HIVE-29598: Fix vectorized outer join wrong results due to stale scratch column values #6486
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
| Original file line number | Diff line number | Diff line change | ||||||
|---|---|---|---|---|---|---|---|---|
| @@ -0,0 +1,264 @@ | ||||||||
| /* | ||||||||
| * Licensed to the Apache Software Foundation (ASF) under one | ||||||||
| * or more contributor license agreements. See the NOTICE file | ||||||||
| * distributed with this work for additional information | ||||||||
| * regarding copyright ownership. The ASF licenses this file | ||||||||
| * to you under the Apache License, Version 2.0 (the | ||||||||
| * "License"); you may not use this file except in compliance | ||||||||
| * with the License. You may obtain a copy of the License at | ||||||||
| * | ||||||||
| * http://www.apache.org/licenses/LICENSE-2.0 | ||||||||
| * | ||||||||
| * Unless required by applicable law or agreed to in writing, software | ||||||||
| * distributed under the License is distributed on an "AS IS" BASIS, | ||||||||
| * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | ||||||||
| * See the License for the specific language governing permissions and | ||||||||
| * limitations under the License. | ||||||||
| */ | ||||||||
|
|
||||||||
| package org.apache.hadoop.hive.ql.exec.vector.mapjoin; | ||||||||
|
|
||||||||
| import org.apache.hadoop.hive.common.type.HiveIntervalDayTime; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.BytesColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.ColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.DecimalColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.DoubleColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.IntervalDayTimeColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.LongColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.TimestampColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.VectorizedRowBatch; | ||||||||
| import org.apache.hadoop.hive.ql.exec.vector.VoidColumnVector; | ||||||||
| import org.apache.hadoop.hive.ql.metadata.HiveException; | ||||||||
| import org.junit.jupiter.api.Test; | ||||||||
| import org.junit.jupiter.params.ParameterizedTest; | ||||||||
| import org.junit.jupiter.params.provider.Arguments; | ||||||||
| import org.junit.jupiter.params.provider.MethodSource; | ||||||||
|
|
||||||||
| import java.io.IOException; | ||||||||
| import java.nio.charset.StandardCharsets; | ||||||||
| import java.util.ArrayList; | ||||||||
| import java.util.Arrays; | ||||||||
| import java.util.List; | ||||||||
| import java.util.stream.Stream; | ||||||||
|
|
||||||||
| import static org.junit.jupiter.api.Assertions.assertEquals; | ||||||||
| import static org.junit.jupiter.api.Assertions.assertFalse; | ||||||||
| import static org.junit.jupiter.api.Assertions.assertNull; | ||||||||
| import static org.junit.jupiter.api.Assertions.assertTrue; | ||||||||
|
|
||||||||
| /** | ||||||||
| * HIVE-29598: verifies {@link VectorMapJoinOuterGenerateResultOperator} clears | ||||||||
| * every small-table slot for unmatched rows, so stale values cannot carry over | ||||||||
| * past the null marking. | ||||||||
| */ | ||||||||
| class TestVectorMapJoinOuterGenerateResultOperator { | ||||||||
|
|
||||||||
| /** Concrete subclass that exposes the generateOuterNulls* methods to tests. */ | ||||||||
| private static final class TestableOuterOp extends VectorMapJoinOuterGenerateResultOperator { | ||||||||
| @Override | ||||||||
| protected String getLoggingPrefix() { | ||||||||
| return "test"; | ||||||||
| } | ||||||||
|
|
||||||||
| @Override | ||||||||
| public void processBatch(VectorizedRowBatch batch) { | ||||||||
| } | ||||||||
|
Contributor
There was a problem hiding this comment. SQ feedback: empty method body.
Suggested change
|
||||||||
| } | ||||||||
|
|
||||||||
| /** | ||||||||
| * Records {@code clearSlotValue} invocations to verify the operator dispatches | ||||||||
| * through {@code clearValue}, not just produces the slot-clearing side effect. | ||||||||
| */ | ||||||||
| private static class TrackingLongColumnVector extends LongColumnVector { | ||||||||
| final List<Integer> clearedIndices = new ArrayList<>(); | ||||||||
|
|
||||||||
| TrackingLongColumnVector(int size) { | ||||||||
| super(size); | ||||||||
| } | ||||||||
|
|
||||||||
| @Override | ||||||||
| protected void clearSlotValue(int elementNum) { | ||||||||
| super.clearSlotValue(elementNum); | ||||||||
| clearedIndices.add(elementNum); | ||||||||
| } | ||||||||
| } | ||||||||
|
|
||||||||
| @Test | ||||||||
| void generateOuterNullsCallsClearValueOnEachMappedColumnForEachUnmatchedRow() throws HiveException, IOException { | ||||||||
| TestableOuterOp op = new TestableOuterOp(); | ||||||||
| op.outerSmallTableKeyColumnMap = new int[] {0}; | ||||||||
| op.smallTableValueColumnMap = new int[] {1, 2}; | ||||||||
|
|
||||||||
| VectorizedRowBatch batch = new VectorizedRowBatch(3, 4); | ||||||||
| TrackingLongColumnVector keyCol = new TrackingLongColumnVector(4); | ||||||||
| TrackingLongColumnVector valCol1 = new TrackingLongColumnVector(4); | ||||||||
| TrackingLongColumnVector valCol2 = new TrackingLongColumnVector(4); | ||||||||
| keyCol.vector[1] = 99L; | ||||||||
| valCol1.vector[1] = 88L; | ||||||||
| valCol2.vector[3] = 77L; | ||||||||
| batch.cols[0] = keyCol; | ||||||||
| batch.cols[1] = valCol1; | ||||||||
| batch.cols[2] = valCol2; | ||||||||
|
|
||||||||
| int[] noMatchs = new int[] {1, 3}; | ||||||||
| op.generateOuterNulls(batch, noMatchs, noMatchs.length); | ||||||||
|
|
||||||||
| assertEquals(Arrays.asList(1, 3), keyCol.clearedIndices); | ||||||||
| assertEquals(Arrays.asList(1, 3), valCol1.clearedIndices); | ||||||||
| assertEquals(Arrays.asList(1, 3), valCol2.clearedIndices); | ||||||||
|
|
||||||||
| assertFalse(keyCol.noNulls); | ||||||||
| assertTrue(keyCol.isNull[1]); | ||||||||
| assertTrue(keyCol.isNull[3]); | ||||||||
| assertFalse(keyCol.isNull[0]); | ||||||||
| assertFalse(keyCol.isNull[2]); | ||||||||
|
|
||||||||
| assertEquals(0L, keyCol.vector[1]); | ||||||||
| assertEquals(0L, valCol1.vector[1]); | ||||||||
| assertEquals(0L, valCol2.vector[3]); | ||||||||
| } | ||||||||
|
|
||||||||
| @Test | ||||||||
| void generateOuterNullsRepeatedAllCallsClearValueAtIndexZeroForEachMappedColumn() throws HiveException { | ||||||||
| TestableOuterOp op = new TestableOuterOp(); | ||||||||
| op.outerSmallTableKeyColumnMap = new int[] {0}; | ||||||||
| op.smallTableValueColumnMap = new int[] {1}; | ||||||||
|
|
||||||||
| VectorizedRowBatch batch = new VectorizedRowBatch(2, 4); | ||||||||
| TrackingLongColumnVector keyCol = new TrackingLongColumnVector(4); | ||||||||
| TrackingLongColumnVector valCol = new TrackingLongColumnVector(4); | ||||||||
| keyCol.vector[0] = 42L; | ||||||||
| valCol.vector[0] = 84L; | ||||||||
| batch.cols[0] = keyCol; | ||||||||
| batch.cols[1] = valCol; | ||||||||
|
|
||||||||
| op.generateOuterNullsRepeatedAll(batch); | ||||||||
|
|
||||||||
| assertEquals(Arrays.asList(0), keyCol.clearedIndices); | ||||||||
| assertEquals(Arrays.asList(0), valCol.clearedIndices); | ||||||||
|
|
||||||||
| // isRepeating is set by the operator, not by clearValue. | ||||||||
| assertFalse(keyCol.noNulls); | ||||||||
| assertTrue(keyCol.isNull[0]); | ||||||||
| assertTrue(keyCol.isRepeating); | ||||||||
| assertFalse(valCol.noNulls); | ||||||||
| assertTrue(valCol.isNull[0]); | ||||||||
| assertTrue(valCol.isRepeating); | ||||||||
|
|
||||||||
| assertEquals(0L, keyCol.vector[0]); | ||||||||
| assertEquals(0L, valCol.vector[0]); | ||||||||
| } | ||||||||
|
|
||||||||
| @Test | ||||||||
| void generateOuterNullsSetsBookkeepingOnTypeWithNoClearSlotValueOverride() throws HiveException, IOException { | ||||||||
| // VoidColumnVector inherits the base no-op clearSlotValue — verifies the | ||||||||
| // operator still drives the null-marking through clearValue() on a type | ||||||||
| // without a per-slot value to zero. | ||||||||
| TestableOuterOp op = new TestableOuterOp(); | ||||||||
| op.outerSmallTableKeyColumnMap = new int[] {}; | ||||||||
| op.smallTableValueColumnMap = new int[] {0}; | ||||||||
|
|
||||||||
| VectorizedRowBatch batch = new VectorizedRowBatch(1, 4); | ||||||||
| VoidColumnVector voidCol = new VoidColumnVector(4); | ||||||||
| batch.cols[0] = voidCol; | ||||||||
|
|
||||||||
| int[] noMatchs = new int[] {1, 3}; | ||||||||
| op.generateOuterNulls(batch, noMatchs, noMatchs.length); | ||||||||
|
|
||||||||
| assertFalse(voidCol.noNulls); | ||||||||
| assertTrue(voidCol.isNull[1]); | ||||||||
| assertTrue(voidCol.isNull[3]); | ||||||||
| assertFalse(voidCol.isNull[0]); | ||||||||
| assertFalse(voidCol.isNull[2]); | ||||||||
| } | ||||||||
|
|
||||||||
| /** | ||||||||
| * For each {@link ColumnVector} subclass whose {@code clearSlotValue} is | ||||||||
| * overridden, verifies the operator's call through {@code clearValue} reaches | ||||||||
| * the override and clears the slot to the type's cleared state. | ||||||||
| */ | ||||||||
| @ParameterizedTest(name = "{0}") | ||||||||
| @MethodSource("modifiedColumnVectorTypes") | ||||||||
| void generateOuterNullsClearsSlotForEachModifiedType( | ||||||||
| String typeName, | ||||||||
| ColumnVector cv, | ||||||||
| Runnable preLoad, | ||||||||
| Runnable assertSlotCleared) throws HiveException, IOException { | ||||||||
|
|
||||||||
| TestableOuterOp op = new TestableOuterOp(); | ||||||||
| op.outerSmallTableKeyColumnMap = new int[] {}; | ||||||||
| op.smallTableValueColumnMap = new int[] {0}; | ||||||||
|
|
||||||||
| VectorizedRowBatch batch = new VectorizedRowBatch(1, 4); | ||||||||
| preLoad.run(); | ||||||||
| batch.cols[0] = cv; | ||||||||
|
|
||||||||
| int[] noMatchs = new int[] {2}; | ||||||||
| op.generateOuterNulls(batch, noMatchs, noMatchs.length); | ||||||||
|
|
||||||||
| assertTrue(cv.isNull[2]); | ||||||||
| assertFalse(cv.noNulls); | ||||||||
| assertSlotCleared.run(); | ||||||||
| } | ||||||||
|
|
||||||||
| static Stream<Arguments> modifiedColumnVectorTypes() { | ||||||||
| final LongColumnVector longCv = new LongColumnVector(4); | ||||||||
| final DoubleColumnVector doubleCv = new DoubleColumnVector(4); | ||||||||
| final BytesColumnVector bytesCv = new BytesColumnVector(4); | ||||||||
| final DecimalColumnVector decCv = new DecimalColumnVector(4, 18, 4); | ||||||||
| final TimestampColumnVector tsCv = new TimestampColumnVector(4); | ||||||||
| final IntervalDayTimeColumnVector ivCv = new IntervalDayTimeColumnVector(4); | ||||||||
|
|
||||||||
| return Stream.of( | ||||||||
| Arguments.of( | ||||||||
| "LongColumnVector", | ||||||||
| longCv, | ||||||||
| (Runnable) () -> longCv.vector[2] = 999L, | ||||||||
| (Runnable) () -> assertEquals(0L, longCv.vector[2])), | ||||||||
| Arguments.of( | ||||||||
| "DoubleColumnVector", | ||||||||
| doubleCv, | ||||||||
| (Runnable) () -> doubleCv.vector[2] = 3.14, | ||||||||
| (Runnable) () -> assertEquals(0.0, doubleCv.vector[2])), | ||||||||
| Arguments.of( | ||||||||
| "BytesColumnVector", | ||||||||
| bytesCv, | ||||||||
| (Runnable) () -> { | ||||||||
| bytesCv.vector[2] = "stale".getBytes(StandardCharsets.UTF_8); | ||||||||
| bytesCv.start[2] = 1; | ||||||||
| bytesCv.length[2] = 3; | ||||||||
| }, | ||||||||
| (Runnable) () -> { | ||||||||
| assertNull(bytesCv.vector[2]); | ||||||||
| assertEquals(0, bytesCv.start[2]); | ||||||||
| assertEquals(0, bytesCv.length[2]); | ||||||||
| }), | ||||||||
| Arguments.of( | ||||||||
| "DecimalColumnVector", | ||||||||
| decCv, | ||||||||
| (Runnable) () -> decCv.vector[2].setFromLong(999L), | ||||||||
| (Runnable) () -> assertEquals(0L, decCv.vector[2].serialize64(decCv.scale))), | ||||||||
| Arguments.of( | ||||||||
| "TimestampColumnVector", | ||||||||
| tsCv, | ||||||||
| (Runnable) () -> { | ||||||||
| tsCv.time[2] = 1234567890000L; | ||||||||
| tsCv.nanos[2] = 999; | ||||||||
| }, | ||||||||
| (Runnable) () -> { | ||||||||
| // setNullValue convention: time = 0, nanos = 1 | ||||||||
| assertEquals(0L, tsCv.time[2]); | ||||||||
| assertEquals(1, tsCv.nanos[2]); | ||||||||
| }), | ||||||||
| Arguments.of( | ||||||||
| "IntervalDayTimeColumnVector", | ||||||||
| ivCv, | ||||||||
| (Runnable) () -> ivCv.set(2, new HiveIntervalDayTime(5, 0)), | ||||||||
| (Runnable) () -> { | ||||||||
| // setNullValue convention: totalSeconds = 0, nanos = 1 | ||||||||
| assertEquals(0L, ivCv.getTotalSeconds(2)); | ||||||||
| assertEquals(1, ivCv.getNanos(2)); | ||||||||
| }) | ||||||||
| ); | ||||||||
| } | ||||||||
| } | ||||||||
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 |
|---|---|---|
| @@ -0,0 +1,43 @@ | ||
| SET hive.auto.convert.join=true; | ||
| SET hive.auto.convert.join.noconditionaltask=true; | ||
|
|
||
| -- SORT_QUERY_RESULTS | ||
|
|
||
| -- HIVE-29598: regression test for stale scratch-slot values in vectorized | ||
| -- outer-join MapJoin. MAX() acts as an aggregation barrier so Calcite cannot | ||
| -- inline the inner expression and simplify the bug surface away. | ||
|
|
||
| CREATE TABLE t (k STRING, v STRING) STORED AS ORC; | ||
|
|
||
| INSERT INTO t VALUES | ||
| ('A','1'),('A','2'),('A','3'), | ||
| ('B','2'),('B','3'), | ||
| ('C','3'), | ||
| ('D','1'),('D','3'); | ||
|
|
||
| WITH | ||
| probe AS ( | ||
| SELECT k, v, (CAST(v AS INT) > 0) AS p_bool | ||
| FROM t WHERE CAST(v AS INT) >= 3 | ||
| ), | ||
| small_side AS ( | ||
| SELECT k, v, (CAST(v AS INT) > 9999) AS s_bool | ||
| FROM t | ||
| ), | ||
| classified AS ( | ||
| SELECT p.k, p.v, CAST((s.s_bool OR p.p_bool) AS INT) AS observed_value | ||
| FROM probe p | ||
| LEFT JOIN small_side s | ||
| ON p.k = s.k | ||
| AND CAST(p.v AS INT) - 1 = CAST(s.v AS INT) | ||
| ), | ||
| diagnosed AS ( | ||
| SELECT k, v, MAX(observed_value) AS observed_value | ||
| FROM classified | ||
| GROUP BY k, v | ||
| ) | ||
| SELECT k, v, | ||
| observed_value AS observed_value_returned_by_select, | ||
| 1 AS required_value_per_sql_semantics | ||
| FROM diagnosed | ||
| WHERE observed_value = 0; |
Oops, something went wrong.
Oops, something went wrong.
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.
SQ feedback: empty/trivial stub body.