26 MotionMark performance tests failing
[WebKit-https.git] / PerformanceTests / ChangeLog
1 2017-01-13  Said Abou-Hallawa  <sabouhallawa@apple.com>
2
3         26 MotionMark performance tests failing
4         https://bugs.webkit.org/show_bug.cgi?id=166854
5
6         Reviewed by Ryosuke Niwa.
7         
8         Skip running MotionMark on WebKit perf bots.
9
10         * Skipped:
11
12 2017-01-06  Said Abou-Hallawa  <sabouhallawa@apple.com>
13
14         Rename the directory and the files of MotionMark from Animometer to MotionMark
15         https://bugs.webkit.org/show_bug.cgi?id=166659
16
17         Reviewed by Jon Lee.
18         
19         Rename the directory and the files of the benchmark to its new name.
20
21         * MotionMark/developer.html: Renamed from PerformanceTests/Animometer/developer.html.
22         * MotionMark/index.html: Renamed from PerformanceTests/Animometer/index.html.
23         * MotionMark/resources/debug-runner/d3.min.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/d3.min.js.
24         * MotionMark/resources/debug-runner/graph.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/graph.js.
25         * MotionMark/resources/debug-runner/motionmark.css: Renamed from PerformanceTests/Animometer/resources/debug-runner/animometer.css.
26         * MotionMark/resources/debug-runner/motionmark.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/animometer.js.
27         * MotionMark/resources/debug-runner/tests.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/tests.js.
28         * MotionMark/resources/extensions.js: Renamed from PerformanceTests/Animometer/resources/extensions.js.
29         * MotionMark/resources/runner/benchmark-runner.js: Renamed from PerformanceTests/Animometer/resources/runner/benchmark-runner.js.
30         * MotionMark/resources/runner/crystal.svg: Renamed from PerformanceTests/Animometer/resources/runner/crystal.svg.
31         * MotionMark/resources/runner/lines.svg: Renamed from PerformanceTests/Animometer/resources/runner/lines.svg.
32         * MotionMark/resources/runner/logo.svg: Renamed from PerformanceTests/Animometer/resources/runner/logo.svg.
33         * MotionMark/resources/runner/motionmark.css: Renamed from PerformanceTests/Animometer/resources/runner/animometer.css.
34         * MotionMark/resources/runner/motionmark.js: Renamed from PerformanceTests/Animometer/resources/runner/animometer.js.
35         * MotionMark/resources/runner/tests.js: Renamed from PerformanceTests/Animometer/resources/runner/tests.js.
36         * MotionMark/resources/statistics.js: Renamed from PerformanceTests/Animometer/resources/statistics.js.
37         * MotionMark/resources/strings.js: Renamed from PerformanceTests/Animometer/resources/strings.js.
38         * MotionMark/tests/3d/resources/webgl.js: Renamed from PerformanceTests/Animometer/tests/3d/resources/webgl.js.
39         * MotionMark/tests/3d/webgl.html: Renamed from PerformanceTests/Animometer/tests/3d/webgl.html.
40         * MotionMark/tests/bouncing-particles/bouncing-canvas-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-canvas-images.html.
41         * MotionMark/tests/bouncing-particles/bouncing-canvas-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html.
42         * MotionMark/tests/bouncing-particles/bouncing-css-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-css-images.html.
43         * MotionMark/tests/bouncing-particles/bouncing-css-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-css-shapes.html.
44         * MotionMark/tests/bouncing-particles/bouncing-svg-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-svg-images.html.
45         * MotionMark/tests/bouncing-particles/bouncing-svg-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-svg-shapes.html.
46         * MotionMark/tests/bouncing-particles/bouncing-tagged-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-tagged-images.html.
47         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-images.js.
48         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-particles.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-particles.js.
49         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-shapes.js.
50         * MotionMark/tests/bouncing-particles/resources/bouncing-css-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-css-images.js.
51         * MotionMark/tests/bouncing-particles/resources/bouncing-css-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-css-shapes.js.
52         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-images.js.
53         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-particles.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-particles.js.
54         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-shapes.js.
55         * MotionMark/tests/bouncing-particles/resources/bouncing-tagged-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js.
56         * MotionMark/tests/bouncing-particles/resources/image1.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image1.jpg.
57         * MotionMark/tests/bouncing-particles/resources/image2.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image2.jpg.
58         * MotionMark/tests/bouncing-particles/resources/image3.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image3.jpg.
59         * MotionMark/tests/bouncing-particles/resources/image4.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image4.jpg.
60         * MotionMark/tests/bouncing-particles/resources/image5.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image5.jpg.
61         * MotionMark/tests/dom/compositing-transforms.html: Renamed from PerformanceTests/Animometer/tests/dom/compositing-transforms.html.
62         * MotionMark/tests/dom/focus.html: Renamed from PerformanceTests/Animometer/tests/dom/focus.html.
63         * MotionMark/tests/dom/leaves.html: Renamed from PerformanceTests/Animometer/tests/dom/leaves.html.
64         * MotionMark/tests/dom/particles.html: Renamed from PerformanceTests/Animometer/tests/dom/particles.html.
65         * MotionMark/tests/dom/resources/compositing-transforms.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/compositing-transforms.js.
66         * MotionMark/tests/dom/resources/dom-particles.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/dom-particles.js.
67         * MotionMark/tests/dom/resources/focus.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/focus.js.
68         * MotionMark/tests/dom/resources/leaves.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/leaves.js.
69         * MotionMark/tests/master/canvas-stage.html: Renamed from PerformanceTests/Animometer/tests/master/canvas-stage.html.
70         * MotionMark/tests/master/focus.html: Renamed from PerformanceTests/Animometer/tests/master/focus.html.
71         * MotionMark/tests/master/image-data.html: Renamed from PerformanceTests/Animometer/tests/master/image-data.html.
72         * MotionMark/tests/master/leaves.html: Renamed from PerformanceTests/Animometer/tests/master/leaves.html.
73         * MotionMark/tests/master/multiply.html: Renamed from PerformanceTests/Animometer/tests/master/multiply.html.
74         * MotionMark/tests/master/resources/canvas-stage.js: Renamed from PerformanceTests/Animometer/tests/master/resources/canvas-stage.js.
75         * MotionMark/tests/master/resources/canvas-tests.js: Renamed from PerformanceTests/Animometer/tests/master/resources/canvas-tests.js.
76         * MotionMark/tests/master/resources/compass.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/compass.svg.
77         * MotionMark/tests/master/resources/compass100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/compass100.png.
78         * MotionMark/tests/master/resources/console.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/console.svg.
79         * MotionMark/tests/master/resources/console100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/console100.png.
80         * MotionMark/tests/master/resources/contribute.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/contribute.svg.
81         * MotionMark/tests/master/resources/contribute100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/contribute100.png.
82         * MotionMark/tests/master/resources/debugger.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/debugger.svg.
83         * MotionMark/tests/master/resources/debugger100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/debugger100.png.
84         * MotionMark/tests/master/resources/focus.js: Renamed from PerformanceTests/Animometer/tests/master/resources/focus.js.
85         * MotionMark/tests/master/resources/image-data.js: Renamed from PerformanceTests/Animometer/tests/master/resources/image-data.js.
86         * MotionMark/tests/master/resources/inspector.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/inspector.svg.
87         * MotionMark/tests/master/resources/inspector100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/inspector100.png.
88         * MotionMark/tests/master/resources/layout.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/layout.svg.
89         * MotionMark/tests/master/resources/layout100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/layout100.png.
90         * MotionMark/tests/master/resources/leaves.js: Renamed from PerformanceTests/Animometer/tests/master/resources/leaves.js.
91         * MotionMark/tests/master/resources/multiply.js: Renamed from PerformanceTests/Animometer/tests/master/resources/multiply.js.
92         * MotionMark/tests/master/resources/particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/particles.js.
93         * MotionMark/tests/master/resources/performance.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/performance.svg.
94         * MotionMark/tests/master/resources/performance100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/performance100.png.
95         * MotionMark/tests/master/resources/script.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/script.svg.
96         * MotionMark/tests/master/resources/script100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/script100.png.
97         * MotionMark/tests/master/resources/shortcuts.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/shortcuts.svg.
98         * MotionMark/tests/master/resources/shortcuts100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/shortcuts100.png.
99         * MotionMark/tests/master/resources/standards.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/standards.svg.
100         * MotionMark/tests/master/resources/standards100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/standards100.png.
101         * MotionMark/tests/master/resources/storage.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/storage.svg.
102         * MotionMark/tests/master/resources/storage100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/storage100.png.
103         * MotionMark/tests/master/resources/styles.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/styles.svg.
104         * MotionMark/tests/master/resources/styles100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/styles100.png.
105         * MotionMark/tests/master/resources/svg-particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/svg-particles.js.
106         * MotionMark/tests/master/resources/text.js: Renamed from PerformanceTests/Animometer/tests/master/resources/text.js.
107         * MotionMark/tests/master/resources/timeline.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/timeline.svg.
108         * MotionMark/tests/master/resources/timeline100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/timeline100.png.
109         * MotionMark/tests/master/svg-particles.html: Renamed from PerformanceTests/Animometer/tests/master/svg-particles.html.
110         * MotionMark/tests/master/text.html: Renamed from PerformanceTests/Animometer/tests/master/text.html.
111         * MotionMark/tests/resources/main.js: Renamed from PerformanceTests/Animometer/tests/resources/main.js.
112         * MotionMark/tests/resources/math.js: Renamed from PerformanceTests/Animometer/tests/resources/math.js.
113         * MotionMark/tests/resources/stage.css: Renamed from PerformanceTests/Animometer/tests/resources/stage.css.
114         * MotionMark/tests/resources/star.svg: Renamed from PerformanceTests/Animometer/tests/resources/star.svg.
115         * MotionMark/tests/resources/yin-yang.png: Renamed from PerformanceTests/Animometer/tests/resources/yin-yang.png.
116         * MotionMark/tests/resources/yin-yang.svg: Renamed from PerformanceTests/Animometer/tests/resources/yin-yang.svg.
117         * MotionMark/tests/simple/resources/simple-canvas-paths.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/simple-canvas-paths.js.
118         * MotionMark/tests/simple/resources/simple-canvas.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/simple-canvas.js.
119         * MotionMark/tests/simple/resources/tiled-canvas-image.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/tiled-canvas-image.js.
120         * MotionMark/tests/simple/simple-canvas-paths.html: Renamed from PerformanceTests/Animometer/tests/simple/simple-canvas-paths.html.
121         * MotionMark/tests/simple/tiled-canvas-image.html: Renamed from PerformanceTests/Animometer/tests/simple/tiled-canvas-image.html.
122         * MotionMark/tests/template/resources/template-canvas.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-canvas.js.
123         * MotionMark/tests/template/resources/template-css.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-css.js.
124         * MotionMark/tests/template/resources/template-svg.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-svg.js.
125         * MotionMark/tests/template/template-canvas.html: Renamed from PerformanceTests/Animometer/tests/template/template-canvas.html.
126         * MotionMark/tests/template/template-css.html: Renamed from PerformanceTests/Animometer/tests/template/template-css.html.
127         * MotionMark/tests/template/template-svg.html: Renamed from PerformanceTests/Animometer/tests/template/template-svg.html.
128
129 2016-12-13  Brady Eidson  <beidson@apple.com>
130
131         Add a PerfTest targeting IDBObjectStore.get().
132         https://bugs.webkit.org/show_bug.cgi?id=165816
133
134         Reviewed by Alex Christensen.
135
136         * IndexedDB/objectstore-get.html: Added.
137
138 2016-12-13  Brady Eidson  <beidson@apple.com>
139
140         Add a PerfTest targeting IDBIndex.get().
141         https://bugs.webkit.org/show_bug.cgi?id=165803
142
143         Reviewed by Alex Christensen.
144
145         * IndexedDB/index-get.html: Added.
146
147 2016-12-12  Brady Eidson  <beidson@apple.com>
148
149         More IndexedDB perf tests.
150         https://bugs.webkit.org/show_bug.cgi?id=165634
151
152         Reviewed by Sam Weinig.
153
154         * IndexedDB/index-multientry.html:
155         * IndexedDB/large-number-of-inserts-responsiveness.html: Added.
156         * IndexedDB/large-number-of-inserts.html: Added.
157         * IndexedDB/objectstore-cursor.html: Added. Runtime cut in 1/4th compared to r209672.
158
159         * resources/runner.js:
160
161 2016-12-12  Commit Queue  <commit-queue@webkit.org>
162
163         Unreviewed, rolling out r209672.
164         https://bugs.webkit.org/show_bug.cgi?id=165766
165
166         IndexedDB/objectstore-cursor.html perf test always times out
167         (Requested by ap on #webkit).
168
169         Reverted changeset:
170
171         "More IndexedDB perf tests."
172         https://bugs.webkit.org/show_bug.cgi?id=165634
173         http://trac.webkit.org/changeset/209672
174
175 2016-12-10  Brady Eidson  <beidson@apple.com>
176
177         More IndexedDB perf tests.
178         https://bugs.webkit.org/show_bug.cgi?id=165634
179
180         Reviewed by Sam Weinig.
181
182         * IndexedDB/index-multientry.html:
183         * IndexedDB/large-number-of-inserts-responsiveness.html: Added.
184         * IndexedDB/large-number-of-inserts.html: Added.
185         * IndexedDB/objectstore-cursor.html: Added.
186         
187         * resources/runner.js: Add "track responsiveness" functionality to PerfTestRunner.
188
189 2016-12-06  Filip Pizlo  <fpizlo@apple.com>
190
191         Concurrent GC should be stable enough to land enabled
192         https://bugs.webkit.org/show_bug.cgi?id=164990
193
194         Reviewed by Geoffrey Garen.
195         
196         Made CDjs more configurable and refined the "large.js" configuration. I was using that one and
197         the new "long.js" configuration to tune concurrent eden GCs.
198         
199         Added a new way of running Splay in browser, which using chartjs to plot the execution times of
200         2000 iterations. This includes the minified chartjs.
201
202         * JetStream/Octane2/splay-detail.html: Added.
203         * JetStream/cdjs/benchmark.js:
204         (benchmarkImpl):
205         (benchmark):
206         * JetStream/cdjs/long.js: Added.
207
208 2016-12-07  Brady Eidson  <beidson@apple.com>
209
210         Add IDB perf tests stressing key size.
211         https://bugs.webkit.org/show_bug.cgi?id=165567
212
213         Reviewed by Alex Christensen.
214
215         * IndexedDB/large-array-keys.html: Added.
216         * IndexedDB/large-binary-keys.html: Added.
217         * IndexedDB/large-string-keys.html: Added.
218
219 2016-12-05  Brady Eidson  <beidson@apple.com>
220
221         Add an IndexedDB perf test to PerformanceTests.
222         https://bugs.webkit.org/show_bug.cgi?id=165430
223
224         Reviewed by Alex Christensen.
225
226         * IndexedDB/index-multientry.html: Added.
227
228 2016-12-02  Filip Pizlo  <fpizlo@apple.com>
229
230         ES6SampleBench should report an average for Steady State so that all of the numbers are comparable
231         https://bugs.webkit.org/show_bug.cgi?id=165325
232
233         Reviewed by Saam Barati.
234
235         This makes all of the numbers that ES6SampleBench reports comparable to each other: they all speak of the time
236         it took to run an iteration of something.
237
238         * ES6SampleBench/results.js:
239         (Results.prototype.reportResult):
240
241 2016-11-18  Jer Noble  <jer.noble@apple.com>
242
243         Add some performance tests for various aspects of media loading and playback.
244         https://bugs.webkit.org/show_bug.cgi?id=164977
245
246         Reviewed by Darin Adler.
247
248         Add new tests for a few aspects of media playback; namely: how quickly  media elements fire
249         the "canplaythrough" event when all data is immediately available, how quickly HLS playback
250         switches resolutions, and at what playback rate MSE-backed video can play without dropping
251         frames. Skip these tests by default, as they're not runnable on all ports, and they require
252         a webserver (run-webkit-httpd) to complete on macOS.
253
254         * Media/HLSCanPlayThrough.html: Added.
255         * Media/HLSGearChange.html: Added.
256         * Media/MSECanPlayThrough.html: Added.
257         * Media/MSEPlaybackRate.html: Added.
258         * Media/StandardCanPlayThrough.html: Added.
259         * Media/hls/1080p/iframe_index.m3u8: Added.
260         * Media/hls/1080p/prog_index.m3u8: Added.
261         * Media/hls/1080p/test.ts: Added.
262         * Media/hls/480p/iframe_index.m3u8: Added.
263         * Media/hls/480p/prog_index.m3u8: Added.
264         * Media/hls/480p/test.ts: Added.
265         * Media/hls/720p/iframe_index.m3u8: Added.
266         * Media/hls/720p/prog_index.m3u8: Added.
267         * Media/hls/720p/test.ts: Added.
268         * Media/hls/720p/test.ts.back: Added.
269         * Media/hls/index.m3u8: Added.
270         * Media/media-source-loader.js: Added.
271         (MediaSourceLoader):
272         (MediaSourceLoader.prototype.loadManifest):
273         (MediaSourceLoader.prototype.loadManifestSucceeded):
274         (MediaSourceLoader.prototype.loadManifestFailed):
275         (MediaSourceLoader.prototype.loadMediaData):
276         (MediaSourceLoader.prototype.loadMediaDataSucceeded):
277         (MediaSourceLoader.prototype.loadMediaDataFailed):
278         (MediaSourceLoader.prototype.get type):
279         (MediaSourceLoader.prototype.get duration):
280         (MediaSourceLoader.prototype.get initSegment):
281         (MediaSourceLoader.prototype.get mediaSegmentsLength):
282         (MediaSourceLoader.prototype.mediaSegments):
283         (MediaSourceLoader.prototype.get everyMediaSegment):
284         * Media/test-fragmented-video.json: Added.
285         * Media/test-fragmented-video.mp4: Added.
286         * Media/test.mp4: Added.
287         * Skipped:
288
289 2016-11-18  Filip Pizlo  <fpizlo@apple.com>
290
291         Concurrent GC should be able to run splay in debug mode and earley/raytrace in release mode with no perf regression
292         https://bugs.webkit.org/show_bug.cgi?id=164282
293
294         Reviewed by Geoffrey Garen and Oliver Hunt.
295         
296         CDjs is a fun benchmark for stressing concurrent GCs, but to really give the GC a good
297         workout you need to increase the amount of work that the test does. This adds a second
298         configuration of the benchmark that has more aircraft. It uses much more memory and causes us
299         to do more GCs and those GCs take longer.
300
301         * JetStream/cdjs/benchmark.js:
302         (benchmarkImpl):
303         (benchmark):
304         * JetStream/cdjs/large.js: Added.
305
306 2016-11-14  Filip Pizlo  <fpizlo@apple.com>
307
308         Unreviewed, revert unintended change.
309
310         * ES6SampleBench/Air/benchmark.js:
311
312 2016-10-03  Saam Barati  <sbarati@apple.com>
313
314         MapHash should speculate on the type of its child node
315         https://bugs.webkit.org/show_bug.cgi?id=161922
316
317         Reviewed by Filip Pizlo.
318
319         I gave the main generator a name so it's easier to see what
320         it is when using the sampling profiler.
321
322         * ES6SampleBench/Basic/ast.js:
323         (Basic.Program):
324
325 2016-09-19  Sergio Villar Senin  <svillar@igalia.com>
326
327         [css-grid] Remove the x2 computation of row sizes with indefinite heights
328         https://bugs.webkit.org/show_bug.cgi?id=162150
329
330         Reviewed by Darin Adler.
331
332         Added a new test case which checks the layout performance of grids inside other grids, i.e,
333         grids acting both as grid container and grid item.
334
335         * Layout/nested-grid.html: Added.
336
337 2016-09-09  Simon Fraser  <simon.fraser@apple.com>
338
339         Perf test Animation/css-accelerated-animation.html failing
340         https://bugs.webkit.org/show_bug.cgi?id=161795
341
342         Skip Animation/css-accelerated-animation.html since this only works as a MobileSafari
343         perf test.
344
345         * Skipped:
346
347 2016-09-08  Simon Fraser  <simon.fraser@apple.com>
348
349         Add a content-animation test with accelerated CSS animations
350         https://bugs.webkit.org/show_bug.cgi?id=161776
351
352         Reviewed by Dean Jackson.
353
354         Add a test that measures the frame rate of accelerated CSS animations.
355
356         This is like css-animation.html (which animates 'left' and 'top') but animates
357         transforms instead. In order to get animation in X and Y we make two nested elements
358         and set transformX() on one, and transformY() on the other.
359
360         * Animation/css-accelerated-animation.html: Added.
361
362 2016-08-23  Saam Barati  <sbarati@apple.com>
363
364         It should be easy to run ES6SampleBench from the jsc shell
365         https://bugs.webkit.org/show_bug.cgi?id=161085
366
367         Reviewed by Yusuke Suzuki.
368
369         This patch makes ES6 sample bench runnable through the `jsc` shell.
370         To do that, you need to be in the PerformanceTests/ES6SampleBench
371         directory and run `jsc cli.js`. To make this work, the benchmark
372         is now aware of being run in two modes: via the browser, and via
373         the shell. Each entry point will set a variable `isInBrowser`,
374         and the benchmark will do different things based on if that
375         variable is true or false.
376
377         * ES6SampleBench/Air/benchmark.js:
378         * ES6SampleBench/Air/stress-test.js:
379         * ES6SampleBench/Basic/benchmark.js:
380         (runBenchmark):
381         * ES6SampleBench/Basic/stress-test.js:
382         * ES6SampleBench/air_benchmark.js:
383         * ES6SampleBench/basic_benchmark.js:
384         * ES6SampleBench/cli.js: Added.
385         (return.doRun):
386         (makeBenchmarkRunner):
387         * ES6SampleBench/driver.js:
388         (Driver):
389         (Driver.prototype._recomputeSummary):
390         (Driver.prototype._iterate.window.setTimeout):
391         (Driver.prototype._iterate):
392         (Driver.prototype._updateIterations):
393         * ES6SampleBench/glue.js:
394         (reportResult):
395         * ES6SampleBench/index.html:
396         * ES6SampleBench/results.js:
397         (Results):
398         (Results.prototype.reportRunning):
399         (Results.prototype.reportDone):
400         (Results.prototype.reportError):
401         * ES6SampleBench/stats.js:
402         (Stats):
403         (Stats.prototype.toString):
404         (Stats.prototype._update):
405
406 2016-08-01  Said Abou-Hallawa  <sabouhallawa@apple.com>
407
408         Add an option to run the MotionMark tests with classic tile size (512x512)
409         https://bugs.webkit.org/show_bug.cgi?id=160371
410
411         Reviewed by Darin Adler.
412
413         Setting the body to large size (3000x3000) and overflow to scroll forces
414         the classic tile size (512x512). This option does not affect the canvas
415         tests because the canvas is displayed on one tile regardless of its size.
416
417         * Animometer/developer.html:
418         * Animometer/resources/debug-runner/animometer.css:
419         (body.showing-test-container.tiles-big):
420         (body.showing-test-container.tiles-classic):
421         * Animometer/resources/debug-runner/animometer.js:
422         * Animometer/resources/runner/animometer.js:
423         (window.benchmarkController.startBenchmark):
424
425 2016-06-29  Filip Pizlo  <fpizlo@apple.com>
426
427         Generators violate bytecode liveness validation
428         https://bugs.webkit.org/show_bug.cgi?id=159279
429
430         Reviewed by Yusuke Suzuki.
431
432         Add Basic to our test harness.
433
434         Also made some cosmetic changes to the benchmark harness.
435
436         * ES6SampleBench/Basic/basic-tests.yaml: Added.
437         * ES6SampleBench/Basic/stress-test.js: Added.
438         (preciseTime):
439         * ES6SampleBench/driver.js:
440         (Driver):
441         (Driver.prototype.start):
442         (Driver.prototype.reportError):
443         * ES6SampleBench/glue.js:
444         * ES6SampleBench/index.html:
445
446 2016-06-28  Filip Pizlo  <fpizlo@apple.com>
447
448         ES6SampleBench should have a harness
449         https://bugs.webkit.org/show_bug.cgi?id=159246
450
451         Reviewed by Saam Barati.
452         
453         This adds a simple web harness for ES6SampleBench. It runs Air and Basic 10 times for 200
454         iterations each and reports three metrics:
455         
456         First iteration: the time it takes for the first iteration to run. This is the first
457         iteration after the benchmark is loaded into the iframe, so it's representative of what
458         would happen if one of these workloads only ran for a short time.
459         
460         Worst 2%: of the last 199 iterations, the average of the worst 2% iterations. If code like
461         any of these workloads was used in an important event handler, you'd want that code to run
462         well in the worst case in addition to having great throughput.
463         
464         Steady state: the total of the last 199 iterations. This is representative of what would
465         happen if you ran code like this for a long time.
466         
467         The total score is the geomean of the firstIteration/worstCase/steadyState numbers of the
468         two benchmarks.
469         
470         The harness does statistics using Student's T-distribution confidence intervals.
471
472         * ES6SampleBench/Basic/benchmark.js:
473         (Benchmark):
474         * ES6SampleBench/air_benchmark.js: Added.
475         * ES6SampleBench/basic_benchmark.js: Added.
476         * ES6SampleBench/driver.js: Added.
477         (Driver):
478         (Driver.prototype.addBenchmark):
479         (Driver.prototype.start):
480         (Driver.prototype.reportResult):
481         (Driver.prototype.reportError):
482         (Driver.prototype._recomputeSummary.Geomean):
483         (Driver.prototype._recomputeSummary.Geomean.prototype.add):
484         (Driver.prototype._recomputeSummary.Geomean.prototype.get result):
485         (Driver.prototype._recomputeSummary):
486         (Driver.prototype._iterate):
487         (Driver.prototype._updateIterations):
488         * ES6SampleBench/glue.js: Added.
489         * ES6SampleBench/index.html: Added.
490         * ES6SampleBench/results.js: Added.
491         (Results):
492         (Results.prototype.get benchmark):
493         (Results.prototype.reset):
494         (Results.prototype.reportRunning):
495         (Results.prototype.reportDone):
496         (Results.prototype.reportResult.averageAbovePercentile):
497         (Results.prototype.reportResult):
498         (Results.prototype.reportError):
499         * ES6SampleBench/stats.js: Added.
500         (Stats):
501         (Stats.prototype.reset):
502         (Stats.prototype.add):
503         (Stats.prototype.get numIterations):
504         (Stats.prototype.valueForIteration):
505         (Stats.get result.tDist):
506         (Stats.prototype.get result):
507         (Stats.prototype.toString):
508         (Stats.prototype._update):
509         * ES6SampleBench/style.css: Added.
510         (body):
511         (body, th, tr):
512         (h1, h2, h3, h4):
513         (h1):
514         (h2):
515         (h3):
516         (hr):
517         (address):
518         (img):
519         (.underline):
520         (ol.loweralpha):
521         (ol.upperalpha):
522         (ol.lowerroman):
523         (ol.upperroman):
524         (ol.arabic):
525         (.banner-link:link, .banner-link:visited):
526         (:link, :visited):
527         (h4 :link, h4 :visited, h5 :link, h5 :visited, h6 :link, h6 :visited):
528         (.anchor:link, .anchor:visited):
529         (* > .anchor:link, * > .anchor:visited):
530         (span:hover .anchor):
531         (a.forbidden, span.forbidden):
532         (a.missing:hover):
533         (a.closed:link, a.closed:visited, span.closed):
534         (pre):
535         (div.code):
536         (div.code pre):
537         (dt):
538         (dd):
539         (dd:last-child):
540         (.site-logo):
541         (.site-logo .tagline):
542         (table):
543         (#contents):
544         (p):
545         (p:last-child):
546         (th):
547         (td):
548
549 2016-06-28  Jon Lee  <jonlee@apple.com>
550
551         Update focus test
552         https://bugs.webkit.org/show_bug.cgi?id=159242
553         rdar://problem/27070007
554
555         Reviewed by Dean Jackson.
556         Provisionally reviewed by Said Abou-Hallawa.
557
558         Move previous test to dom suite, and update the test for better reporting of frame rate, although
559         it uses a different rendering path.
560
561         * Animometer/resources/debug-runner/tests.js: Add to dom suite.
562         * Animometer/tests/dom/focus.html: Copied from PerformanceTests/Animometer/tests/master/focus.html.
563         * Animometer/tests/dom/resources/focus.js: Copied from PerformanceTests/Animometer/tests/master/resources/focus.js.
564         * Animometer/tests/master/focus.html: Remove center element.
565         * Animometer/tests/master/resources/focus.js: Use narrower size range with smaller particles. Remove the
566         container elements. Inline getBlurValue and getOpacityValue since they are only called once.
567
568 2016-06-28  Filip Pizlo  <fpizlo@apple.com>
569
570         Move Air.js and Basic into ES6SampleBench
571
572         Rubber stamped by Geoffrey Garen.
573         
574         We want to group our ES6 benchmarks together, and eventually, we want to give them a
575         common harness.
576
577         * Air.js: Removed.
578         * Air.js/README.md: Removed.
579         * Air.js/airjs-tests.yaml: Removed.
580         * Air.js/all.js: Removed.
581         * Air.js/allocate_stack.js: Removed.
582         * Air.js/arg.js: Removed.
583         * Air.js/basic_block.js: Removed.
584         * Air.js/benchmark.js: Removed.
585         * Air.js/code.js: Removed.
586         * Air.js/custom.js: Removed.
587         * Air.js/frequented_block.js: Removed.
588         * Air.js/insertion_set.js: Removed.
589         * Air.js/inst.js: Removed.
590         * Air.js/liveness.js: Removed.
591         * Air.js/make_dist.sh: Removed.
592         * Air.js/opcode.js: Removed.
593         * Air.js/payload-airjs-ACLj8C.js: Removed.
594         * Air.js/payload-gbemu-executeIteration.js: Removed.
595         * Air.js/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
596         * Air.js/payload-typescript-scanIdentifier.js: Removed.
597         * Air.js/reg.js: Removed.
598         * Air.js/stack_slot.js: Removed.
599         * Air.js/stress-test.js: Removed.
600         * Air.js/strip-hash.rb: Removed.
601         * Air.js/symbols.js: Removed.
602         * Air.js/test.html: Removed.
603         * Air.js/test.js: Removed.
604         * Air.js/tmp.js: Removed.
605         * Air.js/tmp_base.js: Removed.
606         * Air.js/util.js: Removed.
607         * Basic: Removed.
608         * Basic/ast.js: Removed.
609         * Basic/basic.js: Removed.
610         * Basic/benchmark.js: Removed.
611         * Basic/caseless_map.js: Removed.
612         * Basic/lexer.js: Removed.
613         * Basic/number.js: Removed.
614         * Basic/parser.js: Removed.
615         * Basic/random.js: Removed.
616         * Basic/state.js: Removed.
617         * Basic/test.html: Removed.
618         * Basic/test.js: Removed.
619         * Basic/util.js: Removed.
620         * ES6SampleBench: Added.
621         * ES6SampleBench/Air: Copied from PerformanceTests/Air.js.
622         * ES6SampleBench/Basic: Copied from PerformanceTests/Basic.
623         * Skipped:
624
625 2016-06-24  Filip Pizlo  <fpizlo@apple.com>
626
627         Add a ES6 generator benchmark
628         https://bugs.webkit.org/show_bug.cgi?id=159101
629
630         Rubber stamped by Keith Miller.
631         
632         This adds a Basic interpreter loosely based on ECMA-55:
633         
634         http://www.ecma-international.org/publications/files/ECMA-ST-WITHDRAWN/ECMA-55,%201st%20Edition,%20January%201978.pdf
635         
636         It includes a lexer that is a generator, a parser that uses regular expressions, and an AST
637         walk interpreter where the walking functions for statements are generators that call each
638         other with yield*. This enables the interpreter to look like an AST walk even though it can
639         yield at INPUT and PRINT statements.
640         
641         This also uses for-of, classes, Map, and WeakMap. It also uses deprecated-but-awesome RegExp
642         features like RegExp.lastMatch and RegExp.rightContext. I did it that way because this is
643         how I've always written lexers in dynamic languages; see offlineasm's lex() method in
644         parser.rb for example.
645         
646         The benchmark runs a handful of simple Basic programs. The longest-running one computes
647         prime numbers.
648         
649         Includes a command-line and web harness. On my machine it runs in 2-3 seconds.
650
651         * Basic: Added.
652         * Basic/ast.js: Added.
653         (Basic.NumberApply):
654         (Basic.Variable):
655         (Basic.Const):
656         (Basic.NumberPow):
657         (Basic.NumberMul):
658         (Basic.NumberDiv):
659         (Basic.NumberNeg):
660         (Basic.NumberAdd):
661         (Basic.NumberSub):
662         (Basic.StringVar):
663         (Basic.Equals):
664         (Basic.NotEquals):
665         (Basic.LessThan):
666         (Basic.GreaterThan):
667         (Basic.LessEqual):
668         (Basic.GreaterEqual):
669         (Basic.GoTo):
670         (Basic.GoSub):
671         (Basic.Def):
672         (Basic.Let):
673         (Basic.If):
674         (Basic.Return):
675         (Basic.Stop):
676         (Basic.On):
677         (sideState.shouldStop):
678         (Basic.For):
679         (Basic.Next):
680         (Basic.Print):
681         (Basic.Input):
682         (Basic.Read):
683         (Basic.Restore):
684         (Basic.Dim):
685         (Basic.Randomize):
686         (Basic.End):
687         (Basic.Program):
688         * Basic/basic.js: Added.
689         (prepare):
690         (simulate):
691         * Basic/benchmark.js: Added.
692         (Benchmark):
693         (Benchmark.prototype.runIteration.expect):
694         (Benchmark.prototype.runIteration):
695         (runBenchmark):
696         * Basic/caseless_map.js: Added.
697         (CaselessMap):
698         * Basic/lexer.js: Added.
699         (lex.consumeWhitespace):
700         (lex.consume):
701         (lex):
702         * Basic/lexer_test.js: Added.
703         * Basic/number.js: Added.
704         (NumberValue):
705         (NumberValue.prototype.get value):
706         (NumberValue.prototype.apply):
707         (NumberValue.prototype.leftApply):
708         (NumberValue.prototype.assign):
709         (NumberArray.):
710         (NumberArray):
711         (NumberArray.prototype.apply):
712         (NumberArray.prototype.leftApply):
713         (NumberFunction):
714         (NumberFunction.prototype.apply):
715         (NumberFunction.prototype.leftApply):
716         (NativeFunction):
717         (NativeFunction.prototype.apply):
718         (NativeFunction.prototype.leftApply):
719         * Basic/parser.js: Added.
720         (parse):
721         (parse.pushToken):
722         (parse.peekToken):
723         (parse.consumeKind):
724         (parse.consumeToken):
725         (parse.parseVariable):
726         (parse.parseNumericExpression.parsePrimary):
727         (parse.parseNumericExpression.parseFactor):
728         (parse.parseNumericExpression.parseTerm):
729         (parse.parseNumericExpression):
730         (parse.parseConstant):
731         (parse.parseStringExpression):
732         (parse.isStringExpression):
733         (parse.parseRelationalExpression):
734         (parse.parseNonNegativeInteger):
735         (parse.parseGoToStatement):
736         (parse.parseGoSubStatement):
737         (parse.parseStatement):
738         (parse.parseStatements):
739         * Basic/random.js: Added.
740         (createRNG):
741         (createRNGWithFixedSeed):
742         (createRNGWithRandomSeed):
743         * Basic/state.js: Added.
744         (State):
745         (State.prototype.getValue):
746         (State.prototype.getSideState):
747         (State.prototype.abort):
748         (State.prototype.validate):
749         * Basic/test.html: Added.
750         * Basic/test.js: Added.
751         * Basic/util.js: Added.
752         (this.performance.performance.now.currentTime):
753         (else.this.preciseTime.currentTime):
754         (else.currentTime):
755         * Skipped: Make sure that we don't run Basic yet.
756
757 2016-06-21  Jon Lee  <jonlee@apple.com>
758
759         Update canvas size when benchmark begins
760         https://bugs.webkit.org/show_bug.cgi?id=159010
761
762         Reviewed by Dean Jackson.
763
764         Previously we would evaluate the media queries and assign the canvas size to the test
765         on body.onload. Instead, do it once the user starts the benchmark.
766
767         * Animometer/resources/debug-runner/animometer.js: Move benchmarkController.determineCanvasSize()
768         to benchmarkController._startBenchmark.
769         * Animometer/resources/runner/animometer.js: Ditto.
770
771 2016-06-21  Jon Lee  <jonlee@apple.com>
772
773         Add new timestamp option
774         https://bugs.webkit.org/show_bug.cgi?id=159006
775
776         Reviewed by Dean Jackson.
777
778         Add a new option to take timestamps from the rAF callback.
779
780         * Animometer/developer.html: Added option.
781         * Animometer/tests/resources/main.js:
782         (Benchmark): Fall back to using the rAF timestamp if performance.now()
783         is not available.
784         (Benchmark._animateLoop): Update how timestamp is set. Prefer to use
785         local var instead of accessing private var.
786
787 2016-06-21  Jon Lee  <jonlee@apple.com>
788
789         Improvements to Animometer benchmark
790         https://bugs.webkit.org/show_bug.cgi?id=157738
791
792         Reviewed by Dean Jackson.
793         Provisionally reviewed by Said Abou-Hallawa.
794
795         Update tests.
796
797         * Animometer/tests/master/text.html: Ensure only three text sizes for
798         the three canvases.
799         * Animometer/tests/master/focus.html: Reduce the text size to fit with smaller
800         particle sizes.
801         * Animometer/tests/master/resources/focus.js: Remove the quadratic distribution for
802         particle sizes, and make it linear. Reduce the size variance. Shuffle the math to
803         reduce some calculations per frame. Fix the placement of the particles which might
804         otherwise be culled.
805         * Animometer/tests/master/resources/image-data.js: Reduce the particle size
806         to encourage larger scores.
807
808 2016-06-21  Jon Lee  <jonlee@apple.com>
809
810         Improvements to Animometer benchmark
811         https://bugs.webkit.org/show_bug.cgi?id=157738
812
813         Reviewed by Dean Jackson.
814         Provisionally reviewed by Said Abou-Hallawa.
815
816         Include confidence interval for the final score, and store the canvas
817         size in the serialization so that it is accurately shown in results.
818
819         * Animometer/developer.html: Add a "confidence" div.
820         * Animometer/index.html: Ditto. Convert "mean" to "confidence".
821         * Animometer/resources/debug-runner/animometer.js: Look at options, and
822         if the configuration is included, update the body class based on it
823         (similar to what we do when we first load the page). That way, if you
824         drag-and-drop previous results in, that configuration is reflected in
825         the dashboard. Show the full confidence interval.
826
827         * Animometer/resources/debug-runner/animometer.css:
828         * Animometer/resources/debug-runner/animometer.js: Style update.
829         * Animometer/resources/runner/animometer.css:
830
831         * Animometer/resources/runner/animometer.js:
832         (_processData): Propagate the confidence interval values out and calculate
833         the lower and upper bounds. For now, shortcut the aggregate calculation,
834         since we only go through one iteration.
835         (this._processData.calculateScore): Propagate the confidence interval out
836         to be next to the score. Depending on the controller these values are
837         calculated differently.
838         (this._processData._getResultsProperty): Convenience function.
839         (this._processData.get score): Refactor.
840         (this._processData.get scoreLowerBound): Get the aggregate lower bound.
841         (this._processData.get scoreUpperBound): Get the aggregate upper bound.
842         (window.sectionsManager.setSectionScore):
843         (window.benchmarkController._startBenchmark): When the benchmark starts, note
844         the canvas size and add it to options. That way it will be automatically be
845         serialized.
846         (window.benchmarkController.showResults): Include the maximum deviation
847         percentage.
848         * Animometer/resources/runner/lines.svg: Make the background line up with the
849         skew.
850         * Animometer/resources/runner/tests.js:
851         (Headers.details.text): Refactor.
852         * Animometer/resources/statistics.js:
853         (Statistics.largestDeviationPercentage): Convenience function to calculate
854         the deviation percentage on either end and return the largest deviation.
855         * Animometer/resources/strings.js:
856
857         Allow specifying a regression profile to use instead of taking the one
858         with the lowest error.
859
860         Address an issue in the focus test when the regression calculated ends
861         up overestimating the change point, causing a cascade of tougher
862         ramps. The reason behind this is that at max complexity of an initial
863         ramp, the frame length is very high, and it influences the second
864         segment of the piecewise regression strongly, causing it to be very
865         steep. As a result, the first segment, expected to be flat, ends up
866         covering a higher range of complexity. That makes the change point
867         much higher than it should be. To avoid this, we will add a sanity
868         check on the maximum value of the ramp. If the regression's projected
869         value at the maximum complexity of the current ramp is very slow (less
870         than 20 fps), 1) reduce the maximum complexity by 20%, and 2) do not
871         include the regression's change point in the change point estimator.
872         That estimator is used as the midpoint of the next ramp, and including
873         the change point from a poor regression can bake in the error. The
874         controller already knows how to adjust for ramps that are too easy for
875         the system.
876
877         * Animometer/resources/runner/animometer.js:
878         (this._processData.findRegression): Takes a preferred profile and gives that to
879         Regression.
880         (this._processData.calculateScore): With the ramp controller, take the profile
881         of the ramp that was used the most when calculating the ramp's regression. That
882         profile is what is used for the test's score.
883         * Animometer/resources/statistics.js:
884         (Regression.Utilities.createClass): Update to take an options object which can
885         specify a profile to calculate with. Otherwise it will continue to use both and
886         select the one with the lower error.
887         (_calculateRegression): Fix an issue where we claim 0 error if the regression
888         calculation fails due to divide-by-zero. Instead reject that regression calculation
889         by giving it Number.MAX_VALUE.
890         * Animometer/resources/strings.js: New strings for marking a regression as flat
891         or slope.
892         * Animometer/tests/resources/main.js:
893         (RampController): Rename the thresholds for clarity. Add a threshold that, if
894         exceeded, will lower the maximum complexity of the next ramp.
895         (tune): Relax the cdf check to consider whether the interval definitely falls in
896         the desired frame length threshold.
897         (processSamples): Include the profile in the ramp.
898
899         Update ramp controller test. Increase the length of the test to 30 seconds, and extend
900         the interval to 120 ms during sampling. Improve the estimation of the ramp parameters.
901
902         * Animometer/developer.html: Change default to 30 seconds, and don't show the progress bar
903         by default.
904         * Animometer/resources/runner/animometer.js: Change default to 30 seconds.
905         * Animometer/tests/resources/main.js: A number of improvements to the ramp controller, in
906         the order in which they appear in the patch:
907
908         - With a longer test length use longer ramps with longer intervals to get more data at each
909         complexity. Keep the 100 ms interval length during the ramp up phase since we don't need to
910         spend more time there to find the right order of magnitude, but increase it during the
911         ramps to 120 ms.
912         - The ramp linearly interpolates the complexity to render based on its timestamp, but it would
913         never sample the minimum complexity. Instead of lerping max to min complexity from time
914         0 to t where t is the ramp length, instead lerp from 0 to (t - intervalSampleLength) so that
915         we can have at least one interval sample at the min complexity for that ramp.
916         - Some regression calculations only come out with one line segment rather than the two
917         we expect. This could be due to a noisy ramp or the ramp's range is too narrow. If that's the
918         case, influence the minimum complexity of the next ramp towards the lowest bound of 1, so that
919         we ensure that at least part of the ramp is covering a complexity range that the system can
920         handle at full 60.
921         - Remove an assignment to interpolatedFrameLength since that is never subsequently used.
922
923         Update the format used to serialize the results for analysis.
924
925         Each data point used to have named properties for fields like complexity and frame rate.
926         In addition the serialized numbers had rounding errors that took up many characters.
927         Update the format by introducing a new data container called SampleData, which contains a
928         field map. The map maps a string to an array index. Each data point is an array, so, to
929         get a stat, use the field map to get the array index into the data point. This allows future
930         versions to track other data, and reduces the size of the output string by two-thirds.
931
932         * Animometer/resources/extensions.js:
933         (Utilities.toFixedNumber): Add convenience function that truncates the number to a fixed
934         precision, and converts it back to a number.
935         (SampleData): New class that contains sample data and a field map that maps properties to
936         an array index.
937         (get length): Number of data points.
938         (addField): Add a field to the field map.
939         (push): Add a data point.
940         (sort): Sort the data.
941         (slice): Return new SampleData object with sliced data.
942         (forEach): Iterate over the data with provided function.
943         (createDatum):
944         (getFieldInDatum): Returns the data point associated with the field name by looking it up
945         in the field map in the datum provided, which can be the datum object itself (an array) or
946         an index into the data member variable.
947         (setFieldInDatum): Sets the data point associated with the field name.
948         (at): Returns the data point at the provided index.
949         (toArray): Serializes the data where the field map serves as property names for each point.
950
951         * Animometer/resources/debug-runner/graph.js:
952         (updateGraphData): Remove unused _testData. Convert the data to the old array format for the
953         graph to use, since the old format was much easier to work with when displaying the graphs.
954         (onGraphTypeChanged): For some controllers, no alternative score or mean is provided.
955         * Animometer/resources/runner/animometer.css:
956         * Animometer/resources/runner/animometer.js: Refactor to use SampleData. Update JSON output
957         to only go to 3 digits of precision for purposes of reducing the data size.
958         * Animometer/resources/strings.js: Add new strings to put into the field maps.
959         * Animometer/tests/resources/main.js: Refactor to use SampleData.
960
961         * Animometer/developer.html:
962         * Animometer/index.html: Restructure results table for both pages. Add charset attribute to
963         tests.js include.
964         * Animometer/resources/debug-runner/animometer.css: Clear out styles from release runner.
965         * Animometer/resources/debug-runner/graph.js:
966         (onGraphTypeChanged): Update score and mean if bootstrap results are available from the
967         controller, since not all controllers do bootstrapping.
968         * Animometer/resources/debug-runner/tests.js: Update header text.
969         * Animometer/resources/runner/animometer.css: Include confidence interval in results.
970         * Animometer/resources/runner/animometer.js:
971         (ResultsTable._addHeader): Header contents can be HTML, so use innerHTML instead.
972         (ResultsTable._addBody): Add tbody element.
973         (ResultsTable._addTest): Allow a data cell to invoke a JS function to get its contents.
974         (window.benchmarkController.showResults): Add table that includes tests' confidence intervals.
975         * Animometer/resources/runner/tests.js:
976         (Headers.details.text): Add new details table that includes bootstrap confidence interval.
977         The interval can be asymmetric, but for simplicity, report the maximum deviation percentage
978         on either side of the bootstrap median.
979         * Animometer/resources/statistics.js:
980         (bootstrap): Include the confidence percentage in the return object.
981
982         Report canvas size in results.
983
984         * Animometer/developer.html: Add markup to indicate whether a small, medium, or large
985         canvas was used.
986         * Animometer/index.html: Ditto.
987         * Animometer/resources/debug-runner/animometer.js: Call determineCanvasSize().
988         * Animometer/resources/runner/animometer.css: Update styles to set the canvas based on the
989         body class size.
990         * Animometer/resources/runner/animometer.js:
991         (window.benchmarkController.initialize): Update styles to set the canvas based on the
992         body class size.
993         (window.benchmarkController.determineCanvasSize): Run various media queries and set the body
994         class based on the size of the device.
995
996         * Animometer/developer.html: Refactor to include the main CSS file, and redo
997         the layout so that it doesn't rely on flexbox.
998         * Animometer/resources/debug-runner/animometer.css:
999         * Animometer/resources/debug-runner/animometer.js:
1000         (updateDisplay): Since various parts of the script alter the body class, we can't
1001         replace the className directly. Instead, remove all display-based values and then add
1002         the one that was selected.
1003         * Animometer/resources/debug-runner/graph.js:
1004         (updateGraphData): To set the size of the graph, use window.innerHeight.
1005         * Animometer/resources/runner/animometer.js:
1006         (window.sectionsManager.showSection): Since various parts of the script alter the body
1007         class, we can't replace the className directly. Remove all of the section classes
1008         individually and then add the one desired.
1009         * Animometer/tests/resources/stage.css: Remove -apple-system as a font to use in the
1010         stage.
1011
1012 2016-06-12  Filip Pizlo  <fpizlo@apple.com>
1013
1014         Fix round-down goof in Air.js's ShuffleCustom.forEachArg
1015         https://bugs.webkit.org/show_bug.cgi?id=158674
1016
1017         Reviewed by Michael Saboff.
1018         
1019         x / 3 * 3 is not how you round down to multiples of 3 in JavaScript. You need to do
1020         Math.floor(x / 3) * 3 instead.
1021         
1022         This is a benign change, because having extra arguments to a Shuffle is not something we
1023         actually take advantage of yet. But I think it's best to match the original C++ code's
1024         intent.
1025
1026         * Air.js/custom.js:
1027         (const.ShuffleCustom.forEachArg):
1028
1029 2016-06-11  Filip Pizlo  <fpizlo@apple.com>
1030
1031         Unreviewed, skip Air.js. It got unskipped when I renamed it from JSAir.
1032
1033         * Skipped:
1034
1035 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1036
1037         Make it easy to package Air.js
1038         https://bugs.webkit.org/show_bug.cgi?id=158652
1039
1040         Reviewed by Benjamin Poulain.
1041         
1042         If you want to give Air.js to someone, you can now do:
1043         
1044         cd PerformanceTests/Air.js
1045         ./make_dist.sh
1046         
1047         This will create Air.js.tar.gz and a directory called Air.js. If you want to send someone
1048         the benchmark, just send either of those things and tell them to look at index.html.
1049         
1050         You may have to edit make_dist.sh for wherever you have Markdown.pl. You can get that from
1051         https://daringfireball.net/projects/markdown/
1052
1053         * Air.js/make_dist.sh: Added.
1054         * Air.js/test.html:
1055
1056 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1057
1058         Unreviewed, make the link to test.html into an actual link.
1059
1060         * Air.js/README.md:
1061
1062 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1063
1064         Fix some text in Air.js/README.md
1065         https://bugs.webkit.org/show_bug.cgi?id=158650
1066
1067         Reviewed by Benjamin Poulain.
1068         
1069         I read the text again and found bugs:
1070         
1071         - We never actually say how to run the benchmark. This change adds a blurb about how to run
1072           it.
1073
1074         - We both say that allocateStack is responsible for the bulk of the running time and that
1075           we haven't measured where the bulk of the time is spent. This changes the text to say that
1076           it was a goal to make allocateStack be the hottest part of the benchmark, but that we did
1077           not measure this.
1078
1079         * Air.js/README.md:
1080
1081 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1082
1083         Air.js should have some documentation
1084         https://bugs.webkit.org/show_bug.cgi?id=158648
1085
1086         Reviewed by Keith Miller.
1087         
1088         I want to be able to point people at a document if they want to know more about this
1089         benchmark.
1090
1091         * Air.js/README.md: Added.
1092
1093 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1094
1095         Rename JSAir to Air.js.
1096
1097         Rubber stamped by Mark Lam.
1098
1099         * Air.js: Copied from PerformanceTests/JSAir.
1100         * Air.js/airjs-tests.yaml: Copied from PerformanceTests/JSAir/jsair-tests.yaml.
1101         * Air.js/benchmark.js: Copied from PerformanceTests/JSAir/benchmark.js.
1102         (Benchmark):
1103         * Air.js/jsair-tests.yaml: Removed.
1104         * Air.js/payload-airjs-ACLj8C.js: Copied from PerformanceTests/JSAir/payload-jsair-ACLj8C.js.
1105         * Air.js/payload-jsair-ACLj8C.js: Removed.
1106         * Air.js/stress-test.js: Copied from PerformanceTests/JSAir/stress-test.js.
1107         * Air.js/test.html: Copied from PerformanceTests/JSAir/test.html.
1108         * Air.js/test.js: Copied from PerformanceTests/JSAir/test.js.
1109         * JSAir: Removed.
1110         * JSAir/all.js: Removed.
1111         * JSAir/allocate_stack.js: Removed.
1112         * JSAir/arg.js: Removed.
1113         * JSAir/basic_block.js: Removed.
1114         * JSAir/benchmark.js: Removed.
1115         * JSAir/code.js: Removed.
1116         * JSAir/custom.js: Removed.
1117         * JSAir/frequented_block.js: Removed.
1118         * JSAir/insertion_set.js: Removed.
1119         * JSAir/inst.js: Removed.
1120         * JSAir/jsair-tests.yaml: Removed.
1121         * JSAir/liveness.js: Removed.
1122         * JSAir/opcode.js: Removed.
1123         * JSAir/payload-gbemu-executeIteration.js: Removed.
1124         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
1125         * JSAir/payload-jsair-ACLj8C.js: Removed.
1126         * JSAir/payload-typescript-scanIdentifier.js: Removed.
1127         * JSAir/reg.js: Removed.
1128         * JSAir/stack_slot.js: Removed.
1129         * JSAir/stress-test.js: Removed.
1130         * JSAir/strip-hash.rb: Removed.
1131         * JSAir/symbols.js: Removed.
1132         * JSAir/test.html: Removed.
1133         * JSAir/test.js: Removed.
1134         * JSAir/tmp.js: Removed.
1135         * JSAir/tmp_base.js: Removed.
1136         * JSAir/util.js: Removed.
1137
1138 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1139
1140         JSC Stress Test failing: jsair-tests.yaml/test.js.ftl-eager-no-cjit
1141         https://bugs.webkit.org/show_bug.cgi?id=158571
1142
1143         Reviewed by Keith Miller.
1144         
1145         Introduce a different harness for run-jsc-stress-tests, which runs for a minimum of 10
1146         iterations and then tries to do any number of "bonus" iterations until it's been running for
1147         two seconds. Since this is the sort of test that isn't really meant to stress anything in
1148         particular, I think it's OK if it is time-limited in this way. The worst case is that some
1149         of its failures will be flaky, but I think that they would have been flaky anyway given the
1150         complexity of the test.
1151
1152         * JSAir/benchmark.js:
1153         (Benchmark):
1154         (Benchmark.prototype.runIteration):
1155         (benchmark): Deleted.
1156         * JSAir/jsair-tests.yaml:
1157         * JSAir/stress-test.js: Added.
1158         (preciseTime):
1159         * JSAir/test.html:
1160         * JSAir/test.js:
1161
1162 2016-06-09  Filip Pizlo  <fpizlo@apple.com>
1163
1164         Unreviewed, teach the perf bots not to run JSAir.
1165
1166         * Skipped:
1167
1168 2016-06-08  Filip Pizlo  <fpizlo@apple.com>
1169
1170         Use more ES6 features in JSAir
1171         https://bugs.webkit.org/show_bug.cgi?id=158497
1172
1173         Reviewed by Keith Miller.
1174         
1175         This improves JSAir with the following ES6 features suggested by JoePeck:
1176         
1177         - String interpolation.
1178         - Destructuring inside PatchCustom.
1179         - Default arguments.
1180         
1181         All of these things are on hot paths.
1182         
1183         Note that I didn't use string interpolation everywhere that I could, only in those places
1184         where it made the code more readable. In Ruby, I used the style that if the interpolation
1185         expression has any non-trivial stuff (like a ternary operator, a chain of calls, or embedded
1186         strings) then it's better to use regular strcat. I think that's what I carried over to here.
1187         
1188         Note that the previous change (Add result validation to JSAir) also made the Proxy code not
1189         dead, though it's not necessarily on the hot path. The Proxy isn't called into frequently
1190         but it's used from a function that is otherwise hot, so if calling into the Proxy prevents
1191         that function from being optimized then it will hurt so good.
1192         
1193         I also reenabled tail calls in a few places.
1194         
1195         This change doesn't seem to change the performance of the benchmark for us. That's expected
1196         since these ES6 features are cheap. Note that this claim doesn't include Proxy, which was
1197         added in a separate change and that change did make the benchmark overall more expensive.
1198
1199         * JSAir/allocate_stack.js:
1200         (allocateStack):
1201         * JSAir/arg.js:
1202         (Arg.createBitImm64):
1203         (Arg.createAddr):
1204         (Arg.createStack):
1205         (Arg.logScale):
1206         (Arg.createIndex):
1207         * JSAir/basic_block.js:
1208         (BasicBlock.get headerString):
1209         (BasicBlock.prototype.get if):
1210         (BasicBlock):
1211         * JSAir/benchmark.js:
1212         (benchmark):
1213         * JSAir/code.js:
1214         (Code):
1215         (Code.prototype.addBlock):
1216         (Code.prototype.addStackSlot):
1217         (Code.prototype.newTmp):
1218         (Code.prototype.get size):
1219         (Code.prototype.get blocks):
1220         (Code.prototype.get stackSlots):
1221         (Code.prototype.tmps):
1222         (Code.prototype.get callArgAreaSize):
1223         (Code.prototype.toString):
1224         * JSAir/custom.js:
1225         (const.PatchCustom.forEachArg):
1226         * JSAir/inst.js:
1227         (Inst):
1228         * JSAir/reg.js:
1229         (Reg.prototype.toString):
1230         * JSAir/util.js:
1231         (symbolName):
1232         (lowerSymbolName):
1233
1234 2016-06-07  Filip Pizlo  <fpizlo@apple.com>
1235
1236         Add result validation to JSAir
1237         https://bugs.webkit.org/show_bug.cgi?id=158493
1238
1239         Reviewed by Saam Barati.
1240         
1241         This adds the ability to hash a Code in a way that matches the C++ code's hashing of Code.
1242         This allows us to check if the Code that JSAir sees is the code that C++ saw. We use this to
1243         check the Code before and after allocateStack, and compare against hashes we got from C++.
1244         
1245         Doing this uncovered bugs. roundUpToMultipleOf wasn't doing anything. allocateStack was not
1246         allocating things correctly because I was concatting a Set to an Array, which doesn't really
1247         work. Now these bugs are fixed.
1248         
1249         The checking step adds to the running time so I reduced the number of iterations. The
1250         benchmark spends a decent amount of its time computing Code hashes; I think it's around 1/3
1251         total. This is probably OK. It's better to verify the results even if the running time is
1252         not all in the "core" of the algorithm.
1253         
1254         Also add a run-jsc-stress-tests yaml file to allow this to run as a test.
1255
1256         * JSAir/allocate_stack.js:
1257         * JSAir/arg.js:
1258         (Arg.createImm):
1259         (Arg.createBigImm):
1260         (Arg.createBitImm):
1261         (Arg.createBitImm64):
1262         (Arg.createWidth):
1263         (Arg.createSpecial):
1264         (Arg.prototype.get kind):
1265         (Arg.prototype.get isTmp):
1266         (Arg.prototype.get isImm):
1267         (Arg.prototype.get isSomeImm):
1268         (Arg.prototype.get isSomeBigImm):
1269         (Arg.prototype.get isCondition):
1270         (Arg.prototype.get isWidth):
1271         (Arg.prototype.get isSpecial):
1272         (Arg.prototype.get isAlive):
1273         (Arg.prototype.get tmp):
1274         (Arg.prototype.get value):
1275         (Arg.prototype.get lowValue):
1276         (Arg.prototype.get highValue):
1277         (Arg.prototype.get base):
1278         (Arg.prototype.get isGP):
1279         (Arg.prototype.get isFP):
1280         (Arg.prototype.isValidForm):
1281         (Arg.prototype.get isInvertible):
1282         (Arg.kindCode):
1283         (Arg.prototype.hash):
1284         (Arg.prototype.toString):
1285         (Arg):
1286         * JSAir/basic_block.js:
1287         (BasicBlock.get successorBlocks):
1288         * JSAir/benchmark.js:
1289         (benchmark):
1290         * JSAir/code.js:
1291         (Code.prototype.setFrameSize):
1292         (Code.prototype.hash):
1293         (Code.prototype.toString):
1294         (Code):
1295         * JSAir/inst.js:
1296         (Inst.prototype.get hasNonArgEffects):
1297         (Inst.prototype.hash):
1298         (Inst.prototype.toString):
1299         (Inst):
1300         * JSAir/jsair-tests.yaml: Added.
1301         * JSAir/opcode.js:
1302         (Inst_forEachArg):
1303         (Inst_hasNonArgEffects):
1304         (opcodeCode):
1305         * JSAir/payload-gbemu-executeIteration.js:
1306         (createPayloadGbemuExecuteIteration):
1307         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js:
1308         (createPayloadImagingGaussianBlurGaussianBlur):
1309         * JSAir/payload-jsair-ACLj8C.js:
1310         (createPayloadJSAirACLj8C):
1311         * JSAir/payload-typescript-scanIdentifier.js:
1312         (createPayloadTypescriptScanIdentifier):
1313         * JSAir/reg.js:
1314         (Reg.prototype.get isReg):
1315         (Reg.prototype.hash):
1316         (Reg.prototype.toString):
1317         * JSAir/stack_slot.js:
1318         (StackSlot.prototype.setOffsetFromFP):
1319         (StackSlot.prototype.hash):
1320         (StackSlot.prototype.toString):
1321         * JSAir/symbols.js:
1322         (relCondCode):
1323         (resCondCode):
1324         (doubleCondCode):
1325         * JSAir/test.html:
1326         * JSAir/tmp.js:
1327         (Tmp.prototype.get isReg):
1328         (Tmp.prototype.hash):
1329         (Tmp.prototype.toString):
1330         * JSAir/util.js:
1331         (roundUpToMultipleOf):
1332         (symbolName):
1333
1334 2016-06-07  Filip Pizlo  <fpizlo@apple.com>
1335
1336         Implement Air::allocateStack() in ES6 to see how much of a bad idea that is
1337         https://bugs.webkit.org/show_bug.cgi?id=158318
1338
1339         Reviewed by Saam Barati.
1340         
1341         This adds a new benchmark for us to play with called JSAir. It's a complete ES6
1342         implementation of Air's allocateStack() phase along with all of Air needed to run that
1343         phase. This includes things like stack slots, registers, temporaries, basic blocks,
1344         instructions, and all of the code for iterating over, inspecting, and modifying those
1345         things.
1346         
1347         To make this work, JSC can now dump Air just before allocateStack() in the form of JS code
1348         that creates a Code object that matches exactly what C++ Air saw. This benchmark comprises
1349         four Air IRs:
1350         
1351         - Octane/gbemu's largest function, executeIteration.
1352         - Kraken/imaging-gaussian-blur's largest function in OSR entry mode, gaussuanBlur.
1353         - Octane/typescript's largest function that is actually hot, scanIdentifier.
1354         - JSAir's largest hot function, which is anonymous, so we call it by its hash (ACLj8C).
1355         
1356         This runs in about 2 seconds on my machine in JSC trunk. It includes both a commandline
1357         harness and a web harness. JSAir is almost exactly 100x slower in ES6 in WebKit than the
1358         C++ Air::allocateStack() phase on which it is based.
1359         
1360         JSAir uses the following ES6 features:
1361         
1362         - Symbol.
1363         - for-of.
1364         - arrow functions.
1365         - Map/Set.
1366         - let/const.
1367         - classes.
1368         
1369         All of these things are used in anger and should end up on the hot path. There is also code
1370         that uses Proxies, but it ends up being dead.
1371         
1372         We can improve this even more:
1373         I still need to add result validation: https://bugs.webkit.org/show_bug.cgi?id=158493
1374         I want to make it use more ES6 features: https://bugs.webkit.org/show_bug.cgi?id=158497
1375         
1376         * JSAir: Added.
1377         * JSAir/all.js: Added.
1378         * JSAir/allocate_stack.js: Added.
1379         (allocateStack.attemptAssignment):
1380         (allocateStack.assign):
1381         (allocateStack.interfere):
1382         (allocateStack.):
1383         (allocateStack):
1384         * JSAir/arg.js: Added.
1385         (Arg):
1386         (Arg.isAnyUse):
1387         (Arg.isColdUse):
1388         (Arg.isWarmUse):
1389         (Arg.cooled):
1390         (Arg.isEarlyUse):
1391         (Arg.isLateUse):
1392         (Arg.isAnyDef):
1393         (Arg.isEarlyDef):
1394         (Arg.isLateDef):
1395         (Arg.isZDef):
1396         (Arg.typeForB3Type):
1397         (Arg.widthForB3Type):
1398         (Arg.conservativeWidth):
1399         (Arg.minimumWidth):
1400         (Arg.bytes):
1401         (Arg.widthForBytes):
1402         (Arg.createTmp):
1403         (Arg.fromReg):
1404         (Arg.createImm):
1405         (Arg.createBigImm):
1406         (Arg.createBitImm):
1407         (Arg.createBitImm64):
1408         (Arg.createAddr):
1409         (Arg.createStack):
1410         (Arg.createCallArg):
1411         (Arg.createStackAddr):
1412         (Arg.isValidScale):
1413         (Arg.logScale):
1414         (Arg.createIndex):
1415         (Arg.createRelCond):
1416         (Arg.createResCond):
1417         (Arg.createDoubleCond):
1418         (Arg.createWidth):
1419         (Arg.prototype.get kind):
1420         (Arg.prototype.get isTmp):
1421         (Arg.prototype.get isImm):
1422         (Arg.prototype.get isBigImm):
1423         (Arg.prototype.get isBitImm):
1424         (Arg.prototype.get isBitImm64):
1425         (Arg.prototype.get isSomeImm):
1426         (Arg.prototype.get isAddr):
1427         (Arg.prototype.get isStack):
1428         (Arg.prototype.get isCallArg):
1429         (Arg.prototype.get isIndex):
1430         (Arg.prototype.get isMemory):
1431         (Arg.prototype.get isStackMemory):
1432         (Arg.prototype.get isRelCond):
1433         (Arg.prototype.get isResCond):
1434         (Arg.prototype.get isDoubleCond):
1435         (Arg.prototype.get isCondition):
1436         (Arg.prototype.get isWidth):
1437         (Arg.prototype.get isAlive):
1438         (Arg.prototype.get tmp):
1439         (Arg.prototype.get value):
1440         (Arg.prototype.get base):
1441         (Arg.prototype.get hasOffset):
1442         (Arg.prototype.get offset):
1443         (Arg.prototype.get stackSlot):
1444         (Arg.prototype.get index):
1445         (Arg.prototype.get scale):
1446         (Arg.prototype.get logScale):
1447         (Arg.prototype.get width):
1448         (Arg.prototype.get isGPTmp):
1449         (Arg.prototype.get isFPTmp):
1450         (Arg.prototype.get isGP):
1451         (Arg.prototype.get isFP):
1452         (Arg.prototype.get hasType):
1453         (Arg.prototype.get type):
1454         (Arg.prototype.isType):
1455         (Arg.prototype.isCompatibleType):
1456         (Arg.prototype.get isGPR):
1457         (Arg.prototype.get gpr):
1458         (Arg.prototype.get isFPR):
1459         (Arg.prototype.get fpr):
1460         (Arg.prototype.get isReg):
1461         (Arg.prototype.get reg):
1462         (Arg.isValidImmForm):
1463         (Arg.isValidBitImmForm):
1464         (Arg.isValidBitImm64Form):
1465         (Arg.isValidAddrForm):
1466         (Arg.isValidIndexForm):
1467         (Arg.prototype.isValidForm):
1468         (Arg.prototype.forEachTmpFast):
1469         (Arg.prototype.usesTmp):
1470         (Arg.prototype.forEachTmp):
1471         (Arg.prototype.is):
1472         (Arg.prototype.as):
1473         (Arg.prototype.forEachFast):
1474         (Arg.prototype.forEach):
1475         (Arg.extract):
1476         (Arg.forEachFast):
1477         (Arg.forEach):
1478         (Arg.prototype.get condition):
1479         (Arg.prototype.get isInvertible):
1480         (Arg.prototype.toString):
1481         * JSAir/basic_block.js: Added.
1482         (BasicBlock):
1483         (BasicBlock.prototype.get index):
1484         (BasicBlock.prototype.get size):
1485         (BasicBlock.prototype.Symbol.iterator):
1486         (BasicBlock.prototype.at):
1487         (BasicBlock.get last):
1488         (BasicBlock.get insts):
1489         (BasicBlock.get numSuccessors):
1490         (BasicBlock.get successors):
1491         (BasicBlock.get successorBlocks.):
1492         (BasicBlock.get successorBlocks):
1493         (BasicBlock.set get numPredecessors):
1494         (BasicBlock.get predecessors):
1495         (BasicBlock.get frequency):
1496         (BasicBlock.get headerString):
1497         (BasicBlock.get footerString):
1498         * JSAir/benchmark.js: Added.
1499         (benchmark):
1500         * JSAir/code.js: Added.
1501         (Code):
1502         (Code.prototype.addBlock):
1503         (Code.prototype.addStackSlot):
1504         (Code.prototype.newTmp):
1505         (Code.prototype.get size):
1506         (Code.prototype.at):
1507         (Code.prototype.Symbol.iterator):
1508         (Code.prototype.get blocks):
1509         (Code.prototype.get stackSlots):
1510         (Code.prototype.tmps):
1511         (Code.prototype.get callArgAreaSize):
1512         (Code.prototype.requestCallArgAreaSize):
1513         (Code.prototype.get frameSize):
1514         (Code.prototype.setFrameSize):
1515         (Code.prototype.toString):
1516         * JSAir/custom.js: Added.
1517         (const.ShuffleCustom.forEachArg):
1518         (const.ShuffleCustom.hasNonArgNonControlEffects):
1519         (const.PatchCustom.forEachArg):
1520         (const.PatchCustom.hasNonArgNonControlEffects):
1521         (const.CCallCustom.forEachArg):
1522         (const.CCallCustom.hasNonArgNonControlEffects):
1523         (const.ColdCCallCustom.forEachArg):
1524         (const.ColdCCallCustom.hasNonArgNonControlEffects):
1525         * JSAir/frequented_block.js: Added.
1526         (FrequentedBlock):
1527         (FrequentedBlock.prototype.toString):
1528         * JSAir/insertion_set.js: Added.
1529         (Insertion):
1530         (Insertion.prototype.get index):
1531         (Insertion.prototype.get element):
1532         (Insertion.prototype.lessThan):
1533         (InsertionSet):
1534         (InsertionSet.prototype.appendInsertion):
1535         (InsertionSet.prototype.append):
1536         (InsertionSet.prototype.execute):
1537         * JSAir/inst.js: Added.
1538         (Inst):
1539         (Inst.prototype.append):
1540         (Inst.prototype.clear):
1541         (Inst.prototype.get opcode):
1542         (Inst.prototype.get args):
1543         (Inst.prototype.visitArg):
1544         (Inst.prototype.forEachTmpFast):
1545         (Inst.prototype.forEachArg):
1546         (Inst.prototype.forEachTmp):
1547         (Inst.prototype.forEach):
1548         (Inst.forEachDef):
1549         (Inst.forEachDefWithExtraClobberedRegs):
1550         (Inst.prototype.get hasNonArgEffects):
1551         (Inst.prototype.toString):
1552         * JSAir/liveness.js: Added.
1553         (Liveness):
1554         (Liveness.prototype.get thing):
1555         (Liveness.prototype.get code):
1556         (Liveness.prototype.get liveAtHead):
1557         (Liveness.prototype.get liveAtTail):
1558         (Liveness.prototype.localCalc.LocalCalc):
1559         (Liveness.prototype.localCalc.LocalCalc.prototype.get liveSet):
1560         (Liveness.prototype.localCalc.LocalCalc.prototype.execute):
1561         (Liveness.prototype.localCalc):
1562         * JSAir/opcode.js: Added.
1563         (Inst_forEachArg):
1564         (Inst_hasNonArgEffects):
1565         * JSAir/payload-gbemu-executeIteration.js: Added.
1566         (createPayloadGbemuExecuteIteration):
1567         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js: Added.
1568         (createPayloadImagingGaussianBlurGaussianBlur):
1569         * JSAir/payload-jsair-ACLj8C.js: Added.
1570         (createPayloadJSAirACLj8C):
1571         * JSAir/payload-typescript-scanIdentifier.js: Added.
1572         (createPayloadTypescriptScanIdentifier):
1573         * JSAir/reg.js: Added.
1574         (Reg):
1575         (Reg.fromReg):
1576         (Reg.prototype.get index):
1577         (Reg.prototype.get type):
1578         (Reg.prototype.get name):
1579         (Reg.prototype.get isCalleeSave):
1580         (Reg.prototype.get isReg):
1581         (Reg.prototype.toString):
1582         (Reg.extract):
1583         (Reg.forEachFast):
1584         (Reg.forEach):
1585         (newGPR):
1586         (Reg.gprs.Reg.fprs.Reg.calleeSaveGPRs.Reg.calleeSaveFPRs.Reg.calleeSaves):
1587         * JSAir/stack_slot.js: Added.
1588         (StackSlot):
1589         (StackSlot.prototype.get byteSize):
1590         (StackSlot.prototype.get kind):
1591         (StackSlot.prototype.get isLocked):
1592         (StackSlot.prototype.get isSpill):
1593         (StackSlot.prototype.get index):
1594         (StackSlot.prototype.ensureSize):
1595         (StackSlot.prototype.get alignment):
1596         (StackSlot.prototype.get offsetFromFP):
1597         (StackSlot.prototype.setOffsetFromFP):
1598         (StackSlot.prototype.toString):
1599         (StackSlot.extract):
1600         (StackSlot.forEachFast):
1601         (StackSlot.forEach):
1602         * JSAir/symbols.js: Added.
1603         * JSAir/test.html: Added.
1604         * JSAir/test.js: Added.
1605         * JSAir/tmp.js: Added.
1606         (Tmp):
1607         (Tmp.fromReg):
1608         (Tmp.prototype.get index):
1609         (Tmp.prototype.get type):
1610         (Tmp.prototype.get isReg):
1611         (Tmp.prototype.toString):
1612         (Tmp.extract):
1613         (Tmp.forEachFast):
1614         (Tmp.forEach):
1615         * JSAir/tmp_base.js: Added.
1616         (TmpBase.prototype.get isGP):
1617         (TmpBase.prototype.get isFP):
1618         (TmpBase.prototype.get isGPR):
1619         (TmpBase.prototype.get isFPR):
1620         (TmpBase.prototype.get reg):
1621         (TmpBase.prototype.get gpr):
1622         (TmpBase.prototype.get fpr):
1623         (TmpBase):
1624         * JSAir/util.js: Added.
1625         (isRepresentableAsInt32):
1626         (addIndexed):
1627         (roundUpToMultipleOf):
1628         (symbolName):
1629         (mergeIntoSet):
1630         (nonEmptyRangesOverlap):
1631         (rangesOverlap):
1632         (removeAllMatching):
1633         (swap):
1634         (bubble):
1635         (bubbleSort):
1636         (currentTime):
1637         (else.currentTime):
1638
1639 2016-05-26  Yusuke Suzuki  <utatane.tea@gmail.com>
1640
1641         [JSC] Allow JSBench to use precise time
1642         https://bugs.webkit.org/show_bug.cgi?id=158050
1643
1644         Reviewed by Geoffrey Garen.
1645
1646         * JSBench/amazon-chrome-win/urem.js:
1647         (else.window.performance.window.performance.now.currentTimeInMS):
1648         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1649         (else.else.currentTimeInMS):
1650         (onload.cb):
1651         (onload):
1652         * JSBench/amazon-chrome/urem.js:
1653         (else.window.performance.window.performance.now.currentTimeInMS):
1654         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1655         (else.else.currentTimeInMS):
1656         (onload.cb):
1657         (onload):
1658         * JSBench/amazon-firefox-win/urm.js:
1659         (else.window.performance.window.performance.now.currentTimeInMS):
1660         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1661         (else.else.currentTimeInMS):
1662         (onload.cb):
1663         (onload):
1664         * JSBench/amazon-firefox/urm.js:
1665         (else.window.performance.window.performance.now.currentTimeInMS):
1666         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1667         (else.else.currentTimeInMS):
1668         (onload.cb):
1669         (onload):
1670         * JSBench/amazon-safari/urem.js:
1671         (else.window.performance.window.performance.now.currentTimeInMS):
1672         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1673         (else.else.currentTimeInMS):
1674         (onload.cb):
1675         (onload):
1676         * JSBench/facebook-chrome-win/urem.js:
1677         (else.window.performance.window.performance.now.currentTimeInMS):
1678         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1679         (else.else.currentTimeInMS):
1680         (onload.cb):
1681         (onload):
1682         * JSBench/facebook-chrome/urem.js:
1683         (else.window.performance.window.performance.now.currentTimeInMS):
1684         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1685         (else.else.currentTimeInMS):
1686         (onload.cb):
1687         (onload):
1688         * JSBench/facebook-firefox-win/urem.js:
1689         (else.window.performance.window.performance.now.currentTimeInMS):
1690         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1691         (else.else.currentTimeInMS):
1692         (onload.cb):
1693         (onload):
1694         * JSBench/facebook-firefox/urem.js:
1695         (else.window.performance.window.performance.now.currentTimeInMS):
1696         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1697         (else.else.currentTimeInMS):
1698         (onload.cb):
1699         (onload):
1700         * JSBench/facebook-safari/urem.js:
1701         (else.window.performance.window.performance.now.currentTimeInMS):
1702         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1703         (else.else.currentTimeInMS):
1704         (onload.cb):
1705         (onload):
1706         * JSBench/google-chrome-win/urem.js:
1707         (else.window.performance.window.performance.now.currentTimeInMS):
1708         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1709         (else.else.currentTimeInMS):
1710         (onload.cb):
1711         (onload):
1712         * JSBench/google-chrome/urem.js:
1713         (else.window.performance.window.performance.now.currentTimeInMS):
1714         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1715         (else.else.currentTimeInMS):
1716         (onload.cb):
1717         (onload):
1718         * JSBench/google-firefox-win/urem.js:
1719         (else.window.performance.window.performance.now.currentTimeInMS):
1720         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1721         (else.else.currentTimeInMS):
1722         (onload.cb):
1723         (onload):
1724         * JSBench/google-firefox/uem.js:
1725         (else.window.performance.window.performance.now.currentTimeInMS):
1726         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1727         (else.else.currentTimeInMS):
1728         (onload.cb):
1729         (onload):
1730         * JSBench/google-safari/urem.js:
1731         (else.window.performance.window.performance.now.currentTimeInMS):
1732         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1733         (else.else.currentTimeInMS):
1734         (onload.cb):
1735         (onload):
1736         * JSBench/harness.js:
1737         (runBenchmark.window.currentTimeInMS):
1738         (runBenchmark.else.window.currentTimeInMS):
1739         * JSBench/twitter-chrome-win/rem.js:
1740         (else.window.performance.window.performance.now.currentTimeInMS):
1741         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1742         (else.else.currentTimeInMS):
1743         (onload.cb):
1744         (onload):
1745         * JSBench/twitter-chrome/urem.js:
1746         (else.window.performance.window.performance.now.currentTimeInMS):
1747         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1748         (else.else.currentTimeInMS):
1749         (onload.cb):
1750         (onload):
1751         * JSBench/twitter-firefox-win/urem.js:
1752         (else.window.performance.window.performance.now.currentTimeInMS):
1753         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1754         (else.else.currentTimeInMS):
1755         (onload.cb):
1756         (onload):
1757         * JSBench/twitter-firefox/urem.js:
1758         (else.window.performance.window.performance.now.currentTimeInMS):
1759         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1760         (else.else.currentTimeInMS):
1761         (onload.cb):
1762         (onload):
1763         * JSBench/twitter-safari/urem.js:
1764         (else.window.performance.window.performance.now.currentTimeInMS):
1765         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1766         (else.else.currentTimeInMS):
1767         (onload.cb):
1768         (onload):
1769         * JSBench/yahoo-chrome-win/urem.js:
1770         (else.window.performance.window.performance.now.currentTimeInMS):
1771         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1772         (else.else.currentTimeInMS):
1773         (onload.cb):
1774         (onload):
1775         * JSBench/yahoo-chrome/urem.js:
1776         (else.window.performance.window.performance.now.currentTimeInMS):
1777         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1778         (else.else.currentTimeInMS):
1779         (onload.cb):
1780         (onload):
1781         * JSBench/yahoo-firefox-win/urem.js:
1782         (else.window.performance.window.performance.now.currentTimeInMS):
1783         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1784         (else.else.currentTimeInMS):
1785         (onload.cb):
1786         (onload):
1787         * JSBench/yahoo-firefox/urem.js:
1788         (else.window.performance.window.performance.now.currentTimeInMS):
1789         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1790         (else.else.currentTimeInMS):
1791         (onload.cb):
1792         (onload):
1793         * JSBench/yahoo-safari/urem.js:
1794         (else.window.performance.window.performance.now.currentTimeInMS):
1795         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1796         (else.else.currentTimeInMS):
1797         (onload.cb):
1798         (onload):
1799
1800 2016-05-25  Geoffrey Garen  <ggaren@apple.com>
1801
1802         replaceable own properties seem to ignore replacement after property caching
1803         https://bugs.webkit.org/show_bug.cgi?id=158091
1804
1805         Reviewed by Darin Adler.
1806
1807         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
1808         * MallocBench/MallocBench/Benchmark.cpp:
1809         * MallocBench/MallocBench/Interpreter.cpp:
1810         (Interpreter::doMallocOp):
1811         * MallocBench/MallocBench/Interpreter.h:
1812         * MallocBench/MallocBench/fastMallocLog.63316.ops: Added.
1813         * MallocBench/MallocBench/jetstream.cpp: Added.
1814         (benchmark_jetstream):
1815         * MallocBench/MallocBench/jetstream.h: Added.
1816
1817 2016-05-25  Keith Miller  <keith_miller@apple.com>
1818
1819         Unreviewed, add JSBench to the skipped list for now since it doesn't
1820         work currently.
1821
1822         * Skipped:
1823
1824 2016-05-17  Keith Miller  <keith_miller@apple.com>
1825         We should have JSBench in PerformanceTests
1826         https://bugs.webkit.org/show_bug.cgi?id=157952
1827
1828         Rubber-stamped by Saam Barati.
1829
1830         There are some slight changes to the layout of the test directory
1831         to make it work nicely with run-jsc-benchmarks. Before JSBench
1832         had each of the browser specific sub-tests in a sub-directory.
1833         These have been flattened e.g. amazon/safari/ has become
1834         amazon-safari/.
1835
1836         * JSBench/amazon-chrome-win/urem.html: Added.
1837         * JSBench/amazon-chrome-win/urem.js: Added.
1838         * JSBench/amazon-chrome/urem.html: Added.
1839         * JSBench/amazon-chrome/urem.js: Added.
1840         * JSBench/amazon-firefox-win/urm.html: Added.
1841         * JSBench/amazon-firefox-win/urm.js: Added.
1842         * JSBench/amazon-firefox/urm.html: Added.
1843         * JSBench/amazon-firefox/urm.js: Added.
1844         * JSBench/amazon-safari/urem.html: Added.
1845         * JSBench/amazon-safari/urem.js: Added.
1846         * JSBench/browsercheck.js: Added.
1847         * JSBench/facebook-chrome-win/urem.html: Added.
1848         * JSBench/facebook-chrome-win/urem.js: Added.
1849         * JSBench/facebook-chrome/urem.html: Added.
1850         * JSBench/facebook-chrome/urem.js: Added.
1851         * JSBench/facebook-firefox-win/urem.html: Added.
1852         * JSBench/facebook-firefox-win/urem.js: Added.
1853         * JSBench/facebook-firefox/urem.html: Added.
1854         * JSBench/facebook-firefox/urem.js: Added.
1855         * JSBench/facebook-safari/urem.html: Added.
1856         * JSBench/facebook-safari/urem.js: Added.
1857         * JSBench/google-chrome-win/urem.html: Added.
1858         * JSBench/google-chrome-win/urem.js: Added.
1859         * JSBench/google-chrome/urem.html: Added.
1860         * JSBench/google-chrome/urem.js: Added.
1861         * JSBench/google-firefox-win/urem.html: Added.
1862         * JSBench/google-firefox-win/urem.js: Added.
1863         * JSBench/google-firefox/uem.html: Added.
1864         * JSBench/google-firefox/uem.js: Added.
1865         * JSBench/google-safari/urem.html: Added.
1866         * JSBench/google-safari/urem.js: Added.
1867         * JSBench/harness.html: Added.
1868         * JSBench/harness.js: Added.
1869         * JSBench/harness.py: Added.
1870         * JSBench/index.html: Added.
1871         * JSBench/reload.html: Added.
1872         * JSBench/twitter-chrome-win/rem.html: Added.
1873         * JSBench/twitter-chrome-win/rem.js: Added.
1874         * JSBench/twitter-chrome/urem.html: Added.
1875         * JSBench/twitter-chrome/urem.js: Added.
1876         * JSBench/twitter-firefox-win/urem.html: Added.
1877         * JSBench/twitter-firefox-win/urem.js: Added.
1878         * JSBench/twitter-firefox/urem.html: Added.
1879         * JSBench/twitter-firefox/urem.js: Added.
1880         * JSBench/twitter-safari/urem.html: Added.
1881         * JSBench/twitter-safari/urem.js: Added.
1882         * JSBench/yahoo-chrome-win/urem.html: Added.
1883         * JSBench/yahoo-chrome-win/urem.js: Added.
1884         * JSBench/yahoo-chrome/urem.html: Added.
1885         * JSBench/yahoo-chrome/urem.js: Added.
1886         * JSBench/yahoo-firefox-win/urem.html: Added.
1887         * JSBench/yahoo-firefox-win/urem.js: Added.
1888         * JSBench/yahoo-firefox/urem.html: Added.
1889         * JSBench/yahoo-firefox/urem.js: Added.
1890         * JSBench/yahoo-safari/urem.html: Added.
1891         * JSBench/yahoo-safari/urem.js: Added.
1892
1893 2016-05-20  Ryosuke Niwa  <rniwa@webkit.org>
1894
1895         run-benchmark's results should contain Animometer's debug output
1896         https://bugs.webkit.org/show_bug.cgi?id=157941
1897
1898         Reviewed by Stephanie Lewis.
1899
1900         Made developer.html support the JSON generated by run-benchmark which stores everything under debugOutput.
1901
1902         * Animometer/resources/debug-runner/animometer.js:
1903         (Utilities.initialize): Unwrap debugOutput in the case run-benchmark's result JSON is used.
1904
1905 2016-05-18  Timothy Hatcher  <timothy@apple.com>
1906
1907         Make Animometer work in all browsers
1908         https://bugs.webkit.org/show_bug.cgi?id=157855
1909         rdar://problem/26338521
1910
1911         Reviewed by Jon Lee.
1912
1913         * Animometer/resources/runner/logo.svg: Use a symbol instead
1914         of putting the id on the root svg. This works in all browsers.
1915
1916 2016-05-07  Timothy Hatcher  <timothy@apple.com>
1917
1918         Update Animometer design
1919         https://bugs.webkit.org/show_bug.cgi?id=157449
1920
1921         Reviewed by Darin Adler.
1922
1923         * Animometer/developer.html: Stop including runner/animometer.css.
1924         * Animometer/index.html: Udpated elements to fit new styles.
1925
1926         * Animometer/resources/debug-runner/animometer.css:
1927         Copy the old runner/animometer.css to here so the debug runner is unaffected.
1928
1929         * Animometer/resources/runner/animometer.css:
1930         New styles for the design. Drop flex box for broad browser support.
1931
1932         * Animometer/resources/runner/animometer.js:
1933         (ResultsTable.clear): Use textContent.
1934         (window.sectionsManager.showSection): Add class to the body.
1935         (window.sectionsManager.setSectionScore): Use textContent, not innerHTML.
1936
1937         * Animometer/resources/runner/crystal.svg: Added.
1938         * Animometer/resources/runner/lines.svg: Added.
1939         * Animometer/resources/runner/logo.svg: Added.
1940
1941         * Animometer/resources/runner/tests.js: Shorten test name and title case them.
1942
1943 2016-05-06  Manuel Rego Casasnovas  <rego@igalia.com>
1944
1945         [css-grid] Unprefix CSS Grid Layout properties
1946         https://bugs.webkit.org/show_bug.cgi?id=157137
1947
1948         Reviewed by Simon Fraser.
1949
1950         Remove "-webkit" prefix from all the grid layout properties,
1951         including the display value.
1952         Update the source code to remove the prefix where it was used too.
1953
1954         * Layout/auto-grid-lots-of-data.html:
1955         * Layout/fixed-grid-lots-of-data.html:
1956         * Layout/fixed-grid-lots-of-stretched-data.html:
1957
1958 2016-04-25  Geoffrey Garen  <ggaren@apple.com>
1959
1960         bmalloc: Misc improvements to MallocBench
1961         https://bugs.webkit.org/show_bug.cgi?id=157004
1962
1963         Reviewed by Darin Adler.
1964
1965         * MallocBench/run-malloc-benchmarks: Added --memory and --memory_warning
1966         modes for focused memory testing.
1967
1968         * MallocBench/MallocBench/Benchmark.cpp:
1969         (Benchmark::printReport): Clarified output.
1970
1971         (Benchmark::currentMemoryBytes): Added compressed memory because top
1972         does the same. (It always happens to zero in the benchmarks we run. But
1973         this is good for sanity.)
1974
1975         * MallocBench/MallocBench/CommandLine.cpp: Moved up to 8 runs to reduce
1976         variance.
1977
1978         * MallocBench/MallocBench/alloc_free.cpp:
1979         (benchmark_alloc_free): Cycle a single allocation in order to stress
1980         the effect of merging on calls to madvise.
1981
1982         * MallocBench/MallocBench/big.cpp:
1983         (benchmark_big): Graduated to 8kB-128kB because medium tests up to 8 and
1984         our large allocator doesn't kick in until 64kB.
1985
1986         * MallocBench/MallocBench/medium.cpp:
1987         (benchmark_medium): Test all the way down to 1kB because our large
1988         allocator used to service 1kB allocations and 1kB is an interesting
1989         middle size where memory is unusually large but allocation throughput
1990         still matters.
1991
1992         * MallocBench/MallocBench/stress.cpp:
1993         (benchmark_stress): Reduced the churn count to match stress_aligned
1994         because this test was taking too long to complete.
1995
1996         * MallocBench/MallocBench/stress_aligned.cpp:
1997         (benchmark_stress_aligned): Our new large allocator can handle even
1998         more absurdly large values.
1999
2000 2016-04-25  Simon Fraser  <simon.fraser@apple.com>
2001
2002         Add a content animation test that uses SVG animation.
2003         https://bugs.webkit.org/show_bug.cgi?id=156827
2004
2005         * Animation/svg-animation.html: Added.
2006         * Skipped:
2007
2008 2016-04-22  Simon Fraser  <simon.fraser@apple.com>
2009
2010         Skip two content animation tests which are only meant for iOS testing.
2011
2012         * Animation/css-animation.html: Added.
2013         * Animation/raf-animation.html: Added.
2014
2015         * Skipped:
2016
2017 2016-04-20  Simon Fraser  <simon.fraser@apple.com>
2018
2019         Add content animation tests to benchmark_runner, and allow the runner to collect device data as part of the results
2020         https://bugs.webkit.org/show_bug.cgi?id=156827
2021
2022         Add two files missing from the previous commit.
2023
2024         * Animation/css-animation.html: Added.
2025         * Animation/raf-animation.html: Added.
2026
2027 2016-04-20  Simon Fraser  <simon.fraser@apple.com>
2028
2029         Add content animation tests to benchmark_runner, and allow the runner to collect device data as part of the results
2030         https://bugs.webkit.org/show_bug.cgi?id=156827
2031
2032         Reviewed by Stephanie Lewis.
2033
2034         Two new files for the content-animation suite.
2035
2036         After patching for use in the benchmark, these tests are a little different to other benchmarks.
2037         The test content itself does not collect performance data; the tests spit out a boilerplate JSON
2038         result with a placeholder for framerate, which is collected in native code.
2039
2040         When run as a benchmark, test completion involves a couple of bounces between the test and MobileSafari.
2041         Test completion is initiated by a setTimeout() in the test, at which point location.hash is set to "#done".
2042         MobileSafari detects that, and asynchronously dumps performance data. Once this is complete, MobileSafari
2043         changes the URL has to "#submit", which triggers the XHR to the test relay that indicates the test is
2044         complete.
2045
2046         * Animation/css-animation.html: Added.
2047         * Animation/raf-animation.html: Added.
2048
2049 2016-04-15  Jon Lee  <jonlee@apple.com>
2050
2051         Animometer test could report a NaN
2052         https://bugs.webkit.org/show_bug.cgi?id=156646
2053
2054         Reviewed by Darin Adler.
2055         Provisionally reviewed by Said Abou-Hallawa.
2056
2057         * Animometer/tests/resources/main.js:
2058         (didFinishInterval): The ramp controller has a first phase where it ramps up the complexity and
2059         reacts based on how well the system handles the load. The assumption was that it would handle at least
2060         1 particle easily. That is not always the case. As a result, an interpolation calculation could
2061         end up setting an upper bound of NaN.
2062
2063         This occurs because we never get out of the first tier, so this._lastTierComplexity is undefined.
2064         Now that we guarantee a minimum complexity of 1, modify the conditional to check for this._lastTierComplexity
2065         before interpolating the upper bound of the first ramp. In the case where the system struggles with
2066         1 particle, set it to currentComplexity.
2067
2068 2016-04-09  Jon Lee  <jonlee@apple.com>
2069
2070         Update Animometer to accommodate different screens
2071         https://bugs.webkit.org/show_bug.cgi?id=156449
2072
2073         Reviewed by Darin Adler.
2074         Provisionally reviewed by Said Abou-Hallawa.
2075
2076         * Animometer/index.html: Wrap button in a container to add padding at the bottom.
2077         * Animometer/resources/debug-runner/animometer.css:
2078         (@media screen and (min-device-width: 1800px)): Deleted.
2079         (@media screen and (min-width: 1800px)): Cannot use min-device-width since it may match incorrectly.
2080         (screen and (max-device-height: 414px) and (orientation: landscape)): Some devices swap device width
2081         and height with orientation change.
2082         * Animometer/resources/runner/animometer.css: Similar.
2083         (screen and (min-device-width: 1024px) and (orientation: landscape)):
2084         (screen and (max-device-height: 414px) and (orientation: landscape)):
2085         (.frame-container): On smaller iPhones, adding 1px prevents the navigation bars from appearing.
2086         (@media screen and (min-device-width: 768px) and (max-device-width: 1024px)): Deleted.
2087         (@media (min-device-height: 768px) and (max-device-height: 1024px)): Target iPad Airs and similar.
2088         (@media screen and (min-device-width: 1024px) and (max-device-width: 1366px)): Deleted.
2089         (@media screen and (max-device-width: 1024px) and (min-device-height: 1366px)): Target iPad Pro.
2090         (#results footer): Add padding below the button for testing again.
2091         * Animometer/tests/master/multiply.html: Remove the center text.
2092         * Animometer/tests/master/resources/text.js: Update the test so that in every frame the text moves.
2093         * Animometer/tests/master/text.html: Update the text sizing depending on the size of the device.
2094
2095 2016-04-08  Jon Lee  <jonlee@apple.com>
2096
2097         Have Animometer benchmark always start with complexity of 1
2098         https://bugs.webkit.org/show_bug.cgi?id=156432
2099
2100         Reviewed by Ryosuke Niwa.
2101
2102         * Animometer/tests/resources/main.js: Update the default Controller and RampController to
2103         set its minimum complexities to 1 instead of 0.
2104
2105 2016-04-08  Jon Lee  <jonlee@apple.com>
2106
2107         Fix SVG benchmark test
2108         https://bugs.webkit.org/show_bug.cgi?id=156410
2109
2110         Reviewed by Dean Jackson.
2111
2112         * Animometer/resources/extensions.js: Update Point.zero to be a static Point.
2113         * Animometer/tests/simple/resources/tiled-canvas-image.js:
2114         (Stage.call._setupTiles): Refactor.
2115         * Animometer/tests/master/resources/particles.js:
2116         (Particle.prototype.reset): Use Point.center.
2117         (complexity): We are not using a gradient background anymore, so remove the +1.
2118         * Animometer/tests/master/resources/svg-particles.js: Update to use SVG transform
2119         instead of CSS transform.
2120
2121 2016-04-06  Jon Lee  <jonlee@apple.com>
2122
2123         Update master benchmark with SVG test
2124         https://bugs.webkit.org/show_bug.cgi?id=156273
2125
2126         Reviewed by Dean Jackson.
2127         Provisionally reviewed by Said Abou-Hallawa.
2128
2129         Switch masks tests for SVG path test.
2130
2131         * Animometer/resources/debug-runner/tests.js: Move mask test here.
2132         * Animometer/resources/runner/tests.js: Add SVG test here.
2133         * Animometer/tests/dom/particles.html: Renamed from PerformanceTests/Animometer/tests/master/particles.html.
2134         * Animometer/tests/dom/resources/dom-particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/dom-particles.js.
2135
2136         * Animometer/tests/master/resources/particles.js: Add minPosition for bounds checking. Prevents particle from being
2137         partially obscured.
2138         * Animometer/tests/master/resources/svg-particles.js: Added.
2139         (Particle): The particle is either a path object or a rect using a path as a clip. The
2140         same path is used either way. For each particle create a linear gradient with a random
2141         rotation.
2142         (SVGParticlesStage): Look in #shapes to see how many different kinds of paths are available.
2143         This makes the test more generic in case other shapes need to be tested.
2144         * Animometer/tests/master/svg-particles.html: Added. Have two defs, one that houses each
2145         particle's gradient, and one that holds the shape templates.
2146
2147 2016-03-27  Hunseop Jeong  <hs85.jeong@samsung.com>
2148
2149         [EFL] REGRESSION(r188793): It made 200 layout tests and Bindings/event-target-wrapper.html performance test fail
2150         https://bugs.webkit.org/show_bug.cgi?id=148470 
2151
2152         Reviewed by Darin Adler.
2153
2154         * Skipped: Unskip the Bindings/event-target-wrapper test.
2155
2156 2016-03-23  Jon Lee  <jonlee@apple.com>
2157
2158         Update focus and leaves tests
2159         https://bugs.webkit.org/show_bug.cgi?id=155825
2160
2161         Reviewed by Simon Fraser.
2162
2163         Particles would improperly disappear, before they were fully off stage. Update the calculation.
2164         * Animometer/tests/dom/resources/leaves.js:
2165         * Animometer/tests/master/resources/leaves.js:
2166
2167         * Animometer/tests/master/focus.html: Update the center element styling.
2168         * Animometer/tests/master/resources/focus.js:
2169         (FocusElement.Utilities.createClass): Give enough space in the container for each particle for the blur
2170         to spread. It should be about 3x more space than the blur radius on each side.
2171         (Stage.call.initialize): Ditto for the center element.
2172
2173 2016-03-23  Jon Lee  <jonlee@apple.com>
2174
2175         Update support for other platforms
2176         https://bugs.webkit.org/show_bug.cgi?id=155824
2177
2178         Reviewed by Simon Fraser.
2179
2180         * Animometer/resources/debug-runner/animometer.css: Remove extraneous media query. All of it should
2181         be handled in the release stylesheet.
2182         * Animometer/resources/debug-runner/animometer.js: Add support for browsers that return an HTMLCollection
2183         when retrieving radio elements with a name, instead of the selected value.
2184         * Animometer/resources/runner/animometer.css: Adjust the queries for devices of varying widths.
2185         (.frame-container): Move overflow:hidden to #stage.
2186         * Animometer/tests/master/resources/stage.css: Consolidate into tests/resources/stage.css.
2187         * Animometer/tests/resources/stage.css:
2188         (body): Remove overflow:hidden.
2189         (#stage): Set overflow:hidden here.
2190         (#center-text): Moved from master/resources/stage.css.
2191
2192         Update stylesheet hrefs.
2193         * Animometer/tests/master/canvas-stage.html:
2194         * Animometer/tests/master/focus.html:
2195         * Animometer/tests/master/leaves.html:
2196         * Animometer/tests/master/multiply.html:
2197         * Animometer/tests/master/particles.html:
2198         * Animometer/tests/master/text.html: Adjust text size to fit in the device screen.
2199
2200 2016-03-23  Jon Lee  <jonlee@apple.com>
2201
2202         Make the benchmark require device to be in landscape orientation
2203         https://bugs.webkit.org/show_bug.cgi?id=155822
2204         rdar://problem/25258650
2205
2206         Reviewed by Ryosuke Niwa.
2207
2208         Check that the device is in landscape orientation. If not, disable the button to start it,
2209         and show a warning message.
2210
2211         * Animometer/developer.html: Add a message about how the browser should be set up. Include a
2212         warning message that appears if the orientation is incorrect on mobile devices.
2213         * Animometer/index.html: Ditto.
2214
2215         * Animometer/resources/debug-runner/animometer.css: Migrate the .hidden rule into the release
2216         stylesheet. Update the style to accommodate the new UI.
2217         * Animometer/resources/runner/animometer.css:
2218
2219         * Animometer/resources/runner/animometer.js:
2220         (window.benchmarkController.initialize): Add the orientation listener is needed.
2221         (window.benchmarkController.addOrientationListenerIfNecessary): Only mobile devices need this. Check
2222         to see for support of window.orientation.
2223         (window.benchmarkController._orientationChanged): Toggle the warning based on whether we match the
2224         landscape query. Set a state variable, which is needed for the debug harness. Call updateStartButtonState.
2225         (window.benchmarkController.updateStartButtonState):
2226         * Animometer/resources/debug-runner/animometer.js: Change _updateStartButtonState to return
2227         a boolean about whether at least one test is selected. That will be used in
2228         benchmarkController.updateStartButtonState(). Move the load event listener to the release version.
2229         (window.benchmarkController.updateStartButtonState): Override the release version, and also check
2230         that at least one test is selected.
2231
2232 2016-03-22  Geoffrey Garen  <ggaren@apple.com>
2233
2234         MallocBench: consolidate regression testing for aligned allocation
2235         https://bugs.webkit.org/show_bug.cgi?id=155762
2236
2237         Reviewed by Michael Saboff.
2238
2239         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
2240         * MallocBench/MallocBench/Benchmark.cpp:
2241         * MallocBench/MallocBench/memalign.cpp: Removed.
2242         * MallocBench/MallocBench/memalign.h: Removed. The stress_aligned test
2243         covers this and much more.
2244
2245         * MallocBench/MallocBench/stress_aligned.cpp:
2246         (benchmark_stress_aligned): Include specific tests for extreme sizes
2247         and alignments.
2248
2249 2016-03-21  Jon Lee  <jonlee@apple.com>
2250
2251         Update benchmark tests
2252         https://bugs.webkit.org/show_bug.cgi?id=155723
2253
2254         Reviewed by Darin Adler.
2255         Provisionally reviewed by Said Abou-Hallawa.
2256
2257         Add quadratic and bezier segments to the canvas path test.
2258
2259         * Animometer/resources/runner/tests.js: Some of the query strings are unnecessary.
2260         Rename the test.
2261         * Animometer/tests/master/resources/canvas-stage.js:
2262         (tune): Update to be able to take an array of possible constructors. Choose one randomly.
2263         * Animometer/tests/master/resources/canvas-tests.js:
2264         (CanvasLinePoint.Utilities.createClass): Move the point selection out to a separate
2265         function called randomPoint() for reuse. Move X_LOOPS, Y_LOOPS, and offsets into the
2266         class definition.
2267         (randomPoint): Scale the grid down a little bit so that the lines along the edge of the
2268         canvas are not cut off when the stroke size is thick.
2269         (CanvasQuadraticSegment): Added.
2270         (CanvasBezierSegment): Added.
2271         (SimpleCanvasStage): Pass in an array of the different segment types. Since line segments
2272         are short compared to the curved ones, make it twice as likely to render a line segment.
2273         (SimpleCanvasStage.animate): Update the drawing code so that we render all line segments.
2274
2275         Add a helper method that selects a random element from an array.
2276
2277         * Animometer/tests/resources/main.js:
2278         (Stage.randomElementInArray): Select a random element from the provided array.
2279
2280         * Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js: Refactor.
2281         * Animometer/tests/dom/resources/leaves.js: Ditto.
2282         * Animometer/tests/master/resources/dom-particles.js: Ditto.
2283         * Animometer/tests/master/resources/image-data.js: Ditto.
2284         * Animometer/tests/master/resources/leaves.js: Ditto.
2285         * Animometer/tests/simple/resources/simple-canvas-paths.js: Ditto.
2286
2287         Add canvas tests that includes all stroke and fill paths. This makes it possible to avoid having to
2288         include the full simple canvas suite for perf testing.
2289
2290         * Animometer/resources/debug-runner/tests.js: Add new tests. Move the canvas test into the Canvas
2291         suite.
2292         * Animometer/tests/simple/resources/simple-canvas-paths.js: Add a CanvasStroke and CanvasFill particle
2293         that random selects an object to render.
2294
2295         * Animometer/resources/debug-runner/tests.js: Move 3D suite before basic canvas suite.
2296         When updating the perf bot script, we will include the suites up to this one, but exclude
2297         the basic canvas suite.
2298
2299         Move compositing transforms test to HTML suite and remove the empty Miscellaneous suite.
2300
2301         * Animometer/resources/debug-runner/tests.js:
2302         * Animometer/tests/dom/compositing-transforms.html: Renamed from PerformanceTests/Animometer/tests/misc/compositing-transforms.html.
2303         * Animometer/tests/dom/resources/compositing-transforms.js: Renamed from PerformanceTests/Animometer/tests/misc/resources/compositing-transforms.js.
2304
2305         Clean up miscellaneous test suite. Add a canvas ellipse test, and remove the other
2306         canvas tests.
2307
2308         * Animometer/resources/debug-runner/tests.js: Add ellipse tests to the simple suite.
2309         * Animometer/tests/misc/canvas-electrons.html: Removed.
2310         * Animometer/tests/misc/canvas-stars.html: Removed.
2311         * Animometer/tests/misc/resources/canvas-electrons.js: Removed.
2312         * Animometer/tests/misc/resources/canvas-stars.js: Removed.
2313         * Animometer/tests/simple/resources/simple-canvas-paths.js: Add ellipse primitives.
2314
2315         Merge text tests together into one.
2316
2317         * Animometer/resources/runner/tests.js: Remove international.html.
2318         * Animometer/tests/master/international.html: Removed.
2319         * Animometer/tests/master/resources/text.js:
2320         (animate): Update styling. Manually calculate gradients for the shadow particles.
2321         Reduce the step size for y direction to avoid cutting text off at the margins. Increase
2322         step for x since there will be a little more room.
2323         * Animometer/tests/master/text.html: Add more translations and lay it out in a table.
2324
2325 2016-03-20  Jon Lee  <jonlee@apple.com>
2326
2327         Add a link to show debug data
2328         https://bugs.webkit.org/show_bug.cgi?id=155724
2329
2330         Reviewed by Simon Fraser.
2331
2332         * Animometer/developer.html: Attach the onclick handler to the score.
2333         * Animometer/index.html: Ditto.
2334
2335 2016-03-18  Dean Jackson  <dino@apple.com>
2336
2337         Add a basic WebGL test to Animometer
2338         https://bugs.webkit.org/show_bug.cgi?id=155475
2339         <rdar://problem/25156860>
2340
2341         Reviewed by Simon Fraser and Jon Lee.
2342
2343         Add a pretty simple WebGL test to Animometer
2344         which tries to draw a number of animated triangles.
2345         The result still needs a bit of tuning.
2346
2347         * Animometer/resources/debug-runner/tests.js: Add a new
2348         Suite for "3D" tests.
2349         * Animometer/tests/3d/resources/webgl.js: Added.
2350         * Animometer/tests/3d/webgl.html: Added.
2351
2352 2016-03-18  Jon Lee  <jonlee@apple.com>
2353
2354         Add support for statically linking to a specific test
2355         https://bugs.webkit.org/show_bug.cgi?id=155631
2356
2357         Rubber-stamped by Darin Adler.
2358
2359         * Animometer/developer.html: Update to call restartBenchmark instead so that it works for
2360         both the preset and the manually set test sessions.
2361         * Animometer/resources/debug-runner/animometer.css:
2362         (.tree .link): Style the "link" UI.
2363         * Animometer/resources/debug-runner/animometer.js: Move updateDisplay from suitesManager
2364         to optionsManager since this is a visual option and has nothing to do with the suites
2365         settings. Update the construction of the test UI by including a "link" after each test.
2366         Clicking on that link brings up a JS prompt with a URL and a query string with the current
2367         parameters of the controller, and selected test. Pasting this URL into the location bar will
2368         automatically start running the selected test.
2369         (suitesManager.suitesFromQueryString): Iterate through the Suites and tests and find the one
2370         that matches the provided parameters. Returns an object similar to the form in
2371         suitesManager.updateLocalStorageFromUI.
2372         (benchmarkController.initialize): After settings up the events and options, try parsing the
2373         query string and running the benchmark immediately. Otherwise, fall back to the form.
2374         (benchmarkController.startBenchmark): Store the options and suites into member variables for
2375         reuse in restartBenchmark.
2376         (benchmarkController.startBenchmarkImmediatelyIfEncoded): Convert the query string to an object.
2377         If that's successful, find the suite and test referenced in the query string. Start the benchmark if
2378         the search for the test succeeded.
2379         * Animometer/resources/extensions.js:
2380         (Utilities.stripNonASCIICharacters): Helper method to convert the name of the suite and test
2381         into a query-string-friendly version.
2382         (Utilities.convertObjectToQueryString): Helper method to convert an object to query string
2383         format.
2384         (Utilities.convertQueryStringToObject): Helper method to convert query string into an object
2385         with properties and values.
2386
2387 2016-03-16  Jon Lee  <jonlee@apple.com>
2388
2389         Add a new benchmark test
2390         https://bugs.webkit.org/show_bug.cgi?id=155570
2391
2392         Reviewed by Simon Fraser.
2393
2394         New Leaves test includes various image sizes and opacity.
2395
2396         * Animometer/resources/debug-runner/tests.js: Add it to the HTML test suite.
2397         * Animometer/tests/dom/leaves.html: Added.
2398         * Animometer/tests/dom/resources/leaves.js: Added. Override the
2399         (Particle.call.reset): Uses a range of sizes, and opacity.
2400         (Particle.call.animate): Opacity goes up then down. When it hits 0, reset the particle.
2401         (Particle.call.move): Set transform and opacity.
2402         * Animometer/tests/master/resources/leaves.js: Get rid of the closure so that it
2403         can be used in this test. Update the relative path so that it works in both the master
2404         and dom test suite.
2405
2406 2016-03-15  Simon Fraser  <simon.fraser@apple.com>
2407
2408         Add developer Animometer test that bounces P3-tagged images
2409         https://bugs.webkit.org/show_bug.cgi?id=155511
2410
2411         Reviewed by Tim Horton.
2412         
2413         Add a test for rendering performance of tagged images. The 5 images are tagged
2414         with the Display P3 colorspace.
2415
2416         * Animometer/resources/debug-runner/tests.js:
2417         * Animometer/tests/bouncing-particles/bouncing-tagged-images.html: Added.
2418         * Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js: Added.
2419         * Animometer/tests/bouncing-particles/resources/image1.jpg: Added.
2420         * Animometer/tests/bouncing-particles/resources/image2.jpg: Added.
2421         * Animometer/tests/bouncing-particles/resources/image3.jpg: Added.
2422         * Animometer/tests/bouncing-particles/resources/image4.jpg: Added.
2423         * Animometer/tests/bouncing-particles/resources/image5.jpg: Added.
2424
2425 2016-03-09  Jon Lee  <jonlee@apple.com>
2426
2427         Enhance existing Animometer tests
2428         https://bugs.webkit.org/show_bug.cgi?id=155261
2429
2430         Reviewed by Simon Fraser.
2431
2432         * Animometer/tests/master/resources/canvas-tests.js: Add a gradient to
2433         the filled circles. To expose more of the gradient, add another circle.
2434         (SimpleCanvasStage.call.animate): For each frame, create a gradient with
2435         undulating stop points and colors. Fill the circles twice; once with
2436         the solid color, and once with the gradient.
2437
2438         * Animometer/tests/master/resources/dom-particles.js: Refactor the
2439         emission variables into a separate stage for this test. Add a colorOffset
2440         variable to make the colors of each particle slightly different, since
2441         the ramp controller can add large numbers of particles all at once, which
2442         would otherwise get all the same color.
2443         * Animometer/tests/master/resources/particles.js:
2444         (initialize): Remove the code specific to the SVG mask test.
2445
2446         * Animometer/tests/master/resources/image-data.js:
2447         (initialize): Remove unused local variable.
2448         * Animometer/tests/master/resources/multiply.js:
2449         (initialize): Make the test harder by adding more total particles.
2450
2451 2016-03-09  Jon Lee  <jonlee@apple.com>
2452
2453         Add text tests
2454         https://bugs.webkit.org/show_bug.cgi?id=155257
2455
2456         Reviewed by Simon Fraser.
2457
2458         * Animometer/resources/extensions.js:
2459         (UnitBezier.Utilities.createClass): Add a class that computes Bezier points
2460         assuming that two of the control points are at (0,0) and (1,1). Taken from
2461         WebCore/platform/graphics/UnitBezier.h
2462         * Animometer/resources/runner/tests.js: Add a test for English text and one
2463         for international text.
2464         * Animometer/tests/master/international.html: Added.
2465         * Animometer/tests/master/resources/text.js: Added. The test assumes there is
2466         #template div which it will copy. The copies are placed behind the template, and
2467         are set with different colors each frame. They are moved around with CSS transform.
2468         * Animometer/tests/master/text.html: Added.
2469
2470         Remove the other text tests, since these ones cover the same techniques.
2471         * Animometer/tests/text/layering-text.html: Removed.
2472         * Animometer/tests/text/resources/layering-text.js: Removed.
2473         * Animometer/tests/text/resources/text-boxes.js: Removed.
2474         * Animometer/tests/text/text-boxes.html: Removed.
2475         * Animometer/resources/debug-runner/tests.js:
2476
2477 2016-03-09  Jon Lee  <jonlee@apple.com>
2478
2479         Add a new image test
2480         https://bugs.webkit.org/show_bug.cgi?id=155232
2481
2482         Reviewed by Dean Jackson.
2483         Provisionally reviewed by Said Abou-Hallawa.
2484
2485         The image test renders PNGs and moves them with translate and rotate
2486         transforms. Each particle has a lifetime, and when the lifetime ends or
2487         the particle goes offscreen, it resets itself somewhere on the stage.
2488
2489         * Animometer/resources/debug-runner/tests.js: Remove the CSS bouncing PNG
2490         images test, because this one tests the same technique.
2491         * Animometer/resources/runner/tests.js: Added here as "Leaves".
2492         * Animometer/tests/master/leaves.html: Added.
2493         * Animometer/tests/master/resources/particles.js: Refactor out the parts
2494         specific to the DOM particles test. Consequently make velocity a public
2495         member.
2496         (initialize): The options parameter is never used, so remove it.
2497         (animate): Remove unused local variable.
2498         * Animometer/tests/master/resources/leaves.js: Maintains a focal point that
2499         moves back and forth across the canvas. That point affects the velocity
2500         of the particles. When the focal point is on the leftmost side, it is
2501         between [-6, -2], and set in reset(). When the focal point is on the rightmost
2502         side, it will be from [2, 6].
2503         * Animometer/tests/master/resources/dom-particles.js: Move JS specific to
2504         this test here from particles.js.
2505
2506         New images.
2507         * Animometer/tests/master/resources/compass100.png: Added.
2508         * Animometer/tests/master/resources/console100.png: Added.
2509         * Animometer/tests/master/resources/contribute100.png: Added.
2510         * Animometer/tests/master/resources/debugger100.png: Added.
2511         * Animometer/tests/master/resources/inspector100.png: Added.
2512         * Animometer/tests/master/resources/layout100.png: Added.
2513         * Animometer/tests/master/resources/performance100.png: Added.
2514         * Animometer/tests/master/resources/script100.png: Added.
2515         * Animometer/tests/master/resources/shortcuts100.png: Added.
2516         * Animometer/tests/master/resources/standards100.png: Added.
2517         * Animometer/tests/master/resources/storage100.png: Added.
2518         * Animometer/tests/master/resources/styles100.png: Added.
2519         * Animometer/tests/master/resources/timeline100.png: Added.
2520
2521 2016-03-03  Jon Lee  <jonlee@apple.com>
2522
2523         Add ability to retrieve raw data from release harness
2524         https://bugs.webkit.org/show_bug.cgi?id=155026
2525
2526         Reviewed by Simon Fraser.
2527
2528         * Animometer/developer.html: Remove the special UI in the debug
2529         harness.
2530         * Animometer/resources/runner/animometer.css: Add styles for the
2531         overlay.
2532         * Animometer/resources/runner/animometer.js: Let 'j' show the JSON
2533         results, but only if the overlay doesn't exist. Add 'esc' key to dismiss
2534         the overlay.
2535         (window.benchmarkController.selectResults): Cycle the cases around
2536         so that the first 's' press selects both the benchmark score and
2537         the individual test scores.
2538         * Animometer/resources/debug-runner/animometer.css: Remove
2539         unneeded rules.
2540         * Animometer/resources/debug-runner/animometer.js: Make the same
2541         call to handleKeyPress.
2542
2543 2016-03-03  Jon Lee  <jonlee@apple.com>
2544
2545         Make sure multiply test particles have at least some opacity
2546         https://bugs.webkit.org/show_bug.cgi?id=155027
2547
2548         Reviewed by Simon Fraser.
2549
2550         * Animometer/tests/master/resources/multiply.js:
2551         (initialize): Have the elements spin a little faster to make it
2552         more obvious when the system is being stressed.
2553         (_addTile):
2554         (animate): Make sure each element being animated has at least 1%
2555         opacity.
2556
2557 2016-03-03  Jon Lee  <jonlee@apple.com>
2558
2559         Update image test
2560         https://bugs.webkit.org/show_bug.cgi?id=154962
2561
2562         Rubber-stamped by Darin Adler.
2563
2564         * Animometer/tests/master/image-data.html: Make each canvas have a compositing layer.
2565         * Animometer/tests/master/resources/image-data.js:
2566         (initialize): Update to load multiple images serially.
2567         (_loadImage):
2568         (tune): Use display instead of visibility. When showing an element anew, move it somewhere else.
2569         (_createTestElement): Set up the element, refactor out placement of element to _refreshElement.
2570         (_refreshElement): Place element in a tile grid.
2571         (animate):
2572         (_getRandomNeighboringPixelIndex): Use a more conservative distribution to make the effect last a little longer.
2573         * Animometer/tests/master/resources/compass.svg: Added.
2574         * Animometer/tests/master/resources/console.svg: Added.
2575         * Animometer/tests/master/resources/contribute.svg: Added.
2576         * Animometer/tests/master/resources/debugger.svg: Added.
2577         * Animometer/tests/master/resources/inspector.svg: Added.
2578         * Animometer/tests/master/resources/layout.svg: Added.
2579         * Animometer/tests/master/resources/performance.svg: Added.
2580         * Animometer/tests/master/resources/script.svg: Added.
2581         * Animometer/tests/master/resources/shortcuts.svg: Added.
2582         * Animometer/tests/master/resources/standards.svg: Added.
2583         * Animometer/tests/master/resources/storage.svg: Added.
2584         * Animometer/tests/master/resources/styles.svg: Added.
2585         * Animometer/tests/master/resources/timeline.svg: Added.
2586
2587 2016-03-02  Jon Lee  <jonlee@apple.com>
2588
2589         Add some new controllers, and refine tests
2590         https://bugs.webkit.org/show_bug.cgi?id=154914
2591
2592         Reviewed by Simon Fraser.
2593
2594         Improve tests.
2595
2596         * Animometer/tests/master/focus.html: Move each particle into a layer. Put the particle
2597         in a container element. Expand the container element by the max blur radius, and clip
2598         overflow. This way, when the blur is applied, it avoids causing layer resizes.
2599         * Animometer/tests/master/resources/focus.js:
2600         (FocusElement.Utilities.createClass): Create a container element, and put the particle
2601         inside.
2602         (hide): Set display:none.
2603         (show): Set display:block.
2604         (animate): Apply filters to the container element.
2605         (FocusStage.call.initialize): Instead of inserting and removing elements from the DOM,
2606         keep them in the stage, but set the display style instead. Use this._offsetIndex to
2607         keep track of which elements are displayed.
2608         (FocusStage.call.animate): Move some calculations around to avoid doing unneeded math.
2609         (FocusStage.call.getBlurValue): Make sure elements are always blurred.
2610         (FocusStage.call.getOpacityValue): Make sure elements have some opacity.
2611         * Animometer/tests/master/resources/multiply.js: Have the particles on the edge of the
2612         stage appear less black when the complexity gets large enough.
2613         (tune): Have this._distanceFactor calculate the factor. Avoid calculating square root each
2614         frame.
2615
2616 2016-03-02  Jon Lee  <jonlee@apple.com>
2617
2618         Add some new controllers, and refine tests
2619         https://bugs.webkit.org/show_bug.cgi?id=154914
2620
2621         Reviewed by Simon Fraser.
2622
2623         Add a controller that centers around 30 fps instead of 60 fps.
2624
2625         * Animometer/developer.html: Add a new option.
2626         * Animometer/resources/debug-runner/animometer.js:
2627         * Animometer/resources/runner/animometer.js:
2628         (this._processData.findRegression): When calculating the complexity-frameLength regression,
2629         check the controller, and use a 30 fps baseline if needed.
2630         * Animometer/resources/statistics.js:
2631         (Regression.Utilities.createClass): Update to allow clients to specify the baseline
2632         frame length.
2633         * Animometer/tests/resources/main.js:
2634         (tune): Override some of the constants in RampController. Move those constants out for
2635         easier reading.
2636
2637         Add a fixed controller, with no step.
2638
2639         * Animometer/developer.html: Add a controller that takes no step.
2640         * Animometer/resources/debug-runner/animometer.js:
2641         * Animometer/tests/resources/main.js:
2642         (Rotater.Utilities.createClass):
2643
2644         Switch to ramp controller as default.
2645
2646         * Animometer/developer.html: Increase the test length to 20 seconds.
2647         * Animometer/resources/debug-runner/animometer.js: Rename the "adjustment" field to "controller"
2648         since that is a more accurate description.
2649         * Animometer/resources/debug-runner/graph.js:
2650         * Animometer/resources/runner/animometer.js: Update preferences for release suite.
2651
2652 2016-02-26  Jon Lee  <jonlee@apple.com>
2653
2654         Address Dean's comments in 154673.
2655
2656         * Animometer/developer.html:
2657         * Animometer/resources/debug-runner/animometer.js:
2658         * Animometer/resources/debug-runner/graph.js:
2659
2660 2016-02-25  Jon Lee  <jonlee@apple.com>
2661
2662         Update animation benchmark and tests
2663         https://bugs.webkit.org/show_bug.cgi?id=154673
2664
2665         Reviewed by Dean Jackson.
2666
2667         Update test visuals.
2668
2669         * Animometer/tests/master/resources/canvas-tests.js:
2670         (CanvasLineSegment.Utilities.createClass): Line segments near the edge of the stage get clipped.
2671         Update circle position and radius to minimize impact.
2672         * Animometer/tests/master/resources/particles.js: Get rid of rotating gradient background, and
2673         have 3 locations for emitting particles.
2674         * Animometer/tests/master/resources/multiply.js: Update distance metric so that fringe tiles get
2675         more color.
2676         (initialize): Fix some of the math for laying out the tiles.
2677         (animate): When a tile isn't used, set visibility: hidden.
2678
2679 2016-02-24  Jon Lee  <jonlee@apple.com>
2680
2681         Update animation benchmark and tests
2682         https://bugs.webkit.org/show_bug.cgi?id=154673
2683
2684         Reviewed by Dean Jackson.
2685
2686         Update the ramp controller.
2687
2688         The controller refines the complexity interval to test across.
2689
2690         * Animometer/resources/statistics.js: Add functions that estimate cumulative distribution function.
2691         (Regression): For the flat regression, force the first segment to be at 60 fps.
2692         (valueAt): Add convenience function to return interpolated value based on the regression used.
2693         (_calculateRegression): Include the number of points included for both segments, and the piecewise
2694         errors.
2695         * Animometer/tests/resources/math.js: Make the Kalman estimator subclass Experiment, and allow it
2696         to be reset.
2697
2698         * Animometer/tests/resources/main.js: Initialize the tier such that it starts at 10^0 = 1.
2699         Increase the number of ramps. Maintain three FPS thresholds-- the frame rate of interest, a limit
2700         on the lowest FPS we care to go for later interpolation, and a minimum FPS threshold we want to
2701         aim for each ramp. Also keep three estimators: a running average of the change point, a minimum
2702         boundary for each ramp, and an estimator for all the frames within an interval. The first two
2703         are used to determine the parameters of the next ramp, and the latter allows us to refine the
2704         parameters.
2705         (update): During the tier phase, it is possible that the highest complexity possible for a test
2706         won't stress the system enough to trigger stopping the tier phase and transitioning to the ramps.
2707         If the complexity doesn't change when going to the next tier, we've maxed the test out, and move
2708         on. When the tier phase completed, turn off Controller.frameLengthEstimator, which estimates the
2709         FPS at each tier.
2710         (tune): At each interval, look at the confidence distribution of being on the 60 FPS side or the
2711         slow side. If the slowest FPS we achieve at the ramp's maximum complexity is not at least
2712         _fpsRampSlowThreshold, then increase the maximum complexity. If we ever achieve 60 FPS, increase
2713         the ramp's minimum complexity to that level. If, at an even lower complexity, a glitch causes the
2714         FPS to drop, we reset the minimum complexity.
2715
2716         Have the bootstrap calculation occur between tests. Clean up harness.
2717
2718         * Animometer/resources/debug-runner/animometer.js: Run bootstrap after a test has
2719         completed to avoid doing all of it at the end before showing the results. Clean up
2720         parameters being passed around.
2721         * Animometer/resources/debug-runner/tests.js:
2722         (text):
2723         * Animometer/resources/runner/animometer.js:
2724         (this._processData.calculateScore): Save the results to the same object holding the data.
2725         (this._processData._processData): In the case where a file is dragged, calculate the score
2726         serially. Grab the results object and move it to the results variable and remove it from
2727         the data object. This avoids serializing the results into the JSON.
2728         (this._processData.findRegression): Include the samples used for bootstrapping. Reduce the
2729         resample size to shorten the wait.
2730         * Animometer/resources/runner/benchmark-runner.js:
2731         * Animometer/resources/statistics.js:
2732         (bootstrap): Update how bootstrapData is sorted. In some regression results the mix of
2733         floats and integers causes an alphabetical sort to occur.
2734         * Animometer/resources/strings.js:
2735
2736         Add meta charset so that encodings between harness and test match.
2737
2738         * Animometer/tests/bouncing-particles/bouncing-canvas-images.html:
2739         * Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html:
2740         * Animometer/tests/bouncing-particles/bouncing-css-images.html:
2741         * Animometer/tests/bouncing-particles/bouncing-css-shapes.html:
2742         * Animometer/tests/bouncing-particles/bouncing-svg-images.html:
2743         * Animometer/tests/bouncing-particles/bouncing-svg-shapes.html:
2744         * Animometer/tests/master/canvas-stage.html:
2745         * Animometer/tests/master/focus.html:
2746         * Animometer/tests/master/image-data.html:
2747         * Animometer/tests/master/multiply.html:
2748         * Animometer/tests/master/particles.html:
2749         * Animometer/tests/misc/canvas-electrons.html:
2750         * Animometer/tests/misc/canvas-stars.html:
2751         * Animometer/tests/misc/compositing-transforms.html:
2752         * Animometer/tests/simple/simple-canvas-paths.html:
2753         * Animometer/tests/simple/tiled-canvas-image.html:
2754         * Animometer/tests/template/template-canvas.html:
2755         * Animometer/tests/template/template-css.html:
2756         * Animometer/tests/template/template-svg.html:
2757         * Animometer/tests/text/layering-text.html:
2758         * Animometer/tests/text/text-boxes.html:
2759
2760         Update test harness reporting.
2761
2762         * Animometer/developer.html: Add missing meta charset.
2763         * Animometer/index.html: Remove unnecessary utf-8 declaration.
2764         * Animometer/resources/debug-runner/animometer.css: Add convenience classes for
2765         formatting the results table.
2766         * Animometer/resources/debug-runner/animometer.js: Adjust which stats are shown.
2767         * Animometer/resources/debug-runner/tests.js: Display bootstrapping statistics.
2768         * Animometer/resources/strings.js: Move strings not used by the release harness.
2769
2770         Switch to a pseudo-random number generator.
2771
2772         * Animometer/resources/statistics.js: Add a Pseudo class, with a simple
2773         pseudo-random number generator.
2774         (_calculateRegression): Reset the generator before running bootstrap.
2775         (bootstrap): Deleted.
2776
2777         Replace Math.random with Pseudo.random.
2778         * Animometer/tests/master/resources/canvas-tests.js:
2779         * Animometer/tests/master/resources/focus.js:
2780         * Animometer/tests/master/resources/particles.js:
2781         * Animometer/tests/resources/main.js:
2782
2783         Use bootstrapping to get confidence interval in the breakpoint.
2784
2785         For the ramp controller, calculate the piecewise regression, and then use
2786         bootstrapping in order to find the 95% confidence interval. Use the raw data.
2787
2788         * Animometer/developer.html: Default to the complexity graph. Add a legend
2789         checkbox to toggle visibility of the bootstrap score and histogram.
2790         * Animometer/resources/debug-runner/animometer.css: Make some more space to show
2791         the old raw and average scores in the legend. Add new styles for the data.
2792         * Animometer/resources/debug-runner/graph.js:
2793         (_addRegressionLine): Allow passing an array for the variance bar tied to the
2794         regression line. Now |stdev| is |range|.
2795         (createComplexityGraph): Add bootstrap median, and overlay a histogram of
2796         the bootstrap samples. Switch raw samples from circles to X's.
2797         (onComplexityGraphOptionsChanged): Allow toggling of the bootstrap data.
2798         (onGraphTypeChanged): Move the regressions for the raw and average samples to the
2799         legend. In the subtitle use the bootstrap median, and include the 95% confidence
2800         interval.
2801         * Animometer/resources/runner/animometer.js:
2802         (this._processData.findRegression): Factor out the code that determines which
2803         samples to include when calculating the piecewise regression. For series that have
2804         many samples, or a wider range of recorded complexities, throw away the 2.5%
2805         lowest and highest samples before calculating the regression. Keep all samples
2806         if the number of samples to regress is small or the range of complexities is
2807         narrow.
2808         (this._processData._calculateScore): Factor out regression calculation to
2809         findRegression(). Bootstrap the change point of the regression. The score is the
2810         median.
2811         * Animometer/resources/statistics.js:
2812         (_calculateRegression): Correct an issue in the calculation of the regression, where
2813         the denominator can be 0.
2814         (bootstrap): Template for bootstrapping. Create a bootstrap sample array, Create
2815         re-samples by random selection with replacement. Return the 95% confidence samples,
2816         the bootstrap median, mean, and the data itself.
2817         * Animometer/resources/strings.js: Add bootstrap.
2818         * Animometer/tests/resources/main.js:
2819         (processSamples): Don't prematurely cut the sample data.
2820
2821         Fix graph drawing.
2822
2823         * Animometer/resources/debug-runner/animometer.js: Add spacing in the JSON output.
2824         Multiple tests output a lot of JSON and can hang when selecting JSON with no whitespace.
2825         * Animometer/resources/debug-runner/animometer.css:
2826         (#complexity-graph .series.raw circle): Update the color.
2827         * Animometer/resources/debug-runner/graph.js: Use the FPS axis instead of the
2828         complexity axis, which can vary in domain. For determining the complexity domain,
2829         only use samples after samplingTimeOffset.
2830
2831         Allow dropping results JSON.
2832
2833         * Animometer/developer.html: Add a button.
2834         * Animometer/resources/debug-runner/animometer.css:
2835         * Animometer/resources/debug-runner/animometer.js: Read the data and go straight
2836         to the dashboard. With JSON output, write out only the options and the raw data.
2837
2838         Teach the harness to evaluate the samples and determine the test score.
2839
2840         This will allow us to update how the score is calculated separately from the samples recorded.
2841         This also prepares the harness to be able to accept JSON of prior runs.
2842
2843         * Animometer/resources/strings.js: Clean up and remove unneeded strings and reduce some of the
2844         hierarchy.
2845         * Animometer/resources/debug-runner/tests.js: Update to use the new strings.
2846
2847         * Animometer/tests/resources/main.js: Allow all controllers to show a complexity-FPS graph.
2848         (_processComplexitySamples): Factor out some of the sample processing done in the ramp
2849         controller for the benefit of the other controllers. |complexitySamples| contains a list of
2850         samples. Sort the samples by complexity. Optionally remove the top x% of samples.
2851         Group them, and calculate distribution of samples within the same complexity, and add those as
2852         new entries into |complexityAverageSamples|.
2853         (Controller.processSamples): Move the code responsible for determining the complexity and FPS
2854         scores out to ResultsDashboard. The structure of the data returned by the controller is:
2855
2856         {
2857             controller: [time-regression, time-regression, ...], // optional, data specific to controller
2858             marks: [...],
2859             samples: {                    // all of the sample data
2860                 controller: [...],
2861                 complexity: [...],        // processed from controller samples
2862                 complexityAverage: [...], // processed from complexity samples
2863             }
2864         }
2865
2866         (AdaptiveController.processSamples): Adding the target frame length is no longer necessary; we
2867         now pass the test options to the graph.
2868         (Regression): Move to statistics.js.
2869         * Animometer/resources/statistics.js: Move Regression to here. Add a check if the sampling range
2870         only contains one sample, since we cannot calculate a regression from one sample point.
2871
2872         Teach the test harness to evaluate the data.
2873         * Animometer/resources/runner/animometer.js:
2874         (ResultsDashboard): Store the options used to run the test and the computed results/score separately
2875         from the data. The results are stored as:
2876
2877         {
2878             score: /* geomean of iteration score */,
2879             iterationsResults: [
2880                 {
2881                     score: /* geomean of tests */,
2882                     testsResults: {
2883                         suiteName: {
2884                             testName: {
2885                                 controller: {
2886                                     average:
2887                                     concern:
2888                                     stdev:
2889                                     percent:
2890                                 },
2891                                 frameLength: { ... },
2892                                 complexity: {
2893                                     complexity:
2894                                     stdev:
2895                                     segment1:
2896                                     segment2:
2897                                 },
2898                                 complexityAverage: { ... }
2899                             },
2900                             testName: { ... },
2901                         },
2902                         ... next suite ...
2903                     }
2904                 },
2905                 { ...next iteration... }
2906             ]
2907         }
2908
2909         * Animometer/resources/debug-runner/animometer.js: Pass options around instead of relying
2910         on what was selected in the form. This will later allow for dropping previous results, and
2911         using those runs' options when calculating scores.
2912         (ResultsTable._addGraphButton): Simplify button action by using attached test data.
2913         * Animometer/resources/debug-runner/graph.js: Refactor to use the data.
2914
2915         Consolidate JS files, and move statistics out to a separate JS.
2916
2917         Preparation for having the Controller only handle recording and storage of the samples,
2918         and leave the evaluation of the test score out to the harness. Move Experiment to
2919         a new statistics.js, where Regression will also eventually go. Get rid of algorithm.js
2920         and move it to utilities.js since the Heap is used only for Experiments.
2921
2922         * Animometer/tests/resources/algorithm.js: Removed. Heap is in utilities.js.
2923         * Animometer/tests/resources/sampler.js: Removed. Experiment is in statistics.js,
2924         Sampler in main.js.
2925         * Animometer/tests/resources/main.js: Move Sampler here.
2926         * Animometer/resources/statistics.js: Added. Move Statistics and Experiment here.
2927         * Animometer/resources/extensions.js: Move Heap here. Attach static method to create
2928         a max or min heap to Heap, instead of a new Algorithm object.
2929
2930         Update JS files.
2931         * Animometer/developer.html:
2932         * Animometer/index.html:
2933         * Animometer/tests/bouncing-particles/bouncing-canvas-images.html:
2934         * Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html:
2935         * Animometer/tests/bouncing-particles/bouncing-css-images.html:
2936         * Animometer/tests/bouncing-particles/bouncing-css-shapes.html:
2937         * Animometer/tests/bouncing-particles/bouncing-svg-images.html:
2938         * Animometer/tests/bouncing-particles/bouncing-svg-shapes.html:
2939         * Animometer/tests/master/canvas-stage.html:
2940         * Animometer/tests/master/focus.html:
2941         * Animometer/tests/master/image-data.html:
2942         * Animometer/tests/master/multiply.html:
2943         * Animometer/tests/master/particles.html:
2944         * Animometer/tests/misc/canvas-electrons.html:
2945         * Animometer/tests/misc/canvas-stars.html:
2946         * Animometer/tests/misc/compositing-transforms.html:
2947         * Animometer/tests/simple/simple-canvas-paths.html:
2948         * Animometer/tests/simple/tiled-canvas-image.html:
2949         * Animometer/tests/template/template-canvas.html:
2950         * Animometer/tests/template/template-css.html:
2951         * Animometer/tests/template/template-svg.html:
2952         * Animometer/tests/text/layering-text.html:
2953         * Animometer/tests/text/text-boxes.html:
2954
2955         Fix the cursor in the graph analysis when the min
2956         complexity is not 0.
2957
2958         * Animometer/resources/debug-runner/graph.js:
2959         (_addRegression):
2960         (createComplexityGraph):
2961
2962 2016-02-23  Geoffrey Garen  <ggaren@apple.com>
2963
2964         Fix some issues in MallocBench
2965         https://bugs.webkit.org/show_bug.cgi?id=154600
2966
2967         Reviewed by Oliver Hunt.
2968
2969         * MallocBench/MallocBench.xcodeproj/project.pbxproj: Make nimlang.ops
2970         a part of the build so that it copies to the right place for execution.
2971
2972         * MallocBench/MallocBench/Interpreter.cpp:
2973         (Interpreter::Interpreter): Don't try to open .ops files for writing
2974         because we're only going to read and we might need extra permissions to
2975         write.
2976
2977         (Interpreter::~Interpreter): Give more context when opening a file fails
2978         to help with debugging.
2979
2980         * MallocBench/MallocBench/stress.cpp:
2981         (benchmark_stress): Reduce iterations to complete in less than 10 seconds.
2982
2983 2016-02-23  Michael Saboff  <msaboff@apple.com>
2984
2985         Unreviewed change to revert extraneous changes made part of change set 196955.
2986
2987         * MallocBench/MallocBench/Interpreter.cpp:
2988         (Interpreter::doMallocOp):
2989         (Interpreter::Thread::switchTo):
2990         (writeData): Deleted.
2991
2992 2016-02-19  Michael Saboff  <msaboff@apple.com>
2993
2994         MallocBench: Added recording for nimlang website, new recording details and added new options
2995         https://bugs.webkit.org/show_bug.cgi?id=154485
2996
2997         Reviewed by Geoff Garen.
2998
2999         Added new capabilities to MallocBench.  These include:
3000             Added a recording of http://nim-lang.org/docs/lib.html.
3001             Added thread id to the recording and the ability to playback switching threads in MallocBench
3002             Added aligned allocations to recordings and the ability to playback
3003             Added --use-thread-id option to honor recorded thread ids
3004             Added --detailed-report to output remaining allocations by size after playback
3005             Added --no-warmup to not run the warm up iteration
3006
3007         Changed the way that options are passed down to the benchmarks.  Instead of passing individual
3008         boolean or numeric option values, just pass a reference the CommandLine itself.  Each benchmark
3009         can access the options that are appropriate.  The Benchmark class also uses the options for
3010         is parallel, run counts and warm up.
3011
3012         Added thread id and aligned malloc to the Op by noticing that structure padding and Opcode allowed
3013         for another 32 bits of data.  Breaking that unused 32 bits into a 16 bit thread id value and a
3014         16 bit log base 2 of the alignment for aligned malloc allowed for existing recordings to playback
3015         without any incompatibilities.
3016
3017         Threaded operation is simulated by creating threads as needed.  As long as the next Op's thread id
3018         is the same as the last, operation continues as normal.  When the next Op has a different thread id,
3019         we switch to that thread using the shared Op stream to continue playing back.  There is a mutex to
3020         assure that only one thread is really running at a time and a condition variable used to wait
3021         that the current thread id matches each block thread's thread id.  This doesn't simulate true
3022         concurrent threading, but is instead plays back Ops recorded for multiple thread faithfully.
3023
3024         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
3025         * MallocBench/MallocBench/Benchmark.cpp:
3026         (deallocateHeap):
3027         (Benchmark::Benchmark):
3028         (Benchmark::runOnce):
3029         (Benchmark::run):
3030         * MallocBench/MallocBench/Benchmark.h:
3031         (Benchmark::isValid):
3032         * MallocBench/MallocBench/CommandLine.cpp:
3033         (CommandLine::printUsage):
3034         * MallocBench/MallocBench/CommandLine.h:
3035         (CommandLine::isValid):
3036         (CommandLine::benchmarkName):
3037         (CommandLine::isParallel):
3038         (CommandLine::useThreadID):
3039         (CommandLine::detailedReport):
3040         (CommandLine::warmUp):
3041         (CommandLine::heapSize):
3042         (CommandLine::runs):
3043         * MallocBench/MallocBench/Interpreter.cpp:
3044         (Interpreter::Interpreter):
3045         (Interpreter::run):
3046         (Interpreter::readOps):
3047         (Interpreter::doOnSameThread):
3048         (Interpreter::switchToThread):
3049         (Interpreter::detailedReport):
3050         (compute2toPower):
3051         (writeData):
3052         (Interpreter::doMallocOp):
3053         (Interpreter::Thread::Thread):
3054         (Interpreter::Thread::stop):
3055         (Interpreter::Thread::~Thread):
3056         (Interpreter::Thread::runThread):
3057         (Interpreter::Thread::waitToRun):
3058         (Interpreter::Thread::switchTo):
3059         * MallocBench/MallocBench/Interpreter.h:
3060         (Interpreter::Thread::isMainThread):
3061         * MallocBench/MallocBench/alloc_free.cpp: Added.
3062         (benchmark_alloc_free):
3063         * MallocBench/MallocBench/alloc_free.h: Added.
3064         * MallocBench/MallocBench/balloon.cpp:
3065         (benchmark_balloon):
3066         * MallocBench/MallocBench/balloon.h:
3067         * MallocBench/MallocBench/big.cpp:
3068         (benchmark_big):
3069         * MallocBench/MallocBench/big.h:
3070         * MallocBench/MallocBench/churn.cpp:
3071         (benchmark_churn):
3072         * MallocBench/MallocBench/churn.h:
3073         * MallocBench/MallocBench/facebook.cpp:
3074         (benchmark_facebook):
3075         * MallocBench/MallocBench/facebook.h:
3076         * MallocBench/MallocBench/flickr.cpp:
3077         (benchmark_flickr):
3078         (benchmark_flickr_memory_warning):
3079         * MallocBench/MallocBench/flickr.h:
3080         * MallocBench/MallocBench/fragment.cpp:
3081         (validate):
3082         (benchmark_fragment):
3083         (benchmark_fragment_iterate):
3084         * MallocBench/MallocBench/fragment.h:
3085         * MallocBench/MallocBench/list.cpp:
3086         (benchmark_list_allocate):
3087         (benchmark_list_traverse):
3088         * MallocBench/MallocBench/list.h:
3089         * MallocBench/MallocBench/main.cpp:
3090         (main):
3091         * MallocBench/MallocBench/medium.cpp:
3092         (benchmark_medium):
3093         * MallocBench/MallocBench/medium.h:
3094         * MallocBench/MallocBench/memalign.cpp:
3095         (test):
3096         (benchmark_memalign):
3097         * MallocBench/MallocBench/memalign.h:
3098         * MallocBench/MallocBench/message.cpp:
3099         (benchmark_message_one):
3100         (benchmark_message_many):
3101         * MallocBench/MallocBench/message.h:
3102         * MallocBench/MallocBench/nimlang.cpp: Added.
3103         (benchmark_nimlang):
3104         * MallocBench/MallocBench/nimlang.h: Added.
3105         * MallocBench/MallocBench/nimlang.ops: Added.
3106         * MallocBench/MallocBench/realloc.cpp:
3107         (benchmark_realloc):
3108         * MallocBench/MallocBench/realloc.h:
3109         * MallocBench/MallocBench/reddit.cpp:
3110         (benchmark_reddit):
3111         (benchmark_reddit_memory_warning):
3112         * MallocBench/MallocBench/reddit.h:
3113         * MallocBench/MallocBench/stress.cpp:
3114         (deallocate):
3115         (benchmark_stress):
3116         * MallocBench/MallocBench/stress.h:
3117         * MallocBench/MallocBench/stress_aligned.cpp:
3118         (benchmark_stress_aligned):
3119         * MallocBench/MallocBench/stress_aligned.h:
3120         * MallocBench/MallocBench/theverge.cpp:
3121         (benchmark_theverge):
3122         (benchmark_theverge_memory_warning):
3123         * MallocBench/MallocBench/theverge.h:
3124         * MallocBench/MallocBench/tree.cpp:
3125         (benchmark_tree_allocate):
3126         (benchmark_tree_traverse):
3127         (benchmark_tree_churn):
3128         * MallocBench/MallocBench/tree.h:
3129         * MallocBench/run-malloc-benchmarks:
3130
3131 2016-02-11  Jon Lee  <jonlee@apple.com>
3132
3133         Fix a missing refactoring.
3134
3135         * Animometer/tests/master/resources/multiply.js:
3136         (animate): Move to Utilities.lerp.
3137         (_lerp): Deleted.
3138
3139 2016-02-10  Jon Lee  <jonlee@apple.com>
3140
3141         Add new benchmark tests.
3142         https://bugs.webkit.org/show_bug.cgi?id=154063
3143
3144         Provisionally reviewed by Said Abou-Hallawa.
3145
3146         Add tests for get/put image data, filters, opacity, and css transforms.
3147
3148         * Animometer/resources/runner/benchmark-runner.js:
3149         (_runBenchmarkAndRecordResults): Update the body background color to match that of
3150         the stage.
3151         (this._runNextIteration): Clear the background color style for the results page.
3152         * Animometer/resources/runner/tests.js:
3153         * Animometer/tests/master/focus.html: Added.
3154         * Animometer/tests/master/image-data.html: Added.
3155         * Animometer/tests/master/multiply.html: Added.
3156         * Animometer/tests/master/resources/focus.js: Added.
3157         * Animometer/tests/master/resources/image-data.js: Added.
3158         * Animometer/tests/master/resources/multiply.js: Added.
3159         * Animometer/tests/master/resources/stage.css: Move common styles out.
3160         * Animometer/tests/resources/main.js: Update Stage.randomBool to use Math.random.
3161         Add Stage.randomSign for randomly setting a direction. Add the notion of the
3162         current timestamp of the test to Benchmark, since some animations cycle through
3163         colors and rely on an incremental counter like the time.
3164
3165 2016-02-09  Said Abou-Hallawa  <sabouhallawa@apple.com>
3166
3167         Add internal benchmark tests for CSS mix-blend-modes and filters
3168         https://bugs.webkit.org/show_bug.cgi?id=154058
3169
3170         Provisionally reviewed by Jon Lee.
3171
3172         * Animometer/resources/debug-runner/tests.js: Include the new tests in the
3173         "HTML suite" of the debug runner.
3174
3175         * Animometer/resources/extensions.js:
3176         (Utilities.browserPrefix):
3177         (Utilities.setElementPrefixedProperty): Utility functions to allow setting
3178         prefixed style properties.
3179
3180         * Animometer/tests/bouncing-particles/resources/bouncing-css-shapes.js:
3181         Set the mix-blend-mode and the filter to some random values if the options
3182         of the test requested that.
3183
3184         * Animometer/tests/bouncing-particles/resources/bouncing-particles.js:
3185         (parseShapeParameters): Parse the url options "blend" and "filter" and set
3186         the corresponding flags.
3187
3188         * Animometer/tests/resources/main.js:
3189         (randomStyleMixBlendMode):
3190         (randomStyleFilter): Return random mix-blend-mode and filter.
3191
3192 2016-02-08  Jon Lee  <jonlee@apple.com>
3193
3194         Add a ramp controller
3195         https://bugs.webkit.org/show_bug.cgi?id=154028
3196
3197         Provisionally reviewed by Said Abou-Hallawa.
3198
3199         Enhance the graph to include a complexity-fps graph, in addition
3200         to the time graph.
3201
3202         * Animometer/developer.html: Add a ramp option.
3203         * Animometer/resources/debug-runner/animometer.css: Update the style.
3204         * Animometer/resources/strings.js: Flatten the Strings.text constants.
3205         * Animometer/resources/debug-runner/animometer.js:
3206         (ResultsTable.call._addGraphButton): Refactor.
3207         (ResultsTable.call._addTest): Add regression data.
3208         (benchmarkController): Add a form that allows the user to switch between the two forms,
3209         Add a form that allows the user to toggle different data. Hide certain header columns
3210         depending on the selected controller.
3211         * Animometer/resources/debug-runner/graph.js: Add the complexity regressions.
3212         * Animometer/resources/debug-runner/tests.js: Add headers for the ramp results.
3213         * Animometer/resources/runner/animometer.js:
3214         (ResultsTable): If a header is disabled don't include them in _flattenedHeaders.
3215         * Animometer/tests/resources/main.js:
3216         (Controller): Allow options to specify the capacity for sample arrays.
3217         (Regression): A piecewise regression that tries to fit a slope and a flat profile.
3218         (_calculateRegression): Options can fix the slope and bias when calculating the minimal
3219         error. Sweep across the samples in time (which could be backward depending on the controller)
3220         and calculate the intersection point.
3221         (RampController): This controller assumes that the target frame rate is below
3222         58 FPS. It runs in two stages. The first stage quickly determines the order of
3223         magnitude of objects needed to stress the system by the setting the complexity
3224         to increasingly difficult tiers. Perform a series of ramps descending from a
3225         high-water mark of complexity. The complexity needed to reach the target frame
3226         length is done by performing a piecewise regression on each ramp, and track a
3227         running average of these values. For the next ramp, make that running average
3228         the center of the ramp. With a minimum complexity of 0, the high-water mark is
3229         twice that average. The score is based on the highest complexity that can
3230         reach 60 fps.
3231
3232 2016-02-08  Jon Lee  <jonlee@apple.com>
3233
3234         Address Said's comments on the benchmark, and do some clean up.
3235
3236         * Animometer/developer.html:
3237         * Animometer/resources/debug-runner/animometer.css: Add styles for averages.
3238         * Animometer/resources/debug-runner/animometer.js: Use the right
3239         Strings constants.
3240         * Animometer/resources/debug-runner/graph.js:
3241         (_addRegressionLine): Add missing code to draw the line and standard
3242         deviation highlight.
3243         (onGraphTypeChanged): Remove unneeded variables
3244         (onTimeGraphOptionsChanged):
3245         * Animometer/resources/runner/benchmark-runner.js:
3246         (_runBenchmarkAndRecordResults): Rename samplers to suiteResults and
3247         _suitesSamplers to _suitesResults.
3248         * Animometer/tests/resources/main.js:
3249         (results): Call processSamples().
3250         (update): Change sampling timestamp comparison.
3251         (_animateLoop): Move shouldStop call to before the update.
3252         * Animometer/tests/resources/sampler.js:
3253         (process): Rename to processSamples().
3254
3255 2016-02-07  Jon Lee  <jonlee@apple.com>
3256
3257         Teach Controller to measure intervals, and turn off the frame length estimator.
3258
3259         * Animometer/tests/resources/main.js: Default interval length is 100 ms.
3260         (start): Set the first interval.
3261         (_measureAndResetInterval): Reports the average frame length of the interval that just
3262         completed, and sets up the next interval.
3263         (update): If there is no length, then just use the estimator per frame, otherwise the
3264         estimator measures per interval. Add a didFinishInterval for subclasses to process
3265         prior to recording the sample. Update tune() to include whether an interval had
3266         finished.
3267         (StepController): Step controllers don't measure on an interval basis.
3268
3269 2016-02-07  Jon Lee  <jonlee@apple.com>
3270
3271         Minor refactoring. Rename Controller._estimator to Controller._frameLengthEstimator
3272         and switch the parameters for start(), update(), and tune(), so that the timestamp
3273         is first and stage is second.
3274
3275         * Animometer/tests/resources/main.js:
3276
3277 2016-02-07  Jon Lee  <jonlee@apple.com>
3278
3279         Move ResultsTable functionality not needed for release tests out.
3280         Move reporting of score and mean to selection of the time-based graph.
3281
3282         * Animometer/developer.html: Rename graph-options to time-graph-options.
3283         * Animometer/resources/debug-runner/animometer.js:
3284         (DeveloperResultsTable): Moved from runner/animometer.js. Switch from mean
3285         values to "average" objects which can hold stdev. Move graph button and
3286         calculation of noisy measurements here. Sophisticated header processing
3287         is not needed in release suite.
3288         (populateTable): Use DeveloperResultsTable.
3289         * Animometer/resources/debug-runner/graph.js: Pull time graph creation to
3290         its own function, and add a new onGraphTypeChanged handler in preparation
3291         of a complexity graph to be added later.
3292         * Animometer/resources/runner/animometer.js:
3293         (ResultsTable): Simplify to just handle test names and scores.
3294
3295 2016-02-07  Jon Lee  <jonlee@apple.com>
3296
3297         Tests: reuse objects already made.
3298
3299         Avoid thrash of object creation and removal by maintaining an index that
3300         moves along the array as the adjust values change. If the tune value
3301         requires more objects than the maximum size of the object array, then create
3302         new objects. This means that the object array size never decreases.
3303
3304         * Animometer/tests/master/resources/canvas-stage.js: Maintain a separate
3305         offsetIndex. For these tests, we want to avoid drawing the oldest objects,
3306         so the scene will draw the object at offsetIndex to the end of the array.
3307         (tune): Reverse the logic since "removal" of objects is much simpler and
3308         involves simply changing the offsetIndex.
3309         (animate): Update the for loop to draw from offsetIndex to the end.