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The content of this topic has been archived between 26 Feb 2018 and 7 May 2018.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.

Hi @David, long time no talk!

I'm facing an issue with the latest update you published on 7/29 . I did the usual script-based "David compatible" upgrade, and 2 things failed:

A) f2fs is not installing . Error is weird... ( remember that I change the repos to the LEDE ones after installing your image ):

/code
root@RouterACS:/mnt# opkg install kmod-fs-f2fs
Installing kmod-fs-f2fs (4.9.30-1) to root...
Downloading https://davidc502sis.dynamic-dns.net/sn … _vfpv3.ipk
Collected errors:
* satisfy_dependencies_for: Cannot satisfy the following dependencies for kmod-fs-f2fs:
*      kernel (= 4.9.30-1-261bc28a665dffab13fceedc7e0a9e20) *  kernel (= 4.9.30-1-261bc28a665dffab13fceedc7e0a9e20) *
* opkg_install_cmd: Cannot install package kmod-fs-f2fs.
/code

B) CIFS fails to mount a remote drive that I always was able to map...

/code

mount.cifs //192.168.1.102/Disk /mnt/NAS -o unc=\\\\IPHERE\\Disk,ip=192.168.1.102,user=usuaerio,pass="password",dom=workgroup
mount error: cifs filesystem not supported by the system
mount error(19): No such device
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

/code

Both kmod-fs-cifs and cifsmount are installed... Any ideas?

Okay, the modified wifi driver for kernel 4.4 is ready to be tested.

Requesting v1 owners do some testing before and after installing the driver below, and reporting back the results.

https://davidc502sis.dynamic-dns.net/re … _vfpv3.ipk

Hi @david, I upgraded to the last version, and tried to configure openVPN, but there is no build-ca/build-key-server files. All the packages (openvpn-easy-rsa, openvpn-openssl, luci-app-openvpn) are installed. I tried to force reinstall, with no success, I'm still not able to find build-ca.

Any ideas?

Regards,

Thank you for your help, rebooting after flashing the new firmware fixed all of my issues.

Edit
After about 30 min, radio1 is reporting slow speeds and pings are strange. Every other ping is very high

Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=242ms TTL=44
Reply from 8.8.8.8: bytes=32 time=22ms TTL=44
Reply from 8.8.8.8: bytes=32 time=219ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=200ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=189ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=178ms TTL=44

(Last edited by NewImage on 1 Aug 2017, 03:12)

famg wrote:

Hi @david, I upgraded to the last version, and tried to configure openVPN, but there is no build-ca/build-key-server files. All the packages (openvpn-easy-rsa, openvpn-openssl, luci-app-openvpn) are installed. I tried to force reinstall, with no success, I'm still not able to find build-ca.

Any ideas?

Regards,

None of the packages are are available after installing  openvpn-openssl openvpn-easy-rsa.

The following appear to be missing.

build-ca
build-dh
build-key-server my-server
build-key-pkcs12 my-client

I did see a commit to upgrade Openssl, so I have to wonder if this is the reason >   https://github.com/lede-project/source/ … 988dc2be6a

You might check with the package maintainer to see if this is now a known issue or not.

mariano.silva wrote:

Hi @David, long time no talk!

I'm facing an issue with the latest update you published on 7/29 . I did the usual script-based "David compatible" upgrade, and 2 things failed:

A) f2fs is not installing . Error is weird... ( remember that I change the repos to the LEDE ones after installing your image ):

/code
root@RouterACS:/mnt# opkg install kmod-fs-f2fs
Installing kmod-fs-f2fs (4.9.30-1) to root...
Downloading https://davidc502sis.dynamic-dns.net/sn … _vfpv3.ipk
Collected errors:
* satisfy_dependencies_for: Cannot satisfy the following dependencies for kmod-fs-f2fs:
*      kernel (= 4.9.30-1-261bc28a665dffab13fceedc7e0a9e20) *  kernel (= 4.9.30-1-261bc28a665dffab13fceedc7e0a9e20) *
* opkg_install_cmd: Cannot install package kmod-fs-f2fs.
/code

B) CIFS fails to mount a remote drive that I always was able to map...

/code

mount.cifs //192.168.1.102/Disk /mnt/NAS -o unc=\\\\IPHERE\\Disk,ip=192.168.1.102,user=usuaerio,pass="password",dom=workgroup
mount error: cifs filesystem not supported by the system
mount error(19): No such device
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

/code

Both kmod-fs-cifs and cifsmount are installed... Any ideas?

A) for any packages beginning with kmod, you must download them from @david's repo.

I leave my repos configured as follows:

src/gz reboot_core https://davidc502sis.dynamic-dns.net/sn … c/packages
src/gz reboot_base https://downloads.lede-project.org/snap … vfpv3/base
src/gz reboot_luci https://downloads.lede-project.org/snap … vfpv3/luci
src/gz reboot_packages https://downloads.lede-project.org/snap … 3/packages
src/gz reboot_routing https://downloads.lede-project.org/snap … v3/routing
src/gz reboot_telephony https://downloads.lede-project.org/snap … /telephony
src/gz reboot_darkmatter https://davidc502sis.dynamic-dns.net/sn … darkmatter

everything from reboot_core is kernel and build specific.  and there is no darkmatter official repo.  all the rest are safe to set to the lede snapshot repos and upgrade from (except for the odd package every 2 or 3 months that breaks something, and I try and make a post warning everyone when that occurs cause I'm usually one of the first to upgrade my packages smile ).

Edit: I found your problem.  You said you were running the latest build but

Installing kmod-fs-f2fs (4.9.30-1) to root...
Downloading https://davidc502sis.dynamic-dns.net/snapshots/r4323/targets/mvebu/generic/packages/kmod-fs-f2fs_4.9.30-1_arm_cortex-a9_vfpv3.ipk

that link has r4323 in it instead of r4651.  If you are going to change repos, be sure to keep the ones that point to @david's server updated with the correct build number.

(Last edited by starcms on 1 Aug 2017, 04:45)

@david,

I believe you have a 3200ACM.  Did you do the ping test to see if the amsdu problem is present on it?

Because the 3200ACM and any model running kernel 4.4 I am unsure if the issue exists (different reasons why).

All models (except possibly the 3200acm) the problem definitely exists on kernel 4.9.

(Last edited by starcms on 1 Aug 2017, 04:17)

davidc502 wrote:

@starcms

Since it's making a big difference, I had planned on modifying the code until the driver is fixed.

Thanks so much @david!  And thanks for all the time and effort you spend making this build the best it can be!!!

Edit: Would you consider adding the following packages in future builds?

a) joe (text editor, very tiny)
b) ethtool (used to issue command to make SQM work better; for those interested edit /etc/init.d/sqm and add "ethtool -K eth1 tso off gso off gro off" to both the start and boot functions
c) Include the luci-darkmatter theme in addition to the material theme.

(Last edited by starcms on 1 Aug 2017, 04:40)

NewImage wrote:

Thank you for your help, rebooting after flashing the new firmware fixed all of my issues.

Edit
After about 30 min, radio1 is reporting slow speeds and pings are strange. Every other ping is very high

Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=242ms TTL=44
Reply from 8.8.8.8: bytes=32 time=22ms TTL=44
Reply from 8.8.8.8: bytes=32 time=219ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=200ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=189ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=178ms TTL=44

Did you apply the updated driver that @david posted to fix the amsdu issue (not that that should cause this anyway)
What model are you on, and what kernel version are you running?  Did you try doing a clean flash (not saving settings) and see if it still exists?

starcms wrote:
NewImage wrote:

Thank you for your help, rebooting after flashing the new firmware fixed all of my issues.

Edit
After about 30 min, radio1 is reporting slow speeds and pings are strange. Every other ping is very high

Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=242ms TTL=44
Reply from 8.8.8.8: bytes=32 time=22ms TTL=44
Reply from 8.8.8.8: bytes=32 time=219ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=200ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=189ms TTL=44
Reply from 8.8.8.8: bytes=32 time=23ms TTL=44
Reply from 8.8.8.8: bytes=32 time=178ms TTL=44

Did you apply the updated driver that @david posted to fix the amsdu issue (not that that should cause this anyway)
What model are you on, and what kernel version are you running?  Did you try doing a clean flash (not saving settings) and see if it still exists?

Thank you again
I am on a 3200, running the latests firmware build from 7-30 (Lede Reboot SNAPSHOT r4651-a6f6f8d)
There is not an option that I see to save settings in the factory firmware flash proccess.

I am still seeing the same ping issues, and is precistant even on the router itself.

Is the amsdu issue on the 3200?

root@LEDE:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=45 time=63.116 ms
64 bytes from 8.8.8.8: seq=1 ttl=45 time=302.882 ms
64 bytes from 8.8.8.8: seq=2 ttl=45 time=61.981 ms
64 bytes from 8.8.8.8: seq=3 ttl=45 time=302.677 ms

Rebooting router and trying ping again I get
root@LEDE:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=45 time=20.097 ms
64 bytes from 8.8.8.8: seq=1 ttl=45 time=20.512 ms
64 bytes from 8.8.8.8: seq=2 ttl=45 time=19.914 ms
64 bytes from 8.8.8.8: seq=3 ttl=45 time=19.843 ms

(Last edited by NewImage on 1 Aug 2017, 13:03)

starcms wrote:

Hi @David, long time no talk!

I'm facing an issue with the latest update you published on 7/29 . I did the usual script-based "David compatible" upgrade, and 2 things failed:

A) f2fs is not installing . Error is weird... ( remember that I change the repos to the LEDE ones after installing your image ):


Edit: I found your problem.  You said you were running the latest build but


Installing kmod-fs-f2fs (4.9.30-1) to root...
Downloading https://davidc502sis.dynamic-dns.net/sn … _vfpv3.ipk

that link has r4323 in it instead of r4651.  If you are going to change repos, be sure to keep the ones that point to @david's server updated with the correct build number.

You rule! As always smile

Fixed it. THAT is the only issue I've run ( and that happened before ... I just didn't catch that this time... that r???? needs to be adjusted after every upgrade. I'll make sure I'm not as sleepy as I was last night when doing the upgrade wink

starcms wrote:

@david,

I believe you have a 3200ACM.  Did you do the ping test to see if the amsdu problem is present on it?

Because the 3200ACM and any model running kernel 4.4 I am unsure if the issue exists (different reasons why).

All models (except possibly the 3200acm) the problem definitely exists on kernel 4.9.

About DarkMatter --   I'm a bit concerned why this never was accepted into trunk. The source of the code comes from https://apollo.open-resource.org/, which I can't tell what [hat] they are... White/Grey or Black, but advertise themselves as hackerspace...   

Maybe you know more about this hackerspace group?

@David

Thanks for the new wifi driver for 1900 users i'm able to game again!

Toetje583 wrote:

@David

Thanks for the new wifi driver for 1900 users i'm able to game again!

Also thank starcms as he pushed a bit harder to get me to do it smile   

cheers,

NewImage wrote:

Thank you again
I am on a 3200, running the latests firmware build from 7-30 (Lede Reboot SNAPSHOT r4651-a6f6f8d)
There is not an option that I see to save settings in the factory firmware flash proccess.

I am still seeing the same ping issues, and is precistant even on the router itself.

Is the amsdu issue on the 3200?

root@LEDE:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=45 time=63.116 ms
64 bytes from 8.8.8.8: seq=1 ttl=45 time=302.882 ms
64 bytes from 8.8.8.8: seq=2 ttl=45 time=61.981 ms
64 bytes from 8.8.8.8: seq=3 ttl=45 time=302.677 ms

Rebooting router and trying ping again I get
root@LEDE:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=45 time=20.097 ms
64 bytes from 8.8.8.8: seq=1 ttl=45 time=20.512 ms
64 bytes from 8.8.8.8: seq=2 ttl=45 time=19.914 ms
64 bytes from 8.8.8.8: seq=3 ttl=45 time=19.843 ms

The amsdu only affects the ping times from wireless clients.
Pinging directly from the router, the wireless driver is not used.
The high ping times could be caused by your internet connection, provider or general internet congestion.

davidc502 wrote:
starcms wrote:

@david,

I believe you have a 3200ACM.  Did you do the ping test to see if the amsdu problem is present on it?

Because the 3200ACM and any model running kernel 4.4 I am unsure if the issue exists (different reasons why).

All models (except possibly the 3200acm) the problem definitely exists on kernel 4.9.

About DarkMatter --   I'm a bit concerned why this never was accepted into trunk. The source of the code comes from https://apollo.open-resource.org/, which I can't tell what [hat] they are... White/Grey or Black, but advertise themselves as hackerspace...   

Maybe you know more about this hackerspace group?

No idea, but on their page about darkmatter, I definitely remember them saying it was a beta, and I think I remember them saying it was something they had created in their free time.  My guess is they simply never submitted it officially.

NewImage wrote:

Thank you again
I am on a 3200, running the latests firmware build from 7-30 (Lede Reboot SNAPSHOT r4651-a6f6f8d)
There is not an option that I see to save settings in the factory firmware flash proccess.

I am still seeing the same ping issues, and is precistant even on the router itself.

Is the amsdu issue on the 3200?

root@LEDE:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=45 time=63.116 ms
64 bytes from 8.8.8.8: seq=1 ttl=45 time=302.882 ms
64 bytes from 8.8.8.8: seq=2 ttl=45 time=61.981 ms
64 bytes from 8.8.8.8: seq=3 ttl=45 time=302.677 ms

Rebooting router and trying ping again I get
root@LEDE:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=45 time=20.097 ms
64 bytes from 8.8.8.8: seq=1 ttl=45 time=20.512 ms
64 bytes from 8.8.8.8: seq=2 ttl=45 time=19.914 ms
64 bytes from 8.8.8.8: seq=3 ttl=45 time=19.843 ms

Really no explanation that I can see.  I don't know if the amsdu problem affects the 3200, but those aren't the symptoms of it anyway.  Try installing @david's build fresh from the stock Linksys firmware and see if that helps.

edit: oh wait, you said you already tried that.  I don't know.  ISP is the only thing I could think of.

(Last edited by starcms on 1 Aug 2017, 18:16)

davidc502 wrote:
Toetje583 wrote:

@David

Thanks for the new wifi driver for 1900 users i'm able to game again!

Also thank starcms as he pushed a bit harder to get me to do it smile   

cheers,

wink big_smile

Btw, @david, you do have a 3200acm correct?  Does it have the amsdu problem?  Still waiting to hear back from anyone on a mamba if they experience it on kernel 4.4.

We know for sure it affects all 1200/1900 models on kernel 4.9.  It doesn't really matter if the 3200 suffers from it, since the modified kernel 4.9 driver with amsdu disabled will fix it if it does, just curious, because I read that the 3200 handles amsdu differently.

(Last edited by starcms on 1 Aug 2017, 18:34)

starcms wrote:
davidc502 wrote:
Toetje583 wrote:

@David

Thanks for the new wifi driver for 1900 users i'm able to game again!

Also thank starcms as he pushed a bit harder to get me to do it smile   

cheers,

wink big_smile

Btw, @david, you do have a 3200acm correct?  Does it have the amsdu problem?  Still waiting to hear back from anyone on a mamba if they experience it on kernel 4.4.

We know for sure it affects all 1200/1900 models on kernel 4.9.


Hi Starcms,

I will load my WRT1900ACv1 tonight with the 4.4 build and check the ping.  Once I have some results, I will report back

jefftk99 wrote:

Hi Starcms,

I will load my WRT1900ACv1 tonight with the 4.4 build and check the ping.  Once I have some results, I will report back

Big thanks in advance!  This will let @david know if the mwlwifi driver for kernel 4.4 also needs the fix or not.  Before updating to the modified driver with with amsdu disabled, simply start continuously pinging your router (192.168.1.1 unless you changed it), get a few pings (should be approx 1 to 3 ms), then simply load some webpages or load a youtube video and see if the pings jump to approx 30ms or higher.  If they jump, then it does have the amsdu problem and the modified driver @david compiled will fix it.  If pings stay constant at 1 -3 ms regardless of other activity, then it's not required.  It only affects wireless clients so be sure to do it from a laptop or other wireless device.

(Last edited by starcms on 1 Aug 2017, 18:30)

I today tried openwrt firmware and there the networkplayer is spotted right away.
So with Linksys orig and openwrt it works but not with LEDE at least on my 3200acm.

starcms wrote:

@david,

just curious, did you try the simple test with the un-modified driver and then test it with the modified driver you just put up and see the difference?  It may only affect 1200/1900 models because the 3200 implements amsdu differently.

To all

In case you haven't been following the thread, and especially if you have complained or noticed laggyness or bad lag or ping, especially when playing multiplayer games when using wifi (2.4GHz and 5.0GHz), which is when the problem is most noticeable (see posts on page 111 towards the top for a recent complaints), please read the following.

The problem is that if your router (both 2.4 and 5GHz), when doing basically any broadband significant activity, such as watching youtube, downloading a file, or simply browsing a webpage, activates something know as amsdu.  I believe it is some sort of a caching mechanism.  However it doesn't work properly and causes an increase of ping of over 30ms and packet loss when doing anything over wireless.  Just FYI, this is not related to buffer-bloat in any way at all.

A good test is this:  Start pinging your router continuous from your PC, observe the pings (which should be normal, 1 - 3 ms, no packet loss) for a few seconds, then simply refresh a webpage, load a youtube video, do basically anything that uses bandwidth.  You'll see your pings to 192.168.1.1 (the router) jump to 30+ ms when actually using it, and that there is occasional packet loss as well.  Stop all broadband activity and the pings will go back to normal.  Pings don't activate amsdu which is why we can do this test which is extremely quick and easy.

So the consequence is this: the ping when playing a multiplayer game (for example) instead of just being X ms to the gaming server is actually X + Y ms where Y is the usually negligible 1ms ping to the router.  But because of the bug in amsdu, Y (the wireless connection between your computer and the router) becomes approx. 30 ms.  Assuming X is 40ms, then instead of the ping to your game server being 41 ms (which is indistinguishable from 40 ms), it is now 70 ms!  And remember that packet loss is also introduced causing more lag.

In most other uses (except for VoIP and live video streaming services and other similar software), the increased ping is not noticed because if the ping to a website is 100ms, with the amsdu problem, it is now approx 130ms, but not a large enough increase to notice.  Now take VoIP into account, and the increased ping combined with packet losses can make the connection horrible.

So the solution is clear, until amsdu is properly fixed (which won't happen for a long, long time would be my guess, because I think this problem only affects the 1200/1900 models, and all work going into the driver currently is for the 3200), the best solution is simply to disable it, which is what the driver @david just uploaded does.  There are no ill-effects or issues introduced with it being disabled (such as loss of speed, difficulty in connecting, etc)

So please, do yourself a favor and install the updated kmod-mwlwifi .ipk driver file that @david posted and test it out with amsdu disabled.  You can easily use the ping test I mentioned above to clearly see the difference.

amazing detailed information man ! this david's thread brings amazing people !
thank you ! and david for making all this happen !
btw: new build works great!

Im on Linksys WRT1900acV1 "Mamba" 4.4 branch r4651
Im pinging an AppleTV4 connected wirelessly to my 5ghz network from my PC.
Did this test at work running a teamviewer session to my home pc, I could be doing the testing wrong.

Here is ping before patch.

Pinging 192.168.1.198 with 32 bytes of data:
Reply from 192.168.1.198: bytes=32 time=12ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=4ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=14ms TTL=64
Reply from 192.168.1.198: bytes=32 time=7ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=4ms TTL=64
Reply from 192.168.1.198: bytes=32 time=7ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=7ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=7ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64

After the patch with a reboot to router

Pinging 192.168.1.198 with 32 bytes of data:
Reply from 192.168.1.198: bytes=32 time=13ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=7ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=8ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=6ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64
Reply from 192.168.1.198: bytes=32 time=4ms TTL=64
Reply from 192.168.1.198: bytes=32 time=4ms TTL=64
Reply from 192.168.1.198: bytes=32 time=5ms TTL=64

(Last edited by Hazado on 1 Aug 2017, 20:48)

starcms wrote:
davidc502 wrote:
Toetje583 wrote:

@David

Thanks for the new wifi driver for 1900 users i'm able to game again!

Also thank starcms as he pushed a bit harder to get me to do it smile   

cheers,

wink big_smile

Btw, @david, you do have a 3200acm correct?  Does it have the amsdu problem?  Still waiting to hear back from anyone on a mamba if they experience it on kernel 4.4.

We know for sure it affects all 1200/1900 models on kernel 4.9.  It doesn't really matter if the 3200 suffers from it, since the modified kernel 4.9 driver with amsdu disabled will fix it if it does, just curious, because I read that the 3200 handles amsdu differently.

I have a new laptop to test it out on tonight.

Yes, the primary router is the 3200acm.

(Last edited by davidc502 on 1 Aug 2017, 21:26)

I tried both old and modified wifi driver, and they behaved similarly.

Continuous ping to 192.168.2.1 and pings remained between 2 and 6ms during speed tests 300/300mbps.

The change doesn't appear to affect the 3200 either way.

Sorry, posts 2876 to 2875 are missing from our archive.