Re: [PATCH] hashed device lookup (Does NOT meet Linus' sumission policy!)

[Andi Kleen]

On Mon, Jan 08, 2001 at 07:12:09PM +1300, Chris Wedgwood wrote:
> On Mon, Jan 08, 2001 at 06:32:14AM +0100, Andi Kleen wrote:
> 
>     I think it would be better to keep it. The ifa based alias
>     interface emulation adds minor overhead (currently it's only a
>     few lines of code, assuming we need named if addresses for other
>     reasons too, which we do) and removing it it would break a lot of
>     configuration scripts etc., for no really good gain.
> 
> It's ugly and deceptive -- eth0:0 is _not_ a separate device to eth0,
> so why pretend it is?

Who says that it names a device? It names interfaces.
There are good reasons to have names for ifas, and I see no really good
convincing reasons not to put these names into the interface name space.
(in addition it'll save a lot of people a lot of grief) 
When you're proposing a change that breaks thousands of configuration you
need a really good reason for it, and so far I cannot see one. It would 
be different if the older way needed lots of hard to maintain fragile code in 
the kernel, but that's really not the case. 

-Andi