Avoid unnecessary buffer zero-fill in Snappy decompression#9583
Open
Dandandan wants to merge 1 commit intoapache:mainfrom
Open
Avoid unnecessary buffer zero-fill in Snappy decompression#9583Dandandan wants to merge 1 commit intoapache:mainfrom
Dandandan wants to merge 1 commit intoapache:mainfrom
Conversation
Write directly into spare capacity instead of resize+zero-fill, eliminating unnecessary memset for the decompression output buffer. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Contributor
Author
|
run benchmark arrow_reader_clickbench |
|
🤖 Arrow criterion benchmark running (GKE) | trigger |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Details
Resource Usagebase (merge-base)
branch
|
alamb
reviewed
Mar 20, 2026
| let n = self | ||
| .decoder | ||
| .decompress(input_buf, &mut spare_bytes[..len]) | ||
| .map_err(|e| -> ParquetError { e.into() })?; |
Contributor
There was a problem hiding this comment.
If this returns on error before setting len, will the buffer be left in an inconsistent state?
I think the use of the mut slice ensures that the call to decompress won't overwrite the newly allocated bytes.
However, this also basically passes in uninitialized bytes to decompress -- how do we know that the decompress doesn't read them? Maybe we should add a SAFETY warning to the decompress API that says it can't rely on initialized bytes 🤔
Contributor
Author
There was a problem hiding this comment.
Effectively we rely on this:
https://docs.rs/snap/latest/snap/raw/struct.Decoder.html#errors
- output has length less than decompress_len(input).
To not use unsafe we would need to have this feature:
BurntSushi/rust-snappy#62
Contributor
|
run benchmark arrow_reader |
|
🤖 Arrow criterion benchmark running (GKE) | trigger |
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
3 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.
Which issue does this PR close?
Closes #9579
Rationale
Currently, Snappy decompression uses
resize(len, 0)which zero-fills the buffer before writing. Since Snappy will overwrite the entire region on success, this memset is wasted work.1-2% win on snappy e2e decoding of snappy encoded parquet data
What changes are included in this PR?
Write directly into spare capacity using
reserve()+spare_capacity_mut()+set_len(), eliminating the unnecessary zero-fill.Are there any user-facing changes?
No.
🤖 Generated with Claude Code