[Webkit-unassigned] [Bug 27762] [v8] Cache atomic strings in externalized v8 strings

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Aug 3 10:32:21 PDT 2009


https://bugs.webkit.org/show_bug.cgi?id=27762





--- Comment #39 from Eric Seidel <eric at webkit.org>  2009-08-03 10:32:19 PDT ---
Ok.  Talking with Dave Levin over IM:

1.  JSC has this same hot-spot.
2.  JSC based WebKit is 4x as fast as V8 WebKit on this same benchmark.
3.  It's unclear what other things v8 is doing to make it so slow on this
benchmark.
4.  We could add this optimization to JSC in a similar way by converting
UString's hash into a hash/stringimpl* union and keeping around a bit to know
which it was.  Or we could add a StringImpl* to UString.  Neither seems
necessary at this point.

I no-longer object to this change.  I'm still confused as to if this is what
accounts for the 4x slowdown in v8 vs. JSC on this benchmark.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list