Kernel soft lockup occures in `fib6_select_path` function

[Omid]

We operate a Linux-based cluster of edge routers within a highly dynamic setting where BGP-advertised routes constantly evolve due to new next hop destinations frequently appearing and disappearing. In areas experiencing significant IPv6 traffic, we periodically face kernel soft lockups. Upon investigation, these lockups seem to occur during the traversal of the multipath circular linked-list, for finding the optimal path, in the fib6_select_path function, specifically while iterating through the siblings in the multipath linked-list. The problem typically arises when the linked list is unexpectedly deleted (its list-head reference count equals zero), leading to an infinite loop. This results in a soft lockup that triggers a system panic due to the watchdog timer.

It's worth noting that this is a longstanding issue we've been dealing with for about a year. Despite switching between different kernel versions, the problem persists.

Attached please find the `.crash` report generated by `apport-cli`.

Thanks,
Omid

ProblemType: Bug
DistroRelease: Ubuntu 22.04
Package: linux-image-6.5.0-26-generic 6.5.0-26.26~22.04.1
ProcVersionSignature: Ubuntu 6.5.0-26.26~22.04.1-generic 6.5.13
Uname: Linux 6.5.0-26-generic x86_64
ApportVersion: 2.20.11-0ubuntu82.5
Architecture: amd64
CasperMD5CheckResult: unknown
Date: Tue Apr 23 00:00:14 2024
ProcEnviron:
 TERM=xterm-256color
 PATH=(custom, no user)
 LANG=C.UTF-8
 SHELL=/bin/bash
SourcePackage: linux-signed-hwe-6.5
UpgradeStatus: No upgrade log present (probably fresh install)

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.