After upgrading to 2.4.0-test2 I'm suddenly noticing a resurection of an OLD problem I use to see a lot in the Boomerang / Vortex drivers. Get the network congestion up to a certain level and the driver suddenly goes dain-bramaged and you get and never ending stream of theses on the console and in the syslog: Jul 5 10:18:35 alcove kernel: eth0: transmit timed out, tx_status 00 status e000. Jul 5 10:18:35 alcove kernel: Flags; bus-master 1, full 1; dirty 29160(8) current 29176(8). Jul 5 10:18:35 alcove kernel: Transmit list 012d0280 vs. c12d0280. Jul 5 10:18:35 alcove kernel: 0: @c12d0200 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 1: @c12d0210 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 2: @c12d0220 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 3: @c12d0230 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 4: @c12d0240 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 5: @c12d0250 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 6: @c12d0260 length 80000042 status 80000042 Jul 5 10:18:35 alcove kernel: 7: @c12d0270 length 80000042 status 80000042 Jul 5 10:18:35 alcove kernel: 8: @c12d0280 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 9: @c12d0290 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 10: @c12d02a0 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 11: @c12d02b0 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 12: @c12d02c0 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 13: @c12d02d0 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 14: @c12d02e0 length 80000042 status 00000042 Jul 5 10:18:35 alcove kernel: 15: @c12d02f0 length 80000042 status 00000042 At that point, it never recovers and I end up having to sync, unmount, and reboot via Magic SysRq (maybe reboot would work if I gave it a few hours to time out all the processes). Gets REALLY bad if I'm doing any intense NFS work. I don't recall seeing this on test1, but I may not have been stressing it quite so much. I've got two systems running 2.4.0-test2, both with 3c900 cards and both showing the same problem. Thoughts? Anything I should try or examine? Mike -- Michael H. Warfield | (770) 985-6132 | mhw@WittsEnd.com (The Mad Wizard) | (770) 331-2437 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471 | possible worlds. A pessimist is sure of it! - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.rutgers.edu
