On Wed, Feb 28, 2018 at 07:29:15AM +0200, Boris Pismenny wrote: > >>I definitely agree. > >>I've started doing it and realized that changing this here implies changes > >>throughout mlx5 where many types should be unsigned. > > > >?? I don't see it. Almost all the changes are in code you added here: > > What I meant to say is that once I've started doing and realized that this > changes requires too many other changes, then I've stopped and reverted to > the approach you see here. Again, I don't see it. > >Make qpinlen size_t in the create_raw_packet_qp_rq signature > >Make inlen size_t in the create_raw_packet_qp signature > >Make inlen size_t in create_qp_common > > Do that and you would get warnings about passing size_t into function that > take an int argument. For example: > qpinlen is a parameter for mlx5_core_create_qp, and it passes it down > mlx5_cmd_exec. Fixing mlx5_cmd_exec requires fixing at least ~150 lines. Okay, so don't change it in create_qp_common. But you can still use the right type in all the new arguments introduced by this patch. Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
