Re: [Codel] The next slice of cake

[Dave Taht <dave.taht@gmail.com>]

On Wed, Apr 1, 2015 at 9:48 PM, Jonathan Morton <chromatix99@gmail.com> wrote:
>
>> On 30 Mar, 2015, at 22:28, Dave Taht <dave.taht@gmail.com> wrote:
>>
>>> I made a lot of progress over the weekend, and got cake3 running on a sacrificial box today.  No crashes yet!
>>
>> Preliminary patches always appreciated. I plan to do a build thursday
>> for a test cycle this weekend of various things.
>
> Have a go with this, then.  I took the time to do quite a lot of little cleanup things, such as suppressing the statistics output of classes that are not in use.
>
> Tests on the Pentium-MMX show similar throughput and latency capabilities to the first version of cake, but this version has more robust behaviour in high-traffic situations - and the Diffserv logic is also far more useful.  I’m working towards getting OProfile running on at least one of my desktop-type machines; hopefully it’ll work sufficiently well on the Pentium-MMX, since that’s the easiest one to drive into saturation.

Perf is included in every kernel these days, it is way better than
oprofile, and easy to build if you are running natively.
in your linux build dir:

cd tools
make # to get more info


> The kernel patch applies on top of the previous one, but the iproute2 patch should apply to stock.  Note that the current git version of iproute2 seems to have a problem with displaying the stats from an ingress filter.

I think for benefit of the audience a fresh patch would be better. I
will see if this applies on top of my old patches.

>
>  - Jonathan Morton
>



-- 
Dave Täht
Let's make wifi fast, less jittery and reliable again!

https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb