Re: [PATCH 3/5] ASoC: sam9x5_wm8731: Return -ENODEV when probe does not find OF node

[Alexandre Belloni <alexandre.belloni@free-electrons.com>]

On 15/01/2018 at 10:11:14 +0100, Ladislav Michl wrote:
> On Mon, Jan 15, 2018 at 09:20:25AM +0100, Alexandre Belloni wrote:
> > On 15/01/2018 at 08:53:51 +0100, Ladislav Michl wrote:
> > > Probe should fail with -ENODEV when called with NULL pdev->dev.of_node.
> > > 
> > > Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
> > > ---
> > >  sound/soc/atmel/sam9x5_wm8731.c | 6 ++----
> > >  1 file changed, 2 insertions(+), 4 deletions(-)
> > > 
> > > diff --git a/sound/soc/atmel/sam9x5_wm8731.c b/sound/soc/atmel/sam9x5_wm8731.c
> > > index e6c303ab869d..9db08826b32a 100644
> > > --- a/sound/soc/atmel/sam9x5_wm8731.c
> > > +++ b/sound/soc/atmel/sam9x5_wm8731.c
> > > @@ -84,10 +84,8 @@ static int sam9x5_wm8731_driver_probe(struct platform_device *pdev)
> > >  	struct sam9x5_drvdata *priv;
> > >  	int ret;
> > >  
> > > -	if (!np) {
> > > -		dev_err(&pdev->dev, "No device node supplied\n");
> > > -		return -EINVAL;
> > > -	}
> > > +	if (!np)
> > > +		return -ENODEV;
> > >  
> > 
> > This will never happen, you may as well just remove the test.
> 
> Ah, right, this is DT only driver. And the same for the patch 5.
> 
> Btw, all that drivers seems to be for development boards, so what
> about moving to simple card instead? (and let these special drivers
> die silently)
> 

Yes, that would be good, I'm all for it. However, I think some custom
boards are using those drivers and are not upstream and will also need
to be migrated or at least have a proper migration path.


-- 
Alexandre Belloni, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com