Kernel panic (page fault) on 10.3-STABLE with IB & VIMAGE modules

[Justin Clift]

On 21 Apr 2016, at 16:03, Justin Clift <justin at postgresql.org> wrote:
> On 21 Apr 2016, at 15:16, Justin Clift <justin at postgresql.org> wrote:
> <snip>
>> Have been hitting a kernel panic (page fault) with the IB modules loaded
>> on 10.3-STABLE.  (compiled multiple times over the last few days, all panicing)
>> 
>> Spent several hours narrowing down the cause, and it's definitely a bad
>> interaction between the IB modules (unsure which) + the "VIMAGE" module.
>> 
>> I'll fill out a bug report in a bit.
> 
> Btw, bug report filled out:
> 
>  https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=208957
> 
> VIMAGE problems with IB seems to be a large part of the reason why FreeNAS
> don't support Infiniband, and the NAS4Free project is struggling with it.
> 
> Hopefully we can get this resolved without it taking weeks/months, so
> getting IB functional for them can be made to happen. :)

Had some time to look at ddd quickly.  Looks very useful, as it shows the
IPoIB code where the trap seems to spring from.  (and yeah, kgdb can probably
show the same for a clueful person :>)

Added screenshots of the ddd session to the bug report.  Hoping it helps
get this figured out sooner. :)

* ipoib_ib_completion() - https://bugs.freebsd.org/bugzilla/attachment.cgi?id=169532
* ipoib_cm_handle_rx_wc() - https://bugs.freebsd.org/bugzilla/attachment.cgi?id=169533
* netisr_dispatch_src() - https://bugs.freebsd.org/bugzilla/attachment.cgi?id=169534
* ip_input() - https://bugs.freebsd.org/bugzilla/attachment.cgi?id=169535
* calltrap() - https://bugs.freebsd.org/bugzilla/attachment.cgi?id=169536

+ Justin

--
"My grandfather once told me that there are two kinds of people: those
who work and those who take the credit. He told me to try to be in the
first group; there was less competition there."
- Indira Gandhi