[Webkit-unassigned] [Bug 196160] [JSC] Butterfly allocation from LargeAllocation should try "realloc" behavior if corrector thread is not active

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Fri Mar 22 14:47:18 PDT 2019


https://bugs.webkit.org/show_bug.cgi?id=196160

--- Comment #1 from Yusuke Suzuki <ysuzuki at apple.com> ---
I don't think this hurts performance even if we make it by default.
This is simply because allocating LargeAllocation should be already super slow.
And this cost is amortized by the exponential size increase policy.
But anyway, we can measure it. And if we hit some performance problems, we could do,

1. enable it only if sweepSynchronously() is true
2. find some way for mitigation

I think we do not need to do this and believe it is performance-neutral.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20190322/9fe17335/attachment.html>


More information about the webkit-unassigned mailing list