Cake - FQ_codel the next generation
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: cake@lists.bufferbloat.net
Subject: [Cake] Fwd: out of tree "cake" now buildable on multiple linux versions
Date: Wed, 22 Apr 2015 09:29:14 -0700	[thread overview]
Message-ID: <CAA93jw5owYgPA8GxinDEVPf3-pTyWi4YtRws6FQWEMcYNmwDVw@mail.gmail.com> (raw)
In-Reply-To: <CAOaVG17B17CnBBLc6R6SXKnnhMJsq=89V_tT4ZS0X=abW5PkNQ@mail.gmail.com>

jonathon: please do try to meet kernel coding standards....


---------- Forwarded message ----------
From: Stephen Hemminger <stephen@networkplumber.org>
Date: Wed, Apr 22, 2015 at 8:50 AM
Subject: Re: [Cake] out of tree "cake" now buildable on multiple linux versions
To: Dave Taht <dave.taht@gmail.com>


Run checkpatch and scream.

On Wed, Apr 22, 2015 at 8:49 AM, Stephen Hemminger
<stephen@networkplumber.org> wrote:
>
> You do know the sch_cake source needs a pass through the indentor to keep the style gods at bay.
>
> On Wed, Apr 22, 2015 at 1:16 AM, Dave Taht <dave.taht@gmail.com> wrote:
>>
>> Do a:
>>
>> git clone https://github.com/dtaht/sch_cake.git
>>
>>  # IF you have kernel source installed to leverage, adding cake is as easy as:
>>
>> cd sch_cake
>> make; sudo make install
>>
>> # How do you install kernel headers on various linuxes?
>>
>> NOTE: I did not take the time to get the linux version checks
>> exactly correct, but did manage to get it to compile on linux
>> 3.13 and linux 3.18 without error, on x86_64.
>>
>> To use it properly, you will also need to build and install the
>> iproute with cake support:
>>
>> git clone git://kau.toke.dk/cake/iproute2/ iproute2-cake
>> cd iproute2-cake
>> #edit out the sfq_codel entry in tc/Makefile (for now)
>> make
>> make install
>>
>> #Then:
>>
>> tc qdisc add dev whatever root cake # and whatever options
>>
>> Some example results while a rrul test was running:
>>
>> d@nuc-client:~/git/iproute2-cake$ ./tc/tc -s qdisc show dev eth0
>> qdisc cake 8002: root refcnt 2 unlimited diffserv4 flows
>>  Sent 13895939355 bytes 9605458 pkt (dropped 194, overlimits 0 requeues 0)
>>  backlog 318798b 26p requeues 0
>>            Class 0     Class 1     Class 2     Class 3
>>   rate          0bit        0bit        0bit        0bit
>>   target       5.0ms       5.0ms       5.0ms       5.0ms
>> interval     105.0ms     105.0ms     105.0ms     105.0ms
>> Pk delay       6.0ms       5.0ms         1us       2.3ms
>> Av delay       1.5ms       1.4ms         0us       654us
>> Sp delay       317us       201us         0us       141us
>>   pkts        128068      316663          56      236467
>> way inds           0           0           0           0
>> way miss           3          40           3           7
>> way cols           0           0           0           0
>>   bytes     45317244 11531124242        5524  2320607748
>>   drops          173          21           0           0
>>   marks            0           0           0           0
>>
>> A problem I have is in reconciling the netperf-wrapper plots which hit
>> a minimum of 2ms for sparse flows, where I typically see ~200us delay
>> in the qdiscs themselves. I have generally not cared at all about
>> about anything less than 3ms prior to now.
>>
>> qdisc cake 8003: root refcnt 2 bandwidth 920Mbit diffserv4 flows
>>  Sent 2298586363 bytes 1826648 pkt (dropped 45, overlimits 474931 requeues 0)
>>  backlog 468331b 25p requeues 0
>>            Class 0     Class 1     Class 2     Class 3
>>   rate       920Mbit  862500Kbit     690Mbit     230Mbit
>>   target       5.0ms       5.0ms       5.0ms       5.0ms
>> interval     105.0ms     105.0ms     105.0ms     105.0ms
>> Pk delay       4.6ms       3.2ms         0us       5.2ms
>> Av delay       1.5ms       1.2ms         0us       1.6ms
>> Sp delay       303us       290us         0us       191us
>>   pkts        110498      111953           0      181478
>> way inds           0           0           0           0
>> way miss           3          31           0           6
>> way cols           0           0           0           0
>>   bytes    347279080  1377499174           0   574954712
>>   drops           14           0           0          31
>>   marks            0           0           0           0
>>
>> Have I mentioned how much I hate offloads? see the backlog relative to
>> the number of "packets".
>>
>> Still, I do think developing this out of tree will help a lot, after
>> we get kernel versions straightened out more. Next up is trying to get
>> it to build on openwrt, also out of tree.
>>
>> --
>> Dave Täht
>> Open Networking needs **Open Source Hardware**
>>
>> https://plus.google.com/u/0/+EricRaymond/posts/JqxCe2pFr67
>> _______________________________________________
>> Cake mailing list
>> Cake@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cake
>
>



-- 
Dave Täht
Open Networking needs **Open Source Hardware**

https://plus.google.com/u/0/+EricRaymond/posts/JqxCe2pFr67

  parent reply	other threads:[~2015-04-22 16:29 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-22  8:16 [Cake] " Dave Taht
2015-04-22  8:37 ` Toke Høiland-Jørgensen
     [not found] ` <CAOaVG14ESn-Se2T2e6CenuBfKJ-FuWYHNLwJawJtEkJ99ckuWA@mail.gmail.com>
     [not found]   ` <CAOaVG17B17CnBBLc6R6SXKnnhMJsq=89V_tT4ZS0X=abW5PkNQ@mail.gmail.com>
2015-04-22 16:29     ` Dave Taht [this message]
2015-04-22 18:56       ` [Cake] Fwd: " Jonathan Morton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/cake.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAA93jw5owYgPA8GxinDEVPf3-pTyWi4YtRws6FQWEMcYNmwDVw@mail.gmail.com \
    --to=dave.taht@gmail.com \
    --cc=cake@lists.bufferbloat.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox