Return mapped buffer pointer directly for flush, WriteableRegion for textures (#2494)

* Return mapped buffer pointer directly for flush, WriteableRegion for textures

A few changes here to generally improve performance, even for platforms not using the persistent buffer flush.

- Texture and buffer flush now return a ReadOnlySpan<byte>. It's guaranteed that this span is pinned in memory, but it will be overwritten on the next flush from that thread, so it is expected that the data is used before calling again.
- As a result, persistent mappings no longer copy to a new array - rather the persistent map is returned directly as a Span<>. A similar host array is used for the glGet flushes instead of allocating new arrays each time.
- Texture flushes now do their layout conversion into a WriteableRegion when the texture is not MultiRange, which allows the flush to happen directly into guest memory rather than into a temporary span, then copied over. This avoids another copy when doing layout conversion.

Overall, this saves 1 data copy for buffer flush, 1 copy for linear textures with matching source/target stride, and 2 copies for block textures or linear textures with mismatching strides.

* Fix tests

* Fix array pointer for Mesa/Intel path

* Address some feedback

* Update method for getting array pointer.

Ryujinx.Graphics.OpenGL/PersistentBuffers.cs

@@ -1,4 +1,5 @@
 using System;
+using System.Runtime.CompilerServices;
 using System.Runtime.InteropServices;
 using OpenTK.Graphics.OpenGL;
 using Ryujinx.Common.Logging;
@@ -27,6 +28,9 @@ namespace Ryujinx.Graphics.OpenGL
         private int _copyBufferHandle;
         private int _copyBufferSize;
 
+        private byte[] _data;
+        private IntPtr _dataMap;
+
         private void EnsureBuffer(int requiredSize)
         {
             if (_copyBufferSize < requiredSize && _copyBufferHandle != 0)
@@ -48,6 +52,18 @@ namespace Ryujinx.Graphics.OpenGL
             }
         }
 
+        public unsafe IntPtr GetHostArray(int requiredSize)
+        {
+            if (_data == null || _data.Length < requiredSize)
+            {
+                _data = GC.AllocateUninitializedArray<byte>(requiredSize, true);
+
+                _dataMap = (IntPtr)Unsafe.AsPointer(ref MemoryMarshal.GetArrayDataReference(_data));
+            }
+
+            return _dataMap;
+        }
+
         private void Sync()
         {
             GL.MemoryBarrier(MemoryBarrierFlags.ClientMappedBufferBarrierBit);
@@ -63,7 +79,7 @@ namespace Ryujinx.Graphics.OpenGL
             GL.DeleteSync(sync);
         }
 
-        public byte[] GetTextureData(TextureView view, int size)
+        public unsafe ReadOnlySpan<byte> GetTextureData(TextureView view, int size)
         {
             EnsureBuffer(size);
 
@@ -73,16 +89,12 @@ namespace Ryujinx.Graphics.OpenGL
 
             GL.BindBuffer(BufferTarget.PixelPackBuffer, 0);
 
-            byte[] data = new byte[size];
-
             Sync();
 
-            Marshal.Copy(_bufferMap, data, 0, size);
-
-            return data;
+            return new ReadOnlySpan<byte>(_bufferMap.ToPointer(), size);
         }
 
-        public byte[] GetBufferData(BufferHandle buffer, int offset, int size)
+        public unsafe ReadOnlySpan<byte> GetBufferData(BufferHandle buffer, int offset, int size)
         {
             EnsureBuffer(size);
 
@@ -93,13 +105,9 @@ namespace Ryujinx.Graphics.OpenGL
 
             GL.BindBuffer(BufferTarget.CopyWriteBuffer, 0);
 
-            byte[] data = new byte[size];
-
             Sync();
 
-            Marshal.Copy(_bufferMap, data, 0, size);
-
-            return data;
+            return new ReadOnlySpan<byte>(_bufferMap.ToPointer(), size);
         }
 
         public void Dispose()