Backing Store not working in 8 Bit color depth with vesa driver

Patrick O'Donnell pao at ascent.com
Tue Sep 20 06:43:25 PDT 2011


>Date: Mon, 19 Sep 2011 13:49:30 -0500
>From: Doug Kuvaas <doug.kuvaas at gmail.com>
>
>On Mon, Sep 19, 2011 at 1:03 PM, Patrick O'Donnell <pao at ascent.com> wrote:
>>>Date: Mon, 19 Sep 2011 10:20:12 -0500
>>>From: Doug Kuvaas <doug.kuvaas at gmail.com>
>>>
>>>I posted to this list earlier about not getting color in 8 bit depth
>>>using the intel driver.  After some assistance I was able to get 8 bit
>>>color to show in color.  However, now the backing store does not seem
>>>to work, so the application I am attempting to run leaves blank areas
>>>on the display whenever a menu or other window covers that area.
>>>xdpyinfo does report that the backing store is on.
>>
>> We, too, ran afoul of this change.  I turns out, though, that X
>> clients were never allowed to presume the server would provide backing
>> store (or save unders).  Even if the server announced that it
>> supported backing store, it was allowed, if resources became tight, to
>> refuse to provide it for any window that asked.
>>
>> It sounds like the client you're attempting to run is presuming upon
>> the backing store and is ignoring the expose events that are being
>> generated.  This, unfortunately, is broken behavior.
>
>I thought the -wm option was supposed to force all windows to use the
>WhenMapped backing store?  Were you able to come up with a
>work-around?

Well, if the server offers backing store, then that's great.  It can
still choose not to give a window backing store.

The "work-around" we came up with was to fix our client.

It's frustrating when the server doesn't give backing store,
particularly when so many have over the years that client programmers
have come to rely on it, but it's always been optional.

		- Pat



More information about the xorg mailing list