Skip to content
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.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Specialize batched column-wise hashing for RunLengthEncodedBlock #19723

Merged
merged 3 commits into from
Nov 15, 2023
Merged

Specialize batched column-wise hashing for RunLengthEncodedBlock #19723

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
ebe96cb
Short circuit on zero-length hash length
pettyjamesm Nov 13, 2023
5617bce
Special case handling of RLE Blocks for batched hashing
pettyjamesm Nov 13, 2023
9ac52d5
Avoid retaining generated classes in TestFlatHashStrategy
pettyjamesm Nov 13, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
79 changes: 58 additions & 21 deletions core/trino-main/src/main/java/io/trino/operator/FlatHashStrategyCompiler.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,15 +23,18 @@
import io.airlift.bytecode.Variable;
import io.airlift.bytecode.control.ForLoop;
import io.airlift.bytecode.control.IfStatement;
import io.airlift.bytecode.expression.BytecodeExpression;
import io.trino.operator.scalar.CombineHashFunction;
import io.trino.spi.block.Block;
import io.trino.spi.block.BlockBuilder;
import io.trino.spi.block.RunLengthEncodedBlock;
import io.trino.spi.type.Type;
import io.trino.spi.type.TypeOperators;
import io.trino.sql.gen.CallSiteBinder;

import java.lang.invoke.MethodHandle;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
Expand All @@ -51,6 +54,7 @@
import static io.airlift.bytecode.expression.BytecodeExpressions.constantInt;
import static io.airlift.bytecode.expression.BytecodeExpressions.constantLong;
import static io.airlift.bytecode.expression.BytecodeExpressions.constantTrue;
import static io.airlift.bytecode.expression.BytecodeExpressions.equal;
import static io.airlift.bytecode.expression.BytecodeExpressions.invokeDynamic;
import static io.airlift.bytecode.expression.BytecodeExpressions.invokeStatic;
import static io.airlift.bytecode.expression.BytecodeExpressions.lessThan;
Expand Down Expand Up @@ -378,6 +382,8 @@ private static void generateHashBlocksBatched(ClassDefinition definition, List<K
BytecodeBlock body = methodDefinition.getBody();
body.append(invokeStatic(Objects.class, "checkFromIndexSize", int.class, constantInt(0), length, hashes.length()).pop());

BytecodeBlock nonEmptyLength = new BytecodeBlock();

Map<Type, MethodDefinition> typeMethods = new HashMap<>();
for (KeyField keyField : keyFields) {
MethodDefinition method;
Expand All @@ -393,9 +399,13 @@ private static void generateHashBlocksBatched(ClassDefinition definition, List<K
typeMethods.put(keyField.type(), method);
}
}
body.append(invokeStatic(method, blocks.getElement(keyField.index()), hashes, offset, length));
nonEmptyLength.append(invokeStatic(method, blocks.getElement(keyField.index()), hashes, offset, length));
}
body.ret();

body.append(new IfStatement("if (length != 0)")
.condition(equal(length, constantInt(0)))
.ifFalse(nonEmptyLength))
.ret();
}

private static MethodDefinition generateHashBlockVectorized(ClassDefinition definition, KeyField field, CallSiteBinder callSiteBinder)
Expand All @@ -422,35 +432,62 @@ private static MethodDefinition generateHashBlockVectorized(ClassDefinition defi
Variable mayHaveNull = scope.declareVariable(boolean.class, "mayHaveNull");
Variable hash = scope.declareVariable(long.class, "hash");

body.append(mayHaveNull.set(block.invoke("mayHaveNull", boolean.class)));
body.append(position.set(invokeStatic(Objects.class, "checkFromToIndex", int.class, offset, add(offset, length), block.invoke("getPositionCount", int.class))));
body.append(invokeStatic(Objects.class, "checkFromIndexSize", int.class, constantInt(0), length, hashes.length()).pop());

BytecodeBlock loopBody = new BytecodeBlock().append(new IfStatement("if (mayHaveNull && block.isNull(position))")
.condition(and(mayHaveNull, block.invoke("isNull", boolean.class, position)))
.ifTrue(hash.set(constantLong(NULL_HASH_CODE)))
.ifFalse(hash.set(invokeDynamic(
BOOTSTRAP_METHOD,
ImmutableList.of(callSiteBinder.bind(field.hashBlockMethod()).getBindingId()),
"hash",
long.class,
block,
position))));
BytecodeExpression computeHashNonNull = invokeDynamic(
BOOTSTRAP_METHOD,
ImmutableList.of(callSiteBinder.bind(field.hashBlockMethod()).getBindingId()),
"hash",
long.class,
block,
position);

BytecodeExpression setHashExpression;
if (field.index() == 0) {
// hashes[index] = hash;
loopBody.append(hashes.setElement(index, hash));
setHashExpression = hashes.setElement(index, hash);
}
else {
// hashes[index] = CombineHashFunction.getHash(hashes[index], hash);
loopBody.append(hashes.setElement(index, invokeStatic(CombineHashFunction.class, "getHash", long.class, hashes.getElement(index), hash)));
setHashExpression = hashes.setElement(index, invokeStatic(CombineHashFunction.class, "getHash", long.class, hashes.getElement(index), hash));
}

BytecodeBlock rleHandling = new BytecodeBlock()
.append(new IfStatement("hash = block.isNull(position) ? NULL_HASH_CODE : hash(block, position)")
.condition(block.invoke("isNull", boolean.class, position))
.ifTrue(hash.set(constantLong(NULL_HASH_CODE)))
.ifFalse(hash.set(computeHashNonNull)));
if (field.index() == 0) {
// Arrays.fill(hashes, 0, length, hash)
rleHandling.append(invokeStatic(Arrays.class, "fill", void.class, hashes, constantInt(0), length, hash));
}
else {
rleHandling.append(new ForLoop("for (int index = 0; index < length; index++) { hashes[index] = CombineHashFunction.getHash(hashes[index], hash); }")
.initialize(index.set(constantInt(0)))
.condition(lessThan(index, length))
.update(index.increment())
.body(setHashExpression));
}
loopBody.append(position.increment());

body.append(new ForLoop("for (index = 0; index < length; index++)")
.initialize(index.set(constantInt(0)))
.condition(lessThan(index, length))
.update(index.increment())
.body(loopBody))
BytecodeBlock computeHashLoop = new BytecodeBlock()
.append(mayHaveNull.set(block.invoke("mayHaveNull", boolean.class)))
.append(new ForLoop("for (int index = 0; index < length; index++)")
.initialize(index.set(constantInt(0)))
.condition(lessThan(index, length))
.update(index.increment())
.body(new BytecodeBlock()
.append(new IfStatement("if (mayHaveNull && block.isNull(position))")
.condition(and(mayHaveNull, block.invoke("isNull", boolean.class, position)))
.ifTrue(hash.set(constantLong(NULL_HASH_CODE)))
.ifFalse(hash.set(computeHashNonNull)))
dain marked this conversation as resolved.
Show resolved Hide resolved
.append(setHashExpression)
.append(position.increment())));

body.append(new IfStatement("if (block instanceof RunLengthEncodedBlock)")
.condition(block.instanceOf(RunLengthEncodedBlock.class))
.ifTrue(rleHandling)
.ifFalse(computeHashLoop))
.ret();

return methodDefinition;
Expand Down
48 changes: 37 additions & 11 deletions core/trino-main/src/test/java/io/trino/operator/TestFlatHashStrategy.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@

import com.google.common.collect.ImmutableList;
import io.trino.spi.block.Block;
import io.trino.spi.block.RunLengthEncodedBlock;
import io.trino.spi.connector.ConnectorSession;
import io.trino.spi.type.ArrayType;
import io.trino.spi.type.MapType;
Expand Down Expand Up @@ -44,19 +45,31 @@
import static io.trino.spi.type.VarbinaryType.VARBINARY;
import static io.trino.spi.type.VarcharType.VARCHAR;
import static io.trino.type.IpAddressType.IPADDRESS;
import static org.junit.jupiter.api.Assertions.assertDoesNotThrow;
import static org.junit.jupiter.api.Assertions.assertNotNull;
import static org.junit.jupiter.api.Assertions.fail;

public class TestFlatHashStrategy
{
private static final TypeOperators TYPE_OPERATORS = new TypeOperators();
private static final JoinCompiler JOIN_COMPILER = new JoinCompiler(TYPE_OPERATORS);
private final TypeOperators typeOperators = new TypeOperators();
private final JoinCompiler joinCompiler = new JoinCompiler(typeOperators);

@Test
public void testBatchedRawHashesZeroLength()
{
List<Type> types = createTestingTypes(typeOperators);
FlatHashStrategy flatHashStrategy = joinCompiler.getFlatHashStrategy(types);

int positionCount = 10;
// Attempting to touch any of the blocks would result in a NullPointerException
assertDoesNotThrow(() -> flatHashStrategy.hashBlocksBatched(new Block[types.size()], new long[positionCount], 0, 0));
}

@Test
public void testBatchedRawHashesMatchSinglePositionHashes()
{
List<Type> types = createTestingTypes();
FlatHashStrategy flatHashStrategy = JOIN_COMPILER.getFlatHashStrategy(types);
List<Type> types = createTestingTypes(typeOperators);
FlatHashStrategy flatHashStrategy = joinCompiler.getFlatHashStrategy(types);

int positionCount = 1024;
Block[] blocks = new Block[types.size()];
Expand All @@ -66,17 +79,30 @@ public void testBatchedRawHashesMatchSinglePositionHashes()

long[] hashes = new long[positionCount];
flatHashStrategy.hashBlocksBatched(blocks, hashes, 0, positionCount);
for (int position = 0; position < hashes.length; position++) {
long singleRowHash = flatHashStrategy.hash(blocks, position);
if (hashes[position] != singleRowHash) {
fail("Hash mismatch: %s <> %s at position %s - Values: %s".formatted(hashes[position], singleRowHash, position, singleRowTypesAndValues(types, blocks, position)));
}
assertHashesEqual(types, blocks, hashes, flatHashStrategy);

// Convert all blocks to RunLengthEncoded and re-check results match
for (int i = 0; i < blocks.length; i++) {
blocks[i] = RunLengthEncodedBlock.create(blocks[i].getSingleValueBlock(0), positionCount);
}
flatHashStrategy.hashBlocksBatched(blocks, hashes, 0, positionCount);
assertHashesEqual(types, blocks, hashes, flatHashStrategy);

// Ensure the formatting logic produces a real string and doesn't blow up since otherwise this code wouldn't be exercised
assertNotNull(singleRowTypesAndValues(types, blocks, 0));
}

private static List<Type> createTestingTypes()
private static void assertHashesEqual(List<Type> types, Block[] blocks, long[] batchedHashes, FlatHashStrategy flatHashStrategy)
{
for (int position = 0; position < batchedHashes.length; position++) {
long singleRowHash = flatHashStrategy.hash(blocks, position);
if (batchedHashes[position] != singleRowHash) {
fail("Hash mismatch: %s <> %s at position %s - Values: %s".formatted(batchedHashes[position], singleRowHash, position, singleRowTypesAndValues(types, blocks, position)));
}
}
}

private static List<Type> createTestingTypes(TypeOperators typeOperators)
{
List<Type> baseTypes = List.of(
BIGINT,
Expand All @@ -102,7 +128,7 @@ private static List<Type> createTestingTypes()
builder.add(RowType.anonymous(baseTypes));
for (Type baseType : baseTypes) {
builder.add(new ArrayType(baseType));
builder.add(new MapType(baseType, baseType, TYPE_OPERATORS));
builder.add(new MapType(baseType, baseType, typeOperators));
}
return builder.build();
}
Expand Down