[openib-general] Segmentation fault when calling osmv_get_query_svc_rec()

[Steven Wooding]

Hi Hal,

Those changes did not do the trick, I'm afraid.

Cheers,

Steve.


On 22 Nov 2006 07:20:00 -0500, Hal Rosenstock <halr at voltaire.com> wrote:
>
> Hi Steve,
>
> On Wed, 2006-11-22 at 06:22, Steven Wooding wrote:
> > Hi Hal,
> >
> > The debugger reveals that the pointer p_sa_mad is equal to 0xffffffff.
> > When the osmv_get_query_result() returns it tries to use
> > p_sa_mad->data and that's what causes the seg fault.
> >
> > I also noticed that mad_size=0 in the osm_madw_t structure.
> >
> > What could have caused this to only happen in my user code?
>
> I'm not 100% sure it's your user code. I want to eliminate the following
> first prior to looking some more at what is going on:
>
> There are two changes to make and see if this helps:
>
> 1. Save your original versions of osm_vendor_ibumad_sa.c and
> osm_vendor_ibumad.c
>
> 2. Make the following change in osm_vendor_ibumad_sa.c:
>
> Add the following in:
> 182,187c180
>
>   /* put back the request madw */
>   if (p_req_madw)
>     osm_mad_pool_put(p_bind->p_mad_pool, p_req_madw);
>
> after:
> Exit:
>
> /* free the copied query request if found */
> if (p_query_req_copy) cl_free(p_query_req_copy);
>
> 3. Copy OFED 1.1 osm_vendor_ibumad.c over. There are a number of
> important bug fixes which may impact this. (Yes, I know the memory
> allocation is using malloc/free directly but things should work or you
> can hand edit these).
>
> 4. Rebuild
>
> 5. Rerun and see if still have same failure
>
> You can back these out if you don't like but I think they are
> improvements.
>
> -- Hal
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20061122/c7c796e9/attachment.html>