[lvc-project] [PATCH] thunderbolt: handle possible NULL pointer from get_device()

[Greg KH]

On Fri, Jun 09, 2023 at 10:15:36AM +0300, Dmitry Antipov wrote:
> On 6/9/23 09:30, Greg KH wrote:
> 
> > Sorry, but how can that happen?
> 
> Hmmm.. there should be a complete mess in the device tree (e.g. passing
> root device with no parent where the regular device is expected). If
> you're sure that this doesn't worth checking then please disregard this.

Again, how did you test this?

And yes, the driver model guarantees that a child can never have a NULL
parent (except for the root device, but that's a different story...)

thanks,

greg k-h