Getting Started with DPDK and UHD

From Ettus Knowledge Base
Revision as of 15:21, 30 August 2026 by MichaelDickens (Talk | contribs) (→‎Enable hugepages)

Jump to: navigation, search

Application Note Number and Authors

AN-500 by Nate Temple, Alex Williams, Wade Fife, Matt Prost, and Michael Dickens

Overview

This application note walks through the process to get started with the Data Plane Development Kit (DPDK) driver within UHD.

Abstract

UHD's original support for networked devices used the Linux kernel's sockets implementation. Every call to send() or recv() would cause a context switch and invite the kernel's scheduler to replace the UHD thread with something else. Because the typical scheduler is optimized to distribute CPU time fairly across multiple cores, the timing-critical threads might sporadically be hit with sleeping time and the thread might be migrated off its current CPU and forced to run on another. The overhead and random latency spikes caused by these make it difficult to enable reliable real-time streaming at higher rates.

DPDK is a high-speed packet processing framework that enables a kernel bypass for network drivers. By putting the entire driver in user-space, avoiding context switches, and pinning I/O threads to cores, UHD and DPDK combine to largely prevent the latency spikes induced by the Linux thread scheduler. In addition, DPDK lowers the overall overhead for packet processing.

DPDK utilizes dedicated resources -- CPU cores and memory -- to increase data throughput. This tradeoff may be acceptable for some hardware and some use-cases. DPDK is intended for use-cases where standard networking is not performant enough; the dedicated resources are otherwise wasted if not required.

Supported Devices

USRPs

DPDK is supported on the following USRP devices, which are those providing 10 GbE (SFP+) and 100 GbE (QSFP28) links, whether physical or virtual: 

   E320, N300 / N310, N320 / N321, X300 / X310, X410, X420, X440

Host Network Cards

DPDK is supported on many Intel and Mellanox based 10Gb and 100Gb NICs and lots of other NICs. Below is a list of NICs Ettus Research has tested. For a full list of NICs supported by DPDK, please see the DPDK manual.

  • Intel X520-DA1 (1x10Gb)
  • Intel X520-DA2 (2x10Gb)
  • Intel X710-DA2 (2x10Gb)
  • Intel X710-DA4 (4x10Gb)
  • Intel XL710-QDA2 (2x40Gb breakout to 4x10Gb)
  • Intel E810-CQDA1 (1x100Gb and 1x4x10Gb)
  • Intel E810-CQDA2 (1x100Gb and 2x4x10Gb; note that this NIC does not support 2x100Gb)
  • Intel E810-2CQDA2 (1x100Gb and 2x4x10Gb; 2x100Gb with some work)
  • Intel E830-CQDA2 (2x100Gb or 8x10Gb)
  • Mellanox MCX4121A-ACAT ConnectX-4 Lx (2x10Gb)
  • Mellanox MCX515A-CCAT ConnectX-5 EN (2x100Gb or 2x10Gb)
  • Mellanox MCX516A-CCAT ConnectX-5 EN (2x100Gb or 2x10Gb)
  • Mellanox MCX516A-CDAT ConnectX-5 Ex EN (2x100Gb or 2x10Gb)
  • Mellanox MCX623106AN-CDAT ConnectX-6 Dx EN (2x100Gb)
  • NI Dual 100 Gigabit Ethernet PCIe Interface Kit (PN 788216-01)

References

UHD DPDK Version Dependencies

  • UHD 3.x requires DPDK 17.11
  • UHD 4.0 and 4.1 require DPDK 18.11
  • UHD 4.2 to 4.7 can use any version of DPDK from 18.11 to 21.11
  • UHD 4.8 and 4.9 can use any version of DPDK from 18.11 to 24.11
  • UHD 4.10 can use any version of DPDK from 19.11 to 25.11

Installing DPDK

We recommend installing DPDK via the system-provided installer; for example with Ubuntu: 

   sudo apt install dpdk dpdk-dev

While it is possible to install DPDK from source, we recommend using the system-provided install unless there is a very good reason to not do so. DPDK version 18 and older can be challenging to correctly build from source. DPDK version 19 and newer use meson and ninja along with default settings needed by UHD and hence are relatively simple to build and install. If you require installing DPDK from source, the install guide for various versions is noted below:

DPDK releases come three times per year in April (version X.04), July (version X.07), and November (version X.11). The X.11 version is more of a formal release ("long term service" or LTS). While any can be used with UHD, we strongly recommend using just the formal releases.

NOTE: It is sometimes necessary to manually install NIC device drivers because of improved feature sets or just getting a NIC to work (e.g., the system provided NIC drivers are too old). In this case, to use DPDK one must build DPDK and UHD from source. Follow these steps in the order noted to guarantee that DPDK is built with the updated NIC drivers and that UHD is built with the noted DPDK:

  1. Remove the system provided DPDK and DPDK-DEV and all UHD install(s)
  2. Download and install the new NIC drivers, NVM, and anything else that the NIC needs to be updated to the version needed
  3. Download and install DPDK from source
  4. Download and install UHD from source

Installing UHD

Once the dpdk and dpdk-dev packages are installed, UHD will locate them during a build and you should see DPDK in the enabled components lists when running cmake.

NOTE that in general UHD installed from PPA or system packages does not include support for DPDK, and even if DPDK is installed alongside these UHD it will not be used. In order to get UHD with DPDK support, UHD must be built from source.

Enable hugepages

DPDK requires hugepages to be enabled and "large enough". There are multiple ways to enable hugepages. The following is an easy way to enable this feature and allocate enough hugepages For UHD's DPDK needs.

Edit your grub configuration file, /etc/default/grub and add the follow parameters to GRUB_CMDLINE_LINUX_DEFAULT: 

   hugepages=2048

On a vanilla Ubuntu system this entry may read like this: 

   GRUB_CMDLINE_LINUX_DEFAULT="quiet splash iommu=pt intel_iommu=on hugepages=2048"

Close /etc/default/grub and at the command prompt, update your grub configuration with the command: 

   sudo update-grub

For these settings to take effect, reboot your host machine.

Preparing your UHD Configuration File

You must note the MAC addresses for your NICs before proceeding.

The MAC addresses for your NICs can be found by running the command: 

   ip a

You must then create a UHD configuration file.

For UHD 3 the location is /root/.uhd/uhd.conf. 

   sudo su
   mkdir -p /root/.uhd
   nano /root/.uhd/uhd.conf

For UHD 4 the location is /root/.config/uhd.conf. 

   sudo su
   mkdir -p /root/.config
   nano /root/.config/uhd.conf

UHD 3.x

An example uhd.conf file is listed below. Note that field names in UHD 3.x are slightly different from UHD 4.0.

You should update the following fields for your configuration from this example:

  • Update the MAC address variables, dpdk-mac, to match your NIC
  • Update the dpdk-driver if the location is different on your system. /usr/lib/x86_64-linux-gnu/dpdk-17.11-drivers/ is the default location on Ubuntu 18.04.x when dpdk is installed via apt.
  • Update the dpdk-corelist and dpdk-io-cpu fields. In this example, a two port NIC is used. There should be one core for the main dpdk thread (in this example core #2), and then separate cores assigned to each NIC (in this example core #3 for the first port on the NIC, core #4 for the second port on the NIC)
  • Update the dpdk-ipv4 fields to your desired IP range.
    • 192.168.30.2, 192.168.40.2 on a default X3xx system
    • 192.168.10.2, 192.168.20.2 on a default N3xx system
    • 192.168.10.2 on a default E320 system
   [use_dpdk=1]
   dpdk-mtu=9000
   dpdk-driver=/usr/lib/x86_64-linux-gnu/dpdk-17.11-drivers/
   dpdk-corelist=2,3,4
   dpdk-num-mbufs=4095
   dpdk-mbufs-cache-size=315
   
   [dpdk-mac=aa:bb:cc:dd:ee:f1]
   dpdk-io-cpu = 3
   dpdk-ipv4 = 192.168.10.1/24
   
   [dpdk-mac=aa:bb:cc:dd:ee:f2]
   dpdk-io-cpu = 4
   dpdk-ipv4 = 192.168.20.1/24

Note: Additional information on the UHD configuration file can be found here: https://files.ettus.com/manual_archive/v3.15.0.0/html/page_dpdk.html#dpdk_nic_config

UHD 4.x

An example uhd.conf file is listed below. Note that the field names in UHD 4.x are slightly different from UHD 3.x.

You must verify and/or update the following fields for your configuration from this example:

  • The MAC address variables, dpdk_mac, to match your NIC(s) link(s); note that the MAC address info must be lowercase
  • The dpdk_driver if the location is different on your system. /usr/local/lib/ is the default location on when DPDK is built and installed from source.
  • The dpdk_corelist and dpdk_lcore fields. In this example, a two port NIC and both links are used. There must be one core for the main dpdk thread (in this example core #2 because it is not otherwise used in the file) and then separate cores assigned to each NIC link (in this example core #3 for the one of the NIC links and core #4 for the other NIC link).
  • The dpdk_ipv4 fields to your desired IP range(s).
    • 192.168.30.2, 192.168.40.2 on a default X3xx system
    • 192.168.10.2, 192.168.20.2 on a default N3xx system
    • 192.168.10.2 on a default E320 system
    • 192.168.10.2 and 192.168.20.2 on a default X4xx systems
   [use_dpdk=1]
   dpdk_mtu=9000
   dpdk_driver=/usr/lib/x86_64-linux-gnu/dpdk/pmds-20.0/
   dpdk_corelist=2,3,4
   dpdk_num_mbufs=4096
   dpdk_num_desc=4096
   dpdk_mbuf_cache_size=315
   
   [dpdk_mac=aa:bb:cc:dd:ee:f1]
   dpdk_lcore=3
   dpdk_ipv4=192.168.10.1/24
   
   [dpdk_mac=aa:bb:cc:dd:ee:f2]
   dpdk_lcore=4
   dpdk_ipv4=192.168.20.1/24

Notes:

Additional Host Configuration for NIC Vendors

The process for this step is different for Intel and Mellanox NICs and is detailed in individual sections below.

For these NICs DPDK requires the vfio-pci driver, and it must generally be loaded beforehand: 

   sudo modprobe vfio-pci

Intel NICs

First, take the link(s) down; for example for the link name eth1np1: 

   sudo ifconfig eth1np1 down

Note: You may need to turn off the NIC first before doing the rebind. In Ubuntu under System -> Network -> click the switches to off for the link(s).

Note: Make sure to pick the correct link name(s) for your specific system! If there are multiple Intel NIC links to be used by DPDK take them all down, one each in turn.

Next, you will need to manually rebind the NIC to the vfio-pci drivers.

Identify the PCI address your NIC link(s) is/are at: 

   dpdk-devbind.py -s

Make note of the PCI address that your NIC link(s) is/are connected to for the next step; for example for this output: 

   0000:02:00.0 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb' drv=ixgbe unused=vfio-pci
   0000:02:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb' drv=ixgbe unused=vfio-pci

one would use 02:00.0 and 02:00.1, as follows: 

   sudo dpdk-devbind.py --bind=vfio-pci 02:00.0
   sudo dpdk-devbind.py --bind=vfio-pci 02:00.1

Note: Your PCI address will likely be different than 02:00.0 as shown in the example below.

Note: In older DPDK versions the command does not contain the trailing .py .

You should now see the NIC listed under DPDK devices; for example: 

   # dpdk-devbind.py -s
   
   Network devices using DPDK-compatible driver
   ============================================
   0000:02:00.0 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb' drv=vfio-pci unused=ixgbe
   0000:02:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb' drv=vfio-pci unused=ixgbe

Note: More info can be found on the rebinding process here: https://doc.dpdk.org/guides-25.11/linux_gsg/linux_drivers.html#binding-and-unbinding-network-ports-to-from-the-kernel-modules

Once the NIC link(s) is/are binded to the vfio-pci driver they are ready for use by UHD.

Mellanox NICs

The NVIDIA Mellanox ("Mellanox") NICs do not require manual rebinding. Mellanox provides system IC drivers and additional drivers for DPDK to handle the binding to and from the vfio-pci driver.

In general the Mellanox system-provided NIC and DPDK drivers should work and you should not need to install Mellanox drivers. Mellanox provides tools for updating drivers and NIC firmware, if desired.

If for some reason your host OS does not provide Mellanox DPDK drivers, then these can be installed manually. For example: 

   sudo apt install librte-pmd-mlx5
   sudo modprobe -a ib_uverbs mlx5_core mlx5_ib

If for some reason you are running into issues using a Mellanox NIC, then you can download and install the latest Mellanox drivers from the Mellanox website (https://network.nvidia.com/products/infiniband-drivers/linux/mlnx_ofed/).

The MLX5 poll mode driver library (librte_pmd_mlx5) in DPDK provides support for Mellanox ConnextX-4 through -7 NICs. For DPDK 20 and older only this driver must be enabled manually with the build option CONFIG_RTE_LIBRTE_MLX5_PMD=y when building DPDK. We recommend DPDK 21 and newer when possible; these versions are easy to build from source using meson and ninja.

Running UHD Applications with DPDK

UHD based application (including GNU Radio flowgraphs) can now be ran using a DPDK transport by passing in the Device Argument: use_dpdk=1.

Important Note: In order for UHD to use DPDK, the UHD application must be ran as the root user; this can be done either by prepending sudo to the command or becoming the roort user via sudo su.

For example, running the benchmark_rate utility after issuing sudo su (one can also run this command simply as sudo /usr/local/lib/uhd/examples/benchmark_rate .....

NOTE that the text output from UHD 3 or 4 is about the same; the primary difference is in the initial [INFO] line. 

# cd /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=192.168.10.2,second_addr=192.168.20.2,mgmt_addr=10.2.1.19,master_clock_rate=125e6,use_dpdk=1"

[INFO] [UHD] linux; GNU C++ version 7.3.0; Boost_106501; UHD_3.14.0.HEAD-0-gabf0db4e
EAL: Detected 8 lcore(s)
EAL: Some devices want iova as va but pa will be used because.. EAL: IOMMU does not support IOVA as VA
EAL: No free hugepages reported in hugepages-1048576kB
EAL: Probing VFIO support...
EAL: VFIO support initialized
EAL: PCI device 0000:02:00.0 on NUMA socket -1
EAL:   Invalid NUMA socket, default to 0
EAL:   probe driver: 8086:10fb net_ixgbe
EAL:   using IOMMU type 1 (Type 1)
EAL: Ignore mapping IO port bar(2)
EAL: PCI device 0000:02:00.1 on NUMA socket -1
EAL:   Invalid NUMA socket, default to 0
EAL:   probe driver: 8086:10fb net_ixgbe
EAL: Ignore mapping IO port bar(2)
PMD: ixgbe_dev_link_status_print():  Port 0: Link Down
EAL: Port 0 MAC: aa bb cc dd ee f1
EAL: Port 0 UP: 1
PMD: ixgbe_dev_link_status_print():  Port 1: Link Down
EAL: Port 1 MAC: aa bb cc dd ee f2
EAL: Port 1 UP: 1
EAL: Init DONE!
EAL: Starting I/O threads!
USER2: Thread 1 started
[00:00:00.000003] Creating the usrp device with: addr=192.168.10.2,second_addr=192.168.20.2,mgmt_addr=10.2.1.19,master_clock_rate=125e6,use_dpdk=1...
[INFO] [MPMD] Initializing 1 device(s) in parallel with args: mgmt_addr=10.2.1.19,type=n3xx,product=n310,serial=313ABDA,claimed=False,addr=192.168.10.2,second_addr=192.168.20.2,master_clock_rate=125e6,use_dpdk=1
[INFO] [MPM.PeriphManager] init() called with device args 'product=n310,time_source=internal,master_clock_rate=125e6,clock_source=internal,use_dpdk=1,second_addr=192.168.20.2,mgmt_addr=10.2.1.19'.
[INFO] [0/DmaFIFO_0] Initializing block control (NOC ID: 0xF1F0D00000000004)
[INFO] [0/DmaFIFO_0] BIST passed (Throughput: 1344 MB/s)
[INFO] [0/DmaFIFO_0] BIST passed (Throughput: 1341 MB/s)
[INFO] [0/DmaFIFO_0] BIST passed (Throughput: 1348 MB/s)
[INFO] [0/DmaFIFO_0] BIST passed (Throughput: 1347 MB/s)
[INFO] [0/Radio_0] Initializing block control (NOC ID: 0x12AD100000011312)
[INFO] [0/Radio_1] Initializing block control (NOC ID: 0x12AD100000011312)
[INFO] [0/DDC_0] Initializing block control (NOC ID: 0xDDC0000000000000)
[INFO] [0/DDC_1] Initializing block control (NOC ID: 0xDDC0000000000000)
[INFO] [0/DUC_0] Initializing block control (NOC ID: 0xD0C0000000000002)
[INFO] [0/DUC_1] Initializing block control (NOC ID: 0xD0C0000000000002)
Using Device: Single USRP:
  Device: N300-Series Device
  Mboard 0: ni-n3xx-313ABDA
  RX Channel: 0
    RX DSP: 0
    RX Dboard: A
    RX Subdev: Magnesium
  RX Channel: 1
    RX DSP: 0
    RX Dboard: B
    RX Subdev: Magnesium
  TX Channel: 0
    TX DSP: 0
    TX Dboard: A
    TX Subdev: Magnesium
  TX Channel: 1
    TX DSP: 0
    TX Dboard: B
    TX Subdev: Magnesium

[00:00:03.728707] Setting device timestamp to 0...
[INFO] [MULTI_USRP]     1) catch time transition at pps edge
[INFO] [MULTI_USRP]     2) set times next pps (synchronously)
[00:00:05.331920] Testing receive rate 125.000000 Msps on 2 channels
[00:00:05.610789] Testing transmit rate 125.000000 Msps on 2 channels
[00:00:15.878071] Benchmark complete.


Benchmark rate summary:
  Num received samples:     2557247854
  Num dropped samples:      0
  Num overruns detected:    0
  Num transmitted samples:  2504266704
  Num sequence errors (Tx): 0
  Num sequence errors (Rx): 0
  Num underruns detected:   0
  Num late commands:        0
  Num timeouts (Tx):        0
  Num timeouts (Rx):        0


Done!

Tuning Notes

General Host Performance Tuning App Note

Perform the general host performance tuning tips and tricks.

Increasing num_recv_frames

If you experience Overflows at higher data rates, adding the device argument num_recv_frames=### can help, where ### might be 500 for expected link data throughput of 10-20 Gbps and 3000 for 80 Gbps. There is no "correct" value, unlike, for example, the number samples-per-packet which is bound by the Ethernet packet size: 1500 Bytes for 1 GbE links and 9000 Bytes ("jumbo packet size") for 10 GbE and 100 GbE links. Higher values require more memory allocation, which for DPDK means higher dpdk_num_mbufs and sometimes dpdk_mbuf_cache_size and dpdk_num_desc, in the UHD DPDK configuration file.

Full Rate Streaming (UHD 3.x only)

If you're streaming data at the full master clock rate, and there is no interpolation or decimation being performed on the FPGA, you can skip the DUC and DDC blocks within the FPGA with the following parameters: 

   skip_ddc=1
   skip_duc=1

Full Rate on X3xx

If you're streaming two transmit channels at full rate (200e6) on the X3xx platform, you should additionally set the following device arg: 

   enable_tx_dual_eth=1

Isolate Cores/CPUs

NOTE: Isolation cores must be used along with a taskset or other method such that the UHD-based application uses the isolated cores.

Isolating the cores that are used for DPDK can improve performance. This can be done by adding the isolcpus parameter to your /etc/default/grub file in the GRUB_CMDLINE_LINUX_DEFAULT="" (the "GRUB_CONFIG" file). For example, to isolate cores 2 through and including 4, add this entry: 

   isolcpus=2-4

NOTE: After saving the GRUB_CONFIG file, execute sudo update-grub and then reboot the computer for the changes to take effect.

If hyperthreading is enabled for the cores being isolated, also isolate the hyperthreaded cores. Note that some CPUs -- such as the Intel i7 and i9 -- have both performance and efficiency cores, and generally only the performance cores can be hyperthreaded. Thus, for example for an 8 physical core CPU with all cores hyperthreaded -- cores 0-7 as the physical cores and 8-15 are the hyperthreaded virtual cores with core 0 paired with hyperthreaded core 8, 1 paired with 9, and so forth -- this entry would be: 

   isolcpus=2-4,10-12

DPDK handles core isolation during runtime for cores assigned to it, but it is OK to isolate cores used by DPDK via this method.

With whatever cores are listed as isolated in GRUB, the Linux thread scheduler will generally not schedule threads or interrupts to run on them. Using just isolcpus does not entirely isolate cores, nor does it prevent forced context switching on threads running on those cores; these are handled by other GRUB parameters and/or system settings. In Linux kernel 6, better -- but still not full -- isolation would include these entries (for the same cores noted above): 

   isolcpus=nohz,domain,managed_irq,2-4,10-12

When determining the number of cores to isolate for UHD, take into account the number dedicated for DPDK: 1 core per DPDK link as of UHD 4.11. Thus for the possible 8-core CPU, if using 2 DPDK links there are just 6 cores remaining and one (core 0) must be dedicated for the Linux kernel -- and it's wise to dedicate 2 cores for Linux (cores 0 and 1) so that the system does not lock. In this 8-core CPU scenario if using 2 DPDK links, only 4 cores would be available for UHD. This may be enough cores for your UHD-based application; only you know the use-case and can thus determine how many cores the CPU should have to support it.

A good rule of thumb for UHD-based applications is 1 core for the general application, 1 core per RX link, and 2 core per TX link. While not 1:1 thread:core, the actual aggregate thread CPU% used maps well.

If the CPU has enough cores to support 1 thread per core and your application allows for this level of core isolation, then both isolating and assigning these cores will truly utilize the CPU for maximum UHD benefit.

Disable System Interrupts on Cores/CPUs

NOTE: Disabling system interrupts should be used in conjunction with core isolation, which must be used along with a taskset or other method such that the UHD-based application uses the isolated cores.

Disabling system interrupts can improve the jitter and performance. This can be done by adding the parameters nohz_full and rcu_nocbs to your GRUB_CONFIG. For example, to disable the vast majority of system interrupts on cores 2 through and including 4 and their hyperthreaded relations, add this entry: 

    nohz_full=2-4,10-12 rcu_nocbs=2-4,10-12 rcu_nocb_poll

NOTE: After saving the GRUB_CONFIG file, execute sudo update-grub and then reboot the computer for the changes to take effect. It is OK to isolate core used by DPDK via this method.

NOTE: Adding these entries hugely reduces the interrupts being handled by the noted cores but do not prevent forced context switching to handle multiple threads per core.

Streaming on Multiple Channels using 1 Thread Per Stream

If you're streaming on multiple channels simultaneously, you can create multiple streamer objects on separate threads. For example, this can be accomplished with the benchmark_rate example by using the parameter --multi_streamer -- which assigns 1 channel per thread, whether TX or RX.

With your own UHD-based application, you will need to determine how many channels to assign to a given thread / stream. It is not necessarily true that 1 channel per thread will provide the best ratio of throughput to number of threads or cores in use.

Elevated Streaming Thread Priority

In UHD 4, streaming thread priorities can be elevated with the uhd::set_thread_priority_safe() function call. This can be accomplished with the benchmark_rate example by using parameter --priority high. Note that if the Thread Schedule Priority has not been enabled for the current user then this parameter requires sudo to work.

Extra nice Priority

Beyond elevating streaming thread priority, one can also increase the nice priority level to maximum to increase the amount of CPU time for the process and its thread by prepending the following to the command being issued: 

   sudo nice -n -20

Limit Execution CPUs/Cores

With Linux one can limit the cores being used by the threads in a process via a taskset prepended to a command being executed. When combining with the other techniques listed here, one can fairly well constrain a process and its thread to specific cores and give them maximum CPU time. Note that when using DPDK the MAC cores specified in uhd.conf will already be included as part of the taskset (but those cores will not be isolated nor have system interrupts disabled on them unless using that setting as noted above); it generally won't hurt to include the DPDK cores as part of the overall taskset but it's not required. For example, to limit process/thread execution to cores 2 through and including 4, one would prepend the following (note that sudo is not required): 

   taskset -c "2-4"

NOTE: For best performance do not include the cores used by DPDK in the dpdk_corelist argument in the taskset nor the hyperthreaded cores. The "DPDK control" core may be included in the taskset because it is not dedicated by DPDK for its sole use.

Stopping Extraneous Processes

The Linux kernel spawns a number of processes and threads that spend most of their time sleeping; one cannot stop these processes/threads. That said, in a typical Linux install (for example Ubuntu) there will be a graphical desktop (e.g., gdm) and various daemons started up by user request (e.g., containerd, docker, snapd). Most of these daemons can be controlled by systemctl; some must be manually quit. Any process that runs regularly stands a chance of interrupting the UHD and DPDK threads, and thus stopping, quitting, or disabling those processes can increase performance. For example, the following commands stop various daemons that execute regularly on Ubuntu; these need to be executed with sudo and some might not exist on your specific system and for these the command will do nothing so it's OK to run it: 

   systemctl stop containerd containerd.socket
   systemctl stop docker docker.socket
   systemctl stop dbus dbus.socket
   systemctl stop snapd snapd.socket
   systemctl stop udev systemd-udevd-control.socket systemd-udevd-kernel.socket

The following command stops the main Ubuntu desktop GUI, so running it will render the system accessible only via networking (e.g., ssh), so make sure network access is enabled before executing this command: 

   systemctl stop gdm

NOTE: Stopping extraneous process, like disabling hyperthreading, should be a last resort. Using thoughtful core isolation and core process assignments (e.g., "taskset") should work the vast majority of the time.

Disable Hyper-threading

In some applications which require the highest possible CPU performance per core, disabling hyper-threading can provide roughly a 10% increase in core performance, at the cost of having fewer core threads. Hyper-threading is disabled within the BIOS and how to do this varies by motherboard manufacturer. With other techniques listed here, disabling hyper-threading should only be done as a last resort to eek absolute maximum performance from the CPU.

Additional Tuning Notes from Intel

Known Issues / Troubleshooting

Regular Overflows on multi-CPU Systems

On Multi-CPU systems each CPU is a NUMA node. The NUMA node contains all of the cores on the CPU. For best performance when using CPU/core isolation ("isocpus" per the section Isolate Cores/CPUs above) make sure that all of the cores are in the same NUMA node.

Underruns Every Second with DPDK + Ubuntu

With Linux kernels 5.10 and beyond, we have observed periodic underruns on systems that otherwise have no issues. These Linux kernel versions are the default for Ubuntu 20.04.3 LTS and later. The underrun issue is due to the RT_RUNTIME_SHARE feature being disabled by default in these versions of the Linux kernel (shown as NO_RT_RUNTIME_SHARE). The following procedure can be used to enable this feature. This process was tested on Linux kernel version 5.13; the procedure may be slightly different on other kernel versions. To determine the Linux kernel version of your system, in a terminal issue the command uname -r. 

$ sudo -s
$ cd /sys/kernel/debug/sched
$ cat features | tr ' ' '\n' | grep RUNTIME_SHARE
NO_RT_RUNTIME_SHARE
$ echo RT_RUNTIME_SHARE > features
$ cat features | tr ' ' '\n' | grep RUNTIME_SHARE
RT_RUNTIME_SHARE
Retrieved from "https://kb.ettus.com/index.php?title=Getting_Started_with_DPDK_and_UHD&oldid=7231"