This technote describes steps to update a BlueField-2 DPU with the DOCA 2.2.0 firmware image. If you have a BF2-DPU running an older version (e.g. DOCA 1.2.0) and attempted to update the firmware with sdkmanager, you might end-up bricking the DPU. The newer releases require the BlueField mode to be set in the UEFI BIOS, and disable secure boot in order to flash the latest image correctly to an older BF2 DPU.
NVIDIA BlueField-2 DPU VPI - Ethernet Setup Ubuntu-22.04
Overview
This document describes how to setup an NVIDIA BlueField-2 DPU VPI network adapter and configure its ports for Ethernet.
Important:
DOCA 2.2.0 only supports NVIDIA® BlueField®-2 DPUs. For BlueField-3, please refer or upgrade to DOCA 2.2.1.
Procedure
Step 01.00: Configure host computer system.
Step 01.01: BIOS settings.
Configure the following BIOS settings for an Intel 12th Gen Z690-chipset based host computer system motherboard:
CSM support: Enabled.
PCI Express Slew Rate: Fast
Step 02.00: Install DPU hardware into a computer system.
Step 02.01: Install DPU hardware into a PCIe slot.
Install the DPU adapter into a spare PCIe slot.
Step 02.02: Verify DPU hardware detection.
Boot the system and check if the DPU is recognized:
lspci |grep -i nox
01:00.0 Ethernet controller: Mellanox Technologies MT42822 BlueField-2 integrated ConnectX-6 Dx network controller (rev 01)
01:00.1 Ethernet controller: Mellanox Technologies MT42822 BlueField-2 integrated ConnectX-6 Dx network controller (rev 01)
01:00.2 DMA controller: Mellanox Technologies MT42822 BlueField-2 SoC Management Interface (rev 01)
Step 03.00: Install DPU drivers.
You can download the latest DOCA drivers from the following website:
DOCA_WEBSITE: https://developer.nvidia.com/networking/doca
For instructions on how to install DOCA for BlueField-2 DPUs, see section Installing DOCA on BlueField-2 DPU
To install a new DOCA Runtime package, you must first remove the existing package.
For details, please refer to setcion “Upgrading BlueField DPU OS on Host Side” in the BlueField User Manual
(https://docs.mellanox.com/category/bluefieldsw).
sudo -s
# remove all doca packages
for f in $( dpkg --list | grep doca | awk '{print $2}' ); do echo $f ; apt remove --purge $f -y ; done
sudo apt autoremove
Step 03.01: Install DOCA runtime, sdk and tools.
export DOCA_URL=https://www.mellanox.com/downloads/DOCA/DOCA_v2.2.0
export DOCA_INSTALLER=doca-host-repo-ubuntu2204_2.2.0-0.0.3.2.2.0080.1.23.07.0.5.0.0_amd64.deb
# download
wget $DOCA_URL/$DOCA_INSTALLER
sudo dpkg -i $DOCA_INSTALLER
sudo apt update
sudo apt dist-upgrade
sudo apt install doca-runtime
sudo apt install doca-tools
sudo apt install doca-sdk
# install dpdk-doc
sudo apt install mlnx-dpdk-doc
# optional packages: rdnssd linux-image-generic openvswitch-datapath-module
# reboot the system
To remove the doca packages:
dpkg -r $(dpkg -f $DOCA_INSTALLER Package)
# example
sudo dpkg -r doca-host-repo-ubuntu2204
If you run into issues with broken packages, install synaptic to fix broken packages:
sudo apt install synaptic
# select a broken package and mark it for removal
Step 04.00: Configure network interfaces.
Step 04.01: Change Mellanox VPI ports from Infiniband to Ethernet
Mellanox makes three main types of cards: Ethernet only, Infiniband only, and VPI cards capable of both. You need the VPI versions and you may need to check a model number against a spec sheet to ensure you have a VPI capable card.
Swapping from Infiniband to Ethernet or back on a Mellanox ConnectX or DPU VPI card is simple.
First, we see what devices are installed.
# you must be root to use mst tool
sudo -s
# determine device-id
mst start
mst status -v
MST modules:
------------
MST PCI module is not loaded
MST PCI configuration module loaded
PCI devices:
------------
DEVICE_TYPE MST PCI RDMA NET NUMA
BlueField2(rev:1) /dev/mst/mt41686_pciconf0.1 01:00.1 mlx5_1 net-enp1s0f1np1 -1
BlueField2(rev:1) /dev/mst/mt41686_pciconf0 01:00.0 mlx5_0 net-enp1s0f0np0 -1
The device IDs for the BlueField-2 in this example is /dev/mst/mt41686_pciconf0
Display device configuration information:
# help
sudo mlxconfig -h
Examples:
To query configurations : mlxconfig -d /dev/mst/mt4099_pciconf0 query
To set configuration : mlxconfig -d /dev/mst/mt4099_pciconf0 set SRIOV_EN=1 NUM_OF_VFS=16 WOL_MAGIC_EN_P1=1
To set raw configuration : mlxconfig -d /dev/mst/mt4115_pciconf0 -f conf_file set_raw
To reset configuration : mlxconfig -d /dev/mst/mt4099_pciconf0 reset
# display informations about all configurations
sudo mlxconfig -d 01:00.0 i
# query device configuration
sudo mlxconfig -d 01:00.0 query
Device #1:
----------
Device type: BlueField2
Name: MBF2M516A-EECO_Ax_Bx
Description: BlueField-2 E-Series DPU 100GbE/EDR/HDR100 VPI Dual-Port QSFP56; PCIe Gen4 x16; Crypto and Secure Boot Enabled; 16GB on-board DDR; 1GbE OOB management; FHHL
Device: 01:00.0
Configurations: Next Boot
MEMIC_BAR_SIZE 0
MEMIC_SIZE_LIMIT _256KB(1)
HOST_CHAINING_MODE DISABLED(0)
HOST_CHAINING_CACHE_DISABLE False(0)
HOST_CHAINING_DESCRIPTORS Array[0..7]
HOST_CHAINING_TOTAL_BUFFER_SIZE Array[0..7]
INTERNAL_CPU_MODEL EMBEDDED_CPU(1)
FLEX_PARSER_PROFILE_ENABLE 0
FLEX_IPV4_OVER_VXLAN_PORT 0
ROCE_NEXT_PROTOCOL 254
ESWITCH_HAIRPIN_DESCRIPTORS Array[0..7]
ESWITCH_HAIRPIN_TOT_BUFFER_SIZE Array[0..7]
PF_BAR2_SIZE 3
PF_NUM_OF_VF_VALID False(0)
NON_PREFETCHABLE_PF_BAR False(0)
VF_VPD_ENABLE False(0)
PF_NUM_PF_MSIX_VALID False(0)
PER_PF_NUM_SF False(0)
STRICT_VF_MSIX_NUM False(0)
VF_NODNIC_ENABLE False(0)
NUM_PF_MSIX_VALID True(1)
NUM_OF_VFS 16
NUM_OF_PF 2
PF_BAR2_ENABLE True(1)
HIDE_PORT2_PF False(0)
SRIOV_EN True(1)
PF_LOG_BAR_SIZE 5
VF_LOG_BAR_SIZE 1
NUM_PF_MSIX 63
NUM_VF_MSIX 11
INT_LOG_MAX_PAYLOAD_SIZE AUTOMATIC(0)
PCIE_CREDIT_TOKEN_TIMEOUT 0
LAG_RESOURCE_ALLOCATION DEVICE_DEFAULT(0)
PHY_COUNT_LINK_UP_DELAY DELAY_NONE(0)
ACCURATE_TX_SCHEDULER False(0)
PARTIAL_RESET_EN False(0)
RESET_WITH_HOST_ON_ERRORS False(0)
NVME_EMULATION_ENABLE False(0)
NVME_EMULATION_NUM_VF 0
NVME_EMULATION_NUM_PF 1
NVME_EMULATION_VENDOR_ID 5555
NVME_EMULATION_DEVICE_ID 24577
NVME_EMULATION_CLASS_CODE 67586
NVME_EMULATION_REVISION_ID 0
NVME_EMULATION_SUBSYSTEM_VENDOR_ID 0
NVME_EMULATION_SUBSYSTEM_ID 0
NVME_EMULATION_NUM_MSIX 0
PCI_SWITCH_EMULATION_NUM_PORT 0
PCI_SWITCH_EMULATION_ENABLE False(0)
VIRTIO_NET_EMULATION_ENABLE False(0)
VIRTIO_NET_EMULATION_NUM_VF 0
VIRTIO_NET_EMULATION_NUM_PF 0
VIRTIO_NET_EMU_SUBSYSTEM_VENDOR_ID 6900
VIRTIO_NET_EMULATION_SUBSYSTEM_ID 1
VIRTIO_NET_EMULATION_NUM_MSIX 2
VIRTIO_BLK_EMULATION_ENABLE False(0)
VIRTIO_BLK_EMULATION_NUM_VF 0
VIRTIO_BLK_EMULATION_NUM_PF 0
VIRTIO_BLK_EMU_SUBSYSTEM_VENDOR_ID 6900
VIRTIO_BLK_EMULATION_SUBSYSTEM_ID 2
VIRTIO_BLK_EMULATION_NUM_MSIX 2
PCI_DOWNSTREAM_PORT_OWNER Array[0..15]
CQE_COMPRESSION BALANCED(0)
IP_OVER_VXLAN_EN False(0)
MKEY_BY_NAME False(0)
PRIO_TAG_REQUIRED_EN False(0)
UCTX_EN True(1)
REAL_TIME_CLOCK_ENABLE False(0)
RDMA_SELECTIVE_REPEAT_EN False(0)
PCI_ATOMIC_MODE PCI_ATOMIC_DISABLED_EXT_ATOMIC_ENABLED(0)
TUNNEL_ECN_COPY_DISABLE False(0)
LRO_LOG_TIMEOUT0 6
LRO_LOG_TIMEOUT1 7
LRO_LOG_TIMEOUT2 8
LRO_LOG_TIMEOUT3 13
LOG_TX_PSN_WINDOW 7
LOG_MAX_OUTSTANDING_WQE 7
TUNNEL_IP_PROTO_ENTROPY_DISABLE False(0)
ICM_CACHE_MODE DEVICE_DEFAULT(0)
TLS_OPTIMIZE False(0)
TX_SCHEDULER_BURST 0
ZERO_TOUCH_TUNING_ENABLE False(0)
ROCE_CC_LEGACY_DCQCN True(1)
LOG_DCR_HASH_TABLE_SIZE 11
DCR_LIFO_SIZE 16384
LINK_TYPE_P1 IB(1)
LINK_TYPE_P2 IB(1)
ROCE_CC_PRIO_MASK_P1 255
ROCE_CC_PRIO_MASK_P2 255
CLAMP_TGT_RATE_AFTER_TIME_INC_P1 True(1)
CLAMP_TGT_RATE_P1 False(0)
RPG_TIME_RESET_P1 300
RPG_BYTE_RESET_P1 32767
RPG_THRESHOLD_P1 1
RPG_MAX_RATE_P1 0
RPG_AI_RATE_P1 5
RPG_HAI_RATE_P1 50
RPG_GD_P1 11
RPG_MIN_DEC_FAC_P1 50
RPG_MIN_RATE_P1 1
RATE_TO_SET_ON_FIRST_CNP_P1 0
DCE_TCP_G_P1 1019
DCE_TCP_RTT_P1 1
RATE_REDUCE_MONITOR_PERIOD_P1 4
INITIAL_ALPHA_VALUE_P1 1023
MIN_TIME_BETWEEN_CNPS_P1 4
CNP_802P_PRIO_P1 6
CNP_DSCP_P1 48
CLAMP_TGT_RATE_AFTER_TIME_INC_P2 True(1)
CLAMP_TGT_RATE_P2 False(0)
RPG_TIME_RESET_P2 300
RPG_BYTE_RESET_P2 32767
RPG_THRESHOLD_P2 1
RPG_MAX_RATE_P2 0
RPG_AI_RATE_P2 5
RPG_HAI_RATE_P2 50
RPG_GD_P2 11
RPG_MIN_DEC_FAC_P2 50
RPG_MIN_RATE_P2 1
RATE_TO_SET_ON_FIRST_CNP_P2 0
DCE_TCP_G_P2 1019
DCE_TCP_RTT_P2 1
RATE_REDUCE_MONITOR_PERIOD_P2 4
INITIAL_ALPHA_VALUE_P2 1023
MIN_TIME_BETWEEN_CNPS_P2 4
CNP_802P_PRIO_P2 6
CNP_DSCP_P2 48
LLDP_NB_DCBX_P1 False(0)
LLDP_NB_RX_MODE_P1 OFF(0)
LLDP_NB_TX_MODE_P1 OFF(0)
LLDP_NB_DCBX_P2 False(0)
LLDP_NB_RX_MODE_P2 OFF(0)
LLDP_NB_TX_MODE_P2 OFF(0)
DCBX_IEEE_P1 True(1)
DCBX_CEE_P1 True(1)
DCBX_WILLING_P1 True(1)
DCBX_IEEE_P2 True(1)
DCBX_CEE_P2 True(1)
DCBX_WILLING_P2 True(1)
KEEP_ETH_LINK_UP_P1 True(1)
KEEP_IB_LINK_UP_P1 False(0)
KEEP_LINK_UP_ON_BOOT_P1 False(0)
KEEP_LINK_UP_ON_STANDBY_P1 False(0)
DO_NOT_CLEAR_PORT_STATS_P1 False(0)
AUTO_POWER_SAVE_LINK_DOWN_P1 False(0)
KEEP_ETH_LINK_UP_P2 True(1)
KEEP_IB_LINK_UP_P2 False(0)
KEEP_LINK_UP_ON_BOOT_P2 False(0)
KEEP_LINK_UP_ON_STANDBY_P2 False(0)
DO_NOT_CLEAR_PORT_STATS_P2 False(0)
AUTO_POWER_SAVE_LINK_DOWN_P2 False(0)
NUM_OF_VL_P1 _4_VLs(3)
NUM_OF_TC_P1 _8_TCs(0)
NUM_OF_PFC_P1 8
VL15_BUFFER_SIZE_P1 0
NUM_OF_VL_P2 _4_VLs(3)
NUM_OF_TC_P2 _8_TCs(0)
NUM_OF_PFC_P2 8
VL15_BUFFER_SIZE_P2 0
DUP_MAC_ACTION_P1 LAST_CFG(0)
MPFS_MC_LOOPBACK_DISABLE_P1 False(0)
MPFS_UC_LOOPBACK_DISABLE_P1 False(0)
UNKNOWN_UPLINK_MAC_FLOOD_P1 False(0)
SRIOV_IB_ROUTING_MODE_P1 LID(1)
IB_ROUTING_MODE_P1 LID(1)
DUP_MAC_ACTION_P2 LAST_CFG(0)
MPFS_MC_LOOPBACK_DISABLE_P2 False(0)
MPFS_UC_LOOPBACK_DISABLE_P2 False(0)
UNKNOWN_UPLINK_MAC_FLOOD_P2 False(0)
SRIOV_IB_ROUTING_MODE_P2 LID(1)
IB_ROUTING_MODE_P2 LID(1)
PF_TOTAL_SF 0
PF_SF_BAR_SIZE 0
PF_NUM_PF_MSIX 63
ROCE_CONTROL ROCE_ENABLE(2)
PCI_WR_ORDERING per_mkey(0)
MULTI_PORT_VHCA_EN False(0)
PORT_OWNER True(1)
ALLOW_RD_COUNTERS True(1)
RENEG_ON_CHANGE True(1)
TRACER_ENABLE True(1)
IP_VER IPv4(0)
BOOT_UNDI_NETWORK_WAIT 0
UEFI_HII_EN True(1)
BOOT_DBG_LOG False(0)
UEFI_LOGS DISABLED(0)
BOOT_VLAN 1
LEGACY_BOOT_PROTOCOL PXE(1)
BOOT_RETRY_CNT NONE(0)
BOOT_INTERRUPT_DIS False(0)
BOOT_LACP_DIS True(1)
BOOT_VLAN_EN False(0)
BOOT_PKEY 0
P2P_ORDERING_MODE DEVICE_DEFAULT(0)
EXP_ROM_VIRTIO_NET_PXE_ENABLE True(1)
EXP_ROM_VIRTIO_NET_UEFI_x86_ENABLE True(1)
EXP_ROM_VIRTIO_BLK_UEFI_x86_ENABLE True(1)
EXP_ROM_NVME_UEFI_x86_ENABLE True(1)
ATS_ENABLED False(0)
EXP_ROM_UEFI_ARM_ENABLE True(1)
EXP_ROM_UEFI_x86_ENABLE True(1)
EXP_ROM_PXE_ENABLE True(1)
ADVANCED_PCI_SETTINGS False(0)
SAFE_MODE_THRESHOLD 10
SAFE_MODE_ENABLE True(1)
Here, we notice that the link type for both ports are set to InfiniBand.
LINK_TYPE_P1 IB(1)
LINK_TYPE_P2 IB(1)
We can then select the port configuration. Here, 1 is for InfiniBand and 2 is for Ethernet:
sudo mlxconfig -d 01:00.0 set LINK_TYPE_P1=2 LINK_TYPE_P2=2
Device #1:
----------
Device type: BlueField2
Name: MBF2M516A-EECO_Ax_Bx
Description: BlueField-2 E-Series DPU 100GbE/EDR/HDR100 VPI Dual-Port QSFP56; PCIe Gen4 x16; Crypto and Secure Boot Enabled; 16GB on-board DDR; 1GbE OOB management; FHHL
Device: 01:00.0
Configurations: Next Boot New
LINK_TYPE_P1 IB(1) ETH(2)
LINK_TYPE_P2 IB(1) ETH(2)
Apply new Configuration? (y/n) [n] : y
Applying... Done!
-I- Please reboot machine to load new configurations.
Reboot the system for the changes to take effect.
To check which mode a SmartNIC is running on, use the following command in the host:
$ mst start
$ mst status -v # identify the MST device
$ mlxconfig -d /dev/mst/mt41686_pciconf0 q | grep -i internal_cpu_model
INTERNAL_CPU_MODEL EMBEDDED_CPU(1)
Step 04.02: Configure IP address for network interfaces.
sudo -s
cd /etc/netplan
nano bf_config.yaml
Edit the file by typing nano bf_config.yaml:
# This file describes the network interfaces available on your system
# For more information, see netplan(5).
# This file describes the network interfaces available on your system
# For more information, see netplan(5).
# This file describes the network interfaces available on your system
# For more information, see netplan(5).
network:
version: 2
renderer: networkd
ethernets:
tmfifo_net0:
addresses: [192.168.100.1/24]
dhcp4: no
enp1s0f0np0:
mtu: 9000
addresses: [192.168.200.80/24]
dhcp4: no
enp1s0f1np1:
mtu: 9000
addresses: [192.168.200.81/24]
dhcp4: no
Modify the existing 01-network-manager-all.yaml
# Let NetworkManager manage all devices on this system
network:
version: 2
renderer: NetworkManager
to the following configuration for eth0 and eth1 :
# Let NetworkManager manage all devices on this system
network:
version: 2
renderer: networkd
ethernets:
eth0:
dhcp4: no
addresses: [192.168.200.127/24]
eth1:
dhcp4: yes
nameservers:
addresses: [192.168.1.10]
Restart the network interface:
sudo netplan apply
Verify that Rshim is active:
sudo systemctl status rshim
â—Ź rshim.service - rshim driver for BlueField SoC
Loaded: loaded (/lib/systemd/system/rshim.service; enabled; vendor preset: enabled)
Active: active (running) since Sun 2023-11-05 09:35:40 +04; 7min ago
Docs: man:rshim(8)
Main PID: 12153 (rshim)
Tasks: 7 (limit: 154244)
Memory: 1.4M
CPU: 11.663s
CGroup: /system.slice/rshim.service
└─12153 /usr/sbin/rshim
Nov 05 09:35:40 zeus rshim[12153]: Probing pcie-0000:01:00.2(vfio)
Nov 05 09:35:40 zeus rshim[12153]: Create rshim pcie-0000:01:00.2
Nov 05 09:35:40 zeus rshim[12153]: Failed to enable INTx
Nov 05 09:35:40 zeus rshim[12153]: rshim pcie-0000:01:00.2 enable
Nov 05 09:35:41 zeus rshim[12153]: rshim0 attached
Nov 05 09:35:41 zeus rshim[12153]: USB device detected
Nov 05 09:35:45 zeus rshim[12153]: Probing usb-1-b.3
Nov 05 09:35:45 zeus rshim[12153]: create rshim usb-1-b.3
Nov 05 09:35:45 zeus rshim[12153]: another backend already attached
Nov 05 09:35:45 zeus rshim[12153]: rshim usb-1-b.3 deleted
This command is expected to display active (running). If RShim service does not launch automatically, run:
sudo systemctl enable rshim
sudo systemctl start rshim
Step 04.03: Configure VPI Ethernet interface
Assign an IP address and bring the ethernet network interfaces up:
sudo ifconfig enp2s0f0 192.168.253.80 netmask 255.255.255.0 up
sudo ifconfig enp2s0f1 192.168.253.81 netmask 255.255.255.0 up
Check the network link status:
ibstatus
Infiniband device 'mlx5_0' port 1 status:
default gid: fe80:0000:0000:0000:0ac0:ebff:fe88:1924
base lid: 0x0
sm lid: 0x0
state: 4: ACTIVE
phys state: 5: LinkUp
rate: 100 Gb/sec (4X EDR)
link_layer: Ethernet
Infiniband device 'mlx5_1' port 1 status:
default gid: fe80:0000:0000:0000:0ac0:ebff:fe88:1925
base lid: 0x0
sm lid: 0x0
state: 4: ACTIVE
phys state: 5: LinkUp
rate: 100 Gb/sec (4X EDR)
link_layer: Ethernet
Step 04.04: Ensure that the BlueField Mode is correctly set in the UEFI configuration.
Reset the DPU
sudo -s
echo SW_RESET 1 > /dev/rshim0/misc
You can monitor the progress of the installation by typing the following command in a separate terminal:
# to interact with the console
sudo minicom --color on --baudrate 115200 --device /dev/rshim0/console
# to monitor only
cat /dev/rshim0/console 115200
Interrupt the boot process by pressing the ESC key twice and logging into the UEFI menu. The default password will be bluefield. Change it to a different password when logging into the UEFI settings the first time.
After that, ensure that you set the BlueField Mode to a valid setting:
Internal CPU Model: <Embedded>
Host Privilege Level: <Privileged>
Please note that Unavailable is an invalid setting and the BlueField-2 DPU wil not boot.
You will only be able to update the BlueField-2 DPU image if the BlueField Mode is set to a valid configuration.
Step 05.00: Update the BlueField-2 DPU image.
Step 05.01: Download the NVIDIA SDK Manager on the host computer.
Download the NVIDIA SDK Manager and install it on the host computer.
sudo apt install ./sdkmanager_2.0.0-11402_amd64.deb
Step 05.02: Use the NVIDIA SDK Manager to update the BlueField-2 DPU image.
Launch the SDK Manager
sdkmanager
Install the DOCA SDK with the SDK Manager
When SDK manager prompts you to flash the BlueField-2 InfiniBand/VPI DPU, set the initial user details for the DOCA OS:
Username: ubuntu
New Password: ubuntu
Step 05.03: Manually update the BlueField-2 DPU image.
Ubuntu users are prompted to change the default password (ubuntu) for the default user (ubuntu) upon first login. Logging in will not be possible even if the login prompt appears until all services are up (“DPU is ready” message appears in /dev/rshim0/misc).
Ubuntu users can provide a unique password that will be applied at the end of the BFB installation by creating a bf.cfg file which changes the default credentials. The password for the ubuntu user will be defined in the bf.cfg configuration file.
Create password hash. Run:
# openssl passwd -1
Password:
Verifying - Password:
$1$3B0RIrfX$TlHry93NFUJzg3Nya00rE1
Add the password hash in quotes to the bf.cfg file:
nano bf.cfg
ubuntu_PASSWORD='$1$3B0RIrfX$TlHry93NFUJzg3Nya00rE1'
If the Target OS image fails to update automatically using the SDK manager, type the following command to update it manually.
The bf.cfg file will be used with the bfb-install script in the following step.
Install the BFB.
Installing the BFB does not update the firmware. To do that, refer to section Firmware Upgrade.
sudo -s
bfb-install --bfb ./DOCA_2.2.0_BSP_4.2.0_Ubuntu_22.04-2.23-07.prod.bfb --config bf.cfg --rshim rshim0
You should see the following message on the console. It will take approximately 15 minutes to write the new firmware image:
Pushing bfb
1.11GiB 0:01:34 [12.0MiB/s] [ <=> ]
Collecting BlueField booting status. Press Ctrl+C to stop…
INFO[BL2]: start
INFO[BL2]: DDR POST passed
INFO[BL2]: UEFI loaded
INFO[BL31]: start
INFO[BL31]: lifecycle GA Secured
INFO[BL31]: runtime
INFO[UEFI]: UPVS valid
WARN[UEFI]: UPVS full
WARN[UEFI]: UPVS reclaim
INFO[UEFI]: eMMC init
INFO[UEFI]: eMMC probed
WARN[UEFI]: Var reclaim
WARN[UEFI]: Var reclaim done
INFO[UEFI]: PMI: updates started
INFO[UEFI]: PMI: total updates: 1
INFO[UEFI]: PMI: updates completed, status 0
INFO[UEFI]: PCIe enum start
INFO[UEFI]: PCIe enum end
INFO[UEFI]: exit Boot Service
INFO[MISC]: Ubuntu installation started
INFO[MISC]: Installing OS image
INFO[MISC]: Installation finished
Monitor the progress of the firmware update by connecting to the DPU console.
You should see the following messages, if the DPU is being programmed correctly:
Last login: Tue Nov 7 02:48:40 UTC 2023 on hvc0
ubuntu@localhost:~$ echo "nameserver 8.8.8.8" | sudo tee /etc/resolv.conf
nameserver 8.8.8.8
[PMI] Boot image update started.
Trusted Board Boot is enabled (GA)
Note: Installed image will be filtered from 5031688 bytes to 3155832 bytes
Size check good, 3155832 <= 33030144
Verify Trusted Board Boot FW certificate.
Check boot image, Status: Success
...Filtering unneeded executables...
...Preparing boot image...
...Generating boot image...
...Writing the boot image, Size: 3155968 bytes...
...Verify boot image...
Write boot image to partition 1, Status: Success
...Filtering unneeded executables...
...Preparing boot image...
...Generating boot image...
...Writing the boot image, Size: 3155968 bytes...
...Verify boot image...
Write boot image to partition 2, Status: Success
[PMI] Boot Image update completed, Status: Success
[PMI] Total number of updates: 1
[PMI] Errors during updates : 0
DHCP Session Start
Press ESC/F2/DEL twice to enter UEFI Menu.
Press ENTER to skip countdown.
3 seconds remain...
2 seconds remain...
1 seconds remain...
0 seconds remain...
EFI stub: Booting Linux Kernel...
EFI stub: Generating empty DTB
EFI stub: Loaded initrd from command line option
EFI stub: Exiting boot services[ 8.760619] mlxbf2_gpio MLNXBF22:01: IRQ index 0 not found
[ 8.782585] Synopsys Designware Multimedia Card Interface Driver
[ 8.782583] mlxbf2_gpio MLNXBF22:02: IRQ index 0 not found
[ 8.784488] mlxfw: loading out-of-tree module taints kernel.
[ 8.784577] mlxfw: loading out-of-tree module taints kernel.
[ 8.828809] dw_mmc PRP0001:00: IDMAC supports 64-bit address mode.
[ 8.834332] Micrel KSZ9031 Gigabit PHY MLNXBF17:00:03: attached PHY driver (mii_bus:phy_addr=MLNXBF17:00:03, irq=74)
[ 8.841304] dw_mmc PRP0001:00: Using internal DMA controller.
[ 8.874029] dw_mmc PRP0001:00: Version ID is 270a
[ 8.874366] Compat-mlnx-ofed backport release: 4e13edc
[ 8.883525] dw_mmc PRP0001:00: DW MMC controller at irq 22,32 bit host data width,256 deep fifo
[ 8.893814] Backport based on mlnx_ofed/mlnx-ofa_kernel-4.0.git 4e13edc
[ 8.923088] mmc_host mmc0: Bus speed (slot 0) = 50000000Hz (slot req 400000Hz, actual 396825HZ div = 63)
[ 8.924616] compat.git: mlnx_ofed/mlnx-ofa_kernel-4.0.git
[ 8.978887] mlxbf_gige MLNXBF17:00 oob_net0: renamed from eth0
[ 9.033847] mmc_host mmc0: Bus speed (slot 0) = 50000000Hz (slot req 50000000Hz, actual 50000000HZ div = 0)
[ 9.041535] mlx5_core 0000:03:00.0: firmware version: 24.38.1002
[ 9.053820] mmc0: new high speed MMC card at address 0001
[ 9.065722] mlx5_core 0000:03:00.0: 252.048 Gb/s available PCIe bandwidth (16.0 GT/s PCIe x16 link)
[ 9.070733] virtio_net virtio1 tmfifo_net0: renamed from eth0
[ 9.077033] mmcblk0: mmc0:0001 S0J58X 59.3 GiB
[ 9.121232] mmcblk0: p1 p2
[ 9.127384] mmcblk0boot0: mmc0:0001 S0J58X 31.5 MiB
[ 9.138536] mmcblk0boot1: mmc0:0001 S0J58X 31.5 MiB
[ 9.590981] mlx5_core 0000:03:00.0: Rate limit: 127 rates are supported, range: 0Mbps to 97656Mbps
[ 9.610630] mlx5_core 0000:03:00.0: E-Switch: Total vports 83, per vport: max uc(128) max mc(2048)
[ 9.638455] mlx5_core 0000:03:00.0: Port module event: module 0, Cable plugged
[ 9.653141] mlx5_core 0000:03:00.0: mlx5_pcie_event:295:(pid 131): PCIe slot power capability was not advertised.
[ 9.672586] mlx5_core 0000:03:00.0: mlx5e: IPSec ESP acceleration enabled
[ 9.687819] mlx5_core 0000:03:00.0: MLX5E: StrdRq(1) RqSz(8) StrdSz(2048) RxCqeCmprss(0 enhanced)
[ 9.834169] mlx5_core 0000:03:00.1: firmware version: 24.38.1002
[ 9.846334] mlx5_core 0000:03:00.1: 252.048 Gb/s available PCIe bandwidth (16.0 GT/s PCIe x16 link)
[ 10.363913] mlx5_core 0000:03:00.1: Rate limit: 127 rates are supported, range: 0Mbps to 97656Mbps
[ 10.383537] mlx5_core 0000:03:00.1: E-Switch: Total vports 83, per vport: max uc(128) max mc(2048)
[ 10.411552] mlx5_core 0000:03:00.1: Port module event: module 1, Cable unplugged
[ 10.426573] mlx5_core 0000:03:00.1: mlx5_pcie_event:295:(pid 147): PCIe slot power capability was not advertised.
[ 10.444488] mlx5_core 0000:03:00.1: mlx5e: IPSec ESP acceleration enabled
[ 10.461142] mlx5_core 0000:03:00.1: MLX5E: StrdRq(1) RqSz(8) StrdSz(2048) RxCqeCmprss(0 enhanced)
[ 10.679120] mlx5_core 0000:03:00.1 p1: renamed from eth1
[ 10.719535] mlx5_core 0000:03:00.0 p0: renamed from eth0
[ 12.035046] raid6: neonx8 gen() 5608 MB/s
[ 12.111046] raid6: neonx8 xor() 4129 MB/s
[ 12.187045] raid6: neonx4 gen() 5688 MB/s
[ 12.263048] raid6: neonx4 xor() 4266 MB/s
[ 12.339045] raid6: neonx2 gen() 5030 MB/s
[ 12.415047] raid6: neonx2 xor() 3924 MB/s
[ 12.491045] raid6: neonx1 gen() 3910 MB/s
[ 12.567045] raid6: neonx1 xor() 3208 MB/s
[ 12.643050] raid6: int64x8 gen() 3012 MB/s
[ 12.719049] raid6: int64x8 xor() 1841 MB/s
[ 12.795049] raid6: int64x4 gen() 3331 MB/s
[ 12.871046] raid6: int64x4 xor() 1930 MB/s
[ 12.947045] raid6: int64x2 gen() 3146 MB/s
[ 13.023050] raid6: int64x2 xor() 1715 MB/s
[ 13.099048] raid6: int64x1 gen() 2442 MB/s
[ 13.175050] raid6: int64x1 xor() 1277 MB/s
[ 13.183622] raid6: using algorithm neonx4 gen() 5688 MB/s
[ 13.194467] raid6: .... xor() 4266 MB/s, rmw enabled
[ 13.204437] raid6: using neon recovery algorithm
[ 13.215137] xor: measuring software checksum speed
[ 13.226027] 8regs : 8108 MB/sec
[ 13.235803] 32regs : 9643 MB/sec
[ 13.246018] arm64_neon : 6708 MB/sec
[ 13.254766] xor: using function: 32regs (9643 MB/sec)
[ 13.265730] async_tx: api initialized (async)
[ 13.318986] sdhci: Secure Digital Host Controller Interface driver
[ 13.331465] sdhci: Copyright(c) Pierre Ossman
[ 13.341170] sdhci-pltfm: SDHCI platform and OF driver helper
[ 13.387288] Mellanox boot control driver (version 1.5)
[ 13.402265] sbsa-gwdt sbsa-gwdt.0: Initialized with 10s timeout @ 200000000 Hz, action=0.
[ 13.433619]
[ 13.438595]
[ 13.443644]
[17:44:48] INFO: Ubuntu installation started
[ 16.526123] mmcblk0: p1 p2
[ 17.544102] mmcblk0: p1 p2
[17:44:52] Installing OS image
[ 16.526123] mmcblk0: p1 p2
[ 17.544102] mmcblk0: p1 p2
[17:44:52] Installing OS image
[ 89.879281] mmcblk0: p1 p2
[ 89.897485] EXT4-fs (mmcblk0p2): mounted filesystem with ordered data mode. Opts: (null). Quota mode: none.
[ 325.641223] mmcblk0: p1 p2
[ 329.177936] mmcblk0: p1 p2
[ 332.275054] mmcblk0: p1 p2
[17:50:10] INFO: Installation finished
[17:50:13] INFO: Rebooting...
[ 341.433633]
[ 341.438486]
[ 341.443377]
[ 341.448245]
[ 341.452973] kvm: exiting hardware virtualization
[ 341.476309] mlx5_core 0000:03:00.1: Shutdown was called
[ 341.490135] mlx5_core 0000:03:00.0: Shutdown was called
Step 05.04: Enable NAT for the DPU to connect to the Internet
In lab configuration, only eth1 interface in the host is connected to the Internet. Hence, SmartNIC Linux needs to go through the host Linux to access the Internet. This post well explains how to setup NAT between them, so I just summarized the way here.
On the host
sudo -s
ip addr add 192.168.100.1/24 dev tmfifo_net0
echo 1 | sudo tee /proc/sys/net/ipv4/ip_forward
iptables -t nat -A POSTROUTING -o eth1 -j MASQUERADE
In case IP forwarding is not working
iptables -A FORWARD -o eth1 -j ACCEPT
iptables -A FORWARD -m state --state ESTABLISHED,RELATED -i eth1 -j ACCEPT
On the SmartNIC
echo "nameserver 8.8.8.8" | sudo tee /etc/resolv.conf
On the SmartNIC
$ ping google.com -c 2
PING google.com (74.125.21.100) 56(84) bytes of data.
64 bytes from yv-in-f100.1e100.net (74.125.21.100): icmp_seq=1 ttl=108 time=7.19 ms
64 bytes from yv-in-f100.1e100.net (74.125.21.100): icmp_seq=2 ttl=108 time=6.93 ms
Step 05.05: Manually update the target image firmware.
The other endpoint of tmfifo_net0 is in the BlueField-2 DPU, and its IPv4 address is 192.168.100.2/24, hence you can use ssh to access BlueField-2 DPU via this address.
First, we need to assign an address.
ip addr add 192.168.100.1/24 dev tmfifo_net0
ping 192.168.100.2 # check whether it is properly configured
# login to the bluefield-2 dpu
ssh ubuntu@192.168.100.2
# enter the password
ubuntu@192.168.100.2 password: ubuntu
# update the password
# update the target os image
sudo apt update; sudo apt upgrade;
Upgrade the firmware on the DPU, run:
sudo /opt/mellanox/mlnx-fw-updater/mlnx_fw_updater.pl --force-fw-update
Do not run a command to upgrade the target OS release. This will brick the DPU.
# warning: do not run this command !!!
# sudo do-release-upgrade
Downloads
User Guide
-
NVIDIA BlueField-2 InfiniBand/Ethernet DPU User Guide - Hardware Installation
-
NVIDIA BlueField-2 Ethernet DPU User Guide - Hardware Installation
Technotes
-
Changing Mellanox ConnectX VPI Ports to Ethernet or InfiniBand in Linux - STH - 20190407
-
15 Useful “ifconfig” Commands to Configure Network Interface in Linux
-
How to Enable (UP)/Disable (DOWN) Network Interface Port (NIC) in Linux?
-
Configuring NVIDIA BlueField2 SmartNIC - Insu Jang - 20220106