Menu
▾
▴
Re: [E1000-devel] Problem with igb i210 and multiple NICs on one board
|
From: Todor I. <ti...@xp...> - 2016-10-13 14:51:14
|
Thank you very much for the support and guidelines. I already contacted the vendor as you suggested. Regards, Todor Best regards, Todor Ivanov Senior System Administrator Xpeqt Ltd 2, Samokovsko Shose Blvd Sofia 1138, Bulgaria Mobile: *+359 887 383 600* On Thu, Oct 13, 2016 at 5:47 PM, Fujinaka, Todd <tod...@in...> wrote: > igb-5.3.5.4 is the newest. 4.1.2 is from January of 2013. We also don’t > support Debian directly, which is why I asked you to try a live DVD. > > > > Multiple cards with the multiple outcomes points to hardware failure. Or > if you mean multiple controllers on the motherboard, same thing. > > > > You could file a bug on sourceforge and attach the full logs, but I’m > still thinking this is a hardware failure. > > > > *Todd Fujinaka* > > Software Application Engineer > > Networking Division (ND) > > Intel Corporation > > *tod...@in... <tod...@in...>* > > (503) 712-4565 > > > > *From:* Todor Ivanov [mailto:ti...@xp...] > *Sent:* Thursday, October 13, 2016 7:19 AM > *To:* Fujinaka, Todd <tod...@in...> > *Cc:* e10...@li... > *Subject:* Re: [E1000-devel] Problem with igb i210 and multiple NICs on > one board > > > > It is not a cabling issue, as I did tests with same cable on both ports > and one of them is 100% working. We are using this setup with multiple > boards. > > > > Changing the kernel will be a heavy task, as we have this highly > customized distribution to fit our specific production needs: Debian > Squeeze with kernel 2.6.32 > > According to the driver: This driver supports kernel versions 2.6.30 or > newer. > > Before requesting help from the vendor, I hoped a developer could quick > check the e1000_i210 source to see if something could be related to the > specific problem we are experiencing. As I said two identical ethernet > cards on same board behave differently. From what I'm seeing the vendor > Adlink is suggesting the same driver, but older version: igb-4.1.2, which > is behaving in the same way as the newest 3.5.3.5. > > > > > > Todor > > > > > > > > > > On Thu, Oct 13, 2016 at 5:00 PM, Fujinaka, Todd <tod...@in...> > wrote: > > I would try a newer kernel first, maybe a live DVD. Not sure where you got > yours. Also unsure of what OS you're running. > > Sounds like a hardware issue as the ports are coming up but not finding > link. I'd try new cables first and then contact Adlink. > > Todd Fujinaka > Software Application Engineer > Networking Division (ND) > Intel Corporation > tod...@in... > (503) 712-4565 > > -----Original Message----- > From: Todor Ivanov [mailto:ti...@xp...] > Sent: Thursday, October 13, 2016 3:35 AM > To: e10...@li... > Subject: [E1000-devel] Problem with igb i210 and multiple NICs on one board > > Hi, All, > > We bough a new integrated PC from Adlink ePCI-3510BLG, which has > three physical i210 (8086:1533) interfaces with two ethernet RJ45 > terminations. > > We are facing issues getting the second network card to work correctly. > The problem is that we see three network interfaces and only one of them > works. > Our kernel is: Linux 2.6.32-5-686 i686 GNU/Linux > > We tried drivers: igb-4.1.2 (Adlink website) (5.2.9.4/5.2.15/igb-5.3.5.3 > Intel website) > > lspci | grep Ethernet > > 03:00.0 Ethernet controller: Intel Corporation Device 1533 (rev 03) > 04:00.0 Ethernet controller: Intel Corporation Device 1533 (rev 03) > 07:00.0 Ethernet controller: Intel Corporation Device 1533 (rev 03) > > Drivers starts normally: > > [ 4649.365204] Intel(R) Gigabit Ethernet Network Driver - version 4.1.2 [ > 4649.369385] Copyright (c) 2007-2012 Intel Corporation. > [ 4649.373538] igb 0000:03:00.0: PCI INT A -> GSI 21 (level, low) -> IRQ > 21 [ 4649.377788] igb 0000:03:00.0: setting latency timer to 64 [ > 4649.378126] igb: 0000:03:00.0: igb_check_options: VMDq option is not > supported. > [ 4649.382623] igb 0000:03:00.0: irq 31 for MSI/MSI-X [ 4649.382625] igb > 0000:03:00.0: irq 32 for MSI/MSI-X [ 4649.438254] igb 0000:03:00.0: > Intel(R) Gigabit Ethernet Network Connection [ 4649.442769] igb > 0000:03:00.0: eth0: (PCIe:2.5GT/s:Width x1) [ 4649.443006] igb > 0000:03:00.0: eth0: MAC: 00:30:64:16:33:35 [ 4649.451885] igb 0000:03:00.0: > eth0: PBA No: 000400-000 [ 4649.460493] igb 0000:03:00.0: LRO is disabled [ > 4649.464750] igb 0000:03:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx > queue(s) [ 4649.469207] igb 0000:04:00.0: PCI INT A -> GSI 22 (level, low) > -> IRQ 22 [ 4649.473695] igb 0000:04:00.0: setting latency timer to 64 [ > 4649.474037] igb: 0000:04:00.0: igb_check_options: VMDq option is not > supported. > [ 4649.478624] igb 0000:04:00.0: irq 33 for MSI/MSI-X [ 4649.478625] igb > 0000:04:00.0: irq 34 for MSI/MSI-X [ 4649.516885] igb 0000:04:00.0: > Intel(R) Gigabit Ethernet Network Connection [ 4649.521582] igb > 0000:04:00.0: eth1: (PCIe:2.5GT/s:Width x1) [ 4649.521744] igb > 0000:04:00.0: eth1: MAC: 00:30:64:16:33:36 [ 4649.530902] igb 0000:04:00.0: > eth1: PBA No: 000400-000 [ 4649.540338] igb 0000:04:00.0: LRO is disabled [ > 4649.544804] igb 0000:04:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx > queue(s) [ 4649.549409] igb 0000:07:00.0: PCI INT A -> GSI 18 (level, low) > -> IRQ 18 [ 4649.554068] igb 0000:07:00.0: setting latency timer to 64 [ > 4649.554439] igb: 0000:07:00.0: igb_check_options: VMDq option is not > supported. > [ 4649.559170] igb 0000:07:00.0: irq 35 for MSI/MSI-X [ 4649.559171] igb > 0000:07:00.0: irq 36 for MSI/MSI-X [ 4649.596770] igb 0000:07:00.0: > Intel(R) Gigabit Ethernet Network Connection [ 4649.601623] igb > 0000:07:00.0: eth2: (PCIe:2.5GT/s:Width x1) [ 4649.601786] igb > 0000:07:00.0: eth2: MAC: 00:30:64:16:33:37 [ 4649.611346] igb 0000:07:00.0: > eth2: PBA No: 000400-000 [ 4649.620185] igb 0000:07:00.0: LRO is disabled [ > 4649.624866] igb 0000:07:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx > queue(s) [ 4651.172103] ADDRCONF(NETDEV_UP): eth0: link is not ready [ > 4651.990442] ADDRCONF(NETDEV_UP): eth1: link is not ready [ 4652.825065] > ADDRCONF(NETDEV_UP): eth2: link is not ready [ 4655.549405] igb: eth2 NIC > Link is Up 1000 Mbps Full Duplex, Flow > Control: None > [ 4655.554425] ADDRCONF(NETDEV_CHANGE): eth2: link becomes ready > > > But only one link is up (We have cables in both. Also tried them in a row) > > eth0 Link encap:Ethernet HWaddr 00:30:64:16:33:35 > UP BROADCAST MULTICAST MTU:1500 Metric:1 > RX packets:0 errors:0 dropped:0 overruns:0 frame:0 > TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1000 > RX bytes:0 (0.0 B) TX bytes:0 (0.0 B) > > eth1 Link encap:Ethernet HWaddr 00:30:64:16:33:36 > UP BROADCAST MULTICAST MTU:1500 Metric:1 > RX packets:0 errors:0 dropped:0 overruns:0 frame:0 > TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1000 > RX bytes:0 (0.0 B) TX bytes:0 (0.0 B) > > eth2 Link encap:Ethernet HWaddr 00:30:64:16:33:37 > inet addr:10.35.91.197 Bcast:10.35.95.255 Mask:255.255.240.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:47540 errors:0 dropped:0 overruns:0 frame:0 > TX packets:1490 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1000 > RX bytes:4603113 (4.3 MiB) TX bytes:164743 (160.8 KiB) > > > root@XPEQTSIM:/mnt/source/igb-4.1.2-fromAdlink# ip link > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > 14: eth0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state > DOWN qlen 1000 > link/ether 00:30:64:16:33:35 brd ff:ff:ff:ff:ff:ff > 15: eth1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state > DOWN qlen 1000 > link/ether 00:30:64:16:33:36 brd ff:ff:ff:ff:ff:ff > 16: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP > qlen > 1000 > link/ether 00:30:64:16:33:37 brd ff:ff:ff:ff:ff:ff > > > The drivers would not compile unless we put the following code into > /usr/src/linux-headers-2.6.32-5-common/include/linux/pm.h : > > #ifdef CONFIG_PM_RUNTIME > #define SET_RUNTIME_PM_OPS(suspend_fn, resume_fn, idle_fn) \ > .runtime_suspend = suspend_fn, \ > .runtime_resume = resume_fn, \ > .runtime_idle = idle_fn, > #else > #define SET_RUNTIME_PM_OPS(suspend_fn, resume_fn, idle_fn) #endif > > I played with bios, but there are not network related options. > Do you know where the problem might be? > > Best regards, > Todor Ivanov > Senior System Administrator > > Xpeqt Ltd > 2, Samokovsko Shose Blvd > Sofia 1138, Bulgaria > Mobile: *+359 887 383 600* > > > |
