Fuse RLE decoding and view gathering for StringView dictionary#9586
Draft
Dandandan wants to merge 18 commits intoapache:mainfrom
Draft
Fuse RLE decoding and view gathering for StringView dictionary#9586Dandandan wants to merge 18 commits intoapache:mainfrom
Dandandan wants to merge 18 commits intoapache:mainfrom
Conversation
When bit_width guarantees all possible indices fit within the dictionary, use unchecked indexing to allow LLVM to unroll the dict gather loop 4x with paired loads/stores instead of scalar with per-element bounds checks. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add RleDecodedBatch enum and get_batch_direct method that exposes RLE vs bit-packed batches via callback, allowing callers to handle each case optimally without going through the index buffer. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace if/else checked/unchecked branching with a single branchless .min(max_idx) clamp. This prevents UB on corrupt parquet files while avoiding per-element bounds checks. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
These are only used by the arrow dictionary_index decoder. Without the arrow feature, they appear as dead code to clippy. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
When bit_width guarantees all possible indices fit within the dictionary, use unchecked access to eliminate per-element bounds checks. Also skip buffer management when all dictionary views are inlined (<=12 bytes). Generates a clean 8-instruction gather loop for the common case (all_indices_valid + base_buffer_idx=0) and a branchless 14-instruction loop for the non-zero buffer offset case. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Reserve the full output capacity upfront before the decode loop, eliminating per-chunk reallocation checks inside extend. This gives a ~25% speedup for dictionary-encoded StringView reads. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
For RLE runs, look up the dict view once and repeat directly with repeat_n, skipping the index buffer entirely. For bit-packed runs, decode indices to a stack-local buffer and gather immediately. Skip buffer management when all dictionary views are inlined (<=12 bytes). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Apply .min(max_idx) clamping in gather_views to prevent UB on corrupt data while keeping the hot loop branchless. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Reserve offsets capacity upfront before the decode loop to avoid per-chunk reallocation. ~3.5% improvement for StringArray dict reads. 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
|
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…thering # Conflicts: # parquet/src/arrow/array_reader/byte_view_array.rs
Contributor
Author
|
run benchmark arrow_reader_clickbench |
|
🤖 Arrow criterion benchmark running (GKE) | trigger File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
…er_views Precompute all_views_inlined when the dictionary is set instead of recomputing it on every read call. Replace branchless index clamping with proper bounds checking using the deferred error pattern to keep the extend loop fast. 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 File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
Contributor
Author
|
run benchmark arrow_reader_clickbench |
|
🤖 Arrow criterion benchmark running (GKE) | trigger File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
…exing The clamping with get_unchecked + min(max_idx) silently returned wrong values for corrupt data. Normal indexing will panic on invalid keys, which is the correct behavior. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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
2 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 #9582
Rationale
StringView dictionary decoding currently goes through an intermediate index buffer: decode indices → gather views. For RLE runs (which are common), this roundtrip is unnecessary.
What changes are included in this PR?
repeat_nto fill views directly, skipping the index buffer entirelyBased on #9583 (branchless clamping PR).
Are there any user-facing changes?
No.
🤖 Generated with Claude Code