<html>
<head>
<base href="https://bugs.webkit.org/">
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - WebProcess ProcessSwapping problems"
href="https://bugs.webkit.org/show_bug.cgi?id=200967#c16">Comment # 16</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - WebProcess ProcessSwapping problems"
href="https://bugs.webkit.org/show_bug.cgi?id=200967">bug 200967</a>
from <span class="vcard"><a class="email" href="mailto:cgarcia@igalia.com" title="Carlos Garcia Campos <cgarcia@igalia.com>"> <span class="fn">Carlos Garcia Campos</span></a>
</span></b>
<pre>(In reply to Michael Catanzaro from <a href="show_bug.cgi?id=200967#c13">comment #13</a>)
<span class="quote">> (In reply to Carlos Garcia Campos from <a href="show_bug.cgi?id=200967#c12">comment #12</a>)
> > The other alternative would be to disable PSON entirely in 2.26.
>
> I'd do it, but only if we also revert enough of Chris's work on the 2.26
> branch to bring back single-process model for 2.26. Then we can have a
> smooth upgrade to 2.26 so this is no longer an emergency, and continue to
> discuss how to handle this situation for 2.28.</span >
Unfortunately that's not that easy and we already have a workaround. I expect more apps to be broken because of this bug than apps depending on the single process model. I know it would be ideal if apps didn't have to do anything, but we have a very simple workaround for the single process model case. This bug is more complicated and the only way I see right now to not break apps is disabling PSON.
<span class="quote">> We should still try to sort this out now, but it'll be a lot easier to solve
> this at the very beginning of the next release cycle rather than the end of
> the current one.</span >
It's a pity that very few people try early unstable releases, and I didn't notice any issue because I don't use the remember password feature of ephy.
<span class="quote">> We might need to seriously consider an API version bump, though hopefully it
> can be avoided.
>
> (In reply to Carlos Garcia Campos from <a href="show_bug.cgi?id=200967#c10">comment #10</a>)
> > hmm, it's not only suspended pages, there's also the process cache.
>
> What's the impact of process cache on API users?</span >
I guess it will be slower to load pages when switching processes due to history navigation (which is always expected to be fast).</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>