On Wed, May 09, 2018 at 04:41:53PM -0300, Mauro Carvalho Chehab wrote:
> Em Wed, 9 May 2018 15:11:07 -0400 (EDT)
> Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> escreveu:
>
> > On Wed, 9 May 2018, Mauro Carvalho Chehab wrote:
> >
> > > Em Wed, 9 May 2018 19:15:01 +0200
> > > Andrea Parri <parri.andrea@xxxxxxxxx> escreveu:
> >
> > > > > tools/memory-model/README | 10 +++++-----
> > > >
> > > > As mentioned in the previous thread, I am for keeping the current
> > > > references: the REAMDE is listing the doc files, as well as other
> > > > files in tools/memory-model/, relatively to that directory.
> > >
> > > Yeah, at least this hunk deserves some rework, as now some
> > > references are Documentation/.../foo, while others are just
> > > bar.
> > >
> > > As on (almost) all other places (except for tools/memory-model/README),
> > > the references are always from the main directory, I would make all
> > > patches there also relative to main dir. If you're afraid of
> > > not being too clearer, we could prefix all of them with something
> > > like:
> > >
> > > ${LINUX}/tools/memory-model/...
> > >
> > > just like some DT binding files do:
> > >
> > > Documentation/devicetree/bindings/sound/audio-graph-card.txt:see ${LINUX}/Documentation/devicetree/bindings/graph.txt
> > >
> > > A bonus of doing that is that the broken reference detect script can
> > > keep parsing it without changes (well, it wouldn't be hard to make
> > > it also accept a relative file, but doing that just due to
> > > tools/memory-model/README seems overkill).
> > >
> > > Another advantage is that it would allow to easily add references
> > > there from the main kernel Documentation, if needed in the future,
> > > without messing with local x non-local relative namespace.
> >
> > How about changing the relative references so that something like
> > Documentation/recipes.txt becomes ./Documentation/recipes.txt?
>
> Sounds a good idea to me. Andrea?
Counterintuitive ;-) Documentation/abc.txt vs. ./Documentation/abc.txt
But let me sleep over this ;-) for the time being, I renew my vote for
"just don't commit these changes (and keep checking automated changes)".
Andrea
>
> Thanks,
> Mauro
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/dri-devel
