[altq 81] Re: When a class is delayed (undertime)

[Kenjiro Cho <kjc@csl.sony.co.jp>]

Muthu,

> 	1. I also feel that pkt_time should be added to tidle (since the next 
> time a overlimit class is allowed to send is the pkt_time + suspension time).
> (From Appendix B "Regulating overlimit classes: the details" of Experimental 
> results for class based queueing). 

Yes, you are right.

> Will I be correct if the earlier
> calculation is retained with the modification suggested by Fulvio (in the
> discussion [altq 22] and Fulvio got better results)

I'll go through the CBQ code again and make a snapshot next week.
I want to get it right this time :)

> 	2. Regarding the following code (in the patch Kenjiro posted on June
> 18): (My comments are below the code)

[snip]

> 	When a class is overlimit & cannot borrow from the ancestors (ie)
> when the class is regulated, it needs to wait offtime (to maintain the steady
> state burst). The overlimit handler (rmc_delay_action) is invoked for this 
> class (which adds offtime to undertime).

> 	But when it gets here, the estimator would have already added the 
> suspension time to undertime when the class was found overlimit. So the 
> overlimit handler just cancels the suspension time (which was added by the 
> estimator in the previous run(?)) since for a regulated class, the suspension
> time is not to be added (Appendix A1 of "Link sharing and resource management
> models for packet networks)

> 	So I feel that this code needs to be retained.

offtime is the required suspention time when avgidle is 0.
If avgidle is negative, undertime (the idle time required to bring
avgidle back to 0) should be added.

-Kenjiro