fix long offset resolution #3460
Merged
Conversation
This file contains 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
terrelln
requested changes
Jan 27, 2023
daniellerozenblit
force-pushed
the
fix-long-offsets-resolution-pointer
branch
from
fb9e376 to
e10676a
Compare
daniellerozenblit
force-pushed
the
fix-long-offsets-resolution-pointer
branch
from
e10676a to
9e4c66b
Compare
terrelln
reviewed
Jan 27, 2023
There was a problem hiding this comment.
Looks good to me! I will also benchmark this PR to make sure I also don't see regressions.
Thanks for adding the test! We might need to try to reduce the memory usage of the test a little bit, otherwise it could be flaky in 32-bit builds, because we could run out of address space.
terrelln
reviewed
Jan 27, 2023
Co-authored-by: Nick Terrell <[email protected]>
…ellerozenblit/zstd into fix-long-offsets-resolution-pointer
terrelln
reviewed
Jan 27, 2023
terrelln
reviewed
Jan 27, 2023
| size_t const kNbSequences = 4; | ||
| ZSTD_Sequence* sequences = malloc(sizeof(ZSTD_Sequence) * kNbSequences); | ||
| void* const checkBuf = malloc(srcSize); | ||
| const size_t largeDictSize = 1 << 25; |
There was a problem hiding this comment.
Oh nice, you got it to reproduce with 1 << 25!
In that case we don't need to hide it behind bigTests. But I still like having one 32-bit test in CI that does run with --big-tests. That should help improve our 32-bit coverage a bit, without making our users tests flaky.
Co-authored-by: Nick Terrell <[email protected]>
…ellerozenblit/zstd into fix-long-offsets-resolution-pointer
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Outdated suggestions cannot be applied.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.
We currently determine whether or not
longOffsetsare present in a set of input sequences by checking if thewindowLogis larger thanSTREAM_ACCUMULATOR_MIN.This does not account for inputs with a dictionary, where it is legal for offsets to be larger than the window size. In these cases, we don't properly account for large offsets when flushing our bit accumulator in
ZSTD_encodeSequences_body().This PR introduces a
largeOffsetsfield in theZSTD_symbolEncodingTypeStats_twhich records whether or not large offsets are present in the sequences. This value of this field is finalized inZSTD_seqToCodes().I ran some benchmarks for zstd level 1 & 3 and there does not appear to be any measurable change in compression speed.