Re: cat /proc/net/tcp takes 0.5 seconds on x86_64

[Stephen Hemminger <shemminger@vyatta.com>]

On Wed, 27 Aug 2008 14:29:41 -0700 (PDT)
David Miller <davem@davemloft.net> wrote:

> From: Andi Kleen <andi@firstfloor.org>
> Date: Wed, 27 Aug 2008 14:41:52 +0200
> 
> > Dave Jones <davej@redhat.com> writes:
> > 
> > > Just had this bug reported against our development tree..
> > 
> > SUSE had an old patch for this which unfortunately got rejected 
> > some time ago for some bogus reason.
> 
> Really, your patch fixes this specific slowdown that got introduced
> recently?  I really doubt it Andi, so please don't use this as an
> opportunity to toot your own horn, thanks.
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

I do wonder if having large hash table actually helps? When TCP hash
table gets too big, it means every lookup is a cache miss. Assuming
a busy server with 2000 connections and perfect hash. On a 4G mem x86-64
we are doing 512K hash entries which is ridiculous. Something like 64K
entries is more than enough.