Implement fast DMA texture to texture copy (#7299)

* Implement fast DMA texture to texture copy

* PR feedback

Directory.Packages.props

@@ -13,14 +13,14 @@
     <PackageVersion Include="CommandLineParser" Version="2.9.1" />
     <PackageVersion Include="Concentus" Version="2.2.0" />
     <PackageVersion Include="DiscordRichPresence" Version="1.2.1.24" />
-    <PackageVersion Include="DynamicData" Version="8.4.1" />
+    <PackageVersion Include="DynamicData" Version="9.0.4" />
     <PackageVersion Include="FluentAvaloniaUI" Version="2.0.5" />
     <PackageVersion Include="GtkSharp.Dependencies" Version="1.1.1" />
     <PackageVersion Include="GtkSharp.Dependencies.osx" Version="0.0.5" />
     <PackageVersion Include="LibHac" Version="0.19.0" />
     <PackageVersion Include="Microsoft.CodeAnalysis.Analyzers" Version="3.3.4" />
     <PackageVersion Include="Microsoft.CodeAnalysis.CSharp" Version="4.9.2" />
-    <PackageVersion Include="Microsoft.IdentityModel.JsonWebTokens" Version="7.6.0" />
+    <PackageVersion Include="Microsoft.IdentityModel.JsonWebTokens" Version="8.0.1" />
     <PackageVersion Include="Microsoft.NET.Test.Sdk" Version="17.9.0" />
     <PackageVersion Include="Microsoft.IO.RecyclableMemoryStream" Version="3.0.1" />
     <PackageVersion Include="MsgPack.Cli" Version="1.0.1" />
@@ -39,11 +39,11 @@
     <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.Extensions.EXT" Version="2.16.0" />
-    <PackageVersion Include="Silk.NET.Vulkan.Extensions.KHR" Version="2.16.0" />
-    <PackageVersion Include="SixLabors.ImageSharp" Version="2.1.8" />
-    <PackageVersion Include="SixLabors.ImageSharp.Drawing" Version="1.0.0" />
+    <PackageVersion Include="Silk.NET.Vulkan" Version="2.21.0" />
+    <PackageVersion Include="Silk.NET.Vulkan.Extensions.EXT" Version="2.21.0" />
+    <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" />
     <PackageVersion Include="System.IO.Hashing" Version="8.0.0" />
     <PackageVersion Include="System.Management" Version="8.0.0" />

Ryujinx.sln

@@ -87,6 +87,8 @@ Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Ryujinx.Horizon", "src\Ryuj
 EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Ryujinx.Horizon.Kernel.Generators", "src\Ryujinx.Horizon.Kernel.Generators\Ryujinx.Horizon.Kernel.Generators.csproj", "{7F55A45D-4E1D-4A36-ADD3-87F29A285AA2}"
 EndProject
+Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Ryujinx.HLE.Generators", "src\Ryujinx.HLE.Generators\Ryujinx.HLE.Generators.csproj", "{B575BCDE-2FD8-4A5D-8756-31CDD7FE81F0}"
+EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution
 		Debug|Any CPU = Debug|Any CPU
@@ -249,6 +251,10 @@ Global
 		{7F55A45D-4E1D-4A36-ADD3-87F29A285AA2}.Debug|Any CPU.Build.0 = Debug|Any CPU
 		{7F55A45D-4E1D-4A36-ADD3-87F29A285AA2}.Release|Any CPU.ActiveCfg = Release|Any CPU
 		{7F55A45D-4E1D-4A36-ADD3-87F29A285AA2}.Release|Any CPU.Build.0 = Release|Any CPU
+		{B575BCDE-2FD8-4A5D-8756-31CDD7FE81F0}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{B575BCDE-2FD8-4A5D-8756-31CDD7FE81F0}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{B575BCDE-2FD8-4A5D-8756-31CDD7FE81F0}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{B575BCDE-2FD8-4A5D-8756-31CDD7FE81F0}.Release|Any CPU.Build.0 = Release|Any CPU
 	EndGlobalSection
 	GlobalSection(SolutionProperties) = preSolution
 		HideSolutionNode = FALSE

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);
@@ -822,6 +823,10 @@ namespace ARMeilleure.Decoders
             SetA32("<<<<00000100xxxxxxxxxxxx1001xxxx", InstName.Umaal,   InstEmit32.Umaal,   OpCode32AluUmull.Create);
             SetA32("<<<<0000101xxxxxxxxxxxxx1001xxxx", InstName.Umlal,   InstEmit32.Umlal,   OpCode32AluUmull.Create);
             SetA32("<<<<0000100xxxxxxxxxxxxx1001xxxx", InstName.Umull,   InstEmit32.Umull,   OpCode32AluUmull.Create);
+            SetA32("<<<<01100110xxxxxxxx11110001xxxx", InstName.Uqadd16, InstEmit32.Uqadd16, OpCode32AluReg.Create);
+            SetA32("<<<<01100110xxxxxxxx11111001xxxx", InstName.Uqadd8,  InstEmit32.Uqadd8,  OpCode32AluReg.Create);
+            SetA32("<<<<01100110xxxxxxxx11110111xxxx", InstName.Uqsub16, InstEmit32.Uqsub16, OpCode32AluReg.Create);
+            SetA32("<<<<01100110xxxxxxxx11111111xxxx", InstName.Uqsub8,  InstEmit32.Uqsub8,  OpCode32AluReg.Create);
             SetA32("<<<<0110111xxxxxxxxxxxxxxx01xxxx", InstName.Usat,    InstEmit32.Usat,    OpCode32Sat.Create);
             SetA32("<<<<01101110xxxxxxxx11110011xxxx", InstName.Usat16,  InstEmit32.Usat16,  OpCode32Sat16.Create);
             SetA32("<<<<01100101xxxxxxxx11111111xxxx", InstName.Usub8,   InstEmit32.Usub8,   OpCode32AluReg.Create);
@@ -1007,6 +1012,8 @@ namespace ARMeilleure.Decoders
             SetAsimd("111100100x10xxxxxxxx1011xxx0xxxx", InstName.Vqdmulh,     InstEmit32.Vqdmulh,      OpCode32SimdReg.Create,         OpCode32SimdReg.CreateT32);
             SetAsimd("111100111x11<<10xxxx00101xx0xxx0", InstName.Vqmovn,      InstEmit32.Vqmovn,       OpCode32SimdMovn.Create,        OpCode32SimdMovn.CreateT32);
             SetAsimd("111100111x11<<10xxxx001001x0xxx0", InstName.Vqmovun,     InstEmit32.Vqmovun,      OpCode32SimdMovn.Create,        OpCode32SimdMovn.CreateT32);
+            SetAsimd("111100110x01xxxxxxxx1011xxx0xxxx", InstName.Vqrdmulh,    InstEmit32.Vqrdmulh,     OpCode32SimdReg.Create,         OpCode32SimdReg.CreateT32);
+            SetAsimd("111100110x10xxxxxxxx1011xxx0xxxx", InstName.Vqrdmulh,    InstEmit32.Vqrdmulh,     OpCode32SimdReg.Create,         OpCode32SimdReg.CreateT32);
             SetAsimd("1111001x1x>>>xxxxxxx100101x1xxx0", InstName.Vqrshrn,     InstEmit32.Vqrshrn,      OpCode32SimdShImmNarrow.Create, OpCode32SimdShImmNarrow.CreateT32);
             SetAsimd("111100111x>>>xxxxxxx100001x1xxx0", InstName.Vqrshrun,    InstEmit32.Vqrshrun,     OpCode32SimdShImmNarrow.Create, OpCode32SimdShImmNarrow.CreateT32);
             SetAsimd("1111001x1x>>>xxxxxxx100100x1xxx0", InstName.Vqshrn,      InstEmit32.Vqshrn,       OpCode32SimdShImmNarrow.Create, OpCode32SimdShImmNarrow.CreateT32);
@@ -1028,8 +1035,10 @@ 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);
             SetAsimd("1111001x1x>>>xxxxxxx0001>xx1xxxx", InstName.Vsra,        InstEmit32.Vsra,         OpCode32SimdShImm.Create,       OpCode32SimdShImm.CreateT32);
             SetAsimd("111101001x00xxxxxxxx0000xxx0xxxx", InstName.Vst1,        InstEmit32.Vst1,         OpCode32SimdMemSingle.Create,   OpCode32SimdMemSingle.CreateT32);
             SetAsimd("111101001x00xxxxxxxx0100xx0xxxxx", InstName.Vst1,        InstEmit32.Vst1,         OpCode32SimdMemSingle.Create,   OpCode32SimdMemSingle.CreateT32);
@@ -1054,6 +1063,7 @@ namespace ARMeilleure.Decoders
             SetAsimd("111100100x10xxxxxxxx1101xxx0xxxx", InstName.Vsub,        InstEmit32.Vsub_V,       OpCode32SimdReg.Create,         OpCode32SimdReg.CreateT32);
             SetAsimd("1111001x1x<<xxxxxxx00010x0x0xxxx", InstName.Vsubl,       InstEmit32.Vsubl_I,      OpCode32SimdRegLong.Create,     OpCode32SimdRegLong.CreateT32);
             SetAsimd("1111001x1x<<xxxxxxx00011x0x0xxxx", InstName.Vsubw,       InstEmit32.Vsubw_I,      OpCode32SimdRegWide.Create,     OpCode32SimdRegWide.CreateT32);
+            SetAsimd("111100111x110010xxxx00000xx0xxxx", InstName.Vswp,        InstEmit32.Vswp,         OpCode32Simd.Create,            OpCode32Simd.CreateT32);
             SetAsimd("111100111x11xxxxxxxx10xxxxx0xxxx", InstName.Vtbl,        InstEmit32.Vtbl,         OpCode32SimdTbl.Create,         OpCode32SimdTbl.CreateT32);
             SetAsimd("111100111x11<<10xxxx00001xx0xxxx", InstName.Vtrn,        InstEmit32.Vtrn,         OpCode32SimdCmpZ.Create,        OpCode32SimdCmpZ.CreateT32);
             SetAsimd("111100100x<<xxxxxxxx1000xxx1xxxx", InstName.Vtst,        InstEmit32.Vtst,         OpCode32SimdReg.Create,         OpCode32SimdReg.CreateT32);

src/ARMeilleure/Instructions/InstEmitAlu32.cs

@@ -2,6 +2,8 @@ using ARMeilleure.Decoders;
 using ARMeilleure.IntermediateRepresentation;
 using ARMeilleure.State;
 using ARMeilleure.Translation;
+using System;
+using System.Diagnostics;
 using System.Diagnostics.CodeAnalysis;
 using static ARMeilleure.Instructions.InstEmitAluHelper;
 using static ARMeilleure.Instructions.InstEmitHelper;
@@ -290,6 +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);
@@ -558,6 +570,46 @@ namespace ARMeilleure.Instructions
             EmitHsub8(context, unsigned: true);
         }
 
+        public static void Uqadd16(ArmEmitterContext context)
+        {
+            OpCode32AluReg op = (OpCode32AluReg)context.CurrOp;
+
+            SetIntA32(context, op.Rd, EmitUnsigned16BitPair(context, GetIntA32(context, op.Rn), GetIntA32(context, op.Rm), (d, n, m) =>
+            {
+                EmitSaturateUqadd(context, d, context.Add(n, m), 16);
+            }));
+        }
+
+        public static void Uqadd8(ArmEmitterContext context)
+        {
+            OpCode32AluReg op = (OpCode32AluReg)context.CurrOp;
+
+            SetIntA32(context, op.Rd, EmitUnsigned8BitPair(context, GetIntA32(context, op.Rn), GetIntA32(context, op.Rm), (d, n, m) =>
+            {
+                EmitSaturateUqadd(context, d, context.Add(n, m), 8);
+            }));
+        }
+
+        public static void Uqsub16(ArmEmitterContext context)
+        {
+            OpCode32AluReg op = (OpCode32AluReg)context.CurrOp;
+
+            SetIntA32(context, op.Rd, EmitUnsigned16BitPair(context, GetIntA32(context, op.Rn), GetIntA32(context, op.Rm), (d, n, m) =>
+            {
+                EmitSaturateUqsub(context, d, context.Subtract(n, m), 16);
+            }));
+        }
+
+        public static void Uqsub8(ArmEmitterContext context)
+        {
+            OpCode32AluReg op = (OpCode32AluReg)context.CurrOp;
+
+            SetIntA32(context, op.Rd, EmitUnsigned8BitPair(context, GetIntA32(context, op.Rn), GetIntA32(context, op.Rm), (d, n, m) =>
+            {
+                EmitSaturateUqsub(context, d, context.Subtract(n, m), 8);
+            }));
+        }
+
         public static void Usat(ArmEmitterContext context)
         {
             OpCode32Sat op = (OpCode32Sat)context.CurrOp;
@@ -934,6 +986,251 @@ 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);
+
+            if (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;
+            }
+
+            // If the result is 0, the values are equal and we don't need saturation.
+            Operand lblNoSat = Label();
+            context.BranchIfFalse(lblNoSat, context.ShiftRightUI(value, Const((int)saturateTo)));
+
+            // Saturate.
+            context.Copy(result, Const(uint.MaxValue >> (32 - (int)saturateTo)));
+
+            Operand lblExit = Label();
+            context.Branch(lblExit);
+
+            context.MarkLabel(lblNoSat);
+
+            context.Copy(result, value);
+
+            context.MarkLabel(lblExit);
+        }
+
+        private static void EmitSaturateUqsub(ArmEmitterContext context, Operand result, Operand value, uint saturateTo)
+        {
+            Debug.Assert(saturateTo <= 32);
+
+            if (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;
+            }
+
+            // If the result is 0, the values are equal and we don't need saturation.
+            Operand lblNoSat = Label();
+            context.BranchIf(lblNoSat, value, Const(0), Comparison.GreaterOrEqual);
+
+            // Saturate.
+            // Assumes that the value can only underflow, since this is only used for unsigned subtraction.
+            context.Copy(result, Const(0));
+
+            Operand lblExit = Label();
+            context.Branch(lblExit);
+
+            context.MarkLabel(lblNoSat);
+
+            context.Copy(result, value);
+
+            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);
+
+            Operand tempN = context.ZeroExtend16(OperandType.I32, rn);
+            Operand tempM = context.ZeroExtend16(OperandType.I32, rm);
+            elementAction(tempD, tempN, tempM);
+            Operand tempD2 = context.ZeroExtend16(OperandType.I32, tempD);
+
+            tempN = context.ShiftRightUI(rn, Const(16));
+            tempM = context.ShiftRightUI(rm, Const(16));
+            elementAction(tempD, tempN, tempM);
+            return context.BitwiseOr(tempD2, context.ShiftLeft(tempD, Const(16)));
+        }
+
+        private static Operand EmitSigned8BitPair(ArmEmitterContext context, Operand rn, Operand rm, Action<Operand, Operand, Operand> elementAction)
+        {
+            return Emit8BitPair(context, rn, rm, elementAction, unsigned: false);
+        }
+
+        private static Operand EmitUnsigned8BitPair(ArmEmitterContext context, Operand rn, Operand rm, Action<Operand, Operand, Operand> elementAction)
+        {
+            return Emit8BitPair(context, rn, rm, elementAction, unsigned: true);
+        }
+
+        private static Operand Emit8BitPair(ArmEmitterContext context, Operand rn, Operand rm, Action<Operand, Operand, Operand> elementAction, bool unsigned)
+        {
+            Operand tempD = context.AllocateLocal(OperandType.I32);
+            Operand result = default;
+
+            for (int b = 0; b < 4; b++)
+            {
+                Operand nByte = b != 0 ? context.ShiftRightUI(rn, Const(b * 8)) : rn;
+                Operand mByte = b != 0 ? context.ShiftRightUI(rm, Const(b * 8)) : rm;
+
+                if (unsigned)
+                {
+                    nByte = context.ZeroExtend8(OperandType.I32, nByte);
+                    mByte = context.ZeroExtend8(OperandType.I32, mByte);
+                }
+                else
+                {
+                    nByte = context.SignExtend8(OperandType.I32, nByte);
+                    mByte = context.SignExtend8(OperandType.I32, mByte);
+                }
+
+                elementAction(tempD, nByte, mByte);
+
+                if (b == 0)
+                {
+                    result = context.ZeroExtend8(OperandType.I32, tempD);
+                }
+                else if (b < 3)
+                {
+                    result = context.BitwiseOr(result, context.ShiftLeft(context.ZeroExtend8(OperandType.I32, tempD), Const(b * 8)));
+                }
+                else
+                {
+                    result = context.BitwiseOr(result, context.ShiftLeft(tempD, Const(24)));
+                }
+            }
+
+            return result;
+        }
+
         private static void EmitAluStore(ArmEmitterContext context, Operand value)
         {
             IOpCode32Alu op = (IOpCode32Alu)context.CurrOp;

src/ARMeilleure/Instructions/InstEmitSimdArithmetic32.cs

@@ -1246,6 +1246,33 @@ namespace ARMeilleure.Instructions
             EmitVectorUnaryNarrowOp32(context, (op1) => EmitSatQ(context, op1, 8 << op.Size, signedSrc: true, signedDst: false), signed: true);
         }
 
+        public static void Vqrdmulh(ArmEmitterContext context)
+        {
+            OpCode32SimdReg op = (OpCode32SimdReg)context.CurrOp;
+            int eSize = 8 << op.Size;
+
+            EmitVectorBinaryOpI32(context, (op1, op2) =>
+            {
+                if (op.Size == 2)
+                {
+                    op1 = context.SignExtend32(OperandType.I64, op1);
+                    op2 = context.SignExtend32(OperandType.I64, op2);
+                }
+
+                Operand res = context.Multiply(op1, op2);
+                res = context.Add(res, Const(res.Type, 1L << (eSize - 2)));
+                res = context.ShiftRightSI(res, Const(eSize - 1));
+                res = EmitSatQ(context, res, eSize, signedSrc: true, signedDst: true);
+
+                if (op.Size == 2)
+                {
+                    res = context.ConvertI64ToI32(res);
+                }
+
+                return res;
+            }, signed: true);
+        }
+
         public static void Vqsub(ArmEmitterContext context)
         {
             OpCode32SimdReg op = (OpCode32SimdReg)context.CurrOp;

src/ARMeilleure/Instructions/InstEmitSimdMove32.cs

@@ -191,6 +191,26 @@ namespace ARMeilleure.Instructions
             context.Copy(GetVecA32(op.Qd), res);
         }
 
+        public static void Vswp(ArmEmitterContext context)
+        {
+            OpCode32Simd op = (OpCode32Simd)context.CurrOp;
+
+            if (op.Q)
+            {
+                Operand temp = context.Copy(GetVecA32(op.Qd));
+
+                context.Copy(GetVecA32(op.Qd), GetVecA32(op.Qm));
+                context.Copy(GetVecA32(op.Qm), temp);
+            }
+            else
+            {
+                Operand temp = ExtractScalar(context, OperandType.I64, op.Vd);
+
+                InsertScalar(context, op.Vd, ExtractScalar(context, OperandType.I64, op.Vm));
+                InsertScalar(context, op.Vm, temp);
+            }
+        }
+
         public static void Vtbl(ArmEmitterContext context)
         {
             OpCode32SimdTbl op = (OpCode32SimdTbl)context.CurrOp;

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;
@@ -130,6 +162,36 @@ namespace ARMeilleure.Instructions
             EmitVectorUnaryNarrowOp32(context, (op1) => context.ShiftRightUI(op1, Const(shift)));
         }
 
+        public static void Vsli_I(ArmEmitterContext context)
+        {
+            OpCode32SimdShImm op = (OpCode32SimdShImm)context.CurrOp;
+            int shift = op.Shift;
+            int eSize = 8 << op.Size;
+
+            ulong mask = shift != 0 ? ulong.MaxValue >> (64 - shift) : 0UL;
+
+            Operand res = GetVec(op.Qd);
+
+            int elems = op.GetBytesCount() >> op.Size;
+
+            for (int index = 0; index < elems; index++)
+            {
+                Operand me = EmitVectorExtractZx(context, op.Qm, op.Im + index, op.Size);
+
+                Operand neShifted = context.ShiftLeft(me, Const(shift));
+
+                Operand de = EmitVectorExtractZx(context, op.Qd, op.Id + index, op.Size);
+
+                Operand deMasked = context.BitwiseAnd(de, Const(mask));
+
+                Operand e = context.BitwiseOr(neShifted, deMasked);
+
+                res = EmitVectorInsert(context, res, e, op.Id + index, op.Size);
+            }
+
+            context.Copy(GetVec(op.Qd), res);
+        }
+
         public static void Vsra(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,
@@ -571,6 +572,10 @@ namespace ARMeilleure.Instructions
         Umaal,
         Umlal,
         Umull,
+        Uqadd16,
+        Uqadd8,
+        Uqsub16,
+        Uqsub8,
         Usat,
         Usat16,
         Usub8,
@@ -645,6 +650,7 @@ namespace ARMeilleure.Instructions
         Vqdmulh,
         Vqmovn,
         Vqmovun,
+        Vqrdmulh,
         Vqrshrn,
         Vqrshrun,
         Vqshrn,
@@ -666,6 +672,7 @@ namespace ARMeilleure.Instructions
         Vshll,
         Vshr,
         Vshrn,
+        Vsli,
         Vst1,
         Vst2,
         Vst3,
@@ -682,6 +689,7 @@ namespace ARMeilleure.Instructions
         Vsub,
         Vsubl,
         Vsubw,
+        Vswp,
         Vtbl,
         Vtrn,
         Vtst,

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 = Marshal.GetFunctionPointerForDelegate<Delegate>(dlg);
+            FuncPtr = funcPtr;
         }
     }
 }

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

@@ -80,7 +80,10 @@ namespace ARMeilleure.Translation
                         return true;
                     }
 
-                    Monitor.Wait(Sync);
+                    if (!_disposed)
+                    {
+                        Monitor.Wait(Sync);
+                    }
                 }
             }
 

src/Ryujinx.Audio/Renderer/Server/Performance/PerformanceManager.cs

@@ -18,16 +18,12 @@ namespace Ryujinx.Audio.Renderer.Server.Performance
 
             if (version == 2)
             {
-                return (ulong)PerformanceManagerGeneric<PerformanceFrameHeaderVersion2,
-                                                 PerformanceEntryVersion2,
-                                                 PerformanceDetailVersion2>.GetRequiredBufferSizeForPerformanceMetricsPerFrame(ref parameter);
+                return (ulong)PerformanceManagerGeneric<PerformanceFrameHeaderVersion2, PerformanceEntryVersion2, PerformanceDetailVersion2>.GetRequiredBufferSizeForPerformanceMetricsPerFrame(ref parameter);
             }
 
             if (version == 1)
             {
-                return (ulong)PerformanceManagerGeneric<PerformanceFrameHeaderVersion1,
-                    PerformanceEntryVersion1,
-                    PerformanceDetailVersion1>.GetRequiredBufferSizeForPerformanceMetricsPerFrame(ref parameter);
+                return (ulong)PerformanceManagerGeneric<PerformanceFrameHeaderVersion1, PerformanceEntryVersion1, PerformanceDetailVersion1>.GetRequiredBufferSizeForPerformanceMetricsPerFrame(ref parameter);
             }
 
             throw new NotImplementedException($"Unknown Performance metrics data format version {version}");

src/Ryujinx.Audio/Renderer/Server/Performance/PerformanceManagerGeneric.cs

@@ -234,7 +234,7 @@ namespace Ryujinx.Audio.Renderer.Server.Performance
         {
             performanceEntry = null;
 
-            if (_entryDetailIndex > MaxFrameDetailCount)
+            if (_entryDetailIndex >= MaxFrameDetailCount)
             {
                 return false;
             }
@@ -245,7 +245,7 @@ namespace Ryujinx.Audio.Renderer.Server.Performance
                 EntryCountOffset = (uint)CurrentHeader.GetEntryCountOffset(),
             };
 
-            uint baseEntryOffset = (uint)(Unsafe.SizeOf<THeader>() + GetEntriesSize() + Unsafe.SizeOf<IPerformanceDetailEntry>() * _entryDetailIndex);
+            uint baseEntryOffset = (uint)(Unsafe.SizeOf<THeader>() + GetEntriesSize() + Unsafe.SizeOf<TEntryDetail>() * _entryDetailIndex);
 
             ref TEntryDetail entryDetail = ref EntriesDetail[_entryDetailIndex];
 

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());
-            Environment.SetEnvironmentVariable("__GL_THREADED_OPTIMIZATIONS", enabled ? "1" : "0");
+            OsUtils.SetEnvironmentVariableNoCaching("mesa_glthread", enabled.ToString().ToLower());
+            OsUtils.SetEnvironmentVariableNoCaching("__GL_THREADED_OPTIMIZATIONS", enabled ? "1" : "0");
 
             try
             {

src/Ryujinx.Common/Memory/ByteMemoryPool.ByteMemoryPoolBuffer.cs

@@ -1,51 +0,0 @@
-using System;
-using System.Buffers;
-using System.Threading;
-
-namespace Ryujinx.Common.Memory
-{
-    public partial class ByteMemoryPool
-    {
-        /// <summary>
-        /// Represents a <see cref="IMemoryOwner{Byte}"/> that wraps an array rented from
-        /// <see cref="ArrayPool{Byte}.Shared"/> and exposes it as <see cref="Memory{Byte}"/>
-        /// with a length of the requested size.
-        /// </summary>
-        private sealed class ByteMemoryPoolBuffer : IMemoryOwner<byte>
-        {
-            private byte[] _array;
-            private readonly int _length;
-
-            public ByteMemoryPoolBuffer(int length)
-            {
-                _array = ArrayPool<byte>.Shared.Rent(length);
-                _length = length;
-            }
-
-            /// <summary>
-            /// Returns a <see cref="Memory{Byte}"/> belonging to this owner.
-            /// </summary>
-            public Memory<byte> Memory
-            {
-                get
-                {
-                    byte[] array = _array;
-
-                    ObjectDisposedException.ThrowIf(array is null, this);
-
-                    return new Memory<byte>(array, 0, _length);
-                }
-            }
-
-            public void Dispose()
-            {
-                var array = Interlocked.Exchange(ref _array, null);
-
-                if (array != null)
-                {
-                    ArrayPool<byte>.Shared.Return(array);
-                }
-            }
-        }
-    }
-}

src/Ryujinx.Common/Memory/ByteMemoryPool.cs

@@ -1,106 +0,0 @@
-using System;
-using System.Buffers;
-
-namespace Ryujinx.Common.Memory
-{
-    /// <summary>
-    /// Provides a pool of re-usable byte array instances.
-    /// </summary>
-    public static partial class ByteMemoryPool
-    {
-        /// <summary>
-        /// Returns the maximum buffer size supported by this pool.
-        /// </summary>
-        public static int MaxBufferSize => Array.MaxLength;
-
-        /// <summary>
-        /// Rents a byte memory buffer from <see cref="ArrayPool{Byte}.Shared"/>.
-        /// The buffer may contain data from a prior use.
-        /// </summary>
-        /// <param name="length">The buffer's required length in bytes</param>
-        /// <returns>A <see cref="IMemoryOwner{Byte}"/> wrapping the rented memory</returns>
-        /// <exception cref="ArgumentOutOfRangeException"></exception>
-        public static IMemoryOwner<byte> Rent(long length)
-            => RentImpl(checked((int)length));
-
-        /// <summary>
-        /// Rents a byte memory buffer from <see cref="ArrayPool{Byte}.Shared"/>.
-        /// The buffer may contain data from a prior use.
-        /// </summary>
-        /// <param name="length">The buffer's required length in bytes</param>
-        /// <returns>A <see cref="IMemoryOwner{Byte}"/> wrapping the rented memory</returns>
-        /// <exception cref="ArgumentOutOfRangeException"></exception>
-        public static IMemoryOwner<byte> Rent(ulong length)
-            => RentImpl(checked((int)length));
-
-        /// <summary>
-        /// Rents a byte memory buffer from <see cref="ArrayPool{Byte}.Shared"/>.
-        /// The buffer may contain data from a prior use.
-        /// </summary>
-        /// <param name="length">The buffer's required length in bytes</param>
-        /// <returns>A <see cref="IMemoryOwner{Byte}"/> wrapping the rented memory</returns>
-        /// <exception cref="ArgumentOutOfRangeException"></exception>
-        public static IMemoryOwner<byte> Rent(int length)
-            => RentImpl(length);
-
-        /// <summary>
-        /// Rents a byte memory buffer from <see cref="ArrayPool{Byte}.Shared"/>.
-        /// The buffer's contents are cleared (set to all 0s) before returning.
-        /// </summary>
-        /// <param name="length">The buffer's required length in bytes</param>
-        /// <returns>A <see cref="IMemoryOwner{Byte}"/> wrapping the rented memory</returns>
-        /// <exception cref="ArgumentOutOfRangeException"></exception>
-        public static IMemoryOwner<byte> RentCleared(long length)
-            => RentCleared(checked((int)length));
-
-        /// <summary>
-        /// Rents a byte memory buffer from <see cref="ArrayPool{Byte}.Shared"/>.
-        /// The buffer's contents are cleared (set to all 0s) before returning.
-        /// </summary>
-        /// <param name="length">The buffer's required length in bytes</param>
-        /// <returns>A <see cref="IMemoryOwner{Byte}"/> wrapping the rented memory</returns>
-        /// <exception cref="ArgumentOutOfRangeException"></exception>
-        public static IMemoryOwner<byte> RentCleared(ulong length)
-            => RentCleared(checked((int)length));
-
-        /// <summary>
-        /// Rents a byte memory buffer from <see cref="ArrayPool{Byte}.Shared"/>.
-        /// The buffer's contents are cleared (set to all 0s) before returning.
-        /// </summary>
-        /// <param name="length">The buffer's required length in bytes</param>
-        /// <returns>A <see cref="IMemoryOwner{Byte}"/> wrapping the rented memory</returns>
-        /// <exception cref="ArgumentOutOfRangeException"></exception>
-        public static IMemoryOwner<byte> RentCleared(int length)
-        {
-            var buffer = RentImpl(length);
-
-            buffer.Memory.Span.Clear();
-
-            return buffer;
-        }
-
-        /// <summary>
-        /// Copies <paramref name="buffer"/> into a newly rented byte memory buffer.
-        /// </summary>
-        /// <param name="buffer">The byte buffer to copy</param>
-        /// <returns>A <see cref="IMemoryOwner{Byte}"/> wrapping the rented memory with <paramref name="buffer"/> copied to it</returns>
-        public static IMemoryOwner<byte> RentCopy(ReadOnlySpan<byte> buffer)
-        {
-            var copy = RentImpl(buffer.Length);
-
-            buffer.CopyTo(copy.Memory.Span);
-
-            return copy;
-        }
-
-        private static ByteMemoryPoolBuffer RentImpl(int length)
-        {
-            if ((uint)length > Array.MaxLength)
-            {
-                throw new ArgumentOutOfRangeException(nameof(length), length, null);
-            }
-
-            return new ByteMemoryPoolBuffer(length);
-        }
-    }
-}

src/Ryujinx.Common/Utilities/EmbeddedResources.cs

@@ -1,6 +1,6 @@
+using Ryujinx.Common.Memory;
 using Ryujinx.Common.Utilities;
 using System;
-using System.Buffers;
 using System.IO;
 using System.Linq;
 using System.Reflection;
@@ -42,14 +42,14 @@ namespace Ryujinx.Common
             return StreamUtils.StreamToBytes(stream);
         }
 
-        public static IMemoryOwner<byte> ReadFileToRentedMemory(string filename)
+        public static MemoryOwner<byte> ReadFileToRentedMemory(string filename)
         {
             var (assembly, path) = ResolveManifestPath(filename);
 
             return ReadFileToRentedMemory(assembly, path);
         }
 
-        public static IMemoryOwner<byte> ReadFileToRentedMemory(Assembly assembly, string filename)
+        public static MemoryOwner<byte> ReadFileToRentedMemory(Assembly assembly, string filename)
         {
             using var stream = GetStream(assembly, filename);
 

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.Common/Utilities/StreamUtils.cs

@@ -1,6 +1,5 @@
 using Microsoft.IO;
 using Ryujinx.Common.Memory;
-using System.Buffers;
 using System.IO;
 using System.Threading;
 using System.Threading.Tasks;
@@ -16,7 +15,7 @@ namespace Ryujinx.Common.Utilities
             return output.ToArray();
         }
 
-        public static IMemoryOwner<byte> StreamToRentedMemory(Stream input)
+        public static MemoryOwner<byte> StreamToRentedMemory(Stream input)
         {
             if (input is MemoryStream inputMemoryStream)
             {
@@ -26,9 +25,9 @@ namespace Ryujinx.Common.Utilities
             {
                 long bytesExpected = input.Length;
 
-                IMemoryOwner<byte> ownedMemory = ByteMemoryPool.Rent(bytesExpected);
+                MemoryOwner<byte> ownedMemory = MemoryOwner<byte>.Rent(checked((int)bytesExpected));
 
-                var destSpan = ownedMemory.Memory.Span;
+                var destSpan = ownedMemory.Span;
 
                 int totalBytesRead = 0;
 
@@ -66,14 +65,14 @@ namespace Ryujinx.Common.Utilities
             return stream.ToArray();
         }
 
-        private static IMemoryOwner<byte> MemoryStreamToRentedMemory(MemoryStream input)
+        private static MemoryOwner<byte> MemoryStreamToRentedMemory(MemoryStream input)
         {
             input.Position = 0;
 
-            IMemoryOwner<byte> ownedMemory = ByteMemoryPool.Rent(input.Length);
+            MemoryOwner<byte> ownedMemory = MemoryOwner<byte>.Rent(checked((int)input.Length));
 
             // Discard the return value because we assume reading a MemoryStream always succeeds completely.
-            _ = input.Read(ownedMemory.Memory.Span);
+            _ = input.Read(ownedMemory.Span);
 
             return ownedMemory;
         }

src/Ryujinx.Cpu/Jit/MemoryManagerHostTracked.cs

@@ -303,9 +303,9 @@ namespace Ryujinx.Cpu.Jit
             }
             else
             {
-                IMemoryOwner<byte> memoryOwner = ByteMemoryPool.Rent(size);
+                MemoryOwner<byte> memoryOwner = MemoryOwner<byte>.Rent(size);
 
-                Read(va, memoryOwner.Memory.Span);
+                Read(va, memoryOwner.Span);
 
                 return new WritableRegion(this, va, memoryOwner);
             }

src/Ryujinx.Cpu/LightningJit/Arm32/Target/Arm64/InstEmitMove.cs

@@ -1,6 +1,5 @@
 using Ryujinx.Cpu.LightningJit.CodeGen;
 using Ryujinx.Cpu.LightningJit.CodeGen.Arm64;
-using System.Diagnostics;
 
 namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
 {

src/Ryujinx.Cpu/LightningJit/Arm32/Target/Arm64/InstEmitSaturate.cs

@@ -114,7 +114,7 @@ namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
             InstEmitCommon.EmitUnsigned16BitPair(context, rd, rn, rm, (d, n, m) =>
             {
                 context.Arm64Assembler.Add(d, n, m);
-                EmitSaturateUnsignedRange(context, d, 16);
+                EmitSaturateUqadd(context, d, 16);
             });
         }
 
@@ -123,7 +123,7 @@ namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
             InstEmitCommon.EmitUnsigned8BitPair(context, rd, rn, rm, (d, n, m) =>
             {
                 context.Arm64Assembler.Add(d, n, m);
-                EmitSaturateUnsignedRange(context, d, 8);
+                EmitSaturateUqadd(context, d, 8);
             });
         }
 
@@ -140,7 +140,7 @@ namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
                     context.Arm64Assembler.Add(d, n, m);
                 }
 
-                EmitSaturateUnsignedRange(context, d, 16);
+                EmitSaturateUq(context, d, 16, e == 0);
             });
         }
 
@@ -157,25 +157,25 @@ namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
                     context.Arm64Assembler.Sub(d, n, m);
                 }
 
-                EmitSaturateUnsignedRange(context, d, 16);
+                EmitSaturateUq(context, d, 16, e != 0);
             });
         }
 
         public static void Uqsub16(CodeGenContext context, uint rd, uint rn, uint rm)
         {
-            InstEmitCommon.EmitSigned16BitPair(context, rd, rn, rm, (d, n, m) =>
+            InstEmitCommon.EmitUnsigned16BitPair(context, rd, rn, rm, (d, n, m) =>
             {
                 context.Arm64Assembler.Sub(d, n, m);
-                EmitSaturateUnsignedRange(context, d, 16);
+                EmitSaturateUqsub(context, d, 16);
             });
         }
 
         public static void Uqsub8(CodeGenContext context, uint rd, uint rn, uint rm)
         {
-            InstEmitCommon.EmitSigned8BitPair(context, rd, rn, rm, (d, n, m) =>
+            InstEmitCommon.EmitUnsigned8BitPair(context, rd, rn, rm, (d, n, m) =>
             {
                 context.Arm64Assembler.Sub(d, n, m);
-                EmitSaturateUnsignedRange(context, d, 8);
+                EmitSaturateUqsub(context, d, 8);
             });
         }
 
@@ -358,7 +358,17 @@ namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
             }
         }
 
-        private static void EmitSaturateUnsignedRange(CodeGenContext context, Operand value, uint saturateTo)
+        private static void EmitSaturateUqadd(CodeGenContext context, Operand value, uint saturateTo)
+        {
+            EmitSaturateUq(context, value, saturateTo, isSub: false);
+        }
+
+        private static void EmitSaturateUqsub(CodeGenContext context, Operand value, uint saturateTo)
+        {
+            EmitSaturateUq(context, value, saturateTo, isSub: true);
+        }
+
+        private static void EmitSaturateUq(CodeGenContext context, Operand value, uint saturateTo, bool isSub)
         {
             Debug.Assert(saturateTo <= 32);
 
@@ -379,7 +389,7 @@ namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
                 return;
             }
 
-            context.Arm64Assembler.Lsr(tempRegister.Operand, value, InstEmitCommon.Const(32 - (int)saturateTo));
+            context.Arm64Assembler.Lsr(tempRegister.Operand, value, InstEmitCommon.Const((int)saturateTo));
 
             int branchIndex = context.CodeWriter.InstructionPointer;
 
@@ -387,7 +397,7 @@ namespace Ryujinx.Cpu.LightningJit.Arm32.Target.Arm64
             context.Arm64Assembler.Cbz(tempRegister.Operand, 0);
 
             // Saturate.
-            context.Arm64Assembler.Mov(value, uint.MaxValue >> (32 - (int)saturateTo));
+            context.Arm64Assembler.Mov(value, isSub ? 0u : uint.MaxValue >> (32 - (int)saturateTo));
 
             int delta = context.CodeWriter.InstructionPointer - branchIndex;
             context.CodeWriter.WriteInstructionAt(branchIndex, context.CodeWriter.ReadInstructionAt(branchIndex) | (uint)((delta & 0x7ffff) << 5));

src/Ryujinx.Graphics.Device/DeviceMemoryManager.cs

@@ -1,7 +1,6 @@
 using Ryujinx.Common.Memory;
 using Ryujinx.Memory;
 using System;
-using System.Buffers;
 using System.Runtime.CompilerServices;
 using System.Runtime.InteropServices;
 
@@ -145,9 +144,9 @@ namespace Ryujinx.Graphics.Device
             }
             else
             {
-                IMemoryOwner<byte> memoryOwner = ByteMemoryPool.Rent(size);
+                MemoryOwner<byte> memoryOwner = MemoryOwner<byte>.Rent(size);
 
-                GetSpan(va, size).CopyTo(memoryOwner.Memory.Span);
+                ReadImpl(va, memoryOwner.Span);
 
                 return new WritableRegion(this, va, memoryOwner, tracked: true);
             }

src/Ryujinx.Graphics.Device/DeviceState.cs

@@ -39,7 +39,10 @@ namespace Ryujinx.Graphics.Device
             {
                 var field = fields[fieldIndex];
 
-                int sizeOfField = SizeCalculator.SizeOf(field.FieldType);
+                var currentFieldOffset = (int)Marshal.OffsetOf<TState>(field.Name);
+                var nextFieldOffset = fieldIndex + 1 == fields.Length ? Unsafe.SizeOf<TState>() : (int)Marshal.OffsetOf<TState>(fields[fieldIndex + 1].Name);
+
+                int sizeOfField = nextFieldOffset - currentFieldOffset;
 
                 for (int i = 0; i < ((sizeOfField + 3) & ~3); i += 4)
                 {

src/Ryujinx.Graphics.Device/SizeCalculator.cs

@@ -1,63 +0,0 @@
-using System;
-using System.Reflection;
-
-namespace Ryujinx.Graphics.Device
-{
-    public static class SizeCalculator
-    {
-        public static int SizeOf(Type type)
-        {
-            // Is type a enum type?
-            if (type.IsEnum)
-            {
-                type = type.GetEnumUnderlyingType();
-            }
-
-            // Is type a pointer type?
-            if (type.IsPointer || type == typeof(IntPtr) || type == typeof(UIntPtr))
-            {
-                return IntPtr.Size;
-            }
-
-            // Is type a struct type?
-            if (type.IsValueType && !type.IsPrimitive)
-            {
-                // Check if the struct has a explicit size, if so, return that.
-                if (type.StructLayoutAttribute.Size != 0)
-                {
-                    return type.StructLayoutAttribute.Size;
-                }
-
-                // Otherwise we calculate the sum of the sizes of all fields.
-                int size = 0;
-                var fields = type.GetFields(BindingFlags.Public | BindingFlags.NonPublic | BindingFlags.Instance);
-
-                for (int fieldIndex = 0; fieldIndex < fields.Length; fieldIndex++)
-                {
-                    size += SizeOf(fields[fieldIndex].FieldType);
-                }
-
-                return size;
-            }
-
-            // Primitive types.
-            return (Type.GetTypeCode(type)) switch
-            {
-                TypeCode.SByte => sizeof(sbyte),
-                TypeCode.Byte => sizeof(byte),
-                TypeCode.Int16 => sizeof(short),
-                TypeCode.UInt16 => sizeof(ushort),
-                TypeCode.Int32 => sizeof(int),
-                TypeCode.UInt32 => sizeof(uint),
-                TypeCode.Int64 => sizeof(long),
-                TypeCode.UInt64 => sizeof(ulong),
-                TypeCode.Char => sizeof(char),
-                TypeCode.Single => sizeof(float),
-                TypeCode.Double => sizeof(double),
-                TypeCode.Decimal => sizeof(decimal),
-                TypeCode.Boolean => sizeof(bool),
-                _ => throw new ArgumentException($"Length for type \"{type.Name}\" is unknown."),
-            };
-        }
-    }
-}

src/Ryujinx.Graphics.GAL/ResourceLayout.cs

@@ -74,13 +74,15 @@ namespace Ryujinx.Graphics.GAL
         public int ArrayLength { get; }
         public ResourceType Type { get; }
         public ResourceStages Stages { get; }
+        public bool Write { get; }
 
-        public ResourceUsage(int binding, int arrayLength, ResourceType type, ResourceStages stages)
+        public ResourceUsage(int binding, int arrayLength, ResourceType type, ResourceStages stages, bool write)
         {
             Binding = binding;
             ArrayLength = arrayLength;
             Type = type;
             Stages = stages;
+            Write = write;
         }
 
         public override int GetHashCode()

src/Ryujinx.Graphics.GAL/TextureCreateInfo.cs

@@ -1,6 +1,5 @@
 using Ryujinx.Common;
 using System;
-using System.Numerics;
 
 namespace Ryujinx.Graphics.GAL
 {
@@ -113,25 +112,6 @@ namespace Ryujinx.Graphics.GAL
             return 1;
         }
 
-        public int GetLevelsClamped()
-        {
-            int maxSize = Width;
-
-            if (Target != Target.Texture1D &&
-                Target != Target.Texture1DArray)
-            {
-                maxSize = Math.Max(maxSize, Height);
-            }
-
-            if (Target == Target.Texture3D)
-            {
-                maxSize = Math.Max(maxSize, Depth);
-            }
-
-            int maxLevels = BitOperations.Log2((uint)maxSize) + 1;
-            return Math.Min(Levels, maxLevels);
-        }
-
         private static int GetLevelSize(int size, int level)
         {
             return Math.Max(1, size >> level);

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/InlineToMemory/InlineToMemoryClass.cs

@@ -199,7 +199,7 @@ namespace Ryujinx.Graphics.Gpu.Engine.InlineToMemory
                     if (target != null)
                     {
                         target.SynchronizeMemory();
-                        var dataCopy = ByteMemoryPool.RentCopy(data);
+                        var dataCopy = MemoryOwner<byte>.RentCopy(data);
                         target.SetData(dataCopy, 0, 0, new GAL.Rectangle<int>(_dstX, _dstY, _lineLengthIn / target.Info.FormatInfo.BytesPerPixel, _lineCount));
                         target.SignalModified();
 

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

@@ -79,7 +79,10 @@ namespace Ryujinx.Graphics.Gpu.Engine.Threed
             {
                 var field = fields[fieldIndex];
 
-                int sizeOfField = SizeCalculator.SizeOf(field.FieldType);
+                var currentFieldOffset = (int)Marshal.OffsetOf<TState>(field.Name);
+                var nextFieldOffset = fieldIndex + 1 == fields.Length ? Unsafe.SizeOf<TState>() : (int)Marshal.OffsetOf<TState>(fields[fieldIndex + 1].Name);
+
+                int sizeOfField = nextFieldOffset - currentFieldOffset;
 
                 if (fieldToDelegate.TryGetValue(field.Name, out int entryIndex))
                 {

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/Texture.cs

@@ -1,4 +1,5 @@
 using Ryujinx.Common.Logging;
+using Ryujinx.Common.Memory;
 using Ryujinx.Graphics.GAL;
 using Ryujinx.Graphics.Gpu.Memory;
 using Ryujinx.Graphics.Texture;
@@ -805,7 +806,7 @@ namespace Ryujinx.Graphics.Gpu.Image
                         sliceDepth,
                         levels,
                         layers,
-                        out IMemoryOwner<byte> decoded))
+                        out MemoryOwner<byte> decoded))
                     {
                         string texInfo = $"{Info.Target} {Info.FormatInfo.Format} {Info.Width}x{Info.Height}x{Info.DepthOrLayers} levels {Info.Levels}";
 

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

@@ -340,7 +340,7 @@ namespace Ryujinx.Graphics.Gpu.Image
             /// <returns>True if any used entries of the pool might have been modified, false otherwise</returns>
             public bool SamplerPoolModified()
             {
-                return SamplerPool.WasModified(ref _samplerPoolSequence);
+                return SamplerPool != null && SamplerPool.WasModified(ref _samplerPoolSequence);
             }
         }
 
@@ -516,12 +516,15 @@ namespace Ryujinx.Graphics.Gpu.Image
                 }
 
                 // Check if any of our cached samplers changed on the pool.
-                foreach ((int samplerId, (Sampler sampler, SamplerDescriptor descriptor)) in SamplerIds)
+                if (SamplerPool != null)
                 {
-                    if (SamplerPool.GetCachedItem(samplerId) != sampler ||
-                        (sampler == null && SamplerPool.IsValidId(samplerId) && !SamplerPool.GetDescriptorRef(samplerId).Equals(descriptor)))
+                    foreach ((int samplerId, (Sampler sampler, SamplerDescriptor descriptor)) in SamplerIds)
                     {
-                        return true;
+                        if (SamplerPool.GetCachedItem(samplerId) != sampler ||
+                            (sampler == null && SamplerPool.IsValidId(samplerId) && !SamplerPool.GetDescriptorRef(samplerId).Equals(descriptor)))
+                        {
+                            return true;
+                        }
                     }
                 }
 
@@ -899,13 +902,19 @@ namespace Ryujinx.Graphics.Gpu.Image
                     }
                 }
 
-                Sampler sampler = samplerPool?.Get(samplerId);
-
                 entry.TextureIds[textureId] = (texture, descriptor);
-                entry.SamplerIds[samplerId] = (sampler, samplerPool?.GetDescriptorRef(samplerId) ?? default);
 
                 ITexture hostTexture = texture?.GetTargetTexture(bindingInfo.Target);
-                ISampler hostSampler = sampler?.GetHostSampler(texture);
+                ISampler hostSampler = null;
+
+                if (!isImage && bindingInfo.Target != Target.TextureBuffer)
+                {
+                    Sampler sampler = samplerPool?.Get(samplerId);
+
+                    entry.SamplerIds[samplerId] = (sampler, samplerPool?.GetDescriptorRef(samplerId) ?? default);
+
+                    hostSampler = sampler?.GetHostSampler(texture);
+                }
 
                 Format format = bindingInfo.Format;
 

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())
-            {
-                target = Target.Texture3D;
-            }
-            else if ((samplesInX | samplesInY) != 1)
+            if ((samplesInX | samplesInY) != 1)
             {
                 target = size.Depth > 1 && layered
                     ? Target.Texture2DMultisampleArray

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

@@ -6,6 +6,7 @@ using Ryujinx.Memory.Range;
 using System;
 using System.Collections.Concurrent;
 using System.Collections.Generic;
+using System.Numerics;
 using System.Threading;
 
 namespace Ryujinx.Graphics.Gpu.Image
@@ -490,6 +491,8 @@ namespace Ryujinx.Graphics.Gpu.Image
                 levels = (maxLod - minLod) + 1;
             }
 
+            levels = ClampLevels(target, width, height, depthOrLayers, levels);
+
             SwizzleComponent swizzleR = descriptor.UnpackSwizzleR().Convert();
             SwizzleComponent swizzleG = descriptor.UnpackSwizzleG().Convert();
             SwizzleComponent swizzleB = descriptor.UnpackSwizzleB().Convert();
@@ -540,6 +543,34 @@ namespace Ryujinx.Graphics.Gpu.Image
                 swizzleA);
         }
 
+        /// <summary>
+        /// Clamps the amount of mipmap levels to the maximum allowed for the given texture dimensions.
+        /// </summary>
+        /// <param name="target">Number of texture dimensions (1D, 2D, 3D, Cube, etc)</param>
+        /// <param name="width">Width of the texture</param>
+        /// <param name="height">Height of the texture, ignored for 1D textures</param>
+        /// <param name="depthOrLayers">Depth of the texture for 3D textures, otherwise ignored</param>
+        /// <param name="levels">Original amount of mipmap levels</param>
+        /// <returns>Clamped mipmap levels</returns>
+        private static int ClampLevels(Target target, int width, int height, int depthOrLayers, int levels)
+        {
+            int maxSize = width;
+
+            if (target != Target.Texture1D &&
+                target != Target.Texture1DArray)
+            {
+                maxSize = Math.Max(maxSize, height);
+            }
+
+            if (target == Target.Texture3D)
+            {
+                maxSize = Math.Max(maxSize, depthOrLayers);
+            }
+
+            int maxLevels = BitOperations.Log2((uint)maxSize) + 1;
+            return Math.Min(levels, maxLevels);
+        }
+
         /// <summary>
         /// Gets the texture depth-stencil mode, based on the swizzle components of each color channel.
         /// The depth-stencil mode is determined based on how the driver sets those parameters.

src/Ryujinx.Graphics.Gpu/Memory/MemoryManager.cs

@@ -2,7 +2,6 @@ using Ryujinx.Common.Memory;
 using Ryujinx.Memory;
 using Ryujinx.Memory.Range;
 using System;
-using System.Buffers;
 using System.Collections.Generic;
 using System.Runtime.CompilerServices;
 using System.Runtime.InteropServices;
@@ -242,9 +241,9 @@ namespace Ryujinx.Graphics.Gpu.Memory
             }
             else
             {
-                IMemoryOwner<byte> memoryOwner = ByteMemoryPool.Rent(size);
+                MemoryOwner<byte> memoryOwner = MemoryOwner<byte>.Rent(size);
 
-                GetSpan(va, size).CopyTo(memoryOwner.Memory.Span);
+                ReadImpl(va, memoryOwner.Span, tracked);
 
                 return new WritableRegion(this, va, memoryOwner, tracked);
             }

src/Ryujinx.Graphics.Gpu/Memory/PhysicalMemory.cs

@@ -192,9 +192,9 @@ namespace Ryujinx.Graphics.Gpu.Memory
             }
             else
             {
-                IMemoryOwner<byte> memoryOwner = ByteMemoryPool.Rent(range.GetSize());
+                MemoryOwner<byte> memoryOwner = MemoryOwner<byte>.Rent(checked((int)range.GetSize()));
 
-                Memory<byte> memory = memoryOwner.Memory;
+                Span<byte> memorySpan = memoryOwner.Span;
 
                 int offset = 0;
                 for (int i = 0; i < range.Count; i++)
@@ -203,7 +203,7 @@ namespace Ryujinx.Graphics.Gpu.Memory
                     int size = (int)currentRange.Size;
                     if (currentRange.Address != MemoryManager.PteUnmapped)
                     {
-                        GetSpan(currentRange.Address, size).CopyTo(memory.Span.Slice(offset, size));
+                        GetSpan(currentRange.Address, size).CopyTo(memorySpan.Slice(offset, size));
                     }
                     offset += size;
                 }

src/Ryujinx.Graphics.Gpu/Shader/DiskCache/DiskCacheHostStorage.cs

@@ -22,7 +22,7 @@ namespace Ryujinx.Graphics.Gpu.Shader.DiskCache
         private const ushort FileFormatVersionMajor = 1;
         private const ushort FileFormatVersionMinor = 2;
         private const uint FileFormatVersionPacked = ((uint)FileFormatVersionMajor << 16) | FileFormatVersionMinor;
-        private const uint CodeGenVersion = 6921;
+        private const uint CodeGenVersion = 7131;
 
         private const string SharedTocFileName = "shared.toc";
         private const string SharedDataFileName = "shared.data";

src/Ryujinx.Graphics.Gpu/Shader/ShaderInfoBuilder.cs

@@ -78,9 +78,9 @@ namespace Ryujinx.Graphics.Gpu.Shader
             ResourceStages stages = vertexAsCompute ? ResourceStages.Compute | ResourceStages.Vertex : VtgStages;
 
             PopulateDescriptorAndUsages(stages, ResourceType.UniformBuffer, uniformSetIndex, 1, rrc.ReservedConstantBuffers - 1);
-            PopulateDescriptorAndUsages(stages, ResourceType.StorageBuffer, storageSetIndex, 0, rrc.ReservedStorageBuffers);
+            PopulateDescriptorAndUsages(stages, ResourceType.StorageBuffer, storageSetIndex, 0, rrc.ReservedStorageBuffers, true);
             PopulateDescriptorAndUsages(stages, ResourceType.BufferTexture, textureSetIndex, 0, rrc.ReservedTextures);
-            PopulateDescriptorAndUsages(stages, ResourceType.BufferImage, imageSetIndex, 0, rrc.ReservedImages);
+            PopulateDescriptorAndUsages(stages, ResourceType.BufferImage, imageSetIndex, 0, rrc.ReservedImages, true);
         }
 
         /// <summary>
@@ -91,10 +91,11 @@ namespace Ryujinx.Graphics.Gpu.Shader
         /// <param name="setIndex">Resource set index where the resources are used</param>
         /// <param name="start">First binding number</param>
         /// <param name="count">Amount of bindings</param>
-        private void PopulateDescriptorAndUsages(ResourceStages stages, ResourceType type, int setIndex, int start, int count)
+        /// <param name="write">True if the binding is written from the shader, false otherwise</param>
+        private void PopulateDescriptorAndUsages(ResourceStages stages, ResourceType type, int setIndex, int start, int count, bool write = false)
         {
             AddDescriptor(stages, type, setIndex, start, count);
-            AddUsage(stages, type, setIndex, start, count);
+            AddUsage(stages, type, setIndex, start, count, write);
         }
 
         /// <summary>
@@ -216,11 +217,12 @@ namespace Ryujinx.Graphics.Gpu.Shader
         /// <param name="setIndex">Descriptor set number where the resource will be bound</param>
         /// <param name="binding">Binding number where the resource will be bound</param>
         /// <param name="count">Number of resources bound at the binding location</param>
-        private void AddUsage(ResourceStages stages, ResourceType type, int setIndex, int binding, int count)
+        /// <param name="write">True if the binding is written from the shader, false otherwise</param>
+        private void AddUsage(ResourceStages stages, ResourceType type, int setIndex, int binding, int count, bool write = false)
         {
             for (int index = 0; index < count; index++)
             {
-                _resourceUsages[setIndex].Add(new ResourceUsage(binding + index, 1, type, stages));
+                _resourceUsages[setIndex].Add(new ResourceUsage(binding + index, 1, type, stages, write));
             }
         }
 
@@ -238,7 +240,8 @@ namespace Ryujinx.Graphics.Gpu.Shader
                     buffer.Binding,
                     1,
                     isStorage ? ResourceType.StorageBuffer : ResourceType.UniformBuffer,
-                    stages));
+                    stages,
+                    buffer.Flags.HasFlag(BufferUsageFlags.Write)));
             }
         }
 
@@ -254,7 +257,12 @@ namespace Ryujinx.Graphics.Gpu.Shader
             {
                 ResourceType type = GetTextureResourceType(texture, isImage);
 
-                GetUsages(texture.Set).Add(new ResourceUsage(texture.Binding, texture.ArrayLength, type, stages));
+                GetUsages(texture.Set).Add(new ResourceUsage(
+                    texture.Binding,
+                    texture.ArrayLength,
+                    type,
+                    stages,
+                    texture.Flags.HasFlag(TextureUsageFlags.ImageStore)));
             }
         }
 

src/Ryujinx.Graphics.OpenGL/Image/FormatConverter.cs

@@ -1,6 +1,5 @@
 using Ryujinx.Common.Memory;
 using System;
-using System.Buffers;
 using System.Numerics;
 using System.Runtime.InteropServices;
 using System.Runtime.Intrinsics;
@@ -10,11 +9,11 @@ namespace Ryujinx.Graphics.OpenGL.Image
 {
     static class FormatConverter
     {
-        public unsafe static IMemoryOwner<byte> ConvertS8D24ToD24S8(ReadOnlySpan<byte> data)
+        public unsafe static MemoryOwner<byte> ConvertS8D24ToD24S8(ReadOnlySpan<byte> data)
         {
-            IMemoryOwner<byte> outputMemory = ByteMemoryPool.Rent(data.Length);
+            MemoryOwner<byte> outputMemory = MemoryOwner<byte>.Rent(data.Length);
 
-            Span<byte> output = outputMemory.Memory.Span;
+            Span<byte> output = outputMemory.Span;
 
             int start = 0;
 

src/Ryujinx.Graphics.OpenGL/Image/TextureStorage.cs

@@ -48,7 +48,7 @@ namespace Ryujinx.Graphics.OpenGL.Image
                 internalFormat = (SizedInternalFormat)format.PixelInternalFormat;
             }
 
-            int levels = Info.GetLevelsClamped();
+            int levels = Info.Levels;
 
             switch (Info.Target)
             {

src/Ryujinx.Graphics.OpenGL/Image/TextureView.cs

@@ -51,7 +51,7 @@ namespace Ryujinx.Graphics.OpenGL.Image
                 pixelInternalFormat = format.PixelInternalFormat;
             }
 
-            int levels = Info.GetLevelsClamped();
+            int levels = Info.Levels;
 
             GL.TextureView(
                 Handle,
@@ -267,7 +267,7 @@ namespace Ryujinx.Graphics.OpenGL.Image
         public unsafe PinnedSpan<byte> GetData()
         {
             int size = 0;
-            int levels = Info.GetLevelsClamped();
+            int levels = Info.Levels;
 
             for (int level = 0; level < levels; level++)
             {
@@ -426,7 +426,7 @@ namespace Ryujinx.Graphics.OpenGL.Image
                 faces = 6;
             }
 
-            int levels = Info.GetLevelsClamped();
+            int levels = Info.Levels;
 
             for (int level = 0; level < levels; level++)
             {
@@ -716,7 +716,7 @@ namespace Ryujinx.Graphics.OpenGL.Image
             int width = Info.Width;
             int height = Info.Height;
             int depth = Info.Depth;
-            int levels = Info.GetLevelsClamped();
+            int levels = Info.Levels;
 
             int offset = 0;
 

src/Ryujinx.Graphics.Shader/Translation/RegisterUsage.cs

@@ -155,9 +155,14 @@ namespace Ryujinx.Graphics.Shader.Translation
                         localInputs[block.Index] |= GetMask(register) & ~localOutputs[block.Index];
                     }
 
-                    if (operation.Dest != null && operation.Dest.Type == OperandType.Register)
+                    for (int dstIndex = 0; dstIndex < operation.DestsCount; dstIndex++)
                     {
-                        localOutputs[block.Index] |= GetMask(operation.Dest.GetRegister());
+                        Operand dest = operation.GetDest(dstIndex);
+
+                        if (dest != null && dest.Type == OperandType.Register)
+                        {
+                            localOutputs[block.Index] |= GetMask(dest.GetRegister());
+                        }
                     }
                 }
             }

src/Ryujinx.Graphics.Texture/Astc/AstcDecoder.cs

@@ -1,7 +1,6 @@
 using Ryujinx.Common.Memory;
 using Ryujinx.Common.Utilities;
 using System;
-using System.Buffers;
 using System.Diagnostics;
 using System.Linq;
 using System.Runtime.CompilerServices;
@@ -293,9 +292,9 @@ namespace Ryujinx.Graphics.Texture.Astc
             int depth,
             int levels,
             int layers,
-            out IMemoryOwner<byte> decoded)
+            out MemoryOwner<byte> decoded)
         {
-            decoded = ByteMemoryPool.Rent(QueryDecompressedSize(width, height, depth, levels, layers));
+            decoded = MemoryOwner<byte>.Rent(QueryDecompressedSize(width, height, depth, levels, layers));
 
             AstcDecoder decoder = new(data, decoded.Memory, blockWidth, blockHeight, width, height, depth, levels, layers);
 

src/Ryujinx.Graphics.Texture/BCnDecoder.cs

@@ -1,7 +1,6 @@
 using Ryujinx.Common;
 using Ryujinx.Common.Memory;
 using System;
-using System.Buffers;
 using System.Buffers.Binary;
 using System.Runtime.InteropServices;
 using System.Runtime.Intrinsics;
@@ -14,7 +13,7 @@ namespace Ryujinx.Graphics.Texture
         private const int BlockWidth = 4;
         private const int BlockHeight = 4;
 
-        public static IMemoryOwner<byte> DecodeBC1(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> DecodeBC1(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
         {
             int size = 0;
 
@@ -23,12 +22,12 @@ namespace Ryujinx.Graphics.Texture
                 size += Math.Max(1, width >> l) * Math.Max(1, height >> l) * Math.Max(1, depth >> l) * layers * 4;
             }
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
 
             Span<byte> tile = stackalloc byte[BlockWidth * BlockHeight * 4];
 
             Span<uint> tileAsUint = MemoryMarshal.Cast<byte, uint>(tile);
-            Span<uint> outputAsUint = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputAsUint = MemoryMarshal.Cast<byte, uint>(output.Span);
 
             Span<Vector128<byte>> tileAsVector128 = MemoryMarshal.Cast<byte, Vector128<byte>>(tile);
 
@@ -102,7 +101,7 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodeBC2(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> DecodeBC2(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
         {
             int size = 0;
 
@@ -111,12 +110,12 @@ namespace Ryujinx.Graphics.Texture
                 size += Math.Max(1, width >> l) * Math.Max(1, height >> l) * Math.Max(1, depth >> l) * layers * 4;
             }
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
 
             Span<byte> tile = stackalloc byte[BlockWidth * BlockHeight * 4];
 
             Span<uint> tileAsUint = MemoryMarshal.Cast<byte, uint>(tile);
-            Span<uint> outputAsUint = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputAsUint = MemoryMarshal.Cast<byte, uint>(output.Span);
 
             Span<Vector128<byte>> tileAsVector128 = MemoryMarshal.Cast<byte, Vector128<byte>>(tile);
 
@@ -197,7 +196,7 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodeBC3(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> DecodeBC3(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
         {
             int size = 0;
 
@@ -206,13 +205,13 @@ namespace Ryujinx.Graphics.Texture
                 size += Math.Max(1, width >> l) * Math.Max(1, height >> l) * Math.Max(1, depth >> l) * layers * 4;
             }
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
 
             Span<byte> tile = stackalloc byte[BlockWidth * BlockHeight * 4];
             Span<byte> rPal = stackalloc byte[8];
 
             Span<uint> tileAsUint = MemoryMarshal.Cast<byte, uint>(tile);
-            Span<uint> outputAsUint = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputAsUint = MemoryMarshal.Cast<byte, uint>(output.Span);
 
             Span<Vector128<byte>> tileAsVector128 = MemoryMarshal.Cast<byte, Vector128<byte>>(tile);
 
@@ -294,7 +293,7 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodeBC4(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers, bool signed)
+        public static MemoryOwner<byte> DecodeBC4(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers, bool signed)
         {
             int size = 0;
 
@@ -306,8 +305,8 @@ namespace Ryujinx.Graphics.Texture
             // Backends currently expect a stride alignment of 4 bytes, so output width must be aligned.
             int alignedWidth = BitUtils.AlignUp(width, 4);
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
-            Span<byte> outputSpan = output.Memory.Span;
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
+            Span<byte> outputSpan = output.Span;
 
             ReadOnlySpan<ulong> data64 = MemoryMarshal.Cast<byte, ulong>(data);
 
@@ -402,7 +401,7 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodeBC5(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers, bool signed)
+        public static MemoryOwner<byte> DecodeBC5(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers, bool signed)
         {
             int size = 0;
 
@@ -414,7 +413,7 @@ namespace Ryujinx.Graphics.Texture
             // Backends currently expect a stride alignment of 4 bytes, so output width must be aligned.
             int alignedWidth = BitUtils.AlignUp(width, 2);
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
 
             ReadOnlySpan<ulong> data64 = MemoryMarshal.Cast<byte, ulong>(data);
 
@@ -423,7 +422,7 @@ namespace Ryujinx.Graphics.Texture
             Span<byte> rPal = stackalloc byte[8];
             Span<byte> gPal = stackalloc byte[8];
 
-            Span<ushort> outputAsUshort = MemoryMarshal.Cast<byte, ushort>(output.Memory.Span);
+            Span<ushort> outputAsUshort = MemoryMarshal.Cast<byte, ushort>(output.Span);
 
             Span<uint> rTileAsUint = MemoryMarshal.Cast<byte, uint>(rTile);
             Span<uint> gTileAsUint = MemoryMarshal.Cast<byte, uint>(gTile);
@@ -527,7 +526,7 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodeBC6(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers, bool signed)
+        public static MemoryOwner<byte> DecodeBC6(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers, bool signed)
         {
             int size = 0;
 
@@ -536,8 +535,8 @@ namespace Ryujinx.Graphics.Texture
                 size += Math.Max(1, width >> l) * Math.Max(1, height >> l) * Math.Max(1, depth >> l) * layers * 8;
             }
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
-            Span<byte> outputSpan = output.Memory.Span;
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
+            Span<byte> outputSpan = output.Span;
 
             int inputOffset = 0;
             int outputOffset = 0;
@@ -566,7 +565,7 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodeBC7(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> DecodeBC7(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
         {
             int size = 0;
 
@@ -575,8 +574,8 @@ namespace Ryujinx.Graphics.Texture
                 size += Math.Max(1, width >> l) * Math.Max(1, height >> l) * Math.Max(1, depth >> l) * layers * 4;
             }
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
-            Span<byte> outputSpan = output.Memory.Span;
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
+            Span<byte> outputSpan = output.Span;
 
             int inputOffset = 0;
             int outputOffset = 0;

src/Ryujinx.Graphics.Texture/BCnEncoder.cs

@@ -2,7 +2,6 @@ using Ryujinx.Common;
 using Ryujinx.Common.Memory;
 using Ryujinx.Graphics.Texture.Encoders;
 using System;
-using System.Buffers;
 
 namespace Ryujinx.Graphics.Texture
 {
@@ -11,7 +10,7 @@ namespace Ryujinx.Graphics.Texture
         private const int BlockWidth = 4;
         private const int BlockHeight = 4;
 
-        public static IMemoryOwner<byte> EncodeBC7(Memory<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> EncodeBC7(Memory<byte> data, int width, int height, int depth, int levels, int layers)
         {
             int size = 0;
 
@@ -23,7 +22,7 @@ namespace Ryujinx.Graphics.Texture
                 size += w * h * 16 * Math.Max(1, depth >> l) * layers;
             }
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(size);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(size);
             Memory<byte> outputMemory = output.Memory;
 
             int imageBaseIOffs = 0;

src/Ryujinx.Graphics.Texture/ETC2Decoder.cs

@@ -1,7 +1,6 @@
 using Ryujinx.Common;
 using Ryujinx.Common.Memory;
 using System;
-using System.Buffers;
 using System.Buffers.Binary;
 using System.Runtime.InteropServices;
 
@@ -51,15 +50,15 @@ namespace Ryujinx.Graphics.Texture
             new int[] { -3, -5, -7, -9, 2, 4, 6, 8 },
         };
 
-        public static IMemoryOwner<byte> DecodeRgb(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> DecodeRgb(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
         {
             ReadOnlySpan<ulong> dataUlong = MemoryMarshal.Cast<byte, ulong>(data);
 
             int inputOffset = 0;
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(CalculateOutputSize(width, height, depth, levels, layers));
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(CalculateOutputSize(width, height, depth, levels, layers));
 
-            Span<uint> outputUint = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputUint = MemoryMarshal.Cast<byte, uint>(output.Span);
             Span<uint> tile = stackalloc uint[BlockWidth * BlockHeight];
 
             int imageBaseOOffs = 0;
@@ -113,15 +112,15 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodePta(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> DecodePta(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
         {
             ReadOnlySpan<ulong> dataUlong = MemoryMarshal.Cast<byte, ulong>(data);
 
             int inputOffset = 0;
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(CalculateOutputSize(width, height, depth, levels, layers));
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(CalculateOutputSize(width, height, depth, levels, layers));
 
-            Span<uint> outputUint = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputUint = MemoryMarshal.Cast<byte, uint>(output.Span);
             Span<uint> tile = stackalloc uint[BlockWidth * BlockHeight];
 
             int imageBaseOOffs = 0;
@@ -170,15 +169,15 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> DecodeRgba(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
+        public static MemoryOwner<byte> DecodeRgba(ReadOnlySpan<byte> data, int width, int height, int depth, int levels, int layers)
         {
             ReadOnlySpan<ulong> dataUlong = MemoryMarshal.Cast<byte, ulong>(data);
 
             int inputOffset = 0;
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(CalculateOutputSize(width, height, depth, levels, layers));
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(CalculateOutputSize(width, height, depth, levels, layers));
 
-            Span<uint> outputUint = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputUint = MemoryMarshal.Cast<byte, uint>(output.Span);
             Span<uint> tile = stackalloc uint[BlockWidth * BlockHeight];
 
             int imageBaseOOffs = 0;

src/Ryujinx.Graphics.Texture/LayoutConverter.cs

@@ -1,7 +1,6 @@
 using Ryujinx.Common;
 using Ryujinx.Common.Memory;
 using System;
-using System.Buffers;
 using System.Runtime.Intrinsics;
 using static Ryujinx.Graphics.Texture.BlockLinearConstants;
 
@@ -95,7 +94,7 @@ namespace Ryujinx.Graphics.Texture
             };
         }
 
-        public static IMemoryOwner<byte> ConvertBlockLinearToLinear(
+        public static MemoryOwner<byte> ConvertBlockLinearToLinear(
             int width,
             int height,
             int depth,
@@ -121,8 +120,8 @@ namespace Ryujinx.Graphics.Texture
                 blockHeight,
                 bytesPerPixel);
 
-            IMemoryOwner<byte> outputOwner = ByteMemoryPool.Rent(outSize);
-            Span<byte> output = outputOwner.Memory.Span;
+            MemoryOwner<byte> outputOwner = MemoryOwner<byte>.Rent(outSize);
+            Span<byte> output = outputOwner.Span;
 
             int outOffs = 0;
 
@@ -249,7 +248,7 @@ namespace Ryujinx.Graphics.Texture
             return outputOwner;
         }
 
-        public static IMemoryOwner<byte> ConvertLinearStridedToLinear(
+        public static MemoryOwner<byte> ConvertLinearStridedToLinear(
             int width,
             int height,
             int blockWidth,
@@ -265,8 +264,8 @@ namespace Ryujinx.Graphics.Texture
             int outStride = BitUtils.AlignUp(w * bytesPerPixel, HostStrideAlignment);
             lineSize = Math.Min(lineSize, outStride);
 
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(h * outStride);
-            Span<byte> outSpan = output.Memory.Span;
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(h * outStride);
+            Span<byte> outSpan = output.Span;
 
             int outOffs = 0;
             int inOffs = 0;

src/Ryujinx.Graphics.Texture/PixelConverter.cs

@@ -1,7 +1,6 @@
 using Ryujinx.Common;
 using Ryujinx.Common.Memory;
 using System;
-using System.Buffers;
 using System.Runtime.InteropServices;
 using System.Runtime.Intrinsics;
 using System.Runtime.Intrinsics.X86;
@@ -21,13 +20,14 @@ namespace Ryujinx.Graphics.Texture
             return (remainder, outRemainder, length / stride);
         }
 
-        public unsafe static IMemoryOwner<byte> ConvertR4G4ToR4G4B4A4(ReadOnlySpan<byte> data, int width)
+        public unsafe static MemoryOwner<byte> ConvertR4G4ToR4G4B4A4(ReadOnlySpan<byte> data, int width)
         {
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(data.Length * 2);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(data.Length * 2);
+            Span<byte> outputSpan = output.Span;
 
             (int remainder, int outRemainder, int height) = GetLineRemainders(data.Length, width, 1, 2);
 
-            Span<ushort> outputSpan = MemoryMarshal.Cast<byte, ushort>(output.Memory.Span);
+            Span<ushort> outputSpanUInt16 = MemoryMarshal.Cast<byte, ushort>(outputSpan);
 
             if (remainder == 0)
             {
@@ -38,7 +38,7 @@ namespace Ryujinx.Graphics.Texture
                     int sizeTrunc = data.Length & ~7;
                     start = sizeTrunc;
 
-                    fixed (byte* inputPtr = data, outputPtr = output.Memory.Span)
+                    fixed (byte* inputPtr = data, outputPtr = outputSpan)
                     {
                         for (ulong offset = 0; offset < (ulong)sizeTrunc; offset += 8)
                         {
@@ -49,7 +49,7 @@ namespace Ryujinx.Graphics.Texture
 
                 for (int i = start; i < data.Length; i++)
                 {
-                    outputSpan[i] = data[i];
+                    outputSpanUInt16[i] = data[i];
                 }
             }
             else
@@ -61,7 +61,7 @@ namespace Ryujinx.Graphics.Texture
                 {
                     for (int x = 0; x < width; x++)
                     {
-                        outputSpan[outOffset++] = data[offset++];
+                        outputSpanUInt16[outOffset++] = data[offset++];
                     }
 
                     offset += remainder;
@@ -72,16 +72,16 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> ConvertR5G6B5ToR8G8B8A8(ReadOnlySpan<byte> data, int width)
+        public static MemoryOwner<byte> ConvertR5G6B5ToR8G8B8A8(ReadOnlySpan<byte> data, int width)
         {
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(data.Length * 2);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(data.Length * 2);
             int offset = 0;
             int outOffset = 0;
 
             (int remainder, int outRemainder, int height) = GetLineRemainders(data.Length, width, 2, 4);
 
             ReadOnlySpan<ushort> inputSpan = MemoryMarshal.Cast<byte, ushort>(data);
-            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Span);
 
             for (int y = 0; y < height; y++)
             {
@@ -109,16 +109,16 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> ConvertR5G5B5ToR8G8B8A8(ReadOnlySpan<byte> data, int width, bool forceAlpha)
+        public static MemoryOwner<byte> ConvertR5G5B5ToR8G8B8A8(ReadOnlySpan<byte> data, int width, bool forceAlpha)
         {
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(data.Length * 2);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(data.Length * 2);
             int offset = 0;
             int outOffset = 0;
 
             (int remainder, int outRemainder, int height) = GetLineRemainders(data.Length, width, 2, 4);
 
             ReadOnlySpan<ushort> inputSpan = MemoryMarshal.Cast<byte, ushort>(data);
-            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Span);
 
             for (int y = 0; y < height; y++)
             {
@@ -146,16 +146,16 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> ConvertA1B5G5R5ToR8G8B8A8(ReadOnlySpan<byte> data, int width)
+        public static MemoryOwner<byte> ConvertA1B5G5R5ToR8G8B8A8(ReadOnlySpan<byte> data, int width)
         {
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(data.Length * 2);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(data.Length * 2);
             int offset = 0;
             int outOffset = 0;
 
             (int remainder, int outRemainder, int height) = GetLineRemainders(data.Length, width, 2, 4);
 
             ReadOnlySpan<ushort> inputSpan = MemoryMarshal.Cast<byte, ushort>(data);
-            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Span);
 
             for (int y = 0; y < height; y++)
             {
@@ -183,16 +183,16 @@ namespace Ryujinx.Graphics.Texture
             return output;
         }
 
-        public static IMemoryOwner<byte> ConvertR4G4B4A4ToR8G8B8A8(ReadOnlySpan<byte> data, int width)
+        public static MemoryOwner<byte> ConvertR4G4B4A4ToR8G8B8A8(ReadOnlySpan<byte> data, int width)
         {
-            IMemoryOwner<byte> output = ByteMemoryPool.Rent(data.Length * 2);
+            MemoryOwner<byte> output = MemoryOwner<byte>.Rent(data.Length * 2);
             int offset = 0;
             int outOffset = 0;
 
             (int remainder, int outRemainder, int height) = GetLineRemainders(data.Length, width, 2, 4);
 
             ReadOnlySpan<ushort> inputSpan = MemoryMarshal.Cast<byte, ushort>(data);
-            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Memory.Span);
+            Span<uint> outputSpan = MemoryMarshal.Cast<byte, uint>(output.Span);
 
             for (int y = 0; y < height; y++)
             {

src/Ryujinx.Graphics.Vulkan/BackgroundResources.cs

@@ -29,7 +29,14 @@ namespace Ryujinx.Graphics.Vulkan
 
                 lock (queueLock)
                 {
-                    _pool = new CommandBufferPool(_gd.Api, _device, queue, queueLock, _gd.QueueFamilyIndex, isLight: true);
+                    _pool = new CommandBufferPool(
+                        _gd.Api,
+                        _device,
+                        queue,
+                        queueLock,
+                        _gd.QueueFamilyIndex,
+                        _gd.IsQualcommProprietary,
+                        isLight: true);
                 }
             }
 

src/Ryujinx.Graphics.Vulkan/BarrierBatch.cs

@@ -1,6 +1,7 @@
 using Silk.NET.Vulkan;
 using System;
 using System.Collections.Generic;
+using System.Runtime.CompilerServices;
 
 namespace Ryujinx.Graphics.Vulkan
 {
@@ -8,22 +9,55 @@ namespace Ryujinx.Graphics.Vulkan
     {
         private const int MaxBarriersPerCall = 16;
 
+        private const AccessFlags BaseAccess = AccessFlags.ShaderReadBit | AccessFlags.ShaderWriteBit;
+        private const AccessFlags BufferAccess = AccessFlags.IndexReadBit | AccessFlags.VertexAttributeReadBit | AccessFlags.UniformReadBit;
+        private const AccessFlags CommandBufferAccess = AccessFlags.IndirectCommandReadBit;
+
         private readonly VulkanRenderer _gd;
 
         private readonly NativeArray<MemoryBarrier> _memoryBarrierBatch = new(MaxBarriersPerCall);
         private readonly NativeArray<BufferMemoryBarrier> _bufferBarrierBatch = new(MaxBarriersPerCall);
         private readonly NativeArray<ImageMemoryBarrier> _imageBarrierBatch = new(MaxBarriersPerCall);
 
-        private readonly List<BarrierWithStageFlags<MemoryBarrier>> _memoryBarriers = new();
-        private readonly List<BarrierWithStageFlags<BufferMemoryBarrier>> _bufferBarriers = new();
-        private readonly List<BarrierWithStageFlags<ImageMemoryBarrier>> _imageBarriers = new();
+        private readonly List<BarrierWithStageFlags<MemoryBarrier, int>> _memoryBarriers = new();
+        private readonly List<BarrierWithStageFlags<BufferMemoryBarrier, int>> _bufferBarriers = new();
+        private readonly List<BarrierWithStageFlags<ImageMemoryBarrier, TextureStorage>> _imageBarriers = new();
         private int _queuedBarrierCount;
 
+        private enum IncoherentBarrierType
+        {
+            None,
+            Texture,
+            All,
+            CommandBuffer
+        }
+
+        private bool _feedbackLoopActive;
+        private PipelineStageFlags _incoherentBufferWriteStages;
+        private PipelineStageFlags _incoherentTextureWriteStages;
+        private PipelineStageFlags _extraStages;
+        private IncoherentBarrierType _queuedIncoherentBarrier;
+        private bool _queuedFeedbackLoopBarrier;
+
         public BarrierBatch(VulkanRenderer gd)
         {
             _gd = gd;
         }
 
+        public static (AccessFlags Access, PipelineStageFlags Stages) GetSubpassAccessSuperset(VulkanRenderer gd)
+        {
+            AccessFlags access = BufferAccess;
+            PipelineStageFlags stages = PipelineStageFlags.AllGraphicsBit;
+
+            if (gd.TransformFeedbackApi != null)
+            {
+                access |= AccessFlags.TransformFeedbackWriteBitExt;
+                stages |= PipelineStageFlags.TransformFeedbackBitExt;
+            }
+
+            return (access, stages);
+        }
+
         private readonly record struct StageFlags : IEquatable<StageFlags>
         {
             public readonly PipelineStageFlags Source;
@@ -36,47 +70,150 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
-        private readonly struct BarrierWithStageFlags<T> where T : unmanaged
+        private readonly struct BarrierWithStageFlags<T, T2> where T : unmanaged
         {
             public readonly StageFlags Flags;
             public readonly T Barrier;
+            public readonly T2 Resource;
 
             public BarrierWithStageFlags(StageFlags flags, T barrier)
             {
                 Flags = flags;
                 Barrier = barrier;
+                Resource = default;
             }
 
-            public BarrierWithStageFlags(PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags, T barrier)
+            public BarrierWithStageFlags(PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags, T barrier, T2 resource)
             {
                 Flags = new StageFlags(srcStageFlags, dstStageFlags);
                 Barrier = barrier;
+                Resource = resource;
             }
         }
 
-        private void QueueBarrier<T>(List<BarrierWithStageFlags<T>> list, T barrier, PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags) where T : unmanaged
+        private void QueueBarrier<T, T2>(List<BarrierWithStageFlags<T, T2>> list, T barrier, T2 resource, PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags) where T : unmanaged
         {
-            list.Add(new BarrierWithStageFlags<T>(srcStageFlags, dstStageFlags, barrier));
+            list.Add(new BarrierWithStageFlags<T, T2>(srcStageFlags, dstStageFlags, barrier, resource));
             _queuedBarrierCount++;
         }
 
         public void QueueBarrier(MemoryBarrier barrier, PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags)
         {
-            QueueBarrier(_memoryBarriers, barrier, srcStageFlags, dstStageFlags);
+            QueueBarrier(_memoryBarriers, barrier, default, srcStageFlags, dstStageFlags);
         }
 
         public void QueueBarrier(BufferMemoryBarrier barrier, PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags)
         {
-            QueueBarrier(_bufferBarriers, barrier, srcStageFlags, dstStageFlags);
+            QueueBarrier(_bufferBarriers, barrier, default, srcStageFlags, dstStageFlags);
         }
 
-        public void QueueBarrier(ImageMemoryBarrier barrier, PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags)
+        public void QueueBarrier(ImageMemoryBarrier barrier, TextureStorage resource, PipelineStageFlags srcStageFlags, PipelineStageFlags dstStageFlags)
         {
-            QueueBarrier(_imageBarriers, barrier, srcStageFlags, dstStageFlags);
+            QueueBarrier(_imageBarriers, barrier, resource, srcStageFlags, dstStageFlags);
         }
 
-        public unsafe void Flush(CommandBuffer cb, bool insideRenderPass, Action endRenderPass)
+        [MethodImpl(MethodImplOptions.AggressiveInlining)]
+        public unsafe void FlushMemoryBarrier(ShaderCollection program, bool inRenderPass)
         {
+            if (_queuedIncoherentBarrier > IncoherentBarrierType.None)
+            {
+                // We should emit a memory barrier if there's a write access in the program (current program, or program since last barrier)
+                bool hasTextureWrite = _incoherentTextureWriteStages != PipelineStageFlags.None;
+                bool hasBufferWrite = _incoherentBufferWriteStages != PipelineStageFlags.None;
+                bool hasBufferBarrier = _queuedIncoherentBarrier > IncoherentBarrierType.Texture;
+
+                if (hasTextureWrite || (hasBufferBarrier && hasBufferWrite))
+                {
+                    AccessFlags access = BaseAccess;
+
+                    PipelineStageFlags stages = inRenderPass ? PipelineStageFlags.AllGraphicsBit : PipelineStageFlags.AllCommandsBit;
+
+                    if (hasBufferBarrier && hasBufferWrite)
+                    {
+                        access |= BufferAccess;
+
+                        if (_gd.TransformFeedbackApi != null)
+                        {
+                            access |= AccessFlags.TransformFeedbackWriteBitExt;
+                            stages |= PipelineStageFlags.TransformFeedbackBitExt;
+                        }
+                    }
+
+                    if (_queuedIncoherentBarrier == IncoherentBarrierType.CommandBuffer)
+                    {
+                        access |= CommandBufferAccess;
+                        stages |= PipelineStageFlags.DrawIndirectBit;
+                    }
+
+                    MemoryBarrier barrier = new MemoryBarrier()
+                    {
+                        SType = StructureType.MemoryBarrier,
+                        SrcAccessMask = access,
+                        DstAccessMask = access
+                    };
+
+                    QueueBarrier(barrier, stages, stages);
+
+                    _incoherentTextureWriteStages = program?.IncoherentTextureWriteStages ?? PipelineStageFlags.None;
+
+                    if (_queuedIncoherentBarrier > IncoherentBarrierType.Texture)
+                    {
+                        if (program != null)
+                        {
+                            _incoherentBufferWriteStages = program.IncoherentBufferWriteStages | _extraStages;
+                        }
+                        else
+                        {
+                            _incoherentBufferWriteStages = PipelineStageFlags.None;
+                        }
+                    }
+
+                    _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, false, inRenderPass, rpHolder, endRenderPass);
+        }
+
+        public unsafe void Flush(CommandBufferScoped cbs, ShaderCollection program, bool feedbackLoopActive, bool inRenderPass, RenderPassHolder rpHolder, Action endRenderPass)
+        {
+            if (program != null)
+            {
+                _incoherentBufferWriteStages |= program.IncoherentBufferWriteStages | _extraStages;
+                _incoherentTextureWriteStages |= program.IncoherentTextureWriteStages;
+            }
+
+            _feedbackLoopActive |= feedbackLoopActive;
+
+            FlushMemoryBarrier(program, inRenderPass);
+
+            if (!inRenderPass && rpHolder != null)
+            {
+                // Render pass is about to begin. Queue any fences that normally interrupt the pass.
+                rpHolder.InsertForcedFences(cbs);
+            }
+
             while (_queuedBarrierCount > 0)
             {
                 int memoryCount = 0;
@@ -86,20 +223,20 @@ namespace Ryujinx.Graphics.Vulkan
                 bool hasBarrier = false;
                 StageFlags flags = default;
 
-                static void AddBarriers<T>(
+                static void AddBarriers<T, T2>(
                     Span<T> target,
                     ref int queuedBarrierCount,
                     ref bool hasBarrier,
                     ref StageFlags flags,
                     ref int count,
-                    List<BarrierWithStageFlags<T>> list) where T : unmanaged
+                    List<BarrierWithStageFlags<T, T2>> list) where T : unmanaged
                 {
                     int firstMatch = -1;
                     int end = list.Count;
 
                     for (int i = 0; i < list.Count; i++)
                     {
-                        BarrierWithStageFlags<T> barrier = list[i];
+                        BarrierWithStageFlags<T, T2> barrier = list[i];
 
                         if (!hasBarrier)
                         {
@@ -162,21 +299,60 @@ namespace Ryujinx.Graphics.Vulkan
                     }
                 }
 
-                if (insideRenderPass)
+                if (inRenderPass && _imageBarriers.Count > 0)
                 {
                     // Image barriers queued in the batch are meant to be globally scoped,
                     // but inside a render pass they're scoped to just the range of the render pass.
 
                     // On MoltenVK, we just break the rules and always use image barrier.
                     // On desktop GPUs, all barriers are globally scoped, so we just replace it with a generic memory barrier.
-                    // TODO: On certain GPUs, we need to split render pass so the barrier scope is global. When this is done,
-                    //       notify the resource that it should add a barrier as soon as a render pass ends to avoid this in future.
+                    // Generally, we want to avoid this from happening in the future, so flag the texture to immediately
+                    // emit a barrier whenever the current render pass is bound again.
 
-                    if (!_gd.IsMoltenVk)
+                    bool anyIsNonAttachment = false;
+
+                    foreach (BarrierWithStageFlags<ImageMemoryBarrier, TextureStorage> barrier in _imageBarriers)
                     {
+                        // If the binding is an attachment, don't add it as a forced fence.
+                        bool isAttachment = rpHolder.ContainsAttachment(barrier.Resource);
+
+                        if (!isAttachment)
+                        {
+                            rpHolder.AddForcedFence(barrier.Resource, barrier.Flags.Dest);
+                            anyIsNonAttachment = true;
+                        }
+                    }
+
+                    if (_gd.IsTBDR)
+                    {
+                        if (!_gd.IsMoltenVk)
+                        {
+                            if (!anyIsNonAttachment)
+                            {
+                                // This case is a feedback loop. To prevent this from causing an absolute performance disaster,
+                                // remove the barriers entirely.
+                                // If this is not here, there will be a lot of single draw render passes.
+                                // TODO: explicit handling for feedback loops, likely outside this class.
+
+                                _queuedBarrierCount -= _imageBarriers.Count;
+                                _imageBarriers.Clear();
+                            }
+                            else
+                            {
+                                // TBDR GPUs are sensitive to barriers, so we need to end the pass to ensure the data is available.
+                                // Metal already has hazard tracking so MVK doesn't need this.
+                                endRenderPass();
+                                inRenderPass = false;
+                            }
+                        }
+                    }
+                    else
+                    {
+                        // Generic pipeline memory barriers will work for desktop GPUs.
+                        // They do require a few more access flags on the subpass dependency, though.
                         foreach (var barrier in _imageBarriers)
                         {
-                            _memoryBarriers.Add(new BarrierWithStageFlags<MemoryBarrier>(
+                            _memoryBarriers.Add(new BarrierWithStageFlags<MemoryBarrier, int>(
                                 barrier.Flags,
                                 new MemoryBarrier()
                                 {
@@ -190,6 +366,22 @@ namespace Ryujinx.Graphics.Vulkan
                     }
                 }
 
+                if (inRenderPass && _memoryBarriers.Count > 0)
+                {
+                    PipelineStageFlags allFlags = PipelineStageFlags.None;
+
+                    foreach (var barrier in _memoryBarriers)
+                    {
+                        allFlags |= barrier.Flags.Dest;
+                    }
+
+                    if (allFlags.HasFlag(PipelineStageFlags.DrawIndirectBit) || !_gd.SupportsRenderPassBarrier(allFlags))
+                    {
+                        endRenderPass();
+                        inRenderPass = false;
+                    }
+                }
+
                 AddBarriers(_memoryBarrierBatch.AsSpan(), ref _queuedBarrierCount, ref hasBarrier, ref flags, ref memoryCount, _memoryBarriers);
                 AddBarriers(_bufferBarrierBatch.AsSpan(), ref _queuedBarrierCount, ref hasBarrier, ref flags, ref bufferCount, _bufferBarriers);
                 AddBarriers(_imageBarrierBatch.AsSpan(), ref _queuedBarrierCount, ref hasBarrier, ref flags, ref imageCount, _imageBarriers);
@@ -198,14 +390,14 @@ namespace Ryujinx.Graphics.Vulkan
                 {
                     PipelineStageFlags srcStageFlags = flags.Source;
 
-                    if (insideRenderPass)
+                    if (inRenderPass)
                     {
                         // Inside a render pass, barrier stages can only be from rasterization.
                         srcStageFlags &= ~PipelineStageFlags.ComputeShaderBit;
                     }
 
                     _gd.Api.CmdPipelineBarrier(
-                        cb,
+                        cbs.CommandBuffer,
                         srcStageFlags,
                         flags.Dest,
                         0,
@@ -219,6 +411,43 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
+        private void QueueIncoherentBarrier(IncoherentBarrierType type)
+        {
+            if (type > _queuedIncoherentBarrier)
+            {
+                _queuedIncoherentBarrier = type;
+            }
+
+            _queuedFeedbackLoopBarrier = true;
+        }
+
+        public void QueueTextureBarrier()
+        {
+            QueueIncoherentBarrier(IncoherentBarrierType.Texture);
+        }
+
+        public void QueueMemoryBarrier()
+        {
+            QueueIncoherentBarrier(IncoherentBarrierType.All);
+        }
+
+        public void QueueCommandBufferBarrier()
+        {
+            QueueIncoherentBarrier(IncoherentBarrierType.CommandBuffer);
+        }
+
+        public void EnableTfbBarriers(bool enable)
+        {
+            if (enable)
+            {
+                _extraStages |= PipelineStageFlags.TransformFeedbackBitExt;
+            }
+            else
+            {
+                _extraStages &= ~PipelineStageFlags.TransformFeedbackBitExt;
+            }
+        }
+
         public void Dispose()
         {
             _memoryBarrierBatch.Dispose();

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

@@ -103,12 +103,19 @@ namespace Ryujinx.Graphics.Vulkan
                 usage |= BufferUsageFlags.IndirectBufferBit;
             }
 
+            var externalMemoryBuffer = new ExternalMemoryBufferCreateInfo
+            {
+                SType = StructureType.ExternalMemoryBufferCreateInfo,
+                HandleTypes = ExternalMemoryHandleTypeFlags.HostAllocationBitExt,
+            };
+
             var bufferCreateInfo = new BufferCreateInfo
             {
                 SType = StructureType.BufferCreateInfo,
                 Size = (ulong)size,
                 Usage = usage,
                 SharingMode = SharingMode.Exclusive,
+                PNext = &externalMemoryBuffer,
             };
 
             gd.Api.CreateBuffer(_device, in bufferCreateInfo, null, out var buffer).ThrowOnError();
@@ -214,7 +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

@@ -18,6 +18,7 @@ namespace Ryujinx.Graphics.Vulkan
         private readonly Device _device;
         private readonly Queue _queue;
         private readonly object _queueLock;
+        private readonly bool _concurrentFenceWaitUnsupported;
         private readonly CommandPool _pool;
         private readonly Thread _owner;
 
@@ -30,11 +31,9 @@ namespace Ryujinx.Graphics.Vulkan
             public int SubmissionCount;
             public CommandBuffer CommandBuffer;
             public FenceHolder Fence;
-            public SemaphoreHolder Semaphore;
 
             public List<IAuto> Dependants;
             public List<MultiFenceHolder> Waitables;
-            public HashSet<SemaphoreHolder> Dependencies;
 
             public void Initialize(Vk api, Device device, CommandPool pool)
             {
@@ -46,11 +45,10 @@ 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>();
-                Dependencies = new HashSet<SemaphoreHolder>();
             }
         }
 
@@ -61,12 +59,20 @@ namespace Ryujinx.Graphics.Vulkan
         private int _queuedCount;
         private int _inUseCount;
 
-        public unsafe CommandBufferPool(Vk api, Device device, Queue queue, object queueLock, uint queueFamilyIndex, bool isLight = false)
+        public unsafe CommandBufferPool(
+            Vk api,
+            Device device,
+            Queue queue,
+            object queueLock,
+            uint queueFamilyIndex,
+            bool concurrentFenceWaitUnsupported,
+            bool isLight = false)
         {
             _api = api;
             _device = device;
             _queue = queue;
             _queueLock = queueLock;
+            _concurrentFenceWaitUnsupported = concurrentFenceWaitUnsupported;
             _owner = Thread.CurrentThread;
 
             var commandPoolCreateInfo = new CommandPoolCreateInfo
@@ -77,7 +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;
@@ -134,14 +140,6 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
-        public void AddDependency(int cbIndex, CommandBufferScoped dependencyCbs)
-        {
-            Debug.Assert(_commandBuffers[cbIndex].InUse);
-            var semaphoreHolder = _commandBuffers[dependencyCbs.CommandBufferIndex].Semaphore;
-            semaphoreHolder.Get();
-            _commandBuffers[cbIndex].Dependencies.Add(semaphoreHolder);
-        }
-
         public void AddWaitable(int cbIndex, MultiFenceHolder waitable)
         {
             ref var entry = ref _commandBuffers[cbIndex];
@@ -255,7 +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);
                     }
@@ -313,7 +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();
                         }
                     }
                 }
@@ -345,19 +343,13 @@ namespace Ryujinx.Graphics.Vulkan
                 waitable.RemoveBufferUses(cbIndex);
             }
 
-            foreach (var dependency in entry.Dependencies)
-            {
-                dependency.Put();
-            }
-
             entry.Dependants.Clear();
             entry.Waitables.Clear();
-            entry.Dependencies.Clear();
             entry.Fence?.Dispose();
 
             if (refreshFence)
             {
-                entry.Fence = new FenceHolder(_api, _device);
+                entry.Fence = new FenceHolder(_api, _device, _concurrentFenceWaitUnsupported);
             }
             else
             {

src/Ryujinx.Graphics.Vulkan/CommandBufferScoped.cs

@@ -26,11 +26,6 @@ namespace Ryujinx.Graphics.Vulkan
             _pool.AddWaitable(CommandBufferIndex, waitable);
         }
 
-        public void AddDependency(CommandBufferScoped dependencyCbs)
-        {
-            _pool.AddDependency(CommandBufferIndex, dependencyCbs);
-        }
-
         public FenceHolder GetFence()
         {
             return _pool.GetFence(CommandBufferIndex);

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 Auto<DisposableImageView> View;
+            public TextureView 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 Auto<DisposableImageView> View;
+            public TextureView 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;
             }
         }
 
@@ -73,7 +74,6 @@ namespace Ryujinx.Graphics.Vulkan
 
         private readonly VulkanRenderer _gd;
         private readonly Device _device;
-        private readonly PipelineBase _pipeline;
         private ShaderCollection _program;
 
         private readonly BufferRef[] _uniformBufferRefs;
@@ -125,11 +125,12 @@ namespace Ryujinx.Graphics.Vulkan
         private readonly TextureView _dummyTexture;
         private readonly SamplerHolder _dummySampler;
 
-        public DescriptorSetUpdater(VulkanRenderer gd, Device device, PipelineBase pipeline)
+        public List<TextureView> FeedbackLoopHazards { get; private set; }
+
+        public DescriptorSetUpdater(VulkanRenderer gd, Device device)
         {
             _gd = gd;
             _device = device;
-            _pipeline = pipeline;
 
             // Some of the bindings counts needs to be multiplied by 2 because we have buffer and
             // regular textures/images interleaved on the same descriptor set.
@@ -211,10 +212,15 @@ namespace Ryujinx.Graphics.Vulkan
             _templateUpdater = new();
         }
 
-        public void Initialize()
+        public void Initialize(bool isMainPipeline)
         {
-            IMemoryOwner<byte> dummyTextureData = ByteMemoryPool.RentCleared(4);
+            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)
@@ -277,6 +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)
@@ -286,7 +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
@@ -307,7 +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
@@ -387,9 +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
             {
@@ -488,9 +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
             {
@@ -512,7 +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);
             }
@@ -684,7 +708,14 @@ namespace Ryujinx.Graphics.Vulkan
 
             if (_dirty.HasFlag(DirtyFlags.Texture))
             {
-                UpdateAndBind(cbs, program, PipelineBase.TextureSetIndex, pbp);
+                if (program.UpdateTexturesWithoutTemplate)
+                {
+                    UpdateAndBindTexturesWithoutTemplate(cbs, program, pbp);
+                }
+                else
+                {
+                    UpdateAndBind(cbs, program, PipelineBase.TextureSetIndex, pbp);
+                }
             }
 
             if (_dirty.HasFlag(DirtyFlags.Image))
@@ -831,7 +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)
@@ -881,7 +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]);
@@ -918,31 +949,84 @@ namespace Ryujinx.Graphics.Vulkan
             _gd.Api.CmdBindDescriptorSets(cbs.CommandBuffer, pbp, _program.PipelineLayout, (uint)setIndex, 1, sets, 0, ReadOnlySpan<uint>.Empty);
         }
 
-        private unsafe void UpdateBuffers(
-            CommandBufferScoped cbs,
-            PipelineBindPoint pbp,
-            int baseBinding,
-            ReadOnlySpan<DescriptorBufferInfo> bufferInfo,
-            DescriptorType type)
+        private void UpdateAndBindTexturesWithoutTemplate(CommandBufferScoped cbs, ShaderCollection program, PipelineBindPoint pbp)
         {
-            if (bufferInfo.Length == 0)
+            int setIndex = PipelineBase.TextureSetIndex;
+            var bindingSegments = program.BindingSegments[setIndex];
+
+            if (bindingSegments.Length == 0)
             {
                 return;
             }
 
-            fixed (DescriptorBufferInfo* pBufferInfo = bufferInfo)
+            if (_updateDescriptorCacheCbIndex)
             {
-                var writeDescriptorSet = new WriteDescriptorSet
-                {
-                    SType = StructureType.WriteDescriptorSet,
-                    DstBinding = (uint)baseBinding,
-                    DescriptorType = type,
-                    DescriptorCount = (uint)bufferInfo.Length,
-                    PBufferInfo = pBufferInfo,
-                };
-
-                _gd.PushDescriptorApi.CmdPushDescriptorSet(cbs.CommandBuffer, pbp, _program.PipelineLayout, 0, 1, &writeDescriptorSet);
+                _updateDescriptorCacheCbIndex = false;
+                program.UpdateDescriptorCacheCommandBufferIndex(cbs.CommandBufferIndex);
             }
+
+            var dsc = program.GetNewDescriptorSetCollection(setIndex, out _).Get(cbs);
+
+            foreach (ResourceBindingSegment segment in bindingSegments)
+            {
+                int binding = segment.Binding;
+                int count = segment.Count;
+
+                if (!segment.IsArray)
+                {
+                    if (segment.Type != ResourceType.BufferTexture)
+                    {
+                        Span<DescriptorImageInfo> textures = _textures;
+
+                        for (int i = 0; i < count; i++)
+                        {
+                            ref var texture = ref textures[i];
+                            ref var refs = ref _textureRefs[binding + i];
+
+                            texture.ImageView = refs.ImageView?.Get(cbs).Value ?? default;
+                            texture.Sampler = refs.Sampler?.Get(cbs).Value ?? default;
+
+                            if (texture.ImageView.Handle == 0)
+                            {
+                                texture.ImageView = _dummyTexture.GetImageView().Get(cbs).Value;
+                            }
+
+                            if (texture.Sampler.Handle == 0)
+                            {
+                                texture.Sampler = _dummySampler.GetSampler().Get(cbs).Value;
+                            }
+                        }
+
+                        dsc.UpdateImages(0, binding, textures[..count], DescriptorType.CombinedImageSampler);
+                    }
+                    else
+                    {
+                        Span<BufferView> bufferTextures = _bufferTextures;
+
+                        for (int i = 0; i < count; i++)
+                        {
+                            bufferTextures[i] = _bufferTextureRefs[binding + i]?.GetBufferView(cbs, false) ?? default;
+                        }
+
+                        dsc.UpdateBufferImages(0, binding, bufferTextures[..count], DescriptorType.UniformTexelBuffer);
+                    }
+                }
+                else
+                {
+                    if (segment.Type != ResourceType.BufferTexture)
+                    {
+                        dsc.UpdateImages(0, binding, _textureArrayRefs[binding].Array.GetImageInfos(_gd, cbs, _dummyTexture, _dummySampler), DescriptorType.CombinedImageSampler);
+                    }
+                    else
+                    {
+                        dsc.UpdateBufferImages(0, binding, _textureArrayRefs[binding].Array.GetBufferViews(cbs), DescriptorType.UniformTexelBuffer);
+                    }
+                }
+            }
+
+            var sets = dsc.GetSets();
+
+            _gd.Api.CmdBindDescriptorSets(cbs.CommandBuffer, pbp, _program.PipelineLayout, (uint)setIndex, 1, sets, 0, ReadOnlySpan<uint>.Empty);
         }
 
         [MethodImpl(MethodImplOptions.AggressiveInlining)]

src/Ryujinx.Graphics.Vulkan/Effects/FsrScalingFilter.cs

@@ -59,14 +59,14 @@ namespace Ryujinx.Graphics.Vulkan.Effects
             var scalingResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 2)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 1)
-                .Add(ResourceStages.Compute, ResourceType.Image, 0).Build();
+                .Add(ResourceStages.Compute, ResourceType.Image, 0, true).Build();
 
             var sharpeningResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 2)
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 3)
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 4)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 1)
-                .Add(ResourceStages.Compute, ResourceType.Image, 0).Build();
+                .Add(ResourceStages.Compute, ResourceType.Image, 0, true).Build();
 
             _sampler = _renderer.CreateSampler(SamplerCreateInfo.Create(MinFilter.Linear, MagFilter.Linear));
 

src/Ryujinx.Graphics.Vulkan/Effects/FxaaPostProcessingEffect.cs

@@ -42,7 +42,7 @@ namespace Ryujinx.Graphics.Vulkan.Effects
             var resourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 2)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 1)
-                .Add(ResourceStages.Compute, ResourceType.Image, 0).Build();
+                .Add(ResourceStages.Compute, ResourceType.Image, 0, true).Build();
 
             _samplerLinear = _renderer.CreateSampler(SamplerCreateInfo.Create(MinFilter.Linear, MagFilter.Linear));
 

src/Ryujinx.Graphics.Vulkan/Effects/SmaaPostProcessingEffect.cs

@@ -81,20 +81,20 @@ namespace Ryujinx.Graphics.Vulkan.Effects
             var edgeResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 2)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 1)
-                .Add(ResourceStages.Compute, ResourceType.Image, 0).Build();
+                .Add(ResourceStages.Compute, ResourceType.Image, 0, true).Build();
 
             var blendResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 2)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 1)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 3)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 4)
-                .Add(ResourceStages.Compute, ResourceType.Image, 0).Build();
+                .Add(ResourceStages.Compute, ResourceType.Image, 0, true).Build();
 
             var neighbourResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 2)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 1)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 3)
-                .Add(ResourceStages.Compute, ResourceType.Image, 0).Build();
+                .Add(ResourceStages.Compute, ResourceType.Image, 0, true).Build();
 
             _samplerLinear = _renderer.CreateSampler(SamplerCreateInfo.Create(MinFilter.Linear, MagFilter.Linear));
 

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/FenceHolder.cs

@@ -10,12 +10,15 @@ namespace Ryujinx.Graphics.Vulkan
         private readonly Device _device;
         private Fence _fence;
         private int _referenceCount;
+        private int _lock;
+        private readonly bool _concurrentWaitUnsupported;
         private bool _disposed;
 
-        public unsafe FenceHolder(Vk api, Device device)
+        public unsafe FenceHolder(Vk api, Device device, bool concurrentWaitUnsupported)
         {
             _api = api;
             _device = device;
+            _concurrentWaitUnsupported = concurrentWaitUnsupported;
 
             var fenceCreateInfo = new FenceCreateInfo
             {
@@ -47,6 +50,11 @@ namespace Ryujinx.Graphics.Vulkan
             }
             while (Interlocked.CompareExchange(ref _referenceCount, lastValue + 1, lastValue) != lastValue);
 
+            if (_concurrentWaitUnsupported)
+            {
+                AcquireLock();
+            }
+
             fence = _fence;
             return true;
         }
@@ -57,6 +65,16 @@ namespace Ryujinx.Graphics.Vulkan
             return _fence;
         }
 
+        public void PutLock()
+        {
+            Put();
+
+            if (_concurrentWaitUnsupported)
+            {
+                ReleaseLock();
+            }
+        }
+
         public void Put()
         {
             if (Interlocked.Decrement(ref _referenceCount) == 0)
@@ -66,24 +84,67 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
+        private void AcquireLock()
+        {
+            while (!TryAcquireLock())
+            {
+                Thread.SpinWait(32);
+            }
+        }
+
+        private bool TryAcquireLock()
+        {
+            return Interlocked.Exchange(ref _lock, 1) == 0;
+        }
+
+        private void ReleaseLock()
+        {
+            Interlocked.Exchange(ref _lock, 0);
+        }
+
         public void Wait()
         {
-            Span<Fence> fences = stackalloc Fence[]
+            if (_concurrentWaitUnsupported)
             {
-                _fence,
-            };
+                AcquireLock();
 
-            FenceHelper.WaitAllIndefinitely(_api, _device, fences);
+                try
+                {
+                    FenceHelper.WaitAllIndefinitely(_api, _device, stackalloc Fence[] { _fence });
+                }
+                finally
+                {
+                    ReleaseLock();
+                }
+            }
+            else
+            {
+                FenceHelper.WaitAllIndefinitely(_api, _device, stackalloc Fence[] { _fence });
+            }
         }
 
         public bool IsSignaled()
         {
-            Span<Fence> fences = stackalloc Fence[]
+            if (_concurrentWaitUnsupported)
             {
-                _fence,
-            };
+                if (!TryAcquireLock())
+                {
+                    return false;
+                }
 
-            return FenceHelper.AllSignaled(_api, _device, fences);
+                try
+                {
+                    return FenceHelper.AllSignaled(_api, _device, stackalloc Fence[] { _fence });
+                }
+                finally
+                {
+                    ReleaseLock();
+                }
+            }
+            else
+            {
+                return FenceHelper.AllSignaled(_api, _device, stackalloc Fence[] { _fence });
+            }
         }
 
         public void Dispose()

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);
         }
 
@@ -286,10 +286,44 @@ namespace Ryujinx.Graphics.Vulkan
 
             _depthStencil?.Storage?.QueueLoadOpBarrier(cbs, true);
 
-            gd.Barriers.Flush(cbs.CommandBuffer, false, null);
+            gd.Barriers.Flush(cbs, false, null, null);
         }
 
-        public (Auto<DisposableRenderPass> renderPass, Auto<DisposableFramebuffer> framebuffer) GetPassAndFramebuffer(
+        public void AddStoreOpUsage()
+        {
+            if (_colors != null)
+            {
+                foreach (var color in _colors)
+                {
+                    color.Storage?.AddStoreOpUsage(false);
+                }
+            }
+
+            _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,
             CommandBufferScoped cbs)

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/HelperShader.cs

@@ -115,7 +115,7 @@ namespace Ryujinx.Graphics.Vulkan
             var strideChangeResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 0)
                 .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 1)
-                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2).Build();
+                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2, true).Build();
 
             _programStrideChange = gd.CreateProgramWithMinimalLayout(new[]
             {
@@ -125,7 +125,7 @@ namespace Ryujinx.Graphics.Vulkan
             var colorCopyResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 0)
                 .Add(ResourceStages.Compute, ResourceType.TextureAndSampler, 0)
-                .Add(ResourceStages.Compute, ResourceType.Image, 0).Build();
+                .Add(ResourceStages.Compute, ResourceType.Image, 0, true).Build();
 
             _programColorCopyShortening = gd.CreateProgramWithMinimalLayout(new[]
             {
@@ -155,7 +155,7 @@ namespace Ryujinx.Graphics.Vulkan
             var convertD32S8ToD24S8ResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 0)
                 .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 1)
-                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2).Build();
+                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2, true).Build();
 
             _programConvertD32S8ToD24S8 = gd.CreateProgramWithMinimalLayout(new[]
             {
@@ -165,7 +165,7 @@ namespace Ryujinx.Graphics.Vulkan
             var convertIndexBufferResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 0)
                 .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 1)
-                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2).Build();
+                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2, true).Build();
 
             _programConvertIndexBuffer = gd.CreateProgramWithMinimalLayout(new[]
             {
@@ -175,7 +175,7 @@ namespace Ryujinx.Graphics.Vulkan
             var convertIndirectDataResourceLayout = new ResourceLayoutBuilder()
                 .Add(ResourceStages.Compute, ResourceType.UniformBuffer, 0)
                 .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 1)
-                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2)
+                .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 2, true)
                 .Add(ResourceStages.Compute, ResourceType.StorageBuffer, 3).Build();
 
             _programConvertIndirectData = gd.CreateProgramWithMinimalLayout(new[]

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/ImageArray.cs

@@ -95,7 +95,7 @@ namespace Ryujinx.Graphics.Vulkan
         {
             _cachedCommandBufferIndex = -1;
             _storages = null;
-            SetDirty(_gd);
+            SetDirty(_gd, isImage: true);
         }
 
         public void QueueWriteToReadBarriers(CommandBufferScoped cbs, PipelineStageFlags stageFlags)

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/MultiFenceHolder.cs

@@ -196,18 +196,23 @@ namespace Ryujinx.Graphics.Vulkan
 
             bool signaled = true;
 
-            if (hasTimeout)
+            try
             {
-                signaled = FenceHelper.AllSignaled(api, device, fences[..fenceCount], timeout);
+                if (hasTimeout)
+                {
+                    signaled = FenceHelper.AllSignaled(api, device, fences[..fenceCount], timeout);
+                }
+                else
+                {
+                    FenceHelper.WaitAllIndefinitely(api, device, fences[..fenceCount]);
+                }
             }
-            else
+            finally
             {
-                FenceHelper.WaitAllIndefinitely(api, device, fences[..fenceCount]);
-            }
-
-            for (int i = 0; i < fenceCount; i++)
-            {
-                fenceHolders[i].Put();
+                for (int i = 0; i < fenceCount; i++)
+                {
+                    fenceHolders[i].PutLock();
+                }
             }
 
             return signaled;

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;
@@ -55,6 +57,7 @@ namespace Ryujinx.Graphics.Vulkan
 
         protected FramebufferParams FramebufferParams;
         private Auto<DisposableFramebuffer> _framebuffer;
+        private RenderPassHolder _rpHolder;
         private Auto<DisposableRenderPass> _renderPass;
         private RenderPassHolder _nullRenderPass;
         private int _writtenAttachmentCount;
@@ -84,9 +87,10 @@ namespace Ryujinx.Graphics.Vulkan
         private bool _tfEnabled;
         private bool _tfActive;
 
-        private readonly PipelineColorBlendAttachmentState[] _storedBlend;
+        private FeedbackLoopAspects _feedbackLoop;
+        private bool _passWritesDepthStencil;
 
-        private ulong _drawCountSinceBarrier;
+        private readonly PipelineColorBlendAttachmentState[] _storedBlend;
         public ulong DrawCount { get; private set; }
         public bool RenderPassActive { get; private set; }
 
@@ -103,9 +107,9 @@ 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, this);
+            _descriptorSetUpdater = new DescriptorSetUpdater(gd, device);
             _vertexBufferUpdater = new VertexBufferUpdater(gd);
 
             _transformFeedbackBuffers = new BufferState[Constants.MaxTransformFeedbackBuffers];
@@ -127,7 +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);
@@ -135,48 +139,7 @@ namespace Ryujinx.Graphics.Vulkan
 
         public unsafe void Barrier()
         {
-            if (_drawCountSinceBarrier != DrawCount)
-            {
-                _drawCountSinceBarrier = DrawCount;
-
-                // Barriers are not supported inside a render pass on Apple GPUs.
-                // As a workaround, end the render pass.
-                if (Gd.Vendor == Vendor.Apple)
-                {
-                    EndRenderPass();
-                }
-            }
-
-            MemoryBarrier memoryBarrier = new()
-            {
-                SType = StructureType.MemoryBarrier,
-                SrcAccessMask = AccessFlags.MemoryReadBit | AccessFlags.MemoryWriteBit,
-                DstAccessMask = AccessFlags.MemoryReadBit | AccessFlags.MemoryWriteBit,
-            };
-
-            PipelineStageFlags pipelineStageFlags = PipelineStageFlags.VertexShaderBit | PipelineStageFlags.FragmentShaderBit;
-
-            if (Gd.Capabilities.SupportsGeometryShader)
-            {
-                pipelineStageFlags |= PipelineStageFlags.GeometryShaderBit;
-            }
-
-            if (Gd.Capabilities.SupportsTessellationShader)
-            {
-                pipelineStageFlags |= PipelineStageFlags.TessellationControlShaderBit | PipelineStageFlags.TessellationEvaluationShaderBit;
-            }
-
-            Gd.Api.CmdPipelineBarrier(
-                CommandBuffer,
-                pipelineStageFlags,
-                pipelineStageFlags,
-                0,
-                1,
-                memoryBarrier,
-                0,
-                null,
-                0,
-                null);
+            Gd.Barriers.QueueMemoryBarrier();
         }
 
         public void ComputeBarrier()
@@ -203,6 +166,7 @@ namespace Ryujinx.Graphics.Vulkan
 
         public void BeginTransformFeedback(PrimitiveTopology topology)
         {
+            Gd.Barriers.EnableTfbBarriers(true);
             _tfEnabled = true;
         }
 
@@ -249,7 +213,7 @@ namespace Ryujinx.Graphics.Vulkan
                 CreateRenderPass();
             }
 
-            Gd.Barriers.Flush(Cbs.CommandBuffer, RenderPassActive, EndRenderPassDelegate);
+            Gd.Barriers.Flush(Cbs, RenderPassActive, _rpHolder, EndRenderPassDelegate);
 
             BeginRenderPass();
 
@@ -287,7 +251,7 @@ namespace Ryujinx.Graphics.Vulkan
                 CreateRenderPass();
             }
 
-            Gd.Barriers.Flush(Cbs.CommandBuffer, RenderPassActive, EndRenderPassDelegate);
+            Gd.Barriers.Flush(Cbs, RenderPassActive, _rpHolder, EndRenderPassDelegate);
 
             BeginRenderPass();
 
@@ -299,24 +263,7 @@ namespace Ryujinx.Graphics.Vulkan
 
         public unsafe void CommandBufferBarrier()
         {
-            MemoryBarrier memoryBarrier = new()
-            {
-                SType = StructureType.MemoryBarrier,
-                SrcAccessMask = BufferHolder.DefaultAccessFlags,
-                DstAccessMask = AccessFlags.IndirectCommandReadBit,
-            };
-
-            Gd.Api.CmdPipelineBarrier(
-                CommandBuffer,
-                PipelineStageFlags.AllCommandsBit,
-                PipelineStageFlags.DrawIndirectBit,
-                0,
-                1,
-                memoryBarrier,
-                0,
-                null,
-                0,
-                null);
+            Gd.Barriers.QueueCommandBufferBarrier();
         }
 
         public void CopyBuffer(BufferHandle source, BufferHandle destination, int srcOffset, int dstOffset, int size)
@@ -722,6 +669,7 @@ namespace Ryujinx.Graphics.Vulkan
 
         public void EndTransformFeedback()
         {
+            Gd.Barriers.EnableTfbBarriers(false);
             PauseTransformFeedbackInternal();
             _tfEnabled = false;
         }
@@ -871,6 +819,8 @@ namespace Ryujinx.Graphics.Vulkan
             _newState.DepthTestEnable = depthTest.TestEnable;
             _newState.DepthWriteEnable = depthTest.WriteEnable;
             _newState.DepthCompareOp = depthTest.Func.Convert();
+
+            UpdatePassDepthStencil();
             SignalStateChange();
         }
 
@@ -1020,6 +970,13 @@ namespace Ryujinx.Graphics.Vulkan
         {
             _newState.RasterizerDiscardEnable = discard;
             SignalStateChange();
+
+            if (!discard && Gd.IsQualcommProprietary)
+            {
+                // On Adreno, enabling rasterizer discard somehow corrupts the viewport state.
+                // Force it to be updated on next use to work around this bug.
+                DynamicState.ForceAllDirty();
+            }
         }
 
         public void SetRenderTargetColorMasks(ReadOnlySpan<uint> componentMask)
@@ -1129,6 +1086,8 @@ namespace Ryujinx.Graphics.Vulkan
             _newState.StencilFrontPassOp = stencilTest.FrontDpPass.Convert();
             _newState.StencilFrontDepthFailOp = stencilTest.FrontDpFail.Convert();
             _newState.StencilFrontCompareOp = stencilTest.FrontFunc.Convert();
+
+            UpdatePassDepthStencil();
             SignalStateChange();
         }
 
@@ -1401,24 +1360,7 @@ namespace Ryujinx.Graphics.Vulkan
 
         public unsafe void TextureBarrier()
         {
-            MemoryBarrier memoryBarrier = new()
-            {
-                SType = StructureType.MemoryBarrier,
-                SrcAccessMask = AccessFlags.MemoryReadBit | AccessFlags.MemoryWriteBit,
-                DstAccessMask = AccessFlags.MemoryReadBit | AccessFlags.MemoryWriteBit,
-            };
-
-            Gd.Api.CmdPipelineBarrier(
-                CommandBuffer,
-                PipelineStageFlags.FragmentShaderBit,
-                PipelineStageFlags.FragmentShaderBit,
-                0,
-                1,
-                memoryBarrier,
-                0,
-                null,
-                0,
-                null);
+            Gd.Barriers.QueueTextureBarrier();
         }
 
         public void TextureBarrierTiled()
@@ -1493,7 +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();
         }
 
@@ -1525,12 +1483,15 @@ namespace Ryujinx.Graphics.Vulkan
                 // Use the null framebuffer.
                 _nullRenderPass ??= new RenderPassHolder(Gd, Device, new RenderPassCacheKey(), FramebufferParams);
 
+                _rpHolder = _nullRenderPass;
                 _renderPass = _nullRenderPass.GetRenderPass();
                 _framebuffer = _nullRenderPass.GetFramebuffer(Gd, Cbs, FramebufferParams);
             }
             else
             {
-                (_renderPass, _framebuffer) = FramebufferParams.GetPassAndFramebuffer(Gd, Device, Cbs);
+                (_rpHolder, _framebuffer) = FramebufferParams.GetPassAndFramebuffer(Gd, Device, Cbs);
+
+                _renderPass = _rpHolder.GetRenderPass();
             }
         }
 
@@ -1557,11 +1518,82 @@ namespace Ryujinx.Graphics.Vulkan
                 }
             }
 
-            Gd.Barriers.Flush(Cbs.CommandBuffer, RenderPassActive, 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))
@@ -1569,7 +1601,7 @@ namespace Ryujinx.Graphics.Vulkan
                 Gd.FlushAllCommands();
             }
 
-            DynamicState.ReplayIfDirty(Gd.Api, CommandBuffer);
+            DynamicState.ReplayIfDirty(Gd, CommandBuffer);
 
             if (_needsIndexBufferRebind && _indexBufferPattern == null)
             {
@@ -1603,7 +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))
                 {
@@ -1612,17 +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.CommandBuffer, RenderPassActive, EndRenderPassDelegate);
+            Gd.Barriers.Flush(Cbs, _program, _feedbackLoop != 0, RenderPassActive, _rpHolder, EndRenderPassDelegate);
 
             _descriptorSetUpdater.UpdateAndBindDescriptorSets(Cbs, PipelineBindPoint.Graphics);
 
@@ -1692,7 +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;
             }
         }
@@ -1701,6 +1733,8 @@ namespace Ryujinx.Graphics.Vulkan
         {
             if (RenderPassActive)
             {
+                FramebufferParams.AddStoreOpUsage();
+
                 PauseTransformFeedbackInternal();
                 Gd.Api.CmdEndRenderPass(CommandBuffer);
                 SignalRenderPassEnd();

src/Ryujinx.Graphics.Vulkan/PipelineConverter.cs

@@ -9,13 +9,6 @@ namespace Ryujinx.Graphics.Vulkan
 {
     static class PipelineConverter
     {
-        private const AccessFlags SubpassAccessMask =
-            AccessFlags.MemoryReadBit |
-            AccessFlags.MemoryWriteBit |
-            AccessFlags.ShaderReadBit |
-            AccessFlags.ColorAttachmentWriteBit |
-            AccessFlags.DepthStencilAttachmentWriteBit;
-
         public static unsafe DisposableRenderPass ToRenderPass(this ProgramPipelineState state, VulkanRenderer gd, Device device)
         {
             const int MaxAttachments = Constants.MaxRenderTargets + 1;
@@ -108,7 +101,7 @@ namespace Ryujinx.Graphics.Vulkan
                 }
             }
 
-            var subpassDependency = CreateSubpassDependency();
+            var subpassDependency = CreateSubpassDependency(gd);
 
             fixed (AttachmentDescription* pAttachmentDescs = attachmentDescs)
             {
@@ -123,35 +116,39 @@ 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);
             }
         }
 
-        public static SubpassDependency CreateSubpassDependency()
+        public static SubpassDependency CreateSubpassDependency(VulkanRenderer gd)
         {
+            var (access, stages) = BarrierBatch.GetSubpassAccessSuperset(gd);
+
             return new SubpassDependency(
                 0,
                 0,
-                PipelineStageFlags.AllGraphicsBit,
-                PipelineStageFlags.AllGraphicsBit,
-                SubpassAccessMask,
-                SubpassAccessMask,
+                stages,
+                stages,
+                access,
+                access,
                 0);
         }
 
-        public unsafe static SubpassDependency2 CreateSubpassDependency2()
+        public unsafe static SubpassDependency2 CreateSubpassDependency2(VulkanRenderer gd)
         {
+            var (access, stages) = BarrierBatch.GetSubpassAccessSuperset(gd);
+
             return new SubpassDependency2(
                 StructureType.SubpassDependency2,
                 null,
                 0,
                 0,
-                PipelineStageFlags.AllGraphicsBit,
-                PipelineStageFlags.AllGraphicsBit,
-                SubpassAccessMask,
-                SubpassAccessMask,
+                stages,
+                stages,
+                access,
+                access,
                 0);
         }
 

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()
@@ -47,10 +49,11 @@ namespace Ryujinx.Graphics.Vulkan
                 return;
             }
 
-            if (componentMask != 0xf)
+            if (componentMask != 0xf || Gd.IsQualcommProprietary)
             {
                 // We can't use CmdClearAttachments if not writing all components,
                 // because on Vulkan, the pipeline state does not affect clears.
+                // On proprietary Adreno drivers, CmdClearAttachments appears to execute out of order, so it's better to not use it at all.
                 var dstTexture = FramebufferParams.GetColorView(index);
                 if (dstTexture == null)
                 {
@@ -87,10 +90,11 @@ namespace Ryujinx.Graphics.Vulkan
                 return;
             }
 
-            if (stencilMask != 0 && stencilMask != 0xff)
+            if ((stencilMask != 0 && stencilMask != 0xff) || Gd.IsQualcommProprietary)
             {
                 // We can't use CmdClearAttachments if not clearing all (mask is all ones, 0xFF) or none (mask is 0) of the stencil bits,
                 // because on Vulkan, the pipeline state does not affect clears.
+                // On proprietary Adreno drivers, CmdClearAttachments appears to execute out of order, so it's better to not use it at all.
                 var dstTexture = FramebufferParams.GetDepthStencilView();
                 if (dstTexture == null)
                 {
@@ -233,7 +237,7 @@ namespace Ryujinx.Graphics.Vulkan
 
             if (Pipeline != null && Pbp == PipelineBindPoint.Graphics)
             {
-                DynamicState.ReplayIfDirty(Gd.Api, CommandBuffer);
+                DynamicState.ReplayIfDirty(Gd, CommandBuffer);
             }
         }
 
@@ -255,7 +259,7 @@ namespace Ryujinx.Graphics.Vulkan
                 PreloadCbs = null;
             }
 
-            Gd.Barriers.Flush(Cbs.CommandBuffer, false, null);
+            Gd.Barriers.Flush(Cbs, false, null, null);
             CommandBuffer = (Cbs = Gd.CommandBufferPool.ReturnAndRent(Cbs)).CommandBuffer;
             Gd.RegisterFlush();
 

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;
@@ -439,7 +446,7 @@ namespace Ryujinx.Graphics.Vulkan
                 {
                     SType = StructureType.PipelineInputAssemblyStateCreateInfo,
                     PrimitiveRestartEnable = primitiveRestartEnable,
-                    Topology = Topology,
+                    Topology = HasTessellationControlShader ? PrimitiveTopology.PatchList : Topology,
                 };
 
                 var tessellationState = new PipelineTessellationStateCreateInfo
@@ -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

@@ -1,5 +1,7 @@
 using Silk.NET.Vulkan;
 using System;
+using System.Collections.Generic;
+using System.Linq;
 
 namespace Ryujinx.Graphics.Vulkan
 {
@@ -29,10 +31,13 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
+        private readonly record struct ForcedFence(TextureStorage Texture, PipelineStageFlags StageFlags);
+
         private readonly TextureView[] _textures;
         private readonly Auto<DisposableRenderPass> _renderPass;
         private readonly HashTableSlim<FramebufferCacheKey, Auto<DisposableFramebuffer>> _framebuffers;
         private readonly RenderPassCacheKey _key;
+        private readonly List<ForcedFence> _forcedFences;
 
         public unsafe RenderPassHolder(VulkanRenderer gd, Device device, RenderPassCacheKey key, FramebufferParams fb)
         {
@@ -105,7 +110,7 @@ namespace Ryujinx.Graphics.Vulkan
                 }
             }
 
-            var subpassDependency = PipelineConverter.CreateSubpassDependency();
+            var subpassDependency = PipelineConverter.CreateSubpassDependency(gd);
 
             fixed (AttachmentDescription* pAttachmentDescs = attachmentDescs)
             {
@@ -120,7 +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));
             }
@@ -138,6 +143,8 @@ namespace Ryujinx.Graphics.Vulkan
 
             _textures = textures;
             _key = key;
+
+            _forcedFences = new List<ForcedFence>();
         }
 
         public Auto<DisposableFramebuffer> GetFramebuffer(VulkanRenderer gd, CommandBufferScoped cbs, FramebufferParams fb)
@@ -159,6 +166,37 @@ namespace Ryujinx.Graphics.Vulkan
             return _renderPass;
         }
 
+        public void AddForcedFence(TextureStorage storage, PipelineStageFlags stageFlags)
+        {
+            if (!_forcedFences.Any(fence => fence.Texture == storage))
+            {
+                _forcedFences.Add(new ForcedFence(storage, stageFlags));
+            }
+        }
+
+        public void InsertForcedFences(CommandBufferScoped cbs)
+        {
+            if (_forcedFences.Count > 0)
+            {
+                _forcedFences.RemoveAll((entry) =>
+                {
+                    if (entry.Texture.Disposed)
+                    {
+                        return true;
+                    }
+
+                    entry.Texture.QueueWriteToReadBarrier(cbs, AccessFlags.ShaderReadBit, entry.StageFlags);
+
+                    return false;
+                });
+            }
+        }
+
+        public bool ContainsAttachment(TextureStorage storage)
+        {
+            return _textures.Any(view => view.Storage == storage);
+        }
+
         public void Dispose()
         {
             // Dispose all framebuffers.

src/Ryujinx.Graphics.Vulkan/ResourceArray.cs

@@ -14,13 +14,20 @@ namespace Ryujinx.Graphics.Vulkan
 
         private int _bindCount;
 
-        protected void SetDirty(VulkanRenderer gd)
+        protected void SetDirty(VulkanRenderer gd, bool isImage)
         {
             ReleaseDescriptorSet();
 
             if (_bindCount != 0)
             {
-                gd.PipelineInternal.ForceTextureDirty();
+                if (isImage)
+                {
+                    gd.PipelineInternal.ForceImageDirty();
+                }
+                else
+                {
+                    gd.PipelineInternal.ForceTextureDirty();
+                }
             }
         }
 

src/Ryujinx.Graphics.Vulkan/ResourceLayoutBuilder.cs

@@ -23,7 +23,7 @@ namespace Ryujinx.Graphics.Vulkan
             }
         }
 
-        public ResourceLayoutBuilder Add(ResourceStages stages, ResourceType type, int binding)
+        public ResourceLayoutBuilder Add(ResourceStages stages, ResourceType type, int binding, bool write = false)
         {
             int setIndex = type switch
             {
@@ -35,7 +35,7 @@ namespace Ryujinx.Graphics.Vulkan
             };
 
             _resourceDescriptors[setIndex].Add(new ResourceDescriptor(binding, 1, type, stages));
-            _resourceUsages[setIndex].Add(new ResourceUsage(binding, 1, type, stages));
+            _resourceUsages[setIndex].Add(new ResourceUsage(binding, 1, type, stages, write));
 
             return this;
         }

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/SemaphoreHolder.cs

@@ -1,60 +0,0 @@
-using Silk.NET.Vulkan;
-using System;
-using System.Threading;
-using VkSemaphore = Silk.NET.Vulkan.Semaphore;
-
-namespace Ryujinx.Graphics.Vulkan
-{
-    class SemaphoreHolder : IDisposable
-    {
-        private readonly Vk _api;
-        private readonly Device _device;
-        private VkSemaphore _semaphore;
-        private int _referenceCount;
-        private bool _disposed;
-
-        public unsafe SemaphoreHolder(Vk api, Device device)
-        {
-            _api = api;
-            _device = device;
-
-            var semaphoreCreateInfo = new SemaphoreCreateInfo
-            {
-                SType = StructureType.SemaphoreCreateInfo,
-            };
-
-            api.CreateSemaphore(device, in semaphoreCreateInfo, null, out _semaphore).ThrowOnError();
-
-            _referenceCount = 1;
-        }
-
-        public VkSemaphore GetUnsafe()
-        {
-            return _semaphore;
-        }
-
-        public VkSemaphore Get()
-        {
-            Interlocked.Increment(ref _referenceCount);
-            return _semaphore;
-        }
-
-        public unsafe void Put()
-        {
-            if (Interlocked.Decrement(ref _referenceCount) == 0)
-            {
-                _api.DestroySemaphore(_device, _semaphore, null);
-                _semaphore = default;
-            }
-        }
-
-        public void Dispose()
-        {
-            if (!_disposed)
-            {
-                Put();
-                _disposed = true;
-            }
-        }
-    }
-}

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/ShaderCollection.cs

@@ -23,8 +23,13 @@ namespace Ryujinx.Graphics.Vulkan
         public bool IsCompute { get; }
         public bool HasTessellationControlShader => (Stages & (1u << 3)) != 0;
 
+        public bool UpdateTexturesWithoutTemplate { get; }
+
         public uint Stages { get; }
 
+        public PipelineStageFlags IncoherentBufferWriteStages { get; }
+        public PipelineStageFlags IncoherentTextureWriteStages { get; }
+
         public ResourceBindingSegment[][] ClearSegments { get; }
         public ResourceBindingSegment[][] BindingSegments { get; }
         public DescriptorSetTemplate[] Templates { get; }
@@ -127,8 +132,12 @@ namespace Ryujinx.Graphics.Vulkan
             Stages = stages;
 
             ClearSegments = BuildClearSegments(sets);
-            BindingSegments = BuildBindingSegments(resourceLayout.SetUsages);
+            BindingSegments = BuildBindingSegments(resourceLayout.SetUsages, out bool usesBufferTextures);
             Templates = BuildTemplates(usePushDescriptors);
+            (IncoherentBufferWriteStages, IncoherentTextureWriteStages) = BuildIncoherentStages(resourceLayout.SetUsages);
+
+            // Updating buffer texture bindings using template updates crashes the Adreno driver on Windows.
+            UpdateTexturesWithoutTemplate = gd.IsQualcommProprietary && usesBufferTextures;
 
             _compileTask = Task.CompletedTask;
             _firstBackgroundUse = false;
@@ -280,8 +289,10 @@ namespace Ryujinx.Graphics.Vulkan
             return segments;
         }
 
-        private static ResourceBindingSegment[][] BuildBindingSegments(ReadOnlyCollection<ResourceUsageCollection> setUsages)
+        private static ResourceBindingSegment[][] BuildBindingSegments(ReadOnlyCollection<ResourceUsageCollection> setUsages, out bool usesBufferTextures)
         {
+            usesBufferTextures = false;
+
             ResourceBindingSegment[][] segments = new ResourceBindingSegment[setUsages.Count][];
 
             for (int setIndex = 0; setIndex < setUsages.Count; setIndex++)
@@ -295,6 +306,11 @@ namespace Ryujinx.Graphics.Vulkan
                 {
                     ResourceUsage usage = setUsages[setIndex].Usages[index];
 
+                    if (usage.Type == ResourceType.BufferTexture)
+                    {
+                        usesBufferTextures = true;
+                    }
+
                     if (currentUsage.Binding + currentCount != usage.Binding ||
                         currentUsage.Type != usage.Type ||
                         currentUsage.Stages != usage.Stages ||
@@ -365,6 +381,73 @@ namespace Ryujinx.Graphics.Vulkan
             return templates;
         }
 
+        private PipelineStageFlags GetPipelineStages(ResourceStages stages)
+        {
+            PipelineStageFlags result = 0;
+
+            if ((stages & ResourceStages.Compute) != 0)
+            {
+                result |= PipelineStageFlags.ComputeShaderBit;
+            }
+
+            if ((stages & ResourceStages.Vertex) != 0)
+            {
+                result |= PipelineStageFlags.VertexShaderBit;
+            }
+
+            if ((stages & ResourceStages.Fragment) != 0)
+            {
+                result |= PipelineStageFlags.FragmentShaderBit;
+            }
+
+            if ((stages & ResourceStages.Geometry) != 0)
+            {
+                result |= PipelineStageFlags.GeometryShaderBit;
+            }
+
+            if ((stages & ResourceStages.TessellationControl) != 0)
+            {
+                result |= PipelineStageFlags.TessellationControlShaderBit;
+            }
+
+            if ((stages & ResourceStages.TessellationEvaluation) != 0)
+            {
+                result |= PipelineStageFlags.TessellationEvaluationShaderBit;
+            }
+
+            return result;
+        }
+
+        private (PipelineStageFlags Buffer, PipelineStageFlags Texture) BuildIncoherentStages(ReadOnlyCollection<ResourceUsageCollection> setUsages)
+        {
+            PipelineStageFlags buffer = PipelineStageFlags.None;
+            PipelineStageFlags texture = PipelineStageFlags.None;
+
+            foreach (var set in setUsages)
+            {
+                foreach (var range in set.Usages)
+                {
+                    if (range.Write)
+                    {
+                        PipelineStageFlags stages = GetPipelineStages(range.Stages);
+
+                        switch (range.Type)
+                        {
+                            case ResourceType.Image:
+                                texture |= stages;
+                                break;
+                            case ResourceType.StorageBuffer:
+                            case ResourceType.BufferImage:
+                                buffer |= stages;
+                                break;
+                        }
+                    }
+                }
+            }
+
+            return (buffer, texture);
+        }
+
         private async Task BackgroundCompilation()
         {
             await Task.WhenAll(_shaders.Select(shader => shader.CompileTask));

src/Ryujinx.Graphics.Vulkan/TextureArray.cs

@@ -104,7 +104,7 @@ namespace Ryujinx.Graphics.Vulkan
         {
             _cachedCommandBufferIndex = -1;
             _storages = null;
-            SetDirty(_gd);
+            SetDirty(_gd, isImage: false);
         }
 
         public void QueueWriteToReadBarriers(CommandBufferScoped cbs, PipelineStageFlags stageFlags)

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);
@@ -407,7 +407,7 @@ namespace Ryujinx.Graphics.Vulkan
                 ImageLayout.General,
                 ImageLayout.General);
 
-            var subpassDependency = PipelineConverter.CreateSubpassDependency2();
+            var subpassDependency = PipelineConverter.CreateSubpassDependency2(gd);
 
             fixed (AttachmentDescription2* pAttachmentDescs = attachmentDescs)
             {
@@ -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;
 
@@ -38,9 +44,12 @@ namespace Ryujinx.Graphics.Vulkan
 
         public TextureCreateInfo Info => _info;
 
+        public bool Disposed { get; private set; }
+
         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;
@@ -53,6 +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(
@@ -71,6 +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();
 
@@ -78,9 +91,9 @@ 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;
+            var flags = ImageCreateFlags.CreateMutableFormatBit | ImageCreateFlags.CreateExtendedUsageBit;
 
             // This flag causes mipmapped texture arrays to break on AMD GCN, so for that copy dependencies are forced for aliasing as cube.
             bool isCube = info.Target == Target.Cubemap || info.Target == Target.CubemapArray;
@@ -112,7 +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)
             {
@@ -146,6 +159,8 @@ namespace Ryujinx.Graphics.Vulkan
 
                 InitialTransition(ImageLayout.Preinitialized, ImageLayout.General);
             }
+
+            _slices = new TextureSliceInfo[levels * _depthOrLayers];
         }
 
         public TextureStorage CreateAliasedColorForDepthStorageUnsafe(Format format)
@@ -282,7 +297,7 @@ namespace Ryujinx.Graphics.Vulkan
                 0,
                 null,
                 1,
-                barrier);
+                in barrier);
 
             if (useTempCbs)
             {
@@ -290,7 +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;
 
@@ -303,11 +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;
         }
 
@@ -399,11 +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;
@@ -433,6 +456,17 @@ namespace Ryujinx.Graphics.Vulkan
             return FormatCapabilities.IsD24S8(Info.Format) && VkFormat == VkFormat.D32SfloatS8Uint;
         }
 
+        public void AddStoreOpUsage(bool depthStencil)
+        {
+            _lastModificationStage = depthStencil ?
+                PipelineStageFlags.LateFragmentTestsBit :
+                PipelineStageFlags.ColorAttachmentOutputBit;
+
+            _lastModificationAccess = depthStencil ?
+                AccessFlags.DepthStencilAttachmentWriteBit :
+                AccessFlags.ColorAttachmentWriteBit;
+        }
+
         public void QueueLoadOpBarrier(CommandBufferScoped cbs, bool depthStencil)
         {
             PipelineStageFlags srcStageFlags = _lastReadStage | _lastModificationStage;
@@ -458,7 +492,7 @@ namespace Ryujinx.Graphics.Vulkan
                     _info.GetLayers(),
                     _info.Levels);
 
-                _gd.Barriers.QueueBarrier(barrier, srcStageFlags, dstStageFlags);
+                _gd.Barriers.QueueBarrier(barrier, this, srcStageFlags, dstStageFlags);
 
                 _lastReadStage = PipelineStageFlags.None;
                 _lastReadAccess = AccessFlags.None;
@@ -491,12 +525,61 @@ namespace Ryujinx.Graphics.Vulkan
                     _info.GetLayers(),
                     _info.Levels);
 
-                _gd.Barriers.QueueBarrier(barrier, _lastModificationStage, dstStageFlags);
+                _gd.Barriers.QueueBarrier(barrier, this, _lastModificationStage, dstStageFlags);
 
                 _lastModificationAccess = AccessFlags.None;
             }
         }
 
+        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++;
@@ -514,6 +597,8 @@ namespace Ryujinx.Graphics.Vulkan
 
         public void Dispose()
         {
+            Disposed = true;
+
             if (_aliasedStorages != null)
             {
                 foreach (var storage in _aliasedStorages.Values)

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();
 
@@ -100,7 +102,7 @@ namespace Ryujinx.Graphics.Vulkan
 
             unsafe Auto<DisposableImageView> CreateImageView(ComponentMapping cm, ImageSubresourceRange sr, ImageViewType viewType, ImageUsageFlags usageFlags)
             {
-                var usage = new ImageViewUsageCreateInfo
+                var imageViewUsage = new ImageViewUsageCreateInfo
                 {
                     SType = StructureType.ImageViewUsageCreateInfo,
                     Usage = usageFlags,
@@ -114,16 +116,16 @@ namespace Ryujinx.Graphics.Vulkan
                     Format = format,
                     Components = cm,
                     SubresourceRange = sr,
-                    PNext = &usage,
+                    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());
             }
 
             ImageUsageFlags shaderUsage = ImageUsageFlags.SampledBit;
 
-            if (info.Format.IsImageCompatible())
+            if (info.Format.IsImageCompatible() && (_gd.Capabilities.SupportsShaderStorageImageMultisample || !info.Target.IsMultisample()))
             {
                 shaderUsage |= ImageUsageFlags.StorageBit;
             }
@@ -154,7 +156,7 @@ namespace Ryujinx.Graphics.Vulkan
                 }
                 else
                 {
-                    subresourceRange = new ImageSubresourceRange(aspectFlags, (uint)firstLevel, levels, (uint)firstLayer, (uint)info.Depth);
+                    subresourceRange = new ImageSubresourceRange(aspectFlags, (uint)firstLevel, 1, (uint)firstLayer, (uint)info.Depth);
 
                     _imageView2dArray = CreateImageView(identityComponentMapping, subresourceRange, ImageViewType.Type2DArray, usage);
                 }
@@ -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)
@@ -667,8 +669,36 @@ namespace Ryujinx.Graphics.Vulkan
 
             if (PrepareOutputBuffer(cbs, hostSize, buffer, out VkBuffer copyToBuffer, out BufferHolder tempCopyHolder))
             {
+                // No barrier necessary, as this is a temporary copy buffer.
                 offset = 0;
             }
+            else
+            {
+                BufferHolder.InsertBufferBarrier(
+                    _gd,
+                    cbs.CommandBuffer,
+                    copyToBuffer,
+                    BufferHolder.DefaultAccessFlags,
+                    AccessFlags.TransferWriteBit,
+                    PipelineStageFlags.AllCommandsBit,
+                    PipelineStageFlags.TransferBit,
+                    offset,
+                    outSize);
+            }
+
+            InsertImageBarrier(
+                _gd.Api,
+                cbs.CommandBuffer,
+                image,
+                TextureStorage.DefaultAccessMask,
+                AccessFlags.TransferReadBit,
+                PipelineStageFlags.AllCommandsBit,
+                PipelineStageFlags.TransferBit,
+                Info.Format.ConvertAspectFlags(),
+                FirstLayer + layer,
+                FirstLevel + level,
+                1,
+                1);
 
             CopyFromOrToBuffer(cbs.CommandBuffer, copyToBuffer, image, hostSize, true, layer, level, 1, 1, singleSlice: true, offset, stride);
 
@@ -677,6 +707,19 @@ namespace Ryujinx.Graphics.Vulkan
                 CopyDataToOutputBuffer(cbs, tempCopyHolder, autoBuffer, hostSize, range.Offset);
                 tempCopyHolder.Dispose();
             }
+            else
+            {
+                BufferHolder.InsertBufferBarrier(
+                    _gd,
+                    cbs.CommandBuffer,
+                    copyToBuffer,
+                    AccessFlags.TransferWriteBit,
+                    BufferHolder.DefaultAccessFlags,
+                    PipelineStageFlags.TransferBit,
+                    PipelineStageFlags.AllCommandsBit,
+                    offset,
+                    outSize);
+            }
         }
 
         private ReadOnlySpan<byte> GetData(CommandBufferPool cbp, PersistentFlushBuffer flushBuffer)
@@ -908,11 +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;
@@ -969,11 +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);
             }
         }
 
@@ -993,7 +1036,35 @@ namespace Ryujinx.Graphics.Vulkan
             throw new NotImplementedException();
         }
 
-        public (Auto<DisposableRenderPass> renderPass, Auto<DisposableFramebuffer> framebuffer) GetPassAndFramebuffer(
+        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,
             CommandBufferScoped cbs,
@@ -1006,7 +1077,7 @@ namespace Ryujinx.Graphics.Vulkan
                 rpHolder = new RenderPassHolder(gd, device, key, fb);
             }
 
-            return (rpHolder.GetRenderPass(), rpHolder.GetFramebuffer(gd, cbs, fb));
+            return (rpHolder, rpHolder.GetFramebuffer(gd, cbs, fb));
         }
 
         public void AddRenderPass(RenderPassCacheKey key, RenderPassHolder renderPass)

src/Ryujinx.Graphics.Vulkan/Vendor.cs

@@ -69,27 +69,30 @@ namespace Ryujinx.Graphics.Vulkan
             {
                 DriverId.AmdProprietary => "AMD",
                 DriverId.AmdOpenSource => "AMD (Open)",
-                DriverId.ArmProprietary => "ARM",
-                DriverId.BroadcomProprietary => "Broadcom",
-                DriverId.CoreaviProprietary => "CoreAVI",
-                DriverId.GgpProprietary => "GGP",
-                DriverId.GoogleSwiftshader => "SwiftShader",
-                DriverId.ImaginationProprietary => "Imagination",
-                DriverId.IntelOpenSourceMesa => "Intel (Open)",
-                DriverId.IntelProprietaryWindows => "Intel",
-                DriverId.JuiceProprietary => "Juice",
-                DriverId.MesaDozen => "Dozen",
-                DriverId.MesaLlvmpipe => "LLVMpipe",
-                DriverId.MesaPanvk => "PanVK",
                 DriverId.MesaRadv => "RADV",
+                DriverId.NvidiaProprietary => "NVIDIA",
+                DriverId.IntelProprietaryWindows => "Intel",
+                DriverId.IntelOpenSourceMesa => "Intel (Open)",
+                DriverId.ImaginationProprietary => "Imagination",
+                DriverId.QualcommProprietary => "Qualcomm",
+                DriverId.ArmProprietary => "ARM",
+                DriverId.GoogleSwiftshader => "SwiftShader",
+                DriverId.GgpProprietary => "GGP",
+                DriverId.BroadcomProprietary => "Broadcom",
+                DriverId.MesaLlvmpipe => "LLVMpipe",
+                DriverId.Moltenvk => "MoltenVK",
+                DriverId.CoreaviProprietary => "CoreAVI",
+                DriverId.JuiceProprietary => "Juice",
+                DriverId.VerisiliconProprietary => "Verisilicon",
                 DriverId.MesaTurnip => "Turnip",
                 DriverId.MesaV3DV => "V3DV",
-                DriverId.MesaVenus => "Venus",
-                DriverId.Moltenvk => "MoltenVK",
-                DriverId.NvidiaProprietary => "NVIDIA",
-                DriverId.QualcommProprietary => "Qualcomm",
+                DriverId.MesaPanvk => "PanVK",
                 DriverId.SamsungProprietary => "Samsung",
-                DriverId.VerisiliconProprietary => "Verisilicon",
+                DriverId.MesaVenus => "Venus",
+                DriverId.MesaDozen => "Dozen",
+                DriverId.MesaNvk => "NVK",
+                DriverId.ImaginationOpenSourceMesa => "Imagination (Open)",
+                DriverId.MesaAgxv => "Honeykrisp",
                 _ => id.ToString(),
             };
         }

src/Ryujinx.Graphics.Vulkan/VulkanInitialization.cs

@@ -42,6 +42,10 @@ namespace Ryujinx.Graphics.Vulkan
             "VK_EXT_depth_clip_control",
             "VK_KHR_portability_subset", // As per spec, we should enable this if present.
             "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 = {
@@ -355,6 +359,36 @@ 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,
+                PNext = features2.PNext,
+            };
+
+            features2.PNext = &supportedPhysicalDeviceVulkan12Features;
+
             api.GetPhysicalDeviceFeatures2(physicalDevice.PhysicalDevice, &features2);
 
             var supportedFeatures = features2.Features;
@@ -382,6 +416,7 @@ namespace Ryujinx.Graphics.Vulkan
                 TessellationShader = supportedFeatures.TessellationShader,
                 VertexPipelineStoresAndAtomics = supportedFeatures.VertexPipelineStoresAndAtomics,
                 RobustBufferAccess = useRobustBufferAccess,
+                SampleRateShading = supportedFeatures.SampleRateShading,
             };
 
             void* pExtendedFeatures = null;
@@ -451,9 +486,11 @@ namespace Ryujinx.Graphics.Vulkan
             {
                 SType = StructureType.PhysicalDeviceVulkan12Features,
                 PNext = pExtendedFeatures,
-                DescriptorIndexing = physicalDevice.IsDeviceExtensionPresent("VK_EXT_descriptor_indexing"),
-                DrawIndirectCount = physicalDevice.IsDeviceExtensionPresent(KhrDrawIndirectCount.ExtensionName),
-                UniformBufferStandardLayout = physicalDevice.IsDeviceExtensionPresent("VK_KHR_uniform_buffer_standard_layout"),
+                DescriptorIndexing = supportedPhysicalDeviceVulkan12Features.DescriptorIndexing,
+                DrawIndirectCount = supportedPhysicalDeviceVulkan12Features.DrawIndirectCount,
+                UniformBufferStandardLayout = supportedPhysicalDeviceVulkan12Features.UniformBufferStandardLayout,
+                UniformAndStorageBuffer8BitAccess = supportedPhysicalDeviceVulkan12Features.UniformAndStorageBuffer8BitAccess,
+                StorageBuffer8BitAccess = supportedPhysicalDeviceVulkan12Features.StorageBuffer8BitAccess,
             };
 
             pExtendedFeatures = &featuresVk12;
@@ -518,6 +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; }
@@ -87,9 +88,11 @@ namespace Ryujinx.Graphics.Vulkan
         internal bool IsAmdGcn { get; private set; }
         internal bool IsNvidiaPreTuring { get; private set; }
         internal bool IsIntelArc { get; private set; }
+        internal bool IsQualcommProprietary { get; private set; }
         internal bool IsMoltenVk { get; private set; }
         internal bool IsTBDR { get; private set; }
         internal bool IsSharedMemory { get; private set; }
+
         public string GpuVendor { get; private set; }
         public string GpuDriver { get; private set; }
         public string GpuRenderer { get; private set; }
@@ -147,6 +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);
@@ -241,6 +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,
@@ -277,6 +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)
@@ -344,7 +378,7 @@ namespace Ryujinx.Graphics.Vulkan
                 {
                     IsNvidiaPreTuring = gpuNumber < 2000;
                 }
-                else if (GpuDriver.Contains("TITAN") && !GpuDriver.Contains("RTX"))
+                else if (GpuRenderer.Contains("TITAN") && !GpuRenderer.Contains("RTX"))
                 {
                     IsNvidiaPreTuring = true;
                 }
@@ -354,6 +388,8 @@ namespace Ryujinx.Graphics.Vulkan
                 IsIntelArc = GpuRenderer.StartsWith("Intel(R) Arc(TM)");
             }
 
+            IsQualcommProprietary = hasDriverProperties && driverProperties.DriverID == DriverId.QualcommProprietary;
+
             ulong minResourceAlignment = Math.Max(
                 Math.Max(
                     properties.Limits.MinStorageBufferOffsetAlignment,
@@ -397,6 +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,
@@ -411,7 +449,7 @@ namespace Ryujinx.Graphics.Vulkan
             Api.TryGetDeviceExtension(_instance.Instance, _device, out ExtExternalMemoryHost hostMemoryApi);
             HostMemoryAllocator = new HostMemoryAllocator(MemoryAllocator, Api, hostMemoryApi, _device);
 
-            CommandBufferPool = new CommandBufferPool(Api, _device, Queue, QueueLock, queueFamilyIndex);
+            CommandBufferPool = new CommandBufferPool(Api, _device, Queue, QueueLock, queueFamilyIndex, IsQualcommProprietary);
 
             PipelineLayoutCache = new PipelineLayoutCache();
 
@@ -688,7 +726,7 @@ namespace Ryujinx.Graphics.Vulkan
                 GpuVendor,
                 memoryType: memoryType,
                 hasFrontFacingBug: IsIntelWindows,
-                hasVectorIndexingBug: Vendor == Vendor.Qualcomm,
+                hasVectorIndexingBug: IsQualcommProprietary,
                 needsFragmentOutputSpecialization: IsMoltenVk,
                 reduceShaderPrecision: IsMoltenVk,
                 supportsAstcCompression: features2.Features.TextureCompressionAstcLdr && supportsAstcFormats,
@@ -935,6 +973,11 @@ namespace Ryujinx.Graphics.Vulkan
             ScreenCaptured?.Invoke(this, bitmap);
         }
 
+        public bool SupportsRenderPassBarrier(PipelineStageFlags flags)
+        {
+            return !(IsMoltenVk || IsQualcommProprietary);
+        }
+
         public unsafe void Dispose()
         {
             if (!_initialized)

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)
@@ -623,7 +623,8 @@ namespace Ryujinx.Graphics.Vulkan
 
         public override void SetSize(int width, int height)
         {
-            // Not needed as we can get the size from the surface.
+            // We don't need to use width and height as we can get the size from the surface.
+            _swapchainIsDirty = true;
         }
 
         public override void ChangeVSyncMode(bool vsyncEnabled)

src/Ryujinx.Gtk3/Program.cs

@@ -4,15 +4,17 @@ 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;
+using Ryujinx.UI.App.Common;
 using Ryujinx.UI.Common;
 using Ryujinx.UI.Common.Configuration;
 using Ryujinx.UI.Common.Helper;
 using Ryujinx.UI.Common.SystemInfo;
 using Ryujinx.UI.Widgets;
-using SixLabors.ImageSharp.Formats.Jpeg;
 using System;
 using System.Collections.Generic;
 using System.Diagnostics;
@@ -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");
-                setenv("GDK_BACKEND", "x11", 1);
+                OsUtils.SetEnvironmentVariableNoCaching("GDK_BACKEND", "x11");
             }
 
             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);
@@ -161,12 +155,6 @@ namespace Ryujinx
                 });
             };
 
-            // Sets ImageSharp Jpeg Encoder Quality.
-            SixLabors.ImageSharp.Configuration.Default.ImageFormatsManager.SetEncoder(JpegFormat.Instance, new JpegEncoder()
-            {
-                Quality = 100,
-            });
-
             string localConfigurationPath = Path.Combine(AppDomain.CurrentDomain.BaseDirectory, ReleaseInformation.ConfigName);
             string appDataConfigurationPath = Path.Combine(AppDataManager.BaseDirPath, ReleaseInformation.ConfigName);
 
@@ -236,9 +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();
@@ -255,6 +243,12 @@ namespace Ryujinx
             MainWindow mainWindow = new();
             mainWindow.Show();
 
+            // Load the game table if no application was requested by the command line
+            if (CommandLineState.LaunchPathArg == null)
+            {
+                mainWindow.UpdateGameTable();
+            }
+
             if (OperatingSystem.IsLinux())
             {
                 int currentVmMaxMapCount = LinuxHelper.VmMaxMapCount;
@@ -322,7 +316,35 @@ namespace Ryujinx
 
             if (CommandLineState.LaunchPathArg != null)
             {
-                mainWindow.RunApplication(CommandLineState.LaunchPathArg, CommandLineState.StartFullscreenArg);
+                if (mainWindow.ApplicationLibrary.TryGetApplicationsFromFile(CommandLineState.LaunchPathArg, out List<ApplicationData> applications))
+                {
+                    ApplicationData applicationData;
+
+                    if (CommandLineState.LaunchApplicationId != null)
+                    {
+                        applicationData = applications.Find(application => application.IdString == CommandLineState.LaunchApplicationId);
+
+                        if (applicationData != null)
+                        {
+                            mainWindow.RunApplication(applicationData, CommandLineState.StartFullscreenArg);
+                        }
+                        else
+                        {
+                            Logger.Error?.Print(LogClass.Application, $"Couldn't find requested application id '{CommandLineState.LaunchApplicationId}' in '{CommandLineState.LaunchPathArg}'.");
+                            UserErrorDialog.CreateUserErrorDialog(UserError.ApplicationNotFound);
+                        }
+                    }
+                    else
+                    {
+                        applicationData = applications[0];
+                        mainWindow.RunApplication(applicationData, CommandLineState.StartFullscreenArg);
+                    }
+                }
+                else
+                {
+                    Logger.Error?.Print(LogClass.Application, $"Couldn't find any application in '{CommandLineState.LaunchPathArg}'.");
+                    UserErrorDialog.CreateUserErrorDialog(UserError.ApplicationNotFound);
+                }
             }
 
             if (ConfigurationState.Instance.CheckUpdatesOnStart.Value && Updater.CanUpdate(false))

src/Ryujinx.Gtk3/Ryujinx.Gtk3.csproj

@@ -30,7 +30,6 @@
     <PackageReference Include="OpenTK.Graphics" />
     <PackageReference Include="SPB" />
     <PackageReference Include="SharpZipLib" />
-    <PackageReference Include="SixLabors.ImageSharp" />
   </ItemGroup>
 
   <ItemGroup>

src/Ryujinx.Gtk3/UI/MainWindow.cs

@@ -37,7 +37,9 @@ using Ryujinx.UI.Windows;
 using Silk.NET.Vulkan;
 using SPB.Graphics.Vulkan;
 using System;
+using System.Collections.Generic;
 using System.Diagnostics;
+using System.Globalization;
 using System.IO;
 using System.Reflection;
 using System.Threading;
@@ -60,7 +62,6 @@ namespace Ryujinx.UI
 
         private WindowsMultimediaTimerResolution _windowsMultimediaTimerResolution;
 
-        private readonly ApplicationLibrary _applicationLibrary;
         private readonly GtkHostUIHandler _uiHandler;
         private readonly AutoResetEvent _deviceExitStatus;
         private readonly ListStore _tableStore;
@@ -69,11 +70,12 @@ namespace Ryujinx.UI
         private bool _gameLoaded;
         private bool _ending;
 
-        private string _currentEmulatedGamePath = null;
+        private ApplicationData _currentApplicationData = null;
 
         private string _lastScannedAmiiboId = "";
         private bool _lastScannedAmiiboShowAll = false;
 
+        public readonly ApplicationLibrary ApplicationLibrary;
         public RendererWidgetBase RendererWidget;
         public InputManager InputManager;
 
@@ -180,8 +182,15 @@ namespace Ryujinx.UI
             _accountManager = new AccountManager(_libHacHorizonManager.RyujinxClient, CommandLineState.Profile);
             _userChannelPersistence = new UserChannelPersistence();
 
+            IntegrityCheckLevel checkLevel = ConfigurationState.Instance.System.EnableFsIntegrityChecks
+                ? IntegrityCheckLevel.ErrorOnInvalid
+                : IntegrityCheckLevel.None;
+
             // Instantiate GUI objects.
-            _applicationLibrary = new ApplicationLibrary(_virtualFileSystem);
+            ApplicationLibrary = new ApplicationLibrary(_virtualFileSystem, checkLevel)
+            {
+                DesiredLanguage = ConfigurationState.Instance.System.Language,
+            };
             _uiHandler = new GtkHostUIHandler(this);
             _deviceExitStatus = new AutoResetEvent(false);
 
@@ -190,8 +199,8 @@ namespace Ryujinx.UI
             FocusInEvent += MainWindow_FocusInEvent;
             FocusOutEvent += MainWindow_FocusOutEvent;
 
-            _applicationLibrary.ApplicationAdded += Application_Added;
-            _applicationLibrary.ApplicationCountUpdated += ApplicationCount_Updated;
+            ApplicationLibrary.ApplicationAdded += Application_Added;
+            ApplicationLibrary.ApplicationCountUpdated += ApplicationCount_Updated;
 
             _fileMenu.StateChanged += FileMenu_StateChanged;
             _actionMenu.StateChanged += ActionMenu_StateChanged;
@@ -319,7 +328,6 @@ namespace Ryujinx.UI
             _hideUI.Label = _hideUI.Label.Replace("SHOWUIKEY", ConfigurationState.Instance.Hid.Hotkeys.Value.ShowUI.ToString());
 
             UpdateColumns();
-            UpdateGameTable();
 
             ConfigurationState.Instance.UI.GameDirs.Event += (sender, args) =>
             {
@@ -732,7 +740,8 @@ namespace Ryujinx.UI
 
             Thread applicationLibraryThread = new(() =>
             {
-                _applicationLibrary.LoadApplications(ConfigurationState.Instance.UI.GameDirs, ConfigurationState.Instance.System.Language);
+                ApplicationLibrary.DesiredLanguage = ConfigurationState.Instance.System.Language;
+                ApplicationLibrary.LoadApplications(ConfigurationState.Instance.UI.GameDirs);
 
                 _updatingGameTable = false;
             })
@@ -783,7 +792,7 @@ namespace Ryujinx.UI
             }
         }
 
-        private bool LoadApplication(string path, bool isFirmwareTitle)
+        private bool LoadApplication(string path, ulong applicationId, bool isFirmwareTitle)
         {
             SystemVersion firmwareVersion = _contentManager.GetCurrentFirmwareVersion();
 
@@ -857,7 +866,7 @@ namespace Ryujinx.UI
                     case ".xci":
                         Logger.Info?.Print(LogClass.Application, "Loading as XCI.");
 
-                        return _emulationContext.LoadXci(path);
+                        return _emulationContext.LoadXci(path, applicationId);
                     case ".nca":
                         Logger.Info?.Print(LogClass.Application, "Loading as NCA.");
 
@@ -866,7 +875,7 @@ namespace Ryujinx.UI
                     case ".pfs0":
                         Logger.Info?.Print(LogClass.Application, "Loading as NSP.");
 
-                        return _emulationContext.LoadNsp(path);
+                        return _emulationContext.LoadNsp(path, applicationId);
                     default:
                         Logger.Info?.Print(LogClass.Application, "Loading as Homebrew.");
                         try
@@ -887,7 +896,7 @@ namespace Ryujinx.UI
             return false;
         }
 
-        public void RunApplication(string path, bool startFullscreen = false)
+        public void RunApplication(ApplicationData application, bool startFullscreen = false)
         {
             if (_gameLoaded)
             {
@@ -909,14 +918,14 @@ namespace Ryujinx.UI
 
                 bool isFirmwareTitle = false;
 
-                if (path.StartsWith("@SystemContent"))
+                if (application.Path.StartsWith("@SystemContent"))
                 {
-                    path = VirtualFileSystem.SwitchPathToSystemPath(path);
+                    application.Path = VirtualFileSystem.SwitchPathToSystemPath(application.Path);
 
                     isFirmwareTitle = true;
                 }
 
-                if (!LoadApplication(path, isFirmwareTitle))
+                if (!LoadApplication(application.Path, application.Id, isFirmwareTitle))
                 {
                     _emulationContext.Dispose();
                     SwitchToGameTable();
@@ -926,7 +935,7 @@ namespace Ryujinx.UI
 
                 SetupProgressUIHandlers();
 
-                _currentEmulatedGamePath = path;
+                _currentApplicationData = application;
 
                 _deviceExitStatus.Reset();
 
@@ -1165,7 +1174,7 @@ namespace Ryujinx.UI
                 _tableStore.AppendValues(
                     args.AppData.Favorite,
                     new Gdk.Pixbuf(args.AppData.Icon, 75, 75),
-                    $"{args.AppData.TitleName}\n{args.AppData.TitleId.ToUpper()}",
+                    $"{args.AppData.Name}\n{args.AppData.IdString.ToUpper()}",
                     args.AppData.Developer,
                     args.AppData.Version,
                     args.AppData.TimePlayedString,
@@ -1253,9 +1262,22 @@ namespace Ryujinx.UI
         {
             _gameTableSelection.GetSelected(out TreeIter treeIter);
 
-            string path = (string)_tableStore.GetValue(treeIter, 9);
+            ApplicationData application = new()
+            {
+                Favorite = (bool)_tableStore.GetValue(treeIter, 0),
+                Name = ((string)_tableStore.GetValue(treeIter, 2)).Split('\n')[0],
+                Id = ulong.Parse(((string)_tableStore.GetValue(treeIter, 2)).Split('\n')[1], NumberStyles.HexNumber),
+                Developer = (string)_tableStore.GetValue(treeIter, 3),
+                Version = (string)_tableStore.GetValue(treeIter, 4),
+                TimePlayed = ValueFormatUtils.ParseTimeSpan((string)_tableStore.GetValue(treeIter, 5)),
+                LastPlayed = ValueFormatUtils.ParseDateTime((string)_tableStore.GetValue(treeIter, 6)),
+                FileExtension = (string)_tableStore.GetValue(treeIter, 7),
+                FileSize = ValueFormatUtils.ParseFileSize((string)_tableStore.GetValue(treeIter, 8)),
+                Path = (string)_tableStore.GetValue(treeIter, 9),
+                ControlHolder = (BlitStruct<ApplicationControlProperty>)_tableStore.GetValue(treeIter, 10),
+            };
 
-            RunApplication(path);
+            RunApplication(application);
         }
 
         private void VSyncStatus_Clicked(object sender, ButtonReleaseEventArgs args)
@@ -1313,13 +1335,22 @@ namespace Ryujinx.UI
                 return;
             }
 
-            string titleFilePath = _tableStore.GetValue(treeIter, 9).ToString();
-            string titleName = _tableStore.GetValue(treeIter, 2).ToString().Split("\n")[0];
-            string titleId = _tableStore.GetValue(treeIter, 2).ToString().Split("\n")[1].ToLower();
+            ApplicationData application = new()
+            {
+                Favorite = (bool)_tableStore.GetValue(treeIter, 0),
+                Name = ((string)_tableStore.GetValue(treeIter, 2)).Split('\n')[0],
+                Id = ulong.Parse(((string)_tableStore.GetValue(treeIter, 2)).Split('\n')[1], NumberStyles.HexNumber),
+                Developer = (string)_tableStore.GetValue(treeIter, 3),
+                Version = (string)_tableStore.GetValue(treeIter, 4),
+                TimePlayed = ValueFormatUtils.ParseTimeSpan((string)_tableStore.GetValue(treeIter, 5)),
+                LastPlayed = ValueFormatUtils.ParseDateTime((string)_tableStore.GetValue(treeIter, 6)),
+                FileExtension = (string)_tableStore.GetValue(treeIter, 7),
+                FileSize = ValueFormatUtils.ParseFileSize((string)_tableStore.GetValue(treeIter, 8)),
+                Path = (string)_tableStore.GetValue(treeIter, 9),
+                ControlHolder = (BlitStruct<ApplicationControlProperty>)_tableStore.GetValue(treeIter, 10),
+            };
 
-            BlitStruct<ApplicationControlProperty> controlData = (BlitStruct<ApplicationControlProperty>)_tableStore.GetValue(treeIter, 10);
-
-            _ = new GameTableContextMenu(this, _virtualFileSystem, _accountManager, _libHacHorizonManager.RyujinxClient, titleFilePath, titleName, titleId, controlData);
+            _ = new GameTableContextMenu(this, _virtualFileSystem, _accountManager, _libHacHorizonManager.RyujinxClient, application);
         }
 
         private void Load_Application_File(object sender, EventArgs args)
@@ -1341,7 +1372,15 @@ namespace Ryujinx.UI
 
             if (fileChooser.Run() == (int)ResponseType.Accept)
             {
-                RunApplication(fileChooser.Filename);
+                if (ApplicationLibrary.TryGetApplicationsFromFile(fileChooser.Filename,
+                        out List<ApplicationData> applications))
+                {
+                    RunApplication(applications[0]);
+                }
+                else
+                {
+                    GtkDialog.CreateErrorDialog("No applications found in selected file.");
+                }
             }
         }
 
@@ -1351,7 +1390,13 @@ namespace Ryujinx.UI
 
             if (fileChooser.Run() == (int)ResponseType.Accept)
             {
-                RunApplication(fileChooser.Filename);
+                ApplicationData applicationData = new()
+                {
+                    Name = System.IO.Path.GetFileNameWithoutExtension(fileChooser.Filename),
+                    Path = fileChooser.Filename,
+                };
+
+                RunApplication(applicationData);
             }
         }
 
@@ -1366,7 +1411,14 @@ namespace Ryujinx.UI
         {
             string contentPath = _contentManager.GetInstalledContentPath(0x0100000000001009, StorageId.BuiltInSystem, NcaContentType.Program);
 
-            RunApplication(contentPath);
+            ApplicationData applicationData = new()
+            {
+                Name = "miiEdit",
+                Id = 0x0100000000001009ul,
+                Path = contentPath,
+            };
+
+            RunApplication(applicationData);
         }
 
         private void Open_Ryu_Folder(object sender, EventArgs args)
@@ -1646,13 +1698,13 @@ namespace Ryujinx.UI
             {
                 _userChannelPersistence.ShouldRestart = false;
 
-                RunApplication(_currentEmulatedGamePath);
+                RunApplication(_currentApplicationData);
             }
             else
             {
                 // otherwise, clear state.
                 _userChannelPersistence = new UserChannelPersistence();
-                _currentEmulatedGamePath = null;
+                _currentApplicationData = null;
                 _actionMenu.Sensitive = false;
                 _firmwareInstallFile.Sensitive = true;
                 _firmwareInstallDirectory.Sensitive = true;
@@ -1714,7 +1766,7 @@ namespace Ryujinx.UI
                 _emulationContext.Processes.ActiveApplication.ProgramId,
                 _emulationContext.Processes.ActiveApplication.ApplicationControlProperties
                     .Title[(int)_emulationContext.System.State.DesiredTitleLanguage].NameString.ToString(),
-                _currentEmulatedGamePath);
+                _currentApplicationData.Path);
 
             window.Destroyed += CheatWindow_Destroyed;
             window.Show();

src/Ryujinx.Gtk3/UI/RendererWidgetBase.cs

@@ -13,16 +13,13 @@ using Ryujinx.Input.HLE;
 using Ryujinx.UI.Common.Configuration;
 using Ryujinx.UI.Common.Helper;
 using Ryujinx.UI.Widgets;
-using SixLabors.ImageSharp;
-using SixLabors.ImageSharp.Formats.Png;
-using SixLabors.ImageSharp.PixelFormats;
-using SixLabors.ImageSharp.Processing;
+using SkiaSharp;
 using System;
 using System.Diagnostics;
 using System.IO;
+using System.Runtime.InteropServices;
 using System.Threading;
 using System.Threading.Tasks;
-using Image = SixLabors.ImageSharp.Image;
 using Key = Ryujinx.Input.Key;
 using ScalingFilter = Ryujinx.Graphics.GAL.ScalingFilter;
 using Switch = Ryujinx.HLE.Switch;
@@ -404,23 +401,31 @@ namespace Ryujinx.UI
                             return;
                         }
 
-                        Image image = e.IsBgra ? Image.LoadPixelData<Bgra32>(e.Data, e.Width, e.Height)
-                                               : Image.LoadPixelData<Rgba32>(e.Data, e.Width, e.Height);
+                        var colorType = e.IsBgra ? SKColorType.Bgra8888 : SKColorType.Rgba8888;
+                        using var image = new SKBitmap(new SKImageInfo(e.Width, e.Height, colorType, SKAlphaType.Premul));
 
-                        if (e.FlipX)
+                        Marshal.Copy(e.Data, 0, image.GetPixels(), e.Data.Length);
+                        using var surface = SKSurface.Create(image.Info);
+                        var canvas = surface.Canvas;
+
+                        if (e.FlipX || e.FlipY)
                         {
-                            image.Mutate(x => x.Flip(FlipMode.Horizontal));
+                            canvas.Clear(SKColors.Transparent);
+
+                            float scaleX = e.FlipX ? -1 : 1;
+                            float scaleY = e.FlipY ? -1 : 1;
+
+                            var matrix = SKMatrix.CreateScale(scaleX, scaleY, image.Width / 2f, image.Height / 2f);
+
+                            canvas.SetMatrix(matrix);
                         }
+                        canvas.DrawBitmap(image, new SKPoint());
 
-                        if (e.FlipY)
-                        {
-                            image.Mutate(x => x.Flip(FlipMode.Vertical));
-                        }
-
-                        image.SaveAsPng(path, new PngEncoder()
-                        {
-                            ColorType = PngColorType.Rgb,
-                        });
+                        surface.Flush();
+                        using var snapshot = surface.Snapshot();
+                        using var encoded = snapshot.Encode(SKEncodedImageFormat.Png, 80);
+                        using var file = File.OpenWrite(path);
+                        encoded.SaveTo(file);
 
                         image.Dispose();
 

src/Ryujinx.Gtk3/UI/Widgets/GameTableContextMenu.cs

@@ -16,6 +16,8 @@ using Ryujinx.Common.Logging;
 using Ryujinx.HLE.FileSystem;
 using Ryujinx.HLE.HOS;
 using Ryujinx.HLE.HOS.Services.Account.Acc;
+using Ryujinx.HLE.Loaders.Processes.Extensions;
+using Ryujinx.HLE.Utilities;
 using Ryujinx.UI.App.Common;
 using Ryujinx.UI.Common.Configuration;
 using Ryujinx.UI.Common.Helper;
@@ -23,7 +25,6 @@ using Ryujinx.UI.Windows;
 using System;
 using System.Buffers;
 using System.Collections.Generic;
-using System.Globalization;
 using System.IO;
 using System.Reflection;
 using System.Threading;
@@ -36,17 +37,13 @@ namespace Ryujinx.UI.Widgets
         private readonly VirtualFileSystem _virtualFileSystem;
         private readonly AccountManager _accountManager;
         private readonly HorizonClient _horizonClient;
-        private readonly BlitStruct<ApplicationControlProperty> _controlData;
 
-        private readonly string _titleFilePath;
-        private readonly string _titleName;
-        private readonly string _titleIdText;
-        private readonly ulong _titleId;
+        private readonly ApplicationData _applicationData;
 
         private MessageDialog _dialog;
         private bool _cancel;
 
-        public GameTableContextMenu(MainWindow parent, VirtualFileSystem virtualFileSystem, AccountManager accountManager, HorizonClient horizonClient, string titleFilePath, string titleName, string titleId, BlitStruct<ApplicationControlProperty> controlData)
+        public GameTableContextMenu(MainWindow parent, VirtualFileSystem virtualFileSystem, AccountManager accountManager, HorizonClient horizonClient, ApplicationData applicationData)
         {
             _parent = parent;
 
@@ -55,23 +52,22 @@ namespace Ryujinx.UI.Widgets
             _virtualFileSystem = virtualFileSystem;
             _accountManager = accountManager;
             _horizonClient = horizonClient;
-            _titleFilePath = titleFilePath;
-            _titleName = titleName;
-            _titleIdText = titleId;
-            _controlData = controlData;
+            _applicationData = applicationData;
 
-            if (!ulong.TryParse(_titleIdText, NumberStyles.HexNumber, CultureInfo.InvariantCulture, out _titleId))
+            if (!_applicationData.ControlHolder.ByteSpan.IsZeros())
             {
-                GtkDialog.CreateErrorDialog("The selected game did not have a valid Title Id");
-
-                return;
+                _openSaveUserDirMenuItem.Sensitive = _applicationData.ControlHolder.Value.UserAccountSaveDataSize > 0;
+                _openSaveDeviceDirMenuItem.Sensitive = _applicationData.ControlHolder.Value.DeviceSaveDataSize > 0;
+                _openSaveBcatDirMenuItem.Sensitive = _applicationData.ControlHolder.Value.BcatDeliveryCacheStorageSize > 0;
+            }
+            else
+            {
+                _openSaveUserDirMenuItem.Sensitive = false;
+                _openSaveDeviceDirMenuItem.Sensitive = false;
+                _openSaveBcatDirMenuItem.Sensitive = false;
             }
 
-            _openSaveUserDirMenuItem.Sensitive = !Utilities.IsZeros(controlData.ByteSpan) && controlData.Value.UserAccountSaveDataSize > 0;
-            _openSaveDeviceDirMenuItem.Sensitive = !Utilities.IsZeros(controlData.ByteSpan) && controlData.Value.DeviceSaveDataSize > 0;
-            _openSaveBcatDirMenuItem.Sensitive = !Utilities.IsZeros(controlData.ByteSpan) && controlData.Value.BcatDeliveryCacheStorageSize > 0;
-
-            string fileExt = System.IO.Path.GetExtension(_titleFilePath).ToLower();
+            string fileExt = System.IO.Path.GetExtension(_applicationData.Path).ToLower();
             bool hasNca = fileExt == ".nca" || fileExt == ".nsp" || fileExt == ".pfs0" || fileExt == ".xci";
 
             _extractRomFsMenuItem.Sensitive = hasNca;
@@ -137,7 +133,7 @@ namespace Ryujinx.UI.Widgets
 
         private void OpenSaveDir(in SaveDataFilter saveDataFilter)
         {
-            if (!TryFindSaveData(_titleName, _titleId, _controlData, in saveDataFilter, out ulong saveDataId))
+            if (!TryFindSaveData(_applicationData.Name, _applicationData.Id, _applicationData.ControlHolder, in saveDataFilter, out ulong saveDataId))
             {
                 return;
             }
@@ -190,7 +186,7 @@ namespace Ryujinx.UI.Widgets
                         {
                             Title = "Ryujinx - NCA Section Extractor",
                             Icon = new Gdk.Pixbuf(Assembly.GetAssembly(typeof(ConfigurationState)), "Ryujinx.Gtk3.UI.Common.Resources.Logo_Ryujinx.png"),
-                            SecondaryText = $"Extracting {ncaSectionType} section from {System.IO.Path.GetFileName(_titleFilePath)}...",
+                            SecondaryText = $"Extracting {ncaSectionType} section from {System.IO.Path.GetFileName(_applicationData.Path)}...",
                             WindowPosition = WindowPosition.Center,
                         };
 
@@ -202,29 +198,16 @@ namespace Ryujinx.UI.Widgets
                         }
                     });
 
-                    using FileStream file = new(_titleFilePath, FileMode.Open, FileAccess.Read);
+                    using FileStream file = new(_applicationData.Path, FileMode.Open, FileAccess.Read);
 
                     Nca mainNca = null;
                     Nca patchNca = null;
 
-                    if ((System.IO.Path.GetExtension(_titleFilePath).ToLower() == ".nsp") ||
-                        (System.IO.Path.GetExtension(_titleFilePath).ToLower() == ".pfs0") ||
-                        (System.IO.Path.GetExtension(_titleFilePath).ToLower() == ".xci"))
+                    if ((System.IO.Path.GetExtension(_applicationData.Path).ToLower() == ".nsp") ||
+                        (System.IO.Path.GetExtension(_applicationData.Path).ToLower() == ".pfs0") ||
+                        (System.IO.Path.GetExtension(_applicationData.Path).ToLower() == ".xci"))
                     {
-                        IFileSystem pfs;
-
-                        if (System.IO.Path.GetExtension(_titleFilePath) == ".xci")
-                        {
-                            Xci xci = new(_virtualFileSystem.KeySet, file.AsStorage());
-
-                            pfs = xci.OpenPartition(XciPartitionType.Secure);
-                        }
-                        else
-                        {
-                            var pfsTemp = new PartitionFileSystem();
-                            pfsTemp.Initialize(file.AsStorage()).ThrowIfFailure();
-                            pfs = pfsTemp;
-                        }
+                        IFileSystem pfs = PartitionFileSystemUtils.OpenApplicationFileSystem(_applicationData.Path, _virtualFileSystem);
 
                         foreach (DirectoryEntryEx fileEntry in pfs.EnumerateEntries("/", "*.nca"))
                         {
@@ -249,7 +232,7 @@ namespace Ryujinx.UI.Widgets
                             }
                         }
                     }
-                    else if (System.IO.Path.GetExtension(_titleFilePath).ToLower() == ".nca")
+                    else if (System.IO.Path.GetExtension(_applicationData.Path).ToLower() == ".nca")
                     {
                         mainNca = new Nca(_virtualFileSystem.KeySet, file.AsStorage());
                     }
@@ -266,7 +249,11 @@ namespace Ryujinx.UI.Widgets
                         return;
                     }
 
-                    (Nca updatePatchNca, _) = ApplicationLibrary.GetGameUpdateData(_virtualFileSystem, mainNca.Header.TitleId.ToString("x16"), programIndex, out _);
+                    IntegrityCheckLevel checkLevel = ConfigurationState.Instance.System.EnableFsIntegrityChecks
+                        ? IntegrityCheckLevel.ErrorOnInvalid
+                        : IntegrityCheckLevel.None;
+
+                    (Nca updatePatchNca, _) = mainNca.GetUpdateData(_virtualFileSystem, checkLevel, programIndex, out _);
 
                     if (updatePatchNca != null)
                     {
@@ -460,44 +447,44 @@ namespace Ryujinx.UI.Widgets
         private void OpenSaveUserDir_Clicked(object sender, EventArgs args)
         {
             var userId = new LibHac.Fs.UserId((ulong)_accountManager.LastOpenedUser.UserId.High, (ulong)_accountManager.LastOpenedUser.UserId.Low);
-            var saveDataFilter = SaveDataFilter.Make(_titleId, saveType: default, userId, saveDataId: default, index: default);
+            var saveDataFilter = SaveDataFilter.Make(_applicationData.Id, saveType: default, userId, saveDataId: default, index: default);
 
             OpenSaveDir(in saveDataFilter);
         }
 
         private void OpenSaveDeviceDir_Clicked(object sender, EventArgs args)
         {
-            var saveDataFilter = SaveDataFilter.Make(_titleId, SaveDataType.Device, userId: default, saveDataId: default, index: default);
+            var saveDataFilter = SaveDataFilter.Make(_applicationData.Id, SaveDataType.Device, userId: default, saveDataId: default, index: default);
 
             OpenSaveDir(in saveDataFilter);
         }
 
         private void OpenSaveBcatDir_Clicked(object sender, EventArgs args)
         {
-            var saveDataFilter = SaveDataFilter.Make(_titleId, SaveDataType.Bcat, userId: default, saveDataId: default, index: default);
+            var saveDataFilter = SaveDataFilter.Make(_applicationData.Id, SaveDataType.Bcat, userId: default, saveDataId: default, index: default);
 
             OpenSaveDir(in saveDataFilter);
         }
 
         private void ManageTitleUpdates_Clicked(object sender, EventArgs args)
         {
-            new TitleUpdateWindow(_parent, _virtualFileSystem, _titleIdText, _titleName).Show();
+            new TitleUpdateWindow(_parent, _virtualFileSystem, _applicationData).Show();
         }
 
         private void ManageDlc_Clicked(object sender, EventArgs args)
         {
-            new DlcWindow(_virtualFileSystem, _titleIdText, _titleName).Show();
+            new DlcWindow(_virtualFileSystem, _applicationData.IdBaseString, _applicationData).Show();
         }
 
         private void ManageCheats_Clicked(object sender, EventArgs args)
         {
-            new CheatWindow(_virtualFileSystem, _titleId, _titleName, _titleFilePath).Show();
+            new CheatWindow(_virtualFileSystem, _applicationData.Id, _applicationData.Name, _applicationData.Path).Show();
         }
 
         private void OpenTitleModDir_Clicked(object sender, EventArgs args)
         {
             string modsBasePath = ModLoader.GetModsBasePath();
-            string titleModsPath = ModLoader.GetApplicationDir(modsBasePath, _titleIdText);
+            string titleModsPath = ModLoader.GetApplicationDir(modsBasePath, _applicationData.IdString);
 
             OpenHelper.OpenFolder(titleModsPath);
         }
@@ -505,7 +492,7 @@ namespace Ryujinx.UI.Widgets
         private void OpenTitleSdModDir_Clicked(object sender, EventArgs args)
         {
             string sdModsBasePath = ModLoader.GetSdModsBasePath();
-            string titleModsPath = ModLoader.GetApplicationDir(sdModsBasePath, _titleIdText);
+            string titleModsPath = ModLoader.GetApplicationDir(sdModsBasePath, _applicationData.IdString);
 
             OpenHelper.OpenFolder(titleModsPath);
         }
@@ -527,7 +514,7 @@ namespace Ryujinx.UI.Widgets
 
         private void OpenPtcDir_Clicked(object sender, EventArgs args)
         {
-            string ptcDir = System.IO.Path.Combine(AppDataManager.GamesDirPath, _titleIdText, "cache", "cpu");
+            string ptcDir = System.IO.Path.Combine(AppDataManager.GamesDirPath, _applicationData.IdString, "cache", "cpu");
 
             string mainPath = System.IO.Path.Combine(ptcDir, "0");
             string backupPath = System.IO.Path.Combine(ptcDir, "1");
@@ -544,7 +531,7 @@ namespace Ryujinx.UI.Widgets
 
         private void OpenShaderCacheDir_Clicked(object sender, EventArgs args)
         {
-            string shaderCacheDir = System.IO.Path.Combine(AppDataManager.GamesDirPath, _titleIdText, "cache", "shader");
+            string shaderCacheDir = System.IO.Path.Combine(AppDataManager.GamesDirPath, _applicationData.IdString, "cache", "shader");
 
             if (!Directory.Exists(shaderCacheDir))
             {
@@ -556,10 +543,10 @@ namespace Ryujinx.UI.Widgets
 
         private void PurgePtcCache_Clicked(object sender, EventArgs args)
         {
-            DirectoryInfo mainDir = new(System.IO.Path.Combine(AppDataManager.GamesDirPath, _titleIdText, "cache", "cpu", "0"));
-            DirectoryInfo backupDir = new(System.IO.Path.Combine(AppDataManager.GamesDirPath, _titleIdText, "cache", "cpu", "1"));
+            DirectoryInfo mainDir = new(System.IO.Path.Combine(AppDataManager.GamesDirPath, _applicationData.IdString, "cache", "cpu", "0"));
+            DirectoryInfo backupDir = new(System.IO.Path.Combine(AppDataManager.GamesDirPath, _applicationData.IdString, "cache", "cpu", "1"));
 
-            MessageDialog warningDialog = GtkDialog.CreateConfirmationDialog("Warning", $"You are about to queue a PPTC rebuild on the next boot of:\n\n<b>{_titleName}</b>\n\nAre you sure you want to proceed?");
+            MessageDialog warningDialog = GtkDialog.CreateConfirmationDialog("Warning", $"You are about to queue a PPTC rebuild on the next boot of:\n\n<b>{_applicationData.Name}</b>\n\nAre you sure you want to proceed?");
 
             List<FileInfo> cacheFiles = new();
 
@@ -593,9 +580,9 @@ namespace Ryujinx.UI.Widgets
 
         private void PurgeShaderCache_Clicked(object sender, EventArgs args)
         {
-            DirectoryInfo shaderCacheDir = new(System.IO.Path.Combine(AppDataManager.GamesDirPath, _titleIdText, "cache", "shader"));
+            DirectoryInfo shaderCacheDir = new(System.IO.Path.Combine(AppDataManager.GamesDirPath, _applicationData.IdString, "cache", "shader"));
 
-            using MessageDialog warningDialog = GtkDialog.CreateConfirmationDialog("Warning", $"You are about to delete the shader cache for :\n\n<b>{_titleName}</b>\n\nAre you sure you want to proceed?");
+            using MessageDialog warningDialog = GtkDialog.CreateConfirmationDialog("Warning", $"You are about to delete the shader cache for :\n\n<b>{_applicationData.Name}</b>\n\nAre you sure you want to proceed?");
 
             List<DirectoryInfo> oldCacheDirectories = new();
             List<FileInfo> newCacheFiles = new();
@@ -637,8 +624,11 @@ namespace Ryujinx.UI.Widgets
 
         private void CreateShortcut_Clicked(object sender, EventArgs args)
         {
-            byte[] appIcon = new ApplicationLibrary(_virtualFileSystem).GetApplicationIcon(_titleFilePath, ConfigurationState.Instance.System.Language);
-            ShortcutHelper.CreateAppShortcut(_titleFilePath, _titleName, _titleIdText, appIcon);
+            IntegrityCheckLevel checkLevel = ConfigurationState.Instance.System.EnableFsIntegrityChecks
+                ? IntegrityCheckLevel.ErrorOnInvalid
+                : IntegrityCheckLevel.None;
+            byte[] appIcon = new ApplicationLibrary(_virtualFileSystem, checkLevel).GetApplicationIcon(_applicationData.Path, ConfigurationState.Instance.System.Language, _applicationData.Id);
+            ShortcutHelper.CreateAppShortcut(_applicationData.Path, _applicationData.Name, _applicationData.IdString, appIcon);
         }
     }
 }

src/Ryujinx.Gtk3/UI/Windows/AvatarWindow.cs

@@ -9,16 +9,13 @@ using LibHac.Tools.FsSystem.NcaUtils;
 using Ryujinx.Common.Memory;
 using Ryujinx.HLE.FileSystem;
 using Ryujinx.UI.Common.Configuration;
-using SixLabors.ImageSharp;
-using SixLabors.ImageSharp.Formats.Png;
-using SixLabors.ImageSharp.PixelFormats;
-using SixLabors.ImageSharp.Processing;
+using SkiaSharp;
 using System;
 using System.Buffers.Binary;
 using System.Collections.Generic;
 using System.IO;
 using System.Reflection;
-using Image = SixLabors.ImageSharp.Image;
+using System.Runtime.InteropServices;
 
 namespace Ryujinx.UI.Windows
 {
@@ -144,9 +141,11 @@ namespace Ryujinx.UI.Windows
 
                         stream.Position = 0;
 
-                        Image avatarImage = Image.LoadPixelData<Rgba32>(DecompressYaz0(stream), 256, 256);
+                        using var avatarImage = new SKBitmap(new SKImageInfo(256, 256, SKColorType.Rgba8888));
+                        var data = DecompressYaz0(stream);
+                        Marshal.Copy(data, 0, avatarImage.GetPixels(), data.Length);
 
-                        avatarImage.SaveAsPng(streamPng);
+                        avatarImage.Encode(streamPng, SKEncodedImageFormat.Png, 80);
 
                         _avatarDict.Add(item.FullPath, streamPng.ToArray());
                     }
@@ -170,15 +169,23 @@ namespace Ryujinx.UI.Windows
         {
             using MemoryStream streamJpg = MemoryStreamManager.Shared.GetStream();
 
-            Image avatarImage = Image.Load(data, new PngDecoder());
+            using var avatarImage = SKBitmap.Decode(data);
+            using var surface = SKSurface.Create(avatarImage.Info);
 
-            avatarImage.Mutate(x => x.BackgroundColor(new Rgba32(
+            var background = new SKColor(
                 (byte)(_backgroundColor.Red * 255),
                 (byte)(_backgroundColor.Green * 255),
                 (byte)(_backgroundColor.Blue * 255),
                 (byte)(_backgroundColor.Alpha * 255)
-            )));
-            avatarImage.SaveAsJpeg(streamJpg);
+            );
+            var canvas = surface.Canvas;
+            canvas.Clear(background);
+            canvas.DrawBitmap(avatarImage, new SKPoint());
+
+            surface.Flush();
+            using var snapshot = surface.Snapshot();
+            using var encoded = snapshot.Encode(SKEncodedImageFormat.Jpeg, 80);
+            encoded.SaveTo(streamJpg);
 
             return streamJpg.ToArray();
         }