inflate: skip the per-match bounds check in the fast loop#540
Open
jlucaso1 wants to merge 1 commit into
Open
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The inflate fast loop already guarantees
INFLATE_FAST_MIN_LEFTbytes of output headroom before it runs, yetcopy_matchstill recomputes a capacity clamp and does a bounds check on every single match. That work is redundant inside the fast loop.This adds a
copy_match_uncheckedvariant used only there, and bumps the margin to also cover the chunked copy's overshoot (260 + 32), so the per-match check can be safely elided. zlib-ng/libdeflate do the same thing.Measured ~8% fewer instructions on inflate (
perf stat, on a mix of protobuf-like data and the silesia corpus). The window-copy path is untouched.Correctness: the full
test-libz-rs-sysC-compat suite passes, plus a fuzz roundtrip (200k random inputs across sizes/levels) with a canary past the output buffer to catch any overshoot.