[Webkit-unassigned] [Bug 27165] Connections-per-host should be tracked closer to the ResourceHandle level, not the Cache

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=27165





--- Comment #4 from Adam Barth <abarth at webkit.org>  2010-09-17 13:42:44 PST ---
(From update of attachment 67919)
View in context: https://bugs.webkit.org/attachment.cgi?id=67919&action=prettypatch

This is massively more elegant.  Is there going to be an issue now that more loads will count against the open connection limit?

It seems slightly unfortunate to have the OpenConnectionLimiter be static.  It seems like we might want to hang it as an object off the NetworkContext object, but we can think about that in a future patch.

> WebCore/loader/ResourceLoader.cpp:132
> +        m_deferredRequest = clientRequest;
> +        setDefersLoading(true);

Should these two lines be one atomic operation?  I'd have to look at how these two pieces of state interact in the rest of this class.

> WebCore/loader/OpenConnectionLimiter.cpp:70
> +    int requestCount = requestsPerHostMap().isEmpty() ? 1 : requestsPerHostMap().get(host) + 1;

I don't quite understand this line.  Why do we need to branch on requestsPerHostMap().isEmpty() ?  Do you mean to ask if the requestsPerHostMap map has an entry for |host| ?

> WebCore/loader/OpenConnectionLimiter.cpp:97
> +    unsigned requestCount = requestsPerHostMap().isEmpty() ? 0 : requestsPerHostMap().get(url.host());

Same question.

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