From bf77d1cab93467676156ebfbd5cf0ae057266e6f Mon Sep 17 00:00:00 2001 From: riperiperi Date: Sat, 8 Oct 2022 16:04:47 +0100 Subject: GPU: Pass SpanOrArray for Texture SetData to avoid copy (#3745) * GPU: Pass SpanOrArray for Texture SetData to avoid copy Texture data is often converted before upload, meaning that an array was allocated to perform the conversion into. However, the backend SetData methods were being passed a Span of that data, and the Multithreaded layer does `ToArray()` on it so that it can be stored for later! This method can't extract the original array, so it creates a copy. This PR changes the type passed for textures to a new ref struct called SpanOrArray, which is backed by either a ReadOnlySpan or an array. The benefit here is that we can have a ToArray method that doesn't copy if it is originally backed by an array. This will also avoid a copy when running the ASTC decoder. On NieR this was taking 38% of texture upload time, which it does a _lot_ of when you move between areas, so there should be a 1.6x performance boost when strictly uploading textures. No doubt this will also improve texture streaming performance in UE4 games, and maybe a small reduction with video playback. From the numbers, it's probably possible to improve the upload rate by a further 1.6x by performing layout conversion on GPU. I'm not sure if we could improve it further than that - multithreading conversion on CPU would probably result in memory bottleneck. This doesn't extend to buffers, since we don't convert their data on the GPU emulator side. * Remove implicit cast to array. --- Ryujinx.Graphics.Texture/LayoutConverter.cs | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'Ryujinx.Graphics.Texture') diff --git a/Ryujinx.Graphics.Texture/LayoutConverter.cs b/Ryujinx.Graphics.Texture/LayoutConverter.cs index 2b327375..188ae0c1 100644 --- a/Ryujinx.Graphics.Texture/LayoutConverter.cs +++ b/Ryujinx.Graphics.Texture/LayoutConverter.cs @@ -93,7 +93,7 @@ namespace Ryujinx.Graphics.Texture }; } - public static Span ConvertBlockLinearToLinear( + public static byte[] ConvertBlockLinearToLinear( int width, int height, int depth, @@ -119,7 +119,7 @@ namespace Ryujinx.Graphics.Texture blockHeight, bytesPerPixel); - Span output = new byte[outSize]; + byte[] output = new byte[outSize]; int outOffs = 0; @@ -246,7 +246,7 @@ namespace Ryujinx.Graphics.Texture return output; } - public static Span ConvertLinearStridedToLinear( + public static byte[] ConvertLinearStridedToLinear( int width, int height, int blockWidth, @@ -262,14 +262,15 @@ namespace Ryujinx.Graphics.Texture int outStride = BitUtils.AlignUp(w * bytesPerPixel, HostStrideAlignment); lineSize = Math.Min(lineSize, outStride); - Span output = new byte[h * outStride]; + byte[] output = new byte[h * outStride]; + Span outSpan = output; int outOffs = 0; int inOffs = 0; for (int y = 0; y < h; y++) { - data.Slice(inOffs, lineSize).CopyTo(output.Slice(outOffs, lineSize)); + data.Slice(inOffs, lineSize).CopyTo(outSpan.Slice(outOffs, lineSize)); inOffs += stride; outOffs += outStride; -- cgit v1.2.3