From: Rami Rosen [mailto:roszenr...@gmail.com] >Hi Morten, >A good idea, thanks for volunteering! >Several minor comments: >I would consider calling it rte_mbuf_hdr_parse(), to make it more related to >the rte_mbuf* methods, and also I would consider having it in librte_mbuf and >not in librte_net as you suggested.
I considered this, but since it looks inside the packet data, and not just processes metadata, I think it doesn't belong in librte_mbuf. > >Also regarding the bulk method. The first option is indeed faster and better >in terms of performance, which is important since it is intended probably >mostly to the datapath. I would consider having the bulk method iterating over >the rte_hdr_parse() method ,( or rte_mbuf_hdr_parse() if you will agree to my >suggestion), and adding a boolean parameter (mark_malform or something like >that) to this method + removing the const quailifier. The bulk method will set >that flag when calling the rte_hdr_parse(). Thus you will avoid duplicity of >the parsing code. Good points... And regarding avoiding code duplicity, I'm pursuing Olivier about merging packet header validation into rte_net_get_ptype() instead of writing a separate function. > >Regards, >Rami Rosen
