Re: SMICng compile error on LMP MIB 7

["Adrian Farrel" <adrian@olddog.co.uk>]

Martin,

The alternative would be to introduce padding bits such as reservedOne(5), reservedTwo(6)
etc.

However, I agree there is no reason to keep the bits spaced as in the protocol.

Thanks,
Adrian


----- Original Message ----- 
From: "Martin Dubuc" <dubuc.consulting@rogers.com>
To: "Wijnen, Bert (Bert)" <bwijnen@lucent.com>; "Ccamp-wg (E-mail)" <ccamp@ops.ietf.org>
Sent: Wednesday, November 05, 2003 2:45 AM
Subject: Re: SMICng compile error on LMP MIB 7


> Bert,
>
> The definition of this bit field was based on the specification of the
> Verify Transport Mechanism field in the LMP draft. This field is a 2 byte
> bit field (16 bits). This field expresses the transport mechanisms supported
> by an implementation (multiple bits could be set to indicate multiple
> mechanisms supported). This field is technology dependent, but so far, only
> SONET/SDH standards have defined values for this field (see [LMP-TEST]). In
> the spec, there is a special value "payload" value defined as the most
> significant bit value that is supposed to apply to all technologies. I
> actually had misinterpreted the spec and coded the bit field as a 32 bit
> field. This explains why I used bit 31 for this bit. In any case, I wanted
> to use the same value as the spec in the bit field definition for the
> "payload" value. However, I don't think it really makes a difference which
> bit is used to represent the "payload" value. Bit 0 would be as good as bit
> 15 (or 31) for this purpose and would allow all bits to be contiguous. I
> have already digressed for the SONET/SDH test spec by keeping the currently
> defined SONET/SDH bits contiguous. In [LMP-TEST], there are five values
> specified over 8 bits (3 bits are reserved). I could have mapped the bits in
> the bit field the same way (for instance DCC section overhead bytes on bit
> 1), but I think it is better to have them all contiguous, the main reason
> being that it is quite possible that other technologies will reuse the same
> bits. This would prevent us from using the bit fields specified in the
> standards because they are likely to be overloaded. In the end, the
> implementor will have to perform some mapping.
>
> I can update the MIB to reflect this.
>
> Martin
>
> ----- Original Message -----
> From: "Wijnen, Bert (Bert)" <bwijnen@lucent.com>
> To: "Ccamp-wg (E-mail)" <ccamp@ops.ietf.org>
> Sent: Tuesday, November 04, 2003 2:22 PM
> Subject: SMICng compile error on LMP MIB 7
>
>
> > The SMICng compile tells me:
> >
> >   E: f(lmp.mi2), (1430,22) Named bits for BITS must be in contiguous
> positions
> >
> > So why is there a gap ??
> >
> >
> > Thanks,
> > Bert
> >
> >
> >
>
>
>