Re: [Lsr] Flex Algo merge work, IS-IS and OSPF FAD sub-TLVs

[Christian Hopps <[e-mail]>]

We aren't talking generically about TLV space here. When I raised the question I certainly intended it to be limited to times, as is the case here, where we are literally duplicating registries b/c they both refer to the same thing. I didn't realize that we'd already done this with SR IGP Algorithm registry.

I did also include talking about what (if anything) to do with the duplicated containing TLV, but it seems no one wants to go there, which is fine, and I happen to agree probably is going too far.

Thanks,
Chris.

> On May 21, 2018, at 11:11 AM, Acee Lindem (acee) <[e-mail]> wrote:
> 
> Hi Peter, Chris,
> 
> In this particular case, it may be ok as long as we just limit the code point space to the 1 octet type (i.e., 0-255 with reserved values). However, for all the reasons Peter and Les have already articulated, there will be cases where the TLV or Sub-TLV registries cannot be common. So, I have to ask myself just what are we gaining by here? The encodings are not going to be identical (for all the previously mentioned reasons) and this leaves the door open for time wasted on revisiting this issue...
> 
> Thanks,
> Acee
> 
> On 5/20/18, 9:21 AM, "Peter Psenak (ppsenak)" <[e-mail]> wrote:
> 
>    Chris,
> 
>    On 20/05/18 01:47 , Christian Hopps wrote:
>> How about an option 2c
>> 
>>   2c: Leave the encodings the way they are, and use a common registry to define the type/value semantics.
> 
>    having a combined registry that defines FAD Sub-TLVs types is fine with me.
> 
>    thanks,
>    Peter
> 
> 
>