Using the 6.0-CURRENT June 2nd Snapshot, I get a panic pretty quickly when using an original NForce 1 MCP Network Card(new nve driver). This has happened while I was running sysinstall via pxeboot, and starting a ftp install. This is what I copied from the screen, omitting some things later in the stack trace: anic: nve_ifstart: attempted use of a free mbuf! cpuid = 0 KBD: enter: panic [thread pid 35 tid 100022 ] Stopped at kdb_enter+0x2b db>trace kdb_enter(x0855c9c) at kdb_enter 0x2b panic(c082ad05,c08154dc,c1a388fc,c1a38800,ce7ebc80) at panic+0x127 nve_ifstart(c1a38800,) at nve_ifstart+0x366 if_start(c1a38800) at if_start+0x7b ether_output_frame(c1a38800,v1b35300,0,0,0) at ether_output_frame+0x1d9 ether_output(c1a38800,v1b35300,,c1ab67d0,c1ba8c60,v1b58b00) at ether_output+0x384 ip_output() tcp_output() ip_input() netisr_processqueue() swi_net(0) .... I don't have the gear locally to setup a serial console, so I am hoping this is enough to give someone a clue. It looks easy to reproduce, and I am happy to provide more information if needed. I did find a mention of a similar panic back in October of 2004, but it was for the fxp driver: http://lists.freebsd.org/pipermail/freebsd-net/2004-October/005222.html Thanks, -Paul QuernaReceived on Sun Jun 12 2005 - 23:49:43 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:36 UTC