Closed Quackoman closed 3 years ago
This issue is in the FAQ at the end of README.
Quote: "It seems to me that this can be fixed driver side to read that multiple adapters..."
No doubt about it. Trying plugging in two identical adapters with MT7612u chipsets and you will get two interfaces ready to go.
https://github.com/morrownr/USB-WiFi
Back on topic: I haven't tested this but I think there is a work around:
PACKAGE_NAME="rtl88x2bu-2" BUILT_MODULE_NAME[0]="88x2bu-2"
My thoughts on this issue: Why on earth can Realtek not support us with good in-kernel drivers that don't need basic functionality fixed or added?
Most of my time is taken up keeping this and the other drivers I support up with kernel dev as well as fixing some basic functionality. It could be a while before I have time to look at a real fix but you and others are welcome work on it, document it and submit it.
Lastly, if you need really dependable identical AC1200 usb adapters, get a couple of adapters based on the MT7612u chipset. You won't even have to find and compile a driver.
Thanks for the quick feedback and sorry I missed that ending of the readme.
I agree, it is crazy that they sell so many of these chipsets and have no care for bettering or simply stabilizing it's functionality.
I'll check out the MT7612u.
Way back when i first ran into this issue with the multiple 8812bu adapters, i tried that workaround with compiling the driver under the -2/-3 name but when loaded it would still name itself 88xbu if I ran an lsmod. and the two adapters will still load the same driver.
it's one that i gave up on, but when i saw this repo somewhat active and making progress on making it a more stable, I thought maybe let me ask if anyone ever did figure it out.
thanks again!
My memory is bad. Here is copy and paste from a guy in the repo that says this works:
There are these things to change:
/include/autoconf.h: DRV_NAME (I'm not sure if changing it is necessary. It's not the kernel module name for the kernel, but the usb driver name for usbcore) Makefile: MODULE_NAME dkms.conf: BUILD_MODULE_NAME, PACKAGE_NAME (I'm not sure if changing PACKAGE_NAME is necessary) /os_dep/linux/usb_intf.c: Change supported vendors, so only the driver you want to load get loaded when a usb device is detected. Now I don't have to return them to amazon, nice. Thanks for your repo! (200 stars congrats) (Feel free to close this issue. I'm leaving it open for now)
currently my lsmod output is showing both drivers:
$ lsmod | grep cfg80211 cfg80211 704512 5 iwlmvm,88x2bu1,88x2bu2,iwlwifi,mac80211
I haven't tested this but in another later msg he reported stable operation. I realize his documentation is not totally complete and I have not tested this myself.
[vendor_id]:[product_id]
pair, then currently I do not think my solution works for you.You have to find another way to force the OS to load two different instances of drivers.
Hi @Quackoman , I am the "a guy in the repo that says this works"
Currently I am running a very stable hotspot using two 88x2bu adapters. I can tell you that it definitely works, and works well.
@morrownr has already posted how I managed to make it work. If you have any questions you can ask me. You can also check the previous issue #73
My setup is:
------------------- ---------------------------------------------------------- ---------------------------------
|(wireless router)|>>>>|88x2bu adpater 1---(wifi extender pc)---88x2bu adapter 2|>>>>|builtin wifi card---(my laptop)|
------------------- ---------------------------------------------------------- ---------------------------------
As you can see, the internet traffic has to go through 3 different wireless cards to reach the router in my home.
I am using 300Mbps AT&T fiber cable. Since 88x2bu supports 867mbps, I can reach the maximum speed with this setup.
Here is a screenshot of downloading World of Warcraft using Battle.net app. Constantly 30+MB/s
Here is a screenshot of "speed test" on Google:
This is lshw
listing NIC devices on my wifi extender pc.
This is another screenshot of lshw
listing USB devices showing both of the drivers.
This is lsmod | grep cfg80211
The virtual machine method I mentioned under #73 may work for you, but unfortunately letting the virtual machine to control USB devices consumes a lot of CPU resources. If you do not mind you can have a try
If adapters are plugged in such that they get 1-1.1.1, 1-1.3 then it works.
I cannot believe it works this way! I did not even realize USB hubs has something to do with this problem. If one driver can handle multiple devices in this setup, then just use a hub and do not mess with the driver.
Hey Everyone,
i've had an issue for these 8812BU adapters. If i plug in two or more to the same usb hub, wpa_supplicant is only able to control the last one plugged in. I reproduced the same using ubuntu 20.04 arm64-5,4.0-1041raspi and centos 7 arm64-5.10.xx.
To reproduce:
if i unbind and bind the usb port for the adapter that is not working, i can return functionality as it will now be the last one plugged in, and so on for the previous one that no longer works if i need to change ssid i can bind and unbind it to regain control.
It seems to me that this can be fixed driver side to read that multiple adapters are on a single hub using the same root usb address: i.e if adapters are plugged in such that their usb assignment is: 1-1.1.1. 1-1.1.2 then issue arises and only last plugged in adapter is able to be controlled. If adapters are plugged in such that they get 1-1.1.1, 1-1.3 then it works.
Another case tested: on the same usb hub, if you plug in two wifi adapters that load different wifi-drivers: plug in a RTL8812AU and a RTL8812BU adapter, there is no issue to control each of them via wpa_supplicant/nmtui only happens when same driver module 88xbu is loaded with two identical adapters on the hub.