On 05/12/2024 14:15, Houshang wrote:
Hello
It 8s not rj45.it <http://rj45.it> is sfp
If your baseline performance cannot even support 25Msps in "benchmark_rate", it's unlikely that DPDK will push you
  into a territory where you can support 125Msps.

I assume that you've done the performance tuning tricks here:

https://kb.ettus.com/USRP_Host_Performance_Tuning_Tips_and_Tricks

[Forgive me if I've already suggested this].

This note here might possibly be useful, also:

https://kb.ettus.com/Getting_Started_with_DPDK_and_UHD#Underruns_Every_Second_with_DPDK_.2B_Ubuntu

But it looks like when you're using DPDK, UHD can't even see the network interfaces via DPDK, so this is a DPDK   configuration issue, and I'm not familiar enough with DPDK to offer much assistance.

You should confirm that you can "ping" each of the network interfaces on your X410 that are connected to your host.
  This will confirm basic network connectivity independent of DPDK.



On Thu, Dec 5, 2024, 7:47 PM Marcus D. Leech <patchvonbr...@gmail.com> wrote:

    On 05/12/2024 12:27, Houshang wrote:
    Hi Marcus
    Here it is:

    ad@bm-super11-intel:~$ sudo
    /usr/local/lib/uhd/examples/benchmark_rate --args
    "type=n3xx,product=n310,addr=10.10.0.100,master_clock_rate=125e6"
    --rx_rate 25e6 --tx_rate 25e6
    My recollection of your config is that 10.10.0.100 is the RJ-45
    you use for your management port?  What if you use
      one of the SFP+ port addresses for the same thing?

    25Msps is the absolute limit of streaming over a 1GiGe connection,
    and I think 10.10.0.100 was the RJ-45 port on
      your configuration as I recall.   It's likely that the combined
    network traffic of full-duplex streaming and management
      traffic cause your 1GIGe interface to totally saturate and start
    losing critical things.



    [INFO] [UHD] linux; GNU C++ version 11.4.0; Boost_107400;
    DPDK_23.11; UHD_4.7.0.HEAD-0-ga5ed1872
    [00:00:00.000131] Creating the usrp device with:
    type=n3xx,product=n310,addr=10.10.0.100,master_clock_rate=125e6...
    [INFO] [MPMD] Initializing 1 device(s) in parallel with args:
    
mgmt_addr=10.10.0.100,type=n3xx,product=n310,serial=32000F1,name=ni-n3xx-32000F1,fpga=XG,claimed=False,addr=10.10.0.100,master_clock_rate=125e6
    [WARNING] [MPM.RPCServer] A timeout event occured!
    [INFO] [MPM.PeriphManager] init() called with device args
    
`fpga=XG,master_clock_rate=125e6,mgmt_addr=10.10.0.100,name=ni-n3xx-32000F1,product=n310,clock_source=internal,time_source=internal'.
    Using Device: Single USRP:
      Device: N300-Series Device
      Mboard 0: n310
      RX Channel: 0
        RX DSP: 0
        RX Dboard: A
        RX Subdev: Magnesium
      RX Channel: 1
        RX DSP: 1
        RX Dboard: A
        RX Subdev: Magnesium
      RX Channel: 2
        RX DSP: 2
        RX Dboard: B
        RX Subdev: Magnesium
      RX Channel: 3
        RX DSP: 3
        RX Dboard: B
        RX Subdev: Magnesium
      TX Channel: 0
        TX DSP: 0
        TX Dboard: A
        TX Subdev: Magnesium
      TX Channel: 1
        TX DSP: 1
        TX Dboard: A
        TX Subdev: Magnesium
      TX Channel: 2
        TX DSP: 2
        TX Dboard: B
        TX Subdev: Magnesium
      TX Channel: 3
        TX DSP: 3
        TX Dboard: B
        TX Subdev: Magnesium

    [00:00:22.270989651] Setting device timestamp to 0...
    [WARNING] [0/DDC#0] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DDC#0] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DDC#0] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DDC#0] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DDC#1] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DDC#1] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DDC#1] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DDC#1] The requested decimation is odd; the user
    should expect passband CIC rolloff.
    Select an even decimation to ensure that a halfband filter is
    enabled.
    Decimations factorable by 4 will enable 2 halfbands, those
    factorable by 8 will enable 3 halfbands.
    decimation = dsp_rate/samp_rate -> 5
    [WARNING] [0/DUC#0] The requested interpolation is odd; the user
    should expect passband CIC rolloff.
    Select an even interpolation to ensure that a halfband filter is
    enabled.

    [WARNING] [0/DUC#0] The requested interpolation is odd; the user
    should expect passband CIC rolloff.
    Select an even interpolation to ensure that a halfband filter is
    enabled.

    [WARNING] [0/DUC#1] The requested interpolation is odd; the user
    should expect passband CIC rolloff.
    Select an even interpolation to ensure that a halfband filter is
    enabled.

    [WARNING] [0/DUC#1] The requested interpolation is odd; the user
    should expect passband CIC rolloff.
    Select an even interpolation to ensure that a halfband filter is
    enabled.

    Setting TX spp to 364
    [00:00:22.285623308] Testing receive rate 25.000000 Msps on 1
    channels
    [00:00:22.290554120] Testing transmit rate 25.000000 Msps on 1
    channels
    UOUUUUUUUUUUUUUUUUUterminate called after throwing an instance of
    'uhd::op_timeout'
      what():  RfnocError: OpTimeout: Control operation timed out
    waiting for ACK
    Aborted
    ad@bm-super11-intel:~$  ping 10.10.0.100
    PING 10.10.0.100 (10.10.0.100) 56(84) bytes of data.
    64 bytes from 10.10.0.100 <http://10.10.0.100>: icmp_seq=1 ttl=64
    time=0.311 ms
    64 bytes from 10.10.0.100 <http://10.10.0.100>: icmp_seq=2 ttl=64
    time=0.297 ms
    ^C
    --- 10.10.0.100 ping statistics ---
    2 packets transmitted, 2 received, 0% packet loss, time 1006ms
    rtt min/avg/max/mdev = 0.297/0.304/0.311/0.007 ms
    ad@bm-super11-intel:~$  uhd_find_devices
    [INFO] [UHD] linux; GNU C++ version 11.4.0; Boost_107400;
    DPDK_23.11; UHD_4.7.0.HEAD-0-ga5ed1872
    --------------------------------------------------
    -- UHD Device 0
    --------------------------------------------------
    Device Address:
        serial: 32000F1
        addr: 10.10.0.100
        claimed: False
        fpga: XG
        mgmt_addr: 10.10.0.100
        name: ni-n3xx-32000F1
        product: n310
        type: n3xx


    ad@bm-super11-intel:~$

    On Thu, 5 Dec 2024 at 18:23, Marcus D. Leech
    <patchvonbr...@gmail.com> wrote:

        On 05/12/2024 11:44, houshang.az...@accelleran.com wrote:

        Hello

        Have you managed to fix this?

        I have a similar issue as you can see below.

        Thanks

        Well, again, what happens when you do the simple
        benchmark_rate test *WITHOUT* DPDK involved?



        |ad@bm-super11-intel:~/accelleran$ sudo dpdk-devbind.py
        --status|

        |Network devices using DPDK-compatible driver|

        |============================================|

        |0000:43:00.1 'Ethernet Controller X710 for 10GbE SFP+ 1572'
        drv=vfio-pci unused=i40e|

        |0000:43:00.2 'Ethernet Controller X710 for 10GbE SFP+ 1572'
        drv=vfio-pci unused=i40e|

        |Network devices using kernel driver|

        |===================================|

        |0000:04:00.0 'Ethernet Controller 10G X550T 1563' if=eno1
        drv=ixgbe unused=vfio-pci *Active*|

        |0000:04:00.1 'Ethernet Controller 10G X550T 1563' if=eno2
        drv=ixgbe unused=vfio-pci |

        |0000:43:00.0 'Ethernet Controller X710 for 10GbE SFP+ 1572'
        if=ens1f0 drv=i40e unused=vfio-pci |

        |0000:43:00.3 'Ethernet Controller X710 for 10GbE SFP+ 1572'
        if=ens1f3 drv=i40e unused=vfio-pci *Active*|

        |0000:70:00.0 'Ethernet Controller X710 for 10GbE SFP+ 1572'
        if=ens2f0 drv=i40e unused=vfio-pci |

        |0000:70:00.1 'Ethernet Controller X710 for 10GbE SFP+ 1572'
        if=ens2f1 drv=i40e unused=vfio-pci |

        |Other Baseband devices|

        |======================|

        |0000:f7:00.0 'Device 57c0' unused=vfio-pci|

        |No 'Crypto' devices detected|

        |============================|

        |DMA devices using kernel driver|

        |===============================|

        |0000:f6:01.0 'Device 0b25' drv=idxd unused=vfio-pci |

        |No 'Eventdev' devices detected|

        |==============================|

        |No 'Mempool' devices detected|

        |=============================|

        |No 'Compress' devices detected|

        |==============================|

        |No 'Misc (rawdev)' devices detected|

        |===================================|

        |No 'Regex' devices detected|

        |===========================|

        |No 'ML' devices detected|

        |========================|

        |ad@bm-super11-intel:~/accelleran$ |

        |ad@bm-super11-intel:~/accelleran$ uhd_find_devices|

        |[INFO] [UHD] linux; GNU C++ version 11.4.0; Boost_107400;
        DPDK_23.11; UHD_4.7.0.HEAD-0-ga5ed1872|

        |--------------------------------------------------|

        |-- UHD Device 0|

        |--------------------------------------------------|

        |Device Address:|

        |serial: 32000F1|

        |addr: 10.10.0.100|

        |claimed: False|

        |fpga: XG|

        |mgmt_addr: 10.10.0.100|

        |name: ni-n3xx-32000F1|

        |product: n310|

        |type: n3xx|

        |ad@bm-super11-intel:~/accelleran$ sudo
        /usr/local/lib/uhd/examples/benchmark_rate --rx_rate 125e6
        --rx_subdev "A:0 B:0" --rx_channels 0,1 --tx_rate 125e6
        --tx_subdev "A:0 B:0" --tx_channels 0,1 --args
        
"addr=10.10.1.100,second_addr=10.10.2.100,mgmt_addr=10.10.0.100,master_clock_rate=125e6,use_dpdk=1"|

        |[INFO] [UHD] linux; GNU C++ version 11.4.0; Boost_107400;
        DPDK_23.11; UHD_4.7.0.HEAD-0-ga5ed1872|

        |EAL: Detected CPU lcores: 64|

        |EAL: Detected NUMA nodes: 1|

        |EAL: Detected shared linkage of DPDK|

        |EAL: Multi-process socket /var/run/dpdk/rte/mp_socket|

        |EAL: Selected IOVA mode 'VA'|

        |EAL: VFIO support initialized|

        |EAL: Using IOMMU type 1 (Type 1)|

        |EAL: Ignore mapping IO port bar(1)|

        |EAL: Ignore mapping IO port bar(4)|

        |EAL: Probe PCI driver: net_i40e (8086:1572) device:
        0000:43:00.1 (socket 0)|

        |EAL: Ignore mapping IO port bar(1)|

        |EAL: Ignore mapping IO port bar(4)|

        |EAL: Probe PCI driver: net_i40e (8086:1572) device:
        0000:43:00.2 (socket 0)|

        |TELEMETRY: No legacy callbacks, legacy socket not created|

        |i40e_alloc_rx_queue_mbufs(): Failed to allocate mbuf for RX|

        |i40e_dev_rx_queue_start(): Failed to allocate RX queue mbuf|

        |[ERROR] [DPDK] Port 1: Could not start device|

        |EAL: FATAL: already called initialization.|

        |EAL: already called initialization.|

        |[00:00:00.000306] Creating the usrp device with:
        
addr=10.10.1.100,second_addr=10.10.2.100,mgmt_addr=10.10.0.100,master_clock_rate=125e6,use_dpdk=1...|

        |EAL: FATAL: already called initialization.|

        |EAL: already called initialization.|

        |[ERROR] [X300] X300 Network discovery error RuntimeError:
        DPDK: Failure to start device|

        |[ERROR] [DPDK] Error with EAL initialization|

        |[ERROR] [UHD] Device discovery error: RuntimeError: Error
        with EAL initialization|

        |[ERROR] [DPDK] Error with EAL initialization|

        |[ERROR] [UHD] Device discovery error: RuntimeError: Error
        with EAL initialization|

        |EAL: FATAL: already called initialization.|

        |EAL: already called initialization.|

        |Error: LookupError: KeyError: No devices found for ----->|

        |Device Address:|

        |addr: 10.10.1.100|

        |second_addr: 10.10.2.100|

        |mgmt_addr: 10.10.0.100|

        |master_clock_rate: 125e6|

        |use_dpdk: 1|

        |[ERROR] [DPDK] Error with EAL initialization|

        |[ERROR] [X300] X300 Network discovery error RuntimeError:
        Error with EAL initialization|

        |Segmentation fault|

        |ad@bm-super11-intel:~/accelleran$|


        _______________________________________________
        USRP-users mailing list --usrp-users@lists.ettus.com
        To unsubscribe send an email tousrp-users-le...@lists.ettus.com

        _______________________________________________
        USRP-users mailing list -- usrp-users@lists.ettus.com
        To unsubscribe send an email to usrp-users-le...@lists.ettus.com



--
    *Houshang Azizi*

    *Test Engineer*

    logo <https://www.accelleran.com/>

    *(32) 492195241*

    *houshang.az...@accelleran.com <mailto:em...@accelleran.com>*

    *www.accelleran.com* <http://www.accelleran.com/>

    linkedin icon <https://www.linkedin.com/company/accelleran>
    twitter icon <https://twitter.com/accelleran> youtube icon
    <https://www.youtube.com/channel/UCrAEtqWp21cibZgSFVIEx2g?themeRefresh=1>



_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com

Reply via email to