Hello again Don, We've done as you suggested and pulled the cable resulting in some new information bringing us closer to the answer, now it looks as though the issue is actually only one of 'active' cables being used, so back to the switch vendor for now. In the meantime I still have that additional information in case something else grabs your attention ..
[521634.994776] ixgbe: Intel(R) 10 Gigabit PCI Express Network Driver - version 2.0.62-k2 [521634.995192] ixgbe: Copyright (c) 1999-2010 Intel Corporation. [521634.995492] ixgbe 0000:04:00.0: PCI INT A -> GSI 30 (level, low) -> IRQ 30 [521634.995725] ixgbe 0000:04:00.0: setting latency timer to 64 [521635.066141] alloc irq_desc for 83 on node -1 [521635.066352] alloc kstat_irqs on node -1 [521635.066575] ixgbe 0000:04:00.0: irq 83 for MSI/MSI-X [521635.066818] alloc irq_desc for 84 on node -1 [521635.067028] alloc kstat_irqs on node -1 [521635.067239] ixgbe 0000:04:00.0: irq 84 for MSI/MSI-X [521635.067448] alloc irq_desc for 85 on node -1 [521635.067655] alloc kstat_irqs on node -1 [521635.067874] ixgbe 0000:04:00.0: irq 85 for MSI/MSI-X [521635.068091] alloc irq_desc for 86 on node -1 [521635.068299] alloc kstat_irqs on node -1 [521635.068518] ixgbe 0000:04:00.0: irq 86 for MSI/MSI-X [521635.068734] alloc irq_desc for 87 on node -1 [521635.068943] alloc kstat_irqs on node -1 [521635.069153] ixgbe 0000:04:00.0: irq 87 for MSI/MSI-X [521635.069363] alloc irq_desc for 88 on node -1 [521635.069572] alloc kstat_irqs on node -1 [521635.069790] ixgbe 0000:04:00.0: irq 88 for MSI/MSI-X [521635.070006] alloc irq_desc for 89 on node -1 [521635.070217] alloc kstat_irqs on node -1 [521635.070428] ixgbe 0000:04:00.0: irq 89 for MSI/MSI-X [521635.070651] alloc irq_desc for 90 on node -1 [521635.070867] alloc kstat_irqs on node -1 [521635.071080] ixgbe 0000:04:00.0: irq 90 for MSI/MSI-X [521635.071291] alloc irq_desc for 91 on node -1 [521635.071501] alloc kstat_irqs on node -1 [521635.071721] ixgbe 0000:04:00.0: irq 91 for MSI/MSI-X [521635.071938] alloc irq_desc for 92 on node -1 [521635.072146] alloc kstat_irqs on node -1 [521635.072356] ixgbe 0000:04:00.0: irq 92 for MSI/MSI-X [521635.072576] alloc irq_desc for 93 on node -1 [521635.072793] alloc kstat_irqs on node -1 [521635.073006] ixgbe 0000:04:00.0: irq 93 for MSI/MSI-X [521635.073217] alloc irq_desc for 94 on node -1 [521635.073424] alloc kstat_irqs on node -1 [521635.073643] ixgbe 0000:04:00.0: irq 94 for MSI/MSI-X [521635.073861] alloc irq_desc for 95 on node -1 [521635.074072] alloc kstat_irqs on node -1 [521635.074283] ixgbe 0000:04:00.0: irq 95 for MSI/MSI-X [521635.074549] ixgbe: 0000:04:00.0: ixgbe_init_interrupt_scheme: Multiqueue Enabled: Rx Queue count = 12, Tx Queue count = 12 [521635.074977] ixgbe 0000:04:00.0: (PCI Express:2.5Gb/s:Width x8) 00:1b:21:8b:f4:3b [521635.075456] ixgbe 0000:04:00.0: MAC: 1, PHY: 5, PBA No: e37002-010 [521635.076169] ixgbe 0000:04:00.0: Intel(R) 10 Gigabit Network Connection [521635.076388] ixgbe 0000:04:00.1: PCI INT B -> GSI 37 (level, low) -> IRQ 37 [521635.076616] ixgbe 0000:04:00.1: setting latency timer to 64 [521635.139824] alloc irq_desc for 96 on node -1 [521635.140036] alloc kstat_irqs on node -1 [521635.140249] ixgbe 0000:04:00.1: irq 96 for MSI/MSI-X [521635.140471] alloc irq_desc for 97 on node -1 [521635.140688] alloc kstat_irqs on node -1 [521635.140900] ixgbe 0000:04:00.1: irq 97 for MSI/MSI-X [521635.141110] alloc irq_desc for 98 on node -1 [521635.141319] alloc kstat_irqs on node -1 [521635.141589] ixgbe 0000:04:00.1: irq 98 for MSI/MSI-X [521635.141808] alloc irq_desc for 99 on node -1 [521635.142018] alloc kstat_irqs on node -1 [521635.142231] ixgbe 0000:04:00.1: irq 99 for MSI/MSI-X [521635.142498] alloc irq_desc for 100 on node -1 [521635.142715] alloc kstat_irqs on node -1 [521635.142926] ixgbe 0000:04:00.1: irq 100 for MSI/MSI-X [521635.143136] alloc irq_desc for 101 on node -1 [521635.143346] alloc kstat_irqs on node -1 [521635.143613] ixgbe 0000:04:00.1: irq 101 for MSI/MSI-X [521635.143831] alloc irq_desc for 102 on node -1 [521635.144041] alloc kstat_irqs on node -1 [521635.144251] ixgbe 0000:04:00.1: irq 102 for MSI/MSI-X [521635.144516] alloc irq_desc for 103 on node -1 [521635.144733] alloc kstat_irqs on node -1 [521635.144945] ixgbe 0000:04:00.1: irq 103 for MSI/MSI-X [521635.145155] alloc irq_desc for 104 on node -1 [521635.145364] alloc kstat_irqs on node -1 [521635.145631] ixgbe 0000:04:00.1: irq 104 for MSI/MSI-X [521635.145849] alloc irq_desc for 105 on node -1 [521635.146059] alloc kstat_irqs on node -1 [521635.146326] ixgbe 0000:04:00.1: irq 105 for MSI/MSI-X [521635.146544] alloc irq_desc for 106 on node -1 [521635.146754] alloc kstat_irqs on node -1 [521635.146965] ixgbe 0000:04:00.1: irq 106 for MSI/MSI-X [521635.147175] alloc irq_desc for 107 on node -1 [521635.147383] alloc kstat_irqs on node -1 [521635.147654] ixgbe 0000:04:00.1: irq 107 for MSI/MSI-X [521635.147873] alloc irq_desc for 108 on node -1 [521635.148083] alloc kstat_irqs on node -1 [521635.148352] ixgbe 0000:04:00.1: irq 108 for MSI/MSI-X [521635.148613] ixgbe: 0000:04:00.1: ixgbe_init_interrupt_scheme: Multiqueue Enabled: Rx Queue count = 12, Tx Queue count = 12 [521635.149036] ixgbe 0000:04:00.1: (PCI Express:2.5Gb/s:Width x8) 00:1b:21:8b:f4:3a [521635.149527] ixgbe 0000:04:00.1: MAC: 1, PHY: 5, PBA No: e37002-010 [521635.150255] ixgbe 0000:04:00.1: Intel(R) 10 Gigabit Network Connection I can now probe the interfaces: # ethtool -i eth2 driver: ixgbe version: 2.0.62-k2 firmware-version: 1.7-0 bus-info: 0000:04:00.0 # ethtool -i eth3 driver: ixgbe version: 2.0.62-k2 firmware-version: 1.7-0 bus-info: 0000:04:00.1 The same as expected from a dual-port adaptor. I have an eeprom dump here as well if you'd like. We use DA cables - specifically SPP-TWX-0508. And having a further look it appears that these are "active DA" cables which I understand isn't supported by the intel 82598EB at all. So back to the vendor and see if we can't resolve that. All your help has been greatly appreciated by the way and I look forward to resolving this asap and getting into actually testing the driver :) On 12/15/2010 02:23 AM, Skidmore, Donald C wrote: > Hi Mark, > > Well I'm glad we have the correct driver up on the web site. :) But actually > if you want to test with the latest driver I would suggest you move to > 3.1.17. It corrects bug that slipped into the last release having to do with > 82598 NIC's so I believe it would affect you. However I don't think it is > related to the issue you're currently seeing, but you might want to keep it > in mind. > > It makes sense actually that you didn't see a ethX device after we detect an > "unsupported module" we free the netdev. Which is why you have to reload the > driver to try again and you don't see a ethX device. > > You can however remove the cable and reload the driver. The netdev will be > created and you can see it with a "ifconfig -a". What ever that ethX device > is you should be able to do the "ethtool -i" on. > > Thanks, > -Don > >> -----Original Message----- >> From: Mark "Naoki" Rogers [mailto:[email protected]] >> Sent: Monday, December 13, 2010 7:16 PM >> To: [email protected] >> Subject: Re: [E1000-devel] ixgbe: Intel 82598EB 10GbE fails with Brocade >> TurboIron 24x >> >> Hello Don, >> >> Just quickly yes you are correct that I meant 3.1.15 of the driver so no >> problems there :) >> >> Here is the additional output you requested: >> # lspci -ns 0000:04:00.0 >> 04:00.0 0200: 8086:10f1 (rev 01) >> >> However I don't appear to have any defined eth devices for these >> interfaces, so nothing for ethtool to probe beyond the on-board intel >> Intel(R) PRO/1000 devices. I tried with the 3.1.15 driver and the result >> was the same: >> >> [ 830.287140] Intel(R) 10 Gigabit PCI Express Network Driver - version >> 3.1.15-NAPI >> [ 830.294522] Copyright (c) 1999-2010 Intel Corporation. >> [ 830.299846] ixgbe 0000:04:00.0: PCI INT A -> GSI 30 (level, low) -> >> IRQ 30 >> [ 830.306794] ixgbe 0000:04:00.0: setting latency timer to 64 >> [ 830.357221] ixgbe: 0000:04:00.0: ixgbe_probe: failed to load because >> an unsupported SFP+ module type was detected. >> [ 830.357224] Reload the driver after installing a supported module. >> [ 830.373805] ixgbe 0000:04:00.0: PCI INT A disabled >> [ 830.378706] ixgbe 0000:04:00.1: PCI INT B -> GSI 37 (level, low) -> >> IRQ 37 >> [ 830.385654] ixgbe 0000:04:00.1: setting latency timer to 64 >> [ 830.434058] ixgbe: 0000:04:00.1: ixgbe_probe: failed to load because >> an unsupported SFP+ module type was detected. >> [ 830.434061] Reload the driver after installing a supported module. >> [ 830.450654] ixgbe 0000:04:00.1: PCI INT B disabled >> >> Should I be assuming there would be downed ethX devices ? >> >> Much appreciate the assistance by the way. >> >> >> On 12/14/2010 11:13 AM, Skidmore, Donald C wrote: >>> Hi Mark, >>> >>> From the error messages below it looks as if the driver isn't >> recognizing your DA cable is a DA cable. We get this information from the >> cable via the Passive Cable bit as defined in the SFF-8472 spec. What is >> the version of your NIC's eeprom. You can find this with: >>> # ethtool -I ethX >>> >>> Also could you tell me the device id of the NIC just so I can be sure >> what 82598 you're working with? For the logs below you should be able to >> get it via: >>> # lspci -ns 0000:04:00.0 >>> >>> Also when you talk about using the 3.1.5 driver you really meant the >> 3.1.15 driver correct? I only asking to make sure we have it labeled >> correctly on the Intel web site you pulled it off of. :) >>> As far as checking the id we are seeing from the cable, I don't know of a >> utility that will give this value to you, short of hacking the driver to >> printk what we return from our i2c call. >>> Thanks, >>> -Don Skidmore<[email protected]> >>> >>> >>> >>>> -----Original Message----- >>>> From: Mark "Naoki" Rogers [mailto:[email protected]] >>>> Sent: Monday, December 13, 2010 2:20 AM >>>> To: [email protected] >>>> Subject: [E1000-devel] ixgbe: Intel 82598EB 10GbE fails with Brocade >>>> TurboIron 24x >>>> >>>> Hello all, >>>> >>>> I've a problem with the aforementioned dual-port adaptor and the >>>> TurboIron 24x. I've checked with the vendor that the cables are >>>> compliant with MSA, SFF-8431 and SFF-8432 however I get disheartening >>>> unsupported line from the driver: >>>> >>>> >>>> [369167.440737] ixgbe 0000:04:00.0: PCI INT A -> GSI 30 (level, low) -> >>>> IRQ 30 >>>> [369167.447758] ixgbe 0000:04:00.0: setting latency timer to 64 >>>> [369167.500906] ixgbe 0000:04:00.0: failed to initialize because an >>>> unsupported SFP+ module type was detected. >>>> [369167.500909] Reload the driver after installing a supported module. >>>> [369167.516983] ixgbe 0000:04:00.0: PCI INT A disabled >>>> [369167.521936] ixgbe 0000:04:00.1: PCI INT B -> GSI 37 (level, low) -> >>>> IRQ 37 >>>> [369167.528960] ixgbe 0000:04:00.1: setting latency timer to 64 >>>> [369167.580755] ixgbe 0000:04:00.1: failed to initialize because an >>>> unsupported SFP+ module type was detected. >>>> [369167.580757] Reload the driver after installing a supported module. >>>> [369167.596834] ixgbe 0000:04:00.1: PCI INT B disabled >>>> >>>> >>>> It appears very similar to this thread in which the odd resolution seems >>>> to be switching PCIe slots, which isn't really an option for my form >>>> factor: >>>> http://comments.gmane.org/gmane.linux.drivers.e1000.devel/7718 >>>> >>>> It's a Xeon X5650 system running the latest available BIOS (Oct 5th/ >> '10): >>>> 04:00.0 Ethernet controller: Intel Corporation 82598EB 10-Gigabit AF >>>> Dual Port Network Connection (rev 01) >>>> 04:00.1 Ethernet controller: Intel Corporation 82598EB 10-Gigabit AF >>>> Dual Port Network Connection (rev 01) >>>> >>>> I'm seeing the same error with drivers 2.0.62-k2, 2.0.8?, 2.0.84-k2, >>>> (2.6.35, 2.6.36, 2.6.37-rc5) and the 3.1.5 driver from Intel's site. I >>>> understand this is due to an ID in the EEPROM not being recognized(?), >>>> any way I can pull that myself and check it? >>>> >>>> Thank you. >>>> >>>> ------------------------------------------------------------------------ >> --- >>>> --- >>>> Oracle to DB2 Conversion Guide: Learn learn about native support for >>>> PL/SQL, >>>> new data types, scalar functions, improved concurrency, built-in >> packages, >>>> OCI, SQL*Plus, data movement tools, best practices and more. >>>> http://p.sf.net/sfu/oracle-sfdev2dev >>>> _______________________________________________ >>>> E1000-devel mailing list >>>> [email protected] >>>> https://lists.sourceforge.net/lists/listinfo/e1000-devel >>>> To learn more about Intel® Ethernet, visit >>>> http://communities.intel.com/community/wired >> >> --------------------------------------------------------------------------- >> --- >> Lotusphere 2011 >> Register now for Lotusphere 2011 and learn how >> to connect the dots, take your collaborative environment >> to the next level, and enter the era of Social Business. >> http://p.sf.net/sfu/lotusphere-d2d >> _______________________________________________ >> E1000-devel mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/e1000-devel >> To learn more about Intel® Ethernet, visit >> http://communities.intel.com/community/wired ------------------------------------------------------------------------------ Lotusphere 2011 Register now for Lotusphere 2011 and learn how to connect the dots, take your collaborative environment to the next level, and enter the era of Social Business. http://p.sf.net/sfu/lotusphere-d2d _______________________________________________ E1000-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
