[Intel-gfx] [PULL] topic/core-stuff

Dave Airlie airlied at gmail.com
Wed Sep 24 23:07:38 CEST 2014


On 25 September 2014 03:59, Daniel Vetter <daniel.vetter at ffwll.ch> wrote:
> On Wed, Sep 24, 2014 at 6:24 PM, Ilia Mirkin <imirkin at alum.mit.edu> wrote:
>> On Wed, Sep 24, 2014 at 6:24 AM, Daniel Vetter <daniel.vetter at ffwll.ch> wrote:
>>> Hi Dave,
>>>
>>> Just noticed that you've picked up the header rework stuff already, so
>>> I've rebased that out again. Otherwise just two stragglers from the vblank
>>> rework and the universal cursor planes locking fix. Plus sprinkling
>>> container_of all over fbdev emulation from Fabian.
>>
>> I thought it was questionable whether these were desired... the
>> pattern of casting between structs that contain others as bases is a
>> pretty common one, and container_of just makes it harder to read. Did
>> the discussion end in a way that concluded the opposite? I might have
>> missed it.
>
> I like them since you can employ gcc to check mundane stuff for you.
> And container_of is so massively established in the kernel (with
> list_head alone) that I don't think any serious kernel hacker has
> problem's reading it. Alan Cox complained a bit a bit on imo elitist
> grounds so until someone goes on record with an official Nack I think
> they're worth it.

I like them, but I don't want to encourage them, they are over time
going to just be conflict
creators, which makes it sounds like work for me,

So I don't mind this batch for the fbdev, but I don't want to see
anymore unless they provide more value.

also I'm on holidays until Tuesday, I'll have some access to stuff,
but won't be on top of urgent things, so hopefully nothing urgent
happens.

Dave.



More information about the Intel-gfx mailing list