[Webkit-unassigned] [Bug 143851] Inline JSFunction allocation in FTL

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=143851

--- Comment #9 from Basile Clement <basile_clement at apple.com> ---
Re-ran the Octane benchmark to offset the hiccup on the regexp bench.


Benchmark report for Octane on Basiles-MacBook-Pro (MacBookPro11,3).

VMs tested:
"Conf#1" at /Volumes/Data/Baseline/OpenSource/WebKitBuild/Release/jsc (r182899)
"Conf#2" at /Volumes/Data/WIP/OpenSource/WebKitBuild/Release/jsc (r182899)

Collected 6 samples per benchmark/VM, with 6 VM invocations per benchmark. Emitted a call to gc()
between sample measurements. Used 1 benchmark iteration per VM invocation for warm-up. Used the
jsc-specific preciseTime() function to get microsecond-level timing. Reporting benchmark execution
times with 95% confidence intervals in milliseconds.

                             Conf#1                    Conf#2                                      

encrypt                 0.20299+-0.00210    ?     0.20498+-0.00314       ?
decrypt                 3.71181+-0.10318          3.70406+-0.11871       
deltablue      x2       0.17178+-0.00579          0.16550+-0.00078         might be 1.0380x faster
earley                  0.48429+-0.00351    ^     0.45653+-0.00461       ^ definitely 1.0608x faster
boyer                   5.81100+-0.12890    ?     6.01216+-0.19705       ? might be 1.0346x slower
navier-stokes  x2       5.35065+-0.16091          5.24746+-0.07685         might be 1.0197x faster
raytrace       x2       1.09675+-0.04058          1.08612+-0.03352       
richards       x2       0.09570+-0.00121    ?     0.09748+-0.00175       ? might be 1.0186x slower
splay          x2       0.36184+-0.00409    ?     0.36956+-0.00652       ? might be 1.0213x slower
regexp         x2      29.95385+-0.72763         29.12325+-0.68707         might be 1.0285x faster
pdfjs          x2      40.03323+-0.43196         40.02588+-0.17253       
mandreel       x2      48.16510+-0.74271         47.97694+-0.91950       
gbemu          x2      32.91826+-0.65299         32.81293+-0.41711       
closure                 0.51014+-0.00732          0.50741+-0.00279       
jquery                  6.26699+-0.04984    ?     6.33675+-0.05284       ? might be 1.0111x slower
box2d          x2      10.68723+-0.20455         10.61400+-0.09674       
zlib           x2     365.98031+-14.27976   ?   366.00936+-18.78087      ?
typescript     x2     677.13818+-18.99019   ?   678.85726+-8.10687       ?

<geometric>             5.98815+-0.05931          5.95943+-0.04587         might be 1.0048x 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/20150417/5ff3fb7e/attachment.html>