Re: :)

[Andreas B. <and...@mu...>]

On Fri, Jun 23, 2000 at 10:59:17AM -0400, Gord Peters wrote:
> So, with me being biased by this problem, I have to ask the question: why
> wouldn't we want the host controller to always attempt to become the root
> node?  From what I understand, if there are multiple contenders for the
> root node (ie. say two PCs are hooked together), then only one will become
> root after the self-ID phase anyways, and everything should work fine
> even if they are both attempting to become root.  Perhaps I missed a
> discussion on this in my hiatus from the list?

Another bus manager may have selected one node to become root based on
some optimization decision derived from bus topology.  Unconditionally
attempting to become root would sabotage this and does not conform to
1394 for this reason.

Our real problem is that we don't yet have bus management code in the
subsystem which would deal with selecting root, among other things.

-- 
 Andreas E. Bombe <and...@mu...>    DSA key 0x04880A44
http://home.pages.de/~andreas.bombe/    http://linux1394.sourceforge.net/