Hi Stephen, > -----Original Message----- > From: Stephen Hemminger <step...@networkplumber.org> > Sent: Monday, June 13, 2022 12:48 PM > To: Chautru, Nicolas <nicolas.chau...@intel.com> > Cc: dev@dpdk.org; tho...@monjalon.net; maxime.coque...@redhat.com; > t...@redhat.com; m...@ashroe.eu; Richardson, Bruce > <bruce.richard...@intel.com>; hemant.agra...@nxp.com; > david.march...@redhat.com > Subject: Re: [PATCH v1] bbdev: allow operation type enum for growth > > On Mon, 13 Jun 2022 11:24:35 -0700 > Nicolas Chautru <nicolas.chau...@intel.com> wrote: > > > Updating the last enum for rte_bbdev_op_type to allow for enum > > insertion. > > > > Signed-off-by: Nicolas Chautru <nicolas.chau...@intel.com> > > Only allowed if you check now for the any of the reserved types and fail.
Let me try to clarify what you mean. You would enforce such check in which function? Do you mean in any implementation of bbdev function taking enum rte_bbdev_op_type as argument? In that case do you mean having in effect 2 values: - one for the number of supported implemented operation type = could be kept private within rte_bbdev.c implementation, purely to reject using a non-supported operation type - and another higher value for padded maximum numbers of operations allowing for enumeration insertion. That would sound fine to me, but please kindly confirm this is what you are implying. I will also check for Thomas feedback as well, thanks.
