WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
NEW
154894
[JSC] Schedule B3 BasicBlocks to reduce instruction dependencies before lowering to Air
https://bugs.webkit.org/show_bug.cgi?id=154894
Summary
[JSC] Schedule B3 BasicBlocks to reduce instruction dependencies before lower...
Benjamin Poulain
Reported
2016-03-01 16:48:40 PST
[JSC] Schedule B3 BasicBlocks to reduce instruction dependencies before lowering to Air
Attachments
Patch
(28.23 KB, patch)
2016-03-01 17:01 PST
,
Benjamin Poulain
no flags
Details
Formatted Diff
Diff
Patch
(28.72 KB, patch)
2016-03-01 17:10 PST
,
Benjamin Poulain
no flags
Details
Formatted Diff
Diff
Archive of layout-test-results from ews103 for mac-yosemite
(971.44 KB, application/zip)
2016-03-01 17:51 PST
,
Build Bot
no flags
Details
Archive of layout-test-results from ews104 for mac-yosemite-wk2
(1.01 MB, application/zip)
2016-03-01 17:55 PST
,
Build Bot
no flags
Details
Archive of layout-test-results from ews115 for mac-yosemite
(1.17 MB, application/zip)
2016-03-01 18:11 PST
,
Build Bot
no flags
Details
Patch
(28.73 KB, patch)
2016-03-02 22:20 PST
,
Benjamin Poulain
no flags
Details
Formatted Diff
Diff
Show Obsolete
(2)
View All
Add attachment
proposed patch, testcase, etc.
Benjamin Poulain
Comment 1
2016-03-01 17:01:29 PST
Created
attachment 272611
[details]
Patch
Benjamin Poulain
Comment 2
2016-03-01 17:10:52 PST
Created
attachment 272614
[details]
Patch
Build Bot
Comment 3
2016-03-01 17:50:57 PST
Comment on
attachment 272614
[details]
Patch
Attachment 272614
[details]
did not pass mac-ews (mac): Output:
http://webkit-queues.webkit.org/results/908610
Number of test failures exceeded the failure limit.
Build Bot
Comment 4
2016-03-01 17:51:00 PST
Created
attachment 272618
[details]
Archive of layout-test-results from ews103 for mac-yosemite The attached test failures were seen while running run-webkit-tests on the mac-ews. Bot: ews103 Port: mac-yosemite Platform: Mac OS X 10.10.5
Build Bot
Comment 5
2016-03-01 17:55:30 PST
Comment on
attachment 272614
[details]
Patch
Attachment 272614
[details]
did not pass mac-wk2-ews (mac-wk2): Output:
http://webkit-queues.webkit.org/results/908615
Number of test failures exceeded the failure limit.
Build Bot
Comment 6
2016-03-01 17:55:31 PST
Created
attachment 272619
[details]
Archive of layout-test-results from ews104 for mac-yosemite-wk2 The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews. Bot: ews104 Port: mac-yosemite-wk2 Platform: Mac OS X 10.10.5
Build Bot
Comment 7
2016-03-01 18:11:54 PST
Comment on
attachment 272614
[details]
Patch
Attachment 272614
[details]
did not pass mac-debug-ews (mac): Output:
http://webkit-queues.webkit.org/results/908642
Number of test failures exceeded the failure limit.
Build Bot
Comment 8
2016-03-01 18:11:56 PST
Created
attachment 272621
[details]
Archive of layout-test-results from ews115 for mac-yosemite The attached test failures were seen while running run-webkit-tests on the mac-debug-ews. Bot: ews115 Port: mac-yosemite Platform: Mac OS X 10.10.5
Benjamin Poulain
Comment 9
2016-03-02 22:20:34 PST
Created
attachment 272731
[details]
Patch
Benjamin Poulain
Comment 10
2016-03-03 14:36:15 PST
Too much downside at the moment: Conf#1 Conf#2 SunSpider: 3d-cube 4.9344+-0.3938 4.7263+-0.0255 might be 1.0440x faster 3d-morph 5.2747+-0.0473 ? 5.3184+-0.0952 ? 3d-raytrace 5.7188+-0.1167 5.6440+-0.0700 might be 1.0133x faster access-binary-trees 2.1470+-0.0283 ? 2.2310+-0.0914 ? might be 1.0391x slower access-fannkuch 6.0102+-0.0578 ? 6.0972+-0.1576 ? might be 1.0145x slower access-nbody 2.7363+-0.1036 2.6830+-0.0320 might be 1.0199x faster access-nsieve 3.0503+-0.0902 3.0486+-0.0694 bitops-3bit-bits-in-byte 1.1839+-0.0296 1.1756+-0.0168 bitops-bits-in-byte 3.3028+-0.0205 ? 3.3310+-0.0413 ? bitops-bitwise-and 2.1092+-0.0947 2.0386+-0.0197 might be 1.0346x faster bitops-nsieve-bits 3.1010+-0.0743 3.0398+-0.0317 might be 1.0201x faster controlflow-recursive 2.3789+-0.0373 2.3784+-0.0224 crypto-aes 4.1331+-0.1386 4.0756+-0.0789 might be 1.0141x faster crypto-md5 2.6342+-0.0549 2.5908+-0.0189 might be 1.0167x faster crypto-sha1 2.3757+-0.1049 2.3279+-0.0290 might be 1.0205x faster date-format-tofte 6.8442+-0.1312 ? 6.9856+-0.1549 ? might be 1.0207x slower date-format-xparb 4.8725+-0.2531 4.7721+-0.0441 might be 1.0210x faster math-cordic 2.9353+-0.0347 2.9183+-0.0128 math-partial-sums 4.9662+-0.1437 4.9359+-0.0537 math-spectral-norm 2.0121+-0.0106 2.0071+-0.0303 regexp-dna 5.9869+-0.0491 ? 6.0871+-0.1236 ? might be 1.0167x slower string-base64 4.5166+-0.1362 ? 4.5408+-0.1658 ? string-fasta 6.0276+-0.0814 5.9472+-0.0984 might be 1.0135x faster string-tagcloud 8.2414+-0.2179 8.1270+-0.0353 might be 1.0141x faster string-unpack-code 19.5470+-0.7551 18.9912+-0.3473 might be 1.0293x faster string-validate-input 4.2758+-0.0439 ? 4.2988+-0.0635 ? <arithmetic> 4.6660+-0.0397 4.6276+-0.0165 might be 1.0083x faster Conf#1 Conf#2 Octane: encrypt 0.15963+-0.00208 0.15863+-0.00126 decrypt 2.85449+-0.01274 ^ 2.76413+-0.01163 ^ definitely 1.0327x faster deltablue x2 0.13838+-0.00305 0.13716+-0.00086 earley 0.28309+-0.00114 ? 0.28422+-0.00115 ? boyer 4.77231+-0.06165 ? 4.83612+-0.06335 ? might be 1.0134x slower navier-stokes x2 4.95451+-0.01643 ? 4.95482+-0.03136 ? raytrace x2 0.89441+-0.00409 0.89261+-0.00411 richards x2 0.08215+-0.00067 0.08146+-0.00059 splay x2 0.35072+-0.00169 0.34979+-0.00156 regexp x2 23.79369+-0.23216 ? 24.23938+-0.32417 ? might be 1.0187x slower pdfjs x2 38.70897+-0.25576 38.55399+-0.16836 mandreel x2 42.72580+-0.09408 ? 42.92244+-0.18253 ? gbemu x2 24.77811+-0.17252 24.62463+-0.15728 closure 0.55958+-0.00224 ? 0.56002+-0.00170 ? jquery 7.37918+-0.02330 7.32703+-0.03363 box2d x2 9.28732+-0.06684 9.22627+-0.04934 zlib x2 375.35697+-6.52533 ? 375.47352+-6.30236 ? typescript x2 656.96589+-3.48584 656.25776+-2.77985 <geometric> 5.24638+-0.01167 5.23608+-0.01046 might be 1.0020x faster Conf#1 Conf#2 Kraken: ai-astar 96.688+-2.262 ? 97.421+-1.395 ? audio-beat-detection 47.211+-0.401 ? 47.810+-0.201 ? might be 1.0127x slower audio-dft 97.151+-1.159 96.804+-1.249 audio-fft 35.909+-0.482 35.874+-0.027 audio-oscillator 48.565+-0.191 48.532+-0.125 imaging-darkroom 60.460+-0.669 ? 60.988+-2.240 ? imaging-desaturate 44.457+-0.187 44.440+-0.129 imaging-gaussian-blur 67.818+-1.692 ? 68.279+-0.150 ? json-parse-financial 37.383+-0.069 ? 37.459+-0.720 ? json-stringify-tinderbox 25.753+-0.680 ^ 23.989+-0.513 ^ definitely 1.0735x faster stanford-crypto-aes 39.794+-0.197 ? 40.034+-0.507 ? stanford-crypto-ccm 36.686+-0.523 ? 37.086+-0.421 ? might be 1.0109x slower stanford-crypto-pbkdf2 101.568+-1.386 100.372+-0.304 might be 1.0119x faster stanford-crypto-sha256-iterative 38.724+-0.673 38.192+-0.236 might be 1.0139x faster <arithmetic> 55.583+-0.278 55.520+-0.313 might be 1.0011x faster Conf#1 Conf#2 AsmBench: bigfib.cpp 432.7674+-5.9059 426.3848+-1.0091 might be 1.0150x faster cray.c 363.3387+-1.2263 361.4283+-1.3040 dry.c 445.8662+-28.9116 428.2324+-20.6497 might be 1.0412x faster FloatMM.c 712.8815+-6.1757 ! 724.8161+-2.5732 ! definitely 1.0167x slower gcc-loops.cpp 3679.1768+-16.6924 ! 3709.6699+-11.6864 ! definitely 1.0083x slower n-body.c 807.7697+-2.3024 ? 809.5475+-1.4877 ? Quicksort.c 395.8270+-1.5012 394.6316+-1.6192 stepanov_container.cpp 3304.9979+-25.5368 3282.2913+-14.3214 Towers.c 267.7679+-0.9792 ^ 264.7185+-1.7422 ^ definitely 1.0115x faster <geometric> 719.6359+-5.4692 715.4599+-3.0444 might be 1.0058x faster Conf#1 Conf#2 Geomean of preferred means: <scaled-result> 31.4556+-0.1062 31.3214+-0.0620 might be 1.0043x faster
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug