intel/flexran_l1_spree

Verified Publisher

By Intel Corporation

•Updated 11 months ago

Image

Networking

Internet of things

Operating systems

10K+

Overview Tags

intel/flexran_l1_spree repository overview

⁠1. FlexRAN™ software reference stack

Intel® provides a vRAN reference architecture in the form of the FlexRAN™ software reference stack, which demonstrates how to optimize VDU software implementations using Intel® C++ class libraries,leveraging the Intel® Advanced Vector Extensions 512 (Intel® AVX-512) instruction set. The multi-threaded design allows a single VDU software implementation to scale to meet the requirements of multiple deployment scenarios, scaling from single small cells deployments, optimized D-RAN deployments or servicing large number of 5G cells in C-RAN pooled deployments.As a SW implementation, it is also capable of supporting LTE, 5G narrow band and 5G massive MIMO deployments all from the same SW stack using the O-RAN 7.2x split. The FlexRAN™ software reference solution framework by Intel® is shown in below diagram:

⁠2. FlexRAN™ containerd image

Since 2022, Intel® FlexRAN team is publishing docker image to docker hub. From 2024, will publish container image. The purpose is to make more and more potential users can easily enter the door and play the game. The containerd image include only binaries, runtime dependency libraries, configure files and several typical cases. If downloader had already been a NDA customer of Intel®, They can get corresponding source code, more test cases and supports from Intel® FlexRAN team. If you are new entry users and just want to do a quick try, please follow below guides. if you have further intention, please contact Intel® FlexRAN Marketing team.

⁠3. User Guide

⁠3.1. HW list

Category	Details
CPU	4th Generation Intel® Xeon® Scalable Processor with vRAN Boost MCC 32 cores
Memory	16*16GB DDR5
Storage	2TB INTEL® SSDPE2KX020T8
NIC1	Intel® Corporation I350(LoM to CPU-0)
NIC2	Intel® Corporation Ethernet Controller E810-C for SFP
Baseband dev	vRAN Boost embeded in SPR-EE CPU Optional, use software mode only if not present

⁠3.2. SW list

Category	Components	Details
Firmware	IFWI	Includes BIOS, BMC, ME as well as FRUSDR.
	E810	4.50 0x8001d8b5 1.3597.0
OS	Ubuntu 24.04	Ubuntu Server 24.04 Realtime kernel 6.8.1-1002
Drivers	E810	ice 1.14.11
Cloudnative	Kubespray	2.25.0
	kubernetes	1.28.8
	Container runtime	Containerd 1.7.16
DPDK	DPDK release	dpdk-23.11

⁠3.3. Prequisition

⁠3.3.1. RT kernel configuration

Ubuntu realtime kernel is available: https://ubuntu.com/realtime-kernel⁠

Install TuneD:

$ apt install tuned
$ vim /etc/grub.d/00_tuned

Add following line to the end of this file

echo "export tuned_params"

Edit /etc/tuned/realtime-variables.conf to add isolated_cores=1-30, 33-62:

isolated_cores=1-30,33-62

Edit /usr/lib/tuned/realtime/tuned.conf to add add nohz, nohz_full and rcu_nocbs:

cmdline_realtime=+isolcpus=${managed_irq}${isolated_cores} nohz=on nohz_full=${isolated_cores} rcu_nocb_poll rcu_nocbs=${isolated_cores} nosoftlockup tsc=nowatchdog

Activate Real-Time Profile:

$ tuned-adm profile realtime

Modify /etc/default/grub

GRUB_CMDLINE_LINUX="intel_iommu=on iommu=pt vfio_pci.enable_sriov=1 vfio_pci.disable_idle_d3=1 usbcore.autosuspend=-1 selinux=0 enforcing=0 nmi_watchdog=0 crashkernel=auto softlockup_panic=0 audit=0 mce=off hugepagesz=1G hugepages=50 hugepagesz=2M hugepages=0 default_hugepagesz=1G irqaffinity=0,31,32,63" GRUB_CMDLINE_LINUX_DEFAULT="$tuned_params" GRUB_INITRD_OVERLAY="${GRUB_INITRD_OVERLAY:+$GRUB_INITRD_OVERLAY }$tuned_initrd"

$ sudo update-grub
$ sudo reboot

After configuration ,check the kernel parameter, which should look like:

$ cat /proc/cmdline
BOOT_IMAGE=/vmlinuz-6.8.1-1002-realtime root=/dev/mapper/ubuntu--vg-ubuntu--lv ro intel_iommu=on iommu=pt vfio_pci.enable_sriov=1 vfio_pci.disable_idle_d3=1 usbcore.autosuspend=-1 selinux=0 enforcing=0 nmi_watchdog=0 crashkernel=auto softlockup_panic=0 audit=0 tsc=nowatchdog mce=off hugepagesz=1G hugepages=50 hugepagesz=2M hugepages=0 default_hugepagesz=1G irqaffinity=0,31,32,63 skew_tick=1 isolcpus=managed_irq,domain,1-30,33-62 nohz=on nohz_full=1-30,33-62 rcu_nocb_poll rcu_nocbs=1-30,33-62 nosoftlockup rcupdate.rcu_normal_after_boot=1

⁠3.3.2. Configure the CPU Frequency and cstate

Further improve the deterministic and power efficiency

$ cpupower frequency-set -g performance

Set cpu core frequency to 2.5Ghz

$ cpupower frequency-set -d 2500000 -u 2500000

⁠3.3.3. Kubernetes and container installation

Make sure specific version of kubernetes and containerd and docker had been installed.

Configure Containerd to support root-less container device access

$ vim /etc/containerd/config.toml
[plugins]
[plugins."io.containerd.grpc.v1.cri"]
device_ownership_from_security_context = true

$ systemctl restart containerd.service

⁠3.3.4. Kubernetes plugins installation

Except for kubernetes and container, below plugin is required (install from master):

calico:
- follow calico instruction on offical website - http://docs.projectcalico.org⁠

-SRIOV FEC operator (optional) -follow FEC operator on offical website -https://github.com/smart-edge-open/sriov-fec-operator⁠

after SRIOV FEC operator deployed, display plugins as below

{ "intel.com/intel_fec_5g": "0", "intel.com/intel_fec_acc200": "1", }

SRIOV (cni and network device plugin):

follow SRIOV instruction on SRIOV GitHub - https://github.com/k8snetworkplumbingwg/sriov-network-device-plugin⁠.

SRIOV DP configuration
below is an example to cofigure SRIOV DP configure map, including FEC, DU and ORU needed resources. customer can configure according to the actual requirment.

$ cd sriov-network-device-plugin 
$ cat <<EOF > deployments/configMap.yaml
apiVersion: v1  
kind: ConfigMap  
metadata:  
  name: sriovdp-config  
  namespace: kube-system  
data:  
  config.json: |  
    {  
        "resourceList": [  
            {  
              "resourceName": "intel_fec_5g",  
                "deviceType": "accelerator",  
                "selectors": {  
                    "vendors": ["8086"],  
                    "devices": ["57c1"]  
                }  
            },  
            {  
              "resourceName": "intel_sriov_odu",  
                "selectors": {  
                    "vendors": ["8086"],  
                    "devices": ["1889"],  
                    "drivers": ["vfio-pci"],  
                    "pfNames": ["ens9f1"]  
                }  
            },  
            {  
              "resourceName": "intel_sriov_oru",  
                "selectors": {  
                    "vendors": ["8086"],  
                    "devices": ["1889"],  
                    "drivers": ["vfio-pci"],  
                    "pfNames": ["ens9f0"]  
                }  
            }  
        ]  
    }  
EOF  
$ kubectl create -f deployments/configMap.yaml  
$ kubectl create -f deployments/sriovdp-daemonset.yaml

⁠3.4. Prepare env

⁠3.4.1. DPDK pakage

Download DPDK

$ cd /opt/  
$ wget http://static.dpdk.org/rel/dpdk-23.11.tar.xz 
$ tar xf /opt/dpdk-23.11.tar.xz

Configure FEC and FVL SRIOV (example as below)

$ RTE_SDK=/opt/dpdk-23.11
$ modprobe vfio-pci enable_sriov=1 disable_idle_d3=1
$ echo 1 | sudo tee /sys/module/vfio_pci/parameters/enable_sriov
$ echo 1 | tee /sys/module/vfio_pci/parameters/disable_idle_d3
$ $RTE_SDK/usertools/dpdk-devbind.py -b vfio-pci f7:00.0
$ export UUID=00112233-4455-6677-8899-aabbccddeeff
$ echo 1 > /sys/bus/pci/devices/0000:f7:00.0/sriov_numvfs
$ $RTE_SDK/usertools/dpdk-devbind.py -b vfio-pci 0000:f7:00.1
$ git clone https://github.com/intel/pf-bb-config.git
$ cd ./pf-bb-config
$ ./pf_bb_config VRB2 -v $UUID -c vrb2/vrb2_config_vf.cfg

$ echo 4 > /sys/bus/pci/devices/0000:4b:00.0/sriov_numvfs
$ ip link set ens9f0 vf 0 mac 00:11:22:33:00:00
$ ip link set ens9f0 vf 1 mac 00:11:22:33:00:10
$ ip link set ens9f0 vf 2 mac 00:11:22:33:00:20
$ ip link set ens9f0 vf 3 mac 00:11:22:33:00:30
$ $RTE_SDK/usertools/dpdk-devbind.py -b vfio-pci 0000:4b:02.0 0000:4b:02.1
$ $RTE_SDK/usertools/dpdk-devbind.py -b vfio-pci 0000:4b:02.2 0000:4b:02.3

$ echo 4 > /sys/bus/pci/devices/0000:4b:00.1/sriov_numvfs
$ ip link set ens9f1 vf 2 mac 00:11:22:33:00:21
$ ip link set ens9f1 vf 3 mac 00:11:22:33:00:31
$ ip link set ens9f1 vf 0 mac 00:11:22:33:00:01
$ ip link set ens9f1 vf 1 mac 00:11:22:33:00:11
$ $RTE_SDK/usertools/dpdk-devbind.py -b vfio-pci 0000:4b:0a.0 0000:4b:0a.1
$ $RTE_SDK/usertools/dpdk-devbind.py -b vfio-pci 0000:4b:0a.2 0000:4b:0a.3
$ $RTE_SDK/usertools/dpdk-devbind.py -s

After configuration, need to restart SRIOV docker container to make VF resource ready.

$ cat <<EOF > /opt/restart_sriov_container.sh
  #!/bin/bash
  
  docker ps | grep sriov
  docker kill `docker ps | grep sriov | head -n 1 | awk -F ' ' '{print $1}'`
  while ((`docker ps | grep sriov | wc -l` < 2 ))
  do
     sleep 3
     docker ps | grep sriov >/dev/null 2>&1
     echo "..."
  done
EOF
$chmod +x /opt/restart_sriov_container.sh; sh /opt/restart_sriov_container.sh

After restart sriov container, the resources can be seen thru below command:

$ sudo apt install jq 
$ kubectl get node node-name -o json | jq '.status.allocatable'
{  

  "intel.com/intel_fec_5g": "1",  
  "intel.com/intel_sriov_odu": "4",  
  "intel.com/intel_sriov_oru": "4",  

}

label node

$ kubectl label node host_name testnode=worker1

⁠3.5. container image prepare

$ docker pull intel/flexran_l1_spree:v24.07
$ docker save intel/flexran_l1_spree:v24.07 > flexran_image_24.03.tar
$ ctr -n k8s.io image import flexran_image_24.07.tar

⁠3.6. Example yaml file for flexran timer mode test and xran mode test

below gives the diagram of the deployment for flexran timer mode and xRAN mode. based on that, following chapters give the example yaml file for deployment.

⁠3.6.1. Example yaml file for flexran timer mode test

Support SRIOV FEC operator or use pf-bb-config to configure FEC device, below is an example of using SRIOV FEC opertor.

$ cat <<EOF > /opt/flexran_phycfg_timer.yaml  
apiVersion: v1
kind: ConfigMap
metadata:
   name: flexran-configmap-timer
data:
   ENABLE_AUTO_CONFIG: "TRUE"
   ENABLE_CORE_ASSIGN: "FALSE"
   FEC_SRIOV_INFO_NAME: "PCIDEVICE_INTEL_COM_INTEL_FEC_ACC200_INFO"
   FEC_VF_TOKEN: "02bddbbf-bbb0-4d79-886b-91bad3fbb510"
   flexranPhyCfg: |
      {
         "DPDK": {
         "dpdkIovaMode": 1,
         "dpdkBasebandFecMode": 1
         }
      }
EOF

if you use pf-bb-configure then can modify below parameters in flexran_phycfg_timer.yaml and change "intel_fec_acc200" to "intel_fec_5g" in flexran_phycfg_timer.yaml

FEC_SRIOV_INFO_NAME: "PCIDEVICE_INTEL_COM_INTEL_FEC_5G_INFO" FEC_VF_TOKEN: "00112233-4455-6677-8899-aabbccddeeff"

$ kubectl apply -f flexran_phycfg_timer.yaml

$ cat <<EOF > /opt/flexran_timer_mode.yaml  
apiVersion: v1
kind: Pod
metadata:
  labels:
    app: flexran-binary-release
  name: flexran-binary-release
spec:
  securityContext:
    fsGroup: 1250
  nodeSelector:
     testnode: worker1
  containers:
  - securityContext:
      runAsNonRoot: true
      runAsUser: 1250
      capabilities:
        add:
          - IPC_LOCK
          - SYS_NICE
          - DAC_READ_SEARCH
    command: [ "/bin/bash", "-c", "--"]
    args: ["sleep infinity"]
    tty: true
    stdin: true
    env:
    - name: LD_LIBRARY_PATH
      value: /opt/oneapi/lib/intel64
    envFrom:
    - configMapRef:
        name: flexran-configmap-timer
    image: docker.io/intel/flexran_l1_spree:v24.07
    imagePullPolicy: IfNotPresent
    name: flexran-du
    resources:
      requests:
        memory: "44Gi" 
        intel.com/intel_fec_acc200: '1'
        hugepages-1Gi: 30Gi  
      limits:
        memory: "44Gi"
        intel.com/intel_fec_acc200: '1'
        hugepages-1Gi: 30Gi
    volumeMounts:
    - name: hugepage
      mountPath: /hugepages
      readOnly: false
    - name: varrun
      mountPath: /tmp/dpdk
      readOnly: false
      readOnly: false
  volumes:
  - name: hugepage
    emptyDir:
      medium: HugePages
  - name: varrun
    emptyDir: {}
EOF
  
$ kubectl apply -f /opt/flexran_timer_mode.yaml

customer can choose test cases according to server platform, below is an example of test on SPR-MCC

In the first terminal, run the following command:


$ kubectl exec -it flexran-binary-release -- bash
$ cd flexran/bin/nr5g/gnb/l1/
$ ./l1.sh -e

In the second terminal, run the following command or other mcc cases in the same path

$ kubectl exec -it flexran-binary-release -- bash
$ cd flexran/bin/nr5g/gnb/testmac
$ ./l2.sh --testfile=spr-sp-mcc/sprsp_mcc_mu0_10mhz_4x4_hton.cfg

⁠3.6.2. Example yaml file for xran mode test

xRAN mode scenario will involve two hosts, DU host and RU host.

For DU setup, need intel_fec_5g/intel_fec_acc200 and intel_sriov_odu resources,customer can use flexran_phycfg_xran_du.yaml file to configure DU parameters aotumatically, also support set "ENABLE_AUTO_CONFIG" to FALSE manually configure DU paramters.


$ cat <<EOF > /opt/flexran_phycfg_xran_du.yaml  
apiVersion: v1
kind: ConfigMap
metadata:
   name: flexran-configmap-du
data:
   ENABLE_AUTO_CONFIG: "TRUE"
   ENABLE_CORE_ASSIGN: "FALSE"
   FEC_SRIOV_INFO_NAME: "PCIDEVICE_INTEL_COM_INTEL_FEC_ACC200_INFO"
   FEC_VF_TOKEN: "02bddbbf-bbb0-4d79-886b-91bad3fbb510"
   sriovMappingTable: |
    {
      "DU_Local_PCI_01": {
        "peerMAC": "00:11:22:33:00:01",
        "tag": "fronthaul_du",
        "index": "0"
      },
      "<DU_Local_PCI_02": {
        "peerMAC": "00:11:22:33:00:11",
        "tag": "fronthaul_du",
        "index": "1"
      },
      "<DU_Local_PCI_03": {
        "peerMAC": "00:11:22:33:00:21",
        "tag": "fronthaul_du",
        "index": "2"
      },
      "<DU_Local_PCI_04": {
        "peerMAC": "00:11:22:33:00:31",
        "tag": "fronthaul_du",
        "index": "3"
      }
    }
   flexranPhyCfg: |
      {
         "DPDK": {
         "dpdkIovaMode": 1,
         "dpdkBasebandFecMode": 1
         }
      }
EOF

$ kubectl create -f /opt/flexran_phycfg_xran_du.yaml

$ cat <<EOF > /opt/flexran_xran_mode_du.yaml  
apiVersion: v1
kind: Pod
metadata:
  labels:
    app: flexran-vdu
  name: flexran-vdu
spec:
  securityContext:
     fsGroup: 1250
  nodeSelector:
     testnode: worker1
  containers:
  - securityContext:
      runAsNonRoot: true
      runAsUser: 1250
      capabilities:
        add:
          - IPC_LOCK
          - SYS_NICE
          - DAC_READ_SEARCH
    command: [ "/bin/bash", "-c", "--" ]
    args: ["sleep infinity"]
    tty: true
    stdin: true
    env:
    - name: LD_LIBRARY_PATH
      value: /opt/oneapi/lib/intel64
    envFrom:
    - configMapRef:
        name: flexran-configmap-du
    image: docker.io/intel/flexran_l1_spree:v24.07
    name: flexran-vdu
    resources:
      requests:
        memory: "32Gi"
        intel.com/intel_fec_acc200: '1'
        intel.com/intel_sriov_odu: '4'
        hugepages-1Gi: 40Gi
      limits:
        memory: "32Gi"
        intel.com/intel_fec_acc200: '1'
        intel.com/intel_sriov_odu: '4'
        hugepages-1Gi: 40Gi
    volumeMounts:
    - name: hugepage
      mountPath: /hugepages
    - name: varrun
      mountPath: /var/run/dpdk
      readOnly: false
  volumes:
  - name: hugepage
    emptyDir:
      medium: HugePages
  - name: varrun
    emptyDir: {}
EOF  

$ kubectl create -f /opt/flexran_xran_mode_du.yaml

For RU setup, for container setup need intel_sriov_oRu resources, here is an example for ORU deployment file

$ cat <<EOF > /opt/flexran_phycfg_xran_ru.yml
apiVersion: v1
kind: ConfigMap
metadata:
  name: flexran-configmap-ru
data:
  ENABLE_AUTO_CONFIG: "TRUE"
  ENABLE_CORE_ASSIGN: "FALSE"
  ORU_SRIOV_NAME: "PCIDEVICE_INTEL_COM_INTEL_SRIOV_ORU"
  sriovMappingTable: |
    {
      "RU_Local_PCI_01": {
        "peerMAC": "00:11:22:33:00:00",
        "tag": "fronthaul_ru",
        "index": "0"
      },
      "RU_Local_PCI_02": {
        "peerMAC": "00:11:22:33:00:10",
        "tag": "fronthaul_ru",
        "index": "1"
      },
      "RU_Local_PCI_03": {
        "peerMAC": "00:11:22:33:00:20",
        "tag": "fronthaul_ru",
        "index": "2"
      },
      "RU_Local_PCI_04": {
        "peerMAC": "00:11:22:33:00:30",
        "tag": "fronthaul_ru",
        "index": "3"
      }
    }
  flexranPhyCfg: |
    {
      "DPDK": {
        "dpdkIovaMode": 1,
        "dpdkBasebandFecMode": 1
      }
    }

$ cat <<EOF > /opt/flexran_xran_mode_ru.yaml
apiVersion: v1
kind: Pod
metadata:
  labels:
    app: flexran-oru
  name: flexran-oru
spec:
  securityContext:
    fsGroup: 1250
  nodeSelector:
     testnode: worker1
  containers:
  - securityContext:
      runAsNonRoot: true
      runAsUser: 1250
      capabilities:
        add:
          - IPC_LOCK
          - SYS_NICE
          - DAC_READ_SEARCH
    command: [ "/bin/bash", "-c", "--" ]
    args: ["sleep infinity"]
    tty: true
    stdin: true
    env:
    - name: LD_LIBRARY_PATH
      value: /opt/oneapi/lib/intel64
    image: flexran.docker.registry/flexran_vdu:24.07
    name: flexran-oru
    resources:
      requests:
        memory: "24Gi"  
        intel.com/intel_sriov_oru: '4'
        hugepages-1Gi: 20Gi   
      limits:
        memory: "24Gi"
        intel.com/intel_sriov_oru: '4'
        hugepages-1Gi: 20Gi
    volumeMounts:
    - name: hugepage
      mountPath: /hugepages
    - name: varrun
      mountPath: /tmp/dpdk
      readOnly: false
  volumes:
  - name: hugepage
    emptyDir:
      medium: HugePages
  - name: varrun
    emptyDir: {}
EOF  

$ kubectl create -f /opt/flexran_phycfg_xran_ru.yml
$ kubectl create -f /opt/flexran_xran_mode_ru.yaml

Below are configuration of the fronthaul PCI addresses on RU side

$ vim FLEXRAN_SRC_RU_HOST/bin/nr5g/gnb/l1/spree_mcc/sub3_mu0_20mhz_4x4/oru/run_o_ru.cfg
#!/bin/bash
num_eth_vfs=1
vf_addr_o_xu_a=<RU_Local_PCI_01>

⁠Update the fronthaul MAC addresses on DU side

$ vim $FLEXRAN_SRC_RU_HOST/bin/nr5g/gnb/l1/spree_mcc/sub3_mu0_20mhz_4x4/oru/usecase_ru.cfg
# remote O-XU 0 Eth Link 0
oXuRem0Mac0=00:11:22:33:00:00
oXuRem0Mac1=00:11:22:33:00:10
oXuRem1Mac0=00:11:22:33:00:20
oXuRem1Mac1=00:11:22:33:00:30
iovaMode=1

Below chapter give the steps to run xRAN mode test case: "sub3_mu0_20mhz_4x4"

Open a new terminal on DU side, run the following command:

$ kubectl exec -it pod-name -- bash    
$ cd flexran/bin/nr5g/gnb/l1/spree_mcc/sub3_mu0_20mhz_4x4/gnb/  
update the file phycfg_xran.xml for FEC address and token , xrancfg_sub6_oru.xml for vf-pci address if using manually configurations 
$ ./l1.sh -oru

Open another new terminal, run the following command:

$ kubectl exec -it pod-name -- bash    
$ cd flexran/bin/nr5g/gnb/testmac  
$ ./l2.sh --testfile=../l1/spree_mcc/sub3_mu0_20mhz_4x4/gnb/testmac_spree_mu0_20mhz_hton_oru.cfg  

# in RU part，run the following command in container as an example:
$ kubectl exec -it pod-name -- bash

$ cd ~/flexran/bin/nr5g/gnb/oru
$ ./run_o_ru.sh -r spree_mcc sub3_mu0_20mhz_4x4

You can run the same for other two test cases "sub3_mu0_10mhz_4x4" and "sub6_mu1_100mhz_4x4"

⁠3.7. Core pining

Intel docker image also provide the support of core pining feature. In order to enable this feature, you need to make below configuration and change of yaml file.

⁠3.7.1. Configuration

Enable core pining feature：

#!/bin/bash
pathfile=/etc/kubernetes/kubelet-config.yaml
sed -i 's/cpuManagerReconcilePeriod: 0s/cpuManagerReconcilePeriod: 10s/g' $pathfile

cat >> $pathfile << EOF
cpuManagerPolicy: static
systemReserved:
  cpu: 2000m
  memory: 2000Mi
kubeReserved:
  cpu: 1000m
  memory: 1000Mi
EOF
rm -rf /var/lib/kubelet/cpu_manager_state
systemctl restart kubelet

copy above code line to a file and execute:

$ sh core_pining_kubelet_config.sh

Disable core pining feature:

#!/bin/bash

pathfile=/etc/kubernetes/kubelet-config.yaml
#pathfile=config.yaml
sed -i 's/cpuManagerReconcilePeriod: 10s/cpuManagerReconcilePeriod: 0s/g' $pathfile

sed -i 's/cpuManagerPolicy: static/ /g' $pathfile
sed -i 's/systemReserved:/ /g' $pathfile
sed -i 's/cpu: 2000m/ /g' $pathfile
sed -i 's/memory: 2000Mi/ /g' $pathfile
sed -i 's/kubeReserved:/ /g' $pathfile
sed -i 's/cpu: 1000m/ /g' $pathfile
sed -i 's/memory: 1000Mi/ /g' $pathfile
rm -rf /var/lib/kubelet/cpu_manager_state
systemctl restart kubelet

copy above code line into a file and execute:

$ sh uncore_pining_kubelet_config.sh

⁠3.7.2. Example yaml file for flexran timer mode test (with core pining)

Modify 3.6.1 /opt/flexran_phycfg_timer.yaml set ENABLE_CORE_ASSIGN to true.


$ cat <<EOF > /opt/flexran_phycfg_timer.yaml  
...
data:
   ENABLE_CORE_ASSIGN: "TRUE"
...

$ cat <<EOF > /opt/flexran_timer_mode.yaml  
......
    name: flexran-du
    resources:
      requests:
        cpu: 30
......
      limits:
        cpu: 30
......
EOF  
  
$ kubectl create -f /opt/flexran_timer_mode.yaml

for timer mode, case procedure is the same as the one without core pining. configure the max required cores when hyper thread on to ensure high performance cases can pass.if customer needn't run high performance cases, can manually adjust cpu core count in yaml file according to case configurations.

⁠3.7.3. Example yaml file for xran mode test (with core pining)

the same modification with timer mode.

⁠3.8. Legal Disclaimer

For GPL/LGPL open source libs/components used by flexran docker image at run time. User can find the used version in below git hub repo: https://github.com/intel/flexRAN-docker-image-dependencies⁠

⁠3.9. Customer Support Declare

For further support, please contact Intel® flexRAN marketing team and FAE/PAE team and Engineering team.

⁠Note:

Since 23.11 release, the same docker image is also validated by E2E test rather than only component test (timer mode and xRAN mode, described above) . If customer want to know detail on E2E test, please contact Intel® flexRAN marketing team and FAE/PAE team and Engineering team.

Tag summary

Recent tags

Content type

Image

Digest

sha256:d2e4d2e0c…

Size

1.9 GB

Last updated

11 months ago

docker pull intel/flexran_l1_spree

This week's pulls

Pulls:

256

Last week

Learn more⁠