From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.7.6) Gecko/20050325 Firefox/1.0.2 Description of problem: This was first posted as Service Request: 509882 I am getting reproducable Kernel panics with a dual processor AMD64 machine using the Broadcom 5704 on-board network chip (Tyan S2882 motherboard). To reproduce the error all I have to do is stess the network driver by downloading something large via scp. Here is the kernel panic message: Kernel BUG at tg3:2426 invalid operand: 0000 [1] SMP CPU 1 Modules linked in: arpt_mangle arptable_filter arp_tables iptable_filter ip_tabd Pid: 0, comm: swapper Tainted: G M 2.6.9-5.0.3.ELsmp RIP: 0010:[<ffffffffa00a1c4f>] <ffffffffa00a1c4f>{:tg3:tg3_poll+163} RSP: 0000:00000102fffb3e68 EFLAGS: 00010046 RAX: 00000000000000a9 RBX: 00000102fd5cbfd8 RCX: 0000000000000002 RDX: ffffffff804bf680 RSI: 00000002ffff849e RDI: 0000000000000002 RBP: 00000000000000a9 R08: 0000000000000042 R09: 0000000000000001 R10: 0000000000000001 R11: 0000000000000002 R12: 0000000000000000 R13: 00000102fd9c4380 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000041401960(0000) GS:ffffffff804bf680(0000) knlGS:0000000008479be0 CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b CR2: 0000002a9556c000 CR3: 0000000037e62000 CR4: 00000000000006e0 Process swapper (pid: 0, threadinfo 00000102fffb8000, task 000001020000b030) Stack: 00000102fd9c43ec 000000016c67e812 0000010037dd5b80 000000018012fd93 000002dd02dde9e0 0000003f00010000 0000000100000040 0000000000000212 00000100f3d82000 00000102fffb3f1c Call Trace:<IRQ> <ffffffff8029eaa7>{net_rx_action+129} <ffffffff80139cc4>{__do_ <ffffffff80139d6d>{do_softirq+49} <ffffffff80112dc3>{do_IRQ+328} <ffffffff801105d7>{ret_from_intr+0} <EOI> <ffffffff8010e5c9>{default_id <ffffffff8010e5e9>{default_idle+32} <ffffffff8010e65c>{cpu_idle+26} Code: 0f 0b 10 a0 0a a0 ff ff ff ff 7a 09 49 8b 4d 20 41 8b 94 24 RIP <ffffffffa00a1c4f>{:tg3:tg3_poll+163} RSP <00000102fffb3e68> NMI Watchdog detected LOCKUP on CPU1, registers: CPU 1 Modules linked in: arpt_mangle arptable_filter arp_tables iptable_filter ip_tabd Pid: 0, comm: swapper Tainted: G M 2.6.9-5.0.3.ELsmp RIP: 0010:[<ffffffff802f86c4>] <ffffffff802f86c4>{.text.lock.spinlock+14} RSP: 0000:00000102fffb3b60 EFLAGS: 00000086 RAX: 0000000000000000 RBX: 00000102fd9c4380 RCX: 00000000000000c1 RDX: 00000102fffb3c18 RSI: 00000102fd9c4000 RDI: 00000102fd9c4380 RBP: 00000100f3d82000 R08: ffffffff803c7188 R09: 000000000000000f R10: ffffffff801ea63a R11: ffffffff801ea63a R12: 00000102fd9c4380 R13: 0000000000000001 R14: 00000102fffb3c18 R15: 0000000000000000 Version-Release number of selected component (if applicable): kernel-2.6.9-5.0.3.ELsmp How reproducible: Always Steps to Reproduce: 1. Install RHEL4 x86_64 with minimal install 2. Start downloading large file via scp Additional info:
Do you have access to a later kernel? There has been a tg3 update to version 3.22rh. What version does "ethtool -i" show for you? Please attempt to recreate with the latest available kernel. If nothing else, please try the kernels here: http://people.redhat.com/linville/kernels/rhel4/ Please post the results of trying to recreate with the later kernels. Thanks!
Here is the output from ethtool -i: driver: tg3 version: 3.10 firmware-version: bus-info: 0000:02:09.0 I'll give one of your kernels a shot and report back what happens. Thanks for the update.
I still had the same problems with your kernel. I'm going to RMA the mobo and see if that is the problem. If I still have problems with a new mobo than I will report back and reopen the bug. For now I just consider this to be a hardware problem. Thanks for the help.