<html>
<head>
<base href="https://bugs.webkit.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_REOPENED "
title="REOPENED - NetworkProcess: DNS prefetch happens in the Web Process"
href="https://bugs.webkit.org/show_bug.cgi?id=147824#c13">Comment # 13</a>
on <a class="bz_bug_link
bz_status_REOPENED "
title="REOPENED - NetworkProcess: DNS prefetch happens in the Web Process"
href="https://bugs.webkit.org/show_bug.cgi?id=147824">bug 147824</a>
from <span class="vcard"><a class="email" href="mailto:cdumez@apple.com" title="Chris Dumez <cdumez@apple.com>"> <span class="fn">Chris Dumez</span></a>
</span></b>
<pre>(In reply to <a href="show_bug.cgi?id=147824#c12">comment #12</a>)
<span class="quote">> The patch is quite simple, it just sends a message to the network process to
> do the prefetch there. So, my guess is that we are sending too many
> messages. We could probably cache in the web process the hosts already
> prefetched to avoid sending messages to the network process.</span >
Yes, too much IPC can definitely slow down the PLT. This has happened a few months back due to diagnostic logging. Looking at the patch, one other possibility (I haven't looked deeply into it) is that the new document.frame() checks sometimes fail and we end up not doing DNS prefetches in cases where we used to?</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the assignee for the bug.</li>
</ul>
</body>
</html>