[PATCH] dix/window: catch illegal window reconfiguration for composited windows.

Peter Harris pharris at opentext.com
Wed Jun 2 13:37:36 PDT 2010

On 2010-06-02 16:22, Keith Packard wrote:
> On Wed,  2 Jun 2010 17:18:23 +1000, Dave Airlie <airlied at gmail.com> wrote:
>> From: Dave Airlie <airlied at redhat.com>
>> If a windows has been redirected, then it has to respect pixmap size limits,
>> 32kx32k. This checks before resizing or changing border width if the window
>> is redirected and the pixmap size limits are being hit, and returns
>> BadValue.
> Should we make this check unconditional? Is there any reason to create a
> window that large when not redirected?

At least some apps create a giant subwindow and scroll by changing the
origin. Those apps are broken anyway (since their scroll stops working
when the underlying window grows beyond 64k), but allowing the window to
grow beyond 32k hides the brokenness for a little bit longer.

That said, I'm content to make the brokenness visible sooner rather than

(Aside: BadAlloc seems to make more sense than BadValue, since the value
is permitted by the protocol - we just don't want to allocate the space
for it. But I'm just bikeshedding here)

Peter Harris
               Open Text Connectivity Solutions Group
Peter Harris                    http://connectivity.opentext.com/
Research and Development        Phone: +1 905 762 6001
pharris at opentext.com            Toll Free: 1 877 359 4866

More information about the xorg-devel mailing list