Re: [PATCH 0/2] NET: Accurate packet scheduling for ATM/ADSL

[Patrick McHardy <kaber@trash.net>]

Krzysztof Matusik wrote:
> Dnia wtorek, 20 czerwca 2006 17:16, Patrick McHardy napisał:
> 
>>The code wouldn't be very complicated, it just adds some overhead. If
>>you do something like I described in my previous mail the overhead for
>>people not using it would be an additional pointer test before reading
>>skb->len. I guess we could also make it a compile time option.
>>I personally think this is something that really improves our quality
>>of implementation, after all, its "wire" resources qdiscs are meant
>>to manage.
> 
> 
> I'd love to see this one implemented. I'm using HFSC more than a year and it 
> never provides proper QoS on ATM/ADSL links; low delays can never be achieved 
> even with significant throttling below the h/w link bandwidth.

Mhh .. I trust you picked a proper clocksource and timer frequency?

> This would help a lot regarding the amount of adsl users but on the other 
> hand- there's not many HFSC implementations in real-life I guess (users seem 
> to be afraid of it's 'complexity'). 
> This idea doesn't seem look dirty- is there a chance to implement it in the 
> kernel and iproute?

I hacked up a patch yesterday. I want to do a bit more testing before
posting it, but its hard to really test the effectiveness because even
without this patch my DSL line already delivers higher throughput and
much better delay than it should (its a throttled SDSL line sold as
ADSL). I'll try to do some testing on an ethernet link now.