Repository navigation
Integrate remaining JPEG and native CPU codec optimizations - #118
Merged
Merged
Conversation
…version - Decode MCU rows against local copies of the bit reader and DC predictors so the accumulator stays in registers; cold refill paths take and return the reader by value. - Refill the bit accumulator eight bytes at a time when the window has no 0xFF, falling back to the byte-wise path for stuffing, markers, and the input tail. - Replace the per-coefficient touched-index list and activity state machine with one AC flag and an OR of natural indices. - Run NEON YCbCr->RGB in exact 16-bit lanes (constants split as 65536*k + remainder), sixteen pixels per iteration, and test it against the scalar converter for every chroma pair.
…u-codec-optimizations
jcwal1516
changed the base branch from
fix/subsampled-component-origin
to
main
September 27, 2026 01:05
An error occurred while trying to automatically change base from
fix/subsampled-component-origin
to
main
September 27, 2026 01:05
Merged
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Integrate the remaining local CPU codec changes: retained JPEG entropy state and NEON conversion, extended/lossless/progressive JPEG decoding, and native bitplane scanning and subsampled placement. Preserve the nonzero-origin fuzz regression fix from #117 when combining the optimized placement path.
Stacked on #117. The resulting source tree retains the previously uncommitted work; differences from the preserved local snapshot are the decoder regression and the repaired trusted-workflow pins.
Validation: 1,631 tests passed across j2k-jpeg and j2k-native, with 4 existing ignored tests. Formatting and both libraries' all-feature Clippy checks passed. No new speedup claim is made.