 
+        public static void Qadd16(ArmEmitterContext context)
+        {
+            OpCode32AluReg op = (OpCode32AluReg)context.CurrOp;
+
+            SetIntA32(context, op.Rd, EmitSigned16BitPair(context, GetIntA32(context, op.Rn), GetIntA32(context, op.Rm), (d, n, m) =>
+            {
+                EmitSaturateRange(context, d, context.Add(n, m), 16, unsigned: false, setQ: false);
+            }));
+        }
+
         public static void Rbit(ArmEmitterContext context)
         {
             Operand m = GetAluM(context);
@@ -976,6 +986,94 @@ namespace ARMeilleure.Instructions
             }
         }
 
+        private static void EmitSaturateRange(ArmEmitterContext context, Operand result, Operand value, uint saturateTo, bool unsigned, bool setQ = true)
+        {
+            Debug.Assert(saturateTo <= 32);
+            Debug.Assert(!unsigned || saturateTo < 32);
+
+            if (!unsigned && saturateTo == 32)
+            {
+                // No saturation possible for this case.
+
+                context.Copy(result, value);
+
+                return;
+            }
+            else if (saturateTo == 0)
+            {
+                // Result is always zero if we saturate 0 bits.
+
+                context.Copy(result, Const(0));
+
+                return;
+            }
+
+            Operand satValue;
+
+            if (unsigned)
+            {
+                // Negative values always saturate (to zero).
+                // So we must always ignore the sign bit when masking, so that the truncated value will differ from the original one.
+
+                satValue = context.BitwiseAnd(value, Const((int)(uint.MaxValue >> (32 - (int)saturateTo))));
+            }
+            else
+            {
+                satValue = context.ShiftLeft(value, Const(32 - (int)saturateTo));
+                satValue = context.ShiftRightSI(satValue, Const(32 - (int)saturateTo));
+            }
+
+            // If the result is 0, the values are equal and we don't need saturation.
+            Operand lblNoSat = Label();
+            context.BranchIfFalse(lblNoSat, context.Subtract(value, satValue));
+
+            // Saturate and set Q flag.
+            if (unsigned)
+            {
+                if (saturateTo == 31)
+                {
+                    // Only saturation case possible when going from 32 bits signed to 32 or 31 bits unsigned
+                    // is when the signed input is negative, as all positive values are representable on a 31 bits range.
+
+                    satValue = Const(0);
+                }
+                else
+                {
+                    satValue = context.ShiftRightSI(value, Const(31));
+                    satValue = context.BitwiseNot(satValue);
+                    satValue = context.ShiftRightUI(satValue, Const(32 - (int)saturateTo));
+                }
+            }
+            else
+            {
+                if (saturateTo == 1)
+                {
+                    satValue = context.ShiftRightSI(value, Const(31));
+                }
+                else
+                {
+                    satValue = Const(uint.MaxValue >> (33 - (int)saturateTo));
+                    satValue = context.BitwiseExclusiveOr(satValue, context.ShiftRightSI(value, Const(31)));
+                }
+            }
+
+            if (setQ)
+            {
+                SetFlag(context, PState.QFlag, Const(1));
+            }
+
+            context.Copy(result, satValue);
+
+            Operand lblExit = Label();
+            context.Branch(lblExit);
+
+            context.MarkLabel(lblNoSat);
+
+            context.Copy(result, value);
+
+            context.MarkLabel(lblExit);
+        }
+
         private static void EmitSaturateUqadd(ArmEmitterContext context, Operand result, Operand value, uint saturateTo)
         {
             Debug.Assert(saturateTo <= 32);
@@ -1053,6 +1151,21 @@ namespace ARMeilleure.Instructions
             context.MarkLabel(lblExit);
         }
 
+        private static Operand EmitSigned16BitPair(ArmEmitterContext context, Operand rn, Operand rm, Action<Operand, Operand, Operand> elementAction)
+        {
+            Operand tempD = context.AllocateLocal(OperandType.I32);
+
+            Operand tempN = context.SignExtend16(OperandType.I32, rn);
+            Operand tempM = context.SignExtend16(OperandType.I32, rm);
+            elementAction(tempD, tempN, tempM);
+            Operand tempD2 = context.ZeroExtend16(OperandType.I32, tempD);
+
+            tempN = context.ShiftRightSI(rn, Const(16));
+            tempM = context.ShiftRightSI(rm, Const(16));
+            elementAction(tempD, tempN, tempM);
+            return context.BitwiseOr(tempD2, context.ShiftLeft(tempD, Const(16)));
+        }
+
         private static Operand EmitUnsigned16BitPair(ArmEmitterContext context, Operand rn, Operand rm, Action<Operand, Operand, Operand> elementAction)
         {
             Operand tempD = context.AllocateLocal(OperandType.I32);

src/ARMeilleure/Instructions/InstEmitSimdShift32.cs

@@ -106,6 +106,38 @@ namespace ARMeilleure.Instructions
             context.Copy(GetVecA32(op.Qd), res);
         }
 
+        public static void Vshll2(ArmEmitterContext context)
+        {
+            OpCode32Simd op = (OpCode32Simd)context.CurrOp;
+
+            Operand res = context.VectorZero();
+
+            int elems = op.GetBytesCount() >> op.Size;
+
+            for (int index = 0; index < elems; index++)
+            {
+                Operand me = EmitVectorExtract32(context, op.Qm, op.Im + index, op.Size, !op.U);
+
+                if (op.Size == 2)
+                {
+                    if (op.U)
+                    {
+                        me = context.ZeroExtend32(OperandType.I64, me);
+                    }
+                    else
+                    {
+                        me = context.SignExtend32(OperandType.I64, me);
+                    }
+                }
+
+                me = context.ShiftLeft(me, Const(8 << op.Size));
+
+                res = EmitVectorInsert(context, res, me, index, op.Size + 1);
+            }
+
+            context.Copy(GetVecA32(op.Qd), res);
+        }
+
         public static void Vshr(ArmEmitterContext context)
         {
             OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;

src/ARMeilleure/Instructions/InstName.cs

@@ -527,6 +527,7 @@ namespace ARMeilleure.Instructions
         Pld,
         Pop,
         Push,
+        Qadd16,
         Rev,
         Revsh,
         Rsb,

src/ARMeilleure/Translation/DelegateInfo.cs

@@ -1,5 +1,4 @@
 using System;
-using System.Runtime.InteropServices;
 
 namespace ARMeilleure.Translation
 {
@@ -11,11 +10,10 @@ namespace ARMeilleure.Translation
 
         public IntPtr FuncPtr { get; }
 
-        public DelegateInfo(Delegate dlg)
+        public DelegateInfo(Delegate dlg, IntPtr funcPtr)
         {
             _dlg = dlg;
-
+            FuncPtr = funcPtr;
-            FuncPtr = Marshal.GetFunctionPointerForDelegate<Delegate>(dlg);
         }
     }
 }

src/ARMeilleure/Translation/Delegates.cs

@@ -3,6 +3,7 @@ using ARMeilleure.State;
 using System;
 using System.Collections.Generic;
 using System.Reflection;
+using System.Runtime.InteropServices;
 
 namespace ARMeilleure.Translation
 {
@@ -64,11 +65,11 @@ namespace ARMeilleure.Translation
             return index;
         }
 
-        private static void SetDelegateInfo(Delegate dlg)
+        private static void SetDelegateInfo(Delegate dlg, IntPtr funcPtr)
         {
             string key = GetKey(dlg.Method);
 
-            _delegates.Add(key, new DelegateInfo(dlg)); // ArgumentException (key).
+            _delegates.Add(key, new DelegateInfo(dlg, funcPtr)); // ArgumentException (key).
         }
 
         private static string GetKey(MethodInfo info)
@@ -82,179 +83,353 @@ namespace ARMeilleure.Translation
         {
             _delegates = new SortedList<string, DelegateInfo>();
 
-            SetDelegateInfo(new MathAbs(Math.Abs));
+            var dlgMathAbs = new MathAbs(Math.Abs);
-            SetDelegateInfo(new MathCeiling(Math.Ceiling));
+            var dlgMathCeiling = new MathCeiling(Math.Ceiling);
-            SetDelegateInfo(new MathFloor(Math.Floor));
+            var dlgMathFloor = new MathFloor(Math.Floor);
-            SetDelegateInfo(new MathRound(Math.Round));
+            var dlgMathRound = new MathRound(Math.Round);
-            SetDelegateInfo(new MathTruncate(Math.Truncate));
+            var dlgMathTruncate = new MathTruncate(Math.Truncate);
 
-            SetDelegateInfo(new MathFAbs(MathF.Abs));
+            var dlgMathFAbs = new MathFAbs(MathF.Abs);
-            SetDelegateInfo(new MathFCeiling(MathF.Ceiling));
+            var dlgMathFCeiling = new MathFCeiling(MathF.Ceiling);
-            SetDelegateInfo(new MathFFloor(MathF.Floor));
+            var dlgMathFFloor = new MathFFloor(MathF.Floor);
-            SetDelegateInfo(new MathFRound(MathF.Round));
+            var dlgMathFRound = new MathFRound(MathF.Round);
-            SetDelegateInfo(new MathFTruncate(MathF.Truncate));
+            var dlgMathFTruncate = new MathFTruncate(MathF.Truncate);
 
-            SetDelegateInfo(new NativeInterfaceBreak(NativeInterface.Break));
+            var dlgNativeInterfaceBreak = new NativeInterfaceBreak(NativeInterface.Break);
-            SetDelegateInfo(new NativeInterfaceCheckSynchronization(NativeInterface.CheckSynchronization));
+            var dlgNativeInterfaceCheckSynchronization = new NativeInterfaceCheckSynchronization(NativeInterface.CheckSynchronization);
-            SetDelegateInfo(new NativeInterfaceEnqueueForRejit(NativeInterface.EnqueueForRejit));
+            var dlgNativeInterfaceEnqueueForRejit = new NativeInterfaceEnqueueForRejit(NativeInterface.EnqueueForRejit);
-            SetDelegateInfo(new NativeInterfaceGetCntfrqEl0(NativeInterface.GetCntfrqEl0));
+            var dlgNativeInterfaceGetCntfrqEl0 = new NativeInterfaceGetCntfrqEl0(NativeInterface.GetCntfrqEl0);
-            SetDelegateInfo(new NativeInterfaceGetCntpctEl0(NativeInterface.GetCntpctEl0));
+            var dlgNativeInterfaceGetCntpctEl0 = new NativeInterfaceGetCntpctEl0(NativeInterface.GetCntpctEl0);
-            SetDelegateInfo(new NativeInterfaceGetCntvctEl0(NativeInterface.GetCntvctEl0));
+            var dlgNativeInterfaceGetCntvctEl0 = new NativeInterfaceGetCntvctEl0(NativeInterface.GetCntvctEl0);
-            SetDelegateInfo(new NativeInterfaceGetCtrEl0(NativeInterface.GetCtrEl0));
+            var dlgNativeInterfaceGetCtrEl0 = new NativeInterfaceGetCtrEl0(NativeInterface.GetCtrEl0);
-            SetDelegateInfo(new NativeInterfaceGetDczidEl0(NativeInterface.GetDczidEl0));
+            var dlgNativeInterfaceGetDczidEl0 = new NativeInterfaceGetDczidEl0(NativeInterface.GetDczidEl0);
-            SetDelegateInfo(new NativeInterfaceGetFunctionAddress(NativeInterface.GetFunctionAddress));
+            var dlgNativeInterfaceGetFunctionAddress = new NativeInterfaceGetFunctionAddress(NativeInterface.GetFunctionAddress);
-            SetDelegateInfo(new NativeInterfaceInvalidateCacheLine(NativeInterface.InvalidateCacheLine));
+            var dlgNativeInterfaceInvalidateCacheLine = new NativeInterfaceInvalidateCacheLine(NativeInterface.InvalidateCacheLine);
-            SetDelegateInfo(new NativeInterfaceReadByte(NativeInterface.ReadByte));
+            var dlgNativeInterfaceReadByte = new NativeInterfaceReadByte(NativeInterface.ReadByte);
-            SetDelegateInfo(new NativeInterfaceReadUInt16(NativeInterface.ReadUInt16));
+            var dlgNativeInterfaceReadUInt16 = new NativeInterfaceReadUInt16(NativeInterface.ReadUInt16);
-            SetDelegateInfo(new NativeInterfaceReadUInt32(NativeInterface.ReadUInt32));
+            var dlgNativeInterfaceReadUInt32 = new NativeInterfaceReadUInt32(NativeInterface.ReadUInt32);
-            SetDelegateInfo(new NativeInterfaceReadUInt64(NativeInterface.ReadUInt64));
+            var dlgNativeInterfaceReadUInt64 = new NativeInterfaceReadUInt64(NativeInterface.ReadUInt64);
-            SetDelegateInfo(new NativeInterfaceReadVector128(NativeInterface.ReadVector128));
+            var dlgNativeInterfaceReadVector128 = new NativeInterfaceReadVector128(NativeInterface.ReadVector128);
-            SetDelegateInfo(new NativeInterfaceSignalMemoryTracking(NativeInterface.SignalMemoryTracking));
+            var dlgNativeInterfaceSignalMemoryTracking = new NativeInterfaceSignalMemoryTracking(NativeInterface.SignalMemoryTracking);
-            SetDelegateInfo(new NativeInterfaceSupervisorCall(NativeInterface.SupervisorCall));
+            var dlgNativeInterfaceSupervisorCall = new NativeInterfaceSupervisorCall(NativeInterface.SupervisorCall);
-            SetDelegateInfo(new NativeInterfaceThrowInvalidMemoryAccess(NativeInterface.ThrowInvalidMemoryAccess));
+            var dlgNativeInterfaceThrowInvalidMemoryAccess = new NativeInterfaceThrowInvalidMemoryAccess(NativeInterface.ThrowInvalidMemoryAccess);
-            SetDelegateInfo(new NativeInterfaceUndefined(NativeInterface.Undefined));
+            var dlgNativeInterfaceUndefined = new NativeInterfaceUndefined(NativeInterface.Undefined);
-            SetDelegateInfo(new NativeInterfaceWriteByte(NativeInterface.WriteByte));
+            var dlgNativeInterfaceWriteByte = new NativeInterfaceWriteByte(NativeInterface.WriteByte);
-            SetDelegateInfo(new NativeInterfaceWriteUInt16(NativeInterface.WriteUInt16));
+            var dlgNativeInterfaceWriteUInt16 = new NativeInterfaceWriteUInt16(NativeInterface.WriteUInt16);
-            SetDelegateInfo(new NativeInterfaceWriteUInt32(NativeInterface.WriteUInt32));
+            var dlgNativeInterfaceWriteUInt32 = new NativeInterfaceWriteUInt32(NativeInterface.WriteUInt32);
-            SetDelegateInfo(new NativeInterfaceWriteUInt64(NativeInterface.WriteUInt64));
+            var dlgNativeInterfaceWriteUInt64 = new NativeInterfaceWriteUInt64(NativeInterface.WriteUInt64);
-            SetDelegateInfo(new NativeInterfaceWriteVector128(NativeInterface.WriteVector128));
+            var dlgNativeInterfaceWriteVector128 = new NativeInterfaceWriteVector128(NativeInterface.WriteVector128);
 
-            SetDelegateInfo(new SoftFallbackCountLeadingSigns(SoftFallback.CountLeadingSigns));
+            var dlgSoftFallbackCountLeadingSigns = new SoftFallbackCountLeadingSigns(SoftFallback.CountLeadingSigns);
-            SetDelegateInfo(new SoftFallbackCountLeadingZeros(SoftFallback.CountLeadingZeros));
+            var dlgSoftFallbackCountLeadingZeros = new SoftFallbackCountLeadingZeros(SoftFallback.CountLeadingZeros);
-            SetDelegateInfo(new SoftFallbackCrc32b(SoftFallback.Crc32b));
+            var dlgSoftFallbackCrc32b = new SoftFallbackCrc32b(SoftFallback.Crc32b);
-            SetDelegateInfo(new SoftFallbackCrc32cb(SoftFallback.Crc32cb));
+            var dlgSoftFallbackCrc32cb = new SoftFallbackCrc32cb(SoftFallback.Crc32cb);
-            SetDelegateInfo(new SoftFallbackCrc32ch(SoftFallback.Crc32ch));
+            var dlgSoftFallbackCrc32ch = new SoftFallbackCrc32ch(SoftFallback.Crc32ch);
-            SetDelegateInfo(new SoftFallbackCrc32cw(SoftFallback.Crc32cw));
+            var dlgSoftFallbackCrc32cw = new SoftFallbackCrc32cw(SoftFallback.Crc32cw);
-            SetDelegateInfo(new SoftFallbackCrc32cx(SoftFallback.Crc32cx));
+            var dlgSoftFallbackCrc32cx = new SoftFallbackCrc32cx(SoftFallback.Crc32cx);
-            SetDelegateInfo(new SoftFallbackCrc32h(SoftFallback.Crc32h));
+            var dlgSoftFallbackCrc32h = new SoftFallbackCrc32h(SoftFallback.Crc32h);
-            SetDelegateInfo(new SoftFallbackCrc32w(SoftFallback.Crc32w));
+            var dlgSoftFallbackCrc32w = new SoftFallbackCrc32w(SoftFallback.Crc32w);
-            SetDelegateInfo(new SoftFallbackCrc32x(SoftFallback.Crc32x));
+            var dlgSoftFallbackCrc32x = new SoftFallbackCrc32x(SoftFallback.Crc32x);
-            SetDelegateInfo(new SoftFallbackDecrypt(SoftFallback.Decrypt));
+            var dlgSoftFallbackDecrypt = new SoftFallbackDecrypt(SoftFallback.Decrypt);
-            SetDelegateInfo(new SoftFallbackEncrypt(SoftFallback.Encrypt));
+            var dlgSoftFallbackEncrypt = new SoftFallbackEncrypt(SoftFallback.Encrypt);
-            SetDelegateInfo(new SoftFallbackFixedRotate(SoftFallback.FixedRotate));
+            var dlgSoftFallbackFixedRotate = new SoftFallbackFixedRotate(SoftFallback.FixedRotate);
-            SetDelegateInfo(new SoftFallbackHashChoose(SoftFallback.HashChoose));
+            var dlgSoftFallbackHashChoose = new SoftFallbackHashChoose(SoftFallback.HashChoose);
-            SetDelegateInfo(new SoftFallbackHashLower(SoftFallback.HashLower));
+            var dlgSoftFallbackHashLower = new SoftFallbackHashLower(SoftFallback.HashLower);
-            SetDelegateInfo(new SoftFallbackHashMajority(SoftFallback.HashMajority));
+            var dlgSoftFallbackHashMajority = new SoftFallbackHashMajority(SoftFallback.HashMajority);
-            SetDelegateInfo(new SoftFallbackHashParity(SoftFallback.HashParity));
+            var dlgSoftFallbackHashParity = new SoftFallbackHashParity(SoftFallback.HashParity);
-            SetDelegateInfo(new SoftFallbackHashUpper(SoftFallback.HashUpper));
+            var dlgSoftFallbackHashUpper = new SoftFallbackHashUpper(SoftFallback.HashUpper);
-            SetDelegateInfo(new SoftFallbackInverseMixColumns(SoftFallback.InverseMixColumns));
+            var dlgSoftFallbackInverseMixColumns = new SoftFallbackInverseMixColumns(SoftFallback.InverseMixColumns);
-            SetDelegateInfo(new SoftFallbackMixColumns(SoftFallback.MixColumns));
+            var dlgSoftFallbackMixColumns = new SoftFallbackMixColumns(SoftFallback.MixColumns);
-            SetDelegateInfo(new SoftFallbackPolynomialMult64_128(SoftFallback.PolynomialMult64_128));
+            var dlgSoftFallbackPolynomialMult64_128 = new SoftFallbackPolynomialMult64_128(SoftFallback.PolynomialMult64_128);
-            SetDelegateInfo(new SoftFallbackSatF32ToS32(SoftFallback.SatF32ToS32));
+            var dlgSoftFallbackSatF32ToS32 = new SoftFallbackSatF32ToS32(SoftFallback.SatF32ToS32);
-            SetDelegateInfo(new SoftFallbackSatF32ToS64(SoftFallback.SatF32ToS64));
+            var dlgSoftFallbackSatF32ToS64 = new SoftFallbackSatF32ToS64(SoftFallback.SatF32ToS64);
-            SetDelegateInfo(new SoftFallbackSatF32ToU32(SoftFallback.SatF32ToU32));
+            var dlgSoftFallbackSatF32ToU32 = new SoftFallbackSatF32ToU32(SoftFallback.SatF32ToU32);
-            SetDelegateInfo(new SoftFallbackSatF32ToU64(SoftFallback.SatF32ToU64));
+            var dlgSoftFallbackSatF32ToU64 = new SoftFallbackSatF32ToU64(SoftFallback.SatF32ToU64);
-            SetDelegateInfo(new SoftFallbackSatF64ToS32(SoftFallback.SatF64ToS32));
+            var dlgSoftFallbackSatF64ToS32 = new SoftFallbackSatF64ToS32(SoftFallback.SatF64ToS32);
-            SetDelegateInfo(new SoftFallbackSatF64ToS64(SoftFallback.SatF64ToS64));
+            var dlgSoftFallbackSatF64ToS64 = new SoftFallbackSatF64ToS64(SoftFallback.SatF64ToS64);
-            SetDelegateInfo(new SoftFallbackSatF64ToU32(SoftFallback.SatF64ToU32));
+            var dlgSoftFallbackSatF64ToU32 = new SoftFallbackSatF64ToU32(SoftFallback.SatF64ToU32);
-            SetDelegateInfo(new SoftFallbackSatF64ToU64(SoftFallback.SatF64ToU64));
+            var dlgSoftFallbackSatF64ToU64 = new SoftFallbackSatF64ToU64(SoftFallback.SatF64ToU64);
-            SetDelegateInfo(new SoftFallbackSha1SchedulePart1(SoftFallback.Sha1SchedulePart1));
+            var dlgSoftFallbackSha1SchedulePart1 = new SoftFallbackSha1SchedulePart1(SoftFallback.Sha1SchedulePart1);
-            SetDelegateInfo(new SoftFallbackSha1SchedulePart2(SoftFallback.Sha1SchedulePart2));
+            var dlgSoftFallbackSha1SchedulePart2 = new SoftFallbackSha1SchedulePart2(SoftFallback.Sha1SchedulePart2);
-            SetDelegateInfo(new SoftFallbackSha256SchedulePart1(SoftFallback.Sha256SchedulePart1));
+            var dlgSoftFallbackSha256SchedulePart1 = new SoftFallbackSha256SchedulePart1(SoftFallback.Sha256SchedulePart1);
-            SetDelegateInfo(new SoftFallbackSha256SchedulePart2(SoftFallback.Sha256SchedulePart2));
+            var dlgSoftFallbackSha256SchedulePart2 = new SoftFallbackSha256SchedulePart2(SoftFallback.Sha256SchedulePart2);
-            SetDelegateInfo(new SoftFallbackSignedShrImm64(SoftFallback.SignedShrImm64));
+            var dlgSoftFallbackSignedShrImm64 = new SoftFallbackSignedShrImm64(SoftFallback.SignedShrImm64);
-            SetDelegateInfo(new SoftFallbackTbl1(SoftFallback.Tbl1));
+            var dlgSoftFallbackTbl1 = new SoftFallbackTbl1(SoftFallback.Tbl1);
-            SetDelegateInfo(new SoftFallbackTbl2(SoftFallback.Tbl2));
+            var dlgSoftFallbackTbl2 = new SoftFallbackTbl2(SoftFallback.Tbl2);
-            SetDelegateInfo(new SoftFallbackTbl3(SoftFallback.Tbl3));
+            var dlgSoftFallbackTbl3 = new SoftFallbackTbl3(SoftFallback.Tbl3);
-            SetDelegateInfo(new SoftFallbackTbl4(SoftFallback.Tbl4));
+            var dlgSoftFallbackTbl4 = new SoftFallbackTbl4(SoftFallback.Tbl4);
-            SetDelegateInfo(new SoftFallbackTbx1(SoftFallback.Tbx1));
+            var dlgSoftFallbackTbx1 = new SoftFallbackTbx1(SoftFallback.Tbx1);
-            SetDelegateInfo(new SoftFallbackTbx2(SoftFallback.Tbx2));
+            var dlgSoftFallbackTbx2 = new SoftFallbackTbx2(SoftFallback.Tbx2);
-            SetDelegateInfo(new SoftFallbackTbx3(SoftFallback.Tbx3));
+            var dlgSoftFallbackTbx3 = new SoftFallbackTbx3(SoftFallback.Tbx3);
-            SetDelegateInfo(new SoftFallbackTbx4(SoftFallback.Tbx4));
+            var dlgSoftFallbackTbx4 = new SoftFallbackTbx4(SoftFallback.Tbx4);
-            SetDelegateInfo(new SoftFallbackUnsignedShrImm64(SoftFallback.UnsignedShrImm64));
+            var dlgSoftFallbackUnsignedShrImm64 = new SoftFallbackUnsignedShrImm64(SoftFallback.UnsignedShrImm64);
 
-            SetDelegateInfo(new SoftFloat16_32FPConvert(SoftFloat16_32.FPConvert));
+            var dlgSoftFloat16_32FPConvert = new SoftFloat16_32FPConvert(SoftFloat16_32.FPConvert);
-            SetDelegateInfo(new SoftFloat16_64FPConvert(SoftFloat16_64.FPConvert));
+            var dlgSoftFloat16_64FPConvert = new SoftFloat16_64FPConvert(SoftFloat16_64.FPConvert);
 
-            SetDelegateInfo(new SoftFloat32FPAdd(SoftFloat32.FPAdd));
+            var dlgSoftFloat32FPAdd = new SoftFloat32FPAdd(SoftFloat32.FPAdd);
-            SetDelegateInfo(new SoftFloat32FPAddFpscr(SoftFloat32.FPAddFpscr)); // A32 only.
+            var dlgSoftFloat32FPAddFpscr = new SoftFloat32FPAddFpscr(SoftFloat32.FPAddFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPCompare(SoftFloat32.FPCompare));
+            var dlgSoftFloat32FPCompare = new SoftFloat32FPCompare(SoftFloat32.FPCompare);
-            SetDelegateInfo(new SoftFloat32FPCompareEQ(SoftFloat32.FPCompareEQ));
+            var dlgSoftFloat32FPCompareEQ = new SoftFloat32FPCompareEQ(SoftFloat32.FPCompareEQ);
-            SetDelegateInfo(new SoftFloat32FPCompareEQFpscr(SoftFloat32.FPCompareEQFpscr)); // A32 only.
+            var dlgSoftFloat32FPCompareEQFpscr = new SoftFloat32FPCompareEQFpscr(SoftFloat32.FPCompareEQFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPCompareGE(SoftFloat32.FPCompareGE));
+            var dlgSoftFloat32FPCompareGE = new SoftFloat32FPCompareGE(SoftFloat32.FPCompareGE);
-            SetDelegateInfo(new SoftFloat32FPCompareGEFpscr(SoftFloat32.FPCompareGEFpscr)); // A32 only.
+            var dlgSoftFloat32FPCompareGEFpscr = new SoftFloat32FPCompareGEFpscr(SoftFloat32.FPCompareGEFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPCompareGT(SoftFloat32.FPCompareGT));
+            var dlgSoftFloat32FPCompareGT = new SoftFloat32FPCompareGT(SoftFloat32.FPCompareGT);
-            SetDelegateInfo(new SoftFloat32FPCompareGTFpscr(SoftFloat32.FPCompareGTFpscr)); // A32 only.
+            var dlgSoftFloat32FPCompareGTFpscr = new SoftFloat32FPCompareGTFpscr(SoftFloat32.FPCompareGTFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPCompareLE(SoftFloat32.FPCompareLE));
+            var dlgSoftFloat32FPCompareLE = new SoftFloat32FPCompareLE(SoftFloat32.FPCompareLE);
-            SetDelegateInfo(new SoftFloat32FPCompareLEFpscr(SoftFloat32.FPCompareLEFpscr)); // A32 only.
+            var dlgSoftFloat32FPCompareLEFpscr = new SoftFloat32FPCompareLEFpscr(SoftFloat32.FPCompareLEFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPCompareLT(SoftFloat32.FPCompareLT));
+            var dlgSoftFloat32FPCompareLT = new SoftFloat32FPCompareLT(SoftFloat32.FPCompareLT);
-            SetDelegateInfo(new SoftFloat32FPCompareLTFpscr(SoftFloat32.FPCompareLTFpscr)); // A32 only.
+            var dlgSoftFloat32FPCompareLTFpscr = new SoftFloat32FPCompareLTFpscr(SoftFloat32.FPCompareLTFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPDiv(SoftFloat32.FPDiv));
+            var dlgSoftFloat32FPDiv = new SoftFloat32FPDiv(SoftFloat32.FPDiv);
-            SetDelegateInfo(new SoftFloat32FPMax(SoftFloat32.FPMax));
+            var dlgSoftFloat32FPMax = new SoftFloat32FPMax(SoftFloat32.FPMax);
-            SetDelegateInfo(new SoftFloat32FPMaxFpscr(SoftFloat32.FPMaxFpscr)); // A32 only.
+            var dlgSoftFloat32FPMaxFpscr = new SoftFloat32FPMaxFpscr(SoftFloat32.FPMaxFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPMaxNum(SoftFloat32.FPMaxNum));
+            var dlgSoftFloat32FPMaxNum = new SoftFloat32FPMaxNum(SoftFloat32.FPMaxNum);
-            SetDelegateInfo(new SoftFloat32FPMaxNumFpscr(SoftFloat32.FPMaxNumFpscr)); // A32 only.
+            var dlgSoftFloat32FPMaxNumFpscr = new SoftFloat32FPMaxNumFpscr(SoftFloat32.FPMaxNumFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPMin(SoftFloat32.FPMin));
+            var dlgSoftFloat32FPMin = new SoftFloat32FPMin(SoftFloat32.FPMin);
-            SetDelegateInfo(new SoftFloat32FPMinFpscr(SoftFloat32.FPMinFpscr)); // A32 only.
+            var dlgSoftFloat32FPMinFpscr = new SoftFloat32FPMinFpscr(SoftFloat32.FPMinFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPMinNum(SoftFloat32.FPMinNum));
+            var dlgSoftFloat32FPMinNum = new SoftFloat32FPMinNum(SoftFloat32.FPMinNum);
-            SetDelegateInfo(new SoftFloat32FPMinNumFpscr(SoftFloat32.FPMinNumFpscr)); // A32 only.
+            var dlgSoftFloat32FPMinNumFpscr = new SoftFloat32FPMinNumFpscr(SoftFloat32.FPMinNumFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPMul(SoftFloat32.FPMul));
+            var dlgSoftFloat32FPMul = new SoftFloat32FPMul(SoftFloat32.FPMul);
-            SetDelegateInfo(new SoftFloat32FPMulFpscr(SoftFloat32.FPMulFpscr)); // A32 only.
+            var dlgSoftFloat32FPMulFpscr = new SoftFloat32FPMulFpscr(SoftFloat32.FPMulFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPMulAdd(SoftFloat32.FPMulAdd));
+            var dlgSoftFloat32FPMulAdd = new SoftFloat32FPMulAdd(SoftFloat32.FPMulAdd);
-            SetDelegateInfo(new SoftFloat32FPMulAddFpscr(SoftFloat32.FPMulAddFpscr)); // A32 only.
+            var dlgSoftFloat32FPMulAddFpscr = new SoftFloat32FPMulAddFpscr(SoftFloat32.FPMulAddFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPMulSub(SoftFloat32.FPMulSub));
+            var dlgSoftFloat32FPMulSub = new SoftFloat32FPMulSub(SoftFloat32.FPMulSub);
-            SetDelegateInfo(new SoftFloat32FPMulSubFpscr(SoftFloat32.FPMulSubFpscr)); // A32 only.
+            var dlgSoftFloat32FPMulSubFpscr = new SoftFloat32FPMulSubFpscr(SoftFloat32.FPMulSubFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPMulX(SoftFloat32.FPMulX));
+            var dlgSoftFloat32FPMulX = new SoftFloat32FPMulX(SoftFloat32.FPMulX);
-            SetDelegateInfo(new SoftFloat32FPNegMulAdd(SoftFloat32.FPNegMulAdd));
+            var dlgSoftFloat32FPNegMulAdd = new SoftFloat32FPNegMulAdd(SoftFloat32.FPNegMulAdd);
-            SetDelegateInfo(new SoftFloat32FPNegMulSub(SoftFloat32.FPNegMulSub));
+            var dlgSoftFloat32FPNegMulSub = new SoftFloat32FPNegMulSub(SoftFloat32.FPNegMulSub);
-            SetDelegateInfo(new SoftFloat32FPRecipEstimate(SoftFloat32.FPRecipEstimate));
+            var dlgSoftFloat32FPRecipEstimate = new SoftFloat32FPRecipEstimate(SoftFloat32.FPRecipEstimate);
-            SetDelegateInfo(new SoftFloat32FPRecipEstimateFpscr(SoftFloat32.FPRecipEstimateFpscr)); // A32 only.
+            var dlgSoftFloat32FPRecipEstimateFpscr = new SoftFloat32FPRecipEstimateFpscr(SoftFloat32.FPRecipEstimateFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPRecipStep(SoftFloat32.FPRecipStep)); // A32 only.
+            var dlgSoftFloat32FPRecipStep = new SoftFloat32FPRecipStep(SoftFloat32.FPRecipStep); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPRecipStepFused(SoftFloat32.FPRecipStepFused));
+            var dlgSoftFloat32FPRecipStepFused = new SoftFloat32FPRecipStepFused(SoftFloat32.FPRecipStepFused);
-            SetDelegateInfo(new SoftFloat32FPRecpX(SoftFloat32.FPRecpX));
+            var dlgSoftFloat32FPRecpX = new SoftFloat32FPRecpX(SoftFloat32.FPRecpX);
-            SetDelegateInfo(new SoftFloat32FPRSqrtEstimate(SoftFloat32.FPRSqrtEstimate));
+            var dlgSoftFloat32FPRSqrtEstimate = new SoftFloat32FPRSqrtEstimate(SoftFloat32.FPRSqrtEstimate);
-            SetDelegateInfo(new SoftFloat32FPRSqrtEstimateFpscr(SoftFloat32.FPRSqrtEstimateFpscr)); // A32 only.
+            var dlgSoftFloat32FPRSqrtEstimateFpscr = new SoftFloat32FPRSqrtEstimateFpscr(SoftFloat32.FPRSqrtEstimateFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPRSqrtStep(SoftFloat32.FPRSqrtStep)); // A32 only.
+            var dlgSoftFloat32FPRSqrtStep = new SoftFloat32FPRSqrtStep(SoftFloat32.FPRSqrtStep); // A32 only.
-            SetDelegateInfo(new SoftFloat32FPRSqrtStepFused(SoftFloat32.FPRSqrtStepFused));
+            var dlgSoftFloat32FPRSqrtStepFused = new SoftFloat32FPRSqrtStepFused(SoftFloat32.FPRSqrtStepFused);
-            SetDelegateInfo(new SoftFloat32FPSqrt(SoftFloat32.FPSqrt));
+            var dlgSoftFloat32FPSqrt = new SoftFloat32FPSqrt(SoftFloat32.FPSqrt);
-            SetDelegateInfo(new SoftFloat32FPSub(SoftFloat32.FPSub));
+            var dlgSoftFloat32FPSub = new SoftFloat32FPSub(SoftFloat32.FPSub);
 
-            SetDelegateInfo(new SoftFloat32_16FPConvert(SoftFloat32_16.FPConvert));
+            var dlgSoftFloat32_16FPConvert = new SoftFloat32_16FPConvert(SoftFloat32_16.FPConvert);
 
-            SetDelegateInfo(new SoftFloat64FPAdd(SoftFloat64.FPAdd));
+            var dlgSoftFloat64FPAdd = new SoftFloat64FPAdd(SoftFloat64.FPAdd);
-            SetDelegateInfo(new SoftFloat64FPAddFpscr(SoftFloat64.FPAddFpscr)); // A32 only.
+            var dlgSoftFloat64FPAddFpscr = new SoftFloat64FPAddFpscr(SoftFloat64.FPAddFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPCompare(SoftFloat64.FPCompare));
+            var dlgSoftFloat64FPCompare = new SoftFloat64FPCompare(SoftFloat64.FPCompare);
-            SetDelegateInfo(new SoftFloat64FPCompareEQ(SoftFloat64.FPCompareEQ));
+            var dlgSoftFloat64FPCompareEQ = new SoftFloat64FPCompareEQ(SoftFloat64.FPCompareEQ);
-            SetDelegateInfo(new SoftFloat64FPCompareEQFpscr(SoftFloat64.FPCompareEQFpscr)); // A32 only.
+            var dlgSoftFloat64FPCompareEQFpscr = new SoftFloat64FPCompareEQFpscr(SoftFloat64.FPCompareEQFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPCompareGE(SoftFloat64.FPCompareGE));
+            var dlgSoftFloat64FPCompareGE = new SoftFloat64FPCompareGE(SoftFloat64.FPCompareGE);
-            SetDelegateInfo(new SoftFloat64FPCompareGEFpscr(SoftFloat64.FPCompareGEFpscr)); // A32 only.
+            var dlgSoftFloat64FPCompareGEFpscr = new SoftFloat64FPCompareGEFpscr(SoftFloat64.FPCompareGEFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPCompareGT(SoftFloat64.FPCompareGT));
+            var dlgSoftFloat64FPCompareGT = new SoftFloat64FPCompareGT(SoftFloat64.FPCompareGT);
-            SetDelegateInfo(new SoftFloat64FPCompareGTFpscr(SoftFloat64.FPCompareGTFpscr)); // A32 only.
+            var dlgSoftFloat64FPCompareGTFpscr = new SoftFloat64FPCompareGTFpscr(SoftFloat64.FPCompareGTFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPCompareLE(SoftFloat64.FPCompareLE));
+            var dlgSoftFloat64FPCompareLE = new SoftFloat64FPCompareLE(SoftFloat64.FPCompareLE);
-            SetDelegateInfo(new SoftFloat64FPCompareLEFpscr(SoftFloat64.FPCompareLEFpscr)); // A32 only.
+            var dlgSoftFloat64FPCompareLEFpscr = new SoftFloat64FPCompareLEFpscr(SoftFloat64.FPCompareLEFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPCompareLT(SoftFloat64.FPCompareLT));
+            var dlgSoftFloat64FPCompareLT = new SoftFloat64FPCompareLT(SoftFloat64.FPCompareLT);
-            SetDelegateInfo(new SoftFloat64FPCompareLTFpscr(SoftFloat64.FPCompareLTFpscr)); // A32 only.
+            var dlgSoftFloat64FPCompareLTFpscr = new SoftFloat64FPCompareLTFpscr(SoftFloat64.FPCompareLTFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPDiv(SoftFloat64.FPDiv));
+            var dlgSoftFloat64FPDiv = new SoftFloat64FPDiv(SoftFloat64.FPDiv);
-            SetDelegateInfo(new SoftFloat64FPMax(SoftFloat64.FPMax));
+            var dlgSoftFloat64FPMax = new SoftFloat64FPMax(SoftFloat64.FPMax);
-            SetDelegateInfo(new SoftFloat64FPMaxFpscr(SoftFloat64.FPMaxFpscr)); // A32 only.
+            var dlgSoftFloat64FPMaxFpscr = new SoftFloat64FPMaxFpscr(SoftFloat64.FPMaxFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPMaxNum(SoftFloat64.FPMaxNum));
+            var dlgSoftFloat64FPMaxNum = new SoftFloat64FPMaxNum(SoftFloat64.FPMaxNum);
-            SetDelegateInfo(new SoftFloat64FPMaxNumFpscr(SoftFloat64.FPMaxNumFpscr)); // A32 only.
+            var dlgSoftFloat64FPMaxNumFpscr = new SoftFloat64FPMaxNumFpscr(SoftFloat64.FPMaxNumFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPMin(SoftFloat64.FPMin));
+            var dlgSoftFloat64FPMin = new SoftFloat64FPMin(SoftFloat64.FPMin);
-            SetDelegateInfo(new SoftFloat64FPMinFpscr(SoftFloat64.FPMinFpscr)); // A32 only.
+            var dlgSoftFloat64FPMinFpscr = new SoftFloat64FPMinFpscr(SoftFloat64.FPMinFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPMinNum(SoftFloat64.FPMinNum));
+            var dlgSoftFloat64FPMinNum = new SoftFloat64FPMinNum(SoftFloat64.FPMinNum);
-            SetDelegateInfo(new SoftFloat64FPMinNumFpscr(SoftFloat64.FPMinNumFpscr)); // A32 only.
+            var dlgSoftFloat64FPMinNumFpscr = new SoftFloat64FPMinNumFpscr(SoftFloat64.FPMinNumFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPMul(SoftFloat64.FPMul));
+            var dlgSoftFloat64FPMul = new SoftFloat64FPMul(SoftFloat64.FPMul);
-            SetDelegateInfo(new SoftFloat64FPMulFpscr(SoftFloat64.FPMulFpscr)); // A32 only.
+            var dlgSoftFloat64FPMulFpscr = new SoftFloat64FPMulFpscr(SoftFloat64.FPMulFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPMulAdd(SoftFloat64.FPMulAdd));
+            var dlgSoftFloat64FPMulAdd = new SoftFloat64FPMulAdd(SoftFloat64.FPMulAdd);
-            SetDelegateInfo(new SoftFloat64FPMulAddFpscr(SoftFloat64.FPMulAddFpscr)); // A32 only.
+            var dlgSoftFloat64FPMulAddFpscr = new SoftFloat64FPMulAddFpscr(SoftFloat64.FPMulAddFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPMulSub(SoftFloat64.FPMulSub));
+            var dlgSoftFloat64FPMulSub = new SoftFloat64FPMulSub(SoftFloat64.FPMulSub);
-            SetDelegateInfo(new SoftFloat64FPMulSubFpscr(SoftFloat64.FPMulSubFpscr)); // A32 only.
+            var dlgSoftFloat64FPMulSubFpscr = new SoftFloat64FPMulSubFpscr(SoftFloat64.FPMulSubFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPMulX(SoftFloat64.FPMulX));
+            var dlgSoftFloat64FPMulX = new SoftFloat64FPMulX(SoftFloat64.FPMulX);
-            SetDelegateInfo(new SoftFloat64FPNegMulAdd(SoftFloat64.FPNegMulAdd));
+            var dlgSoftFloat64FPNegMulAdd = new SoftFloat64FPNegMulAdd(SoftFloat64.FPNegMulAdd);
-            SetDelegateInfo(new SoftFloat64FPNegMulSub(SoftFloat64.FPNegMulSub));
+            var dlgSoftFloat64FPNegMulSub = new SoftFloat64FPNegMulSub(SoftFloat64.FPNegMulSub);
-            SetDelegateInfo(new SoftFloat64FPRecipEstimate(SoftFloat64.FPRecipEstimate));
+            var dlgSoftFloat64FPRecipEstimate = new SoftFloat64FPRecipEstimate(SoftFloat64.FPRecipEstimate);
-            SetDelegateInfo(new SoftFloat64FPRecipEstimateFpscr(SoftFloat64.FPRecipEstimateFpscr)); // A32 only.
+            var dlgSoftFloat64FPRecipEstimateFpscr = new SoftFloat64FPRecipEstimateFpscr(SoftFloat64.FPRecipEstimateFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPRecipStep(SoftFloat64.FPRecipStep)); // A32 only.
+            var dlgSoftFloat64FPRecipStep = new SoftFloat64FPRecipStep(SoftFloat64.FPRecipStep); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPRecipStepFused(SoftFloat64.FPRecipStepFused));
+            var dlgSoftFloat64FPRecipStepFused = new SoftFloat64FPRecipStepFused(SoftFloat64.FPRecipStepFused);
-            SetDelegateInfo(new SoftFloat64FPRecpX(SoftFloat64.FPRecpX));
+            var dlgSoftFloat64FPRecpX = new SoftFloat64FPRecpX(SoftFloat64.FPRecpX);
-            SetDelegateInfo(new SoftFloat64FPRSqrtEstimate(SoftFloat64.FPRSqrtEstimate));
+            var dlgSoftFloat64FPRSqrtEstimate = new SoftFloat64FPRSqrtEstimate(SoftFloat64.FPRSqrtEstimate);
-            SetDelegateInfo(new SoftFloat64FPRSqrtEstimateFpscr(SoftFloat64.FPRSqrtEstimateFpscr)); // A32 only.
+            var dlgSoftFloat64FPRSqrtEstimateFpscr = new SoftFloat64FPRSqrtEstimateFpscr(SoftFloat64.FPRSqrtEstimateFpscr); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPRSqrtStep(SoftFloat64.FPRSqrtStep)); // A32 only.
+            var dlgSoftFloat64FPRSqrtStep = new SoftFloat64FPRSqrtStep(SoftFloat64.FPRSqrtStep); // A32 only.
-            SetDelegateInfo(new SoftFloat64FPRSqrtStepFused(SoftFloat64.FPRSqrtStepFused));
+            var dlgSoftFloat64FPRSqrtStepFused = new SoftFloat64FPRSqrtStepFused(SoftFloat64.FPRSqrtStepFused);
-            SetDelegateInfo(new SoftFloat64FPSqrt(SoftFloat64.FPSqrt));
+            var dlgSoftFloat64FPSqrt = new SoftFloat64FPSqrt(SoftFloat64.FPSqrt);
-            SetDelegateInfo(new SoftFloat64FPSub(SoftFloat64.FPSub));
+            var dlgSoftFloat64FPSub = new SoftFloat64FPSub(SoftFloat64.FPSub);
 
-            SetDelegateInfo(new SoftFloat64_16FPConvert(SoftFloat64_16.FPConvert));
+            var dlgSoftFloat64_16FPConvert = new SoftFloat64_16FPConvert(SoftFloat64_16.FPConvert);
+
+            SetDelegateInfo(dlgMathAbs, Marshal.GetFunctionPointerForDelegate<MathAbs>(dlgMathAbs));
+            SetDelegateInfo(dlgMathCeiling, Marshal.GetFunctionPointerForDelegate<MathCeiling>(dlgMathCeiling));
+            SetDelegateInfo(dlgMathFloor, Marshal.GetFunctionPointerForDelegate<MathFloor>(dlgMathFloor));
+            SetDelegateInfo(dlgMathRound, Marshal.GetFunctionPointerForDelegate<MathRound>(dlgMathRound));
+            SetDelegateInfo(dlgMathTruncate, Marshal.GetFunctionPointerForDelegate<MathTruncate>(dlgMathTruncate));
+
+            SetDelegateInfo(dlgMathFAbs, Marshal.GetFunctionPointerForDelegate<MathFAbs>(dlgMathFAbs));
+            SetDelegateInfo(dlgMathFCeiling, Marshal.GetFunctionPointerForDelegate<MathFCeiling>(dlgMathFCeiling));
+            SetDelegateInfo(dlgMathFFloor, Marshal.GetFunctionPointerForDelegate<MathFFloor>(dlgMathFFloor));
+            SetDelegateInfo(dlgMathFRound, Marshal.GetFunctionPointerForDelegate<MathFRound>(dlgMathFRound));
+            SetDelegateInfo(dlgMathFTruncate, Marshal.GetFunctionPointerForDelegate<MathFTruncate>(dlgMathFTruncate));
+
+            SetDelegateInfo(dlgNativeInterfaceBreak, Marshal.GetFunctionPointerForDelegate<NativeInterfaceBreak>(dlgNativeInterfaceBreak));
+            SetDelegateInfo(dlgNativeInterfaceCheckSynchronization, Marshal.GetFunctionPointerForDelegate<NativeInterfaceCheckSynchronization>(dlgNativeInterfaceCheckSynchronization));
+            SetDelegateInfo(dlgNativeInterfaceEnqueueForRejit, Marshal.GetFunctionPointerForDelegate<NativeInterfaceEnqueueForRejit>(dlgNativeInterfaceEnqueueForRejit));
+            SetDelegateInfo(dlgNativeInterfaceGetCntfrqEl0, Marshal.GetFunctionPointerForDelegate<NativeInterfaceGetCntfrqEl0>(dlgNativeInterfaceGetCntfrqEl0));
+            SetDelegateInfo(dlgNativeInterfaceGetCntpctEl0, Marshal.GetFunctionPointerForDelegate<NativeInterfaceGetCntpctEl0>(dlgNativeInterfaceGetCntpctEl0));
+            SetDelegateInfo(dlgNativeInterfaceGetCntvctEl0, Marshal.GetFunctionPointerForDelegate<NativeInterfaceGetCntvctEl0>(dlgNativeInterfaceGetCntvctEl0));
+            SetDelegateInfo(dlgNativeInterfaceGetCtrEl0, Marshal.GetFunctionPointerForDelegate<NativeInterfaceGetCtrEl0>(dlgNativeInterfaceGetCtrEl0));
+            SetDelegateInfo(dlgNativeInterfaceGetDczidEl0, Marshal.GetFunctionPointerForDelegate<NativeInterfaceGetDczidEl0>(dlgNativeInterfaceGetDczidEl0));
+            SetDelegateInfo(dlgNativeInterfaceGetFunctionAddress, Marshal.GetFunctionPointerForDelegate<NativeInterfaceGetFunctionAddress>(dlgNativeInterfaceGetFunctionAddress));
+            SetDelegateInfo(dlgNativeInterfaceInvalidateCacheLine, Marshal.GetFunctionPointerForDelegate<NativeInterfaceInvalidateCacheLine>(dlgNativeInterfaceInvalidateCacheLine));
+            SetDelegateInfo(dlgNativeInterfaceReadByte, Marshal.GetFunctionPointerForDelegate<NativeInterfaceReadByte>(dlgNativeInterfaceReadByte));
+            SetDelegateInfo(dlgNativeInterfaceReadUInt16, Marshal.GetFunctionPointerForDelegate<NativeInterfaceReadUInt16>(dlgNativeInterfaceReadUInt16));
+            SetDelegateInfo(dlgNativeInterfaceReadUInt32, Marshal.GetFunctionPointerForDelegate<NativeInterfaceReadUInt32>(dlgNativeInterfaceReadUInt32));
+            SetDelegateInfo(dlgNativeInterfaceReadUInt64, Marshal.GetFunctionPointerForDelegate<NativeInterfaceReadUInt64>(dlgNativeInterfaceReadUInt64));
+            SetDelegateInfo(dlgNativeInterfaceReadVector128, Marshal.GetFunctionPointerForDelegate<NativeInterfaceReadVector128>(dlgNativeInterfaceReadVector128));
+            SetDelegateInfo(dlgNativeInterfaceSignalMemoryTracking, Marshal.GetFunctionPointerForDelegate<NativeInterfaceSignalMemoryTracking>(dlgNativeInterfaceSignalMemoryTracking));
+            SetDelegateInfo(dlgNativeInterfaceSupervisorCall, Marshal.GetFunctionPointerForDelegate<NativeInterfaceSupervisorCall>(dlgNativeInterfaceSupervisorCall));
+            SetDelegateInfo(dlgNativeInterfaceThrowInvalidMemoryAccess, Marshal.GetFunctionPointerForDelegate<NativeInterfaceThrowInvalidMemoryAccess>(dlgNativeInterfaceThrowInvalidMemoryAccess));
+            SetDelegateInfo(dlgNativeInterfaceUndefined, Marshal.GetFunctionPointerForDelegate<NativeInterfaceUndefined>(dlgNativeInterfaceUndefined));
+            SetDelegateInfo(dlgNativeInterfaceWriteByte, Marshal.GetFunctionPointerForDelegate<NativeInterfaceWriteByte>(dlgNativeInterfaceWriteByte));
+            SetDelegateInfo(dlgNativeInterfaceWriteUInt16, Marshal.GetFunctionPointerForDelegate<NativeInterfaceWriteUInt16>(dlgNativeInterfaceWriteUInt16));
+            SetDelegateInfo(dlgNativeInterfaceWriteUInt32, Marshal.GetFunctionPointerForDelegate<NativeInterfaceWriteUInt32>(dlgNativeInterfaceWriteUInt32));
+            SetDelegateInfo(dlgNativeInterfaceWriteUInt64, Marshal.GetFunctionPointerForDelegate<NativeInterfaceWriteUInt64>(dlgNativeInterfaceWriteUInt64));
+            SetDelegateInfo(dlgNativeInterfaceWriteVector128, Marshal.GetFunctionPointerForDelegate<NativeInterfaceWriteVector128>(dlgNativeInterfaceWriteVector128));
+
+            SetDelegateInfo(dlgSoftFallbackCountLeadingSigns, Marshal.GetFunctionPointerForDelegate<SoftFallbackCountLeadingSigns>(dlgSoftFallbackCountLeadingSigns));
+            SetDelegateInfo(dlgSoftFallbackCountLeadingZeros, Marshal.GetFunctionPointerForDelegate<SoftFallbackCountLeadingZeros>(dlgSoftFallbackCountLeadingZeros));
+            SetDelegateInfo(dlgSoftFallbackCrc32b, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32b>(dlgSoftFallbackCrc32b));
+            SetDelegateInfo(dlgSoftFallbackCrc32cb, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32cb>(dlgSoftFallbackCrc32cb));
+            SetDelegateInfo(dlgSoftFallbackCrc32ch, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32ch>(dlgSoftFallbackCrc32ch));
+            SetDelegateInfo(dlgSoftFallbackCrc32cw, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32cw>(dlgSoftFallbackCrc32cw));
+            SetDelegateInfo(dlgSoftFallbackCrc32cx, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32cx>(dlgSoftFallbackCrc32cx));
+            SetDelegateInfo(dlgSoftFallbackCrc32h, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32h>(dlgSoftFallbackCrc32h));
+            SetDelegateInfo(dlgSoftFallbackCrc32w, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32w>(dlgSoftFallbackCrc32w));
+            SetDelegateInfo(dlgSoftFallbackCrc32x, Marshal.GetFunctionPointerForDelegate<SoftFallbackCrc32x>(dlgSoftFallbackCrc32x));
+            SetDelegateInfo(dlgSoftFallbackDecrypt, Marshal.GetFunctionPointerForDelegate<SoftFallbackDecrypt>(dlgSoftFallbackDecrypt));
+            SetDelegateInfo(dlgSoftFallbackEncrypt, Marshal.GetFunctionPointerForDelegate<SoftFallbackEncrypt>(dlgSoftFallbackEncrypt));
+            SetDelegateInfo(dlgSoftFallbackFixedRotate, Marshal.GetFunctionPointerForDelegate<SoftFallbackFixedRotate>(dlgSoftFallbackFixedRotate));
+            SetDelegateInfo(dlgSoftFallbackHashChoose, Marshal.GetFunctionPointerForDelegate<SoftFallbackHashChoose>(dlgSoftFallbackHashChoose));
+            SetDelegateInfo(dlgSoftFallbackHashLower, Marshal.GetFunctionPointerForDelegate<SoftFallbackHashLower>(dlgSoftFallbackHashLower));
+            SetDelegateInfo(dlgSoftFallbackHashMajority, Marshal.GetFunctionPointerForDelegate<SoftFallbackHashMajority>(dlgSoftFallbackHashMajority));
+            SetDelegateInfo(dlgSoftFallbackHashParity, Marshal.GetFunctionPointerForDelegate<SoftFallbackHashParity>(dlgSoftFallbackHashParity));
+            SetDelegateInfo(dlgSoftFallbackHashUpper, Marshal.GetFunctionPointerForDelegate<SoftFallbackHashUpper>(dlgSoftFallbackHashUpper));
+            SetDelegateInfo(dlgSoftFallbackInverseMixColumns, Marshal.GetFunctionPointerForDelegate<SoftFallbackInverseMixColumns>(dlgSoftFallbackInverseMixColumns));
+            SetDelegateInfo(dlgSoftFallbackMixColumns, Marshal.GetFunctionPointerForDelegate<SoftFallbackMixColumns>(dlgSoftFallbackMixColumns));
+            SetDelegateInfo(dlgSoftFallbackPolynomialMult64_128, Marshal.GetFunctionPointerForDelegate<SoftFallbackPolynomialMult64_128>(dlgSoftFallbackPolynomialMult64_128));
+            SetDelegateInfo(dlgSoftFallbackSatF32ToS32, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF32ToS32>(dlgSoftFallbackSatF32ToS32));
+            SetDelegateInfo(dlgSoftFallbackSatF32ToS64, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF32ToS64>(dlgSoftFallbackSatF32ToS64));
+            SetDelegateInfo(dlgSoftFallbackSatF32ToU32, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF32ToU32>(dlgSoftFallbackSatF32ToU32));
+            SetDelegateInfo(dlgSoftFallbackSatF32ToU64, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF32ToU64>(dlgSoftFallbackSatF32ToU64));
+            SetDelegateInfo(dlgSoftFallbackSatF64ToS32, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF64ToS32>(dlgSoftFallbackSatF64ToS32));
+            SetDelegateInfo(dlgSoftFallbackSatF64ToS64, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF64ToS64>(dlgSoftFallbackSatF64ToS64));
+            SetDelegateInfo(dlgSoftFallbackSatF64ToU32, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF64ToU32>(dlgSoftFallbackSatF64ToU32));
+            SetDelegateInfo(dlgSoftFallbackSatF64ToU64, Marshal.GetFunctionPointerForDelegate<SoftFallbackSatF64ToU64>(dlgSoftFallbackSatF64ToU64));
+            SetDelegateInfo(dlgSoftFallbackSha1SchedulePart1, Marshal.GetFunctionPointerForDelegate<SoftFallbackSha1SchedulePart1>(dlgSoftFallbackSha1SchedulePart1));
+            SetDelegateInfo(dlgSoftFallbackSha1SchedulePart2, Marshal.GetFunctionPointerForDelegate<SoftFallbackSha1SchedulePart2>(dlgSoftFallbackSha1SchedulePart2));
+            SetDelegateInfo(dlgSoftFallbackSha256SchedulePart1, Marshal.GetFunctionPointerForDelegate<SoftFallbackSha256SchedulePart1>(dlgSoftFallbackSha256SchedulePart1));
+            SetDelegateInfo(dlgSoftFallbackSha256SchedulePart2, Marshal.GetFunctionPointerForDelegate<SoftFallbackSha256SchedulePart2>(dlgSoftFallbackSha256SchedulePart2));
+            SetDelegateInfo(dlgSoftFallbackSignedShrImm64, Marshal.GetFunctionPointerForDelegate<SoftFallbackSignedShrImm64>(dlgSoftFallbackSignedShrImm64));
+            SetDelegateInfo(dlgSoftFallbackTbl1, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbl1>(dlgSoftFallbackTbl1));
+            SetDelegateInfo(dlgSoftFallbackTbl2, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbl2>(dlgSoftFallbackTbl2));
+            SetDelegateInfo(dlgSoftFallbackTbl3, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbl3>(dlgSoftFallbackTbl3));
+            SetDelegateInfo(dlgSoftFallbackTbl4, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbl4>(dlgSoftFallbackTbl4));
+            SetDelegateInfo(dlgSoftFallbackTbx1, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbx1>(dlgSoftFallbackTbx1));
+            SetDelegateInfo(dlgSoftFallbackTbx2, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbx2>(dlgSoftFallbackTbx2));
+            SetDelegateInfo(dlgSoftFallbackTbx3, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbx3>(dlgSoftFallbackTbx3));
+            SetDelegateInfo(dlgSoftFallbackTbx4, Marshal.GetFunctionPointerForDelegate<SoftFallbackTbx4>(dlgSoftFallbackTbx4));
+            SetDelegateInfo(dlgSoftFallbackUnsignedShrImm64, Marshal.GetFunctionPointerForDelegate<SoftFallbackUnsignedShrImm64>(dlgSoftFallbackUnsignedShrImm64));
+
+            SetDelegateInfo(dlgSoftFloat16_32FPConvert, Marshal.GetFunctionPointerForDelegate<SoftFloat16_32FPConvert>(dlgSoftFloat16_32FPConvert));
+            SetDelegateInfo(dlgSoftFloat16_64FPConvert, Marshal.GetFunctionPointerForDelegate<SoftFloat16_64FPConvert>(dlgSoftFloat16_64FPConvert));
+
+            SetDelegateInfo(dlgSoftFloat32FPAdd, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPAdd>(dlgSoftFloat32FPAdd));
+            SetDelegateInfo(dlgSoftFloat32FPAddFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPAddFpscr>(dlgSoftFloat32FPAddFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPCompare, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompare>(dlgSoftFloat32FPCompare));
+            SetDelegateInfo(dlgSoftFloat32FPCompareEQ, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareEQ>(dlgSoftFloat32FPCompareEQ));
+            SetDelegateInfo(dlgSoftFloat32FPCompareEQFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareEQFpscr>(dlgSoftFloat32FPCompareEQFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPCompareGE, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareGE>(dlgSoftFloat32FPCompareGE));
+            SetDelegateInfo(dlgSoftFloat32FPCompareGEFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareGEFpscr>(dlgSoftFloat32FPCompareGEFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPCompareGT, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareGT>(dlgSoftFloat32FPCompareGT));
+            SetDelegateInfo(dlgSoftFloat32FPCompareGTFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareGTFpscr>(dlgSoftFloat32FPCompareGTFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPCompareLE, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareLE>(dlgSoftFloat32FPCompareLE));
+            SetDelegateInfo(dlgSoftFloat32FPCompareLEFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareLEFpscr>(dlgSoftFloat32FPCompareLEFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPCompareLT, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareLT>(dlgSoftFloat32FPCompareLT));
+            SetDelegateInfo(dlgSoftFloat32FPCompareLTFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPCompareLTFpscr>(dlgSoftFloat32FPCompareLTFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPDiv, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPDiv>(dlgSoftFloat32FPDiv));
+            SetDelegateInfo(dlgSoftFloat32FPMax, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMax>(dlgSoftFloat32FPMax));
+            SetDelegateInfo(dlgSoftFloat32FPMaxFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMaxFpscr>(dlgSoftFloat32FPMaxFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPMaxNum, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMaxNum>(dlgSoftFloat32FPMaxNum));
+            SetDelegateInfo(dlgSoftFloat32FPMaxNumFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMaxNumFpscr>(dlgSoftFloat32FPMaxNumFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPMin, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMin>(dlgSoftFloat32FPMin));
+            SetDelegateInfo(dlgSoftFloat32FPMinFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMinFpscr>(dlgSoftFloat32FPMinFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPMinNum, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMinNum>(dlgSoftFloat32FPMinNum));
+            SetDelegateInfo(dlgSoftFloat32FPMinNumFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMinNumFpscr>(dlgSoftFloat32FPMinNumFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPMul, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMul>(dlgSoftFloat32FPMul));
+            SetDelegateInfo(dlgSoftFloat32FPMulFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMulFpscr>(dlgSoftFloat32FPMulFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPMulAdd, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMulAdd>(dlgSoftFloat32FPMulAdd));
+            SetDelegateInfo(dlgSoftFloat32FPMulAddFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMulAddFpscr>(dlgSoftFloat32FPMulAddFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPMulSub, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMulSub>(dlgSoftFloat32FPMulSub));
+            SetDelegateInfo(dlgSoftFloat32FPMulSubFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMulSubFpscr>(dlgSoftFloat32FPMulSubFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPMulX, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPMulX>(dlgSoftFloat32FPMulX));
+            SetDelegateInfo(dlgSoftFloat32FPNegMulAdd, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPNegMulAdd>(dlgSoftFloat32FPNegMulAdd));
+            SetDelegateInfo(dlgSoftFloat32FPNegMulSub, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPNegMulSub>(dlgSoftFloat32FPNegMulSub));
+            SetDelegateInfo(dlgSoftFloat32FPRecipEstimate, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRecipEstimate>(dlgSoftFloat32FPRecipEstimate));
+            SetDelegateInfo(dlgSoftFloat32FPRecipEstimateFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRecipEstimateFpscr>(dlgSoftFloat32FPRecipEstimateFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPRecipStep, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRecipStep>(dlgSoftFloat32FPRecipStep)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPRecipStepFused, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRecipStepFused>(dlgSoftFloat32FPRecipStepFused));
+            SetDelegateInfo(dlgSoftFloat32FPRecpX, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRecpX>(dlgSoftFloat32FPRecpX));
+            SetDelegateInfo(dlgSoftFloat32FPRSqrtEstimate, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRSqrtEstimate>(dlgSoftFloat32FPRSqrtEstimate));
+            SetDelegateInfo(dlgSoftFloat32FPRSqrtEstimateFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRSqrtEstimateFpscr>(dlgSoftFloat32FPRSqrtEstimateFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPRSqrtStep, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRSqrtStep>(dlgSoftFloat32FPRSqrtStep)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat32FPRSqrtStepFused, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPRSqrtStepFused>(dlgSoftFloat32FPRSqrtStepFused));
+            SetDelegateInfo(dlgSoftFloat32FPSqrt, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPSqrt>(dlgSoftFloat32FPSqrt));
+            SetDelegateInfo(dlgSoftFloat32FPSub, Marshal.GetFunctionPointerForDelegate<SoftFloat32FPSub>(dlgSoftFloat32FPSub));
+
+            SetDelegateInfo(dlgSoftFloat32_16FPConvert, Marshal.GetFunctionPointerForDelegate<SoftFloat32_16FPConvert>(dlgSoftFloat32_16FPConvert));
+
+            SetDelegateInfo(dlgSoftFloat64FPAdd, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPAdd>(dlgSoftFloat64FPAdd));
+            SetDelegateInfo(dlgSoftFloat64FPAddFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPAddFpscr>(dlgSoftFloat64FPAddFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPCompare, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompare>(dlgSoftFloat64FPCompare));
+            SetDelegateInfo(dlgSoftFloat64FPCompareEQ, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareEQ>(dlgSoftFloat64FPCompareEQ));
+            SetDelegateInfo(dlgSoftFloat64FPCompareEQFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareEQFpscr>(dlgSoftFloat64FPCompareEQFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPCompareGE, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareGE>(dlgSoftFloat64FPCompareGE));
+            SetDelegateInfo(dlgSoftFloat64FPCompareGEFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareGEFpscr>(dlgSoftFloat64FPCompareGEFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPCompareGT, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareGT>(dlgSoftFloat64FPCompareGT));
+            SetDelegateInfo(dlgSoftFloat64FPCompareGTFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareGTFpscr>(dlgSoftFloat64FPCompareGTFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPCompareLE, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareLE>(dlgSoftFloat64FPCompareLE));
+            SetDelegateInfo(dlgSoftFloat64FPCompareLEFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareLEFpscr>(dlgSoftFloat64FPCompareLEFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPCompareLT, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareLT>(dlgSoftFloat64FPCompareLT));
+            SetDelegateInfo(dlgSoftFloat64FPCompareLTFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPCompareLTFpscr>(dlgSoftFloat64FPCompareLTFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPDiv, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPDiv>(dlgSoftFloat64FPDiv));
+            SetDelegateInfo(dlgSoftFloat64FPMax, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMax>(dlgSoftFloat64FPMax));
+            SetDelegateInfo(dlgSoftFloat64FPMaxFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMaxFpscr>(dlgSoftFloat64FPMaxFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPMaxNum, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMaxNum>(dlgSoftFloat64FPMaxNum));
+            SetDelegateInfo(dlgSoftFloat64FPMaxNumFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMaxNumFpscr>(dlgSoftFloat64FPMaxNumFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPMin, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMin>(dlgSoftFloat64FPMin));
+            SetDelegateInfo(dlgSoftFloat64FPMinFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMinFpscr>(dlgSoftFloat64FPMinFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPMinNum, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMinNum>(dlgSoftFloat64FPMinNum));
+            SetDelegateInfo(dlgSoftFloat64FPMinNumFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMinNumFpscr>(dlgSoftFloat64FPMinNumFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPMul, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMul>(dlgSoftFloat64FPMul));
+            SetDelegateInfo(dlgSoftFloat64FPMulFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMulFpscr>(dlgSoftFloat64FPMulFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPMulAdd, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMulAdd>(dlgSoftFloat64FPMulAdd));
+            SetDelegateInfo(dlgSoftFloat64FPMulAddFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMulAddFpscr>(dlgSoftFloat64FPMulAddFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPMulSub, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMulSub>(dlgSoftFloat64FPMulSub));
+            SetDelegateInfo(dlgSoftFloat64FPMulSubFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMulSubFpscr>(dlgSoftFloat64FPMulSubFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPMulX, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPMulX>(dlgSoftFloat64FPMulX));
+            SetDelegateInfo(dlgSoftFloat64FPNegMulAdd, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPNegMulAdd>(dlgSoftFloat64FPNegMulAdd));
+            SetDelegateInfo(dlgSoftFloat64FPNegMulSub, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPNegMulSub>(dlgSoftFloat64FPNegMulSub));
+            SetDelegateInfo(dlgSoftFloat64FPRecipEstimate, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRecipEstimate>(dlgSoftFloat64FPRecipEstimate));
+            SetDelegateInfo(dlgSoftFloat64FPRecipEstimateFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRecipEstimateFpscr>(dlgSoftFloat64FPRecipEstimateFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPRecipStep, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRecipStep>(dlgSoftFloat64FPRecipStep)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPRecipStepFused, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRecipStepFused>(dlgSoftFloat64FPRecipStepFused));
+            SetDelegateInfo(dlgSoftFloat64FPRecpX, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRecpX>(dlgSoftFloat64FPRecpX));
+            SetDelegateInfo(dlgSoftFloat64FPRSqrtEstimate, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRSqrtEstimate>(dlgSoftFloat64FPRSqrtEstimate));
+            SetDelegateInfo(dlgSoftFloat64FPRSqrtEstimateFpscr, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRSqrtEstimateFpscr>(dlgSoftFloat64FPRSqrtEstimateFpscr)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPRSqrtStep, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRSqrtStep>(dlgSoftFloat64FPRSqrtStep)); // A32 only.
+            SetDelegateInfo(dlgSoftFloat64FPRSqrtStepFused, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPRSqrtStepFused>(dlgSoftFloat64FPRSqrtStepFused));
+            SetDelegateInfo(dlgSoftFloat64FPSqrt, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPSqrt>(dlgSoftFloat64FPSqrt));
+            SetDelegateInfo(dlgSoftFloat64FPSub, Marshal.GetFunctionPointerForDelegate<SoftFloat64FPSub>(dlgSoftFloat64FPSub));
+
+            SetDelegateInfo(dlgSoftFloat64_16FPConvert, Marshal.GetFunctionPointerForDelegate<SoftFloat64_16FPConvert>(dlgSoftFloat64_16FPConvert));
         }
 
         private delegate double MathAbs(double value);

src/Ryujinx.Common/GraphicsDriver/DriverUtilities.cs

@@ -1,13 +1,33 @@
+using Ryujinx.Common.Utilities;
 using System;
 
 namespace Ryujinx.Common.GraphicsDriver
 {
     public static class DriverUtilities
     {
+        private static void AddMesaFlags(string envVar, string newFlags)
+        {
+            string existingFlags = Environment.GetEnvironmentVariable(envVar);
+
+            string flags = existingFlags == null ? newFlags : $"{existingFlags},{newFlags}";
+
+            OsUtils.SetEnvironmentVariableNoCaching(envVar, flags);
+        }
+
+        public static void InitDriverConfig(bool oglThreading)
+        {
+            if (OperatingSystem.IsLinux())
+            {
+                AddMesaFlags("RADV_DEBUG", "nodcc");
+            }
+
+            ToggleOGLThreading(oglThreading);
+        }
+
         public static void ToggleOGLThreading(bool enabled)
         {
-            Environment.SetEnvironmentVariable("mesa_glthread", enabled.ToString().ToLower());
+            OsUtils.SetEnvironmentVariableNoCaching("mesa_glthread", enabled.ToString().ToLower());
-            Environment.SetEnvironmentVariable("__GL_THREADED_OPTIMIZATIONS", enabled ? "1" : "0");
+            OsUtils.SetEnvironmentVariableNoCaching("__GL_THREADED_OPTIMIZATIONS", enabled ? "1" : "0");
 
             try
             {

src/Ryujinx.Common/Utilities/OsUtils.cs

@@ -0,0 +1,24 @@
+using System;
+using System.Diagnostics;
+using System.Runtime.InteropServices;
+
+namespace Ryujinx.Common.Utilities
+{
+    public partial class OsUtils
+    {
+        [LibraryImport("libc", SetLastError = true)]
+        private static partial int setenv([MarshalAs(UnmanagedType.LPStr)] string name, [MarshalAs(UnmanagedType.LPStr)] string value, int overwrite);
+
+        public static void SetEnvironmentVariableNoCaching(string key, string value)
+        {
+            // Set the value in the cached environment variables, too.
+            Environment.SetEnvironmentVariable(key, value);
+
+            if (!OperatingSystem.IsWindows())
+            {
+                int res = setenv(key, value, 1);
+                Debug.Assert(res != -1);
+            }
+        }
+    }
+}

src/Ryujinx.Graphics.Gpu/Engine/Dma/DmaClass.cs

@@ -276,8 +276,6 @@ namespace Ryujinx.Graphics.Gpu.Engine.Dma
                     dstBaseOffset += dstStride * (yCount - 1);
                 }
 
-                ReadOnlySpan<byte> srcSpan = memoryManager.GetSpan(srcGpuVa + (ulong)srcBaseOffset, srcSize, true);
-
                 // If remapping is disabled, we always copy the components directly, in order.
                 // If it's enabled, but the mapping is just XYZW, we also copy them in order.
                 bool isIdentityRemap = !remap ||
@@ -289,6 +287,52 @@ namespace Ryujinx.Graphics.Gpu.Engine.Dma
                 bool completeSource = IsTextureCopyComplete(src, srcLinear, srcBpp, srcStride, xCount, yCount);
                 bool completeDest = IsTextureCopyComplete(dst, dstLinear, dstBpp, dstStride, xCount, yCount);
 
+                // Check if the source texture exists on the GPU, if it does, do a GPU side copy.
+                // Otherwise, we would need to flush the source texture which is costly.
+                // We don't expect the source to be linear in such cases, as linear source usually indicates buffer or CPU written data.
+
+                if (completeSource && completeDest && !srcLinear && isIdentityRemap)
+                {
+                    var source = memoryManager.Physical.TextureCache.FindTexture(
+                        memoryManager,
+                        srcGpuVa,
+                        srcBpp,
+                        srcStride,
+                        src.Height,
+                        xCount,
+                        yCount,
+                        srcLinear,
+                        src.MemoryLayout.UnpackGobBlocksInY(),
+                        src.MemoryLayout.UnpackGobBlocksInZ());
+
+                    if (source != null && source.Height == yCount)
+                    {
+                        source.SynchronizeMemory();
+
+                        var target = memoryManager.Physical.TextureCache.FindOrCreateTexture(
+                            memoryManager,
+                            source.Info.FormatInfo,
+                            dstGpuVa,
+                            xCount,
+                            yCount,
+                            dstStride,
+                            dstLinear,
+                            dst.MemoryLayout.UnpackGobBlocksInY(),
+                            dst.MemoryLayout.UnpackGobBlocksInZ());
+
+                        if (source.ScaleFactor != target.ScaleFactor)
+                        {
+                            target.PropagateScale(source);
+                        }
+
+                        source.HostTexture.CopyTo(target.HostTexture, 0, 0);
+                        target.SignalModified();
+                        return;
+                    }
+                }
+
+                ReadOnlySpan<byte> srcSpan = memoryManager.GetSpan(srcGpuVa + (ulong)srcBaseOffset, srcSize, true);
+
                 // Try to set the texture data directly,
                 // but only if we are doing a complete copy,
                 // and not for block linear to linear copies, since those are typically accessed from the CPU.

src/Ryujinx.Graphics.Gpu/Engine/Threed/ThreedClassState.cs

@@ -415,7 +415,13 @@ namespace Ryujinx.Graphics.Gpu.Engine.Threed
 #pragma warning disable CS0649 // Field is never assigned to
         public int Width;
         public int Height;
-        public int Depth;
+        public ushort Depth;
+        public ushort Flags;
+
+        public readonly bool UnpackIsLayered()
+        {
+            return (Flags & 1) == 0;
+        }
 #pragma warning restore CS0649
     }
 

src/Ryujinx.Graphics.Gpu/Image/TextureCache.cs

@@ -347,6 +347,53 @@ namespace Ryujinx.Graphics.Gpu.Image
             return texture;
         }
 
+        /// <summary>
+        /// Tries to find an existing texture, or create a new one if not found.
+        /// </summary>
+        /// <param name="memoryManager">GPU memory manager where the texture is mapped</param>
+        /// <param name="formatInfo">Format of the texture</param>
+        /// <param name="gpuAddress">GPU virtual address of the texture</param>
+        /// <param name="xCount">Texture width in bytes</param>
+        /// <param name="yCount">Texture height</param>
+        /// <param name="stride">Texture stride if linear, otherwise ignored</param>
+        /// <param name="isLinear">Indicates if the texture is linear or block linear</param>
+        /// <param name="gobBlocksInY">GOB blocks in Y for block linear textures</param>
+        /// <param name="gobBlocksInZ">GOB blocks in Z for 3D block linear textures</param>
+        /// <returns>The texture</returns>
+        public Texture FindOrCreateTexture(
+            MemoryManager memoryManager,
+            FormatInfo formatInfo,
+            ulong gpuAddress,
+            int xCount,
+            int yCount,
+            int stride,
+            bool isLinear,
+            int gobBlocksInY,
+            int gobBlocksInZ)
+        {
+            TextureInfo info = new(
+                gpuAddress,
+                xCount / formatInfo.BytesPerPixel,
+                yCount,
+                1,
+                1,
+                1,
+                1,
+                stride,
+                isLinear,
+                gobBlocksInY,
+                gobBlocksInZ,
+                1,
+                Target.Texture2D,
+                formatInfo);
+
+            Texture texture = FindOrCreateTexture(memoryManager, TextureSearchFlags.ForCopy, info, 0, sizeHint: new Size(xCount, yCount, 1));
+
+            texture?.SynchronizeMemory();
+
+            return texture;
+        }
+
         /// <summary>
         /// Tries to find an existing texture, or create a new one if not found.
         /// </summary>
@@ -468,13 +515,11 @@ namespace Ryujinx.Graphics.Gpu.Image
             int gobBlocksInY = dsState.MemoryLayout.UnpackGobBlocksInY();
             int gobBlocksInZ = dsState.MemoryLayout.UnpackGobBlocksInZ();
 
+            layered &= size.UnpackIsLayered();
+
             Target target;
 
-            if (dsState.MemoryLayout.UnpackIsTarget3D())
+            if ((samplesInX | samplesInY) != 1)
-            {
-                target = Target.Texture3D;
-            }
-            else if ((samplesInX | samplesInY) != 1)
             {
                 target = size.Depth > 1 && layered
                     ? Target.Texture2DMultisampleArray

src/Ryujinx.Graphics.Vulkan/BarrierBatch.cs

@@ -32,10 +32,12 @@ namespace Ryujinx.Graphics.Vulkan
             CommandBuffer
         }
 
+        private bool _feedbackLoopActive;
         private PipelineStageFlags _incoherentBufferWriteStages;
         private PipelineStageFlags _incoherentTextureWriteStages;
         private PipelineStageFlags _extraStages;
         private IncoherentBarrierType _queuedIncoherentBarrier;
+        private bool _queuedFeedbackLoopBarrier;
 
         public BarrierBatch(VulkanRenderer gd)
         {
@@ -53,17 +55,6 @@ namespace Ryujinx.Graphics.Vulkan
                 stages |= PipelineStageFlags.TransformFeedbackBitExt;
             }
 
-            if (!gd.IsTBDR)
-            {
-                // Desktop GPUs can transform image barriers into memory barriers.
-
-                access |= AccessFlags.DepthStencilAttachmentWriteBit | AccessFlags.ColorAttachmentWriteBit;
-                access |= AccessFlags.DepthStencilAttachmentReadBit | AccessFlags.ColorAttachmentReadBit;
-
-                stages |= PipelineStageFlags.EarlyFragmentTestsBit | PipelineStageFlags.LateFragmentTestsBit;
-                stages |= PipelineStageFlags.ColorAttachmentOutputBit;
-            }
-
             return (access, stages);
         }
 
@@ -178,16 +169,34 @@ namespace Ryujinx.Graphics.Vulkan
                     }
 
                     _queuedIncoherentBarrier = IncoherentBarrierType.None;
+                    _queuedFeedbackLoopBarrier = false;
                 }
+                else if (_feedbackLoopActive && _queuedFeedbackLoopBarrier)
+                {
+                    // Feedback loop barrier.
+
+                    MemoryBarrier barrier = new MemoryBarrier()
+                    {
+                        SType = StructureType.MemoryBarrier,
+                        SrcAccessMask = AccessFlags.ShaderWriteBit,
+                        DstAccessMask = AccessFlags.ShaderReadBit
+                    };
+
+                    QueueBarrier(barrier, PipelineStageFlags.FragmentShaderBit, PipelineStageFlags.AllGraphicsBit);
+
+                    _queuedFeedbackLoopBarrier = false;
+                }
+
+                _feedbackLoopActive = false;
             }
         }
 
         public unsafe void Flush(CommandBufferScoped cbs, bool inRenderPass, RenderPassHolder rpHolder, Action endRenderPass)
         {
-            Flush(cbs, null, inRenderPass, rpHolder, endRenderPass);
+            Flush(cbs, null, false, inRenderPass, rpHolder, endRenderPass);
         }
 
-        public unsafe void Flush(CommandBufferScoped cbs, ShaderCollection program, bool inRenderPass, RenderPassHolder rpHolder, Action endRenderPass)
+        public unsafe void Flush(CommandBufferScoped cbs, ShaderCollection program, bool feedbackLoopActive, bool inRenderPass, RenderPassHolder rpHolder, Action endRenderPass)
         {
             if (program != null)
             {
@@ -195,6 +204,8 @@ namespace Ryujinx.Graphics.Vulkan
                 _incoherentTextureWriteStages |= program.IncoherentTextureWriteStages;
             }
 
+            _feedbackLoopActive |= feedbackLoopActive;
+
             FlushMemoryBarrier(program, inRenderPass);
 
             if (!inRenderPass && rpHolder != null)
@@ -406,6 +417,8 @@ namespace Ryujinx.Graphics.Vulkan
             {
                 _queuedIncoherentBarrier = type;
             }
+
+            _queuedFeedbackLoopBarrier = true;
         }
 
         public void QueueTextureBarrier()

src/Ryujinx.Graphics.Vulkan/BufferHolder.cs

@@ -122,7 +122,7 @@ namespace Ryujinx.Graphics.Vulkan
                 Range = (uint)size,
             };
 
-            _gd.Api.CreateBufferView(_device, bufferViewCreateInfo, null, out var bufferView).ThrowOnError();
+            _gd.Api.CreateBufferView(_device, in bufferViewCreateInfo, null, out var bufferView).ThrowOnError();
 
             return new Auto<DisposableBufferView>(new DisposableBufferView(_gd.Api, _device, bufferView), this, _waitable, _buffer);
         }
@@ -153,7 +153,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PipelineStageFlags.AllCommandsBit,
                     DependencyFlags.DeviceGroupBit,
                     1,
-                    memoryBarrier,
+                    in memoryBarrier,
                     0,
                     null,
                     0,
@@ -770,7 +770,7 @@ namespace Ryujinx.Graphics.Vulkan
                 0,
                 null,
                 1,
-                memoryBarrier,
+                in memoryBarrier,
                 0,
                 null);
         }

src/Ryujinx.Graphics.Vulkan/BufferManager.cs

@@ -221,7 +221,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PBufferBinds = &bufferBind
                 };
 
-                gd.Api.QueueBindSparse(gd.Queue, 1, bindSparseInfo, default).ThrowOnError();
+                gd.Api.QueueBindSparse(gd.Queue, 1, in bindSparseInfo, default).ThrowOnError();
             }
 
             var holder = new BufferHolder(gd, _device, buffer, (int)size, storageAllocations);

src/Ryujinx.Graphics.Vulkan/BufferState.cs

@@ -25,7 +25,10 @@ namespace Ryujinx.Graphics.Vulkan
             {
                 var buffer = _buffer.Get(cbs, _offset, _size, true).Value;
 
-                gd.TransformFeedbackApi.CmdBindTransformFeedbackBuffers(cbs.CommandBuffer, binding, 1, buffer, (ulong)_offset, (ulong)_size);
+                ulong offset = (ulong)_offset;
+                ulong size = (ulong)_size;
+
+                gd.TransformFeedbackApi.CmdBindTransformFeedbackBuffers(cbs.CommandBuffer, binding, 1, in buffer, in offset, in size);
             }
         }
 

src/Ryujinx.Graphics.Vulkan/CommandBufferPool.cs

@@ -45,7 +45,7 @@ namespace Ryujinx.Graphics.Vulkan
                     Level = CommandBufferLevel.Primary,
                 };
 
-                api.AllocateCommandBuffers(device, allocateInfo, out CommandBuffer);
+                api.AllocateCommandBuffers(device, in allocateInfo, out CommandBuffer);
 
                 Dependants = new List<IAuto>();
                 Waitables = new List<MultiFenceHolder>();
@@ -83,7 +83,7 @@ namespace Ryujinx.Graphics.Vulkan
                         CommandPoolCreateFlags.ResetCommandBufferBit,
             };
 
-            api.CreateCommandPool(device, commandPoolCreateInfo, null, out _pool).ThrowOnError();
+            api.CreateCommandPool(device, in commandPoolCreateInfo, null, out _pool).ThrowOnError();
 
             // We need at least 2 command buffers to get texture data in some cases.
             _totalCommandBuffers = isLight ? 2 : MaxCommandBuffers;
@@ -253,7 +253,7 @@ namespace Ryujinx.Graphics.Vulkan
                             SType = StructureType.CommandBufferBeginInfo,
                         };
 
-                        _api.BeginCommandBuffer(entry.CommandBuffer, commandBufferBeginInfo).ThrowOnError();
+                        _api.BeginCommandBuffer(entry.CommandBuffer, in commandBufferBeginInfo).ThrowOnError();
 
                         return new CommandBufferScoped(this, entry.CommandBuffer, cursor);
                     }
@@ -311,7 +311,7 @@ namespace Ryujinx.Graphics.Vulkan
 
                         lock (_queueLock)
                         {
-                            _api.QueueSubmit(_queue, 1, sInfo, entry.Fence.GetUnsafe()).ThrowOnError();
+                            _api.QueueSubmit(_queue, 1, in sInfo, entry.Fence.GetUnsafe()).ThrowOnError();
                         }
                     }
                 }

src/Ryujinx.Graphics.Vulkan/DescriptorSetCollection.cs

@@ -43,7 +43,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PBufferInfo = &bufferInfo,
                 };
 
-                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, writeDescriptorSet, 0, null);
+                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, in writeDescriptorSet, 0, null);
             }
         }
 
@@ -66,7 +66,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PBufferInfo = pBufferInfo,
                 };
 
-                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, writeDescriptorSet, 0, null);
+                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, in writeDescriptorSet, 0, null);
             }
         }
 
@@ -84,7 +84,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PImageInfo = &imageInfo,
                 };
 
-                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, writeDescriptorSet, 0, null);
+                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, in writeDescriptorSet, 0, null);
             }
         }
 
@@ -107,7 +107,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PImageInfo = pImageInfo,
                 };
 
-                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, writeDescriptorSet, 0, null);
+                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, in writeDescriptorSet, 0, null);
             }
         }
 
@@ -144,7 +144,7 @@ namespace Ryujinx.Graphics.Vulkan
                             PImageInfo = pImageInfo,
                         };
 
-                        _holder.Api.UpdateDescriptorSets(_holder.Device, 1, writeDescriptorSet, 0, null);
+                        _holder.Api.UpdateDescriptorSets(_holder.Device, 1, in writeDescriptorSet, 0, null);
 
                         i += count - 1;
                     }
@@ -166,7 +166,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PTexelBufferView = &texelBufferView,
                 };
 
-                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, writeDescriptorSet, 0, null);
+                _holder.Api.UpdateDescriptorSets(_holder.Device, 1, in writeDescriptorSet, 0, null);
             }
         }
 
@@ -200,7 +200,7 @@ namespace Ryujinx.Graphics.Vulkan
                             PTexelBufferView = pTexelBufferView + i,
                         };
 
-                        _holder.Api.UpdateDescriptorSets(_holder.Device, 1, writeDescriptorSet, 0, null);
+                        _holder.Api.UpdateDescriptorSets(_holder.Device, 1, in writeDescriptorSet, 0, null);
                     }
 
                     i += count;

src/Ryujinx.Graphics.Vulkan/DescriptorSetManager.cs

@@ -40,7 +40,7 @@ namespace Ryujinx.Graphics.Vulkan
                         PPoolSizes = pPoolsSize,
                     };
 
-                    Api.CreateDescriptorPool(device, descriptorPoolCreateInfo, null, out _pool).ThrowOnError();
+                    Api.CreateDescriptorPool(device, in descriptorPoolCreateInfo, null, out _pool).ThrowOnError();
                 }
             }
 

src/Ryujinx.Graphics.Vulkan/DescriptorSetUpdater.cs

@@ -4,6 +4,7 @@ using Ryujinx.Graphics.Shader;
 using Silk.NET.Vulkan;
 using System;
 using System.Buffers;
+using System.Collections.Generic;
 using System.Runtime.CompilerServices;
 using System.Runtime.InteropServices;
 using CompareOp = Ryujinx.Graphics.GAL.CompareOp;
@@ -42,15 +43,15 @@ namespace Ryujinx.Graphics.Vulkan
         private record struct TextureRef
         {
             public ShaderStage Stage;
-            public TextureStorage Storage;
+            public TextureView View;
-            public Auto<DisposableImageView> View;
+            public Auto<DisposableImageView> ImageView;
             public Auto<DisposableSampler> Sampler;
 
-            public TextureRef(ShaderStage stage, TextureStorage storage, Auto<DisposableImageView> view, Auto<DisposableSampler> sampler)
+            public TextureRef(ShaderStage stage, TextureView view, Auto<DisposableImageView> imageView, Auto<DisposableSampler> sampler)
             {
                 Stage = stage;
-                Storage = storage;
                 View = view;
+                ImageView = imageView;
                 Sampler = sampler;
             }
         }
@@ -58,14 +59,14 @@ namespace Ryujinx.Graphics.Vulkan
         private record struct ImageRef
         {
             public ShaderStage Stage;
-            public TextureStorage Storage;
+            public TextureView View;
-            public Auto<DisposableImageView> View;
+            public Auto<DisposableImageView> ImageView;
 
-            public ImageRef(ShaderStage stage, TextureStorage storage, Auto<DisposableImageView> view)
+            public ImageRef(ShaderStage stage, TextureView view, Auto<DisposableImageView> imageView)
             {
                 Stage = stage;
-                Storage = storage;
                 View = view;
+                ImageView = imageView;
             }
         }
 
@@ -124,6 +125,8 @@ namespace Ryujinx.Graphics.Vulkan
         private readonly TextureView _dummyTexture;
         private readonly SamplerHolder _dummySampler;
 
+        public List<TextureView> FeedbackLoopHazards { get; private set; }
+
         public DescriptorSetUpdater(VulkanRenderer gd, Device device)
         {
             _gd = gd;
@@ -209,10 +212,15 @@ namespace Ryujinx.Graphics.Vulkan
             _templateUpdater = new();
         }
 
-        public void Initialize()
+        public void Initialize(bool isMainPipeline)
         {
             MemoryOwner<byte> dummyTextureData = MemoryOwner<byte>.RentCleared(4);
             _dummyTexture.SetData(dummyTextureData);
+
+            if (isMainPipeline)
+            {
+                FeedbackLoopHazards = new();
+            }
         }
 
         private static bool BindingOverlaps(ref DescriptorBufferInfo info, int bindingOffset, int offset, int size)
@@ -275,6 +283,18 @@ namespace Ryujinx.Graphics.Vulkan
 
         public void InsertBindingBarriers(CommandBufferScoped cbs)
         {
+            if ((FeedbackLoopHazards?.Count ?? 0) > 0)
+            {
+                // Clear existing hazards - they will be rebuilt.
+
+                foreach (TextureView hazard in FeedbackLoopHazards)
+                {
+                    hazard.DecrementHazardUses();
+                }
+
+                FeedbackLoopHazards.Clear();
+            }
+
             foreach (ResourceBindingSegment segment in _program.BindingSegments[PipelineBase.TextureSetIndex])
             {
                 if (segment.Type == ResourceType.TextureAndSampler)
@@ -284,7 +304,7 @@ namespace Ryujinx.Graphics.Vulkan
                         for (int i = 0; i < segment.Count; i++)
                         {
                             ref var texture = ref _textureRefs[segment.Binding + i];
-                            texture.Storage?.QueueWriteToReadBarrier(cbs, AccessFlags.ShaderReadBit, texture.Stage.ConvertToPipelineStageFlags());
+                            texture.View?.PrepareForUsage(cbs, texture.Stage.ConvertToPipelineStageFlags(), FeedbackLoopHazards);
                         }
                     }
                     else
@@ -305,7 +325,7 @@ namespace Ryujinx.Graphics.Vulkan
                         for (int i = 0; i < segment.Count; i++)
                         {
                             ref var image = ref _imageRefs[segment.Binding + i];
-                            image.Storage?.QueueWriteToReadBarrier(cbs, AccessFlags.ShaderReadBit, image.Stage.ConvertToPipelineStageFlags());
+                            image.View?.PrepareForUsage(cbs, image.Stage.ConvertToPipelineStageFlags(), FeedbackLoopHazards);
                         }
                     }
                     else
@@ -385,9 +405,12 @@ namespace Ryujinx.Graphics.Vulkan
             }
             else if (image is TextureView view)
             {
-                view.Storage.QueueWriteToReadBarrier(cbs, AccessFlags.ShaderReadBit, stage.ConvertToPipelineStageFlags());
+                ref ImageRef iRef = ref _imageRefs[binding];
 
-                _imageRefs[binding] = new(stage, view.Storage, view.GetView(imageFormat).GetIdentityImageView());
+                iRef.View?.ClearUsage(FeedbackLoopHazards);
+                view?.PrepareForUsage(cbs, stage.ConvertToPipelineStageFlags(), FeedbackLoopHazards);
+
+                iRef = new(stage, view, view.GetView(imageFormat).GetIdentityImageView());
             }
             else
             {
@@ -486,9 +509,12 @@ namespace Ryujinx.Graphics.Vulkan
             }
             else if (texture is TextureView view)
             {
-                view.Storage.QueueWriteToReadBarrier(cbs, AccessFlags.ShaderReadBit, stage.ConvertToPipelineStageFlags());
+                ref TextureRef iRef = ref _textureRefs[binding];
 
-                _textureRefs[binding] = new(stage, view.Storage, view.GetImageView(), ((SamplerHolder)sampler)?.GetSampler());
+                iRef.View?.ClearUsage(FeedbackLoopHazards);
+                view?.PrepareForUsage(cbs, stage.ConvertToPipelineStageFlags(), FeedbackLoopHazards);
+
+                iRef = new(stage, view, view.GetImageView(), ((SamplerHolder)sampler)?.GetSampler());
             }
             else
             {
@@ -510,7 +536,7 @@ namespace Ryujinx.Graphics.Vulkan
             {
                 view.Storage.QueueWriteToReadBarrier(cbs, AccessFlags.ShaderReadBit, stage.ConvertToPipelineStageFlags());
 
-                _textureRefs[binding] = new(stage, view.Storage, view.GetIdentityImageView(), ((SamplerHolder)sampler)?.GetSampler());
+                _textureRefs[binding] = new(stage, view, view.GetIdentityImageView(), ((SamplerHolder)sampler)?.GetSampler());
 
                 SignalDirty(DirtyFlags.Texture);
             }
@@ -836,7 +862,7 @@ namespace Ryujinx.Graphics.Vulkan
                                 ref var texture = ref textures[i];
                                 ref var refs = ref _textureRefs[binding + i];
 
-                                texture.ImageView = refs.View?.Get(cbs).Value ?? default;
+                                texture.ImageView = refs.ImageView?.Get(cbs).Value ?? default;
                                 texture.Sampler = refs.Sampler?.Get(cbs).Value ?? default;
 
                                 if (texture.ImageView.Handle == 0)
@@ -886,7 +912,7 @@ namespace Ryujinx.Graphics.Vulkan
 
                             for (int i = 0; i < count; i++)
                             {
-                                images[i].ImageView = _imageRefs[binding + i].View?.Get(cbs).Value ?? default;
+                                images[i].ImageView = _imageRefs[binding + i].ImageView?.Get(cbs).Value ?? default;
                             }
 
                             tu.Push<DescriptorImageInfo>(images[..count]);
@@ -957,7 +983,7 @@ namespace Ryujinx.Graphics.Vulkan
                             ref var texture = ref textures[i];
                             ref var refs = ref _textureRefs[binding + i];
 
-                            texture.ImageView = refs.View?.Get(cbs).Value ?? default;
+                            texture.ImageView = refs.ImageView?.Get(cbs).Value ?? default;
                             texture.Sampler = refs.Sampler?.Get(cbs).Value ?? default;
 
                             if (texture.ImageView.Handle == 0)

src/Ryujinx.Graphics.Vulkan/FeedbackLoopAspects.cs

@@ -0,0 +1,12 @@
+using System;
+
+namespace Ryujinx.Graphics.Vulkan
+{
+    [Flags]
+    internal enum FeedbackLoopAspects
+    {
+        None = 0,
+        Color = 1 << 0,
+        Depth = 1 << 1,
+    }
+}

src/Ryujinx.Graphics.Vulkan/FramebufferParams.cs

@@ -250,7 +250,7 @@ namespace Ryujinx.Graphics.Vulkan
                 Layers = Layers,
             };
 
-            api.CreateFramebuffer(_device, framebufferCreateInfo, null, out var framebuffer).ThrowOnError();
+            api.CreateFramebuffer(_device, in framebufferCreateInfo, null, out var framebuffer).ThrowOnError();
             return new Auto<DisposableFramebuffer>(new DisposableFramebuffer(api, _device, framebuffer), null, _attachments);
         }
 
@@ -302,6 +302,27 @@ namespace Ryujinx.Graphics.Vulkan
             _depthStencil?.Storage?.AddStoreOpUsage(true);
         }
 
+        public void ClearBindings()
+        {
+            _depthStencil?.Storage.ClearBindings();
+
+            for (int i = 0; i < _colorsCanonical.Length; i++)
+            {
+                _colorsCanonical[i]?.Storage.ClearBindings();
+            }
+        }
+
+        public void AddBindings()
+        {
+            _depthStencil?.Storage.AddBinding(_depthStencil);
+
+            for (int i = 0; i < _colorsCanonical.Length; i++)
+            {
+                TextureView color = _colorsCanonical[i];
+                color?.Storage.AddBinding(color);
+            }
+        }
+
         public (RenderPassHolder rpHolder, Auto<DisposableFramebuffer> framebuffer) GetPassAndFramebuffer(
             VulkanRenderer gd,
             Device device,

src/Ryujinx.Graphics.Vulkan/HardwareCapabilities.cs

@@ -46,6 +46,8 @@ namespace Ryujinx.Graphics.Vulkan
         public readonly bool SupportsViewportArray2;
         public readonly bool SupportsHostImportedMemory;
         public readonly bool SupportsDepthClipControl;
+        public readonly bool SupportsAttachmentFeedbackLoop;
+        public readonly bool SupportsDynamicAttachmentFeedbackLoop;
         public readonly uint SubgroupSize;
         public readonly SampleCountFlags SupportedSampleCounts;
         public readonly PortabilitySubsetFlags PortabilitySubset;
@@ -84,6 +86,8 @@ namespace Ryujinx.Graphics.Vulkan
             bool supportsViewportArray2,
             bool supportsHostImportedMemory,
             bool supportsDepthClipControl,
+            bool supportsAttachmentFeedbackLoop,
+            bool supportsDynamicAttachmentFeedbackLoop,
             uint subgroupSize,
             SampleCountFlags supportedSampleCounts,
             PortabilitySubsetFlags portabilitySubset,
@@ -121,6 +125,8 @@ namespace Ryujinx.Graphics.Vulkan
             SupportsViewportArray2 = supportsViewportArray2;
             SupportsHostImportedMemory = supportsHostImportedMemory;
             SupportsDepthClipControl = supportsDepthClipControl;
+            SupportsAttachmentFeedbackLoop = supportsAttachmentFeedbackLoop;
+            SupportsDynamicAttachmentFeedbackLoop = supportsDynamicAttachmentFeedbackLoop;
             SubgroupSize = subgroupSize;
             SupportedSampleCounts = supportedSampleCounts;
             PortabilitySubset = portabilitySubset;

src/Ryujinx.Graphics.Vulkan/HostMemoryAllocator.cs

@@ -115,7 +115,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PNext = &importInfo,
                 };
 
-                Result result = _api.AllocateMemory(_device, memoryAllocateInfo, null, out var deviceMemory);
+                Result result = _api.AllocateMemory(_device, in memoryAllocateInfo, null, out var deviceMemory);
 
                 if (result < Result.Success)
                 {

src/Ryujinx.Graphics.Vulkan/MemoryAllocatorBlockList.cs

@@ -220,7 +220,7 @@ namespace Ryujinx.Graphics.Vulkan
                 MemoryTypeIndex = (uint)MemoryTypeIndex,
             };
 
-            _api.AllocateMemory(_device, memoryAllocateInfo, null, out var deviceMemory).ThrowOnError();
+            _api.AllocateMemory(_device, in memoryAllocateInfo, null, out var deviceMemory).ThrowOnError();
 
             IntPtr hostPointer = IntPtr.Zero;
 

src/Ryujinx.Graphics.Vulkan/MoltenVK/MVKInitialization.cs

@@ -1,3 +1,4 @@
+using Silk.NET.Core.Loader;
 using Silk.NET.Vulkan;
 using System;
 using System.Runtime.InteropServices;
@@ -8,6 +9,8 @@ namespace Ryujinx.Graphics.Vulkan.MoltenVK
     [SupportedOSPlatform("macos")]
     public static partial class MVKInitialization
     {
+        private const string VulkanLib = "libvulkan.dylib";
+
         [LibraryImport("libMoltenVK.dylib")]
         private static partial Result vkGetMoltenVKConfigurationMVK(IntPtr unusedInstance, out MVKConfiguration config, in IntPtr configSize);
 
@@ -29,5 +32,20 @@ namespace Ryujinx.Graphics.Vulkan.MoltenVK
 
             vkSetMoltenVKConfigurationMVK(IntPtr.Zero, config, configSize);
         }
+
+        private static string[] Resolver(string path)
+        {
+            if (path.EndsWith(VulkanLib))
+            {
+                path = path[..^VulkanLib.Length] + "libMoltenVK.dylib";
+                return [path];
+            }
+            return Array.Empty<string>();
+        }
+
+        public static void InitializeResolver()
+        {
+            ((DefaultPathResolver)PathResolver.Default).Resolvers.Insert(0, Resolver);
+        }
     }
 }

src/Ryujinx.Graphics.Vulkan/PipelineBase.cs

@@ -2,6 +2,7 @@ using Ryujinx.Graphics.GAL;
 using Ryujinx.Graphics.Shader;
 using Silk.NET.Vulkan;
 using System;
+using System.Collections.Generic;
 using System.Linq;
 using System.Numerics;
 using System.Runtime.CompilerServices;
@@ -33,6 +34,7 @@ namespace Ryujinx.Graphics.Vulkan
         public readonly Action EndRenderPassDelegate;
 
         protected PipelineDynamicState DynamicState;
+        protected bool IsMainPipeline;
         private PipelineState _newState;
         private bool _graphicsStateDirty;
         private bool _computeStateDirty;
@@ -85,6 +87,9 @@ namespace Ryujinx.Graphics.Vulkan
         private bool _tfEnabled;
         private bool _tfActive;
 
+        private FeedbackLoopAspects _feedbackLoop;
+        private bool _passWritesDepthStencil;
+
         private readonly PipelineColorBlendAttachmentState[] _storedBlend;
         public ulong DrawCount { get; private set; }
         public bool RenderPassActive { get; private set; }
@@ -102,7 +107,7 @@ namespace Ryujinx.Graphics.Vulkan
                 SType = StructureType.PipelineCacheCreateInfo,
             };
 
-            gd.Api.CreatePipelineCache(device, pipelineCacheCreateInfo, null, out PipelineCache).ThrowOnError();
+            gd.Api.CreatePipelineCache(device, in pipelineCacheCreateInfo, null, out PipelineCache).ThrowOnError();
 
             _descriptorSetUpdater = new DescriptorSetUpdater(gd, device);
             _vertexBufferUpdater = new VertexBufferUpdater(gd);
@@ -126,7 +131,7 @@ namespace Ryujinx.Graphics.Vulkan
 
         public void Initialize()
         {
-            _descriptorSetUpdater.Initialize();
+            _descriptorSetUpdater.Initialize(IsMainPipeline);
 
             QuadsToTrisPattern = new IndexBufferPattern(Gd, 4, 6, 0, new[] { 0, 1, 2, 0, 2, 3 }, 4, false);
             TriFanToTrisPattern = new IndexBufferPattern(Gd, 3, 3, 2, new[] { int.MinValue, -1, 0 }, 1, true);
@@ -814,6 +819,8 @@ namespace Ryujinx.Graphics.Vulkan
             _newState.DepthTestEnable = depthTest.TestEnable;
             _newState.DepthWriteEnable = depthTest.WriteEnable;
             _newState.DepthCompareOp = depthTest.Func.Convert();
+
+            UpdatePassDepthStencil();
             SignalStateChange();
         }
 
@@ -1079,6 +1086,8 @@ namespace Ryujinx.Graphics.Vulkan
             _newState.StencilFrontPassOp = stencilTest.FrontDpPass.Convert();
             _newState.StencilFrontDepthFailOp = stencilTest.FrontDpFail.Convert();
             _newState.StencilFrontCompareOp = stencilTest.FrontFunc.Convert();
+
+            UpdatePassDepthStencil();
             SignalStateChange();
         }
 
@@ -1426,7 +1435,23 @@ namespace Ryujinx.Graphics.Vulkan
                 }
             }
 
+            if (IsMainPipeline)
+            {
+                FramebufferParams?.ClearBindings();
+            }
+
             FramebufferParams = new FramebufferParams(Device, colors, depthStencil);
+
+            if (IsMainPipeline)
+            {
+                FramebufferParams.AddBindings();
+
+                _newState.FeedbackLoopAspects = FeedbackLoopAspects.None;
+                _bindingBarriersDirty = true;
+            }
+
+            _passWritesDepthStencil = false;
+            UpdatePassDepthStencil();
             UpdatePipelineAttachmentFormats();
         }
 
@@ -1493,11 +1518,82 @@ namespace Ryujinx.Graphics.Vulkan
                 }
             }
 
-            Gd.Barriers.Flush(Cbs, _program, RenderPassActive, _rpHolder, EndRenderPassDelegate);
+            Gd.Barriers.Flush(Cbs, _program, _feedbackLoop != 0, RenderPassActive, _rpHolder, EndRenderPassDelegate);
 
             _descriptorSetUpdater.UpdateAndBindDescriptorSets(Cbs, PipelineBindPoint.Compute);
         }
 
+        private bool ChangeFeedbackLoop(FeedbackLoopAspects aspects)
+        {
+            if (_feedbackLoop != aspects)
+            {
+                if (Gd.Capabilities.SupportsDynamicAttachmentFeedbackLoop)
+                {
+                    DynamicState.SetFeedbackLoop(aspects);
+                }
+                else
+                {
+                    _newState.FeedbackLoopAspects = aspects;
+                }
+
+                _feedbackLoop = aspects;
+
+                return true;
+            }
+
+            return false;
+        }
+
+        [MethodImpl(MethodImplOptions.AggressiveInlining)]
+        private bool UpdateFeedbackLoop()
+        {
+            List<TextureView> hazards = _descriptorSetUpdater.FeedbackLoopHazards;
+
+            if ((hazards?.Count ?? 0) > 0)
+            {
+                FeedbackLoopAspects aspects = 0;
+
+                foreach (TextureView view in hazards)
+                {
+                    // May need to enforce feedback loop layout here in the future.
+                    // Though technically, it should always work with the general layout.
+
+                    if (view.Info.Format.IsDepthOrStencil())
+                    {
+                        if (_passWritesDepthStencil)
+                        {
+                            // If depth/stencil isn't written in the pass, it doesn't count as a feedback loop.
+
+                            aspects |= FeedbackLoopAspects.Depth;
+                        }
+                    }
+                    else
+                    {
+                        aspects |= FeedbackLoopAspects.Color;
+                    }
+                }
+
+                return ChangeFeedbackLoop(aspects);
+            }
+            else if (_feedbackLoop != 0)
+            {
+                return ChangeFeedbackLoop(FeedbackLoopAspects.None);
+            }
+
+            return false;
+        }
+
+        private void UpdatePassDepthStencil()
+        {
+            if (!RenderPassActive)
+            {
+                _passWritesDepthStencil = false;
+            }
+
+            // Stencil test being enabled doesn't necessarily mean a write, but it's not critical to check.
+            _passWritesDepthStencil |= (_newState.DepthTestEnable && _newState.DepthWriteEnable) || _newState.StencilTestEnable;
+        }
+
         private bool RecreateGraphicsPipelineIfNeeded()
         {
             if (AutoFlush.ShouldFlushDraw(DrawCount))
@@ -1505,7 +1601,7 @@ namespace Ryujinx.Graphics.Vulkan
                 Gd.FlushAllCommands();
             }
 
-            DynamicState.ReplayIfDirty(Gd.Api, CommandBuffer);
+            DynamicState.ReplayIfDirty(Gd, CommandBuffer);
 
             if (_needsIndexBufferRebind && _indexBufferPattern == null)
             {
@@ -1539,7 +1635,15 @@ namespace Ryujinx.Graphics.Vulkan
                 _vertexBufferUpdater.Commit(Cbs);
             }
 
-            if (_graphicsStateDirty || Pbp != PipelineBindPoint.Graphics)
+            if (_bindingBarriersDirty)
+            {
+                // Stale barriers may have been activated by switching program. Emit any that are relevant.
+                _descriptorSetUpdater.InsertBindingBarriers(Cbs);
+
+                _bindingBarriersDirty = false;
+            }
+
+            if (UpdateFeedbackLoop() || _graphicsStateDirty || Pbp != PipelineBindPoint.Graphics)
             {
                 if (!CreatePipeline(PipelineBindPoint.Graphics))
                 {
@@ -1548,17 +1652,9 @@ namespace Ryujinx.Graphics.Vulkan
 
                 _graphicsStateDirty = false;
                 Pbp = PipelineBindPoint.Graphics;
-
-                if (_bindingBarriersDirty)
-                {
-                    // Stale barriers may have been activated by switching program. Emit any that are relevant.
-                    _descriptorSetUpdater.InsertBindingBarriers(Cbs);
-
-                    _bindingBarriersDirty = false;
-                }
             }
 
-            Gd.Barriers.Flush(Cbs, _program, RenderPassActive, _rpHolder, EndRenderPassDelegate);
+            Gd.Barriers.Flush(Cbs, _program, _feedbackLoop != 0, RenderPassActive, _rpHolder, EndRenderPassDelegate);
 
             _descriptorSetUpdater.UpdateAndBindDescriptorSets(Cbs, PipelineBindPoint.Graphics);
 
@@ -1628,7 +1724,7 @@ namespace Ryujinx.Graphics.Vulkan
                     ClearValueCount = 1,
                 };
 
-                Gd.Api.CmdBeginRenderPass(CommandBuffer, renderPassBeginInfo, SubpassContents.Inline);
+                Gd.Api.CmdBeginRenderPass(CommandBuffer, in renderPassBeginInfo, SubpassContents.Inline);
                 RenderPassActive = true;
             }
         }

src/Ryujinx.Graphics.Vulkan/PipelineConverter.cs

@@ -116,7 +116,7 @@ namespace Ryujinx.Graphics.Vulkan
                     DependencyCount = 1,
                 };
 
-                gd.Api.CreateRenderPass(device, renderPassCreateInfo, null, out var renderPass).ThrowOnError();
+                gd.Api.CreateRenderPass(device, in renderPassCreateInfo, null, out var renderPass).ThrowOnError();
 
                 return new DisposableRenderPass(gd.Api, device, renderPass);
             }

src/Ryujinx.Graphics.Vulkan/PipelineDynamicState.cs

@@ -1,5 +1,6 @@
 using Ryujinx.Common.Memory;
 using Silk.NET.Vulkan;
+using Silk.NET.Vulkan.Extensions.EXT;
 
 namespace Ryujinx.Graphics.Vulkan
 {
@@ -21,6 +22,8 @@ namespace Ryujinx.Graphics.Vulkan
 
         private Array4<float> _blendConstants;
 
+        private FeedbackLoopAspects _feedbackLoopAspects;
+
         public uint ViewportsCount;
         public Array16<Viewport> Viewports;
 
@@ -32,7 +35,8 @@ namespace Ryujinx.Graphics.Vulkan
             Scissor = 1 << 2,
             Stencil = 1 << 3,
             Viewport = 1 << 4,
-            All = Blend | DepthBias | Scissor | Stencil | Viewport,
+            FeedbackLoop = 1 << 5,
+            All = Blend | DepthBias | Scissor | Stencil | Viewport | FeedbackLoop,
         }
 
         private DirtyFlags _dirty;
@@ -99,13 +103,22 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
+        public void SetFeedbackLoop(FeedbackLoopAspects aspects)
+        {
+            _feedbackLoopAspects = aspects;
+
+            _dirty |= DirtyFlags.FeedbackLoop;
+        }
+
         public void ForceAllDirty()
         {
             _dirty = DirtyFlags.All;
         }
 
-        public void ReplayIfDirty(Vk api, CommandBuffer commandBuffer)
+        public void ReplayIfDirty(VulkanRenderer gd, CommandBuffer commandBuffer)
         {
+            Vk api = gd.Api;
+
             if (_dirty.HasFlag(DirtyFlags.Blend))
             {
                 RecordBlend(api, commandBuffer);
@@ -131,6 +144,11 @@ namespace Ryujinx.Graphics.Vulkan
                 RecordViewport(api, commandBuffer);
             }
 
+            if (_dirty.HasFlag(DirtyFlags.FeedbackLoop) && gd.Capabilities.SupportsDynamicAttachmentFeedbackLoop)
+            {
+                RecordFeedbackLoop(gd.DynamicFeedbackLoopApi, commandBuffer);
+            }
+
             _dirty = DirtyFlags.None;
         }
 
@@ -169,5 +187,17 @@ namespace Ryujinx.Graphics.Vulkan
                 api.CmdSetViewport(commandBuffer, 0, ViewportsCount, Viewports.AsSpan());
             }
         }
+
+        private readonly void RecordFeedbackLoop(ExtAttachmentFeedbackLoopDynamicState api, CommandBuffer commandBuffer)
+        {
+            ImageAspectFlags aspects = (_feedbackLoopAspects & FeedbackLoopAspects.Color) != 0 ? ImageAspectFlags.ColorBit : 0;
+
+            if ((_feedbackLoopAspects & FeedbackLoopAspects.Depth) != 0)
+            {
+                aspects |= ImageAspectFlags.DepthBit | ImageAspectFlags.StencilBit;
+            }
+
+            api.CmdSetAttachmentFeedbackLoopEnable(commandBuffer, aspects);
+        }
     }
 }

src/Ryujinx.Graphics.Vulkan/PipelineFull.cs

@@ -28,6 +28,8 @@ namespace Ryujinx.Graphics.Vulkan
             _activeBufferMirrors = new();
 
             CommandBuffer = (Cbs = gd.CommandBufferPool.Rent()).CommandBuffer;
+
+            IsMainPipeline = true;
         }
 
         private void CopyPendingQuery()
@@ -235,7 +237,7 @@ namespace Ryujinx.Graphics.Vulkan
 
             if (Pipeline != null && Pbp == PipelineBindPoint.Graphics)
             {
-                DynamicState.ReplayIfDirty(Gd.Api, CommandBuffer);
+                DynamicState.ReplayIfDirty(Gd, CommandBuffer);
             }
         }
 

src/Ryujinx.Graphics.Vulkan/PipelineLayoutFactory.cs

@@ -91,7 +91,7 @@ namespace Ryujinx.Graphics.Vulkan
                         Flags = flags,
                     };
 
-                    gd.Api.CreateDescriptorSetLayout(device, descriptorSetLayoutCreateInfo, null, out layouts[setIndex]).ThrowOnError();
+                    gd.Api.CreateDescriptorSetLayout(device, in descriptorSetLayoutCreateInfo, null, out layouts[setIndex]).ThrowOnError();
                 }
             }
 

src/Ryujinx.Graphics.Vulkan/PipelineState.cs

@@ -8,6 +8,7 @@ namespace Ryujinx.Graphics.Vulkan
     struct PipelineState : IDisposable
     {
         private const int RequiredSubgroupSize = 32;
+        private const int MaxDynamicStatesCount = 9;
 
         public PipelineUid Internal;
 
@@ -299,6 +300,12 @@ namespace Ryujinx.Graphics.Vulkan
             set => Internal.Id8 = (Internal.Id8 & 0xFFFFFFFFFFFFFFBF) | ((value ? 1UL : 0UL) << 6);
         }
 
+        public FeedbackLoopAspects FeedbackLoopAspects
+        {
+            readonly get => (FeedbackLoopAspects)((Internal.Id8 >> 7) & 0x3);
+            set => Internal.Id8 = (Internal.Id8 & 0xFFFFFFFFFFFFFE7F) | (((ulong)value) << 7);
+        }
+
         public bool HasTessellationControlShader;
         public NativeArray<PipelineShaderStageCreateInfo> Stages;
         public PipelineLayout PipelineLayout;
@@ -564,9 +571,11 @@ namespace Ryujinx.Graphics.Vulkan
                 }
 
                 bool supportsExtDynamicState = gd.Capabilities.SupportsExtendedDynamicState;
-                int dynamicStatesCount = supportsExtDynamicState ? 8 : 7;
+                bool supportsFeedbackLoopDynamicState = gd.Capabilities.SupportsDynamicAttachmentFeedbackLoop;
 
-                DynamicState* dynamicStates = stackalloc DynamicState[dynamicStatesCount];
+                DynamicState* dynamicStates = stackalloc DynamicState[MaxDynamicStatesCount];
+
+                int dynamicStatesCount = 7;
 
                 dynamicStates[0] = DynamicState.Viewport;
                 dynamicStates[1] = DynamicState.Scissor;
@@ -578,7 +587,12 @@ namespace Ryujinx.Graphics.Vulkan
 
                 if (supportsExtDynamicState)
                 {
-                    dynamicStates[7] = DynamicState.VertexInputBindingStrideExt;
+                    dynamicStates[dynamicStatesCount++] = DynamicState.VertexInputBindingStrideExt;
+                }
+
+                if (supportsFeedbackLoopDynamicState)
+                {
+                    dynamicStates[dynamicStatesCount++] = DynamicState.AttachmentFeedbackLoopEnableExt;
                 }
 
                 var pipelineDynamicStateCreateInfo = new PipelineDynamicStateCreateInfo
@@ -588,9 +602,27 @@ namespace Ryujinx.Graphics.Vulkan
                     PDynamicStates = dynamicStates,
                 };
 
+                PipelineCreateFlags flags = 0;
+
+                if (gd.Capabilities.SupportsAttachmentFeedbackLoop)
+                {
+                    FeedbackLoopAspects aspects = FeedbackLoopAspects;
+
+                    if ((aspects & FeedbackLoopAspects.Color) != 0)
+                    {
+                        flags |= PipelineCreateFlags.CreateColorAttachmentFeedbackLoopBitExt;
+                    }
+
+                    if ((aspects & FeedbackLoopAspects.Depth) != 0)
+                    {
+                        flags |= PipelineCreateFlags.CreateDepthStencilAttachmentFeedbackLoopBitExt;
+                    }
+                }
+
                 var pipelineCreateInfo = new GraphicsPipelineCreateInfo
                 {
                     SType = StructureType.GraphicsPipelineCreateInfo,
+                    Flags = flags,
                     StageCount = StagesCount,
                     PStages = Stages.Pointer,
                     PVertexInputState = &vertexInputState,

src/Ryujinx.Graphics.Vulkan/Queries/BufferedQuery.cs

@@ -52,7 +52,7 @@ namespace Ryujinx.Graphics.Vulkan.Queries
                     PipelineStatistics = flags,
                 };
 
-                gd.Api.CreateQueryPool(device, queryPoolCreateInfo, null, out _queryPool).ThrowOnError();
+                gd.Api.CreateQueryPool(device, in queryPoolCreateInfo, null, out _queryPool).ThrowOnError();
             }
 
             var buffer = gd.BufferManager.Create(gd, sizeof(long), forConditionalRendering: true);

src/Ryujinx.Graphics.Vulkan/RenderPassHolder.cs

@@ -125,7 +125,7 @@ namespace Ryujinx.Graphics.Vulkan
                     DependencyCount = 1,
                 };
 
-                gd.Api.CreateRenderPass(device, renderPassCreateInfo, null, out var renderPass).ThrowOnError();
+                gd.Api.CreateRenderPass(device, in renderPassCreateInfo, null, out var renderPass).ThrowOnError();
 
                 _renderPass = new Auto<DisposableRenderPass>(new DisposableRenderPass(gd.Api, device, renderPass));
             }

src/Ryujinx.Graphics.Vulkan/SamplerHolder.cs

@@ -68,7 +68,7 @@ namespace Ryujinx.Graphics.Vulkan
                 samplerCreateInfo.BorderColor = BorderColor.FloatCustomExt;
             }
 
-            gd.Api.CreateSampler(device, samplerCreateInfo, null, out var sampler).ThrowOnError();
+            gd.Api.CreateSampler(device, in samplerCreateInfo, null, out var sampler).ThrowOnError();
 
             _sampler = new Auto<DisposableSampler>(new DisposableSampler(gd.Api, device, sampler));
         }

src/Ryujinx.Graphics.Vulkan/Shader.cs

@@ -64,7 +64,7 @@ namespace Ryujinx.Graphics.Vulkan
                         PCode = (uint*)pCode,
                     };
 
-                    api.CreateShaderModule(device, shaderModuleCreateInfo, null, out _module).ThrowOnError();
+                    api.CreateShaderModule(device, in shaderModuleCreateInfo, null, out _module).ThrowOnError();
                 }
 
                 CompileStatus = ProgramLinkStatus.Success;

src/Ryujinx.Graphics.Vulkan/TextureCopy.cs

@@ -88,7 +88,7 @@ namespace Ryujinx.Graphics.Vulkan
                     DstOffsets = dstOffsets,
                 };
 
-                api.CmdBlitImage(commandBuffer, srcImage, ImageLayout.General, dstImage, ImageLayout.General, 1, region, filter);
+                api.CmdBlitImage(commandBuffer, srcImage, ImageLayout.General, dstImage, ImageLayout.General, 1, in region, filter);
 
                 copySrcLevel++;
                 copyDstLevel++;
@@ -320,13 +320,13 @@ namespace Ryujinx.Graphics.Vulkan
                 {
                     var region = new ImageResolve(srcSl, new Offset3D(0, 0, srcZ), dstSl, new Offset3D(0, 0, dstZ), extent);
 
-                    api.CmdResolveImage(commandBuffer, srcImage, ImageLayout.General, dstImage, ImageLayout.General, 1, region);
+                    api.CmdResolveImage(commandBuffer, srcImage, ImageLayout.General, dstImage, ImageLayout.General, 1, in region);
                 }
                 else
                 {
                     var region = new ImageCopy(srcSl, new Offset3D(0, 0, srcZ), dstSl, new Offset3D(0, 0, dstZ), extent);
 
-                    api.CmdCopyImage(commandBuffer, srcImage, ImageLayout.General, dstImage, ImageLayout.General, 1, region);
+                    api.CmdCopyImage(commandBuffer, srcImage, ImageLayout.General, dstImage, ImageLayout.General, 1, in region);
                 }
 
                 width = Math.Max(1, width >> 1);
@@ -422,7 +422,7 @@ namespace Ryujinx.Graphics.Vulkan
                     DependencyCount = 1,
                 };
 
-                gd.Api.CreateRenderPass2(device, renderPassCreateInfo, null, out var renderPass).ThrowOnError();
+                gd.Api.CreateRenderPass2(device, in renderPassCreateInfo, null, out var renderPass).ThrowOnError();
 
                 using var rp = new Auto<DisposableRenderPass>(new DisposableRenderPass(gd.Api, device, renderPass));
 
@@ -445,7 +445,7 @@ namespace Ryujinx.Graphics.Vulkan
                     Layers = (uint)src.Layers,
                 };
 
-                gd.Api.CreateFramebuffer(device, framebufferCreateInfo, null, out var framebuffer).ThrowOnError();
+                gd.Api.CreateFramebuffer(device, in framebufferCreateInfo, null, out var framebuffer).ThrowOnError();
                 using var fb = new Auto<DisposableFramebuffer>(new DisposableFramebuffer(gd.Api, device, framebuffer), null, srcView, dstView);
 
                 var renderArea = new Rect2D(null, new Extent2D((uint)src.Info.Width, (uint)src.Info.Height));
@@ -465,7 +465,7 @@ namespace Ryujinx.Graphics.Vulkan
                 // to resolve the depth-stencil texture.
                 // TODO: Do speculative resolve and part of the same render pass as the draw to avoid
                 // ending the current render pass?
-                gd.Api.CmdBeginRenderPass(cbs.CommandBuffer, renderPassBeginInfo, SubpassContents.Inline);
+                gd.Api.CmdBeginRenderPass(cbs.CommandBuffer, in renderPassBeginInfo, SubpassContents.Inline);
                 gd.Api.CmdEndRenderPass(cbs.CommandBuffer);
             }
         }

src/Ryujinx.Graphics.Vulkan/TextureStorage.cs

@@ -4,6 +4,7 @@ using Silk.NET.Vulkan;
 using System;
 using System.Collections.Generic;
 using System.Numerics;
+using System.Runtime.CompilerServices;
 using Format = Ryujinx.Graphics.GAL.Format;
 using VkBuffer = Silk.NET.Vulkan.Buffer;
 using VkFormat = Silk.NET.Vulkan.Format;
@@ -12,6 +13,11 @@ namespace Ryujinx.Graphics.Vulkan
 {
     class TextureStorage : IDisposable
     {
+        private struct TextureSliceInfo
+        {
+            public int BindCount;
+        }
+
         private const MemoryPropertyFlags DefaultImageMemoryFlags =
             MemoryPropertyFlags.DeviceLocalBit;
 
@@ -43,6 +49,7 @@ namespace Ryujinx.Graphics.Vulkan
         private readonly Image _image;
         private readonly Auto<DisposableImage> _imageAuto;
         private readonly Auto<MemoryAllocation> _allocationAuto;
+        private readonly int _depthOrLayers;
         private Auto<MemoryAllocation> _foreignAllocationAuto;
 
         private Dictionary<Format, TextureStorage> _aliasedStorages;
@@ -55,6 +62,9 @@ namespace Ryujinx.Graphics.Vulkan
         private int _viewsCount;
         private readonly ulong _size;
 
+        private int _bindCount;
+        private readonly TextureSliceInfo[] _slices;
+
         public VkFormat VkFormat { get; }
 
         public unsafe TextureStorage(
@@ -73,6 +83,7 @@ namespace Ryujinx.Graphics.Vulkan
             var depth = (uint)(info.Target == Target.Texture3D ? info.Depth : 1);
 
             VkFormat = format;
+            _depthOrLayers = info.GetDepthOrLayers();
 
             var type = info.Target.Convert();
 
@@ -80,7 +91,7 @@ namespace Ryujinx.Graphics.Vulkan
 
             var sampleCountFlags = ConvertToSampleCountFlags(gd.Capabilities.SupportedSampleCounts, (uint)info.Samples);
 
-            var usage = GetImageUsage(info.Format, info.Target, gd.Capabilities.SupportsShaderStorageImageMultisample);
+            var usage = GetImageUsage(info.Format, info.Target, gd.Capabilities);
 
             var flags = ImageCreateFlags.CreateMutableFormatBit | ImageCreateFlags.CreateExtendedUsageBit;
 
@@ -114,7 +125,7 @@ namespace Ryujinx.Graphics.Vulkan
                 Flags = flags,
             };
 
-            gd.Api.CreateImage(device, imageCreateInfo, null, out _image).ThrowOnError();
+            gd.Api.CreateImage(device, in imageCreateInfo, null, out _image).ThrowOnError();
 
             if (foreignAllocation == null)
             {
@@ -148,6 +159,8 @@ namespace Ryujinx.Graphics.Vulkan
 
                 InitialTransition(ImageLayout.Preinitialized, ImageLayout.General);
             }
+
+            _slices = new TextureSliceInfo[levels * _depthOrLayers];
         }
 
         public TextureStorage CreateAliasedColorForDepthStorageUnsafe(Format format)
@@ -284,7 +297,7 @@ namespace Ryujinx.Graphics.Vulkan
                 0,
                 null,
                 1,
-                barrier);
+                in barrier);
 
             if (useTempCbs)
             {
@@ -292,7 +305,7 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
-        public static ImageUsageFlags GetImageUsage(Format format, Target target, bool supportsMsStorage)
+        public static ImageUsageFlags GetImageUsage(Format format, Target target, in HardwareCapabilities capabilities)
         {
             var usage = DefaultUsageFlags;
 
@@ -305,11 +318,19 @@ namespace Ryujinx.Graphics.Vulkan
                 usage |= ImageUsageFlags.ColorAttachmentBit;
             }
 
+            bool supportsMsStorage = capabilities.SupportsShaderStorageImageMultisample;
+
             if (format.IsImageCompatible() && (supportsMsStorage || !target.IsMultisample()))
             {
                 usage |= ImageUsageFlags.StorageBit;
             }
 
+            if (capabilities.SupportsAttachmentFeedbackLoop &&
+                (usage & (ImageUsageFlags.DepthStencilAttachmentBit | ImageUsageFlags.ColorAttachmentBit)) != 0)
+            {
+                usage |= ImageUsageFlags.AttachmentFeedbackLoopBitExt;
+            }
+
             return usage;
         }
 
@@ -401,11 +422,11 @@ namespace Ryujinx.Graphics.Vulkan
 
                 if (to)
                 {
-                    _gd.Api.CmdCopyImageToBuffer(commandBuffer, image, ImageLayout.General, buffer, 1, region);
+                    _gd.Api.CmdCopyImageToBuffer(commandBuffer, image, ImageLayout.General, buffer, 1, in region);
                 }
                 else
                 {
-                    _gd.Api.CmdCopyBufferToImage(commandBuffer, buffer, image, ImageLayout.General, 1, region);
+                    _gd.Api.CmdCopyBufferToImage(commandBuffer, buffer, image, ImageLayout.General, 1, in region);
                 }
 
                 offset += mipSize;
@@ -510,6 +531,55 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
+        public void AddBinding(TextureView view)
+        {
+            // Assumes a view only has a first level.
+
+            int index = view.FirstLevel * _depthOrLayers + view.FirstLayer;
+            int layers = view.Layers;
+
+            for (int i = 0; i < layers; i++)
+            {
+                ref TextureSliceInfo info = ref _slices[index++];
+
+                info.BindCount++;
+            }
+
+            _bindCount++;
+        }
+
+        public void ClearBindings()
+        {
+            if (_bindCount != 0)
+            {
+                Array.Clear(_slices, 0, _slices.Length);
+
+                _bindCount = 0;
+            }
+        }
+
+        [MethodImpl(MethodImplOptions.AggressiveInlining)]
+        public bool IsBound(TextureView view)
+        {
+            if (_bindCount != 0)
+            {
+                int index = view.FirstLevel * _depthOrLayers + view.FirstLayer;
+                int layers = view.Layers;
+
+                for (int i = 0; i < layers; i++)
+                {
+                    ref TextureSliceInfo info = ref _slices[index++];
+
+                    if (info.BindCount != 0)
+                    {
+                        return true;
+                    }
+                }
+            }
+
+            return false;
+        }
+
         public void IncrementViewsCount()
         {
             _viewsCount++;

src/Ryujinx.Graphics.Vulkan/TextureView.cs

@@ -23,6 +23,8 @@ namespace Ryujinx.Graphics.Vulkan
         private readonly Auto<DisposableImageView> _imageView2dArray;
         private Dictionary<Format, TextureView> _selfManagedViews;
 
+        private int _hazardUses;
+
         private readonly TextureCreateInfo _info;
 
         private HashTableSlim<RenderPassCacheKey, RenderPassHolder> _renderPasses;
@@ -60,7 +62,7 @@ namespace Ryujinx.Graphics.Vulkan
             gd.Textures.Add(this);
 
             var format = _gd.FormatCapabilities.ConvertToVkFormat(info.Format);
-            var usage = TextureStorage.GetImageUsage(info.Format, info.Target, gd.Capabilities.SupportsShaderStorageImageMultisample);
+            var usage = TextureStorage.GetImageUsage(info.Format, info.Target, gd.Capabilities);
             var levels = (uint)info.Levels;
             var layers = (uint)info.GetLayers();
 
@@ -117,7 +119,7 @@ namespace Ryujinx.Graphics.Vulkan
                     PNext = &imageViewUsage,
                 };
 
-                gd.Api.CreateImageView(device, imageCreateInfo, null, out var imageView).ThrowOnError();
+                gd.Api.CreateImageView(device, in imageCreateInfo, null, out var imageView).ThrowOnError();
                 return new Auto<DisposableImageView>(new DisposableImageView(gd.Api, device, imageView), null, storage.GetImage());
             }
 
@@ -492,7 +494,7 @@ namespace Ryujinx.Graphics.Vulkan
                 dstStageMask,
                 DependencyFlags.None,
                 1,
-                memoryBarrier,
+                in memoryBarrier,
                 0,
                 null,
                 0,
@@ -557,7 +559,7 @@ namespace Ryujinx.Graphics.Vulkan
                 0,
                 null,
                 1,
-                memoryBarrier);
+                in memoryBarrier);
         }
 
         public TextureView GetView(Format format)
@@ -949,11 +951,11 @@ namespace Ryujinx.Graphics.Vulkan
 
                 if (to)
                 {
-                    _gd.Api.CmdCopyImageToBuffer(commandBuffer, image, ImageLayout.General, buffer, 1, region);
+                    _gd.Api.CmdCopyImageToBuffer(commandBuffer, image, ImageLayout.General, buffer, 1, in region);
                 }
                 else
                 {
-                    _gd.Api.CmdCopyBufferToImage(commandBuffer, buffer, image, ImageLayout.General, 1, region);
+                    _gd.Api.CmdCopyBufferToImage(commandBuffer, buffer, image, ImageLayout.General, 1, in region);
                 }
 
                 offset += mipSize;
@@ -1010,11 +1012,11 @@ namespace Ryujinx.Graphics.Vulkan
 
             if (to)
             {
-                _gd.Api.CmdCopyImageToBuffer(commandBuffer, image, ImageLayout.General, buffer, 1, region);
+                _gd.Api.CmdCopyImageToBuffer(commandBuffer, image, ImageLayout.General, buffer, 1, in region);
             }
             else
             {
-                _gd.Api.CmdCopyBufferToImage(commandBuffer, buffer, image, ImageLayout.General, 1, region);
+                _gd.Api.CmdCopyBufferToImage(commandBuffer, buffer, image, ImageLayout.General, 1, in region);
             }
         }
 
@@ -1034,6 +1036,34 @@ namespace Ryujinx.Graphics.Vulkan
             throw new NotImplementedException();
         }
 
+        public void PrepareForUsage(CommandBufferScoped cbs, PipelineStageFlags flags, List<TextureView> feedbackLoopHazards)
+        {
+            Storage.QueueWriteToReadBarrier(cbs, AccessFlags.ShaderReadBit, flags);
+
+            if (feedbackLoopHazards != null && Storage.IsBound(this))
+            {
+                feedbackLoopHazards.Add(this);
+                _hazardUses++;
+            }
+        }
+
+        public void ClearUsage(List<TextureView> feedbackLoopHazards)
+        {
+            if (_hazardUses != 0 && feedbackLoopHazards != null)
+            {
+                feedbackLoopHazards.Remove(this);
+                _hazardUses--;
+            }
+        }
+
+        public void DecrementHazardUses()
+        {
+            if (_hazardUses != 0)
+            {
+                _hazardUses--;
+            }
+        }
+
         public (RenderPassHolder rpHolder, Auto<DisposableFramebuffer> framebuffer) GetPassAndFramebuffer(
             VulkanRenderer gd,
             Device device,

src/Ryujinx.Graphics.Vulkan/Vendor.cs

@@ -90,11 +90,9 @@ namespace Ryujinx.Graphics.Vulkan
                 DriverId.SamsungProprietary => "Samsung",
                 DriverId.MesaVenus => "Venus",
                 DriverId.MesaDozen => "Dozen",
-
+                DriverId.MesaNvk => "NVK",
-                // TODO: Use real enum when we have an up to date Silk.NET.
+                DriverId.ImaginationOpenSourceMesa => "Imagination (Open)",
-                (DriverId)24 => "NVK",
+                DriverId.MesaAgxv => "Honeykrisp",
-                (DriverId)25 => "Imagination (Open)",
-                (DriverId)26 => "Honeykrisp",
                 _ => id.ToString(),
             };
         }

src/Ryujinx.Graphics.Vulkan/VulkanInitialization.cs

@@ -44,6 +44,8 @@ namespace Ryujinx.Graphics.Vulkan
             "VK_EXT_4444_formats",
             "VK_KHR_8bit_storage",
             "VK_KHR_maintenance2",
+            "VK_EXT_attachment_feedback_loop_layout",
+            "VK_EXT_attachment_feedback_loop_dynamic_state",
         };
 
         private static readonly string[] _requiredExtensions = {
@@ -357,6 +359,28 @@ namespace Ryujinx.Graphics.Vulkan
                 features2.PNext = &supportedFeaturesDepthClipControl;
             }
 
+            PhysicalDeviceAttachmentFeedbackLoopLayoutFeaturesEXT supportedFeaturesAttachmentFeedbackLoopLayout = new()
+            {
+                SType = StructureType.PhysicalDeviceAttachmentFeedbackLoopLayoutFeaturesExt,
+                PNext = features2.PNext,
+            };
+
+            if (physicalDevice.IsDeviceExtensionPresent("VK_EXT_attachment_feedback_loop_layout"))
+            {
+                features2.PNext = &supportedFeaturesAttachmentFeedbackLoopLayout;
+            }
+
+            PhysicalDeviceAttachmentFeedbackLoopDynamicStateFeaturesEXT supportedFeaturesDynamicAttachmentFeedbackLoopLayout = new()
+            {
+                SType = StructureType.PhysicalDeviceAttachmentFeedbackLoopDynamicStateFeaturesExt,
+                PNext = features2.PNext,
+            };
+
+            if (physicalDevice.IsDeviceExtensionPresent("VK_EXT_attachment_feedback_loop_dynamic_state"))
+            {
+                features2.PNext = &supportedFeaturesDynamicAttachmentFeedbackLoopLayout;
+            }
+
             PhysicalDeviceVulkan12Features supportedPhysicalDeviceVulkan12Features = new()
             {
                 SType = StructureType.PhysicalDeviceVulkan12Features,
@@ -531,6 +555,36 @@ namespace Ryujinx.Graphics.Vulkan
                 pExtendedFeatures = &featuresDepthClipControl;
             }
 
+            PhysicalDeviceAttachmentFeedbackLoopLayoutFeaturesEXT featuresAttachmentFeedbackLoopLayout;
+
+            if (physicalDevice.IsDeviceExtensionPresent("VK_EXT_attachment_feedback_loop_layout") &&
+                supportedFeaturesAttachmentFeedbackLoopLayout.AttachmentFeedbackLoopLayout)
+            {
+                featuresAttachmentFeedbackLoopLayout = new()
+                {
+                    SType = StructureType.PhysicalDeviceAttachmentFeedbackLoopLayoutFeaturesExt,
+                    PNext = pExtendedFeatures,
+                    AttachmentFeedbackLoopLayout = true,
+                };
+
+                pExtendedFeatures = &featuresAttachmentFeedbackLoopLayout;
+            }
+
+            PhysicalDeviceAttachmentFeedbackLoopDynamicStateFeaturesEXT featuresDynamicAttachmentFeedbackLoopLayout;
+
+            if (physicalDevice.IsDeviceExtensionPresent("VK_EXT_attachment_feedback_loop_dynamic_state") &&
+                supportedFeaturesDynamicAttachmentFeedbackLoopLayout.AttachmentFeedbackLoopDynamicState)
+            {
+                featuresDynamicAttachmentFeedbackLoopLayout = new()
+                {
+                    SType = StructureType.PhysicalDeviceAttachmentFeedbackLoopDynamicStateFeaturesExt,
+                    PNext = pExtendedFeatures,
+                    AttachmentFeedbackLoopDynamicState = true,
+                };
+
+                pExtendedFeatures = &featuresDynamicAttachmentFeedbackLoopLayout;
+            }
+
             var enabledExtensions = _requiredExtensions.Union(_desirableExtensions.Intersect(physicalDevice.DeviceExtensions)).ToArray();
 
             IntPtr* ppEnabledExtensions = stackalloc IntPtr[enabledExtensions.Length];

src/Ryujinx.Graphics.Vulkan/VulkanRenderer.cs

@@ -38,6 +38,7 @@ namespace Ryujinx.Graphics.Vulkan
         internal KhrPushDescriptor PushDescriptorApi { get; private set; }
         internal ExtTransformFeedback TransformFeedbackApi { get; private set; }
         internal KhrDrawIndirectCount DrawIndirectCountApi { get; private set; }
+        internal ExtAttachmentFeedbackLoopDynamicState DynamicFeedbackLoopApi { get; private set; }
 
         internal uint QueueFamilyIndex { get; private set; }
         internal Queue Queue { get; private set; }
@@ -149,6 +150,11 @@ namespace Ryujinx.Graphics.Vulkan
                 DrawIndirectCountApi = drawIndirectCountApi;
             }
 
+            if (Api.TryGetDeviceExtension(_instance.Instance, _device, out ExtAttachmentFeedbackLoopDynamicState dynamicFeedbackLoopApi))
+            {
+                DynamicFeedbackLoopApi = dynamicFeedbackLoopApi;
+            }
+
             if (maxQueueCount >= 2)
             {
                 Api.GetDeviceQueue(_device, queueFamilyIndex, 1, out var backgroundQueue);
@@ -243,6 +249,16 @@ namespace Ryujinx.Graphics.Vulkan
                 SType = StructureType.PhysicalDeviceDepthClipControlFeaturesExt,
             };
 
+            PhysicalDeviceAttachmentFeedbackLoopLayoutFeaturesEXT featuresAttachmentFeedbackLoop = new()
+            {
+                SType = StructureType.PhysicalDeviceAttachmentFeedbackLoopLayoutFeaturesExt,
+            };
+
+            PhysicalDeviceAttachmentFeedbackLoopDynamicStateFeaturesEXT featuresDynamicAttachmentFeedbackLoop = new()
+            {
+                SType = StructureType.PhysicalDeviceAttachmentFeedbackLoopDynamicStateFeaturesExt,
+            };
+
             PhysicalDevicePortabilitySubsetFeaturesKHR featuresPortabilitySubset = new()
             {
                 SType = StructureType.PhysicalDevicePortabilitySubsetFeaturesKhr,
@@ -279,6 +295,22 @@ namespace Ryujinx.Graphics.Vulkan
                 features2.PNext = &featuresDepthClipControl;
             }
 
+            bool supportsAttachmentFeedbackLoop = _physicalDevice.IsDeviceExtensionPresent("VK_EXT_attachment_feedback_loop_layout");
+
+            if (supportsAttachmentFeedbackLoop)
+            {
+                featuresAttachmentFeedbackLoop.PNext = features2.PNext;
+                features2.PNext = &featuresAttachmentFeedbackLoop;
+            }
+
+            bool supportsDynamicAttachmentFeedbackLoop = _physicalDevice.IsDeviceExtensionPresent("VK_EXT_attachment_feedback_loop_dynamic_state");
+
+            if (supportsDynamicAttachmentFeedbackLoop)
+            {
+                featuresDynamicAttachmentFeedbackLoop.PNext = features2.PNext;
+                features2.PNext = &featuresDynamicAttachmentFeedbackLoop;
+            }
+
             bool usePortability = _physicalDevice.IsDeviceExtensionPresent("VK_KHR_portability_subset");
 
             if (usePortability)
@@ -401,6 +433,8 @@ namespace Ryujinx.Graphics.Vulkan
                 _physicalDevice.IsDeviceExtensionPresent("VK_NV_viewport_array2"),
                 _physicalDevice.IsDeviceExtensionPresent(ExtExternalMemoryHost.ExtensionName),
                 supportsDepthClipControl && featuresDepthClipControl.DepthClipControl,
+                supportsAttachmentFeedbackLoop && featuresAttachmentFeedbackLoop.AttachmentFeedbackLoopLayout,
+                supportsDynamicAttachmentFeedbackLoop && featuresDynamicAttachmentFeedbackLoop.AttachmentFeedbackLoopDynamicState,
                 propertiesSubgroup.SubgroupSize,
                 supportedSampleCounts,
                 portabilityFlags,

src/Ryujinx.Graphics.Vulkan/Window.cs

@@ -160,7 +160,7 @@ namespace Ryujinx.Graphics.Vulkan
                 SwizzleComponent.Blue,
                 SwizzleComponent.Alpha);
 
-            _gd.SwapchainApi.CreateSwapchain(_device, swapchainCreateInfo, null, out _swapchain).ThrowOnError();
+            _gd.SwapchainApi.CreateSwapchain(_device, in swapchainCreateInfo, null, out _swapchain).ThrowOnError();
 
             _gd.SwapchainApi.GetSwapchainImages(_device, _swapchain, &imageCount, null);
 
@@ -187,14 +187,14 @@ namespace Ryujinx.Graphics.Vulkan
 
             for (int i = 0; i < _imageAvailableSemaphores.Length; i++)
             {
-                _gd.Api.CreateSemaphore(_device, semaphoreCreateInfo, null, out _imageAvailableSemaphores[i]).ThrowOnError();
+                _gd.Api.CreateSemaphore(_device, in semaphoreCreateInfo, null, out _imageAvailableSemaphores[i]).ThrowOnError();
             }
 
             _renderFinishedSemaphores = new Semaphore[imageCount];
 
             for (int i = 0; i < _renderFinishedSemaphores.Length; i++)
             {
-                _gd.Api.CreateSemaphore(_device, semaphoreCreateInfo, null, out _renderFinishedSemaphores[i]).ThrowOnError();
+                _gd.Api.CreateSemaphore(_device, in semaphoreCreateInfo, null, out _renderFinishedSemaphores[i]).ThrowOnError();
             }
         }
 
@@ -220,7 +220,7 @@ namespace Ryujinx.Graphics.Vulkan
                 SubresourceRange = subresourceRange,
             };
 
-            _gd.Api.CreateImageView(_device, imageCreateInfo, null, out var imageView).ThrowOnError();
+            _gd.Api.CreateImageView(_device, in imageCreateInfo, null, out var imageView).ThrowOnError();
 
             return new TextureView(_gd, _device, new DisposableImageView(_gd.Api, _device, imageView), info, format);
         }
@@ -479,7 +479,7 @@ namespace Ryujinx.Graphics.Vulkan
 
             lock (_gd.QueueLock)
             {
-                _gd.SwapchainApi.QueuePresent(_gd.Queue, presentInfo);
+                _gd.SwapchainApi.QueuePresent(_gd.Queue, in presentInfo);
             }
         }
 
@@ -611,7 +611,7 @@ namespace Ryujinx.Graphics.Vulkan
                 0,
                 null,
                 1,
-                barrier);
+                in barrier);
         }
 
         private void CaptureFrame(TextureView texture, int x, int y, int width, int height, bool isBgra, bool flipX, bool flipY)

src/Ryujinx.Gtk3/Program.cs

@@ -4,6 +4,8 @@ using Ryujinx.Common.Configuration;
 using Ryujinx.Common.GraphicsDriver;
 using Ryujinx.Common.Logging;
 using Ryujinx.Common.SystemInterop;
+using Ryujinx.Common.Utilities;
+using Ryujinx.Graphics.Vulkan.MoltenVK;
 using Ryujinx.Modules;
 using Ryujinx.SDL2.Common;
 using Ryujinx.UI;
@@ -40,9 +42,6 @@ namespace Ryujinx
         [LibraryImport("user32.dll", SetLastError = true)]
         public static partial int MessageBoxA(IntPtr hWnd, [MarshalAs(UnmanagedType.LPStr)] string text, [MarshalAs(UnmanagedType.LPStr)] string caption, uint type);
 
-        [LibraryImport("libc", SetLastError = true)]
-        private static partial int setenv([MarshalAs(UnmanagedType.LPStr)] string name, [MarshalAs(UnmanagedType.LPStr)] string value, int overwrite);
-
         private const uint MbIconWarning = 0x30;
 
         static Program()
@@ -104,12 +103,13 @@ namespace Ryujinx
                     throw new NotSupportedException("Failed to initialize multi-threading support.");
                 }
 
-                Environment.SetEnvironmentVariable("GDK_BACKEND", "x11");
+                OsUtils.SetEnvironmentVariableNoCaching("GDK_BACKEND", "x11");
-                setenv("GDK_BACKEND", "x11", 1);
             }
 
             if (OperatingSystem.IsMacOS())
             {
+                MVKInitialization.InitializeResolver();
+
                 string baseDirectory = Path.GetDirectoryName(AppDomain.CurrentDomain.BaseDirectory);
                 string resourcesDataDir;
 
@@ -122,19 +122,13 @@ namespace Ryujinx
                     resourcesDataDir = baseDirectory;
                 }
 
-                static void SetEnvironmentVariableNoCaching(string key, string value)
-                {
-                    int res = setenv(key, value, 1);
-                    Debug.Assert(res != -1);
-                }
-
                 // On macOS, GTK3 needs XDG_DATA_DIRS to be set, otherwise it will try searching for "gschemas.compiled" in system directories.
-                SetEnvironmentVariableNoCaching("XDG_DATA_DIRS", Path.Combine(resourcesDataDir, "share"));
+                OsUtils.SetEnvironmentVariableNoCaching("XDG_DATA_DIRS", Path.Combine(resourcesDataDir, "share"));
 
                 // On macOS, GTK3 needs GDK_PIXBUF_MODULE_FILE to be set, otherwise it will try searching for "loaders.cache" in system directories.
-                SetEnvironmentVariableNoCaching("GDK_PIXBUF_MODULE_FILE", Path.Combine(resourcesDataDir, "lib", "gdk-pixbuf-2.0", "2.10.0", "loaders.cache"));
+                OsUtils.SetEnvironmentVariableNoCaching("GDK_PIXBUF_MODULE_FILE", Path.Combine(resourcesDataDir, "lib", "gdk-pixbuf-2.0", "2.10.0", "loaders.cache"));
 
-                SetEnvironmentVariableNoCaching("GTK_IM_MODULE_FILE", Path.Combine(resourcesDataDir, "lib", "gtk-3.0", "3.0.0", "immodules.cache"));
+                OsUtils.SetEnvironmentVariableNoCaching("GTK_IM_MODULE_FILE", Path.Combine(resourcesDataDir, "lib", "gtk-3.0", "3.0.0", "immodules.cache"));
             }
 
             string systemPath = Environment.GetEnvironmentVariable("Path", EnvironmentVariableTarget.Machine);
@@ -230,9 +224,9 @@ namespace Ryujinx
             // Logging system information.
             PrintSystemInfo();
 
-            // Enable OGL multithreading on the driver, when available.
+            // Enable OGL multithreading on the driver, and some other flags.
             BackendThreading threadingMode = ConfigurationState.Instance.Graphics.BackendThreading;
-            DriverUtilities.ToggleOGLThreading(threadingMode == BackendThreading.Off);
+            DriverUtilities.InitDriverConfig(threadingMode == BackendThreading.Off);
 
             // Initialize Gtk.
             Application.Init();

src/Ryujinx.Headless.SDL2/Program.cs

@@ -7,6 +7,7 @@ using Ryujinx.Common.Configuration.Hid;
 using Ryujinx.Common.Configuration.Hid.Controller;
 using Ryujinx.Common.Configuration.Hid.Controller.Motion;
 using Ryujinx.Common.Configuration.Hid.Keyboard;
+using Ryujinx.Common.GraphicsDriver;
 using Ryujinx.Common.Logging;
 using Ryujinx.Common.Logging.Targets;
 using Ryujinx.Common.SystemInterop;
@@ -18,6 +19,7 @@ using Ryujinx.Graphics.Gpu;
 using Ryujinx.Graphics.Gpu.Shader;
 using Ryujinx.Graphics.OpenGL;
 using Ryujinx.Graphics.Vulkan;
+using Ryujinx.Graphics.Vulkan.MoltenVK;
 using Ryujinx.Headless.SDL2.OpenGL;
 using Ryujinx.Headless.SDL2.Vulkan;
 using Ryujinx.HLE;
@@ -88,6 +90,11 @@ namespace Ryujinx.Headless.SDL2
                 };
             }
 
+            if (OperatingSystem.IsMacOS())
+            {
+                MVKInitialization.InitializeResolver();
+            }
+
             Parser.Default.ParseArguments<Options>(args)
             .WithParsed(Load)
             .WithNotParsed(errors => errors.Output());
@@ -457,6 +464,8 @@ namespace Ryujinx.Headless.SDL2
             GraphicsConfig.ShadersDumpPath = option.GraphicsShadersDumpPath;
             GraphicsConfig.EnableMacroHLE = !option.DisableMacroHLE;
 
+            DriverUtilities.InitDriverConfig(option.BackendThreading == BackendThreading.Off);
+
             while (true)
             {
                 LoadApplication(option);

src/Ryujinx.Tests/Cpu/CpuTestAlu32.cs

@@ -29,6 +29,7 @@ namespace Ryujinx.Tests.Cpu
         {
             return new[]
             {
+                0xe6200f10u, // QADD16  R0, R0, R0
                 0xe6600f10u, // UQADD16 R0, R0, R0
                 0xe6600f70u, // UQSUB16 R0, R0, R0
             };

src/Ryujinx.Tests/Cpu/CpuTestSimd32.cs

@@ -328,6 +328,29 @@ namespace Ryujinx.Tests.Cpu
             CompareAgainstUnicorn();
         }
 
+        [Test, Pairwise, Description("VSHLL.<size> {<Vd>}, <Vm>, #<imm>")]
+        public void Vshll([Values(0u, 2u)] uint rd,
+                          [Values(1u, 0u)] uint rm,
+                          [Values(0u, 1u, 2u)] uint size,
+                          [Random(RndCnt)] ulong z,
+                          [Random(RndCnt)] ulong a,
+                          [Random(RndCnt)] ulong b)
+        {
+            uint opcode = 0xf3b20300u; // VSHLL.I8 Q0, D0, #8
+
+            opcode |= ((rm & 0xf) << 0) | ((rm & 0x10) << 1);
+            opcode |= ((rd & 0xf) << 12) | ((rd & 0x10) << 18);
+            opcode |= size << 18;
+
+            V128 v0 = MakeVectorE0E1(z, z);
+            V128 v1 = MakeVectorE0E1(a, z);
+            V128 v2 = MakeVectorE0E1(b, z);
+
+            SingleOpcode(opcode, v0: v0, v1: v1, v2: v2);
+
+            CompareAgainstUnicorn();
+        }
+
         [Test, Pairwise, Description("VSWP D0, D0")]
         public void Vswp([Values(0u, 1u)] uint rd,
                          [Values(0u, 1u)] uint rm,

src/Ryujinx/Program.cs

@@ -7,6 +7,7 @@ using Ryujinx.Common.Configuration;
 using Ryujinx.Common.GraphicsDriver;
 using Ryujinx.Common.Logging;
 using Ryujinx.Common.SystemInterop;
+using Ryujinx.Graphics.Vulkan.MoltenVK;
 using Ryujinx.Modules;
 using Ryujinx.SDL2.Common;
 using Ryujinx.UI.Common;
@@ -80,6 +81,11 @@ namespace Ryujinx.Ava
             // Parse arguments
             CommandLineState.ParseArguments(args);
 
+            if (OperatingSystem.IsMacOS())
+            {
+                MVKInitialization.InitializeResolver();
+            }
+
             // Delete backup files after updating.
             Task.Run(Updater.CleanupUpdate);
 
@@ -111,8 +117,8 @@ namespace Ryujinx.Ava
             // Logging system information.
             PrintSystemInfo();
 
-            // Enable OGL multithreading on the driver, when available.
+            // Enable OGL multithreading on the driver, and some other flags.
-            DriverUtilities.ToggleOGLThreading(ConfigurationState.Instance.Graphics.BackendThreading == BackendThreading.Off);
+            DriverUtilities.InitDriverConfig(ConfigurationState.Instance.Graphics.BackendThreading == BackendThreading.Off);
 
             // Check if keys exists.
             if (!File.Exists(Path.Combine(AppDataManager.KeysDirPath, "prod.keys")))