nanogui: More...Porting to a custom video device
Subject:
Re: [nanogui] More...Porting to a custom video device
From:
Jordan Crouse ####@####.####
Date:
24 May 2001 14:03:20 -0000
Message-Id: <01052408022000.22437@cosmic>
You shouldn't have any problems writing to a virtual framebuffer, but why
would you want to do this? No matter how many devices you stick in between,
you still need some way to transfer the memory to your hardware device.
Tossing in a virtual framebuffer just adds another memory copy:
[ Microwindows ] ---> [ Virtual Framebuffer ] --->
[ Video driver ] ---> [ Video hardware ]
When you can just do:
[ Microwindows ] ---> [ Video driver ] ---> [ Video hardware ]
You should have all the correct elements to make your kernel video driver,
and you would then automatically support all the structures that Microwindows
normally gets from the kernel. As long as you have a way to access your
hardware from the kernel (PCI, ISA, etc...), you should be good to go.
Jordan
On Wednesday 23 May 2001 18:05, Gordon Minns mentioned:
> Thanks, Jordan....
>
> This LCD video device employs no standard graphics chip, so unless we
> emulate a VESA or other standard hardware type in our custom hardware, we
> would not be able to use an existing driver. HOWEVER...
>
> What do you think about configuring MicroWindows to write to a virtual
> frame buffer (kernel option). Then, if we understand the data structure of
> the resulting image, we should be able to read pixel data from the virtual
> frame buffer and send it on out to our custom display hardware.
>
> Do you know of any problems setting up MicroWindows for use with a virtual
> frame buffer?
>
> Gordon Minns
