[PATCH:xf86-video-mga] Fix pointer cast warning.

[Alan Coopersmith]

On 07/31/15 09:02 PM, Connor Behan wrote:
> On 28/07/15 11:59 AM, Thomas Klausner wrote:
>> On Thu, Feb 26, 2015 at 02:59:49AM -0500, Connor Behan wrote:
>>> On 25/02/15 06:41 PM, Tormod Volden wrote:
>>>>> And then disable the DownloadFromScreen() implementation, or perhaps
>>>>> even disable EXA altogether.
>>>> If nobody can find the real problem that is maybe all that can be done.
>>> The fix should be:
>>>
>>> char *src = pExa->memoryBase + exaGetPixmapOffset(pSrc);
>>>
>>> When I first saw exaGetPixmapFirstPixel() in savage and mga, I assumed
>>> this was just a shorthand for the above. But evidently it's not. As for
>>> other broken EXA drivers, I don't have hardware to test but some basic
>>> suggestions are:
>>>
>>> 1. Use the above instead of pPix->devPrivate.ptr.
>>> 2. Make all hooks start by verifying that pSrc is non-NULL.
>>> 3. Remove UTS / DFS hooks unless they actually do a DRM blit.
>>>
>> So what can we do to make this progress?
>>   Thomas
> I just looked through the drivers on the main site and found 7
> problematic ones.
>
> * Geode is not careful enough with pSrc and its drawables
> * Mga is not careful enough either, plus it has a useless DFS hook
> * Radeonhd is not careful enough with pSrc / drawables
> * Siliconmotion is not careful enough with the drawable
> * Sis has useless UTS / DFS hooks but its wrappers do a good job of
> hiding this
> * Via is not careful enough with pSrc / drawables and it uses devPrivate
> all over the place
> * Xgi's UTS / DFS hooks are useless (except maybe on ARM) and it uses
> devPrivate

Some of those drivers are no longer maintained or being updated.

For instance, http://cgit.freedesktop.org/xorg/driver/xf86-video-radeonhd
has no commits in the past 5 years, and -via was last touched in 2009.
http://cgit.freedesktop.org/xorg/driver/xf86-video-xgi/log/ shows the
occasional push, but no tag/release since 2010.


-- 
	-Alan Coopersmith-              alan.coopersmith at oracle.com
	 Oracle Solaris Engineering - http://blogs.oracle.com/alanc