[Cerowrt-devel] an accurate broadband speed test convo on slashdot

[Dave Taht]

sigh

http://ask.slashdot.org/story/14/10/09/1241236/ask-slashdot-an-accurate-broadband-speed-test

-- 
Dave Täht

https://www.bufferbloat.net/projects/make-wifi-fast
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

Re: [Cerowrt-devel] Fwd: next round of SQM changes

[Sebastian Moeller]

Hi Valdis,


On Oct 9, 2014, at 22:54 , valdis.kletni...@vt.edu wrote:

> On Thu, 09 Oct 2014 22:37:30 +0200, Sebastian Moeller said:
>> I am again in need of testers for SQM.
> 
> Sure, what's needed to get the testable bits onto a 3.10.50-1 box?
> 

The attached archaic contains all SQM related files that I changed compared to 
stock cerowrt 3.10.50-1. So all that should be required is to replace the files 
in /usr with the files from the archive. The change in /usr/lib/lua just 
exposes more interfaces to the GUI (and I am not yet sure whether that is a 
sane idea or not). The changes in /usr/lib/sqm implement the changes: more 
robots tear down of active SQM instances, and a switch from indexes IFB devices 
ala ifb0 to names fib devices ala ifb4ge00 to sanitize IFB handling somewhat. 
Hopefully these last changes make cerowrt way more robust against accidental 
left over shaping instances.
Thanks for volunteering to test…

Best Regards
Sebastian




sqm_test_141009.tar.gz
Description: GNU Zip compressed data
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

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

[Sebastian Moeller]

Hi Luis,


On Oct 9, 2014, at 20:05 , Luis E. Garcia  wrote:

> Please excuse my dumb question or just point me to the right documentation.
> But is there an easy way to run AQM/Codel on BB 14.07 ?? I'm running BB on a 
> MyNETN750 that looks like it meets the HW requirements.

There probably is, I assume you need to install ip-full from the BB 
repository, and then copy 
http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages/luci-app-sqm_3-1_all.ipk
 and 
http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages/sqm-scripts_6-1_all.ipk
 onto your router and try:
opkg install sqm-scripts_6-1_all.ipk
and:
opkg install luci-app-sqm_3-1_all.ipk

You might need to play with opkg options, but in theory it should work… There 
probably also is a nicer way like adding a stanza similar to:
src/gz toronto 
http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages
to /etc/opkg.conf
but I do not know the exact incantation of that, maybe hnyman, or Dave know the 
correct answer to this..

Best Regards
Sebastian

> 
> Regards,
> Luis Garcia
> 
> On Thu, Oct 9, 2014 at 12:57 PM, Sebastian Moeller  wrote:
> Hi Richard,
> 
> On Oct 9, 2014, at 18:42 , Richard Smith  wrote:
> 
> > On 10/06/2014 03:41 PM, Sebastian Moeller wrote:
> >
> >>>
> >>> I have spare routers that I can run OpenWRT or CeroWRt on and I'm
> >>> setup to test with netperf, netperf-wrapper on my local network (
> >>> desktop -> router -> laptop )  it's Gbit so I can easily saturate
> >>> the router.
> >>
> >> That sounds great. I think the first test should be to run SQM under
> >> cerowrt, so you get a feel of how things should look. I typically run
> >> netperf-wrapper rrul tests (for ipv4 and if available for ipv6)
> >> through cerowrt with different settings for SQM. A second step then
> >> is to instal SQM-scripts under openwrt and check whether the same
> >> settings produce the same results ;)
> >
> > Ok.  I'll get my spare WNDR3700v2 up running the latest cerowrt.  Should be 
> > in the next few days.  I'm a bit strapped for time right now.
> 
> No hurry, I really appreciate your help and there is no time pressure 
> (rather the other way around, I do this for a hobby, and real life leaves 
> almost no time for  that ;) )
> 
> >
> >> So just let me know what you are willing/ready to test and we will
> >> take it from there okay? (I would already be a happy camper if you
> >> could just install the current SQM-scripts on openwrt and just send
> >> me the output of “logread” after installing and activating SQM, as
> >> well as the output from “tc -d qdisc” before and after enabling SQM,
> >> and finally the output of running “/etc/init.d/sqm stop ;
> >> /etc/init.d/sqm start” on the router’s console; that hopefully works
> >> or at least gives some indication what might be off. If you could
> >> throw in a quick netperf-wrapper RRUL test through the router I will
> >> be most delighted ;))
> >
> > Ok. That sounds like a good starter step.  First question: Where are the 
> > current SQM scripts located and what's the process for installing them?  
> > Are they packages or stuff that I scp over and manually install?
> 
> So the repository lives on:
> https://github.com/dtaht/ceropackages-3.10
>  I just followed Dave’s advice and used:
> git clone g...@github.com:dtaht/ceropackages-3.3.git
> to get my own version to play around with. Oh, SQM basically is the work of 
> Dave (who wrote the scripts doing the actual AQM&QOS work) and Toke you 
> packaged all this nicely and created the GUI for it (all I ever did was 
> shuffling the GUI around a bit). I just started hacking around the files from 
> sqm-scripts and luck-app-sqm on my cerowrt (so I could immediately do some 
> testing), so I never got around building actual packages (I assume you need 
> to be setup to build whole openwrt/cerowrt images and that is outside the 
> scope of my recreational coding), But that said if you copy the files from 
> ceropackages-3.10/net/sqm-scripts/files to the matching directory on your 
> router you should have a working sqm-script. Or just use a cerowrt built as 
> SQM is standard and potentially hnyman’s build ( 
> https://forum.openwrt.org/viewtopic.php?id=28392 ) that also includes SQM.
> 
> Best Regards
> Sebastian
> 
> >
> > I'll let you know when I've taken a stab at getting that working.
> >
> > > Richard A. Smith  
> > > Former One Laptop per Child
> >
> > Just FYI: I'm slowly switching my list subs over from my previous 
> > @laptop.org address to my personal address.  Just switched this one over 
> > thus the email address change.
> >
> > --
> > Richard A. Smith
> ___
> openwrt-devel mailing list
> openwrt-de...@lists.openwrt.org
> https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
> 
> ___
> openwrt-devel mailing list
> openwrt-de...@lists.openwrt.org
> https://list

Re: [Cerowrt-devel] Fwd: next round of SQM changes

[Valdis . Kletnieks]

On Thu, 09 Oct 2014 22:37:30 +0200, Sebastian Moeller said:
> I am again in need of testers for SQM.

Sure, what's needed to get the testable bits onto a 3.10.50-1 box?



pgpr8SjZhldAC.pgp
Description: PGP signature
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

[Cerowrt-devel] next round of SQM changes

[Sebastian Moeller]

Hi List,

I am again in need of testers for SQM. So far SQM tried to deduce from 
/etc/config/sqm which interfaces it potentially needed to stop itself on, which 
did not work very reliably, since it is easy to just change the interface name 
in an existing config file and all traces of the interface on which SQM was 
started originally disappear. So instead now, SQM will stop itself on all 
interfaces it currently runs on, and then will only start itself on those 
interfaces that are actually configured and enabled in the GUI/UCI file. That 
hopefully gets rid of the spurious leftover HTBs and IFBs that some users had 
noticed.
I would be great if anybody experiences problems with the new code to 
holler so I can go fix things up…

Best Regards
Sebastian
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

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

[Luis E. Garcia]

Please excuse my dumb question or just point me to the right documentation.
But is there an easy way to run AQM/Codel on BB 14.07 ?? I'm running BB on
a MyNETN750 that looks like it meets the HW requirements.

Regards,
Luis Garcia

On Thu, Oct 9, 2014 at 12:57 PM, Sebastian Moeller  wrote:

> Hi Richard,
>
> On Oct 9, 2014, at 18:42 , Richard Smith  wrote:
>
> > On 10/06/2014 03:41 PM, Sebastian Moeller wrote:
> >
> >>>
> >>> I have spare routers that I can run OpenWRT or CeroWRt on and I'm
> >>> setup to test with netperf, netperf-wrapper on my local network (
> >>> desktop -> router -> laptop )  it's Gbit so I can easily saturate
> >>> the router.
> >>
> >> That sounds great. I think the first test should be to run SQM under
> >> cerowrt, so you get a feel of how things should look. I typically run
> >> netperf-wrapper rrul tests (for ipv4 and if available for ipv6)
> >> through cerowrt with different settings for SQM. A second step then
> >> is to instal SQM-scripts under openwrt and check whether the same
> >> settings produce the same results ;)
> >
> > Ok.  I'll get my spare WNDR3700v2 up running the latest cerowrt.  Should
> be in the next few days.  I'm a bit strapped for time right now.
>
> No hurry, I really appreciate your help and there is no time
> pressure (rather the other way around, I do this for a hobby, and real life
> leaves almost no time for  that ;) )
>
> >
> >> So just let me know what you are willing/ready to test and we will
> >> take it from there okay? (I would already be a happy camper if you
> >> could just install the current SQM-scripts on openwrt and just send
> >> me the output of “logread” after installing and activating SQM, as
> >> well as the output from “tc -d qdisc” before and after enabling SQM,
> >> and finally the output of running “/etc/init.d/sqm stop ;
> >> /etc/init.d/sqm start” on the router’s console; that hopefully works
> >> or at least gives some indication what might be off. If you could
> >> throw in a quick netperf-wrapper RRUL test through the router I will
> >> be most delighted ;))
> >
> > Ok. That sounds like a good starter step.  First question: Where are the
> current SQM scripts located and what's the process for installing them?
> Are they packages or stuff that I scp over and manually install?
>
> So the repository lives on:
> https://github.com/dtaht/ceropackages-3.10
>  I just followed Dave’s advice and used:
> git clone g...@github.com:dtaht/ceropackages-3.3.git
> to get my own version to play around with. Oh, SQM basically is the work
> of Dave (who wrote the scripts doing the actual AQM&QOS work) and Toke you
> packaged all this nicely and created the GUI for it (all I ever did was
> shuffling the GUI around a bit). I just started hacking around the files
> from sqm-scripts and luck-app-sqm on my cerowrt (so I could immediately do
> some testing), so I never got around building actual packages (I assume you
> need to be setup to build whole openwrt/cerowrt images and that is outside
> the scope of my recreational coding), But that said if you copy the files
> from ceropackages-3.10/net/sqm-scripts/files to the matching directory on
> your router you should have a working sqm-script. Or just use a cerowrt
> built as SQM is standard and potentially hnyman’s build (
> https://forum.openwrt.org/viewtopic.php?id=28392 ) that also includes SQM.
>
> Best Regards
> Sebastian
>
> >
> > I'll let you know when I've taken a stab at getting that working.
> >
> > > Richard A. Smith  
> > > Former One Laptop per Child
> >
> > Just FYI: I'm slowly switching my list subs over from my previous @
> laptop.org address to my personal address.  Just switched this one over
> thus the email address change.
> >
> > --
> > Richard A. Smith
> ___
> openwrt-devel mailing list
> openwrt-de...@lists.openwrt.org
> https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
>
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

[Cerowrt-devel] Fwd: next round of SQM changes

[Sebastian Moeller]

Hi List,

I am again in need of testers for SQM. So far SQM tried to deduce from 
/etc/config/sqm which interfaces it potentially needed to stop itself on, which 
did not work very reliably, since it is easy to just change the interface name 
in an existing config file and all traces of the interface on which SQM was 
started originally disappear. So instead now, SQM will stop itself on all 
interfaces it currently runs on, and then will only start itself on those 
interfaces that are actually configured and enabled in the GUI/UCI file. That 
hopefully gets rid of the spurious leftover HTBs and IFBs that some users had 
noticed.
I would be great if anybody experiences problems with the new code to 
holler so I can go fix things up…

Best Regards
Sebastian

P.S.: Dave I first sent this from the wrong emial account, please just delete 
the message held in moderation ;)
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

Re: [Cerowrt-devel] bulk packet transmission

[Dave Taht]

I have some hope that the skb->xmit_more API could be used to make
aggregating packets in wifi on an AP saner. (my vision for it was that
the overlying qdisc would set xmit_more while it still had packets
queued up for a given station and then stop and switch to the next.
But the rest of the infrastructure ended up pretty closely tied to
BQL)

Jesper just wrote a nice piece about it also.
http://netoptimizer.blogspot.com/2014/10/unlocked-10gbps-tx-wirespeed-smallest.html

It was nice to fool around at 10GigE for a while! And netperf-wrapper
scales to this speed also! :wow:

I do worry that once sch_fq and fq_codel support is added that there
will be side effects. I would really like - now that there are al
these people profiling things at this level to see profiles including
those qdiscs.

/me goes grumbling back to thinking about wifi.

On Thu, Oct 9, 2014 at 12:40 PM, David Lang  wrote:
> lwn.net has an article about a set of new patches that avoid some locking
> overhead by transmitting multiple packets at once.
>
> It doesn't work for things with multiple queues (like fq_codel) in it's
> current iteration, but it sounds like something that should be looked at and
> watched for latency related issues.
>
> http://lwn.net/Articles/615238/
>
> David Lang
> ___
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Dave Täht

https://www.bufferbloat.net/projects/make-wifi-fast
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

[Cerowrt-devel] bulk packet transmission

[David Lang]

lwn.net has an article about a set of new patches that avoid some locking 
overhead by transmitting multiple packets at once.


It doesn't work for things with multiple queues (like fq_codel) in it's current 
iteration, but it sounds like something that should be looked at and watched for 
latency related issues.


http://lwn.net/Articles/615238/

David Lang
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

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

[Sebastian Moeller]

Ooops,


On Oct 9, 2014, at 20:13 , Dave Taht  wrote:

> 3.3 is obsolete, use 3.10.

What Dave says, I should have actually looked at the copied link…

Sorry
Sebastian


> 
> On Thu, Oct 9, 2014 at 10:57 AM, Sebastian Moeller  wrote:
>> Hi Richard,
>> 
>> On Oct 9, 2014, at 18:42 , Richard Smith  wrote:
>> 
>>> On 10/06/2014 03:41 PM, Sebastian Moeller wrote:
>>> 
> 
> I have spare routers that I can run OpenWRT or CeroWRt on and I'm
> setup to test with netperf, netperf-wrapper on my local network (
> desktop -> router -> laptop )  it's Gbit so I can easily saturate
> the router.
 
 That sounds great. I think the first test should be to run SQM under
 cerowrt, so you get a feel of how things should look. I typically run
 netperf-wrapper rrul tests (for ipv4 and if available for ipv6)
 through cerowrt with different settings for SQM. A second step then
 is to instal SQM-scripts under openwrt and check whether the same
 settings produce the same results ;)
>>> 
>>> Ok.  I'll get my spare WNDR3700v2 up running the latest cerowrt.  Should be 
>>> in the next few days.  I'm a bit strapped for time right now.
>> 
>>No hurry, I really appreciate your help and there is no time pressure 
>> (rather the other way around, I do this for a hobby, and real life leaves 
>> almost no time for  that ;) )
>> 
>>> 
 So just let me know what you are willing/ready to test and we will
 take it from there okay? (I would already be a happy camper if you
 could just install the current SQM-scripts on openwrt and just send
 me the output of “logread” after installing and activating SQM, as
 well as the output from “tc -d qdisc” before and after enabling SQM,
 and finally the output of running “/etc/init.d/sqm stop ;
 /etc/init.d/sqm start” on the router’s console; that hopefully works
 or at least gives some indication what might be off. If you could
 throw in a quick netperf-wrapper RRUL test through the router I will
 be most delighted ;))
>>> 
>>> Ok. That sounds like a good starter step.  First question: Where are the 
>>> current SQM scripts located and what's the process for installing them?  
>>> Are they packages or stuff that I scp over and manually install?
>> 
>>So the repository lives on:
>> https://github.com/dtaht/ceropackages-3.10
>> I just followed Dave’s advice and used:
>> git clone g...@github.com:dtaht/ceropackages-3.3.git
>> to get my own version to play around with. Oh, SQM basically is the work of 
>> Dave (who wrote the scripts doing the actual AQM&QOS work) and Toke you 
>> packaged all this nicely and created the GUI for it (all I ever did was 
>> shuffling the GUI around a bit). I just started hacking around the files 
>> from sqm-scripts and luck-app-sqm on my cerowrt (so I could immediately do 
>> some testing), so I never got around building actual packages (I assume you 
>> need to be setup to build whole openwrt/cerowrt images and that is outside 
>> the scope of my recreational coding), But that said if you copy the files 
>> from ceropackages-3.10/net/sqm-scripts/files to the matching directory on 
>> your router you should have a working sqm-script. Or just use a cerowrt 
>> built as SQM is standard and potentially hnyman’s build ( 
>> https://forum.openwrt.org/viewtopic.php?id=28392 ) that also includes SQM.
>> 
>> Best Regards
>>Sebastian
>> 
>>> 
>>> I'll let you know when I've taken a stab at getting that working.
>>> 
 Richard A. Smith  
 Former One Laptop per Child
>>> 
>>> Just FYI: I'm slowly switching my list subs over from my previous 
>>> @laptop.org address to my personal address.  Just switched this one over 
>>> thus the email address change.
>>> 
>>> --
>>> Richard A. Smith
>> 
>> ___
>> Cerowrt-devel mailing list
>> Cerowrt-devel@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cerowrt-devel
> 
> 
> 
> -- 
> Dave Täht
> 
> https://www.bufferbloat.net/projects/make-wifi-fast

___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

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

[Dave Taht]

3.3 is obsolete, use 3.10.

On Thu, Oct 9, 2014 at 10:57 AM, Sebastian Moeller  wrote:
> Hi Richard,
>
> On Oct 9, 2014, at 18:42 , Richard Smith  wrote:
>
>> On 10/06/2014 03:41 PM, Sebastian Moeller wrote:
>>

 I have spare routers that I can run OpenWRT or CeroWRt on and I'm
 setup to test with netperf, netperf-wrapper on my local network (
 desktop -> router -> laptop )  it's Gbit so I can easily saturate
 the router.
>>>
>>> That sounds great. I think the first test should be to run SQM under
>>> cerowrt, so you get a feel of how things should look. I typically run
>>> netperf-wrapper rrul tests (for ipv4 and if available for ipv6)
>>> through cerowrt with different settings for SQM. A second step then
>>> is to instal SQM-scripts under openwrt and check whether the same
>>> settings produce the same results ;)
>>
>> Ok.  I'll get my spare WNDR3700v2 up running the latest cerowrt.  Should be 
>> in the next few days.  I'm a bit strapped for time right now.
>
> No hurry, I really appreciate your help and there is no time pressure 
> (rather the other way around, I do this for a hobby, and real life leaves 
> almost no time for  that ;) )
>
>>
>>> So just let me know what you are willing/ready to test and we will
>>> take it from there okay? (I would already be a happy camper if you
>>> could just install the current SQM-scripts on openwrt and just send
>>> me the output of “logread” after installing and activating SQM, as
>>> well as the output from “tc -d qdisc” before and after enabling SQM,
>>> and finally the output of running “/etc/init.d/sqm stop ;
>>> /etc/init.d/sqm start” on the router’s console; that hopefully works
>>> or at least gives some indication what might be off. If you could
>>> throw in a quick netperf-wrapper RRUL test through the router I will
>>> be most delighted ;))
>>
>> Ok. That sounds like a good starter step.  First question: Where are the 
>> current SQM scripts located and what's the process for installing them?  Are 
>> they packages or stuff that I scp over and manually install?
>
> So the repository lives on:
> https://github.com/dtaht/ceropackages-3.10
>  I just followed Dave’s advice and used:
> git clone g...@github.com:dtaht/ceropackages-3.3.git
> to get my own version to play around with. Oh, SQM basically is the work of 
> Dave (who wrote the scripts doing the actual AQM&QOS work) and Toke you 
> packaged all this nicely and created the GUI for it (all I ever did was 
> shuffling the GUI around a bit). I just started hacking around the files from 
> sqm-scripts and luck-app-sqm on my cerowrt (so I could immediately do some 
> testing), so I never got around building actual packages (I assume you need 
> to be setup to build whole openwrt/cerowrt images and that is outside the 
> scope of my recreational coding), But that said if you copy the files from 
> ceropackages-3.10/net/sqm-scripts/files to the matching directory on your 
> router you should have a working sqm-script. Or just use a cerowrt built as 
> SQM is standard and potentially hnyman’s build ( 
> https://forum.openwrt.org/viewtopic.php?id=28392 ) that also includes SQM.
>
> Best Regards
> Sebastian
>
>>
>> I'll let you know when I've taken a stab at getting that working.
>>
>> > Richard A. Smith  
>> > Former One Laptop per Child
>>
>> Just FYI: I'm slowly switching my list subs over from my previous 
>> @laptop.org address to my personal address.  Just switched this one over 
>> thus the email address change.
>>
>> --
>> Richard A. Smith
>
> ___
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Dave Täht

https://www.bufferbloat.net/projects/make-wifi-fast
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

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

[Sebastian Moeller]

Hi Richard,

On Oct 9, 2014, at 18:42 , Richard Smith  wrote:

> On 10/06/2014 03:41 PM, Sebastian Moeller wrote:
> 
>>> 
>>> I have spare routers that I can run OpenWRT or CeroWRt on and I'm
>>> setup to test with netperf, netperf-wrapper on my local network (
>>> desktop -> router -> laptop )  it's Gbit so I can easily saturate
>>> the router.
>> 
>> That sounds great. I think the first test should be to run SQM under
>> cerowrt, so you get a feel of how things should look. I typically run
>> netperf-wrapper rrul tests (for ipv4 and if available for ipv6)
>> through cerowrt with different settings for SQM. A second step then
>> is to instal SQM-scripts under openwrt and check whether the same
>> settings produce the same results ;)
> 
> Ok.  I'll get my spare WNDR3700v2 up running the latest cerowrt.  Should be 
> in the next few days.  I'm a bit strapped for time right now.

No hurry, I really appreciate your help and there is no time pressure 
(rather the other way around, I do this for a hobby, and real life leaves 
almost no time for  that ;) )

> 
>> So just let me know what you are willing/ready to test and we will
>> take it from there okay? (I would already be a happy camper if you
>> could just install the current SQM-scripts on openwrt and just send
>> me the output of “logread” after installing and activating SQM, as
>> well as the output from “tc -d qdisc” before and after enabling SQM,
>> and finally the output of running “/etc/init.d/sqm stop ;
>> /etc/init.d/sqm start” on the router’s console; that hopefully works
>> or at least gives some indication what might be off. If you could
>> throw in a quick netperf-wrapper RRUL test through the router I will
>> be most delighted ;))
> 
> Ok. That sounds like a good starter step.  First question: Where are the 
> current SQM scripts located and what's the process for installing them?  Are 
> they packages or stuff that I scp over and manually install?

So the repository lives on:
https://github.com/dtaht/ceropackages-3.10
 I just followed Dave’s advice and used:
git clone g...@github.com:dtaht/ceropackages-3.3.git
to get my own version to play around with. Oh, SQM basically is the work of 
Dave (who wrote the scripts doing the actual AQM&QOS work) and Toke you 
packaged all this nicely and created the GUI for it (all I ever did was 
shuffling the GUI around a bit). I just started hacking around the files from 
sqm-scripts and luck-app-sqm on my cerowrt (so I could immediately do some 
testing), so I never got around building actual packages (I assume you need to 
be setup to build whole openwrt/cerowrt images and that is outside the scope of 
my recreational coding), But that said if you copy the files from 
ceropackages-3.10/net/sqm-scripts/files to the matching directory on your 
router you should have a working sqm-script. Or just use a cerowrt built as SQM 
is standard and potentially hnyman’s build ( 
https://forum.openwrt.org/viewtopic.php?id=28392 ) that also includes SQM. 

Best Regards
Sebastian

> 
> I'll let you know when I've taken a stab at getting that working.
> 
> > Richard A. Smith  
> > Former One Laptop per Child
> 
> Just FYI: I'm slowly switching my list subs over from my previous @laptop.org 
> address to my personal address.  Just switched this one over thus the email 
> address change.
> 
> -- 
> Richard A. Smith

___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

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

[Richard Smith]

On 10/08/2014 11:01 PM, Dave Taht wrote:


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.


I can do ingress shaping tests at above 80mbits at work since we now 
have a 100mbit downlink, but I don't have cerowrt hardware that runs 
fast enough to keep up.  Also I would only be able to do them on the 
weekend or late at night when no one else is in the office when I can 
swap out the router.  I can't leave experimental setups alone running 
our office link.


FYI: I've purchased a Nighthawk x6 for some testing.  I did x6 even 
though its CPU speed is slower than the x4 because I want to see if the 
twin 5Ghz radios and the promise of automatic load balancing helps on 
the wireless side (we have a lot of 5Ghz wireless clients)  If it 
doesn't hold up as a router we may still use it as just an AP.


Like you mentioned the stock firmware is not the greatest and there's no 
actual bandwidth control settings.  Hopefully openwrt support will show 
up soon.


My initial tests with Gbit pipes on either side are certainly impressive 
as it can route at over 900 Mbs while keeping latency at 12ms or lower. 
 Tonight I'm planning on testing it with my home Comcast cable link and 
see if the low-latency holds.


--
Richard A. Smith
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

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

[Richard Smith]

On 10/06/2014 03:41 PM, Sebastian Moeller wrote:



I have spare routers that I can run OpenWRT or CeroWRt on and I'm
setup to test with netperf, netperf-wrapper on my local network (
desktop -> router -> laptop )  it's Gbit so I can easily saturate
the router.


That sounds great. I think the first test should be to run SQM under
cerowrt, so you get a feel of how things should look. I typically run
netperf-wrapper rrul tests (for ipv4 and if available for ipv6)
through cerowrt with different settings for SQM. A second step then
is to instal SQM-scripts under openwrt and check whether the same
settings produce the same results ;)


Ok.  I'll get my spare WNDR3700v2 up running the latest cerowrt.  Should 
be in the next few days.  I'm a bit strapped for time right now.



So just let me know what you are willing/ready to test and we will
take it from there okay? (I would already be a happy camper if you
could just install the current SQM-scripts on openwrt and just send
me the output of “logread” after installing and activating SQM, as
well as the output from “tc -d qdisc” before and after enabling SQM,
and finally the output of running “/etc/init.d/sqm stop ;
/etc/init.d/sqm start” on the router’s console; that hopefully works
or at least gives some indication what might be off. If you could
throw in a quick netperf-wrapper RRUL test through the router I will
be most delighted ;))


Ok. That sounds like a good starter step.  First question: Where are the 
current SQM scripts located and what's the process for installing them? 
 Are they packages or stuff that I scp over and manually install?


I'll let you know when I've taken a stab at getting that working.

> Richard A. Smith  
> Former One Laptop per Child

Just FYI: I'm slowly switching my list subs over from my previous 
@laptop.org address to my personal address.  Just switched this one over 
thus the email address change.


--
Richard A. Smith
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

Re: [Cerowrt-devel] I'd like to add you to my professional network on LinkedIn

[Kenneth Finnegan]

It would appear that I pocket dial added my entire Gmail contacts list to
LinkedIn.

I am so tremendously sorry about this. Please ignore this invite with my
utmost apologies.

--
Kenneth Finnegan
http://blog.thelifeofkenneth.com/

On Thu, Oct 9, 2014 at 4:52 AM, Kenneth Finnegan <
kennethfinnegan2...@gmail.com> wrote:

>
>
>[image: LinkedIn]
> 
>
>
>
>
>
>
>  I'd like to add you to my professional network on LinkedIn.
>
>- Kenneth
>
>
>   Confirm that you know Kenneth
> 
>
>
>
>  [image: Kenneth Finnegan]
>
> Kenneth Finnegan  Founder at PhirePhly Design  San Francisco Bay Area
>
>
>
>  You received an invitation to connect. LinkedIn will use your email
> address to make suggestions to our members in features like People You May
> Know. Unsubscribe
> 
>   Learn
> why we included this.
> 
>   If
> you need assistance or have questions, please contact LinkedIn Customer
> Service
> 
> .
>
>© 2014, LinkedIn Corporation. 2029 Stierlin Ct. Mountain View, CA
> 94043, USA
>
>
> ___
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel
>
>
___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

Re: [Cerowrt-devel] wifi over narrow channels

[dpreed]

Wideband is far better for scaling than narrowband, though.  This may seem 
counterintuitive, but narrowband systems are extremely inefficient.  They 
appeal to 0/1 thinking intuitively, but in actual fact the wider the bandwidth 
the more sharing and the more scaling is possible (and not be "balkanization" 
or "exclusive channel negotiation").
 
Two Internets are far worse than a single Internet that combines both.  That's 
because you have more degrees of freedom in a single network than you can in 
two distinct networks, by a combinatorial factor.
 
The analogy holds that one wide band is far better than two disjoint bands in 
terms of scaling and adaptation. The situation only gets better because of the 
physics of "multipath", which creates more problems the more narrowband the 
signal, and when the signal is a single frequency, multipath is disastrous.
 
The same is true if you try to divide space into disjoint "channels" (as 
cellular tries to).
 
So in the near term, narrowband wifi might be a short-term benefit, but 
long-term it is 180 degress away from where you want to go.
 
(the listen-before-talk protocol in WiFi is pragmatic because it is built into 
hardware today, but terrible for wideband signals, because you can't shorten 
the 4 usec. pre-transmit delay, and probably need to lengthen it, since 4 usec. 
is about 1.25 km or 0.8 miles, and  holds 40 bits at 10 Mb/s, or 4000 bits at 1 
Gb/sec).
 
Either for distance or for rate, the "Ethernet MAC+PHY" was designed for short 
"coax" or "hub" domains. Its not good for digital wireless Internet, except for 
one thing: it is based on distributed control that does not require any advance 
planning.
 
If you want to improve open wireless, you have to a) go wide, b) maintain 
distributed control, c) get rid of listen-before-talk to replace it with a 
mixture of co-channel decoding and propagation negotiation.  Then you can beat 
cellular trivially.
 
I wish I could attract investment away from the "short term" WiFi thinking, but 
in the last 15 years, I've failed.  Meanwhile WiFi also attracts those people 
who want to add bufferbloat into the routers because they don't understand 
congestion control.
 
Sad.


On Wednesday, October 8, 2014 6:14pm, "Dave Taht"  said:



> https://www.usenix.org/system/files/conference/nsdi12/nsdi12-final142.pdf
> 
> I've had 5mhz channels working in the ath9k at various points in
> cerowrt's lifetime. (using it for meshy stuff) After digesting most of
> the 802.11ac standard I do find myself wishing they'd gone towards
> narrower channels rather than wider.
> 
> The netgear x4 defaults to a 160mhz wide channel. :sigh:
> 
> The above paper has some nifty ideas in it.
> 
> --
> Dave Täht
> 
> https://www.bufferbloat.net/projects/make-wifi-fast
> ___
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel
> ___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

[Cerowrt-devel] I'd like to add you to my professional network on LinkedIn

[Kenneth Finnegan]

Hi,

I'd like to add you to my professional network on LinkedIn.

- Kenneth

Accept: 
http://www.linkedin.com/blink?simpleRedirect=0SejgUdPARcPcMd3oRd3ARczARfkh9rCZFt65QqnpKqipJdTdQimFFrzRVpkJApn9xq7cCt7dBtmtvpnhFtCVFfmJB9C5QoORBt6BSrCAJt7dBtmsJr6RBfmtKqmJzon9Q9ClQqnpKimtBkClOs3Rx9DwPbm9PrD0Ncz5FbmENtmVDtzRBfP9SbSkLrmZzbCVFp6lHrCBIbDtTtOYLeDdMt7hE&msgID=I8092929768_1&markAsRead=


You received an invitation to connect. LinkedIn will use your email address to 
make suggestions to our members in features like People You May Know. 
Unsubscribe here: 
http://www.linkedin.com/blink?simpleRedirect=u3cJoDdKs34OcmAJqz5RrCtSfmhFpipVmChEhRB8e6lJgAESi4FmgPxahCFIrCFWtQBOimVpiBkSimh1pA5qokNLlRhbp75VrQBie51me6lyhChzsD5Dhllqh5kQpQgSt6phoCBftP1Glk51gmtjcl9jjzARpSl2i551fmhFrSMCt7dBtmtvpnhFtCVFfmJB9B4ScSh5hSxmhD9JhB51fmVBqSZkp6BJ9CVRr3RQ9DwPbm9PrD0Ncz5FbmENtmVDtzRBfP9SbSkLrmZzbCVFp6lHrCBIbDtTtOYLeDdMt7hE&msgID=I8092929768_1&markAsRead=
 Learn why we included this at the following link: 
http://www.linkedin.com/blink?simpleRedirect=3wUdPgZp4BBr6dFt79x9DhPpnlDnSlQqnpKqjRHpipIsBlBoSBSsCljsClJrThPtmcZoipUcORysSVMcj8NqiRGcnlKpToZpjYOtyZBbSRLoOVKqmhBqSVFr2VTtTsLbPFMt7hE&msgID=I8092929768_1&markAsRead=
© 2014, LinkedIn Corporation. 2029 Stierlin Ct. Mountain View, CA 94043, 
USA___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel

Re: [Cerowrt-devel] Other router models for OpenWireless deployment

[Sebastian Moeller]

Hi Ragna,

On Oct 9, 2014, at 01:51 , Ranga Krishnan  wrote:

> 
> As folks on this list might know the OpenWireless project at EFF adopted
> CeroWRT as a base for the hacker release of the OpenWireless router
> in July  -- which as a result is based on the WNDR 3800.  Thanks very much 
> to the CeroWRT community for providing such a solid base on which to 
> build. We could not have got the release out without the help of this 
> community. 
> 
> For the OpenWIreless project we want to move toward wider deployment. 
> I have started to look into how the EFF could support more router models. 
> The idea is to find a few routers with OpenWRT (Barrier Breaker or later) 
> that 
> are widely deployed in the field today. By releasing images for an additional
> 3-5 widely deployed router models we hope to reach more people with a 
> software only solution for openwireless  -- i.e remove the need to buy a 
> specific router model. 

So, what about packaging the Openwireless magic into a series of 
openwrt packages so all openwrt supported routers (if hardware capabilities 
allow) are ready?

> 
> We do want one very modern high end platform  that can support 100+ Mbps
> on the WAN link. This might be the one people on this list have been 
> searching 
> for over the last few months.



> However, we also want to pick 3 or 4 previous 
> generation routers with just 802.11n (no need for 11ac) and processor / 
> memory 
> equivalent or just a bit better than the 3800.

From my totally unscientific monitoring of 
https://forum.openwrt.org/viewforum.php?id=10 the TP-link wdr4900 v1.3 seems to 
be in the right ballpark:
https://forum.openwrt.org/viewtopic.php?pid=244167#p244167, the used powerpc 
SoC seems to be quite capable. No idea how long this will still be available 
and what hardware will power the successor (it is just me or do other’s also 
dislike the router manufacturer’s inclination to keep the same name for wildly 
hardware platforms, e.g. wndr3700 v1 to v4)


> 
> With OpenWRT support, presumably it would be simple to port over the 
> Openwireless code. However whether that is true is TBD. Ease of porting 
> would be an important consideration. 
> 
>  Are there any router models folks on this list can recommend for me to look 
> at  ? Large existing installed base and good OpenWRT support are the main 
> metrics for ranking candidates. 

I would assume that the size of the installed base should 
anti-correlate with price ;)

Best Regards
Sebastian

> 
> Thanks,
> Ranga
> 
> 
> 
> 
> ___
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel

___
Cerowrt-devel mailing list
Cerowrt-devel@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cerowrt-devel