Links with empty hrefs should not be drag sources
[WebKit-https.git] / PerformanceTests / ChangeLog
1 2017-03-08  Yusuke Suzuki  <utatane.tea@gmail.com>
2
3         [JSC] Add helpers to run Six Speed to run-jsc-benchmarks
4         https://bugs.webkit.org/show_bug.cgi?id=169351
5
6         Reviewed by Saam Barati.
7
8         * SixSpeed/wrappers/wrapper.js: Added.
9         (assertEqual):
10         (test):
11         (jscRun):
12
13 2017-03-08  Jon Davis  <jond@apple.com>
14
15         Fixed ARES-6 animations for other browsers
16
17         Rubber stamped by Saam Barati.
18
19         * ARES-6/styles.css:
20         (.swoop):
21         (.logo):
22
23 2017-03-07  Filip Pizlo  <fpizlo@apple.com>
24
25         Revise the ARES-6 explainer text
26         https://bugs.webkit.org/show_bug.cgi?id=169287
27
28         Reviewed by Saam Barati.
29         
30         Just did some copy-editing.
31
32         * ARES-6/about.html:
33
34 2017-03-06  Jon Davis  <jond@apple.com>
35
36         Refined the ARES-6 experience
37         https://bugs.webkit.org/show_bug.cgi?id=169205
38
39         Reviewed by Saam Barati.
40
41         * ARES-6/about.html: Copied from PerformanceTests/ARES-6/index.html.
42         * ARES-6/driver.js:
43         (Driver.prototype.start):
44         (Driver.prototype._updateIterations):
45         (Driver):
46         * ARES-6/index.html:
47         * ARES-6/stats.js:
48         (Stats.prototype.toString.span.span):
49         * ARES-6/styles.css:
50         (body):
51         (.swoop):
52         (p):
53         (.logo):
54         (.start):
55         (.test):
56         (.test:nth-child(odd)):
57         (.overall):
58         (.about h2):
59         (.button):
60         (.button:hover):
61         (.button.return:before):
62         (@keyframes fade-in):
63         (@keyframes scale-in):
64         (@keyframes flicker-in):
65         (@keyframes wipe-in):
66         (.flip): Deleted.
67         (.testing, .about): Deleted.
68         (.testing): Deleted.
69         (.about): Deleted.
70         (#about:target): Deleted.
71
72 2017-03-06  Jeremy Jones  <jeremyj@apple.com>
73
74         StandardCanPlayThrough.html does not work on iOS.
75         https://bugs.webkit.org/show_bug.cgi?id=169207
76
77         Reviewed by Jer Noble.
78
79         On iOS only, add muted=true and autoplay=allowed. The loading policy on iOS requires these attributes to load enough
80         data to trigger canPlayThrough.
81
82         * Media/StandardCanPlayThrough.html:
83
84 2017-02-13  Zalan Bujtas <zalan@apple.com>
85
86         Simple line layout: Add performance test with text-rendering: optimizeSpeed
87         https://bugs.webkit.org/show_bug.cgi?id=168248
88
89         Reviewed by Simon Fraser.
90
91         * Layout/simple-line-layout-with-varying-content-and-optimized-speed.html: Added.
92
93 2017-02-10  Zalan Bujtas  <zalan@apple.com>
94
95         Mail hangs when removing multiple rows from large table.
96         https://bugs.webkit.org/show_bug.cgi?id=168103
97         <rdar://problem/30090186>
98
99         Reviewed by Ryosuke Niwa.
100
101         * DOM/large-table-edit.html: Added.
102
103 2017-02-05  Filip Pizlo  <fpizlo@apple.com>
104
105         Beef up the ARES-6 explainer text
106         https://bugs.webkit.org/show_bug.cgi?id=167867
107
108         Reviewed by Saam Barati.
109         
110         Added appropriate links.
111
112         Added text that explains the subscores.
113
114         * ARES-6/index.html:
115
116 2017-02-06  Jer Noble  <jer.noble@apple.com>
117
118         Playback stalls when a SourceBuffer append causes frame eviction
119         https://bugs.webkit.org/show_bug.cgi?id=167834
120
121         Reviewed by Eric Carlson.
122
123         Add an in-page performance test measuring the amount of time required
124         to append a large amount of media data to a SourceBuffer, and then to
125         completely remove that data 30s at a time.
126
127         Add a microbenchmark for MediaTime which measures the amount of time
128         required to create a 1M entry std::map and traverse the map 1M times.
129
130         * Media/MSERemoveCodedFrames.html: Added.
131         * Media/media-source-loader.js:
132         (MediaSourceLoader.prototype.get duration):
133         * MediaTime/Configurations/Base.xcconfig: Added.
134         * MediaTime/Configurations/DebugRelease.xcconfig: Added.
135         * MediaTime/Makefile: Added.
136         * MediaTime/MediaTime.xcodeproj/project.pbxproj: Added.
137         * MediaTime/main.cpp: Added.
138         (performTest):
139         (test):
140         (main):
141         * Skipped:
142
143 2017-02-06  Saam Barati  <sbarati@apple.com>
144
145         Make ARES-6 work from the CLI again
146         https://bugs.webkit.org/show_bug.cgi?id=167895
147
148         Reviewed by Michael Saboff.
149
150         * ARES-6/driver.js:
151         (Driver.prototype.readyTrigger):
152         (Driver.prototype.disableTrigger):
153         * ARES-6/stats.js:
154         (Stats.prototype.toString.span.span):
155         (Stats.prototype.toString.return.span):
156         (Stats.prototype.toString):
157
158 2017-02-06  Ryan Haddad  <ryanhaddad@apple.com>
159
160         Skip ARES-6 on performance bots
161         https://bugs.webkit.org/show_bug.cgi?id=167863
162
163         Unreviewed test gardening.
164
165         ES6SampleBench was renamed to ARES-6, so this should be updated in PerformanceTests/Skipped as well.
166
167         * Skipped:
168
169 2017-02-05  Filip Pizlo  <fpizlo@apple.com>
170
171         Change ES6SampleBench into ARES-6
172         https://bugs.webkit.org/show_bug.cgi?id=167863
173
174         Rubber stamped by Saam Barati.
175         
176         This imports changes that Jonathan Davis (jond@apple.com) made to ES6SampleBench to make it look
177         super awesome. Its now called ARES-6.
178
179         * ARES-6: Copied from PerformanceTests/ES6SampleBench.
180         * ARES-6/ARES-6.svg: Added.
181         * ARES-6/driver.js:
182         (Driver):
183         (Driver.prototype.readyTrigger):
184         (Driver.prototype.disableTrigger):
185         (Driver.prototype.start):
186         (Driver.prototype._updateIterations):
187         * ARES-6/glue.js:
188         * ARES-6/index.html:
189         * ARES-6/results.js:
190         (Results.prototype.reportRunning):
191         (Results.prototype.reportDone):
192         * ARES-6/stats.js:
193         (Stats.prototype.toString.return.span):
194         (Stats.prototype.toString.span.span):
195         (Stats.prototype.toString): Deleted.
196         * ARES-6/style.css: Removed.
197         * ARES-6/styles.css: Added.
198         (html):
199         (body):
200         (header,):
201         (p a):
202         (p a:hover):
203         (.flip):
204         (header):
205         (h2,):
206         (p):
207         (.about h2):
208         (.logo):
209         (.start):
210         (.start.ready):
211         (.start.ready:hover):
212         (.start:after):
213         (.start.ready .testrun):
214         (.tests):
215         (.test):
216         (.overall h2,):
217         (.test .indicator):
218         (.test .indicator.running):
219         (@keyframes test-running):
220         (100%):
221         (.score):
222         (.score label):
223         (.score .value):
224         (.score .units):
225         (.score .margin):
226         (.overall):
227         (.overall .score):
228         (.overall .margin):
229         (.testing, .about):
230         (.testing):
231         (.about):
232         (#about:target):
233         (@media only screen and (max-width: 784px)):
234         * ARES-6/swoop.svg: Added.
235         * ES6SampleBench: Removed.
236         * ES6SampleBench/Air: Removed.
237         * ES6SampleBench/Air/README.md: Removed.
238         * ES6SampleBench/Air/airjs-tests.yaml: Removed.
239         * ES6SampleBench/Air/all.js: Removed.
240         * ES6SampleBench/Air/allocate_stack.js: Removed.
241         * ES6SampleBench/Air/arg.js: Removed.
242         * ES6SampleBench/Air/basic_block.js: Removed.
243         * ES6SampleBench/Air/benchmark.js: Removed.
244         * ES6SampleBench/Air/code.js: Removed.
245         * ES6SampleBench/Air/custom.js: Removed.
246         * ES6SampleBench/Air/frequented_block.js: Removed.
247         * ES6SampleBench/Air/insertion_set.js: Removed.
248         * ES6SampleBench/Air/inst.js: Removed.
249         * ES6SampleBench/Air/liveness.js: Removed.
250         * ES6SampleBench/Air/make_dist.sh: Removed.
251         * ES6SampleBench/Air/opcode.js: Removed.
252         * ES6SampleBench/Air/payload-airjs-ACLj8C.js: Removed.
253         * ES6SampleBench/Air/payload-gbemu-executeIteration.js: Removed.
254         * ES6SampleBench/Air/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
255         * ES6SampleBench/Air/payload-typescript-scanIdentifier.js: Removed.
256         * ES6SampleBench/Air/reg.js: Removed.
257         * ES6SampleBench/Air/stack_slot.js: Removed.
258         * ES6SampleBench/Air/stress-test.js: Removed.
259         * ES6SampleBench/Air/strip-hash.rb: Removed.
260         * ES6SampleBench/Air/symbols.js: Removed.
261         * ES6SampleBench/Air/test.html: Removed.
262         * ES6SampleBench/Air/test.js: Removed.
263         * ES6SampleBench/Air/tmp.js: Removed.
264         * ES6SampleBench/Air/tmp_base.js: Removed.
265         * ES6SampleBench/Air/util.js: Removed.
266         * ES6SampleBench/Basic: Removed.
267         * ES6SampleBench/Basic/ast.js: Removed.
268         * ES6SampleBench/Basic/basic-tests.yaml: Removed.
269         * ES6SampleBench/Basic/basic.js: Removed.
270         * ES6SampleBench/Basic/benchmark.js: Removed.
271         * ES6SampleBench/Basic/caseless_map.js: Removed.
272         * ES6SampleBench/Basic/lexer.js: Removed.
273         * ES6SampleBench/Basic/number.js: Removed.
274         * ES6SampleBench/Basic/parser.js: Removed.
275         * ES6SampleBench/Basic/random.js: Removed.
276         * ES6SampleBench/Basic/state.js: Removed.
277         * ES6SampleBench/Basic/stress-test.js: Removed.
278         * ES6SampleBench/Basic/test.html: Removed.
279         * ES6SampleBench/Basic/test.js: Removed.
280         * ES6SampleBench/Basic/util.js: Removed.
281         * ES6SampleBench/air_benchmark.js: Removed.
282         * ES6SampleBench/basic_benchmark.js: Removed.
283         * ES6SampleBench/cli.js: Removed.
284         * ES6SampleBench/driver.js: Removed.
285         * ES6SampleBench/glue.js: Removed.
286         * ES6SampleBench/index.html: Removed.
287         * ES6SampleBench/results.js: Removed.
288         * ES6SampleBench/stats.js: Removed.
289         * ES6SampleBench/style.css: Removed.
290
291 2017-02-05  Zalan Bujtas  <zalan@apple.com>
292
293         Simple line layout: Add performance test with varying string and font-size to minimize width-cache hit.
294         https://bugs.webkit.org/show_bug.cgi?id=167859
295         <rdar://problem/30368070>
296
297         Reviewed by Antti Koivisto.
298
299         This is in preparation for webkit.org/b/167843.
300
301         * Layout/simple-line-layout-with-varying-content.html: Added.
302
303 2017-01-16  Filip Pizlo  <fpizlo@apple.com>
304
305         Add framerate baselines to the splay-detail test
306         https://bugs.webkit.org/show_bug.cgi?id=167659
307
308         Reviewed by Geoffrey Garen.
309
310         * JetStream/Octane2/splay-detail.html:
311
312 2017-01-31  Carlos Alberto Lopez Perez  <clopez@igalia.com>
313
314         [EFL][GTK] Skip some IndexedDB tests that time out on the Perf bots.
315
316         Unreviewed performance test gardening.
317
318         * Skipped:
319
320 2017-01-30  Zalan Bujtas  <zalan@apple.com>
321
322         Simple line layout: Small tweaks to improve performance.
323         https://bugs.webkit.org/show_bug.cgi?id=167611
324         <rdar://problem/30274294>
325
326         Reviewed by Simon Fraser.
327
328         * Layout/simple-line-layout-non-repeating-text.html: Added.
329
330 2017-01-13  Said Abou-Hallawa  <sabouhallawa@apple.com>
331
332         26 MotionMark performance tests failing
333         https://bugs.webkit.org/show_bug.cgi?id=166854
334
335         Reviewed by Ryosuke Niwa.
336         
337         Skip running MotionMark on WebKit perf bots.
338
339         * Skipped:
340
341 2017-01-06  Said Abou-Hallawa  <sabouhallawa@apple.com>
342
343         Rename the directory and the files of MotionMark from Animometer to MotionMark
344         https://bugs.webkit.org/show_bug.cgi?id=166659
345
346         Reviewed by Jon Lee.
347         
348         Rename the directory and the files of the benchmark to its new name.
349
350         * MotionMark/developer.html: Renamed from PerformanceTests/Animometer/developer.html.
351         * MotionMark/index.html: Renamed from PerformanceTests/Animometer/index.html.
352         * MotionMark/resources/debug-runner/d3.min.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/d3.min.js.
353         * MotionMark/resources/debug-runner/graph.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/graph.js.
354         * MotionMark/resources/debug-runner/motionmark.css: Renamed from PerformanceTests/Animometer/resources/debug-runner/animometer.css.
355         * MotionMark/resources/debug-runner/motionmark.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/animometer.js.
356         * MotionMark/resources/debug-runner/tests.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/tests.js.
357         * MotionMark/resources/extensions.js: Renamed from PerformanceTests/Animometer/resources/extensions.js.
358         * MotionMark/resources/runner/benchmark-runner.js: Renamed from PerformanceTests/Animometer/resources/runner/benchmark-runner.js.
359         * MotionMark/resources/runner/crystal.svg: Renamed from PerformanceTests/Animometer/resources/runner/crystal.svg.
360         * MotionMark/resources/runner/lines.svg: Renamed from PerformanceTests/Animometer/resources/runner/lines.svg.
361         * MotionMark/resources/runner/logo.svg: Renamed from PerformanceTests/Animometer/resources/runner/logo.svg.
362         * MotionMark/resources/runner/motionmark.css: Renamed from PerformanceTests/Animometer/resources/runner/animometer.css.
363         * MotionMark/resources/runner/motionmark.js: Renamed from PerformanceTests/Animometer/resources/runner/animometer.js.
364         * MotionMark/resources/runner/tests.js: Renamed from PerformanceTests/Animometer/resources/runner/tests.js.
365         * MotionMark/resources/statistics.js: Renamed from PerformanceTests/Animometer/resources/statistics.js.
366         * MotionMark/resources/strings.js: Renamed from PerformanceTests/Animometer/resources/strings.js.
367         * MotionMark/tests/3d/resources/webgl.js: Renamed from PerformanceTests/Animometer/tests/3d/resources/webgl.js.
368         * MotionMark/tests/3d/webgl.html: Renamed from PerformanceTests/Animometer/tests/3d/webgl.html.
369         * MotionMark/tests/bouncing-particles/bouncing-canvas-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-canvas-images.html.
370         * MotionMark/tests/bouncing-particles/bouncing-canvas-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html.
371         * MotionMark/tests/bouncing-particles/bouncing-css-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-css-images.html.
372         * MotionMark/tests/bouncing-particles/bouncing-css-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-css-shapes.html.
373         * MotionMark/tests/bouncing-particles/bouncing-svg-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-svg-images.html.
374         * MotionMark/tests/bouncing-particles/bouncing-svg-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-svg-shapes.html.
375         * MotionMark/tests/bouncing-particles/bouncing-tagged-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-tagged-images.html.
376         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-images.js.
377         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-particles.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-particles.js.
378         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-shapes.js.
379         * MotionMark/tests/bouncing-particles/resources/bouncing-css-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-css-images.js.
380         * MotionMark/tests/bouncing-particles/resources/bouncing-css-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-css-shapes.js.
381         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-images.js.
382         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-particles.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-particles.js.
383         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-shapes.js.
384         * MotionMark/tests/bouncing-particles/resources/bouncing-tagged-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js.
385         * MotionMark/tests/bouncing-particles/resources/image1.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image1.jpg.
386         * MotionMark/tests/bouncing-particles/resources/image2.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image2.jpg.
387         * MotionMark/tests/bouncing-particles/resources/image3.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image3.jpg.
388         * MotionMark/tests/bouncing-particles/resources/image4.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image4.jpg.
389         * MotionMark/tests/bouncing-particles/resources/image5.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image5.jpg.
390         * MotionMark/tests/dom/compositing-transforms.html: Renamed from PerformanceTests/Animometer/tests/dom/compositing-transforms.html.
391         * MotionMark/tests/dom/focus.html: Renamed from PerformanceTests/Animometer/tests/dom/focus.html.
392         * MotionMark/tests/dom/leaves.html: Renamed from PerformanceTests/Animometer/tests/dom/leaves.html.
393         * MotionMark/tests/dom/particles.html: Renamed from PerformanceTests/Animometer/tests/dom/particles.html.
394         * MotionMark/tests/dom/resources/compositing-transforms.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/compositing-transforms.js.
395         * MotionMark/tests/dom/resources/dom-particles.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/dom-particles.js.
396         * MotionMark/tests/dom/resources/focus.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/focus.js.
397         * MotionMark/tests/dom/resources/leaves.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/leaves.js.
398         * MotionMark/tests/master/canvas-stage.html: Renamed from PerformanceTests/Animometer/tests/master/canvas-stage.html.
399         * MotionMark/tests/master/focus.html: Renamed from PerformanceTests/Animometer/tests/master/focus.html.
400         * MotionMark/tests/master/image-data.html: Renamed from PerformanceTests/Animometer/tests/master/image-data.html.
401         * MotionMark/tests/master/leaves.html: Renamed from PerformanceTests/Animometer/tests/master/leaves.html.
402         * MotionMark/tests/master/multiply.html: Renamed from PerformanceTests/Animometer/tests/master/multiply.html.
403         * MotionMark/tests/master/resources/canvas-stage.js: Renamed from PerformanceTests/Animometer/tests/master/resources/canvas-stage.js.
404         * MotionMark/tests/master/resources/canvas-tests.js: Renamed from PerformanceTests/Animometer/tests/master/resources/canvas-tests.js.
405         * MotionMark/tests/master/resources/compass.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/compass.svg.
406         * MotionMark/tests/master/resources/compass100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/compass100.png.
407         * MotionMark/tests/master/resources/console.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/console.svg.
408         * MotionMark/tests/master/resources/console100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/console100.png.
409         * MotionMark/tests/master/resources/contribute.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/contribute.svg.
410         * MotionMark/tests/master/resources/contribute100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/contribute100.png.
411         * MotionMark/tests/master/resources/debugger.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/debugger.svg.
412         * MotionMark/tests/master/resources/debugger100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/debugger100.png.
413         * MotionMark/tests/master/resources/focus.js: Renamed from PerformanceTests/Animometer/tests/master/resources/focus.js.
414         * MotionMark/tests/master/resources/image-data.js: Renamed from PerformanceTests/Animometer/tests/master/resources/image-data.js.
415         * MotionMark/tests/master/resources/inspector.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/inspector.svg.
416         * MotionMark/tests/master/resources/inspector100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/inspector100.png.
417         * MotionMark/tests/master/resources/layout.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/layout.svg.
418         * MotionMark/tests/master/resources/layout100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/layout100.png.
419         * MotionMark/tests/master/resources/leaves.js: Renamed from PerformanceTests/Animometer/tests/master/resources/leaves.js.
420         * MotionMark/tests/master/resources/multiply.js: Renamed from PerformanceTests/Animometer/tests/master/resources/multiply.js.
421         * MotionMark/tests/master/resources/particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/particles.js.
422         * MotionMark/tests/master/resources/performance.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/performance.svg.
423         * MotionMark/tests/master/resources/performance100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/performance100.png.
424         * MotionMark/tests/master/resources/script.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/script.svg.
425         * MotionMark/tests/master/resources/script100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/script100.png.
426         * MotionMark/tests/master/resources/shortcuts.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/shortcuts.svg.
427         * MotionMark/tests/master/resources/shortcuts100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/shortcuts100.png.
428         * MotionMark/tests/master/resources/standards.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/standards.svg.
429         * MotionMark/tests/master/resources/standards100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/standards100.png.
430         * MotionMark/tests/master/resources/storage.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/storage.svg.
431         * MotionMark/tests/master/resources/storage100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/storage100.png.
432         * MotionMark/tests/master/resources/styles.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/styles.svg.
433         * MotionMark/tests/master/resources/styles100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/styles100.png.
434         * MotionMark/tests/master/resources/svg-particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/svg-particles.js.
435         * MotionMark/tests/master/resources/text.js: Renamed from PerformanceTests/Animometer/tests/master/resources/text.js.
436         * MotionMark/tests/master/resources/timeline.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/timeline.svg.
437         * MotionMark/tests/master/resources/timeline100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/timeline100.png.
438         * MotionMark/tests/master/svg-particles.html: Renamed from PerformanceTests/Animometer/tests/master/svg-particles.html.
439         * MotionMark/tests/master/text.html: Renamed from PerformanceTests/Animometer/tests/master/text.html.
440         * MotionMark/tests/resources/main.js: Renamed from PerformanceTests/Animometer/tests/resources/main.js.
441         * MotionMark/tests/resources/math.js: Renamed from PerformanceTests/Animometer/tests/resources/math.js.
442         * MotionMark/tests/resources/stage.css: Renamed from PerformanceTests/Animometer/tests/resources/stage.css.
443         * MotionMark/tests/resources/star.svg: Renamed from PerformanceTests/Animometer/tests/resources/star.svg.
444         * MotionMark/tests/resources/yin-yang.png: Renamed from PerformanceTests/Animometer/tests/resources/yin-yang.png.
445         * MotionMark/tests/resources/yin-yang.svg: Renamed from PerformanceTests/Animometer/tests/resources/yin-yang.svg.
446         * MotionMark/tests/simple/resources/simple-canvas-paths.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/simple-canvas-paths.js.
447         * MotionMark/tests/simple/resources/simple-canvas.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/simple-canvas.js.
448         * MotionMark/tests/simple/resources/tiled-canvas-image.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/tiled-canvas-image.js.
449         * MotionMark/tests/simple/simple-canvas-paths.html: Renamed from PerformanceTests/Animometer/tests/simple/simple-canvas-paths.html.
450         * MotionMark/tests/simple/tiled-canvas-image.html: Renamed from PerformanceTests/Animometer/tests/simple/tiled-canvas-image.html.
451         * MotionMark/tests/template/resources/template-canvas.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-canvas.js.
452         * MotionMark/tests/template/resources/template-css.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-css.js.
453         * MotionMark/tests/template/resources/template-svg.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-svg.js.
454         * MotionMark/tests/template/template-canvas.html: Renamed from PerformanceTests/Animometer/tests/template/template-canvas.html.
455         * MotionMark/tests/template/template-css.html: Renamed from PerformanceTests/Animometer/tests/template/template-css.html.
456         * MotionMark/tests/template/template-svg.html: Renamed from PerformanceTests/Animometer/tests/template/template-svg.html.
457
458 2016-12-13  Brady Eidson  <beidson@apple.com>
459
460         Add a PerfTest targeting IDBObjectStore.get().
461         https://bugs.webkit.org/show_bug.cgi?id=165816
462
463         Reviewed by Alex Christensen.
464
465         * IndexedDB/objectstore-get.html: Added.
466
467 2016-12-13  Brady Eidson  <beidson@apple.com>
468
469         Add a PerfTest targeting IDBIndex.get().
470         https://bugs.webkit.org/show_bug.cgi?id=165803
471
472         Reviewed by Alex Christensen.
473
474         * IndexedDB/index-get.html: Added.
475
476 2016-12-12  Brady Eidson  <beidson@apple.com>
477
478         More IndexedDB perf tests.
479         https://bugs.webkit.org/show_bug.cgi?id=165634
480
481         Reviewed by Sam Weinig.
482
483         * IndexedDB/index-multientry.html:
484         * IndexedDB/large-number-of-inserts-responsiveness.html: Added.
485         * IndexedDB/large-number-of-inserts.html: Added.
486         * IndexedDB/objectstore-cursor.html: Added. Runtime cut in 1/4th compared to r209672.
487
488         * resources/runner.js:
489
490 2016-12-12  Commit Queue  <commit-queue@webkit.org>
491
492         Unreviewed, rolling out r209672.
493         https://bugs.webkit.org/show_bug.cgi?id=165766
494
495         IndexedDB/objectstore-cursor.html perf test always times out
496         (Requested by ap on #webkit).
497
498         Reverted changeset:
499
500         "More IndexedDB perf tests."
501         https://bugs.webkit.org/show_bug.cgi?id=165634
502         http://trac.webkit.org/changeset/209672
503
504 2016-12-10  Brady Eidson  <beidson@apple.com>
505
506         More IndexedDB perf tests.
507         https://bugs.webkit.org/show_bug.cgi?id=165634
508
509         Reviewed by Sam Weinig.
510
511         * IndexedDB/index-multientry.html:
512         * IndexedDB/large-number-of-inserts-responsiveness.html: Added.
513         * IndexedDB/large-number-of-inserts.html: Added.
514         * IndexedDB/objectstore-cursor.html: Added.
515         
516         * resources/runner.js: Add "track responsiveness" functionality to PerfTestRunner.
517
518 2016-12-06  Filip Pizlo  <fpizlo@apple.com>
519
520         Concurrent GC should be stable enough to land enabled
521         https://bugs.webkit.org/show_bug.cgi?id=164990
522
523         Reviewed by Geoffrey Garen.
524         
525         Made CDjs more configurable and refined the "large.js" configuration. I was using that one and
526         the new "long.js" configuration to tune concurrent eden GCs.
527         
528         Added a new way of running Splay in browser, which using chartjs to plot the execution times of
529         2000 iterations. This includes the minified chartjs.
530
531         * JetStream/Octane2/splay-detail.html: Added.
532         * JetStream/cdjs/benchmark.js:
533         (benchmarkImpl):
534         (benchmark):
535         * JetStream/cdjs/long.js: Added.
536
537 2016-12-07  Brady Eidson  <beidson@apple.com>
538
539         Add IDB perf tests stressing key size.
540         https://bugs.webkit.org/show_bug.cgi?id=165567
541
542         Reviewed by Alex Christensen.
543
544         * IndexedDB/large-array-keys.html: Added.
545         * IndexedDB/large-binary-keys.html: Added.
546         * IndexedDB/large-string-keys.html: Added.
547
548 2016-12-05  Brady Eidson  <beidson@apple.com>
549
550         Add an IndexedDB perf test to PerformanceTests.
551         https://bugs.webkit.org/show_bug.cgi?id=165430
552
553         Reviewed by Alex Christensen.
554
555         * IndexedDB/index-multientry.html: Added.
556
557 2016-12-02  Filip Pizlo  <fpizlo@apple.com>
558
559         ES6SampleBench should report an average for Steady State so that all of the numbers are comparable
560         https://bugs.webkit.org/show_bug.cgi?id=165325
561
562         Reviewed by Saam Barati.
563
564         This makes all of the numbers that ES6SampleBench reports comparable to each other: they all speak of the time
565         it took to run an iteration of something.
566
567         * ES6SampleBench/results.js:
568         (Results.prototype.reportResult):
569
570 2016-11-18  Jer Noble  <jer.noble@apple.com>
571
572         Add some performance tests for various aspects of media loading and playback.
573         https://bugs.webkit.org/show_bug.cgi?id=164977
574
575         Reviewed by Darin Adler.
576
577         Add new tests for a few aspects of media playback; namely: how quickly  media elements fire
578         the "canplaythrough" event when all data is immediately available, how quickly HLS playback
579         switches resolutions, and at what playback rate MSE-backed video can play without dropping
580         frames. Skip these tests by default, as they're not runnable on all ports, and they require
581         a webserver (run-webkit-httpd) to complete on macOS.
582
583         * Media/HLSCanPlayThrough.html: Added.
584         * Media/HLSGearChange.html: Added.
585         * Media/MSECanPlayThrough.html: Added.
586         * Media/MSEPlaybackRate.html: Added.
587         * Media/StandardCanPlayThrough.html: Added.
588         * Media/hls/1080p/iframe_index.m3u8: Added.
589         * Media/hls/1080p/prog_index.m3u8: Added.
590         * Media/hls/1080p/test.ts: Added.
591         * Media/hls/480p/iframe_index.m3u8: Added.
592         * Media/hls/480p/prog_index.m3u8: Added.
593         * Media/hls/480p/test.ts: Added.
594         * Media/hls/720p/iframe_index.m3u8: Added.
595         * Media/hls/720p/prog_index.m3u8: Added.
596         * Media/hls/720p/test.ts: Added.
597         * Media/hls/720p/test.ts.back: Added.
598         * Media/hls/index.m3u8: Added.
599         * Media/media-source-loader.js: Added.
600         (MediaSourceLoader):
601         (MediaSourceLoader.prototype.loadManifest):
602         (MediaSourceLoader.prototype.loadManifestSucceeded):
603         (MediaSourceLoader.prototype.loadManifestFailed):
604         (MediaSourceLoader.prototype.loadMediaData):
605         (MediaSourceLoader.prototype.loadMediaDataSucceeded):
606         (MediaSourceLoader.prototype.loadMediaDataFailed):
607         (MediaSourceLoader.prototype.get type):
608         (MediaSourceLoader.prototype.get duration):
609         (MediaSourceLoader.prototype.get initSegment):
610         (MediaSourceLoader.prototype.get mediaSegmentsLength):
611         (MediaSourceLoader.prototype.mediaSegments):
612         (MediaSourceLoader.prototype.get everyMediaSegment):
613         * Media/test-fragmented-video.json: Added.
614         * Media/test-fragmented-video.mp4: Added.
615         * Media/test.mp4: Added.
616         * Skipped:
617
618 2016-11-18  Filip Pizlo  <fpizlo@apple.com>
619
620         Concurrent GC should be able to run splay in debug mode and earley/raytrace in release mode with no perf regression
621         https://bugs.webkit.org/show_bug.cgi?id=164282
622
623         Reviewed by Geoffrey Garen and Oliver Hunt.
624         
625         CDjs is a fun benchmark for stressing concurrent GCs, but to really give the GC a good
626         workout you need to increase the amount of work that the test does. This adds a second
627         configuration of the benchmark that has more aircraft. It uses much more memory and causes us
628         to do more GCs and those GCs take longer.
629
630         * JetStream/cdjs/benchmark.js:
631         (benchmarkImpl):
632         (benchmark):
633         * JetStream/cdjs/large.js: Added.
634
635 2016-11-14  Filip Pizlo  <fpizlo@apple.com>
636
637         Unreviewed, revert unintended change.
638
639         * ES6SampleBench/Air/benchmark.js:
640
641 2016-10-03  Saam Barati  <sbarati@apple.com>
642
643         MapHash should speculate on the type of its child node
644         https://bugs.webkit.org/show_bug.cgi?id=161922
645
646         Reviewed by Filip Pizlo.
647
648         I gave the main generator a name so it's easier to see what
649         it is when using the sampling profiler.
650
651         * ES6SampleBench/Basic/ast.js:
652         (Basic.Program):
653
654 2016-09-19  Sergio Villar Senin  <svillar@igalia.com>
655
656         [css-grid] Remove the x2 computation of row sizes with indefinite heights
657         https://bugs.webkit.org/show_bug.cgi?id=162150
658
659         Reviewed by Darin Adler.
660
661         Added a new test case which checks the layout performance of grids inside other grids, i.e,
662         grids acting both as grid container and grid item.
663
664         * Layout/nested-grid.html: Added.
665
666 2016-09-09  Simon Fraser  <simon.fraser@apple.com>
667
668         Perf test Animation/css-accelerated-animation.html failing
669         https://bugs.webkit.org/show_bug.cgi?id=161795
670
671         Skip Animation/css-accelerated-animation.html since this only works as a MobileSafari
672         perf test.
673
674         * Skipped:
675
676 2016-09-08  Simon Fraser  <simon.fraser@apple.com>
677
678         Add a content-animation test with accelerated CSS animations
679         https://bugs.webkit.org/show_bug.cgi?id=161776
680
681         Reviewed by Dean Jackson.
682
683         Add a test that measures the frame rate of accelerated CSS animations.
684
685         This is like css-animation.html (which animates 'left' and 'top') but animates
686         transforms instead. In order to get animation in X and Y we make two nested elements
687         and set transformX() on one, and transformY() on the other.
688
689         * Animation/css-accelerated-animation.html: Added.
690
691 2016-08-23  Saam Barati  <sbarati@apple.com>
692
693         It should be easy to run ES6SampleBench from the jsc shell
694         https://bugs.webkit.org/show_bug.cgi?id=161085
695
696         Reviewed by Yusuke Suzuki.
697
698         This patch makes ES6 sample bench runnable through the `jsc` shell.
699         To do that, you need to be in the PerformanceTests/ES6SampleBench
700         directory and run `jsc cli.js`. To make this work, the benchmark
701         is now aware of being run in two modes: via the browser, and via
702         the shell. Each entry point will set a variable `isInBrowser`,
703         and the benchmark will do different things based on if that
704         variable is true or false.
705
706         * ES6SampleBench/Air/benchmark.js:
707         * ES6SampleBench/Air/stress-test.js:
708         * ES6SampleBench/Basic/benchmark.js:
709         (runBenchmark):
710         * ES6SampleBench/Basic/stress-test.js:
711         * ES6SampleBench/air_benchmark.js:
712         * ES6SampleBench/basic_benchmark.js:
713         * ES6SampleBench/cli.js: Added.
714         (return.doRun):
715         (makeBenchmarkRunner):
716         * ES6SampleBench/driver.js:
717         (Driver):
718         (Driver.prototype._recomputeSummary):
719         (Driver.prototype._iterate.window.setTimeout):
720         (Driver.prototype._iterate):
721         (Driver.prototype._updateIterations):
722         * ES6SampleBench/glue.js:
723         (reportResult):
724         * ES6SampleBench/index.html:
725         * ES6SampleBench/results.js:
726         (Results):
727         (Results.prototype.reportRunning):
728         (Results.prototype.reportDone):
729         (Results.prototype.reportError):
730         * ES6SampleBench/stats.js:
731         (Stats):
732         (Stats.prototype.toString):
733         (Stats.prototype._update):
734
735 2016-08-01  Said Abou-Hallawa  <sabouhallawa@apple.com>
736
737         Add an option to run the MotionMark tests with classic tile size (512x512)
738         https://bugs.webkit.org/show_bug.cgi?id=160371
739
740         Reviewed by Darin Adler.
741
742         Setting the body to large size (3000x3000) and overflow to scroll forces
743         the classic tile size (512x512). This option does not affect the canvas
744         tests because the canvas is displayed on one tile regardless of its size.
745
746         * Animometer/developer.html:
747         * Animometer/resources/debug-runner/animometer.css:
748         (body.showing-test-container.tiles-big):
749         (body.showing-test-container.tiles-classic):
750         * Animometer/resources/debug-runner/animometer.js:
751         * Animometer/resources/runner/animometer.js:
752         (window.benchmarkController.startBenchmark):
753
754 2016-06-29  Filip Pizlo  <fpizlo@apple.com>
755
756         Generators violate bytecode liveness validation
757         https://bugs.webkit.org/show_bug.cgi?id=159279
758
759         Reviewed by Yusuke Suzuki.
760
761         Add Basic to our test harness.
762
763         Also made some cosmetic changes to the benchmark harness.
764
765         * ES6SampleBench/Basic/basic-tests.yaml: Added.
766         * ES6SampleBench/Basic/stress-test.js: Added.
767         (preciseTime):
768         * ES6SampleBench/driver.js:
769         (Driver):
770         (Driver.prototype.start):
771         (Driver.prototype.reportError):
772         * ES6SampleBench/glue.js:
773         * ES6SampleBench/index.html:
774
775 2016-06-28  Filip Pizlo  <fpizlo@apple.com>
776
777         ES6SampleBench should have a harness
778         https://bugs.webkit.org/show_bug.cgi?id=159246
779
780         Reviewed by Saam Barati.
781         
782         This adds a simple web harness for ES6SampleBench. It runs Air and Basic 10 times for 200
783         iterations each and reports three metrics:
784         
785         First iteration: the time it takes for the first iteration to run. This is the first
786         iteration after the benchmark is loaded into the iframe, so it's representative of what
787         would happen if one of these workloads only ran for a short time.
788         
789         Worst 2%: of the last 199 iterations, the average of the worst 2% iterations. If code like
790         any of these workloads was used in an important event handler, you'd want that code to run
791         well in the worst case in addition to having great throughput.
792         
793         Steady state: the total of the last 199 iterations. This is representative of what would
794         happen if you ran code like this for a long time.
795         
796         The total score is the geomean of the firstIteration/worstCase/steadyState numbers of the
797         two benchmarks.
798         
799         The harness does statistics using Student's T-distribution confidence intervals.
800
801         * ES6SampleBench/Basic/benchmark.js:
802         (Benchmark):
803         * ES6SampleBench/air_benchmark.js: Added.
804         * ES6SampleBench/basic_benchmark.js: Added.
805         * ES6SampleBench/driver.js: Added.
806         (Driver):
807         (Driver.prototype.addBenchmark):
808         (Driver.prototype.start):
809         (Driver.prototype.reportResult):
810         (Driver.prototype.reportError):
811         (Driver.prototype._recomputeSummary.Geomean):
812         (Driver.prototype._recomputeSummary.Geomean.prototype.add):
813         (Driver.prototype._recomputeSummary.Geomean.prototype.get result):
814         (Driver.prototype._recomputeSummary):
815         (Driver.prototype._iterate):
816         (Driver.prototype._updateIterations):
817         * ES6SampleBench/glue.js: Added.
818         * ES6SampleBench/index.html: Added.
819         * ES6SampleBench/results.js: Added.
820         (Results):
821         (Results.prototype.get benchmark):
822         (Results.prototype.reset):
823         (Results.prototype.reportRunning):
824         (Results.prototype.reportDone):
825         (Results.prototype.reportResult.averageAbovePercentile):
826         (Results.prototype.reportResult):
827         (Results.prototype.reportError):
828         * ES6SampleBench/stats.js: Added.
829         (Stats):
830         (Stats.prototype.reset):
831         (Stats.prototype.add):
832         (Stats.prototype.get numIterations):
833         (Stats.prototype.valueForIteration):
834         (Stats.get result.tDist):
835         (Stats.prototype.get result):
836         (Stats.prototype.toString):
837         (Stats.prototype._update):
838         * ES6SampleBench/style.css: Added.
839         (body):
840         (body, th, tr):
841         (h1, h2, h3, h4):
842         (h1):
843         (h2):
844         (h3):
845         (hr):
846         (address):
847         (img):
848         (.underline):
849         (ol.loweralpha):
850         (ol.upperalpha):
851         (ol.lowerroman):
852         (ol.upperroman):
853         (ol.arabic):
854         (.banner-link:link, .banner-link:visited):
855         (:link, :visited):
856         (h4 :link, h4 :visited, h5 :link, h5 :visited, h6 :link, h6 :visited):
857         (.anchor:link, .anchor:visited):
858         (* > .anchor:link, * > .anchor:visited):
859         (span:hover .anchor):
860         (a.forbidden, span.forbidden):
861         (a.missing:hover):
862         (a.closed:link, a.closed:visited, span.closed):
863         (pre):
864         (div.code):
865         (div.code pre):
866         (dt):
867         (dd):
868         (dd:last-child):
869         (.site-logo):
870         (.site-logo .tagline):
871         (table):
872         (#contents):
873         (p):
874         (p:last-child):
875         (th):
876         (td):
877
878 2016-06-28  Jon Lee  <jonlee@apple.com>
879
880         Update focus test
881         https://bugs.webkit.org/show_bug.cgi?id=159242
882         rdar://problem/27070007
883
884         Reviewed by Dean Jackson.
885         Provisionally reviewed by Said Abou-Hallawa.
886
887         Move previous test to dom suite, and update the test for better reporting of frame rate, although
888         it uses a different rendering path.
889
890         * Animometer/resources/debug-runner/tests.js: Add to dom suite.
891         * Animometer/tests/dom/focus.html: Copied from PerformanceTests/Animometer/tests/master/focus.html.
892         * Animometer/tests/dom/resources/focus.js: Copied from PerformanceTests/Animometer/tests/master/resources/focus.js.
893         * Animometer/tests/master/focus.html: Remove center element.
894         * Animometer/tests/master/resources/focus.js: Use narrower size range with smaller particles. Remove the
895         container elements. Inline getBlurValue and getOpacityValue since they are only called once.
896
897 2016-06-28  Filip Pizlo  <fpizlo@apple.com>
898
899         Move Air.js and Basic into ES6SampleBench
900
901         Rubber stamped by Geoffrey Garen.
902         
903         We want to group our ES6 benchmarks together, and eventually, we want to give them a
904         common harness.
905
906         * Air.js: Removed.
907         * Air.js/README.md: Removed.
908         * Air.js/airjs-tests.yaml: Removed.
909         * Air.js/all.js: Removed.
910         * Air.js/allocate_stack.js: Removed.
911         * Air.js/arg.js: Removed.
912         * Air.js/basic_block.js: Removed.
913         * Air.js/benchmark.js: Removed.
914         * Air.js/code.js: Removed.
915         * Air.js/custom.js: Removed.
916         * Air.js/frequented_block.js: Removed.
917         * Air.js/insertion_set.js: Removed.
918         * Air.js/inst.js: Removed.
919         * Air.js/liveness.js: Removed.
920         * Air.js/make_dist.sh: Removed.
921         * Air.js/opcode.js: Removed.
922         * Air.js/payload-airjs-ACLj8C.js: Removed.
923         * Air.js/payload-gbemu-executeIteration.js: Removed.
924         * Air.js/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
925         * Air.js/payload-typescript-scanIdentifier.js: Removed.
926         * Air.js/reg.js: Removed.
927         * Air.js/stack_slot.js: Removed.
928         * Air.js/stress-test.js: Removed.
929         * Air.js/strip-hash.rb: Removed.
930         * Air.js/symbols.js: Removed.
931         * Air.js/test.html: Removed.
932         * Air.js/test.js: Removed.
933         * Air.js/tmp.js: Removed.
934         * Air.js/tmp_base.js: Removed.
935         * Air.js/util.js: Removed.
936         * Basic: Removed.
937         * Basic/ast.js: Removed.
938         * Basic/basic.js: Removed.
939         * Basic/benchmark.js: Removed.
940         * Basic/caseless_map.js: Removed.
941         * Basic/lexer.js: Removed.
942         * Basic/number.js: Removed.
943         * Basic/parser.js: Removed.
944         * Basic/random.js: Removed.
945         * Basic/state.js: Removed.
946         * Basic/test.html: Removed.
947         * Basic/test.js: Removed.
948         * Basic/util.js: Removed.
949         * ES6SampleBench: Added.
950         * ES6SampleBench/Air: Copied from PerformanceTests/Air.js.
951         * ES6SampleBench/Basic: Copied from PerformanceTests/Basic.
952         * Skipped:
953
954 2016-06-24  Filip Pizlo  <fpizlo@apple.com>
955
956         Add a ES6 generator benchmark
957         https://bugs.webkit.org/show_bug.cgi?id=159101
958
959         Rubber stamped by Keith Miller.
960         
961         This adds a Basic interpreter loosely based on ECMA-55:
962         
963         http://www.ecma-international.org/publications/files/ECMA-ST-WITHDRAWN/ECMA-55,%201st%20Edition,%20January%201978.pdf
964         
965         It includes a lexer that is a generator, a parser that uses regular expressions, and an AST
966         walk interpreter where the walking functions for statements are generators that call each
967         other with yield*. This enables the interpreter to look like an AST walk even though it can
968         yield at INPUT and PRINT statements.
969         
970         This also uses for-of, classes, Map, and WeakMap. It also uses deprecated-but-awesome RegExp
971         features like RegExp.lastMatch and RegExp.rightContext. I did it that way because this is
972         how I've always written lexers in dynamic languages; see offlineasm's lex() method in
973         parser.rb for example.
974         
975         The benchmark runs a handful of simple Basic programs. The longest-running one computes
976         prime numbers.
977         
978         Includes a command-line and web harness. On my machine it runs in 2-3 seconds.
979
980         * Basic: Added.
981         * Basic/ast.js: Added.
982         (Basic.NumberApply):
983         (Basic.Variable):
984         (Basic.Const):
985         (Basic.NumberPow):
986         (Basic.NumberMul):
987         (Basic.NumberDiv):
988         (Basic.NumberNeg):
989         (Basic.NumberAdd):
990         (Basic.NumberSub):
991         (Basic.StringVar):
992         (Basic.Equals):
993         (Basic.NotEquals):
994         (Basic.LessThan):
995         (Basic.GreaterThan):
996         (Basic.LessEqual):
997         (Basic.GreaterEqual):
998         (Basic.GoTo):
999         (Basic.GoSub):
1000         (Basic.Def):
1001         (Basic.Let):
1002         (Basic.If):
1003         (Basic.Return):
1004         (Basic.Stop):
1005         (Basic.On):
1006         (sideState.shouldStop):
1007         (Basic.For):
1008         (Basic.Next):
1009         (Basic.Print):
1010         (Basic.Input):
1011         (Basic.Read):
1012         (Basic.Restore):
1013         (Basic.Dim):
1014         (Basic.Randomize):
1015         (Basic.End):
1016         (Basic.Program):
1017         * Basic/basic.js: Added.
1018         (prepare):
1019         (simulate):
1020         * Basic/benchmark.js: Added.
1021         (Benchmark):
1022         (Benchmark.prototype.runIteration.expect):
1023         (Benchmark.prototype.runIteration):
1024         (runBenchmark):
1025         * Basic/caseless_map.js: Added.
1026         (CaselessMap):
1027         * Basic/lexer.js: Added.
1028         (lex.consumeWhitespace):
1029         (lex.consume):
1030         (lex):
1031         * Basic/lexer_test.js: Added.
1032         * Basic/number.js: Added.
1033         (NumberValue):
1034         (NumberValue.prototype.get value):
1035         (NumberValue.prototype.apply):
1036         (NumberValue.prototype.leftApply):
1037         (NumberValue.prototype.assign):
1038         (NumberArray.):
1039         (NumberArray):
1040         (NumberArray.prototype.apply):
1041         (NumberArray.prototype.leftApply):
1042         (NumberFunction):
1043         (NumberFunction.prototype.apply):
1044         (NumberFunction.prototype.leftApply):
1045         (NativeFunction):
1046         (NativeFunction.prototype.apply):
1047         (NativeFunction.prototype.leftApply):
1048         * Basic/parser.js: Added.
1049         (parse):
1050         (parse.pushToken):
1051         (parse.peekToken):
1052         (parse.consumeKind):
1053         (parse.consumeToken):
1054         (parse.parseVariable):
1055         (parse.parseNumericExpression.parsePrimary):
1056         (parse.parseNumericExpression.parseFactor):
1057         (parse.parseNumericExpression.parseTerm):
1058         (parse.parseNumericExpression):
1059         (parse.parseConstant):
1060         (parse.parseStringExpression):
1061         (parse.isStringExpression):
1062         (parse.parseRelationalExpression):
1063         (parse.parseNonNegativeInteger):
1064         (parse.parseGoToStatement):
1065         (parse.parseGoSubStatement):
1066         (parse.parseStatement):
1067         (parse.parseStatements):
1068         * Basic/random.js: Added.
1069         (createRNG):
1070         (createRNGWithFixedSeed):
1071         (createRNGWithRandomSeed):
1072         * Basic/state.js: Added.
1073         (State):
1074         (State.prototype.getValue):
1075         (State.prototype.getSideState):
1076         (State.prototype.abort):
1077         (State.prototype.validate):
1078         * Basic/test.html: Added.
1079         * Basic/test.js: Added.
1080         * Basic/util.js: Added.
1081         (this.performance.performance.now.currentTime):
1082         (else.this.preciseTime.currentTime):
1083         (else.currentTime):
1084         * Skipped: Make sure that we don't run Basic yet.
1085
1086 2016-06-21  Jon Lee  <jonlee@apple.com>
1087
1088         Update canvas size when benchmark begins
1089         https://bugs.webkit.org/show_bug.cgi?id=159010
1090
1091         Reviewed by Dean Jackson.
1092
1093         Previously we would evaluate the media queries and assign the canvas size to the test
1094         on body.onload. Instead, do it once the user starts the benchmark.
1095
1096         * Animometer/resources/debug-runner/animometer.js: Move benchmarkController.determineCanvasSize()
1097         to benchmarkController._startBenchmark.
1098         * Animometer/resources/runner/animometer.js: Ditto.
1099
1100 2016-06-21  Jon Lee  <jonlee@apple.com>
1101
1102         Add new timestamp option
1103         https://bugs.webkit.org/show_bug.cgi?id=159006
1104
1105         Reviewed by Dean Jackson.
1106
1107         Add a new option to take timestamps from the rAF callback.
1108
1109         * Animometer/developer.html: Added option.
1110         * Animometer/tests/resources/main.js:
1111         (Benchmark): Fall back to using the rAF timestamp if performance.now()
1112         is not available.
1113         (Benchmark._animateLoop): Update how timestamp is set. Prefer to use
1114         local var instead of accessing private var.
1115
1116 2016-06-21  Jon Lee  <jonlee@apple.com>
1117
1118         Improvements to Animometer benchmark
1119         https://bugs.webkit.org/show_bug.cgi?id=157738
1120
1121         Reviewed by Dean Jackson.
1122         Provisionally reviewed by Said Abou-Hallawa.
1123
1124         Update tests.
1125
1126         * Animometer/tests/master/text.html: Ensure only three text sizes for
1127         the three canvases.
1128         * Animometer/tests/master/focus.html: Reduce the text size to fit with smaller
1129         particle sizes.
1130         * Animometer/tests/master/resources/focus.js: Remove the quadratic distribution for
1131         particle sizes, and make it linear. Reduce the size variance. Shuffle the math to
1132         reduce some calculations per frame. Fix the placement of the particles which might
1133         otherwise be culled.
1134         * Animometer/tests/master/resources/image-data.js: Reduce the particle size
1135         to encourage larger scores.
1136
1137 2016-06-21  Jon Lee  <jonlee@apple.com>
1138
1139         Improvements to Animometer benchmark
1140         https://bugs.webkit.org/show_bug.cgi?id=157738
1141
1142         Reviewed by Dean Jackson.
1143         Provisionally reviewed by Said Abou-Hallawa.
1144
1145         Include confidence interval for the final score, and store the canvas
1146         size in the serialization so that it is accurately shown in results.
1147
1148         * Animometer/developer.html: Add a "confidence" div.
1149         * Animometer/index.html: Ditto. Convert "mean" to "confidence".
1150         * Animometer/resources/debug-runner/animometer.js: Look at options, and
1151         if the configuration is included, update the body class based on it
1152         (similar to what we do when we first load the page). That way, if you
1153         drag-and-drop previous results in, that configuration is reflected in
1154         the dashboard. Show the full confidence interval.
1155
1156         * Animometer/resources/debug-runner/animometer.css:
1157         * Animometer/resources/debug-runner/animometer.js: Style update.
1158         * Animometer/resources/runner/animometer.css:
1159
1160         * Animometer/resources/runner/animometer.js:
1161         (_processData): Propagate the confidence interval values out and calculate
1162         the lower and upper bounds. For now, shortcut the aggregate calculation,
1163         since we only go through one iteration.
1164         (this._processData.calculateScore): Propagate the confidence interval out
1165         to be next to the score. Depending on the controller these values are
1166         calculated differently.
1167         (this._processData._getResultsProperty): Convenience function.
1168         (this._processData.get score): Refactor.
1169         (this._processData.get scoreLowerBound): Get the aggregate lower bound.
1170         (this._processData.get scoreUpperBound): Get the aggregate upper bound.
1171         (window.sectionsManager.setSectionScore):
1172         (window.benchmarkController._startBenchmark): When the benchmark starts, note
1173         the canvas size and add it to options. That way it will be automatically be
1174         serialized.
1175         (window.benchmarkController.showResults): Include the maximum deviation
1176         percentage.
1177         * Animometer/resources/runner/lines.svg: Make the background line up with the
1178         skew.
1179         * Animometer/resources/runner/tests.js:
1180         (Headers.details.text): Refactor.
1181         * Animometer/resources/statistics.js:
1182         (Statistics.largestDeviationPercentage): Convenience function to calculate
1183         the deviation percentage on either end and return the largest deviation.
1184         * Animometer/resources/strings.js:
1185
1186         Allow specifying a regression profile to use instead of taking the one
1187         with the lowest error.
1188
1189         Address an issue in the focus test when the regression calculated ends
1190         up overestimating the change point, causing a cascade of tougher
1191         ramps. The reason behind this is that at max complexity of an initial
1192         ramp, the frame length is very high, and it influences the second
1193         segment of the piecewise regression strongly, causing it to be very
1194         steep. As a result, the first segment, expected to be flat, ends up
1195         covering a higher range of complexity. That makes the change point
1196         much higher than it should be. To avoid this, we will add a sanity
1197         check on the maximum value of the ramp. If the regression's projected
1198         value at the maximum complexity of the current ramp is very slow (less
1199         than 20 fps), 1) reduce the maximum complexity by 20%, and 2) do not
1200         include the regression's change point in the change point estimator.
1201         That estimator is used as the midpoint of the next ramp, and including
1202         the change point from a poor regression can bake in the error. The
1203         controller already knows how to adjust for ramps that are too easy for
1204         the system.
1205
1206         * Animometer/resources/runner/animometer.js:
1207         (this._processData.findRegression): Takes a preferred profile and gives that to
1208         Regression.
1209         (this._processData.calculateScore): With the ramp controller, take the profile
1210         of the ramp that was used the most when calculating the ramp's regression. That
1211         profile is what is used for the test's score.
1212         * Animometer/resources/statistics.js:
1213         (Regression.Utilities.createClass): Update to take an options object which can
1214         specify a profile to calculate with. Otherwise it will continue to use both and
1215         select the one with the lower error.
1216         (_calculateRegression): Fix an issue where we claim 0 error if the regression
1217         calculation fails due to divide-by-zero. Instead reject that regression calculation
1218         by giving it Number.MAX_VALUE.
1219         * Animometer/resources/strings.js: New strings for marking a regression as flat
1220         or slope.
1221         * Animometer/tests/resources/main.js:
1222         (RampController): Rename the thresholds for clarity. Add a threshold that, if
1223         exceeded, will lower the maximum complexity of the next ramp.
1224         (tune): Relax the cdf check to consider whether the interval definitely falls in
1225         the desired frame length threshold.
1226         (processSamples): Include the profile in the ramp.
1227
1228         Update ramp controller test. Increase the length of the test to 30 seconds, and extend
1229         the interval to 120 ms during sampling. Improve the estimation of the ramp parameters.
1230
1231         * Animometer/developer.html: Change default to 30 seconds, and don't show the progress bar
1232         by default.
1233         * Animometer/resources/runner/animometer.js: Change default to 30 seconds.
1234         * Animometer/tests/resources/main.js: A number of improvements to the ramp controller, in
1235         the order in which they appear in the patch:
1236
1237         - With a longer test length use longer ramps with longer intervals to get more data at each
1238         complexity. Keep the 100 ms interval length during the ramp up phase since we don't need to
1239         spend more time there to find the right order of magnitude, but increase it during the
1240         ramps to 120 ms.
1241         - The ramp linearly interpolates the complexity to render based on its timestamp, but it would
1242         never sample the minimum complexity. Instead of lerping max to min complexity from time
1243         0 to t where t is the ramp length, instead lerp from 0 to (t - intervalSampleLength) so that
1244         we can have at least one interval sample at the min complexity for that ramp.
1245         - Some regression calculations only come out with one line segment rather than the two
1246         we expect. This could be due to a noisy ramp or the ramp's range is too narrow. If that's the
1247         case, influence the minimum complexity of the next ramp towards the lowest bound of 1, so that
1248         we ensure that at least part of the ramp is covering a complexity range that the system can
1249         handle at full 60.
1250         - Remove an assignment to interpolatedFrameLength since that is never subsequently used.
1251
1252         Update the format used to serialize the results for analysis.
1253
1254         Each data point used to have named properties for fields like complexity and frame rate.
1255         In addition the serialized numbers had rounding errors that took up many characters.
1256         Update the format by introducing a new data container called SampleData, which contains a
1257         field map. The map maps a string to an array index. Each data point is an array, so, to
1258         get a stat, use the field map to get the array index into the data point. This allows future
1259         versions to track other data, and reduces the size of the output string by two-thirds.
1260
1261         * Animometer/resources/extensions.js:
1262         (Utilities.toFixedNumber): Add convenience function that truncates the number to a fixed
1263         precision, and converts it back to a number.
1264         (SampleData): New class that contains sample data and a field map that maps properties to
1265         an array index.
1266         (get length): Number of data points.
1267         (addField): Add a field to the field map.
1268         (push): Add a data point.
1269         (sort): Sort the data.
1270         (slice): Return new SampleData object with sliced data.
1271         (forEach): Iterate over the data with provided function.
1272         (createDatum):
1273         (getFieldInDatum): Returns the data point associated with the field name by looking it up
1274         in the field map in the datum provided, which can be the datum object itself (an array) or
1275         an index into the data member variable.
1276         (setFieldInDatum): Sets the data point associated with the field name.
1277         (at): Returns the data point at the provided index.
1278         (toArray): Serializes the data where the field map serves as property names for each point.
1279
1280         * Animometer/resources/debug-runner/graph.js:
1281         (updateGraphData): Remove unused _testData. Convert the data to the old array format for the
1282         graph to use, since the old format was much easier to work with when displaying the graphs.
1283         (onGraphTypeChanged): For some controllers, no alternative score or mean is provided.
1284         * Animometer/resources/runner/animometer.css:
1285         * Animometer/resources/runner/animometer.js: Refactor to use SampleData. Update JSON output
1286         to only go to 3 digits of precision for purposes of reducing the data size.
1287         * Animometer/resources/strings.js: Add new strings to put into the field maps.
1288         * Animometer/tests/resources/main.js: Refactor to use SampleData.
1289
1290         * Animometer/developer.html:
1291         * Animometer/index.html: Restructure results table for both pages. Add charset attribute to
1292         tests.js include.
1293         * Animometer/resources/debug-runner/animometer.css: Clear out styles from release runner.
1294         * Animometer/resources/debug-runner/graph.js:
1295         (onGraphTypeChanged): Update score and mean if bootstrap results are available from the
1296         controller, since not all controllers do bootstrapping.
1297         * Animometer/resources/debug-runner/tests.js: Update header text.
1298         * Animometer/resources/runner/animometer.css: Include confidence interval in results.
1299         * Animometer/resources/runner/animometer.js:
1300         (ResultsTable._addHeader): Header contents can be HTML, so use innerHTML instead.
1301         (ResultsTable._addBody): Add tbody element.
1302         (ResultsTable._addTest): Allow a data cell to invoke a JS function to get its contents.
1303         (window.benchmarkController.showResults): Add table that includes tests' confidence intervals.
1304         * Animometer/resources/runner/tests.js:
1305         (Headers.details.text): Add new details table that includes bootstrap confidence interval.
1306         The interval can be asymmetric, but for simplicity, report the maximum deviation percentage
1307         on either side of the bootstrap median.
1308         * Animometer/resources/statistics.js:
1309         (bootstrap): Include the confidence percentage in the return object.
1310
1311         Report canvas size in results.
1312
1313         * Animometer/developer.html: Add markup to indicate whether a small, medium, or large
1314         canvas was used.
1315         * Animometer/index.html: Ditto.
1316         * Animometer/resources/debug-runner/animometer.js: Call determineCanvasSize().
1317         * Animometer/resources/runner/animometer.css: Update styles to set the canvas based on the
1318         body class size.
1319         * Animometer/resources/runner/animometer.js:
1320         (window.benchmarkController.initialize): Update styles to set the canvas based on the
1321         body class size.
1322         (window.benchmarkController.determineCanvasSize): Run various media queries and set the body
1323         class based on the size of the device.
1324
1325         * Animometer/developer.html: Refactor to include the main CSS file, and redo
1326         the layout so that it doesn't rely on flexbox.
1327         * Animometer/resources/debug-runner/animometer.css:
1328         * Animometer/resources/debug-runner/animometer.js:
1329         (updateDisplay): Since various parts of the script alter the body class, we can't
1330         replace the className directly. Instead, remove all display-based values and then add
1331         the one that was selected.
1332         * Animometer/resources/debug-runner/graph.js:
1333         (updateGraphData): To set the size of the graph, use window.innerHeight.
1334         * Animometer/resources/runner/animometer.js:
1335         (window.sectionsManager.showSection): Since various parts of the script alter the body
1336         class, we can't replace the className directly. Remove all of the section classes
1337         individually and then add the one desired.
1338         * Animometer/tests/resources/stage.css: Remove -apple-system as a font to use in the
1339         stage.
1340
1341 2016-06-12  Filip Pizlo  <fpizlo@apple.com>
1342
1343         Fix round-down goof in Air.js's ShuffleCustom.forEachArg
1344         https://bugs.webkit.org/show_bug.cgi?id=158674
1345
1346         Reviewed by Michael Saboff.
1347         
1348         x / 3 * 3 is not how you round down to multiples of 3 in JavaScript. You need to do
1349         Math.floor(x / 3) * 3 instead.
1350         
1351         This is a benign change, because having extra arguments to a Shuffle is not something we
1352         actually take advantage of yet. But I think it's best to match the original C++ code's
1353         intent.
1354
1355         * Air.js/custom.js:
1356         (const.ShuffleCustom.forEachArg):
1357
1358 2016-06-11  Filip Pizlo  <fpizlo@apple.com>
1359
1360         Unreviewed, skip Air.js. It got unskipped when I renamed it from JSAir.
1361
1362         * Skipped:
1363
1364 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1365
1366         Make it easy to package Air.js
1367         https://bugs.webkit.org/show_bug.cgi?id=158652
1368
1369         Reviewed by Benjamin Poulain.
1370         
1371         If you want to give Air.js to someone, you can now do:
1372         
1373         cd PerformanceTests/Air.js
1374         ./make_dist.sh
1375         
1376         This will create Air.js.tar.gz and a directory called Air.js. If you want to send someone
1377         the benchmark, just send either of those things and tell them to look at index.html.
1378         
1379         You may have to edit make_dist.sh for wherever you have Markdown.pl. You can get that from
1380         https://daringfireball.net/projects/markdown/
1381
1382         * Air.js/make_dist.sh: Added.
1383         * Air.js/test.html:
1384
1385 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1386
1387         Unreviewed, make the link to test.html into an actual link.
1388
1389         * Air.js/README.md:
1390
1391 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1392
1393         Fix some text in Air.js/README.md
1394         https://bugs.webkit.org/show_bug.cgi?id=158650
1395
1396         Reviewed by Benjamin Poulain.
1397         
1398         I read the text again and found bugs:
1399         
1400         - We never actually say how to run the benchmark. This change adds a blurb about how to run
1401           it.
1402
1403         - We both say that allocateStack is responsible for the bulk of the running time and that
1404           we haven't measured where the bulk of the time is spent. This changes the text to say that
1405           it was a goal to make allocateStack be the hottest part of the benchmark, but that we did
1406           not measure this.
1407
1408         * Air.js/README.md:
1409
1410 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1411
1412         Air.js should have some documentation
1413         https://bugs.webkit.org/show_bug.cgi?id=158648
1414
1415         Reviewed by Keith Miller.
1416         
1417         I want to be able to point people at a document if they want to know more about this
1418         benchmark.
1419
1420         * Air.js/README.md: Added.
1421
1422 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1423
1424         Rename JSAir to Air.js.
1425
1426         Rubber stamped by Mark Lam.
1427
1428         * Air.js: Copied from PerformanceTests/JSAir.
1429         * Air.js/airjs-tests.yaml: Copied from PerformanceTests/JSAir/jsair-tests.yaml.
1430         * Air.js/benchmark.js: Copied from PerformanceTests/JSAir/benchmark.js.
1431         (Benchmark):
1432         * Air.js/jsair-tests.yaml: Removed.
1433         * Air.js/payload-airjs-ACLj8C.js: Copied from PerformanceTests/JSAir/payload-jsair-ACLj8C.js.
1434         * Air.js/payload-jsair-ACLj8C.js: Removed.
1435         * Air.js/stress-test.js: Copied from PerformanceTests/JSAir/stress-test.js.
1436         * Air.js/test.html: Copied from PerformanceTests/JSAir/test.html.
1437         * Air.js/test.js: Copied from PerformanceTests/JSAir/test.js.
1438         * JSAir: Removed.
1439         * JSAir/all.js: Removed.
1440         * JSAir/allocate_stack.js: Removed.
1441         * JSAir/arg.js: Removed.
1442         * JSAir/basic_block.js: Removed.
1443         * JSAir/benchmark.js: Removed.
1444         * JSAir/code.js: Removed.
1445         * JSAir/custom.js: Removed.
1446         * JSAir/frequented_block.js: Removed.
1447         * JSAir/insertion_set.js: Removed.
1448         * JSAir/inst.js: Removed.
1449         * JSAir/jsair-tests.yaml: Removed.
1450         * JSAir/liveness.js: Removed.
1451         * JSAir/opcode.js: Removed.
1452         * JSAir/payload-gbemu-executeIteration.js: Removed.
1453         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
1454         * JSAir/payload-jsair-ACLj8C.js: Removed.
1455         * JSAir/payload-typescript-scanIdentifier.js: Removed.
1456         * JSAir/reg.js: Removed.
1457         * JSAir/stack_slot.js: Removed.
1458         * JSAir/stress-test.js: Removed.
1459         * JSAir/strip-hash.rb: Removed.
1460         * JSAir/symbols.js: Removed.
1461         * JSAir/test.html: Removed.
1462         * JSAir/test.js: Removed.
1463         * JSAir/tmp.js: Removed.
1464         * JSAir/tmp_base.js: Removed.
1465         * JSAir/util.js: Removed.
1466
1467 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1468
1469         JSC Stress Test failing: jsair-tests.yaml/test.js.ftl-eager-no-cjit
1470         https://bugs.webkit.org/show_bug.cgi?id=158571
1471
1472         Reviewed by Keith Miller.
1473         
1474         Introduce a different harness for run-jsc-stress-tests, which runs for a minimum of 10
1475         iterations and then tries to do any number of "bonus" iterations until it's been running for
1476         two seconds. Since this is the sort of test that isn't really meant to stress anything in
1477         particular, I think it's OK if it is time-limited in this way. The worst case is that some
1478         of its failures will be flaky, but I think that they would have been flaky anyway given the
1479         complexity of the test.
1480
1481         * JSAir/benchmark.js:
1482         (Benchmark):
1483         (Benchmark.prototype.runIteration):
1484         (benchmark): Deleted.
1485         * JSAir/jsair-tests.yaml:
1486         * JSAir/stress-test.js: Added.
1487         (preciseTime):
1488         * JSAir/test.html:
1489         * JSAir/test.js:
1490
1491 2016-06-09  Filip Pizlo  <fpizlo@apple.com>
1492
1493         Unreviewed, teach the perf bots not to run JSAir.
1494
1495         * Skipped:
1496
1497 2016-06-08  Filip Pizlo  <fpizlo@apple.com>
1498
1499         Use more ES6 features in JSAir
1500         https://bugs.webkit.org/show_bug.cgi?id=158497
1501
1502         Reviewed by Keith Miller.
1503         
1504         This improves JSAir with the following ES6 features suggested by JoePeck:
1505         
1506         - String interpolation.
1507         - Destructuring inside PatchCustom.
1508         - Default arguments.
1509         
1510         All of these things are on hot paths.
1511         
1512         Note that I didn't use string interpolation everywhere that I could, only in those places
1513         where it made the code more readable. In Ruby, I used the style that if the interpolation
1514         expression has any non-trivial stuff (like a ternary operator, a chain of calls, or embedded
1515         strings) then it's better to use regular strcat. I think that's what I carried over to here.
1516         
1517         Note that the previous change (Add result validation to JSAir) also made the Proxy code not
1518         dead, though it's not necessarily on the hot path. The Proxy isn't called into frequently
1519         but it's used from a function that is otherwise hot, so if calling into the Proxy prevents
1520         that function from being optimized then it will hurt so good.
1521         
1522         I also reenabled tail calls in a few places.
1523         
1524         This change doesn't seem to change the performance of the benchmark for us. That's expected
1525         since these ES6 features are cheap. Note that this claim doesn't include Proxy, which was
1526         added in a separate change and that change did make the benchmark overall more expensive.
1527
1528         * JSAir/allocate_stack.js:
1529         (allocateStack):
1530         * JSAir/arg.js:
1531         (Arg.createBitImm64):
1532         (Arg.createAddr):
1533         (Arg.createStack):
1534         (Arg.logScale):
1535         (Arg.createIndex):
1536         * JSAir/basic_block.js:
1537         (BasicBlock.get headerString):
1538         (BasicBlock.prototype.get if):
1539         (BasicBlock):
1540         * JSAir/benchmark.js:
1541         (benchmark):
1542         * JSAir/code.js:
1543         (Code):
1544         (Code.prototype.addBlock):
1545         (Code.prototype.addStackSlot):
1546         (Code.prototype.newTmp):
1547         (Code.prototype.get size):
1548         (Code.prototype.get blocks):
1549         (Code.prototype.get stackSlots):
1550         (Code.prototype.tmps):
1551         (Code.prototype.get callArgAreaSize):
1552         (Code.prototype.toString):
1553         * JSAir/custom.js:
1554         (const.PatchCustom.forEachArg):
1555         * JSAir/inst.js:
1556         (Inst):
1557         * JSAir/reg.js:
1558         (Reg.prototype.toString):
1559         * JSAir/util.js:
1560         (symbolName):
1561         (lowerSymbolName):
1562
1563 2016-06-07  Filip Pizlo  <fpizlo@apple.com>
1564
1565         Add result validation to JSAir
1566         https://bugs.webkit.org/show_bug.cgi?id=158493
1567
1568         Reviewed by Saam Barati.
1569         
1570         This adds the ability to hash a Code in a way that matches the C++ code's hashing of Code.
1571         This allows us to check if the Code that JSAir sees is the code that C++ saw. We use this to
1572         check the Code before and after allocateStack, and compare against hashes we got from C++.
1573         
1574         Doing this uncovered bugs. roundUpToMultipleOf wasn't doing anything. allocateStack was not
1575         allocating things correctly because I was concatting a Set to an Array, which doesn't really
1576         work. Now these bugs are fixed.
1577         
1578         The checking step adds to the running time so I reduced the number of iterations. The
1579         benchmark spends a decent amount of its time computing Code hashes; I think it's around 1/3
1580         total. This is probably OK. It's better to verify the results even if the running time is
1581         not all in the "core" of the algorithm.
1582         
1583         Also add a run-jsc-stress-tests yaml file to allow this to run as a test.
1584
1585         * JSAir/allocate_stack.js:
1586         * JSAir/arg.js:
1587         (Arg.createImm):
1588         (Arg.createBigImm):
1589         (Arg.createBitImm):
1590         (Arg.createBitImm64):
1591         (Arg.createWidth):
1592         (Arg.createSpecial):
1593         (Arg.prototype.get kind):
1594         (Arg.prototype.get isTmp):
1595         (Arg.prototype.get isImm):
1596         (Arg.prototype.get isSomeImm):
1597         (Arg.prototype.get isSomeBigImm):
1598         (Arg.prototype.get isCondition):
1599         (Arg.prototype.get isWidth):
1600         (Arg.prototype.get isSpecial):
1601         (Arg.prototype.get isAlive):
1602         (Arg.prototype.get tmp):
1603         (Arg.prototype.get value):
1604         (Arg.prototype.get lowValue):
1605         (Arg.prototype.get highValue):
1606         (Arg.prototype.get base):
1607         (Arg.prototype.get isGP):
1608         (Arg.prototype.get isFP):
1609         (Arg.prototype.isValidForm):
1610         (Arg.prototype.get isInvertible):
1611         (Arg.kindCode):
1612         (Arg.prototype.hash):
1613         (Arg.prototype.toString):
1614         (Arg):
1615         * JSAir/basic_block.js:
1616         (BasicBlock.get successorBlocks):
1617         * JSAir/benchmark.js:
1618         (benchmark):
1619         * JSAir/code.js:
1620         (Code.prototype.setFrameSize):
1621         (Code.prototype.hash):
1622         (Code.prototype.toString):
1623         (Code):
1624         * JSAir/inst.js:
1625         (Inst.prototype.get hasNonArgEffects):
1626         (Inst.prototype.hash):
1627         (Inst.prototype.toString):
1628         (Inst):
1629         * JSAir/jsair-tests.yaml: Added.
1630         * JSAir/opcode.js:
1631         (Inst_forEachArg):
1632         (Inst_hasNonArgEffects):
1633         (opcodeCode):
1634         * JSAir/payload-gbemu-executeIteration.js:
1635         (createPayloadGbemuExecuteIteration):
1636         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js:
1637         (createPayloadImagingGaussianBlurGaussianBlur):
1638         * JSAir/payload-jsair-ACLj8C.js:
1639         (createPayloadJSAirACLj8C):
1640         * JSAir/payload-typescript-scanIdentifier.js:
1641         (createPayloadTypescriptScanIdentifier):
1642         * JSAir/reg.js:
1643         (Reg.prototype.get isReg):
1644         (Reg.prototype.hash):
1645         (Reg.prototype.toString):
1646         * JSAir/stack_slot.js:
1647         (StackSlot.prototype.setOffsetFromFP):
1648         (StackSlot.prototype.hash):
1649         (StackSlot.prototype.toString):
1650         * JSAir/symbols.js:
1651         (relCondCode):
1652         (resCondCode):
1653         (doubleCondCode):
1654         * JSAir/test.html:
1655         * JSAir/tmp.js:
1656         (Tmp.prototype.get isReg):
1657         (Tmp.prototype.hash):
1658         (Tmp.prototype.toString):
1659         * JSAir/util.js:
1660         (roundUpToMultipleOf):
1661         (symbolName):
1662
1663 2016-06-07  Filip Pizlo  <fpizlo@apple.com>
1664
1665         Implement Air::allocateStack() in ES6 to see how much of a bad idea that is
1666         https://bugs.webkit.org/show_bug.cgi?id=158318
1667
1668         Reviewed by Saam Barati.
1669         
1670         This adds a new benchmark for us to play with called JSAir. It's a complete ES6
1671         implementation of Air's allocateStack() phase along with all of Air needed to run that
1672         phase. This includes things like stack slots, registers, temporaries, basic blocks,
1673         instructions, and all of the code for iterating over, inspecting, and modifying those
1674         things.
1675         
1676         To make this work, JSC can now dump Air just before allocateStack() in the form of JS code
1677         that creates a Code object that matches exactly what C++ Air saw. This benchmark comprises
1678         four Air IRs:
1679         
1680         - Octane/gbemu's largest function, executeIteration.
1681         - Kraken/imaging-gaussian-blur's largest function in OSR entry mode, gaussuanBlur.
1682         - Octane/typescript's largest function that is actually hot, scanIdentifier.
1683         - JSAir's largest hot function, which is anonymous, so we call it by its hash (ACLj8C).
1684         
1685         This runs in about 2 seconds on my machine in JSC trunk. It includes both a commandline
1686         harness and a web harness. JSAir is almost exactly 100x slower in ES6 in WebKit than the
1687         C++ Air::allocateStack() phase on which it is based.
1688         
1689         JSAir uses the following ES6 features:
1690         
1691         - Symbol.
1692         - for-of.
1693         - arrow functions.
1694         - Map/Set.
1695         - let/const.
1696         - classes.
1697         
1698         All of these things are used in anger and should end up on the hot path. There is also code
1699         that uses Proxies, but it ends up being dead.
1700         
1701         We can improve this even more:
1702         I still need to add result validation: https://bugs.webkit.org/show_bug.cgi?id=158493
1703         I want to make it use more ES6 features: https://bugs.webkit.org/show_bug.cgi?id=158497
1704         
1705         * JSAir: Added.
1706         * JSAir/all.js: Added.
1707         * JSAir/allocate_stack.js: Added.
1708         (allocateStack.attemptAssignment):
1709         (allocateStack.assign):
1710         (allocateStack.interfere):
1711         (allocateStack.):
1712         (allocateStack):
1713         * JSAir/arg.js: Added.
1714         (Arg):
1715         (Arg.isAnyUse):
1716         (Arg.isColdUse):
1717         (Arg.isWarmUse):
1718         (Arg.cooled):
1719         (Arg.isEarlyUse):
1720         (Arg.isLateUse):
1721         (Arg.isAnyDef):
1722         (Arg.isEarlyDef):
1723         (Arg.isLateDef):
1724         (Arg.isZDef):
1725         (Arg.typeForB3Type):
1726         (Arg.widthForB3Type):
1727         (Arg.conservativeWidth):
1728         (Arg.minimumWidth):
1729         (Arg.bytes):
1730         (Arg.widthForBytes):
1731         (Arg.createTmp):
1732         (Arg.fromReg):
1733         (Arg.createImm):
1734         (Arg.createBigImm):
1735         (Arg.createBitImm):
1736         (Arg.createBitImm64):
1737         (Arg.createAddr):
1738         (Arg.createStack):
1739         (Arg.createCallArg):
1740         (Arg.createStackAddr):
1741         (Arg.isValidScale):
1742         (Arg.logScale):
1743         (Arg.createIndex):
1744         (Arg.createRelCond):
1745         (Arg.createResCond):
1746         (Arg.createDoubleCond):
1747         (Arg.createWidth):
1748         (Arg.prototype.get kind):
1749         (Arg.prototype.get isTmp):
1750         (Arg.prototype.get isImm):
1751         (Arg.prototype.get isBigImm):
1752         (Arg.prototype.get isBitImm):
1753         (Arg.prototype.get isBitImm64):
1754         (Arg.prototype.get isSomeImm):
1755         (Arg.prototype.get isAddr):
1756         (Arg.prototype.get isStack):
1757         (Arg.prototype.get isCallArg):
1758         (Arg.prototype.get isIndex):
1759         (Arg.prototype.get isMemory):
1760         (Arg.prototype.get isStackMemory):
1761         (Arg.prototype.get isRelCond):
1762         (Arg.prototype.get isResCond):
1763         (Arg.prototype.get isDoubleCond):
1764         (Arg.prototype.get isCondition):
1765         (Arg.prototype.get isWidth):
1766         (Arg.prototype.get isAlive):
1767         (Arg.prototype.get tmp):
1768         (Arg.prototype.get value):
1769         (Arg.prototype.get base):
1770         (Arg.prototype.get hasOffset):
1771         (Arg.prototype.get offset):
1772         (Arg.prototype.get stackSlot):
1773         (Arg.prototype.get index):
1774         (Arg.prototype.get scale):
1775         (Arg.prototype.get logScale):
1776         (Arg.prototype.get width):
1777         (Arg.prototype.get isGPTmp):
1778         (Arg.prototype.get isFPTmp):
1779         (Arg.prototype.get isGP):
1780         (Arg.prototype.get isFP):
1781         (Arg.prototype.get hasType):
1782         (Arg.prototype.get type):
1783         (Arg.prototype.isType):
1784         (Arg.prototype.isCompatibleType):
1785         (Arg.prototype.get isGPR):
1786         (Arg.prototype.get gpr):
1787         (Arg.prototype.get isFPR):
1788         (Arg.prototype.get fpr):
1789         (Arg.prototype.get isReg):
1790         (Arg.prototype.get reg):
1791         (Arg.isValidImmForm):
1792         (Arg.isValidBitImmForm):
1793         (Arg.isValidBitImm64Form):
1794         (Arg.isValidAddrForm):
1795         (Arg.isValidIndexForm):
1796         (Arg.prototype.isValidForm):
1797         (Arg.prototype.forEachTmpFast):
1798         (Arg.prototype.usesTmp):
1799         (Arg.prototype.forEachTmp):
1800         (Arg.prototype.is):
1801         (Arg.prototype.as):
1802         (Arg.prototype.forEachFast):
1803         (Arg.prototype.forEach):
1804         (Arg.extract):
1805         (Arg.forEachFast):
1806         (Arg.forEach):
1807         (Arg.prototype.get condition):
1808         (Arg.prototype.get isInvertible):
1809         (Arg.prototype.toString):
1810         * JSAir/basic_block.js: Added.
1811         (BasicBlock):
1812         (BasicBlock.prototype.get index):
1813         (BasicBlock.prototype.get size):
1814         (BasicBlock.prototype.Symbol.iterator):
1815         (BasicBlock.prototype.at):
1816         (BasicBlock.get last):
1817         (BasicBlock.get insts):
1818         (BasicBlock.get numSuccessors):
1819         (BasicBlock.get successors):
1820         (BasicBlock.get successorBlocks.):
1821         (BasicBlock.get successorBlocks):
1822         (BasicBlock.set get numPredecessors):
1823         (BasicBlock.get predecessors):
1824         (BasicBlock.get frequency):
1825         (BasicBlock.get headerString):
1826         (BasicBlock.get footerString):
1827         * JSAir/benchmark.js: Added.
1828         (benchmark):
1829         * JSAir/code.js: Added.
1830         (Code):
1831         (Code.prototype.addBlock):
1832         (Code.prototype.addStackSlot):
1833         (Code.prototype.newTmp):
1834         (Code.prototype.get size):
1835         (Code.prototype.at):
1836         (Code.prototype.Symbol.iterator):
1837         (Code.prototype.get blocks):
1838         (Code.prototype.get stackSlots):
1839         (Code.prototype.tmps):
1840         (Code.prototype.get callArgAreaSize):
1841         (Code.prototype.requestCallArgAreaSize):
1842         (Code.prototype.get frameSize):
1843         (Code.prototype.setFrameSize):
1844         (Code.prototype.toString):
1845         * JSAir/custom.js: Added.
1846         (const.ShuffleCustom.forEachArg):
1847         (const.ShuffleCustom.hasNonArgNonControlEffects):
1848         (const.PatchCustom.forEachArg):
1849         (const.PatchCustom.hasNonArgNonControlEffects):
1850         (const.CCallCustom.forEachArg):
1851         (const.CCallCustom.hasNonArgNonControlEffects):
1852         (const.ColdCCallCustom.forEachArg):
1853         (const.ColdCCallCustom.hasNonArgNonControlEffects):
1854         * JSAir/frequented_block.js: Added.
1855         (FrequentedBlock):
1856         (FrequentedBlock.prototype.toString):
1857         * JSAir/insertion_set.js: Added.
1858         (Insertion):
1859         (Insertion.prototype.get index):
1860         (Insertion.prototype.get element):
1861         (Insertion.prototype.lessThan):
1862         (InsertionSet):
1863         (InsertionSet.prototype.appendInsertion):
1864         (InsertionSet.prototype.append):
1865         (InsertionSet.prototype.execute):
1866         * JSAir/inst.js: Added.
1867         (Inst):
1868         (Inst.prototype.append):
1869         (Inst.prototype.clear):
1870         (Inst.prototype.get opcode):
1871         (Inst.prototype.get args):
1872         (Inst.prototype.visitArg):
1873         (Inst.prototype.forEachTmpFast):
1874         (Inst.prototype.forEachArg):
1875         (Inst.prototype.forEachTmp):
1876         (Inst.prototype.forEach):
1877         (Inst.forEachDef):
1878         (Inst.forEachDefWithExtraClobberedRegs):
1879         (Inst.prototype.get hasNonArgEffects):
1880         (Inst.prototype.toString):
1881         * JSAir/liveness.js: Added.
1882         (Liveness):
1883         (Liveness.prototype.get thing):
1884         (Liveness.prototype.get code):
1885         (Liveness.prototype.get liveAtHead):
1886         (Liveness.prototype.get liveAtTail):
1887         (Liveness.prototype.localCalc.LocalCalc):
1888         (Liveness.prototype.localCalc.LocalCalc.prototype.get liveSet):
1889         (Liveness.prototype.localCalc.LocalCalc.prototype.execute):
1890         (Liveness.prototype.localCalc):
1891         * JSAir/opcode.js: Added.
1892         (Inst_forEachArg):
1893         (Inst_hasNonArgEffects):
1894         * JSAir/payload-gbemu-executeIteration.js: Added.
1895         (createPayloadGbemuExecuteIteration):
1896         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js: Added.
1897         (createPayloadImagingGaussianBlurGaussianBlur):
1898         * JSAir/payload-jsair-ACLj8C.js: Added.
1899         (createPayloadJSAirACLj8C):
1900         * JSAir/payload-typescript-scanIdentifier.js: Added.
1901         (createPayloadTypescriptScanIdentifier):
1902         * JSAir/reg.js: Added.
1903         (Reg):
1904         (Reg.fromReg):
1905         (Reg.prototype.get index):
1906         (Reg.prototype.get type):
1907         (Reg.prototype.get name):
1908         (Reg.prototype.get isCalleeSave):
1909         (Reg.prototype.get isReg):
1910         (Reg.prototype.toString):
1911         (Reg.extract):
1912         (Reg.forEachFast):
1913         (Reg.forEach):
1914         (newGPR):
1915         (Reg.gprs.Reg.fprs.Reg.calleeSaveGPRs.Reg.calleeSaveFPRs.Reg.calleeSaves):
1916         * JSAir/stack_slot.js: Added.
1917         (StackSlot):
1918         (StackSlot.prototype.get byteSize):
1919         (StackSlot.prototype.get kind):
1920         (StackSlot.prototype.get isLocked):
1921         (StackSlot.prototype.get isSpill):
1922         (StackSlot.prototype.get index):
1923         (StackSlot.prototype.ensureSize):
1924         (StackSlot.prototype.get alignment):
1925         (StackSlot.prototype.get offsetFromFP):
1926         (StackSlot.prototype.setOffsetFromFP):
1927         (StackSlot.prototype.toString):
1928         (StackSlot.extract):
1929         (StackSlot.forEachFast):
1930         (StackSlot.forEach):
1931         * JSAir/symbols.js: Added.
1932         * JSAir/test.html: Added.
1933         * JSAir/test.js: Added.
1934         * JSAir/tmp.js: Added.
1935         (Tmp):
1936         (Tmp.fromReg):
1937         (Tmp.prototype.get index):
1938         (Tmp.prototype.get type):
1939         (Tmp.prototype.get isReg):
1940         (Tmp.prototype.toString):
1941         (Tmp.extract):
1942         (Tmp.forEachFast):
1943         (Tmp.forEach):
1944         * JSAir/tmp_base.js: Added.
1945         (TmpBase.prototype.get isGP):
1946         (TmpBase.prototype.get isFP):
1947         (TmpBase.prototype.get isGPR):
1948         (TmpBase.prototype.get isFPR):
1949         (TmpBase.prototype.get reg):
1950         (TmpBase.prototype.get gpr):
1951         (TmpBase.prototype.get fpr):
1952         (TmpBase):
1953         * JSAir/util.js: Added.
1954         (isRepresentableAsInt32):
1955         (addIndexed):
1956         (roundUpToMultipleOf):
1957         (symbolName):
1958         (mergeIntoSet):
1959         (nonEmptyRangesOverlap):
1960         (rangesOverlap):
1961         (removeAllMatching):
1962         (swap):
1963         (bubble):
1964         (bubbleSort):
1965         (currentTime):
1966         (else.currentTime):
1967
1968 2016-05-26  Yusuke Suzuki  <utatane.tea@gmail.com>
1969
1970         [JSC] Allow JSBench to use precise time
1971         https://bugs.webkit.org/show_bug.cgi?id=158050
1972
1973         Reviewed by Geoffrey Garen.
1974
1975         * JSBench/amazon-chrome-win/urem.js:
1976         (else.window.performance.window.performance.now.currentTimeInMS):
1977         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1978         (else.else.currentTimeInMS):
1979         (onload.cb):
1980         (onload):
1981         * JSBench/amazon-chrome/urem.js:
1982         (else.window.performance.window.performance.now.currentTimeInMS):
1983         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1984         (else.else.currentTimeInMS):
1985         (onload.cb):
1986         (onload):
1987         * JSBench/amazon-firefox-win/urm.js:
1988         (else.window.performance.window.performance.now.currentTimeInMS):
1989         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1990         (else.else.currentTimeInMS):
1991         (onload.cb):
1992         (onload):
1993         * JSBench/amazon-firefox/urm.js:
1994         (else.window.performance.window.performance.now.currentTimeInMS):
1995         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
1996         (else.else.currentTimeInMS):
1997         (onload.cb):
1998         (onload):
1999         * JSBench/amazon-safari/urem.js:
2000         (else.window.performance.window.performance.now.currentTimeInMS):
2001         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2002         (else.else.currentTimeInMS):
2003         (onload.cb):
2004         (onload):
2005         * JSBench/facebook-chrome-win/urem.js:
2006         (else.window.performance.window.performance.now.currentTimeInMS):
2007         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2008         (else.else.currentTimeInMS):
2009         (onload.cb):
2010         (onload):
2011         * JSBench/facebook-chrome/urem.js:
2012         (else.window.performance.window.performance.now.currentTimeInMS):
2013         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2014         (else.else.currentTimeInMS):
2015         (onload.cb):
2016         (onload):
2017         * JSBench/facebook-firefox-win/urem.js:
2018         (else.window.performance.window.performance.now.currentTimeInMS):
2019         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2020         (else.else.currentTimeInMS):
2021         (onload.cb):
2022         (onload):
2023         * JSBench/facebook-firefox/urem.js:
2024         (else.window.performance.window.performance.now.currentTimeInMS):
2025         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2026         (else.else.currentTimeInMS):
2027         (onload.cb):
2028         (onload):
2029         * JSBench/facebook-safari/urem.js:
2030         (else.window.performance.window.performance.now.currentTimeInMS):
2031         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2032         (else.else.currentTimeInMS):
2033         (onload.cb):
2034         (onload):
2035         * JSBench/google-chrome-win/urem.js:
2036         (else.window.performance.window.performance.now.currentTimeInMS):
2037         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2038         (else.else.currentTimeInMS):
2039         (onload.cb):
2040         (onload):
2041         * JSBench/google-chrome/urem.js:
2042         (else.window.performance.window.performance.now.currentTimeInMS):
2043         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2044         (else.else.currentTimeInMS):
2045         (onload.cb):
2046         (onload):
2047         * JSBench/google-firefox-win/urem.js:
2048         (else.window.performance.window.performance.now.currentTimeInMS):
2049         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2050         (else.else.currentTimeInMS):
2051         (onload.cb):
2052         (onload):
2053         * JSBench/google-firefox/uem.js:
2054         (else.window.performance.window.performance.now.currentTimeInMS):
2055         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2056         (else.else.currentTimeInMS):
2057         (onload.cb):
2058         (onload):
2059         * JSBench/google-safari/urem.js:
2060         (else.window.performance.window.performance.now.currentTimeInMS):
2061         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2062         (else.else.currentTimeInMS):
2063         (onload.cb):
2064         (onload):
2065         * JSBench/harness.js:
2066         (runBenchmark.window.currentTimeInMS):
2067         (runBenchmark.else.window.currentTimeInMS):
2068         * JSBench/twitter-chrome-win/rem.js:
2069         (else.window.performance.window.performance.now.currentTimeInMS):
2070         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2071         (else.else.currentTimeInMS):
2072         (onload.cb):
2073         (onload):
2074         * JSBench/twitter-chrome/urem.js:
2075         (else.window.performance.window.performance.now.currentTimeInMS):
2076         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2077         (else.else.currentTimeInMS):
2078         (onload.cb):
2079         (onload):
2080         * JSBench/twitter-firefox-win/urem.js:
2081         (else.window.performance.window.performance.now.currentTimeInMS):
2082         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2083         (else.else.currentTimeInMS):
2084         (onload.cb):
2085         (onload):
2086         * JSBench/twitter-firefox/urem.js:
2087         (else.window.performance.window.performance.now.currentTimeInMS):
2088         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2089         (else.else.currentTimeInMS):
2090         (onload.cb):
2091         (onload):
2092         * JSBench/twitter-safari/urem.js:
2093         (else.window.performance.window.performance.now.currentTimeInMS):
2094         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2095         (else.else.currentTimeInMS):
2096         (onload.cb):
2097         (onload):
2098         * JSBench/yahoo-chrome-win/urem.js:
2099         (else.window.performance.window.performance.now.currentTimeInMS):
2100         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2101         (else.else.currentTimeInMS):
2102         (onload.cb):
2103         (onload):
2104         * JSBench/yahoo-chrome/urem.js:
2105         (else.window.performance.window.performance.now.currentTimeInMS):
2106         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2107         (else.else.currentTimeInMS):
2108         (onload.cb):
2109         (onload):
2110         * JSBench/yahoo-firefox-win/urem.js:
2111         (else.window.performance.window.performance.now.currentTimeInMS):
2112         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2113         (else.else.currentTimeInMS):
2114         (onload.cb):
2115         (onload):
2116         * JSBench/yahoo-firefox/urem.js:
2117         (else.window.performance.window.performance.now.currentTimeInMS):
2118         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2119         (else.else.currentTimeInMS):
2120         (onload.cb):
2121         (onload):
2122         * JSBench/yahoo-safari/urem.js:
2123         (else.window.performance.window.performance.now.currentTimeInMS):
2124         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2125         (else.else.currentTimeInMS):
2126         (onload.cb):
2127         (onload):
2128
2129 2016-05-25  Geoffrey Garen  <ggaren@apple.com>
2130
2131         replaceable own properties seem to ignore replacement after property caching
2132         https://bugs.webkit.org/show_bug.cgi?id=158091
2133
2134         Reviewed by Darin Adler.
2135
2136         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
2137         * MallocBench/MallocBench/Benchmark.cpp:
2138         * MallocBench/MallocBench/Interpreter.cpp:
2139         (Interpreter::doMallocOp):
2140         * MallocBench/MallocBench/Interpreter.h:
2141         * MallocBench/MallocBench/fastMallocLog.63316.ops: Added.
2142         * MallocBench/MallocBench/jetstream.cpp: Added.
2143         (benchmark_jetstream):
2144         * MallocBench/MallocBench/jetstream.h: Added.
2145
2146 2016-05-25  Keith Miller  <keith_miller@apple.com>
2147
2148         Unreviewed, add JSBench to the skipped list for now since it doesn't
2149         work currently.
2150
2151         * Skipped:
2152
2153 2016-05-17  Keith Miller  <keith_miller@apple.com>
2154         We should have JSBench in PerformanceTests
2155         https://bugs.webkit.org/show_bug.cgi?id=157952
2156
2157         Rubber-stamped by Saam Barati.
2158
2159         There are some slight changes to the layout of the test directory
2160         to make it work nicely with run-jsc-benchmarks. Before JSBench
2161         had each of the browser specific sub-tests in a sub-directory.
2162         These have been flattened e.g. amazon/safari/ has become
2163         amazon-safari/.
2164
2165         * JSBench/amazon-chrome-win/urem.html: Added.
2166         * JSBench/amazon-chrome-win/urem.js: Added.
2167         * JSBench/amazon-chrome/urem.html: Added.
2168         * JSBench/amazon-chrome/urem.js: Added.
2169         * JSBench/amazon-firefox-win/urm.html: Added.
2170         * JSBench/amazon-firefox-win/urm.js: Added.
2171         * JSBench/amazon-firefox/urm.html: Added.
2172         * JSBench/amazon-firefox/urm.js: Added.
2173         * JSBench/amazon-safari/urem.html: Added.
2174         * JSBench/amazon-safari/urem.js: Added.
2175         * JSBench/browsercheck.js: Added.
2176         * JSBench/facebook-chrome-win/urem.html: Added.
2177         * JSBench/facebook-chrome-win/urem.js: Added.
2178         * JSBench/facebook-chrome/urem.html: Added.
2179         * JSBench/facebook-chrome/urem.js: Added.
2180         * JSBench/facebook-firefox-win/urem.html: Added.
2181         * JSBench/facebook-firefox-win/urem.js: Added.
2182         * JSBench/facebook-firefox/urem.html: Added.
2183         * JSBench/facebook-firefox/urem.js: Added.
2184         * JSBench/facebook-safari/urem.html: Added.
2185         * JSBench/facebook-safari/urem.js: Added.
2186         * JSBench/google-chrome-win/urem.html: Added.
2187         * JSBench/google-chrome-win/urem.js: Added.
2188         * JSBench/google-chrome/urem.html: Added.
2189         * JSBench/google-chrome/urem.js: Added.
2190         * JSBench/google-firefox-win/urem.html: Added.
2191         * JSBench/google-firefox-win/urem.js: Added.
2192         * JSBench/google-firefox/uem.html: Added.
2193         * JSBench/google-firefox/uem.js: Added.
2194         * JSBench/google-safari/urem.html: Added.
2195         * JSBench/google-safari/urem.js: Added.
2196         * JSBench/harness.html: Added.
2197         * JSBench/harness.js: Added.
2198         * JSBench/harness.py: Added.
2199         * JSBench/index.html: Added.
2200         * JSBench/reload.html: Added.
2201         * JSBench/twitter-chrome-win/rem.html: Added.
2202         * JSBench/twitter-chrome-win/rem.js: Added.
2203         * JSBench/twitter-chrome/urem.html: Added.
2204         * JSBench/twitter-chrome/urem.js: Added.
2205         * JSBench/twitter-firefox-win/urem.html: Added.
2206         * JSBench/twitter-firefox-win/urem.js: Added.
2207         * JSBench/twitter-firefox/urem.html: Added.
2208         * JSBench/twitter-firefox/urem.js: Added.
2209         * JSBench/twitter-safari/urem.html: Added.
2210         * JSBench/twitter-safari/urem.js: Added.
2211         * JSBench/yahoo-chrome-win/urem.html: Added.
2212         * JSBench/yahoo-chrome-win/urem.js: Added.
2213         * JSBench/yahoo-chrome/urem.html: Added.
2214         * JSBench/yahoo-chrome/urem.js: Added.
2215         * JSBench/yahoo-firefox-win/urem.html: Added.
2216         * JSBench/yahoo-firefox-win/urem.js: Added.
2217         * JSBench/yahoo-firefox/urem.html: Added.
2218         * JSBench/yahoo-firefox/urem.js: Added.
2219         * JSBench/yahoo-safari/urem.html: Added.
2220         * JSBench/yahoo-safari/urem.js: Added.
2221
2222 2016-05-20  Ryosuke Niwa  <rniwa@webkit.org>
2223
2224         run-benchmark's results should contain Animometer's debug output
2225         https://bugs.webkit.org/show_bug.cgi?id=157941
2226
2227         Reviewed by Stephanie Lewis.
2228
2229         Made developer.html support the JSON generated by run-benchmark which stores everything under debugOutput.
2230
2231         * Animometer/resources/debug-runner/animometer.js:
2232         (Utilities.initialize): Unwrap debugOutput in the case run-benchmark's result JSON is used.
2233
2234 2016-05-18  Timothy Hatcher  <timothy@apple.com>
2235
2236         Make Animometer work in all browsers
2237         https://bugs.webkit.org/show_bug.cgi?id=157855
2238         rdar://problem/26338521
2239
2240         Reviewed by Jon Lee.
2241
2242         * Animometer/resources/runner/logo.svg: Use a symbol instead
2243         of putting the id on the root svg. This works in all browsers.
2244
2245 2016-05-07  Timothy Hatcher  <timothy@apple.com>
2246
2247         Update Animometer design
2248         https://bugs.webkit.org/show_bug.cgi?id=157449
2249
2250         Reviewed by Darin Adler.
2251
2252         * Animometer/developer.html: Stop including runner/animometer.css.
2253         * Animometer/index.html: Udpated elements to fit new styles.
2254
2255         * Animometer/resources/debug-runner/animometer.css:
2256         Copy the old runner/animometer.css to here so the debug runner is unaffected.
2257
2258         * Animometer/resources/runner/animometer.css:
2259         New styles for the design. Drop flex box for broad browser support.
2260
2261         * Animometer/resources/runner/animometer.js:
2262         (ResultsTable.clear): Use textContent.
2263         (window.sectionsManager.showSection): Add class to the body.
2264         (window.sectionsManager.setSectionScore): Use textContent, not innerHTML.
2265
2266         * Animometer/resources/runner/crystal.svg: Added.
2267         * Animometer/resources/runner/lines.svg: Added.
2268         * Animometer/resources/runner/logo.svg: Added.
2269
2270         * Animometer/resources/runner/tests.js: Shorten test name and title case them.
2271
2272 2016-05-06  Manuel Rego Casasnovas  <rego@igalia.com>
2273
2274         [css-grid] Unprefix CSS Grid Layout properties
2275         https://bugs.webkit.org/show_bug.cgi?id=157137
2276
2277         Reviewed by Simon Fraser.
2278
2279         Remove "-webkit" prefix from all the grid layout properties,
2280         including the display value.
2281         Update the source code to remove the prefix where it was used too.
2282
2283         * Layout/auto-grid-lots-of-data.html:
2284         * Layout/fixed-grid-lots-of-data.html:
2285         * Layout/fixed-grid-lots-of-stretched-data.html:
2286
2287 2016-04-25  Geoffrey Garen  <ggaren@apple.com>
2288
2289         bmalloc: Misc improvements to MallocBench
2290         https://bugs.webkit.org/show_bug.cgi?id=157004
2291
2292         Reviewed by Darin Adler.
2293
2294         * MallocBench/run-malloc-benchmarks: Added --memory and --memory_warning
2295         modes for focused memory testing.
2296
2297         * MallocBench/MallocBench/Benchmark.cpp:
2298         (Benchmark::printReport): Clarified output.
2299
2300         (Benchmark::currentMemoryBytes): Added compressed memory because top
2301         does the same. (It always happens to zero in the benchmarks we run. But
2302         this is good for sanity.)
2303
2304         * MallocBench/MallocBench/CommandLine.cpp: Moved up to 8 runs to reduce
2305         variance.
2306
2307         * MallocBench/MallocBench/alloc_free.cpp:
2308         (benchmark_alloc_free): Cycle a single allocation in order to stress
2309         the effect of merging on calls to madvise.
2310
2311         * MallocBench/MallocBench/big.cpp:
2312         (benchmark_big): Graduated to 8kB-128kB because medium tests up to 8 and
2313         our large allocator doesn't kick in until 64kB.
2314
2315         * MallocBench/MallocBench/medium.cpp:
2316         (benchmark_medium): Test all the way down to 1kB because our large
2317         allocator used to service 1kB allocations and 1kB is an interesting
2318         middle size where memory is unusually large but allocation throughput
2319         still matters.
2320
2321         * MallocBench/MallocBench/stress.cpp:
2322         (benchmark_stress): Reduced the churn count to match stress_aligned
2323         because this test was taking too long to complete.
2324
2325         * MallocBench/MallocBench/stress_aligned.cpp:
2326         (benchmark_stress_aligned): Our new large allocator can handle even
2327         more absurdly large values.
2328
2329 2016-04-25  Simon Fraser  <simon.fraser@apple.com>
2330
2331         Add a content animation test that uses SVG animation.
2332         https://bugs.webkit.org/show_bug.cgi?id=156827
2333
2334         * Animation/svg-animation.html: Added.
2335         * Skipped:
2336
2337 2016-04-22  Simon Fraser  <simon.fraser@apple.com>
2338
2339         Skip two content animation tests which are only meant for iOS testing.
2340
2341         * Animation/css-animation.html: Added.
2342         * Animation/raf-animation.html: Added.
2343
2344         * Skipped:
2345
2346 2016-04-20  Simon Fraser  <simon.fraser@apple.com>
2347
2348         Add content animation tests to benchmark_runner, and allow the runner to collect device data as part of the results
2349         https://bugs.webkit.org/show_bug.cgi?id=156827
2350
2351         Add two files missing from the previous commit.
2352
2353         * Animation/css-animation.html: Added.
2354         * Animation/raf-animation.html: Added.
2355
2356 2016-04-20  Simon Fraser  <simon.fraser@apple.com>
2357
2358         Add content animation tests to benchmark_runner, and allow the runner to collect device data as part of the results
2359         https://bugs.webkit.org/show_bug.cgi?id=156827
2360
2361         Reviewed by Stephanie Lewis.
2362
2363         Two new files for the content-animation suite.
2364
2365         After patching for use in the benchmark, these tests are a little different to other benchmarks.
2366         The test content itself does not collect performance data; the tests spit out a boilerplate JSON
2367         result with a placeholder for framerate, which is collected in native code.
2368
2369         When run as a benchmark, test completion involves a couple of bounces between the test and MobileSafari.
2370         Test completion is initiated by a setTimeout() in the test, at which point location.hash is set to "#done".
2371         MobileSafari detects that, and asynchronously dumps performance data. Once this is complete, MobileSafari
2372         changes the URL has to "#submit", which triggers the XHR to the test relay that indicates the test is
2373         complete.
2374
2375         * Animation/css-animation.html: Added.
2376         * Animation/raf-animation.html: Added.
2377
2378 2016-04-15  Jon Lee  <jonlee@apple.com>
2379
2380         Animometer test could report a NaN
2381         https://bugs.webkit.org/show_bug.cgi?id=156646
2382
2383         Reviewed by Darin Adler.
2384         Provisionally reviewed by Said Abou-Hallawa.
2385
2386         * Animometer/tests/resources/main.js:
2387         (didFinishInterval): The ramp controller has a first phase where it ramps up the complexity and
2388         reacts based on how well the system handles the load. The assumption was that it would handle at least
2389         1 particle easily. That is not always the case. As a result, an interpolation calculation could
2390         end up setting an upper bound of NaN.
2391
2392         This occurs because we never get out of the first tier, so this._lastTierComplexity is undefined.
2393         Now that we guarantee a minimum complexity of 1, modify the conditional to check for this._lastTierComplexity
2394         before interpolating the upper bound of the first ramp. In the case where the system struggles with
2395         1 particle, set it to currentComplexity.
2396
2397 2016-04-09  Jon Lee  <jonlee@apple.com>
2398
2399         Update Animometer to accommodate different screens
2400         https://bugs.webkit.org/show_bug.cgi?id=156449
2401
2402         Reviewed by Darin Adler.
2403         Provisionally reviewed by Said Abou-Hallawa.
2404
2405         * Animometer/index.html: Wrap button in a container to add padding at the bottom.
2406         * Animometer/resources/debug-runner/animometer.css:
2407         (@media screen and (min-device-width: 1800px)): Deleted.
2408         (@media screen and (min-width: 1800px)): Cannot use min-device-width since it may match incorrectly.
2409         (screen and (max-device-height: 414px) and (orientation: landscape)): Some devices swap device width
2410         and height with orientation change.
2411         * Animometer/resources/runner/animometer.css: Similar.
2412         (screen and (min-device-width: 1024px) and (orientation: landscape)):
2413         (screen and (max-device-height: 414px) and (orientation: landscape)):
2414         (.frame-container): On smaller iPhones, adding 1px prevents the navigation bars from appearing.
2415         (@media screen and (min-device-width: 768px) and (max-device-width: 1024px)): Deleted.
2416         (@media (min-device-height: 768px) and (max-device-height: 1024px)): Target iPad Airs and similar.
2417         (@media screen and (min-device-width: 1024px) and (max-device-width: 1366px)): Deleted.
2418         (@media screen and (max-device-width: 1024px) and (min-device-height: 1366px)): Target iPad Pro.
2419         (#results footer): Add padding below the button for testing again.
2420         * Animometer/tests/master/multiply.html: Remove the center text.
2421         * Animometer/tests/master/resources/text.js: Update the test so that in every frame the text moves.
2422         * Animometer/tests/master/text.html: Update the text sizing depending on the size of the device.
2423
2424 2016-04-08  Jon Lee  <jonlee@apple.com>
2425
2426         Have Animometer benchmark always start with complexity of 1
2427         https://bugs.webkit.org/show_bug.cgi?id=156432
2428
2429         Reviewed by Ryosuke Niwa.
2430
2431         * Animometer/tests/resources/main.js: Update the default Controller and RampController to
2432         set its minimum complexities to 1 instead of 0.
2433
2434 2016-04-08  Jon Lee  <jonlee@apple.com>
2435
2436         Fix SVG benchmark test
2437         https://bugs.webkit.org/show_bug.cgi?id=156410
2438
2439         Reviewed by Dean Jackson.
2440
2441         * Animometer/resources/extensions.js: Update Point.zero to be a static Point.
2442         * Animometer/tests/simple/resources/tiled-canvas-image.js:
2443         (Stage.call._setupTiles): Refactor.
2444         * Animometer/tests/master/resources/particles.js:
2445         (Particle.prototype.reset): Use Point.center.
2446         (complexity): We are not using a gradient background anymore, so remove the +1.
2447         * Animometer/tests/master/resources/svg-particles.js: Update to use SVG transform
2448         instead of CSS transform.
2449
2450 2016-04-06  Jon Lee  <jonlee@apple.com>
2451
2452         Update master benchmark with SVG test
2453         https://bugs.webkit.org/show_bug.cgi?id=156273
2454
2455         Reviewed by Dean Jackson.
2456         Provisionally reviewed by Said Abou-Hallawa.
2457
2458         Switch masks tests for SVG path test.
2459
2460         * Animometer/resources/debug-runner/tests.js: Move mask test here.
2461         * Animometer/resources/runner/tests.js: Add SVG test here.
2462         * Animometer/tests/dom/particles.html: Renamed from PerformanceTests/Animometer/tests/master/particles.html.
2463         * Animometer/tests/dom/resources/dom-particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/dom-particles.js.
2464
2465         * Animometer/tests/master/resources/particles.js: Add minPosition for bounds checking. Prevents particle from being
2466         partially obscured.
2467         * Animometer/tests/master/resources/svg-particles.js: Added.
2468         (Particle): The particle is either a path object or a rect using a path as a clip. The
2469         same path is used either way. For each particle create a linear gradient with a random
2470         rotation.
2471         (SVGParticlesStage): Look in #shapes to see how many different kinds of paths are available.
2472         This makes the test more generic in case other shapes need to be tested.
2473         * Animometer/tests/master/svg-particles.html: Added. Have two defs, one that houses each
2474         particle's gradient, and one that holds the shape templates.
2475
2476 2016-03-27  Hunseop Jeong  <hs85.jeong@samsung.com>
2477
2478         [EFL] REGRESSION(r188793): It made 200 layout tests and Bindings/event-target-wrapper.html performance test fail
2479         https://bugs.webkit.org/show_bug.cgi?id=148470 
2480
2481         Reviewed by Darin Adler.
2482
2483         * Skipped: Unskip the Bindings/event-target-wrapper test.
2484
2485 2016-03-23  Jon Lee  <jonlee@apple.com>
2486
2487         Update focus and leaves tests
2488         https://bugs.webkit.org/show_bug.cgi?id=155825
2489
2490         Reviewed by Simon Fraser.
2491
2492         Particles would improperly disappear, before they were fully off stage. Update the calculation.
2493         * Animometer/tests/dom/resources/leaves.js:
2494         * Animometer/tests/master/resources/leaves.js:
2495
2496         * Animometer/tests/master/focus.html: Update the center element styling.
2497         * Animometer/tests/master/resources/focus.js:
2498         (FocusElement.Utilities.createClass): Give enough space in the container for each particle for the blur
2499         to spread. It should be about 3x more space than the blur radius on each side.
2500         (Stage.call.initialize): Ditto for the center element.
2501
2502 2016-03-23  Jon Lee  <jonlee@apple.com>
2503
2504         Update support for other platforms
2505         https://bugs.webkit.org/show_bug.cgi?id=155824
2506
2507         Reviewed by Simon Fraser.
2508
2509         * Animometer/resources/debug-runner/animometer.css: Remove extraneous media query. All of it should
2510         be handled in the release stylesheet.
2511         * Animometer/resources/debug-runner/animometer.js: Add support for browsers that return an HTMLCollection
2512         when retrieving radio elements with a name, instead of the selected value.
2513         * Animometer/resources/runner/animometer.css: Adjust the queries for devices of varying widths.
2514         (.frame-container): Move overflow:hidden to #stage.
2515         * Animometer/tests/master/resources/stage.css: Consolidate into tests/resources/stage.css.
2516         * Animometer/tests/resources/stage.css:
2517         (body): Remove overflow:hidden.
2518         (#stage): Set overflow:hidden here.
2519         (#center-text): Moved from master/resources/stage.css.
2520
2521         Update stylesheet hrefs.
2522         * Animometer/tests/master/canvas-stage.html:
2523         * Animometer/tests/master/focus.html:
2524         * Animometer/tests/master/leaves.html:
2525         * Animometer/tests/master/multiply.html:
2526         * Animometer/tests/master/particles.html:
2527         * Animometer/tests/master/text.html: Adjust text size to fit in the device screen.
2528
2529 2016-03-23  Jon Lee  <jonlee@apple.com>
2530
2531         Make the benchmark require device to be in landscape orientation
2532         https://bugs.webkit.org/show_bug.cgi?id=155822
2533         rdar://problem/25258650
2534
2535         Reviewed by Ryosuke Niwa.
2536
2537         Check that the device is in landscape orientation. If not, disable the button to start it,
2538         and show a warning message.
2539
2540         * Animometer/developer.html: Add a message about how the browser should be set up. Include a
2541         warning message that appears if the orientation is incorrect on mobile devices.
2542         * Animometer/index.html: Ditto.
2543
2544         * Animometer/resources/debug-runner/animometer.css: Migrate the .hidden rule into the release
2545         stylesheet. Update the style to accommodate the new UI.
2546         * Animometer/resources/runner/animometer.css:
2547
2548         * Animometer/resources/runner/animometer.js:
2549         (window.benchmarkController.initialize): Add the orientation listener is needed.
2550         (window.benchmarkController.addOrientationListenerIfNecessary): Only mobile devices need this. Check
2551         to see for support of window.orientation.
2552         (window.benchmarkController._orientationChanged): Toggle the warning based on whether we match the
2553         landscape query. Set a state variable, which is needed for the debug harness. Call updateStartButtonState.
2554         (window.benchmarkController.updateStartButtonState):
2555         * Animometer/resources/debug-runner/animometer.js: Change _updateStartButtonState to return
2556         a boolean about whether at least one test is selected. That will be used in
2557         benchmarkController.updateStartButtonState(). Move the load event listener to the release version.
2558         (window.benchmarkController.updateStartButtonState): Override the release version, and also check
2559         that at least one test is selected.
2560
2561 2016-03-22  Geoffrey Garen  <ggaren@apple.com>
2562
2563         MallocBench: consolidate regression testing for aligned allocation
2564         https://bugs.webkit.org/show_bug.cgi?id=155762
2565
2566         Reviewed by Michael Saboff.
2567
2568         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
2569         * MallocBench/MallocBench/Benchmark.cpp:
2570         * MallocBench/MallocBench/memalign.cpp: Removed.
2571         * MallocBench/MallocBench/memalign.h: Removed. The stress_aligned test
2572         covers this and much more.
2573
2574         * MallocBench/MallocBench/stress_aligned.cpp:
2575         (benchmark_stress_aligned): Include specific tests for extreme sizes
2576         and alignments.
2577
2578 2016-03-21  Jon Lee  <jonlee@apple.com>
2579
2580         Update benchmark tests
2581         https://bugs.webkit.org/show_bug.cgi?id=155723
2582
2583         Reviewed by Darin Adler.
2584         Provisionally reviewed by Said Abou-Hallawa.
2585
2586         Add quadratic and bezier segments to the canvas path test.
2587
2588         * Animometer/resources/runner/tests.js: Some of the query strings are unnecessary.
2589         Rename the test.
2590         * Animometer/tests/master/resources/canvas-stage.js:
2591         (tune): Update to be able to take an array of possible constructors. Choose one randomly.
2592         * Animometer/tests/master/resources/canvas-tests.js:
2593         (CanvasLinePoint.Utilities.createClass): Move the point selection out to a separate
2594         function called randomPoint() for reuse. Move X_LOOPS, Y_LOOPS, and offsets into the
2595         class definition.
2596         (randomPoint): Scale the grid down a little bit so that the lines along the edge of the
2597         canvas are not cut off when the stroke size is thick.
2598         (CanvasQuadraticSegment): Added.
2599         (CanvasBezierSegment): Added.
2600         (SimpleCanvasStage): Pass in an array of the different segment types. Since line segments
2601         are short compared to the curved ones, make it twice as likely to render a line segment.
2602         (SimpleCanvasStage.animate): Update the drawing code so that we render all line segments.
2603
2604         Add a helper method that selects a random element from an array.
2605
2606         * Animometer/tests/resources/main.js:
2607         (Stage.randomElementInArray): Select a random element from the provided array.
2608
2609         * Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js: Refactor.
2610         * Animometer/tests/dom/resources/leaves.js: Ditto.
2611         * Animometer/tests/master/resources/dom-particles.js: Ditto.
2612         * Animometer/tests/master/resources/image-data.js: Ditto.
2613         * Animometer/tests/master/resources/leaves.js: Ditto.
2614         * Animometer/tests/simple/resources/simple-canvas-paths.js: Ditto.
2615
2616         Add canvas tests that includes all stroke and fill paths. This makes it possible to avoid having to
2617         include the full simple canvas suite for perf testing.
2618
2619         * Animometer/resources/debug-runner/tests.js: Add new tests. Move the canvas test into the Canvas
2620         suite.
2621         * Animometer/tests/simple/resources/simple-canvas-paths.js: Add a CanvasStroke and CanvasFill particle
2622         that random selects an object to render.
2623
2624         * Animometer/resources/debug-runner/tests.js: Move 3D suite before basic canvas suite.
2625         When updating the perf bot script, we will include the suites up to this one, but exclude
2626         the basic canvas suite.
2627
2628         Move compositing transforms test to HTML suite and remove the empty Miscellaneous suite.
2629
2630         * Animometer/resources/debug-runner/tests.js:
2631         * Animometer/tests/dom/compositing-transforms.html: Renamed from PerformanceTests/Animometer/tests/misc/compositing-transforms.html.
2632         * Animometer/tests/dom/resources/compositing-transforms.js: Renamed from PerformanceTests/Animometer/tests/misc/resources/compositing-transforms.js.
2633
2634         Clean up miscellaneous test suite. Add a canvas ellipse test, and remove the other
2635         canvas tests.
2636
2637         * Animometer/resources/debug-runner/tests.js: Add ellipse tests to the simple suite.
2638         * Animometer/tests/misc/canvas-electrons.html: Removed.
2639         * Animometer/tests/misc/canvas-stars.html: Removed.
2640         * Animometer/tests/misc/resources/canvas-electrons.js: Removed.
2641         * Animometer/tests/misc/resources/canvas-stars.js: Removed.
2642         * Animometer/tests/simple/resources/simple-canvas-paths.js: Add ellipse primitives.
2643
2644         Merge text tests together into one.
2645
2646         * Animometer/resources/runner/tests.js: Remove international.html.
2647         * Animometer/tests/master/international.html: Removed.
2648         * Animometer/tests/master/resources/text.js:
2649         (animate): Update styling. Manually calculate gradients for the shadow particles.
2650         Reduce the step size for y direction to avoid cutting text off at the margins. Increase
2651         step for x since there will be a little more room.
2652         * Animometer/tests/master/text.html: Add more translations and lay it out in a table.
2653
2654 2016-03-20  Jon Lee  <jonlee@apple.com>
2655
2656         Add a link to show debug data
2657         https://bugs.webkit.org/show_bug.cgi?id=155724
2658
2659         Reviewed by Simon Fraser.
2660
2661         * Animometer/developer.html: Attach the onclick handler to the score.
2662         * Animometer/index.html: Ditto.
2663
2664 2016-03-18  Dean Jackson  <dino@apple.com>
2665
2666         Add a basic WebGL test to Animometer
2667         https://bugs.webkit.org/show_bug.cgi?id=155475
2668         <rdar://problem/25156860>
2669
2670         Reviewed by Simon Fraser and Jon Lee.
2671
2672         Add a pretty simple WebGL test to Animometer
2673         which tries to draw a number of animated triangles.
2674         The result still needs a bit of tuning.
2675
2676         * Animometer/resources/debug-runner/tests.js: Add a new
2677         Suite for "3D" tests.
2678         * Animometer/tests/3d/resources/webgl.js: Added.
2679         * Animometer/tests/3d/webgl.html: Added.
2680
2681 2016-03-18  Jon Lee  <jonlee@apple.com>
2682
2683         Add support for statically linking to a specific test
2684         https://bugs.webkit.org/show_bug.cgi?id=155631
2685
2686         Rubber-stamped by Darin Adler.
2687
2688         * Animometer/developer.html: Update to call restartBenchmark instead so that it works for
2689         both the preset and the manually set test sessions.
2690         * Animometer/resources/debug-runner/animometer.css:
2691         (.tree .link): Style the "link" UI.
2692         * Animometer/resources/debug-runner/animometer.js: Move updateDisplay from suitesManager
2693         to optionsManager since this is a visual option and has nothing to do with the suites
2694         settings. Update the construction of the test UI by including a "link" after each test.
2695         Clicking on that link brings up a JS prompt with a URL and a query string with the current
2696         parameters of the controller, and selected test. Pasting this URL into the location bar will
2697         automatically start running the selected test.
2698         (suitesManager.suitesFromQueryString): Iterate through the Suites and tests and find the one
2699         that matches the provided parameters. Returns an object similar to the form in
2700         suitesManager.updateLocalStorageFromUI.
2701         (benchmarkController.initialize): After settings up the events and options, try parsing the
2702         query string and running the benchmark immediately. Otherwise, fall back to the form.
2703         (benchmarkController.startBenchmark): Store the options and suites into member variables for
2704         reuse in restartBenchmark.
2705         (benchmarkController.startBenchmarkImmediatelyIfEncoded): Convert the query string to an object.
2706         If that's successful, find the suite and test referenced in the query string. Start the benchmark if
2707         the search for the test succeeded.
2708         * Animometer/resources/extensions.js:
2709         (Utilities.stripNonASCIICharacters): Helper method to convert the name of the suite and test
2710         into a query-string-friendly version.
2711         (Utilities.convertObjectToQueryString): Helper method to convert an object to query string
2712         format.
2713         (Utilities.convertQueryStringToObject): Helper method to convert query string into an object
2714         with properties and values.
2715
2716 2016-03-16  Jon Lee  <jonlee@apple.com>
2717
2718         Add a new benchmark test
2719         https://bugs.webkit.org/show_bug.cgi?id=155570
2720
2721         Reviewed by Simon Fraser.
2722
2723         New Leaves test includes various image sizes and opacity.
2724
2725         * Animometer/resources/debug-runner/tests.js: Add it to the HTML test suite.
2726         * Animometer/tests/dom/leaves.html: Added.
2727         * Animometer/tests/dom/resources/leaves.js: Added. Override the
2728         (Particle.call.reset): Uses a range of sizes, and opacity.
2729         (Particle.call.animate): Opacity goes up then down. When it hits 0, reset the particle.
2730         (Particle.call.move): Set transform and opacity.
2731         * Animometer/tests/master/resources/leaves.js: Get rid of the closure so that it
2732         can be used in this test. Update the relative path so that it works in both the master
2733         and dom test suite.
2734
2735 2016-03-15  Simon Fraser  <simon.fraser@apple.com>
2736
2737         Add developer Animometer test that bounces P3-tagged images
2738         https://bugs.webkit.org/show_bug.cgi?id=155511
2739
2740         Reviewed by Tim Horton.
2741         
2742         Add a test for rendering performance of tagged images. The 5 images are tagged
2743         with the Display P3 colorspace.
2744
2745         * Animometer/resources/debug-runner/tests.js:
2746         * Animometer/tests/bouncing-particles/bouncing-tagged-images.html: Added.
2747         * Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js: Added.
2748         * Animometer/tests/bouncing-particles/resources/image1.jpg: Added.
2749         * Animometer/tests/bouncing-particles/resources/image2.jpg: Added.
2750         * Animometer/tests/bouncing-particles/resources/image3.jpg: Added.
2751         * Animometer/tests/bouncing-particles/resources/image4.jpg: Added.
2752         * Animometer/tests/bouncing-particles/resources/image5.jpg: Added.
2753
2754 2016-03-09  Jon Lee  <jonlee@apple.com>
2755
2756         Enhance existing Animometer tests
2757         https://bugs.webkit.org/show_bug.cgi?id=155261
2758
2759         Reviewed by Simon Fraser.
2760
2761         * Animometer/tests/master/resources/canvas-tests.js: Add a gradient to
2762         the filled circles. To expose more of the gradient, add another circle.
2763         (SimpleCanvasStage.call.animate): For each frame, create a gradient with
2764         undulating stop points and colors. Fill the circles twice; once with
2765         the solid color, and once with the gradient.
2766
2767         * Animometer/tests/master/resources/dom-particles.js: Refactor the
2768         emission variables into a separate stage for this test. Add a colorOffset
2769         variable to make the colors of each particle slightly different, since
2770         the ramp controller can add large numbers of particles all at once, which
2771         would otherwise get all the same color.
2772         * Animometer/tests/master/resources/particles.js:
2773         (initialize): Remove the code specific to the SVG mask test.
2774
2775         * Animometer/tests/master/resources/image-data.js:
2776         (initialize): Remove unused local variable.
2777         * Animometer/tests/master/resources/multiply.js:
2778         (initialize): Make the test harder by adding more total particles.
2779
2780 2016-03-09  Jon Lee  <jonlee@apple.com>
2781
2782         Add text tests
2783         https://bugs.webkit.org/show_bug.cgi?id=155257
2784
2785         Reviewed by Simon Fraser.
2786
2787         * Animometer/resources/extensions.js:
2788         (UnitBezier.Utilities.createClass): Add a class that computes Bezier points
2789         assuming that two of the control points are at (0,0) and (1,1). Taken from
2790         WebCore/platform/graphics/UnitBezier.h
2791         * Animometer/resources/runner/tests.js: Add a test for English text and one
2792         for international text.
2793         * Animometer/tests/master/international.html: Added.
2794         * Animometer/tests/master/resources/text.js: Added. The test assumes there is
2795         #template div which it will copy. The copies are placed behind the template, and
2796         are set with different colors each frame. They are moved around with CSS transform.
2797         * Animometer/tests/master/text.html: Added.
2798
2799         Remove the other text tests, since these ones cover the same techniques.
2800         * Animometer/tests/text/layering-text.html: Removed.
2801         * Animometer/tests/text/resources/layering-text.js: Removed.
2802         * Animometer/tests/text/resources/text-boxes.js: Removed.
2803         * Animometer/tests/text/text-boxes.html: Removed.
2804         * Animometer/resources/debug-runner/tests.js:
2805
2806 2016-03-09  Jon Lee  <jonlee@apple.com>
2807
2808         Add a new image test
2809         https://bugs.webkit.org/show_bug.cgi?id=155232
2810
2811         Reviewed by Dean Jackson.
2812         Provisionally reviewed by Said Abou-Hallawa.
2813
2814         The image test renders PNGs and moves them with translate and rotate
2815         transforms. Each particle has a lifetime, and when the lifetime ends or
2816         the particle goes offscreen, it resets itself somewhere on the stage.
2817
2818         * Animometer/resources/debug-runner/tests.js: Remove the CSS bouncing PNG
2819         images test, because this one tests the same technique.
2820         * Animometer/resources/runner/tests.js: Added here as "Leaves".
2821         * Animometer/tests/master/leaves.html: Added.
2822         * Animometer/tests/master/resources/particles.js: Refactor out the parts
2823         specific to the DOM particles test. Consequently make velocity a public
2824         member.
2825         (initialize): The options parameter is never used, so remove it.
2826         (animate): Remove unused local variable.
2827         * Animometer/tests/master/resources/leaves.js: Maintains a focal point that
2828         moves back and forth across the canvas. That point affects the velocity
2829         of the particles. When the focal point is on the leftmost side, it is
2830         between [-6, -2], and set in reset(). When the focal point is on the rightmost
2831         side, it will be from [2, 6].
2832         * Animometer/tests/master/resources/dom-particles.js: Move JS specific to
2833         this test here from particles.js.
2834
2835         New images.
2836         * Animometer/tests/master/resources/compass100.png: Added.
2837         * Animometer/tests/master/resources/console100.png: Added.
2838         * Animometer/tests/master/resources/contribute100.png: Added.
2839         * Animometer/tests/master/resources/debugger100.png: Added.
2840         * Animometer/tests/master/resources/inspector100.png: Added.
2841         * Animometer/tests/master/resources/layout100.png: Added.
2842         * Animometer/tests/master/resources/performance100.png: Added.
2843         * Animometer/tests/master/resources/script100.png: Added.
2844         * Animometer/tests/master/resources/shortcuts100.png: Added.
2845         * Animometer/tests/master/resources/standards100.png: Added.
2846         * Animometer/tests/master/resources/storage100.png: Added.
2847         * Animometer/tests/master/resources/styles100.png: Added.
2848         * Animometer/tests/master/resources/timeline100.png: Added.
2849
2850 2016-03-03  Jon Lee  <jonlee@apple.com>
2851
2852         Add ability to retrieve raw data from release harness
2853         https://bugs.webkit.org/show_bug.cgi?id=155026
2854
2855         Reviewed by Simon Fraser.
2856
2857         * Animometer/developer.html: Remove the special UI in the debug
2858         harness.
2859         * Animometer/resources/runner/animometer.css: Add styles for the
2860         overlay.
2861         * Animometer/resources/runner/animometer.js: Let 'j' show the JSON
2862         results, but only if the overlay doesn't exist. Add 'esc' key to dismiss
2863         the overlay.
2864         (window.benchmarkController.selectResults): Cycle the cases around
2865         so that the first 's' press selects both the benchmark score and
2866         the individual test scores.
2867         * Animometer/resources/debug-runner/animometer.css: Remove
2868         unneeded rules.
2869         * Animometer/resources/debug-runner/animometer.js: Make the same
2870         call to handleKeyPress.
2871
2872 2016-03-03  Jon Lee  <jonlee@apple.com>
2873
2874         Make sure multiply test particles have at least some opacity
2875         https://bugs.webkit.org/show_bug.cgi?id=155027
2876
2877         Reviewed by Simon Fraser.
2878
2879         * Animometer/tests/master/resources/multiply.js:
2880         (initialize): Have the elements spin a little faster to make it
2881         more obvious when the system is being stressed.
2882         (_addTile):
2883         (animate): Make sure each element being animated has at least 1%
2884         opacity.
2885
2886 2016-03-03  Jon Lee  <jonlee@apple.com>
2887
2888         Update image test
2889         https://bugs.webkit.org/show_bug.cgi?id=154962
2890
2891         Rubber-stamped by Darin Adler.
2892
2893         * Animometer/tests/master/image-data.html: Make each canvas have a compositing layer.
2894         * Animometer/tests/master/resources/image-data.js:
2895         (initialize): Update to load multiple images serially.
2896         (_loadImage):
2897         (tune): Use display instead of visibility. When showing an element anew, move it somewhere else.
2898         (_createTestElement): Set up the element, refactor out placement of element to _refreshElement.
2899         (_refreshElement): Place element in a tile grid.
2900         (animate):
2901         (_getRandomNeighboringPixelIndex): Use a more conservative distribution to make the effect last a little longer.
2902         * Animometer/tests/master/resources/compass.svg: Added.
2903         * Animometer/tests/master/resources/console.svg: Added.
2904         * Animometer/tests/master/resources/contribute.svg: Added.
2905         * Animometer/tests/master/resources/debugger.svg: Added.
2906         * Animometer/tests/master/resources/inspector.svg: Added.
2907         * Animometer/tests/master/resources/layout.svg: Added.
2908         * Animometer/tests/master/resources/performance.svg: Added.
2909         * Animometer/tests/master/resources/script.svg: Added.
2910         * Animometer/tests/master/resources/shortcuts.svg: Added.
2911         * Animometer/tests/master/resources/standards.svg: Added.
2912         * Animometer/tests/master/resources/storage.svg: Added.
2913         * Animometer/tests/master/resources/styles.svg: Added.
2914         * Animometer/tests/master/resources/timeline.svg: Added.
2915
2916 2016-03-02  Jon Lee  <jonlee@apple.com>
2917
2918         Add some new controllers, and refine tests
2919         https://bugs.webkit.org/show_bug.cgi?id=154914
2920
2921         Reviewed by Simon Fraser.
2922
2923         Improve tests.
2924
2925         * Animometer/tests/master/focus.html: Move each particle into a layer. Put the particle
2926         in a container element. Expand the container element by the max blur radius, and clip
2927         overflow. This way, when the blur is applied, it avoids causing layer resizes.
2928         * Animometer/tests/master/resources/focus.js:
2929         (FocusElement.Utilities.createClass): Create a container element, and put the particle
2930         inside.
2931         (hide): Set display:none.
2932         (show): Set display:block.
2933         (animate): Apply filters to the container element.
2934         (FocusStage.call.initialize): Instead of inserting and removing elements from the DOM,
2935         keep them in the stage, but set the display style instead. Use this._offsetIndex to
2936         keep track of which elements are displayed.
2937         (FocusStage.call.animate): Move some calculations around to avoid doing unneeded math.
2938         (FocusStage.call.getBlurValue): Make sure elements are always blurred.
2939         (FocusStage.call.getOpacityValue): Make sure elements have some opacity.
2940         * Animometer/tests/master/resources/multiply.js: Have the particles on the edge of the
2941         stage appear less black when the complexity gets large enough.
2942         (tune): Have this._distanceFactor calculate the factor. Avoid calculating square root each
2943         frame.
2944
2945 2016-03-02  Jon Lee  <jonlee@apple.com>
2946
2947         Add some new controllers, and refine tests
2948         https://bugs.webkit.org/show_bug.cgi?id=154914
2949
2950         Reviewed by Simon Fraser.
2951
2952         Add a controller that centers around 30 fps instead of 60 fps.
2953
2954         * Animometer/developer.html: Add a new option.
2955         * Animometer/resources/debug-runner/animometer.js:
2956         * Animometer/resources/runner/animometer.js:
2957         (this._processData.findRegression): When calculating the complexity-frameLength regression,
2958         check the controller, and use a 30 fps baseline if needed.
2959         * Animometer/resources/statistics.js:
2960         (Regression.Utilities.createClass): Update to allow clients to specify the baseline
2961         frame length.
2962         * Animometer/tests/resources/main.js:
2963         (tune): Override some of the constants in RampController. Move those constants out for
2964         easier reading.
2965
2966         Add a fixed controller, with no step.
2967
2968         * Animometer/developer.html: Add a controller that takes no step.
2969         * Animometer/resources/debug-runner/animometer.js:
2970         * Animometer/tests/resources/main.js:
2971         (Rotater.Utilities.createClass):
2972
2973         Switch to ramp controller as default.
2974
2975         * Animometer/developer.html: Increase the test length to 20 seconds.
2976         * Animometer/resources/debug-runner/animometer.js: Rename the "adjustment" field to "controller"
2977         since that is a more accurate description.
2978         * Animometer/resources/debug-runner/graph.js:
2979         * Animometer/resources/runner/animometer.js: Update preferences for release suite.
2980
2981 2016-02-26  Jon Lee  <jonlee@apple.com>
2982
2983         Address Dean's comments in 154673.
2984
2985         * Animometer/developer.html:
2986         * Animometer/resources/debug-runner/animometer.js:
2987         * Animometer/resources/debug-runner/graph.js:
2988
2989 2016-02-25  Jon Lee  <jonlee@apple.com>
2990
2991         Update animation benchmark and tests
2992         https://bugs.webkit.org/show_bug.cgi?id=154673
2993
2994         Reviewed by Dean Jackson.
2995
2996         Update test visuals.
2997
2998         * Animometer/tests/master/resources/canvas-tests.js:
2999         (CanvasLineSegment.Utilities.createClass): Line segments near the edge of the stage get clipped.
3000         Update circle position and radius to minimize impact.
3001         * Animometer/tests/master/resources/particles.js: Get rid of rotating gradient background, and
3002         have 3 locations for emitting particles.
3003         * Animometer/tests/master/resources/multiply.js: Update distance metric so that fringe tiles get
3004         more color.
3005         (initialize): Fix some of the math for laying out the tiles.
3006         (animate): When a tile isn't used, set visibility: hidden.
3007
3008 2016-02-24  Jon Lee  <jonlee@apple.com>
3009
3010         Update animation benchmark and tests
3011         https://bugs.webkit.org/show_bug.cgi?id=154673
3012
3013         Reviewed by Dean Jackson.
3014
3015         Update the ramp controller.
3016
3017         The controller refines the complexity interval to test across.
3018
3019         * Animometer/resources/statistics.js: Add functions that estimate cumulative distribution function.
3020         (Regression): For the flat regression, force the first segment to be at 60 fps.
3021         (valueAt): Add convenience function to return interpolated value based on the regression used.
3022         (_calculateRegression): Include the number of points included for both segments, and the piecewise
3023         errors.
3024         * Animometer/tests/resources/math.js: Make the Kalman estimator subclass Experiment, and allow it
3025         to be reset.
3026
3027         * Animometer/tests/resources/main.js: Initialize the tier such that it starts at 10^0 = 1.
3028         Increase the number of ramps. Maintain three FPS thresholds-- the frame rate of interest, a limit
3029         on the lowest FPS we care to go for later interpolation, and a minimum FPS threshold we want to
3030         aim for each ramp. Also keep three estimators: a running average of the change point, a minimum
3031         boundary for each ramp, and an estimator for all the frames within an interval. The first two
3032         are used to determine the parameters of the next ramp, and the latter allows us to refine the
3033         parameters.
3034         (update): During the tier phase, it is possible that the highest complexity possible for a test
3035         won't stress the system enough to trigger stopping the tier phase and transitioning to the ramps.
3036         If the complexity doesn't change when going to the next tier, we've maxed the test out, and move
3037         on. When the tier phase completed, turn off Controller.frameLengthEstimator, which estimates the
3038         FPS at each tier.
3039         (tune): At each interval, look at the confidence distribution of being on the 60 FPS side or the
3040         slow side. If the slowest FPS we achieve at the ramp's maximum complexity is not at least
3041         _fpsRampSlowThreshold, then increase the maximum complexity. If we ever achieve 60 FPS, increase
3042         the ramp's minimum complexity to that level. If, at an even lower complexity, a glitch causes the
3043         FPS to drop, we reset the minimum complexity.
3044
3045         Have the bootstrap calculation occur between tests. Clean up harness.
3046
3047         * Animometer/resources/debug-runner/animometer.js: Run bootstrap after a test has
3048         completed to avoid doing all of it at the end before showing the results. Clean up
3049         parameters being passed around.
3050         * Animometer/resources/debug-runner/tests.js:
3051         (text):
3052         * Animometer/resources/runner/animometer.js:
3053         (this._processData.calculateScore): Save the results to the same object holding the data.
3054         (this._processData._processData): In the case where a file is dragged, calculate the score
3055         serially. Grab the results object and move it to the results variable and remove it from
3056         the data object. This avoids serializing the results into the JSON.
3057         (this._processData.findRegression): Include the samples used for bootstrapping. Reduce the
3058         resample size to shorten the wait.
3059         * Animometer/resources/runner/benchmark-runner.js:
3060         * Animometer/resources/statistics.js:
3061         (bootstrap): Update how bootstrapData is sorted. In some regression results the mix of
3062         floats and integers causes an alphabetical sort to occur.
3063         * Animometer/resources/strings.js:
3064
3065         Add meta charset so that encodings between harness and test match.
3066
3067         * Animometer/tests/bouncing-particles/bouncing-canvas-images.html:
3068         * Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html:
3069         * Animometer/tests/bouncing-particles/bouncing-css-images.html:
3070         * Animometer/tests/bouncing-particles/bouncing-css-shapes.html:
3071         * Animometer/tests/bouncing-particles/bouncing-svg-images.html:
3072         * Animometer/tests/bouncing-particles/bouncing-svg-shapes.html:
3073         * Animometer/tests/master/canvas-stage.html:
3074         * Animometer/tests/master/focus.html:
3075         * Animometer/tests/master/image-data.html:
3076         * Animometer/tests/master/multiply.html:
3077         * Animometer/tests/master/particles.html:
3078         * Animometer/tests/misc/canvas-electrons.html:
3079         * Animometer/tests/misc/canvas-stars.html:
3080         * Animometer/tests/misc/compositing-transforms.html:
3081         * Animometer/tests/simple/simple-canvas-paths.html:
3082         * Animometer/tests/simple/tiled-canvas-image.html:
3083         * Animometer/tests/template/template-canvas.html:
3084         * Animometer/tests/template/template-css.html:
3085         * Animometer/tests/template/template-svg.html:
3086         * Animometer/tests/text/layering-text.html:
3087         * Animometer/tests/text/text-boxes.html:
3088
3089         Update test harness reporting.
3090
3091         * Animometer/developer.html: Add missing meta charset.
3092         * Animometer/index.html: Remove unnecessary utf-8 declaration.
3093         * Animometer/resources/debug-runner/animometer.css: Add convenience classes for
3094         formatting the results table.
3095         * Animometer/resources/debug-runner/animometer.js: Adjust which stats are shown.
3096         * Animometer/resources/debug-runner/tests.js: Display bootstrapping statistics.
3097         * Animometer/resources/strings.js: Move strings not used by the release harness.
3098
3099         Switch to a pseudo-random number generator.
3100
3101         * Animometer/resources/statistics.js: Add a Pseudo class, with a simple
3102         pseudo-random number generator.
3103         (_calculateRegression): Reset the generator before running bootstrap.
3104         (bootstrap): Deleted.
3105
3106         Replace Math.random with Pseudo.random.
3107         * Animometer/tests/master/resources/canvas-tests.js:
3108         * Animometer/tests/master/resources/focus.js:
3109         * Animometer/tests/master/resources/particles.js:
3110         * Animometer/tests/resources/main.js:
3111
3112         Use bootstrapping to get confidence interval in the breakpoint.
3113
3114         For the ramp controller, calculate the piecewise regression, and then use
3115         bootstrapping in order to find the 95% confidence interval. Use the raw data.
3116
3117         * Animometer/developer.html: Default to the complexity graph. Add a legend
3118         checkbox to toggle visibility of the bootstrap score and histogram.
3119         * Animometer/resources/debug-runner/animometer.css: Make some more space to show
3120         the old raw and average scores in the legend. Add new styles for the data.
3121         * Animometer/resources/debug-runner/graph.js:
3122         (_addRegressionLine): Allow passing an array for the variance bar tied to the
3123         regression line. Now |stdev| is |range|.
3124         (createComplexityGraph): Add bootstrap median, and overlay a histogram of
3125         the bootstrap samples. Switch raw samples from circles to X's.
3126         (onComplexityGraphOptionsChanged): Allow toggling of the bootstrap data.
3127         (onGraphTypeChanged): Move the regressions for the raw and average samples to the
3128         legend. In the subtitle use the bootstrap median, and include the 95% confidence
3129         interval.
3130         * Animometer/resources/runner/animometer.js:
3131         (this._processData.findRegression): Factor out the code that determines which
3132         samples to include when calculating the piecewise regression. For series that have
3133         many samples, or a wider range of recorded complexities, throw away the 2.5%
3134         lowest and highest samples before calculating the regression. Keep all samples
3135         if the number of samples to regress is small or the range of complexities is
3136         narrow.
3137         (this._processData._calculateScore): Factor out regression calculation to
3138         findRegression(). Bootstrap the change point of the regression. The score is the
3139         median.
3140         * Animometer/resources/statistics.js:
3141         (_calculateRegression): Correct an issue in the calculation of the regression, where
3142         the denominator can be 0.
3143         (bootstrap): Template for bootstrapping. Create a bootstrap sample array, Create
3144         re-samples by random selection with replacement. Return the 95% confidence samples,
3145         the bootstrap median, mean, and the data itself.
3146         * Animometer/resources/strings.js: Add bootstrap.
3147         * Animometer/tests/resources/main.js:
3148         (processSamples): Don't prematurely cut the sample data.
3149
3150         Fix graph drawing.
3151
3152         * Animometer/resources/debug-runner/animometer.js: Add spacing in the JSON output.
3153         Multiple tests output a lot of JSON and can hang when selecting JSON with no whitespace.
3154         * Animometer/resources/debug-runner/animometer.css:
3155         (#complexity-graph .series.raw circle): Update the color.
3156         * Animometer/resources/debug-runner/graph.js: Use the FPS axis instead of the
3157         complexity axis, which can vary in domain. For determining the complexity domain,
3158         only use samples after samplingTimeOffset.
3159
3160         Allow dropping results JSON.
3161
3162         * Animometer/developer.html: Add a button.
3163         * Animometer/resources/debug-runner/animometer.css:
3164         * Animometer/resources/debug-runner/animometer.js: Read the data and go straight
3165         to the dashboard. With JSON output, write out only the options and the raw data.
3166
3167         Teach the harness to evaluate the samples and determine the test score.
3168
3169         This will allow us to update how the score is calculated separately from the samples recorded.
3170         This also prepares the harness to be able to accept JSON of prior runs.
3171
3172         * Animometer/resources/strings.js: Clean up and remove unneeded strings and reduce some of the
3173         hierarchy.
3174         * Animometer/resources/debug-runner/tests.js: Update to use the new strings.
3175
3176         * Animometer/tests/resources/main.js: Allow all controllers to show a complexity-FPS graph.
3177         (_processComplexitySamples): Factor out some of the sample processing done in the ramp
3178         controller for the benefit of the other controllers. |complexitySamples| contains a list of
3179         samples. Sort the samples by complexity. Optionally remove the top x% of samples.
3180         Group them, and calculate distribution of samples within the same complexity, and add those as
3181         new entries into |complexityAverageSamples|.
3182         (Controller.processSamples): Move the code responsible for determining the complexity and FPS
3183         scores out to ResultsDashboard. The structure of the data returned by the controller is:
3184
3185         {
3186             controller: [time-regression, time-regression, ...], // optional, data specific to controller
3187             marks: [...],
3188             samples: {                    // all of the sample data
3189                 controller: [...],
3190                 complexity: [...],        // processed from controller samples
3191                 complexityAverage: [...], // processed from complexity samples
3192             }
3193         }
3194
3195         (AdaptiveController.processSamples): Adding the target frame length is no longer necessary; we
3196         now pass the test options to the graph.
3197         (Regression): Move to statistics.js.
3198         * Animometer/resources/statistics.js: Move Regression to here. Add a check if the sampling range
3199         only contains one sample, since we cannot calculate a regression from one sample point.
3200
3201         Teach the test harness to evaluate the data.
3202         * Animometer/resources/runner/animometer.js:
3203         (ResultsDashboard): Store the options used to run the test and the computed results/score separately
3204         from the data. The results are stored as:
3205
3206         {
3207             score: /* geomean of iteration score */,
3208             iterationsResults: [
3209                 {
3210                     score: /* geomean of tests */,
3211                     testsResults: {
3212                         suiteName: {
3213                             testName: {
3214                                 controller: {
3215                                     average:
3216                                     concern:
3217                                     stdev:
3218                                     percent:
3219                                 },
3220                                 frameLength: { ... },
3221                                 complexity: {
3222                                     complexity:
3223                                     stdev:
3224                                     segment1:
3225                                     segment2:
3226                                 },
3227                                 complexityAverage: { ... }
3228                             },
3229                             testName: { ... },
3230                         },
3231                         ... next suite ...
3232                     }
3233                 },
3234                 { ...next iteration... }
3235             ]
3236         }
3237
3238         * Animometer/resources/debug-runner/animometer.js: Pass options around instead of relying
3239         on what was selected in the form. This will later allow for dropping previous results, and
3240         using those runs' options when calculating scores.
3241         (ResultsTable._addGraphButton): Simplify button action by using attached test data.
3242         * Animometer/resources/debug-runner/graph.js: Refactor to use the data.
3243
3244         Consolidate JS files, and move statistics out to a separate JS.
3245
3246         Preparation for having the Controller only handle recording and storage of the samples,
3247         and leave the evaluation of the test score out to the harness. Move Experiment to
3248         a new statistics.js, where Regression will also eventually go. Get rid of algorithm.js
3249         and move it to utilities.js since the Heap is used only for Experiments.
3250
3251         * Animometer/tests/resources/algorithm.js: Removed. Heap is in utilities.js.
3252         * Animometer/tests/resources/sampler.js: Removed. Experiment is in statistics.js,
3253         Sampler in main.js.
3254         * Animometer/tests/resources/main.js: Move Sampler here.
3255         * Animometer/resources/statistics.js: Added. Move Statistics and Experiment here.
3256         * Animometer/resources/extensions.js: Move Heap here. Attach static method to create
3257         a max or min heap to Heap, instead of a new Algorithm object.
3258
3259         Update JS files.
3260         * Animometer/developer.html:
3261         * Animometer/index.html:
3262         * Animometer/tests/bouncing-particles/bouncing-canvas-images.html:
3263         * Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html:
3264         * Animometer/tests/bouncing-particles/bouncing-css-images.html:
3265         * Animometer/tests/bouncing-particles/bouncing-css-shapes.html:
3266         * Animometer/tests/bouncing-particles/bouncing-svg-images.html:
3267         * Animometer/tests/bouncing-particles/bouncing-svg-shapes.html:
3268         * Animometer/tests/master/canvas-stage.html:
3269         * Animometer/tests/master/focus.html:
3270         * Animometer/tests/master/image-data.html:
3271         * Animometer/tests/master/multiply.html:
3272         * Animometer/tests/master/particles.html:
3273         * Animometer/tests/misc/canvas-electrons.html:
3274         * Animometer/tests/misc/canvas-stars.html:
3275         * Animometer/tests/misc/compositing-transforms.html:
3276         * Animometer/tests/simple/simple-canvas-paths.html:
3277         * Animometer/tests/simple/tiled-canvas-image.html:
3278         * Animometer/tests/template/template-canvas.html:
3279         * Animometer/tests/template/template-css.html:
3280         * Animometer/tests/template/template-svg.html:
3281         * Animometer/tests/text/layering-text.html:
3282         * Animometer/tests/text/text-boxes.html:
3283
3284         Fix the cursor in the graph analysis when the min
3285         complexity is not 0.
3286
3287         * Animometer/resources/debug-runner/graph.js:
3288         (_addRegression):
3289         (createComplexityGraph):
3290
3291 2016-02-23  Geoffrey Garen  <ggaren@apple.com>
3292
3293         Fix some issues in MallocBench
3294         https://bugs.webkit.org/show_bug.cgi?id=154600
3295
3296         Reviewed by Oliver Hunt.
3297
3298         * MallocBench/MallocBench.xcodeproj/project.pbxproj: Make nimlang.ops
3299         a part of the build so that it copies to the right place for execution.
3300
3301         * MallocBench/MallocBench/Interpreter.cpp:
3302         (Interpreter::Interpreter): Don't try to open .ops files for writing
3303         because we're only going to read and we might need extra permissions to
3304         write.
3305
3306         (Interpreter::~Interpreter): Give more context when opening a file fails
3307         to help with debugging.
3308
3309         * MallocBench/MallocBench/stress.cpp:
3310         (benchmark_stress): Reduce iterations to complete in less than 10 seconds.
3311
3312 2016-02-23  Michael Saboff  <msaboff@apple.com>
3313
3314         Unreviewed change to revert extraneous changes made part of change set 196955.
3315
3316         * MallocBench/MallocBench/Interpreter.cpp:
3317         (Interpreter::doMallocOp):
3318         (Interpreter::Thread::switchTo):
3319         (writeData): Deleted.
3320
3321 2016-02-19  Michael Saboff  <msaboff@apple.com>
3322
3323         MallocBench: Added recording for nimlang website, new recording details and added new options
3324         https://bugs.webkit.org/show_bug.cgi?id=154485
3325
3326         Reviewed by Geoff Garen.
3327
3328         Added new capabilities to MallocBench.  These include:
3329             Added a recording of http://nim-lang.org/docs/lib.html.
3330             Added thread id to the recording and the ability to playback switching threads in MallocBench
3331             Added aligned allocations to recordings and the ability to playback
3332             Added --use-thread-id option to honor recorded thread ids
3333             Added --detailed-report to output remaining allocations by size after playback
3334             Added --no-warmup to not run the warm up iteration
3335
3336         Changed the way that options are passed down to the benchmarks.  Instead of passing individual
3337         boolean or numeric option values, just pass a reference the CommandLine itself.  Each benchmark
3338         can access the options that are appropriate.  The Benchmark class also uses the options for