In the opinion, I don't see any place that the hack is specifically
called out in the exported interfaces. Nearly all the VLAN specific
interfaces are given Contract Private binding, except for
DL_SUN_VLAN_BIND, which is Unstable.
Notably, I don't know if the hack has ever been formally published to
customers (although it probably has been in qfe and ce manuals -- I just
haven't looked there.)
-- Garrett
James Carlson wrote:
> Garrett D'Amore writes:
>
>> Has any thought been given to the notion of just eliminating the vlan
>> hack altogether?
>>
>
> More than once, sure.
>
>
>> It seems like there are better ways to express vlans, and the "hack" has
>> always been a bit unsettling to me. Now that we're on a
>> pseudo-major-release boundary (OpenSolaris), couldn't this be a time to
>> break that compatibility?
>>
>
> My understanding is that it's not open season on all possible
> incompatibilities, or even those that just please us, but rather only
> those few that are strictly necessary to achieve some higher level of
> "familiarity" (or perhaps "parity") with Linux.
>
> I don't see how the existence of the VLAN hack is related to Linux in
> any substantial way.
>
> I agree with you that it's an ugly thing and certainly does cause
> problems, but that's not something I can see solving just because of
> the Linux compatibility goal.
>
> More importantly, eliminating the VLAN hack does nothing to solve the
> problem I've described, so I've got nothing to gain by it.
>
> (I've heard that Crossbow aims to remove the hack. I don't know
> whether that's the final goal, a strict requirement, or just something
> that discussed with the ARC. In any event, that's not this project.)
>
>
>> (Actually, although everyone knows about, I'm
>> having trouble locating any documentation that specifically refers to
>> the PPA-based VLAN hack. So maybe its not technically interface breakage
>> anyway.)
>>
>
> I cited it in the previous message -- it's defined in PSARC 2000/147.
>
>
_______________________________________________
networking-discuss mailing list
[email protected]
