Re: libpcap capture performance drop

[Rick Jones <rick.jones2 () hp com>]

On Fri, 2011-05-27 at 14:20 +0200, rixed () happyleptic org wrote:
> -[ Mon, May 23, 2011 at 12:38:43AM -0700, Guy Harris ]----
> > On May 23, 2011, at 12:31 AM, rixed () happyleptic org wrote:
> >
> > > Which brings the question: how one could find out the MTU of a
> > > pcap_handle in order no to set caplen to 65535 ?
> >
> > See pcap-linux.c in the top of the trunk or of the 1.2 branch.  (Short answer: SIOCGIFMTU, as in the 
> > iface_get_mtu() routine.)-
>
> If I understand this code correctly, in the next release of the libpcap
> if a client program ask for a capture length bigger than the MTU then
> the size allocated for each frame in the ring buffer will be sized down
> to avoid wasting space ?
>
> If so then I just have to wait for the new libpcap :-)

Is this new libpcap going to be guaranteed that the underlying NIC HW
isn't doing Large Receive Offload, or that the tracepoint in the stack
is below any stack's attempt to do Generic Receive Offload?

rick jones


-
This is the tcpdump-workers list.
Visit https://cod.sandelman.ca/ to unsubscribe.