Run srsRAN on OrangePi 5 Plus

[HaoxinSEU]

Hi community,

We are trying to run srsRAN gNB on OrangePi 5 Plus, which has an ARM CPU. And currently we meet some problems.

We successfully build and launch the gNB on it, and the UE can connect to it. But when we run ping on the UE, the first few works, then it disconnects and the gNB gives an error.

We notice that the gNB has a lot of underflows, and the error rate is very high. Besides, the successful pings have very large latency, like 1 sec.

We are wondering if there is any optimization we can make? Because we see that in the workshop, you show it's possible to run on Raspberry Pi. Thanks for your help!

Here is the console output of gNB:

root@orangepi-srs-image:/# gnb -c /opt/gnb.yaml 
Warning: Scheduling priority of thread "gnb_ue" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "gnb_ctrl" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "du_cell#0" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "up_phy_ul#0#0" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "phy_prach#0" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "up_phy_dl#0#0" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "up_phy_dl#0#1" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "pdsch#0#0" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "pusch#0#0" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "pdsch#0#2" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "pdsch#0#3" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "pdsch#0#1" not changed. Cause: Not enough privileges.
Lower PHY in quad executor mode.
Warning: Scheduling priority of thread "lower_phy_dl#0" not changed. Cause: Not enough privileges.
Warning: Scheduling priority of thread "lower_phy_rx#0" not changed. Cause: Not enough privileges.
           -----------------DL-----------------------|------------------UL--------------------
 pci rnti  cqi  ri  mcs  brate   ok  nok  (%)  dl_bs | pusch  mcs  brate   ok  nok  (%)    bsr
   1 4601   15   1   16    13k   15    8  34%      0 |  14.8   23    39k    9    8  47%      0
Late: 275; Underflow: 110; Overflow: 3;
   1 4601    3   1    3   238k  119  262  68%  1.62k |   5.6   18    97k   20   21  51%  3.91k
Late: 244; Underflow: 152; Overflow: 3;
   1 4601   15   1    5   218k  101  216  68%      0 | -10.6    7    96k   30  105  77%    276
Late: 60; Underflow: 58; Overflow: 2;
   1 4601   15   1   12   679k  132  297  69%      0 |  -2.5   10   108k   39   47  54%      0
Late: 24; Underflow: 68; Overflow: 2;
   1 4601   15   1   11   912k  128  236  64%      0 |  -5.4    4   178k   71   75  51%  39.8k
Late: 6; Underflow: 21; Overflow: 0;
   1 4601   15   1   19   123k   61   42  40%      0 |   7.1    2    31k   13    2  13%   108k
Late: 30; Underflow: 46; Overflow: 1;
   1 4609    3   1    2   5.9k    8    0   0%      0 |  -3.3   17    19k    6   35  85%   108k
 pci rnti  cqi  ri  mcs  brate   ok  nok  (%)  dl_bs | pusch  mcs  brate   ok  nok  (%)    bsr
   1 4609    9   1    5   1.6M  291  440  60%      0 | -13.1    0   385k  306  291  48%  39.8k
Late: 199; Underflow: 113; Overflow: 1;
   1 4609   10   1    4   909k  221  519  70%      0 | -12.1    0   429k  340  264  43%  14.7k
Late: 147; Underflow: 127; Overflow: 1;
   1 4609    6   1    2   817k  273  563  67%  6.22k | -14.9    0   255k  209  290  58%  2.81k
Error: exceeded maximum number of timed out transmissions.
Error: exceeded maximum number of timed out transmissions.

And our gNB configuration is:

gnb_id: 147

slicing:
 - sst: 1
   sd: 0
   
amf:
  addr: 10.43.84.19
  bind_addr: 10.42.6.17

ru_sdr:
  device_driver: uhd
  device_args: 
  # clock:
  # sync:
  srate: 23.04
  otw_format: sc16
  tx_gain: 66
  rx_gain: 15

cell_cfg:
  dl_arfcn: 632628
  band: 78
  channel_bandwidth_MHz: 20
  common_scs: 30
  plmn: "00101"
  tac: 1
  pci: 1
  tdd_ul_dl_cfg:
    dl_ul_tx_period: 10
    nof_dl_slots: 6
    nof_dl_symbols: 6
    nof_ul_slots: 3
    nof_ul_symbols: 6
  nof_antennas_ul: 1
  nof_antennas_dl: 1
  pdsch:
    mcs_table: qam256
    #min_ue_mcs: 22
  pusch:
    mcs_table: qam64
  pucch:
    sr_period_ms: 10

e2:
  enable_du_e2: false
  e2sm_kpm_enabled: true
  addr: 127.0.0.1
  bind_addr: 10.42.1.41
  port: 36421

log:
  filename: /tmp/gnb.log
  all_level: info

pcap:
  mac_enable: false
  mac_filename: /tmp/gnb_mac.pcap
  ngap_enable: false
  ngap_filename: /tmp/gnb_ngap.pcap

[pgawlowicz]

Could you reduce the BW and see if it helps?

[HaoxinSEU]

Hi @pgawlowicz ,

Thanks for your reply!

We also thought about the BW. But this discussion said we had to use 30.72 as the sampling rate for 10 MHz BW, which is even higher than 20 MHz BW. So it seems to add more working load to the CPU...

[HaoxinSEU]

And we also tried to use 5 MHz BW FDD at n3 band, the results are quite similar. We still got a lot of underflows, and large error rate.

[andrepuschmann]

The Rockchip RK3588 is a big-little CPU so you want to make sure that the most demanding tasks run on the four A76 cores and not on the smaller A55 cores.
You need to explore the expert_execution section a bit more and experiment with the settings.
I am afraid this section has changed quite a bit from the previous release and might as well change again. You might also want to try the test branch to have more recent code.
Here is an example:

expert_execution:
 affinities:
    l1_dl_cpus: 0-1
    l1_ul_cpus: 2
    l2_cell_cpus: 3
    ru_cpus: 6-11
    low_priority_cpus: 0-3

 threads:
  upper_phy:
   nof_ul_threads: 1
   nof_pdsch_threads: 2
   nof_dl_threads: 2
  ofh:
   enable_dl_parallelization: true

[HaoxinSEU]

Thanks! This sounds super reasonable!

We will update to the latest test branch, and play with these configurations.

[ismagom]

Hi @HaoxinSEU , you should also check this warning:

Warning: Scheduling priority of thread "gnb_ue" not changed. Cause: Not enough privileges.

It seems the real-time scheduling policy can't be set due to not enough privileges

[HaoxinSEU]

Hi @ismagom ,

Thanks for your reply!

Indeed, this is a problem. And we found that it's because we are using the custom Ubuntu provided by Orange Pi. After we use a default ubuntu 22.04, we don't have these warnings.

And there are very few underflow now when using 5 MHz BW FDD. But still, the error rate is very high, and ping gets 1 sec latency.

[HaoxinSEU]

Hi all,

Thanks for your help @andrepuschmann @ismagom @pgawlowicz ! We are now able to run the gNB with 5 MHz BW FDD and 20 MHz BW TDD.

Our configurations now are:

1. Don't use the Ubuntu provided by Orange Pi, but instead we use this repository
2. Set CPU cores and threads allocated to different procedures

expert_execution:
 affinities:
    l1_dl_cpus: 5-6
    l1_ul_cpus: 0-2
    l2_cell_cpus: 6-7
    ru_cpus: 3-4
    low_priority_cpus: 4-7

 threads:
  upper_phy:
   nof_ul_threads: 4
   nof_pdsch_threads: 2
   nof_dl_threads: 2
  ofh:
   enable_dl_parallelization: true

Now we can achieve ~10 Mbps DL throughput in both cases. And the latency is reasonable, ping from the phone to 8.8.8.8 is ~30 ms. Our observation is that the big-little cores are very important, we have to pu L1_UL and RU in these big cores, otherwise the error rate is very high.

[faratell]

this

This is what I got

          |--------------------DL---------------------|-------------------------UL------------------------------
 pci rnti | cqi  ri  mcs  brate   ok  nok  (%)  dl_bs | pusch  rsrp  mcs  brate   ok  nok  (%)    bsr     ta  phr
   1 4601 |  13 1.3   21    94k   47    0   0%      0 |  23.5 -10.3   21   2.1M   90    2   2%      0   169n   -2
   1 4601 |  15 1.0   23   278k   72    0   0%      0 |  28.9  -5.5   26   326k   53   13  19%  2.81k   168n    2
   1 4601 |  15 1.1   23   428k  129    0   0%      0 |  29.2  -3.8   26   579k   73    9  10%      0   177n   -4
   1 4601 |  10 2.0   14   211k   69    0   0%      0 |  28.9  -1.2   26   150k   45   21  31%      0   341n   -5
   1 4601 |  10 1.9   14   349k   86    1   1%      0 |  27.6  -4.3   26   612k   69   22  24%    276   165n   -9
   1 4601 |  11 2.0   16   912k  148    1   0%      7 |  29.9  -2.7   26   485k   96    6   5%  1.04k   155n   -8
   1 4601 |  11 2.0   17    39M  876    4   0%  22.9k |  27.9  -4.9   25   844k   94   16  14%      0   167n   -5
   1 4601 |  11 2.0   17    22M  566    6   1%      0 |  26.3  -5.5   25   1.2M  103   48  31%  20.5k   207n  -14
   1 4601 |  11 2.0   17    21M  532    1   0%     11 |  27.0  -4.3   25   886k   87   38  30%  14.7k   166n  -11
   1 4601 |  11 2.0   17    31M  701    3   0%      0 |  27.3  -4.8   25   910k  100   21  17%  2.81k   150n   -1
   1 4601 |  11 2.0   17    18M  417    3   0%      0 |  28.2  -3.4   25   740k   91   21  18%     10   155n    1

          |--------------------DL---------------------|-------------------------UL------------------------------
 pci rnti | cqi  ri  mcs  brate   ok  nok  (%)  dl_bs | pusch  rsrp  mcs  brate   ok  nok  (%)    bsr     ta  phr
   1 4601 |  11 2.0   18    20M  474    8   1%      0 |  27.4  -4.5   25   989k  100   24  19%  2.01k   162n  -10
   1 4601 |  11 2.0   16    20M  547   14   2%  61.8k |  26.7  -6.0   25   1.9M  141   22  13%  2.01k   237n  -13
   1 4601 |  10 2.0   14    21M  628    2   0%      0 |  26.6  -5.4   25   1.1M   94   32  25%  5.45k   175n   -5
   1 4601 |  10 2.0   15    15M  450    2   0%      0 |  29.7  -2.3   26   454k   93    8   7%     53   156n   -7
   1 4601 |   9 2.0   14    14M  466    0   0%      0 |  28.4  -3.3   25   582k   75   10  11%  3.91k   160n   -5
   1 4601 |  10 2.0   16    19M  548    5   0%  1.41k |  28.2  -3.6   25   610k   73    8   9%     53   168n    1
   1 4601 |  11 2.0   17    13M  405    7   1%  64.6k |  27.7  -3.8   25   449k   62   15  19%    198   204n   -4
   1 4601 |  10 2.0   15    23M  595    1   0%      0 |  28.7  -4.7   26   459k   71   10  12%      0   155n   -5
   1 4601 |   9 2.0   15    19M  549    1   0%  67.1k |  28.7  -5.1   26   543k   76    4   5%  1.45k   171n   -9
   1 4601 |  10 1.9   17   3.3M   99    1   1%      0 |  30.9  -1.5   26   157k   49    0   0%      0   157n   -5
   1 4601 |  13 1.5   25    80k   30    1   3%      0 |  29.8  -1.9   26   171k   41    4   8%     74   206n   -1

          |--------------------DL---------------------|-------------------------UL------------------------------
 pci rnti | cqi  ri  mcs  brate   ok  nok  (%)  dl_bs | pusch  rsrp  mcs  brate   ok  nok  (%)    bsr     ta  phr
   1 4601 |  13 1.4   23    79k   51    4   7%      0 |  26.5  -4.0   24   658k   46    8  14%  2.01k   238n  -12
   1 4601 |  10 2.0   16   873k  284    3   1%    754 |  23.6  -7.4   23   9.7M  404  109  21%   300k   155n   -9
   1 4601 |  10 2.0   16   398k  247    1   0%      7 |  23.7  -7.5   23   9.0M  371  111  23%   300k   159n  -13
   1 4601 |  10 2.0   16   506k  246    0   0%      0 |  23.6  -8.0   23   9.9M  385   71  15%   300k   164n  -13
   1 4601 |  10 2.0   16   608k  259    2   0%     58 |  23.4  -7.6   23   9.2M  371   96  20%   300k   146n  -13
   1 4601 |  10 2.0   16   475k  257    5   1%      0 |  23.6  -8.1   23   9.4M  378   95  20%   300k   153n  -13
   1 4601 |  10 2.0   16   567k  241    1   0%      0 |  23.6  -8.4   23   8.9M  349  107  23%   300k   158n  -13
Late: 0; Underflow: 1; Overflow: 0;

I did not use low latency kernel. The speed is a lot lower than when I run all in a I5 computer. On the I5 computer, I got about 100 Mbs. I have to idea where the bottleneck is. Maybe, WiFi? `

Thanks. Are you tested srsRAN_4G in Rpi5 for BW=20MHZ?

[ALISROURR]

HI @fllay ,
Good JOB!
Could you please share the conf. files used for the Pi 4,Pi 5, X4?
Did you do any configuration on the expert_execution section?

[fllay]

Sure, Let start with network interface and IP address of each machine

Raspberry pi 5 (Core network) : wlan0 with IP 192.168.68.42

4: wlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 2c:cf:67:2b:03:f4 brd ff:ff:ff:ff:ff:ff
    inet 192.168.68.42/24 brd 192.168.68.255 scope global wlan0

Raspberry pi 5 (CU) : wlan0 with IP 192.168.68.210 and 192.168.68.213. Please note that this interface has two IPS assigned to it

3: wlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 2c:cf:67:2b:00:91 brd ff:ff:ff:ff:ff:ff
    inet 192.168.68.210/24 brd 192.168.68.255 scope global wlan0
       valid_lft forever preferred_lft forever
    inet 192.168.68.213/24 brd 192.168.68.255 scope global secondary wlan0

Radxa X4 (DU) : wlan0 with IP 192.168.68.19

3: wlp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether a8:41:f4:bc:45:74 brd ff:ff:ff:ff:ff:ff
    inet 192.168.68.19/24 brd 192.168.68.255 scope global wlp1s0
       valid_lft forever preferred_lft forever

Open5GS config file

db_uri: mongodb://localhost/open5gs

logger:

test:
  serving:
    - plmn_id:
        mcc: 001
        mnc: 01

global:
  parameter:
#    no_nrf: true
#    no_scp: true
    no_sepp: true
#    no_amf: true
#    no_smf: true
#    no_upf: true
#    no_ausf: true
#    no_udm: true
#    no_pcf: true
#    no_nssf: true
#    no_bsf: true
#    no_udr: true
#    no_mme: true
#    no_sgwc: true
#    no_sgwu: true
#    no_pcrf: true
#    no_hss: true

mme:
  freeDiameter:
    identity: mme.localdomain
    realm: localdomain
    listen_on: 127.0.0.2
    no_fwd: true
    load_extension:
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dbg_msg_dumps.fdx
        conf: 0x8888
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_rfc5777.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_mip6i.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nasreq.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nas_mipv6.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca_3gpp/dict_dcca_3gpp.fdx
    connect:
      - identity: hss.localdomain
        address: 127.0.0.8

  s1ap:
    server:
      - address: 127.0.0.2
  gtpc:
    server:
      - address: 127.0.0.2
    client:
      sgwc:
        - address: 127.0.0.3
      smf:
        - address: 127.0.0.4
  metrics:
    server:
      - address: 127.0.0.2
        port: 9090
  gummei:
    - plmn_id:
        mcc: 001
        mnc: 01
      mme_gid: 2
      mme_code: 1
  tai:
    - plmn_id:
        mcc: 001
        mnc: 01
      tac: 7
  security:
    integrity_order : [ EIA2, EIA1, EIA0 ]
    ciphering_order : [ EEA0, EEA1, EEA2 ]
  network_name:
    full: Open5GS
  time:
    t3412:
      value: 540

sgwc:
  gtpc:
    server:
      - address: 127.0.0.3
  pfcp:
    server:
      - address: 127.0.0.3
    client:
      sgwu:
        - address: 127.0.0.6

smf:
  sbi:
    server:
      - address: 127.0.0.4
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777
  pfcp:
    server:
      - address: 127.0.0.4
    client:
      upf:
        - address: 127.0.0.7
  gtpc:
    server:
      - address: 127.0.0.4
  gtpu:
    server:
      - address: 127.0.0.4
  metrics:
    server:
      - address: 127.0.0.4
        port: 9090
  session:
    - subnet: 10.45.0.0/16
      gateway: 10.45.0.1
    - subnet: 2001:db8:cafe::/48
      gateway: 2001:db8:cafe::1
  dns:
    - 8.8.8.8
    - 8.8.4.4
    - 2001:4860:4860::8888
    - 2001:4860:4860::8844
  mtu: 1400
  freeDiameter:
    identity: smf.localdomain
    realm: localdomain
    listen_on: 127.0.0.4
    no_fwd: true
    load_extension:
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dbg_msg_dumps.fdx
        conf: 0x8888
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_rfc5777.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_mip6i.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nasreq.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nas_mipv6.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca_3gpp/dict_dcca_3gpp.fdx
    connect:
      - identity: pcrf.localdomain
        address: 127.0.0.9

amf:
  sbi:
    server:
      - address: 127.0.0.5
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777
  ngap:
    server:
      - address: 192.168.68.42 #127.0.0.5 #192.168.68.42
  metrics:
    server:
      - address: 127.0.0.5
        port: 9090
  guami:
    - plmn_id:
        mcc: 001
        mnc: 01
      amf_id:
        region: 2
        set: 1
  tai:
    - plmn_id:
        mcc: 001
        mnc: 01
      tac: 7
  plmn_support:
    - plmn_id:
        mcc: 001
        mnc: 01
      s_nssai:
        - sst: 1
  security:
    integrity_order : [ NIA2, NIA1, NIA0 ]
    ciphering_order : [ NEA0, NEA1, NEA2 ]
  network_name:
    full: Open5GS
  amf_name: open5gs-amf0
  time:
    t3512:
      value: 540     # 9 mintues * 60 = 540 seconds

sgwu:
  pfcp:
    server:
      - address: 127.0.0.6
  gtpu:
    server:
      - address: 127.0.0.6

upf:
  pfcp:
    server:
      - address: 127.0.0.7
  gtpu:
    server:
      - address: 192.168.68.42
  session:
    - subnet: 10.45.0.0/16
      gateway: 10.45.0.1
    - subnet: 2001:db8:cafe::/48
      gateway: 2001:db8:cafe::1
  metrics:
    server:
      - address: 127.0.0.7
        port: 9090

hss:
  freeDiameter:
    identity: hss.localdomain
    realm: localdomain
    listen_on: 127.0.0.8
    no_fwd: true
    load_extension:
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dbg_msg_dumps.fdx
        conf: 0x8888
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_rfc5777.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_mip6i.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nasreq.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nas_mipv6.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca_3gpp/dict_dcca_3gpp.fdx
    connect:
      - identity: mme.localdomain
        address: 127.0.0.2
pcrf:
  freeDiameter:
    identity: pcrf.localdomain
    realm: localdomain
    listen_on: 127.0.0.9
    no_fwd: true
    load_extension:
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dbg_msg_dumps.fdx
        conf: 0x8888
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_rfc5777.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_mip6i.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nasreq.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_nas_mipv6.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca.fdx
      - module: /home/pi/src/open5gs/build/subprojects/freeDiameter/extensions/dict_dcca_3gpp/dict_dcca_3gpp.fdx
    connect:
      - identity: smf.localdomain
        address: 127.0.0.4

nrf:
  sbi:
    server:
      - address: 127.0.0.10
        port: 7777

scp:
  sbi:
    server:
      - address: 127.0.0.200
        port: 7777
    client:
      nrf:
        - uri: http://127.0.0.10:7777

ausf:
  sbi:
    server:
      - address: 127.0.0.11
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777

udm:
  hnet:
    - id: 1
      scheme: 1
      key: /home/pi/src/open5gs/build/configs/open5gs/hnet/curve25519-1.key
    - id: 2
      scheme: 2
      key: /home/pi/src/open5gs/build/configs/open5gs/hnet/secp256r1-2.key
  sbi:
    server:
      - address: 127.0.0.12
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777

pcf:
  sbi:
    server:
      - address: 127.0.0.13
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777
  metrics:
    server:
      - address: 127.0.0.13
        port: 9090

nssf:
  sbi:
    server:
      - address: 127.0.0.14
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777
      nsi:
        - uri: http://127.0.0.10:7777
          s_nssai:
            sst: 1
bsf:
  sbi:
    server:
      - address: 127.0.0.15
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777

udr:
  sbi:
    server:
      - address: 127.0.0.20
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777

I think I change only in AMF section in NGAP server address

amf:
  sbi:
    server:
      - address: 127.0.0.5
        port: 7777
    client:
      scp:
        - uri: http://127.0.0.200:7777
  ngap:
    server:
      - address: 192.168.68.42 #127.0.0.5 #192.168.68.42

CU Config file

gnb_id: 531
gnb_id_bit_length: 32
ran_node_name: cu_cp_01

#slicing:                           
#  -                                   
#    sst: 1                               
#    sd: 0                              

#amf:
#  addr:  192.168.68.42
#  bind_addr: 192.168.68.210

cu_cp:
  inactivity_timer: 7200
  amf:
    addr: 192.168.68.42
    port: 38412
    bind_addr: 192.168.68.210
    supported_tracking_areas:
      - tac: 7
        plmn_list:
          - plmn: "00101"
            tai_slice_support_list:
              - sst: 1
  mobility:
    trigger_handover_from_measurements: true
    cells:
      - nr_cell_id: 0x2130
        periodic_report_cfg_id: 1
        ncells:
          - nr_cell_id: 0x2131
            report_configs: [ 2 ]
      - nr_cell_id: 0x2131
        periodic_report_cfg_id: 1
        ncells:
          - nr_cell_id: 0x2130
            report_configs: [ 2 ]
    report_configs:
      - report_cfg_id: 1
        report_type: periodical
        report_interval_ms: 1024
      - report_cfg_id: 2
        report_type: event_triggered
        event_triggered_report_type: a3
        meas_trigger_quantity: rsrp
        meas_trigger_quantity_offset_db: 2
        hysteresis_db: 0
        time_to_trigger_ms: 100
        report_interval_ms: 1024

  f1ap:
    bind_addr: 192.168.68.210


cu_up:
  nru:
    bind_addr: 192.168.68.213
    udp_max_rx_msgs: 256

log:
  filename: /tmp/cu.log
  all_level: debug


pcap:
  ngap_filename: /tmp/cu_ngap.pcap     # Optional TEXT (/tmp/gnb_ngap.pcap). Path for NGAP PCAPs. 
  ngap_enable: true                    # Optional BOOLEAN (false). Enable/disable NGAP packet capture. Supported: [false, true].
  n3_filename: /tmp/cu_n3.pcap         # Optional TEXT (/tmp/gnb_n3.pcap). Path for N3 PCAPs. 
  n3_enable: true                      # Optional BOOLEAN (false). Enable/disable N3 packet capture. Supported: [false, true].
  e1ap_filename: /tmp/cu_e1ap.pcap     # Optional TEXT (/tmp/gnb_e1ap.pcap). Path for E1AP PCAPs. 
  e1ap_enable: true                    # Optional BOOLEAN (false). Enable/disable E1AP packet capture. Supported: [false, true].
  f1ap_filename: /tmp/cu_f1ap.pcap     # Optional TEXT (/tmp/gnb_f1ap.pcap). Path for F1AP PCAPs. 
  f1ap_enable: true                    # Optional BOOLEAN (false). Enable/disable F1AP packet capture. Supported: [false, true].
  f1u_filename: /tmp/cu_f1u.pcap       # Optional TEXT (/tmp/gnb_f1ap.pcap). Path for F1AP PCAPs. 
  f1u_enable: true                     # Optional BOOLEAN (false). Enable/disable F1AP packet capture. Supported: [false, true].

and DU config file

gnb_id: 531
gnb_id_bit_length: 32
gnb_du_id: 0

f1ap:
  cu_cp_addr: 192.168.68.210
  bind_addr: 192.168.68.19
nru:
  bind_addr: 192.168.68.19

ru_sdr:
  device_driver: uhd
  device_args: type=b200,num_recv_frames=64,num_send_frames=64
  srate: 23.04
  otw_format: sc12
  tx_gain: 82
  rx_gain: 50


cell_cfg:
  dl_arfcn: 632628                                                # ARFCN of the downlink carrier>
  band: 78                                                        # The NR band.
  channel_bandwidth_MHz: 20                                       # Bandwith in MHz. Number of PR>
  common_scs: 30                                                  # Subcarrier spacing in kHz use>
  plmn: "00101"                                                   # PLMN broadcasted by the gNB.
  tac: 7                                                          # Tracking area code (needs to >
  pci: 1                                                          # Physical cell ID.
  nof_antennas_dl: 2
  nof_antennas_ul: 2
  sector_id: 0
  pdsch:
    mcs_table: qam256

log:
  filename: /tmp/du.log
  all_level: warning

pcap:
  mac_enable: false
  mac_filename: /tmp/du_mac.pcap
  f1ap_enable: false
  f1ap_filename: /tmp/du_f1ap.pcap
  f1u_enable: false
  f1u_filename: /tmp/du_f1u.pcap

[ALISROURR]

Thanks!

[Legend-Dd]

Could you help me? I managed to set up Open5GS and SRSRAN on a Raspberry Pi 5 for 5G SA, and when I register the B210 base station, this error pops up, do you know if there is any solution?

/home/core5g/srsRAN_Project/lib/pdcp/pdcp_entity_rx.cpp:540: srsran::security::security_result srsran::pdcp_entity_rx::apply_deciphering_and_integrity_check(srsran::byte_buffer, uint32_t): Assertion failed - Worker index exceeds number of crypto workers. worker_idx=4 max_nof_crypto_workers=4 Aborted

[faratell]

i used rpi5 with rpi os and run srsran_4g on it. for nprb=100 it shows overflow and underflow then disconnected. How can improve its performance? using RT kernel can be effective?

[ALISROURR]

@fllay I was using x310, it was working with Pi5, however when i switched to usrp b210.
The gnb stops here:

Do you have any recommendations?
gnb conf:
cu_cp:
amf:
addr: 127.0.0.1 # The address or hostname of the AMF.
bind_addr: 127.0.0.1 # A local IP that the gNB binds to for traffic from the AMF.
supported_tracking_areas: # Configure the TA associated with the CU-CP
- tac: 1
plmn_list:
- plmn: "00101"
tai_slice_support_list:
- sst: 1

ru_sdr:
device_driver: uhd # The RF driver name.
device_args: type=b200,num_recv_frames=32,num_send_frames=32 # Optionally pass arguments to the selected RF driver.

srate: 23.04 # RF sample rate might need to be adjusted according to selected bandwidth.
otw_format: sc12
tx_gain: 30 # Transmit gain of the RF might need to adjusted to the given situation.
rx_gain: 37 # Receive gain of the RF might need to adjusted to the given situation.

cell_cfg:
dl_arfcn: 663334 # ARFCN of the downlink carrier (center frequency).
band: 77 # The NR band.
channel_bandwidth_MHz: 20 # Bandwith in MHz. Number of PRBs will be automatically derived.
common_scs: 30 # Subcarrier spacing in kHz used for data.
plmn: "00101" # PLMN broadcasted by the gNB.
tac: 1 # Tracking area code (needs to match the core configuration).
pci: 1 # Physical cell ID.

pcap:
mac_enable: false # Set to true to enable MAC-layer PCAPs.
mac_filename: /tmp/gnb_mac.pcap # Path where the MAC PCAP is stored.
ngap_enable: false # Set to true to enable NGAP PCAPs.
ngap_filename: /tmp/gnb_ngap.pcap # Path where the NGAP PCAP is stored.
metrics:
autostart_stdout_metrics: true

[fllay]

@fllay I was using x310, it was working with Pi5, however when i switched to usrp b210. The gnb stops here: Do you have any recommendations? gnb conf: cu_cp: amf: addr: 127.0.0.1 # The address or hostname of the AMF. bind_addr: 127.0.0.1 # A local IP that the gNB binds to for traffic from the AMF. supported_tracking_areas: # Configure the TA associated with the CU-CP - tac: 1 plmn_list: - plmn: "00101" tai_slice_support_list: - sst: 1

ru_sdr: device_driver: uhd # The RF driver name. device_args: type=b200,num_recv_frames=32,num_send_frames=32 # Optionally pass arguments to the selected RF driver.

srate: 23.04 # RF sample rate might need to be adjusted according to selected bandwidth. otw_format: sc12 tx_gain: 30 # Transmit gain of the RF might need to adjusted to the given situation. rx_gain: 37 # Receive gain of the RF might need to adjusted to the given situation.

cell_cfg: dl_arfcn: 663334 # ARFCN of the downlink carrier (center frequency). band: 77 # The NR band. channel_bandwidth_MHz: 20 # Bandwith in MHz. Number of PRBs will be automatically derived. common_scs: 30 # Subcarrier spacing in kHz used for data. plmn: "00101" # PLMN broadcasted by the gNB. tac: 1 # Tracking area code (needs to match the core configuration). pci: 1 # Physical cell ID.

pcap: mac_enable: false # Set to true to enable MAC-layer PCAPs. mac_filename: /tmp/gnb_mac.pcap # Path where the MAC PCAP is stored. ngap_enable: false # Set to true to enable NGAP PCAPs. ngap_filename: /tmp/gnb_ngap.pcap # Path where the NGAP PCAP is stored. metrics: autostart_stdout_metrics: true

I think that is normal. Just press 't' to show the metrics.

[ALISROURR]

Which UHD version are you using?

[Legend-Dd]

Could you help me? I managed to set up Open5GS and SRSRAN on a Raspberry Pi 5 for 5G SA, and when I register the B210 base station, this error pops up, do you know if there is any solution?

/home/core5g/srsRAN_Project/lib/pdcp/pdcp_entity_rx.cpp:540: srsran::security::security_result srsran::pdcp_entity_rx::apply_deciphering_and_integrity_check(srsran::byte_buffer, uint32_t):  Assertion failed - Worker index exceeds number of crypto workers. worker_idx=4 max_nof_crypto_workers=4 Aborted

[Legend-Dd]

Could you help me? I managed to set up Open5GS and SRSRAN on a Raspberry Pi 5 for 5G SA, and when I register the B210 base station, this error pops up, do you know if there is any solution?

/home/core5g/srsRAN_Project/lib/pdcp/pdcp_entity_rx.cpp:540: srsran::security::security_result srsran::pdcp_entity_rx::apply_deciphering_and_integrity_check(srsran::byte_buffer, uint32_t): Assertion failed - Worker index exceeds number of crypto workers. worker_idx=4 max_nof_crypto_workers=4 Aborted

[robertfalkenberg]

It is a known issue with the determination of available CPUs on the system. It is fixed already and will be available on next push to main.

As a temporary "dirty" fix, you can put a sufficiently-large hard-coded number (e.g. 8) instead of max_nof_crypto_workers_) in the constructor of pdcp_entity_rx:

pdcp_entity_rx::pdcp_entity_rx(uint32_t                        ue_index,
                               rb_id_t                         rb_id_,
                               pdcp_rx_config                  cfg_,
                               pdcp_rx_upper_data_notifier&    upper_dn_,
                               pdcp_rx_upper_control_notifier& upper_cn_,
                               timer_factory                   ue_ul_timer_factory_,
                               task_executor&                  ue_ul_executor_,
                               task_executor&                  crypto_executor_,
                               uint32_t                        max_nof_crypto_workers_,
                               pdcp_metrics_aggregator&        metrics_agg_) :
  pdcp_entity_tx_rx_base(rb_id_, cfg_.rb_type, cfg_.rlc_mode, cfg_.sn_size),
  logger("PDCP", {ue_index, rb_id_, "UL"}),
  cfg(cfg_),
  rx_window(logger, pdcp_window_size(pdcp_sn_size_to_uint(cfg.sn_size))),
  upper_dn(upper_dn_),
  upper_cn(upper_cn_),
  ue_ul_timer_factory(ue_ul_timer_factory_),
  ue_ul_executor(ue_ul_executor_),
  crypto_executor(crypto_executor_),
  max_nof_crypto_workers(max_nof_crypto_workers_),   ///<<<---- replace "max_nof_crypto_workers_" with e.g. 8 according to your system.

[Legend-Dd]

Could this work on my Raspberry Pi 5? I only have 4 cores.

[robertfalkenberg]

yes