[Ecls-list] 10.3.1 slime issues

[Matthew Mondor]

On Sun, 7 Mar 2010 23:09:40 +0100
Juan Jose Garcia-Ripoll <juanjose.garciaripoll at googlemail.com> wrote:

> Can you explain it a bit better? I am confused about the "timeout".

Basically if I ktruss the ecl process I see select(2) exiting with 0
(no error, so not EINTR, nor an actual FD event) and being called again
at a rate of about four times per second.  Seeing that the timeval
parameter has a valid pointer suggested to me that most probably a very
short timeout is specified, but if that's the case it could be because
another event source which cannot be polled using select(2) also must
be checked often.

But I've not read the whole swank nor serve-event code yet, so I'm not
sure what's going on exactly.  It's a very minor issue afterall, as the
rate at which select is recalled is a relaxed one and it doesn't
consume much CPU anyway.

Thanks,
-- 
Matt