Morphic slowness (was Re: Does *anyone* use MVC?)
Dan Ingalls
Dan at SqueakLand.org
Fri Aug 2 02:31:52 UTC 2002
Andreas -
>#startSteppingSubmorphsOf: is called from various places and not - like
>#intoWorld: - from a few well-defined points in the life-time of a
>morph. Meaning that there are many cases in which the method is run even
>though the morph tree is already known to be stepping or not (can't say
>for sure if that's the case here but I have a strong feeling that this
>may be the case).
Ah. OK, that may be. It suggests running
MessageTally tallySends: []
^^^^^^
to get *exact counts* from one of these menu creations.
>And then, there's the issue that
>startSteppingSubmorphsOf: is less efficient than intoWorld: in itself -
>due to the nature of our profiler which uses stochastic sampling it is
>quite possible that you see slowness show up in other places - after all
>this place was sampled at most 20 times (possibly less if there have
>been any process switches) which doesn't make for a very good confidence
>interval. If I were *really* interested in this issue (which I am not
>;-) I'd tally the sends to see what is going on for sure.
Ya. Just what I was getting at.
- Dan
More information about the Squeak-dev
mailing list
|