forked from Mirror/Ryujinx
a10b2c5ff2
* Ground work for separate GPU channels * Rename TextureManager to TextureCache * Decouple texture bindings management from the texture cache * Rename BufferManager to BufferCache * Decouple buffer bindings management from the buffer cache * More comments and proper disposal * PR feedback * Force host state update on channel switch * Typo * PR feedback * Missing using
210 lines
8.3 KiB
C#
210 lines
8.3 KiB
C#
using Ryujinx.Common.Logging;
|
|
using Ryujinx.Graphics.GAL;
|
|
using Ryujinx.Graphics.Gpu.State;
|
|
using Ryujinx.Graphics.Shader;
|
|
|
|
namespace Ryujinx.Graphics.Gpu.Shader
|
|
{
|
|
/// <summary>
|
|
/// Represents a GPU state and memory accessor.
|
|
/// </summary>
|
|
class GpuAccessor : TextureDescriptorCapableGpuAccessor, IGpuAccessor
|
|
{
|
|
private readonly GpuContext _context;
|
|
private readonly GpuState _state;
|
|
private readonly int _stageIndex;
|
|
private readonly bool _compute;
|
|
private readonly int _localSizeX;
|
|
private readonly int _localSizeY;
|
|
private readonly int _localSizeZ;
|
|
private readonly int _localMemorySize;
|
|
private readonly int _sharedMemorySize;
|
|
|
|
/// <summary>
|
|
/// Creates a new instance of the GPU state accessor for graphics shader translation.
|
|
/// </summary>
|
|
/// <param name="context">GPU context</param>
|
|
/// <param name="state">Current GPU state</param>
|
|
/// <param name="stageIndex">Graphics shader stage index (0 = Vertex, 4 = Fragment)</param>
|
|
public GpuAccessor(GpuContext context, GpuState state, int stageIndex)
|
|
{
|
|
_context = context;
|
|
_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="state">Current GPU state</param>
|
|
/// <param name="localSizeX">Local group size X of the compute shader</param>
|
|
/// <param name="localSizeY">Local group size Y of the compute shader</param>
|
|
/// <param name="localSizeZ">Local group size Z of the compute shader</param>
|
|
/// <param name="localMemorySize">Local memory size of the compute shader</param>
|
|
/// <param name="sharedMemorySize">Shared memory size of the compute shader</param>
|
|
public GpuAccessor(
|
|
GpuContext context,
|
|
GpuState state,
|
|
int localSizeX,
|
|
int localSizeY,
|
|
int localSizeZ,
|
|
int localMemorySize,
|
|
int sharedMemorySize)
|
|
{
|
|
_context = context;
|
|
_state = state;
|
|
_compute = true;
|
|
_localSizeX = localSizeX;
|
|
_localSizeY = localSizeY;
|
|
_localSizeZ = localSizeZ;
|
|
_localMemorySize = localMemorySize;
|
|
_sharedMemorySize = sharedMemorySize;
|
|
}
|
|
|
|
/// <summary>
|
|
/// Prints a log message.
|
|
/// </summary>
|
|
/// <param name="message">Message to print</param>
|
|
public void Log(string message)
|
|
{
|
|
Logger.Warning?.Print(LogClass.Gpu, $"Shader translator: {message}");
|
|
}
|
|
|
|
/// <summary>
|
|
/// Reads data from GPU memory.
|
|
/// </summary>
|
|
/// <typeparam name="T">Type of the data to be read</typeparam>
|
|
/// <param name="address">GPU virtual address of the data</param>
|
|
/// <returns>Data at the memory location</returns>
|
|
public override T MemoryRead<T>(ulong address)
|
|
{
|
|
return _context.MemoryManager.Read<T>(address);
|
|
}
|
|
|
|
/// <summary>
|
|
/// Checks if a given memory address is mapped.
|
|
/// </summary>
|
|
/// <param name="address">GPU virtual address to be checked</param>
|
|
/// <returns>True if the address is mapped, false otherwise</returns>
|
|
public bool MemoryMapped(ulong address)
|
|
{
|
|
return _context.MemoryManager.IsMapped(address);
|
|
}
|
|
|
|
/// <summary>
|
|
/// Queries Local Size X for compute shaders.
|
|
/// </summary>
|
|
/// <returns>Local Size X</returns>
|
|
public int QueryComputeLocalSizeX() => _localSizeX;
|
|
|
|
/// <summary>
|
|
/// Queries Local Size Y for compute shaders.
|
|
/// </summary>
|
|
/// <returns>Local Size Y</returns>
|
|
public int QueryComputeLocalSizeY() => _localSizeY;
|
|
|
|
/// <summary>
|
|
/// Queries Local Size Z for compute shaders.
|
|
/// </summary>
|
|
/// <returns>Local Size Z</returns>
|
|
public int QueryComputeLocalSizeZ() => _localSizeZ;
|
|
|
|
/// <summary>
|
|
/// Queries Local Memory size in bytes for compute shaders.
|
|
/// </summary>
|
|
/// <returns>Local Memory size in bytes</returns>
|
|
public int QueryComputeLocalMemorySize() => _localMemorySize;
|
|
|
|
/// <summary>
|
|
/// Queries Shared Memory size in bytes for compute shaders.
|
|
/// </summary>
|
|
/// <returns>Shared Memory size in bytes</returns>
|
|
public int QueryComputeSharedMemorySize() => _sharedMemorySize;
|
|
|
|
/// <summary>
|
|
/// Queries Constant Buffer usage information.
|
|
/// </summary>
|
|
/// <returns>A mask where each bit set indicates a bound constant buffer</returns>
|
|
public uint QueryConstantBufferUse()
|
|
{
|
|
return _compute
|
|
? _state.Channel.BufferManager.GetComputeUniformBufferUseMask()
|
|
: _state.Channel.BufferManager.GetGraphicsUniformBufferUseMask(_stageIndex);
|
|
}
|
|
|
|
/// <summary>
|
|
/// Queries current primitive topology for geometry shaders.
|
|
/// </summary>
|
|
/// <returns>Current primitive topology</returns>
|
|
public InputTopology QueryPrimitiveTopology()
|
|
{
|
|
switch (_context.Methods.Topology)
|
|
{
|
|
case PrimitiveTopology.Points:
|
|
return InputTopology.Points;
|
|
case PrimitiveTopology.Lines:
|
|
case PrimitiveTopology.LineLoop:
|
|
case PrimitiveTopology.LineStrip:
|
|
return InputTopology.Lines;
|
|
case PrimitiveTopology.LinesAdjacency:
|
|
case PrimitiveTopology.LineStripAdjacency:
|
|
return InputTopology.LinesAdjacency;
|
|
case PrimitiveTopology.Triangles:
|
|
case PrimitiveTopology.TriangleStrip:
|
|
case PrimitiveTopology.TriangleFan:
|
|
return InputTopology.Triangles;
|
|
case PrimitiveTopology.TrianglesAdjacency:
|
|
case PrimitiveTopology.TriangleStripAdjacency:
|
|
return InputTopology.TrianglesAdjacency;
|
|
}
|
|
|
|
return InputTopology.Points;
|
|
}
|
|
|
|
/// <summary>
|
|
/// Queries host storage buffer alignment required.
|
|
/// </summary>
|
|
/// <returns>Host storage buffer alignment in bytes</returns>
|
|
public int QueryStorageBufferOffsetAlignment() => _context.Capabilities.StorageBufferOffsetAlignment;
|
|
|
|
/// <summary>
|
|
/// Queries host support for readable images without a explicit format declaration on the shader.
|
|
/// </summary>
|
|
/// <returns>True if formatted image load is supported, false otherwise</returns>
|
|
public bool QuerySupportsImageLoadFormatted() => _context.Capabilities.SupportsImageLoadFormatted;
|
|
|
|
/// <summary>
|
|
/// Queries host GPU non-constant texture offset support.
|
|
/// </summary>
|
|
/// <returns>True if the GPU and driver supports non-constant texture offsets, false otherwise</returns>
|
|
public bool QuerySupportsNonConstantTextureOffset() => _context.Capabilities.SupportsNonConstantTextureOffset;
|
|
|
|
/// <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>
|
|
public override Image.ITextureDescriptor GetTextureDescriptor(int handle, int cbufSlot)
|
|
{
|
|
if (_compute)
|
|
{
|
|
return _state.Channel.TextureManager.GetComputeTextureDescriptor(_state, handle, cbufSlot);
|
|
}
|
|
else
|
|
{
|
|
return _state.Channel.TextureManager.GetGraphicsTextureDescriptor(_state, _stageIndex, handle, cbufSlot);
|
|
}
|
|
}
|
|
|
|
/// <summary>
|
|
/// Queries if host state forces early depth testing.
|
|
/// </summary>
|
|
/// <returns>True if early depth testing is forced</returns>
|
|
public bool QueryEarlyZForce()
|
|
{
|
|
return _state.Get<bool>(MethodOffset.EarlyZForce);
|
|
}
|
|
}
|
|
}
|