mirror of
https://github.com/dart-lang/sdk
synced 2024-10-14 10:48:25 +00:00
b6aa2976dc
The benchmarks were using a range of antipatterns that did not do what the authors thought they did. It seems that the authors thought the warmup method has to run for a while and do the full warmup, but the truth is that the harness will do that for you by running the warmup function in a timed loop. Instead these patterns just wasted time by making the warmup more expensive and complex than needed. This change just removes the warmup overrides since none of them do anything positive. This change prepares us for future improvements to the benchmark harness. Fixes: b/324874055 Change-Id: Ib7c282123a2151614bc95a105a30e67221f11315 Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/352022 Reviewed-by: William Hesse <whesse@google.com> Commit-Queue: Jonas Termansen <sortie@google.com>
105 lines
2.8 KiB
Dart
105 lines
2.8 KiB
Dart
// Copyright (c) 2020, the Dart project authors. Please see the AUTHORS file
|
|
// for details. All rights reserved. Use of this source code is governed by a
|
|
// BSD-style license that can be found in the LICENSE file.
|
|
//
|
|
// Benchmark for UTF-8 encoding
|
|
|
|
// @dart=2.9
|
|
|
|
import 'dart:convert';
|
|
|
|
import 'package:benchmark_harness/benchmark_harness.dart';
|
|
|
|
import 'datext_latin1_10k.dart';
|
|
import 'entext_ascii_10k.dart';
|
|
import 'netext_3_10k.dart';
|
|
import 'rutext_2_10k.dart';
|
|
import 'sktext_10k.dart';
|
|
import 'zhtext_10k.dart';
|
|
|
|
class Utf8Encode extends BenchmarkBase {
|
|
final String language;
|
|
final String originalText;
|
|
// Size is measured in number of runes rather than number of bytes.
|
|
// This differs from the Utf8Decode benchmark, but runes are the input
|
|
// to the encode function which makes them more natural than bytes here.
|
|
final int size;
|
|
List<String> benchmarkTextChunks;
|
|
|
|
static String _makeName(String language, int size) {
|
|
String name = 'Utf8Encode.$language.';
|
|
name += size >= 1000000
|
|
? '${size ~/ 1000000}M'
|
|
: size >= 1000
|
|
? '${size ~/ 1000}k'
|
|
: '$size';
|
|
return name;
|
|
}
|
|
|
|
Utf8Encode(this.language, this.originalText, this.size)
|
|
: super(_makeName(language, size));
|
|
|
|
@override
|
|
void setup() {
|
|
final int nRunes = originalText.runes.toList().length;
|
|
final String repeatedText = originalText * (size / nRunes).ceil();
|
|
final List<int> runes = repeatedText.runes.toList();
|
|
final int nChunks = (size < nRunes) ? (nRunes / size).floor() : 1;
|
|
benchmarkTextChunks = List<String>.filled(nChunks, null);
|
|
for (int i = 0; i < nChunks; i++) {
|
|
final offset = i * size;
|
|
benchmarkTextChunks[i] =
|
|
String.fromCharCodes(runes.sublist(offset, offset + size));
|
|
}
|
|
}
|
|
|
|
@override
|
|
void run() {
|
|
for (int i = 0; i < benchmarkTextChunks.length; i++) {
|
|
final encoded = utf8.encode(benchmarkTextChunks[i]);
|
|
if (encoded.length < benchmarkTextChunks[i].length) {
|
|
throw 'There should be at least as many encoded bytes as runes';
|
|
}
|
|
}
|
|
}
|
|
|
|
@override
|
|
void exercise() {
|
|
// Only a single run per measurement.
|
|
run();
|
|
}
|
|
|
|
@override
|
|
double measure() {
|
|
// Report time per input rune.
|
|
return super.measure() / size / benchmarkTextChunks.length;
|
|
}
|
|
|
|
@override
|
|
void report() {
|
|
// Report time in nanoseconds.
|
|
final double score = measure() * 1000.0;
|
|
print('$name(RunTime): $score ns.');
|
|
}
|
|
}
|
|
|
|
void main(List<String> args) {
|
|
const texts = {
|
|
'en': en,
|
|
'da': da,
|
|
'sk': sk,
|
|
'ru': ru,
|
|
'ne': ne,
|
|
'zh': zh,
|
|
};
|
|
final benchmarks = [
|
|
for (int size in [10, 10000, 10000000])
|
|
for (String language in texts.keys)
|
|
() => Utf8Encode(language, texts[language], size)
|
|
];
|
|
|
|
for (var bm in benchmarks) {
|
|
bm().report();
|
|
}
|
|
}
|