Re: Network-transparent IPC (was Re: vk_l4 -- CVS Setup)

[Farid Hajji]

> However, developers generally deemed the clans-and-chiefs model too
> limiting and uncomfortable, so most L4v2 implementations do not
> support it, and L4v4 will sport a new redirection model (which will
> also allow network-transparent IPC).
Will L4v4 provide for (local) fault-IPC? I'm meaning here that an IPC
to a TID without thread listening on the other end would be redirected
by the kernel to another address, say a proxy task?

> As an aside, network-transparent IPC for L4 would be rather complex to
> implement: If it was to support messages containing flexpage mappings
> (and not just memory-copying IPC), it would have to provide some sort
> of distributed shared memory.
Oh yeah, this is a very hairy issue. I've very carefully looked at
XMM/NORMA on Mach some time ago. The point here was to export/copy
pages and mark them read-only on the remote host. A page-fault would
then induce the remote pager to engage in a protocol with the originating
local pager to retain consistency. This is not easy at all, but a good
_additional_ feature for something like a user-level L4-based UVM-pager ;-).

> Michael

-Farid.

-- 
Farid Hajji -- Unix Systems and Network Admin | Phone: +49-2131-67-555
Broicherdorfstr. 83, D-41564 Kaarst, Germany  | address@hidden
- - - - - - - - - - - - - - - - - - - - - - - + - - - - - - - - - - - -
One OS To Rule Them All And In The Darkness Bind Them... --Bill Gates.