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