Aaron Stone wrote: > Good enough for me. I think we should go for it. Paul, did you have an > objection? I'm also not terribly sure if we should build towards 2.0.1 > while Ilja is out of town, although I'd like to. Perhaps a bugtracker > thread with all of the patches we want in 2.0.1 so that we can test > them > for a while?
Question: will 2.0.1 be based on current 2.0 code with patches, but without regard for CVS development? I have a reason for asking. I have a FETCH-speedup idea that will not need many changes in the current 2.0 code; in fact code outside _fetch_ic() is not affected. It's not as fast as my "real" idea, but it's probably a 3 or 4 times speedup nevertheless. (Brief summary of the idea. Prefetch everything in all messages, except the body in one query; this code is already present for the no-parsing case, I'll just use some of it in all cases. Then fetch the body, or just headers, as it was done before). I can implement that for 2.0.1 today. But this means a patch against the 2.0 _ic_fetch() , not CVS HEAD - so I need to make sure this is useful for 2.0.1. Yours, Mikhail Ramendik
