[openib-general] [RFC] [PATCH 1/3] RDMA CM:add rdma_get/set_optioncalls to get/set path records

[Roland Dreier]

    Sean> What I'm more concerned with is whether it's acceptable to
    Sean> call copy_to_user / copy_from_user on kernel memory?

No, I don't think you should do that.  It might work by chance but it
really sounds like something is wrong with the design if you want to
do that.  For one thing it will make it impossible to have sparse
check __user annotations.

Why wouldn't you just do the copying in the place where you dispatch
user functions to the kernel handlers?

 - R.