[Webkit-unassigned] [Bug 55973] Provide a WebView knob to disable DNS prefetching
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Wed Mar 9 03:57:12 PST 2011
https://bugs.webkit.org/show_bug.cgi?id=55973
--- Comment #3 from marco <marco at peereboom.us> 2011-03-09 03:57:12 PST ---
Updated the patch with all your suggestions, thanks :-)
My non-scientific setup is as follows:
* OpenBSD 4.9
* WebKitGTK 1.2.7
* xxxterm browser 1.348
* adsuck 2.1 (DNS server that spoofs NXDOMAIN for unwanted servers)
I rigged the browser to display total page load time. Where total page load is
defined as time spent between WEBKIT_LOAD_PROVISIONAL and WEBKIT_LOAD_FINISHED.
I ran it through a number of websites and got the following results:
prefetch 0: page load time: 2.148816 -> http://slashdot.org/
prefetch 1: page load time: 7.705360 -> http://slashdot.org/
prefetch 0: page load time: 0.973012 -> http://news.google.com/
prefetch 1: page load time: 19.757806 -> http://news.google.com/
prefetch 0: page load time: 1.504187 -> http://www.wired.com/
prefetch 1: page load time: 3.965751 -> http://www.wired.com/
prefetch 0: page load time: 1.283165 -> http://www.theregister.co.uk/
prefetch 1: page load time: 1.759814 -> http://www.theregister.co.uk/
prefetch 0: page load time: 2.288612 -> http://www.techreview.com/blog/arxiv/
prefetch 1: page load time: 3.196511 -> http://www.techreview.com/blog/arxiv/
prefetch 0: page load time: 2.149060 -> http://www.cnn.com/
prefetch 1: page load time: 5.169388 -> http://www.cnn.com/
prefetch 0: page load time: 2.192299 -> http://espn.go.com/
prefetch 1: page load time: 7.610725 -> http://espn.go.com/
That is what I mean by "quite a bit snappier". The browser seems to load pages
at a linear speed instead of bumping along (by observing the completion bar).
This is just the performance aspect for me. I can go into the security issues
as well but I think those have been described before in other bugs.
--
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