[Cerowrt-devel] [OpenWrt-Devel] [RFC PATCH] packages: Smart Queue Management for AQM Packet Scheduling and Qos from CeroWrt

[Dave Taht]

On Wed, Oct 08, 2014 at 11:10:46PM +0300, Hannu Nyman wrote:
> Dave Taht wrote on Thu Oct 2 03:49:15 CEST 2014:
> > So I don't know where to go. Certainly I'd like to see the battle
> hardened sqm scripts (which are more flexible than the C code above)
> get more widely used and in BB.
> 
> SQM seems to work ok with the current Chaos Calmer trunk.

Well, for now... I'd expect the delta to break eventually...

> 
> I have included your SQM in my trunk WNDR3700/3800 community build (
> https://forum.openwrt.org/viewtopic.php?id=28392 ) and it works ok.
> Arokh has also picked it up for his build (
> https://forum.openwrt.org/viewtopic.php?id=50914 ). So you might get
> more feedback sooner or later, if users of our builds really do try
> it.

Excellent. I'd *really* like to get some testers doing ingress shaping
at above 60-80mbits, which seems to be a brick wall we've hit on
the ar71xx and octeon, on other platforms like the arm and x86.

A tool we use a lot is netperf-wrapper.

> I feel that the best way to get wider testing and real-life usage
> for SQM would be to create a pull request in the new packages feed
> in Github, to get both the SQM itself and the Luci frontend included
> there. Being available in the packages feed would create more
> interest for the package. If it proves to work ok, the devs might
> then import it into the core Openwrt repo (where the old qos-scripts
> is).

I went to sleep pre BBrc1. I woke up, and everything was in github.

It's still not clear to me how I'd build cero from frozen BB sources,
rather than the evolving github packages.

> 
> 
> Ps. For my own build I made a slight modification to the Luci menu> item, as pure "SQM" does not say much. I changed it to "SQM QoS".
> 

Yes, SQM by itself doesn't mean enough.