OpenWrt Forum Archive

Topic: davidc502 1900ac 3200acm builds

The content of this topic has been archived between 26 Feb 2018 and 7 May 2018. Unfortunately there are posts – most likely complete pages – missing.

davidc502 wrote:

There is a new build uploaded to the site r4164.  For those on the 3200acm, there was a wifi commit a few days back that changes "Corrected the rx buffer size". There were also commits for "ClearFog".

I've only been up, on the new image, for around 15 minutes, and the wifi is absolutely great so far. I'm interested in hearing from others on the 3200acm that are testing the new wifi driver.

Thanks,

One odd thing I've noticed (very minor).  I see you took my advice and included darkmatter in addition to material.  But when selecting darkmatter, the theme stays on material.

Edit: Wait, I see, darkmatter was never actually installed.  But there was a selection in the drop-down list to choose it.  I wonder if it had to do with restoring my previous configuration?  Never happened before though...  SImply installed darkmatter, all good to go.  Running perfectly so far!

(Last edited by starcms on 21 May 2017, 00:56)

I picked up a 3200acm a few days ago, and built my own image based on the david502c config.seed and latest mwlwifi commit (i.e., including the rx buffer fix.)

Everything seems to work nicely, except when I try to enable multiple SSID on the same radio.  When I do that, my devices are lucky to connect at all - and even when it does connect, it is unstable.  I saw a related issue report on the mwlwifi github, but didn't find any mention of the issue here on this thread.  I'm curious if it's working for others, or if it's just not a feature people are commonly using.

Thanks for your efforts in getting this model working with LEDE.

starcms wrote:

Yep, no issues selecting any of the DFS channels.

Right. No more DFS issues, as well as 5G is now 30dbm.

But other issue come to play with David's firmware - Luci lost wireless tab under "networks", as well as wireless stats in "Overview" is empty.
Is anyone have an idea how it happens? Any clue about the place to look for Luci logs? Interfaces working ok and visible in collectd, for example, but not in Luci web UI.

BTW, signal quality (for me) is not so different in compare if you just use "BM" for country code, so not a big reason to do it, according to bug mentioned above.

David

Lantis (over on the Gargoyle website) has very kindly provided me with a copy of gargoyle for my WRT1900ACS with the iptables-ndpi package included. It works well in that it blocks a whole load of websites/apps (like layer 7 should) but it has made my router unstable and it randomly reboots every day or two. I see there was a version of the package compiled for CC but not for LEDE or the latest OpenWRT snapshot. Is there any chance of adding this as an available package to download to try with your LEDE snapshot to see if it runs stably under LEDE  as its a really useful package with kids in the house ?
Thanks
Roger.

T-Troll wrote:
starcms wrote:

Yep, no issues selecting any of the DFS channels.

Right. No more DFS issues, as well as 5G is now 30dbm.

But other issue come to play with David's firmware - Luci lost wireless tab under "networks", as well as wireless stats in "Overview" is empty.
Is anyone have an idea how it happens? Any clue about the place to look for Luci logs? Interfaces working ok and visible in collectd, for example, but not in Luci web UI.

BTW, signal quality (for me) is not so different in compare if you just use "BM" for country code, so not a big reason to do it, according to bug mentioned above.

Are you talking about a 3200ACM?  Wireless stats (signal level, bandwidth, etc) are not shown yet.  A known issue with the driver that will eventually be fixed.

Or are you talking about applying the hack I mentioned to disable DFS on a 1200 V1, 1900V1, 1900V2, or 1900ACS V1?  If talking about the hack, the issues you mentioned aren't related to that  Or shouldn't be anyway.. I don't know what would cause the wireless tab under networks to disappear, or anything else you mentioned.

But you could always try restoring the backup of the original cfg80211.ko file or reflashing @david's build and see if things go back to normal. Did you follow the directions exactly?  I hope you didn't copy the modified cfg80211.ko file from an older build of @david's to a new one?  You need to apply the hack to the original cfg80211.ko file in each updated build.

I still am not sure if I'm even answering your question.  What you quoted me saying is so ambiguous all by itself.

(Last edited by starcms on 21 May 2017, 11:50)

davidc502 wrote:

There is a new build uploaded to the site r4164.  For those on the 3200acm, there was a wifi commit a few days back that changes "Corrected the rx buffer size". There were also commits for "ClearFog".

I've only been up, on the new image, for around 15 minutes, and the wifi is absolutely great so far. I'm interested in hearing from others on the 3200acm that are testing the new wifi driver.

Thanks,

r4164 has been running completely stable on my 3200ACM for some hours now!

@starcms

Does your reghack method work for older devices to unlock channels 12-13 and extend power limits?

An example would be applying this hack on a WNR1000v2, which supports Openwrt and LEDE to this day. I would appreciate this greatly as the range on it is terrible, although not my main router it supplements the WRT1200AC's terrible 2.4ghz performance.

davidc502 wrote:

There is a new build uploaded to the site r4164.  I'm interested in hearing from others on the 3200acm that are testing the new wifi driver.

got it, no problems with the update.

Jim A.
KB3TBX

P.S> I have also started a new thread on the DFS issues. Linksys made some choices when they went for certification. We'll see.

T-Troll wrote:

But other issue come to play with David's firmware - Luci lost wireless tab under "networks", as well as wireless stats in "Overview" is empty.
Is anyone have an idea how it happens? Any clue about the place to look for Luci logs? Interfaces working ok and visible in collectd, for example, but not in Luci web UI.

Which skin? this is all OK in Material, darkmatter looks OK too.

Right now I am just very disappointed in the WRT1200AC, especially wuth the 2.4G band

Get this - I plugged in an AR9271 USB and it even came up as a 802.11 Generic Controller, and STILL pulled better perfomance than the Marvell 88W8864 could muster up at VERY close range and from afar. Even with worse signal the AR9271 performance was BETTER than the 88W8864 which gave me better signal.

This is also the behavior for the 5G band as well, which only pulls N performance at moderate distances.

I need suggestions, is there ANY way I can improve the speed? Perhaps use another firmware? Please do NOT give me basic suggestions like "Make sure the channel is 1, 6 or 11" I've already debunked those because in the same conditions the AR9271 was performing better than the 88W8864, almost in the exact same spot. I need fixes because I am getting angry that I wasted money on a product that's inferior to my N150 device. The only thing I can do is get a replacement from Linksys, but I am scared that I'll get the same results, and I'll be very peeved by then.

Note: I also get the same results on the stock firmware so I've ruled that out.

I saw someone else that reported in this thread the same problem, I am with you sir, this thing is terrible. Maybe Linksys delivered on the old school looks, but not the speed and reliability, shame on you Linksys.

Sizeable Swiss wrote:

Right now I am just very disappointed in the WRT1200AC, especially wuth the 2.4G band

Get this - I plugged in an AR9271 USB and it even came up as a 802.11 Generic Controller, and STILL pulled better perfomance than the Marvell 88W8864 could muster up at VERY close range and from afar. Even with worse signal the AR9271 performance was BETTER than the 88W8864 which gave me better signal.

This is also the behavior for the 5G band as well, which only pulls N performance at moderate distances.

I need suggestions, is there ANY way I can improve the speed? Perhaps use another firmware? Please do NOT give me basic suggestions like "Make sure the channel is 1, 6 or 11" I've already debunked those because in the same conditions the AR9271 was performing better than the 88W8864, almost in the exact same spot. I need fixes because I am getting angry that I wasted money on a product that's inferior to my N150 device. The only thing I can do is get a replacement from Linksys, but I am scared that I'll get the same results, and I'll be very peeved by then.

Note: I also get the same results on the stock firmware so I've ruled that out.

I saw someone else that reported in this thread the same problem, I am with you sir, this thing is terrible. Maybe Linksys delivered on the old school looks, but not the speed and reliability, shame on you Linksys.

I'm curious to see what your signal to noise ratio is on 2.4 and 5ghz.  Have you ran the same throughput tests early in the morning to see if you're getting the same result? I know in my case, between 5pm and 10pm, the best I can do on the 3200acm/1900acs is around 50mbps/50mbps... However around midnight it goes above 100-200mbps up/down, and early in the morning 6am I can pull maximum bandwidth, which in my case is 300mbps/300mbps.

Through the day, the noise level continues to increase and peaks around 7pm to which I get the worst performance, even if I'm right next to the antennas.

Also, the Lede image includes iperf, so can you install iperf on your laptop/device and run some throughput tests on 2.4 and 5Ghz?

(Last edited by davidc502 on 21 May 2017, 23:00)

Sizeable Swiss wrote:

@starcms

Does your reghack method work for older devices to unlock channels 12-13 and extend power limits?

An example would be applying this hack on a WNR1000v2, which supports Openwrt and LEDE to this day. I would appreciate this greatly as the range on it is terrible, although not my main router it supplements the WRT1200AC's terrible 2.4ghz performance.

Highly unlikely, the reghack2 file I linked was compiled just for the WRT1200/1900ac/acs series of routers.  If you follow the thread I had linked on the very bottom of that post, it has the source code and could be compiled to work with other routers.

(Last edited by starcms on 21 May 2017, 23:18)

Sizeable Swiss wrote:

Right now I am just very disappointed in the WRT1200AC, especially wuth the 2.4G band

Get this - I plugged in an AR9271 USB and it even came up as a 802.11 Generic Controller, and STILL pulled better perfomance than the Marvell 88W8864 could muster up at VERY close range and from afar. Even with worse signal the AR9271 performance was BETTER than the 88W8864 which gave me better signal.

This is also the behavior for the 5G band as well, which only pulls N performance at moderate distances.

I need suggestions, is there ANY way I can improve the speed? Perhaps use another firmware? Please do NOT give me basic suggestions like "Make sure the channel is 1, 6 or 11" I've already debunked those because in the same conditions the AR9271 was performing better than the 88W8864, almost in the exact same spot. I need fixes because I am getting angry that I wasted money on a product that's inferior to my N150 device. The only thing I can do is get a replacement from Linksys, but I am scared that I'll get the same results, and I'll be very peeved by then.

Note: I also get the same results on the stock firmware so I've ruled that out.

I saw someone else that reported in this thread the same problem, I am with you sir, this thing is terrible. Maybe Linksys delivered on the old school looks, but not the speed and reliability, shame on you Linksys.

That's funny, because I also upgraded from a WNR1000 (N150) to my WRT1200AC and the performance isn't even comparable.  The 1200AC kills it on 2.4GHz (which I don't even use most of the time), and the WNR1000 didn't even do 5GHz on 802.11n nor 802.11ac at all.

Edit: nevermind, I had a WNR2000 (N300).  Everything else still applies though smile

Just curious, why are you so concerned with 2.4GHz performance anyway?  It's overcrowded and impossible to get a clear channel basically anywhere.  Plus, 2.4GHz doesn't support 802.11ac, and 802.11n devices run much faster and more reliably (due to lack of interference most likely) at 5GHz in my experience.

(Last edited by starcms on 21 May 2017, 23:32)

@david, it's a shame you just released a build.  Kernel has been upgraded to 4.9.29.

starcms wrote:

@david, it's a shame you just released a build.  Kernel has been upgraded to 4.9.29.

It's okay, stuff changes all the time. I'll get it this coming weekend.

(Last edited by davidc502 on 22 May 2017, 00:26)

Btw, these are my iperf3 results over 5GHz:

WRT1200AC router.  Laptop has 2x2:2 802.11ac WiFi chip.

Download (router to laptop):

Accepted connection from 192.168.1.40, port 50330
[  5] local 192.168.1.1 port 5201 connected to 192.168.1.40 port 50331
[ ID] Interval           Transfer     Bandwidth       Retr  Cwnd
[  5]   0.00-1.00   sec  51.5 MBytes   432 Mbits/sec    0   2.40 MBytes
[  5]   1.00-2.00   sec  47.5 MBytes   398 Mbits/sec    0   3.62 MBytes
[  5]   2.00-3.00   sec  71.2 MBytes   598 Mbits/sec    0   3.81 MBytes
[  5]   3.00-4.00   sec  71.2 MBytes   596 Mbits/sec    0   3.81 MBytes
[  5]   4.00-5.00   sec  71.2 MBytes   600 Mbits/sec    0   3.81 MBytes
[  5]   5.00-6.00   sec  70.0 MBytes   587 Mbits/sec    0   3.81 MBytes
[  5]   6.00-7.00   sec  68.8 MBytes   575 Mbits/sec    0   3.81 MBytes
[  5]   7.00-8.00   sec  67.5 MBytes   567 Mbits/sec    0   3.81 MBytes
[  5]   8.00-9.00   sec  70.0 MBytes   588 Mbits/sec    0   3.81 MBytes
[  5]   9.00-10.00  sec  63.8 MBytes   535 Mbits/sec    0   3.81 MBytes
[  5]  10.00-10.14  sec  8.75 MBytes   528 Mbits/sec    0   3.81 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth       Retr
[  5]   0.00-10.14  sec   661 MBytes   547 Mbits/sec    0             sender
[  5]   0.00-10.14  sec  0.00 Bytes  0.00 bits/sec                  receiver

Upload (laptop to router):

Accepted connection from 192.168.1.40, port 50336
[  5] local 192.168.1.1 port 5201 connected to 192.168.1.40 port 50337
[ ID] Interval           Transfer     Bandwidth
[  5]   0.00-1.00   sec  29.3 MBytes   246 Mbits/sec
[  5]   1.00-2.00   sec  31.5 MBytes   264 Mbits/sec
[  5]   2.00-3.00   sec  30.4 MBytes   255 Mbits/sec
[  5]   3.00-4.00   sec  31.3 MBytes   263 Mbits/sec
[  5]   4.00-5.00   sec  27.0 MBytes   227 Mbits/sec
[  5]   5.00-6.00   sec  13.5 MBytes   114 Mbits/sec
[  5]   6.00-7.00   sec  30.0 MBytes   251 Mbits/sec
[  5]   7.00-8.00   sec  31.0 MBytes   260 Mbits/sec
[  5]   8.00-9.00   sec  30.1 MBytes   252 Mbits/sec
[  5]   9.00-10.00  sec  31.8 MBytes   267 Mbits/sec
[  5]  10.00-10.04  sec  1.12 MBytes   262 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  5]   0.00-10.04  sec  0.00 Bytes  0.00 bits/sec                  sender
[  5]   0.00-10.04  sec   287 MBytes   240 Mbits/sec                  receiver

Obviously, theoretical maximum should be 866Mbits.  How do these speeds compare to everyone else?  Pretty decent?  Is it normal for upload to be less than half of download?  I personally had never run iperf3 before.  I also had 4 other devices connected to the 5GHz AP during the test (3 smartphones and 1 tablet).

Edit: And what does the CWND column represent?

Edit2: Test was performed at 7:15pm local time (central time).  I'll try later on and earlier in the morning to see if it increases any.

(Last edited by starcms on 22 May 2017, 01:42)

davidc502 wrote:
starcms wrote:

@david, it's a shame you just released a build.  Kernel has been upgraded to 4.9.29.

It's okay, stuff changes all the time. I'll get it this coming weekend.

Great to have the new build coming in yesterday/today. Did the sysupgrade, no issues on 1900ACS v2. In fact, thanks to all the help I got from you guys here, I was able to do the full sysupgrade in 10 minutes. Aiming to lower that time next time by scripting it smile

Ethernet (Desktop to router)

Accepted connection from 192.168.1.130, port 51923
[  5] local 192.168.1.1 port 5201 connected to 192.168.1.130 port 51924
[ ID] Interval           Transfer     Bandwidth
[  5]   0.00-1.00   sec  90.0 MBytes   755 Mbits/sec
[  5]   1.00-2.00   sec   112 MBytes   941 Mbits/sec
[  5]   2.00-3.00   sec   112 MBytes   942 Mbits/sec
[  5]   3.00-4.00   sec   112 MBytes   942 Mbits/sec
[  5]   4.00-5.00   sec   112 MBytes   942 Mbits/sec
[  5]   5.00-6.00   sec   112 MBytes   936 Mbits/sec
[  5]   6.00-7.00   sec   112 MBytes   942 Mbits/sec
[  5]   7.00-8.00   sec   112 MBytes   942 Mbits/sec
[  5]   8.00-9.00   sec   112 MBytes   942 Mbits/sec
[  5]   9.00-10.00  sec   112 MBytes   942 Mbits/sec
[  5]  10.00-10.21  sec  23.5 MBytes   941 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  5]   0.00-10.21  sec  0.00 Bytes  0.00 bits/sec                  sender
[  5]   0.00-10.21  sec  1.10 GBytes   923 Mbits/sec                  receiver

Wireless AC - -65-71db

Card: Qualcomm Atheros QCA61x4 Wireless Network Adapter

Connecting to host 192.168.1.1, port 5201
[  4] local 192.168.1.84 port 51957 connected to 192.168.1.1 port 5201
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-1.02   sec  2.00 MBytes  16.5 Mbits/sec
[  4]   1.02-2.00   sec  5.00 MBytes  42.5 Mbits/sec
[  4]   2.00-3.01   sec  4.38 MBytes  36.2 Mbits/sec
[  4]   3.01-4.01   sec  2.75 MBytes  23.1 Mbits/sec
[  4]   4.01-5.01   sec  2.75 MBytes  23.1 Mbits/sec
[  4]   5.01-6.01   sec  2.88 MBytes  24.2 Mbits/sec
[  4]   6.01-7.01   sec  2.62 MBytes  22.0 Mbits/sec
[  4]   7.01-8.00   sec  3.25 MBytes  27.6 Mbits/sec
[  4]   8.00-9.00   sec   256 KBytes  2.09 Mbits/sec
[  4]   9.00-10.00  sec  2.62 MBytes  22.1 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-10.00  sec  28.5 MBytes  23.9 Mbits/sec                  sender
[  4]   0.00-10.00  sec  28.3 MBytes  23.8 Mbits/sec                  receiver

Wireless N - -55 -60db

Card: Qualcomm Atheros QCA61x4 Wireless Network Adapter

Connecting to host 192.168.1.1, port 5201
[  4] local 192.168.1.84 port 51980 connected to 192.168.1.1 port 5201
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-1.00   sec   256 KBytes  2.10 Mbits/sec
[  4]   1.00-2.00   sec  0.00 Bytes  0.00 bits/sec
[  4]   2.00-3.01   sec  0.00 Bytes  0.00 bits/sec
[  4]   3.01-4.00   sec  0.00 Bytes  0.00 bits/sec
[  4]   4.00-5.00   sec  0.00 Bytes  0.00 bits/sec
[  4]   5.00-6.00   sec  0.00 Bytes  0.00 bits/sec
[  4]   6.00-7.00   sec  0.00 Bytes  0.00 bits/sec
[  4]   7.00-8.00   sec  0.00 Bytes  0.00 bits/sec
[  4]   8.00-9.00   sec  0.00 Bytes  0.00 bits/sec
[  4]   9.00-10.00  sec  0.00 Bytes  0.00 bits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-10.00  sec   256 KBytes   210 Kbits/sec                  sender
[  4]   0.00-10.00  sec  4.28 KBytes  3.50 Kbits/sec                  receiver


@StarCMS, now do you see why I'm so worried? These speeds are insanely anemic, even for the mediocre signal.
I haven't tested next to the router.

Noise is usually around -85dB to -90dB, but just according to LEDE, if there is any other way to find this out let me know.

Sizeable Swiss wrote:


@StarCMS, now do you see why I'm so worried? These speeds are insanely anemic, even for the mediocre signal.
I haven't tested next to the router.

Noise is usuually around -85dB to -90dB, but just according to LEDE, if there is any other way to find this out let me know.

Ok, this is perfect for troubleshooting because we have the same router (WRT1200AC) and my laptop has the same Qualcomm Atheros QCA61x4 Wireless adapter as you.  I assume you are on @david's firmware, or you wouldn't be in this thread.  So everything is the same.

Lets focus on 5GHz for now.  My signal is approx -60 to -65dBm, noise is -88dBm.  So even that is almost identical to yours.

In LEDE (when you go to 192.168.1.1 and login), and go to network --> wireless and click on radio0 (the 5GHz radio), do you have mode set to AC and width set to 80MHz?  Set transmit power at 23dBm (should be the highest it will go without the hack).  Make sure channel is NOT set to auto.  Set it to 157.

Below that, click on Wireless Security.  make sure encryption is set to WPA2-PSK and cipher is set to Force CCMP (AES).  This is extremely important for speed.

Lastly, make sure the drivers for your QCA61x4 on your laptop are updated to the very latest available.

Then our settings and hardware will be identical.  If that doesn't help, I don't know...

Edit: Also make sure any power saving features on your laptop for the wifi adapter are turned off.  Check both in device manager and under Power Options --> Change Plan Setting --> Change Advanced Power Settings.  (I'm assuming you are running Windows).

(Last edited by starcms on 22 May 2017, 03:52)

I have actually had all of those, settings and all. Except that I'm connected to my
other network as a client, but I still get the same results if I'm not. Same channel too (or close to it) maybe I really have gotten a dud. But mines is v2 and yours is v1. How does your 1200ac perform?

The only thing I can think of is the placement of my router which is VERY congested on the 2.4 band (but 5g is still meh?) and my N150 was in the same spot so I dropped that theory.

And I actually am on official LEDE on an older build. I don't know what the problem is but all of the newer builds seem to have this problem where the requests process very slowly (YouTube takes longer to load and such) and it's VERY noticeable. Ugh I really need to get to a computer. Too bad there's no mobile optimized version for this site.

So either I have a dud or the v2 is just rubbish (only difference between setups) and I'm leaning towards the first one, but I saw someone post on here about how the speeds were bad for his v2, so maybe the v2's drivers are the problem. I wish I had a v1, those sound great sad

Maybe I'm on to nothing, but maybe it's the drivers and nobody has touched it since it's release. Only one update to the driver for official Linksys firmware which may only include a minor change.

The grand part of this all is how it can connect to my network as a client and get PERFECT connection, full speed! This is a step.

hmm. @starcms stated config on a WRT3200ACM, latest r4164 LEDE, a Linksys USB AC adapter on a laptop as client.

wireless connection shows 433 Mbs link, I get 67.6 Mbit/s with iperf3 server on WRT. Only one CPU gets loaded.

Sizeable Swiss wrote:

So either I have a dud or the v2 is just rubbish (only difference between setups) and I'm leaning towards the first one, but I saw someone post on here about how the speeds were bad for his v2, so maybe the v2's drivers are the problem. I wish I had a v1, those sound great sad

Maybe I'm on to nothing, but maybe it's the drivers and nobody has touched it since it's release. Only one update to the driver for official Linksys firmware which may only include a minor change.

The grand part of this all is how it can connect to my network as a client and get PERFECT connection, full speed! This is a step.

1200 V1 and V2 use the same drivers.  They are identical in every way except the V2 has the power tables written to EEPROM so that they cannot be modified (to make the FCC happy and pass their new requirements).  This would have nothing to do with your issue.  And if you are talking about the firmware updates available from Linksys's website -- that isn't drivers, that is firmware.  If you are running LEDE, then you aren't using that anyway.

But a) why are you running an older LEDE build which comes with older wifi drivers and b) this thread is for questions about people using @david's builds.  If you are running official LEDE, there are forums for that...and that could also be the cause of your issue...

(Last edited by starcms on 22 May 2017, 06:12)

jalles wrote:

hmm. @starcms stated config on a WRT3200ACM, latest r4164 LEDE, a Linksys USB AC adapter on a laptop as client.

wireless connection shows 433 Mbs link, I get 67.6 Mbit/s with iperf3 server on WRT. Only one CPU gets loaded.

I believe the performance of the 3200ACM is still slower than the rest, due to the mwlwifi driver still being worked on and the 3200ACM having the 88W8964 wifi chips instead of the 88W8864 like all the other WRT models.

Last I heard, I believe it is now at least stable and doesn't eat up memory, but is still slower than it should be.  I'm sure @david or someone with a 3200 will chime in.