[PATCH v8 1/3] mbuf: de-inline sanity checking a reinitialized mbuf

[Morten Brørup]

> From: Thomas Monjalon [mailto:thomas at monjalon.net]
> Sent: Thursday, 9 October 2025 19.29
> 
> 09/10/2025 19:12, Morten Brørup:
> > From: Bruce Richardson [mailto:bruce.richardson at intel.com]
> > > On Sat, Aug 23, 2025 at 06:30:00AM +0000, Morten Brørup wrote:
> > > > +/**  check reinitialized mbuf type in debug mode */
> > >
> > > This is in release mode, not debug mode. Comment below seems wrong
> too.
> >
> > Yes, I noticed the comment was present in both debug and release
> mode,
> > which I couldn't understand. So I guessed it was for Doxygen or some
> other parser.
> > I have seen weird stuff for Doxygen, e.g. "#ifdef __DOXYGEN__"
> > for documenting a function [1],
> > so I didn't attempt to understand the reason for it,
> > but just followed the same pattern.
> 
> Hum, as a maintainer, I would prefer you try to understand, or ask
> please.
> Note: we can use Slack for such questions.

Point taken. Good guidance!
I'll try to broaden my scope of knowledge to include preprocessing for Doxygen.

> 
> __DOXYGEN__ is defined only by Doxygen,
> so any code inside #ifdef __DOXYGEN__ is for documentation only.
> It was supposed to be used in lib/eal/include/generic
> for functions which are really defined inline per CPU implementation.
> 
Yes, I get that.
But I don't get - when there are two definitions of a macro - why is the same comment/documentation present in both instances?
And if the instances have different comments/documentation, how is that reflected in the documentation output from Doxygen?

In this specific case, how should the code look to both 1) get the wanted Doxygen documentation output, and 2) have relevant comments inline in the source code for both instances?