Re: [PATCH] ASoC: Intel: avs: Add NULL check in avs_component_probe()

[Ethan Carter Edwards]

On 25/04/02 02:00PM, Markus Elfring wrote:
> > devm_kasprintf() returns NULL when memory allocation fails. Currently,
> …
>                 call?                                  failed?
> 
> 
> > Add NULL check after devm_kasprintf() to prevent this issue.
> 
> Do you generally propose here to improve the error handling?
> 
> 
> …
> > +++ b/sound/soc/intel/avs/pcm.c
> > @@ -927,7 +927,8 @@ static int avs_component_probe(struct snd_soc_component *component)
> >  		else
> >  			mach->tplg_filename = devm_kasprintf(adev->dev, GFP_KERNEL,
> >  							     "hda-generic-tplg.bin");
> > -
> > +		if (!mach->tplg_filename)
> > +			return -ENOMEM;
> 
> Can a blank line be desirable after such a statement?
> 
> 
> Would another source code transformation become helpful with an additional update step?
> 
> -		if (((vendor_id >> 16) & 0xFFFF) == 0x8086)
> -			mach->tplg_filename = devm_kasprintf(adev->dev, GFP_KERNEL,
> -							     "hda-8086-generic-tplg.bin");
> -		else
> -			mach->tplg_filename = devm_kasprintf(adev->dev, GFP_KERNEL,
> -							     "hda-generic-tplg.bin");
> +		mach->tplg_filename = devm_kasprintf(adev->dev, GFP_KERNEL,
> +						     (((vendor_id >> 16) & 0xFFFF) == 0x8086)
> +						     ? "hda-8086-generic-tplg.bin"
> +						     : "hda-generic-tplg.bin");
> 
> 
> Regards,
> Markus


Please feel free to ignore Markus. He has a tendency to provide unclear
and nonsense feedback. He has done this several times with my patch
sets. I almost wonder if he is an internet troll.

In any case, this change makes sense to me.

Reviewed-by: Ethan Carter Edwards <ethan@xxxxxxxxxxxxxxxxx>