Bug Fixes in This Version

4564535

Description: Under high flow-insertion rates on BlueField-3, pipe resize operations may time out (exceeding 10 seconds). This can cause OVS threads (ovs_rcu) to block, indicating a stall in the hardware offload process.

Keyword: OVS; Pipe Resize; Timeout

Detected in version: 3.1.0

4564538

Description: A pipe resize operation on BlueField-3, often triggered by dynamic flow limits, may fail. This can result in memory allocation errors and severe delays (10+ seconds), leading to a loss of network connectivity, such as BGP peering sessions dropping.

Keyword: Pipe Resize; Flow Limit; Memory Allocation

Detected in version: 3.1.0

4621473

Description: A race condition during flow deletion may cause application instability or incorrect counter values.

Keyword: Flow; delete

Detected in version: 3.1.0

4366800

Description: Creating a packet filtering rule with match criteria on certain unsupported fields (e.g., specific VXLAN-GPE flags) does not return an error as expected.

Keyword: Packet filtering; DOCA Flow

Detected in version: 3.1.0

4627850

Description: The RTC timestamp is not correctly parsed in the doca_telemetry_diag sample.

Keyword: RTC; PTP; DOCA Telemetry

Detected in version: 3.1.0

4548383

Description: The doca_flow_header_ib_bth structure includes fields for which match and action operations are not supported. Supported fields:

• Match and action: opcode, dest_qp

• Match only: AckReq (in flags1)

• Action only: psn

Keyword: DOCA Flow

Detected in version: 3.1.0

4560966

Description: Enabling the RX Rate flag in the ZTR RTTCC algorithm causes compilation to fail because the private DOCA PCC device header is not included.

Keyword: DOCA PCC; Zero-touch RoCE; round-trip time congestion control

Detected in version: 3.1.0

4223392

Description: CT miss pipe receives configured IP version only.

Keyword: DOCA Flow

Detected in version: 3.1.0

4539156

Description: Fixed an issue on Debian/Ubuntu where DKMS kernel modules post-install scripts did not set autoinstall, causing modules to appear as "added" instead of "installed".

Keyword: DKMS kernel modules, Debian/Ubuntu, post-install scripts

Detected in version: 3.1.0

4548661

Description: Resolved startup warning in MFT caused by missing dimax-init component, which was not included in the EN+RoCE Bundle.

Keyword: MFT

Detected in version: 3.1.0

4578268

Description: Resolved missing kernel symbol dependencies in kmod-* packages by switching to DKMS-based modules in DOCA Host 3.2.0, ensuring proper module load on custom kernels.

Keyword: kmod-* packages, DKMS-based modules

Detected in version: 3.1.0

4648210

Description: Fixed an issue preventing DNF/YUM upgrades of rebuilt mlnx-ofa_kernel-devel packages (via doca-kernel-support) due to version sorting errors. The rebuilt package now appends ".1" and the kernel version to ensure proper RPM comparison and smooth upgrades.

Keyword: mlnx-ofa_kernel-devel package

Detected in version: 3.1.0

4651525

Description: Fixed an issue on ARM64 systems where the write-combining capability might not have been properly detected during driver initialization.

Now, the driver performs an internal check at boot time to correctly determine write-combining support, improving data path efficiency and traffic latency.

Keyword: Firmware, kernel lock

Detected in version: 3.1.0

4642754

Description: Fixed a non-responsive firmware issue that could cause the kernel to lock up for an extended period.

Keyword: Firmware, kernel lock

Detected in version: 3.1.0

4567695 / 4606693

Description: Fixed a rare crash that could occur when a MAD completion queue (CQ) was destroyed while RDMA CM traffic was still active.

Keyword: MAD CQ

Detected in version: 3.0.0

4222720

Description: Fixed an issue where, when running on DPU Arm cores, ptp4l could lose synchronization due to delayed transmit port timestamps, causing the tx_ts_timeout to expire and resulting in the error message: ptp4l[XXXX]: timed out while waiting for tx timestamp.

Keyword: DOCA Firefly Service

Detected in version: 3.1.0

4553499

Description: Fixed an issue where, on devices that do not support BlueFlame, allocation of a new Transport Domain (TD) could fail when attempting to allocate a dedicated UAR.

Keyword: BlueFlame, Transport Domain (TD), UAR

Detected in version: 3.0.0

4625955

Description: BFB installation via doca-installer fails if the process exceeds the hardcoded 45-minute timeout limit.

Keyword: DOCA installer; BFB installation; timeout

Detected in version: 4.13.0

4765027

Description: mlnx-sf fails to apply the -r parameter (disabling RoCE for SF) as the mlxdevm API changed.

Keyword: RoCE; SF

Detected in version: 4.13.0

4712344

Description: Automatic activation is not expected when performing a SimpleUpdate with "Stage":"true". To maintain backward compatibility with BMC FW 25.07 while using the new bf-upgrade behavior on the DPU side, set COMPATIBILITY_ACTIVATION=yes in bf.cfg.

Keyword: SimpleUpdate

Detected in version: 4.13.0

4654183

Description: After doca-installer installs the new BMC/CEC images, the "Post Upgrade (Pending)" has zero padding in the BMC version (e.g., BF-25.10-08). Therefore, it does not match the "Target Version" (BF-25.10-8).

Keywords: doca-installer

Detected in version: 4.13.0

4691172

Description: Swissbit eMMC firmware update may fail when using the default mmc-utils version that comes with Ubuntu 22.04, leading to potential update compatibility issues.

Keywords: eMMC; firmware

Detected in version: 4.13.0

4707555

Description: Restarting the openibd service causes the openvswitch service to stop. However, openvswitch does not automatically restart when openibd is brought back up.

Keywords: openibd; openvswitch

Detected in version: 4.13.0

4669177

Description: RShim command line will print "Timeout" when doing bfdump and Arm is unresponsive which causes bfdump to time out prematurely.

Keywords: RShim; dump

Detected in version: 4.13.0

4715798

Description: The BMC does not reset automatically when a DPU reset occurs to activate new BMC firmware after a deferred update.

Keywords: BMC; firmware

Detected in version: 4.13.0

4662191

Description: The BMC pending version may be displayed in a non-standard format (e.g., the version BF-24.10-36 may appear as 02410036).

Reported in version: 25.10

4658475

Description: The BMC pending version string generated by the BMC may include an unwanted leading zero in the last segment of the version (e.g., 00250804 is exposed as BF-25.08-04 instead of BF-25.08-4).

Reported in version: 25.10

4686938

Description: After a successful BFB update via SCP, subsequent updates may show incorrect progress messages (e.g., “Transfer progress: 100%”) that do not reflect the actual update status.

Reported in version: 25.10

4664471

Description: When creating a VLAN using ipmitool, the interface incorrectly appeared as a second entry in the DedicatedNetworkPorts collection. This results in redundant port entries showing identical LLDP data.

Reported in version: 25.10

4570205

Description: Fixed a firmware issue where the ZTR_RTTCC algorithm parameters AI and HAI did not support a sufficient range.

Keywords: PCC, ZTR_RTTCC

Detected in version: 32.46.1006

Fixed in Release: 32.47.1026

4650314

Description: The SBR hot reset error during mlxfwreset level 4 is resolved by skipping configuration of the second (disabled) link via mlxconfig. With only the active link configured, the reset flow completes successfully.

Keywords: mlxfwreset

Detected in version: 32.46.1006

Fixed in Release: 32.47.1026

4629077

Description: Fixed an issue where coalescing regular SX events with SX RTT events under ZTR_RTTCC could keep improper event fields, which could impact congestion control behavior.

Keywords: PCC, ZTR_RTTCC

Detected in version: 32.46.1006

Fixed in Release: 32.47.1026

4683328

Description: Fixed an issue in the ZTR_RTTCC algorithm where probe-abortion handling could behave improperly under high-stress network conditions, ensuring proper congestion control and stable traffic performance.

Keywords: PCC, ZTR_RTTCC

Detected in version: 32.46.1006

Fixed in Release: 32.47.1026

4501554

Description: Fixed an assertion failure that could occur with the E-Switch uplink in specific configurations where the e-switch was disabled and Path Migration was active or GVMIs were using SRQ loopback in SQs. The issue occurred because the firmware attempted to perform cleanup operations when the uplink configuration lacked sufficient capacity.

Now, when the E-Switch is disabled and no actions are available in the uplink STE, the firmware connects to the uplink STE instead of copying it.

Keywords: Path migration, steering

Detected in version: 32.46.1006

Fixed in Release: 32.47.1026

4506854

Description: Added Scaling Factor "read" field. To obtain correct values in mlxlink, MFT version 4.33.0 or later is required.

Keywords: Scaling Factor, mlxlink, MFT

Detected in version: 32.46.1006

Fixed in Release: 32.47.1026

4500365

Description: Fixed the adaptive PCIe gain setting.

Keywords: PCIe gain

Discovered in Version: 32.43.2402

Fixed in Release: 32.47.1026

4571079

Description: Fixed an issue where invoking the resourcedump tool with segment type DPA_PROCESS_LST returned invalid data when the parameter n1 == 1 and no processes existed on the current vhca_id.

The fix adds a proper check, and the resourcedump tool now reports the correct error in this scenario.

Keywords: DPA PROCESS, RESOURCE DUMP

Discovered in Version: 32.45.1020

Fixed in Release: 32.47.1026

4529293

Description: Fixed an issue where, during failover or restart, the SM sending a PortInfo MAD to the HCA firmware triggered reinitialization of port buffers, momentarily halting ingress traffic and causing packet drops.

The firmware now avoids reconfiguring port buffers when the new configuration matches the current one.

Keywords: OpenSM

Discovered in Version: 32.46.1006

Fixed in Release: 32.47.1026

4683346

Description: Fixed an issue where, under the ZTR_RTTCC algorithm, a flow that reached its minimum rate due to heavy congestion would not recover its rate once the congestion cleared.

Keywords: PCC, ZTR_RTTCC

Discovered in Version: 32.46.1006

Fixed in Release: 32.47.1026

4458670

Description: Fixed an issue where writing to nic_interface could set the parent function to DISABLED without propagating the state to its child VFs. The firmware now ensures that children inherit the DISABLED state correctly.

Keywords: VF health

Discovered in Version: 32.45.1020

Fixed in Release: 32.47.1026

4517037

Description: Fixed a rare issue where performing an FLR on a released ICM address from the DPA process could cause the DPA kernel to hang due to a race condition.

Keywords: FLR, ICM address

Discovered in Version: 32.46.1006

Fixed in Release: 32.47.1026

4618972

Description: Fixed an issue that could cause false-positive detection of DEAD_IRISC during DPA destroy_process or destroy_mem operations on highly loaded systems.

Keywords: IRISC

Discovered in Version: 32.43.2402

Fixed in Release: 32.47.1026

4133425

Description: Fixed an issue where PTP was not supported when the port speed was configured to 1G.

Keywords: PTP

Discovered in Version: 32.46.1006

Fixed in Release: 32.47.1026

4500507

Description: Fixed an issue where parallel execution of CREATE_NVME_SQ or CREATE_NVME_SQ_BE objects could lead to a fatal assert (0x8184).

Keywords: NVME, CREATE_NVME_SQ, CREATE_NVME_SQ_BE

Discovered in Version: 32.46.1006

Fixed in Release: 32.47.1026

4501554

Description: Fixed an assertion failure that could occur with the E-Switch uplink in specific configurations where the e-switch was disabled and Path Migration was active or GVMIs were using SRQ loopback in SQs. The issue occurred because the firmware attempted to perform cleanup operations when the uplink configuration lacked sufficient capacity.

Now, when the E-Switch is disabled and no actions are available in the uplink STE, the firmware connects to the uplink STE instead of copying it.

Keywords: Path migration, steering

Discovered in Version: 24.46.1006

Fixed in Release: 24.47.1026

4529293

Description: Fixed an issue where, during failover or restart, the SM sending a PortInfo MAD to the HCA firmware triggered reinitialization of port buffers, momentarily halting ingress traffic and causing packet drops.

The firmware now avoids reconfiguring port buffers when the new configuration matches the current one.

Keywords: OpenSM

Discovered in Version: 24.45.1020

Fixed in Release: 24.47.1026

4366117

Description: Configuring a small MTU leads to fragmentation of packets critical for the PXE boot process. As a result, the PXE boot filters mistakenly discard these packets, causing the PXE boot to fail.

Keywords: PXE boot filters

Discovered in Version: 24.45.1020

Fixed in Release: 24.47.1026