[Webkit-unassigned] [Bug 158083] LLInt should support other types of prototype GetById caching.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=158083

--- Comment #76 from Caio Lima <ticaiolima at gmail.com> ---
(In reply to comment #75)
> Created attachment 281540 [details]
> speedometer run comparison with and without patch applied
> 
> It is the result of speedometer benchmark comparing baseline and the patch
> applied.

I noticed that there is no real gain in this benchmark in terms of performance for the custom accessor caching. So I decided to check the cache miss ratio of it to verify the behavior of this benchmark. The summary results I've got are:

----------------------------------------------------------------------------
Current CacheMiss Ratio:145467/319000
----------------------------------------------------------------------------

This Ration is 0.45600940438871473. I am not convinced that this benchmark is a good choice to test the performance gain of this patch. What I am doing right now is logging my tonight usage to collect the CacheMiss Ration using facebook, youtube and other real web applications.

I am going to send you the results soon.

-- 
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/20160617/2fa85a31/attachment.html>