> Perhaps the best way of accomplishing this is like a multicomputer.
>
> Each processor runs a seperate kernel, and they all talk via ultrafast,
> over the bus, modified TCP/IP.
I've always wondered why the lo interface had room for 16M IP addresses...
:-)
> In a shared (or semi shared) it would be tough.. You'd have a master
> processor allocate chunks for the other chips.. This would only work if
> they all had decent MMUs I guess... :)
Things are even worse -- there are machines out there that have their MMUs
shared between two processors... :-) But I guess support for this will not
really be added in the first place... :-)
CU
Simon
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu