Ryujinx/Ryujinx.Graphics.OpenGL/Renderer.cs
gdkchan 43ebd7a9bb
New shader cache implementation (#3194)
* 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>
2022-04-10 10:49:44 -03:00

233 lines
7.9 KiB
C#

using OpenTK.Graphics.OpenGL;
using Ryujinx.Common.Configuration;
using Ryujinx.Common.Logging;
using Ryujinx.Graphics.GAL;
using Ryujinx.Graphics.OpenGL.Image;
using Ryujinx.Graphics.OpenGL.Queries;
using Ryujinx.Graphics.Shader.Translation;
using System;
namespace Ryujinx.Graphics.OpenGL
{
public sealed class Renderer : IRenderer
{
private readonly Pipeline _pipeline;
public IPipeline Pipeline => _pipeline;
private readonly Counters _counters;
private readonly Window _window;
public IWindow Window => _window;
private TextureCopy _textureCopy;
private TextureCopy _backgroundTextureCopy;
internal TextureCopy TextureCopy => BackgroundContextWorker.InBackground ? _backgroundTextureCopy : _textureCopy;
private Sync _sync;
public event EventHandler<ScreenCaptureImageInfo> ScreenCaptured;
internal PersistentBuffers PersistentBuffers { get; }
internal ResourcePool ResourcePool { get; }
internal int BufferCount { get; private set; }
public string GpuVendor { get; private set; }
public string GpuRenderer { get; private set; }
public string GpuVersion { get; private set; }
public bool PreferThreading => true;
public Renderer()
{
_pipeline = new Pipeline();
_counters = new Counters();
_window = new Window(this);
_textureCopy = new TextureCopy(this);
_backgroundTextureCopy = new TextureCopy(this);
_sync = new Sync();
PersistentBuffers = new PersistentBuffers();
ResourcePool = new ResourcePool();
}
public BufferHandle CreateBuffer(int size)
{
BufferCount++;
return Buffer.Create(size);
}
public IProgram CreateProgram(ShaderSource[] shaders, ShaderInfo info)
{
return new Program(shaders, info.FragmentOutputMap);
}
public ISampler CreateSampler(SamplerCreateInfo info)
{
return new Sampler(info);
}
public ITexture CreateTexture(TextureCreateInfo info, float scaleFactor)
{
if (info.Target == Target.TextureBuffer)
{
return new TextureBuffer(this, info);
}
else
{
return ResourcePool.GetTextureOrNull(info, scaleFactor) ?? new TextureStorage(this, info, scaleFactor).CreateDefaultView();
}
}
public void DeleteBuffer(BufferHandle buffer)
{
Buffer.Delete(buffer);
}
public ReadOnlySpan<byte> GetBufferData(BufferHandle buffer, int offset, int size)
{
return Buffer.GetData(this, buffer, offset, size);
}
public Capabilities GetCapabilities()
{
return new Capabilities(
api: TargetApi.OpenGL,
vendorName: GpuVendor,
hasFrontFacingBug: HwCapabilities.Vendor == HwCapabilities.GpuVendor.IntelWindows,
hasVectorIndexingBug: HwCapabilities.Vendor == HwCapabilities.GpuVendor.AmdWindows,
supportsAstcCompression: HwCapabilities.SupportsAstcCompression,
supports3DTextureCompression: false,
supportsBgraFormat: false,
supportsR4G4Format: false,
supportsFragmentShaderInterlock: HwCapabilities.SupportsFragmentShaderInterlock,
supportsFragmentShaderOrderingIntel: HwCapabilities.SupportsFragmentShaderOrdering,
supportsImageLoadFormatted: HwCapabilities.SupportsImageLoadFormatted,
supportsMismatchingViewFormat: HwCapabilities.SupportsMismatchingViewFormat,
supportsNonConstantTextureOffset: HwCapabilities.SupportsNonConstantTextureOffset,
supportsShaderBallot: HwCapabilities.SupportsShaderBallot,
supportsTextureShadowLod: HwCapabilities.SupportsTextureShadowLod,
supportsViewportSwizzle: HwCapabilities.SupportsViewportSwizzle,
supportsIndirectParameters: HwCapabilities.SupportsIndirectParameters,
maximumComputeSharedMemorySize: HwCapabilities.MaximumComputeSharedMemorySize,
maximumSupportedAnisotropy: HwCapabilities.MaximumSupportedAnisotropy,
storageBufferOffsetAlignment: HwCapabilities.StorageBufferOffsetAlignment);
}
public void SetBufferData(BufferHandle buffer, int offset, ReadOnlySpan<byte> data)
{
Buffer.SetData(buffer, offset, data);
}
public void UpdateCounters()
{
_counters.Update();
}
public void PreFrame()
{
_sync.Cleanup();
ResourcePool.Tick();
}
public ICounterEvent ReportCounter(CounterType type, EventHandler<ulong> resultHandler, bool hostReserved)
{
return _counters.QueueReport(type, resultHandler, _pipeline.DrawCount, hostReserved);
}
public void Initialize(GraphicsDebugLevel glLogLevel)
{
Debugger.Initialize(glLogLevel);
PrintGpuInformation();
if (HwCapabilities.SupportsParallelShaderCompile)
{
GL.Arb.MaxShaderCompilerThreads(Math.Min(Environment.ProcessorCount, 8));
}
_pipeline.Initialize(this);
_counters.Initialize();
// This is required to disable [0, 1] clamping for SNorm outputs on compatibility profiles.
// This call is expected to fail if we're running with a core profile,
// as this clamp target was deprecated, but that's fine as a core profile
// should already have the desired behaviour were outputs are not clamped.
GL.ClampColor(ClampColorTarget.ClampFragmentColor, ClampColorMode.False);
}
private void PrintGpuInformation()
{
GpuVendor = GL.GetString(StringName.Vendor);
GpuRenderer = GL.GetString(StringName.Renderer);
GpuVersion = GL.GetString(StringName.Version);
Logger.Notice.Print(LogClass.Gpu, $"{GpuVendor} {GpuRenderer} ({GpuVersion})");
}
public void ResetCounter(CounterType type)
{
_counters.QueueReset(type);
}
public void BackgroundContextAction(Action action, bool alwaysBackground = false)
{
// alwaysBackground is ignored, since we cannot switch from the current context.
if (IOpenGLContext.HasContext())
{
action(); // We have a context already - use that (assuming it is the main one).
}
else
{
_window.BackgroundContext.Invoke(action);
}
}
public void InitializeBackgroundContext(IOpenGLContext baseContext)
{
_window.InitializeBackgroundContext(baseContext);
}
public void Dispose()
{
_textureCopy.Dispose();
_backgroundTextureCopy.Dispose();
PersistentBuffers.Dispose();
ResourcePool.Dispose();
_pipeline.Dispose();
_window.Dispose();
_counters.Dispose();
_sync.Dispose();
}
public IProgram LoadProgramBinary(byte[] programBinary, bool hasFragmentShader, ShaderInfo info)
{
return new Program(programBinary, hasFragmentShader, info.FragmentOutputMap);
}
public void CreateSync(ulong id)
{
_sync.Create(id);
}
public void WaitSync(ulong id)
{
_sync.Wait(id);
}
public void Screenshot()
{
_window.ScreenCaptureRequested = true;
}
public void OnScreenCaptured(ScreenCaptureImageInfo bitmap)
{
ScreenCaptured?.Invoke(this, bitmap);
}
}
}