36093d9070c2761c120f3fc0594c4b4e97038389
[WebKit-https.git] / PerformanceTests / ChangeLog
1 2017-05-11  Addy Osmani  <addyosmani@gmail.com>
2
3         Speedometer: Update the AngularJS implementation to a more recent library version
4         https://bugs.webkit.org/show_bug.cgi?id=171307
5
6         Reviewed by Ryosuke Niwa.
7
8         Updates the AngularJS implementation of TodoMVC to 1.4.x
9
10         * Speedometer/resources/todomvc/architecture-examples/angularjs/bower.json: Removed.
11         * Speedometer/resources/todomvc/architecture-examples/angularjs/bower_components/angular-mocks/angular-mocks.js: Removed.
12         * Speedometer/resources/todomvc/architecture-examples/angularjs/bower_components/angular/angular.js: Removed.
13         * Speedometer/resources/todomvc/architecture-examples/angularjs/bower_components/todomvc-common/base.css: Removed.
14         * Speedometer/resources/todomvc/architecture-examples/angularjs/bower_components/todomvc-common/base.js: Removed.
15         * Speedometer/resources/todomvc/architecture-examples/angularjs/bower_components/todomvc-common/bg.png: Removed.
16         * Speedometer/resources/todomvc/architecture-examples/angularjs/index.html:
17         * Speedometer/resources/todomvc/architecture-examples/angularjs/js/app.js:
18         * Speedometer/resources/todomvc/architecture-examples/angularjs/js/controllers/todoCtrl.js:
19         * Speedometer/resources/todomvc/architecture-examples/angularjs/js/directives/todoBlur.js: Removed.
20         * Speedometer/resources/todomvc/architecture-examples/angularjs/js/directives/todoEscape.js:
21         * Speedometer/resources/todomvc/architecture-examples/angularjs/js/directives/todoFocus.js:
22         * Speedometer/resources/todomvc/architecture-examples/angularjs/js/services/todoStorage.js:
23         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/angular-resource/angular-resource.js: Added.
24         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/angular-route/angular-route.js: Added.
25         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/angular/angular.js: Added.
26         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/todomvc-app-css/index.css: Added.
27         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/todomvc-app-css/package.json: Added.
28         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/todomvc-app-css/readme.md: Added.
29         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/todomvc-common/base.css: Added.
30         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/todomvc-common/base.js: Added.
31         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/todomvc-common/package.json: Added.
32         * Speedometer/resources/todomvc/architecture-examples/angularjs/node_modules/todomvc-common/readme.md: Added.
33         * Speedometer/resources/todomvc/architecture-examples/angularjs/package.json: Added.
34         * Speedometer/resources/todomvc/architecture-examples/angularjs/readme.md:
35         * Speedometer/resources/todomvc/architecture-examples/angularjs/test/config/testacular.conf.js: Removed.
36         * Speedometer/resources/todomvc/architecture-examples/angularjs/test/package.json: Removed.
37         * Speedometer/resources/todomvc/architecture-examples/angularjs/test/readme.md: Removed.
38         * Speedometer/resources/todomvc/architecture-examples/angularjs/test/unit/directivesSpec.js: Removed.
39         * Speedometer/resources/todomvc/architecture-examples/angularjs/test/unit/todoCtrlSpec.js: Removed.
40         * Speedometer/resources/todomvc/architecture-examples/angularjs/todomvc-index.html: Added.
41
42 2017-05-09  Jon Davis  <jond@apple.com>
43
44         [ARES6] Improved the running test indicator
45         https://bugs.webkit.org/show_bug.cgi?id=171400
46         
47         Running tests show a triangle before the test title and the titles are white.
48
49         Reviewed by Joseph Pecoraro.
50
51         * ARES-6/index.html:
52         * ARES-6/styles.css:
53         (.test .running):
54         (.test .running:before):
55         (.test .indicator): Deleted.
56         (.test .indicator.running): Deleted.
57         (@keyframes test-running): Deleted.
58
59 2017-05-04  Andy VanWagoner  <thetalecrafter@gmail.com>
60
61         [INTL] Add PerformanceTests for Intl objects
62         https://bugs.webkit.org/show_bug.cgi?id=171695
63
64         Reviewed by Benjamin Poulain.
65
66         * Intl/collator-compare-all-options.html: Added.
67         * Intl/collator-compare-default.html: Added.
68         * Intl/collator-create-all-options.html: Added.
69         * Intl/collator-create-default.html: Added.
70         * Intl/datetimeformat-create-all-options.html: Added.
71         * Intl/datetimeformat-create-default.html: Added.
72         * Intl/datetimeformat-format-all-options.html: Added.
73         * Intl/datetimeformat-format-default.html: Added.
74         * Intl/numberformat-create-all-options.html: Added.
75         * Intl/numberformat-create-default.html: Added.
76         * Intl/numberformat-format-all-options.html: Added.
77         * Intl/numberformat-format-default.html: Added.
78
79 2017-04-27  Michael Saboff  <msaboff@apple.com>
80
81         Add back tests inadvertently deleted in r205032.
82
83         Rubber stamped by Geoffrey Garen.
84
85         * MallocBench/run-malloc-benchmarks:
86
87 2017-04-25  Myles C. Maxfield  <mmaxfield@apple.com>
88
89         Add performance test for FontCache::systemFallbackForCharacters()
90         https://bugs.webkit.org/show_bug.cgi?id=170842
91
92         Reviewed by Tim Horton.
93
94         * Layout/word-joiner.html: Added.
95
96 2017-04-24  Saam Barati  <sbarati@apple.com>
97
98         Add ML to ARES6
99         https://bugs.webkit.org/show_bug.cgi?id=171206
100
101         Rubber stamped by Filip Pizlo.
102
103         This patch adds a new test to ARES6 called ML. ML is an implementation of
104         a feedforward neural network: https://github.com/mljs/feedforward-neural-networks.
105         It makes heavy use of classes, and does non-trivial matrix math using the
106         ml-matrix library: https://github.com/mljs/matrix
107
108         * ARES-6/about.html:
109         * ARES-6/cli.js:
110         * ARES-6/glue.js:
111         * ARES-6/index.html:
112         * ARES-6/ml: Added.
113         * ARES-6/ml/benchmark.js: Added.
114         * ARES-6/ml/index.js: Added.
115         * ARES-6/ml_benchmark.js: Added.
116
117 2017-04-21  Zalan Bujtas  <zalan@apple.com>
118
119         Simple line layout: Add performance test to measure mid-word line breaking with long text.
120         https://bugs.webkit.org/show_bug.cgi?id=171136
121
122         Reviewed by Ryosuke Niwa.
123
124         * Layout/simple-line-layout-word-break-long-text.html: Added.
125
126 2017-04-17  Ryan Haddad  <ryanhaddad@apple.com>
127
128         Unreviewed, rolling out r215366.
129
130         This test is failing on performance bots.
131
132         Reverted changeset:
133
134         "Add performance test for asking the platform for a font for
135         U+2060 WORD JOINER"
136         https://bugs.webkit.org/show_bug.cgi?id=170842
137         http://trac.webkit.org/changeset/215366
138
139 2017-04-14  Myles C. Maxfield  <mmaxfield@apple.com>
140
141         Add performance test for asking the platform for a font for U+2060 WORD JOINER
142         https://bugs.webkit.org/show_bug.cgi?id=170842
143
144         Reviewed by Tim Horton.
145
146         * Layout/word-joiner.html: Added.
147
148 2017-04-03  Saam Barati  <sbarati@apple.com>
149
150         Add a new test to ARES6
151         https://bugs.webkit.org/show_bug.cgi?id=170077
152
153         Rubber stamped by Filip Pizlo.
154
155         This patch adds a new test to ARES6. The test is called Babylon. Babylon is
156         the JavaScript parser that is part of the Babel transpiler. The test runs
157         the parser on 4 different JS scripts.
158
159         * ARES-6/Babylon: Added.
160         * ARES-6/Babylon/AUTHORS: Added.
161         * ARES-6/Babylon/air-blob.js: Added.
162         (Reg):
163         (Reg.fromReg):
164         (Reg.prototype.get index):
165         (Reg.prototype.get type):
166         (Reg.prototype.get name):
167         (Reg.prototype.get isCalleeSave):
168         (Reg.prototype.get isReg):
169         (Reg.prototype.hash):
170         (Reg.prototype.toString):
171         (Reg.extract):
172         (Reg.forEachFast):
173         (Reg.forEach):
174         (newGPR):
175         (Reg.gprs.Reg.fprs.Reg.calleeSaveGPRs.Reg.calleeSaveFPRs.Reg.calleeSaves):
176         * ARES-6/Babylon/babylon-blob.js: Added.
177         (export.default.Parser):
178         (export.default.Parser.prototype.isReservedWord):
179         (export.default.Parser.prototype.hasPlugin):
180         (export.default.Parser.prototype.extend):
181         (export.default.Parser.prototype.loadPlugins):
182         (export.default.Parser.prototype.parse):
183         * ARES-6/Babylon/basic-blob.js: Added.
184         (Basic.NumberApply):
185         (Basic.Variable):
186         (Basic.Const):
187         (Basic.NumberPow):
188         (Basic.NumberMul):
189         (Basic.NumberDiv):
190         (Basic.NumberNeg):
191         (Basic.NumberAdd):
192         (Basic.NumberSub):
193         (Basic.StringVar):
194         (Basic.Equals):
195         (Basic.NotEquals):
196         (Basic.LessThan):
197         (Basic.GreaterThan):
198         (Basic.LessEqual):
199         (Basic.GreaterEqual):
200         (Basic.GoTo):
201         (Basic.GoSub):
202         (Basic.Def):
203         (Basic.Let):
204         (Basic.If):
205         (Basic.Return):
206         (Basic.Stop):
207         (Basic.On):
208         (sideState.shouldStop):
209         (Basic.For):
210         (Basic.Next):
211         (Basic.Print):
212         (Basic.Input):
213         (Basic.Read):
214         (Basic.Restore):
215         (Basic.Dim):
216         (Basic.Randomize):
217         (Basic.End):
218         (Basic.Program):
219         * ARES-6/Babylon/benchmark.js: Added.
220         (this.performance.performance.now.currentTime):
221         (else.this.preciseTime.currentTime):
222         (else.currentTime):
223         (BabylonBenchmark.appendSource):
224         (BabylonBenchmark):
225         (BabylonBenchmark.prototype.runIteration.parse):
226         (BabylonBenchmark.prototype.runIteration.parseExpression):
227         (BabylonBenchmark.prototype.runIteration):
228         (runBenchmark):
229         * ARES-6/Babylon/index.js: Added.
230         (isIdentifierStart):
231         (isIdentifierChar):
232         (isNewLine):
233         (Position):
234         (SourceLocation):
235         (getLineInfo):
236         (TokenType):
237         (KeywordTokenType):
238         (BinopTokenType):
239         (TokContext):
240         (tt.parenR.updateContext.tt.braceR.updateContext):
241         (tt.name.updateContext):
242         (tt.braceL.updateContext):
243         (tt.dollarBraceL.updateContext):
244         (tt.parenL.updateContext):
245         (tt.incDec.updateContext):
246         (tt._function.updateContext):
247         (tt.backQuote.updateContext):
248         (State.prototype.init):
249         (State.prototype.curPosition):
250         (State.prototype.clone):
251         (State):
252         (Token):
253         (codePointToString):
254         (Tokenizer):
255         (Tokenizer.prototype.next):
256         (Tokenizer.prototype.eat):
257         (Tokenizer.prototype.match):
258         (Tokenizer.prototype.isKeyword):
259         (Tokenizer.prototype.lookahead):
260         (Tokenizer.prototype.setStrict):
261         (Tokenizer.prototype.curContext):
262         (Tokenizer.prototype.nextToken):
263         (Tokenizer.prototype.readToken):
264         (Tokenizer.prototype.fullCharCodeAtPos):
265         (Tokenizer.prototype.pushComment):
266         (Tokenizer.prototype.skipBlockComment):
267         (Tokenizer.prototype.skipLineComment):
268         (Tokenizer.prototype.skipSpace):
269         (Tokenizer.prototype.finishToken):
270         (Tokenizer.prototype.readToken_dot):
271         (Tokenizer.prototype.readToken_slash):
272         (Tokenizer.prototype.readToken_mult_modulo):
273         (Tokenizer.prototype.readToken_pipe_amp):
274         (Tokenizer.prototype.readToken_caret):
275         (Tokenizer.prototype.readToken_plus_min):
276         (Tokenizer.prototype.readToken_lt_gt):
277         (Tokenizer.prototype.readToken_eq_excl):
278         (Tokenizer.prototype.getTokenFromCode):
279         (Tokenizer.prototype.finishOp):
280         (Tokenizer.prototype.readRegexp):
281         (Tokenizer.prototype.readInt):
282         (Tokenizer.prototype.readRadixNumber):
283         (Tokenizer.prototype.readNumber):
284         (Tokenizer.prototype.readCodePoint):
285         (Tokenizer.prototype.readString):
286         (Tokenizer.prototype.readTmplToken):
287         (Tokenizer.prototype.readEscapedChar):
288         (Tokenizer.prototype.readHexChar):
289         (Tokenizer.prototype.readWord1):
290         (Tokenizer.prototype.readWord):
291         (Tokenizer.prototype.braceIsBlock):
292         (Tokenizer.prototype.updateContext):
293         (getOptions):
294         (prototype.isReservedWord):
295         (prototype.hasPlugin):
296         (prototype.extend):
297         (prototype.loadPlugins):
298         (prototype.parse):
299         (last):
300         (pp.addComment):
301         (pp.processComment):
302         (pp.checkPropClash):
303         (pp.getExpression):
304         (pp.parseExpression):
305         (pp.parseMaybeAssign):
306         (pos.set i):
307         * ARES-6/Babylon/inspector-blob.js: Added.
308         (WebInspector.ScriptSyntaxTree):
309         (WebInspector.ScriptSyntaxTree.prototype.get parsedSuccessfully):
310         (WebInspector.ScriptSyntaxTree.prototype.forEachNode):
311         (WebInspector.ScriptSyntaxTree.prototype.filter):
312         (WebInspector.ScriptSyntaxTree.prototype.containersOfOffset):
313         (WebInspector.ScriptSyntaxTree.prototype.filterByRange.filterForNodesInRange):
314         (WebInspector.ScriptSyntaxTree.prototype.filterByRange):
315         (WebInspector.ScriptSyntaxTree.prototype.containsNonEmptyReturnStatement.removeFunctionsFilter):
316         (WebInspector.ScriptSyntaxTree.prototype.containsNonEmptyReturnStatement):
317         (WebInspector.ScriptSyntaxTree.functionReturnDivot):
318         (WebInspector.ScriptSyntaxTree.prototype.updateTypes):
319         (WebInspector.ScriptSyntaxTree.prototype._gatherIdentifiersInDeclaration.gatherIdentifiers):
320         (WebInspector.ScriptSyntaxTree.prototype._gatherIdentifiersInDeclaration):
321         (WebInspector.ScriptSyntaxTree.prototype._defaultParserState):
322         (WebInspector.ScriptSyntaxTree.prototype._recurse):
323         (WebInspector.ScriptSyntaxTree.prototype._recurseArray):
324         (WebInspector.ScriptSyntaxTree.prototype._createInternalSyntaxTree):
325         * ARES-6/about.html:
326         * ARES-6/babylon_benchmark.js: Added.
327         * ARES-6/cli.js:
328         * ARES-6/glue.js:
329         * ARES-6/index.html:
330         * ARES-6/results.js:
331         (Results.prototype.reportResult):
332         (Results.prototype.reportResult.averageAbovePercentile): Deleted.
333         * ARES-6/styles.css:
334         (.test):
335         (@media only screen and (max-width: 784px)):
336
337 2017-03-29  Zalan Bujtas <zalan@apple.com>
338
339         [Arabic text] Add whitespace preserve/word wrap performance test.
340         https://bugs.webkit.org/show_bug.cgi?id=170243
341
342         Reviewed by Simon Fraser.
343
344         * Layout/ArabicLineLayout-pre-and-wrap.html: Added.
345
346 2017-03-08  Yusuke Suzuki  <utatane.tea@gmail.com>
347
348         [JSC] Add helpers to run Six Speed to run-jsc-benchmarks
349         https://bugs.webkit.org/show_bug.cgi?id=169351
350
351         Reviewed by Saam Barati.
352
353         * SixSpeed/wrappers/wrapper.js: Added.
354         (assertEqual):
355         (test):
356         (jscRun):
357
358 2017-03-08  Jon Davis  <jond@apple.com>
359
360         Fixed ARES-6 animations for other browsers
361
362         Rubber stamped by Saam Barati.
363
364         * ARES-6/styles.css:
365         (.swoop):
366         (.logo):
367
368 2017-03-07  Filip Pizlo  <fpizlo@apple.com>
369
370         Revise the ARES-6 explainer text
371         https://bugs.webkit.org/show_bug.cgi?id=169287
372
373         Reviewed by Saam Barati.
374         
375         Just did some copy-editing.
376
377         * ARES-6/about.html:
378
379 2017-03-06  Jon Davis  <jond@apple.com>
380
381         Refined the ARES-6 experience
382         https://bugs.webkit.org/show_bug.cgi?id=169205
383
384         Reviewed by Saam Barati.
385
386         * ARES-6/about.html: Copied from PerformanceTests/ARES-6/index.html.
387         * ARES-6/driver.js:
388         (Driver.prototype.start):
389         (Driver.prototype._updateIterations):
390         (Driver):
391         * ARES-6/index.html:
392         * ARES-6/stats.js:
393         (Stats.prototype.toString.span.span):
394         * ARES-6/styles.css:
395         (body):
396         (.swoop):
397         (p):
398         (.logo):
399         (.start):
400         (.test):
401         (.test:nth-child(odd)):
402         (.overall):
403         (.about h2):
404         (.button):
405         (.button:hover):
406         (.button.return:before):
407         (@keyframes fade-in):
408         (@keyframes scale-in):
409         (@keyframes flicker-in):
410         (@keyframes wipe-in):
411         (.flip): Deleted.
412         (.testing, .about): Deleted.
413         (.testing): Deleted.
414         (.about): Deleted.
415         (#about:target): Deleted.
416
417 2017-03-06  Jeremy Jones  <jeremyj@apple.com>
418
419         StandardCanPlayThrough.html does not work on iOS.
420         https://bugs.webkit.org/show_bug.cgi?id=169207
421
422         Reviewed by Jer Noble.
423
424         On iOS only, add muted=true and autoplay=allowed. The loading policy on iOS requires these attributes to load enough
425         data to trigger canPlayThrough.
426
427         * Media/StandardCanPlayThrough.html:
428
429 2017-02-13  Zalan Bujtas <zalan@apple.com>
430
431         Simple line layout: Add performance test with text-rendering: optimizeSpeed
432         https://bugs.webkit.org/show_bug.cgi?id=168248
433
434         Reviewed by Simon Fraser.
435
436         * Layout/simple-line-layout-with-varying-content-and-optimized-speed.html: Added.
437
438 2017-02-10  Zalan Bujtas  <zalan@apple.com>
439
440         Mail hangs when removing multiple rows from large table.
441         https://bugs.webkit.org/show_bug.cgi?id=168103
442         <rdar://problem/30090186>
443
444         Reviewed by Ryosuke Niwa.
445
446         * DOM/large-table-edit.html: Added.
447
448 2017-02-05  Filip Pizlo  <fpizlo@apple.com>
449
450         Beef up the ARES-6 explainer text
451         https://bugs.webkit.org/show_bug.cgi?id=167867
452
453         Reviewed by Saam Barati.
454         
455         Added appropriate links.
456
457         Added text that explains the subscores.
458
459         * ARES-6/index.html:
460
461 2017-02-06  Jer Noble  <jer.noble@apple.com>
462
463         Playback stalls when a SourceBuffer append causes frame eviction
464         https://bugs.webkit.org/show_bug.cgi?id=167834
465
466         Reviewed by Eric Carlson.
467
468         Add an in-page performance test measuring the amount of time required
469         to append a large amount of media data to a SourceBuffer, and then to
470         completely remove that data 30s at a time.
471
472         Add a microbenchmark for MediaTime which measures the amount of time
473         required to create a 1M entry std::map and traverse the map 1M times.
474
475         * Media/MSERemoveCodedFrames.html: Added.
476         * Media/media-source-loader.js:
477         (MediaSourceLoader.prototype.get duration):
478         * MediaTime/Configurations/Base.xcconfig: Added.
479         * MediaTime/Configurations/DebugRelease.xcconfig: Added.
480         * MediaTime/Makefile: Added.
481         * MediaTime/MediaTime.xcodeproj/project.pbxproj: Added.
482         * MediaTime/main.cpp: Added.
483         (performTest):
484         (test):
485         (main):
486         * Skipped:
487
488 2017-02-06  Saam Barati  <sbarati@apple.com>
489
490         Make ARES-6 work from the CLI again
491         https://bugs.webkit.org/show_bug.cgi?id=167895
492
493         Reviewed by Michael Saboff.
494
495         * ARES-6/driver.js:
496         (Driver.prototype.readyTrigger):
497         (Driver.prototype.disableTrigger):
498         * ARES-6/stats.js:
499         (Stats.prototype.toString.span.span):
500         (Stats.prototype.toString.return.span):
501         (Stats.prototype.toString):
502
503 2017-02-06  Ryan Haddad  <ryanhaddad@apple.com>
504
505         Skip ARES-6 on performance bots
506         https://bugs.webkit.org/show_bug.cgi?id=167863
507
508         Unreviewed test gardening.
509
510         ES6SampleBench was renamed to ARES-6, so this should be updated in PerformanceTests/Skipped as well.
511
512         * Skipped:
513
514 2017-02-05  Filip Pizlo  <fpizlo@apple.com>
515
516         Change ES6SampleBench into ARES-6
517         https://bugs.webkit.org/show_bug.cgi?id=167863
518
519         Rubber stamped by Saam Barati.
520         
521         This imports changes that Jonathan Davis (jond@apple.com) made to ES6SampleBench to make it look
522         super awesome. Its now called ARES-6.
523
524         * ARES-6: Copied from PerformanceTests/ES6SampleBench.
525         * ARES-6/ARES-6.svg: Added.
526         * ARES-6/driver.js:
527         (Driver):
528         (Driver.prototype.readyTrigger):
529         (Driver.prototype.disableTrigger):
530         (Driver.prototype.start):
531         (Driver.prototype._updateIterations):
532         * ARES-6/glue.js:
533         * ARES-6/index.html:
534         * ARES-6/results.js:
535         (Results.prototype.reportRunning):
536         (Results.prototype.reportDone):
537         * ARES-6/stats.js:
538         (Stats.prototype.toString.return.span):
539         (Stats.prototype.toString.span.span):
540         (Stats.prototype.toString): Deleted.
541         * ARES-6/style.css: Removed.
542         * ARES-6/styles.css: Added.
543         (html):
544         (body):
545         (header,):
546         (p a):
547         (p a:hover):
548         (.flip):
549         (header):
550         (h2,):
551         (p):
552         (.about h2):
553         (.logo):
554         (.start):
555         (.start.ready):
556         (.start.ready:hover):
557         (.start:after):
558         (.start.ready .testrun):
559         (.tests):
560         (.test):
561         (.overall h2,):
562         (.test .indicator):
563         (.test .indicator.running):
564         (@keyframes test-running):
565         (100%):
566         (.score):
567         (.score label):
568         (.score .value):
569         (.score .units):
570         (.score .margin):
571         (.overall):
572         (.overall .score):
573         (.overall .margin):
574         (.testing, .about):
575         (.testing):
576         (.about):
577         (#about:target):
578         (@media only screen and (max-width: 784px)):
579         * ARES-6/swoop.svg: Added.
580         * ES6SampleBench: Removed.
581         * ES6SampleBench/Air: Removed.
582         * ES6SampleBench/Air/README.md: Removed.
583         * ES6SampleBench/Air/airjs-tests.yaml: Removed.
584         * ES6SampleBench/Air/all.js: Removed.
585         * ES6SampleBench/Air/allocate_stack.js: Removed.
586         * ES6SampleBench/Air/arg.js: Removed.
587         * ES6SampleBench/Air/basic_block.js: Removed.
588         * ES6SampleBench/Air/benchmark.js: Removed.
589         * ES6SampleBench/Air/code.js: Removed.
590         * ES6SampleBench/Air/custom.js: Removed.
591         * ES6SampleBench/Air/frequented_block.js: Removed.
592         * ES6SampleBench/Air/insertion_set.js: Removed.
593         * ES6SampleBench/Air/inst.js: Removed.
594         * ES6SampleBench/Air/liveness.js: Removed.
595         * ES6SampleBench/Air/make_dist.sh: Removed.
596         * ES6SampleBench/Air/opcode.js: Removed.
597         * ES6SampleBench/Air/payload-airjs-ACLj8C.js: Removed.
598         * ES6SampleBench/Air/payload-gbemu-executeIteration.js: Removed.
599         * ES6SampleBench/Air/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
600         * ES6SampleBench/Air/payload-typescript-scanIdentifier.js: Removed.
601         * ES6SampleBench/Air/reg.js: Removed.
602         * ES6SampleBench/Air/stack_slot.js: Removed.
603         * ES6SampleBench/Air/stress-test.js: Removed.
604         * ES6SampleBench/Air/strip-hash.rb: Removed.
605         * ES6SampleBench/Air/symbols.js: Removed.
606         * ES6SampleBench/Air/test.html: Removed.
607         * ES6SampleBench/Air/test.js: Removed.
608         * ES6SampleBench/Air/tmp.js: Removed.
609         * ES6SampleBench/Air/tmp_base.js: Removed.
610         * ES6SampleBench/Air/util.js: Removed.
611         * ES6SampleBench/Basic: Removed.
612         * ES6SampleBench/Basic/ast.js: Removed.
613         * ES6SampleBench/Basic/basic-tests.yaml: Removed.
614         * ES6SampleBench/Basic/basic.js: Removed.
615         * ES6SampleBench/Basic/benchmark.js: Removed.
616         * ES6SampleBench/Basic/caseless_map.js: Removed.
617         * ES6SampleBench/Basic/lexer.js: Removed.
618         * ES6SampleBench/Basic/number.js: Removed.
619         * ES6SampleBench/Basic/parser.js: Removed.
620         * ES6SampleBench/Basic/random.js: Removed.
621         * ES6SampleBench/Basic/state.js: Removed.
622         * ES6SampleBench/Basic/stress-test.js: Removed.
623         * ES6SampleBench/Basic/test.html: Removed.
624         * ES6SampleBench/Basic/test.js: Removed.
625         * ES6SampleBench/Basic/util.js: Removed.
626         * ES6SampleBench/air_benchmark.js: Removed.
627         * ES6SampleBench/basic_benchmark.js: Removed.
628         * ES6SampleBench/cli.js: Removed.
629         * ES6SampleBench/driver.js: Removed.
630         * ES6SampleBench/glue.js: Removed.
631         * ES6SampleBench/index.html: Removed.
632         * ES6SampleBench/results.js: Removed.
633         * ES6SampleBench/stats.js: Removed.
634         * ES6SampleBench/style.css: Removed.
635
636 2017-02-05  Zalan Bujtas  <zalan@apple.com>
637
638         Simple line layout: Add performance test with varying string and font-size to minimize width-cache hit.
639         https://bugs.webkit.org/show_bug.cgi?id=167859
640         <rdar://problem/30368070>
641
642         Reviewed by Antti Koivisto.
643
644         This is in preparation for webkit.org/b/167843.
645
646         * Layout/simple-line-layout-with-varying-content.html: Added.
647
648 2017-01-16  Filip Pizlo  <fpizlo@apple.com>
649
650         Add framerate baselines to the splay-detail test
651         https://bugs.webkit.org/show_bug.cgi?id=167659
652
653         Reviewed by Geoffrey Garen.
654
655         * JetStream/Octane2/splay-detail.html:
656
657 2017-01-31  Carlos Alberto Lopez Perez  <clopez@igalia.com>
658
659         [EFL][GTK] Skip some IndexedDB tests that time out on the Perf bots.
660
661         Unreviewed performance test gardening.
662
663         * Skipped:
664
665 2017-01-30  Zalan Bujtas  <zalan@apple.com>
666
667         Simple line layout: Small tweaks to improve performance.
668         https://bugs.webkit.org/show_bug.cgi?id=167611
669         <rdar://problem/30274294>
670
671         Reviewed by Simon Fraser.
672
673         * Layout/simple-line-layout-non-repeating-text.html: Added.
674
675 2017-01-13  Said Abou-Hallawa  <sabouhallawa@apple.com>
676
677         26 MotionMark performance tests failing
678         https://bugs.webkit.org/show_bug.cgi?id=166854
679
680         Reviewed by Ryosuke Niwa.
681         
682         Skip running MotionMark on WebKit perf bots.
683
684         * Skipped:
685
686 2017-01-06  Said Abou-Hallawa  <sabouhallawa@apple.com>
687
688         Rename the directory and the files of MotionMark from Animometer to MotionMark
689         https://bugs.webkit.org/show_bug.cgi?id=166659
690
691         Reviewed by Jon Lee.
692         
693         Rename the directory and the files of the benchmark to its new name.
694
695         * MotionMark/developer.html: Renamed from PerformanceTests/Animometer/developer.html.
696         * MotionMark/index.html: Renamed from PerformanceTests/Animometer/index.html.
697         * MotionMark/resources/debug-runner/d3.min.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/d3.min.js.
698         * MotionMark/resources/debug-runner/graph.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/graph.js.
699         * MotionMark/resources/debug-runner/motionmark.css: Renamed from PerformanceTests/Animometer/resources/debug-runner/animometer.css.
700         * MotionMark/resources/debug-runner/motionmark.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/animometer.js.
701         * MotionMark/resources/debug-runner/tests.js: Renamed from PerformanceTests/Animometer/resources/debug-runner/tests.js.
702         * MotionMark/resources/extensions.js: Renamed from PerformanceTests/Animometer/resources/extensions.js.
703         * MotionMark/resources/runner/benchmark-runner.js: Renamed from PerformanceTests/Animometer/resources/runner/benchmark-runner.js.
704         * MotionMark/resources/runner/crystal.svg: Renamed from PerformanceTests/Animometer/resources/runner/crystal.svg.
705         * MotionMark/resources/runner/lines.svg: Renamed from PerformanceTests/Animometer/resources/runner/lines.svg.
706         * MotionMark/resources/runner/logo.svg: Renamed from PerformanceTests/Animometer/resources/runner/logo.svg.
707         * MotionMark/resources/runner/motionmark.css: Renamed from PerformanceTests/Animometer/resources/runner/animometer.css.
708         * MotionMark/resources/runner/motionmark.js: Renamed from PerformanceTests/Animometer/resources/runner/animometer.js.
709         * MotionMark/resources/runner/tests.js: Renamed from PerformanceTests/Animometer/resources/runner/tests.js.
710         * MotionMark/resources/statistics.js: Renamed from PerformanceTests/Animometer/resources/statistics.js.
711         * MotionMark/resources/strings.js: Renamed from PerformanceTests/Animometer/resources/strings.js.
712         * MotionMark/tests/3d/resources/webgl.js: Renamed from PerformanceTests/Animometer/tests/3d/resources/webgl.js.
713         * MotionMark/tests/3d/webgl.html: Renamed from PerformanceTests/Animometer/tests/3d/webgl.html.
714         * MotionMark/tests/bouncing-particles/bouncing-canvas-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-canvas-images.html.
715         * MotionMark/tests/bouncing-particles/bouncing-canvas-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html.
716         * MotionMark/tests/bouncing-particles/bouncing-css-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-css-images.html.
717         * MotionMark/tests/bouncing-particles/bouncing-css-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-css-shapes.html.
718         * MotionMark/tests/bouncing-particles/bouncing-svg-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-svg-images.html.
719         * MotionMark/tests/bouncing-particles/bouncing-svg-shapes.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-svg-shapes.html.
720         * MotionMark/tests/bouncing-particles/bouncing-tagged-images.html: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/bouncing-tagged-images.html.
721         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-images.js.
722         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-particles.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-particles.js.
723         * MotionMark/tests/bouncing-particles/resources/bouncing-canvas-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-canvas-shapes.js.
724         * MotionMark/tests/bouncing-particles/resources/bouncing-css-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-css-images.js.
725         * MotionMark/tests/bouncing-particles/resources/bouncing-css-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-css-shapes.js.
726         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-images.js.
727         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-particles.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-particles.js.
728         * MotionMark/tests/bouncing-particles/resources/bouncing-svg-shapes.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-svg-shapes.js.
729         * MotionMark/tests/bouncing-particles/resources/bouncing-tagged-images.js: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js.
730         * MotionMark/tests/bouncing-particles/resources/image1.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image1.jpg.
731         * MotionMark/tests/bouncing-particles/resources/image2.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image2.jpg.
732         * MotionMark/tests/bouncing-particles/resources/image3.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image3.jpg.
733         * MotionMark/tests/bouncing-particles/resources/image4.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image4.jpg.
734         * MotionMark/tests/bouncing-particles/resources/image5.jpg: Renamed from PerformanceTests/Animometer/tests/bouncing-particles/resources/image5.jpg.
735         * MotionMark/tests/dom/compositing-transforms.html: Renamed from PerformanceTests/Animometer/tests/dom/compositing-transforms.html.
736         * MotionMark/tests/dom/focus.html: Renamed from PerformanceTests/Animometer/tests/dom/focus.html.
737         * MotionMark/tests/dom/leaves.html: Renamed from PerformanceTests/Animometer/tests/dom/leaves.html.
738         * MotionMark/tests/dom/particles.html: Renamed from PerformanceTests/Animometer/tests/dom/particles.html.
739         * MotionMark/tests/dom/resources/compositing-transforms.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/compositing-transforms.js.
740         * MotionMark/tests/dom/resources/dom-particles.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/dom-particles.js.
741         * MotionMark/tests/dom/resources/focus.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/focus.js.
742         * MotionMark/tests/dom/resources/leaves.js: Renamed from PerformanceTests/Animometer/tests/dom/resources/leaves.js.
743         * MotionMark/tests/master/canvas-stage.html: Renamed from PerformanceTests/Animometer/tests/master/canvas-stage.html.
744         * MotionMark/tests/master/focus.html: Renamed from PerformanceTests/Animometer/tests/master/focus.html.
745         * MotionMark/tests/master/image-data.html: Renamed from PerformanceTests/Animometer/tests/master/image-data.html.
746         * MotionMark/tests/master/leaves.html: Renamed from PerformanceTests/Animometer/tests/master/leaves.html.
747         * MotionMark/tests/master/multiply.html: Renamed from PerformanceTests/Animometer/tests/master/multiply.html.
748         * MotionMark/tests/master/resources/canvas-stage.js: Renamed from PerformanceTests/Animometer/tests/master/resources/canvas-stage.js.
749         * MotionMark/tests/master/resources/canvas-tests.js: Renamed from PerformanceTests/Animometer/tests/master/resources/canvas-tests.js.
750         * MotionMark/tests/master/resources/compass.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/compass.svg.
751         * MotionMark/tests/master/resources/compass100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/compass100.png.
752         * MotionMark/tests/master/resources/console.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/console.svg.
753         * MotionMark/tests/master/resources/console100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/console100.png.
754         * MotionMark/tests/master/resources/contribute.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/contribute.svg.
755         * MotionMark/tests/master/resources/contribute100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/contribute100.png.
756         * MotionMark/tests/master/resources/debugger.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/debugger.svg.
757         * MotionMark/tests/master/resources/debugger100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/debugger100.png.
758         * MotionMark/tests/master/resources/focus.js: Renamed from PerformanceTests/Animometer/tests/master/resources/focus.js.
759         * MotionMark/tests/master/resources/image-data.js: Renamed from PerformanceTests/Animometer/tests/master/resources/image-data.js.
760         * MotionMark/tests/master/resources/inspector.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/inspector.svg.
761         * MotionMark/tests/master/resources/inspector100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/inspector100.png.
762         * MotionMark/tests/master/resources/layout.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/layout.svg.
763         * MotionMark/tests/master/resources/layout100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/layout100.png.
764         * MotionMark/tests/master/resources/leaves.js: Renamed from PerformanceTests/Animometer/tests/master/resources/leaves.js.
765         * MotionMark/tests/master/resources/multiply.js: Renamed from PerformanceTests/Animometer/tests/master/resources/multiply.js.
766         * MotionMark/tests/master/resources/particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/particles.js.
767         * MotionMark/tests/master/resources/performance.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/performance.svg.
768         * MotionMark/tests/master/resources/performance100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/performance100.png.
769         * MotionMark/tests/master/resources/script.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/script.svg.
770         * MotionMark/tests/master/resources/script100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/script100.png.
771         * MotionMark/tests/master/resources/shortcuts.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/shortcuts.svg.
772         * MotionMark/tests/master/resources/shortcuts100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/shortcuts100.png.
773         * MotionMark/tests/master/resources/standards.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/standards.svg.
774         * MotionMark/tests/master/resources/standards100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/standards100.png.
775         * MotionMark/tests/master/resources/storage.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/storage.svg.
776         * MotionMark/tests/master/resources/storage100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/storage100.png.
777         * MotionMark/tests/master/resources/styles.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/styles.svg.
778         * MotionMark/tests/master/resources/styles100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/styles100.png.
779         * MotionMark/tests/master/resources/svg-particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/svg-particles.js.
780         * MotionMark/tests/master/resources/text.js: Renamed from PerformanceTests/Animometer/tests/master/resources/text.js.
781         * MotionMark/tests/master/resources/timeline.svg: Renamed from PerformanceTests/Animometer/tests/master/resources/timeline.svg.
782         * MotionMark/tests/master/resources/timeline100.png: Renamed from PerformanceTests/Animometer/tests/master/resources/timeline100.png.
783         * MotionMark/tests/master/svg-particles.html: Renamed from PerformanceTests/Animometer/tests/master/svg-particles.html.
784         * MotionMark/tests/master/text.html: Renamed from PerformanceTests/Animometer/tests/master/text.html.
785         * MotionMark/tests/resources/main.js: Renamed from PerformanceTests/Animometer/tests/resources/main.js.
786         * MotionMark/tests/resources/math.js: Renamed from PerformanceTests/Animometer/tests/resources/math.js.
787         * MotionMark/tests/resources/stage.css: Renamed from PerformanceTests/Animometer/tests/resources/stage.css.
788         * MotionMark/tests/resources/star.svg: Renamed from PerformanceTests/Animometer/tests/resources/star.svg.
789         * MotionMark/tests/resources/yin-yang.png: Renamed from PerformanceTests/Animometer/tests/resources/yin-yang.png.
790         * MotionMark/tests/resources/yin-yang.svg: Renamed from PerformanceTests/Animometer/tests/resources/yin-yang.svg.
791         * MotionMark/tests/simple/resources/simple-canvas-paths.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/simple-canvas-paths.js.
792         * MotionMark/tests/simple/resources/simple-canvas.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/simple-canvas.js.
793         * MotionMark/tests/simple/resources/tiled-canvas-image.js: Renamed from PerformanceTests/Animometer/tests/simple/resources/tiled-canvas-image.js.
794         * MotionMark/tests/simple/simple-canvas-paths.html: Renamed from PerformanceTests/Animometer/tests/simple/simple-canvas-paths.html.
795         * MotionMark/tests/simple/tiled-canvas-image.html: Renamed from PerformanceTests/Animometer/tests/simple/tiled-canvas-image.html.
796         * MotionMark/tests/template/resources/template-canvas.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-canvas.js.
797         * MotionMark/tests/template/resources/template-css.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-css.js.
798         * MotionMark/tests/template/resources/template-svg.js: Renamed from PerformanceTests/Animometer/tests/template/resources/template-svg.js.
799         * MotionMark/tests/template/template-canvas.html: Renamed from PerformanceTests/Animometer/tests/template/template-canvas.html.
800         * MotionMark/tests/template/template-css.html: Renamed from PerformanceTests/Animometer/tests/template/template-css.html.
801         * MotionMark/tests/template/template-svg.html: Renamed from PerformanceTests/Animometer/tests/template/template-svg.html.
802
803 2016-12-13  Brady Eidson  <beidson@apple.com>
804
805         Add a PerfTest targeting IDBObjectStore.get().
806         https://bugs.webkit.org/show_bug.cgi?id=165816
807
808         Reviewed by Alex Christensen.
809
810         * IndexedDB/objectstore-get.html: Added.
811
812 2016-12-13  Brady Eidson  <beidson@apple.com>
813
814         Add a PerfTest targeting IDBIndex.get().
815         https://bugs.webkit.org/show_bug.cgi?id=165803
816
817         Reviewed by Alex Christensen.
818
819         * IndexedDB/index-get.html: Added.
820
821 2016-12-12  Brady Eidson  <beidson@apple.com>
822
823         More IndexedDB perf tests.
824         https://bugs.webkit.org/show_bug.cgi?id=165634
825
826         Reviewed by Sam Weinig.
827
828         * IndexedDB/index-multientry.html:
829         * IndexedDB/large-number-of-inserts-responsiveness.html: Added.
830         * IndexedDB/large-number-of-inserts.html: Added.
831         * IndexedDB/objectstore-cursor.html: Added. Runtime cut in 1/4th compared to r209672.
832
833         * resources/runner.js:
834
835 2016-12-12  Commit Queue  <commit-queue@webkit.org>
836
837         Unreviewed, rolling out r209672.
838         https://bugs.webkit.org/show_bug.cgi?id=165766
839
840         IndexedDB/objectstore-cursor.html perf test always times out
841         (Requested by ap on #webkit).
842
843         Reverted changeset:
844
845         "More IndexedDB perf tests."
846         https://bugs.webkit.org/show_bug.cgi?id=165634
847         http://trac.webkit.org/changeset/209672
848
849 2016-12-10  Brady Eidson  <beidson@apple.com>
850
851         More IndexedDB perf tests.
852         https://bugs.webkit.org/show_bug.cgi?id=165634
853
854         Reviewed by Sam Weinig.
855
856         * IndexedDB/index-multientry.html:
857         * IndexedDB/large-number-of-inserts-responsiveness.html: Added.
858         * IndexedDB/large-number-of-inserts.html: Added.
859         * IndexedDB/objectstore-cursor.html: Added.
860         
861         * resources/runner.js: Add "track responsiveness" functionality to PerfTestRunner.
862
863 2016-12-06  Filip Pizlo  <fpizlo@apple.com>
864
865         Concurrent GC should be stable enough to land enabled
866         https://bugs.webkit.org/show_bug.cgi?id=164990
867
868         Reviewed by Geoffrey Garen.
869         
870         Made CDjs more configurable and refined the "large.js" configuration. I was using that one and
871         the new "long.js" configuration to tune concurrent eden GCs.
872         
873         Added a new way of running Splay in browser, which using chartjs to plot the execution times of
874         2000 iterations. This includes the minified chartjs.
875
876         * JetStream/Octane2/splay-detail.html: Added.
877         * JetStream/cdjs/benchmark.js:
878         (benchmarkImpl):
879         (benchmark):
880         * JetStream/cdjs/long.js: Added.
881
882 2016-12-07  Brady Eidson  <beidson@apple.com>
883
884         Add IDB perf tests stressing key size.
885         https://bugs.webkit.org/show_bug.cgi?id=165567
886
887         Reviewed by Alex Christensen.
888
889         * IndexedDB/large-array-keys.html: Added.
890         * IndexedDB/large-binary-keys.html: Added.
891         * IndexedDB/large-string-keys.html: Added.
892
893 2016-12-05  Brady Eidson  <beidson@apple.com>
894
895         Add an IndexedDB perf test to PerformanceTests.
896         https://bugs.webkit.org/show_bug.cgi?id=165430
897
898         Reviewed by Alex Christensen.
899
900         * IndexedDB/index-multientry.html: Added.
901
902 2016-12-02  Filip Pizlo  <fpizlo@apple.com>
903
904         ES6SampleBench should report an average for Steady State so that all of the numbers are comparable
905         https://bugs.webkit.org/show_bug.cgi?id=165325
906
907         Reviewed by Saam Barati.
908
909         This makes all of the numbers that ES6SampleBench reports comparable to each other: they all speak of the time
910         it took to run an iteration of something.
911
912         * ES6SampleBench/results.js:
913         (Results.prototype.reportResult):
914
915 2016-11-18  Jer Noble  <jer.noble@apple.com>
916
917         Add some performance tests for various aspects of media loading and playback.
918         https://bugs.webkit.org/show_bug.cgi?id=164977
919
920         Reviewed by Darin Adler.
921
922         Add new tests for a few aspects of media playback; namely: how quickly  media elements fire
923         the "canplaythrough" event when all data is immediately available, how quickly HLS playback
924         switches resolutions, and at what playback rate MSE-backed video can play without dropping
925         frames. Skip these tests by default, as they're not runnable on all ports, and they require
926         a webserver (run-webkit-httpd) to complete on macOS.
927
928         * Media/HLSCanPlayThrough.html: Added.
929         * Media/HLSGearChange.html: Added.
930         * Media/MSECanPlayThrough.html: Added.
931         * Media/MSEPlaybackRate.html: Added.
932         * Media/StandardCanPlayThrough.html: Added.
933         * Media/hls/1080p/iframe_index.m3u8: Added.
934         * Media/hls/1080p/prog_index.m3u8: Added.
935         * Media/hls/1080p/test.ts: Added.
936         * Media/hls/480p/iframe_index.m3u8: Added.
937         * Media/hls/480p/prog_index.m3u8: Added.
938         * Media/hls/480p/test.ts: Added.
939         * Media/hls/720p/iframe_index.m3u8: Added.
940         * Media/hls/720p/prog_index.m3u8: Added.
941         * Media/hls/720p/test.ts: Added.
942         * Media/hls/720p/test.ts.back: Added.
943         * Media/hls/index.m3u8: Added.
944         * Media/media-source-loader.js: Added.
945         (MediaSourceLoader):
946         (MediaSourceLoader.prototype.loadManifest):
947         (MediaSourceLoader.prototype.loadManifestSucceeded):
948         (MediaSourceLoader.prototype.loadManifestFailed):
949         (MediaSourceLoader.prototype.loadMediaData):
950         (MediaSourceLoader.prototype.loadMediaDataSucceeded):
951         (MediaSourceLoader.prototype.loadMediaDataFailed):
952         (MediaSourceLoader.prototype.get type):
953         (MediaSourceLoader.prototype.get duration):
954         (MediaSourceLoader.prototype.get initSegment):
955         (MediaSourceLoader.prototype.get mediaSegmentsLength):
956         (MediaSourceLoader.prototype.mediaSegments):
957         (MediaSourceLoader.prototype.get everyMediaSegment):
958         * Media/test-fragmented-video.json: Added.
959         * Media/test-fragmented-video.mp4: Added.
960         * Media/test.mp4: Added.
961         * Skipped:
962
963 2016-11-18  Filip Pizlo  <fpizlo@apple.com>
964
965         Concurrent GC should be able to run splay in debug mode and earley/raytrace in release mode with no perf regression
966         https://bugs.webkit.org/show_bug.cgi?id=164282
967
968         Reviewed by Geoffrey Garen and Oliver Hunt.
969         
970         CDjs is a fun benchmark for stressing concurrent GCs, but to really give the GC a good
971         workout you need to increase the amount of work that the test does. This adds a second
972         configuration of the benchmark that has more aircraft. It uses much more memory and causes us
973         to do more GCs and those GCs take longer.
974
975         * JetStream/cdjs/benchmark.js:
976         (benchmarkImpl):
977         (benchmark):
978         * JetStream/cdjs/large.js: Added.
979
980 2016-11-14  Filip Pizlo  <fpizlo@apple.com>
981
982         Unreviewed, revert unintended change.
983
984         * ES6SampleBench/Air/benchmark.js:
985
986 2016-10-03  Saam Barati  <sbarati@apple.com>
987
988         MapHash should speculate on the type of its child node
989         https://bugs.webkit.org/show_bug.cgi?id=161922
990
991         Reviewed by Filip Pizlo.
992
993         I gave the main generator a name so it's easier to see what
994         it is when using the sampling profiler.
995
996         * ES6SampleBench/Basic/ast.js:
997         (Basic.Program):
998
999 2016-09-19  Sergio Villar Senin  <svillar@igalia.com>
1000
1001         [css-grid] Remove the x2 computation of row sizes with indefinite heights
1002         https://bugs.webkit.org/show_bug.cgi?id=162150
1003
1004         Reviewed by Darin Adler.
1005
1006         Added a new test case which checks the layout performance of grids inside other grids, i.e,
1007         grids acting both as grid container and grid item.
1008
1009         * Layout/nested-grid.html: Added.
1010
1011 2016-09-09  Simon Fraser  <simon.fraser@apple.com>
1012
1013         Perf test Animation/css-accelerated-animation.html failing
1014         https://bugs.webkit.org/show_bug.cgi?id=161795
1015
1016         Skip Animation/css-accelerated-animation.html since this only works as a MobileSafari
1017         perf test.
1018
1019         * Skipped:
1020
1021 2016-09-08  Simon Fraser  <simon.fraser@apple.com>
1022
1023         Add a content-animation test with accelerated CSS animations
1024         https://bugs.webkit.org/show_bug.cgi?id=161776
1025
1026         Reviewed by Dean Jackson.
1027
1028         Add a test that measures the frame rate of accelerated CSS animations.
1029
1030         This is like css-animation.html (which animates 'left' and 'top') but animates
1031         transforms instead. In order to get animation in X and Y we make two nested elements
1032         and set transformX() on one, and transformY() on the other.
1033
1034         * Animation/css-accelerated-animation.html: Added.
1035
1036 2016-08-23  Saam Barati  <sbarati@apple.com>
1037
1038         It should be easy to run ES6SampleBench from the jsc shell
1039         https://bugs.webkit.org/show_bug.cgi?id=161085
1040
1041         Reviewed by Yusuke Suzuki.
1042
1043         This patch makes ES6 sample bench runnable through the `jsc` shell.
1044         To do that, you need to be in the PerformanceTests/ES6SampleBench
1045         directory and run `jsc cli.js`. To make this work, the benchmark
1046         is now aware of being run in two modes: via the browser, and via
1047         the shell. Each entry point will set a variable `isInBrowser`,
1048         and the benchmark will do different things based on if that
1049         variable is true or false.
1050
1051         * ES6SampleBench/Air/benchmark.js:
1052         * ES6SampleBench/Air/stress-test.js:
1053         * ES6SampleBench/Basic/benchmark.js:
1054         (runBenchmark):
1055         * ES6SampleBench/Basic/stress-test.js:
1056         * ES6SampleBench/air_benchmark.js:
1057         * ES6SampleBench/basic_benchmark.js:
1058         * ES6SampleBench/cli.js: Added.
1059         (return.doRun):
1060         (makeBenchmarkRunner):
1061         * ES6SampleBench/driver.js:
1062         (Driver):
1063         (Driver.prototype._recomputeSummary):
1064         (Driver.prototype._iterate.window.setTimeout):
1065         (Driver.prototype._iterate):
1066         (Driver.prototype._updateIterations):
1067         * ES6SampleBench/glue.js:
1068         (reportResult):
1069         * ES6SampleBench/index.html:
1070         * ES6SampleBench/results.js:
1071         (Results):
1072         (Results.prototype.reportRunning):
1073         (Results.prototype.reportDone):
1074         (Results.prototype.reportError):
1075         * ES6SampleBench/stats.js:
1076         (Stats):
1077         (Stats.prototype.toString):
1078         (Stats.prototype._update):
1079
1080 2016-08-01  Said Abou-Hallawa  <sabouhallawa@apple.com>
1081
1082         Add an option to run the MotionMark tests with classic tile size (512x512)
1083         https://bugs.webkit.org/show_bug.cgi?id=160371
1084
1085         Reviewed by Darin Adler.
1086
1087         Setting the body to large size (3000x3000) and overflow to scroll forces
1088         the classic tile size (512x512). This option does not affect the canvas
1089         tests because the canvas is displayed on one tile regardless of its size.
1090
1091         * Animometer/developer.html:
1092         * Animometer/resources/debug-runner/animometer.css:
1093         (body.showing-test-container.tiles-big):
1094         (body.showing-test-container.tiles-classic):
1095         * Animometer/resources/debug-runner/animometer.js:
1096         * Animometer/resources/runner/animometer.js:
1097         (window.benchmarkController.startBenchmark):
1098
1099 2016-06-29  Filip Pizlo  <fpizlo@apple.com>
1100
1101         Generators violate bytecode liveness validation
1102         https://bugs.webkit.org/show_bug.cgi?id=159279
1103
1104         Reviewed by Yusuke Suzuki.
1105
1106         Add Basic to our test harness.
1107
1108         Also made some cosmetic changes to the benchmark harness.
1109
1110         * ES6SampleBench/Basic/basic-tests.yaml: Added.
1111         * ES6SampleBench/Basic/stress-test.js: Added.
1112         (preciseTime):
1113         * ES6SampleBench/driver.js:
1114         (Driver):
1115         (Driver.prototype.start):
1116         (Driver.prototype.reportError):
1117         * ES6SampleBench/glue.js:
1118         * ES6SampleBench/index.html:
1119
1120 2016-06-28  Filip Pizlo  <fpizlo@apple.com>
1121
1122         ES6SampleBench should have a harness
1123         https://bugs.webkit.org/show_bug.cgi?id=159246
1124
1125         Reviewed by Saam Barati.
1126         
1127         This adds a simple web harness for ES6SampleBench. It runs Air and Basic 10 times for 200
1128         iterations each and reports three metrics:
1129         
1130         First iteration: the time it takes for the first iteration to run. This is the first
1131         iteration after the benchmark is loaded into the iframe, so it's representative of what
1132         would happen if one of these workloads only ran for a short time.
1133         
1134         Worst 2%: of the last 199 iterations, the average of the worst 2% iterations. If code like
1135         any of these workloads was used in an important event handler, you'd want that code to run
1136         well in the worst case in addition to having great throughput.
1137         
1138         Steady state: the total of the last 199 iterations. This is representative of what would
1139         happen if you ran code like this for a long time.
1140         
1141         The total score is the geomean of the firstIteration/worstCase/steadyState numbers of the
1142         two benchmarks.
1143         
1144         The harness does statistics using Student's T-distribution confidence intervals.
1145
1146         * ES6SampleBench/Basic/benchmark.js:
1147         (Benchmark):
1148         * ES6SampleBench/air_benchmark.js: Added.
1149         * ES6SampleBench/basic_benchmark.js: Added.
1150         * ES6SampleBench/driver.js: Added.
1151         (Driver):
1152         (Driver.prototype.addBenchmark):
1153         (Driver.prototype.start):
1154         (Driver.prototype.reportResult):
1155         (Driver.prototype.reportError):
1156         (Driver.prototype._recomputeSummary.Geomean):
1157         (Driver.prototype._recomputeSummary.Geomean.prototype.add):
1158         (Driver.prototype._recomputeSummary.Geomean.prototype.get result):
1159         (Driver.prototype._recomputeSummary):
1160         (Driver.prototype._iterate):
1161         (Driver.prototype._updateIterations):
1162         * ES6SampleBench/glue.js: Added.
1163         * ES6SampleBench/index.html: Added.
1164         * ES6SampleBench/results.js: Added.
1165         (Results):
1166         (Results.prototype.get benchmark):
1167         (Results.prototype.reset):
1168         (Results.prototype.reportRunning):
1169         (Results.prototype.reportDone):
1170         (Results.prototype.reportResult.averageAbovePercentile):
1171         (Results.prototype.reportResult):
1172         (Results.prototype.reportError):
1173         * ES6SampleBench/stats.js: Added.
1174         (Stats):
1175         (Stats.prototype.reset):
1176         (Stats.prototype.add):
1177         (Stats.prototype.get numIterations):
1178         (Stats.prototype.valueForIteration):
1179         (Stats.get result.tDist):
1180         (Stats.prototype.get result):
1181         (Stats.prototype.toString):
1182         (Stats.prototype._update):
1183         * ES6SampleBench/style.css: Added.
1184         (body):
1185         (body, th, tr):
1186         (h1, h2, h3, h4):
1187         (h1):
1188         (h2):
1189         (h3):
1190         (hr):
1191         (address):
1192         (img):
1193         (.underline):
1194         (ol.loweralpha):
1195         (ol.upperalpha):
1196         (ol.lowerroman):
1197         (ol.upperroman):
1198         (ol.arabic):
1199         (.banner-link:link, .banner-link:visited):
1200         (:link, :visited):
1201         (h4 :link, h4 :visited, h5 :link, h5 :visited, h6 :link, h6 :visited):
1202         (.anchor:link, .anchor:visited):
1203         (* > .anchor:link, * > .anchor:visited):
1204         (span:hover .anchor):
1205         (a.forbidden, span.forbidden):
1206         (a.missing:hover):
1207         (a.closed:link, a.closed:visited, span.closed):
1208         (pre):
1209         (div.code):
1210         (div.code pre):
1211         (dt):
1212         (dd):
1213         (dd:last-child):
1214         (.site-logo):
1215         (.site-logo .tagline):
1216         (table):
1217         (#contents):
1218         (p):
1219         (p:last-child):
1220         (th):
1221         (td):
1222
1223 2016-06-28  Jon Lee  <jonlee@apple.com>
1224
1225         Update focus test
1226         https://bugs.webkit.org/show_bug.cgi?id=159242
1227         rdar://problem/27070007
1228
1229         Reviewed by Dean Jackson.
1230         Provisionally reviewed by Said Abou-Hallawa.
1231
1232         Move previous test to dom suite, and update the test for better reporting of frame rate, although
1233         it uses a different rendering path.
1234
1235         * Animometer/resources/debug-runner/tests.js: Add to dom suite.
1236         * Animometer/tests/dom/focus.html: Copied from PerformanceTests/Animometer/tests/master/focus.html.
1237         * Animometer/tests/dom/resources/focus.js: Copied from PerformanceTests/Animometer/tests/master/resources/focus.js.
1238         * Animometer/tests/master/focus.html: Remove center element.
1239         * Animometer/tests/master/resources/focus.js: Use narrower size range with smaller particles. Remove the
1240         container elements. Inline getBlurValue and getOpacityValue since they are only called once.
1241
1242 2016-06-28  Filip Pizlo  <fpizlo@apple.com>
1243
1244         Move Air.js and Basic into ES6SampleBench
1245
1246         Rubber stamped by Geoffrey Garen.
1247         
1248         We want to group our ES6 benchmarks together, and eventually, we want to give them a
1249         common harness.
1250
1251         * Air.js: Removed.
1252         * Air.js/README.md: Removed.
1253         * Air.js/airjs-tests.yaml: Removed.
1254         * Air.js/all.js: Removed.
1255         * Air.js/allocate_stack.js: Removed.
1256         * Air.js/arg.js: Removed.
1257         * Air.js/basic_block.js: Removed.
1258         * Air.js/benchmark.js: Removed.
1259         * Air.js/code.js: Removed.
1260         * Air.js/custom.js: Removed.
1261         * Air.js/frequented_block.js: Removed.
1262         * Air.js/insertion_set.js: Removed.
1263         * Air.js/inst.js: Removed.
1264         * Air.js/liveness.js: Removed.
1265         * Air.js/make_dist.sh: Removed.
1266         * Air.js/opcode.js: Removed.
1267         * Air.js/payload-airjs-ACLj8C.js: Removed.
1268         * Air.js/payload-gbemu-executeIteration.js: Removed.
1269         * Air.js/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
1270         * Air.js/payload-typescript-scanIdentifier.js: Removed.
1271         * Air.js/reg.js: Removed.
1272         * Air.js/stack_slot.js: Removed.
1273         * Air.js/stress-test.js: Removed.
1274         * Air.js/strip-hash.rb: Removed.
1275         * Air.js/symbols.js: Removed.
1276         * Air.js/test.html: Removed.
1277         * Air.js/test.js: Removed.
1278         * Air.js/tmp.js: Removed.
1279         * Air.js/tmp_base.js: Removed.
1280         * Air.js/util.js: Removed.
1281         * Basic: Removed.
1282         * Basic/ast.js: Removed.
1283         * Basic/basic.js: Removed.
1284         * Basic/benchmark.js: Removed.
1285         * Basic/caseless_map.js: Removed.
1286         * Basic/lexer.js: Removed.
1287         * Basic/number.js: Removed.
1288         * Basic/parser.js: Removed.
1289         * Basic/random.js: Removed.
1290         * Basic/state.js: Removed.
1291         * Basic/test.html: Removed.
1292         * Basic/test.js: Removed.
1293         * Basic/util.js: Removed.
1294         * ES6SampleBench: Added.
1295         * ES6SampleBench/Air: Copied from PerformanceTests/Air.js.
1296         * ES6SampleBench/Basic: Copied from PerformanceTests/Basic.
1297         * Skipped:
1298
1299 2016-06-24  Filip Pizlo  <fpizlo@apple.com>
1300
1301         Add a ES6 generator benchmark
1302         https://bugs.webkit.org/show_bug.cgi?id=159101
1303
1304         Rubber stamped by Keith Miller.
1305         
1306         This adds a Basic interpreter loosely based on ECMA-55:
1307         
1308         http://www.ecma-international.org/publications/files/ECMA-ST-WITHDRAWN/ECMA-55,%201st%20Edition,%20January%201978.pdf
1309         
1310         It includes a lexer that is a generator, a parser that uses regular expressions, and an AST
1311         walk interpreter where the walking functions for statements are generators that call each
1312         other with yield*. This enables the interpreter to look like an AST walk even though it can
1313         yield at INPUT and PRINT statements.
1314         
1315         This also uses for-of, classes, Map, and WeakMap. It also uses deprecated-but-awesome RegExp
1316         features like RegExp.lastMatch and RegExp.rightContext. I did it that way because this is
1317         how I've always written lexers in dynamic languages; see offlineasm's lex() method in
1318         parser.rb for example.
1319         
1320         The benchmark runs a handful of simple Basic programs. The longest-running one computes
1321         prime numbers.
1322         
1323         Includes a command-line and web harness. On my machine it runs in 2-3 seconds.
1324
1325         * Basic: Added.
1326         * Basic/ast.js: Added.
1327         (Basic.NumberApply):
1328         (Basic.Variable):
1329         (Basic.Const):
1330         (Basic.NumberPow):
1331         (Basic.NumberMul):
1332         (Basic.NumberDiv):
1333         (Basic.NumberNeg):
1334         (Basic.NumberAdd):
1335         (Basic.NumberSub):
1336         (Basic.StringVar):
1337         (Basic.Equals):
1338         (Basic.NotEquals):
1339         (Basic.LessThan):
1340         (Basic.GreaterThan):
1341         (Basic.LessEqual):
1342         (Basic.GreaterEqual):
1343         (Basic.GoTo):
1344         (Basic.GoSub):
1345         (Basic.Def):
1346         (Basic.Let):
1347         (Basic.If):
1348         (Basic.Return):
1349         (Basic.Stop):
1350         (Basic.On):
1351         (sideState.shouldStop):
1352         (Basic.For):
1353         (Basic.Next):
1354         (Basic.Print):
1355         (Basic.Input):
1356         (Basic.Read):
1357         (Basic.Restore):
1358         (Basic.Dim):
1359         (Basic.Randomize):
1360         (Basic.End):
1361         (Basic.Program):
1362         * Basic/basic.js: Added.
1363         (prepare):
1364         (simulate):
1365         * Basic/benchmark.js: Added.
1366         (Benchmark):
1367         (Benchmark.prototype.runIteration.expect):
1368         (Benchmark.prototype.runIteration):
1369         (runBenchmark):
1370         * Basic/caseless_map.js: Added.
1371         (CaselessMap):
1372         * Basic/lexer.js: Added.
1373         (lex.consumeWhitespace):
1374         (lex.consume):
1375         (lex):
1376         * Basic/lexer_test.js: Added.
1377         * Basic/number.js: Added.
1378         (NumberValue):
1379         (NumberValue.prototype.get value):
1380         (NumberValue.prototype.apply):
1381         (NumberValue.prototype.leftApply):
1382         (NumberValue.prototype.assign):
1383         (NumberArray.):
1384         (NumberArray):
1385         (NumberArray.prototype.apply):
1386         (NumberArray.prototype.leftApply):
1387         (NumberFunction):
1388         (NumberFunction.prototype.apply):
1389         (NumberFunction.prototype.leftApply):
1390         (NativeFunction):
1391         (NativeFunction.prototype.apply):
1392         (NativeFunction.prototype.leftApply):
1393         * Basic/parser.js: Added.
1394         (parse):
1395         (parse.pushToken):
1396         (parse.peekToken):
1397         (parse.consumeKind):
1398         (parse.consumeToken):
1399         (parse.parseVariable):
1400         (parse.parseNumericExpression.parsePrimary):
1401         (parse.parseNumericExpression.parseFactor):
1402         (parse.parseNumericExpression.parseTerm):
1403         (parse.parseNumericExpression):
1404         (parse.parseConstant):
1405         (parse.parseStringExpression):
1406         (parse.isStringExpression):
1407         (parse.parseRelationalExpression):
1408         (parse.parseNonNegativeInteger):
1409         (parse.parseGoToStatement):
1410         (parse.parseGoSubStatement):
1411         (parse.parseStatement):
1412         (parse.parseStatements):
1413         * Basic/random.js: Added.
1414         (createRNG):
1415         (createRNGWithFixedSeed):
1416         (createRNGWithRandomSeed):
1417         * Basic/state.js: Added.
1418         (State):
1419         (State.prototype.getValue):
1420         (State.prototype.getSideState):
1421         (State.prototype.abort):
1422         (State.prototype.validate):
1423         * Basic/test.html: Added.
1424         * Basic/test.js: Added.
1425         * Basic/util.js: Added.
1426         (this.performance.performance.now.currentTime):
1427         (else.this.preciseTime.currentTime):
1428         (else.currentTime):
1429         * Skipped: Make sure that we don't run Basic yet.
1430
1431 2016-06-21  Jon Lee  <jonlee@apple.com>
1432
1433         Update canvas size when benchmark begins
1434         https://bugs.webkit.org/show_bug.cgi?id=159010
1435
1436         Reviewed by Dean Jackson.
1437
1438         Previously we would evaluate the media queries and assign the canvas size to the test
1439         on body.onload. Instead, do it once the user starts the benchmark.
1440
1441         * Animometer/resources/debug-runner/animometer.js: Move benchmarkController.determineCanvasSize()
1442         to benchmarkController._startBenchmark.
1443         * Animometer/resources/runner/animometer.js: Ditto.
1444
1445 2016-06-21  Jon Lee  <jonlee@apple.com>
1446
1447         Add new timestamp option
1448         https://bugs.webkit.org/show_bug.cgi?id=159006
1449
1450         Reviewed by Dean Jackson.
1451
1452         Add a new option to take timestamps from the rAF callback.
1453
1454         * Animometer/developer.html: Added option.
1455         * Animometer/tests/resources/main.js:
1456         (Benchmark): Fall back to using the rAF timestamp if performance.now()
1457         is not available.
1458         (Benchmark._animateLoop): Update how timestamp is set. Prefer to use
1459         local var instead of accessing private var.
1460
1461 2016-06-21  Jon Lee  <jonlee@apple.com>
1462
1463         Improvements to Animometer benchmark
1464         https://bugs.webkit.org/show_bug.cgi?id=157738
1465
1466         Reviewed by Dean Jackson.
1467         Provisionally reviewed by Said Abou-Hallawa.
1468
1469         Update tests.
1470
1471         * Animometer/tests/master/text.html: Ensure only three text sizes for
1472         the three canvases.
1473         * Animometer/tests/master/focus.html: Reduce the text size to fit with smaller
1474         particle sizes.
1475         * Animometer/tests/master/resources/focus.js: Remove the quadratic distribution for
1476         particle sizes, and make it linear. Reduce the size variance. Shuffle the math to
1477         reduce some calculations per frame. Fix the placement of the particles which might
1478         otherwise be culled.
1479         * Animometer/tests/master/resources/image-data.js: Reduce the particle size
1480         to encourage larger scores.
1481
1482 2016-06-21  Jon Lee  <jonlee@apple.com>
1483
1484         Improvements to Animometer benchmark
1485         https://bugs.webkit.org/show_bug.cgi?id=157738
1486
1487         Reviewed by Dean Jackson.
1488         Provisionally reviewed by Said Abou-Hallawa.
1489
1490         Include confidence interval for the final score, and store the canvas
1491         size in the serialization so that it is accurately shown in results.
1492
1493         * Animometer/developer.html: Add a "confidence" div.
1494         * Animometer/index.html: Ditto. Convert "mean" to "confidence".
1495         * Animometer/resources/debug-runner/animometer.js: Look at options, and
1496         if the configuration is included, update the body class based on it
1497         (similar to what we do when we first load the page). That way, if you
1498         drag-and-drop previous results in, that configuration is reflected in
1499         the dashboard. Show the full confidence interval.
1500
1501         * Animometer/resources/debug-runner/animometer.css:
1502         * Animometer/resources/debug-runner/animometer.js: Style update.
1503         * Animometer/resources/runner/animometer.css:
1504
1505         * Animometer/resources/runner/animometer.js:
1506         (_processData): Propagate the confidence interval values out and calculate
1507         the lower and upper bounds. For now, shortcut the aggregate calculation,
1508         since we only go through one iteration.
1509         (this._processData.calculateScore): Propagate the confidence interval out
1510         to be next to the score. Depending on the controller these values are
1511         calculated differently.
1512         (this._processData._getResultsProperty): Convenience function.
1513         (this._processData.get score): Refactor.
1514         (this._processData.get scoreLowerBound): Get the aggregate lower bound.
1515         (this._processData.get scoreUpperBound): Get the aggregate upper bound.
1516         (window.sectionsManager.setSectionScore):
1517         (window.benchmarkController._startBenchmark): When the benchmark starts, note
1518         the canvas size and add it to options. That way it will be automatically be
1519         serialized.
1520         (window.benchmarkController.showResults): Include the maximum deviation
1521         percentage.
1522         * Animometer/resources/runner/lines.svg: Make the background line up with the
1523         skew.
1524         * Animometer/resources/runner/tests.js:
1525         (Headers.details.text): Refactor.
1526         * Animometer/resources/statistics.js:
1527         (Statistics.largestDeviationPercentage): Convenience function to calculate
1528         the deviation percentage on either end and return the largest deviation.
1529         * Animometer/resources/strings.js:
1530
1531         Allow specifying a regression profile to use instead of taking the one
1532         with the lowest error.
1533
1534         Address an issue in the focus test when the regression calculated ends
1535         up overestimating the change point, causing a cascade of tougher
1536         ramps. The reason behind this is that at max complexity of an initial
1537         ramp, the frame length is very high, and it influences the second
1538         segment of the piecewise regression strongly, causing it to be very
1539         steep. As a result, the first segment, expected to be flat, ends up
1540         covering a higher range of complexity. That makes the change point
1541         much higher than it should be. To avoid this, we will add a sanity
1542         check on the maximum value of the ramp. If the regression's projected
1543         value at the maximum complexity of the current ramp is very slow (less
1544         than 20 fps), 1) reduce the maximum complexity by 20%, and 2) do not
1545         include the regression's change point in the change point estimator.
1546         That estimator is used as the midpoint of the next ramp, and including
1547         the change point from a poor regression can bake in the error. The
1548         controller already knows how to adjust for ramps that are too easy for
1549         the system.
1550
1551         * Animometer/resources/runner/animometer.js:
1552         (this._processData.findRegression): Takes a preferred profile and gives that to
1553         Regression.
1554         (this._processData.calculateScore): With the ramp controller, take the profile
1555         of the ramp that was used the most when calculating the ramp's regression. That
1556         profile is what is used for the test's score.
1557         * Animometer/resources/statistics.js:
1558         (Regression.Utilities.createClass): Update to take an options object which can
1559         specify a profile to calculate with. Otherwise it will continue to use both and
1560         select the one with the lower error.
1561         (_calculateRegression): Fix an issue where we claim 0 error if the regression
1562         calculation fails due to divide-by-zero. Instead reject that regression calculation
1563         by giving it Number.MAX_VALUE.
1564         * Animometer/resources/strings.js: New strings for marking a regression as flat
1565         or slope.
1566         * Animometer/tests/resources/main.js:
1567         (RampController): Rename the thresholds for clarity. Add a threshold that, if
1568         exceeded, will lower the maximum complexity of the next ramp.
1569         (tune): Relax the cdf check to consider whether the interval definitely falls in
1570         the desired frame length threshold.
1571         (processSamples): Include the profile in the ramp.
1572
1573         Update ramp controller test. Increase the length of the test to 30 seconds, and extend
1574         the interval to 120 ms during sampling. Improve the estimation of the ramp parameters.
1575
1576         * Animometer/developer.html: Change default to 30 seconds, and don't show the progress bar
1577         by default.
1578         * Animometer/resources/runner/animometer.js: Change default to 30 seconds.
1579         * Animometer/tests/resources/main.js: A number of improvements to the ramp controller, in
1580         the order in which they appear in the patch:
1581
1582         - With a longer test length use longer ramps with longer intervals to get more data at each
1583         complexity. Keep the 100 ms interval length during the ramp up phase since we don't need to
1584         spend more time there to find the right order of magnitude, but increase it during the
1585         ramps to 120 ms.
1586         - The ramp linearly interpolates the complexity to render based on its timestamp, but it would
1587         never sample the minimum complexity. Instead of lerping max to min complexity from time
1588         0 to t where t is the ramp length, instead lerp from 0 to (t - intervalSampleLength) so that
1589         we can have at least one interval sample at the min complexity for that ramp.
1590         - Some regression calculations only come out with one line segment rather than the two
1591         we expect. This could be due to a noisy ramp or the ramp's range is too narrow. If that's the
1592         case, influence the minimum complexity of the next ramp towards the lowest bound of 1, so that
1593         we ensure that at least part of the ramp is covering a complexity range that the system can
1594         handle at full 60.
1595         - Remove an assignment to interpolatedFrameLength since that is never subsequently used.
1596
1597         Update the format used to serialize the results for analysis.
1598
1599         Each data point used to have named properties for fields like complexity and frame rate.
1600         In addition the serialized numbers had rounding errors that took up many characters.
1601         Update the format by introducing a new data container called SampleData, which contains a
1602         field map. The map maps a string to an array index. Each data point is an array, so, to
1603         get a stat, use the field map to get the array index into the data point. This allows future
1604         versions to track other data, and reduces the size of the output string by two-thirds.
1605
1606         * Animometer/resources/extensions.js:
1607         (Utilities.toFixedNumber): Add convenience function that truncates the number to a fixed
1608         precision, and converts it back to a number.
1609         (SampleData): New class that contains sample data and a field map that maps properties to
1610         an array index.
1611         (get length): Number of data points.
1612         (addField): Add a field to the field map.
1613         (push): Add a data point.
1614         (sort): Sort the data.
1615         (slice): Return new SampleData object with sliced data.
1616         (forEach): Iterate over the data with provided function.
1617         (createDatum):
1618         (getFieldInDatum): Returns the data point associated with the field name by looking it up
1619         in the field map in the datum provided, which can be the datum object itself (an array) or
1620         an index into the data member variable.
1621         (setFieldInDatum): Sets the data point associated with the field name.
1622         (at): Returns the data point at the provided index.
1623         (toArray): Serializes the data where the field map serves as property names for each point.
1624
1625         * Animometer/resources/debug-runner/graph.js:
1626         (updateGraphData): Remove unused _testData. Convert the data to the old array format for the
1627         graph to use, since the old format was much easier to work with when displaying the graphs.
1628         (onGraphTypeChanged): For some controllers, no alternative score or mean is provided.
1629         * Animometer/resources/runner/animometer.css:
1630         * Animometer/resources/runner/animometer.js: Refactor to use SampleData. Update JSON output
1631         to only go to 3 digits of precision for purposes of reducing the data size.
1632         * Animometer/resources/strings.js: Add new strings to put into the field maps.
1633         * Animometer/tests/resources/main.js: Refactor to use SampleData.
1634
1635         * Animometer/developer.html:
1636         * Animometer/index.html: Restructure results table for both pages. Add charset attribute to
1637         tests.js include.
1638         * Animometer/resources/debug-runner/animometer.css: Clear out styles from release runner.
1639         * Animometer/resources/debug-runner/graph.js:
1640         (onGraphTypeChanged): Update score and mean if bootstrap results are available from the
1641         controller, since not all controllers do bootstrapping.
1642         * Animometer/resources/debug-runner/tests.js: Update header text.
1643         * Animometer/resources/runner/animometer.css: Include confidence interval in results.
1644         * Animometer/resources/runner/animometer.js:
1645         (ResultsTable._addHeader): Header contents can be HTML, so use innerHTML instead.
1646         (ResultsTable._addBody): Add tbody element.
1647         (ResultsTable._addTest): Allow a data cell to invoke a JS function to get its contents.
1648         (window.benchmarkController.showResults): Add table that includes tests' confidence intervals.
1649         * Animometer/resources/runner/tests.js:
1650         (Headers.details.text): Add new details table that includes bootstrap confidence interval.
1651         The interval can be asymmetric, but for simplicity, report the maximum deviation percentage
1652         on either side of the bootstrap median.
1653         * Animometer/resources/statistics.js:
1654         (bootstrap): Include the confidence percentage in the return object.
1655
1656         Report canvas size in results.
1657
1658         * Animometer/developer.html: Add markup to indicate whether a small, medium, or large
1659         canvas was used.
1660         * Animometer/index.html: Ditto.
1661         * Animometer/resources/debug-runner/animometer.js: Call determineCanvasSize().
1662         * Animometer/resources/runner/animometer.css: Update styles to set the canvas based on the
1663         body class size.
1664         * Animometer/resources/runner/animometer.js:
1665         (window.benchmarkController.initialize): Update styles to set the canvas based on the
1666         body class size.
1667         (window.benchmarkController.determineCanvasSize): Run various media queries and set the body
1668         class based on the size of the device.
1669
1670         * Animometer/developer.html: Refactor to include the main CSS file, and redo
1671         the layout so that it doesn't rely on flexbox.
1672         * Animometer/resources/debug-runner/animometer.css:
1673         * Animometer/resources/debug-runner/animometer.js:
1674         (updateDisplay): Since various parts of the script alter the body class, we can't
1675         replace the className directly. Instead, remove all display-based values and then add
1676         the one that was selected.
1677         * Animometer/resources/debug-runner/graph.js:
1678         (updateGraphData): To set the size of the graph, use window.innerHeight.
1679         * Animometer/resources/runner/animometer.js:
1680         (window.sectionsManager.showSection): Since various parts of the script alter the body
1681         class, we can't replace the className directly. Remove all of the section classes
1682         individually and then add the one desired.
1683         * Animometer/tests/resources/stage.css: Remove -apple-system as a font to use in the
1684         stage.
1685
1686 2016-06-12  Filip Pizlo  <fpizlo@apple.com>
1687
1688         Fix round-down goof in Air.js's ShuffleCustom.forEachArg
1689         https://bugs.webkit.org/show_bug.cgi?id=158674
1690
1691         Reviewed by Michael Saboff.
1692         
1693         x / 3 * 3 is not how you round down to multiples of 3 in JavaScript. You need to do
1694         Math.floor(x / 3) * 3 instead.
1695         
1696         This is a benign change, because having extra arguments to a Shuffle is not something we
1697         actually take advantage of yet. But I think it's best to match the original C++ code's
1698         intent.
1699
1700         * Air.js/custom.js:
1701         (const.ShuffleCustom.forEachArg):
1702
1703 2016-06-11  Filip Pizlo  <fpizlo@apple.com>
1704
1705         Unreviewed, skip Air.js. It got unskipped when I renamed it from JSAir.
1706
1707         * Skipped:
1708
1709 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1710
1711         Make it easy to package Air.js
1712         https://bugs.webkit.org/show_bug.cgi?id=158652
1713
1714         Reviewed by Benjamin Poulain.
1715         
1716         If you want to give Air.js to someone, you can now do:
1717         
1718         cd PerformanceTests/Air.js
1719         ./make_dist.sh
1720         
1721         This will create Air.js.tar.gz and a directory called Air.js. If you want to send someone
1722         the benchmark, just send either of those things and tell them to look at index.html.
1723         
1724         You may have to edit make_dist.sh for wherever you have Markdown.pl. You can get that from
1725         https://daringfireball.net/projects/markdown/
1726
1727         * Air.js/make_dist.sh: Added.
1728         * Air.js/test.html:
1729
1730 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1731
1732         Unreviewed, make the link to test.html into an actual link.
1733
1734         * Air.js/README.md:
1735
1736 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1737
1738         Fix some text in Air.js/README.md
1739         https://bugs.webkit.org/show_bug.cgi?id=158650
1740
1741         Reviewed by Benjamin Poulain.
1742         
1743         I read the text again and found bugs:
1744         
1745         - We never actually say how to run the benchmark. This change adds a blurb about how to run
1746           it.
1747
1748         - We both say that allocateStack is responsible for the bulk of the running time and that
1749           we haven't measured where the bulk of the time is spent. This changes the text to say that
1750           it was a goal to make allocateStack be the hottest part of the benchmark, but that we did
1751           not measure this.
1752
1753         * Air.js/README.md:
1754
1755 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1756
1757         Air.js should have some documentation
1758         https://bugs.webkit.org/show_bug.cgi?id=158648
1759
1760         Reviewed by Keith Miller.
1761         
1762         I want to be able to point people at a document if they want to know more about this
1763         benchmark.
1764
1765         * Air.js/README.md: Added.
1766
1767 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1768
1769         Rename JSAir to Air.js.
1770
1771         Rubber stamped by Mark Lam.
1772
1773         * Air.js: Copied from PerformanceTests/JSAir.
1774         * Air.js/airjs-tests.yaml: Copied from PerformanceTests/JSAir/jsair-tests.yaml.
1775         * Air.js/benchmark.js: Copied from PerformanceTests/JSAir/benchmark.js.
1776         (Benchmark):
1777         * Air.js/jsair-tests.yaml: Removed.
1778         * Air.js/payload-airjs-ACLj8C.js: Copied from PerformanceTests/JSAir/payload-jsair-ACLj8C.js.
1779         * Air.js/payload-jsair-ACLj8C.js: Removed.
1780         * Air.js/stress-test.js: Copied from PerformanceTests/JSAir/stress-test.js.
1781         * Air.js/test.html: Copied from PerformanceTests/JSAir/test.html.
1782         * Air.js/test.js: Copied from PerformanceTests/JSAir/test.js.
1783         * JSAir: Removed.
1784         * JSAir/all.js: Removed.
1785         * JSAir/allocate_stack.js: Removed.
1786         * JSAir/arg.js: Removed.
1787         * JSAir/basic_block.js: Removed.
1788         * JSAir/benchmark.js: Removed.
1789         * JSAir/code.js: Removed.
1790         * JSAir/custom.js: Removed.
1791         * JSAir/frequented_block.js: Removed.
1792         * JSAir/insertion_set.js: Removed.
1793         * JSAir/inst.js: Removed.
1794         * JSAir/jsair-tests.yaml: Removed.
1795         * JSAir/liveness.js: Removed.
1796         * JSAir/opcode.js: Removed.
1797         * JSAir/payload-gbemu-executeIteration.js: Removed.
1798         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js: Removed.
1799         * JSAir/payload-jsair-ACLj8C.js: Removed.
1800         * JSAir/payload-typescript-scanIdentifier.js: Removed.
1801         * JSAir/reg.js: Removed.
1802         * JSAir/stack_slot.js: Removed.
1803         * JSAir/stress-test.js: Removed.
1804         * JSAir/strip-hash.rb: Removed.
1805         * JSAir/symbols.js: Removed.
1806         * JSAir/test.html: Removed.
1807         * JSAir/test.js: Removed.
1808         * JSAir/tmp.js: Removed.
1809         * JSAir/tmp_base.js: Removed.
1810         * JSAir/util.js: Removed.
1811
1812 2016-06-10  Filip Pizlo  <fpizlo@apple.com>
1813
1814         JSC Stress Test failing: jsair-tests.yaml/test.js.ftl-eager-no-cjit
1815         https://bugs.webkit.org/show_bug.cgi?id=158571
1816
1817         Reviewed by Keith Miller.
1818         
1819         Introduce a different harness for run-jsc-stress-tests, which runs for a minimum of 10
1820         iterations and then tries to do any number of "bonus" iterations until it's been running for
1821         two seconds. Since this is the sort of test that isn't really meant to stress anything in
1822         particular, I think it's OK if it is time-limited in this way. The worst case is that some
1823         of its failures will be flaky, but I think that they would have been flaky anyway given the
1824         complexity of the test.
1825
1826         * JSAir/benchmark.js:
1827         (Benchmark):
1828         (Benchmark.prototype.runIteration):
1829         (benchmark): Deleted.
1830         * JSAir/jsair-tests.yaml:
1831         * JSAir/stress-test.js: Added.
1832         (preciseTime):
1833         * JSAir/test.html:
1834         * JSAir/test.js:
1835
1836 2016-06-09  Filip Pizlo  <fpizlo@apple.com>
1837
1838         Unreviewed, teach the perf bots not to run JSAir.
1839
1840         * Skipped:
1841
1842 2016-06-08  Filip Pizlo  <fpizlo@apple.com>
1843
1844         Use more ES6 features in JSAir
1845         https://bugs.webkit.org/show_bug.cgi?id=158497
1846
1847         Reviewed by Keith Miller.
1848         
1849         This improves JSAir with the following ES6 features suggested by JoePeck:
1850         
1851         - String interpolation.
1852         - Destructuring inside PatchCustom.
1853         - Default arguments.
1854         
1855         All of these things are on hot paths.
1856         
1857         Note that I didn't use string interpolation everywhere that I could, only in those places
1858         where it made the code more readable. In Ruby, I used the style that if the interpolation
1859         expression has any non-trivial stuff (like a ternary operator, a chain of calls, or embedded
1860         strings) then it's better to use regular strcat. I think that's what I carried over to here.
1861         
1862         Note that the previous change (Add result validation to JSAir) also made the Proxy code not
1863         dead, though it's not necessarily on the hot path. The Proxy isn't called into frequently
1864         but it's used from a function that is otherwise hot, so if calling into the Proxy prevents
1865         that function from being optimized then it will hurt so good.
1866         
1867         I also reenabled tail calls in a few places.
1868         
1869         This change doesn't seem to change the performance of the benchmark for us. That's expected
1870         since these ES6 features are cheap. Note that this claim doesn't include Proxy, which was
1871         added in a separate change and that change did make the benchmark overall more expensive.
1872
1873         * JSAir/allocate_stack.js:
1874         (allocateStack):
1875         * JSAir/arg.js:
1876         (Arg.createBitImm64):
1877         (Arg.createAddr):
1878         (Arg.createStack):
1879         (Arg.logScale):
1880         (Arg.createIndex):
1881         * JSAir/basic_block.js:
1882         (BasicBlock.get headerString):
1883         (BasicBlock.prototype.get if):
1884         (BasicBlock):
1885         * JSAir/benchmark.js:
1886         (benchmark):
1887         * JSAir/code.js:
1888         (Code):
1889         (Code.prototype.addBlock):
1890         (Code.prototype.addStackSlot):
1891         (Code.prototype.newTmp):
1892         (Code.prototype.get size):
1893         (Code.prototype.get blocks):
1894         (Code.prototype.get stackSlots):
1895         (Code.prototype.tmps):
1896         (Code.prototype.get callArgAreaSize):
1897         (Code.prototype.toString):
1898         * JSAir/custom.js:
1899         (const.PatchCustom.forEachArg):
1900         * JSAir/inst.js:
1901         (Inst):
1902         * JSAir/reg.js:
1903         (Reg.prototype.toString):
1904         * JSAir/util.js:
1905         (symbolName):
1906         (lowerSymbolName):
1907
1908 2016-06-07  Filip Pizlo  <fpizlo@apple.com>
1909
1910         Add result validation to JSAir
1911         https://bugs.webkit.org/show_bug.cgi?id=158493
1912
1913         Reviewed by Saam Barati.
1914         
1915         This adds the ability to hash a Code in a way that matches the C++ code's hashing of Code.
1916         This allows us to check if the Code that JSAir sees is the code that C++ saw. We use this to
1917         check the Code before and after allocateStack, and compare against hashes we got from C++.
1918         
1919         Doing this uncovered bugs. roundUpToMultipleOf wasn't doing anything. allocateStack was not
1920         allocating things correctly because I was concatting a Set to an Array, which doesn't really
1921         work. Now these bugs are fixed.
1922         
1923         The checking step adds to the running time so I reduced the number of iterations. The
1924         benchmark spends a decent amount of its time computing Code hashes; I think it's around 1/3
1925         total. This is probably OK. It's better to verify the results even if the running time is
1926         not all in the "core" of the algorithm.
1927         
1928         Also add a run-jsc-stress-tests yaml file to allow this to run as a test.
1929
1930         * JSAir/allocate_stack.js:
1931         * JSAir/arg.js:
1932         (Arg.createImm):
1933         (Arg.createBigImm):
1934         (Arg.createBitImm):
1935         (Arg.createBitImm64):
1936         (Arg.createWidth):
1937         (Arg.createSpecial):
1938         (Arg.prototype.get kind):
1939         (Arg.prototype.get isTmp):
1940         (Arg.prototype.get isImm):
1941         (Arg.prototype.get isSomeImm):
1942         (Arg.prototype.get isSomeBigImm):
1943         (Arg.prototype.get isCondition):
1944         (Arg.prototype.get isWidth):
1945         (Arg.prototype.get isSpecial):
1946         (Arg.prototype.get isAlive):
1947         (Arg.prototype.get tmp):
1948         (Arg.prototype.get value):
1949         (Arg.prototype.get lowValue):
1950         (Arg.prototype.get highValue):
1951         (Arg.prototype.get base):
1952         (Arg.prototype.get isGP):
1953         (Arg.prototype.get isFP):
1954         (Arg.prototype.isValidForm):
1955         (Arg.prototype.get isInvertible):
1956         (Arg.kindCode):
1957         (Arg.prototype.hash):
1958         (Arg.prototype.toString):
1959         (Arg):
1960         * JSAir/basic_block.js:
1961         (BasicBlock.get successorBlocks):
1962         * JSAir/benchmark.js:
1963         (benchmark):
1964         * JSAir/code.js:
1965         (Code.prototype.setFrameSize):
1966         (Code.prototype.hash):
1967         (Code.prototype.toString):
1968         (Code):
1969         * JSAir/inst.js:
1970         (Inst.prototype.get hasNonArgEffects):
1971         (Inst.prototype.hash):
1972         (Inst.prototype.toString):
1973         (Inst):
1974         * JSAir/jsair-tests.yaml: Added.
1975         * JSAir/opcode.js:
1976         (Inst_forEachArg):
1977         (Inst_hasNonArgEffects):
1978         (opcodeCode):
1979         * JSAir/payload-gbemu-executeIteration.js:
1980         (createPayloadGbemuExecuteIteration):
1981         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js:
1982         (createPayloadImagingGaussianBlurGaussianBlur):
1983         * JSAir/payload-jsair-ACLj8C.js:
1984         (createPayloadJSAirACLj8C):
1985         * JSAir/payload-typescript-scanIdentifier.js:
1986         (createPayloadTypescriptScanIdentifier):
1987         * JSAir/reg.js:
1988         (Reg.prototype.get isReg):
1989         (Reg.prototype.hash):
1990         (Reg.prototype.toString):
1991         * JSAir/stack_slot.js:
1992         (StackSlot.prototype.setOffsetFromFP):
1993         (StackSlot.prototype.hash):
1994         (StackSlot.prototype.toString):
1995         * JSAir/symbols.js:
1996         (relCondCode):
1997         (resCondCode):
1998         (doubleCondCode):
1999         * JSAir/test.html:
2000         * JSAir/tmp.js:
2001         (Tmp.prototype.get isReg):
2002         (Tmp.prototype.hash):
2003         (Tmp.prototype.toString):
2004         * JSAir/util.js:
2005         (roundUpToMultipleOf):
2006         (symbolName):
2007
2008 2016-06-07  Filip Pizlo  <fpizlo@apple.com>
2009
2010         Implement Air::allocateStack() in ES6 to see how much of a bad idea that is
2011         https://bugs.webkit.org/show_bug.cgi?id=158318
2012
2013         Reviewed by Saam Barati.
2014         
2015         This adds a new benchmark for us to play with called JSAir. It's a complete ES6
2016         implementation of Air's allocateStack() phase along with all of Air needed to run that
2017         phase. This includes things like stack slots, registers, temporaries, basic blocks,
2018         instructions, and all of the code for iterating over, inspecting, and modifying those
2019         things.
2020         
2021         To make this work, JSC can now dump Air just before allocateStack() in the form of JS code
2022         that creates a Code object that matches exactly what C++ Air saw. This benchmark comprises
2023         four Air IRs:
2024         
2025         - Octane/gbemu's largest function, executeIteration.
2026         - Kraken/imaging-gaussian-blur's largest function in OSR entry mode, gaussuanBlur.
2027         - Octane/typescript's largest function that is actually hot, scanIdentifier.
2028         - JSAir's largest hot function, which is anonymous, so we call it by its hash (ACLj8C).
2029         
2030         This runs in about 2 seconds on my machine in JSC trunk. It includes both a commandline
2031         harness and a web harness. JSAir is almost exactly 100x slower in ES6 in WebKit than the
2032         C++ Air::allocateStack() phase on which it is based.
2033         
2034         JSAir uses the following ES6 features:
2035         
2036         - Symbol.
2037         - for-of.
2038         - arrow functions.
2039         - Map/Set.
2040         - let/const.
2041         - classes.
2042         
2043         All of these things are used in anger and should end up on the hot path. There is also code
2044         that uses Proxies, but it ends up being dead.
2045         
2046         We can improve this even more:
2047         I still need to add result validation: https://bugs.webkit.org/show_bug.cgi?id=158493
2048         I want to make it use more ES6 features: https://bugs.webkit.org/show_bug.cgi?id=158497
2049         
2050         * JSAir: Added.
2051         * JSAir/all.js: Added.
2052         * JSAir/allocate_stack.js: Added.
2053         (allocateStack.attemptAssignment):
2054         (allocateStack.assign):
2055         (allocateStack.interfere):
2056         (allocateStack.):
2057         (allocateStack):
2058         * JSAir/arg.js: Added.
2059         (Arg):
2060         (Arg.isAnyUse):
2061         (Arg.isColdUse):
2062         (Arg.isWarmUse):
2063         (Arg.cooled):
2064         (Arg.isEarlyUse):
2065         (Arg.isLateUse):
2066         (Arg.isAnyDef):
2067         (Arg.isEarlyDef):
2068         (Arg.isLateDef):
2069         (Arg.isZDef):
2070         (Arg.typeForB3Type):
2071         (Arg.widthForB3Type):
2072         (Arg.conservativeWidth):
2073         (Arg.minimumWidth):
2074         (Arg.bytes):
2075         (Arg.widthForBytes):
2076         (Arg.createTmp):
2077         (Arg.fromReg):
2078         (Arg.createImm):
2079         (Arg.createBigImm):
2080         (Arg.createBitImm):
2081         (Arg.createBitImm64):
2082         (Arg.createAddr):
2083         (Arg.createStack):
2084         (Arg.createCallArg):
2085         (Arg.createStackAddr):
2086         (Arg.isValidScale):
2087         (Arg.logScale):
2088         (Arg.createIndex):
2089         (Arg.createRelCond):
2090         (Arg.createResCond):
2091         (Arg.createDoubleCond):
2092         (Arg.createWidth):
2093         (Arg.prototype.get kind):
2094         (Arg.prototype.get isTmp):
2095         (Arg.prototype.get isImm):
2096         (Arg.prototype.get isBigImm):
2097         (Arg.prototype.get isBitImm):
2098         (Arg.prototype.get isBitImm64):
2099         (Arg.prototype.get isSomeImm):
2100         (Arg.prototype.get isAddr):
2101         (Arg.prototype.get isStack):
2102         (Arg.prototype.get isCallArg):
2103         (Arg.prototype.get isIndex):
2104         (Arg.prototype.get isMemory):
2105         (Arg.prototype.get isStackMemory):
2106         (Arg.prototype.get isRelCond):
2107         (Arg.prototype.get isResCond):
2108         (Arg.prototype.get isDoubleCond):
2109         (Arg.prototype.get isCondition):
2110         (Arg.prototype.get isWidth):
2111         (Arg.prototype.get isAlive):
2112         (Arg.prototype.get tmp):
2113         (Arg.prototype.get value):
2114         (Arg.prototype.get base):
2115         (Arg.prototype.get hasOffset):
2116         (Arg.prototype.get offset):
2117         (Arg.prototype.get stackSlot):
2118         (Arg.prototype.get index):
2119         (Arg.prototype.get scale):
2120         (Arg.prototype.get logScale):
2121         (Arg.prototype.get width):
2122         (Arg.prototype.get isGPTmp):
2123         (Arg.prototype.get isFPTmp):
2124         (Arg.prototype.get isGP):
2125         (Arg.prototype.get isFP):
2126         (Arg.prototype.get hasType):
2127         (Arg.prototype.get type):
2128         (Arg.prototype.isType):
2129         (Arg.prototype.isCompatibleType):
2130         (Arg.prototype.get isGPR):
2131         (Arg.prototype.get gpr):
2132         (Arg.prototype.get isFPR):
2133         (Arg.prototype.get fpr):
2134         (Arg.prototype.get isReg):
2135         (Arg.prototype.get reg):
2136         (Arg.isValidImmForm):
2137         (Arg.isValidBitImmForm):
2138         (Arg.isValidBitImm64Form):
2139         (Arg.isValidAddrForm):
2140         (Arg.isValidIndexForm):
2141         (Arg.prototype.isValidForm):
2142         (Arg.prototype.forEachTmpFast):
2143         (Arg.prototype.usesTmp):
2144         (Arg.prototype.forEachTmp):
2145         (Arg.prototype.is):
2146         (Arg.prototype.as):
2147         (Arg.prototype.forEachFast):
2148         (Arg.prototype.forEach):
2149         (Arg.extract):
2150         (Arg.forEachFast):
2151         (Arg.forEach):
2152         (Arg.prototype.get condition):
2153         (Arg.prototype.get isInvertible):
2154         (Arg.prototype.toString):
2155         * JSAir/basic_block.js: Added.
2156         (BasicBlock):
2157         (BasicBlock.prototype.get index):
2158         (BasicBlock.prototype.get size):
2159         (BasicBlock.prototype.Symbol.iterator):
2160         (BasicBlock.prototype.at):
2161         (BasicBlock.get last):
2162         (BasicBlock.get insts):
2163         (BasicBlock.get numSuccessors):
2164         (BasicBlock.get successors):
2165         (BasicBlock.get successorBlocks.):
2166         (BasicBlock.get successorBlocks):
2167         (BasicBlock.set get numPredecessors):
2168         (BasicBlock.get predecessors):
2169         (BasicBlock.get frequency):
2170         (BasicBlock.get headerString):
2171         (BasicBlock.get footerString):
2172         * JSAir/benchmark.js: Added.
2173         (benchmark):
2174         * JSAir/code.js: Added.
2175         (Code):
2176         (Code.prototype.addBlock):
2177         (Code.prototype.addStackSlot):
2178         (Code.prototype.newTmp):
2179         (Code.prototype.get size):
2180         (Code.prototype.at):
2181         (Code.prototype.Symbol.iterator):
2182         (Code.prototype.get blocks):
2183         (Code.prototype.get stackSlots):
2184         (Code.prototype.tmps):
2185         (Code.prototype.get callArgAreaSize):
2186         (Code.prototype.requestCallArgAreaSize):
2187         (Code.prototype.get frameSize):
2188         (Code.prototype.setFrameSize):
2189         (Code.prototype.toString):
2190         * JSAir/custom.js: Added.
2191         (const.ShuffleCustom.forEachArg):
2192         (const.ShuffleCustom.hasNonArgNonControlEffects):
2193         (const.PatchCustom.forEachArg):
2194         (const.PatchCustom.hasNonArgNonControlEffects):
2195         (const.CCallCustom.forEachArg):
2196         (const.CCallCustom.hasNonArgNonControlEffects):
2197         (const.ColdCCallCustom.forEachArg):
2198         (const.ColdCCallCustom.hasNonArgNonControlEffects):
2199         * JSAir/frequented_block.js: Added.
2200         (FrequentedBlock):
2201         (FrequentedBlock.prototype.toString):
2202         * JSAir/insertion_set.js: Added.
2203         (Insertion):
2204         (Insertion.prototype.get index):
2205         (Insertion.prototype.get element):
2206         (Insertion.prototype.lessThan):
2207         (InsertionSet):
2208         (InsertionSet.prototype.appendInsertion):
2209         (InsertionSet.prototype.append):
2210         (InsertionSet.prototype.execute):
2211         * JSAir/inst.js: Added.
2212         (Inst):
2213         (Inst.prototype.append):
2214         (Inst.prototype.clear):
2215         (Inst.prototype.get opcode):
2216         (Inst.prototype.get args):
2217         (Inst.prototype.visitArg):
2218         (Inst.prototype.forEachTmpFast):
2219         (Inst.prototype.forEachArg):
2220         (Inst.prototype.forEachTmp):
2221         (Inst.prototype.forEach):
2222         (Inst.forEachDef):
2223         (Inst.forEachDefWithExtraClobberedRegs):
2224         (Inst.prototype.get hasNonArgEffects):
2225         (Inst.prototype.toString):
2226         * JSAir/liveness.js: Added.
2227         (Liveness):
2228         (Liveness.prototype.get thing):
2229         (Liveness.prototype.get code):
2230         (Liveness.prototype.get liveAtHead):
2231         (Liveness.prototype.get liveAtTail):
2232         (Liveness.prototype.localCalc.LocalCalc):
2233         (Liveness.prototype.localCalc.LocalCalc.prototype.get liveSet):
2234         (Liveness.prototype.localCalc.LocalCalc.prototype.execute):
2235         (Liveness.prototype.localCalc):
2236         * JSAir/opcode.js: Added.
2237         (Inst_forEachArg):
2238         (Inst_hasNonArgEffects):
2239         * JSAir/payload-gbemu-executeIteration.js: Added.
2240         (createPayloadGbemuExecuteIteration):
2241         * JSAir/payload-imaging-gaussian-blur-gaussianBlur.js: Added.
2242         (createPayloadImagingGaussianBlurGaussianBlur):
2243         * JSAir/payload-jsair-ACLj8C.js: Added.
2244         (createPayloadJSAirACLj8C):
2245         * JSAir/payload-typescript-scanIdentifier.js: Added.
2246         (createPayloadTypescriptScanIdentifier):
2247         * JSAir/reg.js: Added.
2248         (Reg):
2249         (Reg.fromReg):
2250         (Reg.prototype.get index):
2251         (Reg.prototype.get type):
2252         (Reg.prototype.get name):
2253         (Reg.prototype.get isCalleeSave):
2254         (Reg.prototype.get isReg):
2255         (Reg.prototype.toString):
2256         (Reg.extract):
2257         (Reg.forEachFast):
2258         (Reg.forEach):
2259         (newGPR):
2260         (Reg.gprs.Reg.fprs.Reg.calleeSaveGPRs.Reg.calleeSaveFPRs.Reg.calleeSaves):
2261         * JSAir/stack_slot.js: Added.
2262         (StackSlot):
2263         (StackSlot.prototype.get byteSize):
2264         (StackSlot.prototype.get kind):
2265         (StackSlot.prototype.get isLocked):
2266         (StackSlot.prototype.get isSpill):
2267         (StackSlot.prototype.get index):
2268         (StackSlot.prototype.ensureSize):
2269         (StackSlot.prototype.get alignment):
2270         (StackSlot.prototype.get offsetFromFP):
2271         (StackSlot.prototype.setOffsetFromFP):
2272         (StackSlot.prototype.toString):
2273         (StackSlot.extract):
2274         (StackSlot.forEachFast):
2275         (StackSlot.forEach):
2276         * JSAir/symbols.js: Added.
2277         * JSAir/test.html: Added.
2278         * JSAir/test.js: Added.
2279         * JSAir/tmp.js: Added.
2280         (Tmp):
2281         (Tmp.fromReg):
2282         (Tmp.prototype.get index):
2283         (Tmp.prototype.get type):
2284         (Tmp.prototype.get isReg):
2285         (Tmp.prototype.toString):
2286         (Tmp.extract):
2287         (Tmp.forEachFast):
2288         (Tmp.forEach):
2289         * JSAir/tmp_base.js: Added.
2290         (TmpBase.prototype.get isGP):
2291         (TmpBase.prototype.get isFP):
2292         (TmpBase.prototype.get isGPR):
2293         (TmpBase.prototype.get isFPR):
2294         (TmpBase.prototype.get reg):
2295         (TmpBase.prototype.get gpr):
2296         (TmpBase.prototype.get fpr):
2297         (TmpBase):
2298         * JSAir/util.js: Added.
2299         (isRepresentableAsInt32):
2300         (addIndexed):
2301         (roundUpToMultipleOf):
2302         (symbolName):
2303         (mergeIntoSet):
2304         (nonEmptyRangesOverlap):
2305         (rangesOverlap):
2306         (removeAllMatching):
2307         (swap):
2308         (bubble):
2309         (bubbleSort):
2310         (currentTime):
2311         (else.currentTime):
2312
2313 2016-05-26  Yusuke Suzuki  <utatane.tea@gmail.com>
2314
2315         [JSC] Allow JSBench to use precise time
2316         https://bugs.webkit.org/show_bug.cgi?id=158050
2317
2318         Reviewed by Geoffrey Garen.
2319
2320         * JSBench/amazon-chrome-win/urem.js:
2321         (else.window.performance.window.performance.now.currentTimeInMS):
2322         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2323         (else.else.currentTimeInMS):
2324         (onload.cb):
2325         (onload):
2326         * JSBench/amazon-chrome/urem.js:
2327         (else.window.performance.window.performance.now.currentTimeInMS):
2328         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2329         (else.else.currentTimeInMS):
2330         (onload.cb):
2331         (onload):
2332         * JSBench/amazon-firefox-win/urm.js:
2333         (else.window.performance.window.performance.now.currentTimeInMS):
2334         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2335         (else.else.currentTimeInMS):
2336         (onload.cb):
2337         (onload):
2338         * JSBench/amazon-firefox/urm.js:
2339         (else.window.performance.window.performance.now.currentTimeInMS):
2340         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2341         (else.else.currentTimeInMS):
2342         (onload.cb):
2343         (onload):
2344         * JSBench/amazon-safari/urem.js:
2345         (else.window.performance.window.performance.now.currentTimeInMS):
2346         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2347         (else.else.currentTimeInMS):
2348         (onload.cb):
2349         (onload):
2350         * JSBench/facebook-chrome-win/urem.js:
2351         (else.window.performance.window.performance.now.currentTimeInMS):
2352         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2353         (else.else.currentTimeInMS):
2354         (onload.cb):
2355         (onload):
2356         * JSBench/facebook-chrome/urem.js:
2357         (else.window.performance.window.performance.now.currentTimeInMS):
2358         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2359         (else.else.currentTimeInMS):
2360         (onload.cb):
2361         (onload):
2362         * JSBench/facebook-firefox-win/urem.js:
2363         (else.window.performance.window.performance.now.currentTimeInMS):
2364         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2365         (else.else.currentTimeInMS):
2366         (onload.cb):
2367         (onload):
2368         * JSBench/facebook-firefox/urem.js:
2369         (else.window.performance.window.performance.now.currentTimeInMS):
2370         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2371         (else.else.currentTimeInMS):
2372         (onload.cb):
2373         (onload):
2374         * JSBench/facebook-safari/urem.js:
2375         (else.window.performance.window.performance.now.currentTimeInMS):
2376         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2377         (else.else.currentTimeInMS):
2378         (onload.cb):
2379         (onload):
2380         * JSBench/google-chrome-win/urem.js:
2381         (else.window.performance.window.performance.now.currentTimeInMS):
2382         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2383         (else.else.currentTimeInMS):
2384         (onload.cb):
2385         (onload):
2386         * JSBench/google-chrome/urem.js:
2387         (else.window.performance.window.performance.now.currentTimeInMS):
2388         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2389         (else.else.currentTimeInMS):
2390         (onload.cb):
2391         (onload):
2392         * JSBench/google-firefox-win/urem.js:
2393         (else.window.performance.window.performance.now.currentTimeInMS):
2394         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2395         (else.else.currentTimeInMS):
2396         (onload.cb):
2397         (onload):
2398         * JSBench/google-firefox/uem.js:
2399         (else.window.performance.window.performance.now.currentTimeInMS):
2400         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2401         (else.else.currentTimeInMS):
2402         (onload.cb):
2403         (onload):
2404         * JSBench/google-safari/urem.js:
2405         (else.window.performance.window.performance.now.currentTimeInMS):
2406         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2407         (else.else.currentTimeInMS):
2408         (onload.cb):
2409         (onload):
2410         * JSBench/harness.js:
2411         (runBenchmark.window.currentTimeInMS):
2412         (runBenchmark.else.window.currentTimeInMS):
2413         * JSBench/twitter-chrome-win/rem.js:
2414         (else.window.performance.window.performance.now.currentTimeInMS):
2415         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2416         (else.else.currentTimeInMS):
2417         (onload.cb):
2418         (onload):
2419         * JSBench/twitter-chrome/urem.js:
2420         (else.window.performance.window.performance.now.currentTimeInMS):
2421         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2422         (else.else.currentTimeInMS):
2423         (onload.cb):
2424         (onload):
2425         * JSBench/twitter-firefox-win/urem.js:
2426         (else.window.performance.window.performance.now.currentTimeInMS):
2427         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2428         (else.else.currentTimeInMS):
2429         (onload.cb):
2430         (onload):
2431         * JSBench/twitter-firefox/urem.js:
2432         (else.window.performance.window.performance.now.currentTimeInMS):
2433         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2434         (else.else.currentTimeInMS):
2435         (onload.cb):
2436         (onload):
2437         * JSBench/twitter-safari/urem.js:
2438         (else.window.performance.window.performance.now.currentTimeInMS):
2439         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2440         (else.else.currentTimeInMS):
2441         (onload.cb):
2442         (onload):
2443         * JSBench/yahoo-chrome-win/urem.js:
2444         (else.window.performance.window.performance.now.currentTimeInMS):
2445         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2446         (else.else.currentTimeInMS):
2447         (onload.cb):
2448         (onload):
2449         * JSBench/yahoo-chrome/urem.js:
2450         (else.window.performance.window.performance.now.currentTimeInMS):
2451         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2452         (else.else.currentTimeInMS):
2453         (onload.cb):
2454         (onload):
2455         * JSBench/yahoo-firefox-win/urem.js:
2456         (else.window.performance.window.performance.now.currentTimeInMS):
2457         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2458         (else.else.currentTimeInMS):
2459         (onload.cb):
2460         (onload):
2461         * JSBench/yahoo-firefox/urem.js:
2462         (else.window.performance.window.performance.now.currentTimeInMS):
2463         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2464         (else.else.currentTimeInMS):
2465         (onload.cb):
2466         (onload):
2467         * JSBench/yahoo-safari/urem.js:
2468         (else.window.performance.window.performance.now.currentTimeInMS):
2469         (else.else.typeof.preciseTime.string_appeared_here.currentTimeInMS):
2470         (else.else.currentTimeInMS):
2471         (onload.cb):
2472         (onload):
2473
2474 2016-05-25  Geoffrey Garen  <ggaren@apple.com>
2475
2476         replaceable own properties seem to ignore replacement after property caching
2477         https://bugs.webkit.org/show_bug.cgi?id=158091
2478
2479         Reviewed by Darin Adler.
2480
2481         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
2482         * MallocBench/MallocBench/Benchmark.cpp:
2483         * MallocBench/MallocBench/Interpreter.cpp:
2484         (Interpreter::doMallocOp):
2485         * MallocBench/MallocBench/Interpreter.h:
2486         * MallocBench/MallocBench/fastMallocLog.63316.ops: Added.
2487         * MallocBench/MallocBench/jetstream.cpp: Added.
2488         (benchmark_jetstream):
2489         * MallocBench/MallocBench/jetstream.h: Added.
2490
2491 2016-05-25  Keith Miller  <keith_miller@apple.com>
2492
2493         Unreviewed, add JSBench to the skipped list for now since it doesn't
2494         work currently.
2495
2496         * Skipped:
2497
2498 2016-05-17  Keith Miller  <keith_miller@apple.com>
2499         We should have JSBench in PerformanceTests
2500         https://bugs.webkit.org/show_bug.cgi?id=157952
2501
2502         Rubber-stamped by Saam Barati.
2503
2504         There are some slight changes to the layout of the test directory
2505         to make it work nicely with run-jsc-benchmarks. Before JSBench
2506         had each of the browser specific sub-tests in a sub-directory.
2507         These have been flattened e.g. amazon/safari/ has become
2508         amazon-safari/.
2509
2510         * JSBench/amazon-chrome-win/urem.html: Added.
2511         * JSBench/amazon-chrome-win/urem.js: Added.
2512         * JSBench/amazon-chrome/urem.html: Added.
2513         * JSBench/amazon-chrome/urem.js: Added.
2514         * JSBench/amazon-firefox-win/urm.html: Added.
2515         * JSBench/amazon-firefox-win/urm.js: Added.
2516         * JSBench/amazon-firefox/urm.html: Added.
2517         * JSBench/amazon-firefox/urm.js: Added.
2518         * JSBench/amazon-safari/urem.html: Added.
2519         * JSBench/amazon-safari/urem.js: Added.
2520         * JSBench/browsercheck.js: Added.
2521         * JSBench/facebook-chrome-win/urem.html: Added.
2522         * JSBench/facebook-chrome-win/urem.js: Added.
2523         * JSBench/facebook-chrome/urem.html: Added.
2524         * JSBench/facebook-chrome/urem.js: Added.
2525         * JSBench/facebook-firefox-win/urem.html: Added.
2526         * JSBench/facebook-firefox-win/urem.js: Added.
2527         * JSBench/facebook-firefox/urem.html: Added.
2528         * JSBench/facebook-firefox/urem.js: Added.
2529         * JSBench/facebook-safari/urem.html: Added.
2530         * JSBench/facebook-safari/urem.js: Added.
2531         * JSBench/google-chrome-win/urem.html: Added.
2532         * JSBench/google-chrome-win/urem.js: Added.
2533         * JSBench/google-chrome/urem.html: Added.
2534         * JSBench/google-chrome/urem.js: Added.
2535         * JSBench/google-firefox-win/urem.html: Added.
2536         * JSBench/google-firefox-win/urem.js: Added.
2537         * JSBench/google-firefox/uem.html: Added.
2538         * JSBench/google-firefox/uem.js: Added.
2539         * JSBench/google-safari/urem.html: Added.
2540         * JSBench/google-safari/urem.js: Added.
2541         * JSBench/harness.html: Added.
2542         * JSBench/harness.js: Added.
2543         * JSBench/harness.py: Added.
2544         * JSBench/index.html: Added.
2545         * JSBench/reload.html: Added.
2546         * JSBench/twitter-chrome-win/rem.html: Added.
2547         * JSBench/twitter-chrome-win/rem.js: Added.
2548         * JSBench/twitter-chrome/urem.html: Added.
2549         * JSBench/twitter-chrome/urem.js: Added.
2550         * JSBench/twitter-firefox-win/urem.html: Added.
2551         * JSBench/twitter-firefox-win/urem.js: Added.
2552         * JSBench/twitter-firefox/urem.html: Added.
2553         * JSBench/twitter-firefox/urem.js: Added.
2554         * JSBench/twitter-safari/urem.html: Added.
2555         * JSBench/twitter-safari/urem.js: Added.
2556         * JSBench/yahoo-chrome-win/urem.html: Added.
2557         * JSBench/yahoo-chrome-win/urem.js: Added.
2558         * JSBench/yahoo-chrome/urem.html: Added.
2559         * JSBench/yahoo-chrome/urem.js: Added.
2560         * JSBench/yahoo-firefox-win/urem.html: Added.
2561         * JSBench/yahoo-firefox-win/urem.js: Added.
2562         * JSBench/yahoo-firefox/urem.html: Added.
2563         * JSBench/yahoo-firefox/urem.js: Added.
2564         * JSBench/yahoo-safari/urem.html: Added.
2565         * JSBench/yahoo-safari/urem.js: Added.
2566
2567 2016-05-20  Ryosuke Niwa  <rniwa@webkit.org>
2568
2569         run-benchmark's results should contain Animometer's debug output
2570         https://bugs.webkit.org/show_bug.cgi?id=157941
2571
2572         Reviewed by Stephanie Lewis.
2573
2574         Made developer.html support the JSON generated by run-benchmark which stores everything under debugOutput.
2575
2576         * Animometer/resources/debug-runner/animometer.js:
2577         (Utilities.initialize): Unwrap debugOutput in the case run-benchmark's result JSON is used.
2578
2579 2016-05-18  Timothy Hatcher  <timothy@apple.com>
2580
2581         Make Animometer work in all browsers
2582         https://bugs.webkit.org/show_bug.cgi?id=157855
2583         rdar://problem/26338521
2584
2585         Reviewed by Jon Lee.
2586
2587         * Animometer/resources/runner/logo.svg: Use a symbol instead
2588         of putting the id on the root svg. This works in all browsers.
2589
2590 2016-05-07  Timothy Hatcher  <timothy@apple.com>
2591
2592         Update Animometer design
2593         https://bugs.webkit.org/show_bug.cgi?id=157449
2594
2595         Reviewed by Darin Adler.
2596
2597         * Animometer/developer.html: Stop including runner/animometer.css.
2598         * Animometer/index.html: Udpated elements to fit new styles.
2599
2600         * Animometer/resources/debug-runner/animometer.css:
2601         Copy the old runner/animometer.css to here so the debug runner is unaffected.
2602
2603         * Animometer/resources/runner/animometer.css:
2604         New styles for the design. Drop flex box for broad browser support.
2605
2606         * Animometer/resources/runner/animometer.js:
2607         (ResultsTable.clear): Use textContent.
2608         (window.sectionsManager.showSection): Add class to the body.
2609         (window.sectionsManager.setSectionScore): Use textContent, not innerHTML.
2610
2611         * Animometer/resources/runner/crystal.svg: Added.
2612         * Animometer/resources/runner/lines.svg: Added.
2613         * Animometer/resources/runner/logo.svg: Added.
2614
2615         * Animometer/resources/runner/tests.js: Shorten test name and title case them.
2616
2617 2016-05-06  Manuel Rego Casasnovas  <rego@igalia.com>
2618
2619         [css-grid] Unprefix CSS Grid Layout properties
2620         https://bugs.webkit.org/show_bug.cgi?id=157137
2621
2622         Reviewed by Simon Fraser.
2623
2624         Remove "-webkit" prefix from all the grid layout properties,
2625         including the display value.
2626         Update the source code to remove the prefix where it was used too.
2627
2628         * Layout/auto-grid-lots-of-data.html:
2629         * Layout/fixed-grid-lots-of-data.html:
2630         * Layout/fixed-grid-lots-of-stretched-data.html:
2631
2632 2016-04-25  Geoffrey Garen  <ggaren@apple.com>
2633
2634         bmalloc: Misc improvements to MallocBench
2635         https://bugs.webkit.org/show_bug.cgi?id=157004
2636
2637         Reviewed by Darin Adler.
2638
2639         * MallocBench/run-malloc-benchmarks: Added --memory and --memory_warning
2640         modes for focused memory testing.
2641
2642         * MallocBench/MallocBench/Benchmark.cpp:
2643         (Benchmark::printReport): Clarified output.
2644
2645         (Benchmark::currentMemoryBytes): Added compressed memory because top
2646         does the same. (It always happens to zero in the benchmarks we run. But
2647         this is good for sanity.)
2648
2649         * MallocBench/MallocBench/CommandLine.cpp: Moved up to 8 runs to reduce
2650         variance.
2651
2652         * MallocBench/MallocBench/alloc_free.cpp:
2653         (benchmark_alloc_free): Cycle a single allocation in order to stress
2654         the effect of merging on calls to madvise.
2655
2656         * MallocBench/MallocBench/big.cpp:
2657         (benchmark_big): Graduated to 8kB-128kB because medium tests up to 8 and
2658         our large allocator doesn't kick in until 64kB.
2659
2660         * MallocBench/MallocBench/medium.cpp:
2661         (benchmark_medium): Test all the way down to 1kB because our large
2662         allocator used to service 1kB allocations and 1kB is an interesting
2663         middle size where memory is unusually large but allocation throughput
2664         still matters.
2665
2666         * MallocBench/MallocBench/stress.cpp:
2667         (benchmark_stress): Reduced the churn count to match stress_aligned
2668         because this test was taking too long to complete.
2669
2670         * MallocBench/MallocBench/stress_aligned.cpp:
2671         (benchmark_stress_aligned): Our new large allocator can handle even
2672         more absurdly large values.
2673
2674 2016-04-25  Simon Fraser  <simon.fraser@apple.com>
2675
2676         Add a content animation test that uses SVG animation.
2677         https://bugs.webkit.org/show_bug.cgi?id=156827
2678
2679         * Animation/svg-animation.html: Added.
2680         * Skipped:
2681
2682 2016-04-22  Simon Fraser  <simon.fraser@apple.com>
2683
2684         Skip two content animation tests which are only meant for iOS testing.
2685
2686         * Animation/css-animation.html: Added.
2687         * Animation/raf-animation.html: Added.
2688
2689         * Skipped:
2690
2691 2016-04-20  Simon Fraser  <simon.fraser@apple.com>
2692
2693         Add content animation tests to benchmark_runner, and allow the runner to collect device data as part of the results
2694         https://bugs.webkit.org/show_bug.cgi?id=156827
2695
2696         Add two files missing from the previous commit.
2697
2698         * Animation/css-animation.html: Added.
2699         * Animation/raf-animation.html: Added.
2700
2701 2016-04-20  Simon Fraser  <simon.fraser@apple.com>
2702
2703         Add content animation tests to benchmark_runner, and allow the runner to collect device data as part of the results
2704         https://bugs.webkit.org/show_bug.cgi?id=156827
2705
2706         Reviewed by Stephanie Lewis.
2707
2708         Two new files for the content-animation suite.
2709
2710         After patching for use in the benchmark, these tests are a little different to other benchmarks.
2711         The test content itself does not collect performance data; the tests spit out a boilerplate JSON
2712         result with a placeholder for framerate, which is collected in native code.
2713
2714         When run as a benchmark, test completion involves a couple of bounces between the test and MobileSafari.
2715         Test completion is initiated by a setTimeout() in the test, at which point location.hash is set to "#done".
2716         MobileSafari detects that, and asynchronously dumps performance data. Once this is complete, MobileSafari
2717         changes the URL has to "#submit", which triggers the XHR to the test relay that indicates the test is
2718         complete.
2719
2720         * Animation/css-animation.html: Added.
2721         * Animation/raf-animation.html: Added.
2722
2723 2016-04-15  Jon Lee  <jonlee@apple.com>
2724
2725         Animometer test could report a NaN
2726         https://bugs.webkit.org/show_bug.cgi?id=156646
2727
2728         Reviewed by Darin Adler.
2729         Provisionally reviewed by Said Abou-Hallawa.
2730
2731         * Animometer/tests/resources/main.js:
2732         (didFinishInterval): The ramp controller has a first phase where it ramps up the complexity and
2733         reacts based on how well the system handles the load. The assumption was that it would handle at least
2734         1 particle easily. That is not always the case. As a result, an interpolation calculation could
2735         end up setting an upper bound of NaN.
2736
2737         This occurs because we never get out of the first tier, so this._lastTierComplexity is undefined.
2738         Now that we guarantee a minimum complexity of 1, modify the conditional to check for this._lastTierComplexity
2739         before interpolating the upper bound of the first ramp. In the case where the system struggles with
2740         1 particle, set it to currentComplexity.
2741
2742 2016-04-09  Jon Lee  <jonlee@apple.com>
2743
2744         Update Animometer to accommodate different screens
2745         https://bugs.webkit.org/show_bug.cgi?id=156449
2746
2747         Reviewed by Darin Adler.
2748         Provisionally reviewed by Said Abou-Hallawa.
2749
2750         * Animometer/index.html: Wrap button in a container to add padding at the bottom.
2751         * Animometer/resources/debug-runner/animometer.css:
2752         (@media screen and (min-device-width: 1800px)): Deleted.
2753         (@media screen and (min-width: 1800px)): Cannot use min-device-width since it may match incorrectly.
2754         (screen and (max-device-height: 414px) and (orientation: landscape)): Some devices swap device width
2755         and height with orientation change.
2756         * Animometer/resources/runner/animometer.css: Similar.
2757         (screen and (min-device-width: 1024px) and (orientation: landscape)):
2758         (screen and (max-device-height: 414px) and (orientation: landscape)):
2759         (.frame-container): On smaller iPhones, adding 1px prevents the navigation bars from appearing.
2760         (@media screen and (min-device-width: 768px) and (max-device-width: 1024px)): Deleted.
2761         (@media (min-device-height: 768px) and (max-device-height: 1024px)): Target iPad Airs and similar.
2762         (@media screen and (min-device-width: 1024px) and (max-device-width: 1366px)): Deleted.
2763         (@media screen and (max-device-width: 1024px) and (min-device-height: 1366px)): Target iPad Pro.
2764         (#results footer): Add padding below the button for testing again.
2765         * Animometer/tests/master/multiply.html: Remove the center text.
2766         * Animometer/tests/master/resources/text.js: Update the test so that in every frame the text moves.
2767         * Animometer/tests/master/text.html: Update the text sizing depending on the size of the device.
2768
2769 2016-04-08  Jon Lee  <jonlee@apple.com>
2770
2771         Have Animometer benchmark always start with complexity of 1
2772         https://bugs.webkit.org/show_bug.cgi?id=156432
2773
2774         Reviewed by Ryosuke Niwa.
2775
2776         * Animometer/tests/resources/main.js: Update the default Controller and RampController to
2777         set its minimum complexities to 1 instead of 0.
2778
2779 2016-04-08  Jon Lee  <jonlee@apple.com>
2780
2781         Fix SVG benchmark test
2782         https://bugs.webkit.org/show_bug.cgi?id=156410
2783
2784         Reviewed by Dean Jackson.
2785
2786         * Animometer/resources/extensions.js: Update Point.zero to be a static Point.
2787         * Animometer/tests/simple/resources/tiled-canvas-image.js:
2788         (Stage.call._setupTiles): Refactor.
2789         * Animometer/tests/master/resources/particles.js:
2790         (Particle.prototype.reset): Use Point.center.
2791         (complexity): We are not using a gradient background anymore, so remove the +1.
2792         * Animometer/tests/master/resources/svg-particles.js: Update to use SVG transform
2793         instead of CSS transform.
2794
2795 2016-04-06  Jon Lee  <jonlee@apple.com>
2796
2797         Update master benchmark with SVG test
2798         https://bugs.webkit.org/show_bug.cgi?id=156273
2799
2800         Reviewed by Dean Jackson.
2801         Provisionally reviewed by Said Abou-Hallawa.
2802
2803         Switch masks tests for SVG path test.
2804
2805         * Animometer/resources/debug-runner/tests.js: Move mask test here.
2806         * Animometer/resources/runner/tests.js: Add SVG test here.
2807         * Animometer/tests/dom/particles.html: Renamed from PerformanceTests/Animometer/tests/master/particles.html.
2808         * Animometer/tests/dom/resources/dom-particles.js: Renamed from PerformanceTests/Animometer/tests/master/resources/dom-particles.js.
2809
2810         * Animometer/tests/master/resources/particles.js: Add minPosition for bounds checking. Prevents particle from being
2811         partially obscured.
2812         * Animometer/tests/master/resources/svg-particles.js: Added.
2813         (Particle): The particle is either a path object or a rect using a path as a clip. The
2814         same path is used either way. For each particle create a linear gradient with a random
2815         rotation.
2816         (SVGParticlesStage): Look in #shapes to see how many different kinds of paths are available.
2817         This makes the test more generic in case other shapes need to be tested.
2818         * Animometer/tests/master/svg-particles.html: Added. Have two defs, one that houses each
2819         particle's gradient, and one that holds the shape templates.
2820
2821 2016-03-27  Hunseop Jeong  <hs85.jeong@samsung.com>
2822
2823         [EFL] REGRESSION(r188793): It made 200 layout tests and Bindings/event-target-wrapper.html performance test fail
2824         https://bugs.webkit.org/show_bug.cgi?id=148470 
2825
2826         Reviewed by Darin Adler.
2827
2828         * Skipped: Unskip the Bindings/event-target-wrapper test.
2829
2830 2016-03-23  Jon Lee  <jonlee@apple.com>
2831
2832         Update focus and leaves tests
2833         https://bugs.webkit.org/show_bug.cgi?id=155825
2834
2835         Reviewed by Simon Fraser.
2836
2837         Particles would improperly disappear, before they were fully off stage. Update the calculation.
2838         * Animometer/tests/dom/resources/leaves.js:
2839         * Animometer/tests/master/resources/leaves.js:
2840
2841         * Animometer/tests/master/focus.html: Update the center element styling.
2842         * Animometer/tests/master/resources/focus.js:
2843         (FocusElement.Utilities.createClass): Give enough space in the container for each particle for the blur
2844         to spread. It should be about 3x more space than the blur radius on each side.
2845         (Stage.call.initialize): Ditto for the center element.
2846
2847 2016-03-23  Jon Lee  <jonlee@apple.com>
2848
2849         Update support for other platforms
2850         https://bugs.webkit.org/show_bug.cgi?id=155824
2851
2852         Reviewed by Simon Fraser.
2853
2854         * Animometer/resources/debug-runner/animometer.css: Remove extraneous media query. All of it should
2855         be handled in the release stylesheet.
2856         * Animometer/resources/debug-runner/animometer.js: Add support for browsers that return an HTMLCollection
2857         when retrieving radio elements with a name, instead of the selected value.
2858         * Animometer/resources/runner/animometer.css: Adjust the queries for devices of varying widths.
2859         (.frame-container): Move overflow:hidden to #stage.
2860         * Animometer/tests/master/resources/stage.css: Consolidate into tests/resources/stage.css.
2861         * Animometer/tests/resources/stage.css:
2862         (body): Remove overflow:hidden.
2863         (#stage): Set overflow:hidden here.
2864         (#center-text): Moved from master/resources/stage.css.
2865
2866         Update stylesheet hrefs.
2867         * Animometer/tests/master/canvas-stage.html:
2868         * Animometer/tests/master/focus.html:
2869         * Animometer/tests/master/leaves.html:
2870         * Animometer/tests/master/multiply.html:
2871         * Animometer/tests/master/particles.html:
2872         * Animometer/tests/master/text.html: Adjust text size to fit in the device screen.
2873
2874 2016-03-23  Jon Lee  <jonlee@apple.com>
2875
2876         Make the benchmark require device to be in landscape orientation
2877         https://bugs.webkit.org/show_bug.cgi?id=155822
2878         rdar://problem/25258650
2879
2880         Reviewed by Ryosuke Niwa.
2881
2882         Check that the device is in landscape orientation. If not, disable the button to start it,
2883         and show a warning message.
2884
2885         * Animometer/developer.html: Add a message about how the browser should be set up. Include a
2886         warning message that appears if the orientation is incorrect on mobile devices.
2887         * Animometer/index.html: Ditto.
2888
2889         * Animometer/resources/debug-runner/animometer.css: Migrate the .hidden rule into the release
2890         stylesheet. Update the style to accommodate the new UI.
2891         * Animometer/resources/runner/animometer.css:
2892
2893         * Animometer/resources/runner/animometer.js:
2894         (window.benchmarkController.initialize): Add the orientation listener is needed.
2895         (window.benchmarkController.addOrientationListenerIfNecessary): Only mobile devices need this. Check
2896         to see for support of window.orientation.
2897         (window.benchmarkController._orientationChanged): Toggle the warning based on whether we match the
2898         landscape query. Set a state variable, which is needed for the debug harness. Call updateStartButtonState.
2899         (window.benchmarkController.updateStartButtonState):
2900         * Animometer/resources/debug-runner/animometer.js: Change _updateStartButtonState to return
2901         a boolean about whether at least one test is selected. That will be used in
2902         benchmarkController.updateStartButtonState(). Move the load event listener to the release version.
2903         (window.benchmarkController.updateStartButtonState): Override the release version, and also check
2904         that at least one test is selected.
2905
2906 2016-03-22  Geoffrey Garen  <ggaren@apple.com>
2907
2908         MallocBench: consolidate regression testing for aligned allocation
2909         https://bugs.webkit.org/show_bug.cgi?id=155762
2910
2911         Reviewed by Michael Saboff.
2912
2913         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
2914         * MallocBench/MallocBench/Benchmark.cpp:
2915         * MallocBench/MallocBench/memalign.cpp: Removed.
2916         * MallocBench/MallocBench/memalign.h: Removed. The stress_aligned test
2917         covers this and much more.
2918
2919         * MallocBench/MallocBench/stress_aligned.cpp:
2920         (benchmark_stress_aligned): Include specific tests for extreme sizes
2921         and alignments.
2922
2923 2016-03-21  Jon Lee  <jonlee@apple.com>
2924
2925         Update benchmark tests
2926         https://bugs.webkit.org/show_bug.cgi?id=155723
2927
2928         Reviewed by Darin Adler.
2929         Provisionally reviewed by Said Abou-Hallawa.
2930
2931         Add quadratic and bezier segments to the canvas path test.
2932
2933         * Animometer/resources/runner/tests.js: Some of the query strings are unnecessary.
2934         Rename the test.
2935         * Animometer/tests/master/resources/canvas-stage.js:
2936         (tune): Update to be able to take an array of possible constructors. Choose one randomly.
2937         * Animometer/tests/master/resources/canvas-tests.js:
2938         (CanvasLinePoint.Utilities.createClass): Move the point selection out to a separate
2939         function called randomPoint() for reuse. Move X_LOOPS, Y_LOOPS, and offsets into the
2940         class definition.
2941         (randomPoint): Scale the grid down a little bit so that the lines along the edge of the
2942         canvas are not cut off when the stroke size is thick.
2943         (CanvasQuadraticSegment): Added.
2944         (CanvasBezierSegment): Added.
2945         (SimpleCanvasStage): Pass in an array of the different segment types. Since line segments
2946         are short compared to the curved ones, make it twice as likely to render a line segment.
2947         (SimpleCanvasStage.animate): Update the drawing code so that we render all line segments.
2948
2949         Add a helper method that selects a random element from an array.
2950
2951         * Animometer/tests/resources/main.js:
2952         (Stage.randomElementInArray): Select a random element from the provided array.
2953
2954         * Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js: Refactor.
2955         * Animometer/tests/dom/resources/leaves.js: Ditto.
2956         * Animometer/tests/master/resources/dom-particles.js: Ditto.
2957         * Animometer/tests/master/resources/image-data.js: Ditto.
2958         * Animometer/tests/master/resources/leaves.js: Ditto.
2959         * Animometer/tests/simple/resources/simple-canvas-paths.js: Ditto.
2960
2961         Add canvas tests that includes all stroke and fill paths. This makes it possible to avoid having to
2962         include the full simple canvas suite for perf testing.
2963
2964         * Animometer/resources/debug-runner/tests.js: Add new tests. Move the canvas test into the Canvas
2965         suite.
2966         * Animometer/tests/simple/resources/simple-canvas-paths.js: Add a CanvasStroke and CanvasFill particle
2967         that random selects an object to render.
2968
2969         * Animometer/resources/debug-runner/tests.js: Move 3D suite before basic canvas suite.
2970         When updating the perf bot script, we will include the suites up to this one, but exclude
2971         the basic canvas suite.
2972
2973         Move compositing transforms test to HTML suite and remove the empty Miscellaneous suite.
2974
2975         * Animometer/resources/debug-runner/tests.js:
2976         * Animometer/tests/dom/compositing-transforms.html: Renamed from PerformanceTests/Animometer/tests/misc/compositing-transforms.html.
2977         * Animometer/tests/dom/resources/compositing-transforms.js: Renamed from PerformanceTests/Animometer/tests/misc/resources/compositing-transforms.js.
2978
2979         Clean up miscellaneous test suite. Add a canvas ellipse test, and remove the other
2980         canvas tests.
2981
2982         * Animometer/resources/debug-runner/tests.js: Add ellipse tests to the simple suite.
2983         * Animometer/tests/misc/canvas-electrons.html: Removed.
2984         * Animometer/tests/misc/canvas-stars.html: Removed.
2985         * Animometer/tests/misc/resources/canvas-electrons.js: Removed.
2986         * Animometer/tests/misc/resources/canvas-stars.js: Removed.
2987         * Animometer/tests/simple/resources/simple-canvas-paths.js: Add ellipse primitives.
2988
2989         Merge text tests together into one.
2990
2991         * Animometer/resources/runner/tests.js: Remove international.html.
2992         * Animometer/tests/master/international.html: Removed.
2993         * Animometer/tests/master/resources/text.js:
2994         (animate): Update styling. Manually calculate gradients for the shadow particles.
2995         Reduce the step size for y direction to avoid cutting text off at the margins. Increase
2996         step for x since there will be a little more room.
2997         * Animometer/tests/master/text.html: Add more translations and lay it out in a table.
2998
2999 2016-03-20  Jon Lee  <jonlee@apple.com>
3000
3001         Add a link to show debug data
3002         https://bugs.webkit.org/show_bug.cgi?id=155724
3003
3004         Reviewed by Simon Fraser.
3005
3006         * Animometer/developer.html: Attach the onclick handler to the score.
3007         * Animometer/index.html: Ditto.
3008
3009 2016-03-18  Dean Jackson  <dino@apple.com>
3010
3011         Add a basic WebGL test to Animometer
3012         https://bugs.webkit.org/show_bug.cgi?id=155475
3013         <rdar://problem/25156860>
3014
3015         Reviewed by Simon Fraser and Jon Lee.
3016
3017         Add a pretty simple WebGL test to Animometer
3018         which tries to draw a number of animated triangles.
3019         The result still needs a bit of tuning.
3020
3021         * Animometer/resources/debug-runner/tests.js: Add a new
3022         Suite for "3D" tests.
3023         * Animometer/tests/3d/resources/webgl.js: Added.
3024         * Animometer/tests/3d/webgl.html: Added.
3025
3026 2016-03-18  Jon Lee  <jonlee@apple.com>
3027
3028         Add support for statically linking to a specific test
3029         https://bugs.webkit.org/show_bug.cgi?id=155631
3030
3031         Rubber-stamped by Darin Adler.
3032
3033         * Animometer/developer.html: Update to call restartBenchmark instead so that it works for
3034         both the preset and the manually set test sessions.
3035         * Animometer/resources/debug-runner/animometer.css:
3036         (.tree .link): Style the "link" UI.
3037         * Animometer/resources/debug-runner/animometer.js: Move updateDisplay from suitesManager
3038         to optionsManager since this is a visual option and has nothing to do with the suites
3039         settings. Update the construction of the test UI by including a "link" after each test.
3040         Clicking on that link brings up a JS prompt with a URL and a query string with the current
3041         parameters of the controller, and selected test. Pasting this URL into the location bar will
3042         automatically start running the selected test.
3043         (suitesManager.suitesFromQueryString): Iterate through the Suites and tests and find the one
3044         that matches the provided parameters. Returns an object similar to the form in
3045         suitesManager.updateLocalStorageFromUI.
3046         (benchmarkController.initialize): After settings up the events and options, try parsing the
3047         query string and running the benchmark immediately. Otherwise, fall back to the form.
3048         (benchmarkController.startBenchmark): Store the options and suites into member variables for
3049         reuse in restartBenchmark.
3050         (benchmarkController.startBenchmarkImmediatelyIfEncoded): Convert the query string to an object.
3051         If that's successful, find the suite and test referenced in the query string. Start the benchmark if
3052         the search for the test succeeded.
3053         * Animometer/resources/extensions.js:
3054         (Utilities.stripNonASCIICharacters): Helper method to convert the name of the suite and test
3055         into a query-string-friendly version.
3056         (Utilities.convertObjectToQueryString): Helper method to convert an object to query string
3057         format.
3058         (Utilities.convertQueryStringToObject): Helper method to convert query string into an object
3059         with properties and values.
3060
3061 2016-03-16  Jon Lee  <jonlee@apple.com>
3062
3063         Add a new benchmark test
3064         https://bugs.webkit.org/show_bug.cgi?id=155570
3065
3066         Reviewed by Simon Fraser.
3067
3068         New Leaves test includes various image sizes and opacity.
3069
3070         * Animometer/resources/debug-runner/tests.js: Add it to the HTML test suite.
3071         * Animometer/tests/dom/leaves.html: Added.
3072         * Animometer/tests/dom/resources/leaves.js: Added. Override the
3073         (Particle.call.reset): Uses a range of sizes, and opacity.
3074         (Particle.call.animate): Opacity goes up then down. When it hits 0, reset the particle.
3075         (Particle.call.move): Set transform and opacity.
3076         * Animometer/tests/master/resources/leaves.js: Get rid of the closure so that it
3077         can be used in this test. Update the relative path so that it works in both the master
3078         and dom test suite.
3079
3080 2016-03-15  Simon Fraser  <simon.fraser@apple.com>
3081
3082         Add developer Animometer test that bounces P3-tagged images
3083         https://bugs.webkit.org/show_bug.cgi?id=155511
3084
3085         Reviewed by Tim Horton.
3086         
3087         Add a test for rendering performance of tagged images. The 5 images are tagged
3088         with the Display P3 colorspace.
3089
3090         * Animometer/resources/debug-runner/tests.js:
3091         * Animometer/tests/bouncing-particles/bouncing-tagged-images.html: Added.
3092         * Animometer/tests/bouncing-particles/resources/bouncing-tagged-images.js: Added.
3093         * Animometer/tests/bouncing-particles/resources/image1.jpg: Added.
3094         * Animometer/tests/bouncing-particles/resources/image2.jpg: Added.
3095         * Animometer/tests/bouncing-particles/resources/image3.jpg: Added.
3096         * Animometer/tests/bouncing-particles/resources/image4.jpg: Added.
3097         * Animometer/tests/bouncing-particles/resources/image5.jpg: Added.
3098
3099 2016-03-09  Jon Lee  <jonlee@apple.com>
3100
3101         Enhance existing Animometer tests
3102         https://bugs.webkit.org/show_bug.cgi?id=155261
3103
3104         Reviewed by Simon Fraser.
3105
3106         * Animometer/tests/master/resources/canvas-tests.js: Add a gradient to
3107         the filled circles. To expose more of the gradient, add another circle.
3108         (SimpleCanvasStage.call.animate): For each frame, create a gradient with
3109         undulating stop points and colors. Fill the circles twice; once with
3110         the solid color, and once with the gradient.
3111
3112         * Animometer/tests/master/resources/dom-particles.js: Refactor the
3113         emission variables into a separate stage for this test. Add a colorOffset
3114         variable to make the colors of each particle slightly different, since
3115         the ramp controller can add large numbers of particles all at once, which
3116         would otherwise get all the same color.
3117         * Animometer/tests/master/resources/particles.js:
3118         (initialize): Remove the code specific to the SVG mask test.
3119
3120         * Animometer/tests/master/resources/image-data.js:
3121         (initialize): Remove unused local variable.
3122         * Animometer/tests/master/resources/multiply.js:
3123         (initialize): Make the test harder by adding more total particles.
3124
3125 2016-03-09  Jon Lee  <jonlee@apple.com>
3126
3127         Add text tests
3128         https://bugs.webkit.org/show_bug.cgi?id=155257
3129
3130         Reviewed by Simon Fraser.
3131
3132         * Animometer/resources/extensions.js:
3133         (UnitBezier.Utilities.createClass): Add a class that computes Bezier points
3134         assuming that two of the control points are at (0,0) and (1,1). Taken from
3135         WebCore/platform/graphics/UnitBezier.h
3136         * Animometer/resources/runner/tests.js: Add a test for English text and one
3137         for international text.
3138         * Animometer/tests/master/international.html: Added.
3139         * Animometer/tests/master/resources/text.js: Added. The test assumes there is
3140         #template div which it will copy. The copies are placed behind the template, and
3141         are set with different colors each frame. They are moved around with CSS transform.
3142         * Animometer/tests/master/text.html: Added.
3143
3144         Remove the other text tests, since these ones cover the same techniques.
3145         * Animometer/tests/text/layering-text.html: Removed.
3146         * Animometer/tests/text/resources/layering-text.js: Removed.
3147         * Animometer/tests/text/resources/text-boxes.js: Removed.
3148         * Animometer/tests/text/text-boxes.html: Removed.
3149         * Animometer/resources/debug-runner/tests.js:
3150
3151 2016-03-09  Jon Lee  <jonlee@apple.com>
3152
3153         Add a new image test
3154         https://bugs.webkit.org/show_bug.cgi?id=155232
3155
3156         Reviewed by Dean Jackson.
3157         Provisionally reviewed by Said Abou-Hallawa.
3158
3159         The image test renders PNGs and moves them with translate and rotate
3160         transforms. Each particle has a lifetime, and when the lifetime ends or
3161         the particle goes offscreen, it resets itself somewhere on the stage.
3162
3163         * Animometer/resources/debug-runner/tests.js: Remove the CSS bouncing PNG
3164         images test, because this one tests the same technique.
3165         * Animometer/resources/runner/tests.js: Added here as "Leaves".
3166         * Animometer/tests/master/leaves.html: Added.
3167         * Animometer/tests/master/resources/particles.js: Refactor out the parts
3168         specific to the DOM particles test. Consequently make velocity a public
3169         member.
3170         (initialize): The options parameter is never used, so remove it.
3171         (animate): Remove unused local variable.
3172         * Animometer/tests/master/resources/leaves.js: Maintains a focal point that
3173         moves back and forth across the canvas. That point affects the velocity
3174         of the particles. When the focal point is on the leftmost side, it is
3175         between [-6, -2], and set in reset(). When the focal point is on the rightmost
3176         side, it will be from [2, 6].
3177         * Animometer/tests/master/resources/dom-particles.js: Move JS specific to
3178         this test here from particles.js.
3179
3180         New images.
3181         * Animometer/tests/master/resources/compass100.png: Added.
3182         * Animometer/tests/master/resources/console100.png: Added.
3183         * Animometer/tests/master/resources/contribute100.png: Added.
3184         * Animometer/tests/master/resources/debugger100.png: Added.
3185         * Animometer/tests/master/resources/inspector100.png: Added.
3186         * Animometer/tests/master/resources/layout100.png: Added.
3187         * Animometer/tests/master/resources/performance100.png: Added.
3188         * Animometer/tests/master/resources/script100.png: Added.
3189         * Animometer/tests/master/resources/shortcuts100.png: Added.
3190         * Animometer/tests/master/resources/standards100.png: Added.
3191         * Animometer/tests/master/resources/storage100.png: Added.
3192         * Animometer/tests/master/resources/styles100.png: Added.
3193         * Animometer/tests/master/resources/timeline100.png: Added.
3194
3195 2016-03-03  Jon Lee  <jonlee@apple.com>
3196
3197         Add ability to retrieve raw data from release harness
3198         https://bugs.webkit.org/show_bug.cgi?id=155026
3199
3200         Reviewed by Simon Fraser.
3201
3202         * Animometer/developer.html: Remove the special UI in the debug
3203         harness.
3204         * Animometer/resources/runner/animometer.css: Add styles for the
3205         overlay.
3206         * Animometer/resources/runner/animometer.js: Let 'j' show the JSON
3207         results, but only if the overlay doesn't exist. Add 'esc' key to dismiss
3208         the overlay.
3209         (window.benchmarkController.selectResults): Cycle the cases around
3210         so that the first 's' press selects both the benchmark score and
3211         the individual test scores.
3212         * Animometer/resources/debug-runner/animometer.css: Remove
3213         unneeded rules.
3214         * Animometer/resources/debug-runner/animometer.js: Make the same
3215         call to handleKeyPress.
3216
3217 2016-03-03  Jon Lee  <jonlee@apple.com>
3218
3219         Make sure multiply test particles have at least some opacity
3220         https://bugs.webkit.org/show_bug.cgi?id=155027
3221
3222         Reviewed by Simon Fraser.
3223
3224         * Animometer/tests/master/resources/multiply.js:
3225         (initialize): Have the elements spin a little faster to make it
3226         more obvious when the system is being stressed.
3227         (_addTile):
3228         (animate): Make sure each element being animated has at least 1%
3229         opacity.
3230
3231 2016-03-03  Jon Lee  <jonlee@apple.com>
3232
3233         Update image test
3234         https://bugs.webkit.org/show_bug.cgi?id=154962
3235
3236         Rubber-stamped by Darin Adler.
3237
3238         * Animometer/tests/master/image-data.html: Make each canvas have a compositing layer.
3239         * Animometer/tests/master/resources/image-data.js:
3240         (initialize): Update to load multiple images serially.
3241         (_loadImage):
3242         (tune): Use display instead of visibility. When showing an element anew, move it somewhere else.
3243         (_createTestElement): Set up the element, refactor out placement of element to _refreshElement.
3244         (_refreshElement): Place element in a tile grid.
3245         (animate):
3246         (_getRandomNeighboringPixelIndex): Use a more conservative distribution to make the effect last a little longer.
3247         * Animometer/tests/master/resources/compass.svg: Added.
3248         * Animometer/tests/master/resources/console.svg: Added.
3249         * Animometer/tests/master/resources/contribute.svg: Added.
3250         * Animometer/tests/master/resources/debugger.svg: Added.
3251         * Animometer/tests/master/resources/inspector.svg: Added.
3252         * Animometer/tests/master/resources/layout.svg: Added.
3253         * Animometer/tests/master/resources/performance.svg: Added.
3254         * Animometer/tests/master/resources/script.svg: Added.
3255         * Animometer/tests/master/resources/shortcuts.svg: Added.
3256         * Animometer/tests/master/resources/standards.svg: Added.
3257         * Animometer/tests/master/resources/storage.svg: Added.
3258         * Animometer/tests/master/resources/styles.svg: Added.
3259         * Animometer/tests/master/resources/timeline.svg: Added.
3260
3261 2016-03-02  Jon Lee  <jonlee@apple.com>
3262
3263         Add some new controllers, and refine tests
3264         https://bugs.webkit.org/show_bug.cgi?id=154914
3265
3266         Reviewed by Simon Fraser.
3267
3268         Improve tests.
3269
3270         * Animometer/tests/master/focus.html: Move each particle into a layer. Put the particle
3271         in a container element. Expand the container element by the max blur radius, and clip
3272         overflow. This way, when the blur is applied, it avoids causing layer resizes.
3273         * Animometer/tests/master/resources/focus.js:
3274         (FocusElement.Utilities.createClass): Create a container element, and put the particle
3275         inside.
3276         (hide): Set display:none.
3277         (show): Set display:block.
3278         (animate): Apply filters to the container element.
3279         (FocusStage.call.initialize): Instead of inserting and removing elements from the DOM,
3280         keep them in the stage, but set the display style instead. Use this._offsetIndex to
3281         keep track of which elements are displayed.
3282         (FocusStage.call.animate): Move some calculations around to avoid doing unneeded math.
3283         (FocusStage.call.getBlurValue): Make sure elements are always blurred.
3284         (FocusStage.call.getOpacityValue): Make sure elements have some opacity.
3285         * Animometer/tests/master/resources/multiply.js: Have the particles on the edge of the
3286         stage appear less black when the complexity gets large enough.
3287         (tune): Have this._distanceFactor calculate the factor. Avoid calculating square root each
3288         frame.
3289
3290 2016-03-02  Jon Lee  <jonlee@apple.com>
3291
3292         Add some new controllers, and refine tests
3293         https://bugs.webkit.org/show_bug.cgi?id=154914
3294
3295         Reviewed by Simon Fraser.
3296
3297         Add a controller that centers around 30 fps instead of 60 fps.
3298
3299         * Animometer/developer.html: Add a new option.
3300         * Animometer/resources/debug-runner/animometer.js:
3301         * Animometer/resources/runner/animometer.js:
3302         (this._processData.findRegression): When calculating the complexity-frameLength regression,
3303         check the controller, and use a 30 fps baseline if needed.
3304         * Animometer/resources/statistics.js:
3305         (Regression.Utilities.createClass): Update to allow clients to specify the baseline
3306         frame length.
3307         * Animometer/tests/resources/main.js:
3308         (tune): Override some of the constants in RampController. Move those constants out for
3309         easier reading.
3310
3311         Add a fixed controller, with no step.
3312
3313         * Animometer/developer.html: Add a controller that takes no step.
3314         * Animometer/resources/debug-runner/animometer.js:
3315         * Animometer/tests/resources/main.js:
3316         (Rotater.Utilities.createClass):
3317
3318         Switch to ramp controller as default.
3319
3320         * Animometer/developer.html: Increase the test length to 20 seconds.
3321         * Animometer/resources/debug-runner/animometer.js: Rename the "adjustment" field to "controller"
3322         since that is a more accurate description.
3323         * Animometer/resources/debug-runner/graph.js:
3324         * Animometer/resources/runner/animometer.js: Update preferences for release suite.
3325
3326 2016-02-26  Jon Lee  <jonlee@apple.com>
3327
3328         Address Dean's comments in 154673.
3329
3330         * Animometer/developer.html:
3331         * Animometer/resources/debug-runner/animometer.js:
3332         * Animometer/resources/debug-runner/graph.js:
3333
3334 2016-02-25  Jon Lee  <jonlee@apple.com>
3335
3336         Update animation benchmark and tests
3337         https://bugs.webkit.org/show_bug.cgi?id=154673
3338
3339         Reviewed by Dean Jackson.
3340
3341         Update test visuals.
3342
3343         * Animometer/tests/master/resources/canvas-tests.js:
3344         (CanvasLineSegment.Utilities.createClass): Line segments near the edge of the stage get clipped.
3345         Update circle position and radius to minimize impact.
3346         * Animometer/tests/master/resources/particles.js: Get rid of rotating gradient background, and
3347         have 3 locations for emitting particles.
3348         * Animometer/tests/master/resources/multiply.js: Update distance metric so that fringe tiles get
3349         more color.
3350         (initialize): Fix some of the math for laying out the tiles.
3351         (animate): When a tile isn't used, set visibility: hidden.
3352
3353 2016-02-24  Jon Lee  <jonlee@apple.com>
3354
3355         Update animation benchmark and tests
3356         https://bugs.webkit.org/show_bug.cgi?id=154673
3357
3358         Reviewed by Dean Jackson.
3359
3360         Update the ramp controller.
3361
3362         The controller refines the complexity interval to test across.
3363
3364         * Animometer/resources/statistics.js: Add functions that estimate cumulative distribution function.
3365         (Regression): For the flat regression, force the first segment to be at 60 fps.
3366         (valueAt): Add convenience function to return interpolated value based on the regression used.
3367         (_calculateRegression): Include the number of points included for both segments, and the piecewise
3368         errors.
3369         * Animometer/tests/resources/math.js: Make the Kalman estimator subclass Experiment, and allow it
3370         to be reset.
3371
3372         * Animometer/tests/resources/main.js: Initialize the tier such that it starts at 10^0 = 1.
3373         Increase the number of ramps. Maintain three FPS thresholds-- the frame rate of interest, a limit
3374         on the lowest FPS we care to go for later interpolation, and a minimum FPS threshold we want to
3375         aim for each ramp. Also keep three estimators: a running average of the change point, a minimum
3376         boundary for each ramp, and an estimator for all the frames within an interval. The first two
3377         are used to determine the parameters of the next ramp, and the latter allows us to refine the
3378         parameters.
3379         (update): During the tier phase, it is possible that the highest complexity possible for a test
3380         won't stress the system enough to trigger stopping the tier phase and transitioning to the ramps.
3381         If the complexity doesn't change when going to the next tier, we've maxed the test out, and move
3382         on. When the tier phase completed, turn off Controller.frameLengthEstimator, which estimates the
3383         FPS at each tier.
3384         (tune): At each interval, look at the confidence distribution of being on the 60 FPS side or the
3385         slow side. If the slowest FPS we achieve at the ramp's maximum complexity is not at least
3386         _fpsRampSlowThreshold, then increase the maximum complexity. If we ever achieve 60 FPS, increase
3387         the ramp's minimum complexity to that level. If, at an even lower complexity, a glitch causes the
3388         FPS to drop, we reset the minimum complexity.
3389
3390         Have the bootstrap calculation occur between tests. Clean up harness.
3391
3392         * Animometer/resources/debug-runner/animometer.js: Run bootstrap after a test has
3393         completed to avoid doing all of it at the end before showing the results. Clean up
3394         parameters being passed around.
3395         * Animometer/resources/debug-runner/tests.js:
3396         (text):
3397         * Animometer/resources/runner/animometer.js:
3398         (this._processData.calculateScore): Save the results to the same object holding the data.
3399         (this._processData._processData): In the case where a file is dragged, calculate the score
3400         serially. Grab the results object and move it to the results variable and remove it from
3401         the data object. This avoids serializing the results into the JSON.
3402         (this._processData.findRegression): Include the samples used for bootstrapping. Reduce the
3403         resample size to shorten the wait.
3404         * Animometer/resources/runner/benchmark-runner.js:
3405         * Animometer/resources/statistics.js:
3406         (bootstrap): Update how bootstrapData is sorted. In some regression results the mix of
3407         floats and integers causes an alphabetical sort to occur.
3408         * Animometer/resources/strings.js:
3409
3410         Add meta charset so that encodings between harness and test match.
3411
3412         * Animometer/tests/bouncing-particles/bouncing-canvas-images.html:
3413         * Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html:
3414         * Animometer/tests/bouncing-particles/bouncing-css-images.html:
3415         * Animometer/tests/bouncing-particles/bouncing-css-shapes.html:
3416         * Animometer/tests/bouncing-particles/bouncing-svg-images.html:
3417         * Animometer/tests/bouncing-particles/bouncing-svg-shapes.html:
3418         * Animometer/tests/master/canvas-stage.html:
3419         * Animometer/tests/master/focus.html:
3420         * Animometer/tests/master/image-data.html:
3421         * Animometer/tests/master/multiply.html:
3422         * Animometer/tests/master/particles.html:
3423         * Animometer/tests/misc/canvas-electrons.html:
3424         * Animometer/tests/misc/canvas-stars.html:
3425         * Animometer/tests/misc/compositing-transforms.html:
3426         * Animometer/tests/simple/simple-canvas-paths.html:
3427         * Animometer/tests/simple/tiled-canvas-image.html:
3428         * Animometer/tests/template/template-canvas.html:
3429         * Animometer/tests/template/template-css.html:
3430         * Animometer/tests/template/template-svg.html:
3431         * Animometer/tests/text/layering-text.html:
3432         * Animometer/tests/text/text-boxes.html:
3433
3434         Update test harness reporting.
3435
3436         * Animometer/developer.html: Add missing meta charset.
3437         * Animometer/index.html: Remove unnecessary utf-8 declaration.
3438         * Animometer/resources/debug-runner/animometer.css: Add convenience classes for
3439         formatting the results table.
3440         * Animometer/resources/debug-runner/animometer.js: Adjust which stats are shown.
3441         * Animometer/resources/debug-runner/tests.js: Display bootstrapping statistics.
3442         * Animometer/resources/strings.js: Move strings not used by the release harness.
3443
3444         Switch to a pseudo-random number generator.
3445
3446         * Animometer/resources/statistics.js: Add a Pseudo class, with a simple
3447         pseudo-random number generator.
3448         (_calculateRegression): Reset the generator before running bootstrap.
3449         (bootstrap): Deleted.
3450
3451         Replace Math.random with Pseudo.random.
3452         * Animometer/tests/master/resources/canvas-tests.js:
3453         * Animometer/tests/master/resources/focus.js:
3454         * Animometer/tests/master/resources/particles.js:
3455         * Animometer/tests/resources/main.js:
3456
3457         Use bootstrapping to get confidence interval in the breakpoint.
3458
3459         For the ramp controller, calculate the piecewise regression, and then use
3460         bootstrapping in order to find the 95% confidence interval. Use the raw data.
3461
3462         * Animometer/developer.html: Default to the complexity graph. Add a legend
3463         checkbox to toggle visibility of the bootstrap score and histogram.
3464         * Animometer/resources/debug-runner/animometer.css: Make some more space to show
3465         the old raw and average scores in the legend. Add new styles for the data.
3466         * Animometer/resources/debug-runner/graph.js:
3467         (_addRegressionLine): Allow passing an array for the variance bar tied to the
3468         regression line. Now |stdev| is |range|.
3469         (createComplexityGraph): Add bootstrap median, and overlay a histogram of
3470         the bootstrap samples. Switch raw samples from circles to X's.
3471         (onComplexityGraphOptionsChanged): Allow toggling of the bootstrap data.
3472         (onGraphTypeChanged): Move the regressions for the raw and average samples to the
3473         legend. In the subtitle use the bootstrap median, and include the 95% confidence
3474         interval.
3475         * Animometer/resources/runner/animometer.js:
3476         (this._processData.findRegression): Factor out the code that determines which
3477         samples to include when calculating the piecewise regression. For series that have
3478         many samples, or a wider range of recorded complexities, throw away the 2.5%
3479         lowest and highest samples before calculating the regression. Keep all samples
3480         if the number of samples to regress is small or the range of complexities is
3481         narrow.
3482         (this._processData._calculateScore): Factor out regression calculation to
3483         findRegression(). Bootstrap the change point of the regression. The score is the
3484         median.
3485         * Animometer/resources/statistics.js:
3486         (_calculateRegression): Correct an issue in the calculation of the regression, where
3487         the denominator can be 0.
3488         (bootstrap): Template for bootstrapping. Create a bootstrap sample array, Create
3489         re-samples by random selection with replacement. Return the 95% confidence samples,
3490         the bootstrap median, mean, and the data itself.
3491         * Animometer/resources/strings.js: Add bootstrap.
3492         * Animometer/tests/resources/main.js:
3493         (processSamples): Don't prematurely cut the sample data.
3494
3495         Fix graph drawing.
3496
3497         * Animometer/resources/debug-runner/animometer.js: Add spacing in the JSON output.
3498         Multiple tests output a lot of JSON and can hang when selecting JSON with no whitespace.
3499         * Animometer/resources/debug-runner/animometer.css:
3500         (#complexity-graph .series.raw circle): Update the color.
3501         * Animometer/resources/debug-runner/graph.js: Use the FPS axis instead of the
3502         complexity axis, which can vary in domain. For determining the complexity domain,
3503         only use samples after samplingTimeOffset.
3504
3505         Allow dropping results JSON.
3506
3507         * Animometer/developer.html: Add a button.
3508         * Animometer/resources/debug-runner/animometer.css:
3509         * Animometer/resources/debug-runner/animometer.js: Read the data and go straight
3510         to the dashboard. With JSON output, write out only the options and the raw data.
3511
3512         Teach the harness to evaluate the samples and determine the test score.
3513
3514         This will allow us to update how the score is calculated separately from the samples recorded.
3515         This also prepares the harness to be able to accept JSON of prior runs.
3516
3517         * Animometer/resources/strings.js: Clean up and remove unneeded strings and reduce some of the
3518         hierarchy.
3519         * Animometer/resources/debug-runner/tests.js: Update to use the new strings.
3520
3521         * Animometer/tests/resources/main.js: Allow all controllers to show a complexity-FPS graph.
3522         (_processComplexitySamples): Factor out some of the sample processing done in the ramp
3523         controller for the benefit of the other controllers. |complexitySamples| contains a list of
3524         samples. Sort the samples by complexity. Optionally remove the top x% of samples.
3525         Group them, and calculate distribution of samples within the same complexity, and add those as
3526         new entries into |complexityAverageSamples|.
3527         (Controller.processSamples): Move the code responsible for determining the complexity and FPS
3528         scores out to ResultsDashboard. The structure of the data returned by the controller is:
3529
3530         {
3531             controller: [time-regression, time-regression, ...], // optional, data specific to controller
3532             marks: [...],
3533             samples: {                    // all of the sample data
3534                 controller: [...],
3535                 complexity: [...],        // processed from controller samples
3536                 complexityAverage: [...], // processed from complexity samples
3537             }
3538         }
3539
3540         (AdaptiveController.processSamples): Adding the target frame length is no longer necessary; we
3541         now pass the test options to the graph.
3542         (Regression): Move to statistics.js.
3543         * Animometer/resources/statistics.js: Move Regression to here. Add a check if the sampling range
3544         only contains one sample, since we cannot calculate a regression from one sample point.
3545
3546         Teach the test harness to evaluate the data.
3547         * Animometer/resources/runner/animometer.js:
3548         (ResultsDashboard): Store the options used to run the test and the computed results/score separately
3549         from the data. The results are stored as:
3550
3551         {
3552             score: /* geomean of iteration score */,
3553             iterationsResults: [
3554                 {
3555                     score: /* geomean of tests */,
3556                     testsResults: {
3557                         suiteName: {
3558                             testName: {
3559                                 controller: {
3560                                     average:
3561                                     concern:
3562                                     stdev:
3563                                     percent:
3564                                 },
3565                                 frameLength: { ... },
3566                                 complexity: {
3567                                     complexity:
3568                                     stdev:
3569                                     segment1:
3570                                     segment2:
3571                                 },
3572                                 complexityAverage: { ... }
3573                             },
3574                             testName: { ... },
3575                         },
3576                         ... next suite ...
3577                     }
3578                 },
3579                 { ...next iteration... }
3580             ]
3581         }
3582
3583         * Animometer/resources/debug-runner/animometer.js: Pass options around instead of relying
3584         on what was selected in the form. This will later allow for dropping previous results, and
3585         using those runs' options when calculating scores.
3586         (ResultsTable._addGraphButton): Simplify button action by using attached test data.
3587         * Animometer/resources/debug-runner/graph.js: Refactor to use the data.
3588
3589         Consolidate JS files, and move statistics out to a separate JS.
3590
3591         Preparation for having the Controller only handle recording and storage of the samples,
3592         and leave the evaluation of the test score out to the harness. Move Experiment to
3593         a new statistics.js, where Regression will also eventually go. Get rid of algorithm.js
3594         and move it to utilities.js since the Heap is used only for Experiments.
3595
3596         * Animometer/tests/resources/algorithm.js: Removed. Heap is in utilities.js.
3597         * Animometer/tests/resources/sampler.js: Removed. Experiment is in statistics.js,
3598         Sampler in main.js.
3599         * Animometer/tests/resources/main.js: Move Sampler here.
3600         * Animometer/resources/statistics.js: Added. Move Statistics and Experiment here.
3601         * Animometer/resources/extensions.js: Move Heap here. Attach static method to create
3602         a max or min heap to Heap, instead of a new Algorithm object.
3603
3604         Update JS files.
3605         * Animometer/developer.html:
3606         * Animometer/index.html:
3607         * Animometer/tests/bouncing-particles/bouncing-canvas-images.html:
3608         * Animometer/tests/bouncing-particles/bouncing-canvas-shapes.html:
3609         * Animometer/tests/bouncing-particles/bouncing-css-images.html:
3610         * Animometer/tests/bouncing-particles/bouncing-css-shapes.html:
3611         * Animometer/tests/bouncing-particles/bouncing-svg-images.html:
3612         * Animometer/tests/bouncing-particles/bouncing-svg-shapes.html:
3613         * Animometer/tests/master/canvas-stage.html:
3614         * Animometer/tests/master/focus.html:
3615         * Animometer/tests/master/image-data.html:
3616         * Animometer/tests/master/multiply.html:
3617         * Animometer/tests/master/particles.html:
3618         * Animometer/tests/misc/canvas-electrons.html:
3619         * Animometer/tests/misc/canvas-stars.html:
3620         * Animometer/tests/misc/compositing-transforms.html:
3621         * Animometer/tests/simple/simple-canvas-paths.html:
3622         * Animometer/tests/simple/tiled-canvas-image.html:
3623         * Animometer/tests/template/template-canvas.html:
3624         * Animometer/tests/template/template-css.html:
3625         * Animometer/tests/template/template-svg.html:
3626         * Animometer/tests/text/layering-text.html:
3627         * Animometer/tests/text/text-boxes.html:
3628
3629         Fix the cursor in the graph analysis when the min
3630         complexity is not 0.
3631
3632         * Animometer/resources/debug-runner/graph.js:
3633         (_addRegression):
3634         (createComplexityGraph):
3635
3636 2016-02-23  Geoffrey Garen  <ggaren@apple.com>
3637
3638         Fix some issues in MallocBench
3639         https://bugs.webkit.org/show_bug.cgi?id=154600
3640
3641         Reviewed by Oliver Hunt.
3642
3643         * MallocBench/MallocBench.xcodeproj/project.pbxproj: Make nimlang.ops
3644         a part of the build so that it copies to the right place for execution.
3645
3646         * MallocBench/MallocBench/Interpreter.cpp:
3647         (Interpreter::Interpreter): Don't try to open .ops files for writing
3648         because we're only going to read and we might need extra permissions to
3649         write.
3650
3651         (Interpreter::~Interpreter): Give more context when opening a file fails
3652         to help with debugging.
3653
3654         * MallocBench/MallocBench/stress.cpp:
3655         (benchmark_stress): Reduce iterations to complete in less than 10 seconds.
3656
3657 2016-02-23  Michael Saboff  <msaboff@apple.com>
3658
3659         Unreviewed change to revert extraneous changes made part of change set 196955.
3660
3661         * MallocBench/MallocBench/Interpreter.cpp:
3662         (Interpreter::doMallocOp):
3663         (Interpreter::Thread::switchTo):
3664         (writeData): Deleted.
3665
3666 2016-02-19  Michael Saboff  <msaboff@apple.com>
3667
3668         MallocBench: Added recording for nimlang website, new recording details and added new options
3669         https://bugs.webkit.org/show_bug.cgi?id=154485
3670
3671         Reviewed by Geoff Garen.
3672
3673         Added new capabilities to MallocBench.  These include:
3674             Added a recording of http://nim-lang.org/docs/lib.html.
3675             Added thread id to the recording and the ability to playback switching threads in MallocBench
3676             Added aligned allocations to recordings and the ability to playback
3677             Added --use-thread-id option to honor recorded thread ids
3678             Added --detailed-report to output remaining allocations by size after playback
3679             Added --no-warmup to not run the warm up iteration
3680
3681         Changed the way that options are passed down to the benchmarks.  Instead of passing individual
3682         boolean or numeric option values, just pass a reference the CommandLine itself.  Each benchmark
3683         can access the options that are appropriate.  The Benchmark class also uses the options for
3684         is parallel, run counts and warm up.
3685
3686         Added thread id and aligned malloc to the Op by noticing that structure padding and Opcode allowed
3687         for another 32 bits of data.  Breaking that unused 32 bits into a 16 bit thread id value and a
3688         16 bit log base 2 of the alignment for aligned malloc allowed for existing recordings to playback
3689         without any incompatibilities.
3690
3691         Threaded operation is simulated by creating threads as needed.  As long as the next Op's thread id
3692         is the same as the last, operation continues as normal.  When the next Op has a different thread id,
3693         we switch to that thread using the shared Op stream to continue playing back.  There is a mutex to
3694         assure that only one thread is really running at a time and a condition variable used to wait
3695         that the current thread id matches each block thread's thread id.  This doesn't simulate true
3696         concurrent threading, but is instead plays back Ops recorded for multiple thread faithfully.
3697
3698         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
3699         * MallocBench/MallocBench/Benchmark.cpp:
3700         (deallocateHeap):
3701         (Benchmark::Benchmark):
3702         (Benchmark::runOnce):
3703         (Benchmark::run):
3704         * MallocBench/MallocBench/Benchmark.h:
3705         (Benchmark::isValid):
3706         * MallocBench/MallocBench/CommandLine.cpp:
3707         (CommandLine::printUsage):
3708         * MallocBench/MallocBench/CommandLine.h:
3709         (CommandLine::isValid):
3710         (CommandLine::benchmarkName):
3711         (CommandLine::isParallel):
3712         (CommandLine::useThreadID):
3713         (CommandLine::detailedReport):
3714         (CommandLine::warmUp):
3715         (CommandLine::heapSize):
3716         (CommandLine::runs):
3717         * MallocBench/MallocBench/Interpreter.cpp:
3718         (Interpreter::Interpreter):
3719         (Interpreter::run):
3720         (Interpreter::readOps):
3721         (Interpreter::doOnSameThread):
3722         (Interpreter::switchToThread):
3723         (Interpreter::detailedReport):
3724         (compute2toPower):
3725         (writeData):
3726         (Interpreter::doMallocOp):
3727         (Interpreter::Thread::Thread):
3728         (Interpreter::Thread::stop):
3729         (Interpreter::Thread::~Thread):
3730         (Interpreter::Thread::runThread):
3731         (Interpreter::Thread::waitToRun):
3732         (Interpreter::Thread::switchTo):
3733         * MallocBench/MallocBench/Interpreter.h:
3734         (Interpreter::Thread::isMainThread):
3735         * MallocBench/MallocBench/alloc_free.cpp: Added.
3736         (benchmark_alloc_free):
3737         * MallocBench/MallocBench/alloc_free.h: Added.
3738         * MallocBench/MallocBench/balloon.cpp:
3739         (benchmark_balloon):
3740         * MallocBench/MallocBench/balloon.h:
3741         * MallocBench/MallocBench/big.cpp:
3742         (benchmark_big):
3743         * MallocBench/MallocBench/big.h:
3744         * MallocBench/MallocBench/churn.cpp:
3745         (benchmark_churn):
3746         * MallocBench/MallocBench/churn.h:
3747         * MallocBench/MallocBench/facebook.cpp:
3748         (benchmark_facebook):
3749         * MallocBench/MallocBench/facebook.h:
3750         * MallocBench/MallocBench/flickr.cpp:
3751         (benchmark_flickr):
3752         (benchmark_flickr_memory_warning):
3753         * MallocBench/MallocBench/flickr.h:
3754         * MallocBench/MallocBench/fragment.cpp:
3755         (validate):
3756         (benchmark_fragment):
3757         (benchmark_fragment_iterate):
3758         * MallocBench/MallocBench/fragment.h:
3759         * MallocBench/MallocBench/list.cpp:
3760         (benchmark_list_allocate):
3761         (benchmark_list_traverse):
3762         * MallocBench/MallocBench/list.h:
3763         * MallocBench/MallocBench/main.cpp:
3764         (main):
3765         * MallocBench/MallocBench/medium.cpp:
3766         (benchmark_medium):
3767         * MallocBench/MallocBench/medium.h:
3768         * MallocBench/MallocBench/memalign.cpp:
3769         (test):
3770         (benchmark_memalign):
3771         * MallocBench/MallocBench/memalign.h:
3772         * MallocBench/MallocBench/message.cpp:
3773         (benchmark_message_one):
3774         (benchmark_message_many):
3775         * MallocBench/MallocBench/message.h:
3776         * MallocBench/MallocBench/nimlang.cpp: Added.
3777         (benchmark_nimlang):
3778         * MallocBench/MallocBench/nimlang.h: Added.
3779         * MallocBench/MallocBench/nimlang.ops: Added.
3780         * MallocBench/MallocBench/realloc.cpp:
3781         (benchmark_realloc):
3782         * MallocBench/MallocBench/realloc.h:
3783         * MallocBench/MallocBench/reddit.cpp:
3784         (benchmark_reddit):
3785         (benchmark_reddit_memory_warning):
3786         * MallocBench/MallocBench/reddit.h:
3787         * MallocBench/MallocBench/stress.cpp:
3788         (deallocate):
3789         (benchmark_stress):
3790         * MallocBench/MallocBench/stress.h:
3791         * MallocBench/MallocBench/stress_aligned.cpp:
3792         (benchmark_stress_aligned):
3793         * MallocBench/MallocBench/stress_aligned.h:
3794         * MallocBench/MallocBench/theverge.cpp:
3795         (benchmark_theverge):
3796         (benchmark_theverge_memory_warning):
3797         * MallocBench/MallocBench/theverge.h:
3798         * MallocBench/MallocBench/tree.cpp:
3799         (benchmark_tree_allocate):
3800         (benchmark_tree_traverse):
3801         (benchmark_tree_churn):
3802         * MallocBench/MallocBench/tree.h:
3803         * MallocBench/run-malloc-benchmarks:
3804
3805 2016-02-11  Jon Lee  <jonlee@apple.com>
3806
3807         Fix a missing refactoring.
3808
3809         * Animometer/tests/master/resources/multiply.js:
3810         (animate): Move to Utilities.lerp.
3811         (_lerp): Deleted.
3812
3813 2016-02-10  Jon Lee  <jonlee@apple.com>
3814
3815         Add new benchmark tests.
3816         https://bugs.webkit.org/show_bug.cgi?id=154063
3817
3818         Provisionally reviewed by Said Abou-Hallawa.
3819
3820         Add tests for get/put image data, filters, opacity, and css transforms.
3821
3822         * Animometer/resources/runner/benchmark-runner.js:
3823         (_runBenchmarkAndRecordResults): Update the body background color to match that of
3824         the stage.
3825         (this._runNextIteration): Clear the background color style for the results page.
3826         * Animometer/resources/runner/tests.js:
3827         * Animometer/tests/master/focus.html: Added.
3828         * Animometer/tests/master/image-data.html: Added.
3829         * Animometer/tests/master/multiply.html: Added.
3830         * Animometer/tests/master/resources/focus.js: Added.
3831         * Animometer/tests/master/resources/image-data.js: Added.
3832         * Animometer/tests/master/resources/multiply.js: Added.
3833         * Animometer/tests/master/resources/stage.css: Move common styles out.
3834         * Animometer/tests/resources/main.js: Update Stage.randomBool to use Math.random.
3835         Add Stage.randomSign for randomly setting a direction. Add the notion of the
3836         current timestamp of the test to Benchmark, since some animations cycle through
3837         colors and rely on an incremental counter like the time.
3838
3839 2016-02-09  Said Abou-Hallawa  <sabouhallawa@apple.com>
3840
3841         Add internal benchmark tests for CSS mix-blend-modes and filters
3842         https://bugs.webkit.org/show_bug.cgi?id=154058
3843
3844         Provisionally reviewed by Jon Lee.
3845
3846         * Animometer/resources/debug-runner/tests.js: Include the new tests in the
3847         "HTML suite" of the debug runner.
3848
3849         * Animometer/resources/extensions.js:
3850         (Utilities.browserPrefix):
3851         (Utilities.setElementPrefixedProperty): Utility functions to allow setting
3852         prefixed style properties.
3853
3854         * Animometer/tests/bouncing-particles/resources/bouncing-css-shapes.js:
3855         Set the mix-blend-mode and the filter to some random values if the options
3856         of the test requested that.
3857
3858         * Animometer/tests/bouncing-particles/resources/bouncing-particles.js:
3859         (parseShapeParameters): Parse the url options "blend" and "filter" and set
3860         the corresponding flags.
3861
3862         * Animometer/tests/resources/main.js:
3863         (randomStyleMixBlendMode):
3864         (randomStyleFilter): Return random mix-blend-mode and filter.
3865
3866 2016-02-08  Jon Lee  <jonlee@apple.com>
3867
3868         Add a ramp controller
3869         https://bugs.webkit.org/show_bug.cgi?id=154028
3870
3871         Provisionally reviewed by Said Abou-Hallawa.
3872
3873         Enhance the graph to include a complexity-fps graph, in addition
3874         to the time graph.
3875
3876         * Animometer/developer.html: Add a ramp option.
3877         * Animometer/resources/debug-runner/animometer.css: Update the style.
3878         * Animometer/resources/strings.js: Flatten the Strings.text constants.
3879         * Animometer/resources/debug-runner/animometer.js:
3880         (ResultsTable.call._addGraphButton): Refactor.
3881         (ResultsTable.call._addTest): Add regression data.
3882         (benchmarkController): Add a form that allows the user to switch between the two forms,
3883         Add a form that allows the user to toggle different data. Hide certain header columns
3884         depending on the selected controller.
3885         * Animometer/resources/debug-runner/graph.js: Add the complexity regressions.
3886         * Animometer/resources/debug-runner/tests.js: Add headers for the ramp results.
3887         * Animometer/resources/runner/animometer.js:
3888         (ResultsTable): If a header is disabled don't include them in _flattenedHeaders.
3889         * Animometer/tests/resources/main.js:
3890         (Controller): Allow options to specify the capacity for sample arrays.
3891         (Regression): A piecewise regression that tries to fit a slope and a flat profile.
3892         (_calculateRegression): Options can fix the slope and bias when calculating the minimal
3893         error. Sweep across the samples in time (which could be backward depending on the controller)
3894         and calculate the intersection point.
3895         (RampController): This controller assumes that the target frame rate is below
3896         58 FPS. It runs in two stages. The first stage quickly determines the order of
3897         magnitude of objects needed to stress the system by the setting the complexity
3898         to increasingly difficult tiers. Perform a series of ramps descending from a
3899         high-water mark of complexity. The complexity needed to reach the target frame
3900         length is done by performing a piecewise regression on each ramp, and track a
3901         running average of these values. For the next ramp, make that running average
3902         the center of the ramp. With a minimum complexity of 0, the high-water mark is
3903         twice that average. The score is based on the highest complexity that can
3904         reach 60&nb