Public SourceMage Discussion List

[seth AT swoolley.homeip.net]

My point was that "if you choose a mac-based system, here's a possible 
implementation"  This is an interface-centric viewpoint.

There are other choices too.  a pci- or usb-id-based system, for 
example.  usb-id-based systems depend on the order the item is inserted.  
pci-id-based systems depend on the order on the motherboard.
This is a slot-oriented viewpoint.

If you want to resolve this where both are supported by a flag, that 
might be what you're going after.  Having a slot-oriented view would get 
you duplicate macs, but wouldn't get you a mac-based configuration, 
which is actually the more natural of the two given a richer hotplugging 
setup.  It's a trade-off.  The trade-off can even be made 
configuration-file specific for complete flexibility.

Why not handle both?

On Thu, Jun 22, 2006 at 08:41:59AM +0200, Pieter Lenaerts wrote:
> On Wed, Jun 21, 2006 at 11:01:40PM -0700, seth AT swoolley.homeip.net wrote:
> 
> snipping heavily
> 
> > If a user messes with a MAC outside 
> > of this framework, they are on their own, of course.
> > 
> > This of course assumes that the driver is loaded and not reloaded (which 
> > would bring it back to its NATIVE_MAC state.
> > 
> > That covers almost any situation and still allows it to be a flexible 
> > field.  It doesn't allow duplicate names, nor duplicate MACs, but that's 
> > the flaw when you want both to be unique.
> 
> it's this (imo) abuse of the word "any" that we're trying to solve here. i
> clearly interpret "any situation" differently* ;)
> 
> the implementation right now is a rough cut of what you describe.
> it pretends to offer some advanced features but doesn't really. using
> the mac will never work in "any" situation, exactly because we know that
> we can't guarantee that we can track it.
> 
> anything that assumes that the user knows exactly when it will work but
> pretends to work at all time is crap, must be removed from our master
> branch and can return when we agree that it's ready.
> 
> 
> 
> 
> ps: * : let's agree that we keep "any" within the boundaries of present,
> working devices, so that noone starts throwing ridiculous reasons like
> kernel configs or attacking martians at anyone (not esp. you, Seth, but
> rather to all in the discussion)
> 
> --
> Pieter Lenaerts
> Source Mage GNU/Linux
> http://www.sourcemage.org - "Linux so advanced it may well be magic"



> _______________________________________________
> SM-Discuss mailing list
> SM-Discuss AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/sm-discuss