[ofa-general] Re: [PATCH V2] mlx4: check for FW version which properly supports resize_cq

[Roland Dreier]

 > Not so (unfortunately).  The problem is that doing it the "correct" way
 > (as you describe above) results in userspace apps getting "EINVAL" as
 > the return value, and not ENOSYS.  This means that there is no way to
 > differentiate between a real error in the call, and simply a not-implemented
 > verb.  This really clobbers MPI.

Hmm... seems like a generic problem, since eg nes doesn't implement
resize_cq() ever.  I guess the cleanest fix would be to add handling for
unset device->resize_cq into uverbs_cmd.c, so drivers can add that to
their allowed userspace commands without actually implementing that.  We

Or maybe it's cleaner to add a stub resize_cq method that just returns
ENOSYS that drivers can set when they don't actually implement it...

 - R.