Hi all, I have a pretty straightforward setup. Hypervisor: dual xeon e5205 running Gentoo Linux kernel 2.6.27 with virtio devices enabled kvm 84 libvirt 0.5.1 Guest: 32-bit, virtio for nic and disk, qcow2. Linux 2.6.28. Network is bridged using tap and brctl. I'm running Apache on the guest. Whenever I send enough data through the virtual NIC, I get a panic in skb_copy_bits. I've tried using the e1000 driver instead of the virtio one, but that makes no difference. Has anyone else seen this behavior before? I got this on 2.6.27 and 2.6.28. Here's a snippet: [280204.340016] [<c02253e2>] panic+0x4e/0xea [280204.340016] [<c05906b9>] oops_end+0x8f/0xa3 [280204.340016] [<c0204e94>] die+0x57/0x5f [280204.340016] [<c0592192>] do_page_fault+0x605/0x6bc [280204.340016] [<c059010d>] ? _spin_lock+0x15/0x18 [280204.340016] [<c04bfad8>] ? __qdisc_run+0xe6/0x1a7 [280204.340016] [<c0591b8d>] ? do_page_fault+0x0/0x6bc [280204.340016] [<c0590482>] error_code+0x72/0x78 [280204.340016] [<c04ace8c>] ? skb_copy_bits+0x4f/0x1c4 [280204.340016] [<c0215864>] ? kvm_set_pte+0x26/0x29 [280204.340016] [<c054889c>] xdr_skb_read_bits+0x1f/0x37 [280204.340016] [<c054872b>] xdr_partial_copy_from_skb+ 0x117/0x16c [280204.340016] [<c0549ec4>] xs_tcp_data_recv+0x245/0x3de [280204.340016] [<c054887d>] ? xdr_skb_read_bits+0x0/0x37 [280204.340016] [<c04e07d6>] tcp_read_sock+0x8c/0x1e2 [280204.340016] [<c0549c7f>] ? xs_tcp_data_recv+0x0/0x3de [280204.340016] [<c054a5d1>] xs_tcp_data_ready+0x54/0x64 [280204.340016] [<c04e9469>] tcp_rcv_established+0x524/0x7b7 [280204.340016] [<c04ee4b2>] tcp_v4_do_rcv+0x173/0x2dc -- Justin Dossey -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
