Yes, that’s right, the crazy project of mine to have FreeBSD and OpenBSD WiFi drivers coexist on Haiku is a success, and as of hrev56168 it has been merged out of my working branches and into Haiku itself!
As of right now, only the idualwifi7260 (equivalent to iwm on OpenBSD) driver is enabled, but in the coming days, the brand-new iaxwifi200 (equivalent to iwx on OpenBSD) driver should also get enabled.
I already commented on various tickets of relevance, but if you want to try out the new driver, note that you must upgrade the intel_wifi_firmwares package to at least version 2022_01_11 and wpa_supplicant to at least version 2.10.haiku.1; these are not yet bundled with the default nightly builds, and without the first the driver will not even start, and without the second it will not be able to connect to secured networks at all.
There are a few TODOs (wpa_supplicant does not interact perfectly with the new drivers; for example it won’t save networks though it will remember ones you already saved; and it won’t notice if connecting fails and ask you to enter a different password) but on the whole, things seem much better than the old version of the driver. I did not experiment much to see if 802.11ac mode worked, but I was getting much faster speeds connecting to my 2.4GHz (i.e. non-ac) AP, so I think that means 802.11n mode was working at least.
That’s really great news
I’ll have to try that on my Medion Akoya soon,at the last try there was an FreeBSD driver available but it couldn’t connect.
Maybe the OpenBSD driver works
Wifi card Intel N6235 in the revision 56168 works poorly, unstable.
It spontaneously disconnects and loses its wifi access point.
Revision 56147 works perfectly, a stable wifi connection.
It is likely that the bug appeared in the revision 56148 - 56158.
vendor 8086: Intel Corporation
device 088e: Centrino Advanced-N 6235 https://dev.haiku-os.org/ticket/17778
This (the iprowifi4965) driver is still the FreeBSD one and on the FreeBSD WiFi stack, so the only way it could be affected recently is if the general changes to the FreeBSD compatibility layer broke it. I was hoping I didn’t inadvertently introduce regressions here, but it’s quite possible I did.
Try again with tomorrow’s (well, tonight’s) nightly build; there were some problems related to scan results that are now committed. Though it could be that the device errored somehow, so you may need to file a ticket.
