Re: [tcpm] Comments on draft-blanton-tcpm-3517bis-01

[" Ilpo Järvinen " <[e-mail]>]

On Wed, 13 Apr 2011, Mark Allman wrote:

> 
> > ...I see, now we're back in discussing what this DupAcks counter is 
> > for, right? It was added in sack-recovery-entry ID and had proper 
> > explation of its purpose there in place. The DupAcks counter is to handle 
> > case where the sender is sending smaller than SMSS sized segments. 
> > ...Perhaps it would be useful to explain it like the sack-recovery-entry 
> > did because otherwise this will just keep confusing people?
> 
> Thanks for the reminder... Ethan also noted that this is for the small
> packet case.  I forgot about that.  Why don't we just add a quick note
> to the document after the (1) test.  I.e., something like ...
> 
>    (1) If DupAcks >= DupThresh, go to step (4).
>    
>        Note: This check covers the case when a TCP is not sending
>        full-sized packets and therefore IsLost() (next step) may be

packets => segments

>        hard-pressed to declare a segment as lost.

Something simple like that would certainly be useful. 

-- 
 i.