Re: [dpdk-users] Failed to initialize one specific port of X710-DA4 in i40e pmd

[Hitoshi Masutani <masutani.hitoshi@lab.ntt.co.jp>]

Hi Aaron,

There is still problem.

Although I changed some DPDK versions, including 17.11.1, or 18.02
  with linux-image-4.4.0-87-generic,
the same problem has occurred.

As I have upgraded the latest firmware version of X710-DA4
in our test setup, probably such problem is not related to the firmware.

Thanks,

Hitoshi

On 2018/08/24 4:54, Aaron Hughes wrote:
> Hi Hitoshi,
> 
> Did you ever figure out a solution to your problem with the X710?  I'm
> seeing the same problem with an XL710 using DPDK version 17.05 on Redhat
> 7.5:
> 
> validate_tcmap_parameter(): Enabled TC map 0x80 not applicable to HW
> support 0x1
> i40e_vsi_setup(): Failed to update TC bandwidth
> i40e_pf_setup(): Setup of main vsi failed
> eth_i40e_dev_init(): Failed to setup pf switch: -63
> 
> Thanks,
> Aaron
> 
> I am facing a failure of dpdk(16.11.3) initialization with testpmd
>> using i40e pmd on a specific physical port of X710-DA4.
>>
>> Without such specific port, testpmd can run successfully.
>>
>> My server box has two X710-DA4 cards and
>> these two cards have the same failure.
>>
>> To clarify whether it is caused by hardware failure or
>> by the lack of appropriate nic configuration,
>> the following build options of dpdk are enabled.
>>
>> --------------
>> CONFIG_RTE_LIBRTE_ETHDEV_DEBUG=y
>> CONFIG_RTE_LIBRTE_I40E_DEBUG_INIT=y
>> CONFIG_RTE_LIBRTE_I40E_DEBUG_DRIVER=y
>> --------------
>>
>> My server box spec. is as follows :
>>
>> CPU        : Intel(R) Xeon(R) Gold 6136
>> MEMORY     : 128GB (16G x 8)
>> Motherboard: SYS-1029U-TR4
>> NIC1       : I350 x 4 (onboard nic)
>> NIC2       : XL710 dual
>> NIC3       : X710-DA4
>> NIC4       : X710-DA4
>> ------------------------------------------------------------
>> OS         : ubnutu 16.04.3 LTS(linux-image-4.4.0-87-generic)
>> DPDK       : 16.11.3
>>
>> testpmd's boot message is as follows :
>> --------------------------------------------
>> $ sudo ./testpmd -- -i
>>
>> EAL: Detected 24 lcore(s)
>> EAL: Probing VFIO support...
>> EAL: PCI device 0000:18:00.0 on NUMA socket 0
>> EAL:   probe driver: 8086:1521 net_e1000_igb
>> EAL: PCI device 0000:18:00.1 on NUMA socket 0
>> EAL:   probe driver: 8086:1521 net_e1000_igb
>> EAL: PCI device 0000:18:00.2 on NUMA socket 0
>> EAL:   probe driver: 8086:1521 net_e1000_igb
>> EAL: PCI device 0000:18:00.3 on NUMA socket 0
>> EAL:   probe driver: 8086:1521 net_e1000_igb
>> EAL: PCI device 0000:86:00.0 on NUMA socket 1
>> EAL:   probe driver: 8086:1583 net_i40e
>> EAL: PCI device 0000:86:00.1 on NUMA socket 1
>> EAL:   probe driver: 8086:1583 net_i40e
>> EAL: PCI device 0000:af:00.0 on NUMA socket 1
>> EAL:   probe driver: 8086:1572 net_i40e
>> PMD: i40e_set_symmetric_hash_enable_per_port(): Symmetric hash has
>> already been disabled
>> PMD: eth_i40e_dev_init(): FW 5.0 API 1.5 NVM 05.00.05 eetrack 80002919
>> PMD: i40e_update_default_filter_setting(): Cannot remove the default
>> macvlan filter
>> PMD: i40e_pf_setup(): i40e_vsi_setup finished
>> PMD: i40e_pf_setup(): Hardware capability of hash lookup table size: 512
>>
>> PMD: i40e_update_flow_control(): Link auto negotiation not completed
>> EAL: PCI device 0000:af:00.1 on NUMA socket 1
>> EAL:   probe driver: 8086:1572 net_i40e
>> EAL: PCI device 0000:af:00.2 on NUMA socket 1
>> EAL:   probe driver: 8086:1572 net_i40e
>> PMD: i40e_set_symmetric_hash_enable_per_port(): Symmetric hash has
>> already been disabled
>> PMD: eth_i40e_dev_init(): FW 5.0 API 1.5 NVM 05.00.05 eetrack 80002919
>> PMD: validate_tcmap_parameter(): Enabled TC map 0x80 not applicable to
>> HW support 0x1
>> PMD: i40e_vsi_setup(): Failed to update TC bandwidth
>> PMD: i40e_pf_setup(): Setup of main vsi failed
>> PMD: eth_i40e_dev_init(): Failed to setup pf switch: -63
>> rte_eth_dev_pci_probe: driver net_i40e: eth_dev_init(vendor_id=0x8086
>> device_id=0x1572) failed
>> EAL: Error - exiting with code: 1
>>       Cause: Requested device 0000:af:00.2 cannot be used
>> ---------------------------
>>
>> If you have solutions, could you tell me how this problem can be resolved ?
>>
>> Regards,
>> Hitoshi
>>
>>
>