Re: Max tcp connections

Stephen C. Tweedie (sct@redhat.com)
Fri, 12 Nov 1999 17:49:41 +0000 (GMT)


Hi,

On Thu, 11 Nov 1999 23:47:35 -0800, Dan Kegel <dank@alumni.caltech.edu>
said:

> The signal model gives you the best of both worlds: low overhead
> when only a few fd's at a time need attention (no linear scan),
> and low overhead when they all do (same as old poll() method,
> but with the extra overhead of a single SIGIO).

> sct must be sleeping, or he would have answered like this, I think :-)

You did it fine yourself. :) One thing I do want to add is a
per-process limit to the sigqueue depth, just to make the model degrade
even better under high signal load, but otherwise you summed it up
pretty well.

--Stephen

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/