Re: [PATCH] net: introduce big and little endian types

[Thomas Monjalon <thomas.monjalon@6wind.com>]

2016-12-08 13:59, Wiles, Keith:
> 
> > On Dec 8, 2016, at 3:30 AM, Nélio Laranjeiro <nelio.laranjeiro@6wind.com> wrote:
> > 
> > Hi all,
> > 
> > Following previous discussions, I would like to gather requirements for
> > v2, currently we have:
> > 
> > 1. Introduction of new typedefs.
> > 2. Modification of network headers.
> > 3. Modification of rte_*_to_*() functions.
> > 
> > Point 1. seems not to be an issue, everyone seems to agree on the fact
> > having those types could help to document some parts of the code.
> 
> I never stated these new types were useful in any way, I still believe documentation of the code is the better solution then forcing yet another restriction in submitting patches. 

It would not be a restriction, just a help for those wanting to document
some tricky parts by using these types.

I see 2 usages:

- in a struct:

rte_be32_t speed; /**< 0 for speed negotiation */
instead of
uint32_t speed; /**< [big endian] 0 for speed negotiation */

- in a function:

rte_be32_t decode_speed (void *);
[...]
speed = rte_be_to_cpu_32(decode_speed());

It is difficult to reject something which could help a bit.
Do you really think it would bring some confusion to have some code
using these endianed-types?