don't say that, a distraction can happen to everyone. you helped openwrt a lot with drivers.
@eduperez i cannot find the drivers for that kernel anymore, for 5.4 .66, if that is not a problem, can you upload that if you still have it?
firmware and driver for wrt 3200, i want test the thagabe version.
I took the packages down to avoid confusions, I will upload them again, no problem.
I have just restored the packages for the latest SNAPSHOT 
Packages for 19.07.4 will come shortly, this time for real, I promise... or maybe not 
I need to come up with a better naming scheme for the files, because the firmware packages are essentially the same for all versions of OpenWrt, but they have different dependencies, and must have different names. I open to ideas here 
Packages for 19.07.4 are also available now. Please, be careful when downloading the packages for your version of OpenWrt; it is very easy to confuse the names, I know.
it's working now. thanks.
how could i use this new driver on the current firmware compilation for wrt1900acs v2, I could change it from the command line?
thanks eduperez, i insatalled that and are working good, really i cannont see any different with driver in master, but anyway alot people complain about mwlwifi, i know are not perfect but i never had all this problem. also i'm using the device, (wrt3200) as a true repeter, like the official firmware, and is working very very nice, something that on my ath10k devices i'm not able to do in a stable way
Thanks! The 19.07.4 packages are working fine on my WRT3200ACM.
I managed to update the driver/firmware for my WRT3200ACM on 19.07.4 Stable. Seems to be working fine.
What exactly is the improvement between the old 2019-03-02 files and your 2020-07-29 files?
don't be so hard on you! your work is flawless!
Hi, anyone has a patch for the latest drivers (10.3.8.0-thagabe) for the snapshot?
From what I can tell people with WRT1900V2 (ACS) are the ones having issues with the new firmware.
Poll: Anyone with a 3200ACM/32X have any issues with wireless drops?
I might just create two separate PR one for stability improvements on Intel cards (please let me know if you have specifically 7260 AC).
And the second to be if the firmware update is even worth the hassle of making it device specific. Anyone testes it only on 1900AC v1? I am currently using it exclusively on the V1 and it works fine.
no issues on wrt3200, radio on channel 100 wds-sta and ap .
As far as I know, the latest stable driver was pushed to OpenWrt a long time ago, and is included in the 19.07 series. There is a newer version, and some experimental parallel versions, that have been causing trouble to some people.
Anyway, I think you should ask the official maintainer of the drivers at OpenWrt, I just package new or experimental versions, so people can install them directly on top fo a stable release.
I'm seeing drops on 5Ghz Radio.
Devices lock up and cannot reconnect. Hard wired devices operating fine. Takes some wiggling to get the 5Ghz band to operate again...
Which device?
I have the WRT3200ACM
I use the current snapshot 5.4.80? The driver with the snapshot does not obtain an IP So I downgraded back to 4.19 - that one works for me
After some research I found the following to compile the latest mwlwifi:
package/kernel/mwlwifi/Makefile:PKG_VERSION:=10.3.0.17-20160601
PKG_SOURCE_VERSION:=4bb95ba1aeccce506a95499b49b9b844ecfae8a1
PKG_VERSION is used to give a version to the package, and you are mostly free to put here whatever makes sense to you.
PKG_SOURCE_VERSION is the commit identifier at kaloz's repo that you want to biuld (https://github.com/kaloz/mwlwifi/commits/master).
So I changed it to:
PKG_SOURCE_VERSION:=111118dc2ea3b592a5f7dff18c82d57a651970e7
Rename package/kernel/mwlwifi/patches to package/kernel/mwlwifi/patches.tmp.
Run the following commands:
make package/kernel/mwlwifi/clean
and compile the whole firmware again
make -j $((1+
nproc))
So far so good it seems to work
Great job on the driver
Well after 1 day of uptime it did the same thing again and did not assign an IP address anymore. Right now I down graded again. (I still have the second partition setup with the old 4.19 firmware) I might wait for the next update of the driver/firmware for mwlwifi.
The end result of the above is that you built the same mwlwifi code as what is currently built by master, just by using the latest commit from the upstream repo in favour of the current patch in master. The upstream repo has been dead for quite some time so I would not expect any improvements for mwlwifi to come any time soon. As this is the first report I have seen of the issue you describe, I suspect that you are probably pointing at the wrong culprit.