Bug 155347

Summary:

[JSC] Remove a few jumps from DFG

Product:

WebKit

Reporter:

Benjamin Poulain <benjamin>

Component:

New Bugs

Assignee:

Benjamin Poulain <benjamin>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

buildbot, commit-queue, keith_miller, mark.lam, msaboff, rniwa, saam

Priority:

Version:

WebKit Nightly Build

Hardware:

Unspecified

OS:

Unspecified

Attachments:

Description	Flags
Patch	none
Archive of layout-test-results from ews101 for mac-yosemite	none
Archive of layout-test-results from ews107 for mac-yosemite-wk2	none
Archive of layout-test-results from ews125 for ios-simulator-wk2	none
Archive of layout-test-results from ews116 for mac-yosemite	none
Patch	none

Benjamin Poulain

Reported 2016-03-10 20:56:08 PST

[JSC] Remove a few jumps from DFG

Attachments
Patch (3.22 KB, patch) 2016-03-10 21:00 PST, Benjamin Poulain	no flags	Details Formatted Diff Diff
Archive of layout-test-results from ews101 for mac-yosemite (1.01 MB, application/zip) 2016-03-10 21:51 PST, Build Bot	no flags	Details
Archive of layout-test-results from ews107 for mac-yosemite-wk2 (945.58 KB, application/zip) 2016-03-10 21:56 PST, Build Bot	no flags	Details
Archive of layout-test-results from ews125 for ios-simulator-wk2 (588.51 KB, application/zip) 2016-03-10 21:59 PST, Build Bot	no flags	Details
Archive of layout-test-results from ews116 for mac-yosemite (1.02 MB, application/zip) 2016-03-10 22:09 PST, Build Bot	no flags	Details
Patch (3.24 KB, patch) 2016-03-11 17:42 PST, Benjamin Poulain	no flags	Details Formatted Diff Diff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.

Benjamin Poulain

Comment 1 2016-03-10 21:00:53 PST

Created attachment 273679 [details] Patch

Mark Lam

Comment 2 2016-03-10 21:05:51 PST

Comment on attachment 273679 [details] Patch r=me

Saam Barati

Comment 3 2016-03-10 21:06:43 PST

Comment on attachment 273679 [details] Patch r=me

Build Bot

Comment 4 2016-03-10 21:51:10 PST

Comment on attachment 273679 [details] Patch Attachment 273679 [details] did not pass mac-ews (mac): Output: http://webkit-queues.webkit.org/results/958530 New failing tests: imported/w3c/web-platform-tests/dom/traversal/NodeIterator.html imported/w3c/web-platform-tests/dom/ranges/Range-surroundContents.html imported/w3c/web-platform-tests/dom/ranges/Range-insertNode.html js/regress/cmpeq-obj-to-obj-other.html imported/w3c/web-platform-tests/dom/ranges/Range-mutations.html imported/w3c/web-platform-tests/dom/traversal/NodeIterator-removal.html

Build Bot

Comment 5 2016-03-10 21:51:12 PST

Created attachment 273687 [details] Archive of layout-test-results from ews101 for mac-yosemite The attached test failures were seen while running run-webkit-tests on the mac-ews. Bot: ews101 Port: mac-yosemite Platform: Mac OS X 10.10.5

Build Bot

Comment 6 2016-03-10 21:56:19 PST

Comment on attachment 273679 [details] Patch Attachment 273679 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.webkit.org/results/958546 New failing tests: imported/w3c/web-platform-tests/dom/traversal/NodeIterator.html imported/w3c/web-platform-tests/dom/ranges/Range-surroundContents.html imported/w3c/web-platform-tests/dom/ranges/Range-insertNode.html js/regress/cmpeq-obj-to-obj-other.html imported/w3c/web-platform-tests/dom/ranges/Range-mutations.html imported/w3c/web-platform-tests/dom/traversal/NodeIterator-removal.html

Build Bot

Comment 7 2016-03-10 21:56:21 PST

Created attachment 273688 [details] Archive of layout-test-results from ews107 for mac-yosemite-wk2 The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews. Bot: ews107 Port: mac-yosemite-wk2 Platform: Mac OS X 10.10.5

Build Bot

Comment 8 2016-03-10 21:59:14 PST

Comment on attachment 273679 [details] Patch Attachment 273679 [details] did not pass ios-sim-ews (ios-simulator-wk2): Output: http://webkit-queues.webkit.org/results/958544 New failing tests: imported/w3c/web-platform-tests/dom/traversal/NodeIterator.html imported/w3c/web-platform-tests/dom/ranges/Range-surroundContents.html imported/w3c/web-platform-tests/dom/ranges/Range-insertNode.html js/regress/cmpeq-obj-to-obj-other.html imported/w3c/web-platform-tests/dom/ranges/Range-mutations.html imported/w3c/web-platform-tests/dom/traversal/NodeIterator-removal.html

Build Bot

Comment 9 2016-03-10 21:59:17 PST

Created attachment 273689 [details] Archive of layout-test-results from ews125 for ios-simulator-wk2 The attached test failures were seen while running run-webkit-tests on the ios-sim-ews. Bot: ews125 Port: ios-simulator-wk2 Platform: Mac OS X 10.10.5

Build Bot

Comment 10 2016-03-10 22:09:51 PST

Comment on attachment 273679 [details] Patch Attachment 273679 [details] did not pass mac-debug-ews (mac): Output: http://webkit-queues.webkit.org/results/958547 New failing tests: imported/w3c/web-platform-tests/dom/traversal/NodeIterator.html imported/w3c/web-platform-tests/dom/ranges/Range-surroundContents.html imported/w3c/web-platform-tests/dom/ranges/Range-insertNode.html js/regress/cmpeq-obj-to-obj-other.html imported/w3c/web-platform-tests/dom/traversal/NodeIterator-removal.html

Build Bot

Comment 11 2016-03-10 22:09:54 PST

Created attachment 273690 [details] Archive of layout-test-results from ews116 for mac-yosemite The attached test failures were seen while running run-webkit-tests on the mac-debug-ews. Bot: ews116 Port: mac-yosemite Platform: Mac OS X 10.10.5

Benjamin Poulain

Comment 12 2016-03-11 17:42:16 PST

Created attachment 273789 [details] Patch

Benjamin Poulain

Comment 13 2016-03-11 17:42:51 PST

On Haswell: Conf#1 Conf#2 SunSpider: 3d-cube 4.3975+-0.0868 4.3949+-0.0566 3d-morph 4.8105+-0.0389 4.7847+-0.0325 3d-raytrace 5.2768+-0.1391 5.1633+-0.0442 might be 1.0220x faster access-binary-trees 1.9632+-0.0276 1.9467+-0.0154 access-fannkuch 5.7381+-0.0410 ? 5.8177+-0.1148 ? might be 1.0139x slower access-nbody 2.4241+-0.0260 ? 2.4602+-0.0479 ? might be 1.0149x slower access-nsieve 2.7753+-0.0489 ? 2.8413+-0.0774 ? might be 1.0238x slower bitops-3bit-bits-in-byte 1.0757+-0.0264 1.0576+-0.0065 might be 1.0172x faster bitops-bits-in-byte 2.5632+-0.0360 ? 2.5909+-0.0448 ? might be 1.0108x slower bitops-bitwise-and 1.8789+-0.0166 ? 1.8799+-0.0119 ? bitops-nsieve-bits 2.8969+-0.0310 ? 2.9621+-0.1303 ? might be 1.0225x slower controlflow-recursive 2.1674+-0.0123 ? 2.1947+-0.0418 ? might be 1.0126x slower crypto-aes 3.7352+-0.0385 ? 3.7709+-0.1133 ? crypto-md5 2.3493+-0.0551 2.3366+-0.0169 crypto-sha1 2.1776+-0.0221 ? 2.2102+-0.0521 ? might be 1.0150x slower date-format-tofte 6.3400+-0.1871 6.1980+-0.0910 might be 1.0229x faster date-format-xparb 4.5988+-0.1484 4.4214+-0.0375 might be 1.0401x faster math-cordic 2.6503+-0.0594 ? 2.7414+-0.1776 ? might be 1.0344x slower math-partial-sums 4.5984+-0.0942 4.5541+-0.0640 math-spectral-norm 1.8646+-0.0515 1.8383+-0.0109 might be 1.0143x faster regexp-dna 5.6087+-0.1762 5.5817+-0.0725 string-base64 4.1238+-0.0254 4.0932+-0.0287 string-fasta 5.4404+-0.0472 ? 5.5032+-0.0939 ? might be 1.0115x slower string-tagcloud 7.5335+-0.0448 ? 7.5541+-0.0569 ? string-unpack-code 17.4060+-0.2698 ^ 16.9562+-0.1411 ^ definitely 1.0265x faster string-validate-input 4.0099+-0.2177 4.0064+-0.1764 <arithmetic> 4.2463+-0.0179 4.2254+-0.0156 might be 1.0050x faster Conf#1 Conf#2 Octane: encrypt 0.15201+-0.00243 0.14992+-0.00165 might be 1.0140x faster decrypt 2.74856+-0.02373 2.74432+-0.02447 deltablue x2 0.13111+-0.00090 0.13066+-0.00104 earley 0.30426+-0.00336 ? 0.30471+-0.00288 ? boyer 4.77107+-0.02118 4.75521+-0.03198 navier-stokes x2 4.68973+-0.00583 4.68460+-0.00606 raytrace x2 0.92400+-0.01217 0.92062+-0.00797 richards x2 0.07825+-0.00072 0.07802+-0.00055 splay x2 0.39018+-0.00282 0.38938+-0.00284 regexp x2 20.03725+-0.07426 ^ 19.40154+-0.29439 ^ definitely 1.0328x faster pdfjs x2 37.67401+-0.30669 37.38918+-0.22493 mandreel x2 39.90832+-0.26069 ? 40.00725+-0.18733 ? gbemu x2 22.55758+-0.08268 ? 22.63318+-0.17271 ? closure 0.54608+-0.00191 ? 0.54701+-0.00222 ? jquery 6.94196+-0.03282 6.92659+-0.04162 box2d x2 8.65679+-0.09918 8.65554+-0.03955 zlib x2 369.59111+-6.66368 ? 370.27905+-6.15758 ? typescript x2 620.85905+-5.91149 618.28597+-5.00259 <geometric> 5.05026+-0.01400 5.03067+-0.01243 might be 1.0039x faster Conf#1 Conf#2 Kraken: ai-astar 83.442+-2.153 ? 83.466+-1.085 ? audio-beat-detection 42.859+-0.208 42.716+-0.133 audio-dft 88.167+-0.275 ? 88.392+-0.312 ? audio-fft 33.850+-0.027 33.848+-0.021 audio-oscillator 44.938+-0.426 44.662+-0.162 imaging-darkroom 57.162+-0.199 ? 57.370+-0.403 ? imaging-desaturate 41.869+-0.100 ? 42.058+-0.484 ? imaging-gaussian-blur 61.225+-2.815 59.725+-3.049 might be 1.0251x faster json-parse-financial 35.464+-0.795 34.789+-0.168 might be 1.0194x faster json-stringify-tinderbox 21.720+-0.321 ? 22.108+-0.401 ? might be 1.0179x slower stanford-crypto-aes 38.376+-0.295 ? 38.689+-0.370 ? stanford-crypto-ccm 34.440+-0.644 ? 34.763+-0.842 ? stanford-crypto-pbkdf2 95.371+-0.346 95.323+-0.492 stanford-crypto-sha256-iterative 36.661+-0.138 ? 36.694+-0.209 ? <arithmetic> 51.110+-0.319 51.043+-0.214 might be 1.0013x faster Conf#1 Conf#2 AsmBench: bigfib.cpp 409.5055+-6.1927 ? 410.1419+-5.2106 ? cray.c 340.3798+-2.2207 338.2782+-1.1468 dry.c 400.1723+-5.5608 ? 403.1098+-19.3978 ? FloatMM.c 678.5357+-0.4145 ? 680.0313+-1.3960 ? gcc-loops.cpp 3493.6306+-10.6787 ? 3503.0929+-8.4453 ? n-body.c 766.7423+-1.7096 ? 767.1165+-2.3526 ? Quicksort.c 370.4900+-0.9764 369.2836+-0.8938 stepanov_container.cpp 3029.1641+-15.6930 3012.2418+-8.3204 Towers.c 254.5035+-1.2592 254.0570+-0.6536 <geometric> 674.6211+-2.0117 674.2918+-3.1648 might be 1.0005x faster Conf#1 Conf#2 Geomean of preferred means: <scaled-result> 29.3236+-0.0691 29.2458+-0.0592 might be 1.0027x faster

Mark Lam

Comment 14 2016-03-11 18:39:14 PST

Comment on attachment 273789 [details] Patch r=me

WebKit Commit Bot

Comment 15 2016-03-11 19:28:21 PST

Comment on attachment 273789 [details] Patch Clearing flags on attachment: 273789 Committed r198057: <http://trac.webkit.org/changeset/198057>

WebKit Commit Bot

Comment 16 2016-03-11 19:28:27 PST

All reviewed patches have been landed. Closing bug.

Note You need to log in before you can comment on or make changes to this bug.