Bug Fixes History

Warning

This section includes history of bug fixes of 3 major releases back. For older releases history, please refer to the relevant firmware versions Release Notes inhttps://docs.mellanox.com/category/adapterfw.

Internal Ref.

Issue

2450264

Description: Fixed an issue that caused TX PRBS not to change after reconfiguring it. Now all PRBS mode are enabled in test mode.

Keywords: PRBS

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2603793

Description: Fixed an assert that was caused when trying to open 1024 functions on the device. The maximum number of functions is 1023.

Keywords: Max GVMI, sub-functions

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2648336

Description: Disabled the CNP counter “rp_cnp_ignored " (triggered by OOS (out-of-sequence)) when all ports are IB.

Note: For mixed IB/ETH scenario, the behavior depends on the RoCE configuration, the counter on the IB port may still increase but will not affect the regular use.

Keywords: CNP counter, IB

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2667272

Description: Fixed the TMP421 sensor temperature reporting.

Keywords: Sensor temperature

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2641734

Description: Fixed the rate select mechanism in QSFP modules.

Keywords: Cables

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2600783

Description: Fixed classification issues for "Passive" cables to be more robust.

Keywords: Cables

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2574322

Description: Fixed an issue that occasionally caused some performance issues related to RC QPs using E2E-credits (not connected to SRQ and doing send/receive traffic) when the ROCE_ACCL tx_window was enabled.

Keywords: Bandwidth, performance

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2391109

Description: Fixed an issue that caused a fatal error, and eventually resulted in the HCA hanging when a packet was larger than a strided receive WQE that was being scattered.

Keywords: Strided RQ, MTU

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2569999

Description: Fixed a rare issue that caused RX pipe to hang.

Keywords: RX pipe

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2621704

Description: Fixed the resource number size (a 64 bit number) to avoid a scenario where it overwrote it with a 32 bit number and erased the high bits when de-allocating the resource number.

In this scenario, when two resource numbers had identical low 32 bits, and because the high bits were cleared, it resulted in the same idx. Consequently, when two idxes were identical, then it freed the same idx twice.

Keywords: Resource number size, free_4k page

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2619161

Description: Initialized the rate table in the static configuration so it will be configured at the link-not-up scenarios.

Keywords: RoCE, static configuration, rate table

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2589430

Description: CRT_DCR with index larger than 1 << 21 can collide with the CRT_SW_RESERVED address.

Keywords: DCR

Discovered in Version: 16.30.1004

Fixed in Release: 16.31.1014

2684071

Description: Changing the default host chaining buffer size or WQE size (HOST_CHAINING_DESCRIPTORS, HOST_CHAINING_TOTAL_BUFFER_SIZE) using NVconfig might result in driver initialization failure.

Keywords: Host chaining

Discovered in Version: 16.29.2002

Fixed in Release: 16.31.1014

2565218

Description: Fixed an issue that caused the TX queue to hang when the VF rate limiter was set and it was leaded as NODNIC.

Keywords: NODNIC

Discovered in Version: 16.27.2008

Fixed in Release: 16.31.1014

2799269

Description: Tunnel Atomics is not functional when using UMR.

Keywords: UMR, Tunneled Atomic

Discovered in Version: 16.29.1016

Fixed in Release: 16.30.1004

2507096

Description: Removed the option to create unnecessary internal CNP operation for the Lossy ADP retransmission feature.

Keywords: RoCE, Lossy, Adp_retrans

Discovered in Version: 16.29.1016

Fixed in Release: 16.30.1004

2444837

Description: Set the cap to 0 for high index functions to avoid too many parallel VF NODNIC functions.

Keywords: NODNIC, VF, ETH PXE

Discovered in Version: 16.29.1016

Fixed in Release: 16.30.1004

2455041

Description: Fixed an issue that prevented PF from sending out packets. A new trigger (every ~1sec) was added to trigger the VQoS algorithm to run full iteration on all the VQoS tree.

Keywords: PF, packets, VQoS

Discovered in Version: 16.29.1016

Fixed in Release: 16.30.1004

2325629

Description: Occasionally, Tag Matching RNDV and NVME emulation may hang.

Keywords: Tag Matching

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2301590

Description: Congestion Control may not work properly if the card supports two ports and each PF for each port is not raised at the same time.

Keywords: Congestion Control

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2411542

Description: Multi-APP QoS is not supported when LAG is configured.

Keywords: Multi-APP QoS, LAG

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2403654

Description: Increased the default number of outstanding read bytes on the PCIe link for PCIe Gen4 devices when working in PCIe Gen3 servers. This will enable the NIC to maximize the PCIe link and achieve maximum bandwidth.

Keywords: PCIe, performance

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2339971

Description: Fixed an issue that prevented MCAM from reporting support for MFBA, MFBE, MFPA registry keys although they were available through the CMDIF interface.

Keywords: MCAM

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2410395

Description: Fixed an issue that prevented a SFP28 cable from linking up in a 25GbE speed.

Keywords: Cables

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2385117

Description: Modified the calculation of NUM_VF_MSIX to take into account NVME, Virtio Net/Blk, HotPlug PFs & VFs. Since max_total_msix is the maximum number used for all PFs and VFs (Port, NVME. Virtio Net/Blk, HotPlug), if there are not enough MSIX for all the devices, the number of port VF MSIX may be lowered (less than NUM_VF_MSIX) in order to not exceed the max_total_msix.

Note: In case of compatibility issues with ab old driver requiring more than 4 MSI-X, you should consider lowering number of PFs/VFs on any of the configurable functions (NVME. Virtio Net/Blk).

Keywords: MSIX

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2385117

Description: Added protection for decapsulated packets with invalid IP (bad length). Such packets are now discarded in the hardware.

Keywords: Decap ip_bad_length

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2392225

Description: When PER_PF_NUM_SF=1 (per PF configurations are used for SFs), if the number of SFs configured for a PF is 0 (PF_TOTAL_SF=0), than the firmware wrongly opens BAR2 with size 128KB.

Keywords: BAR2, Sub-functions, SF

Discovered in Version: 16.29.1016

Fixed in Release: 16.29.2002

2245422

Description: When MKEY_BY_NAME is enabled by NVCONFIG and a large number of VFs are configured, VM restart (VF/PF FLR) will take longer than when MKEY_BY_NAME is disabled

Keywords: SR-IOV

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

1911080

Description: Fixed a rare race condition that caused an erroneous write to the firmware image during certain power-down scenarios. This resulted in firmware being recognized as corrupted and prevented the adapter card from being recognized by the system due to missing valid Flash images.

Keywords: Flash images, firmware corruption

Discovered in Version: 16.24.1000

Fixed in Release: 16.29.1016

2360496

Description: Changed the default value of DCQCN's NP parameter min_time_between_cnps to 4 on all devices to support larger scalability of cluster.

Keywords: RoCE, Congestion control, DCQCN

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

2200824

Description: Fixed an issue that prevented VXLAN packets with svlan/cvlan tag from being matched.

Keywords: VXLAN

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

2355328

Description: Fixed an issue that caused the eth_wqe_too_small counter to count ODP page faults.

Keywords: Counters

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

2245422

Description: When MKEY_BY_NAME is enabled by NVCONFIG and a large number of VFs are configured, VM restart (VF/PF FLR) will take longer than when MKEY_BY_NAME is disabled.

Keywords: SR-IOV

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

2282225/2241765

Description: Fixed an issue that resulted in low performance after enabling the RoCE Accelerator capability.

Note: The fix is available when all ports are set as Ethernet.

Keywords: Performance, RoCE

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

2252559

Description: On rare cases, a fatal error related to errors from the PCI transport layer might be reported during FLR.

Keywords: FLR, PCI transport layer, errors

Discovered in Version: 16.26.1040

Fixed in Release: 16.29.1016

2127946

Description: Fixed the chassis manager calculation for Multi-Host and Socket-Direct adapter cards to allow running NC-SI commands by the chassis manager BMC. Now the chassis manager is count as BMC with index 0, regardless of how many BMC there are.

Keywords: Chassis manager, BMC

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

2204520

Description: When a cable module is plugged in but the Fiber is not connected, the link down counter may increase.

Keywords: Cables, link down

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

2321713

Description: Fixed an issue that caused caused the device to go to dead IRISC as one of the firmware semaphores could not be released when a speed change or port state change was triggered.

Keywords: IRISC, firmware semaphore,

Discovered in Version: 16.28.1002

Fixed in Release: 16.29.1016

1979562

Description: Fixed an issue that prevented the DHCP from assigning IPv6 address to the BMC during the initialization phase.

Keywords: DHCP, IPv6 address, BMC

Discovered in Version: 16.28.1002

Fixed in Release: 16.28.2006

2215104

Description: Updated the following Mellanox OEM NC-SI commands to fix an issue that caused the "Port swap" capability not to function properly:

  • Get Temperature

  • Get Module Serial Data

  • Set Module Serial Data

Keywords: Port swap

Discovered in Version: 16.28.1002

Fixed in Release: 16.28.2006

2080917

Description: Fixed and issue that resulted in driver startup failure when working in pass-through mode and dual port devices.

Keywords: Pass-through mode, dual port devices

Discovered in Version: 16.28.1002

Fixed in Release: 16.28.2006

2108543

Description: Enabled Bar configuration bytewise by applying the write_en bitmask.

Keywords: Bytewise BAR Programming

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2089896

Description: Fixed an issue in the UC traffic, that prevented out-of-sequence packets at a responder from being counted.

Keywords: Out-of-Sequence Counters

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2119975

Description: Fixed low PXE performance while using the VSC to trigger the send_ring_doorbells.

Keywords: NODNIC, DOORBELL

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2127535

Description: Updated multi_prio_sq is as following:

  • SmartNIC: ECPF only

  • Multi host/Single Host: support for PFs. VFs should also be enabled unless explicitly disabled by the PF

  • UID = 0 only

Keywords: multi_prio_sq

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2065624

Description: Fixed an issue related to counting CNP packets.

Keywords: CNP packets

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2149674

Description: Fixed an issue that caused packets to get stuck when the Rate Limiter was enabled.

Keywords: Rate Limiter

Discovered in Version: 16.27.6008

Fixed in Release: 16.28.1002

2100377

Description: Fixed a rare issue related to MCTP pass-through packet that caused the PCI boot sequence to fail if sent when the PCI link was not ready.

Keywords: MCTP pass-through packet

Discovered in Version: 16.27.1016

Fixed in Release: 16.28.1002

2181246

Description: Fixed an issue related to iRISC processor internal cache mechanisms that caused context corruption.

Keywords: iRISC processor

Discovered in Version: 16.27.1016

Fixed in Release: 16.28.1002

2108543

Description: Enabled Bar configuration bytewise by applying the write_en bitmask.

Keywords: Bytewise BAR Programming

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2058677

Description: In Socket Direct supported cards, after performing mlxfwreset, the expansion ROM register might be writable on all hosts for less than 1 second.

Keywords: Expansion ROM, Socket Direct

Discovered in Version: 16.27.1016

Fixed in Release: 16.28.1002

2089896

Description: Fixed an issue in the UC traffic, that prevented out-of-sequence packets at a responder from being counted.

Keywords: Out-of-Sequence Counters

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2119975

Description: Fixed low PXE performance while using the VSC to trigger the send_ring_doorbells.

Keywords: NODNIC, DOORBELL

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2127535

Description: Updated multi_prio_sq is as following:

  • SmartNIC: ECPF only

  • Multi host/Single Host: support for PFs. VFs should also be enabled unless explicitly disabled by the PF

  • UID = 0 only

Keywords: multi_prio_sq

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2065624

Description: Fixed an issue related to counting CNP packets.

Keywords: CNP packets

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2119135

Description: Fixed an issue that cause fragmented IP packets to drop.

Keywords: Fragmented IP packet

Discovered in Version: 16.27.2008

Fixed in Release: 16.28.1002

2165169

Description: Added the option to use the unicast MAC from the NC-SI cmd Set MAC Address to establish OS to BMC passthrough.

Keywords: OS to BMC passthrough

Discovered in Version: 16.27.1016

Fixed in Release: 16.28.1002

2169365

Description: Fixed an issue that caused PortCounters.PortRcvErr / PPCNT.infiniband_counters.PortRcvErr not to report port icrc errors.

Keywords: InfiniBand, ICRC, PortRcvErr, PortCounters

Discovered in Version: 16.27.1016

Fixed in Release: 16.28.1002

2136952

Description: Fixed an issue that caused PCI atomic over ETH to be posted as regular atomic operations.

Keywords: PCI Atomic, ETH

Discovered in Version: 16.27.4000

Fixed in Release: 16.28.1002

© Copyright 2023, NVIDIA. Last updated on May 23, 2023.