CVS Update: xc (branch: trunk)

[Thomas Winischhofer]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Alan Hourihane wrote:
> On Fri, 2005-10-07 at 14:04 -0700, Aaron Plattner wrote:
> 
>>Aaron Plattner wrote:
>>
>>>=====================
>>>This message appears to be from an NVIDIA.COM e-mail address, yet did not originate from an NVIDIA server.
>>>Please be cautious of any attachments or links in this message.
>>>Thank you, NVIDIA IT Team.
>>>=====================
>>>Alan Hourihane wrote:
>>>
>>>
>>>>CVSROOT:	/cvs/xorg
>>>>Module name:	xc
>>>>Changes by:	alanh at gabe.freedesktop.org	05/10/07 08:39:52
>>>>
>>>>Log message:
>>>> 2005-10-07 Alan Hourihane <alanh at fairlite.demon.co.uk>
>>>>         * programs/Xserver/hw/xfree86/common/xf86RandR.c
>>>> 	Use PointerMoved instead of SetCursorPosition, as PointerMoved will
>>>> 	call AdjustFrame to reposition the window if necessary and avoid the
>>>> 	cursor ending up offscreen.
>>>>
>>>>Modified files:
>>>>     ./:
>>>>       ChangeLog 
>>>>     xc/programs/Xserver/hw/xfree86/common/:
>>>>       xf86RandR.c 
>>>> 
>>>> Revision      Changes    Path
>>>> 1.1403        +6 -0      xc/ChangeLog
>>>> http://cvs.freedesktop.org/xorg/xc/ChangeLog
>>>> 1.14          +1 -1      xc/programs/Xserver/hw/xfree86/common/xf86RandR.c
>>>> http://cvs.freedesktop.org/xorg/xc/programs/Xserver/hw/xfree86/common/xf86RandR.c
>>>>
>>>>_______________________________________________
>>>>xorg-commit mailing list
>>>>xorg-commit at lists.freedesktop.org
>>>>http://lists.freedesktop.org/mailman/listinfo/xorg-commit
>>>
>>>
>>>Alan,
>>>
>>>This change breaks the rotation support for the nv driver that I was about to check in.
>>>http://people.freedesktop.org/~aplattner/nv-randr-1.patch.  If the cursor's in the
>>>lower-right corner of the screen, an "xrandr -o left" leaves the viewport panned partway
>>>off the root window.  I think you need to compare against scrp->virtualX and
>>>scrp->virtualY instead of size->width and size->height.
>>>
>>>-- Aaron
>>
>>Er, I meant pScreen->width and pScreen->height, rather than srcp->virtual[XY], as the
>>former are the rotated dimensions.
> 
> 
> Good catch Aaron, I'll update it.
> 

While we are at it: Could we "broaden" the cursor update upon a RandR
(rotation) change even further and cause/force a reload of the (hw)
cursor image? Reason being that the cursor image will have to change,
too, if the screen is rotated. Won't do anything for current CVS, but I
think that issue may come up when driver actually support rotation,
including rotating the hwcursor (as the current development sis driver
already does - and I am currently fiddling with a quick and smart way to
actually force an update).

Thomas

- --
Thomas Winischhofer
Vienna/Austria
thomas AT winischhofer DOT net	       *** http://www.winischhofer.net
twini AT xfree86 DOT org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)

iD8DBQFDRwG1zydIRAktyUcRAk+JAKC9+ukon6dC/r+uoOTwhLVM5qHtwgCff986
8gDAoWocxAulgEtSahMSoWc=
=MFAN
-----END PGP SIGNATURE-----