Keyboard mapping

Adam Jackson ajax at nwnk.net
Sun May 22 15:29:18 PDT 2005


On Sunday 22 May 2005 18:14, Jon Smirl wrote:
> I know that VCs are not required to run the X server. The problem
> comes from people who run consoles in some sessions and the X server
> in another. On VT swap nasty things happen. Number one, the current X
> server disables all VGA devices in the system. Disabling all VGA
> devices is bad for the other users since they are using them and they
> haven't swapped VTs.
>
> The other problem with VT swap is that historically you are allowed to
> do anything you want to the video card in another VT session. That
> includes changing the engine mode and wiping VRAM. This plays havoc
> with things like monitor hotplug - the interrupt may be enabled and
> the other VT session replaces the driver. Or it may disable the
> interupt and you miss the event.

Doctor, it hurts when I go like this.

> My personal preference is to see in-kernel VTs eliminated and replaced
> with a user mode scheme that looks just like them. A single device
> driver (merged fbdev/DRI) would be loaded and control the state of the
> video card. The user space VT implementation would then use this
> single driver. A primitive in-kernel console would be maintained but
> it would only be used for system emergencies, normal login would use
> the user space console.

I don't disagree with any of this.  Text consoles are an anachronism.  But I 
have no idea how this relates to either keyboard mapping, or EGL.

- ajax
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/dri-egl/attachments/20050522/094fe4c5/attachment.pgp


More information about the dri-egl mailing list