On Thu, Mar 10, 2022 at 01:56:19PM -0500, Dave Voutila wrote: > > Stefan Sperling <[email protected]> writes: > > > On Thu, Mar 10, 2022 at 12:25:17PM +0100, Stefan Sperling wrote: > >> Unless anyone else finds a problem, this patch can be considered ready > >> for review and commit. > > > > Of course, I forgot to apply my sysassert fix to the second phy context > > command function... Fixed here. > > > > Looking good on the 3 devices I've tested, with one minor nit. All 3 > devices are similar AX200s like so: > > iwx0 at pci3 dev 0 function 0 "Intel Wi-Fi 6 AX200" rev 0x1a, msix > iwx0: hw rev 0x340, fw ver 67.8f59b80b.0, address ... > > For some reason, on my lenovo x13 I had to manually enable 11ac mode via > ifconfig. No idea why. It was sitting on the table next to my Go 2 that > automatically picked up on 11ac mode.
Which channel was it on while not in 11ac mode? Only the 5GHz channel range (chan 36+) can use 11ac. In the 2GHz channel range (chan 1-12 for the US) 11ac cannot be used. In that range it should autoselect mode 11n, or 11g/11b for very old APs. We could tweak access point selection heuristics to prefer 11ac over 11n if both are available for a given SSID. But that can be done in a later patch. > In all 3 cases I tested with tpcbench between my apu2 gateway connected > to my Synology RT2600ac and the client machine. In all 3 cases I saw > close to a 100% increase in average bandwidth reported. :-) Very nice. Thanks for testing!
