gf2121 commented on a change in pull request #541: URL: https://github.com/apache/lucene/pull/541#discussion_r779797718
########## File path: lucene/core/src/java/org/apache/lucene/util/bkd/BKDForUtil.java ########## @@ -0,0 +1,126 @@ +// This file has been automatically generated, DO NOT EDIT +/* + * 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.lucene.util.bkd; + +import java.io.IOException; +import org.apache.lucene.store.DataInput; +import org.apache.lucene.store.DataOutput; + +// Inspired from https://fulmicoton.com/posts/bitpacking/ +// Encodes multiple integers in a long to get SIMD-like speedups. +// If bitsPerValue <= 8 then we pack 8 ints per long +// else if bitsPerValue <= 16 we pack 4 ints per long +// else we pack 2 ints per long +final class BKDForUtil { + + static final int BLOCK_SIZE = 512; + private static final long MASK32_8 = mask32(8); + private static final long MASK32_24 = mask32(24); + + private static long expandMask32(long mask32) { + return mask32 | (mask32 << 32); + } + + private static long mask32(int bitsPerValue) { + return expandMask32((1L << bitsPerValue) - 1); + } + + private static void expand16(long[] arr) { + for (int i = 0; i < 128; ++i) { + long l = arr[i]; + arr[i] = (l >>> 48) & 0xFFFFL; + arr[128 + i] = (l >>> 32) & 0xFFFFL; + arr[256 + i] = (l >>> 16) & 0xFFFFL; + arr[384 + i] = l & 0xFFFFL; + } + } + + private static void collapse16(long[] arr) { + for (int i = 0; i < 128; ++i) { + arr[i] = (arr[i] << 48) | (arr[128 + i] << 32) | (arr[256 + i] << 16) | arr[384 + i]; + } + } + + private static void expand32(long[] arr) { + for (int i = 0; i < 256; ++i) { + long l = arr[i]; + arr[i] = l >>> 32; + arr[256 + i] = l & 0xFFFFFFFFL; + } + } + + private static void collapse32(long[] arr) { + for (int i = 0; i < 256; ++i) { + arr[i] = (arr[i] << 32) | arr[256 + i]; + } + } + + private final long[] tmp = new long[256]; + + void encode16(long[] longs, DataOutput out) throws IOException { + collapse16(longs); + for (int i = 0; i < 128; i++) { + out.writeLong(longs[i]); + } + } + + void encode32(long[] longs, DataOutput out) throws IOException { + collapse32(longs); + for (int i = 0; i < 256; i++) { + out.writeLong(longs[i]); + } + } + + void encode24(long[] longs, DataOutput out) throws IOException { + collapse32(longs); + for (int i = 0; i < 192; ++i) { + tmp[i] = longs[i] << 8; + } + for (int i = 0; i < 64; i++) { Review comment: This encoding logic is a bit different from the `ForUtil` we used in postings, as i want to make the remainder decoding can also trigger SIMD. Here is the JMH result: ``` Benchmark (bitsPerValue) (byteOrder) Mode Cnt Score Error Units PackedIntsDecodeBenchmark.baseline 24 LE thrpt 5 7756530.597 ± 1654468.198 ops/s PackedIntsDecodeBenchmark.candidate 24 LE thrpt 5 9681438.494 ± 2528482.525 ops/s ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
