QMI IPv6 crashes

[Bjørn Mork]

Sorry, but it looks like I'll be unable to test this today. After surviving all the crazy skiing last week, I ended up with a broken shoulder after a motorcycle accident on my way to work today. Well, well...


Bjørn

On 17 April 2014 18:29:14 CEST, "Bjørn Mork" <bjorn at mork.no> wrote:
>Sorry for the top posting. It's just too difficult to do proper quoting
>on this android thing...
>
>In good Norwegian tradition I'm up in the mountains for easter, trying
>to find the last remains of winter now that spring is finally here :-)
>
>But I'll test the branch on my ipv6 capable qmi devices (MC7710 and
>E392) when I'm back in Tuesday. 
>
>As for the bug, that looks like it's failing on a ND packet for ::?
>Don't remember if we send something like that. I guess we should not,
>since qmi devices appear as real Ethernet devices, if we ignore the
>various firmware bugs. But maybe a USB and/or IP packet snoop could
>reveal something suspicious?
>
>
>Bjørn
>
>On 16 April 2014 18:54:24 CEST, Dan Williams <dcbw at redhat.com> wrote:
>>Hi,
>>
>>So I implemented QMI IPv6 stuff in the dcbw/icera-ipv6 branch first as
>>DHCP-based, and second as static using GetCurrentSettings.  Those
>parts
>>work fine.  But like before, with my Sierra MC7750, immediately after
>>NetworkManager sets the IP addresses on the interface, it crashes. 
>>I've
>>observed this with other QMI devices too, so the MC7750 is not unusual
>>here.  However, all these devices work fine on Verizon where IPv6 is
>>normally used under Windows.
>>
>>The Sierra devices include some additional diagnostic commands, so
>I've
>>captured that and the culprit is:
>>
>>Src: FatalError
>>Dat: 00000DFE
>>Str: ps_icmp6_nd.c
>>Fmt: Can't find the ND control block
>>00000000 00000000 00000000 00000000
>>Task: DS
>>Time: 0000CCE3
>>
>>ps_icmp6_nd.c is part of the AMSS sources of the firmware running on
>>the
>>modem.  I can provide more of the traceback if needed.
>>
>>Ben, do you have an Qualcomm contacts that you could ping about this
>>issue?  Or does anyone have any further ideas to isolate what the host
>>is doing that the firmware doesn't like?
>>
>>Could somebody else test the dcbw/icera-ipv6 branch with a QMI device
>>that supports IPv6 and see if they also have crashes?
>>
>>Dan
>>
>>
>>_______________________________________________
>>ModemManager-devel mailing list
>>ModemManager-devel at lists.freedesktop.org
>>http://lists.freedesktop.org/mailman/listinfo/modemmanager-devel
>
>_______________________________________________
>ModemManager-devel mailing list
>ModemManager-devel at lists.freedesktop.org
>http://lists.freedesktop.org/mailman/listinfo/modemmanager-devel