Re: [Make-wifi-fast] Updated LEDE build with airtime patches

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 604 bytes --]

Wish I wasn't out of town still...  I could have got some time with the new
one.  Had been running the 11-18 build version of yours,  on a C7 v2 in my
mixed home network enviroment.  N devices, AC devices,  a semi rogue N
range extender, all running WPA2-AES.

I did see some occasional large amounts of bloat,  most of the time good
behavior.  Occasional odd, regular thru put drops to 1/4 speed.   I
mentioned this in earlier posts,  have done some more detailed testing
since,  that's at home and unavailable  till I'm home again.  You could see
a few DSLReports logs in my posts.

Hope that helps...

[-- Attachment #2: Type: text/html, Size: 681 bytes --]

[Make-wifi-fast] Updated LEDE build with airtime patches

[Leonidas Spyropoulos]

Hello,

Sorry for not posting on the same thread, I just joined the list and
can't find a way to list-reply on [1] thread.
I tried to use the sysupgrade image for TP-Link Archer C7 v2.0 to
upgrade my OpenWrt firmware to LEDE with the pathes. I was surprised to
find out that the iamges don't include PPPoE support so I couldn't
connect.

I'm willing to build my own builds using vagrant [2] so if you can share the
.config for that device I can add the PPPoE and rebuild - right?

As alternative if you can add PPPoE support I can test the images on
that router as I'm looking for something to upgrade from the OpenWrt
release from March (more question on why I'm looking on LEDE here [3])

[1]:
https://lists.bufferbloat.net/pipermail/make-wifi-fast/2016-November/001140.html
[2]: https://github.com/inglor/lede-vagrant
[3]:
https://forum.lede-project.org/t/tp-link-archer-c7-v2-0-upgrade-from-owrt/462

Thanks in advance,

-- 
Leonidas Spyropoulos

A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 8400 bytes --]

Here's some straight eth to router test runs.

Wierdly, I'm getting worse performance on ethernet, than 15 minutes ago
over the wifi link. This computer is older and slower though, and with a HD
rather than a SSD.

UL speed is slower overall, in the first two of the three tests,  with many
more high latency events on the bloat graph occuring on both the DL and the
UL.  Last one was cleaner and faster, though with a few large excursions.

C7 gbit eth - gbit PC eth

Somewhat slower Win 10 PC than the USB adapter platform.

http://www.dslreports.com/speedtest/6881431  slow UL speeds, many high
latency events, bloat score B
http://www.dslreports.com/speedtest/6881473               "         "
http://www.dslreports.com/speedtest/6881491  better than the first two,
both in UL speed and low latency

Hope this is helpful...


On Sun, Dec 4, 2016 at 1:39 AM, Jon Pike <jonpike54@gmail.com> wrote:

> Did a new set on the same PC located on the other side of the house.  Will
> also add the requested ethernet attached PC runs later.  Just discovered
> that you can click on the bufferbloat graph items on the DSLReports
> Speedtest results, and they actually expand into a detailed picture vs time
> graph.  Never knew about that!
>
> I disabled my virus scanner (Kaspersky) for the first time.  Don't know if
> the smoother and faster results are more due to that, vs maybe it's pretty
> quiet on the home network front right now...
>
> One thing I've been wondering about, is the speed setting in the SQM,
> i.e. the 55mbit/5mbit in my case.  Is it supposed to cap the data rates
> below the setting?  If so, why am I still getting my upload speed surge of
> 10 or more mbit, even with it set to 5?
>
>
> New set of tests on the C7, Toke's 11-24 build.
> Same cake and piece of cake SQM as before, Noah's suggestion of removing
> the extra SQM settings has been done.
> SQM cap speeds: 55/5mbit
> Other new change, virus scanner disabled.
>
> Very smooth, near perfect speeds, only a very few high latency spikes. No
> neighboring PC running videos, or gaming, like there was in earlier tests.
> Maybe the upstairs roommate also wasn't as busy as well?
>
> 2.4Ghz AC adapter - Netgear A6200 AC1200
>
> http://www.dslreports.com/speedtest/6880534
> http://www.dslreports.com/speedtest/6880607
> http://www.dslreports.com/speedtest/6880642
>
> 5Ghz AC adapter - Netgear A6200 AC1200
>
> http://www.dslreports.com/speedtest/6880694
> http://www.dslreports.com/speedtest/6880747
> http://www.dslreports.com/speedtest/6880827
>
> 2.4Ghz N adapter - Netgear WNA3100 N300
>
> http://www.dslreports.com/speedtest/6880915
> http://www.dslreports.com/speedtest/6880950
> http://www.dslreports.com/speedtest/6880968
>
>>
>>
>> ---------- Forwarded message ----------
>> From: Noah Causin <n0manletter@gmail.com>
>> To: make-wifi-fast@lists.bufferbloat.net
>> Cc:
>> Date: Fri, 2 Dec 2016 13:41:11 -0500
>> Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches
>>
>> Are you using the dual-dsthost and dual-srchost options?  If so, try
>> removing those settings from SQM and uncheck all the checkboxes you
>> selected to reach them.
>>
>> Then try a test on Ethernet with SQM on.
>>
>> Noah
>>
>> On 12/2/2016 1:14 PM, Jon Pike wrote:
>>
>> Hello all...
>>
>> Well, with medical issues, and having a busy week,  I didn't get back
>> right away.  Here's the DSLReports Speedtest runs I took.
>>
>> My results here are pretty variable,  so I don't know how trustworthy
>> they are or how worth while to you guys,  vs  FLENT runs.
>> The lower block is using Toke's 11-18 C7 FW build,  the upper block is
>> with the 11-24 build.
>>
>> Upshot,  on my Archer C7,  both ran for days with no apparent issues,  on
>> my home network.  Interesting note,  I ran both the C7 v2 and the C7 v3,
>> both ran fine.  So, if you end up with a v3,  it shouldn't be an issue.
>> Further side note,  the different configuration of the v3 model,  with both
>> 2.4Ghz and 5Ghz on the external "rabbit ears" antennas, seems to have a bit
>> stronger signal on 2.4Ghz on the other end of my house,  than the v2 model
>> that has 3 separate, small 2.4Ghz antennas inside the box.  Caveat,  my
>> testing of that wasn't very thourough..
>>
>> My overall home network setup is:  Cox cable - SB6146 modem - Archer C7
>> router - various clients.   I did testing on a PC plugged directely in a
>> eth port of the C7,  to PC's using Netgear Netgear WNA3100  N300, and A6200
>> AC1200 USB adapters.
>>
>> Below are the links to DSLReports,  on the tests.  Most have cake, and
>> piece of cake, and Noah's suggested settings, with speed caps of 55 and 5
>> mbit.  My cable is supposed to be 60/6,  I see 65 often, with wild speed
>> variation on the UL, settling to 5-6mbits.
>>
>> I have the Speedtest settings set for hi res bufferbloat,  test time
>> extended to 20sec, which is about as long as you can go without pushing the
>> start of the test off the end of that graph.
>>
>> Let me know if this is in any way useful,  or what I can do to make it
>> more worth looking at.  It seems so variable and random, that I wonder
>> about the worth,  as well as you don't get a good look at what the latencey
>> is over time,  vs FLENT.
>>
>> -------------------------------------------
>> eth0 to router   -  no SQM
>>
>> http://www.dslreports.com/speedtest/6656688  Starts near 60, plateau
>> 64-65 (hits 66) then drops to 52-54 at the 10sec mark. UL hits 18, before
>> falling down to a 4.5-5.5 average  Bloat C
>>
>> http://www.dslreports.com/speedtest/6657058  very similar,  but bloat an
>> A this time
>>
>> http://www.dslreports.com/speedtest/6657916    "        Bloat a C this
>> time
>>
>>
>> now C7 V2, with Toke 11-24 FW  SQM 55/5
>>
>> http://www.dslreports.com/speedtest/6659367 51.2 1.65, bloat B
>>
>> http://www.dslreports.com/speedtest/6659493 52.4 3.01  bloat A  peak of
>> 300-400ms  on all including idle?
>>
>> --------------------------------
>> Earlier tests with Toke's 11-18 FW,  SQM 55/5 mbits
>>
>> 2.4Ghz N adapter  Netgear WNA3100  N300 adapter
>>
>> http://www.dslreports.com/speedtest/6440808
>>
>> http://www.dslreports.com/speedtest/6440846
>>
>> http://www.dslreports.com/speedtest/6440864
>>
>>
>> 2.4Ghz AC adapter  Netgear A6200 AC1200 adapter
>>
>> http://www.dslreports.com/speedtest/6440940
>>
>> http://www.dslreports.com/speedtest/6440963
>>
>> http://www.dslreports.com/speedtest/6440987
>>
>> 5Ghz AC adapter  Netgear A6200 AC1200 adapter
>>
>> http://www.dslreports.com/speedtest/6441015
>>
>> http://www.dslreports.com/speedtest/6441038
>>
>> http://www.dslreports.com/speedtest/6441056
>>
>> http://www.dslreports.com/speedtest/6441132   (first A grade of the
>> batch)
>>
>> eth1 to C7
>>
>> http://www.dslreports.com/speedtest/6441483
>>
>> http://www.dslreports.com/speedtest/6441570
>>
>> http://www.dslreports.com/speedtest/6441586
>>
>>
>> ---------- Forwarded message ----------
>> From: Jon Pike <jonpike54@gmail.com>
>> To: "Toke Høiland-Jørgensen" <toke@toke.dk>
>> Cc: make-wifi-fast@lists.bufferbloat.net
>> Date: Fri, 25 Nov 2016 21:58:46 -0800
>> Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches
>>
>> Wish I wasn't out of town still...  I could have got some time with the
>> new one.  Had been running the 11-18 build version of yours,  on a C7 v2 in
>> my mixed home network enviroment.  N devices, AC devices,  a semi rogue N
>> range extender, all running WPA2-AES.
>>
>> I did see some occasional large amounts of bloat,  most of the time good
>> behavior.  Occasional odd, regular thru put drops to 1/4 speed.   I
>> mentioned this in earlier posts,  have done some more detailed testing
>> since,  that's at home and unavailable  till I'm home again.  You could see
>> a few DSLReports logs in my posts.
>>
>> Hope that helps...
>>
>>
>>
>>
>> _______________________________________________
>> Make-wifi-fast mailing listMake-wifi-fast@lists.bufferbloat.nethttps://lists.bufferbloat.net/listinfo/make-wifi-fast
>>
>>
>>
>> _______________________________________________
>> Make-wifi-fast mailing list
>> Make-wifi-fast@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>>
>
>

[-- Attachment #2: Type: text/html, Size: 14600 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Noah Causin]

[-- Attachment #1: Type: text/plain, Size: 11584 bytes --]

Maybe your ISP gives you a burst of bandwidth and then takes it away 
shortly after.  Comcast was known for doing this with their Powerboost.

Try these settings:

Download: 42500

Upload:  4000


On 12/4/2016 5:11 AM, Jon Pike wrote:
> Here's some straight eth to router test runs.
>
> Wierdly, I'm getting worse performance on ethernet, than 15 minutes 
> ago over the wifi link. This computer is older and slower though, and 
> with a HD rather than a SSD.
>
> UL speed is slower overall, in the first two of the three tests,  with 
> many more high latency events on the bloat graph occuring on both the 
> DL and the UL.  Last one was cleaner and faster, though with a few 
> large excursions.
>
> C7 gbit eth - gbit PC eth
>
> Somewhat slower Win 10 PC than the USB adapter platform.
>
> http://www.dslreports.com/speedtest/6881431 slow UL speeds, many high 
> latency events, bloat score B
> http://www.dslreports.com/speedtest/6881473 "         "
> http://www.dslreports.com/speedtest/6881491 better than the first two, 
> both in UL speed and low latency
>
> Hope this is helpful...
>
>
> On Sun, Dec 4, 2016 at 1:39 AM, Jon Pike <jonpike54@gmail.com 
> <mailto:jonpike54@gmail.com>> wrote:
>
>     Did a new set on the same PC located on the other side of the
>     house.  Will also add the requested ethernet attached PC runs
>     later.  Just discovered that you can click on the bufferbloat
>     graph items on the DSLReports Speedtest results, and they actually
>     expand into a detailed picture vs time graph.  Never knew about that!
>
>     I disabled my virus scanner (Kaspersky) for the first time.  Don't
>     know if the smoother and faster results are more due to that, vs
>     maybe it's pretty quiet on the home network front right now...
>
>     One thing I've been wondering about, is the speed setting in the
>     SQM,  i.e. the 55mbit/5mbit in my case.  Is it supposed to cap the
>     data rates below the setting?  If so, why am I still getting my
>     upload speed surge of 10 or more mbit, even with it set to 5?
>
>
>     New set of tests on the C7, Toke's 11-24 build.
>     Same cake and piece of cake SQM as before, Noah's suggestion of
>     removing the extra SQM settings has been done.
>     SQM cap speeds: 55/5mbit
>     Other new change, virus scanner disabled.
>
>     Very smooth, near perfect speeds, only a very few high latency
>     spikes. No neighboring PC running videos, or gaming, like there
>     was in earlier tests. Maybe the upstairs roommate also wasn't as
>     busy as well?
>
>     2.4Ghz AC adapter - Netgear A6200 AC1200
>
>     http://www.dslreports.com/speedtest/6880534
>     <http://www.dslreports.com/speedtest/6880534>
>     http://www.dslreports.com/speedtest/6880607
>     <http://www.dslreports.com/speedtest/6880607>
>     http://www.dslreports.com/speedtest/6880642
>     <http://www.dslreports.com/speedtest/6880642>
>
>     5Ghz AC adapter - Netgear A6200 AC1200
>
>     http://www.dslreports.com/speedtest/6880694
>     <http://www.dslreports.com/speedtest/6880694>
>     http://www.dslreports.com/speedtest/6880747
>     <http://www.dslreports.com/speedtest/6880747>
>     http://www.dslreports.com/speedtest/6880827
>     <http://www.dslreports.com/speedtest/6880827>
>
>     2.4Ghz N adapter - Netgear WNA3100 N300
>
>     http://www.dslreports.com/speedtest/6880915
>     <http://www.dslreports.com/speedtest/6880915>
>     http://www.dslreports.com/speedtest/6880950
>     <http://www.dslreports.com/speedtest/6880950>
>     http://www.dslreports.com/speedtest/6880968
>     <http://www.dslreports.com/speedtest/6880968>
>
>
>
>         ---------- Forwarded message ----------
>         From: Noah Causin <n0manletter@gmail.com
>         <mailto:n0manletter@gmail.com>>
>         To: make-wifi-fast@lists.bufferbloat.net
>         <mailto:make-wifi-fast@lists.bufferbloat.net>
>         Cc:
>         Date: Fri, 2 Dec 2016 13:41:11 -0500
>         Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime
>         patches
>
>         Are you using the dual-dsthost and dual-srchost options?  If
>         so, try removing those settings from SQM and uncheck all the
>         checkboxes you selected to reach them.
>
>         Then try a test on Ethernet with SQM on.
>
>         Noah
>
>
>         On 12/2/2016 1:14 PM, Jon Pike wrote:
>>         Hello all...
>>
>>         Well, with medical issues, and having a busy week,  I didn't
>>         get back right away. Here's the DSLReports Speedtest runs I took.
>>
>>         My results here are pretty variable,  so I don't know how
>>         trustworthy they are or how worth while to you guys,  vs
>>         FLENT runs.
>>         The lower block is using Toke's 11-18 C7 FW build,  the upper
>>         block is with the 11-24 build.
>>
>>         Upshot,  on my Archer C7, both ran for days with no apparent
>>         issues,  on my home network.  Interesting note,  I ran both
>>         the C7 v2 and the C7 v3,  both ran fine.  So, if you end up
>>         with a v3,  it shouldn't be an issue.  Further side note, 
>>         the different configuration of the v3 model, with both 2.4Ghz
>>         and 5Ghz on the external "rabbit ears" antennas, seems to
>>         have a bit stronger signal on 2.4Ghz on the other end of my
>>         house,  than the v2 model that has 3 separate, small 2.4Ghz
>>         antennas inside the box. Caveat,  my testing of that wasn't
>>         very thourough..
>>
>>         My overall home network setup is:  Cox cable - SB6146 modem -
>>         Archer C7 router - various clients.   I did testing on a PC
>>         plugged directely in a eth port of the C7,  to PC's using
>>         Netgear Netgear WNA3100  N300, and A6200 AC1200 USB adapters.
>>
>>         Below are the links to DSLReports,  on the tests.  Most have
>>         cake, and piece of cake, and Noah's suggested settings, with
>>         speed caps of 55 and 5 mbit.  My cable is supposed to be
>>         60/6,  I see 65 often, with wild speed variation on the UL,
>>         settling to 5-6mbits.
>>
>>         I have the Speedtest settings set for hi res bufferbloat,
>>         test time extended to 20sec, which is about as long as you
>>         can go without pushing the start of the test off the end of
>>         that graph.
>>
>>         Let me know if this is in any way useful,  or what I can do
>>         to make it more worth looking at. It seems so variable and
>>         random, that I wonder about the worth, as well as you don't
>>         get a good look at what the latencey is over time,  vs FLENT.
>>
>>         -------------------------------------------
>>         eth0 to router   -  no SQM
>>
>>         http://www.dslreports.com/speedtest/6656688
>>         <http://www.dslreports.com/speedtest/6656688> Starts near 60,
>>         plateau 64-65 (hits 66) then drops to 52-54 at the 10sec
>>         mark. UL hits 18, before falling down to a 4.5-5.5 average 
>>         Bloat C
>>
>>         http://www.dslreports.com/speedtest/6657058
>>         <http://www.dslreports.com/speedtest/6657058> very similar, 
>>         but bloat an A this time
>>
>>         http://www.dslreports.com/speedtest/6657916
>>         <http://www.dslreports.com/speedtest/6657916> "        Bloat
>>         a C this time
>>
>>
>>         now C7 V2, with Toke 11-24 FW SQM 55/5
>>
>>         http://www.dslreports.com/speedtest/6659367
>>         <http://www.dslreports.com/speedtest/6659367> 51.2 1.65, bloat B
>>
>>         http://www.dslreports.com/speedtest/6659493
>>         <http://www.dslreports.com/speedtest/6659493> 52.4 3.01 
>>         bloat A  peak of 300-400ms  on all including idle?
>>
>>         --------------------------------
>>         Earlier tests with Toke's 11-18 FW,  SQM 55/5 mbits
>>
>>         2.4Ghz N adapter  Netgear WNA3100  N300 adapter
>>
>>         http://www.dslreports.com/speedtest/6440808
>>         <http://www.dslreports.com/speedtest/6440808>
>>
>>         http://www.dslreports.com/speedtest/6440846
>>         <http://www.dslreports.com/speedtest/6440846>
>>
>>         http://www.dslreports.com/speedtest/6440864
>>         <http://www.dslreports.com/speedtest/6440864>
>>
>>
>>         2.4Ghz AC adapter  Netgear A6200 AC1200 adapter
>>
>>         http://www.dslreports.com/speedtest/6440940
>>         <http://www.dslreports.com/speedtest/6440940>
>>
>>         http://www.dslreports.com/speedtest/6440963
>>         <http://www.dslreports.com/speedtest/6440963>
>>
>>         http://www.dslreports.com/speedtest/6440987
>>         <http://www.dslreports.com/speedtest/6440987>
>>
>>         5Ghz AC adapter  Netgear A6200 AC1200 adapter
>>
>>         http://www.dslreports.com/speedtest/6441015
>>         <http://www.dslreports.com/speedtest/6441015>
>>
>>         http://www.dslreports.com/speedtest/6441038
>>         <http://www.dslreports.com/speedtest/6441038>
>>
>>         http://www.dslreports.com/speedtest/6441056
>>         <http://www.dslreports.com/speedtest/6441056>
>>
>>         http://www.dslreports.com/speedtest/6441132
>>         <http://www.dslreports.com/speedtest/6441132> (first A grade
>>         of the batch)
>>
>>         eth1 to C7
>>
>>         http://www.dslreports.com/speedtest/6441483
>>         <http://www.dslreports.com/speedtest/6441483>
>>
>>         http://www.dslreports.com/speedtest/6441570
>>         <http://www.dslreports.com/speedtest/6441570>
>>
>>         http://www.dslreports.com/speedtest/6441586
>>         <http://www.dslreports.com/speedtest/6441586>
>>
>>
>>         ---------- Forwarded message ----------
>>         From: Jon Pike <jonpike54@gmail.com <mailto:jonpike54@gmail.com>>
>>         To: "Toke Høiland-Jørgensen" <toke@toke.dk <mailto:toke@toke.dk>>
>>         Cc: make-wifi-fast@lists.bufferbloat.net
>>         <mailto:make-wifi-fast@lists.bufferbloat.net>
>>         Date: Fri, 25 Nov 2016 21:58:46 -0800
>>         Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime
>>         patches
>>
>>         Wish I wasn't out of town still...  I could have got some
>>         time with the new one.  Had been running the 11-18 build
>>         version of yours, on a C7 v2 in my mixed home network
>>         enviroment.  N devices, AC devices,  a semi rogue N range
>>         extender, all running WPA2-AES.
>>
>>         I did see some occasional large amounts of bloat,  most of
>>         the time good behavior.  Occasional odd, regular thru put
>>         drops to 1/4 speed.   I mentioned this in earlier posts, 
>>         have done some more detailed testing since, that's at home
>>         and unavailable  till I'm home again.  You could see a few
>>         DSLReports logs in my posts.
>>
>>         Hope that helps...
>>
>>
>>
>>
>>
>>         _______________________________________________
>>         Make-wifi-fast mailing list
>>         Make-wifi-fast@lists.bufferbloat.net
>>         <mailto:Make-wifi-fast@lists.bufferbloat.net>
>>         https://lists.bufferbloat.net/listinfo/make-wifi-fast
>>         <https://lists.bufferbloat.net/listinfo/make-wifi-fast>
>         _______________________________________________ Make-wifi-fast
>         mailing list Make-wifi-fast@lists.bufferbloat.net
>         <mailto:Make-wifi-fast@lists.bufferbloat.net>
>         https://lists.bufferbloat.net/listinfo/make-wifi-fast
>         <https://lists.bufferbloat.net/listinfo/make-wifi-fast> 
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast

[-- Attachment #2: Type: text/html, Size: 28422 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

[-- Attachment #1: Type: text/plain, Size: 540 bytes --]

My new test setup is
rpi2--archer c7----(wifi)-fedora laptop (atheros wifi and fq_codel by default)

The results are somewhat not what I expected. I am seeing relatively
large latencies at close range for the rpi2 server which is plugged
into the archer v2 in-built switch.

Any pointer as to what I need to tweak to reach 40ms latency under load ?

root@lede:/sys# cat kernel/debug/ieee80211/phy1/netdev\:wlan1/aqm
ac backlog-bytes backlog-packets new-flows drops marks overlimit
collisions tx-bytes tx-packets
2 0 0 187 0 0 0 0 39948 187

[-- Attachment #2: rpi2_4.png --]
[-- Type: image/png, Size: 61029 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

the ath10k remains unfixed. the 2.4ghz ath9k has all the needed patches

On Tue, Dec 6, 2016 at 12:54 PM, Loganaden Velvindron
<loganaden@gmail.com> wrote:
> My new test setup is
> rpi2--archer c7----(wifi)-fedora laptop (atheros wifi and fq_codel by default)
>
> The results are somewhat not what I expected. I am seeing relatively
> large latencies at close range for the rpi2 server which is plugged
> into the archer v2 in-built switch.
>
> Any pointer as to what I need to tweak to reach 40ms latency under load ?
>
> root@lede:/sys# cat kernel/debug/ieee80211/phy1/netdev\:wlan1/aqm
> ac backlog-bytes backlog-packets new-flows drops marks overlimit
> collisions tx-bytes tx-packets
> 2 0 0 187 0 0 0 0 39948 187
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

secondly you are measuring the rrul test. we have not fixed wifi qos.
rrul_be will give you vastly better results on both chipsets.

On Tue, Dec 6, 2016 at 12:58 PM, Dave Taht <dave.taht@gmail.com> wrote:
> the ath10k remains unfixed. the 2.4ghz ath9k has all the needed patches
>
> On Tue, Dec 6, 2016 at 12:54 PM, Loganaden Velvindron
> <loganaden@gmail.com> wrote:
>> My new test setup is
>> rpi2--archer c7----(wifi)-fedora laptop (atheros wifi and fq_codel by default)
>>
>> The results are somewhat not what I expected. I am seeing relatively
>> large latencies at close range for the rpi2 server which is plugged
>> into the archer v2 in-built switch.
>>
>> Any pointer as to what I need to tweak to reach 40ms latency under load ?
>>
>> root@lede:/sys# cat kernel/debug/ieee80211/phy1/netdev\:wlan1/aqm
>> ac backlog-bytes backlog-packets new-flows drops marks overlimit
>> collisions tx-bytes tx-packets
>> 2 0 0 187 0 0 0 0 39948 187
>>
>> _______________________________________________
>> Make-wifi-fast mailing list
>> Make-wifi-fast@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>
>
>
> --
> Dave Täht
> Let's go make home routers and wifi faster! With better software!
> http://blog.cerowrt.org



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

On Wed, Dec 7, 2016 at 12:58 AM, Dave Taht <dave.taht@gmail.com> wrote:
> the ath10k remains unfixed. the 2.4ghz ath9k has all the needed patches
>

That's the ath9k with all of the patches applied.

I forgot to mention that in my diagram.

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

[-- Attachment #1: Type: text/plain, Size: 112 bytes --]

Using rrul_be gives me better latency.

Please see attached screenshot. This is closer to what Dave is getting.

[-- Attachment #2: rpi2_13.png --]
[-- Type: image/png, Size: 57310 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

that is a correct looking result. We can still do better.

The variance on the icmp side is due to the kernel on one side
responding to icmp packets faster (on the bottom half of the
distribution), and the cost of the context switch (and some in-kernel
deprioritization) on the top half.

On Tue, Dec 6, 2016 at 1:08 PM, Loganaden Velvindron
<loganaden@gmail.com> wrote:
> Using rrul_be gives me better latency.
>
> Please see attached screenshot. This is closer to what Dave is getting.



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 7239 bytes --]

Did a new set on the same PC located on the other side of the house.  Will
also add the requested ethernet attached PC runs later.  Just discovered
that you can click on the bufferbloat graph items on the DSLReports
Speedtest results, and they actually expand into a detailed picture vs time
graph.  Never knew about that!

I disabled my virus scanner (Kaspersky) for the first time.  Don't know if
the smoother and faster results are more due to that, vs maybe it's pretty
quiet on the home network front right now...

One thing I've been wondering about, is the speed setting in the SQM,  i.e.
the 55mbit/5mbit in my case.  Is it supposed to cap the data rates below
the setting?  If so, why am I still getting my upload speed surge of 10 or
more mbit, even with it set to 5?


New set of tests on the C7, Toke's 11-24 build.
Same cake and piece of cake SQM as before, Noah's suggestion of removing
the extra SQM settings has been done.
SQM cap speeds: 55/5mbit
Other new change, virus scanner disabled.

Very smooth, near perfect speeds, only a very few high latency spikes. No
neighboring PC running videos, or gaming, like there was in earlier tests.
Maybe the upstairs roommate also wasn't as busy as well?

2.4Ghz AC adapter - Netgear A6200 AC1200

http://www.dslreports.com/speedtest/6880534
http://www.dslreports.com/speedtest/6880607
http://www.dslreports.com/speedtest/6880642

5Ghz AC adapter - Netgear A6200 AC1200

http://www.dslreports.com/speedtest/6880694
http://www.dslreports.com/speedtest/6880747
http://www.dslreports.com/speedtest/6880827

2.4Ghz N adapter - Netgear WNA3100 N300

http://www.dslreports.com/speedtest/6880915
http://www.dslreports.com/speedtest/6880950
http://www.dslreports.com/speedtest/6880968

>
>
> ---------- Forwarded message ----------
> From: Noah Causin <n0manletter@gmail.com>
> To: make-wifi-fast@lists.bufferbloat.net
> Cc:
> Date: Fri, 2 Dec 2016 13:41:11 -0500
> Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches
>
> Are you using the dual-dsthost and dual-srchost options?  If so, try
> removing those settings from SQM and uncheck all the checkboxes you
> selected to reach them.
>
> Then try a test on Ethernet with SQM on.
>
> Noah
>
> On 12/2/2016 1:14 PM, Jon Pike wrote:
>
> Hello all...
>
> Well, with medical issues, and having a busy week,  I didn't get back
> right away.  Here's the DSLReports Speedtest runs I took.
>
> My results here are pretty variable,  so I don't know how trustworthy they
> are or how worth while to you guys,  vs  FLENT runs.
> The lower block is using Toke's 11-18 C7 FW build,  the upper block is
> with the 11-24 build.
>
> Upshot,  on my Archer C7,  both ran for days with no apparent issues,  on
> my home network.  Interesting note,  I ran both the C7 v2 and the C7 v3,
> both ran fine.  So, if you end up with a v3,  it shouldn't be an issue.
> Further side note,  the different configuration of the v3 model,  with both
> 2.4Ghz and 5Ghz on the external "rabbit ears" antennas, seems to have a bit
> stronger signal on 2.4Ghz on the other end of my house,  than the v2 model
> that has 3 separate, small 2.4Ghz antennas inside the box.  Caveat,  my
> testing of that wasn't very thourough..
>
> My overall home network setup is:  Cox cable - SB6146 modem - Archer C7
> router - various clients.   I did testing on a PC plugged directely in a
> eth port of the C7,  to PC's using Netgear Netgear WNA3100  N300, and A6200
> AC1200 USB adapters.
>
> Below are the links to DSLReports,  on the tests.  Most have cake, and
> piece of cake, and Noah's suggested settings, with speed caps of 55 and 5
> mbit.  My cable is supposed to be 60/6,  I see 65 often, with wild speed
> variation on the UL, settling to 5-6mbits.
>
> I have the Speedtest settings set for hi res bufferbloat,  test time
> extended to 20sec, which is about as long as you can go without pushing the
> start of the test off the end of that graph.
>
> Let me know if this is in any way useful,  or what I can do to make it
> more worth looking at.  It seems so variable and random, that I wonder
> about the worth,  as well as you don't get a good look at what the latencey
> is over time,  vs FLENT.
>
> -------------------------------------------
> eth0 to router   -  no SQM
>
> http://www.dslreports.com/speedtest/6656688  Starts near 60, plateau
> 64-65 (hits 66) then drops to 52-54 at the 10sec mark. UL hits 18, before
> falling down to a 4.5-5.5 average  Bloat C
>
> http://www.dslreports.com/speedtest/6657058  very similar,  but bloat an
> A this time
>
> http://www.dslreports.com/speedtest/6657916    "        Bloat a C this
> time
>
>
> now C7 V2, with Toke 11-24 FW  SQM 55/5
>
> http://www.dslreports.com/speedtest/6659367 51.2 1.65, bloat B
>
> http://www.dslreports.com/speedtest/6659493 52.4 3.01  bloat A  peak of
> 300-400ms  on all including idle?
>
> --------------------------------
> Earlier tests with Toke's 11-18 FW,  SQM 55/5 mbits
>
> 2.4Ghz N adapter  Netgear WNA3100  N300 adapter
>
> http://www.dslreports.com/speedtest/6440808
>
> http://www.dslreports.com/speedtest/6440846
>
> http://www.dslreports.com/speedtest/6440864
>
>
> 2.4Ghz AC adapter  Netgear A6200 AC1200 adapter
>
> http://www.dslreports.com/speedtest/6440940
>
> http://www.dslreports.com/speedtest/6440963
>
> http://www.dslreports.com/speedtest/6440987
>
> 5Ghz AC adapter  Netgear A6200 AC1200 adapter
>
> http://www.dslreports.com/speedtest/6441015
>
> http://www.dslreports.com/speedtest/6441038
>
> http://www.dslreports.com/speedtest/6441056
>
> http://www.dslreports.com/speedtest/6441132   (first A grade of the batch)
>
> eth1 to C7
>
> http://www.dslreports.com/speedtest/6441483
>
> http://www.dslreports.com/speedtest/6441570
>
> http://www.dslreports.com/speedtest/6441586
>
>
> ---------- Forwarded message ----------
> From: Jon Pike <jonpike54@gmail.com>
> To: "Toke Høiland-Jørgensen" <toke@toke.dk>
> Cc: make-wifi-fast@lists.bufferbloat.net
> Date: Fri, 25 Nov 2016 21:58:46 -0800
> Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches
>
> Wish I wasn't out of town still...  I could have got some time with the
> new one.  Had been running the 11-18 build version of yours,  on a C7 v2 in
> my mixed home network enviroment.  N devices, AC devices,  a semi rogue N
> range extender, all running WPA2-AES.
>
> I did see some occasional large amounts of bloat,  most of the time good
> behavior.  Occasional odd, regular thru put drops to 1/4 speed.   I
> mentioned this in earlier posts,  have done some more detailed testing
> since,  that's at home and unavailable  till I'm home again.  You could see
> a few DSLReports logs in my posts.
>
> Hope that helps...
>
>
>
>
> _______________________________________________
> Make-wifi-fast mailing listMake-wifi-fast@lists.bufferbloat.nethttps://lists.bufferbloat.net/listinfo/make-wifi-fast
>
>
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>

[-- Attachment #2: Type: text/html, Size: 12604 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 4353 bytes --]

Hello all...

Well, with medical issues, and having a busy week,  I didn't get back right
away.  Here's the DSLReports Speedtest runs I took.

My results here are pretty variable,  so I don't know how trustworthy they
are or how worth while to you guys,  vs  FLENT runs.
The lower block is using Toke's 11-18 C7 FW build,  the upper block is with
the 11-24 build.

Upshot,  on my Archer C7,  both ran for days with no apparent issues,  on
my home network.  Interesting note,  I ran both the C7 v2 and the C7 v3,
both ran fine.  So, if you end up with a v3,  it shouldn't be an issue.
Further side note,  the different configuration of the v3 model,  with both
2.4Ghz and 5Ghz on the external "rabbit ears" antennas, seems to have a bit
stronger signal on 2.4Ghz on the other end of my house,  than the v2 model
that has 3 separate, small 2.4Ghz antennas inside the box.  Caveat,  my
testing of that wasn't very thourough..

My overall home network setup is:  Cox cable - SB6146 modem - Archer C7
router - various clients.   I did testing on a PC plugged directely in a
eth port of the C7,  to PC's using Netgear Netgear WNA3100  N300, and A6200
AC1200 USB adapters.

Below are the links to DSLReports,  on the tests.  Most have cake, and
piece of cake, and Noah's suggested settings, with speed caps of 55 and 5
mbit.  My cable is supposed to be 60/6,  I see 65 often, with wild speed
variation on the UL, settling to 5-6mbits.

I have the Speedtest settings set for hi res bufferbloat,  test time
extended to 20sec, which is about as long as you can go without pushing the
start of the test off the end of that graph.

Let me know if this is in any way useful,  or what I can do to make it more
worth looking at.  It seems so variable and random, that I wonder about the
worth,  as well as you don't get a good look at what the latencey is over
time,  vs FLENT.

-------------------------------------------
eth0 to router   -  no SQM

http://www.dslreports.com/speedtest/6656688  Starts near 60, plateau 64-65
(hits 66) then drops to 52-54 at the 10sec mark. UL hits 18, before falling
down to a 4.5-5.5 average  Bloat C

http://www.dslreports.com/speedtest/6657058  very similar,  but bloat an A
this time

http://www.dslreports.com/speedtest/6657916    "        Bloat a C this time


now C7 V2, with Toke 11-24 FW  SQM 55/5

http://www.dslreports.com/speedtest/6659367 51.2 1.65, bloat B

http://www.dslreports.com/speedtest/6659493 52.4 3.01  bloat A  peak of
300-400ms  on all including idle?

--------------------------------
Earlier tests with Toke's 11-18 FW,  SQM 55/5 mbits

2.4Ghz N adapter  Netgear WNA3100  N300 adapter

http://www.dslreports.com/speedtest/6440808

http://www.dslreports.com/speedtest/6440846

http://www.dslreports.com/speedtest/6440864


2.4Ghz AC adapter  Netgear A6200 AC1200 adapter

http://www.dslreports.com/speedtest/6440940

http://www.dslreports.com/speedtest/6440963

http://www.dslreports.com/speedtest/6440987

5Ghz AC adapter  Netgear A6200 AC1200 adapter

http://www.dslreports.com/speedtest/6441015

http://www.dslreports.com/speedtest/6441038

http://www.dslreports.com/speedtest/6441056

http://www.dslreports.com/speedtest/6441132   (first A grade of the batch)

eth1 to C7

http://www.dslreports.com/speedtest/6441483

http://www.dslreports.com/speedtest/6441570

http://www.dslreports.com/speedtest/6441586


---------- Forwarded message ----------
From: Jon Pike <jonpike54@gmail.com>
To: "Toke Høiland-Jørgensen" <toke@toke.dk>
Cc: make-wifi-fast@lists.bufferbloat.net
Date: Fri, 25 Nov 2016 21:58:46 -0800
Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches

Wish I wasn't out of town still...  I could have got some time with the new
one.  Had been running the 11-18 build version of yours,  on a C7 v2 in my
mixed home network enviroment.  N devices, AC devices,  a semi rogue N
range extender, all running WPA2-AES.

I did see some occasional large amounts of bloat,  most of the time good
behavior.  Occasional odd, regular thru put drops to 1/4 speed.   I
mentioned this in earlier posts,  have done some more detailed testing
since,  that's at home and unavailable  till I'm home again.  You could see
a few DSLReports logs in my posts.

Hope that helps...

[-- Attachment #2: Type: text/html, Size: 6061 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Noah Causin]

[-- Attachment #1: Type: text/plain, Size: 4994 bytes --]

Are you using the dual-dsthost and dual-srchost options?  If so, try 
removing those settings from SQM and uncheck all the checkboxes you 
selected to reach them.

Then try a test on Ethernet with SQM on.

Noah


On 12/2/2016 1:14 PM, Jon Pike wrote:
> Hello all...
>
> Well, with medical issues, and having a busy week,  I didn't get back 
> right away.  Here's the DSLReports Speedtest runs I took.
>
> My results here are pretty variable,  so I don't know how trustworthy 
> they are or how worth while to you guys,  vs FLENT runs.
> The lower block is using Toke's 11-18 C7 FW build,  the upper block is 
> with the 11-24 build.
>
> Upshot,  on my Archer C7,  both ran for days with no apparent issues,  
> on my home network.  Interesting note,  I ran both the C7 v2 and the 
> C7 v3,  both ran fine.  So, if you end up with a v3,  it shouldn't be 
> an issue.  Further side note,  the different configuration of the v3 
> model,  with both 2.4Ghz and 5Ghz on the external "rabbit ears" 
> antennas, seems to have a bit stronger signal on 2.4Ghz on the other 
> end of my house,  than the v2 model that has 3 separate, small 2.4Ghz 
> antennas inside the box.  Caveat,  my testing of that wasn't very 
> thourough..
>
> My overall home network setup is:  Cox cable - SB6146 modem - Archer 
> C7 router - various clients.   I did testing on a PC plugged directely 
> in a eth port of the C7,  to PC's using Netgear Netgear WNA3100  N300, 
> and A6200 AC1200 USB adapters.
>
> Below are the links to DSLReports,  on the tests.  Most have cake, and 
> piece of cake, and Noah's suggested settings, with speed caps of 55 
> and 5 mbit.  My cable is supposed to be 60/6,  I see 65 often, with 
> wild speed variation on the UL, settling to 5-6mbits.
>
> I have the Speedtest settings set for hi res bufferbloat, test time 
> extended to 20sec, which is about as long as you can go without 
> pushing the start of the test off the end of that graph.
>
> Let me know if this is in any way useful,  or what I can do to make it 
> more worth looking at.  It seems so variable and random, that I wonder 
> about the worth,  as well as you don't get a good look at what the 
> latencey is over time,  vs FLENT.
>
> -------------------------------------------
> eth0 to router   -  no SQM
>
> http://www.dslreports.com/speedtest/6656688 Starts near 60, plateau 
> 64-65 (hits 66) then drops to 52-54 at the 10sec mark. UL hits 18, 
> before falling down to a 4.5-5.5 average  Bloat C
>
> http://www.dslreports.com/speedtest/6657058 very similar,  but bloat 
> an A this time
>
> http://www.dslreports.com/speedtest/6657916 "        Bloat a C this time
>
>
> now C7 V2, with Toke 11-24 FW  SQM 55/5
>
> http://www.dslreports.com/speedtest/6659367 51.2 1.65, bloat B
>
> http://www.dslreports.com/speedtest/6659493 52.4 3.01  bloat A  peak 
> of 300-400ms  on all including idle?
>
> --------------------------------
> Earlier tests with Toke's 11-18 FW,  SQM 55/5 mbits
>
> 2.4Ghz N adapter  Netgear WNA3100  N300 adapter
>
> http://www.dslreports.com/speedtest/6440808
>
> http://www.dslreports.com/speedtest/6440846
>
> http://www.dslreports.com/speedtest/6440864
>
>
> 2.4Ghz AC adapter  Netgear A6200 AC1200 adapter
>
> http://www.dslreports.com/speedtest/6440940
>
> http://www.dslreports.com/speedtest/6440963
>
> http://www.dslreports.com/speedtest/6440987
>
> 5Ghz AC adapter  Netgear A6200 AC1200 adapter
>
> http://www.dslreports.com/speedtest/6441015
>
> http://www.dslreports.com/speedtest/6441038
>
> http://www.dslreports.com/speedtest/6441056
>
> http://www.dslreports.com/speedtest/6441132 (first A grade of the batch)
>
> eth1 to C7
>
> http://www.dslreports.com/speedtest/6441483
>
> http://www.dslreports.com/speedtest/6441570
>
> http://www.dslreports.com/speedtest/6441586
>
>
> ---------- Forwarded message ----------
> From: Jon Pike <jonpike54@gmail.com <mailto:jonpike54@gmail.com>>
> To: "Toke Høiland-Jørgensen" <toke@toke.dk <mailto:toke@toke.dk>>
> Cc: make-wifi-fast@lists.bufferbloat.net 
> <mailto:make-wifi-fast@lists.bufferbloat.net>
> Date: Fri, 25 Nov 2016 21:58:46 -0800
> Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches
>
> Wish I wasn't out of town still...  I could have got some time with 
> the new one.  Had been running the 11-18 build version of yours,  on a 
> C7 v2 in my mixed home network enviroment.  N devices, AC devices,  a 
> semi rogue N range extender, all running WPA2-AES.
>
> I did see some occasional large amounts of bloat,  most of the time 
> good behavior.  Occasional odd, regular thru put drops to 1/4 speed.   
> I mentioned this in earlier posts,  have done some more detailed 
> testing since, that's at home and unavailable  till I'm home again.  
> You could see a few DSLReports logs in my posts.
>
> Hope that helps...
>
>
>
>
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast


[-- Attachment #2: Type: text/html, Size: 9303 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 3108 bytes --]

And let me know what I might do to take more useful data with DSLReports...

On Nov 19, 2016 7:20 PM, "Jon Pike" <jonpike54@gmail.com> wrote:

> Ok, hosed that first one up pretty good.   Let's see how badly I can do
> this time...
>
>
> Taking it around the block... so far so good on my Archer C7...  ;-)
>
> Chickened out a bit,  didn't try a sysupgrade,  I had been running a
> custom build from Noah and didn't know if that was going to be an issue.
> Also didn't try to save settings, let it clean everything out.
> Installed my most recent settings,  that seemed to work. Someone let me
> know if there could be version issues there I might not have noticed yet.
>
> I'm running on the settings Noah mentioned a while back, cake and piece
> of cake and the below:
> > Simply select cake as the qdisc and a sqm script that uses cake. Under
> > the "Queue Discipline" tab, enable a set of two checkboxes.
> >
> > Under "Advanced option string to pass to the ingress queueing
> > disciplines," put in "dual-dsthost nat"
> >
> > Under "Advanced option string to pass to the egress queueing
> > disciplines," put in "dual-srchost nat"
> >
> > Because of the issues with WMM in the WiFi stack, I recommend using the
> > following custom iptables rules:
> >
> > iptables -t mangle -A FORWARD -o br-lan -j DSCP --set-dscp-class be
> > iptables -t mangle -A OUTPUT -o br-lan -j DSCP --set-dscp-class be
> > ip6tables -t mangle -A FORWARD -o br-lan -j DSCP --set-dscp-class be
> > ip6tables -t mangle -A OUTPUT -o br-lan -j DSCP --set-dscp-class be
> >
> > You still need WMM to get the full speeds on the newer WiFi modes.
> >
> > Noah
>
> I'm a little uncertain of what parts of what are in what state of
> working.  Current setup is 65/6Mbit Cable Modem - C7 - variety of Win
> PC's and other things.  I have a couple of AC1200 usb adapters, a few
> phones are AC as well,  as well as a lot of N devices.   Been trying to
> keep up with what's working with which interface.
>
> I'm getting mostly decent numbers for bufferbloat on DSLreports. Only
> have DSLreports as a test method now, so I'm not much good for decent data.
> My PC to eth connection has next to nothing,  and most of the time the
> wifi looks very good with bloat below 30ms.
>
> I have been seeing an anomaly, though, in that I get a fast startup,
> with very low latency,  download speed in the 50's,  then after 5-10 sec
> the speed drops drastically to the 10's for a few seconds,  then rises
> back up to the 50's again.  I was getting ready to write this up the
> other night,  then trying it again it was gone.  Full, flat speed thru
> the whole test,  over many tries. Tonight,  it's back again.  This was
> on Noah's build, as well as tonight's change to Toke's 11-18 C7 build.
> Not sure what to make of it.  Here's an example URL to DSLreports,  if
> that helps: http://www.dslreports.com/speedtest/6276874
>
> It can do this on 2.4Ghz and on 5Ghz with the AC1200 dongle, believe it
> did it on the N dongle,  will try that shortly.  Also, the game player
> is home from work now... we'll see what his comments are...
>
> Jon
>

[-- Attachment #2: Type: text/html, Size: 3963 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Noah Causin]

[-- Attachment #1: Type: text/plain, Size: 4109 bytes --]

What I think you should do is completely reset the router's settings in 
the update firmware tab on that new build, and just setup the WiFi names 
and passwords.

Then go to luci-app-sqm, set your speeds, select cake, and select 
piece_of_cake.qos.

The settings file I set you was for a very thorough setup.  I think you 
should just go with basic WiFi and Cake to avoid conflicts.


On 11/19/2016 10:24 PM, Jon Pike wrote:
>
> And let me know what I might do to take more useful data with 
> DSLReports...
>
>
> On Nov 19, 2016 7:20 PM, "Jon Pike" <jonpike54@gmail.com 
> <mailto:jonpike54@gmail.com>> wrote:
>
>     Ok, hosed that first one up pretty good. Let's see how badly I can
>     do this time...
>
>
>     Taking it around the block... so far so good on my Archer C7...  ;-)
>
>     Chickened out a bit,  didn't try a sysupgrade, I had been running a
>     custom build from Noah and didn't know if that was going to be an
>     issue.
>     Also didn't try to save settings, let it clean everything out.
>     Installed my most recent settings,  that seemed to work. Someone
>     let me
>     know if there could be version issues there I might not have
>     noticed yet.
>
>     I'm running on the settings Noah mentioned a while back, cake and
>     piece
>     of cake and the below:
>     > Simply select cake as the qdisc and a sqm script that uses cake.
>     Under
>     > the "Queue Discipline" tab, enable a set of two checkboxes.
>     >
>     > Under "Advanced option string to pass to the ingress queueing
>     > disciplines," put in "dual-dsthost nat"
>     >
>     > Under "Advanced option string to pass to the egress queueing
>     > disciplines," put in "dual-srchost nat"
>     >
>     > Because of the issues with WMM in the WiFi stack, I recommend
>     using the
>     > following custom iptables rules:
>     >
>     > iptables -t mangle -A FORWARD -o br-lan -j DSCP --set-dscp-class be
>     > iptables -t mangle -A OUTPUT -o br-lan -j DSCP --set-dscp-class be
>     > ip6tables -t mangle -A FORWARD -o br-lan -j DSCP --set-dscp-class be
>     > ip6tables -t mangle -A OUTPUT -o br-lan -j DSCP --set-dscp-class be
>     >
>     > You still need WMM to get the full speeds on the newer WiFi modes.
>     >
>     > Noah
>
>     I'm a little uncertain of what parts of what are in what state of
>     working.  Current setup is 65/6Mbit Cable Modem - C7 - variety of Win
>     PC's and other things.  I have a couple of AC1200 usb adapters, a few
>     phones are AC as well,  as well as a lot of N devices. Been trying to
>     keep up with what's working with which interface.
>
>     I'm getting mostly decent numbers for bufferbloat on DSLreports. Only
>     have DSLreports as a test method now, so I'm not much good for
>     decent data.
>     My PC to eth connection has next to nothing,  and most of the time
>     the
>     wifi looks very good with bloat below 30ms.
>
>     I have been seeing an anomaly, though, in that I get a fast startup,
>     with very low latency,  download speed in the 50's,  then after
>     5-10 sec
>     the speed drops drastically to the 10's for a few seconds,  then
>     rises
>     back up to the 50's again.  I was getting ready to write this up the
>     other night,  then trying it again it was gone.  Full, flat speed
>     thru
>     the whole test,  over many tries. Tonight,  it's back again.  This
>     was
>     on Noah's build, as well as tonight's change to Toke's 11-18 C7
>     build.
>     Not sure what to make of it.  Here's an example URL to
>     DSLreports,  if
>     that helps: http://www.dslreports.com/speedtest/6276874
>     <http://www.dslreports.com/speedtest/6276874>
>
>     It can do this on 2.4Ghz and on 5Ghz with the AC1200 dongle,
>     believe it
>     did it on the N dongle,  will try that shortly.  Also, the game
>     player
>     is home from work now... we'll see what his comments are...
>
>     Jon
>
>
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast


[-- Attachment #2: Type: text/html, Size: 6512 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 4288 bytes --]

I should have said that I stopped using your config,  went back to
default.  Was having issues, as I was learning LEDE and luci, (probably
more me than it) so config'ed from scratch,  and THAT is what I used when
going from your build to Toke's.

I could imagine eventual problems as what's in a LEDE version changes too
much,  question is what is too much and how do you know?

Has what I done pretty much accomplished the same thing?

---------------
From: Noah Causin <n0manletter@gmail.com>
To: make-wifi-fast@lists.bufferbloat.net
Cc:
Date: Sat, 19 Nov 2016 22:32:14 -0500
Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches
What I think you should do is completely reset the router's settings in the
update firmware tab on that new build, and just setup the WiFi names and
passwords.

Then go to luci-app-sqm, set your speeds, select cake, and select
piece_of_cake.qos.

The settings file I set you was for a very thorough setup.  I think you
should just go with basic WiFi and Cake to avoid conflicts

On Nov 19, 2016 7:32 PM, <make-wifi-fast-request@lists.bufferbloat.net>
wrote:

> And let me know what I might do to take more useful data with DSLReports...
>
> On Nov 19, 2016 7:20 PM, "Jon Pike" <jonpike54@gmail.com> wrote:
>
>> Ok, hosed that first one up pretty good.   Let's see how badly I can do
>> this time...
>>
>>
>> Taking it around the block... so far so good on my Archer C7...  ;-)
>>
>> Chickened out a bit,  didn't try a sysupgrade,  I had been running a
>> custom build from Noah and didn't know if that was going to be an issue.
>> Also didn't try to save settings, let it clean everything out.
>> Installed my most recent settings,  that seemed to work. Someone let me
>> know if there could be version issues there I might not have noticed yet.
>>
>> I'm running on the settings Noah mentioned a while back, cake and piece
>> of cake and the below:
>> > Simply select cake as the qdisc and a sqm script that uses cake. Under
>> > the "Queue Discipline" tab, enable a set of two checkboxes.
>> >
>> > Under "Advanced option string to pass to the ingress queueing
>> > disciplines," put in "dual-dsthost nat"
>> >
>> > Under "Advanced option string to pass to the egress queueing
>> > disciplines," put in "dual-srchost nat"
>> >
>> > Because of the issues with WMM in the WiFi stack, I recommend using the
>> > following custom iptables rules:
>> >
>> > iptables -t mangle -A FORWARD -o br-lan -j DSCP --set-dscp-class be
>> > iptables -t mangle -A OUTPUT -o br-lan -j DSCP --set-dscp-class be
>> > ip6tables -t mangle -A FORWARD -o br-lan -j DSCP --set-dscp-class be
>> > ip6tables -t mangle -A OUTPUT -o br-lan -j DSCP --set-dscp-class be
>> >
>> > You still need WMM to get the full speeds on the newer WiFi modes.
>> >
>> > Noah
>>
>> I'm a little uncertain of what parts of what are in what state of
>> working.  Current setup is 65/6Mbit Cable Modem - C7 - variety of Win
>> PC's and other things.  I have a couple of AC1200 usb adapters, a few
>> phones are AC as well,  as well as a lot of N devices.   Been trying to
>> keep up with what's working with which interface.
>>
>> I'm getting mostly decent numbers for bufferbloat on DSLreports. Only
>> have DSLreports as a test method now, so I'm not much good for decent
>> data.
>> My PC to eth connection has next to nothing,  and most of the time the
>> wifi looks very good with bloat below 30ms.
>>
>> I have been seeing an anomaly, though, in that I get a fast startup,
>> with very low latency,  download speed in the 50's,  then after 5-10 sec
>> the speed drops drastically to the 10's for a few seconds,  then rises
>> back up to the 50's again.  I was getting ready to write this up the
>> other night,  then trying it again it was gone.  Full, flat speed thru
>> the whole test,  over many tries. Tonight,  it's back again.  This was
>> on Noah's build, as well as tonight's change to Toke's 11-18 C7 build.
>> Not sure what to make of it.  Here's an example URL to DSLreports,  if
>> that helps: http://www.dslreports.com/speedtest/6276874
>>
>> It can do this on 2.4Ghz and on 5Ghz with the AC1200 dongle, believe it
>> did it on the N dongle,  will try that shortly.  Also, the game player
>> is home from work now... we'll see what his comments are...
>>
>> Jon
>>
>

[-- Attachment #2: Type: text/html, Size: 5638 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Noah Causin]

[-- Attachment #1: Type: text/plain, Size: 5981 bytes --]

The tweaks I usually do are for fine-tuning and can be too difficult for 
the average user.  I made you a custom build and a config file, so you 
could easily have a finely-tuned build.

Toke's build should work well for you.  You have improvements to your 
WiFi stack in your router.  The custom firewall rules I recommended 
before only work if you have a specific set of packages installed, so I 
recommend just removing them.  If you have the dual-src and dual-dst 
settings in your SQM settings, keeping that will benefit you.


On 11/19/2016 10:52 PM, Jon Pike wrote:
>
> I should have said that I stopped using your config, went back to 
> default.  Was having issues, as I was learning LEDE and luci, 
> (probably more me than it) so config'ed from scratch, and THAT is what 
> I used when going from your build to Toke's.
>
> I could imagine eventual problems as what's in a LEDE version changes 
> too much,  question is what is too much and how do you know?
>
> Has what I done pretty much accomplished the same thing?
>
> ---------------
> From: Noah Causin <n0manletter@gmail.com <mailto:n0manletter@gmail.com>>
> To: make-wifi-fast@lists.bufferbloat.net 
> <mailto:make-wifi-fast@lists.bufferbloat.net>
> Cc:
> Date: Sat, 19 Nov 2016 22:32:14 -0500
> Subject: Re: [Make-wifi-fast] Updated LEDE build with airtime patches
> What I think you should do is completely reset the router's settings 
> in the update firmware tab on that new build, and just setup the WiFi 
> names and passwords.
>
> Then go to luci-app-sqm, set your speeds, select cake, and select 
> piece_of_cake.qos.
>
> The settings file I set you was for a very thorough setup.  I think 
> you should just go with basic WiFi and Cake to avoid conflicts
>
>
> On Nov 19, 2016 7:32 PM, <make-wifi-fast-request@lists.bufferbloat.net 
> <mailto:make-wifi-fast-request@lists.bufferbloat.net>> wrote:
>
>     And let me know what I might do to take more useful data with
>     DSLReports...
>
>
>     On Nov 19, 2016 7:20 PM, "Jon Pike" <jonpike54@gmail.com
>     <mailto:jonpike54@gmail.com>> wrote:
>
>         Ok, hosed that first one up pretty good.   Let's see how badly
>         I can do this time...
>
>
>         Taking it around the block... so far so good on my Archer
>         C7...  ;-)
>
>         Chickened out a bit,  didn't try a sysupgrade,  I had been
>         running a
>         custom build from Noah and didn't know if that was going to be
>         an issue.
>         Also didn't try to save settings, let it clean everything out.
>         Installed my most recent settings,  that seemed to work.
>         Someone let me
>         know if there could be version issues there I might not have
>         noticed yet.
>
>         I'm running on the settings Noah mentioned a while back, cake
>         and piece
>         of cake and the below:
>         > Simply select cake as the qdisc and a sqm script that uses
>         cake. Under
>         > the "Queue Discipline" tab, enable a set of two checkboxes.
>         >
>         > Under "Advanced option string to pass to the ingress queueing
>         > disciplines," put in "dual-dsthost nat"
>         >
>         > Under "Advanced option string to pass to the egress queueing
>         > disciplines," put in "dual-srchost nat"
>         >
>         > Because of the issues with WMM in the WiFi stack, I
>         recommend using the
>         > following custom iptables rules:
>         >
>         > iptables -t mangle -A FORWARD -o br-lan -j DSCP
>         --set-dscp-class be
>         > iptables -t mangle -A OUTPUT -o br-lan -j DSCP
>         --set-dscp-class be
>         > ip6tables -t mangle -A FORWARD -o br-lan -j DSCP
>         --set-dscp-class be
>         > ip6tables -t mangle -A OUTPUT -o br-lan -j DSCP
>         --set-dscp-class be
>         >
>         > You still need WMM to get the full speeds on the newer WiFi
>         modes.
>         >
>         > Noah
>
>         I'm a little uncertain of what parts of what are in what state of
>         working.  Current setup is 65/6Mbit Cable Modem - C7 - variety
>         of Win
>         PC's and other things.  I have a couple of AC1200 usb
>         adapters, a few
>         phones are AC as well,  as well as a lot of N devices.   Been
>         trying to
>         keep up with what's working with which interface.
>
>         I'm getting mostly decent numbers for bufferbloat on
>         DSLreports. Only
>         have DSLreports as a test method now, so I'm not much good for
>         decent data.
>         My PC to eth connection has next to nothing,  and most of the
>         time the
>         wifi looks very good with bloat below 30ms.
>
>         I have been seeing an anomaly, though, in that I get a fast
>         startup,
>         with very low latency,  download speed in the 50's, then after
>         5-10 sec
>         the speed drops drastically to the 10's for a few seconds, 
>         then rises
>         back up to the 50's again.  I was getting ready to write this
>         up the
>         other night,  then trying it again it was gone. Full, flat
>         speed thru
>         the whole test,  over many tries. Tonight,  it's back again. 
>         This was
>         on Noah's build, as well as tonight's change to Toke's 11-18
>         C7 build.
>         Not sure what to make of it.  Here's an example URL to
>         DSLreports,  if
>         that helps: http://www.dslreports.com/speedtest/6276874
>         <http://www.dslreports.com/speedtest/6276874>
>
>         It can do this on 2.4Ghz and on 5Ghz with the AC1200 dongle,
>         believe it
>         did it on the N dongle,  will try that shortly. Also, the game
>         player
>         is home from work now... we'll see what his comments are...
>
>         Jon
>
>
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast


[-- Attachment #2: Type: text/html, Size: 9317 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 255 bytes --]

Taking it for a spin.     So far, Toke's 11-18 build and my C7 is a pretty
nice ride...

Waiting to see if the extensive post I made last night made it through,
before going into detail.  Have some DSLReports links to share,  FWIW...

Waiting to see if o

[-- Attachment #2: Type: text/html, Size: 322 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jon Pike]

[-- Attachment #1: Type: text/plain, Size: 1140 bytes --]

Ok,  here's my earlier post.  Let me know if there are better measurements
I could make...

On Nov 19, 2016 5:59 PM, "Jon" <jhpike@cox.net> wrote:

> Yup,  this one showing up pretty insantaneously answers my question...
> I'll repost the stuff that didn't go...
>
> On Nov 19, 2016 11:51 AM, "Dave Taht" <dave.taht@gmail.com> wrote:
>
>> ? I don't see any other traffic from you in the last 24 hours.
>>
>> On Sat, Nov 19, 2016 at 11:49 AM, Jon Pike <jonpike54@gmail.com> wrote:
>> > Taking it for a spin.     So far, Toke's 11-18 build and my C7 is a
>> pretty
>> > nice ride...
>> >
>> > Waiting to see if the extensive post I made last night made it through,
>> > before going into detail.  Have some DSLReports links to share,  FWIW...
>> >
>> > Waiting to see if o
>> >
>> >
>> > _______________________________________________
>> > Make-wifi-fast mailing list
>> > Make-wifi-fast@lists.bufferbloat.net
>> > https://lists.bufferbloat.net/listinfo/make-wifi-fast
>> >
>>
>>
>>
>> --
>> Dave Täht
>> Let's go make home routers and wifi faster! With better software!
>> http://blog.cerowrt.org
>>
>

[-- Attachment #2: Type: text/html, Size: 2106 bytes --]

[Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

I created an updated build of current LEDE head with added airtime
fairness patches, and the patch to re-enable the intermediate queues for
ath10k. I.e., the contents of my repo here:

https://kau.toke.dk/git/lede/

Builds for some common ar71xx devices are here:

https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/

There's an assortment of packages included in the build. I *think* it's
possible to install more packages from the LEDE nightly builds by simply
doing an 'opkg update'.

I was going to do a build for the Turris Omnia as well, but turns out
they haven't upstreamed their code. On the other hand, they seem to be
tracking both LEDE and OpenWrt fairly closely, so the intermediate
queueing patches for ath9k are included in their builds.

I make no promises of regular updated builds, but feel free to take this
one for a spin :)

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

:puppy dog eyes:

ubnt uap-lite ?


On Thu, Nov 17, 2016 at 8:30 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> I created an updated build of current LEDE head with added airtime
> fairness patches, and the patch to re-enable the intermediate queues for
> ath10k. I.e., the contents of my repo here:
>
> https://kau.toke.dk/git/lede/
>
> Builds for some common ar71xx devices are here:
>
> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>
> There's an assortment of packages included in the build. I *think* it's
> possible to install more packages from the LEDE nightly builds by simply
> doing an 'opkg update'.
>
> I was going to do a build for the Turris Omnia as well, but turns out
> they haven't upstreamed their code. On the other hand, they seem to be
> tracking both LEDE and OpenWrt fairly closely, so the intermediate
> queueing patches for ath9k are included in their builds.
>
> I make no promises of regular updated builds, but feel free to take this
> one for a spin :)
>
> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Dave Taht <dave.taht@gmail.com> writes:

> :puppy dog eyes:
>
> ubnt uap-lite ?

The ubnt images fail to build because they get too large with the
current package selection...

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Noah Causin]

[-- Attachment #1: Type: text/plain, Size: 680 bytes --]

Toke,

Maybe previous packages from previous builds for different router models 
are getting added to the image and increasing the file size.

Maybe try a *make clean* followed by a *rm .config* to completely reset 
the config but keep the toolchain.


On 11/18/2016 10:30 AM, Toke Høiland-Jørgensen wrote:
> Dave Taht <dave.taht@gmail.com> writes:
>
>> :puppy dog eyes:
>>
>> ubnt uap-lite ?
> The ubnt images fail to build because they get too large with the
> current package selection...
>
> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast


[-- Attachment #2: Type: text/html, Size: 1426 bytes --]

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Noah Causin <n0manletter@gmail.com> writes:

> Toke,
>
> Maybe previous packages from previous builds for different router
> models are getting added to the image and increasing the file size.

No, these are packages I selected on purpose. They just don't fit in the
ubnt images (I'm building multiple images in one go). And for now I
don't want to manage multiple configs, so I'm afraid there'll be no ubnt
images from me...

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Noah Causin]

Hi Dave,

I built firmware for your Ubiquiti AC Lite from Toke's updated repository.

https://drive.google.com/drive/folders/0B86IEgnrkAk0YkxMalhlUUsyOFU?usp=sharing

There are two versions.

I made one using LEDE's stock ath10k firmware and one using 
Candelatech's ath10k firmware.

Packages:

luci-ssl-openssl kmod-sched-cake luci-app-sqm luci-app-ntpc nano ethtool 
ebtables-firewall ebtables-netfilter openssh-sftp-server dnsmasq wpad 
basic-iputils (ping, ping6, etc...)
iperf iperf3 netperf tcpdump

I changed the Cake and sqm-scripts Makefiles to point to their latest 
stable git versions.

I changed the ath10k-firmware and ath10k-ct Makefiles to point to 
updated versions.

LEDE stock ath10k firmware:

From
firmware-5.bin_10.2.4.70.54
To
firmware-5.bin_10.2.4.70.58

Candelatech ath10k firmware:

From
firmware-2-ct-full-community.bin-18.rc1-lede
To
firmware-2-ct-full-community.bin-18.rc2-lede

Noah Causin

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

Thx noah! I just tried this, and aside from lacking babeld by default
(I need this to dynamically reroute the testbed), your QCA build works
on the uap-lite.

On the ath10k side, I get ~300mbit download rates from the osx box,
(vs 150-200 on the archer) via the uap-lite, with about 12ms latency
(in line with expectations). (in both test series I was using vht80).
However, I do not see codel induced drops on that side:

9a:d9# cat aqm
tid ac backlog-bytes backlog-packets new-flows drops marks overlimit
collisions tx-bytes tx-packets
0 2 0 0 204580 0 0 0 0 1463032915 954453

And latencies get way higher if I use vht40, so I kind of suspect we
still have multiple issues here.

as for the first issue: dropping at the ag71xx? NAPI bug (as pointed
out by eric dumazet)? Not enough patches?

As for the latter, I was concerned about the NAPI patches recently
added in for the ath10k, but don't know if those are in the 4.4
backport to lede.

I'd rather like to try a napi budget of like 16, rather than 64 on
router platforms that can't achieve a gbit in the first place.

...

There is a great deal more work to do on testing more stations (I
usually have to shut down the yurtlab in the winter, and cluttering up
the apt with dozens of wifi stations is a matter of ongoing
negotiation with gf), channels, rates, etc - I'm merely happy to have
a smoke test succeed enough to be able to encourage more to try this
stuff - and my first priority is nailing the ath9k to perfection.....

the snapon build server has long been standing idle, if you want to
use that for future builds.

...

Testing the 2.4ghz (ath9k HT20) side of the uap, I get similar results
as to the archer, and  sane codel drops, etc, with about 40Mbits of
throughput and sane latency (given how crowded the air is at the apt
I'm surprised it's even this good)

/sys/kernel/debug/ieee80211/phy1/netdev:wlan1/stations/48:d7:05:c0:9a:d9

# cat aqm
tid ac backlog-bytes backlog-packets new-flows drops marks overlimit
collisions tx-bytes tx-packets
0 2 168740 110 11230 644 0 0 0 239846710 156783



On Fri, Nov 18, 2016 at 7:00 PM, Noah Causin <n0manletter@gmail.com> wrote:
> Hi Dave,
>
> I built firmware for your Ubiquiti AC Lite from Toke's updated repository.
>
> https://drive.google.com/drive/folders/0B86IEgnrkAk0YkxMalhlUUsyOFU?usp=sharing
>
> There are two versions.
>
> I made one using LEDE's stock ath10k firmware and one using Candelatech's
> ath10k firmware.
>
> Packages:
>
> luci-ssl-openssl kmod-sched-cake luci-app-sqm luci-app-ntpc nano ethtool
> ebtables-firewall ebtables-netfilter openssh-sftp-server dnsmasq wpad
> basic-iputils (ping, ping6, etc...)
> iperf iperf3 netperf tcpdump
>
> I changed the Cake and sqm-scripts Makefiles to point to their latest stable
> git versions.
>
> I changed the ath10k-firmware and ath10k-ct Makefiles to point to updated
> versions.
>
> LEDE stock ath10k firmware:
>
> From
> firmware-5.bin_10.2.4.70.54
> To
> firmware-5.bin_10.2.4.70.58
>
> Candelatech ath10k firmware:
>
> From
> firmware-2-ct-full-community.bin-18.rc1-lede
> To
> firmware-2-ct-full-community.bin-18.rc2-lede
>
> Noah Causin
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Noah Causin]

Does babeld install by opkg, or would you like me to integrate it into 
the firmware?

If it needs to be integrated into the firmware, please let me know what 
packages you would need installed and any extras you would like.

On 11/19/2016 12:18 PM, Dave Taht wrote:
> Thx noah! I just tried this, and aside from lacking babeld by default
> (I need this to dynamically reroute the testbed), your QCA build works
> on the uap-lite.

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

I tested this against the archer c7v2 just now.

It doesn't crash after an hour of extensive testing.

For testers, I note the airtime fairness portion of these patches is ath9k only.

The ath10k firmware build on toke's site is out of date, but the qca
firmware can be installed for the ath10k regardless after modifying
/etc/opkg/distfeeds.conf to point to his site.

Unencrypted, using an osx box as the client, using flent with
tcp_ndown and --test-parameter=download_streams=1 8, and 24

The ath10k works - with download rates in vht80 ranging from 210 to
140, eating 97% of cpu while forwarding, with less than < 30ms latency
at the 90th percentile.
The ath9k works in HT20 mode, eating 40% of cpu while forwarding, with
less than <30ms also, speed in the 50mbit range.

I am about 10 ft from the AP in the SF apt, in a fairly noisy space
(can hear over 50 other APs). The only sane way to get a before/after
comparison is for me to reflash...

but: I will try a bridged connection next, then try crypted, then go
back to looking at the TSQ issue on x86.


On Thu, Nov 17, 2016 at 8:30 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> I created an updated build of current LEDE head with added airtime
> fairness patches, and the patch to re-enable the intermediate queues for
> ath10k. I.e., the contents of my repo here:
>
> https://kau.toke.dk/git/lede/
>
> Builds for some common ar71xx devices are here:
>
> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>
> There's an assortment of packages included in the build. I *think* it's
> possible to install more packages from the LEDE nightly builds by simply
> doing an 'opkg update'.
>
> I was going to do a build for the Turris Omnia as well, but turns out
> they haven't upstreamed their code. On the other hand, they seem to be
> tracking both LEDE and OpenWrt fairly closely, so the intermediate
> queueing patches for ath9k are included in their builds.
>
> I make no promises of regular updated builds, but feel free to take this
> one for a spin :)
>
> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

stats collection doesn't make a lot of sense.

root@archer-2:/sys/kernel/debug/ieee80211/phy1/ath9k# cat xmit
                            BE         BK        VI        VO

MPDUs Queued:                0          0         0       811
MPDUs Completed:            78          1         1       828
MPDUs XRetried:              0          0         0         1
Aggregates:              46528         99     13023         0 ?
AMPDUs Queued HW:            0          0         0         0
AMPDUs Completed:       986000       4544    244768         0 ??
AMPDUs Retried:          18377         23      4073         0

On Thu, Nov 17, 2016 at 11:08 AM, Dave Taht <dave.taht@gmail.com> wrote:
> I tested this against the archer c7v2 just now.
>
> It doesn't crash after an hour of extensive testing.
>
> For testers, I note the airtime fairness portion of these patches is ath9k only.
>
> The ath10k firmware build on toke's site is out of date, but the qca
> firmware can be installed for the ath10k regardless after modifying
> /etc/opkg/distfeeds.conf to point to his site.
>
> Unencrypted, using an osx box as the client, using flent with
> tcp_ndown and --test-parameter=download_streams=1 8, and 24
>
> The ath10k works - with download rates in vht80 ranging from 210 to
> 140, eating 97% of cpu while forwarding, with less than < 30ms latency
> at the 90th percentile.
> The ath9k works in HT20 mode, eating 40% of cpu while forwarding, with
> less than <30ms also, speed in the 50mbit range.
>
> I am about 10 ft from the AP in the SF apt, in a fairly noisy space
> (can hear over 50 other APs). The only sane way to get a before/after
> comparison is for me to reflash...
>
> but: I will try a bridged connection next, then try crypted, then go
> back to looking at the TSQ issue on x86.
>
>
> On Thu, Nov 17, 2016 at 8:30 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>> I created an updated build of current LEDE head with added airtime
>> fairness patches, and the patch to re-enable the intermediate queues for
>> ath10k. I.e., the contents of my repo here:
>>
>> https://kau.toke.dk/git/lede/
>>
>> Builds for some common ar71xx devices are here:
>>
>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>
>> There's an assortment of packages included in the build. I *think* it's
>> possible to install more packages from the LEDE nightly builds by simply
>> doing an 'opkg update'.
>>
>> I was going to do a build for the Turris Omnia as well, but turns out
>> they haven't upstreamed their code. On the other hand, they seem to be
>> tracking both LEDE and OpenWrt fairly closely, so the intermediate
>> queueing patches for ath9k are included in their builds.
>>
>> I make no promises of regular updated builds, but feel free to take this
>> one for a spin :)
>>
>> -Toke
>> _______________________________________________
>> Make-wifi-fast mailing list
>> Make-wifi-fast@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>
>
>
> --
> Dave Täht
> Let's go make home routers and wifi faster! With better software!
> http://blog.cerowrt.org



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Kevin Darbyshire-Bryant]

On 17/11/16 19:08, Dave Taht wrote:

> The ath10k firmware build on toke's site is out of date, but the qca
> firmware can be installed for the ath10k regardless after modifying
> /etc/opkg/distfeeds.conf to point to his site.

Can you clarify the above please Dave, 'cos it doesn't make sense to 
me...it's out of date on his site but point to his site to get the latest?

Kevin

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

opkg install ath10k-firmware-qca988x kmod-ath10k works

after changing /etc/opkg/distfeeds.conf to have this as a first line

src/gz reboot_core
https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/packages/

On Thu, Nov 17, 2016 at 11:36 AM, Kevin Darbyshire-Bryant
<kevin@darbyshire-bryant.me.uk> wrote:
>
>
> On 17/11/16 19:08, Dave Taht wrote:
>
>> The ath10k firmware build on toke's site is out of date, but the qca
>> firmware can be installed for the ath10k regardless after modifying
>> /etc/opkg/distfeeds.conf to point to his site.
>
>
> Can you clarify the above please Dave, 'cos it doesn't make sense to
> me...it's out of date on his site but point to his site to get the latest?
>
> Kevin
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

in other news, the 6 "c.h.i.p"s I'd intended to do more multistation
testing with (back in august) finally arrived today.

On Thu, Nov 17, 2016 at 11:45 AM, Dave Taht <dave.taht@gmail.com> wrote:
> opkg install ath10k-firmware-qca988x kmod-ath10k works
>
> after changing /etc/opkg/distfeeds.conf to have this as a first line
>
> src/gz reboot_core
> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/packages/
>
> On Thu, Nov 17, 2016 at 11:36 AM, Kevin Darbyshire-Bryant
> <kevin@darbyshire-bryant.me.uk> wrote:
>>
>>
>> On 17/11/16 19:08, Dave Taht wrote:
>>
>>> The ath10k firmware build on toke's site is out of date, but the qca
>>> firmware can be installed for the ath10k regardless after modifying
>>> /etc/opkg/distfeeds.conf to point to his site.
>>
>>
>> Can you clarify the above please Dave, 'cos it doesn't make sense to
>> me...it's out of date on his site but point to his site to get the latest?
>>
>> Kevin
>>
>> _______________________________________________
>> Make-wifi-fast mailing list
>> Make-wifi-fast@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>
>
>
> --
> Dave Täht
> Let's go make home routers and wifi faster! With better software!
> http://blog.cerowrt.org



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Dave Taht <dave.taht@gmail.com> writes:

> I tested this against the archer c7v2 just now.
>
> It doesn't crash after an hour of extensive testing.
>
> For testers, I note the airtime fairness portion of these patches is ath9k only.
>
> The ath10k firmware build on toke's site is out of date, but the qca
> firmware can be installed for the ath10k regardless after modifying
> /etc/opkg/distfeeds.conf to point to his site.

Updated the builds to include the ath10k driver and firmware by default.

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Jonathan Morton]

> On 17 Nov, 2016, at 18:30, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> 
> I created an updated build of current LEDE head with added airtime
> fairness patches, and the patch to re-enable the intermediate queues for
> ath10k. I.e., the contents of my repo here:

I could do with getting my Buffalo router working again.  One of my own LEDE builds appeared to semi-brick it, and it’s apparently *very* finicky about the recovery ramdisk images.  As a result, I have rather fewer dual-GigE machines available than I’d like.

 - Jonathan Morton

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Toke Høiland-Jørgensen <toke@toke.dk> writes:

> I created an updated build of current LEDE head with added airtime
> fairness patches, and the patch to re-enable the intermediate queues for
> ath10k. I.e., the contents of my repo here:
>
> https://kau.toke.dk/git/lede/
>
> Builds for some common ar71xx devices are here:
>
> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/


These builds have now been updated to a new version of the airtime
fairness patch (that was also posted to the list a few hours ago).
Please take them for a spin. Once I have confirmation that they don't
blow up catastrophically, I'll submit the patch for inclusion in LEDE :)

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

It looks like you dropped the quantum 1514/codel 5ms from the patchset?

On Thu, Nov 24, 2016 at 7:01 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>
>> I created an updated build of current LEDE head with added airtime
>> fairness patches, and the patch to re-enable the intermediate queues for
>> ath10k. I.e., the contents of my repo here:
>>
>> https://kau.toke.dk/git/lede/
>>
>> Builds for some common ar71xx devices are here:
>>
>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>
>
> These builds have now been updated to a new version of the airtime
> fairness patch (that was also posted to the list a few hours ago).
> Please take them for a spin. Once I have confirmation that they don't
> blow up catastrophically, I'll submit the patch for inclusion in LEDE :)
>
> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Yup. Planning to look into the codel tuning issue next. Until then I figured I'd keep the patch set small...

On 24 November 2016 18:01:32 CET, Dave Taht <dave.taht@gmail.com> wrote:
>It looks like you dropped the quantum 1514/codel 5ms from the patchset?
>
>On Thu, Nov 24, 2016 at 7:01 AM, Toke Høiland-Jørgensen <toke@toke.dk>
>wrote:
>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>
>>> I created an updated build of current LEDE head with added airtime
>>> fairness patches, and the patch to re-enable the intermediate queues
>for
>>> ath10k. I.e., the contents of my repo here:
>>>
>>> https://kau.toke.dk/git/lede/
>>>
>>> Builds for some common ar71xx devices are here:
>>>
>>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>
>>
>> These builds have now been updated to a new version of the airtime
>> fairness patch (that was also posted to the list a few hours ago).
>> Please take them for a spin. Once I have confirmation that they don't
>> blow up catastrophically, I'll submit the patch for inclusion in LEDE
>:)
>>
>> -Toke
>> _______________________________________________
>> Make-wifi-fast mailing list
>> Make-wifi-fast@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/make-wifi-fast

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

It's more the quantum than the target that is an issue on *wrt, IMHO.

(I got no problem with testing each of these in isolation)

I am starting a test run now. So far no crashes on the ath9k side on the archer.

Two things I am not testing on this platform are HT40 and crypto. (I
was using the x86 versions for that). Nor was I monitoring cpu usage
on the previous archer test runs (sigh), and for some value of
consistency, am not doing it on these either. I'd *really* like more
ht40.....

I will finish bootstrapping up my "c.h.i.p." nodes to do better
airtime fairness testing soon via the rtt_var tests.

Ideally I'd like more to join in on this hopefully last push to get
the core patches upstream.

Note that in the US...

it's the thanksgiving holiday which can be interpreted as either
people spending more time with family - or being free enough from work
to dig in and test the heck out of this stuff (or in my case, fire off
a big automated test, while spending time with family). :)


On Thu, Nov 24, 2016 at 9:14 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Yup. Planning to look into the codel tuning issue next. Until then I figured I'd keep the patch set small...
>
> On 24 November 2016 18:01:32 CET, Dave Taht <dave.taht@gmail.com> wrote:
>>It looks like you dropped the quantum 1514/codel 5ms from the patchset?
>>
>>On Thu, Nov 24, 2016 at 7:01 AM, Toke Høiland-Jørgensen <toke@toke.dk>
>>wrote:
>>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>>
>>>> I created an updated build of current LEDE head with added airtime
>>>> fairness patches, and the patch to re-enable the intermediate queues
>>for
>>>> ath10k. I.e., the contents of my repo here:
>>>>
>>>> https://kau.toke.dk/git/lede/
>>>>
>>>> Builds for some common ar71xx devices are here:
>>>>
>>>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>>
>>>
>>> These builds have now been updated to a new version of the airtime
>>> fairness patch (that was also posted to the list a few hours ago).
>>> Please take them for a spin. Once I have confirmation that they don't
>>> blow up catastrophically, I'll submit the patch for inclusion in LEDE
>>:)
>>>
>>> -Toke
>>> _______________________________________________
>>> Make-wifi-fast mailing list
>>> Make-wifi-fast@lists.bufferbloat.net
>>> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

on the rrul test, the BK queue got no connection, and I lost the
connection several times during it. While I was testing
the ath9k, a new message showed up in the ath10k log.

It would not surprise me if a shorter codel target made the hardware
queues work better.

Otherwise the short test runs I just did (tcp_nup, tcp_ndown) were fine.

 0x00001000
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.777307] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.784831] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.792409] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.799962] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.807518] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.815016] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.822565] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.830113] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.837657] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon
Thu Nov 24 17:50:58 2016 kern.warn kernel: [ 2760.845181] ath10k_pci
0000:01:00.0: SWBA overrun on vdev 0, skipped old beacon

On Thu, Nov 24, 2016 at 9:35 AM, Dave Taht <dave.taht@gmail.com> wrote:
> It's more the quantum than the target that is an issue on *wrt, IMHO.
>
> (I got no problem with testing each of these in isolation)
>
> I am starting a test run now. So far no crashes on the ath9k side on the archer.
>
> Two things I am not testing on this platform are HT40 and crypto. (I
> was using the x86 versions for that). Nor was I monitoring cpu usage
> on the previous archer test runs (sigh), and for some value of
> consistency, am not doing it on these either. I'd *really* like more
> ht40.....
>
> I will finish bootstrapping up my "c.h.i.p." nodes to do better
> airtime fairness testing soon via the rtt_var tests.
>
> Ideally I'd like more to join in on this hopefully last push to get
> the core patches upstream.
>
> Note that in the US...
>
> it's the thanksgiving holiday which can be interpreted as either
> people spending more time with family - or being free enough from work
> to dig in and test the heck out of this stuff (or in my case, fire off
> a big automated test, while spending time with family). :)
>
>
> On Thu, Nov 24, 2016 at 9:14 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>> Yup. Planning to look into the codel tuning issue next. Until then I figured I'd keep the patch set small...
>>
>> On 24 November 2016 18:01:32 CET, Dave Taht <dave.taht@gmail.com> wrote:
>>>It looks like you dropped the quantum 1514/codel 5ms from the patchset?
>>>
>>>On Thu, Nov 24, 2016 at 7:01 AM, Toke Høiland-Jørgensen <toke@toke.dk>
>>>wrote:
>>>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>>>
>>>>> I created an updated build of current LEDE head with added airtime
>>>>> fairness patches, and the patch to re-enable the intermediate queues
>>>for
>>>>> ath10k. I.e., the contents of my repo here:
>>>>>
>>>>> https://kau.toke.dk/git/lede/
>>>>>
>>>>> Builds for some common ar71xx devices are here:
>>>>>
>>>>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>>>
>>>>
>>>> These builds have now been updated to a new version of the airtime
>>>> fairness patch (that was also posted to the list a few hours ago).
>>>> Please take them for a spin. Once I have confirmation that they don't
>>>> blow up catastrophically, I'll submit the patch for inclusion in LEDE
>>>:)
>>>>
>>>> -Toke
>>>> _______________________________________________
>>>> Make-wifi-fast mailing list
>>>> Make-wifi-fast@lists.bufferbloat.net
>>>> https://lists.bufferbloat.net/listinfo/make-wifi-fast
>>
>
>
>
> --
> Dave Täht
> Let's go make home routers and wifi faster! With better software!
> http://blog.cerowrt.org



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Dave Taht]

tested on the archer ath9k, all day, no crashes, latency and
throughput performance within expected parameters on rrul_be, rrul,
tcp_nup, tcp_ndown, against 1-3 stations (linux ath9k + osx + roku),
against video streaming, and various combinations of the above tests
against the above platform.

you can add my:

tested-by: Dave Taht <dave.taht@gmail.com>


On Thu, Nov 24, 2016 at 7:01 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>
>> I created an updated build of current LEDE head with added airtime
>> fairness patches, and the patch to re-enable the intermediate queues for
>> ath10k. I.e., the contents of my repo here:
>>
>> https://kau.toke.dk/git/lede/
>>
>> Builds for some common ar71xx devices are here:
>>
>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>
>
> These builds have now been updated to a new version of the airtime
> fairness patch (that was also posted to the list a few hours ago).
> Please take them for a spin. Once I have confirmation that they don't
> blow up catastrophically, I'll submit the patch for inclusion in LEDE :)
>
> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Dave Taht <dave.taht@gmail.com> writes:

> tested on the archer ath9k, all day, no crashes, latency and
> throughput performance within expected parameters on rrul_be, rrul,
> tcp_nup, tcp_ndown, against 1-3 stations (linux ath9k + osx + roku),
> against video streaming, and various combinations of the above tests
> against the above platform.
>
> you can add my:
>
> tested-by: Dave Taht <dave.taht@gmail.com>

Awesome, thanks. Submitted to upstream LEDE :)

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

On Thu, Nov 24, 2016 at 7:01 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>
>> I created an updated build of current LEDE head with added airtime
>> fairness patches, and the patch to re-enable the intermediate queues for
>> ath10k. I.e., the contents of my repo here:
>>
>> https://kau.toke.dk/git/lede/
>>
>> Builds for some common ar71xx devices are here:
>>
>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>

Hi Toke, I tested with 6 mobile phones connected on a 20 MB/s
Internet. No crashes so far on ath9k. No pppoe support out of the box
for the firmware for the WAN interface.


>
> These builds have now been updated to a new version of the airtime
> fairness patch (that was also posted to the list a few hours ago).
> Please take them for a spin. Once I have confirmation that they don't
> blow up catastrophically, I'll submit the patch for inclusion in LEDE :)
>

It would be nice if we could get firmware images, and also please can
you ship it with mod-pppoe modules built-in for bridged connections ?


> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Loganaden Velvindron <loganaden@gmail.com> writes:

> On Thu, Nov 24, 2016 at 7:01 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>
>>> I created an updated build of current LEDE head with added airtime
>>> fairness patches, and the patch to re-enable the intermediate queues for
>>> ath10k. I.e., the contents of my repo here:
>>>
>>> https://kau.toke.dk/git/lede/
>>>
>>> Builds for some common ar71xx devices are here:
>>>
>>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>
>
> Hi Toke, I tested with 6 mobile phones connected on a 20 MB/s
> Internet. No crashes so far on ath9k. No pppoe support out of the box
> for the firmware for the WAN interface.

Cool, thanks :)

>>
>> These builds have now been updated to a new version of the airtime
>> fairness patch (that was also posted to the list a few hours ago).
>> Please take them for a spin. Once I have confirmation that they don't
>> blow up catastrophically, I'll submit the patch for inclusion in LEDE :)
>>
>
> It would be nice if we could get firmware images, and also please can
> you ship it with mod-pppoe modules built-in for bridged connections ?

I'm hoping that I won't have to keep building images for so much longer.
But I'll add pppoe the next time round, sure :)

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

On Fri, Nov 25, 2016 at 6:51 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Loganaden Velvindron <loganaden@gmail.com> writes:
>
[...]
>
> I'm hoping that I won't have to keep building images for so much longer.
> But I'll add pppoe the next time round, sure :)
>
> -Toke

ping :)

I would be nice to get a fresh image just before you go on holidays
with pppoe, the latest patch for ath9k & ath10k !

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Toke Høiland-Jørgensen <toke@toke.dk> writes:

> Loganaden Velvindron <loganaden@gmail.com> writes:
>
>> On Thu, Nov 24, 2016 at 7:01 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>>
>>>> I created an updated build of current LEDE head with added airtime
>>>> fairness patches, and the patch to re-enable the intermediate queues for
>>>> ath10k. I.e., the contents of my repo here:
>>>>
>>>> https://kau.toke.dk/git/lede/
>>>>
>>>> Builds for some common ar71xx devices are here:
>>>>
>>>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>>
>>
>> Hi Toke, I tested with 6 mobile phones connected on a 20 MB/s
>> Internet. No crashes so far on ath9k. No pppoe support out of the box
>> for the firmware for the WAN interface.
>
> Cool, thanks :)
>
>>>
>>> These builds have now been updated to a new version of the airtime
>>> fairness patch (that was also posted to the list a few hours ago).
>>> Please take them for a spin. Once I have confirmation that they don't
>>> blow up catastrophically, I'll submit the patch for inclusion in LEDE :)
>>>
>>
>> It would be nice if we could get firmware images, and also please can
>> you ship it with mod-pppoe modules built-in for bridged connections ?
>
> I'm hoping that I won't have to keep building images for so much longer.
> But I'll add pppoe the next time round, sure :)


Since the airtime fairness patches haven't been merged yet, here's an
updated build, rebased on upstream LEDE head:

https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/


Note that the file names have changed. This build should also include
the kmod-pppoe package.

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

On Sun, Dec 11, 2016 at 2:28 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>
>> Loganaden Velvindron <loganaden@gmail.com> writes:
>>
>>> On Thu, Nov 24, 2016 at 7:01 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>>>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>>>
>>>>> I created an updated build of current LEDE head with added airtime
>>>>> fairness patches, and the patch to re-enable the intermediate queues for
>>>>> ath10k. I.e., the contents of my repo here:
>>>>>
>>>>> https://kau.toke.dk/git/lede/
>>>>>
>>>>> Builds for some common ar71xx devices are here:
>>>>>
>>>>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>>>
>>>
>>> Hi Toke, I tested with 6 mobile phones connected on a 20 MB/s
>>> Internet. No crashes so far on ath9k. No pppoe support out of the box
>>> for the firmware for the WAN interface.
>>
>> Cool, thanks :)
>>
>>>>
>>>> These builds have now been updated to a new version of the airtime
>>>> fairness patch (that was also posted to the list a few hours ago).
>>>> Please take them for a spin. Once I have confirmation that they don't
>>>> blow up catastrophically, I'll submit the patch for inclusion in LEDE :)
>>>>
>>>
>>> It would be nice if we could get firmware images, and also please can
>>> you ship it with mod-pppoe modules built-in for bridged connections ?
>>
>> I'm hoping that I won't have to keep building images for so much longer.
>> But I'll add pppoe the next time round, sure :)
>
>
> Since the airtime fairness patches haven't been merged yet, here's an
> updated build, rebased on upstream LEDE head:
>
> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>
>
> Note that the file names have changed. This build should also include
> the kmod-pppoe package.
>
> -Toke
Thank you.

Tested with 4 mobile phones, while stressed with flent on a linux
laptop having an ath5k chipset. Pretty good.

Btw, I was wondering about replacing the ath5k radio with an ath9k
card. Any recommendation ?

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Loganaden Velvindron <loganaden@gmail.com> writes:

> On Sun, Dec 11, 2016 at 2:28 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>
>>> Loganaden Velvindron <loganaden@gmail.com> writes:
>>>
>>>> On Thu, Nov 24, 2016 at 7:01 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>>>>> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>>>>>
>>>>>> I created an updated build of current LEDE head with added airtime
>>>>>> fairness patches, and the patch to re-enable the intermediate queues for
>>>>>> ath10k. I.e., the contents of my repo here:
>>>>>>
>>>>>> https://kau.toke.dk/git/lede/
>>>>>>
>>>>>> Builds for some common ar71xx devices are here:
>>>>>>
>>>>>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>>>>
>>>>
>>>> Hi Toke, I tested with 6 mobile phones connected on a 20 MB/s
>>>> Internet. No crashes so far on ath9k. No pppoe support out of the box
>>>> for the firmware for the WAN interface.
>>>
>>> Cool, thanks :)
>>>
>>>>>
>>>>> These builds have now been updated to a new version of the airtime
>>>>> fairness patch (that was also posted to the list a few hours ago).
>>>>> Please take them for a spin. Once I have confirmation that they don't
>>>>> blow up catastrophically, I'll submit the patch for inclusion in LEDE :)
>>>>>
>>>>
>>>> It would be nice if we could get firmware images, and also please can
>>>> you ship it with mod-pppoe modules built-in for bridged connections ?
>>>
>>> I'm hoping that I won't have to keep building images for so much longer.
>>> But I'll add pppoe the next time round, sure :)
>>
>>
>> Since the airtime fairness patches haven't been merged yet, here's an
>> updated build, rebased on upstream LEDE head:
>>
>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>
>>
>> Note that the file names have changed. This build should also include
>> the kmod-pppoe package.
>>
>> -Toke
> Thank you.
>
> Tested with 4 mobile phones, while stressed with flent on a linux
> laptop having an ath5k chipset. Pretty good.

Awesome, thanks!

> Btw, I was wondering about replacing the ath5k radio with an ath9k
> card. Any recommendation ?

Anything that fits in your laptop should work. The cards mainly differ
in whether they are dual-band, the connection and form factor, and how
many antennas they use. This is probably a good starting point:
https://wireless.wiki.kernel.org/en/users/drivers/ath9k/products

-Toke

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Leonidas Spyropoulos]

On 10/12/16, Toke Høiland-Jørgensen wrote:
> Since the airtime fairness patches haven't been merged yet, here's an
> updated build, rebased on upstream LEDE head:
> 
> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
> 
> 
> Note that the file names have changed. This build should also include
> the kmod-pppoe package.
> 

I tried the build on Archer C7 v2.0 and indeed the kmod-pppoe package
was there but was missing the ppp-mod-pppoe package. I'm not sure where
would that be I think it's part of 'base'.

@Toke is it possible for you to share the config for Archer C7 v2.0 so I
can build it myself?

Thanks,

-- 
Leonidas Spyropoulos

A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

On Tue, Dec 13, 2016 at 9:50 PM, Leonidas Spyropoulos
<artafinde@gmail.com> wrote:
> On 10/12/16, Toke Høiland-Jørgensen wrote:
>> Since the airtime fairness patches haven't been merged yet, here's an
>> updated build, rebased on upstream LEDE head:
>>
>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>>
>>
>> Note that the file names have changed. This build should also include
>> the kmod-pppoe package.
>>
>
> I tried the build on Archer C7 v2.0 and indeed the kmod-pppoe package
> was there but was missing the ppp-mod-pppoe package. I'm not sure where
> would that be I think it's part of 'base'.
>
> @Toke is it possible for you to share the config for Archer C7 v2.0 so I
> can build it myself?
>
> Thanks,
>

Hi Leonidas,

You can change the distribution feed to point to toke's directory.
Then, you get ppp-mod-pppoe package.


> --
> Leonidas Spyropoulos
>
> A: Because it messes up the order in which people normally read text.
> Q: Why is it such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing on usenet and in e-mail?
>
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Leonidas Spyropoulos]

On Tue, Dec 13, 2016 at 5:52 PM, Loganaden Velvindron
<loganaden@gmail.com> wrote:
>
>
> You can change the distribution feed to point to toke's directory.
> Then, you get ppp-mod-pppoe package.
>

Do you mean prior of building the image? Because after even if I do
change the opkg feeds I won't be able to update them as I cannot
connect to internet.

-- 
Leonidas Spyropoulos

A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Loganaden Velvindron]

On Tue, Dec 13, 2016 at 9:57 PM, Leonidas Spyropoulos
<artafinde@gmail.com> wrote:
> On Tue, Dec 13, 2016 at 5:52 PM, Loganaden Velvindron
> <loganaden@gmail.com> wrote:
>>
>>
>> You can change the distribution feed to point to toke's directory.
>> Then, you get ppp-mod-pppoe package.
>>
>
> Do you mean prior of building the image? Because after even if I do
> change the opkg feeds I won't be able to update them as I cannot
> connect to internet.
>

You can use a router to do the pppoe negociation, and then, connect
the tp-link in NAT mode. Then, you download the package via luci.
That's how I did it.


> --
> Leonidas Spyropoulos
>
> A: Because it messes up the order in which people normally read text.
> Q: Why is it such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing on usenet and in e-mail?

Re: [Make-wifi-fast] Updated LEDE build with airtime patches

[Toke Høiland-Jørgensen]

Leonidas Spyropoulos <artafinde@gmail.com> writes:

> On 10/12/16, Toke Høiland-Jørgensen wrote:
>> Since the airtime fairness patches haven't been merged yet, here's an
>> updated build, rebased on upstream LEDE head:
>> 
>> https://kau.toke.dk/lede/airtime-fairness-builds/ar71xx/generic/
>> 
>> 
>> Note that the file names have changed. This build should also include
>> the kmod-pppoe package.
>> 
>
> I tried the build on Archer C7 v2.0 and indeed the kmod-pppoe package
> was there but was missing the ppp-mod-pppoe package. I'm not sure where
> would that be I think it's part of 'base'.
>
> @Toke is it possible for you to share the config for Archer C7 v2.0 so
> I can build it myself?

There's no special config needed to enable the patches, you can just
pull my tree (https://kau.toke.dk/git/lede/) and build with whatever
config you prefer :)

-Toke