[Webkit-unassigned] [Bug 144787] REGRESSION(r180595): same-callee profiling no longer works
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Fri May 8 21:31:07 PDT 2015
https://bugs.webkit.org/show_bug.cgi?id=144787
--- Comment #2 from Ryosuke Niwa <rniwa at webkit.org> ---
Created attachment 252768
--> https://bugs.webkit.org/attachment.cgi?id=252768&action=review
Benchmark results (summary)
Here's the output from two runs of run-jsc-benchmarks. Signinificantly, I'm seeing Octane to be 1.5 - 2.0% faster:
Octane:
encrypt 0.19000+-0.00046 ? 0.19081+-0.00221 ?
decrypt 3.17850+-0.03612 ? 3.22190+-0.18735 ? might be 1.0137x slower
deltablue x2 0.15496+-0.00104 0.15385+-0.00452
earley 0.39098+-0.00314 ^ 0.36285+-0.00449 ^ definitely 1.0775x faster
boyer 5.33157+-0.17280 ^ 4.00897+-0.01914 ^ definitely 1.3299x faster
navier-stokes x2 4.79455+-0.06503 4.78718+-0.05907
raytrace x2 0.96876+-0.06991 ? 0.97763+-0.06643 ?
richards x2 0.08811+-0.00207 ? 0.08890+-0.00181 ?
splay x2 0.33256+-0.00298 ? 0.33877+-0.01250 ? might be 1.0187x slower
regexp x2 26.54050+-0.47729 26.15802+-1.54154 might be 1.0146x faster
pdfjs x2 37.21557+-0.79409 36.78595+-1.51987 might be 1.0117x faster
mandreel x2 42.97181+-0.99295 ? 43.05130+-1.28202 ?
gbemu x2 32.85735+-3.09244 31.79627+-0.51182 might be 1.0334x faster
closure 0.47578+-0.00959 0.47554+-0.00437
jquery 5.86110+-0.10876 ? 5.89380+-0.09717 ?
box2d x2 10.24121+-0.23015 10.11899+-0.35676 might be 1.0121x faster
zlib x2 361.80455+-3.84013 338.13940+-36.79902 might be 1.0700x faster
typescript x2 634.92981+-20.34847 632.39490+-14.44438
<geometric> * 5.50521+-0.04661 5.40241+-0.07164 might be 1.0190x faster
Octane:
encrypt 0.19202+-0.00500 0.18906+-0.00097 might be 1.0157x faster
decrypt 3.18267+-0.02299 ? 3.23126+-0.04715 ? might be 1.0153x slower
deltablue x2 0.15525+-0.00097 ^ 0.15088+-0.00121 ^ definitely 1.0290x faster
earley 0.39824+-0.00512 ^ 0.36348+-0.00596 ^ definitely 1.0956x faster
boyer 5.30302+-0.08364 ^ 3.97578+-0.03324 ^ definitely 1.3338x faster
navier-stokes x2 4.86913+-0.26404 4.81151+-0.09188 might be 1.0120x faster
raytrace x2 0.99406+-0.06960 0.98159+-0.05859 might be 1.0127x faster
richards x2 0.08730+-0.00150 0.08704+-0.00248
splay x2 0.33225+-0.00364 ? 0.33411+-0.00372 ?
regexp x2 26.20238+-1.44534 ? 26.47932+-0.62304 ? might be 1.0106x slower
pdfjs x2 37.22585+-0.36108 36.85741+-1.11554
mandreel x2 43.15461+-1.84598 ? 43.64833+-1.23361 ? might be 1.0114x slower
gbemu x2 31.74233+-0.42850 31.55462+-0.28688
closure 0.47360+-0.00342 ? 0.47582+-0.00683 ?
jquery 5.85984+-0.09226 ? 5.88440+-0.06700 ?
box2d x2 9.95725+-0.09875 ? 10.00289+-0.23351 ?
zlib x2 357.65324+-12.99726 349.95835+-27.65437 might be 1.0220x faster
typescript x2 645.34644+-18.24724 643.10883+-7.71424
<geometric> * 5.49677+-0.01248 ^ 5.40585+-0.04412 ^ definitely 1.0168x faster
--
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20150509/5a4daa49/attachment-0001.html>
More information about the webkit-unassigned
mailing list