Have a CLI version of JetStream 2 to run on watchOS
[WebKit-https.git] / PerformanceTests / ChangeLog
1 2019-10-17  Saam Barati  <sbarati@apple.com>
2
3         Have a CLI version of JetStream 2 to run on watchOS
4         https://bugs.webkit.org/show_bug.cgi?id=202998
5         <rdar://problem/56208554>
6
7         Reviewed by Tadeu Zagallo.
8
9         This patch adds a CLI version of JS2 to run on watchOS. We run most subtests
10         from the browser version of JS2, but skip a few tests that are particularly
11         long running, like WSL and some of WTB. We also don't run the Wasm tests
12         as we don't JIT on watchOS. Each test runs for fewer iterations in the watch
13         version. 15 is the default iteration count. This benchmark runs in 7 minutes on
14         a Series 4 watch.
15
16         * JetStream2/JetStreamDriver.js:
17         (getIterationCount):
18         (getWorstCaseCount):
19         (Driver.prototype.async.start):
20         (Driver.prototype.resultsJSON):
21         (Driver.prototype.dumpJSONResultsIfNeeded):
22         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
23         (DefaultBenchmark):
24         * JetStream2/RexBench/UniPoker/benchmark.js:
25         (Benchmark.prototype.validate):
26         (Benchmark):
27         * JetStream2/watch-cli.js: Added.
28
29 2019-10-17  Adrian Perez de Castro  <aperez@igalia.com>
30
31         [Linux] Avoid usage of <sys/sysctl.h> in MallocBench
32         https://bugs.webkit.org/show_bug.cgi?id=203109
33
34         Reviewed by Carlos Alberto Lopez Perez.
35
36         * MallocBench/MallocBench/CPUCount.cpp: Only include <sys/sysctl.h> if
37         the sysctl() function will be used.
38
39 2019-10-17  Carlos Alberto Lopez Perez  <clopez@igalia.com>
40
41         [GTK] Perf test IndexedDB/basic/index-cursor-delete.html timeouts
42         https://bugs.webkit.org/show_bug.cgi?id=203083
43
44         Unreviewed gardening.
45
46         * Skipped: Skip a test that is timing out on the GTK perf bot.
47
48 2019-09-24  Caio Lima  <ticaiolima@gmail.com>
49
50         [BigInt] Add ValueBitRShift into DFG
51         https://bugs.webkit.org/show_bug.cgi?id=192663
52
53         Reviewed by Robin Morisset.
54
55         * BigIntBench/big-int-simple-rshift.js: Added.
56         (bigInt):
57
58 2019-09-09  Jon Lee  <jonlee@apple.com>
59
60         Upload triangles tests for 3D suite in MotionMark
61         https://bugs.webkit.org/show_bug.cgi?id=201606
62
63         Reviewed by Simon Fraser.
64
65         * MotionMark/resources/debug-runner/motionmark.css:
66         * MotionMark/resources/debug-runner/tests.js:
67         * MotionMark/tests/3d/resources/webgl.js:
68         * MotionMark/tests/3d/resources/webgpu.js: Added.
69         * MotionMark/tests/3d/triangles-webgl.html: Renamed from Websites/browserbench.org/MotionMark1.1/tests/3d/webgl.html.
70         * MotionMark/tests/3d/triangles-webgpu.html: Copied from PerformanceTests/MotionMark/tests/3d/webgl.html.
71
72 2019-08-29  Keith Rollin  <krollin@apple.com>
73
74         Update .xcconfig symbols to reflect the current set of past and future product versions.
75         https://bugs.webkit.org/show_bug.cgi?id=200720
76         <rdar://problem/54305032>
77
78         Reviewed by Alex Christensen.
79
80         Remove version symbols related to old OS's we no longer support,
81         ensure that version symbols are defined for OS's we do support.
82
83         * MediaTime/Configurations/Base.xcconfig:
84         * MediaTime/Configurations/DebugRelease.xcconfig:
85
86 2019-08-21  Sihui Liu  <sihui_liu@apple.com>
87
88         Add some IDB performance tests
89         https://bugs.webkit.org/show_bug.cgi?id=200906
90
91         Reviewed by Geoffrey Garen.
92
93         Add test coverage for IDB performance, and organize IDB tests into two categories.
94
95         * IndexedDB/basic/index-count-key.html: Added.
96         * IndexedDB/basic/index-count.html: Added.
97         * IndexedDB/basic/index-cursor-advance.html: Added.
98         * IndexedDB/basic/index-cursor-continue.html: Added.
99         * IndexedDB/basic/index-cursor-continuePrimaryKey.html: Added.
100         * IndexedDB/basic/index-cursor-delete.html: Added.
101         * IndexedDB/basic/index-cursor-update.html: Added.
102         * IndexedDB/basic/index-get.html: Added.
103         * IndexedDB/basic/index-multientry.html: Renamed from PerformanceTests/IndexedDB/index-multientry.html.
104         * IndexedDB/basic/objectstore-add.html: Added.
105         * IndexedDB/basic/objectstore-clear.html: Added.
106         * IndexedDB/basic/objectstore-count.html: Added.
107         * IndexedDB/basic/objectstore-cursor-advance.html: Added.
108         * IndexedDB/basic/objectstore-cursor-continue.html: Added.
109         * IndexedDB/basic/objectstore-cursor-delete.html: Added.
110         * IndexedDB/basic/objectstore-cursor-update.html: Added.
111         * IndexedDB/basic/objectstore-cursor.html: Renamed from PerformanceTests/IndexedDB/objectstore-cursor.html.
112         * IndexedDB/basic/objectstore-delete.html: Renamed from PerformanceTests/IndexedDB/index-get.html.
113         * IndexedDB/basic/objectstore-get.html: Added.
114         * IndexedDB/basic/objectstore-put.html: Renamed from PerformanceTests/IndexedDB/objectstore-get.html.
115         * IndexedDB/stress/large-array-keys.html: Renamed from PerformanceTests/IndexedDB/large-array-keys.html.
116         * IndexedDB/stress/large-binary-keys.html: Renamed from PerformanceTests/IndexedDB/large-binary-keys.html.
117         * IndexedDB/stress/large-number-of-inserts-responsiveness.html: Renamed from PerformanceTests/IndexedDB/large-number-of-inserts-responsiveness.html.
118         * IndexedDB/stress/large-number-of-inserts.html: Renamed from PerformanceTests/IndexedDB/large-number-of-inserts.html.
119         * IndexedDB/stress/large-string-keys.html: Renamed from PerformanceTests/IndexedDB/large-string-keys.html.
120         * Skipped:
121
122 2019-08-19  Sihui Liu  <sihui_liu@apple.com>
123
124         Remove IDB-specific quota
125         https://bugs.webkit.org/show_bug.cgi?id=196545
126         <rdar://problem/54201783>
127
128         Reviewed by Youenn Fablet.
129
130         * IndexedDB/large-binary-keys.html:
131         * IndexedDB/large-string-keys.html:
132
133 2019-08-14  Keith Rollin  <krollin@apple.com>
134
135         Remove support for macOS < 10.13
136         https://bugs.webkit.org/show_bug.cgi?id=200694
137         <rdar://problem/54278851>
138
139         Reviewed by Youenn Fablet.
140
141         Update conditionals that reference __MAC_OS_X_VERSION_MIN_REQUIRED and
142         __MAC_OS_X_VERSION_MAX_ALLOWED, assuming that they both have values >=
143         101300. This means that expressions like
144         "__MAC_OS_X_VERSION_MIN_REQUIRED < 101300" are always False and
145         "__MAC_OS_X_VERSION_MIN_REQUIRED >= 101300" are always True.
146
147         * StitchMarker/wtf/Platform.h:
148         * StitchMarker/wtf/dependencies/bmalloc/BPlatform.h:
149         * StitchMarker/wtf/text/icu/TextBreakIteratorICU.h:
150         (WTF::TextBreakIteratorICU::TextBreakIteratorICU):
151         (WTF::caretRules): Deleted.
152
153 2019-08-06  Michael Saboff  <msaboff@apple.com>
154
155         Make JSC memory benchmark available on OpenSource
156         https://bugs.webkit.org/show_bug.cgi?id=195012
157
158         Reviewed by Saam Barati.
159
160         This patch upstreams the RAMification memory benchmark to open source.
161
162         The prior RAMification.py driver script in the Internal directory has been moved here
163         with the device execution code seperated out to a new Internal driver script.
164         As a result, command line argument parsing as well as acting on the resulting argument
165         values has changed.  The remote device Runner creates an ArgumentParser object that is
166         passed through main().  In main(), if we have an ArgumentParser, we use it otherwise we
167         create one for benchmarking locally.  The arguments object returned from argument
168         parsing is now passed to the Runner constructor so it can extract the settings appropriate
169         for that particular runner.
170
171         Added some sane argument default values for the JSC command patch and the JetStream2
172         source directory.
173
174         Updated the Lua tests to eliminate the testing harness and its overhead, matching what was
175         done in the Internal tree.
176
177         * JetStream2/LuaJSFight/hello_world.js:
178         (Benchmark.prototype.runIteration): Deleted.
179         (Benchmark): Deleted.
180         * JetStream2/LuaJSFight/list_search.js:
181         (run): Deleted.
182         (Benchmark.prototype.runIteration): Deleted.
183         (Benchmark): Deleted.
184         * JetStream2/LuaJSFight/lists.js:
185         (): Deleted.
186         (Benchmark.prototype.runIteration): Deleted.
187         (Benchmark): Deleted.
188         * JetStream2/LuaJSFight/richards.js: Added.
189         (runRichards):
190         (Scheduler):
191         (Scheduler.prototype.addIdleTask):
192         (Scheduler.prototype.addWorkerTask):
193         (Scheduler.prototype.addHandlerTask):
194         (Scheduler.prototype.addDeviceTask):
195         (Scheduler.prototype.addRunningTask):
196         (Scheduler.prototype.addTask):
197         (Scheduler.prototype.schedule):
198         (Scheduler.prototype.release):
199         (Scheduler.prototype.holdCurrent):
200         (Scheduler.prototype.suspendCurrent):
201         (Scheduler.prototype.queue):
202         (TaskControlBlock):
203         (TaskControlBlock.prototype.setRunning):
204         (TaskControlBlock.prototype.markAsNotHeld):
205         (TaskControlBlock.prototype.markAsHeld):
206         (TaskControlBlock.prototype.isHeldOrSuspended):
207         (TaskControlBlock.prototype.markAsSuspended):
208         (TaskControlBlock.prototype.markAsRunnable):
209         (TaskControlBlock.prototype.run):
210         (TaskControlBlock.prototype.checkPriorityAdd):
211         (TaskControlBlock.prototype.toString):
212         (IdleTask):
213         (IdleTask.prototype.run):
214         (IdleTask.prototype.toString):
215         (DeviceTask):
216         (DeviceTask.prototype.run):
217         (DeviceTask.prototype.toString):
218         (WorkerTask):
219         (WorkerTask.prototype.run):
220         (WorkerTask.prototype.toString):
221         (HandlerTask):
222         (HandlerTask.prototype.run):
223         (HandlerTask.prototype.toString):
224         (Packet):
225         (Packet.prototype.addTo):
226         (Packet.prototype.toString):
227         * JetStream2/LuaJSFight/string_lists.js:
228         (): Deleted.
229         (Benchmark.prototype.runIteration): Deleted.
230         (Benchmark): Deleted.
231         * JetStream2/RAMification.py: Added.
232         (mean):
233         (geomean):
234         (frameworkPathFromExecutablePath):
235         (parseArgs):
236         (parseArgs.optStrToBool):
237         (BaseRunner):
238         (BaseRunner.__init__):
239         (BaseRunner.setup):
240         (BaseRunner.setEnv):
241         (BaseRunner.unsetEnv):
242         (BaseRunner.resetForTest):
243         (BaseRunner.processLine):
244         (BaseRunner.setReturnCode):
245         (BaseRunner.getResults):
246         (LocalRunner):
247         (LocalRunner.__init__):
248         (LocalRunner.runOneTest):
249         (main):
250         (main.runTestList):
251         * JetStream2/__init__.py: Added.
252
253 2019-07-23  Keith Rollin  <krollin@apple.com>
254
255         Remove rest of NavigatorContentUtils support
256         https://bugs.webkit.org/show_bug.cgi?id=200052
257         <rdar://problem/53467357>
258
259         Reviewed by Alex Christensen.
260
261         Bug 196070 removes most of the support for the never-used
262         NavigatorContentUtils facility. However, there were still a couple of
263         left-over references after that change. This patch sweeps those up.
264
265         * StitchMarker/wtf/FeatureDefines.h:
266
267 2019-07-12  Caio Lima  <ticaiolima@gmail.com>
268
269         [BigInt] Add ValueBitLShift into DFG
270         https://bugs.webkit.org/show_bug.cgi?id=192664
271
272         Reviewed by Saam Barati.
273
274         * BigIntBench/big-int-simple-lshift.js: Added.
275
276 2019-06-28  Konstantin Tokarev  <annulen@yandex.ru>
277
278         Remove traces of ENABLE_ICONDATABASE remaining after its removal in 219733
279         https://bugs.webkit.org/show_bug.cgi?id=199317
280
281         Reviewed by Michael Catanzaro.
282
283         While IconDatabase and all code using it was removed,
284         ENABLE_ICONDATABASE still exists as build option and C++ macro.
285
286         * StitchMarker/wtf/FeatureDefines.h:
287
288 2019-06-27  Beth Dakin  <bdakin@apple.com>
289
290         Upstream use of MACCATALYST
291         https://bugs.webkit.org/show_bug.cgi?id=199245
292         rdar://problem/51687723
293
294         Reviewed by Tim Horton.
295
296         * MediaTime/Configurations/SDKVariant.xcconfig:
297
298 2019-06-20  Carlos Garcia Campos  <cgarcia@igalia.com>
299
300         [GTK] Remove support for GTK2 plugins
301         https://bugs.webkit.org/show_bug.cgi?id=199065
302
303         Reviewed by Sergio Villar Senin.
304
305         * StitchMarker/wtf/Platform.h:
306
307 2019-05-08  Caio Lima  <ticaiolima@gmail.com>
308
309         [BigInt] Add ValueMod into DFG
310         https://bugs.webkit.org/show_bug.cgi?id=186174
311
312         Reviewed by Saam Barati.
313
314         * BigIntBench/big-int-simple-mod.js: Added.
315
316 2019-05-06  Saam Barati  <sbarati@apple.com>
317
318         JS2 should print scores for different categories
319         https://bugs.webkit.org/show_bug.cgi?id=197605
320
321         Reviewed by Yusuke Suzuki.
322
323         * JetStream2/JetStreamDriver.js:
324         (displayCategoryScores):
325         (isInBrowser.document.onkeydown):
326         (Driver.prototype.async.start):
327         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
328
329 2019-04-16  Jon Lee  <jonlee@apple.com>
330
331         [MotionMark] The text "kkkj" is causing Firefox console warning: `unreachable code after return statement`
332         https://bugs.webkit.org/show_bug.cgi?id=196814
333
334         Reviewed by Wenson Hsieh.
335
336         * MotionMark/resources/extensions.js:
337         (subtract.subtract.sampleY):
338
339 2019-04-15  Myles C. Maxfield  <mmaxfield@apple.com>
340
341         [Cocoa] FontPlatformData objects aren't cached at all when using font-family:system-ui
342         https://bugs.webkit.org/show_bug.cgi?id=196846
343         <rdar://problem/49499971>
344
345         Reviewed by Simon Fraser and Darin Adler.
346
347         * Layout/system-ui-rebuild-emoji.html: Added.
348
349 2019-04-03  Myles C. Maxfield  <mmaxfield@apple.com>
350
351         -apple-trailing-word is needed for browser detection
352         https://bugs.webkit.org/show_bug.cgi?id=196575
353
354         Unreviewed.
355
356         * MotionMark/resources/debug-runner/motionmark.css:
357         (#intro .start-benchmark p):
358
359 2019-03-27  Ryan Haddad  <ryanhaddad@apple.com>
360
361         REGRESSION (r242911?): High Sierra Release WK2 Perf bot timing out while running IndexedDB/large-number-of-inserts.html
362         https://bugs.webkit.org/show_bug.cgi?id=195952
363
364         Unreviewed test gardening.
365
366         * Skipped: Skip the affected test so the bot will finish the run.
367
368 2019-03-27  Saam Barati  <sbarati@apple.com>
369
370         JetStream 2 should not report time values as scores
371         https://bugs.webkit.org/show_bug.cgi?id=196334
372
373         Reviewed by Yusuke Suzuki.
374
375         * JetStream2/JetStreamDriver.js:
376         (toTimeValue):
377         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
378
379 2019-03-27  Saam barati  <sbarati@apple.com>
380
381         Unreviewed. Fix individual benchmark description urls to go to in-depth.html instead of about.html
382
383         * JetStream2/JetStreamDriver.js:
384         (Driver.prototype.prepareToRun.text.div.id.string_appeared_here.h3):
385
386 2019-03-25  Saam barati  <sbarati@apple.com>
387
388         Update the blurb describing JetStream2 and fix a broken link in the in-depth page.
389
390         Rubber-stamped by Filip Pizlo.
391
392         * JetStream2/in-depth.html:
393         * JetStream2/index.html:
394
395 2019-03-18  Saam Barati  <sbarati@apple.com>
396
397         Add a plan file for JetStream 2
398         https://bugs.webkit.org/show_bug.cgi?id=190734
399
400         Reviewed by Mark Lam.
401
402         Make the delay when starting via run-benchmark longer to allow
403         the page's animation to finish.
404
405         * JetStream2/JetStreamDriver.js:
406         (Driver.prototype.async.initialize):
407
408 2019-03-15  Jon Davis  <jond@apple.com>
409
410         Fixed a bug preventing the overall sore to be displayed on JetStream2
411         https://bugs.webkit.org/show_bug.cgi?id=195817
412
413         Rubber-stamped by Saam Barati.
414
415         * JetStream2/JetStreamDriver.js:
416         (Driver.prototype.async.start):
417
418 2019-03-14  Jon Davis  <jond@apple.com>
419
420         Implement a new design for JetStream2
421         https://bugs.webkit.org/show_bug.cgi?id=195492
422
423         Reviewed by Saam Barati.
424
425         * JetStream2/JetStream.css:
426         (html):
427         (body):
428         (::selection):
429         (main):
430         (.logo):
431         (.logo .logo-image):
432         (#jetstreams):
433         (#jetstreams svg):
434         (.summary):
435         (.summary + .summary):
436         (.summary:empty):
437         (article, .summary):
438         (h1):
439         (h2, h3, h4, h5, h6):
440         (h4, h5, h6):
441         (p):
442         (h5, h6):
443         (h6):
444         (a:link,):
445         (a:hover,):
446         (#status):
447         (#status label,):
448         (a.button):
449         (#status.loading):
450         (#status.error):
451         (.error h2, .error p):
452         (.error h2):
453         (#result-summary):
454         (#result-summary label):
455         (#result-summary .score):
456         (#result-summary .score .interval):
457         (#results):
458         (.benchmark):
459         (.benchmark h3, .benchmark h4, .benchmark .result, .benchmark label):
460         (.benchmark-running h4, .benchmark-running .result, .benchmark-running label):
461         (.benchmark-done h3, .benchmark-done h4, .benchmark-done .result, .benchmark-done label):
462         (.benchmark h3):
463         (.benchmark-running h3):
464         (.benchmark-done h3):
465         (.benchmark h3 a,):
466         (.benchmark-done h3 a:hover):
467         (.benchmark h4):
468         (.benchmark-done h4):
469         (.benchmark p,):
470         (.benchmark .result):
471         (.benchmark-done .result):
472         (.benchmark label):
473         (.benchmark-done label):
474         (@keyframes fadein):
475         (to):
476         (@keyframes scaledown):
477         (@keyframes shine):
478         (100%):
479         (@keyframes swingin):
480         (@media (max-width: 415px)):
481         (#logo): Deleted.
482         (p.summary): Deleted.
483         (p.summary + p.summary): Deleted.
484         (p.summary:empty): Deleted.
485         (p:first-of-type): Deleted.
486         (#status a:link): Deleted.
487         (.interval): Deleted.
488         (a:link, a:visited): Deleted.
489         (a:link:hover): Deleted.
490         (table): Deleted.
491         (td, th): Deleted.
492         (th): Deleted.
493         (tr:first-child > th:nth-child(even)): Deleted.
494         (tr:nth-child(even):not(:first-child, .benchmark-running)): Deleted.
495         (.result): Deleted.
496         (.benchmark-running): Deleted.
497         (:not(.benchmark-running) .result): Deleted.
498         (.benchmark-running .result): Deleted.
499         (.benchmark-name): Deleted.
500         (.result.category,): Deleted.
501         (.benchmark-name:not(.category):not(.geometric-mean)): Deleted.
502         (.geometric-mean): Deleted.
503         (.benchmark-name:not(:first-child)): Deleted.
504         (.benchmark-name a:link,): Deleted.
505         (.highlighted-result): Deleted.
506         * JetStream2/JetStream2Logo.svg: Added.
507         * JetStream2/JetStreamDriver.js:
508         (updateUI):
509         (Driver.prototype.async.start):
510         (Driver.prototype.prepareToRun.text.div.id.string_appeared_here.h3):
511         (Driver.prototype.prepareToRun.text.span):
512         (Driver.prototype.prepareToRun.a.h3.h4):
513         (Driver.prototype.prepareToRun.text.div):
514         (Driver.prototype.async.initialize):
515         (Driver.prototype.async.fetchResources.prototype.statusElement.onclick):
516         (prototype.fetchResources):
517         (prototype.updateUIBeforeRun):
518         (prototype.updateUIAfterRun):
519         (Driver.prototype.async.fetchResources.statusElement.innerHTML.a.href.string_appeared_here):
520         (Driver.prototype.async.fetchResources):
521         (Driver.prototype.prepareToRun.text.tr.id.string_appeared_here.FIXME.link.to.benchmark.explanation.td): Deleted.
522         * JetStream2/clouds.svg: Added.
523         * JetStream2/in-depth.html:
524         * JetStream2/index.html:
525         * JetStream2/jetstreams.svg: Added.
526
527 2019-03-11  Caio Lima  <ticaiolima@gmail.com>
528
529         [ESNext][BigInt] Implement "~" unary operation
530         https://bugs.webkit.org/show_bug.cgi?id=182216
531
532         Reviewed by Keith Miller.
533
534         * BigIntBench/big-int-simple-bit-not.js: Added.
535
536 2019-03-04  Saam Barati  <sbarati@apple.com>
537
538         Add a detailed summary page for JetStream 2
539         https://bugs.webkit.org/show_bug.cgi?id=195014
540
541         Rubber-stamped by Darin Adler.
542
543         * JetStream2/JetStream.css:
544         (a:link, a:visited):
545         * JetStream2/in-depth.html: Added.
546
547 2019-02-08  Saam barati  <sbarati@apple.com>
548
549         Update JetStream2 CLI to not print "ms" since we now just print the score values
550         https://bugs.webkit.org/show_bug.cgi?id=193981
551
552         Reviewed by Yusuke Suzuki.
553
554         * JetStream2/JetStreamDriver.js:
555         (DefaultBenchmark.prototype.updateUIAfterRun):
556         (DefaultBenchmark):
557         (WSLBenchmark.prototype.updateUIAfterRun):
558         (WSLBenchmark):
559         (WasmBenchmark.prototype.updateUIAfterRun):
560         (WasmBenchmark):
561
562 2019-01-31  Sihui Liu  <sihui_liu@apple.com>
563
564         REGRESSION (r240358): IndexedDB/large-binary-keys.html and IndexedDB/large-string-keys.html perf tests failing
565         https://bugs.webkit.org/show_bug.cgi?id=193956
566         <rdar://problem/47623109>
567
568         Reviewed by Ryosuke Niwa.
569
570         r240358 sets default IDB quota as 50 MB for single test, we need to increase that for big tests. 
571
572         * IndexedDB/large-binary-keys.html:
573         * IndexedDB/large-string-keys.html:
574
575 2019-01-18  Saam Barati  <sbarati@apple.com>
576
577         Use scores everywhere in JetStream2's UI
578         https://bugs.webkit.org/show_bug.cgi?id=193593
579
580         Reviewed by Keith Miller.
581
582         * JetStream2/JetStreamDriver.js:
583         (DefaultBenchmark.prototype.processResults):
584         (DefaultBenchmark.prototype.get score):
585         (DefaultBenchmark.scoreDescription):
586         (WSLBenchmark.prototype.processResults):
587         (WSLBenchmark.prototype.get score):
588         (WSLBenchmark.scoreDescription):
589         (WasmBenchmark.prototype.processResults):
590         (WasmBenchmark.prototype.get score):
591         (WasmBenchmark.scoreDescription):
592         * JetStream2/index.html:
593
594 2019-01-18  Jer Noble  <jer.noble@apple.com>
595
596         SDK_VARIANT build destinations should be separate from non-SDK_VARIANT builds
597         https://bugs.webkit.org/show_bug.cgi?id=189553
598
599         Reviewed by Tim Horton.
600
601         * MediaTime/Configurations/Base.xcconfig:
602         * MediaTime/Configurations/SDKVariant.xcconfig: Added.
603
604 2019-01-17  Truitt Savell  <tsavell@apple.com>
605
606         Unreviewed, rolling out r240124.
607
608         This commit broke an internal build.
609
610         Reverted changeset:
611
612         "SDK_VARIANT build destinations should be separate from non-
613         SDK_VARIANT builds"
614         https://bugs.webkit.org/show_bug.cgi?id=189553
615         https://trac.webkit.org/changeset/240124
616
617 2019-01-17  Jer Noble  <jer.noble@apple.com>
618
619         SDK_VARIANT build destinations should be separate from non-SDK_VARIANT builds
620         https://bugs.webkit.org/show_bug.cgi?id=189553
621
622         Reviewed by Tim Horton.
623
624         * MediaTime/Configurations/Base.xcconfig:
625         * MediaTime/Configurations/SDKVariant.xcconfig: Added.
626
627 2018-12-20  Caio Lima  <ticaiolima@gmail.com>
628
629         [BigInt] We should enable CSE into arithmetic operations that speculate BigIntUse
630         https://bugs.webkit.org/show_bug.cgi?id=192723
631
632         Reviewed by Yusuke Suzuki.
633
634         * BigIntBench/big-int-cse.js: Added.
635         * BigIntBench/big-int-global-cse.js: Added.
636         * BigIntBench/big-int-licm.js: Added.
637
638 2018-12-19  Commit Queue  <commit-queue@webkit.org>
639
640         Unreviewed, rolling out r239377.
641         https://bugs.webkit.org/show_bug.cgi?id=192921
642
643         broke 32-bit JSC tests (Requested by keith_miller on #webkit).
644
645         Reverted changeset:
646
647         "[BigInt] We should enable CSE into arithmetic operations that
648         speculate BigIntUse"
649         https://bugs.webkit.org/show_bug.cgi?id=192723
650         https://trac.webkit.org/changeset/239377
651
652 2018-12-19  Caio Lima  <ticaiolima@gmail.com>
653
654         [BigInt] We should enable CSE into arithmetic operations that speculate BigIntUse
655         https://bugs.webkit.org/show_bug.cgi?id=192723
656
657         Reviewed by Saam Barati.
658
659         * BigIntBench/big-int-cse.js: Added.
660         * BigIntBench/big-int-global-cse.js: Added.
661         * BigIntBench/big-int-licm.js: Added.
662
663 2018-12-17  Suresh Koppisetty  <skoppisetty@apple.com>
664
665         Add "-o/--output" option to startup.py and new_tab.py benchmark scripts to save the results in json format.
666         https://bugs.webkit.org/show_bug.cgi?id=192385
667
668         Reviewed by Ryosuke Niwa.
669
670         Sample json output for new tab benchmark script after running for 2 iterations and 2 groups. Values are in milliseconds.
671         {
672                 "NewTabBenchmark": {
673                         "metrics": {
674                                 "Time": {
675                                         "current": [
676                                                 [
677                                                         410.2939453125,
678                                                         307.81494140625
679                                                 ],
680                                                 [
681                                                         340.616943359375,
682                                                         265.94384765625
683                                                 ]
684                                         ]
685                                 }
686                         }
687                 }
688         }
689
690         Sample json output for startup time benchmark script after running for 2 iterations. Values are in milliseconds.
691         {
692                 "StartupBenchmark": {
693                         "metrics": {
694                                 "Time": {
695                                         "current": [
696                                                 [
697                                                         1415.2099609375,
698                                                         1439.552978515625
699                                                 ]
700                                         ]
701                                 }
702                         }
703                 }
704         }
705
706         * LaunchTime/launch_time.py:
707         * LaunchTime/new_tab.py:
708         (NewTabBenchmark.get_test_name):
709         * LaunchTime/startup.py:
710         (StartupBenchmark.get_test_name):
711
712 2018-12-17  Suresh Koppisetty  <skoppisetty@apple.com>
713
714         Import FeedbackServer only if "-f/--feedback-in-browser" option is enabled.
715         https://bugs.webkit.org/show_bug.cgi?id=192378
716
717         Reviewed by Ryosuke Niwa.
718
719         FeedbackServer currently depends on Tornado-5.1, which further adds
720         dependency of "singledispatch", "backports-abc" and "futures" python libraries.
721         Importing FeedbackServer only if "-f/--feedback-in-browser" option is enabled
722         will let us run the benchmark scripts without installing any new python libraries.
723
724         * LaunchTime/launch_time.py:
725         * LaunchTime/new_tab.py:
726
727 2018-12-13  Caio Lima  <ticaiolima@gmail.com>
728
729         [BigInt] Add ValueDiv into DFG
730         https://bugs.webkit.org/show_bug.cgi?id=186178
731
732         Reviewed by Yusuke Suzuki.
733
734         * BigIntBench/big-int-simple-div.js: Added.
735         * BigIntBench/value-div-type-propagation.js: Added.
736
737 2018-12-10  Caio Lima  <ticaiolima@gmail.com>
738
739         [BigInt] Add ValueMul into DFG
740         https://bugs.webkit.org/show_bug.cgi?id=186175
741
742         Reviewed by Yusuke Suzuki.
743
744         * BigIntBench/big-int-simple-mul.js: Added.
745         * BigIntBench/value-mul-type-propagation.js: Added.
746
747 2018-11-07  Caio Lima  <ticaiolima@gmail.com>
748
749         [BigInt] Add support to BigInt into ValueAdd
750         https://bugs.webkit.org/show_bug.cgi?id=186177
751
752         Reviewed by Keith Miller.
753
754         The idea of BigIntBench is to provide a set of microbenchmarks and
755         benchmarks to evaluate how fast BigInt computations are happening on
756         JSC implementation.
757
758         Now, we are adding microbenchmarks in this set,
759         but the plan is to move these tests to "JSTest/microbenchmarks" when
760         BigInt is enabled by default. After that, the focus of Bigint bench is
761         to provide a set of tests that represents real use cases of BigInt in
762         JS programs.
763
764         * BigIntBench/big-int-add-prediction-propagation.js: Added.
765         * BigIntBench/big-int-simple-add.js: Added.
766         * BigIntBench/big-int-simple-sub.js: Added.
767
768 2018-11-07  Tadeu Zagallo  <tzagallo@apple.com>
769
770         REGRESSION(r237547): Test failures on 32-bit JSC since the JIT was disabled
771         https://bugs.webkit.org/show_bug.cgi?id=191184
772
773         Reviewed by Saam Barati.
774
775         Skip test when the JIT is disabled, it takes too long to run on CLoop.
776
777         * JetStream/cdjs/main.js:
778
779 2018-11-05  Myles C. Maxfield  <mmaxfield@apple.com>
780
781         Cache glyph paths and share underline skipping code between all the ports
782         https://bugs.webkit.org/show_bug.cgi?id=191239
783
784         Reviewed by Alex Christensen.
785
786         Measures the performance of drawing a whole lot of underlines
787
788         * Layout/underline.html: Added.
789
790 2018-10-25  Saam Barati  <sbarati@apple.com>
791
792         Check in corresponding C code in JetStream 2
793         https://bugs.webkit.org/show_bug.cgi?id=190733
794
795         Rubber-stamped by Filip Pizlo.
796
797         * JetStream2/simple/float-mm.c: Added.
798         * JetStream2/wasm/HashSet.cpp: Added.
799         * JetStream2/wasm/gcc-loops.cpp: Added.
800         * JetStream2/wasm/quicksort.c: Added.
801         * JetStream2/wasm/TSF: Added.
802         * JetStream2/wasm/TSF/build.sh: Added.
803         * JetStream2/wasm/TSF/config.h: Added.
804         * JetStream2/wasm/TSF/gpc.h: Added.
805         * JetStream2/wasm/TSF/gpc_code_gen_util.c: Added.
806         * JetStream2/wasm/TSF/gpc_instruction.c: Added.
807         * JetStream2/wasm/TSF/gpc_instruction_dispatch.gen: Added.
808         * JetStream2/wasm/TSF/gpc_instruction_size.gen: Added.
809         * JetStream2/wasm/TSF/gpc_instruction_stack_effects.gen: Added.
810         * JetStream2/wasm/TSF/gpc_instruction_static_size.gen: Added.
811         * JetStream2/wasm/TSF/gpc_instruction_to_string.gen: Added.
812         * JetStream2/wasm/TSF/gpc_int_common.h: Added.
813         * JetStream2/wasm/TSF/gpc_intable.c: Added.
814         * JetStream2/wasm/TSF/gpc_internal.h: Added.
815         * JetStream2/wasm/TSF/gpc_interpreter.gen: Added.
816         * JetStream2/wasm/TSF/gpc_program.c: Added.
817         * JetStream2/wasm/TSF/gpc_proto.c: Added.
818         * JetStream2/wasm/TSF/gpc_stack_height.c: Added.
819         * JetStream2/wasm/TSF/gpc_threaded.c: Added.
820         * JetStream2/wasm/TSF/gpc_worklist.h: Added.
821         * JetStream2/wasm/TSF/tsf.h: Added.
822         * JetStream2/wasm/TSF/tsf_adaptive_reader.c: Added.
823         * JetStream2/wasm/TSF/tsf_asprintf.c: Added.
824         * JetStream2/wasm/TSF/tsf_atomics.h: Added.
825         * JetStream2/wasm/TSF/tsf_buf_reader.c: Added.
826         * JetStream2/wasm/TSF/tsf_buf_writer.c: Added.
827         * JetStream2/wasm/TSF/tsf_buffer.c: Added.
828         * JetStream2/wasm/TSF/tsf_build_defines.h: Added.
829         * JetStream2/wasm/TSF/tsf_config.h: Added.
830         * JetStream2/wasm/TSF/tsf_config_stub.h: Added.
831         * JetStream2/wasm/TSF/tsf_copier.c: Added.
832         * JetStream2/wasm/TSF/tsf_define_helpers.c: Added.
833         * JetStream2/wasm/TSF/tsf_define_helpers.h: Added.
834         * JetStream2/wasm/TSF/tsf_destructor.c: Added.
835         * JetStream2/wasm/TSF/tsf_error.c: Added.
836         * JetStream2/wasm/TSF/tsf_format.h: Added.
837         * JetStream2/wasm/TSF/tsf_fsdb.c: Added.
838         * JetStream2/wasm/TSF/tsf_fsdb_protocol.c: Added.
839         * JetStream2/wasm/TSF/tsf_fsdb_protocol.h: Added.
840         * JetStream2/wasm/TSF/tsf_generator.c: Added.
841         * JetStream2/wasm/TSF/tsf_gpc_code_gen.c: Added.
842         * JetStream2/wasm/TSF/tsf_indent.h: Added.
843         * JetStream2/wasm/TSF/tsf_internal.h: Added.
844         * JetStream2/wasm/TSF/tsf_internal_config.h: Added.
845         * JetStream2/wasm/TSF/tsf_internal_config_stub.h: Added.
846         * JetStream2/wasm/TSF/tsf_inttypes.h: Added.
847         * JetStream2/wasm/TSF/tsf_io.c: Added.
848         * JetStream2/wasm/TSF/tsf_io_utils.c: Added.
849         * JetStream2/wasm/TSF/tsf_ir.c: Added.
850         * JetStream2/wasm/TSF/tsf_ir.h: Added.
851         * JetStream2/wasm/TSF/tsf_ir_different.c: Added.
852         * JetStream2/wasm/TSF/tsf_ir_different.h: Added.
853         * JetStream2/wasm/TSF/tsf_ir_speed.c: Added.
854         * JetStream2/wasm/TSF/tsf_limits.c: Added.
855         * JetStream2/wasm/TSF/tsf_named_type.c: Added.
856         * JetStream2/wasm/TSF/tsf_native.c: Added.
857         * JetStream2/wasm/TSF/tsf_parser.c: Added.
858         * JetStream2/wasm/TSF/tsf_primitive.c: Added.
859         * JetStream2/wasm/TSF/tsf_ra_type_man.c: Added.
860         * JetStream2/wasm/TSF/tsf_reflect.c: Added.
861         * JetStream2/wasm/TSF/tsf_region.h: Added.
862         * JetStream2/wasm/TSF/tsf_serial_in_man.c: Added.
863         * JetStream2/wasm/TSF/tsf_serial_out_man.c: Added.
864         * JetStream2/wasm/TSF/tsf_serial_protocol.h: Added.
865         * JetStream2/wasm/TSF/tsf_sha1.c: Added.
866         * JetStream2/wasm/TSF/tsf_sha1.h: Added.
867         * JetStream2/wasm/TSF/tsf_sha1_writer.c: Added.
868         * JetStream2/wasm/TSF/tsf_sort.c: Added.
869         * JetStream2/wasm/TSF/tsf_st.c: Added.
870         * JetStream2/wasm/TSF/tsf_st.h: Added.
871         * JetStream2/wasm/TSF/tsf_st_typetable.c: Added.
872         * JetStream2/wasm/TSF/tsf_stream_file_input.c: Added.
873         * JetStream2/wasm/TSF/tsf_stream_file_output.c: Added.
874         * JetStream2/wasm/TSF/tsf_type.c: Added.
875         * JetStream2/wasm/TSF/tsf_type_in_map.c: Added.
876         * JetStream2/wasm/TSF/tsf_type_out_map.c: Added.
877         * JetStream2/wasm/TSF/tsf_type_table.c: Added.
878         * JetStream2/wasm/TSF/tsf_types.h: Added.
879         * JetStream2/wasm/TSF/tsf_util.h: Added.
880         * JetStream2/wasm/TSF/tsf_version.c: Added.
881         * JetStream2/wasm/TSF/tsf_zip_abstract.c: Added.
882         * JetStream2/wasm/TSF/tsf_zip_abstract.h: Added.
883         * JetStream2/wasm/TSF/tsf_zip_attr.c: Added.
884         * JetStream2/wasm/TSF/tsf_zip_reader.c: Added.
885         * JetStream2/wasm/TSF/tsf_zip_writer.c: Added.
886
887 2018-10-18  Saam Barati  <sbarati@apple.com>
888
889         Make JetStream 2
890         https://bugs.webkit.org/show_bug.cgi?id=187829
891
892         Rubber-stamped by Mark Lam.
893
894         This patch checks in the new JetStream 2 benchmark. JetStream 2 is
895         a new JavaScript and Web Assembly benchmark. JetStream 2's goal
896         is to measure the startup, worst case, and peak throughput performance
897         of the JavaScript engine. JetStream 2 incorporates these previous
898         benchmarks:
899         - JetStream
900         - ARES-6
901         - Kraken
902         - Web Tooling Benchmark
903         - WasmBench
904         - RexBench
905         
906         JetStream 2 also adds some new benchmarks:
907         - Two tests emphasizing web worker performance.
908         - One test emphasizing Promise, async iteration, and DataView performance.
909         - Two new code load tests.
910         - WSL: a test measuring all kinds of things, especially emphasizing exception performance.
911
912         * JetStream2: Added.
913
914 2018-10-02  Chris Dumez  <cdumez@apple.com>
915
916         Regression(r236613): Parser/html-parser.html performance test is failing
917         https://bugs.webkit.org/show_bug.cgi?id=190201
918
919         Unreviewed, apply the same fix as in r236699 to make sure the iframe is treated as
920         same origin, so that calls to document.open() / document.write() succeed.
921
922         * Parser/html-parser.html:
923
924 2018-10-01  Chris Dumez  <cdumez@apple.com>
925
926         Regression(r236613): Parser/HTML5-8266-ParseOnly.html performance tests no longer runs
927         https://bugs.webkit.org/show_bug.cgi?id=190174
928
929         Reviewed by Ryosuke Niwa.
930
931         Update iframe's sandbox to use 'allow-same-origin' so that the iframe is treated as same-origin and so that
932         the top frame can call document.open() / document.write() on the subframe. This test was also failing in
933         Firefox without this change.
934
935         * Parser/HTML5-8266-ParseOnly.html:
936
937 2018-08-27  Aditya Keerthi  <akeerthi@apple.com>
938
939         Consolidate ENABLE_INPUT_TYPE_COLOR and ENABLE_INPUT_TYPE_COLOR_POPOVER
940         https://bugs.webkit.org/show_bug.cgi?id=188931
941
942         Reviewed by Wenson Hsieh.
943
944         * StitchMarker/wtf/FeatureDefines.h: Removed ENABLE_INPUT_TYPE_COLOR_POPOVER.
945
946 2018-08-23  Don Olmstead  <don.olmstead@sony.com>
947
948         [CMake] Have checks are not getting set inside CMake properly
949         https://bugs.webkit.org/show_bug.cgi?id=188901
950
951         Reviewed by Michael Catanzaro.
952
953         Add notification that MallocBench is disabled.
954
955         * CMakeLists.txt:
956
957 2018-08-23  Don Olmstead  <don.olmstead@sony.com>
958
959         [CMake] Add HAVE_MALLOC_TRIM definition
960         https://bugs.webkit.org/show_bug.cgi?id=188897
961
962         Reviewed by Konstantin Tokarev.
963
964         MallocBench should only be built on Apple platforms and platforms that
965         have malloc_trim.
966
967         * CMakeLists.txt:
968
969 2018-08-13  Jon Lee  <jonlee@apple.com>
970
971         Update MotionMark version number
972         https://bugs.webkit.org/show_bug.cgi?id=188535
973         <rdar://problem/43254078>
974
975         Reviewed by Said Abou-Hallawa.
976
977         * MotionMark/resources/strings.js: To 1.1.
978
979 2018-08-13  Jon Lee  <jonlee@apple.com>
980
981         [MotionMark] Update Multiply test
982         https://bugs.webkit.org/show_bug.cgi?id=188532
983         <rdar://problem/43252151>
984
985         Reviewed by Said Abou-Hallawa.
986
987         Update Multiply test to have a larger upper limit of number of particles. Cycle through three
988         different ways of hiding an element. Add new Multiple suite that isolates those components for
989         debugging purposes.
990
991         * MotionMark/resources/debug-runner/tests.js:
992         * MotionMark/tests/dom/multiply.html: Added.
993         * MotionMark/tests/dom/resources/multiply.js: Added.
994         * MotionMark/tests/master/resources/multiply.js:
995
996 2018-08-13  Jon Lee  <jonlee@apple.com>
997
998         [MotionMark] Update Leaves test
999         https://bugs.webkit.org/show_bug.cgi?id=188530
1000         <rdar://problem/43251862>
1001
1002         Reviewed by Said Abou-Hallawa.
1003
1004         Update Leaves test to include opacity and scale. Add new Leaves suite that isolate those components
1005         for debugging purposes.
1006
1007         * MotionMark/resources/debug-runner/tests.js:
1008         * MotionMark/tests/dom/leaves.html:
1009         * MotionMark/tests/dom/resources/leaves.js:
1010         * MotionMark/tests/master/resources/leaves.js:
1011
1012 2018-08-10  Ben Richards  <benton_richards@apple.com>
1013
1014         Add ability to ignore process prewarming for launch time benchmark
1015         https://bugs.webkit.org/show_bug.cgi?id=188462
1016
1017         Reviewed by Ryosuke Niwa.
1018
1019         Added a flag to the new tab benchmark that will open a blank tab before the tab to be measured in order to ignore process prewarming.
1020
1021         * LaunchTime/launch_time.py:
1022         (DefaultLaunchTimeHandler):
1023         (DefaultLaunchTimeHandler.on_receive_stop_time): Deleted.
1024         (DefaultLaunchTimeHandler.on_receive_stop_signal):
1025         (DefaultLaunchTimeHandler.do_HEAD):
1026         (DefaultLaunchTimeHandler.do_GET):
1027         (DefaultLaunchTimeHandler.do_POST):
1028         (LaunchTimeBenchmark):
1029         (LaunchTimeBenchmark._standard_deviation): Fixed divide by zero bug when '-n' is set to 1
1030         (LaunchTimeBenchmark.open_tab): Added option to open a blank tab
1031         (LaunchTimeBenchmark.run):
1032         * LaunchTime/new_tab.py:
1033         (NewTabBenchmark.initialize):
1034         (NewTabBenchmark.run_iteration):
1035         (NewTabBenchmark.will_parse_arguments):
1036         (NewTabBenchmark.did_parse_arguments):
1037         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1038         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1039         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1040         (NewTabBenchmark):
1041         * LaunchTime/startup.py:
1042         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1043         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1044         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1045         (StartupBenchmark):
1046
1047 2018-08-10  Truitt Savell  <tsavell@apple.com>
1048
1049         Unreviewed, rolling out r234750.
1050
1051         Caused 185 perf test failures.
1052
1053         Reverted changeset:
1054
1055         "Add ability to ignore process prewarming for launch time
1056         benchmark"
1057         https://bugs.webkit.org/show_bug.cgi?id=188462
1058         https://trac.webkit.org/changeset/234750
1059
1060 2018-08-09  Ben Richards  <benton_richards@apple.com>
1061
1062         Add ability to ignore process prewarming for launch time benchmark
1063         https://bugs.webkit.org/show_bug.cgi?id=188462
1064
1065         Reviewed by Ryosuke Niwa.
1066
1067         Added a flag to the new tab benchmark that will open a blank tab before the tab to be measured in order to ignore process prewarming.
1068
1069         * LaunchTime/launch_time.py:
1070         (DefaultLaunchTimeHandler):
1071         (DefaultLaunchTimeHandler.on_receive_stop_time): Deleted.
1072         (DefaultLaunchTimeHandler.on_receive_stop_signal):
1073         (DefaultLaunchTimeHandler.do_HEAD):
1074         (DefaultLaunchTimeHandler.do_GET):
1075         (DefaultLaunchTimeHandler.do_POST):
1076         (LaunchTimeBenchmark):
1077         (LaunchTimeBenchmark._standard_deviation): Fixed divide by zero bug when '-n' is set to 1
1078         (LaunchTimeBenchmark.open_tab): Added option to open a blank tab
1079         (LaunchTimeBenchmark.run):
1080         * LaunchTime/new_tab.py:
1081         (NewTabBenchmark.initialize):
1082         (NewTabBenchmark.run_iteration):
1083         (NewTabBenchmark.will_parse_arguments):
1084         (NewTabBenchmark.did_parse_arguments):
1085         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1086         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1087         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1088         (NewTabBenchmark):
1089         * LaunchTime/startup.py:
1090         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1091         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1092         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1093         (StartupBenchmark):
1094
1095 2018-07-26  Andy VanWagoner  <andy@vanwagoner.family>
1096
1097         [INTL] Remove INTL sub-feature compile flags
1098         https://bugs.webkit.org/show_bug.cgi?id=188081
1099
1100         Reviewed by Michael Catanzaro.
1101
1102         Removed old and unused i18n flag.
1103
1104         * StitchMarker/wtf/FeatureDefines.h:
1105
1106 2018-07-20  David Fenton  <david_fenton@apple.com>
1107
1108         REGRESSION (234006) ERROR: Blocked access to external URL http://code.jquery.com/jquery-1.9.1.min.js
1109         https://bugs.webkit.org/show_bug.cgi?id=187858
1110
1111         Unreviewed test gardening, skipped test.
1112
1113         * Skipped: LaunchTime
1114
1115 2018-07-19  Ben Richards  <benton_richards@apple.com>
1116
1117         Add benchmark for WebKit process launch times
1118         https://bugs.webkit.org/show_bug.cgi?id=186414
1119
1120         Reviewed by Ryosuke Niwa.
1121
1122         Added two benchmarks, one for measuring browser new tab launch time and one for browser startup time.
1123
1124         * LaunchTime/.gitignore: Added.
1125         * LaunchTime/feedback_client.html: Added.
1126         Displays benchmark progress in browser
1127         * LaunchTime/feedback_server.py: Added.
1128         (FeedbackServer): Sends data to feedback_client via websocket
1129         (FeedbackServer.__init__):
1130         (FeedbackServer._create_app):
1131         (FeedbackServer._start_server):
1132         (FeedbackServer._send_all_messages):
1133         (FeedbackServer.start):
1134         (FeedbackServer.stop):
1135         (FeedbackServer.send_message): Send a message to the feedback_client
1136         (FeedbackServer.wait_until_client_has_loaded): Wait until the feedback_client has opened a websocket connection to the feedback_server
1137         (FeedbackServer.MainHandler): Handler factory to create handler that serves feedback_client.html
1138         (FeedbackServer.MainHandler.Handler):
1139         (FeedbackServer.MainHandler.Handler.get):
1140         (FeedbackServer.WSHandler): Handler factory to create handler that sends data to feedback client
1141         (FeedbackServer.WSHandler.Handler):
1142         (FeedbackServer.WSHandler.Handler.open): On websocket connection opened
1143         (FeedbackServer.WSHandler.Handler.on_close): On websocket connection closed
1144         * LaunchTime/launch_time.py: Added.
1145         (DefaultLaunchTimeHandler): Abstract HTTP request handler for launch time benchmarks
1146         (DefaultLaunchTimeHandler.get_test_page): Default test page to be overridden by benchmarks
1147         (DefaultLaunchTimeHandler.get_blank_page):
1148         (DefaultLaunchTimeHandler.on_receive_stop_time):
1149         (DefaultLaunchTimeHandler.do_HEAD):
1150         (DefaultLaunchTimeHandler.do_GET):
1151         (DefaultLaunchTimeHandler.do_POST):
1152         (DefaultLaunchTimeHandler.log_message): Suppresses HTTP logs from SimpleHTTPRequestHandler
1153         (LaunchTimeBenchmark): Abstract class which launch time benchmarks inherit from and override methods desired to customize
1154         (LaunchTimeBenchmark.__init__):
1155         (LaunchTimeBenchmark._parse_browser_bundle_path): Parser for bundle path option
1156         (LaunchTimeBenchmark._parse_args):
1157         (LaunchTimeBenchmark._run_server): Target function for main server thread
1158         (LaunchTimeBenchmark._setup_servers):
1159         (LaunchTimeBenchmark._clean_up):
1160         (LaunchTimeBenchmark._exit_due_to_exception):
1161         (LaunchTimeBenchmark._geometric_mean):
1162         (LaunchTimeBenchmark._standard_deviation):
1163         (LaunchTimeBenchmark._compute_results): Returns mean and std dev of list of results and pretty prints if should_print=True is specified
1164         (LaunchTimeBenchmark._wait_times): Mimic numpy.linspace
1165         (LaunchTimeBenchmark.open_tab): Open a browser tab with the html given by self.response_handler.get_test_page
1166         (LaunchTimeBenchmark.launch_browser): Open a broser to either the feedback client (if option is set) or a blank page
1167         (LaunchTimeBenchmark.quit_browser):
1168         (LaunchTimeBenchmark.quit_browser.quit_app):
1169         (LaunchTimeBenchmark.quit_browser.is_app_closed):
1170         (LaunchTimeBenchmark.close_tab):
1171         (LaunchTimeBenchmark.wait):
1172         (LaunchTimeBenchmark.log): Print to console and send to feedback client if --feedback-in-browser flag is used
1173         (LaunchTimeBenchmark.log_verbose): Only logs if --verbose flag is used
1174         (LaunchTimeBenchmark.run):
1175         (LaunchTimeBenchmark.group_init): Initialization done before each round of iterations
1176         (LaunchTimeBenchmark.run_iteration):
1177         (LaunchTimeBenchmark.initialize): Convenience method to be overriden by subclasses which is called at the end of __init__
1178         (LaunchTimeBenchmark.will_parse_arguments): Called before argparse.parse_args to let subclasses add new command line arguments
1179         (LaunchTimeBenchmark.did_parse_arguments): Called after argparse.parse_args to let subclass initialize based on command line arguments
1180         * LaunchTime/new_tab.py: Added
1181         (NewTabBenchmark):
1182         (NewTabBenchmark._parse_wait_time): Parser for wait time option
1183         (NewTabBenchmark.initialize):
1184         (NewTabBenchmark.run_iteration):
1185         (NewTabBenchmark.group_init):
1186         (NewTabBenchmark.will_parse_arguments):
1187         (NewTabBenchmark.did_parse_arguments):
1188         (NewTabBenchmark.ResponseHandler):
1189         (NewTabBenchmark.ResponseHandler.Handler):
1190         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1191         * LaunchTime/startup.py: Added
1192         (StartupBenchmark): This benchmark measures browser startup time and initial page load time
1193         (StartupBenchmark.initialize):
1194         (StartupBenchmark.run_iteration):
1195         (StartupBenchmark.ResponseHandler):
1196         (StartupBenchmark.ResponseHandler.Handler):
1197         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1198         * LaunchTime/thirdparty/__init__.py: Added.
1199         (AutoinstallImportHook): Auto installs tornado package for feedback server
1200         (AutoinstallImportHook.__init__):
1201         (AutoinstallImportHook._ensure_autoinstalled_dir_is_in_sys_path):
1202         (AutoinstallImportHook.find_module):
1203         (AutoinstallImportHook._install_tornado):
1204         (AutoinstallImportHook.greater_than_equal_to_version):
1205         (AutoinstallImportHook._install):
1206         (AutoinstallImportHook.get_latest_pypi_url):
1207         (AutoinstallImportHook.install_binary):
1208         (autoinstall_everything):
1209         (get_os_info):
1210
1211 2018-06-25  Jon Lee  <jonlee@apple.com>
1212
1213         [MotionMark] Add support for version numbers
1214         https://bugs.webkit.org/show_bug.cgi?id=186479
1215
1216         Reviewed by Said Abou-Hallawa.
1217
1218         Add support for displaying the version number as well as including it in the JSON results.
1219
1220         When loading the front page, script replaces any element with classname version with the
1221         version number of the benchmark, which is stored in Strings.version.
1222
1223         The JSON structure for the results includes a new version property:
1224             {
1225               "version": "1.0",
1226               "options": { ... },
1227               "data": [ ... ]
1228             }
1229
1230         When dragging a results file, the version listed will come from the JSON file. Older
1231         results will not have had the version property, in which case it will default to "1.0".
1232
1233         * MotionMark/index.html: Update title to some other default. Script will update it.
1234         Include the version number in the logo title.
1235         * MotionMark/developer.html: Ditto.
1236         * MotionMark/about.html: Ditto.
1237
1238         * MotionMark/resources/runner/motionmark.js:
1239         (ResultsDashboard): Update constructor to include version. This is used when serializing
1240         results out to JSON, and displaying the results panel in developer mode.
1241         (ResultsDashboard._processData): When running the benchmark, include benchmark version string
1242         in the results object.
1243         (ResultsDashboard.version):
1244         (window.benchmarkRunnerClient.willStartFirstIteration): When running the benchmark, pass the
1245         benchmark version string to the dashboard, which holds the results.
1246         (window.sectionsManager.setSectionVersion): Helper function to update the element in the
1247         section with the class name version.
1248         (window.benchmarkController.initialize): Populate all DOM elements with class name "version"
1249         with the version string. Update the page title.
1250         (window.benchmarkController.showResults): When showing results, update the version string
1251         based on what is included in the JSON results, which would be the same as the benchmark version.
1252         * MotionMark/resources/runner/motionmark.css: Include missing copyright. Wrap the SVG logo
1253         in a div and include the version string.
1254         * MotionMark/resources/strings.js: Add strings for the page title template, and the version.
1255
1256         * MotionMark/resources/debug-runner/motionmark.css:
1257         * MotionMark/resources/debug-runner/motionmark.js:
1258         (window.benchmarkRunnerClient.willStartFirstIteration): When running the benchmark, pass the
1259         benchmark version string to the dashboard, which holds the results.
1260         (window.benchmarkController.initialize): Populate all DOM elements with class name "version"
1261         with the version string. Update the page title. When dragging in JSON results, look for
1262         version to pass to the dashboard. If it doesn't exist, default to "1.0".
1263         (window.benchmarkController.showResults): When showing results, update the version string
1264         based on what is included in the JSON results, instead of the current benchmark version.
1265         * MotionMark/resources/debug-runner/tests.js: Update page title template.
1266
1267 2018-06-08  Jon Lee  <jonlee@apple.com>
1268
1269         [MotionMark] Rename Suits test files
1270         https://bugs.webkit.org/show_bug.cgi?id=186447
1271
1272         Reviewed by Said Abou-Hallawa.
1273
1274         * MotionMark/resources/runner/tests.js:
1275         * MotionMark/tests/master/resources/suits.js: Renamed from PerformanceTests/MotionMark/tests/master/resources/svg-particles.js.
1276         * MotionMark/tests/master/suits.html: Renamed from PerformanceTests/MotionMark/tests/master/svg-particles.html.
1277         * MotionMark/tests/svg/suits.html:
1278
1279 2018-06-08  Jon Lee  <jonlee@apple.com>
1280
1281         Add sub-tests based on Suits
1282         https://bugs.webkit.org/show_bug.cgi?id=186260
1283
1284         Reviewed by Said Abou-Hallawa.
1285
1286         Add a new developer Suits suite with sub-tests that isolate parts of each particle.
1287         The sub-tests are:
1288             - Particles using only clip paths
1289             - Particles using only shapes
1290             - Particles that have no gradients
1291             - Particles that have no rotation
1292             - Particles that do not move around at all (but all of the physics calculations are
1293             still performed)
1294
1295         * MotionMark/resources/debug-runner/tests.js:
1296         * MotionMark/tests/master/resources/svg-particles.js: Rename the stage, particle, and
1297         benchmark to "Suits" from "SVG". Change SuitsParticle so that we guarantee 50% clip and
1298         shape paths rather than relying on random chance. Wrap a check around creation of the
1299         gradient element, and use a simple fill color when we aren't using a gradient.
1300         * MotionMark/tests/svg/suits.html: Added.
1301         * MotionMark/tests/svg/suits.js: Added.
1302         Look for the query string and set the particle based on what is selected.
1303
1304 2018-06-06  Jon Lee  <jonlee@apple.com>
1305
1306         Allow the ramp controller to run tests that take less time than the initial ramp-up phase
1307         https://bugs.webkit.org/show_bug.cgi?id=186257
1308
1309         Reviewed by Said Abou-Hallawa.
1310
1311         * MotionMark/tests/resources/main.js: Extend the desired end timestamp when ramp-up occurs.
1312         Previously the controller assumed that the duration of the test was always longer than the
1313         time it took to figure out the order of magnitude of particles to render. When that range
1314         is determined, the end timestamp is then extended. If the duration is too short, the controller
1315         would throw an exception. The fix is to continue extending the timestamp every time the order of
1316         magnitude tiers up.
1317
1318         In general testing with really short durations isn't practical, but sometimes it is
1319         desired for debugging.
1320
1321         To avoid confusion of the private variables, I renamed Benchmark._startTimestamp to
1322         Benchmark._benchmarkStartTimestamp, to disambiguate from Controller._startTimestamp.
1323         Benchmark._startTimestamp is set once and never changed, after the initial 100ms warm-up.
1324         Controller._startTimestamp is initialized to 0 (which means "not initialized"), and then after
1325         the warm up is complete, set in Controller.start() to the same wall clock timestamp, and never
1326         changed afterwards.
1327
1328 2018-06-06  Jon Lee  <jonlee@apple.com>
1329
1330         Remove unneeded data processing in MotionMark
1331         https://bugs.webkit.org/show_bug.cgi?id=186256
1332
1333         Reviewed by Said Abou-Hallawa.
1334
1335         Remove calculation of average frame rate based on multiple samples at a given complexity.
1336         It is not used in the calculation of the score, and is just extra work to do at the end
1337         of a test.
1338
1339         * MotionMark/developer.html: Remove options to show average data and regression based on
1340         that data.
1341         * MotionMark/resources/debug-runner/graph.js: Remove data. Update the time graph that
1342         shows the relationship between the break point in the regression against the complexity
1343         graph.
1344         * MotionMark/resources/debug-runner/motionmark.css: Make the confidence intervals easier
1345         to see.
1346         * MotionMark/resources/runner/motionmark.js: Simplify score calculation now that we are
1347         only calculating regressions on the raw data. We were calculating it based on the averages,
1348         but that was just for analysis, and not used in the actual score. Get rid of the
1349         |complexitySamples| local variable.
1350         * MotionMark/resources/strings.js: Remove unneeded string constant.
1351         * MotionMark/tests/resources/main.js:
1352         (processSamples): Refactor so that more of it can be shared between Controller and
1353         RampController. It also avoids processing the raw samples (controllerSamples) twice for
1354         the ramp controller.
1355         (_processComplexitySamples): Remove logic to calculate averages based on complexity.
1356         (_processMarks): Split this out from the old processSamples.
1357         (_processControllerSamples): Split this out from the old processSamples.
1358
1359 2018-06-06  Jon Lee  <jonlee@apple.com>
1360
1361         Strip out more characters when creating permalinks
1362         https://bugs.webkit.org/show_bug.cgi?id=186259
1363
1364         Reviewed by Said Abou-Hallawa.
1365
1366         Tests and suites may have names that make permalinks to the tests less readable when
1367         shared. Expand the set of characters that are stripped out of the names.
1368
1369         * MotionMark/resources/debug-runner/motionmark.js: Refactor to call
1370         stripUnwantedCharactersForURL. For existing links, decode the suite and test names.
1371         * MotionMark/resources/extensions.js:
1372         (Utilities.stripUnwantedCharactersForURL): Rename from stripNonASCIICharacters.
1373         Remove any non-alphanumeric character.
1374
1375 2018-06-04  Jon Lee  <jonlee@apple.com>
1376
1377         Remove unnecessary MotionMark controllers
1378         https://bugs.webkit.org/show_bug.cgi?id=186255
1379
1380         Reviewed by Said Abou-Hallawa.
1381
1382         Get rid of the ramp30 and the fixed-with-a-step controllers, which aren't used for testing
1383         or debugging.
1384
1385         * MotionMark/developer.html: Move the "ramp" option as first, since that is the default
1386         controller for the benchmark.
1387         * MotionMark/resources/debug-runner/motionmark.js:
1388         * MotionMark/resources/runner/motionmark.js: Added missing copyright notice.
1389         * MotionMark/tests/resources/main.js:
1390
1391 2018-05-25  Myles C. Maxfield  <mmaxfield@apple.com>
1392
1393         Improve the performance of Font::canRenderCombiningCharacterSequence()
1394         https://bugs.webkit.org/show_bug.cgi?id=185933
1395
1396         Reviewed by Ryosuke Niwa.
1397
1398         * Layout/ComplexLongUnique.html: Added.
1399
1400 2018-05-25  Saam Barati  <sbarati@apple.com>
1401
1402         Have a memory test where we can validate JSCs mini memory mode
1403         https://bugs.webkit.org/show_bug.cgi?id=185932
1404
1405         Reviewed by Mark Lam.
1406
1407         We add a directory here with the contents of the testmem benchmark.
1408         To run it, use `Tools/Scripts/run-testmem`. To add new tests in the future,
1409         you just need to add JS files to this directory.
1410
1411         * testmem: Added.
1412         * testmem/air.js: Added.
1413         * testmem/base64.js: Added.
1414         * testmem/basic.js: Added.
1415         * testmem/box2d.js: Added.
1416         * testmem/crypto-md5.js: Added.
1417         * testmem/date-format-tofte.js: Added.
1418         * testmem/earley-boyer.js: Added.
1419         * testmem/hash-map.js: Added.
1420         * testmem/regex-dna.js: Added.
1421         * testmem/splay.js: Added.
1422         * testmem/tagcloud.js: Added.
1423
1424 2018-05-22  Ryan Haddad  <ryanhaddad@apple.com>
1425
1426         Unreviewed, rolling out r232052.
1427
1428         Breaks internal builds.
1429
1430         Reverted changeset:
1431
1432         "Use more C++17"
1433         https://bugs.webkit.org/show_bug.cgi?id=185176
1434         https://trac.webkit.org/changeset/232052
1435
1436 2018-05-21  Yusuke Suzuki  <utatane.tea@gmail.com>
1437
1438         Use more C++17
1439         https://bugs.webkit.org/show_bug.cgi?id=185176
1440
1441         Reviewed by JF Bastien.
1442
1443         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
1444         * MediaTime/Configurations/Base.xcconfig:
1445         * MediaTime/MediaTime.xcodeproj/project.pbxproj:
1446
1447 2018-04-30  Myles C. Maxfield  <mmaxfield@apple.com>
1448
1449         Improve the performance of FontCascadeDescription's effectiveFamilies
1450         https://bugs.webkit.org/show_bug.cgi?id=184720
1451
1452         Reviewed by Simon Fraser.
1453
1454         This performance test calls SystemFontDatabase::systemFontCascadeList() around 2,000,000 times (before
1455         this patch is applied), which is roughly equivalent to the page we found the performance problem on.
1456         The calling pattern is roughly equivalent in this test.
1457
1458         * Layout/system-ui.html: Added.
1459
1460 2018-03-08  Antti Koivisto  <antti@apple.com>
1461
1462         Update StyleBench version number in page title to 0.3
1463
1464         * StyleBench/index.html:
1465
1466         There has been a bunch of changes, most notably the addition of attribute selector test step.
1467
1468 2018-02-16  Dean Jackson  <dino@apple.com>
1469
1470         Use OPENGL macros to be more clear about which OpenGL/ES WebGL uses on Cocoa
1471         https://bugs.webkit.org/show_bug.cgi?id=182894
1472
1473         Reviewed by Tim Horton.
1474
1475         Rename OPENGL_ES_2 to OPENGL_ES.
1476
1477         * StitchMarker/wtf/Platform.h:
1478
1479 2018-02-09  Ryosuke Niwa  <rniwa@webkit.org>
1480
1481         Make run-perf-tests work with Speedometer 2.0 and re-enable the test
1482         https://bugs.webkit.org/show_bug.cgi?id=182088
1483
1484         Reviewed by Antti Koivisto.
1485
1486         This patch updates the test harness to report new metrics of Speedometer 2.0.
1487
1488         Replaced the total time by the geometric mean of the time spent in each suite, and added scores
1489         as a metric in addition to time. Also, updated the PerfTestRunner harness and the results template
1490         to support "Score" metric with "pt" as its unit.
1491
1492         * Skipped: Unskipped the test.
1493         * Speedometer/resources/benchmark-report.js:
1494         (window.benchmarkClient.willStartFirstIteration..createTest): Added the unit as an optional argument
1495         to report the score in addition to time. Also report the geomean of time instead of the total time.
1496         (window.benchmarkClient.willStartFirstIteration): Report the geomean of time.
1497         (window.benchmarkClient.didRunSuites): Record the entire measuredValues instead of just tests.
1498         (window.benchmarkClient.didFinishLastIteration): Report the scores.
1499
1500 2018-02-09  Ross Kirsling  <ross.kirsling@sony.com>
1501
1502         Use REQUEST_ANIMATION_FRAME_DISPLAY_MONITOR on any non-Windows port.
1503         https://bugs.webkit.org/show_bug.cgi?id=182623
1504
1505         Reviewed by Alex Christensen.
1506
1507         * StitchMarker/wtf/Platform.h:
1508         Simplify #if.
1509
1510 2018-02-01  Antti Koivisto  <antti@apple.com>
1511
1512         StyleBench: Attribute selectors and other improvements
1513         https://bugs.webkit.org/show_bug.cgi?id=182387
1514
1515         Reviewed by Joseph Pecoraro.
1516
1517         - Add some attributes to elements in all tests
1518         - Add some attribute selectors to stylesheets in all tests
1519         - Also add some * selectors to all stylesheets.
1520         - Add attribute mutation step to all suites
1521         - Make test steps do more mutations (25->100) and reduce the number of steps to keep testing time in check.
1522           Too fast steps were running into timer resolution limits.
1523
1524         * StyleBench/resources/style-bench.js:
1525         (defaultConfiguration):
1526         (prototype.randomAttributeName):
1527         (prototype.randomAttributeValue):
1528         (prototype.randomAttributeSelector):
1529         (prototype.makeCompoundSelector):
1530         (prototype.makeElement):
1531         (prototype.addClasses):
1532         (prototype.removeClasses):
1533         (prototype.mutateAttributes):
1534         (prototype.async.runForever):
1535         * StyleBench/resources/tests.js:
1536         (makeSteps):
1537         (makeSuite):
1538
1539 2018-02-01  Geoffrey Garen  <ggaren@apple.com>
1540
1541         Make MallocBench easier for non-WebKit engineers to run
1542         https://bugs.webkit.org/show_bug.cgi?id=182415
1543
1544         Reviewed by Saam Barati.
1545
1546         * MallocBench/MallocBench.xcodeproj/project.pbxproj: Use c++14 so we
1547         can make_unique.
1548
1549         Specify that we support all Darwin platforms so you can test them.
1550
1551         * MallocBench/run-malloc-benchmarks: Specify the path to MallocBench
1552         and libmbmalloc explicitly, rather than computing them implicitly
1553         using webkitdirs. Non-WebKit folks don't have the directory structure
1554         required by webkitdirs.
1555
1556         Remove Linux-specific and cmake-specific behaviors because we only
1557         needed them in the world of implicit path computation.
1558
1559 2018-01-31  Ryosuke Niwa  <rniwa@webkit.org>
1560
1561         Make run-perf-tests work with StyleBench and re-enable it
1562         https://bugs.webkit.org/show_bug.cgi?id=182322
1563
1564         Reviewed by Antti Koivisto.
1565
1566         This patch updates the test harness to report new metrics of StyleBench.
1567
1568         Replaced the total time by the geometric mean of the time spent in each suite, and added scores
1569         as a metric in addition to time. Also, updated the PerfTestRunner harness and the results template
1570         to support "Score" metric with "pt" as its unit.
1571
1572         * Skipped: Unskipped the test.
1573         * StyleBench/resources/benchmark-report.js:
1574         (window.benchmarkClient.willStartFirstIteration..createTest): Added the unit as an optional argument
1575         to report the score in addition to time. Also report the geomean of time instead of the total time.
1576         (window.benchmarkClient.willStartFirstIteration): Report the geomean of time.
1577         (window.benchmarkClient.didRunSuites): Record the entire measuredValues instead of just tests.
1578         (window.benchmarkClient.didFinishLastIteration): Report the scores.
1579         * resources/results-template.html: Added "Score" as a metric and "pt" as its unit.
1580         * resources/runner.js:
1581         (PerfTestRunner..finish): Ditto.
1582
1583 2018-01-30  Ryosuke Niwa  <rniwa@webkit.org>
1584
1585         StyleBench: Generate more mutation steps
1586         https://bugs.webkit.org/show_bug.cgi?id=182291
1587
1588         Rubber-stamped by Antti Koivisto
1589
1590         Use zero-based index instead of one-based index.
1591
1592         * StyleBench/resources/tests.js:
1593         (makeSteps):
1594
1595 2018-01-30  Antti Koivisto  <antti@apple.com>
1596
1597         StyleBench: Generate more mutation steps
1598         https://bugs.webkit.org/show_bug.cgi?id=182291
1599
1600         Reviewed by Ryosuke Niwa.
1601
1602         The test should be doing multiple rounds of mutations in a single document to test what is supposed to test (like it did before r227756).
1603         This also makes test results more stable and less dependent on possible first-time costs.
1604
1605         * StyleBench/resources/benchmark-runner.js:
1606
1607         Adjust correctionFactor to match the new steps count. With 1.5 Firefox is ~72rpm.
1608
1609         * StyleBench/resources/tests.js:
1610         (makeSteps):
1611
1612         Generate 10 rounds of class/element mutation steps.
1613         Give each generated step unique name so reporting works.
1614         Make individual steps do fewer mutations.
1615
1616 2018-01-29  Antti Koivisto  <antti@apple.com>
1617
1618         StyleBench: Remove : from a test name
1619         https://bugs.webkit.org/show_bug.cgi?id=182258
1620
1621         Reviewed by Ryosuke Niwa.
1622
1623         It breaks reporting.
1624
1625         * StyleBench/resources/style-bench.js:
1626         (nthPseudoClassConfiguration):
1627
1628 2018-01-29  Ryosuke Niwa  <rniwa@webkit.org>
1629
1630         Temporarily skip StyleBench until webkit.org/b/182088 is fixed.
1631
1632         * Skipped:
1633
1634 2018-01-29  Ryosuke Niwa  <rniwa@webkit.org>
1635
1636         Make StyleBench compatible with run-benchmark and run-perf-tests
1637         https://bugs.webkit.org/show_bug.cgi?id=182262
1638
1639         Reviewed by Antti Koivisto.
1640
1641         Copied resource files referenced from Speedometer directory since run-benchmark needs to be able
1642         to checkout each benchmark separately.
1643
1644         Removed the code to create tests of the same name five times in makeSteps since this
1645         won't be compatible with either run-benchmark or run-perf-tests.
1646
1647         * StyleBench/index.html: Removed the code to show warnings for local files since run-benchmark
1648         doesn't use HTTP server in WebDriver mode.
1649         * StyleBench/resources/benchmark-report.js: Copied from resources/benchmark-report.js.
1650         * StyleBench/resources/benchmark-runner.js: Copied from resources/benchmark-runner.js.
1651         (BenchmarkRunner.prototype._finalize): Use the correction factor of 8 instead of 5 in StyleBench.
1652         * StyleBench/resources/gauge.png: Copied from resources/gauge.png.
1653         * StyleBench/resources/gauge@2x.png: Copied from resources/gauge@2x.png.
1654         * StyleBench/resources/main.css: Copied from resources/main.css.
1655         * StyleBench/resources/main.js: Copied from resources/main.js.
1656         * StyleBench/resources/tests.js:
1657         (makeSteps): Only make each test once.
1658
1659 2018-01-26  Chris Nardi  <cnardi@chromium.org>
1660
1661         Addressing post-review comments after r226614
1662         https://bugs.webkit.org/show_bug.cgi?id=182151
1663
1664         Reviewed by Myles C. Maxfield.
1665
1666         * StitchMarker/wtf/text/StringImpl.h:
1667         (WTF::isSpaceOrNewline):
1668         * StitchMarker/wtf/text/TextBreakIterator.cpp:
1669         (WTF::numCodeUnitsInGraphemeClusters):
1670         * StitchMarker/wtf/text/TextBreakIterator.h:
1671
1672 2018-01-22  Antti Koivisto  <antti@apple.com>
1673
1674         StyleBench: Separate test for :nth pseudo classes
1675         https://bugs.webkit.org/show_bug.cgi?id=181932
1676
1677         Reviewed by Sam Weinig.
1678
1679         Optimization strategies for nth are different from other structural pseudo classes (:first-child etc). Move them to a separate test.
1680
1681         * StyleBench/resources/style-bench.js:
1682         (structuralPseudoClassConfiguration):
1683         (nthPseudoClassConfiguration):
1684
1685             Add a new configuration.
1686
1687         (predefinedConfigurations):
1688         (pseudoClassConfiguration): Deleted.
1689
1690 2018-01-17  Ryosuke Niwa  <rniwa@webkit.org>
1691
1692         Temporarily skip Speedometer until we fix the metrics.
1693
1694         * Skipped:
1695
1696 2018-01-16  Alex Christensen  <achristensen@webkit.org>
1697
1698         Fix Mac CMake build after r222900
1699
1700         * MallocBench/MallocBench/CMakeLists.txt:
1701         The mbmalloc target doesn't exist if we're not in developer mode, so only use it if we are.
1702
1703 2018-01-09  Ryosuke Niwa  <rniwa@webkit.org>
1704
1705         Speedometer 2.0: Update the About page.
1706         https://bugs.webkit.org/show_bug.cgi?id=175918
1707
1708         Reviewed by Keith Miller.
1709
1710         Updated the about page for Speedometer 2.0.
1711
1712         * Speedometer/index.html:
1713
1714 == Rolled over to ChangeLog-2018-01-01 ==