> On Fri, 18 Mar 2005 10:29:02 +0100 (CET)
> [EMAIL PROTECTED] wrote:
>
>> You can't do that. It could means : software filtering that is slow,
>> and/or kernel system call that is also slow.
>>
>> Why do you need that ? If you provided hardware access to a safe
>> interface
>> of the card, there is no problem. (this interface context should be
>> saved
>> and restored by the kernel when a task switch occur, the card could also
>> manage "few" context (3/4, the opengl programme and the windows manager
>> at
>> least))
>
> You don't want to access the graphics card on every task switch.
> On a current linux system you have something between 100 and 2000
> context switches per second. If you do something like this, then
> the driver has to intercept every access and check whether
> the registers need to be swapped.
>

no it juste have to use it's normal VM mechanism. You could also manage a
kind of unique access to the 3D part of the card. I don't know how dri
handle two 3D app in the same time on the screen.

But you could imagine, having 4 duplicate register bank. Each discribe the
same user interface, but not the same graphic context. Each user have one
register bank mapped by the VM. You could imagine one for xorg, and 3 for
app. Kernel could also simulate more context by saving the content but it
could be slow.
Here each app access this own register set without anoying other app.

The tricky point is how to manage the video memory : maybe by a kernel call.


_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to