I will take care of fixing this upstream (and backporting the fix to OpenWrt)
I sent a fix upstream and pushed it to my staging tree at https://git.openwrt.org/?p=openwrt/staging/nbd.git;a=summary
Please run a quick test to make sure it fixes the issue
9 Likes
@cotequeiroz Would you be able to create a mac80211 ipk for 21.02.1 based on the upstreamed patch from Felix?
I would like to test it thoroughly on 21.02.1 but simply don’t have a build environment. Thank you.
1 Like
Shouldn't the package get a PKG_RELEASE bump?
I've bumped it when building the 21.02.1 package for testing, to avoid the update problem @WildByDesign reported earlier.
This time, the kmod-cfg80211 package has changed as well. The packages are saved here:
https://drive.google.com/drive/folders/1bF_N41aQ3w58dRLn9sE28954McDh7pC9?usp=sharing
1 Like
I will bump PKG_RELEASE when I backport this to the 21.02 branch
2 Likes
Just got a confirmation that the fix works. I've pushed it to OpenWrt master and 21.02
12 Likes
Thank you so much guys. Looking forward to trying this out on the next OpenWRT release.
3 Likes
do you know if this patch was accepted into stable linux branch? I wish to see it reviewed before we have it openwrt.
I personally don't have any insight into the Linux kernel, sorry.
1 Like
I sent my patch to linux-wireless already. However, from here on it takes quite a bit of time until it hits Linux stable trees. First it has to be picked by Johannes Berg for and added to the mac80211 tree for 5.16. From there it has to go into Linus' tree, and only then can it be picked up for inclusion into stable trees. That process often takes more than just a few days.
I tagged the patch with Cc: stable@..., so it will automatically be picked up once it hits mainline.
That said, I expect my patch to be accepted, since it's a simple and obvious fix for the regression, that's why I pushed it to OpenWrt master and 21.02 already.
10 Likes
@nbd thanks a lot for taking this to the linux-stable@. I didn't look into, though hope you did it right and we will get your patch upstreamed, thus credible openwrt changes. I have couple of 3200acm I'd like to run on a recent openwrt release. Sorry I couldn't test patches, though you guys made an incredible work to narrow down the issue. Looking forward to see fixed and working upcoming release. <3 Thanks!
An outstanding piece of troubleshooting.
Just wanted to say as a WRT3200ACM user I have been following the progress of this thread and dedication of the troubleshooting and fault finding undertaken to find the issue has been amazing.
Thanks to everyone who tested and helped find the offending commit, certainly looks promising for using 21.02 on the WRT3200ACM in the next release in a more stable fashion for wireless when it gets backported.
Amazing work.
2 Likes
Thanks to each and every one of you for pouring your time into troubleshooting and finally fixing the issue upstream. I added information on the main post with @nbd's replies.
Cheers!
2 Likes
Wow, that is really good news!
Just out of curiosity I extracted mac80211.ko from the ipk file and copied it to the /lib/modules/5.4.143/ folder, overwriting the existing mac80211.ko on my WRT3200ACM with Openwrt v21.02.0. After a restart Wifi was stable and any cutouts had gone.
I want to thank all involved in resolving this issue for their dedication and invested time and efforts. Thank you so much!!
Due to my lack of technical knowledge I did not understand all of the technical details discussed here. So, please, allow me one more question:
Retrospectively, was that issue caused by faulty Linksys firmware and this could be overcome by adapting the Linux kernel module? Or was the issue caused only by a regression flaw in the kernel module?
Thanks again to all
Alexander
1 Like
It was caused by a regression in upstream kernel module (mac80211), which regressions only affected some wifi drivers. nbd specified mwlwifi and ath9k in the commit message...
The patch fixes this upstream commit:
5 Likes
Latency seems to improved significantly on my WRT3200ACM. Previously, I would always get something like Unloaded 70 ms and around 500-700 ms Loaded latency on Fast speedtest.
After the fix, I just got Unloaded 14 ms and 78 ms Loaded latency. I didn't believe it at first until I ran it a few times.
Same settings as always, the only difference is the applied fix.
1 Like
The most interesting fact is that the regression affected a rather specific group of devices, most notably Apple devices, while others were unaffected. I wonder how that can be explained.
3 Likes
I was actually hoping that somebody could explain that as well. I've been very curious as to why Apple devices trigger this issue more than any other devices.
And within the family, never saw the issue with an IPad Air 2, but an IPad Air 4 suffers.