[openib-general] [PATCHv2] IPoIB CM Experimental support

[Michael S. Tsirkin]

> >I also noticed that when I set the MTU to 16K and ran some stressful MPI tests,
> >that my system seemed to get un-responsive like IPoIB was taking up too
> >much kernel memory. 
> 
> Correction, I saw the strange behavior when I had the MPU set to 64K,
> not 16K MTU,
> and I cannot be sure that it was IPoIB_CM that was causing the problem,
> so I think it would be good for others to give this some airtime and
> report
> their experiences to the list.

That's the setup I'm mostly testing at I haven't seen this yet.

Are you running this together with Sean's multicast patches and the sa cache?

Are you seeing something in the log? What about when you
set debug_level to 1? Does increasing the RQ size help?

-- 
MST