Re: [2.0.35+] query regarding backlog size
Alan Cox (alan@lxorguk.ukuu.org.uk)
Wed, 28 Oct 1998 00:56:56 +0000 (GMT)
> The number "300" appears to be sole proprietor controlling the size we
> allow the backlog to grow before dropping. Is this number in some way
> magical, or is it safe to fiddle with? Ultimately, I'm trying to appraise
> a kernel patch which replaces the magic number 300 with a function that
> returns a dynamically changing max backlog value based on available memory.
> I'm hoping someone (perhaps Alan) can comment further.
Its deep magic. Memory size isnt involved either.
300 is simply about the right value, its a pulled out of the air number
that people have since verified has about the right results.
You don't want the backlog queue too long as it is there to smooth bursts
not to jam the machine in net_bh
-
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/