> You didn't read what I said. I don't have a gigabit ethernet switch. > I only have cards. Therefore the *only* way I can test the operation > of the driver and adapters is to connect two machines with gigabit > cards back to back with a patch cable. This automatically implies 'using > gb end-to-end.' > > As for corruption due to TCP sequence number wrapping, I don't know > what to tell you. I never noticed such behavior in my tests, but that's > why I'm asking for feedback from other people.
The obvious answer to the TCP sequence number problem is RFC 1323. I assume anybody who wants to use gigabit Ethernet over significant distances *will* use RFC 1323, if they are interested in any performance at all. Otherwise the 64 kbyte window will kill you. As for me, I have tested the driver with Netgear cards. Works great here, I got 470 Mbps (effective application to application) with ttcp, running back to back on a PII-350 and a Celeron 300A (overclocked to 337, thus PCI bus clocked at 37.5 Mhz). The limit in my case is clearly the CPU. However I did *not* see any better performance when I turned on jumbo frames. Next I'll put one card in an old PPro-200 and see what I can get from that. Steinar Haug, Nethelp consulting, sth...@nethelp.no To Unsubscribe: send mail to majord...@freebsd.org with "unsubscribe freebsd-current" in the body of the message
