mirror of
https://github.com/Ryujinx/Ryujinx.git
synced 2024-08-20 15:48:43 +02:00
* New shader cache implementation * Remove some debug code * Take transform feedback varying count into account * Create shader cache directory if it does not exist + fragment output map related fixes * Remove debug code * Only check texture descriptors if the constant buffer is bound * Also check CPU VA on GetSpanMapped * Remove more unused code and move cache related code * XML docs + remove more unused methods * Better codegen for TransformFeedbackDescriptor.AsSpan * Support migration from old cache format, remove more unused code Shader cache rebuild now also rewrites the shared toc and data files * Fix migration error with BRX shaders * Add a limit to the async translation queue Avoid async translation threads not being able to keep up and the queue growing very large * Re-create specialization state on recompile This might be required if a new version of the shader translator requires more or less state, or if there is a bug related to the GPU state access * Make shader cache more error resilient * Add some missing XML docs and move GpuAccessor docs to the interface/use inheritdoc * Address early PR feedback * Fix rebase * Remove IRenderer.CompileShader and IShader interface, replace with new ShaderSource struct passed to CreateProgram directly * Handle some missing exceptions * Make shader cache purge delete both old and new shader caches * Register textures on new specialization state * Translate and compile shaders in forward order (eliminates diffs due to different binding numbers) * Limit in-flight shader compilation to the maximum number of compilation threads * Replace ParallelDiskCacheLoader state changed event with a callback function * Better handling for invalid constant buffer 1 data length * Do not create the old cache directory structure if the old cache does not exist * Constant buffer use should be per-stage. This change will invalidate existing new caches (file format version was incremented) * Replace rectangle texture with just coordinate normalization * Skip incompatible shaders that are missing texture information, instead of crashing This is required if we, for example, support new texture instruction to the shader translator, and then they allow access to textures that were not accessed before. In this scenario, the old cache entry is no longer usable * Fix coordinates normalization on cubemap textures * Check if title ID is null before combining shader cache path * More robust constant buffer address validation on spec state * More robust constant buffer address validation on spec state (2) * Regenerate shader cache with one stream, rather than one per shader. * Only create shader cache directory during initialization * Logging improvements * Proper shader program disposal * PR feedback, and add a comment on serialized structs * XML docs for RegisterTexture Co-authored-by: riperiperi <rhy3756547@hotmail.com>
227 lines
8 KiB
C#
227 lines
8 KiB
C#
using Ryujinx.Common.Logging;
|
|
using Ryujinx.Graphics.Gpu.Image;
|
|
using Ryujinx.Graphics.Shader;
|
|
using System;
|
|
using System.Runtime.InteropServices;
|
|
|
|
namespace Ryujinx.Graphics.Gpu.Shader
|
|
{
|
|
/// <summary>
|
|
/// Represents a GPU state and memory accessor.
|
|
/// </summary>
|
|
class GpuAccessor : GpuAccessorBase, IGpuAccessor
|
|
{
|
|
private readonly GpuChannel _channel;
|
|
private readonly GpuAccessorState _state;
|
|
private readonly int _stageIndex;
|
|
private readonly bool _compute;
|
|
|
|
/// <summary>
|
|
/// Creates a new instance of the GPU state accessor for graphics shader translation.
|
|
/// </summary>
|
|
/// <param name="context">GPU context</param>
|
|
/// <param name="channel">GPU channel</param>
|
|
/// <param name="state">Current GPU state</param>
|
|
/// <param name="stageIndex">Graphics shader stage index (0 = Vertex, 4 = Fragment)</param>
|
|
public GpuAccessor(GpuContext context, GpuChannel channel, GpuAccessorState state, int stageIndex) : base(context)
|
|
{
|
|
_channel = channel;
|
|
_state = state;
|
|
_stageIndex = stageIndex;
|
|
}
|
|
|
|
/// <summary>
|
|
/// Creates a new instance of the GPU state accessor for compute shader translation.
|
|
/// </summary>
|
|
/// <param name="context">GPU context</param>
|
|
/// <param name="channel">GPU channel</param>
|
|
/// <param name="state">Current GPU state</param>
|
|
public GpuAccessor(GpuContext context, GpuChannel channel, GpuAccessorState state) : base(context)
|
|
{
|
|
_channel = channel;
|
|
_state = state;
|
|
_compute = true;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public uint ConstantBuffer1Read(int offset)
|
|
{
|
|
ulong baseAddress = _compute
|
|
? _channel.BufferManager.GetComputeUniformBufferAddress(1)
|
|
: _channel.BufferManager.GetGraphicsUniformBufferAddress(_stageIndex, 1);
|
|
|
|
return _channel.MemoryManager.Physical.Read<uint>(baseAddress + (ulong)offset);
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public void Log(string message)
|
|
{
|
|
Logger.Warning?.Print(LogClass.Gpu, $"Shader translator: {message}");
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public ReadOnlySpan<ulong> GetCode(ulong address, int minimumSize)
|
|
{
|
|
int size = Math.Max(minimumSize, 0x1000 - (int)(address & 0xfff));
|
|
return MemoryMarshal.Cast<byte, ulong>(_channel.MemoryManager.GetSpan(address, size));
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryBindingConstantBuffer(int index)
|
|
{
|
|
return _state.ResourceCounts.UniformBuffersCount++;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryBindingStorageBuffer(int index)
|
|
{
|
|
return _state.ResourceCounts.StorageBuffersCount++;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryBindingTexture(int index)
|
|
{
|
|
return _state.ResourceCounts.TexturesCount++;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryBindingImage(int index)
|
|
{
|
|
return _state.ResourceCounts.ImagesCount++;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryComputeLocalSizeX() => _state.ComputeState.LocalSizeX;
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryComputeLocalSizeY() => _state.ComputeState.LocalSizeY;
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryComputeLocalSizeZ() => _state.ComputeState.LocalSizeZ;
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryComputeLocalMemorySize() => _state.ComputeState.LocalMemorySize;
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryComputeSharedMemorySize() => _state.ComputeState.SharedMemorySize;
|
|
|
|
/// <inheritdoc/>
|
|
public uint QueryConstantBufferUse()
|
|
{
|
|
uint useMask = _compute
|
|
? _channel.BufferManager.GetComputeUniformBufferUseMask()
|
|
: _channel.BufferManager.GetGraphicsUniformBufferUseMask(_stageIndex);
|
|
|
|
_state.SpecializationState?.RecordConstantBufferUse(_stageIndex, useMask);
|
|
return useMask;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public InputTopology QueryPrimitiveTopology()
|
|
{
|
|
_state.SpecializationState?.RecordPrimitiveTopology();
|
|
return ConvertToInputTopology(_state.GraphicsState.Topology, _state.GraphicsState.TessellationMode);
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public bool QueryTessCw()
|
|
{
|
|
return _state.GraphicsState.TessellationMode.UnpackCw();
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public TessPatchType QueryTessPatchType()
|
|
{
|
|
return _state.GraphicsState.TessellationMode.UnpackPatchType();
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public TessSpacing QueryTessSpacing()
|
|
{
|
|
return _state.GraphicsState.TessellationMode.UnpackSpacing();
|
|
}
|
|
|
|
//// <inheritdoc/>
|
|
public TextureFormat QueryTextureFormat(int handle, int cbufSlot)
|
|
{
|
|
_state.SpecializationState?.RecordTextureFormat(_stageIndex, handle, cbufSlot);
|
|
var descriptor = GetTextureDescriptor(handle, cbufSlot);
|
|
return ConvertToTextureFormat(descriptor.UnpackFormat(), descriptor.UnpackSrgb());
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public SamplerType QuerySamplerType(int handle, int cbufSlot)
|
|
{
|
|
_state.SpecializationState?.RecordTextureSamplerType(_stageIndex, handle, cbufSlot);
|
|
return GetTextureDescriptor(handle, cbufSlot).UnpackTextureTarget().ConvertSamplerType();
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public bool QueryTextureCoordNormalized(int handle, int cbufSlot)
|
|
{
|
|
_state.SpecializationState?.RecordTextureCoordNormalized(_stageIndex, handle, cbufSlot);
|
|
return GetTextureDescriptor(handle, cbufSlot).UnpackTextureCoordNormalized();
|
|
}
|
|
|
|
/// <summary>
|
|
/// Gets the texture descriptor for a given texture on the pool.
|
|
/// </summary>
|
|
/// <param name="handle">Index of the texture (this is the word offset of the handle in the constant buffer)</param>
|
|
/// <param name="cbufSlot">Constant buffer slot for the texture handle</param>
|
|
/// <returns>Texture descriptor</returns>
|
|
private Image.TextureDescriptor GetTextureDescriptor(int handle, int cbufSlot)
|
|
{
|
|
if (_compute)
|
|
{
|
|
return _channel.TextureManager.GetComputeTextureDescriptor(
|
|
_state.PoolState.TexturePoolGpuVa,
|
|
_state.PoolState.TextureBufferIndex,
|
|
_state.PoolState.TexturePoolMaximumId,
|
|
handle,
|
|
cbufSlot);
|
|
}
|
|
else
|
|
{
|
|
return _channel.TextureManager.GetGraphicsTextureDescriptor(
|
|
_state.PoolState.TexturePoolGpuVa,
|
|
_state.PoolState.TextureBufferIndex,
|
|
_state.PoolState.TexturePoolMaximumId,
|
|
_stageIndex,
|
|
handle,
|
|
cbufSlot);
|
|
}
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public bool QueryTransformFeedbackEnabled()
|
|
{
|
|
return _state.TransformFeedbackDescriptors != null;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public ReadOnlySpan<byte> QueryTransformFeedbackVaryingLocations(int bufferIndex)
|
|
{
|
|
return _state.TransformFeedbackDescriptors[bufferIndex].AsSpan();
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public int QueryTransformFeedbackStride(int bufferIndex)
|
|
{
|
|
return _state.TransformFeedbackDescriptors[bufferIndex].Stride;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public bool QueryEarlyZForce()
|
|
{
|
|
_state.SpecializationState?.RecordEarlyZForce();
|
|
return _state.GraphicsState.EarlyZForce;
|
|
}
|
|
|
|
/// <inheritdoc/>
|
|
public void RegisterTexture(int handle, int cbufSlot)
|
|
{
|
|
_state.SpecializationState?.RegisterTexture(_stageIndex, handle, cbufSlot, GetTextureDescriptor(handle, cbufSlot));
|
|
}
|
|
}
|
|
}
|