allocation optimization for lz4frame compression #1158
Merged
+37
−8
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
as noted by @yixiutt, the temporary buffer inside lz4frame compression is being invalidated at end of compression, forcing it to be re-allocated at next compression job. This shouldn't be necessary. This change was introduced in #236, as a way to fix #232, but neither the issue is explained, nor why the patch fixes it. This patch revert to previous behavior, where temporary buffer is kept between compression calls. This results in a net reduction of allocation workload. Additionally, the temporary buffer should only need malloc(), not calloc(), thus saving initialization. This diff implements both changes. Long fuzzer tests will be run to ensure that no sanitizer warning get triggered. Additionally : fixed a minor ubsan warning in LZ4F_decompress().
|
LGTM |
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Outdated suggestions cannot be applied.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.
as noted by @yixiutt in #1157, the temporary buffer managed by lz4frame compression context is being invalidated at end of compression, forcing it to be re-allocated at next compression job.
This shouldn't be necessary.
This behavior was introduced in #236, as a way to fix #232, but neither the issue is explained, nor why the patch fixes it.
This patch reverts to previous behavior,
where temporary buffer is reused between compression calls.
This results in a net reduction of allocation workload.
Additionally, the temporary buffer should only need
malloc(), notcalloc(), thus saving some potential 0-initialization cost.This diff implements both changes. It's expected to improve compression speed when repetitively compressing small data.
Performance impact on M1 laptop :
As expected, performance difference is only perceptible for small data, but it can matter a lot in this case.
Once this diff is merged, long fuzzer tests will be run to ensure that no sanitizer warning gets triggered.
Additionally :
LZ4F_decompress()LZ4F_compressUpdate()test tofullbench