Repository navigation
Optimize CPU uint8 GatherBlockQuantized dequantization - #32621
Merged
kunal-vaishnavi merged 2 commits intoSep 17, 2026
Merged
Conversation
Co-authored-by: kunal-vaishnavi <115581922+kunal-vaishnavi@users.noreply.github.com>
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
Copilot created this pull request from a session on behalf of
kunal-vaishnavi
September 15, 2026 22:56
kunal-vaishnavi
marked this pull request as ready for review
September 15, 2026 22:58
Contributor
There was a problem hiding this comment.
🟡 Changes recommended
The new 256-element batch boundary lacks regression coverage.
Get a fresh assessment by requesting another Copilot review.
Pull request overview
Optimizes CPU dequantization for packed uint8 GatherBlockQuantized data.
Changes:
- Batch-unpacks 2-, 4-, and 8-bit values.
- Uses MLAS for vectorized dequantization and FP16 conversion.
- Preserves existing 4-bit wrapper paths.
File summaries
| File | Description |
|---|---|
onnxruntime/contrib_ops/cpu/quantization/gather_block_quantized.cc |
Implements batched MLAS dequantization for uint8. |
Review details
- Files reviewed: 1/1 changed files
- Comments generated: 1
- Review effort level: Balanced
💡 Configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
Co-authored-by: kunal-vaishnavi <115581922+kunal-vaishnavi@users.noreply.github.com>
Akshay Sonawane (apsonawane)
approved these changes
Sep 16, 2026
kunal-vaishnavi
approved these changes
Sep 16, 2026
kunal-vaishnavi
enabled auto-merge (squash)
September 16, 2026 01:11
kunal-vaishnavi
deleted the
copilot/recover-revert-gatherblockquantized-edits
branch
September 17, 2026 03:03
This was referenced Oct 9, 2026
This was referenced Oct 14, 2026
This was referenced Oct 21, 2026
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
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.
Description
GatherBlockQuantizedoptimization from Optimize CPU uint8 GatherBlockQuantized dequantization #32598.uint8values.Int4x2andUInt4x2paths.Motivation and Context
The stacked-PR automation moved #32598’s head to its merged base, making the PR empty without incorporating its optimization into
main. This replacement reapplies the recoverable one-file diff from its last complete head.