Re: [dpdk-dev] [PATCH v6 3/5] eal: make OS shims internal

[Thomas Monjalon <thomas@monjalon.net>]

02/04/2021 01:10, Dmitry Kozlyuk:
> 2021-03-31 23:55 (UTC+0200), Thomas Monjalon:
> > 31/03/2021 23:45, Nick Connolly:
> > > >
> > > > I don't understand your point.
> > > > I am just proposing to allow some apps to explicitly include the shim
> > > > for their convenience in case they are fully based on DPDK and
> > > > understand the risk of conflict with some other code.  
> > > 
> > > 
> > > Agreed - there’s no harm in doing so.
> > > 
> > > My point was simply that for a non-trivial application I suspect it will
> > > have limited value due to the sorts of conflicts that are likely to arise.  
> > 
> > OK yes.
> > If we expose this header, we should bring it with a disclaimer.
> 
> Let's now avoid "internal" (rte_os_shim?), but expose it later, if ever:
> 
> 1. The concept of "not so public" API is new, probably a techboard subject.
> 2. We may want devtools checks against including it from public headers.

Yes it should be well discussed and checked with scripts.
For now, let's not expose such shim, but as you said,
adopt a more descriptive name avoiding "internal" qualifier.