[dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[Alex Zelichenko]

Hi, I am running provided *bond_app* for 2 ports (bound to igb_uio), and
the log I get is as follows:
alex@alex-lab:~/dpdk-stable-18.05.1$ sudo ./examples/bond/build/bond_app
-l 2-7 -n 2 -m --huge-dir /mnt/huge -- -p 3 -P
--config="(0,2,3,7),(1,4,5,6)"
EAL: Detected 8 lcore(s)
EAL: Detected 1 NUMA nodes
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: No free hugepages reported in hugepages-1048576kB
EAL: Probing VFIO support...
EAL: PCI device 0000:00:1f.6 on NUMA socket -1
EAL:   Invalid NUMA socket, default to 0
EAL:   probe driver: 8086:15b7 net_e1000_em
EAL: PCI device 0000:03:00.0 on NUMA socket -1
EAL:   Invalid NUMA socket, default to 0
EAL:   probe driver: 8086:1528 net_ixgbe
EAL: PCI device 0000:03:00.1 on NUMA socket -1
EAL:   Invalid NUMA socket, default to 0
EAL:   probe driver: 8086:1528 net_ixgbe
User device list:
Port 0 MAC: b4:96:91:21:a2:6c
Port 1 MAC: b4:96:91:21:a2:6e
ethdev port_id=2 requested Rx offloads 0x1000 doesn't match Rx offloads
capabilities 0x0 in rte_eth_dev_configure()

Port 2 MAC: b4:96:91:21:a2:6c
Starting lcore_main on core 3:0 Our IP:7.0.0.10
bond6>

How do I get rid of that error on bond port? How do I get that bond
interface? I can't see anything in the system that is related to
*net_bonding0*
-- 
Best,
Alex Z

Re: [dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[Cliff Burdick]

Can you try zeroing out the offloads flag at the top?

        .rxmode = {
                .mq_mode = ETH_MQ_RX_NONE,
                .max_rx_pkt_len = ETHER_MAX_LEN,
                .split_hdr_size = 0,
                .offloads = 0,
        },


On Tue, Dec 4, 2018 at 12:53 AM Alex Zelichenko <alex@scadafence.com> wrote:

> Hi, I am running provided *bond_app* for 2 ports (bound to igb_uio), and
> the log I get is as follows:
> alex@alex-lab:~/dpdk-stable-18.05.1$ sudo ./examples/bond/build/bond_app
> -l 2-7 -n 2 -m --huge-dir /mnt/huge -- -p 3 -P
> --config="(0,2,3,7),(1,4,5,6)"
> EAL: Detected 8 lcore(s)
> EAL: Detected 1 NUMA nodes
> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
> EAL: No free hugepages reported in hugepages-1048576kB
> EAL: Probing VFIO support...
> EAL: PCI device 0000:00:1f.6 on NUMA socket -1
> EAL:   Invalid NUMA socket, default to 0
> EAL:   probe driver: 8086:15b7 net_e1000_em
> EAL: PCI device 0000:03:00.0 on NUMA socket -1
> EAL:   Invalid NUMA socket, default to 0
> EAL:   probe driver: 8086:1528 net_ixgbe
> EAL: PCI device 0000:03:00.1 on NUMA socket -1
> EAL:   Invalid NUMA socket, default to 0
> EAL:   probe driver: 8086:1528 net_ixgbe
> User device list:
> Port 0 MAC: b4:96:91:21:a2:6c
> Port 1 MAC: b4:96:91:21:a2:6e
> ethdev port_id=2 requested Rx offloads 0x1000 doesn't match Rx offloads
> capabilities 0x0 in rte_eth_dev_configure()
>
> Port 2 MAC: b4:96:91:21:a2:6c
> Starting lcore_main on core 3:0 Our IP:7.0.0.10
> bond6>
>
> How do I get rid of that error on bond port? How do I get that bond
> interface? I can't see anything in the system that is related to
> *net_bonding0*
> --
> Best,
> Alex Z
>

Re: [dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[Alex Zelichenko]

Great, it worked (10x!). There's no error any more. But how do I interact
with that bond port? For now it behaves like network stub. Just collects
stats. Can I feed it to kni?

On Tue, Dec 4, 2018 at 6:59 PM Cliff Burdick <shaklee3@gmail.com> wrote:

> Can you try zeroing out the offloads flag at the top?
>
>         .rxmode = {
>                 .mq_mode = ETH_MQ_RX_NONE,
>                 .max_rx_pkt_len = ETHER_MAX_LEN,
>                 .split_hdr_size = 0,
>                 .offloads = 0,
>         },
>
>
> On Tue, Dec 4, 2018 at 12:53 AM Alex Zelichenko <alex@scadafence.com>
> wrote:
>
>> Hi, I am running provided *bond_app* for 2 ports (bound to igb_uio), and
>> the log I get is as follows:
>> alex@alex-lab:~/dpdk-stable-18.05.1$ sudo ./examples/bond/build/bond_app
>> -l 2-7 -n 2 -m --huge-dir /mnt/huge -- -p 3 -P
>> --config="(0,2,3,7),(1,4,5,6)"
>> EAL: Detected 8 lcore(s)
>> EAL: Detected 1 NUMA nodes
>> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
>> EAL: No free hugepages reported in hugepages-1048576kB
>> EAL: Probing VFIO support...
>> EAL: PCI device 0000:00:1f.6 on NUMA socket -1
>> EAL:   Invalid NUMA socket, default to 0
>> EAL:   probe driver: 8086:15b7 net_e1000_em
>> EAL: PCI device 0000:03:00.0 on NUMA socket -1
>> EAL:   Invalid NUMA socket, default to 0
>> EAL:   probe driver: 8086:1528 net_ixgbe
>> EAL: PCI device 0000:03:00.1 on NUMA socket -1
>> EAL:   Invalid NUMA socket, default to 0
>> EAL:   probe driver: 8086:1528 net_ixgbe
>> User device list:
>> Port 0 MAC: b4:96:91:21:a2:6c
>> Port 1 MAC: b4:96:91:21:a2:6e
>> ethdev port_id=2 requested Rx offloads 0x1000 doesn't match Rx offloads
>> capabilities 0x0 in rte_eth_dev_configure()
>>
>> Port 2 MAC: b4:96:91:21:a2:6c
>> Starting lcore_main on core 3:0 Our IP:7.0.0.10
>> bond6>
>>
>> How do I get rid of that error on bond port? How do I get that bond
>> interface? I can't see anything in the system that is related to
>> *net_bonding0*
>> --
>> Best,
>> Alex Z
>>
>

-- 
Best Regards,
*Alex Zelichenko
<https://www.linkedin.com/in/alexander-zelichenko-3a6005111/>, Senior
Researcher*
+972-53-7737006
SCADAfence.com

Re: [dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[Cliff Burdick]

Before going there, can you describe your compiler version? The code should
have worked as-is I think, so it would be good to understand what broke.

On Wed, Dec 5, 2018, 02:36 Alex Zelichenko <alex@scadafence.com wrote:

> Great, it worked (10x!). There's no error any more. But how do I interact
> with that bond port? For now it behaves like network stub. Just collects
> stats. Can I feed it to kni?
>
> On Tue, Dec 4, 2018 at 6:59 PM Cliff Burdick <shaklee3@gmail.com> wrote:
>
>> Can you try zeroing out the offloads flag at the top?
>>
>>         .rxmode = {
>>                 .mq_mode = ETH_MQ_RX_NONE,
>>                 .max_rx_pkt_len = ETHER_MAX_LEN,
>>                 .split_hdr_size = 0,
>>                 .offloads = 0,
>>         },
>>
>>
>> On Tue, Dec 4, 2018 at 12:53 AM Alex Zelichenko <alex@scadafence.com>
>> wrote:
>>
>>> Hi, I am running provided *bond_app* for 2 ports (bound to igb_uio), and
>>> the log I get is as follows:
>>> alex@alex-lab:~/dpdk-stable-18.05.1$ sudo ./examples/bond/build/bond_app
>>> -l 2-7 -n 2 -m --huge-dir /mnt/huge -- -p 3 -P
>>> --config="(0,2,3,7),(1,4,5,6)"
>>> EAL: Detected 8 lcore(s)
>>> EAL: Detected 1 NUMA nodes
>>> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
>>> EAL: No free hugepages reported in hugepages-1048576kB
>>> EAL: Probing VFIO support...
>>> EAL: PCI device 0000:00:1f.6 on NUMA socket -1
>>> EAL:   Invalid NUMA socket, default to 0
>>> EAL:   probe driver: 8086:15b7 net_e1000_em
>>> EAL: PCI device 0000:03:00.0 on NUMA socket -1
>>> EAL:   Invalid NUMA socket, default to 0
>>> EAL:   probe driver: 8086:1528 net_ixgbe
>>> EAL: PCI device 0000:03:00.1 on NUMA socket -1
>>> EAL:   Invalid NUMA socket, default to 0
>>> EAL:   probe driver: 8086:1528 net_ixgbe
>>> User device list:
>>> Port 0 MAC: b4:96:91:21:a2:6c
>>> Port 1 MAC: b4:96:91:21:a2:6e
>>> ethdev port_id=2 requested Rx offloads 0x1000 doesn't match Rx offloads
>>> capabilities 0x0 in rte_eth_dev_configure()
>>>
>>> Port 2 MAC: b4:96:91:21:a2:6c
>>> Starting lcore_main on core 3:0 Our IP:7.0.0.10
>>> bond6>
>>>
>>> How do I get rid of that error on bond port? How do I get that bond
>>> interface? I can't see anything in the system that is related to
>>> *net_bonding0*
>>> --
>>> Best,
>>> Alex Z
>>>
>>
>
> --
> Best Regards,
> *Alex Zelichenko
> <https://www.linkedin.com/in/alexander-zelichenko-3a6005111/>, Senior
> Researcher*
> +972-53-7737006
> SCADAfence.com
>

Re: [dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[Alex Zelichenko]

It's *gcc (Ubuntu 4.8.4-2ubuntu1~14.04.4) 4.8.4*


On Wed, Dec 5, 2018 at 3:53 PM Cliff Burdick <shaklee3@gmail.com> wrote:

> Before going there, can you describe your compiler version? The code
> should have worked as-is I think, so it would be good to understand what
> broke.
>
> On Wed, Dec 5, 2018, 02:36 Alex Zelichenko <alex@scadafence.com wrote:
>
>> Great, it worked (10x!). There's no error any more. But how do I interact
>> with that bond port? For now it behaves like network stub. Just collects
>> stats. Can I feed it to kni?
>>
>> On Tue, Dec 4, 2018 at 6:59 PM Cliff Burdick <shaklee3@gmail.com> wrote:
>>
>>> Can you try zeroing out the offloads flag at the top?
>>>
>>>         .rxmode = {
>>>                 .mq_mode = ETH_MQ_RX_NONE,
>>>                 .max_rx_pkt_len = ETHER_MAX_LEN,
>>>                 .split_hdr_size = 0,
>>>                 .offloads = 0,
>>>         },
>>>
>>>
>>> On Tue, Dec 4, 2018 at 12:53 AM Alex Zelichenko <alex@scadafence.com>
>>> wrote:
>>>
>>>> Hi, I am running provided *bond_app* for 2 ports (bound to igb_uio), and
>>>> the log I get is as follows:
>>>> alex@alex-lab:~/dpdk-stable-18.05.1$ sudo
>>>> ./examples/bond/build/bond_app
>>>> -l 2-7 -n 2 -m --huge-dir /mnt/huge -- -p 3 -P
>>>> --config="(0,2,3,7),(1,4,5,6)"
>>>> EAL: Detected 8 lcore(s)
>>>> EAL: Detected 1 NUMA nodes
>>>> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
>>>> EAL: No free hugepages reported in hugepages-1048576kB
>>>> EAL: Probing VFIO support...
>>>> EAL: PCI device 0000:00:1f.6 on NUMA socket -1
>>>> EAL:   Invalid NUMA socket, default to 0
>>>> EAL:   probe driver: 8086:15b7 net_e1000_em
>>>> EAL: PCI device 0000:03:00.0 on NUMA socket -1
>>>> EAL:   Invalid NUMA socket, default to 0
>>>> EAL:   probe driver: 8086:1528 net_ixgbe
>>>> EAL: PCI device 0000:03:00.1 on NUMA socket -1
>>>> EAL:   Invalid NUMA socket, default to 0
>>>> EAL:   probe driver: 8086:1528 net_ixgbe
>>>> User device list:
>>>> Port 0 MAC: b4:96:91:21:a2:6c
>>>> Port 1 MAC: b4:96:91:21:a2:6e
>>>> ethdev port_id=2 requested Rx offloads 0x1000 doesn't match Rx offloads
>>>> capabilities 0x0 in rte_eth_dev_configure()
>>>>
>>>> Port 2 MAC: b4:96:91:21:a2:6c
>>>> Starting lcore_main on core 3:0 Our IP:7.0.0.10
>>>> bond6>
>>>>
>>>> How do I get rid of that error on bond port? How do I get that bond
>>>> interface? I can't see anything in the system that is related to
>>>> *net_bonding0*
>>>> --
>>>> Best,
>>>> Alex Z
>>>>
>>>
>>

Re: [dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[Alex Zelichenko]

Hi Cliff,
Can you help me with the second part of the issue? How do I get the packets
from bonded interface into my tcpdump-like application? Is that interface
(net_bonding0) accessible from outside the application that created it?
Best,
Alex Z

On Wed, Dec 5, 2018 at 4:59 PM Alex Zelichenko <alex@scadafence.com> wrote:

> It's *gcc (Ubuntu 4.8.4-2ubuntu1~14.04.4) 4.8.4*
>
>
> On Wed, Dec 5, 2018 at 3:53 PM Cliff Burdick <shaklee3@gmail.com> wrote:
>
>> Before going there, can you describe your compiler version? The code
>> should have worked as-is I think, so it would be good to understand what
>> broke.
>>
>> On Wed, Dec 5, 2018, 02:36 Alex Zelichenko <alex@scadafence.com wrote:
>>
>>> Great, it worked (10x!). There's no error any more. But how do I
>>> interact with that bond port? For now it behaves like network stub. Just
>>> collects stats. Can I feed it to kni?
>>>
>>> On Tue, Dec 4, 2018 at 6:59 PM Cliff Burdick <shaklee3@gmail.com> wrote:
>>>
>>>> Can you try zeroing out the offloads flag at the top?
>>>>
>>>>         .rxmode = {
>>>>                 .mq_mode = ETH_MQ_RX_NONE,
>>>>                 .max_rx_pkt_len = ETHER_MAX_LEN,
>>>>                 .split_hdr_size = 0,
>>>>                 .offloads = 0,
>>>>         },
>>>>
>>>>
>>>> On Tue, Dec 4, 2018 at 12:53 AM Alex Zelichenko <alex@scadafence.com>
>>>> wrote:
>>>>
>>>>> Hi, I am running provided *bond_app* for 2 ports (bound to igb_uio),
>>>>> and
>>>>> the log I get is as follows:
>>>>> alex@alex-lab:~/dpdk-stable-18.05.1$ sudo
>>>>> ./examples/bond/build/bond_app
>>>>> -l 2-7 -n 2 -m --huge-dir /mnt/huge -- -p 3 -P
>>>>> --config="(0,2,3,7),(1,4,5,6)"
>>>>> EAL: Detected 8 lcore(s)
>>>>> EAL: Detected 1 NUMA nodes
>>>>> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
>>>>> EAL: No free hugepages reported in hugepages-1048576kB
>>>>> EAL: Probing VFIO support...
>>>>> EAL: PCI device 0000:00:1f.6 on NUMA socket -1
>>>>> EAL:   Invalid NUMA socket, default to 0
>>>>> EAL:   probe driver: 8086:15b7 net_e1000_em
>>>>> EAL: PCI device 0000:03:00.0 on NUMA socket -1
>>>>> EAL:   Invalid NUMA socket, default to 0
>>>>> EAL:   probe driver: 8086:1528 net_ixgbe
>>>>> EAL: PCI device 0000:03:00.1 on NUMA socket -1
>>>>> EAL:   Invalid NUMA socket, default to 0
>>>>> EAL:   probe driver: 8086:1528 net_ixgbe
>>>>> User device list:
>>>>> Port 0 MAC: b4:96:91:21:a2:6c
>>>>> Port 1 MAC: b4:96:91:21:a2:6e
>>>>> ethdev port_id=2 requested Rx offloads 0x1000 doesn't match Rx offloads
>>>>> capabilities 0x0 in rte_eth_dev_configure()
>>>>>
>>>>> Port 2 MAC: b4:96:91:21:a2:6c
>>>>> Starting lcore_main on core 3:0 Our IP:7.0.0.10
>>>>> bond6>
>>>>>
>>>>> How do I get rid of that error on bond port? How do I get that bond
>>>>> interface? I can't see anything in the system that is related to
>>>>> *net_bonding0*
>>>>> --
>>>>> Best,
>>>>> Alex Z
>>>>>
>>>>
>>>

Re: [dpdk-users] error running bonding example - Rx offloads capabilities 0x0 in rte_eth_dev_configure()

[Gowrishankar Muthukrishnan]

I think you can check dpdk pdump:
https://doc.dpdk.org/guides/howto/packet_capture_framework.html

Thanks,
Gowrishankar

On Mon, Dec 10, 2018 at 11:43 AM Alex Zelichenko <alex@scadafence.com>
wrote:

> Hi Cliff,
> Can you help me with the second part of the issue? How do I get the packets
> from bonded interface into my tcpdump-like application? Is that interface
> (net_bonding0) accessible from outside the application that created it?
> Best,
> Alex Z
>
> On Wed, Dec 5, 2018 at 4:59 PM Alex Zelichenko <alex@scadafence.com>
> wrote:
>
> > It's *gcc (Ubuntu 4.8.4-2ubuntu1~14.04.4) 4.8.4*
> >
> >
> > On Wed, Dec 5, 2018 at 3:53 PM Cliff Burdick <shaklee3@gmail.com> wrote:
> >
> >> Before going there, can you describe your compiler version? The code
> >> should have worked as-is I think, so it would be good to understand what
> >> broke.
> >>
> >> On Wed, Dec 5, 2018, 02:36 Alex Zelichenko <alex@scadafence.com wrote:
> >>
> >>> Great, it worked (10x!). There's no error any more. But how do I
> >>> interact with that bond port? For now it behaves like network stub.
> Just
> >>> collects stats. Can I feed it to kni?
> >>>
> >>> On Tue, Dec 4, 2018 at 6:59 PM Cliff Burdick <shaklee3@gmail.com>
> wrote:
> >>>
> >>>> Can you try zeroing out the offloads flag at the top?
> >>>>
> >>>>         .rxmode = {
> >>>>                 .mq_mode = ETH_MQ_RX_NONE,
> >>>>                 .max_rx_pkt_len = ETHER_MAX_LEN,
> >>>>                 .split_hdr_size = 0,
> >>>>                 .offloads = 0,
> >>>>         },
> >>>>
> >>>>
> >>>> On Tue, Dec 4, 2018 at 12:53 AM Alex Zelichenko <alex@scadafence.com>
> >>>> wrote:
> >>>>
> >>>>> Hi, I am running provided *bond_app* for 2 ports (bound to igb_uio),
> >>>>> and
> >>>>> the log I get is as follows:
> >>>>> alex@alex-lab:~/dpdk-stable-18.05.1$ sudo
> >>>>> ./examples/bond/build/bond_app
> >>>>> -l 2-7 -n 2 -m --huge-dir /mnt/huge -- -p 3 -P
> >>>>> --config="(0,2,3,7),(1,4,5,6)"
> >>>>> EAL: Detected 8 lcore(s)
> >>>>> EAL: Detected 1 NUMA nodes
> >>>>> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
> >>>>> EAL: No free hugepages reported in hugepages-1048576kB
> >>>>> EAL: Probing VFIO support...
> >>>>> EAL: PCI device 0000:00:1f.6 on NUMA socket -1
> >>>>> EAL:   Invalid NUMA socket, default to 0
> >>>>> EAL:   probe driver: 8086:15b7 net_e1000_em
> >>>>> EAL: PCI device 0000:03:00.0 on NUMA socket -1
> >>>>> EAL:   Invalid NUMA socket, default to 0
> >>>>> EAL:   probe driver: 8086:1528 net_ixgbe
> >>>>> EAL: PCI device 0000:03:00.1 on NUMA socket -1
> >>>>> EAL:   Invalid NUMA socket, default to 0
> >>>>> EAL:   probe driver: 8086:1528 net_ixgbe
> >>>>> User device list:
> >>>>> Port 0 MAC: b4:96:91:21:a2:6c
> >>>>> Port 1 MAC: b4:96:91:21:a2:6e
> >>>>> ethdev port_id=2 requested Rx offloads 0x1000 doesn't match Rx
> offloads
> >>>>> capabilities 0x0 in rte_eth_dev_configure()
> >>>>>
> >>>>> Port 2 MAC: b4:96:91:21:a2:6c
> >>>>> Starting lcore_main on core 3:0 Our IP:7.0.0.10
> >>>>> bond6>
> >>>>>
> >>>>> How do I get rid of that error on bond port? How do I get that bond
> >>>>> interface? I can't see anything in the system that is related to
> >>>>> *net_bonding0*
> >>>>> --
> >>>>> Best,
> >>>>> Alex Z
> >>>>>
> >>>>
> >>>
>