[Bug 270859] dmesg message: mce0: a looped back NS message is detected during DAD

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=270859

            Bug ID: 270859
           Summary: dmesg message: mce0: a looped back NS message is
                    detected during DAD
           Product: Base System
           Version: 13.2-RELEASE
          Hardware: Any
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: kern
          Assignee: bugs@FreeBSD.org
          Reporter: benoitc@enki-multimedia.eu

When I am using the mce driver with srv-io enabled I get the following message:

mce0: a looped back NS message is detected during DAD for
fe80:2::2c79:38ff:fe0d:f9e.  Another DAD probes are being sent.

wich is repeated until I am using the command line : 

 sudo sysctl net.inet6.ip6.dad_count=0

While the work around work I would really prefer to kepp this dupicate address
check. Is there anything that can be done? 

Looking at the dmesg log I can see it is happening when the interface goes
first down then up:

```
mlx5_core0: <mlx5_core> mem 0xc0000000-0xc00fffff at device 2.0 on pci0
mlx5: Mellanox Core driver 3.7.1 (November 2021)mlx5_core0: WARN:
mlx5_init_once:962:(pid 0): Unable to find vendor specific capabilities
mlx5_core0: WARN: mlx5_fwdump_prep:94:(pid 0): Unable to find vendor-specific
capability, error 2
mce0: Ethernet address: 2e:79:38:0d:0f:9e
mce0: link state changed to DOWN
Trying to mount root from zfs:zroot/ROOT/default []...
lo0: link state changed to UP
mce0: link state changed to UP
```

or another type of message:

```
mce0: Ethernet address: 9c:dc:71:4c:84:f0
mce0: link state changed to DOWN
mce1: Ethernet address: 9c:dc:71:4c:84:f1
mce1: link state changed to DOWN
mce0: link state changed to UP
mce0: a looped back NS message is detected during DAD for
fe80:4::9edc:71ff:fe4c:84f0.  Another DAD probes are being sent.
```

rc.conf is pretty straightforward for now:

```
hostname="myhost.domain.tld"
keymap="fr.macbook.kbd"
ifconfig_mce0="mtu 9000 up"
ifconfig_mce0_ipv6="inet6 .../64"
ipv6_defaultrouter="...::1"
sshd_enable="YES"
ntpd_enable="YES"
# Set dumpdev to "AUTO" to enable crash dumps, "NO" to disable
dumpdev="AUTO"
zfs_enable="YES"
bastille_enable="YES"
iovctl_files="/etc/iov/mce0.conf"
vm_enable="YES"
vm_dir="zfs:zdata/vms"
```

Some google research shows thatthis error is triggered when two ifconfig 
command on the interface are done quickly one after the other:
https://network.nvidia.com/pdf/prod_software/Mellanox_FreeBSD_Release_Notes_3.0.0.pdf

I'm using updated HPE rebranded card (so with their firmware)

-- 
You are receiving this mail because:
You are the assignee for the bug.