Date: Thu, 21 Mar 2002 00:40:12 -0600 (CST) From: Haryadi Gunawi <haryadi@cs.wisc.edu> To: net@freebsd.org Subject: Re: incorrect checksums with xl? Message-ID: <Pine.LNX.4.21.0203210032380.6631-100000@tux19.cs.wisc.edu>
next in thread | raw e-mail | index | archive | help
Hi, Can someone help me point the source code where "the TCP/IP stack knows that the NIC is going to insert the checksums"? Is there anyway to disable checksum in the hardware so that the checksum will be done in software or to force the TCP/IP stack do the checksumming? My current project is to measure the TCP/IP processing overhead, I want to measure the performance with checksumming in the software. I use FreeBSD 4.4. thanks in advance Haryadi ps: I am not in the milis, so can you please cc me. > You are running tcpdump on the system that's sending the "bad" > packets, right? It makes perfect sense. The TCP/IP stack knows > that the NIC is going to insert the checksums, so it doesn't bother > calculating them itself. So BPF is handed packets in mbufs without > any checksums. On the way out to the wire, the NIC hardware/firmware > adds the checksums, but they don't show up in the mbufs that are being > sent to BPF. If you run tcpdump on the receiving system instead, > you'll find that the checksums seen there are valid. > The same thing happens in the bge driver, because it also offloads > checksum generation to the NIC. > John To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.21.0203210032380.6631-100000>