I only got a WRT3200ACM a week ago despite talk of poor WiFi and it's still returnable but it appears to be worth the risk given there's not much else in its price bracket to match the stats. I'm getting 420mbs over 5ghz wireless on a 400mbs PPPoE to the ISP.
Previous driver had some stability issues not found in Linksys stock so got high hopes.
I now want to build this into a new image but never done it. I'm going hunting for answers but any pointers welcome.
And I use your driver in combo with a unregulatory wireless-regdb (no dfs, 1000mw and unlocked channels).... And on a macbook pro 2020...I get 890mbps and 870 up from my fiber isp with a ping of 3.7-4.2ms tested with a distance of 30 meters....and still over 500 up/down, at my neighbor two floors down...
I told my friends and now I'm installing wrt1900acs and wrt3200 for a lot of openwrt and wifi speedfreaks, from airbnb locations to a owner of expensive super apartment...he has three 3200...with radius and vpn tunnel and multiple fiber wan. He has 2gbit bandwidth!
Now running 4 weeks without a glitch!
I'm on 21 specialized build with your mwlwifi, low latency kernel through hardware devcrypto and openssl libs and range over 200 meters from my house, urban environment!
I share a gigabit fiber with 2 neighbors, each with wrt1900acs for own wifi, each with adblock, dnscrypt-proxy2, adblock, IPv6, multiple ssid and VLAN isolated networks...running for 2 two weeks straight!
Unfortunately, it does not look like NXP plans to play nice.
Looks like they do not even want to answer this publicly but here it goes anyway:
From: NXP Tech Support email@example.com
Subject: NXP support case: #00378643: “[Community] Plan to Update Public Wi-Fi Firmware Available on Kernel.org”: Project “”
Date: Fri, 09 Jul 2021 08:38:08 +0000
Do you recall if you ever offered an ipk sans commit, and if so did it resolve issue.
There seems to be a a lot of people stating the same regarding intermittent wifi on the venom and rango with current master or 21.xRCs, but things are fine with 19.x or old community build (wo said commit). That would seem to be a likely candidate to point ones finger at, especially given its method of arrival in the mwlwifi repo.
Edit: But I suppose it would be easy enough for those afflicted to just replace the FW on the live system they are currently running to test theory.
It is the same as it has always been for this device, now it feels more like a new generation cycle or something that have found the original fault. If you don’t believe me you can search this forum for “wrt3200acm wifi problems”.
I have rango (actually two) and wifi didn’t worked (especially 5GHz in any form or variant) from factory, it didn’t work on 18.06 and it didn’t work on 19.07.
I haven’t tested the wifi on 21.02 but I see no reason why it would work there since the original problem is still there.
It was hopeless to specify country code and that setting messed up ever single radio and DTS never finished. Every time I came home the smartphone never connected to the WiFi.
The only thing that acually worked was 2,4GHz.
Then I bought a TP-Link EAP access point and I haven’t had a single wifi problem since.
Not really, plus you are conflating things. I never had an issue with WLAN cutting out on the rango back when I used the radios on the device. I am not talking about cessation of development, CC, WPA3 or any of the myriad of issues.
There are quite a number of voices saying they have no cutout issue with 19.x, but experience issue with 21.x.RC, so backup to version 19.x where things work. At any rate, for those that have the issue on 21.x.RC / master and might want to try the previous FW, you could try:
mv 88W8964.bin 88W8964.bin.ORG
#### 88W8964 firmware 126.96.36.199 -- 11896425a9fe6d576b52864d8d0945656db3d551
wget https://github.com/kaloz/mwlwifi/raw/11896425a/bin/firmware/88W8964.bin -O 88W8964.bin.TST
ln -s 88W8964.bin.TST 88W8964.bin
## for back to original
## rm 88W8964.bin && ln -s 88W8964.bin.ORG 88W8964.bin
to see if it changes anything.
On setting the CC, there was a patch (I still have ti here), plus CC with the third radio was a well known issue, and resolvable if dealt with correctly, at least it WFM on a NA(CA) unit.
Looking back at the original issue 378 on the mwlwifi github when the OEM BLOB was being extracted and tested, eventually there are people reporting the intermittent wifi issue and backing out to the, then current, previous BLOB. Appears to be possibly client related.
So choice appears to be the more performant BLOB if no issue, otherwise go back to previous. Seems simple enough for those with the issue to use something like the above; problem is awareness I guess.
To offer an easy way to test things I have put up an image that has both BLOBs included, with the link upon installation pointing at the previous 188.8.131.52 BLOB, you can change the link to the current 184.108.40.206 BLOB to test things.
Just the image as I run things here, master with today's 5.10.51 kernel push, r17161-853e8465a7.51051BLOB in my drop.
I pushed a PR to openwrt-21.02 trunk to downgrade the firmware to 220.127.116.11. Affected devices are Linksys WRT3200ACM & WRT32X. It'd be really helpful to get the PR merged if more people can confirm 18.104.22.168 solves the Wi-Fi cut-out issue.
I’m not sure. This was intended to be a hotfix to be shipped with 21.02 stable release until actual fixes on the trunk are backported to the 21.02 branch. I think it will be better to just backport those instead of downgrading the firmware. I’m working on submitting a patch for it.
Some people seems to be having issues after upgrading to 21.02.0, so I decided to grab the older drivers from 19.07, and make packages to install them on 21.02.0-rc4; this is obviously not a long term solution, but I hope it might be useful to debug the issue.