Re: drivers for l4-hurd

[Jörg Sonnenberger]

On Thu, 28 Nov 2002 23:46:03 +0100
address@hidden (Peter 'p2' De Schrijver) wrote:
[snip]
> > > > In fact, I would think that for some things (Bus driver + device 
> > > > drivers)
> > > > it's easier to have it all in one driver program with maybe dynamically
> > > > loadable shared object modules.
> > > > 
> > > 
> > > This split depends on a few things : speed of L4 messaging and memory
> > > mapping, device throughput, CPU speed, ... . 
> > 
> > Oh, yes, I don't want to give a premature answer.  As repeatably said, my
> > message is that an answer should not be given now ;)
> > 
> 
> The API's should probably be designed so that drivers can share an AS or
> live in a different AS. This would allow us to use seperate AS's when
> developing/testing and move to 1 AS when speed is required.

First of all the problem with bus driver and device driver is the dependency of
IDE on PCI, of SCSI on PCI, of USB on PCI. You got the point. The solution with
shared AS sounds partly reasonable. I don't think all device drivers using PCI 
should
shared AS, but they could embed the PCI bus driver AS and some memory for IO 
messages.
This would give good performance and isolate the drivers. The AS of the bus 
driver
should be r/o of course (with the above exception).

- Joerg Sonnenberger

> 
> Cheers,
> 
> Peter.
> 
> 
> _______________________________________________
> L4-hurd mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/l4-hurd
>