diff options
| author | gdkchan <gab.dark.100@gmail.com> | 2024-04-07 18:25:55 -0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-04-07 18:25:55 -0300 |
| commit | 3e6e0e4afaa3c3ffb118cb17b61feb16966a7eeb (patch) | |
| tree | a4652499c089b0853e39c382cad82a9db4d6ad08 /src/Ryujinx.Graphics.Shader/Translation/Optimizations/BindlessToArray.cs | |
| parent | 808803d97a0c06809bf000687c252f960048fcf0 (diff) | |
Add support for large sampler arrays on Vulkan (#6489)
* Add support for large sampler arrays on Vulkan
* Shader cache version bump
* Format whitespace
* Move DescriptorSetManager to PipelineLayoutCacheEntry to allow different pool sizes per layout
* Handle array textures with different types on the same buffer
* Somewhat better caching system
* Avoid useless buffer data modification checks
* Move redundant bindings update checking to the backend
* Fix an issue where texture arrays would get the same bindings across stages on Vulkan
* Backport some fixes from part 2
* Fix typo
* PR feedback
* Format whitespace
* Add some missing XML docs
Diffstat (limited to 'src/Ryujinx.Graphics.Shader/Translation/Optimizations/BindlessToArray.cs')
| -rw-r--r-- | src/Ryujinx.Graphics.Shader/Translation/Optimizations/BindlessToArray.cs | 236 |
1 files changed, 236 insertions, 0 deletions
diff --git a/src/Ryujinx.Graphics.Shader/Translation/Optimizations/BindlessToArray.cs b/src/Ryujinx.Graphics.Shader/Translation/Optimizations/BindlessToArray.cs new file mode 100644 index 00000000..7543d1c2 --- /dev/null +++ b/src/Ryujinx.Graphics.Shader/Translation/Optimizations/BindlessToArray.cs @@ -0,0 +1,236 @@ +using Ryujinx.Graphics.Shader.IntermediateRepresentation; +using System; +using System.Collections.Generic; +using static Ryujinx.Graphics.Shader.IntermediateRepresentation.OperandHelper; + +namespace Ryujinx.Graphics.Shader.Translation.Optimizations +{ + static class BindlessToArray + { + private const int NvnTextureBufferIndex = 2; + private const int HardcodedArrayLengthOgl = 4; + + // 1 and 0 elements are not considered arrays anymore. + private const int MinimumArrayLength = 2; + + public static void RunPassOgl(BasicBlock block, ResourceManager resourceManager) + { + // We can turn a bindless texture access into a indexed access, + // as long the following conditions are true: + // - The handle is loaded using a LDC instruction. + // - The handle is loaded from the constant buffer with the handles (CB2 for NVN). + // - The load has a constant offset. + // The base offset of the array of handles on the constant buffer is the constant offset. + for (LinkedListNode<INode> node = block.Operations.First; node != null; node = node.Next) + { + if (node.Value is not TextureOperation texOp) + { + continue; + } + + if ((texOp.Flags & TextureFlags.Bindless) == 0) + { + continue; + } + + if (texOp.GetSource(0).AsgOp is not Operation handleAsgOp) + { + continue; + } + + if (handleAsgOp.Inst != Instruction.Load || + handleAsgOp.StorageKind != StorageKind.ConstantBuffer || + handleAsgOp.SourcesCount != 4) + { + continue; + } + + Operand ldcSrc0 = handleAsgOp.GetSource(0); + + if (ldcSrc0.Type != OperandType.Constant || + !resourceManager.TryGetConstantBufferSlot(ldcSrc0.Value, out int src0CbufSlot) || + src0CbufSlot != NvnTextureBufferIndex) + { + continue; + } + + Operand ldcSrc1 = handleAsgOp.GetSource(1); + + // We expect field index 0 to be accessed. + if (ldcSrc1.Type != OperandType.Constant || ldcSrc1.Value != 0) + { + continue; + } + + Operand ldcSrc2 = handleAsgOp.GetSource(2); + + // FIXME: This is missing some checks, for example, a check to ensure that the shift value is 2. + // Might be not worth fixing since if that doesn't kick in, the result will be no texture + // to access anyway which is also wrong. + // Plus this whole transform is fundamentally flawed as-is since we have no way to know the array size. + // Eventually, this should be entirely removed in favor of a implementation that supports true bindless + // texture access. + if (ldcSrc2.AsgOp is not Operation shrOp || shrOp.Inst != Instruction.ShiftRightU32) + { + continue; + } + + if (shrOp.GetSource(0).AsgOp is not Operation shrOp2 || shrOp2.Inst != Instruction.ShiftRightU32) + { + continue; + } + + if (shrOp2.GetSource(0).AsgOp is not Operation addOp || addOp.Inst != Instruction.Add) + { + continue; + } + + Operand addSrc1 = addOp.GetSource(1); + + if (addSrc1.Type != OperandType.Constant) + { + continue; + } + + TurnIntoArray(resourceManager, texOp, NvnTextureBufferIndex, addSrc1.Value / 4, HardcodedArrayLengthOgl); + + Operand index = Local(); + + Operand source = addOp.GetSource(0); + + Operation shrBy3 = new(Instruction.ShiftRightU32, index, source, Const(3)); + + block.Operations.AddBefore(node, shrBy3); + + texOp.SetSource(0, index); + } + } + + public static void RunPass(BasicBlock block, ResourceManager resourceManager, IGpuAccessor gpuAccessor) + { + // We can turn a bindless texture access into a indexed access, + // as long the following conditions are true: + // - The handle is loaded using a LDC instruction. + // - The handle is loaded from the constant buffer with the handles (CB2 for NVN). + // - The load has a constant offset. + // The base offset of the array of handles on the constant buffer is the constant offset. + for (LinkedListNode<INode> node = block.Operations.First; node != null; node = node.Next) + { + if (node.Value is not TextureOperation texOp) + { + continue; + } + + if ((texOp.Flags & TextureFlags.Bindless) == 0) + { + continue; + } + + if (texOp.GetSource(0).AsgOp is not Operation handleAsgOp) + { + continue; + } + + int secondaryCbufSlot = 0; + int secondaryCbufOffset = 0; + bool hasSecondaryHandle = false; + + if (handleAsgOp.Inst == Instruction.BitwiseOr) + { + Operand src0 = handleAsgOp.GetSource(0); + Operand src1 = handleAsgOp.GetSource(1); + + if (src0.Type == OperandType.ConstantBuffer && src1.AsgOp is Operation) + { + handleAsgOp = src1.AsgOp as Operation; + secondaryCbufSlot = src0.GetCbufSlot(); + secondaryCbufOffset = src0.GetCbufOffset(); + hasSecondaryHandle = true; + } + else if (src0.AsgOp is Operation && src1.Type == OperandType.ConstantBuffer) + { + handleAsgOp = src0.AsgOp as Operation; + secondaryCbufSlot = src1.GetCbufSlot(); + secondaryCbufOffset = src1.GetCbufOffset(); + hasSecondaryHandle = true; + } + } + + if (handleAsgOp.Inst != Instruction.Load || + handleAsgOp.StorageKind != StorageKind.ConstantBuffer || + handleAsgOp.SourcesCount != 4) + { + continue; + } + + Operand ldcSrc0 = handleAsgOp.GetSource(0); + + if (ldcSrc0.Type != OperandType.Constant || + !resourceManager.TryGetConstantBufferSlot(ldcSrc0.Value, out int src0CbufSlot)) + { + continue; + } + + Operand ldcSrc1 = handleAsgOp.GetSource(1); + + // We expect field index 0 to be accessed. + if (ldcSrc1.Type != OperandType.Constant || ldcSrc1.Value != 0) + { + continue; + } + + Operand ldcVecIndex = handleAsgOp.GetSource(2); + Operand ldcElemIndex = handleAsgOp.GetSource(3); + + if (ldcVecIndex.Type != OperandType.LocalVariable || ldcElemIndex.Type != OperandType.LocalVariable) + { + continue; + } + + int cbufSlot; + int handleIndex; + + if (hasSecondaryHandle) + { + cbufSlot = TextureHandle.PackSlots(src0CbufSlot, secondaryCbufSlot); + handleIndex = TextureHandle.PackOffsets(0, secondaryCbufOffset, TextureHandleType.SeparateSamplerHandle); + } + else + { + cbufSlot = src0CbufSlot; + handleIndex = 0; + } + + int length = Math.Max(MinimumArrayLength, gpuAccessor.QueryTextureArrayLengthFromBuffer(src0CbufSlot)); + + TurnIntoArray(resourceManager, texOp, cbufSlot, handleIndex, length); + + Operand vecIndex = Local(); + Operand elemIndex = Local(); + Operand index = Local(); + Operand indexMin = Local(); + + block.Operations.AddBefore(node, new Operation(Instruction.ShiftLeft, vecIndex, ldcVecIndex, Const(1))); + block.Operations.AddBefore(node, new Operation(Instruction.ShiftRightU32, elemIndex, ldcElemIndex, Const(1))); + block.Operations.AddBefore(node, new Operation(Instruction.Add, index, vecIndex, elemIndex)); + block.Operations.AddBefore(node, new Operation(Instruction.MinimumU32, indexMin, index, Const(length - 1))); + + texOp.SetSource(0, indexMin); + } + } + + private static void TurnIntoArray(ResourceManager resourceManager, TextureOperation texOp, int cbufSlot, int handleIndex, int length) + { + int binding = resourceManager.GetTextureOrImageBinding( + texOp.Inst, + texOp.Type, + texOp.Format, + texOp.Flags & ~TextureFlags.Bindless, + cbufSlot, + handleIndex, + length); + + texOp.TurnIntoArray(binding); + } + } +} |
