At 01:10 PM 4/18/2005, Mladen Turk wrote: >William A. Rowe, Jr. wrote: >>I do have a productive suggestion though that we kicked around once or >>twice. Spin up helper threads (we can even keep a cache of them) to >>handle each group of 64 events, and have them pop an event to the >>parent thread once finished. at 64x63 events, this could be quite >>respectible. > >How did you think to catch the threads? >Sharing them among multiple pollsets, or keeping them around per >pollset?
Yes, we would have to signal on an event (perhaps, the thread itself) to inform the master Waiter-upon-Wait's that a specific tangle of wait events had popped. >>Keep in mind, as you consider new designs for apr_poll, that the >>longest standing issue is polling on files. If you can solve both >>issues in one whack, we would trip over each other getting your >>patch committed :) >Well in that case it would be switch among WSA and Object events, >not a big deal actually ;). It's a rather huge deal - compare the API for waiting on File objects with the unix poll() behavior :( I had it working, at one point, and came up short in two or three respects. Somewhere on one of these boxes sits that patch :-/ Bill
