[git pull] randr - reviewed patches.

[Aaron Plattner]

On 04/30/2013 05:42 AM, Dave Airlie wrote:
> On Tue, Apr 30, 2013 at 1:51 PM, Aaron Plattner <aplattner at nvidia.com> wrote:
>> On 04/29/13 17:15, Dave Airlie wrote:
>>>
>>> The following changes since commit
>>> 451ba4bd41b82acd4aec6236ba121e00cfeb311b:
>>>
>>>     hw/xfree86: Only report SetDesiredModes() failed if at least one
>>> modeset fails (2013-04-29 09:10:06 -0700)
>>>
>>> are available in the git repository at:
>>>
>>>     ssh://people.freedesktop.org/~airlied/xserver for-keithp
>>>
>>> for you to fetch changes up to f2fd8ec3725a61abbc831f0a9ec28fa2b7020c47:
>>>
>>>     gpu: call CreateScreenResources for GPU screens (2013-04-30 10:10:51
>>> +1000)
>>>
>>> ----------------------------------------------------------------
>>> Dave Airlie (8):
>>>         randr: don't directly set changed bits in randr screen
>>>         randr: make SetChanged modify the main protocol screen not the gpu
>>> screen
>>>         randr: only respected changed on the protocol screen
>>>         randr: report changes when we disconnect a GPU slave
>>>         dix/gpu: remove asserts for output/offload from same slave
>>>         xf86crtc: don't use scrn->display for gpu screens
>>>         dix: allow pixmap dirty helper to be used for non-shared pixmaps
>>>         gpu: call CreateScreenResources for GPU screens
>>
>>
>> Nak, please.  This one breaks normal prime display offload for me (and
>> everyone else using Ubuntu 13.04).  See
>> http://lists.x.org/archives/xorg-devel/2013-April/036087.html
>>
>
> But its already in Ubuntu and Fedora 18 + 19, and it works for
> non-binary driver.

It breaks intel + modesetting(udl) too.  The problem is that the GPU 
screen starts with a mode set on its outputs, so xrandr --auto doesn't 
actually do anything unless you explicitly turn it off first.

> So I'm not sure not putting it into the X server matters, if distros
> already have it.

Distros are broken.

-- 
Aaron