On Tue, 31 Oct 2017 15:57:41 -0400, Eric Garver wrote: > On Mon, Oct 30, 2017 at 09:29:34AM +0800, Yi Yang wrote: > > + if (WARN_ON(is_push_nsh && is_mask)) > > + return -EINVAL; > > OVS_NLERR() is probably more appropriate.
No, not here. If this happens, it's a bug in the kernel and WARN_ON is what we need. This is not triggerable from user space and user space has no way to fix it if this happens. Jiri
