71e6bf56c819f1f1f628ced70b983ab2f90d05ab
[WebKit-https.git] / PerformanceTests / ChangeLog
1 2020-03-15  Yusuke Suzuki  <ysuzuki@apple.com>
2
3         Should not use variable-length-array (VLA)
4         https://bugs.webkit.org/show_bug.cgi?id=209043
5
6         Reviewed by Mark Lam.
7
8         * MediaTime/Configurations/Base.xcconfig:
9
10 2020-02-17  Don Olmstead  <don.olmstead@sony.com>
11
12         [CMake] Use builtin targets
13         https://bugs.webkit.org/show_bug.cgi?id=205166
14
15         Reviewed by Darin Adler.
16
17         * MallocBench/MallocBench/CMakeLists.txt:
18
19 2020-01-28  Sihui Liu  <sihui_liu@apple.com>
20
21         IndexedDB: speed up index records deletion
22         https://bugs.webkit.org/show_bug.cgi?id=206196
23
24         Reviewed by Brady Eidson.
25
26         * IndexedDB/basic/index-cursor-delete-2.html: Added.
27
28 2020-01-08  Sihui Liu  <sihui_liu@apple.com>
29
30         REGRESSION (r242911?): High Sierra Release WK2 Perf bot timing out while running IndexedDB/large-number-of-inserts.html
31         https://bugs.webkit.org/show_bug.cgi?id=195952
32         <rdar://problem/49137688>
33
34         Re-enable the test after recent IDB performance improvement.
35
36         Reviewed by Alexey Proskuryakov.
37
38         * Skipped:
39
40 2020-01-07  Carlos Garcia Campos  <cgarcia@igalia.com>
41
42         [GTK][WPE] Add API to set purpose and hints of active editable element to input methods
43         https://bugs.webkit.org/show_bug.cgi?id=205605
44
45         Reviewed by Žan Doberšek.
46
47         Split ENABLE_IOS_AUTOCORRECT_AND_AUTOCAPITALIZE and remove the IOS prefix.
48
49         * StitchMarker/wtf/FeatureDefines.h:
50
51 2020-01-06  Mark Lam  <mark.lam@apple.com>
52
53         Convert ASSERT_DISABLED to ASSERT_ENABLED, and fix some tests of NDEBUG that should actually test for ASSERT_ENABLED.
54         https://bugs.webkit.org/show_bug.cgi?id=205776
55
56         Reviewed by Saam Barati.
57
58         * JetStream2/wasm/HashSet.cpp:
59         * StitchMarker/wtf/Assertions.h:
60         * StitchMarker/wtf/DateMath.cpp:
61         (WTF::initializeDates):
62         * StitchMarker/wtf/HashTable.h:
63         * StitchMarker/wtf/Hasher.h:
64         (WTF::StringHasher::addCharacters):
65         * StitchMarker/wtf/NeverDestroyed.h:
66         (WTF::LazyNeverDestroyed::construct):
67         * StitchMarker/wtf/StackBounds.h:
68         (WTF::StackBounds::checkConsistency const):
69         * StitchMarker/wtf/ValueCheck.h:
70         * StitchMarker/wtf/Vector.h:
71         (WTF::minCapacity>::checkConsistency):
72         * StitchMarker/wtf/text/AtomicStringImpl.cpp:
73         * StitchMarker/wtf/text/AtomicStringImpl.h:
74         * StitchMarker/wtf/text/StringCommon.h:
75         (WTF::hasPrefixWithLettersIgnoringASCIICaseCommon):
76         * StitchMarker/wtf/text/StringImpl.h:
77         * StitchMarker/wtf/text/SymbolImpl.h:
78         * StitchMarker/wtf/text/UniquedStringImpl.h:
79
80 2020-01-05  Dean Jackson  <dino@apple.com>
81
82         Rename GraphicsContext3D to GraphicsContextGL
83         https://bugs.webkit.org/show_bug.cgi?id=205778
84         <rdar://problem/58327597>
85
86         Reviewed by Sam Weinig.
87
88         Rename all the GraphicsContext3D things to GraphicsContextGL
89         (includes Extensions3D and GRAPHICS_CONTEXT_3D).
90
91         GraphicsContext3DBase now becomes GraphicsContextGL.
92         GraphicsContext3D is now GraphicsContextGLOpenGL (since it represents
93         the OpenGL implementation of the API).
94
95         * StitchMarker/wtf/FeatureDefines.h:
96         * StitchMarker/wtf/Platform.h:
97
98 2019-12-16  Antti Koivisto  <antti@apple.com>
99
100         Add StyleBench subtest for dynamic media query performance
101         https://bugs.webkit.org/show_bug.cgi?id=205263
102
103         Reviewed by Zalan Bujtas.
104
105         Add a subtest that contains a small number of rules inside min/max-width media queries.
106         The test is executed by resizing the test frame to various widths.
107
108         * StyleBench/index.html:
109         * StyleBench/resources/style-bench.js:
110         (Random.prototype.chance):
111         (defaultConfiguration):
112         (mediaQueryConfiguration):
113         (predefinedConfigurations):
114         * StyleBench/resources/tests.js:
115         (makeSteps):
116
117 2019-10-26  Chris Lord  <clord@igalia.com>
118
119         Put OffscreenCanvas behind a build flag
120         https://bugs.webkit.org/show_bug.cgi?id=203146
121
122         Reviewed by Ryosuke Niwa.
123
124         * StitchMarker/wtf/FeatureDefines.h:
125
126 2019-10-21  Carlos Alberto Lopez Perez  <clopez@igalia.com>
127
128         [GTK] Perf test SVG/UnderTheSeeBenchmark.html timeouts.
129         https://bugs.webkit.org/show_bug.cgi?id=203229
130
131         Unreviewed gardening.
132
133         * Skipped: Mark 2 new tests timing out in the GTK perf bot.
134
135 2019-10-17  Saam Barati  <sbarati@apple.com>
136
137         Have a CLI version of JetStream 2 to run on watchOS
138         https://bugs.webkit.org/show_bug.cgi?id=202998
139         <rdar://problem/56208554>
140
141         Reviewed by Tadeu Zagallo.
142
143         This patch adds a CLI version of JS2 to run on watchOS. We run most subtests
144         from the browser version of JS2, but skip a few tests that are particularly
145         long running, like WSL and some of WTB. We also don't run the Wasm tests
146         as we don't JIT on watchOS. Each test runs for fewer iterations in the watch
147         version. 15 is the default iteration count. This benchmark runs in 7 minutes on
148         a Series 4 watch.
149
150         * JetStream2/JetStreamDriver.js:
151         (getIterationCount):
152         (getWorstCaseCount):
153         (Driver.prototype.async.start):
154         (Driver.prototype.resultsJSON):
155         (Driver.prototype.dumpJSONResultsIfNeeded):
156         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
157         (DefaultBenchmark):
158         * JetStream2/RexBench/UniPoker/benchmark.js:
159         (Benchmark.prototype.validate):
160         (Benchmark):
161         * JetStream2/watch-cli.js: Added.
162
163 2019-10-17  Adrian Perez de Castro  <aperez@igalia.com>
164
165         [Linux] Avoid usage of <sys/sysctl.h> in MallocBench
166         https://bugs.webkit.org/show_bug.cgi?id=203109
167
168         Reviewed by Carlos Alberto Lopez Perez.
169
170         * MallocBench/MallocBench/CPUCount.cpp: Only include <sys/sysctl.h> if
171         the sysctl() function will be used.
172
173 2019-10-17  Carlos Alberto Lopez Perez  <clopez@igalia.com>
174
175         [GTK] Perf test IndexedDB/basic/index-cursor-delete.html timeouts
176         https://bugs.webkit.org/show_bug.cgi?id=203083
177
178         Unreviewed gardening.
179
180         * Skipped: Skip a test that is timing out on the GTK perf bot.
181
182 2019-09-24  Caio Lima  <ticaiolima@gmail.com>
183
184         [BigInt] Add ValueBitRShift into DFG
185         https://bugs.webkit.org/show_bug.cgi?id=192663
186
187         Reviewed by Robin Morisset.
188
189         * BigIntBench/big-int-simple-rshift.js: Added.
190         (bigInt):
191
192 2019-09-09  Jon Lee  <jonlee@apple.com>
193
194         Upload triangles tests for 3D suite in MotionMark
195         https://bugs.webkit.org/show_bug.cgi?id=201606
196
197         Reviewed by Simon Fraser.
198
199         * MotionMark/resources/debug-runner/motionmark.css:
200         * MotionMark/resources/debug-runner/tests.js:
201         * MotionMark/tests/3d/resources/webgl.js:
202         * MotionMark/tests/3d/resources/webgpu.js: Added.
203         * MotionMark/tests/3d/triangles-webgl.html: Renamed from Websites/browserbench.org/MotionMark1.1/tests/3d/webgl.html.
204         * MotionMark/tests/3d/triangles-webgpu.html: Copied from PerformanceTests/MotionMark/tests/3d/webgl.html.
205
206 2019-08-29  Keith Rollin  <krollin@apple.com>
207
208         Update .xcconfig symbols to reflect the current set of past and future product versions.
209         https://bugs.webkit.org/show_bug.cgi?id=200720
210         <rdar://problem/54305032>
211
212         Reviewed by Alex Christensen.
213
214         Remove version symbols related to old OS's we no longer support,
215         ensure that version symbols are defined for OS's we do support.
216
217         * MediaTime/Configurations/Base.xcconfig:
218         * MediaTime/Configurations/DebugRelease.xcconfig:
219
220 2019-08-21  Sihui Liu  <sihui_liu@apple.com>
221
222         Add some IDB performance tests
223         https://bugs.webkit.org/show_bug.cgi?id=200906
224
225         Reviewed by Geoffrey Garen.
226
227         Add test coverage for IDB performance, and organize IDB tests into two categories.
228
229         * IndexedDB/basic/index-count-key.html: Added.
230         * IndexedDB/basic/index-count.html: Added.
231         * IndexedDB/basic/index-cursor-advance.html: Added.
232         * IndexedDB/basic/index-cursor-continue.html: Added.
233         * IndexedDB/basic/index-cursor-continuePrimaryKey.html: Added.
234         * IndexedDB/basic/index-cursor-delete.html: Added.
235         * IndexedDB/basic/index-cursor-update.html: Added.
236         * IndexedDB/basic/index-get.html: Added.
237         * IndexedDB/basic/index-multientry.html: Renamed from PerformanceTests/IndexedDB/index-multientry.html.
238         * IndexedDB/basic/objectstore-add.html: Added.
239         * IndexedDB/basic/objectstore-clear.html: Added.
240         * IndexedDB/basic/objectstore-count.html: Added.
241         * IndexedDB/basic/objectstore-cursor-advance.html: Added.
242         * IndexedDB/basic/objectstore-cursor-continue.html: Added.
243         * IndexedDB/basic/objectstore-cursor-delete.html: Added.
244         * IndexedDB/basic/objectstore-cursor-update.html: Added.
245         * IndexedDB/basic/objectstore-cursor.html: Renamed from PerformanceTests/IndexedDB/objectstore-cursor.html.
246         * IndexedDB/basic/objectstore-delete.html: Renamed from PerformanceTests/IndexedDB/index-get.html.
247         * IndexedDB/basic/objectstore-get.html: Added.
248         * IndexedDB/basic/objectstore-put.html: Renamed from PerformanceTests/IndexedDB/objectstore-get.html.
249         * IndexedDB/stress/large-array-keys.html: Renamed from PerformanceTests/IndexedDB/large-array-keys.html.
250         * IndexedDB/stress/large-binary-keys.html: Renamed from PerformanceTests/IndexedDB/large-binary-keys.html.
251         * IndexedDB/stress/large-number-of-inserts-responsiveness.html: Renamed from PerformanceTests/IndexedDB/large-number-of-inserts-responsiveness.html.
252         * IndexedDB/stress/large-number-of-inserts.html: Renamed from PerformanceTests/IndexedDB/large-number-of-inserts.html.
253         * IndexedDB/stress/large-string-keys.html: Renamed from PerformanceTests/IndexedDB/large-string-keys.html.
254         * Skipped:
255
256 2019-08-19  Sihui Liu  <sihui_liu@apple.com>
257
258         Remove IDB-specific quota
259         https://bugs.webkit.org/show_bug.cgi?id=196545
260         <rdar://problem/54201783>
261
262         Reviewed by Youenn Fablet.
263
264         * IndexedDB/large-binary-keys.html:
265         * IndexedDB/large-string-keys.html:
266
267 2019-08-14  Keith Rollin  <krollin@apple.com>
268
269         Remove support for macOS < 10.13
270         https://bugs.webkit.org/show_bug.cgi?id=200694
271         <rdar://problem/54278851>
272
273         Reviewed by Youenn Fablet.
274
275         Update conditionals that reference __MAC_OS_X_VERSION_MIN_REQUIRED and
276         __MAC_OS_X_VERSION_MAX_ALLOWED, assuming that they both have values >=
277         101300. This means that expressions like
278         "__MAC_OS_X_VERSION_MIN_REQUIRED < 101300" are always False and
279         "__MAC_OS_X_VERSION_MIN_REQUIRED >= 101300" are always True.
280
281         * StitchMarker/wtf/Platform.h:
282         * StitchMarker/wtf/dependencies/bmalloc/BPlatform.h:
283         * StitchMarker/wtf/text/icu/TextBreakIteratorICU.h:
284         (WTF::TextBreakIteratorICU::TextBreakIteratorICU):
285         (WTF::caretRules): Deleted.
286
287 2019-08-06  Michael Saboff  <msaboff@apple.com>
288
289         Make JSC memory benchmark available on OpenSource
290         https://bugs.webkit.org/show_bug.cgi?id=195012
291
292         Reviewed by Saam Barati.
293
294         This patch upstreams the RAMification memory benchmark to open source.
295
296         The prior RAMification.py driver script in the Internal directory has been moved here
297         with the device execution code seperated out to a new Internal driver script.
298         As a result, command line argument parsing as well as acting on the resulting argument
299         values has changed.  The remote device Runner creates an ArgumentParser object that is
300         passed through main().  In main(), if we have an ArgumentParser, we use it otherwise we
301         create one for benchmarking locally.  The arguments object returned from argument
302         parsing is now passed to the Runner constructor so it can extract the settings appropriate
303         for that particular runner.
304
305         Added some sane argument default values for the JSC command patch and the JetStream2
306         source directory.
307
308         Updated the Lua tests to eliminate the testing harness and its overhead, matching what was
309         done in the Internal tree.
310
311         * JetStream2/LuaJSFight/hello_world.js:
312         (Benchmark.prototype.runIteration): Deleted.
313         (Benchmark): Deleted.
314         * JetStream2/LuaJSFight/list_search.js:
315         (run): Deleted.
316         (Benchmark.prototype.runIteration): Deleted.
317         (Benchmark): Deleted.
318         * JetStream2/LuaJSFight/lists.js:
319         (): Deleted.
320         (Benchmark.prototype.runIteration): Deleted.
321         (Benchmark): Deleted.
322         * JetStream2/LuaJSFight/richards.js: Added.
323         (runRichards):
324         (Scheduler):
325         (Scheduler.prototype.addIdleTask):
326         (Scheduler.prototype.addWorkerTask):
327         (Scheduler.prototype.addHandlerTask):
328         (Scheduler.prototype.addDeviceTask):
329         (Scheduler.prototype.addRunningTask):
330         (Scheduler.prototype.addTask):
331         (Scheduler.prototype.schedule):
332         (Scheduler.prototype.release):
333         (Scheduler.prototype.holdCurrent):
334         (Scheduler.prototype.suspendCurrent):
335         (Scheduler.prototype.queue):
336         (TaskControlBlock):
337         (TaskControlBlock.prototype.setRunning):
338         (TaskControlBlock.prototype.markAsNotHeld):
339         (TaskControlBlock.prototype.markAsHeld):
340         (TaskControlBlock.prototype.isHeldOrSuspended):
341         (TaskControlBlock.prototype.markAsSuspended):
342         (TaskControlBlock.prototype.markAsRunnable):
343         (TaskControlBlock.prototype.run):
344         (TaskControlBlock.prototype.checkPriorityAdd):
345         (TaskControlBlock.prototype.toString):
346         (IdleTask):
347         (IdleTask.prototype.run):
348         (IdleTask.prototype.toString):
349         (DeviceTask):
350         (DeviceTask.prototype.run):
351         (DeviceTask.prototype.toString):
352         (WorkerTask):
353         (WorkerTask.prototype.run):
354         (WorkerTask.prototype.toString):
355         (HandlerTask):
356         (HandlerTask.prototype.run):
357         (HandlerTask.prototype.toString):
358         (Packet):
359         (Packet.prototype.addTo):
360         (Packet.prototype.toString):
361         * JetStream2/LuaJSFight/string_lists.js:
362         (): Deleted.
363         (Benchmark.prototype.runIteration): Deleted.
364         (Benchmark): Deleted.
365         * JetStream2/RAMification.py: Added.
366         (mean):
367         (geomean):
368         (frameworkPathFromExecutablePath):
369         (parseArgs):
370         (parseArgs.optStrToBool):
371         (BaseRunner):
372         (BaseRunner.__init__):
373         (BaseRunner.setup):
374         (BaseRunner.setEnv):
375         (BaseRunner.unsetEnv):
376         (BaseRunner.resetForTest):
377         (BaseRunner.processLine):
378         (BaseRunner.setReturnCode):
379         (BaseRunner.getResults):
380         (LocalRunner):
381         (LocalRunner.__init__):
382         (LocalRunner.runOneTest):
383         (main):
384         (main.runTestList):
385         * JetStream2/__init__.py: Added.
386
387 2019-07-23  Keith Rollin  <krollin@apple.com>
388
389         Remove rest of NavigatorContentUtils support
390         https://bugs.webkit.org/show_bug.cgi?id=200052
391         <rdar://problem/53467357>
392
393         Reviewed by Alex Christensen.
394
395         Bug 196070 removes most of the support for the never-used
396         NavigatorContentUtils facility. However, there were still a couple of
397         left-over references after that change. This patch sweeps those up.
398
399         * StitchMarker/wtf/FeatureDefines.h:
400
401 2019-07-12  Caio Lima  <ticaiolima@gmail.com>
402
403         [BigInt] Add ValueBitLShift into DFG
404         https://bugs.webkit.org/show_bug.cgi?id=192664
405
406         Reviewed by Saam Barati.
407
408         * BigIntBench/big-int-simple-lshift.js: Added.
409
410 2019-06-28  Konstantin Tokarev  <annulen@yandex.ru>
411
412         Remove traces of ENABLE_ICONDATABASE remaining after its removal in 219733
413         https://bugs.webkit.org/show_bug.cgi?id=199317
414
415         Reviewed by Michael Catanzaro.
416
417         While IconDatabase and all code using it was removed,
418         ENABLE_ICONDATABASE still exists as build option and C++ macro.
419
420         * StitchMarker/wtf/FeatureDefines.h:
421
422 2019-06-27  Beth Dakin  <bdakin@apple.com>
423
424         Upstream use of MACCATALYST
425         https://bugs.webkit.org/show_bug.cgi?id=199245
426         rdar://problem/51687723
427
428         Reviewed by Tim Horton.
429
430         * MediaTime/Configurations/SDKVariant.xcconfig:
431
432 2019-06-20  Carlos Garcia Campos  <cgarcia@igalia.com>
433
434         [GTK] Remove support for GTK2 plugins
435         https://bugs.webkit.org/show_bug.cgi?id=199065
436
437         Reviewed by Sergio Villar Senin.
438
439         * StitchMarker/wtf/Platform.h:
440
441 2019-05-08  Caio Lima  <ticaiolima@gmail.com>
442
443         [BigInt] Add ValueMod into DFG
444         https://bugs.webkit.org/show_bug.cgi?id=186174
445
446         Reviewed by Saam Barati.
447
448         * BigIntBench/big-int-simple-mod.js: Added.
449
450 2019-05-06  Saam Barati  <sbarati@apple.com>
451
452         JS2 should print scores for different categories
453         https://bugs.webkit.org/show_bug.cgi?id=197605
454
455         Reviewed by Yusuke Suzuki.
456
457         * JetStream2/JetStreamDriver.js:
458         (displayCategoryScores):
459         (isInBrowser.document.onkeydown):
460         (Driver.prototype.async.start):
461         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
462
463 2019-04-16  Jon Lee  <jonlee@apple.com>
464
465         [MotionMark] The text "kkkj" is causing Firefox console warning: `unreachable code after return statement`
466         https://bugs.webkit.org/show_bug.cgi?id=196814
467
468         Reviewed by Wenson Hsieh.
469
470         * MotionMark/resources/extensions.js:
471         (subtract.subtract.sampleY):
472
473 2019-04-15  Myles C. Maxfield  <mmaxfield@apple.com>
474
475         [Cocoa] FontPlatformData objects aren't cached at all when using font-family:system-ui
476         https://bugs.webkit.org/show_bug.cgi?id=196846
477         <rdar://problem/49499971>
478
479         Reviewed by Simon Fraser and Darin Adler.
480
481         * Layout/system-ui-rebuild-emoji.html: Added.
482
483 2019-04-03  Myles C. Maxfield  <mmaxfield@apple.com>
484
485         -apple-trailing-word is needed for browser detection
486         https://bugs.webkit.org/show_bug.cgi?id=196575
487
488         Unreviewed.
489
490         * MotionMark/resources/debug-runner/motionmark.css:
491         (#intro .start-benchmark p):
492
493 2019-03-27  Ryan Haddad  <ryanhaddad@apple.com>
494
495         REGRESSION (r242911?): High Sierra Release WK2 Perf bot timing out while running IndexedDB/large-number-of-inserts.html
496         https://bugs.webkit.org/show_bug.cgi?id=195952
497
498         Unreviewed test gardening.
499
500         * Skipped: Skip the affected test so the bot will finish the run.
501
502 2019-03-27  Saam Barati  <sbarati@apple.com>
503
504         JetStream 2 should not report time values as scores
505         https://bugs.webkit.org/show_bug.cgi?id=196334
506
507         Reviewed by Yusuke Suzuki.
508
509         * JetStream2/JetStreamDriver.js:
510         (toTimeValue):
511         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
512
513 2019-03-27  Saam barati  <sbarati@apple.com>
514
515         Unreviewed. Fix individual benchmark description urls to go to in-depth.html instead of about.html
516
517         * JetStream2/JetStreamDriver.js:
518         (Driver.prototype.prepareToRun.text.div.id.string_appeared_here.h3):
519
520 2019-03-25  Saam barati  <sbarati@apple.com>
521
522         Update the blurb describing JetStream2 and fix a broken link in the in-depth page.
523
524         Rubber-stamped by Filip Pizlo.
525
526         * JetStream2/in-depth.html:
527         * JetStream2/index.html:
528
529 2019-03-18  Saam Barati  <sbarati@apple.com>
530
531         Add a plan file for JetStream 2
532         https://bugs.webkit.org/show_bug.cgi?id=190734
533
534         Reviewed by Mark Lam.
535
536         Make the delay when starting via run-benchmark longer to allow
537         the page's animation to finish.
538
539         * JetStream2/JetStreamDriver.js:
540         (Driver.prototype.async.initialize):
541
542 2019-03-15  Jon Davis  <jond@apple.com>
543
544         Fixed a bug preventing the overall sore to be displayed on JetStream2
545         https://bugs.webkit.org/show_bug.cgi?id=195817
546
547         Rubber-stamped by Saam Barati.
548
549         * JetStream2/JetStreamDriver.js:
550         (Driver.prototype.async.start):
551
552 2019-03-14  Jon Davis  <jond@apple.com>
553
554         Implement a new design for JetStream2
555         https://bugs.webkit.org/show_bug.cgi?id=195492
556
557         Reviewed by Saam Barati.
558
559         * JetStream2/JetStream.css:
560         (html):
561         (body):
562         (::selection):
563         (main):
564         (.logo):
565         (.logo .logo-image):
566         (#jetstreams):
567         (#jetstreams svg):
568         (.summary):
569         (.summary + .summary):
570         (.summary:empty):
571         (article, .summary):
572         (h1):
573         (h2, h3, h4, h5, h6):
574         (h4, h5, h6):
575         (p):
576         (h5, h6):
577         (h6):
578         (a:link,):
579         (a:hover,):
580         (#status):
581         (#status label,):
582         (a.button):
583         (#status.loading):
584         (#status.error):
585         (.error h2, .error p):
586         (.error h2):
587         (#result-summary):
588         (#result-summary label):
589         (#result-summary .score):
590         (#result-summary .score .interval):
591         (#results):
592         (.benchmark):
593         (.benchmark h3, .benchmark h4, .benchmark .result, .benchmark label):
594         (.benchmark-running h4, .benchmark-running .result, .benchmark-running label):
595         (.benchmark-done h3, .benchmark-done h4, .benchmark-done .result, .benchmark-done label):
596         (.benchmark h3):
597         (.benchmark-running h3):
598         (.benchmark-done h3):
599         (.benchmark h3 a,):
600         (.benchmark-done h3 a:hover):
601         (.benchmark h4):
602         (.benchmark-done h4):
603         (.benchmark p,):
604         (.benchmark .result):
605         (.benchmark-done .result):
606         (.benchmark label):
607         (.benchmark-done label):
608         (@keyframes fadein):
609         (to):
610         (@keyframes scaledown):
611         (@keyframes shine):
612         (100%):
613         (@keyframes swingin):
614         (@media (max-width: 415px)):
615         (#logo): Deleted.
616         (p.summary): Deleted.
617         (p.summary + p.summary): Deleted.
618         (p.summary:empty): Deleted.
619         (p:first-of-type): Deleted.
620         (#status a:link): Deleted.
621         (.interval): Deleted.
622         (a:link, a:visited): Deleted.
623         (a:link:hover): Deleted.
624         (table): Deleted.
625         (td, th): Deleted.
626         (th): Deleted.
627         (tr:first-child > th:nth-child(even)): Deleted.
628         (tr:nth-child(even):not(:first-child, .benchmark-running)): Deleted.
629         (.result): Deleted.
630         (.benchmark-running): Deleted.
631         (:not(.benchmark-running) .result): Deleted.
632         (.benchmark-running .result): Deleted.
633         (.benchmark-name): Deleted.
634         (.result.category,): Deleted.
635         (.benchmark-name:not(.category):not(.geometric-mean)): Deleted.
636         (.geometric-mean): Deleted.
637         (.benchmark-name:not(:first-child)): Deleted.
638         (.benchmark-name a:link,): Deleted.
639         (.highlighted-result): Deleted.
640         * JetStream2/JetStream2Logo.svg: Added.
641         * JetStream2/JetStreamDriver.js:
642         (updateUI):
643         (Driver.prototype.async.start):
644         (Driver.prototype.prepareToRun.text.div.id.string_appeared_here.h3):
645         (Driver.prototype.prepareToRun.text.span):
646         (Driver.prototype.prepareToRun.a.h3.h4):
647         (Driver.prototype.prepareToRun.text.div):
648         (Driver.prototype.async.initialize):
649         (Driver.prototype.async.fetchResources.prototype.statusElement.onclick):
650         (prototype.fetchResources):
651         (prototype.updateUIBeforeRun):
652         (prototype.updateUIAfterRun):
653         (Driver.prototype.async.fetchResources.statusElement.innerHTML.a.href.string_appeared_here):
654         (Driver.prototype.async.fetchResources):
655         (Driver.prototype.prepareToRun.text.tr.id.string_appeared_here.FIXME.link.to.benchmark.explanation.td): Deleted.
656         * JetStream2/clouds.svg: Added.
657         * JetStream2/in-depth.html:
658         * JetStream2/index.html:
659         * JetStream2/jetstreams.svg: Added.
660
661 2019-03-11  Caio Lima  <ticaiolima@gmail.com>
662
663         [ESNext][BigInt] Implement "~" unary operation
664         https://bugs.webkit.org/show_bug.cgi?id=182216
665
666         Reviewed by Keith Miller.
667
668         * BigIntBench/big-int-simple-bit-not.js: Added.
669
670 2019-03-04  Saam Barati  <sbarati@apple.com>
671
672         Add a detailed summary page for JetStream 2
673         https://bugs.webkit.org/show_bug.cgi?id=195014
674
675         Rubber-stamped by Darin Adler.
676
677         * JetStream2/JetStream.css:
678         (a:link, a:visited):
679         * JetStream2/in-depth.html: Added.
680
681 2019-02-08  Saam barati  <sbarati@apple.com>
682
683         Update JetStream2 CLI to not print "ms" since we now just print the score values
684         https://bugs.webkit.org/show_bug.cgi?id=193981
685
686         Reviewed by Yusuke Suzuki.
687
688         * JetStream2/JetStreamDriver.js:
689         (DefaultBenchmark.prototype.updateUIAfterRun):
690         (DefaultBenchmark):
691         (WSLBenchmark.prototype.updateUIAfterRun):
692         (WSLBenchmark):
693         (WasmBenchmark.prototype.updateUIAfterRun):
694         (WasmBenchmark):
695
696 2019-01-31  Sihui Liu  <sihui_liu@apple.com>
697
698         REGRESSION (r240358): IndexedDB/large-binary-keys.html and IndexedDB/large-string-keys.html perf tests failing
699         https://bugs.webkit.org/show_bug.cgi?id=193956
700         <rdar://problem/47623109>
701
702         Reviewed by Ryosuke Niwa.
703
704         r240358 sets default IDB quota as 50 MB for single test, we need to increase that for big tests. 
705
706         * IndexedDB/large-binary-keys.html:
707         * IndexedDB/large-string-keys.html:
708
709 2019-01-18  Saam Barati  <sbarati@apple.com>
710
711         Use scores everywhere in JetStream2's UI
712         https://bugs.webkit.org/show_bug.cgi?id=193593
713
714         Reviewed by Keith Miller.
715
716         * JetStream2/JetStreamDriver.js:
717         (DefaultBenchmark.prototype.processResults):
718         (DefaultBenchmark.prototype.get score):
719         (DefaultBenchmark.scoreDescription):
720         (WSLBenchmark.prototype.processResults):
721         (WSLBenchmark.prototype.get score):
722         (WSLBenchmark.scoreDescription):
723         (WasmBenchmark.prototype.processResults):
724         (WasmBenchmark.prototype.get score):
725         (WasmBenchmark.scoreDescription):
726         * JetStream2/index.html:
727
728 2019-01-18  Jer Noble  <jer.noble@apple.com>
729
730         SDK_VARIANT build destinations should be separate from non-SDK_VARIANT builds
731         https://bugs.webkit.org/show_bug.cgi?id=189553
732
733         Reviewed by Tim Horton.
734
735         * MediaTime/Configurations/Base.xcconfig:
736         * MediaTime/Configurations/SDKVariant.xcconfig: Added.
737
738 2019-01-17  Truitt Savell  <tsavell@apple.com>
739
740         Unreviewed, rolling out r240124.
741
742         This commit broke an internal build.
743
744         Reverted changeset:
745
746         "SDK_VARIANT build destinations should be separate from non-
747         SDK_VARIANT builds"
748         https://bugs.webkit.org/show_bug.cgi?id=189553
749         https://trac.webkit.org/changeset/240124
750
751 2019-01-17  Jer Noble  <jer.noble@apple.com>
752
753         SDK_VARIANT build destinations should be separate from non-SDK_VARIANT builds
754         https://bugs.webkit.org/show_bug.cgi?id=189553
755
756         Reviewed by Tim Horton.
757
758         * MediaTime/Configurations/Base.xcconfig:
759         * MediaTime/Configurations/SDKVariant.xcconfig: Added.
760
761 2018-12-20  Caio Lima  <ticaiolima@gmail.com>
762
763         [BigInt] We should enable CSE into arithmetic operations that speculate BigIntUse
764         https://bugs.webkit.org/show_bug.cgi?id=192723
765
766         Reviewed by Yusuke Suzuki.
767
768         * BigIntBench/big-int-cse.js: Added.
769         * BigIntBench/big-int-global-cse.js: Added.
770         * BigIntBench/big-int-licm.js: Added.
771
772 2018-12-19  Commit Queue  <commit-queue@webkit.org>
773
774         Unreviewed, rolling out r239377.
775         https://bugs.webkit.org/show_bug.cgi?id=192921
776
777         broke 32-bit JSC tests (Requested by keith_miller on #webkit).
778
779         Reverted changeset:
780
781         "[BigInt] We should enable CSE into arithmetic operations that
782         speculate BigIntUse"
783         https://bugs.webkit.org/show_bug.cgi?id=192723
784         https://trac.webkit.org/changeset/239377
785
786 2018-12-19  Caio Lima  <ticaiolima@gmail.com>
787
788         [BigInt] We should enable CSE into arithmetic operations that speculate BigIntUse
789         https://bugs.webkit.org/show_bug.cgi?id=192723
790
791         Reviewed by Saam Barati.
792
793         * BigIntBench/big-int-cse.js: Added.
794         * BigIntBench/big-int-global-cse.js: Added.
795         * BigIntBench/big-int-licm.js: Added.
796
797 2018-12-17  Suresh Koppisetty  <skoppisetty@apple.com>
798
799         Add "-o/--output" option to startup.py and new_tab.py benchmark scripts to save the results in json format.
800         https://bugs.webkit.org/show_bug.cgi?id=192385
801
802         Reviewed by Ryosuke Niwa.
803
804         Sample json output for new tab benchmark script after running for 2 iterations and 2 groups. Values are in milliseconds.
805         {
806                 "NewTabBenchmark": {
807                         "metrics": {
808                                 "Time": {
809                                         "current": [
810                                                 [
811                                                         410.2939453125,
812                                                         307.81494140625
813                                                 ],
814                                                 [
815                                                         340.616943359375,
816                                                         265.94384765625
817                                                 ]
818                                         ]
819                                 }
820                         }
821                 }
822         }
823
824         Sample json output for startup time benchmark script after running for 2 iterations. Values are in milliseconds.
825         {
826                 "StartupBenchmark": {
827                         "metrics": {
828                                 "Time": {
829                                         "current": [
830                                                 [
831                                                         1415.2099609375,
832                                                         1439.552978515625
833                                                 ]
834                                         ]
835                                 }
836                         }
837                 }
838         }
839
840         * LaunchTime/launch_time.py:
841         * LaunchTime/new_tab.py:
842         (NewTabBenchmark.get_test_name):
843         * LaunchTime/startup.py:
844         (StartupBenchmark.get_test_name):
845
846 2018-12-17  Suresh Koppisetty  <skoppisetty@apple.com>
847
848         Import FeedbackServer only if "-f/--feedback-in-browser" option is enabled.
849         https://bugs.webkit.org/show_bug.cgi?id=192378
850
851         Reviewed by Ryosuke Niwa.
852
853         FeedbackServer currently depends on Tornado-5.1, which further adds
854         dependency of "singledispatch", "backports-abc" and "futures" python libraries.
855         Importing FeedbackServer only if "-f/--feedback-in-browser" option is enabled
856         will let us run the benchmark scripts without installing any new python libraries.
857
858         * LaunchTime/launch_time.py:
859         * LaunchTime/new_tab.py:
860
861 2018-12-13  Caio Lima  <ticaiolima@gmail.com>
862
863         [BigInt] Add ValueDiv into DFG
864         https://bugs.webkit.org/show_bug.cgi?id=186178
865
866         Reviewed by Yusuke Suzuki.
867
868         * BigIntBench/big-int-simple-div.js: Added.
869         * BigIntBench/value-div-type-propagation.js: Added.
870
871 2018-12-10  Caio Lima  <ticaiolima@gmail.com>
872
873         [BigInt] Add ValueMul into DFG
874         https://bugs.webkit.org/show_bug.cgi?id=186175
875
876         Reviewed by Yusuke Suzuki.
877
878         * BigIntBench/big-int-simple-mul.js: Added.
879         * BigIntBench/value-mul-type-propagation.js: Added.
880
881 2018-11-07  Caio Lima  <ticaiolima@gmail.com>
882
883         [BigInt] Add support to BigInt into ValueAdd
884         https://bugs.webkit.org/show_bug.cgi?id=186177
885
886         Reviewed by Keith Miller.
887
888         The idea of BigIntBench is to provide a set of microbenchmarks and
889         benchmarks to evaluate how fast BigInt computations are happening on
890         JSC implementation.
891
892         Now, we are adding microbenchmarks in this set,
893         but the plan is to move these tests to "JSTest/microbenchmarks" when
894         BigInt is enabled by default. After that, the focus of Bigint bench is
895         to provide a set of tests that represents real use cases of BigInt in
896         JS programs.
897
898         * BigIntBench/big-int-add-prediction-propagation.js: Added.
899         * BigIntBench/big-int-simple-add.js: Added.
900         * BigIntBench/big-int-simple-sub.js: Added.
901
902 2018-11-07  Tadeu Zagallo  <tzagallo@apple.com>
903
904         REGRESSION(r237547): Test failures on 32-bit JSC since the JIT was disabled
905         https://bugs.webkit.org/show_bug.cgi?id=191184
906
907         Reviewed by Saam Barati.
908
909         Skip test when the JIT is disabled, it takes too long to run on CLoop.
910
911         * JetStream/cdjs/main.js:
912
913 2018-11-05  Myles C. Maxfield  <mmaxfield@apple.com>
914
915         Cache glyph paths and share underline skipping code between all the ports
916         https://bugs.webkit.org/show_bug.cgi?id=191239
917
918         Reviewed by Alex Christensen.
919
920         Measures the performance of drawing a whole lot of underlines
921
922         * Layout/underline.html: Added.
923
924 2018-10-25  Saam Barati  <sbarati@apple.com>
925
926         Check in corresponding C code in JetStream 2
927         https://bugs.webkit.org/show_bug.cgi?id=190733
928
929         Rubber-stamped by Filip Pizlo.
930
931         * JetStream2/simple/float-mm.c: Added.
932         * JetStream2/wasm/HashSet.cpp: Added.
933         * JetStream2/wasm/gcc-loops.cpp: Added.
934         * JetStream2/wasm/quicksort.c: Added.
935         * JetStream2/wasm/TSF: Added.
936         * JetStream2/wasm/TSF/build.sh: Added.
937         * JetStream2/wasm/TSF/config.h: Added.
938         * JetStream2/wasm/TSF/gpc.h: Added.
939         * JetStream2/wasm/TSF/gpc_code_gen_util.c: Added.
940         * JetStream2/wasm/TSF/gpc_instruction.c: Added.
941         * JetStream2/wasm/TSF/gpc_instruction_dispatch.gen: Added.
942         * JetStream2/wasm/TSF/gpc_instruction_size.gen: Added.
943         * JetStream2/wasm/TSF/gpc_instruction_stack_effects.gen: Added.
944         * JetStream2/wasm/TSF/gpc_instruction_static_size.gen: Added.
945         * JetStream2/wasm/TSF/gpc_instruction_to_string.gen: Added.
946         * JetStream2/wasm/TSF/gpc_int_common.h: Added.
947         * JetStream2/wasm/TSF/gpc_intable.c: Added.
948         * JetStream2/wasm/TSF/gpc_internal.h: Added.
949         * JetStream2/wasm/TSF/gpc_interpreter.gen: Added.
950         * JetStream2/wasm/TSF/gpc_program.c: Added.
951         * JetStream2/wasm/TSF/gpc_proto.c: Added.
952         * JetStream2/wasm/TSF/gpc_stack_height.c: Added.
953         * JetStream2/wasm/TSF/gpc_threaded.c: Added.
954         * JetStream2/wasm/TSF/gpc_worklist.h: Added.
955         * JetStream2/wasm/TSF/tsf.h: Added.
956         * JetStream2/wasm/TSF/tsf_adaptive_reader.c: Added.
957         * JetStream2/wasm/TSF/tsf_asprintf.c: Added.
958         * JetStream2/wasm/TSF/tsf_atomics.h: Added.
959         * JetStream2/wasm/TSF/tsf_buf_reader.c: Added.
960         * JetStream2/wasm/TSF/tsf_buf_writer.c: Added.
961         * JetStream2/wasm/TSF/tsf_buffer.c: Added.
962         * JetStream2/wasm/TSF/tsf_build_defines.h: Added.
963         * JetStream2/wasm/TSF/tsf_config.h: Added.
964         * JetStream2/wasm/TSF/tsf_config_stub.h: Added.
965         * JetStream2/wasm/TSF/tsf_copier.c: Added.
966         * JetStream2/wasm/TSF/tsf_define_helpers.c: Added.
967         * JetStream2/wasm/TSF/tsf_define_helpers.h: Added.
968         * JetStream2/wasm/TSF/tsf_destructor.c: Added.
969         * JetStream2/wasm/TSF/tsf_error.c: Added.
970         * JetStream2/wasm/TSF/tsf_format.h: Added.
971         * JetStream2/wasm/TSF/tsf_fsdb.c: Added.
972         * JetStream2/wasm/TSF/tsf_fsdb_protocol.c: Added.
973         * JetStream2/wasm/TSF/tsf_fsdb_protocol.h: Added.
974         * JetStream2/wasm/TSF/tsf_generator.c: Added.
975         * JetStream2/wasm/TSF/tsf_gpc_code_gen.c: Added.
976         * JetStream2/wasm/TSF/tsf_indent.h: Added.
977         * JetStream2/wasm/TSF/tsf_internal.h: Added.
978         * JetStream2/wasm/TSF/tsf_internal_config.h: Added.
979         * JetStream2/wasm/TSF/tsf_internal_config_stub.h: Added.
980         * JetStream2/wasm/TSF/tsf_inttypes.h: Added.
981         * JetStream2/wasm/TSF/tsf_io.c: Added.
982         * JetStream2/wasm/TSF/tsf_io_utils.c: Added.
983         * JetStream2/wasm/TSF/tsf_ir.c: Added.
984         * JetStream2/wasm/TSF/tsf_ir.h: Added.
985         * JetStream2/wasm/TSF/tsf_ir_different.c: Added.
986         * JetStream2/wasm/TSF/tsf_ir_different.h: Added.
987         * JetStream2/wasm/TSF/tsf_ir_speed.c: Added.
988         * JetStream2/wasm/TSF/tsf_limits.c: Added.
989         * JetStream2/wasm/TSF/tsf_named_type.c: Added.
990         * JetStream2/wasm/TSF/tsf_native.c: Added.
991         * JetStream2/wasm/TSF/tsf_parser.c: Added.
992         * JetStream2/wasm/TSF/tsf_primitive.c: Added.
993         * JetStream2/wasm/TSF/tsf_ra_type_man.c: Added.
994         * JetStream2/wasm/TSF/tsf_reflect.c: Added.
995         * JetStream2/wasm/TSF/tsf_region.h: Added.
996         * JetStream2/wasm/TSF/tsf_serial_in_man.c: Added.
997         * JetStream2/wasm/TSF/tsf_serial_out_man.c: Added.
998         * JetStream2/wasm/TSF/tsf_serial_protocol.h: Added.
999         * JetStream2/wasm/TSF/tsf_sha1.c: Added.
1000         * JetStream2/wasm/TSF/tsf_sha1.h: Added.
1001         * JetStream2/wasm/TSF/tsf_sha1_writer.c: Added.
1002         * JetStream2/wasm/TSF/tsf_sort.c: Added.
1003         * JetStream2/wasm/TSF/tsf_st.c: Added.
1004         * JetStream2/wasm/TSF/tsf_st.h: Added.
1005         * JetStream2/wasm/TSF/tsf_st_typetable.c: Added.
1006         * JetStream2/wasm/TSF/tsf_stream_file_input.c: Added.
1007         * JetStream2/wasm/TSF/tsf_stream_file_output.c: Added.
1008         * JetStream2/wasm/TSF/tsf_type.c: Added.
1009         * JetStream2/wasm/TSF/tsf_type_in_map.c: Added.
1010         * JetStream2/wasm/TSF/tsf_type_out_map.c: Added.
1011         * JetStream2/wasm/TSF/tsf_type_table.c: Added.
1012         * JetStream2/wasm/TSF/tsf_types.h: Added.
1013         * JetStream2/wasm/TSF/tsf_util.h: Added.
1014         * JetStream2/wasm/TSF/tsf_version.c: Added.
1015         * JetStream2/wasm/TSF/tsf_zip_abstract.c: Added.
1016         * JetStream2/wasm/TSF/tsf_zip_abstract.h: Added.
1017         * JetStream2/wasm/TSF/tsf_zip_attr.c: Added.
1018         * JetStream2/wasm/TSF/tsf_zip_reader.c: Added.
1019         * JetStream2/wasm/TSF/tsf_zip_writer.c: Added.
1020
1021 2018-10-18  Saam Barati  <sbarati@apple.com>
1022
1023         Make JetStream 2
1024         https://bugs.webkit.org/show_bug.cgi?id=187829
1025
1026         Rubber-stamped by Mark Lam.
1027
1028         This patch checks in the new JetStream 2 benchmark. JetStream 2 is
1029         a new JavaScript and Web Assembly benchmark. JetStream 2's goal
1030         is to measure the startup, worst case, and peak throughput performance
1031         of the JavaScript engine. JetStream 2 incorporates these previous
1032         benchmarks:
1033         - JetStream
1034         - ARES-6
1035         - Kraken
1036         - Web Tooling Benchmark
1037         - WasmBench
1038         - RexBench
1039         
1040         JetStream 2 also adds some new benchmarks:
1041         - Two tests emphasizing web worker performance.
1042         - One test emphasizing Promise, async iteration, and DataView performance.
1043         - Two new code load tests.
1044         - WSL: a test measuring all kinds of things, especially emphasizing exception performance.
1045
1046         * JetStream2: Added.
1047
1048 2018-10-02  Chris Dumez  <cdumez@apple.com>
1049
1050         Regression(r236613): Parser/html-parser.html performance test is failing
1051         https://bugs.webkit.org/show_bug.cgi?id=190201
1052
1053         Unreviewed, apply the same fix as in r236699 to make sure the iframe is treated as
1054         same origin, so that calls to document.open() / document.write() succeed.
1055
1056         * Parser/html-parser.html:
1057
1058 2018-10-01  Chris Dumez  <cdumez@apple.com>
1059
1060         Regression(r236613): Parser/HTML5-8266-ParseOnly.html performance tests no longer runs
1061         https://bugs.webkit.org/show_bug.cgi?id=190174
1062
1063         Reviewed by Ryosuke Niwa.
1064
1065         Update iframe's sandbox to use 'allow-same-origin' so that the iframe is treated as same-origin and so that
1066         the top frame can call document.open() / document.write() on the subframe. This test was also failing in
1067         Firefox without this change.
1068
1069         * Parser/HTML5-8266-ParseOnly.html:
1070
1071 2018-08-27  Aditya Keerthi  <akeerthi@apple.com>
1072
1073         Consolidate ENABLE_INPUT_TYPE_COLOR and ENABLE_INPUT_TYPE_COLOR_POPOVER
1074         https://bugs.webkit.org/show_bug.cgi?id=188931
1075
1076         Reviewed by Wenson Hsieh.
1077
1078         * StitchMarker/wtf/FeatureDefines.h: Removed ENABLE_INPUT_TYPE_COLOR_POPOVER.
1079
1080 2018-08-23  Don Olmstead  <don.olmstead@sony.com>
1081
1082         [CMake] Have checks are not getting set inside CMake properly
1083         https://bugs.webkit.org/show_bug.cgi?id=188901
1084
1085         Reviewed by Michael Catanzaro.
1086
1087         Add notification that MallocBench is disabled.
1088
1089         * CMakeLists.txt:
1090
1091 2018-08-23  Don Olmstead  <don.olmstead@sony.com>
1092
1093         [CMake] Add HAVE_MALLOC_TRIM definition
1094         https://bugs.webkit.org/show_bug.cgi?id=188897
1095
1096         Reviewed by Konstantin Tokarev.
1097
1098         MallocBench should only be built on Apple platforms and platforms that
1099         have malloc_trim.
1100
1101         * CMakeLists.txt:
1102
1103 2018-08-13  Jon Lee  <jonlee@apple.com>
1104
1105         Update MotionMark version number
1106         https://bugs.webkit.org/show_bug.cgi?id=188535
1107         <rdar://problem/43254078>
1108
1109         Reviewed by Said Abou-Hallawa.
1110
1111         * MotionMark/resources/strings.js: To 1.1.
1112
1113 2018-08-13  Jon Lee  <jonlee@apple.com>
1114
1115         [MotionMark] Update Multiply test
1116         https://bugs.webkit.org/show_bug.cgi?id=188532
1117         <rdar://problem/43252151>
1118
1119         Reviewed by Said Abou-Hallawa.
1120
1121         Update Multiply test to have a larger upper limit of number of particles. Cycle through three
1122         different ways of hiding an element. Add new Multiple suite that isolates those components for
1123         debugging purposes.
1124
1125         * MotionMark/resources/debug-runner/tests.js:
1126         * MotionMark/tests/dom/multiply.html: Added.
1127         * MotionMark/tests/dom/resources/multiply.js: Added.
1128         * MotionMark/tests/master/resources/multiply.js:
1129
1130 2018-08-13  Jon Lee  <jonlee@apple.com>
1131
1132         [MotionMark] Update Leaves test
1133         https://bugs.webkit.org/show_bug.cgi?id=188530
1134         <rdar://problem/43251862>
1135
1136         Reviewed by Said Abou-Hallawa.
1137
1138         Update Leaves test to include opacity and scale. Add new Leaves suite that isolate those components
1139         for debugging purposes.
1140
1141         * MotionMark/resources/debug-runner/tests.js:
1142         * MotionMark/tests/dom/leaves.html:
1143         * MotionMark/tests/dom/resources/leaves.js:
1144         * MotionMark/tests/master/resources/leaves.js:
1145
1146 2018-08-10  Ben Richards  <benton_richards@apple.com>
1147
1148         Add ability to ignore process prewarming for launch time benchmark
1149         https://bugs.webkit.org/show_bug.cgi?id=188462
1150
1151         Reviewed by Ryosuke Niwa.
1152
1153         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.
1154
1155         * LaunchTime/launch_time.py:
1156         (DefaultLaunchTimeHandler):
1157         (DefaultLaunchTimeHandler.on_receive_stop_time): Deleted.
1158         (DefaultLaunchTimeHandler.on_receive_stop_signal):
1159         (DefaultLaunchTimeHandler.do_HEAD):
1160         (DefaultLaunchTimeHandler.do_GET):
1161         (DefaultLaunchTimeHandler.do_POST):
1162         (LaunchTimeBenchmark):
1163         (LaunchTimeBenchmark._standard_deviation): Fixed divide by zero bug when '-n' is set to 1
1164         (LaunchTimeBenchmark.open_tab): Added option to open a blank tab
1165         (LaunchTimeBenchmark.run):
1166         * LaunchTime/new_tab.py:
1167         (NewTabBenchmark.initialize):
1168         (NewTabBenchmark.run_iteration):
1169         (NewTabBenchmark.will_parse_arguments):
1170         (NewTabBenchmark.did_parse_arguments):
1171         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1172         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1173         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1174         (NewTabBenchmark):
1175         * LaunchTime/startup.py:
1176         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1177         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1178         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1179         (StartupBenchmark):
1180
1181 2018-08-10  Truitt Savell  <tsavell@apple.com>
1182
1183         Unreviewed, rolling out r234750.
1184
1185         Caused 185 perf test failures.
1186
1187         Reverted changeset:
1188
1189         "Add ability to ignore process prewarming for launch time
1190         benchmark"
1191         https://bugs.webkit.org/show_bug.cgi?id=188462
1192         https://trac.webkit.org/changeset/234750
1193
1194 2018-08-09  Ben Richards  <benton_richards@apple.com>
1195
1196         Add ability to ignore process prewarming for launch time benchmark
1197         https://bugs.webkit.org/show_bug.cgi?id=188462
1198
1199         Reviewed by Ryosuke Niwa.
1200
1201         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.
1202
1203         * LaunchTime/launch_time.py:
1204         (DefaultLaunchTimeHandler):
1205         (DefaultLaunchTimeHandler.on_receive_stop_time): Deleted.
1206         (DefaultLaunchTimeHandler.on_receive_stop_signal):
1207         (DefaultLaunchTimeHandler.do_HEAD):
1208         (DefaultLaunchTimeHandler.do_GET):
1209         (DefaultLaunchTimeHandler.do_POST):
1210         (LaunchTimeBenchmark):
1211         (LaunchTimeBenchmark._standard_deviation): Fixed divide by zero bug when '-n' is set to 1
1212         (LaunchTimeBenchmark.open_tab): Added option to open a blank tab
1213         (LaunchTimeBenchmark.run):
1214         * LaunchTime/new_tab.py:
1215         (NewTabBenchmark.initialize):
1216         (NewTabBenchmark.run_iteration):
1217         (NewTabBenchmark.will_parse_arguments):
1218         (NewTabBenchmark.did_parse_arguments):
1219         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1220         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1221         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1222         (NewTabBenchmark):
1223         * LaunchTime/startup.py:
1224         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1225         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1226         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1227         (StartupBenchmark):
1228
1229 2018-07-26  Andy VanWagoner  <andy@vanwagoner.family>
1230
1231         [INTL] Remove INTL sub-feature compile flags
1232         https://bugs.webkit.org/show_bug.cgi?id=188081
1233
1234         Reviewed by Michael Catanzaro.
1235
1236         Removed old and unused i18n flag.
1237
1238         * StitchMarker/wtf/FeatureDefines.h:
1239
1240 2018-07-20  David Fenton  <david_fenton@apple.com>
1241
1242         REGRESSION (234006) ERROR: Blocked access to external URL http://code.jquery.com/jquery-1.9.1.min.js
1243         https://bugs.webkit.org/show_bug.cgi?id=187858
1244
1245         Unreviewed test gardening, skipped test.
1246
1247         * Skipped: LaunchTime
1248
1249 2018-07-19  Ben Richards  <benton_richards@apple.com>
1250
1251         Add benchmark for WebKit process launch times
1252         https://bugs.webkit.org/show_bug.cgi?id=186414
1253
1254         Reviewed by Ryosuke Niwa.
1255
1256         Added two benchmarks, one for measuring browser new tab launch time and one for browser startup time.
1257
1258         * LaunchTime/.gitignore: Added.
1259         * LaunchTime/feedback_client.html: Added.
1260         Displays benchmark progress in browser
1261         * LaunchTime/feedback_server.py: Added.
1262         (FeedbackServer): Sends data to feedback_client via websocket
1263         (FeedbackServer.__init__):
1264         (FeedbackServer._create_app):
1265         (FeedbackServer._start_server):
1266         (FeedbackServer._send_all_messages):
1267         (FeedbackServer.start):
1268         (FeedbackServer.stop):
1269         (FeedbackServer.send_message): Send a message to the feedback_client
1270         (FeedbackServer.wait_until_client_has_loaded): Wait until the feedback_client has opened a websocket connection to the feedback_server
1271         (FeedbackServer.MainHandler): Handler factory to create handler that serves feedback_client.html
1272         (FeedbackServer.MainHandler.Handler):
1273         (FeedbackServer.MainHandler.Handler.get):
1274         (FeedbackServer.WSHandler): Handler factory to create handler that sends data to feedback client
1275         (FeedbackServer.WSHandler.Handler):
1276         (FeedbackServer.WSHandler.Handler.open): On websocket connection opened
1277         (FeedbackServer.WSHandler.Handler.on_close): On websocket connection closed
1278         * LaunchTime/launch_time.py: Added.
1279         (DefaultLaunchTimeHandler): Abstract HTTP request handler for launch time benchmarks
1280         (DefaultLaunchTimeHandler.get_test_page): Default test page to be overridden by benchmarks
1281         (DefaultLaunchTimeHandler.get_blank_page):
1282         (DefaultLaunchTimeHandler.on_receive_stop_time):
1283         (DefaultLaunchTimeHandler.do_HEAD):
1284         (DefaultLaunchTimeHandler.do_GET):
1285         (DefaultLaunchTimeHandler.do_POST):
1286         (DefaultLaunchTimeHandler.log_message): Suppresses HTTP logs from SimpleHTTPRequestHandler
1287         (LaunchTimeBenchmark): Abstract class which launch time benchmarks inherit from and override methods desired to customize
1288         (LaunchTimeBenchmark.__init__):
1289         (LaunchTimeBenchmark._parse_browser_bundle_path): Parser for bundle path option
1290         (LaunchTimeBenchmark._parse_args):
1291         (LaunchTimeBenchmark._run_server): Target function for main server thread
1292         (LaunchTimeBenchmark._setup_servers):
1293         (LaunchTimeBenchmark._clean_up):
1294         (LaunchTimeBenchmark._exit_due_to_exception):
1295         (LaunchTimeBenchmark._geometric_mean):
1296         (LaunchTimeBenchmark._standard_deviation):
1297         (LaunchTimeBenchmark._compute_results): Returns mean and std dev of list of results and pretty prints if should_print=True is specified
1298         (LaunchTimeBenchmark._wait_times): Mimic numpy.linspace
1299         (LaunchTimeBenchmark.open_tab): Open a browser tab with the html given by self.response_handler.get_test_page
1300         (LaunchTimeBenchmark.launch_browser): Open a broser to either the feedback client (if option is set) or a blank page
1301         (LaunchTimeBenchmark.quit_browser):
1302         (LaunchTimeBenchmark.quit_browser.quit_app):
1303         (LaunchTimeBenchmark.quit_browser.is_app_closed):
1304         (LaunchTimeBenchmark.close_tab):
1305         (LaunchTimeBenchmark.wait):
1306         (LaunchTimeBenchmark.log): Print to console and send to feedback client if --feedback-in-browser flag is used
1307         (LaunchTimeBenchmark.log_verbose): Only logs if --verbose flag is used
1308         (LaunchTimeBenchmark.run):
1309         (LaunchTimeBenchmark.group_init): Initialization done before each round of iterations
1310         (LaunchTimeBenchmark.run_iteration):
1311         (LaunchTimeBenchmark.initialize): Convenience method to be overriden by subclasses which is called at the end of __init__
1312         (LaunchTimeBenchmark.will_parse_arguments): Called before argparse.parse_args to let subclasses add new command line arguments
1313         (LaunchTimeBenchmark.did_parse_arguments): Called after argparse.parse_args to let subclass initialize based on command line arguments
1314         * LaunchTime/new_tab.py: Added
1315         (NewTabBenchmark):
1316         (NewTabBenchmark._parse_wait_time): Parser for wait time option
1317         (NewTabBenchmark.initialize):
1318         (NewTabBenchmark.run_iteration):
1319         (NewTabBenchmark.group_init):
1320         (NewTabBenchmark.will_parse_arguments):
1321         (NewTabBenchmark.did_parse_arguments):
1322         (NewTabBenchmark.ResponseHandler):
1323         (NewTabBenchmark.ResponseHandler.Handler):
1324         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1325         * LaunchTime/startup.py: Added
1326         (StartupBenchmark): This benchmark measures browser startup time and initial page load time
1327         (StartupBenchmark.initialize):
1328         (StartupBenchmark.run_iteration):
1329         (StartupBenchmark.ResponseHandler):
1330         (StartupBenchmark.ResponseHandler.Handler):
1331         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1332         * LaunchTime/thirdparty/__init__.py: Added.
1333         (AutoinstallImportHook): Auto installs tornado package for feedback server
1334         (AutoinstallImportHook.__init__):
1335         (AutoinstallImportHook._ensure_autoinstalled_dir_is_in_sys_path):
1336         (AutoinstallImportHook.find_module):
1337         (AutoinstallImportHook._install_tornado):
1338         (AutoinstallImportHook.greater_than_equal_to_version):
1339         (AutoinstallImportHook._install):
1340         (AutoinstallImportHook.get_latest_pypi_url):
1341         (AutoinstallImportHook.install_binary):
1342         (autoinstall_everything):
1343         (get_os_info):
1344
1345 2018-06-25  Jon Lee  <jonlee@apple.com>
1346
1347         [MotionMark] Add support for version numbers
1348         https://bugs.webkit.org/show_bug.cgi?id=186479
1349
1350         Reviewed by Said Abou-Hallawa.
1351
1352         Add support for displaying the version number as well as including it in the JSON results.
1353
1354         When loading the front page, script replaces any element with classname version with the
1355         version number of the benchmark, which is stored in Strings.version.
1356
1357         The JSON structure for the results includes a new version property:
1358             {
1359               "version": "1.0",
1360               "options": { ... },
1361               "data": [ ... ]
1362             }
1363
1364         When dragging a results file, the version listed will come from the JSON file. Older
1365         results will not have had the version property, in which case it will default to "1.0".
1366
1367         * MotionMark/index.html: Update title to some other default. Script will update it.
1368         Include the version number in the logo title.
1369         * MotionMark/developer.html: Ditto.
1370         * MotionMark/about.html: Ditto.
1371
1372         * MotionMark/resources/runner/motionmark.js:
1373         (ResultsDashboard): Update constructor to include version. This is used when serializing
1374         results out to JSON, and displaying the results panel in developer mode.
1375         (ResultsDashboard._processData): When running the benchmark, include benchmark version string
1376         in the results object.
1377         (ResultsDashboard.version):
1378         (window.benchmarkRunnerClient.willStartFirstIteration): When running the benchmark, pass the
1379         benchmark version string to the dashboard, which holds the results.
1380         (window.sectionsManager.setSectionVersion): Helper function to update the element in the
1381         section with the class name version.
1382         (window.benchmarkController.initialize): Populate all DOM elements with class name "version"
1383         with the version string. Update the page title.
1384         (window.benchmarkController.showResults): When showing results, update the version string
1385         based on what is included in the JSON results, which would be the same as the benchmark version.
1386         * MotionMark/resources/runner/motionmark.css: Include missing copyright. Wrap the SVG logo
1387         in a div and include the version string.
1388         * MotionMark/resources/strings.js: Add strings for the page title template, and the version.
1389
1390         * MotionMark/resources/debug-runner/motionmark.css:
1391         * MotionMark/resources/debug-runner/motionmark.js:
1392         (window.benchmarkRunnerClient.willStartFirstIteration): When running the benchmark, pass the
1393         benchmark version string to the dashboard, which holds the results.
1394         (window.benchmarkController.initialize): Populate all DOM elements with class name "version"
1395         with the version string. Update the page title. When dragging in JSON results, look for
1396         version to pass to the dashboard. If it doesn't exist, default to "1.0".
1397         (window.benchmarkController.showResults): When showing results, update the version string
1398         based on what is included in the JSON results, instead of the current benchmark version.
1399         * MotionMark/resources/debug-runner/tests.js: Update page title template.
1400
1401 2018-06-08  Jon Lee  <jonlee@apple.com>
1402
1403         [MotionMark] Rename Suits test files
1404         https://bugs.webkit.org/show_bug.cgi?id=186447
1405
1406         Reviewed by Said Abou-Hallawa.
1407
1408         * MotionMark/resources/runner/tests.js:
1409         * MotionMark/tests/master/resources/suits.js: Renamed from PerformanceTests/MotionMark/tests/master/resources/svg-particles.js.
1410         * MotionMark/tests/master/suits.html: Renamed from PerformanceTests/MotionMark/tests/master/svg-particles.html.
1411         * MotionMark/tests/svg/suits.html:
1412
1413 2018-06-08  Jon Lee  <jonlee@apple.com>
1414
1415         Add sub-tests based on Suits
1416         https://bugs.webkit.org/show_bug.cgi?id=186260
1417
1418         Reviewed by Said Abou-Hallawa.
1419
1420         Add a new developer Suits suite with sub-tests that isolate parts of each particle.
1421         The sub-tests are:
1422             - Particles using only clip paths
1423             - Particles using only shapes
1424             - Particles that have no gradients
1425             - Particles that have no rotation
1426             - Particles that do not move around at all (but all of the physics calculations are
1427             still performed)
1428
1429         * MotionMark/resources/debug-runner/tests.js:
1430         * MotionMark/tests/master/resources/svg-particles.js: Rename the stage, particle, and
1431         benchmark to "Suits" from "SVG". Change SuitsParticle so that we guarantee 50% clip and
1432         shape paths rather than relying on random chance. Wrap a check around creation of the
1433         gradient element, and use a simple fill color when we aren't using a gradient.
1434         * MotionMark/tests/svg/suits.html: Added.
1435         * MotionMark/tests/svg/suits.js: Added.
1436         Look for the query string and set the particle based on what is selected.
1437
1438 2018-06-06  Jon Lee  <jonlee@apple.com>
1439
1440         Allow the ramp controller to run tests that take less time than the initial ramp-up phase
1441         https://bugs.webkit.org/show_bug.cgi?id=186257
1442
1443         Reviewed by Said Abou-Hallawa.
1444
1445         * MotionMark/tests/resources/main.js: Extend the desired end timestamp when ramp-up occurs.
1446         Previously the controller assumed that the duration of the test was always longer than the
1447         time it took to figure out the order of magnitude of particles to render. When that range
1448         is determined, the end timestamp is then extended. If the duration is too short, the controller
1449         would throw an exception. The fix is to continue extending the timestamp every time the order of
1450         magnitude tiers up.
1451
1452         In general testing with really short durations isn't practical, but sometimes it is
1453         desired for debugging.
1454
1455         To avoid confusion of the private variables, I renamed Benchmark._startTimestamp to
1456         Benchmark._benchmarkStartTimestamp, to disambiguate from Controller._startTimestamp.
1457         Benchmark._startTimestamp is set once and never changed, after the initial 100ms warm-up.
1458         Controller._startTimestamp is initialized to 0 (which means "not initialized"), and then after
1459         the warm up is complete, set in Controller.start() to the same wall clock timestamp, and never
1460         changed afterwards.
1461
1462 2018-06-06  Jon Lee  <jonlee@apple.com>
1463
1464         Remove unneeded data processing in MotionMark
1465         https://bugs.webkit.org/show_bug.cgi?id=186256
1466
1467         Reviewed by Said Abou-Hallawa.
1468
1469         Remove calculation of average frame rate based on multiple samples at a given complexity.
1470         It is not used in the calculation of the score, and is just extra work to do at the end
1471         of a test.
1472
1473         * MotionMark/developer.html: Remove options to show average data and regression based on
1474         that data.
1475         * MotionMark/resources/debug-runner/graph.js: Remove data. Update the time graph that
1476         shows the relationship between the break point in the regression against the complexity
1477         graph.
1478         * MotionMark/resources/debug-runner/motionmark.css: Make the confidence intervals easier
1479         to see.
1480         * MotionMark/resources/runner/motionmark.js: Simplify score calculation now that we are
1481         only calculating regressions on the raw data. We were calculating it based on the averages,
1482         but that was just for analysis, and not used in the actual score. Get rid of the
1483         |complexitySamples| local variable.
1484         * MotionMark/resources/strings.js: Remove unneeded string constant.
1485         * MotionMark/tests/resources/main.js:
1486         (processSamples): Refactor so that more of it can be shared between Controller and
1487         RampController. It also avoids processing the raw samples (controllerSamples) twice for
1488         the ramp controller.
1489         (_processComplexitySamples): Remove logic to calculate averages based on complexity.
1490         (_processMarks): Split this out from the old processSamples.
1491         (_processControllerSamples): Split this out from the old processSamples.
1492
1493 2018-06-06  Jon Lee  <jonlee@apple.com>
1494
1495         Strip out more characters when creating permalinks
1496         https://bugs.webkit.org/show_bug.cgi?id=186259
1497
1498         Reviewed by Said Abou-Hallawa.
1499
1500         Tests and suites may have names that make permalinks to the tests less readable when
1501         shared. Expand the set of characters that are stripped out of the names.
1502
1503         * MotionMark/resources/debug-runner/motionmark.js: Refactor to call
1504         stripUnwantedCharactersForURL. For existing links, decode the suite and test names.
1505         * MotionMark/resources/extensions.js:
1506         (Utilities.stripUnwantedCharactersForURL): Rename from stripNonASCIICharacters.
1507         Remove any non-alphanumeric character.
1508
1509 2018-06-04  Jon Lee  <jonlee@apple.com>
1510
1511         Remove unnecessary MotionMark controllers
1512         https://bugs.webkit.org/show_bug.cgi?id=186255
1513
1514         Reviewed by Said Abou-Hallawa.
1515
1516         Get rid of the ramp30 and the fixed-with-a-step controllers, which aren't used for testing
1517         or debugging.
1518
1519         * MotionMark/developer.html: Move the "ramp" option as first, since that is the default
1520         controller for the benchmark.
1521         * MotionMark/resources/debug-runner/motionmark.js:
1522         * MotionMark/resources/runner/motionmark.js: Added missing copyright notice.
1523         * MotionMark/tests/resources/main.js:
1524
1525 2018-05-25  Myles C. Maxfield  <mmaxfield@apple.com>
1526
1527         Improve the performance of Font::canRenderCombiningCharacterSequence()
1528         https://bugs.webkit.org/show_bug.cgi?id=185933
1529
1530         Reviewed by Ryosuke Niwa.
1531
1532         * Layout/ComplexLongUnique.html: Added.
1533
1534 2018-05-25  Saam Barati  <sbarati@apple.com>
1535
1536         Have a memory test where we can validate JSCs mini memory mode
1537         https://bugs.webkit.org/show_bug.cgi?id=185932
1538
1539         Reviewed by Mark Lam.
1540
1541         We add a directory here with the contents of the testmem benchmark.
1542         To run it, use `Tools/Scripts/run-testmem`. To add new tests in the future,
1543         you just need to add JS files to this directory.
1544
1545         * testmem: Added.
1546         * testmem/air.js: Added.
1547         * testmem/base64.js: Added.
1548         * testmem/basic.js: Added.
1549         * testmem/box2d.js: Added.
1550         * testmem/crypto-md5.js: Added.
1551         * testmem/date-format-tofte.js: Added.
1552         * testmem/earley-boyer.js: Added.
1553         * testmem/hash-map.js: Added.
1554         * testmem/regex-dna.js: Added.
1555         * testmem/splay.js: Added.
1556         * testmem/tagcloud.js: Added.
1557
1558 2018-05-22  Ryan Haddad  <ryanhaddad@apple.com>
1559
1560         Unreviewed, rolling out r232052.
1561
1562         Breaks internal builds.
1563
1564         Reverted changeset:
1565
1566         "Use more C++17"
1567         https://bugs.webkit.org/show_bug.cgi?id=185176
1568         https://trac.webkit.org/changeset/232052
1569
1570 2018-05-21  Yusuke Suzuki  <utatane.tea@gmail.com>
1571
1572         Use more C++17
1573         https://bugs.webkit.org/show_bug.cgi?id=185176
1574
1575         Reviewed by JF Bastien.
1576
1577         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
1578         * MediaTime/Configurations/Base.xcconfig:
1579         * MediaTime/MediaTime.xcodeproj/project.pbxproj:
1580
1581 2018-04-30  Myles C. Maxfield  <mmaxfield@apple.com>
1582
1583         Improve the performance of FontCascadeDescription's effectiveFamilies
1584         https://bugs.webkit.org/show_bug.cgi?id=184720
1585
1586         Reviewed by Simon Fraser.
1587
1588         This performance test calls SystemFontDatabase::systemFontCascadeList() around 2,000,000 times (before
1589         this patch is applied), which is roughly equivalent to the page we found the performance problem on.
1590         The calling pattern is roughly equivalent in this test.
1591
1592         * Layout/system-ui.html: Added.
1593
1594 2018-03-08  Antti Koivisto  <antti@apple.com>
1595
1596         Update StyleBench version number in page title to 0.3
1597
1598         * StyleBench/index.html:
1599
1600         There has been a bunch of changes, most notably the addition of attribute selector test step.
1601
1602 2018-02-16  Dean Jackson  <dino@apple.com>
1603
1604         Use OPENGL macros to be more clear about which OpenGL/ES WebGL uses on Cocoa
1605         https://bugs.webkit.org/show_bug.cgi?id=182894
1606
1607         Reviewed by Tim Horton.
1608
1609         Rename OPENGL_ES_2 to OPENGL_ES.
1610
1611         * StitchMarker/wtf/Platform.h:
1612
1613 2018-02-09  Ryosuke Niwa  <rniwa@webkit.org>
1614
1615         Make run-perf-tests work with Speedometer 2.0 and re-enable the test
1616         https://bugs.webkit.org/show_bug.cgi?id=182088
1617
1618         Reviewed by Antti Koivisto.
1619
1620         This patch updates the test harness to report new metrics of Speedometer 2.0.
1621
1622         Replaced the total time by the geometric mean of the time spent in each suite, and added scores
1623         as a metric in addition to time. Also, updated the PerfTestRunner harness and the results template
1624         to support "Score" metric with "pt" as its unit.
1625
1626         * Skipped: Unskipped the test.
1627         * Speedometer/resources/benchmark-report.js:
1628         (window.benchmarkClient.willStartFirstIteration..createTest): Added the unit as an optional argument
1629         to report the score in addition to time. Also report the geomean of time instead of the total time.
1630         (window.benchmarkClient.willStartFirstIteration): Report the geomean of time.
1631         (window.benchmarkClient.didRunSuites): Record the entire measuredValues instead of just tests.
1632         (window.benchmarkClient.didFinishLastIteration): Report the scores.
1633
1634 2018-02-09  Ross Kirsling  <ross.kirsling@sony.com>
1635
1636         Use REQUEST_ANIMATION_FRAME_DISPLAY_MONITOR on any non-Windows port.
1637         https://bugs.webkit.org/show_bug.cgi?id=182623
1638
1639         Reviewed by Alex Christensen.
1640
1641         * StitchMarker/wtf/Platform.h:
1642         Simplify #if.
1643
1644 2018-02-01  Antti Koivisto  <antti@apple.com>
1645
1646         StyleBench: Attribute selectors and other improvements
1647         https://bugs.webkit.org/show_bug.cgi?id=182387
1648
1649         Reviewed by Joseph Pecoraro.
1650
1651         - Add some attributes to elements in all tests
1652         - Add some attribute selectors to stylesheets in all tests
1653         - Also add some * selectors to all stylesheets.
1654         - Add attribute mutation step to all suites
1655         - Make test steps do more mutations (25->100) and reduce the number of steps to keep testing time in check.
1656           Too fast steps were running into timer resolution limits.
1657
1658         * StyleBench/resources/style-bench.js:
1659         (defaultConfiguration):
1660         (prototype.randomAttributeName):
1661         (prototype.randomAttributeValue):
1662         (prototype.randomAttributeSelector):
1663         (prototype.makeCompoundSelector):
1664         (prototype.makeElement):
1665         (prototype.addClasses):
1666         (prototype.removeClasses):
1667         (prototype.mutateAttributes):
1668         (prototype.async.runForever):
1669         * StyleBench/resources/tests.js:
1670         (makeSteps):
1671         (makeSuite):
1672
1673 2018-02-01  Geoffrey Garen  <ggaren@apple.com>
1674
1675         Make MallocBench easier for non-WebKit engineers to run
1676         https://bugs.webkit.org/show_bug.cgi?id=182415
1677
1678         Reviewed by Saam Barati.
1679
1680         * MallocBench/MallocBench.xcodeproj/project.pbxproj: Use c++14 so we
1681         can make_unique.
1682
1683         Specify that we support all Darwin platforms so you can test them.
1684
1685         * MallocBench/run-malloc-benchmarks: Specify the path to MallocBench
1686         and libmbmalloc explicitly, rather than computing them implicitly
1687         using webkitdirs. Non-WebKit folks don't have the directory structure
1688         required by webkitdirs.
1689
1690         Remove Linux-specific and cmake-specific behaviors because we only
1691         needed them in the world of implicit path computation.
1692
1693 2018-01-31  Ryosuke Niwa  <rniwa@webkit.org>
1694
1695         Make run-perf-tests work with StyleBench and re-enable it
1696         https://bugs.webkit.org/show_bug.cgi?id=182322
1697
1698         Reviewed by Antti Koivisto.
1699
1700         This patch updates the test harness to report new metrics of StyleBench.
1701
1702         Replaced the total time by the geometric mean of the time spent in each suite, and added scores
1703         as a metric in addition to time. Also, updated the PerfTestRunner harness and the results template
1704         to support "Score" metric with "pt" as its unit.
1705
1706         * Skipped: Unskipped the test.
1707         * StyleBench/resources/benchmark-report.js:
1708         (window.benchmarkClient.willStartFirstIteration..createTest): Added the unit as an optional argument
1709         to report the score in addition to time. Also report the geomean of time instead of the total time.
1710         (window.benchmarkClient.willStartFirstIteration): Report the geomean of time.
1711         (window.benchmarkClient.didRunSuites): Record the entire measuredValues instead of just tests.
1712         (window.benchmarkClient.didFinishLastIteration): Report the scores.
1713         * resources/results-template.html: Added "Score" as a metric and "pt" as its unit.
1714         * resources/runner.js:
1715         (PerfTestRunner..finish): Ditto.
1716
1717 2018-01-30  Ryosuke Niwa  <rniwa@webkit.org>
1718
1719         StyleBench: Generate more mutation steps
1720         https://bugs.webkit.org/show_bug.cgi?id=182291
1721
1722         Rubber-stamped by Antti Koivisto
1723
1724         Use zero-based index instead of one-based index.
1725
1726         * StyleBench/resources/tests.js:
1727         (makeSteps):
1728
1729 2018-01-30  Antti Koivisto  <antti@apple.com>
1730
1731         StyleBench: Generate more mutation steps
1732         https://bugs.webkit.org/show_bug.cgi?id=182291
1733
1734         Reviewed by Ryosuke Niwa.
1735
1736         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).
1737         This also makes test results more stable and less dependent on possible first-time costs.
1738
1739         * StyleBench/resources/benchmark-runner.js:
1740
1741         Adjust correctionFactor to match the new steps count. With 1.5 Firefox is ~72rpm.
1742
1743         * StyleBench/resources/tests.js:
1744         (makeSteps):
1745
1746         Generate 10 rounds of class/element mutation steps.
1747         Give each generated step unique name so reporting works.
1748         Make individual steps do fewer mutations.
1749
1750 2018-01-29  Antti Koivisto  <antti@apple.com>
1751
1752         StyleBench: Remove : from a test name
1753         https://bugs.webkit.org/show_bug.cgi?id=182258
1754
1755         Reviewed by Ryosuke Niwa.
1756
1757         It breaks reporting.
1758
1759         * StyleBench/resources/style-bench.js:
1760         (nthPseudoClassConfiguration):
1761
1762 2018-01-29  Ryosuke Niwa  <rniwa@webkit.org>
1763
1764         Temporarily skip StyleBench until webkit.org/b/182088 is fixed.
1765
1766         * Skipped:
1767
1768 2018-01-29  Ryosuke Niwa  <rniwa@webkit.org>
1769
1770         Make StyleBench compatible with run-benchmark and run-perf-tests
1771         https://bugs.webkit.org/show_bug.cgi?id=182262
1772
1773         Reviewed by Antti Koivisto.
1774
1775         Copied resource files referenced from Speedometer directory since run-benchmark needs to be able
1776         to checkout each benchmark separately.
1777
1778         Removed the code to create tests of the same name five times in makeSteps since this
1779         won't be compatible with either run-benchmark or run-perf-tests.
1780
1781         * StyleBench/index.html: Removed the code to show warnings for local files since run-benchmark
1782         doesn't use HTTP server in WebDriver mode.
1783         * StyleBench/resources/benchmark-report.js: Copied from resources/benchmark-report.js.
1784         * StyleBench/resources/benchmark-runner.js: Copied from resources/benchmark-runner.js.
1785         (BenchmarkRunner.prototype._finalize): Use the correction factor of 8 instead of 5 in StyleBench.
1786         * StyleBench/resources/gauge.png: Copied from resources/gauge.png.
1787         * StyleBench/resources/gauge@2x.png: Copied from resources/gauge@2x.png.
1788         * StyleBench/resources/main.css: Copied from resources/main.css.
1789         * StyleBench/resources/main.js: Copied from resources/main.js.
1790         * StyleBench/resources/tests.js:
1791         (makeSteps): Only make each test once.
1792
1793 2018-01-26  Chris Nardi  <cnardi@chromium.org>
1794
1795         Addressing post-review comments after r226614
1796         https://bugs.webkit.org/show_bug.cgi?id=182151
1797
1798         Reviewed by Myles C. Maxfield.
1799
1800         * StitchMarker/wtf/text/StringImpl.h:
1801         (WTF::isSpaceOrNewline):
1802         * StitchMarker/wtf/text/TextBreakIterator.cpp:
1803         (WTF::numCodeUnitsInGraphemeClusters):
1804         * StitchMarker/wtf/text/TextBreakIterator.h:
1805
1806 2018-01-22  Antti Koivisto  <antti@apple.com>
1807
1808         StyleBench: Separate test for :nth pseudo classes
1809         https://bugs.webkit.org/show_bug.cgi?id=181932
1810
1811         Reviewed by Sam Weinig.
1812
1813         Optimization strategies for nth are different from other structural pseudo classes (:first-child etc). Move them to a separate test.
1814
1815         * StyleBench/resources/style-bench.js:
1816         (structuralPseudoClassConfiguration):
1817         (nthPseudoClassConfiguration):
1818
1819             Add a new configuration.
1820
1821         (predefinedConfigurations):
1822         (pseudoClassConfiguration): Deleted.
1823
1824 2018-01-17  Ryosuke Niwa  <rniwa@webkit.org>
1825
1826         Temporarily skip Speedometer until we fix the metrics.
1827
1828         * Skipped:
1829
1830 2018-01-16  Alex Christensen  <achristensen@webkit.org>
1831
1832         Fix Mac CMake build after r222900
1833
1834         * MallocBench/MallocBench/CMakeLists.txt:
1835         The mbmalloc target doesn't exist if we're not in developer mode, so only use it if we are.
1836
1837 2018-01-09  Ryosuke Niwa  <rniwa@webkit.org>
1838
1839         Speedometer 2.0: Update the About page.
1840         https://bugs.webkit.org/show_bug.cgi?id=175918
1841
1842         Reviewed by Keith Miller.
1843
1844         Updated the about page for Speedometer 2.0.
1845
1846         * Speedometer/index.html:
1847
1848 == Rolled over to ChangeLog-2018-01-01 ==