On Sat, 10 Aug 2002, Mike Tancsa wrote: Dear Mike.
I reported a similar problem on August the 9th and I must say, sorry, if I would have offered the dmesg output all should have seen that we use only Intel NICs around here! Another workstation with a 3Com NIC does not show this strange behaviour. > > I did some more tests with the new driver with 2 machines on a totally > different segment and the same sorts of packet loss on simple pings. If I > revert to the previous version, no packet loss at all. > > $FreeBSD: src/sys/dev/fxp/if_fxp.c,v 1.110.2.24 2002/08/09 02:04:20 luigi Exp $ > vs > $FreeBSD: src/sys/dev/fxp/if_fxp.c,v 1.110.2.23 2002/07/29 02:50:59 luigi Exp $ > > I confirmed this on 2 separate machines. > > > The fxp nics in question are > > fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0xa800-0xa81f mem > 0xf9000000-0xf90fffff,0xfd000000-0xfd000fff irq 11 at device 13.0 on pci0 > fxp0: Ethernet address 00:a0:c9:89:d5:ba > inphy0: <i82555 10/100 media interface> on miibus0 > inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto > > and > > fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0xa000-0xa03f mem > 0xe9500000-0xe95fffff,0xe9600000-0xe9600fff irq 10 at device 4.0 on pci2 > fxp0: Ethernet address 00:d0:b7:db:69:6a > inphy0: <i82555 10/100 media interface> on miibus0 > inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto > > Both nics on both machines in question do have shard IRQs, but like I said, > with the previous version of the driver it works fine. If I can provide any > more information, please let me know. I do have a serial console on one > machine and will have physical access to it once again on Monday. > > ---Mike > > > At 08:53 PM 8/10/2002 -0400, Mike Tancsa wrote: > > >Actually, I am still seeing a lot of strange things. The patch certainly > >fixed the panics, but now I am seeing packet loss on links off the same subnet. > > > >e.g. > >32 packets transmitted, 23 packets received, 28% packet loss > >round-trip min/avg/max/stddev = 0.333/0.348/0.389/0.015 ms > > > >If I reboot with an old kernel its fine. > > > >Even stranger..... If I do a dump from a machine a few hops away, I get my > >normal throughput. > > > >netstat -ni does not show any problems, nor do the switch counters show > >any issues. These are with fxp cards. Could it be the latest fxp commit ? > > > > ---Mike > > > > > > > >At 07:45 AM 8/10/2002 -0700, Randy Bush wrote: > >> > Could someone put the word out when this issue is fixed? This problem > >> > just hosed my web server after I forgot the prime directive: test on a > >> > non-critical machine. > >> > >>it was fixed > >> > >>randy > >> > >> > >>To Unsubscribe: send mail to [EMAIL PROTECTED] > >>with "unsubscribe freebsd-stable" in the body of the message > > > >-------------------------------------------------------------------- > >Mike Tancsa, tel +1 519 651 3400 > >Sentex Communications, [EMAIL PROTECTED] > >Providing Internet since 1994 www.sentex.net > >Cambridge, Ontario Canada www.sentex.net/mike > > > > > >To Unsubscribe: send mail to [EMAIL PROTECTED] > >with "unsubscribe freebsd-stable" in the body of the message > > -------------------------------------------------------------------- > Mike Tancsa, tel +1 519 651 3400 > Sentex Communications, [EMAIL PROTECTED] > Providing Internet since 1994 www.sentex.net > Cambridge, Ontario Canada www.sentex.net/mike > > > To Unsubscribe: send mail to [EMAIL PROTECTED] > with "unsubscribe freebsd-stable" in the body of the message > -- MfG O. Hartmann [EMAIL PROTECTED] ---------------------------------------------------------------- IT-Administration des Institut fuer Physik der Atmosphaere (IPA) ---------------------------------------------------------------- Johannes Gutenberg Universitaet Mainz Becherweg 21 55099 Mainz Tel: +496131/3924662 (Maschinensaal) Tel: +496131/3924144 FAX: +496131/3923532 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
