Prev: [PATCH] failslab: add ability to filter slab caches
Next: [PATCH -next] staging: rtl8192* needs semaphore.h
From: Dmitry Torokhov on 24 Feb 2010 14:10
On Wed, Feb 24, 2010 at 10:25:49AM +0000, Mark Brown wrote:
> On Tue, Feb 23, 2010 at 11:38:50PM -0800, Dmitry Torokhov wrote:
> > Driver data set by platform_set_drvdata() is for private use of
> > the driver currently bound to teh device and not for use by parent,
> > subsystem and anyone else.
> >
> > Also have wm8400_register_regulator() accept 'sturct wm8400 *'
> > instead of generic device structure.
>
> Nack due to this change - this change would make it impossible for
> callers to actually call the function. Note that nothing including only
> wm8400.h even has a struct declaration, much less defniition, for struct
> wm8400.

If you notice I added forward declaration of "struct wm8400" to wm8400.h
thus users can pass around pointer to the structure.

I really think we should refrain from passing naked 'struct device *'
pointers as much as possible since it is error prone. Otherwise
wm8400_register_regulator has no way of ensuting that passed 'dev' is
indeed wm8400.

--
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Dmitry Torokhov on 24 Feb 2010 14:30
On Wed, Feb 24, 2010 at 07:14:03PM +0000, Mark Brown wrote:
> On Wed, Feb 24, 2010 at 11:02:34AM -0800, Dmitry Torokhov wrote:
> > On Wed, Feb 24, 2010 at 10:25:49AM +0000, Mark Brown wrote:
>
> > > Nack due to this change - this change would make it impossible for
> > > callers to actually call the function. Note that nothing including only
> > > wm8400.h even has a struct declaration, much less defniition, for struct
> > > wm8400.
>
> > If you notice I added forward declaration of "struct wm8400" to wm8400.h
> > thus users can pass around pointer to the structure.
>
> This doesn't help unless you also provide a way for users to obtain a
> struct wm8400.

Why would they need it? Only code that creates instances of wm8400 needs
to know the definition of the sturcture, the rest can simply pass the
pointer around.

I guess there is disconnect between us and I do not see any users of
wm8400_register_regulator() in linux-next... Is there another tree I
could peek at?

>
> > I really think we should refrain from passing naked 'struct device *'
> > pointers as much as possible since it is error prone. Otherwise
> > wm8400_register_regulator has no way of ensuting that passed 'dev' is
> > indeed wm8400.
>
> Right, there's no type safety here (and this whole method of registering
> regulators is fairly deprecated anyway in favour of just straight
> platform data) but really it doesn't buy us much - the users get exactly
> the struct device they need to use passed in to them, there's really
> very little chance for them to get confused about what they're talking
> about.

--
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 | 
Pages: 1
Prev: [PATCH] failslab: add ability to filter slab caches
Next: [PATCH -next] staging: rtl8192* needs semaphore.h