[openib-general] on calling rdma_disconnect from non sleepablecontext
Roland Dreier
rdreier at cisco.com
Wed Feb 1 09:20:29 PST 2006
> I have now instrumented iser code to always call rdma_disconnect /
> rdma_destory_qp etc from sleepable context. Before doing so i was
> getting this oops few times. My interpertation was that
> cma_modify_qp_err is not supposted to get called when in_atomic is
> true, am i correct?
Yes, the modify QP operation might sleep. For example, on Mellanox
hardware, modifying a QP requires a firmware command, which allocates
a mailbox with GFP_KERNEL and also sleeps until the command completes.
I tried to document these sorts of rules in Documentation/infiniband/core_locking.txt
- R.
More information about the general
mailing list