Re: The GGI and EvStack debate -- Linus and such persons please reard.

Jon M. Taylor (taylorj@ecs.csus.edu)
Thu, 26 Mar 1998 09:44:47 -0800 (PST)


On 26 Mar 1998, Linus Torvalds wrote:

> In article <3519C152.2B438E37@onramp.net>,
> Trever Adams <arabian@onramp.net> wrote:
> >
> >It seems Linus (though I respect him, think he is intelligent, I don't
> >believe he is infallible! No one is!) doesn't like the EvStack due to
> >the ideas of multiple keyboards.
>
> No.
>
> I don't like EvStack because I don't see the point. We can already
> support multiple keyboards and multiple displays, and we don't need
> EvStack to do so.

AFAIK there is a lot more to EvStack than multiheading stuff.
That isn't even the main point, really. The main point is extending to
all console IO devices a generic layered message passing system.
Multiheading is just a good example of what EvStack can do easily.

I like the EvStack concept of consoles consisting of a series of
pluggable event handler/router modules. I like the idea of the kernel
having a generic, flexible and fast way to pass abstract messages between
itself and userspace. And I like the idea that any console IO device, no
matter how wierd, can have all its capabilities encapsulated into a set of
abstracted messages.

That last is very important for graphics hardware, where the
feature set that can be exported by the kernel to userspace varies widely
from card to card. But the same message-based paradigm can allow any sort
of odd device to be easily supported by Linux - just write a kernel driver
that wraps a thin message layer over the raw hardware functionality and
let the userspace stacks contain the logic the glues the devices together
into a "console".

Jon

---
'Cloning and the reprogramming of DNA is the first serious step in 
becoming one with God.'
	- Scientist G. Richard Seed

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu