We have motherboard with Marvell 88E8056 ethernet controllers. ------ 02:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8056 PCI-E Gigabit Ethernet Controller (rev 14) 03:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8056 PCI-E Gigabit Ethernet Controller (rev 14) ------ It is used for routing network packets between two interfaces. This system has been tested for several days using pktgen with different packet size and bandwidth settings and found problem similar to http://marc.info/?l=linux-net&m=121967539203137&w=2 Firmware was updated according to that thread instructions, thereafter rev. changed from 12 to 14. But interfaces periodically starting to loose a large number of packets and netdev_rx_csum_fault threw. --------------------------------------------------------------------------------- sky2 eth1: rx length error: status 0x5d60100 length 1702 Pid: 0, comm: swapper Tainted: G W 2.6.30 #2 Call Trace: <IRQ> [<ffffffff803fe19c>] ? netdev_rx_csum_fault+0x29/0x31 [<ffffffff803f9445>] ? __skb_checksum_complete_head+0x43/0x55 [<ffffffffa029a203>] ? udp_error+0x121/0x179 [nf_conntrack] [<ffffffffa029688e>] ? nf_conntrack_in+0x122/0x5ac [nf_conntrack] [<ffffffff803544d1>] ? swiotlb_map_page+0xac/0x134 [<ffffffff80354425>] ? swiotlb_map_page+0x0/0x134 [<ffffffff80353b1f>] ? swiotlb_virt_to_bus+0xc/0x18 [<ffffffff80353b59>] ? swiotlb_dma_mapping_error+0x10/0x1b [<ffffffff80419232>] ? nf_iterate+0x41/0x7d [<ffffffff8041ef30>] ? ip_rcv_finish+0x0/0x2f0 [<ffffffff804192d0>] ? nf_hook_slow+0x62/0xc3 [<ffffffff8041ef30>] ? ip_rcv_finish+0x0/0x2f0 [<ffffffff802506f5>] ? ktime_get_real+0xc/0x43 [<ffffffff8041f454>] ? ip_rcv+0x234/0x279 [<ffffffffa01adb8f>] ? sky2_poll+0x80c/0xb7d [sky2] [<ffffffff80475f4f>] ? _spin_lock_bh+0x9/0x1f [<ffffffffa02964a0>] ? destroy_conntrack+0xef/0xf3 [nf_conntrack] [<ffffffff803fc0e5>] ? net_rx_action+0xa8/0x190 [<ffffffff8023ff85>] ? __do_softirq+0xac/0x173 [<ffffffff8020cbcc>] ? call_softirq+0x1c/0x28 [<ffffffff8020e49e>] ? do_softirq+0x3a/0x7e [<ffffffff8023fc82>] ? irq_exit+0x3f/0x7c [<ffffffff8020dd72>] ? do_IRQ+0xa9/0xbf [<ffffffff8020c453>] ? ret_from_intr+0x0/0x11 <EOI> [<ffffffff80603140>] ? early_idt_handler+0x0/0x71 [<ffffffff8021286b>] ? mwait_idle+0xab/0xe8 [<ffffffff80250e21>] ? notifier_call_chain+0x29/0x4c [<ffffffff80603140>] ? early_idt_handler+0x0/0x71 [<ffffffff80603140>] ? early_idt_handler+0x0/0x71 [<ffffffff8020adda>] ? cpu_idle+0x50/0x91 [<ffffffff80603c39>] ? start_kernel+0x37a/0x386 [<ffffffff8060339b>] ? x86_64_start_kernel+0xf9/0x106 --------------------------------------------------------------------------------- At that time two interfaces routed symmetric 150mbit/s traffic flow with different packet size, approximately with 100-200 kpps throughput. There is a default debian 2.6.30 x86_64 kernel and dual-core intel cpu. We've tried to simulate this situation using pktgen. Our pktgen has made a heavy load, great than real, but all NICs worked fine. May be yukon chip is going crazy only by special formed packet or sequence of packets, so it is not clear yet how this situation can be modeled. Does anybody have similar problems? Is this a hardware problem of 88E8056? -- WBR, Tsisyk Roman -- To unsubscribe from this list: send the line "unsubscribe linux-net" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
