Re: [PATCH] staging: greybus: codecs: fix type confusion with dedicated list iterator variable

[Jakob Koschel <jakobkoschel@gmail.com>]

> On 21. Mar 2022, at 10:21, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> 
> On Mon, Mar 21, 2022 at 10:06:13AM +0100, Jakob Koschel wrote:
>> 
>>> On 21. Mar 2022, at 09:48, Dan Carpenter <dan.carpenter@oracle.com> wrote:
>>> 
>>> The subject says that it fixes a bug but it does not.
>> 
>> Thank you for your review!
>> 
>> I don't agree that this doesn't fix a bug:
>> 
>>> +		}
>>> 	}
>>> 	if (!data) {
>>> -		dev_err(dai->dev, "%s:%s DATA connection missing\n",
>>> -			dai->name, module->name);
>> 
>> Using 'module' when data == NULL is *guaranteed* to be a type confused
>> bogus pointer. It fundamentally can never be correct.
> 
> Ah.  I did not read all the way to the end of the patch.
> 
> The bugfix needs to be sent as it's own patch.  Just the one liner.  It
> needs a fixes tag as well.
> 
> [PATCH] staging: greybus: fix Oops in error message
> 
> The "module" pointer is invalid here.  It's the list iterator and we
> exited the loop without finding a valid entry.
> 
> Fixes: 6dd67645f22c ("greybus: audio: Use single codec driver registration")
> Signed-off-by: You
> 
> 	if (!data) {
> -		dev_err(dai->dev, "%s:%s DATA connection missing\n",
> -			dai->name, module->name);
> +		dev_err(dai->dev, "%s DATA connection missing\n",
> +			dai->name);
> 		mutex_unlock(&codec->lock);
> 
> We're happy to apply the other stuff as well, but we don't mix cleanups
> and bug fixes like that.

ok great, I'll separate and resubmit both. Thanks!

> 
> regards,
> dan carpenter

	Jakob