Re: [PATCH v2] [media] of: move graph helpers from drivers/media/v4l2-core to drivers/media

[Philipp Zabel <p.zabel@pengutronix.de>]

Hi Mauro,

Am Mittwoch, den 12.02.2014, 22:35 +0900 schrieb Mauro Carvalho Chehab:
> Em Wed, 12 Feb 2014 10:11:54 +0100
> Philipp Zabel <p.zabel@pengutronix.de> escreveu:
> 
> > Hi Mauro,
> > 
> > Am Mittwoch, den 12.02.2014, 06:53 +0900 schrieb Mauro Carvalho Chehab:
> > [...]
> > > > diff --git a/include/media/of_graph.h b/include/media/of_graph.h
> > > > new file mode 100644
> > > > index 0000000..3bbeb60
> > > > --- /dev/null
> > > > +++ b/include/media/of_graph.h
> > > > @@ -0,0 +1,46 @@
> > > > +/*
> > > > + * OF graph binding parsing helpers
> > > > + *
> > > > + * Copyright (C) 2012 - 2013 Samsung Electronics Co., Ltd.
> > > > + * Author: Sylwester Nawrocki <s.nawrocki@samsung.com>
> > > > + *
> > > > + * Copyright (C) 2012 Renesas Electronics Corp.
> > > > + * Author: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
> > > > + *
> > > > + * This program is free software; you can redistribute it and/or modify
> > > > + * it under the terms of version 2 of the GNU General Public License as
> > > > + * published by the Free Software Foundation.
> > > > + */
> > > > +#ifndef __LINUX_OF_GRAPH_H
> > > > +#define __LINUX_OF_GRAPH_H
> > > > +
> > > > +#ifdef CONFIG_OF
> > > 
> > > As a matter of consistency, it would be better to test here for
> > > CONFIG_OF_GRAPH instead, to reflect the same symbol that enables such
> > > functions as used on Kconfig/Makefile.
> > 
> > Maybe I'm trying to be too clever for my own good, but my reasoning was
> > as follows:
> > 
> > Suppose I newly use the of_graph_ helpers in a subsystem that does not
> > yet select OF_GRAPH. In that case I'd rather get linking errors earlier
> > rather than stubbed out functions that silently fail to parse the DT
> > later.
> 
> I see your point, but, imagining that someone pushed a patch using those
> symbols upstream, that would break compilation and git bisection, with 
> will hurt everyone, and not only the very few of us that would actually
> need the OF_GRAPH symbol for an specific driver.
> 
> Also, such push would mean that someone forgot to do his homework and
> to test if the committed functionality is actually working.
> 
> So, it seems more fair that the one that did the mistake will be the one
> that will suffer the consequences for his errors instead of applying a
> penalty to everybody's else ;)

point taken.

> > Since there is
> > config VIDEO_DEV
> > 	select OF_GRAPH if OF
> > already and the same should be added for other users of device tree
> > graphs, I think stubbing out the functions only if OF is disabled should
> > be enough.
> 
> Well, if you want to be sure that the graph will always be there if OF, then
> you could do, instead:
> 
> config OF_GRAPH
> 	bool
> 	default OF
> 
> (that would actually make OF_GRAPH just an alias to OF - so we could just
> use OF instead).
> 
> In any case, I think that we should use the same config name at Makefile, 
> Kconfig and of_graph.h (either OF_GRAPH or just OF).

If nobody states a strong preference, I'll just get rid of the
CONFIG_OF_GRAPH option altogether and use CONFIG_OF directly.

regards
Philipp