On Thu, May 03, 2007 at 06:28:27PM -0600, Jordan Crouse wrote:
I meant that the code in the kernel would be completely native to the hardware. No spoofing.
So what would handle the several dozen MSRs needed to set up routing - some of which are for essential devices that don't have drivers?
Good question. I guess most of the world depends on PCI today?
There's HT, but I've understood that it works a lot like PCI.
There's VIA's V-Link too. Will be interesting to see how that works out in the C7 port. (Sorry Corey, don't know anything except that it's a proprietary north-south interconnect.)
And we would have to invent a new bus just for Geode devices (some of which share code with other entities - amd74xxx IDE for example),
Well, the new bus was invented by the hardware designers. (NatSemi or maybe even Cyrix I guess.)
It's interesting because it's a first. The Geodes are really different.
because without PCI, you wouldn't be able to match vendor/device IDs, or control the devices, or assign I/O or memory resources in virtual space, at least not easily.
Yes, I guess it would be a completely new architecture, since that's what the hardware is after all, except that it's still x86. :)
So basically it would take the Geode from a pretty well supported platform to one that would be completely custom and on the fringe, and certainly not suitable for the mainstream kernel. I think the SMM handler is a small price to pay. :)
I understand that completely. If geode-vsa is as critical as you say then there is certainly no other option in the foreseeable future.
(But someone (Tom?) wrote something to give me the impression that the LX would actually run well also without geode-vsa?)
//Peter