- Notifications
You must be signed in to change notification settings - Fork868
perf: optimize ObjectID.ReadFrom with direct io.ReadFull#1738
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to ourterms of service andprivacy statement. We’ll occasionally send you account related emails.
Already on GitHub?Sign in to your account
Open
Bluebugs wants to merge2 commits intogo-git:mainChoose a base branch
base:main
Could not load branches
Branch not found:{{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline, and old review comments may become outdated.
Uh oh!
There was an error while loading.Please reload this page.
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
Replace encoding/binary.Read with io.ReadFull in ObjectID.ReadFrom forbetter performance, eliminating reflection overhead.Background:-----------The ObjectID.ReadFrom method used encoding/binary.Read, which employsreflection to copy bytes into the hash array. This adds unnecessary overheadin hot paths, particularly during Tree.Decode operations that are calledthousands of times during clone operations.Changes:--------1. Updated ObjectID.ReadFrom implementation to use io.ReadFull directly instead of encoding/binary.Read2. Added error handling to zero out the hash on read errors for cleaner state3. Updated test coverage to verify the optimized implementation: - TestReadFrom: validates existing functionality still works - TestReadFromOptimization: validates direct io.ReadFull behaviorTechnical Notes:---------------This is a backward-compatible change. The method signature remains identicaland it still implements io.ReaderFrom. For byte arrays (like hashes),binary.Read's BigEndian parameter is a no-op since no byte-order conversionoccurs. io.ReadFull produces identical results while being significantlyfaster by avoiding reflection.Performance Impact:------------------Benchmark results from 10 iterations (AMD Ryzen 7 6800U):Speed improvements (statistically significant, p<0.05): TreeDecode: -2.91% (1429ns → 1387ns) TreeDecodeSmall: -8.56% (308ns → 282ns) TreeDecodeLarge: -5.47% (16.95µs → 16.02µs) HashReadFrom: -2.47% (65.18ns → 63.57ns) Overall geometric: -4.02%Memory: Allocations: unchanged (same count) Bytes/op: +6.81% to +14.45% increase (acceptable tradeoff)The speed improvements are consistent across all tree sizes, with the mostsignificant gains on small trees (-8.56%). CPU profiling showed hash readingoverhead reduced from 13% to 7% of total Tree.Decode time.This optimization benefits all callers automatically with no code changesrequired, as the method signature and behavior remain identical.
Contributor
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
Pull request overview
This PR optimizesObjectID.ReadFrom by replacingencoding/binary.Read withio.ReadFull to eliminate reflection overhead. The change is motivated by profiling data showing hash reading operations consuming significant CPU time during Tree.Decode operations in hot paths like git clone.
Key changes:
- Replaced
binary.Readwithio.ReadFullinObjectID.ReadFrommethod - Added hash zeroing on read errors for cleaner error state
- Added comprehensive test coverage for the optimization with both success and error cases
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| plumbing/objectid.go | Replacedbinary.Read withio.ReadFull in ReadFrom method, added error handling to zero hash on failure, updated documentation |
| plumbing/objectid_test.go | AddedTestReadFromOptimization with comprehensive test cases for SHA1/SHA256 valid reads, partial reads, and empty reader scenarios |
💡Add Copilot custom instructions for smarter, more guided reviews.Learn how to get started.
Uh oh!
There was an error while loading.Please reload this page.
Sign up for freeto join this conversation on GitHub. Already have an account?Sign in to comment
Labels
None yet
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.
Replace encoding/binary.Read with io.ReadFull in ObjectID.ReadFrom for better performance, eliminating reflection overhead.
Background:
The ObjectID.ReadFrom method used encoding/binary.Read, which employs reflection to copy bytes into the hash array. This adds unnecessary overhead in hot paths, particularly during Tree.Decode operations that are called thousands of times during clone operations.
Changes:
Updated ObjectID.ReadFrom implementation to use io.ReadFull directly instead of encoding/binary.Read
Added error handling to zero out the hash on read errors for cleaner state
Updated test coverage to verify the optimized implementation:
Technical Notes:
This is a backward-compatible change. The method signature remains identical and it still implements io.ReaderFrom. For byte arrays (like hashes), binary.Read's BigEndian parameter is a no-op since no byte-order conversion occurs. io.ReadFull produces identical results while being significantly faster by avoiding reflection.
Performance Impact:
Benchmark results from 10 iterations (AMD Ryzen 7 6800U):
Speed improvements (statistically significant, p<0.05):
TreeDecode: -2.91% (1429ns → 1387ns)
TreeDecodeSmall: -8.56% (308ns → 282ns)
TreeDecodeLarge: -5.47% (16.95µs → 16.02µs)
HashReadFrom: -2.47% (65.18ns → 63.57ns)
Overall geometric: -4.02%
Memory:
Allocations: unchanged (same count)
Bytes/op: +6.81% to +14.45% increase (acceptable tradeoff)
The speed improvements are consistent across all tree sizes, with the most significant gains on small trees (-8.56%). CPU profiling showed hash reading overhead reduced from 13% to 7% of total Tree.Decode time.
This optimization benefits all callers automatically with no code changes required, as the method signature and behavior remain identical.