On Mon, 2005-05-09 at 10:46 -0400, Hal Rosenstock wrote: > On Mon, 2005-05-09 at 10:39, James Lentini wrote: > > The slab corruption message below seems to involve IBAT's > > req_comp_work function. Any ideas? > > I looked at this over the weekend and still don't have a working theory. > req_comp_work calls req_free which based on the pend->type frees the (in > this case) path request (calls free_path_req). It is the same as the one > which was allocated from the path request cache in > ib_at_paths_by_route. We are still looking at this.
Can you reproduce this on your end? I am still seeing it every time I run the kdapltest program on x86_64 with slab debugging turned on. With the latest bits (v2305). -tduffy
signature.asc
Description: This is a digitally signed message part
_______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
