[ofa-general] [PATCH] RDMA/nes: Check cm_node before using it

[Tung, Chien Tin]

>This patch is fine -- but I've never seen the oops the current code
>would cause, and I'm guessing you haven't either.  Is there 
>any way that
>schedule_nes_timer() gets passed a NULL cm_node?

I checked all caller of schedule_nes_timer() and didn't see any
instances of cm_node being NULL.  We have a few cm related patches 
coming, I will roll this change into one of them.

Thanks,

Chien