Hi, I tried turning off EEE, sadly it didn't change anything. I was going to upload these files to a bug on sourceforge but I didn't see one open for this issue and it would not let me create one.
The logs are here, for both good and bad boots: http://repo.ezzi.net/igblog/ The hardware in question is this board for the logs I posted: http://www.supermicro.com/products/motherboard/Atom/X10/A1SRi-2758F.cfm I also had the same issue occur on this board previously: http://www.supermicro.com/products/motherboard/Xeon/C220/X10SL7-F.cfm but I don't run Linux on that one anymore and works fine under FreeBSD, but when I noticed it on the Xeon board it was running Debian 7.5 For reproducing it just putting latest OpenWRT on it with static ip set on the first card, then reboot the machine a few times and 40-60% of the time the NIC will fail to come up. FreeBSD worked fine on this machine also. BIOS is updated to latest version. Both machines are on completely different switches as well, so that is definitely not related. Thanks -----Original Message----- From: Fujinaka, Todd [mailto:todd.fujin...@intel.com] Sent: Monday, January 11, 2016 11:22 AM To: Fujinaka, Todd <todd.fujin...@intel.com>; Len White <lwh...@nrw.ca> Cc: e1000-devel@lists.sourceforge.net Subject: RE: C2758 i354 link issue Oh, and someone just reminded me that this might be an EEE interoperability issue. Have you tried turning off EEE? Todd Fujinaka Software Application Engineer Networking Division (ND) Intel Corporation todd.fujin...@intel.com (503) 712-4565 -----Original Message----- From: Fujinaka, Todd [mailto:todd.fujin...@intel.com] Sent: Monday, January 11, 2016 6:30 AM To: Len White Cc: e1000-devel@lists.sourceforge.net Subject: Re: [E1000-devel] C2758 i354 link issue While that's useful, we would also like reproduction steps. We'll need exact hardware and software setups and the steps you took to get there. Send the full dmesg, "lspci -vvv", eeprom dump, and registers dumps of both the failed and passing cases. It would be easiest if you send those as separate files and attach them to a bug opened on sourceforge. Thanks. Todd Fujinaka Software Application Engineer Networking Division (ND) Intel Corporation todd.fujin...@intel.com (503) 712-4565 From: Len White [mailto:lwh...@nrw.ca] Sent: Sunday, January 10, 2016 10:04 PM To: Fujinaka, Todd Cc: e1000-devel@lists.sourceforge.net Subject: RE: C2758 i354 link issue So I've compiled the driver with debug mode turned on, and I have two boot logs and a log of workaround to share with you, the first one is it booting normally and the link comes up normally without any workarounds, the second is when it fails. Good boot: http://repo.ezzi.net/igb1.txt Fail boot: http://repo.ezzi.net/igb2.txt For the most part they are very similar, until it hits e1000_check_downshift_generic and then its totally different. The time it worked correctly it hit: "e1000_check_downshift_generic" e1000_put_hw_semaphore_generic e1000_read_phy_reg_mdic e1000_release_phy_82575 e1000_release_swfw_sync_82575 e1000_get_hw_semaphore_generic e1000_put_hw_semaphore_generic e1000_check_downshift_generic e1000_read_phy_reg_82580 e1000_acquire_phy_82575 e1000_acquire_swfw_sync_82575 VS: e1000_put_hw_semaphore_generic e1000_read_phy_reg_mdic e1000_release_phy_82575 e1000_release_swfw_sync_82575 e1000_get_hw_semaphore_generic e1000_put_hw_semaphore_generic <this is where it changes on the failure boot> e1000_check_for_link_82575 e1000_check_for_copper_link e1000_phy_has_link_generic e1000_read_phy_reg_82580 e1000_acquire_phy_82575 e1000_acquire_swfw_sync_82575 e1000_get_hw_semaphore_generic and it will sit here looping forever: [ 1513.308318] e1000_put_hw_semaphore_generic [ 1513.313244] e1000_read_phy_reg_82580 [ 1513.317666] e1000_acquire_phy_82575 [ 1513.321996] e1000_acquire_swfw_sync_82575 [ 1513.326851] e1000_get_hw_semaphore_generic [ 1513.331798] e1000_put_hw_semaphore_generic [ 1513.336732] e1000_read_phy_reg_mdic [ 1513.341118] e1000_release_phy_82575 [ 1513345449] e1000_release_swfw_sync_82575 [ 1513.350303] e1000_get_hw_semaphore_generic [ 1513.355250] e1000_put_hw_semaphore_generic [ 1514.233267] e1000_check_for_link_82575 [ 1514.237881] e1000_check_for_copper_link [ 1514.242558] e1000_phy_has_link_generic [ 1514.247153] e1000_read_phy_reg_82580 [ 1514.251571] e1000_acquire_phy_82575 [ 1514.255895] e1000_acquire_swfw_sync_82575 [ 1514.260753] e1000_get_hw_semaphore_generic [ 1514.265699] e1000_put_hw_semaphore_generic Over and over and over. If I apply my workaround with ethtool, here is the output of how it changes: http://repo.ezzi.net/igb_workaround.txt and then it works normally. You can also see e1000_check_downshift_generic gets called this time. Just for a test in e1000_macc, I commented this out: // if (!link) // return E1000_SUCCESS; /* No link detected */ And it did allow the network card to come up on boot on what otherwise would have been a failed boot, but there was still a interruption in the link (it went down then it came back up), and every 60-90 seconds it would go down again for 10-15 seconds in a loop. Hopefully this helps narrow it down From: Fujinaka, Todd [mailto:todd.fujin...@intel.com] Sent: Sunday, January 10, 2016 8:17 PM To: Len White <lwh...@nrw.ca<mailto:lwh...@nrw.ca>> Cc: e1000-devel@lists.sourceforge.net<mailto:e1000-devel@lists.sourceforge.net> Subject: RE: C2758 i354 link issue Kristian didn't reply to the mailing list but if you check e1000-bugs you'll see he resolved his issues. I'm still not clear on what the problem was, but you can check there for more details. In any case, I'm guessing your problem isn't the same as his problem. I think I asked him to try a different distro and I don't think he did that. I'm still not convinced this isn't a problem with OpenWRT but then again you've never said what OS you're running. If it is OpenWRT, have you tried a different distro? Todd Fujinaka Software Application Engineer Networking Division (ND) Intel Corporation todd.fujin...@intel.com<mailto:todd.fujin...@intel.com> (503) 712-4565 From: Len White [mailto:lwh...@nrw.ca] Sent: Sunday, January 10, 2016 3:01 PM To: Fujinaka, Todd Cc: e1000-devel@lists.sourceforge.net<mailto:e1000-devel@lists.sourceforge.net> Subject: C2758 i354 link issue Hi, I'm having the same problem as Kristian is with his igb link issue. I also have a different machine with an i210 card in it, that also has the same issue on Linux ONLY. FreeBSD works fine. It's about 40% chance on boot that the issue will happen (but I did find a workaround). I believe it's something to do with auto-negotiation code. I can either unplug/replug the cable and it will come back up, or I can force the link speed/duplex and it will bring the link back up when the issue is triggered. Ethtool on a boot with issue: root@moffat-gw / # ethtool eth0 Settings for eth0: Supported ports: [ TP ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Supported pause frame use: Symmetric Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Advertised pause frame use: Symmetric Advertised auto-negotiation: Yes Speed: Unknown! Duplex: Unknown! (255) Port: Twisted Pair PHYAD: 0 Transceiver: internal Auto-negotiation: on MDI-X: off (auto) Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000007 (7) drv probe link Link detected: no root@moffat-gw / # ethtool -s eth0 speed 1000 duplex full root@moffat-gw / # ethtool eth0[ 59.285152] igb 0000:00:14.0 eth0: igb: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX [ 59.295788] br-lan: port 1(eth0) entered forwarding state [ 59.302438] br-lan: port 1(eth0) entered forwarding state [ 59.309274] IPv6: ADDRCONF(NETDEV_CHANGE): br-lan: link becomes ready Settings for eth0: Supported ports: [ TP ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Supported pause frame use: Symmetric Supports auto-negotiation: Yes Advertised link modes: 1000baseT/Full Advertised pause frame use: Symmetric Advertised auto-negotiation: Yes Speed: 1000Mb/s Duplex: Full Port: Twisted Pair PHYAD: 0 Transceiver: internal Auto-negotiation: on MDI-X: off (auto) Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000007 (7) drv probe link Link detected: yes root@moffat-gw / # ethtool -i eth0 driver: igb version: 5.3.3.5 firmware-version: 0.0.0 expansion-rom-version: bus-info: 0000:00:14.0 supports-statistics: yes supports-test: yes supports-eeprom-access: yes supports-register-dump: yes supports-priv-flags: no root@moffat-gw / # uname -a Linux moffat-gw 4.1.15 #12 SMP Sun Jan 10 22:09:02 EST 2016 x86_64 GNU/Linux root@moffat-gw / # ethtool -S eth0 NIC statistics: rx_packets: 12463 tx_packets: 9238 rx_bytes: 2548091 tx_bytes: 1699471 rx_broadcast: 191 tx_broadcast: 6 rx_multicast: 3073 tx_multicast: 91 multicast: 3073 collisions: 0 rx_crc_errors: 0 rx_no_buffer_count: 0 rx_missed_errors: 0 tx_aborted_errors: 0 tx_carrier_errors: 0 tx_window_errors: 0 tx_abort_late_coll: 0 tx_deferred_ok: 0 tx_single_coll_ok: 0 tx_multi_coll_ok: 0 tx_timeout_count: 0 rx_long_length_errors: 0 rx_short_length_errors: 0 rx_align_errors: 0 tx_tcp_seg_good: 0 tx_tcp_seg_failed: 0 rx_flow_control_xon: 0 rx_flow_control_xoff: 0 tx_flow_control_xon: 0 tx_flow_control_xoff: 0 rx_long_byte_count: 2548091 tx_dma_out_of_sync: 0 lro_aggregated: 0 lro_flushed: 0 tx_smbus: 3068 rx_smbus: 3106 dropped_smbus: 0 os2bmc_rx_by_bmc: 2 os2bmc_tx_by_bmc: 3068 os2bmc_tx_by_host: 2 os2bmc_rx_by_host: 3068 tx_hwtstamp_timeouts: 0 rx_hwtstamp_cleared: 0 rx_errors: 0 tx_errors: 0 tx_dropped: 0 rx_length_errors: 0 rx_over_errors: 0 rx_frame_errors: 0 rx_fifo_errors: 0 tx_fifo_errors: 0 tx_heartbeat_errors: 0 tx_queue_0_packets: 6170 tx_queue_0_bytes: 400752 tx_queue_0_restart: 0 rx_queue_0_packets: 12434 rx_queue_0_bytes: 2916614 rx_queue_0_drops: 0 rx_queue_0_csum_err: 0 rx_queue_0_alloc_failed: 0 ________________________________ No viruses found in this message ________________________________ No viruses found in this message ________________________________ No viruses found in this message ----- No viruses found in this message ----- No viruses found in this message ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 _______________________________________________ E1000-devel mailing list E1000-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
