Prevent concurrent access to local breaker in rerank #128162
There are no files selected for viewing
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -19,14 +19,15 @@ | |
| import org.elasticsearch.compute.operator.DriverContext; | ||
| import org.elasticsearch.compute.operator.EvalOperator.ExpressionEvaluator; | ||
| import org.elasticsearch.compute.operator.Operator; | ||
| import org.elasticsearch.core.Releasable; | ||
| import org.elasticsearch.core.Releasables; | ||
| import org.elasticsearch.inference.TaskType; | ||
| import org.elasticsearch.xpack.core.inference.action.InferenceAction; | ||
| import org.elasticsearch.xpack.core.inference.results.RankedDocsResults; | ||
|
|
||
| import java.util.List; | ||
|
|
||
| public class RerankOperator extends AsyncOperator<RerankOperator.OngoingRerank> { | ||
|
|
||
| // Move to a setting. | ||
| private static final int MAX_INFERENCE_WORKER = 10; | ||
|
|
@@ -85,20 +86,16 @@ public RerankOperator( | |
| } | ||
|
|
||
| @Override | ||
| protected void performAsync(Page inputPage, ActionListener<OngoingRerank> listener) { | ||
| // Ensure input page blocks are released when the listener is called. | ||
| listener = listener.delegateResponse((l, e) -> { | ||
| releasePageOnAnyThread(inputPage); | ||
| l.onFailure(e); | ||
| }); | ||
| try { | ||
|
Contributor
There was a problem hiding this comment. If it isn't safe to create blocks here (And therefore update the CB), why is it safe to read a block from transport like we do in
Member
Author
There was a problem hiding this comment.
|
||
| inferenceRunner.doInference(buildInferenceRequest(inputPage), listener.map(resp -> new OngoingRerank(inputPage, resp))); | ||
| } catch (Exception e) { | ||
| listener.onFailure(e); | ||
| } | ||
| } | ||
|
|
||
|
|
@@ -108,91 +105,106 @@ protected void doClose() { | |
| } | ||
|
|
||
| @Override | ||
| protected void releaseFetchedOnAnyThread(OngoingRerank result) { | ||
| releasePageOnAnyThread(result.inputPage); | ||
| } | ||
|
|
||
| @Override | ||
| public Page getOutput() { | ||
| var fetched = fetchFromBuffer(); | ||
| if (fetched == null) { | ||
| return null; | ||
| } else { | ||
| return fetched.buildOutput(blockFactory, scoreChannel); | ||
| } | ||
| } | ||
|
|
||
| @Override | ||
| public String toString() { | ||
| return "RerankOperator[inference_id=[" + inferenceId + "], query=[" + queryText + "], score_channel=[" + scoreChannel + "]]"; | ||
| } | ||
|
|
||
| private InferenceAction.Request buildInferenceRequest(Page inputPage) { | ||
| try (BytesRefBlock encodedRowsBlock = (BytesRefBlock) rowEncoder.eval(inputPage)) { | ||
| assert (encodedRowsBlock.getPositionCount() == inputPage.getPositionCount()); | ||
| String[] inputs = new String[inputPage.getPositionCount()]; | ||
| BytesRef buffer = new BytesRef(); | ||
|
|
||
| for (int pos = 0; pos < inputPage.getPositionCount(); pos++) { | ||
| if (encodedRowsBlock.isNull(pos)) { | ||
| inputs[pos] = ""; | ||
| } else { | ||
| buffer = encodedRowsBlock.getBytesRef(encodedRowsBlock.getFirstValueIndex(pos), buffer); | ||
| inputs[pos] = BytesRefs.toString(buffer); | ||
| } | ||
| } | ||
|
|
||
| return InferenceAction.Request.builder(inferenceId, TaskType.RERANK).setInput(List.of(inputs)).setQuery(queryText).build(); | ||
| } | ||
| } | ||
|
|
||
| public static final class OngoingRerank { | ||
| final Page inputPage; | ||
| final Double[] rankedScores; | ||
|
|
||
| OngoingRerank(Page inputPage, InferenceAction.Response resp) { | ||
| if (resp.getResults() instanceof RankedDocsResults == false) { | ||
| releasePageOnAnyThread(inputPage); | ||
| throw new IllegalStateException( | ||
| "Inference result has wrong type. Got [" | ||
| + resp.getResults().getClass() | ||
| + "] while expecting [" | ||
| + RankedDocsResults.class | ||
| + "]" | ||
| ); | ||
|
|
||
| } | ||
| final var results = (RankedDocsResults) resp.getResults(); | ||
| this.inputPage = inputPage; | ||
| this.rankedScores = extractRankedScores(inputPage.getPositionCount(), results); | ||
| } | ||
|
|
||
| private static Double[] extractRankedScores(int positionCount, RankedDocsResults rankedDocsResults) { | ||
| Double[] sortedRankedDocsScores = new Double[positionCount]; | ||
| for (RankedDocsResults.RankedDoc rankedDoc : rankedDocsResults.getRankedDocs()) { | ||
| sortedRankedDocsScores[rankedDoc.index()] = (double) rankedDoc.relevanceScore(); | ||
| } | ||
| return sortedRankedDocsScores; | ||
| } | ||
|
|
||
| Page buildOutput(BlockFactory blockFactory, int scoreChannel) { | ||
| int blockCount = Integer.max(inputPage.getBlockCount(), scoreChannel + 1); | ||
| Block[] blocks = new Block[blockCount]; | ||
| Page outputPage = null; | ||
| try (Releasable ignored = inputPage::releaseBlocks) { | ||
| for (int b = 0; b < blockCount; b++) { | ||
| if (b == scoreChannel) { | ||
| blocks[b] = buildScoreBlock(blockFactory); | ||
| } else { | ||
| blocks[b] = inputPage.getBlock(b); | ||
| blocks[b].incRef(); | ||
| } | ||
| } | ||
| outputPage = new Page(blocks); | ||
| return outputPage; | ||
| } finally { | ||
| if (outputPage == null) { | ||
| Releasables.closeExpectNoException(blocks); | ||
| } | ||
| } | ||
| } | ||
|
|
||
| private Block buildScoreBlock(BlockFactory blockFactory) { | ||
| try (DoubleBlock.Builder scoreBlockFactory = blockFactory.newDoubleBlockBuilder(rankedScores.length)) { | ||
| for (Double rankedScore : rankedScores) { | ||
| if (rankedScore != null) { | ||
| scoreBlockFactory.appendDouble(rankedScore); | ||
| } else { | ||
| scoreBlockFactory.appendNull(); | ||
| } | ||
| } | ||
| return scoreBlockFactory.build(); | ||
| } | ||
| } | ||
| } | ||
| } | ||
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
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.
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.
Should we wrap this in something that
asserts that we're on the same thread every time?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.
++ I pushed ce54583.