Make it so RAMification can be run with python 3 and 2 and that it recognizes the...
[WebKit-https.git] / PerformanceTests / ChangeLog
1 2020-03-26  Saam Barati  <sbarati@apple.com>
2
3         Make it so RAMification can be run with python 3 and 2 and that it recognizes the new JavaScriptCore.framework directory structure
4         https://bugs.webkit.org/show_bug.cgi?id=209636
5
6         Reviewed by Yusuke Suzuki.
7
8         - JavaScriptCore builds now put the jsc shell under JavaScriptCore.framework/Helpers/jsc,
9           not JavaScriptCore.framework/Resources/jsc
10         - It's also helpful to be able to run RAMification with python3.7 and 2.7,
11           since there are some contexts where 3.7 is all we have.
12
13         * JetStream2/RAMification.py:
14         (frameworkPathFromExecutablePath):
15         (BaseRunner.processLine):
16         (LocalRunner.runOneTest):
17         (main.runTestList):
18         (main):
19
20 2020-03-15  Yusuke Suzuki  <ysuzuki@apple.com>
21
22         Should not use variable-length-array (VLA)
23         https://bugs.webkit.org/show_bug.cgi?id=209043
24
25         Reviewed by Mark Lam.
26
27         * MediaTime/Configurations/Base.xcconfig:
28
29 2020-02-17  Don Olmstead  <don.olmstead@sony.com>
30
31         [CMake] Use builtin targets
32         https://bugs.webkit.org/show_bug.cgi?id=205166
33
34         Reviewed by Darin Adler.
35
36         * MallocBench/MallocBench/CMakeLists.txt:
37
38 2020-01-28  Sihui Liu  <sihui_liu@apple.com>
39
40         IndexedDB: speed up index records deletion
41         https://bugs.webkit.org/show_bug.cgi?id=206196
42
43         Reviewed by Brady Eidson.
44
45         * IndexedDB/basic/index-cursor-delete-2.html: Added.
46
47 2020-01-08  Sihui Liu  <sihui_liu@apple.com>
48
49         REGRESSION (r242911?): High Sierra Release WK2 Perf bot timing out while running IndexedDB/large-number-of-inserts.html
50         https://bugs.webkit.org/show_bug.cgi?id=195952
51         <rdar://problem/49137688>
52
53         Re-enable the test after recent IDB performance improvement.
54
55         Reviewed by Alexey Proskuryakov.
56
57         * Skipped:
58
59 2020-01-07  Carlos Garcia Campos  <cgarcia@igalia.com>
60
61         [GTK][WPE] Add API to set purpose and hints of active editable element to input methods
62         https://bugs.webkit.org/show_bug.cgi?id=205605
63
64         Reviewed by Žan Doberšek.
65
66         Split ENABLE_IOS_AUTOCORRECT_AND_AUTOCAPITALIZE and remove the IOS prefix.
67
68         * StitchMarker/wtf/FeatureDefines.h:
69
70 2020-01-06  Mark Lam  <mark.lam@apple.com>
71
72         Convert ASSERT_DISABLED to ASSERT_ENABLED, and fix some tests of NDEBUG that should actually test for ASSERT_ENABLED.
73         https://bugs.webkit.org/show_bug.cgi?id=205776
74
75         Reviewed by Saam Barati.
76
77         * JetStream2/wasm/HashSet.cpp:
78         * StitchMarker/wtf/Assertions.h:
79         * StitchMarker/wtf/DateMath.cpp:
80         (WTF::initializeDates):
81         * StitchMarker/wtf/HashTable.h:
82         * StitchMarker/wtf/Hasher.h:
83         (WTF::StringHasher::addCharacters):
84         * StitchMarker/wtf/NeverDestroyed.h:
85         (WTF::LazyNeverDestroyed::construct):
86         * StitchMarker/wtf/StackBounds.h:
87         (WTF::StackBounds::checkConsistency const):
88         * StitchMarker/wtf/ValueCheck.h:
89         * StitchMarker/wtf/Vector.h:
90         (WTF::minCapacity>::checkConsistency):
91         * StitchMarker/wtf/text/AtomicStringImpl.cpp:
92         * StitchMarker/wtf/text/AtomicStringImpl.h:
93         * StitchMarker/wtf/text/StringCommon.h:
94         (WTF::hasPrefixWithLettersIgnoringASCIICaseCommon):
95         * StitchMarker/wtf/text/StringImpl.h:
96         * StitchMarker/wtf/text/SymbolImpl.h:
97         * StitchMarker/wtf/text/UniquedStringImpl.h:
98
99 2020-01-05  Dean Jackson  <dino@apple.com>
100
101         Rename GraphicsContext3D to GraphicsContextGL
102         https://bugs.webkit.org/show_bug.cgi?id=205778
103         <rdar://problem/58327597>
104
105         Reviewed by Sam Weinig.
106
107         Rename all the GraphicsContext3D things to GraphicsContextGL
108         (includes Extensions3D and GRAPHICS_CONTEXT_3D).
109
110         GraphicsContext3DBase now becomes GraphicsContextGL.
111         GraphicsContext3D is now GraphicsContextGLOpenGL (since it represents
112         the OpenGL implementation of the API).
113
114         * StitchMarker/wtf/FeatureDefines.h:
115         * StitchMarker/wtf/Platform.h:
116
117 2019-12-16  Antti Koivisto  <antti@apple.com>
118
119         Add StyleBench subtest for dynamic media query performance
120         https://bugs.webkit.org/show_bug.cgi?id=205263
121
122         Reviewed by Zalan Bujtas.
123
124         Add a subtest that contains a small number of rules inside min/max-width media queries.
125         The test is executed by resizing the test frame to various widths.
126
127         * StyleBench/index.html:
128         * StyleBench/resources/style-bench.js:
129         (Random.prototype.chance):
130         (defaultConfiguration):
131         (mediaQueryConfiguration):
132         (predefinedConfigurations):
133         * StyleBench/resources/tests.js:
134         (makeSteps):
135
136 2019-10-26  Chris Lord  <clord@igalia.com>
137
138         Put OffscreenCanvas behind a build flag
139         https://bugs.webkit.org/show_bug.cgi?id=203146
140
141         Reviewed by Ryosuke Niwa.
142
143         * StitchMarker/wtf/FeatureDefines.h:
144
145 2019-10-21  Carlos Alberto Lopez Perez  <clopez@igalia.com>
146
147         [GTK] Perf test SVG/UnderTheSeeBenchmark.html timeouts.
148         https://bugs.webkit.org/show_bug.cgi?id=203229
149
150         Unreviewed gardening.
151
152         * Skipped: Mark 2 new tests timing out in the GTK perf bot.
153
154 2019-10-17  Saam Barati  <sbarati@apple.com>
155
156         Have a CLI version of JetStream 2 to run on watchOS
157         https://bugs.webkit.org/show_bug.cgi?id=202998
158         <rdar://problem/56208554>
159
160         Reviewed by Tadeu Zagallo.
161
162         This patch adds a CLI version of JS2 to run on watchOS. We run most subtests
163         from the browser version of JS2, but skip a few tests that are particularly
164         long running, like WSL and some of WTB. We also don't run the Wasm tests
165         as we don't JIT on watchOS. Each test runs for fewer iterations in the watch
166         version. 15 is the default iteration count. This benchmark runs in 7 minutes on
167         a Series 4 watch.
168
169         * JetStream2/JetStreamDriver.js:
170         (getIterationCount):
171         (getWorstCaseCount):
172         (Driver.prototype.async.start):
173         (Driver.prototype.resultsJSON):
174         (Driver.prototype.dumpJSONResultsIfNeeded):
175         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
176         (DefaultBenchmark):
177         * JetStream2/RexBench/UniPoker/benchmark.js:
178         (Benchmark.prototype.validate):
179         (Benchmark):
180         * JetStream2/watch-cli.js: Added.
181
182 2019-10-17  Adrian Perez de Castro  <aperez@igalia.com>
183
184         [Linux] Avoid usage of <sys/sysctl.h> in MallocBench
185         https://bugs.webkit.org/show_bug.cgi?id=203109
186
187         Reviewed by Carlos Alberto Lopez Perez.
188
189         * MallocBench/MallocBench/CPUCount.cpp: Only include <sys/sysctl.h> if
190         the sysctl() function will be used.
191
192 2019-10-17  Carlos Alberto Lopez Perez  <clopez@igalia.com>
193
194         [GTK] Perf test IndexedDB/basic/index-cursor-delete.html timeouts
195         https://bugs.webkit.org/show_bug.cgi?id=203083
196
197         Unreviewed gardening.
198
199         * Skipped: Skip a test that is timing out on the GTK perf bot.
200
201 2019-09-24  Caio Lima  <ticaiolima@gmail.com>
202
203         [BigInt] Add ValueBitRShift into DFG
204         https://bugs.webkit.org/show_bug.cgi?id=192663
205
206         Reviewed by Robin Morisset.
207
208         * BigIntBench/big-int-simple-rshift.js: Added.
209         (bigInt):
210
211 2019-09-09  Jon Lee  <jonlee@apple.com>
212
213         Upload triangles tests for 3D suite in MotionMark
214         https://bugs.webkit.org/show_bug.cgi?id=201606
215
216         Reviewed by Simon Fraser.
217
218         * MotionMark/resources/debug-runner/motionmark.css:
219         * MotionMark/resources/debug-runner/tests.js:
220         * MotionMark/tests/3d/resources/webgl.js:
221         * MotionMark/tests/3d/resources/webgpu.js: Added.
222         * MotionMark/tests/3d/triangles-webgl.html: Renamed from Websites/browserbench.org/MotionMark1.1/tests/3d/webgl.html.
223         * MotionMark/tests/3d/triangles-webgpu.html: Copied from PerformanceTests/MotionMark/tests/3d/webgl.html.
224
225 2019-08-29  Keith Rollin  <krollin@apple.com>
226
227         Update .xcconfig symbols to reflect the current set of past and future product versions.
228         https://bugs.webkit.org/show_bug.cgi?id=200720
229         <rdar://problem/54305032>
230
231         Reviewed by Alex Christensen.
232
233         Remove version symbols related to old OS's we no longer support,
234         ensure that version symbols are defined for OS's we do support.
235
236         * MediaTime/Configurations/Base.xcconfig:
237         * MediaTime/Configurations/DebugRelease.xcconfig:
238
239 2019-08-21  Sihui Liu  <sihui_liu@apple.com>
240
241         Add some IDB performance tests
242         https://bugs.webkit.org/show_bug.cgi?id=200906
243
244         Reviewed by Geoffrey Garen.
245
246         Add test coverage for IDB performance, and organize IDB tests into two categories.
247
248         * IndexedDB/basic/index-count-key.html: Added.
249         * IndexedDB/basic/index-count.html: Added.
250         * IndexedDB/basic/index-cursor-advance.html: Added.
251         * IndexedDB/basic/index-cursor-continue.html: Added.
252         * IndexedDB/basic/index-cursor-continuePrimaryKey.html: Added.
253         * IndexedDB/basic/index-cursor-delete.html: Added.
254         * IndexedDB/basic/index-cursor-update.html: Added.
255         * IndexedDB/basic/index-get.html: Added.
256         * IndexedDB/basic/index-multientry.html: Renamed from PerformanceTests/IndexedDB/index-multientry.html.
257         * IndexedDB/basic/objectstore-add.html: Added.
258         * IndexedDB/basic/objectstore-clear.html: Added.
259         * IndexedDB/basic/objectstore-count.html: Added.
260         * IndexedDB/basic/objectstore-cursor-advance.html: Added.
261         * IndexedDB/basic/objectstore-cursor-continue.html: Added.
262         * IndexedDB/basic/objectstore-cursor-delete.html: Added.
263         * IndexedDB/basic/objectstore-cursor-update.html: Added.
264         * IndexedDB/basic/objectstore-cursor.html: Renamed from PerformanceTests/IndexedDB/objectstore-cursor.html.
265         * IndexedDB/basic/objectstore-delete.html: Renamed from PerformanceTests/IndexedDB/index-get.html.
266         * IndexedDB/basic/objectstore-get.html: Added.
267         * IndexedDB/basic/objectstore-put.html: Renamed from PerformanceTests/IndexedDB/objectstore-get.html.
268         * IndexedDB/stress/large-array-keys.html: Renamed from PerformanceTests/IndexedDB/large-array-keys.html.
269         * IndexedDB/stress/large-binary-keys.html: Renamed from PerformanceTests/IndexedDB/large-binary-keys.html.
270         * IndexedDB/stress/large-number-of-inserts-responsiveness.html: Renamed from PerformanceTests/IndexedDB/large-number-of-inserts-responsiveness.html.
271         * IndexedDB/stress/large-number-of-inserts.html: Renamed from PerformanceTests/IndexedDB/large-number-of-inserts.html.
272         * IndexedDB/stress/large-string-keys.html: Renamed from PerformanceTests/IndexedDB/large-string-keys.html.
273         * Skipped:
274
275 2019-08-19  Sihui Liu  <sihui_liu@apple.com>
276
277         Remove IDB-specific quota
278         https://bugs.webkit.org/show_bug.cgi?id=196545
279         <rdar://problem/54201783>
280
281         Reviewed by Youenn Fablet.
282
283         * IndexedDB/large-binary-keys.html:
284         * IndexedDB/large-string-keys.html:
285
286 2019-08-14  Keith Rollin  <krollin@apple.com>
287
288         Remove support for macOS < 10.13
289         https://bugs.webkit.org/show_bug.cgi?id=200694
290         <rdar://problem/54278851>
291
292         Reviewed by Youenn Fablet.
293
294         Update conditionals that reference __MAC_OS_X_VERSION_MIN_REQUIRED and
295         __MAC_OS_X_VERSION_MAX_ALLOWED, assuming that they both have values >=
296         101300. This means that expressions like
297         "__MAC_OS_X_VERSION_MIN_REQUIRED < 101300" are always False and
298         "__MAC_OS_X_VERSION_MIN_REQUIRED >= 101300" are always True.
299
300         * StitchMarker/wtf/Platform.h:
301         * StitchMarker/wtf/dependencies/bmalloc/BPlatform.h:
302         * StitchMarker/wtf/text/icu/TextBreakIteratorICU.h:
303         (WTF::TextBreakIteratorICU::TextBreakIteratorICU):
304         (WTF::caretRules): Deleted.
305
306 2019-08-06  Michael Saboff  <msaboff@apple.com>
307
308         Make JSC memory benchmark available on OpenSource
309         https://bugs.webkit.org/show_bug.cgi?id=195012
310
311         Reviewed by Saam Barati.
312
313         This patch upstreams the RAMification memory benchmark to open source.
314
315         The prior RAMification.py driver script in the Internal directory has been moved here
316         with the device execution code seperated out to a new Internal driver script.
317         As a result, command line argument parsing as well as acting on the resulting argument
318         values has changed.  The remote device Runner creates an ArgumentParser object that is
319         passed through main().  In main(), if we have an ArgumentParser, we use it otherwise we
320         create one for benchmarking locally.  The arguments object returned from argument
321         parsing is now passed to the Runner constructor so it can extract the settings appropriate
322         for that particular runner.
323
324         Added some sane argument default values for the JSC command patch and the JetStream2
325         source directory.
326
327         Updated the Lua tests to eliminate the testing harness and its overhead, matching what was
328         done in the Internal tree.
329
330         * JetStream2/LuaJSFight/hello_world.js:
331         (Benchmark.prototype.runIteration): Deleted.
332         (Benchmark): Deleted.
333         * JetStream2/LuaJSFight/list_search.js:
334         (run): Deleted.
335         (Benchmark.prototype.runIteration): Deleted.
336         (Benchmark): Deleted.
337         * JetStream2/LuaJSFight/lists.js:
338         (): Deleted.
339         (Benchmark.prototype.runIteration): Deleted.
340         (Benchmark): Deleted.
341         * JetStream2/LuaJSFight/richards.js: Added.
342         (runRichards):
343         (Scheduler):
344         (Scheduler.prototype.addIdleTask):
345         (Scheduler.prototype.addWorkerTask):
346         (Scheduler.prototype.addHandlerTask):
347         (Scheduler.prototype.addDeviceTask):
348         (Scheduler.prototype.addRunningTask):
349         (Scheduler.prototype.addTask):
350         (Scheduler.prototype.schedule):
351         (Scheduler.prototype.release):
352         (Scheduler.prototype.holdCurrent):
353         (Scheduler.prototype.suspendCurrent):
354         (Scheduler.prototype.queue):
355         (TaskControlBlock):
356         (TaskControlBlock.prototype.setRunning):
357         (TaskControlBlock.prototype.markAsNotHeld):
358         (TaskControlBlock.prototype.markAsHeld):
359         (TaskControlBlock.prototype.isHeldOrSuspended):
360         (TaskControlBlock.prototype.markAsSuspended):
361         (TaskControlBlock.prototype.markAsRunnable):
362         (TaskControlBlock.prototype.run):
363         (TaskControlBlock.prototype.checkPriorityAdd):
364         (TaskControlBlock.prototype.toString):
365         (IdleTask):
366         (IdleTask.prototype.run):
367         (IdleTask.prototype.toString):
368         (DeviceTask):
369         (DeviceTask.prototype.run):
370         (DeviceTask.prototype.toString):
371         (WorkerTask):
372         (WorkerTask.prototype.run):
373         (WorkerTask.prototype.toString):
374         (HandlerTask):
375         (HandlerTask.prototype.run):
376         (HandlerTask.prototype.toString):
377         (Packet):
378         (Packet.prototype.addTo):
379         (Packet.prototype.toString):
380         * JetStream2/LuaJSFight/string_lists.js:
381         (): Deleted.
382         (Benchmark.prototype.runIteration): Deleted.
383         (Benchmark): Deleted.
384         * JetStream2/RAMification.py: Added.
385         (mean):
386         (geomean):
387         (frameworkPathFromExecutablePath):
388         (parseArgs):
389         (parseArgs.optStrToBool):
390         (BaseRunner):
391         (BaseRunner.__init__):
392         (BaseRunner.setup):
393         (BaseRunner.setEnv):
394         (BaseRunner.unsetEnv):
395         (BaseRunner.resetForTest):
396         (BaseRunner.processLine):
397         (BaseRunner.setReturnCode):
398         (BaseRunner.getResults):
399         (LocalRunner):
400         (LocalRunner.__init__):
401         (LocalRunner.runOneTest):
402         (main):
403         (main.runTestList):
404         * JetStream2/__init__.py: Added.
405
406 2019-07-23  Keith Rollin  <krollin@apple.com>
407
408         Remove rest of NavigatorContentUtils support
409         https://bugs.webkit.org/show_bug.cgi?id=200052
410         <rdar://problem/53467357>
411
412         Reviewed by Alex Christensen.
413
414         Bug 196070 removes most of the support for the never-used
415         NavigatorContentUtils facility. However, there were still a couple of
416         left-over references after that change. This patch sweeps those up.
417
418         * StitchMarker/wtf/FeatureDefines.h:
419
420 2019-07-12  Caio Lima  <ticaiolima@gmail.com>
421
422         [BigInt] Add ValueBitLShift into DFG
423         https://bugs.webkit.org/show_bug.cgi?id=192664
424
425         Reviewed by Saam Barati.
426
427         * BigIntBench/big-int-simple-lshift.js: Added.
428
429 2019-06-28  Konstantin Tokarev  <annulen@yandex.ru>
430
431         Remove traces of ENABLE_ICONDATABASE remaining after its removal in 219733
432         https://bugs.webkit.org/show_bug.cgi?id=199317
433
434         Reviewed by Michael Catanzaro.
435
436         While IconDatabase and all code using it was removed,
437         ENABLE_ICONDATABASE still exists as build option and C++ macro.
438
439         * StitchMarker/wtf/FeatureDefines.h:
440
441 2019-06-27  Beth Dakin  <bdakin@apple.com>
442
443         Upstream use of MACCATALYST
444         https://bugs.webkit.org/show_bug.cgi?id=199245
445         rdar://problem/51687723
446
447         Reviewed by Tim Horton.
448
449         * MediaTime/Configurations/SDKVariant.xcconfig:
450
451 2019-06-20  Carlos Garcia Campos  <cgarcia@igalia.com>
452
453         [GTK] Remove support for GTK2 plugins
454         https://bugs.webkit.org/show_bug.cgi?id=199065
455
456         Reviewed by Sergio Villar Senin.
457
458         * StitchMarker/wtf/Platform.h:
459
460 2019-05-08  Caio Lima  <ticaiolima@gmail.com>
461
462         [BigInt] Add ValueMod into DFG
463         https://bugs.webkit.org/show_bug.cgi?id=186174
464
465         Reviewed by Saam Barati.
466
467         * BigIntBench/big-int-simple-mod.js: Added.
468
469 2019-05-06  Saam Barati  <sbarati@apple.com>
470
471         JS2 should print scores for different categories
472         https://bugs.webkit.org/show_bug.cgi?id=197605
473
474         Reviewed by Yusuke Suzuki.
475
476         * JetStream2/JetStreamDriver.js:
477         (displayCategoryScores):
478         (isInBrowser.document.onkeydown):
479         (Driver.prototype.async.start):
480         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
481
482 2019-04-16  Jon Lee  <jonlee@apple.com>
483
484         [MotionMark] The text "kkkj" is causing Firefox console warning: `unreachable code after return statement`
485         https://bugs.webkit.org/show_bug.cgi?id=196814
486
487         Reviewed by Wenson Hsieh.
488
489         * MotionMark/resources/extensions.js:
490         (subtract.subtract.sampleY):
491
492 2019-04-15  Myles C. Maxfield  <mmaxfield@apple.com>
493
494         [Cocoa] FontPlatformData objects aren't cached at all when using font-family:system-ui
495         https://bugs.webkit.org/show_bug.cgi?id=196846
496         <rdar://problem/49499971>
497
498         Reviewed by Simon Fraser and Darin Adler.
499
500         * Layout/system-ui-rebuild-emoji.html: Added.
501
502 2019-04-03  Myles C. Maxfield  <mmaxfield@apple.com>
503
504         -apple-trailing-word is needed for browser detection
505         https://bugs.webkit.org/show_bug.cgi?id=196575
506
507         Unreviewed.
508
509         * MotionMark/resources/debug-runner/motionmark.css:
510         (#intro .start-benchmark p):
511
512 2019-03-27  Ryan Haddad  <ryanhaddad@apple.com>
513
514         REGRESSION (r242911?): High Sierra Release WK2 Perf bot timing out while running IndexedDB/large-number-of-inserts.html
515         https://bugs.webkit.org/show_bug.cgi?id=195952
516
517         Unreviewed test gardening.
518
519         * Skipped: Skip the affected test so the bot will finish the run.
520
521 2019-03-27  Saam Barati  <sbarati@apple.com>
522
523         JetStream 2 should not report time values as scores
524         https://bugs.webkit.org/show_bug.cgi?id=196334
525
526         Reviewed by Yusuke Suzuki.
527
528         * JetStream2/JetStreamDriver.js:
529         (toTimeValue):
530         (Driver.prototype.async.reportScoreToRunBenchmarkRunner):
531
532 2019-03-27  Saam barati  <sbarati@apple.com>
533
534         Unreviewed. Fix individual benchmark description urls to go to in-depth.html instead of about.html
535
536         * JetStream2/JetStreamDriver.js:
537         (Driver.prototype.prepareToRun.text.div.id.string_appeared_here.h3):
538
539 2019-03-25  Saam barati  <sbarati@apple.com>
540
541         Update the blurb describing JetStream2 and fix a broken link in the in-depth page.
542
543         Rubber-stamped by Filip Pizlo.
544
545         * JetStream2/in-depth.html:
546         * JetStream2/index.html:
547
548 2019-03-18  Saam Barati  <sbarati@apple.com>
549
550         Add a plan file for JetStream 2
551         https://bugs.webkit.org/show_bug.cgi?id=190734
552
553         Reviewed by Mark Lam.
554
555         Make the delay when starting via run-benchmark longer to allow
556         the page's animation to finish.
557
558         * JetStream2/JetStreamDriver.js:
559         (Driver.prototype.async.initialize):
560
561 2019-03-15  Jon Davis  <jond@apple.com>
562
563         Fixed a bug preventing the overall sore to be displayed on JetStream2
564         https://bugs.webkit.org/show_bug.cgi?id=195817
565
566         Rubber-stamped by Saam Barati.
567
568         * JetStream2/JetStreamDriver.js:
569         (Driver.prototype.async.start):
570
571 2019-03-14  Jon Davis  <jond@apple.com>
572
573         Implement a new design for JetStream2
574         https://bugs.webkit.org/show_bug.cgi?id=195492
575
576         Reviewed by Saam Barati.
577
578         * JetStream2/JetStream.css:
579         (html):
580         (body):
581         (::selection):
582         (main):
583         (.logo):
584         (.logo .logo-image):
585         (#jetstreams):
586         (#jetstreams svg):
587         (.summary):
588         (.summary + .summary):
589         (.summary:empty):
590         (article, .summary):
591         (h1):
592         (h2, h3, h4, h5, h6):
593         (h4, h5, h6):
594         (p):
595         (h5, h6):
596         (h6):
597         (a:link,):
598         (a:hover,):
599         (#status):
600         (#status label,):
601         (a.button):
602         (#status.loading):
603         (#status.error):
604         (.error h2, .error p):
605         (.error h2):
606         (#result-summary):
607         (#result-summary label):
608         (#result-summary .score):
609         (#result-summary .score .interval):
610         (#results):
611         (.benchmark):
612         (.benchmark h3, .benchmark h4, .benchmark .result, .benchmark label):
613         (.benchmark-running h4, .benchmark-running .result, .benchmark-running label):
614         (.benchmark-done h3, .benchmark-done h4, .benchmark-done .result, .benchmark-done label):
615         (.benchmark h3):
616         (.benchmark-running h3):
617         (.benchmark-done h3):
618         (.benchmark h3 a,):
619         (.benchmark-done h3 a:hover):
620         (.benchmark h4):
621         (.benchmark-done h4):
622         (.benchmark p,):
623         (.benchmark .result):
624         (.benchmark-done .result):
625         (.benchmark label):
626         (.benchmark-done label):
627         (@keyframes fadein):
628         (to):
629         (@keyframes scaledown):
630         (@keyframes shine):
631         (100%):
632         (@keyframes swingin):
633         (@media (max-width: 415px)):
634         (#logo): Deleted.
635         (p.summary): Deleted.
636         (p.summary + p.summary): Deleted.
637         (p.summary:empty): Deleted.
638         (p:first-of-type): Deleted.
639         (#status a:link): Deleted.
640         (.interval): Deleted.
641         (a:link, a:visited): Deleted.
642         (a:link:hover): Deleted.
643         (table): Deleted.
644         (td, th): Deleted.
645         (th): Deleted.
646         (tr:first-child > th:nth-child(even)): Deleted.
647         (tr:nth-child(even):not(:first-child, .benchmark-running)): Deleted.
648         (.result): Deleted.
649         (.benchmark-running): Deleted.
650         (:not(.benchmark-running) .result): Deleted.
651         (.benchmark-running .result): Deleted.
652         (.benchmark-name): Deleted.
653         (.result.category,): Deleted.
654         (.benchmark-name:not(.category):not(.geometric-mean)): Deleted.
655         (.geometric-mean): Deleted.
656         (.benchmark-name:not(:first-child)): Deleted.
657         (.benchmark-name a:link,): Deleted.
658         (.highlighted-result): Deleted.
659         * JetStream2/JetStream2Logo.svg: Added.
660         * JetStream2/JetStreamDriver.js:
661         (updateUI):
662         (Driver.prototype.async.start):
663         (Driver.prototype.prepareToRun.text.div.id.string_appeared_here.h3):
664         (Driver.prototype.prepareToRun.text.span):
665         (Driver.prototype.prepareToRun.a.h3.h4):
666         (Driver.prototype.prepareToRun.text.div):
667         (Driver.prototype.async.initialize):
668         (Driver.prototype.async.fetchResources.prototype.statusElement.onclick):
669         (prototype.fetchResources):
670         (prototype.updateUIBeforeRun):
671         (prototype.updateUIAfterRun):
672         (Driver.prototype.async.fetchResources.statusElement.innerHTML.a.href.string_appeared_here):
673         (Driver.prototype.async.fetchResources):
674         (Driver.prototype.prepareToRun.text.tr.id.string_appeared_here.FIXME.link.to.benchmark.explanation.td): Deleted.
675         * JetStream2/clouds.svg: Added.
676         * JetStream2/in-depth.html:
677         * JetStream2/index.html:
678         * JetStream2/jetstreams.svg: Added.
679
680 2019-03-11  Caio Lima  <ticaiolima@gmail.com>
681
682         [ESNext][BigInt] Implement "~" unary operation
683         https://bugs.webkit.org/show_bug.cgi?id=182216
684
685         Reviewed by Keith Miller.
686
687         * BigIntBench/big-int-simple-bit-not.js: Added.
688
689 2019-03-04  Saam Barati  <sbarati@apple.com>
690
691         Add a detailed summary page for JetStream 2
692         https://bugs.webkit.org/show_bug.cgi?id=195014
693
694         Rubber-stamped by Darin Adler.
695
696         * JetStream2/JetStream.css:
697         (a:link, a:visited):
698         * JetStream2/in-depth.html: Added.
699
700 2019-02-08  Saam barati  <sbarati@apple.com>
701
702         Update JetStream2 CLI to not print "ms" since we now just print the score values
703         https://bugs.webkit.org/show_bug.cgi?id=193981
704
705         Reviewed by Yusuke Suzuki.
706
707         * JetStream2/JetStreamDriver.js:
708         (DefaultBenchmark.prototype.updateUIAfterRun):
709         (DefaultBenchmark):
710         (WSLBenchmark.prototype.updateUIAfterRun):
711         (WSLBenchmark):
712         (WasmBenchmark.prototype.updateUIAfterRun):
713         (WasmBenchmark):
714
715 2019-01-31  Sihui Liu  <sihui_liu@apple.com>
716
717         REGRESSION (r240358): IndexedDB/large-binary-keys.html and IndexedDB/large-string-keys.html perf tests failing
718         https://bugs.webkit.org/show_bug.cgi?id=193956
719         <rdar://problem/47623109>
720
721         Reviewed by Ryosuke Niwa.
722
723         r240358 sets default IDB quota as 50 MB for single test, we need to increase that for big tests. 
724
725         * IndexedDB/large-binary-keys.html:
726         * IndexedDB/large-string-keys.html:
727
728 2019-01-18  Saam Barati  <sbarati@apple.com>
729
730         Use scores everywhere in JetStream2's UI
731         https://bugs.webkit.org/show_bug.cgi?id=193593
732
733         Reviewed by Keith Miller.
734
735         * JetStream2/JetStreamDriver.js:
736         (DefaultBenchmark.prototype.processResults):
737         (DefaultBenchmark.prototype.get score):
738         (DefaultBenchmark.scoreDescription):
739         (WSLBenchmark.prototype.processResults):
740         (WSLBenchmark.prototype.get score):
741         (WSLBenchmark.scoreDescription):
742         (WasmBenchmark.prototype.processResults):
743         (WasmBenchmark.prototype.get score):
744         (WasmBenchmark.scoreDescription):
745         * JetStream2/index.html:
746
747 2019-01-18  Jer Noble  <jer.noble@apple.com>
748
749         SDK_VARIANT build destinations should be separate from non-SDK_VARIANT builds
750         https://bugs.webkit.org/show_bug.cgi?id=189553
751
752         Reviewed by Tim Horton.
753
754         * MediaTime/Configurations/Base.xcconfig:
755         * MediaTime/Configurations/SDKVariant.xcconfig: Added.
756
757 2019-01-17  Truitt Savell  <tsavell@apple.com>
758
759         Unreviewed, rolling out r240124.
760
761         This commit broke an internal build.
762
763         Reverted changeset:
764
765         "SDK_VARIANT build destinations should be separate from non-
766         SDK_VARIANT builds"
767         https://bugs.webkit.org/show_bug.cgi?id=189553
768         https://trac.webkit.org/changeset/240124
769
770 2019-01-17  Jer Noble  <jer.noble@apple.com>
771
772         SDK_VARIANT build destinations should be separate from non-SDK_VARIANT builds
773         https://bugs.webkit.org/show_bug.cgi?id=189553
774
775         Reviewed by Tim Horton.
776
777         * MediaTime/Configurations/Base.xcconfig:
778         * MediaTime/Configurations/SDKVariant.xcconfig: Added.
779
780 2018-12-20  Caio Lima  <ticaiolima@gmail.com>
781
782         [BigInt] We should enable CSE into arithmetic operations that speculate BigIntUse
783         https://bugs.webkit.org/show_bug.cgi?id=192723
784
785         Reviewed by Yusuke Suzuki.
786
787         * BigIntBench/big-int-cse.js: Added.
788         * BigIntBench/big-int-global-cse.js: Added.
789         * BigIntBench/big-int-licm.js: Added.
790
791 2018-12-19  Commit Queue  <commit-queue@webkit.org>
792
793         Unreviewed, rolling out r239377.
794         https://bugs.webkit.org/show_bug.cgi?id=192921
795
796         broke 32-bit JSC tests (Requested by keith_miller on #webkit).
797
798         Reverted changeset:
799
800         "[BigInt] We should enable CSE into arithmetic operations that
801         speculate BigIntUse"
802         https://bugs.webkit.org/show_bug.cgi?id=192723
803         https://trac.webkit.org/changeset/239377
804
805 2018-12-19  Caio Lima  <ticaiolima@gmail.com>
806
807         [BigInt] We should enable CSE into arithmetic operations that speculate BigIntUse
808         https://bugs.webkit.org/show_bug.cgi?id=192723
809
810         Reviewed by Saam Barati.
811
812         * BigIntBench/big-int-cse.js: Added.
813         * BigIntBench/big-int-global-cse.js: Added.
814         * BigIntBench/big-int-licm.js: Added.
815
816 2018-12-17  Suresh Koppisetty  <skoppisetty@apple.com>
817
818         Add "-o/--output" option to startup.py and new_tab.py benchmark scripts to save the results in json format.
819         https://bugs.webkit.org/show_bug.cgi?id=192385
820
821         Reviewed by Ryosuke Niwa.
822
823         Sample json output for new tab benchmark script after running for 2 iterations and 2 groups. Values are in milliseconds.
824         {
825                 "NewTabBenchmark": {
826                         "metrics": {
827                                 "Time": {
828                                         "current": [
829                                                 [
830                                                         410.2939453125,
831                                                         307.81494140625
832                                                 ],
833                                                 [
834                                                         340.616943359375,
835                                                         265.94384765625
836                                                 ]
837                                         ]
838                                 }
839                         }
840                 }
841         }
842
843         Sample json output for startup time benchmark script after running for 2 iterations. Values are in milliseconds.
844         {
845                 "StartupBenchmark": {
846                         "metrics": {
847                                 "Time": {
848                                         "current": [
849                                                 [
850                                                         1415.2099609375,
851                                                         1439.552978515625
852                                                 ]
853                                         ]
854                                 }
855                         }
856                 }
857         }
858
859         * LaunchTime/launch_time.py:
860         * LaunchTime/new_tab.py:
861         (NewTabBenchmark.get_test_name):
862         * LaunchTime/startup.py:
863         (StartupBenchmark.get_test_name):
864
865 2018-12-17  Suresh Koppisetty  <skoppisetty@apple.com>
866
867         Import FeedbackServer only if "-f/--feedback-in-browser" option is enabled.
868         https://bugs.webkit.org/show_bug.cgi?id=192378
869
870         Reviewed by Ryosuke Niwa.
871
872         FeedbackServer currently depends on Tornado-5.1, which further adds
873         dependency of "singledispatch", "backports-abc" and "futures" python libraries.
874         Importing FeedbackServer only if "-f/--feedback-in-browser" option is enabled
875         will let us run the benchmark scripts without installing any new python libraries.
876
877         * LaunchTime/launch_time.py:
878         * LaunchTime/new_tab.py:
879
880 2018-12-13  Caio Lima  <ticaiolima@gmail.com>
881
882         [BigInt] Add ValueDiv into DFG
883         https://bugs.webkit.org/show_bug.cgi?id=186178
884
885         Reviewed by Yusuke Suzuki.
886
887         * BigIntBench/big-int-simple-div.js: Added.
888         * BigIntBench/value-div-type-propagation.js: Added.
889
890 2018-12-10  Caio Lima  <ticaiolima@gmail.com>
891
892         [BigInt] Add ValueMul into DFG
893         https://bugs.webkit.org/show_bug.cgi?id=186175
894
895         Reviewed by Yusuke Suzuki.
896
897         * BigIntBench/big-int-simple-mul.js: Added.
898         * BigIntBench/value-mul-type-propagation.js: Added.
899
900 2018-11-07  Caio Lima  <ticaiolima@gmail.com>
901
902         [BigInt] Add support to BigInt into ValueAdd
903         https://bugs.webkit.org/show_bug.cgi?id=186177
904
905         Reviewed by Keith Miller.
906
907         The idea of BigIntBench is to provide a set of microbenchmarks and
908         benchmarks to evaluate how fast BigInt computations are happening on
909         JSC implementation.
910
911         Now, we are adding microbenchmarks in this set,
912         but the plan is to move these tests to "JSTest/microbenchmarks" when
913         BigInt is enabled by default. After that, the focus of Bigint bench is
914         to provide a set of tests that represents real use cases of BigInt in
915         JS programs.
916
917         * BigIntBench/big-int-add-prediction-propagation.js: Added.
918         * BigIntBench/big-int-simple-add.js: Added.
919         * BigIntBench/big-int-simple-sub.js: Added.
920
921 2018-11-07  Tadeu Zagallo  <tzagallo@apple.com>
922
923         REGRESSION(r237547): Test failures on 32-bit JSC since the JIT was disabled
924         https://bugs.webkit.org/show_bug.cgi?id=191184
925
926         Reviewed by Saam Barati.
927
928         Skip test when the JIT is disabled, it takes too long to run on CLoop.
929
930         * JetStream/cdjs/main.js:
931
932 2018-11-05  Myles C. Maxfield  <mmaxfield@apple.com>
933
934         Cache glyph paths and share underline skipping code between all the ports
935         https://bugs.webkit.org/show_bug.cgi?id=191239
936
937         Reviewed by Alex Christensen.
938
939         Measures the performance of drawing a whole lot of underlines
940
941         * Layout/underline.html: Added.
942
943 2018-10-25  Saam Barati  <sbarati@apple.com>
944
945         Check in corresponding C code in JetStream 2
946         https://bugs.webkit.org/show_bug.cgi?id=190733
947
948         Rubber-stamped by Filip Pizlo.
949
950         * JetStream2/simple/float-mm.c: Added.
951         * JetStream2/wasm/HashSet.cpp: Added.
952         * JetStream2/wasm/gcc-loops.cpp: Added.
953         * JetStream2/wasm/quicksort.c: Added.
954         * JetStream2/wasm/TSF: Added.
955         * JetStream2/wasm/TSF/build.sh: Added.
956         * JetStream2/wasm/TSF/config.h: Added.
957         * JetStream2/wasm/TSF/gpc.h: Added.
958         * JetStream2/wasm/TSF/gpc_code_gen_util.c: Added.
959         * JetStream2/wasm/TSF/gpc_instruction.c: Added.
960         * JetStream2/wasm/TSF/gpc_instruction_dispatch.gen: Added.
961         * JetStream2/wasm/TSF/gpc_instruction_size.gen: Added.
962         * JetStream2/wasm/TSF/gpc_instruction_stack_effects.gen: Added.
963         * JetStream2/wasm/TSF/gpc_instruction_static_size.gen: Added.
964         * JetStream2/wasm/TSF/gpc_instruction_to_string.gen: Added.
965         * JetStream2/wasm/TSF/gpc_int_common.h: Added.
966         * JetStream2/wasm/TSF/gpc_intable.c: Added.
967         * JetStream2/wasm/TSF/gpc_internal.h: Added.
968         * JetStream2/wasm/TSF/gpc_interpreter.gen: Added.
969         * JetStream2/wasm/TSF/gpc_program.c: Added.
970         * JetStream2/wasm/TSF/gpc_proto.c: Added.
971         * JetStream2/wasm/TSF/gpc_stack_height.c: Added.
972         * JetStream2/wasm/TSF/gpc_threaded.c: Added.
973         * JetStream2/wasm/TSF/gpc_worklist.h: Added.
974         * JetStream2/wasm/TSF/tsf.h: Added.
975         * JetStream2/wasm/TSF/tsf_adaptive_reader.c: Added.
976         * JetStream2/wasm/TSF/tsf_asprintf.c: Added.
977         * JetStream2/wasm/TSF/tsf_atomics.h: Added.
978         * JetStream2/wasm/TSF/tsf_buf_reader.c: Added.
979         * JetStream2/wasm/TSF/tsf_buf_writer.c: Added.
980         * JetStream2/wasm/TSF/tsf_buffer.c: Added.
981         * JetStream2/wasm/TSF/tsf_build_defines.h: Added.
982         * JetStream2/wasm/TSF/tsf_config.h: Added.
983         * JetStream2/wasm/TSF/tsf_config_stub.h: Added.
984         * JetStream2/wasm/TSF/tsf_copier.c: Added.
985         * JetStream2/wasm/TSF/tsf_define_helpers.c: Added.
986         * JetStream2/wasm/TSF/tsf_define_helpers.h: Added.
987         * JetStream2/wasm/TSF/tsf_destructor.c: Added.
988         * JetStream2/wasm/TSF/tsf_error.c: Added.
989         * JetStream2/wasm/TSF/tsf_format.h: Added.
990         * JetStream2/wasm/TSF/tsf_fsdb.c: Added.
991         * JetStream2/wasm/TSF/tsf_fsdb_protocol.c: Added.
992         * JetStream2/wasm/TSF/tsf_fsdb_protocol.h: Added.
993         * JetStream2/wasm/TSF/tsf_generator.c: Added.
994         * JetStream2/wasm/TSF/tsf_gpc_code_gen.c: Added.
995         * JetStream2/wasm/TSF/tsf_indent.h: Added.
996         * JetStream2/wasm/TSF/tsf_internal.h: Added.
997         * JetStream2/wasm/TSF/tsf_internal_config.h: Added.
998         * JetStream2/wasm/TSF/tsf_internal_config_stub.h: Added.
999         * JetStream2/wasm/TSF/tsf_inttypes.h: Added.
1000         * JetStream2/wasm/TSF/tsf_io.c: Added.
1001         * JetStream2/wasm/TSF/tsf_io_utils.c: Added.
1002         * JetStream2/wasm/TSF/tsf_ir.c: Added.
1003         * JetStream2/wasm/TSF/tsf_ir.h: Added.
1004         * JetStream2/wasm/TSF/tsf_ir_different.c: Added.
1005         * JetStream2/wasm/TSF/tsf_ir_different.h: Added.
1006         * JetStream2/wasm/TSF/tsf_ir_speed.c: Added.
1007         * JetStream2/wasm/TSF/tsf_limits.c: Added.
1008         * JetStream2/wasm/TSF/tsf_named_type.c: Added.
1009         * JetStream2/wasm/TSF/tsf_native.c: Added.
1010         * JetStream2/wasm/TSF/tsf_parser.c: Added.
1011         * JetStream2/wasm/TSF/tsf_primitive.c: Added.
1012         * JetStream2/wasm/TSF/tsf_ra_type_man.c: Added.
1013         * JetStream2/wasm/TSF/tsf_reflect.c: Added.
1014         * JetStream2/wasm/TSF/tsf_region.h: Added.
1015         * JetStream2/wasm/TSF/tsf_serial_in_man.c: Added.
1016         * JetStream2/wasm/TSF/tsf_serial_out_man.c: Added.
1017         * JetStream2/wasm/TSF/tsf_serial_protocol.h: Added.
1018         * JetStream2/wasm/TSF/tsf_sha1.c: Added.
1019         * JetStream2/wasm/TSF/tsf_sha1.h: Added.
1020         * JetStream2/wasm/TSF/tsf_sha1_writer.c: Added.
1021         * JetStream2/wasm/TSF/tsf_sort.c: Added.
1022         * JetStream2/wasm/TSF/tsf_st.c: Added.
1023         * JetStream2/wasm/TSF/tsf_st.h: Added.
1024         * JetStream2/wasm/TSF/tsf_st_typetable.c: Added.
1025         * JetStream2/wasm/TSF/tsf_stream_file_input.c: Added.
1026         * JetStream2/wasm/TSF/tsf_stream_file_output.c: Added.
1027         * JetStream2/wasm/TSF/tsf_type.c: Added.
1028         * JetStream2/wasm/TSF/tsf_type_in_map.c: Added.
1029         * JetStream2/wasm/TSF/tsf_type_out_map.c: Added.
1030         * JetStream2/wasm/TSF/tsf_type_table.c: Added.
1031         * JetStream2/wasm/TSF/tsf_types.h: Added.
1032         * JetStream2/wasm/TSF/tsf_util.h: Added.
1033         * JetStream2/wasm/TSF/tsf_version.c: Added.
1034         * JetStream2/wasm/TSF/tsf_zip_abstract.c: Added.
1035         * JetStream2/wasm/TSF/tsf_zip_abstract.h: Added.
1036         * JetStream2/wasm/TSF/tsf_zip_attr.c: Added.
1037         * JetStream2/wasm/TSF/tsf_zip_reader.c: Added.
1038         * JetStream2/wasm/TSF/tsf_zip_writer.c: Added.
1039
1040 2018-10-18  Saam Barati  <sbarati@apple.com>
1041
1042         Make JetStream 2
1043         https://bugs.webkit.org/show_bug.cgi?id=187829
1044
1045         Rubber-stamped by Mark Lam.
1046
1047         This patch checks in the new JetStream 2 benchmark. JetStream 2 is
1048         a new JavaScript and Web Assembly benchmark. JetStream 2's goal
1049         is to measure the startup, worst case, and peak throughput performance
1050         of the JavaScript engine. JetStream 2 incorporates these previous
1051         benchmarks:
1052         - JetStream
1053         - ARES-6
1054         - Kraken
1055         - Web Tooling Benchmark
1056         - WasmBench
1057         - RexBench
1058         
1059         JetStream 2 also adds some new benchmarks:
1060         - Two tests emphasizing web worker performance.
1061         - One test emphasizing Promise, async iteration, and DataView performance.
1062         - Two new code load tests.
1063         - WSL: a test measuring all kinds of things, especially emphasizing exception performance.
1064
1065         * JetStream2: Added.
1066
1067 2018-10-02  Chris Dumez  <cdumez@apple.com>
1068
1069         Regression(r236613): Parser/html-parser.html performance test is failing
1070         https://bugs.webkit.org/show_bug.cgi?id=190201
1071
1072         Unreviewed, apply the same fix as in r236699 to make sure the iframe is treated as
1073         same origin, so that calls to document.open() / document.write() succeed.
1074
1075         * Parser/html-parser.html:
1076
1077 2018-10-01  Chris Dumez  <cdumez@apple.com>
1078
1079         Regression(r236613): Parser/HTML5-8266-ParseOnly.html performance tests no longer runs
1080         https://bugs.webkit.org/show_bug.cgi?id=190174
1081
1082         Reviewed by Ryosuke Niwa.
1083
1084         Update iframe's sandbox to use 'allow-same-origin' so that the iframe is treated as same-origin and so that
1085         the top frame can call document.open() / document.write() on the subframe. This test was also failing in
1086         Firefox without this change.
1087
1088         * Parser/HTML5-8266-ParseOnly.html:
1089
1090 2018-08-27  Aditya Keerthi  <akeerthi@apple.com>
1091
1092         Consolidate ENABLE_INPUT_TYPE_COLOR and ENABLE_INPUT_TYPE_COLOR_POPOVER
1093         https://bugs.webkit.org/show_bug.cgi?id=188931
1094
1095         Reviewed by Wenson Hsieh.
1096
1097         * StitchMarker/wtf/FeatureDefines.h: Removed ENABLE_INPUT_TYPE_COLOR_POPOVER.
1098
1099 2018-08-23  Don Olmstead  <don.olmstead@sony.com>
1100
1101         [CMake] Have checks are not getting set inside CMake properly
1102         https://bugs.webkit.org/show_bug.cgi?id=188901
1103
1104         Reviewed by Michael Catanzaro.
1105
1106         Add notification that MallocBench is disabled.
1107
1108         * CMakeLists.txt:
1109
1110 2018-08-23  Don Olmstead  <don.olmstead@sony.com>
1111
1112         [CMake] Add HAVE_MALLOC_TRIM definition
1113         https://bugs.webkit.org/show_bug.cgi?id=188897
1114
1115         Reviewed by Konstantin Tokarev.
1116
1117         MallocBench should only be built on Apple platforms and platforms that
1118         have malloc_trim.
1119
1120         * CMakeLists.txt:
1121
1122 2018-08-13  Jon Lee  <jonlee@apple.com>
1123
1124         Update MotionMark version number
1125         https://bugs.webkit.org/show_bug.cgi?id=188535
1126         <rdar://problem/43254078>
1127
1128         Reviewed by Said Abou-Hallawa.
1129
1130         * MotionMark/resources/strings.js: To 1.1.
1131
1132 2018-08-13  Jon Lee  <jonlee@apple.com>
1133
1134         [MotionMark] Update Multiply test
1135         https://bugs.webkit.org/show_bug.cgi?id=188532
1136         <rdar://problem/43252151>
1137
1138         Reviewed by Said Abou-Hallawa.
1139
1140         Update Multiply test to have a larger upper limit of number of particles. Cycle through three
1141         different ways of hiding an element. Add new Multiple suite that isolates those components for
1142         debugging purposes.
1143
1144         * MotionMark/resources/debug-runner/tests.js:
1145         * MotionMark/tests/dom/multiply.html: Added.
1146         * MotionMark/tests/dom/resources/multiply.js: Added.
1147         * MotionMark/tests/master/resources/multiply.js:
1148
1149 2018-08-13  Jon Lee  <jonlee@apple.com>
1150
1151         [MotionMark] Update Leaves test
1152         https://bugs.webkit.org/show_bug.cgi?id=188530
1153         <rdar://problem/43251862>
1154
1155         Reviewed by Said Abou-Hallawa.
1156
1157         Update Leaves test to include opacity and scale. Add new Leaves suite that isolate those components
1158         for debugging purposes.
1159
1160         * MotionMark/resources/debug-runner/tests.js:
1161         * MotionMark/tests/dom/leaves.html:
1162         * MotionMark/tests/dom/resources/leaves.js:
1163         * MotionMark/tests/master/resources/leaves.js:
1164
1165 2018-08-10  Ben Richards  <benton_richards@apple.com>
1166
1167         Add ability to ignore process prewarming for launch time benchmark
1168         https://bugs.webkit.org/show_bug.cgi?id=188462
1169
1170         Reviewed by Ryosuke Niwa.
1171
1172         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.
1173
1174         * LaunchTime/launch_time.py:
1175         (DefaultLaunchTimeHandler):
1176         (DefaultLaunchTimeHandler.on_receive_stop_time): Deleted.
1177         (DefaultLaunchTimeHandler.on_receive_stop_signal):
1178         (DefaultLaunchTimeHandler.do_HEAD):
1179         (DefaultLaunchTimeHandler.do_GET):
1180         (DefaultLaunchTimeHandler.do_POST):
1181         (LaunchTimeBenchmark):
1182         (LaunchTimeBenchmark._standard_deviation): Fixed divide by zero bug when '-n' is set to 1
1183         (LaunchTimeBenchmark.open_tab): Added option to open a blank tab
1184         (LaunchTimeBenchmark.run):
1185         * LaunchTime/new_tab.py:
1186         (NewTabBenchmark.initialize):
1187         (NewTabBenchmark.run_iteration):
1188         (NewTabBenchmark.will_parse_arguments):
1189         (NewTabBenchmark.did_parse_arguments):
1190         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1191         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1192         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1193         (NewTabBenchmark):
1194         * LaunchTime/startup.py:
1195         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1196         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1197         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1198         (StartupBenchmark):
1199
1200 2018-08-10  Truitt Savell  <tsavell@apple.com>
1201
1202         Unreviewed, rolling out r234750.
1203
1204         Caused 185 perf test failures.
1205
1206         Reverted changeset:
1207
1208         "Add ability to ignore process prewarming for launch time
1209         benchmark"
1210         https://bugs.webkit.org/show_bug.cgi?id=188462
1211         https://trac.webkit.org/changeset/234750
1212
1213 2018-08-09  Ben Richards  <benton_richards@apple.com>
1214
1215         Add ability to ignore process prewarming for launch time benchmark
1216         https://bugs.webkit.org/show_bug.cgi?id=188462
1217
1218         Reviewed by Ryosuke Niwa.
1219
1220         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.
1221
1222         * LaunchTime/launch_time.py:
1223         (DefaultLaunchTimeHandler):
1224         (DefaultLaunchTimeHandler.on_receive_stop_time): Deleted.
1225         (DefaultLaunchTimeHandler.on_receive_stop_signal):
1226         (DefaultLaunchTimeHandler.do_HEAD):
1227         (DefaultLaunchTimeHandler.do_GET):
1228         (DefaultLaunchTimeHandler.do_POST):
1229         (LaunchTimeBenchmark):
1230         (LaunchTimeBenchmark._standard_deviation): Fixed divide by zero bug when '-n' is set to 1
1231         (LaunchTimeBenchmark.open_tab): Added option to open a blank tab
1232         (LaunchTimeBenchmark.run):
1233         * LaunchTime/new_tab.py:
1234         (NewTabBenchmark.initialize):
1235         (NewTabBenchmark.run_iteration):
1236         (NewTabBenchmark.will_parse_arguments):
1237         (NewTabBenchmark.did_parse_arguments):
1238         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1239         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1240         (NewTabBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1241         (NewTabBenchmark):
1242         * LaunchTime/startup.py:
1243         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1244         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_time): Deleted.
1245         (StartupBenchmark.ResponseHandler.Handler.on_receive_stop_signal):
1246         (StartupBenchmark):
1247
1248 2018-07-26  Andy VanWagoner  <andy@vanwagoner.family>
1249
1250         [INTL] Remove INTL sub-feature compile flags
1251         https://bugs.webkit.org/show_bug.cgi?id=188081
1252
1253         Reviewed by Michael Catanzaro.
1254
1255         Removed old and unused i18n flag.
1256
1257         * StitchMarker/wtf/FeatureDefines.h:
1258
1259 2018-07-20  David Fenton  <david_fenton@apple.com>
1260
1261         REGRESSION (234006) ERROR: Blocked access to external URL http://code.jquery.com/jquery-1.9.1.min.js
1262         https://bugs.webkit.org/show_bug.cgi?id=187858
1263
1264         Unreviewed test gardening, skipped test.
1265
1266         * Skipped: LaunchTime
1267
1268 2018-07-19  Ben Richards  <benton_richards@apple.com>
1269
1270         Add benchmark for WebKit process launch times
1271         https://bugs.webkit.org/show_bug.cgi?id=186414
1272
1273         Reviewed by Ryosuke Niwa.
1274
1275         Added two benchmarks, one for measuring browser new tab launch time and one for browser startup time.
1276
1277         * LaunchTime/.gitignore: Added.
1278         * LaunchTime/feedback_client.html: Added.
1279         Displays benchmark progress in browser
1280         * LaunchTime/feedback_server.py: Added.
1281         (FeedbackServer): Sends data to feedback_client via websocket
1282         (FeedbackServer.__init__):
1283         (FeedbackServer._create_app):
1284         (FeedbackServer._start_server):
1285         (FeedbackServer._send_all_messages):
1286         (FeedbackServer.start):
1287         (FeedbackServer.stop):
1288         (FeedbackServer.send_message): Send a message to the feedback_client
1289         (FeedbackServer.wait_until_client_has_loaded): Wait until the feedback_client has opened a websocket connection to the feedback_server
1290         (FeedbackServer.MainHandler): Handler factory to create handler that serves feedback_client.html
1291         (FeedbackServer.MainHandler.Handler):
1292         (FeedbackServer.MainHandler.Handler.get):
1293         (FeedbackServer.WSHandler): Handler factory to create handler that sends data to feedback client
1294         (FeedbackServer.WSHandler.Handler):
1295         (FeedbackServer.WSHandler.Handler.open): On websocket connection opened
1296         (FeedbackServer.WSHandler.Handler.on_close): On websocket connection closed
1297         * LaunchTime/launch_time.py: Added.
1298         (DefaultLaunchTimeHandler): Abstract HTTP request handler for launch time benchmarks
1299         (DefaultLaunchTimeHandler.get_test_page): Default test page to be overridden by benchmarks
1300         (DefaultLaunchTimeHandler.get_blank_page):
1301         (DefaultLaunchTimeHandler.on_receive_stop_time):
1302         (DefaultLaunchTimeHandler.do_HEAD):
1303         (DefaultLaunchTimeHandler.do_GET):
1304         (DefaultLaunchTimeHandler.do_POST):
1305         (DefaultLaunchTimeHandler.log_message): Suppresses HTTP logs from SimpleHTTPRequestHandler
1306         (LaunchTimeBenchmark): Abstract class which launch time benchmarks inherit from and override methods desired to customize
1307         (LaunchTimeBenchmark.__init__):
1308         (LaunchTimeBenchmark._parse_browser_bundle_path): Parser for bundle path option
1309         (LaunchTimeBenchmark._parse_args):
1310         (LaunchTimeBenchmark._run_server): Target function for main server thread
1311         (LaunchTimeBenchmark._setup_servers):
1312         (LaunchTimeBenchmark._clean_up):
1313         (LaunchTimeBenchmark._exit_due_to_exception):
1314         (LaunchTimeBenchmark._geometric_mean):
1315         (LaunchTimeBenchmark._standard_deviation):
1316         (LaunchTimeBenchmark._compute_results): Returns mean and std dev of list of results and pretty prints if should_print=True is specified
1317         (LaunchTimeBenchmark._wait_times): Mimic numpy.linspace
1318         (LaunchTimeBenchmark.open_tab): Open a browser tab with the html given by self.response_handler.get_test_page
1319         (LaunchTimeBenchmark.launch_browser): Open a broser to either the feedback client (if option is set) or a blank page
1320         (LaunchTimeBenchmark.quit_browser):
1321         (LaunchTimeBenchmark.quit_browser.quit_app):
1322         (LaunchTimeBenchmark.quit_browser.is_app_closed):
1323         (LaunchTimeBenchmark.close_tab):
1324         (LaunchTimeBenchmark.wait):
1325         (LaunchTimeBenchmark.log): Print to console and send to feedback client if --feedback-in-browser flag is used
1326         (LaunchTimeBenchmark.log_verbose): Only logs if --verbose flag is used
1327         (LaunchTimeBenchmark.run):
1328         (LaunchTimeBenchmark.group_init): Initialization done before each round of iterations
1329         (LaunchTimeBenchmark.run_iteration):
1330         (LaunchTimeBenchmark.initialize): Convenience method to be overriden by subclasses which is called at the end of __init__
1331         (LaunchTimeBenchmark.will_parse_arguments): Called before argparse.parse_args to let subclasses add new command line arguments
1332         (LaunchTimeBenchmark.did_parse_arguments): Called after argparse.parse_args to let subclass initialize based on command line arguments
1333         * LaunchTime/new_tab.py: Added
1334         (NewTabBenchmark):
1335         (NewTabBenchmark._parse_wait_time): Parser for wait time option
1336         (NewTabBenchmark.initialize):
1337         (NewTabBenchmark.run_iteration):
1338         (NewTabBenchmark.group_init):
1339         (NewTabBenchmark.will_parse_arguments):
1340         (NewTabBenchmark.did_parse_arguments):
1341         (NewTabBenchmark.ResponseHandler):
1342         (NewTabBenchmark.ResponseHandler.Handler):
1343         (NewTabBenchmark.ResponseHandler.Handler.get_test_page):
1344         * LaunchTime/startup.py: Added
1345         (StartupBenchmark): This benchmark measures browser startup time and initial page load time
1346         (StartupBenchmark.initialize):
1347         (StartupBenchmark.run_iteration):
1348         (StartupBenchmark.ResponseHandler):
1349         (StartupBenchmark.ResponseHandler.Handler):
1350         (StartupBenchmark.ResponseHandler.Handler.get_test_page):
1351         * LaunchTime/thirdparty/__init__.py: Added.
1352         (AutoinstallImportHook): Auto installs tornado package for feedback server
1353         (AutoinstallImportHook.__init__):
1354         (AutoinstallImportHook._ensure_autoinstalled_dir_is_in_sys_path):
1355         (AutoinstallImportHook.find_module):
1356         (AutoinstallImportHook._install_tornado):
1357         (AutoinstallImportHook.greater_than_equal_to_version):
1358         (AutoinstallImportHook._install):
1359         (AutoinstallImportHook.get_latest_pypi_url):
1360         (AutoinstallImportHook.install_binary):
1361         (autoinstall_everything):
1362         (get_os_info):
1363
1364 2018-06-25  Jon Lee  <jonlee@apple.com>
1365
1366         [MotionMark] Add support for version numbers
1367         https://bugs.webkit.org/show_bug.cgi?id=186479
1368
1369         Reviewed by Said Abou-Hallawa.
1370
1371         Add support for displaying the version number as well as including it in the JSON results.
1372
1373         When loading the front page, script replaces any element with classname version with the
1374         version number of the benchmark, which is stored in Strings.version.
1375
1376         The JSON structure for the results includes a new version property:
1377             {
1378               "version": "1.0",
1379               "options": { ... },
1380               "data": [ ... ]
1381             }
1382
1383         When dragging a results file, the version listed will come from the JSON file. Older
1384         results will not have had the version property, in which case it will default to "1.0".
1385
1386         * MotionMark/index.html: Update title to some other default. Script will update it.
1387         Include the version number in the logo title.
1388         * MotionMark/developer.html: Ditto.
1389         * MotionMark/about.html: Ditto.
1390
1391         * MotionMark/resources/runner/motionmark.js:
1392         (ResultsDashboard): Update constructor to include version. This is used when serializing
1393         results out to JSON, and displaying the results panel in developer mode.
1394         (ResultsDashboard._processData): When running the benchmark, include benchmark version string
1395         in the results object.
1396         (ResultsDashboard.version):
1397         (window.benchmarkRunnerClient.willStartFirstIteration): When running the benchmark, pass the
1398         benchmark version string to the dashboard, which holds the results.
1399         (window.sectionsManager.setSectionVersion): Helper function to update the element in the
1400         section with the class name version.
1401         (window.benchmarkController.initialize): Populate all DOM elements with class name "version"
1402         with the version string. Update the page title.
1403         (window.benchmarkController.showResults): When showing results, update the version string
1404         based on what is included in the JSON results, which would be the same as the benchmark version.
1405         * MotionMark/resources/runner/motionmark.css: Include missing copyright. Wrap the SVG logo
1406         in a div and include the version string.
1407         * MotionMark/resources/strings.js: Add strings for the page title template, and the version.
1408
1409         * MotionMark/resources/debug-runner/motionmark.css:
1410         * MotionMark/resources/debug-runner/motionmark.js:
1411         (window.benchmarkRunnerClient.willStartFirstIteration): When running the benchmark, pass the
1412         benchmark version string to the dashboard, which holds the results.
1413         (window.benchmarkController.initialize): Populate all DOM elements with class name "version"
1414         with the version string. Update the page title. When dragging in JSON results, look for
1415         version to pass to the dashboard. If it doesn't exist, default to "1.0".
1416         (window.benchmarkController.showResults): When showing results, update the version string
1417         based on what is included in the JSON results, instead of the current benchmark version.
1418         * MotionMark/resources/debug-runner/tests.js: Update page title template.
1419
1420 2018-06-08  Jon Lee  <jonlee@apple.com>
1421
1422         [MotionMark] Rename Suits test files
1423         https://bugs.webkit.org/show_bug.cgi?id=186447
1424
1425         Reviewed by Said Abou-Hallawa.
1426
1427         * MotionMark/resources/runner/tests.js:
1428         * MotionMark/tests/master/resources/suits.js: Renamed from PerformanceTests/MotionMark/tests/master/resources/svg-particles.js.
1429         * MotionMark/tests/master/suits.html: Renamed from PerformanceTests/MotionMark/tests/master/svg-particles.html.
1430         * MotionMark/tests/svg/suits.html:
1431
1432 2018-06-08  Jon Lee  <jonlee@apple.com>
1433
1434         Add sub-tests based on Suits
1435         https://bugs.webkit.org/show_bug.cgi?id=186260
1436
1437         Reviewed by Said Abou-Hallawa.
1438
1439         Add a new developer Suits suite with sub-tests that isolate parts of each particle.
1440         The sub-tests are:
1441             - Particles using only clip paths
1442             - Particles using only shapes
1443             - Particles that have no gradients
1444             - Particles that have no rotation
1445             - Particles that do not move around at all (but all of the physics calculations are
1446             still performed)
1447
1448         * MotionMark/resources/debug-runner/tests.js:
1449         * MotionMark/tests/master/resources/svg-particles.js: Rename the stage, particle, and
1450         benchmark to "Suits" from "SVG". Change SuitsParticle so that we guarantee 50% clip and
1451         shape paths rather than relying on random chance. Wrap a check around creation of the
1452         gradient element, and use a simple fill color when we aren't using a gradient.
1453         * MotionMark/tests/svg/suits.html: Added.
1454         * MotionMark/tests/svg/suits.js: Added.
1455         Look for the query string and set the particle based on what is selected.
1456
1457 2018-06-06  Jon Lee  <jonlee@apple.com>
1458
1459         Allow the ramp controller to run tests that take less time than the initial ramp-up phase
1460         https://bugs.webkit.org/show_bug.cgi?id=186257
1461
1462         Reviewed by Said Abou-Hallawa.
1463
1464         * MotionMark/tests/resources/main.js: Extend the desired end timestamp when ramp-up occurs.
1465         Previously the controller assumed that the duration of the test was always longer than the
1466         time it took to figure out the order of magnitude of particles to render. When that range
1467         is determined, the end timestamp is then extended. If the duration is too short, the controller
1468         would throw an exception. The fix is to continue extending the timestamp every time the order of
1469         magnitude tiers up.
1470
1471         In general testing with really short durations isn't practical, but sometimes it is
1472         desired for debugging.
1473
1474         To avoid confusion of the private variables, I renamed Benchmark._startTimestamp to
1475         Benchmark._benchmarkStartTimestamp, to disambiguate from Controller._startTimestamp.
1476         Benchmark._startTimestamp is set once and never changed, after the initial 100ms warm-up.
1477         Controller._startTimestamp is initialized to 0 (which means "not initialized"), and then after
1478         the warm up is complete, set in Controller.start() to the same wall clock timestamp, and never
1479         changed afterwards.
1480
1481 2018-06-06  Jon Lee  <jonlee@apple.com>
1482
1483         Remove unneeded data processing in MotionMark
1484         https://bugs.webkit.org/show_bug.cgi?id=186256
1485
1486         Reviewed by Said Abou-Hallawa.
1487
1488         Remove calculation of average frame rate based on multiple samples at a given complexity.
1489         It is not used in the calculation of the score, and is just extra work to do at the end
1490         of a test.
1491
1492         * MotionMark/developer.html: Remove options to show average data and regression based on
1493         that data.
1494         * MotionMark/resources/debug-runner/graph.js: Remove data. Update the time graph that
1495         shows the relationship between the break point in the regression against the complexity
1496         graph.
1497         * MotionMark/resources/debug-runner/motionmark.css: Make the confidence intervals easier
1498         to see.
1499         * MotionMark/resources/runner/motionmark.js: Simplify score calculation now that we are
1500         only calculating regressions on the raw data. We were calculating it based on the averages,
1501         but that was just for analysis, and not used in the actual score. Get rid of the
1502         |complexitySamples| local variable.
1503         * MotionMark/resources/strings.js: Remove unneeded string constant.
1504         * MotionMark/tests/resources/main.js:
1505         (processSamples): Refactor so that more of it can be shared between Controller and
1506         RampController. It also avoids processing the raw samples (controllerSamples) twice for
1507         the ramp controller.
1508         (_processComplexitySamples): Remove logic to calculate averages based on complexity.
1509         (_processMarks): Split this out from the old processSamples.
1510         (_processControllerSamples): Split this out from the old processSamples.
1511
1512 2018-06-06  Jon Lee  <jonlee@apple.com>
1513
1514         Strip out more characters when creating permalinks
1515         https://bugs.webkit.org/show_bug.cgi?id=186259
1516
1517         Reviewed by Said Abou-Hallawa.
1518
1519         Tests and suites may have names that make permalinks to the tests less readable when
1520         shared. Expand the set of characters that are stripped out of the names.
1521
1522         * MotionMark/resources/debug-runner/motionmark.js: Refactor to call
1523         stripUnwantedCharactersForURL. For existing links, decode the suite and test names.
1524         * MotionMark/resources/extensions.js:
1525         (Utilities.stripUnwantedCharactersForURL): Rename from stripNonASCIICharacters.
1526         Remove any non-alphanumeric character.
1527
1528 2018-06-04  Jon Lee  <jonlee@apple.com>
1529
1530         Remove unnecessary MotionMark controllers
1531         https://bugs.webkit.org/show_bug.cgi?id=186255
1532
1533         Reviewed by Said Abou-Hallawa.
1534
1535         Get rid of the ramp30 and the fixed-with-a-step controllers, which aren't used for testing
1536         or debugging.
1537
1538         * MotionMark/developer.html: Move the "ramp" option as first, since that is the default
1539         controller for the benchmark.
1540         * MotionMark/resources/debug-runner/motionmark.js:
1541         * MotionMark/resources/runner/motionmark.js: Added missing copyright notice.
1542         * MotionMark/tests/resources/main.js:
1543
1544 2018-05-25  Myles C. Maxfield  <mmaxfield@apple.com>
1545
1546         Improve the performance of Font::canRenderCombiningCharacterSequence()
1547         https://bugs.webkit.org/show_bug.cgi?id=185933
1548
1549         Reviewed by Ryosuke Niwa.
1550
1551         * Layout/ComplexLongUnique.html: Added.
1552
1553 2018-05-25  Saam Barati  <sbarati@apple.com>
1554
1555         Have a memory test where we can validate JSCs mini memory mode
1556         https://bugs.webkit.org/show_bug.cgi?id=185932
1557
1558         Reviewed by Mark Lam.
1559
1560         We add a directory here with the contents of the testmem benchmark.
1561         To run it, use `Tools/Scripts/run-testmem`. To add new tests in the future,
1562         you just need to add JS files to this directory.
1563
1564         * testmem: Added.
1565         * testmem/air.js: Added.
1566         * testmem/base64.js: Added.
1567         * testmem/basic.js: Added.
1568         * testmem/box2d.js: Added.
1569         * testmem/crypto-md5.js: Added.
1570         * testmem/date-format-tofte.js: Added.
1571         * testmem/earley-boyer.js: Added.
1572         * testmem/hash-map.js: Added.
1573         * testmem/regex-dna.js: Added.
1574         * testmem/splay.js: Added.
1575         * testmem/tagcloud.js: Added.
1576
1577 2018-05-22  Ryan Haddad  <ryanhaddad@apple.com>
1578
1579         Unreviewed, rolling out r232052.
1580
1581         Breaks internal builds.
1582
1583         Reverted changeset:
1584
1585         "Use more C++17"
1586         https://bugs.webkit.org/show_bug.cgi?id=185176
1587         https://trac.webkit.org/changeset/232052
1588
1589 2018-05-21  Yusuke Suzuki  <utatane.tea@gmail.com>
1590
1591         Use more C++17
1592         https://bugs.webkit.org/show_bug.cgi?id=185176
1593
1594         Reviewed by JF Bastien.
1595
1596         * MallocBench/MallocBench.xcodeproj/project.pbxproj:
1597         * MediaTime/Configurations/Base.xcconfig:
1598         * MediaTime/MediaTime.xcodeproj/project.pbxproj:
1599
1600 2018-04-30  Myles C. Maxfield  <mmaxfield@apple.com>
1601
1602         Improve the performance of FontCascadeDescription's effectiveFamilies
1603         https://bugs.webkit.org/show_bug.cgi?id=184720
1604
1605         Reviewed by Simon Fraser.
1606
1607         This performance test calls SystemFontDatabase::systemFontCascadeList() around 2,000,000 times (before
1608         this patch is applied), which is roughly equivalent to the page we found the performance problem on.
1609         The calling pattern is roughly equivalent in this test.
1610
1611         * Layout/system-ui.html: Added.
1612
1613 2018-03-08  Antti Koivisto  <antti@apple.com>
1614
1615         Update StyleBench version number in page title to 0.3
1616
1617         * StyleBench/index.html:
1618
1619         There has been a bunch of changes, most notably the addition of attribute selector test step.
1620
1621 2018-02-16  Dean Jackson  <dino@apple.com>
1622
1623         Use OPENGL macros to be more clear about which OpenGL/ES WebGL uses on Cocoa
1624         https://bugs.webkit.org/show_bug.cgi?id=182894
1625
1626         Reviewed by Tim Horton.
1627
1628         Rename OPENGL_ES_2 to OPENGL_ES.
1629
1630         * StitchMarker/wtf/Platform.h:
1631
1632 2018-02-09  Ryosuke Niwa  <rniwa@webkit.org>
1633
1634         Make run-perf-tests work with Speedometer 2.0 and re-enable the test
1635         https://bugs.webkit.org/show_bug.cgi?id=182088
1636
1637         Reviewed by Antti Koivisto.
1638
1639         This patch updates the test harness to report new metrics of Speedometer 2.0.
1640
1641         Replaced the total time by the geometric mean of the time spent in each suite, and added scores
1642         as a metric in addition to time. Also, updated the PerfTestRunner harness and the results template
1643         to support "Score" metric with "pt" as its unit.
1644
1645         * Skipped: Unskipped the test.
1646         * Speedometer/resources/benchmark-report.js:
1647         (window.benchmarkClient.willStartFirstIteration..createTest): Added the unit as an optional argument
1648         to report the score in addition to time. Also report the geomean of time instead of the total time.
1649         (window.benchmarkClient.willStartFirstIteration): Report the geomean of time.
1650         (window.benchmarkClient.didRunSuites): Record the entire measuredValues instead of just tests.
1651         (window.benchmarkClient.didFinishLastIteration): Report the scores.
1652
1653 2018-02-09  Ross Kirsling  <ross.kirsling@sony.com>
1654
1655         Use REQUEST_ANIMATION_FRAME_DISPLAY_MONITOR on any non-Windows port.
1656         https://bugs.webkit.org/show_bug.cgi?id=182623
1657
1658         Reviewed by Alex Christensen.
1659
1660         * StitchMarker/wtf/Platform.h:
1661         Simplify #if.
1662
1663 2018-02-01  Antti Koivisto  <antti@apple.com>
1664
1665         StyleBench: Attribute selectors and other improvements
1666         https://bugs.webkit.org/show_bug.cgi?id=182387
1667
1668         Reviewed by Joseph Pecoraro.
1669
1670         - Add some attributes to elements in all tests
1671         - Add some attribute selectors to stylesheets in all tests
1672         - Also add some * selectors to all stylesheets.
1673         - Add attribute mutation step to all suites
1674         - Make test steps do more mutations (25->100) and reduce the number of steps to keep testing time in check.
1675           Too fast steps were running into timer resolution limits.
1676
1677         * StyleBench/resources/style-bench.js:
1678         (defaultConfiguration):
1679         (prototype.randomAttributeName):
1680         (prototype.randomAttributeValue):
1681         (prototype.randomAttributeSelector):
1682         (prototype.makeCompoundSelector):
1683         (prototype.makeElement):
1684         (prototype.addClasses):
1685         (prototype.removeClasses):
1686         (prototype.mutateAttributes):
1687         (prototype.async.runForever):
1688         * StyleBench/resources/tests.js:
1689         (makeSteps):
1690         (makeSuite):
1691
1692 2018-02-01  Geoffrey Garen  <ggaren@apple.com>
1693
1694         Make MallocBench easier for non-WebKit engineers to run
1695         https://bugs.webkit.org/show_bug.cgi?id=182415
1696
1697         Reviewed by Saam Barati.
1698
1699         * MallocBench/MallocBench.xcodeproj/project.pbxproj: Use c++14 so we
1700         can make_unique.
1701
1702         Specify that we support all Darwin platforms so you can test them.
1703
1704         * MallocBench/run-malloc-benchmarks: Specify the path to MallocBench
1705         and libmbmalloc explicitly, rather than computing them implicitly
1706         using webkitdirs. Non-WebKit folks don't have the directory structure
1707         required by webkitdirs.
1708
1709         Remove Linux-specific and cmake-specific behaviors because we only
1710         needed them in the world of implicit path computation.
1711
1712 2018-01-31  Ryosuke Niwa  <rniwa@webkit.org>
1713
1714         Make run-perf-tests work with StyleBench and re-enable it
1715         https://bugs.webkit.org/show_bug.cgi?id=182322
1716
1717         Reviewed by Antti Koivisto.
1718
1719         This patch updates the test harness to report new metrics of StyleBench.
1720
1721         Replaced the total time by the geometric mean of the time spent in each suite, and added scores
1722         as a metric in addition to time. Also, updated the PerfTestRunner harness and the results template
1723         to support "Score" metric with "pt" as its unit.
1724
1725         * Skipped: Unskipped the test.
1726         * StyleBench/resources/benchmark-report.js:
1727         (window.benchmarkClient.willStartFirstIteration..createTest): Added the unit as an optional argument
1728         to report the score in addition to time. Also report the geomean of time instead of the total time.
1729         (window.benchmarkClient.willStartFirstIteration): Report the geomean of time.
1730         (window.benchmarkClient.didRunSuites): Record the entire measuredValues instead of just tests.
1731         (window.benchmarkClient.didFinishLastIteration): Report the scores.
1732         * resources/results-template.html: Added "Score" as a metric and "pt" as its unit.
1733         * resources/runner.js:
1734         (PerfTestRunner..finish): Ditto.
1735
1736 2018-01-30  Ryosuke Niwa  <rniwa@webkit.org>
1737
1738         StyleBench: Generate more mutation steps
1739         https://bugs.webkit.org/show_bug.cgi?id=182291
1740
1741         Rubber-stamped by Antti Koivisto
1742
1743         Use zero-based index instead of one-based index.
1744
1745         * StyleBench/resources/tests.js:
1746         (makeSteps):
1747
1748 2018-01-30  Antti Koivisto  <antti@apple.com>
1749
1750         StyleBench: Generate more mutation steps
1751         https://bugs.webkit.org/show_bug.cgi?id=182291
1752
1753         Reviewed by Ryosuke Niwa.
1754
1755         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).
1756         This also makes test results more stable and less dependent on possible first-time costs.
1757
1758         * StyleBench/resources/benchmark-runner.js:
1759
1760         Adjust correctionFactor to match the new steps count. With 1.5 Firefox is ~72rpm.
1761
1762         * StyleBench/resources/tests.js:
1763         (makeSteps):
1764
1765         Generate 10 rounds of class/element mutation steps.
1766         Give each generated step unique name so reporting works.
1767         Make individual steps do fewer mutations.
1768
1769 2018-01-29  Antti Koivisto  <antti@apple.com>
1770
1771         StyleBench: Remove : from a test name
1772         https://bugs.webkit.org/show_bug.cgi?id=182258
1773
1774         Reviewed by Ryosuke Niwa.
1775
1776         It breaks reporting.
1777
1778         * StyleBench/resources/style-bench.js:
1779         (nthPseudoClassConfiguration):
1780
1781 2018-01-29  Ryosuke Niwa  <rniwa@webkit.org>
1782
1783         Temporarily skip StyleBench until webkit.org/b/182088 is fixed.
1784
1785         * Skipped:
1786
1787 2018-01-29  Ryosuke Niwa  <rniwa@webkit.org>
1788
1789         Make StyleBench compatible with run-benchmark and run-perf-tests
1790         https://bugs.webkit.org/show_bug.cgi?id=182262
1791
1792         Reviewed by Antti Koivisto.
1793
1794         Copied resource files referenced from Speedometer directory since run-benchmark needs to be able
1795         to checkout each benchmark separately.
1796
1797         Removed the code to create tests of the same name five times in makeSteps since this
1798         won't be compatible with either run-benchmark or run-perf-tests.
1799
1800         * StyleBench/index.html: Removed the code to show warnings for local files since run-benchmark
1801         doesn't use HTTP server in WebDriver mode.
1802         * StyleBench/resources/benchmark-report.js: Copied from resources/benchmark-report.js.
1803         * StyleBench/resources/benchmark-runner.js: Copied from resources/benchmark-runner.js.
1804         (BenchmarkRunner.prototype._finalize): Use the correction factor of 8 instead of 5 in StyleBench.
1805         * StyleBench/resources/gauge.png: Copied from resources/gauge.png.
1806         * StyleBench/resources/gauge@2x.png: Copied from resources/gauge@2x.png.
1807         * StyleBench/resources/main.css: Copied from resources/main.css.
1808         * StyleBench/resources/main.js: Copied from resources/main.js.
1809         * StyleBench/resources/tests.js:
1810         (makeSteps): Only make each test once.
1811
1812 2018-01-26  Chris Nardi  <cnardi@chromium.org>
1813
1814         Addressing post-review comments after r226614
1815         https://bugs.webkit.org/show_bug.cgi?id=182151
1816
1817         Reviewed by Myles C. Maxfield.
1818
1819         * StitchMarker/wtf/text/StringImpl.h:
1820         (WTF::isSpaceOrNewline):
1821         * StitchMarker/wtf/text/TextBreakIterator.cpp:
1822         (WTF::numCodeUnitsInGraphemeClusters):
1823         * StitchMarker/wtf/text/TextBreakIterator.h:
1824
1825 2018-01-22  Antti Koivisto  <antti@apple.com>
1826
1827         StyleBench: Separate test for :nth pseudo classes
1828         https://bugs.webkit.org/show_bug.cgi?id=181932
1829
1830         Reviewed by Sam Weinig.
1831
1832         Optimization strategies for nth are different from other structural pseudo classes (:first-child etc). Move them to a separate test.
1833
1834         * StyleBench/resources/style-bench.js:
1835         (structuralPseudoClassConfiguration):
1836         (nthPseudoClassConfiguration):
1837
1838             Add a new configuration.
1839
1840         (predefinedConfigurations):
1841         (pseudoClassConfiguration): Deleted.
1842
1843 2018-01-17  Ryosuke Niwa  <rniwa@webkit.org>
1844
1845         Temporarily skip Speedometer until we fix the metrics.
1846
1847         * Skipped:
1848
1849 2018-01-16  Alex Christensen  <achristensen@webkit.org>
1850
1851         Fix Mac CMake build after r222900
1852
1853         * MallocBench/MallocBench/CMakeLists.txt:
1854         The mbmalloc target doesn't exist if we're not in developer mode, so only use it if we are.
1855
1856 2018-01-09  Ryosuke Niwa  <rniwa@webkit.org>
1857
1858         Speedometer 2.0: Update the About page.
1859         https://bugs.webkit.org/show_bug.cgi?id=175918
1860
1861         Reviewed by Keith Miller.
1862
1863         Updated the about page for Speedometer 2.0.
1864
1865         * Speedometer/index.html:
1866
1867 == Rolled over to ChangeLog-2018-01-01 ==