Improve SSE4.1 ImagePngLoader decode-line handlers - #949
Merged
Conversation
Process whole SSE vectors in Avg/Paeth/AvgFirst unfilters instead of per-pixel LoadN/StoreN, enable SIMD decode handlers only when width >= A, and add n=1 paths for those filters. Co-authored-by: igor.ermolaev <igor.ermolaev@altezza.org>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Process whole SSE vectors in PNG unfilters instead of per-pixel
LoadN/StoreN.The previous SSE4.1
_decodeLineAvg/Paeth/AvgFirst paths decodedsrcNbytes at a time. That is slow for RGB (srcN=3) because each pixel does scalar-style 3-byte loads/stores around SIMD arithmetic.Changes (
SimdSse41ImageLoadPng.cpp)KeepFirst+Average), thenStoreStepof 8/12/15/16 bytes, matching the existing Sub-filter approach.PaethPack, assemble and store the vector._width >= A; narrower images keep the Base scalar handlers.srcN == 1(Gray8). A 16-step serial chain in SIMD was slower than Base.Tests
Test::ImageLoadFromMemoryAutoTestandImageLoadFromMemorySpecialTestboth pass.AutoTest (Release,
-ts=1, 1920x1080 PNG), Sse41 vs Base speedup (Bs/S4):SpecialTest: PNG suite under
data/image/png/for Gray8, Bgr24, Bgra32, Rgb24, Rgba32. Base vsSimdImageLoadFromMemory(AVX-512BW dispatches toSse41::ImagePngLoader).