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