On 1 October 2014 01:22, Jeff Pinner <jpinner@twitter.com> wrote: > What are these other benefits? I certainly don't believe there are any > -- having to know the size of the dynamic table to compute a static > index offset? > At least one benefit is that you have the ability to pre compute values for statics fields, or even series of static fields, so they do not need to be calculated on every generation. Now that we don't copy static entries to the dynamic field this generate once use many times extends to fields using a static name and custom value. Why was the change put in, without a discussion on the list, when it > admittedly had a negative impact on compression in your testing? There was definitely discussion on the list at the time. It had also been discussed previously and there were reasons not to make the change earlier, but they evaporated with the removal of the reference set. My tests (also published in the list) with test headers available (and used for other decisions here) showed a small but positive change. cheers -- Greg Wilkins <gregw@intalio.com> http://eclipse.org/jetty HTTP, SPDY, Websocket server and client that scales http://www.webtide.com advice and support for jetty and cometd.
Received on Tuesday, 30 September 2014 15:40:14 UTC