On Mon, Feb 4, 2013 at 1:29 AM, David Laight <david.lai...@aculab.com> wrote: >> I agree. Honestly I think a perfectly reasonable stance to take is >> requesting that the filters get packets *as seen on the wire/nic*. I >> think that's the mental model everyone uses, and any deviation from >> that model is prone to bugs in the kernel, libpcap, and for the enduser. > > TX and RX segmentation offload also confuse matters here. > I think Linux can give libpcap large TCP fragments even when the > hardware isn't doing segmentation offload. > This also breaks the mental model. >
It would be nice if someone can initiate a discussion/debate on netdev on the points that we are raising here. I tried one time with vlan issues and it met with lot of resistance. May be another person will have better success than me :-) _______________________________________________ tcpdump-workers mailing list tcpdump-workers@lists.tcpdump.org https://lists.sandelman.ca/mailman/listinfo/tcpdump-workers
