[PATCH] drm/vgem: Fix vgem_init to get drm device avaliable.

[Emil Velikov]

On 15 November 2017 at 21:25, Sharma, Deepak <Deepak.Sharma at amd.com> wrote:

>>
>> I'd still recommend updating the two functions' documentation, ideally coupled with enforcing for *parent to be non NULL.
>> Could be code as follow-up though.
>>
> If I got it correctly you are referring   "Note that for purely virtual devices @parent can be NULL" for said two functions.
> I think changes might be required if it was "should/must be NULL"?


Since you're changing the behaviour:
The statement "Note that for purely virtual devices @parent can be
NULL" is never true and should be dropped.

Additionally, you want to update the functions to error out when
parent is NULL since it indicates a driver bug.
For the drm_dev_set_unique hunk one can drop the comment (it's NA) and
simplify to:

ret = drm_dev_set_unique(dev, dev_name(parent));

As mentioned before - it can be code as follow-up.


>>
>> Most of which use i915 <> vgem. If you don't have the HW to test, one can use the Intel GFX trybot.
>> Just keep [1] in the To/CC list and you'll get a report with the results.
>>
>
> Thanks. I have added Intel GFX trybot in CC, that should be sufficient or I need to send patch again using git send-mail?
>
Right, should have been clearer - the actual patches should be send/cc'd.
Otherwise one has know way of retrieving (and thus testing) the patch ;-)

HTH
Emil