arp(8) performance w/ many aliases assigned to an interface

[Max Laier]

On Wednesday 09 June 2010 23:56:27 Nick Rogers wrote:
> Thanks a lot, I have been trying to get a meaningful response on this issue
> for about a week now. Guess I finally got the right list. Thanks again!

I saw your post before on -net etc, just never found the time ... and then I 
forgot.  So it was more the right timing, than the right list :-)

> Can this change make it into 8.1-RELEASE?

It looks like it might be too late, I'll ask once the HEAD commit is done, but 
I wouldn't count on it.

> On Wed, Jun 9, 2010 at 5:30 PM, Max Laier <max at love2party.net> wrote:
> > On Wednesday 09 June 2010 23:21:05 Max Laier wrote:
> > > Hello Nick,
> > > 
> > > On Wednesday 09 June 2010 20:17:43 Nick Rogers wrote:
> > > > Is there something that can be done to speedup the call to
> > > > if_indextoname(), or would it be worthwhile for me to submit a patch
> > 
> > that
> > 
> > > > adds the ability to skip the interface lookup as an arp(8) option?
> > > 
> > > how about the attached:
> > > 
> > > # time arp -an > /dev/null
> > > 0.171u 0.462s 0:00.63 100.0%    21+1538k 0+0io 0pf+0w
> > > # time arp.patched -an > /dev/null
> > > 0.005u 0.000s 0:00.00 0.0%      0+0k 0+0io 0pf+0w
> > > 
> > > # arp -an | wc -l
> > > 
> > >     1095
> > > 
> > > I'll commit this soon unless somebody objects.
> > 
> > ... make that s/long/int64_t/ ... I keep forgetting that long is 32bit
> > sometimes.
> 
> !DSPAM:4c100e18981892214220659!