Re: [dpdk-users] error running l3fwd on kvm

[Mario Carrillo <mario.alfredo.c.arevalo@intel.com>]

thanks Arseniy Zaostrovnykh, I have tried your suggestion and I got a 
segmentation fault.

EAL: Detected lcore 0 as core 0 on socket 0
EAL: Detected lcore 1 as core 0 on socket 0
EAL: Detected lcore 2 as core 1 on socket 0
EAL: Detected lcore 3 as core 1 on socket 0
EAL: Detected lcore 4 as core 2 on socket 0
EAL: Detected lcore 5 as core 2 on socket 0
EAL: Detected lcore 6 as core 3 on socket 0
EAL: Detected lcore 7 as core 3 on socket 0
EAL: Support maximum 128 logical core(s) by configuration.
EAL: Detected 8 lcore(s)
EAL: VFIO modules not all loaded, skip VFIO support...
EAL: Setting up physically contiguous memory...
EAL: Ask a virtual area of 0x47000000 bytes
EAL: Virtual area found at 0x7f009c600000 (size = 0x47000000)
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f009c200000 (size = 0x200000)
EAL: Ask a virtual area of 0x38c00000 bytes
EAL: Virtual area found at 0x7f0063400000 (size = 0x38c00000)
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f0063000000 (size = 0x200000)
EAL: Requesting 1024 pages of size 2MB from socket 0
EAL: TSC frequency is ~2294687 KHz
EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using 
unreliable clock cycles !
EAL: open shared lib librte_pmd_virtio.so
EAL: Master lcore 0 is ready (tid=e44d31c0;cpuset=[0])
EAL: PCI device 0000:00:02.0 on NUMA socket -1
EAL:   probe driver: 1af4:1000 rte_virtio_pmd
EAL: PCI device 0000:00:03.0 on NUMA socket -1
EAL:   probe driver: 1af4:1000 rte_virtio_pmd
EAL: PCI device 0000:00:04.0 on NUMA socket -1
EAL:   probe driver: 1af4:1000 rte_virtio_pmd
EAL: PCI device 0000:00:05.0 on NUMA socket -1
EAL:   probe driver: 1af4:1000 rte_virtio_pmd
Initializing port 0 ... Segmentation fault (core dumped)


I'm going to try with other examples in order to know if it's the same 
behavior :).

On 11/12/2015 12:25 PM, Arseniy Zaostrovnykh wrote:
> It seems like there should be enough memory. However you use 2 lcores, 
> and this may imply, that you will be using 2 TX queues, look at the 
> line 2796 in examples/l3fwd/main.c :
>         n_tx_queue = nb_lcores;
>
> I had that problem, I fixed it by using just a single lcore, e.g.:
>
> -c 0x1 .... --config="(0,0,0),(1,0,0)"
>
>
> On 12.11.2015 18:12, Mario Carrillo wrote:
>> This is the configuration that I'm using for running the l3fwd example
>> ./build/l3fwd -c 0x3 -n 1 -d librte_pmd_virtio.so -- -p 0x3 
>> --config="(0,0,0),(1,0,1)"
>>
>> and about environment
>>  Hugepagesize:       2048 kB
>>
>> echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
>>
>> echo 1024 > 
>> /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
>>
>> what do you think?
>>
>> Thanks.
>> Mario.
>>
>>
>> On 11/12/2015 10:48 AM, Arseniy Zaostrovnykh wrote:
>>> Another possible cause just come to my mind:
>>> some drivers do not support more than 1 queues in each direction. 
>>> You have 2 transmission queues (nb_txq=2) here.
>>>
>>> On 12.11.2015 17:46, Arseniy Zaostrovnykh wrote:
>>>> You may lack memory. Make sure you are allocating enough hugepages 
>>>> (at least 1Gb)
>>>>
>>>> On 12.11.2015 17:13, Arevalo, Mario Alfredo C wrote:
>>>>> From: Arevalo, Mario Alfredo C
>>>>> Sent: Thursday, November 12, 2015 7:54 AM
>>>>> To: users@dpdk.org
>>>>> Subject: error running l3fwd on kvm
>>>>>
>>>>>      Hello guys, I'm trying to run l3fwd, it works very well when 
>>>>> I run it in bare-metal, however
>>>>> when I try to run in a Virtual Machine I get the next error:
>>>>>
>>>>> EAL: Detected lcore 0 as core 0 on socket 0
>>>>> EAL: Detected lcore 1 as core 0 on socket 0
>>>>> EAL: Detected lcore 2 as core 1 on socket 0
>>>>> EAL: Detected lcore 3 as core 1 on socket 0
>>>>> EAL: Detected lcore 4 as core 2 on socket 0
>>>>> EAL: Detected lcore 5 as core 2 on socket 0
>>>>> EAL: Detected lcore 6 as core 3 on socket 0
>>>>> EAL: Detected lcore 7 as core 3 on socket 0
>>>>> EAL: Support maximum 128 logical core(s) by configuration.
>>>>> EAL: Detected 8 lcore(s)
>>>>> EAL: VFIO modules not all loaded, skip VFIO support...
>>>>> EAL: Setting up physically contiguous memory...
>>>>> EAL: Ask a virtual area of 0x600000 bytes
>>>>> EAL: Virtual area found at 0x7f9905000000 (size = 0x600000)
>>>>> EAL: Ask a virtual area of 0x200000 bytes
>>>>> EAL: Virtual area found at 0x7f9904c00000 (size = 0x200000)
>>>>> EAL: Ask a virtual area of 0x35400000 bytes
>>>>> EAL: Virtual area found at 0x7f98cf600000 (size = 0x35400000)
>>>>> EAL: Requesting 430 pages of size 2MB from socket 0
>>>>> EAL: TSC frequency is ~2294687 KHz
>>>>> EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using 
>>>>> unreliable clock cycles !
>>>>> EAL: open shared lib librte_pmd_virtio.so
>>>>> EAL: Master lcore 0 is ready (tid=634d1c0;cpuset=[0])
>>>>> EAL: lcore 1 is ready (tid=cf3f6700;cpuset=[1])
>>>>> EAL: PCI device 0000:00:02.0 on NUMA socket -1
>>>>> EAL:   probe driver: 1af4:1000 rte_virtio_pmd
>>>>> EAL: PCI device 0000:00:03.0 on NUMA socket -1
>>>>> EAL:   probe driver: 1af4:1000 rte_virtio_pmd
>>>>> EAL: PCI device 0000:00:04.0 on NUMA socket -1
>>>>> EAL:   probe driver: 1af4:1000 rte_virtio_pmd
>>>>> EAL: PCI device 0000:00:05.0 on NUMA socket -1
>>>>> EAL:   probe driver: 1af4:1000 rte_virtio_pmd
>>>>> Initializing port 0 ... Creating queues: nb_rxq=1 nb_txq=2... EAL: 
>>>>> Error - exiting with code: 1
>>>>>    Cause: Cannot configure device: err=-22, port=0
>>>>>
>>>>> has  someone seen  this error before?
>>>>>
>>>>> Thank you.
>>>>> Mario.
>>>>>
>>>>>
>>>>
>>>
>>
>
> -- 
> Respectfully,
> Arseniy.