Question IWLWIFI + AX210 / Linux - needs some work

[Tech Junky]

So, I've bene playing around with my TMHI 5G setup for a couple of months now trying to get the best speeds possible from it. I've stripped it from the case and took off anything that doesn't need to be on the "modem" and added some different antennas ranging from 3db-7db and converted the pigtails from SMA to RP-SMA to kill off the B71/N71 600mhz slow band .

I've come across a couple of postings around the web showing people hitting upwards of 700mbps+ and got curious so I took the gateway on the road and did some speed testing around town. Most of them averaged 200mbps DL which is about what I normally get too. Knowing I have a cell site mounted on the side of my building though and digging through other tech postings I decided to pull into the parking lot next door and run a speed test just to see on a hunch if LOS at a slight distance of ~50-100ft would make a difference compared to my window 20ft under the antennas on the side of the building.

I'm really wishing I didn't do this speed test because the damned thing hit 600mbps DL speeds and now I'm thinking of ways to get it to do that inside the house. So, I started by moving the GW over to the window closest to the antennas and didn't really see much improvement. I moved the GW off to the west ~15ft to see if cancelling out some of the signal would improve the throughput.

Anyway... so, moving the GW into a different room meant either plugging the server/router into the wall and jumping it through the Ethernet hub in the closet or simply use the wifi card to connect to the GW on AX. This is where things get interesting.... So, I've used an AX210 on my laptop w/ Linux and not had any issues and thought nothing of doing it from the server >> GW. SMH this is odd and slow and a bit of a PITA to setup with my unique network configuration using bonding / bridging.

Got it working w/ the GW / hidden SSID (needed to add a line to wpa to make it work). So, I pull up waveform to see what the hell it's doing and RX is sitting at 6mps while TX is over a 1gbps. WTH is going on here. Never an issue in the laptop/s over the past couple of years so, why is the stupid server being so difficult? Tried playing with different kernel versions down to 5.16.x and nothing changed. Found some "options" though that you can put into /etc/modprobe.d/iwlwifi.conf based on the output from "iw list".

There's a few options you can invoke when probing the module to disable / enable / power save / etc. A few posts were mentioning using "11n_disable=" which didn't do much for the speed issue. "disable_11ac=" did bring the RX up but then dropped the TX down. " disable_11ax=" brought both to AC top level speeds of 866.7mbps which "works" but kind of defeats the purpose of using an AX card for AC speeds.

I'm considering grabbing a QCA2066-5 / WCN6856 based card for internalizing the AP function to the server again vs using an AP with a POE injector. I've done this before with a different card but, it took awhile for AX cards other than Intel to come to market and they're still about double the cost @ $67 but still far cheaper than a router w/ 6ghz. Switching to ath11k drivers should perform better than iwlwifi does but, If there's anyone that has seen / fixed this w/o using the flags to disable functions to band aid the speeds it would be appreciated.

 

[Tech Junky]

Ordered a couple of EmWicon WMX7205 cards to internalize the AP function to add 6ghz and ditch the external AP / POE injector. Hopefully testing with these will show better performance w/o the need to hobble AX speeds to get higher RX speed.

I have a plan on using 1 to put in place of the AX210 for WIFI backup to the TMHI to be able to move the GW around and find better speeds. The 2nd one for the AP function using hostapd like I have in past with AC.

This does a couple of things which for one is getting away from using IWLWIFI by using ath11k instead to circumvent the speed issue. Also reduces the need for 2 devices / cables to do WIFI from the DIY server / router.

The unknown though is whether I can get the 160mhz working on the AX cards as the last time I used AC it was being temperamental but, since the physical AP has been stable @ 160mhz for quite awhile I suspect the interference causing it to fila on the prior attempts should be gone.

Just have to wait for the cards to show up along with some other stuff for the TMHI gateway as I picked up some directional antennas and bases to see if I can snag some more speed by aiming the antennas out the window with a narrow beam to catch the higher bandwidth and realize 600mbps DL consistently.

 

[Tech Junky]

modinfo iwlwifi

parm:           swcrypto:using crypto in software (default 0 [hardware]) (int)
parm:           11n_disable:disable 11n functionality, bitmap: 1: full, 2: disable agg TX, 4: disable agg RX, 8 enable agg TX (uint)
parm:           amsdu_size:amsdu size 0: 12K for multi Rx queue devices, 2K for AX210 devices, 4K for other devices 1:4K 2:8K 3:12K (16K buffers) 4: 2K (default 0) (int)
parm:           fw_restart:restart firmware in case of error (default true) (bool)
parm:           nvm_file:NVM file name (charp)
parm:           uapsd_disable:disable U-APSD functionality bitmap 1: BSS 2: P2P Client (default: 3) (uint)
parm:           enable_ini:0:disable, 1-15:FW_DBG_PRESET Values, 16:enabled without preset value defined,Debug INI TLV FW debug infrastructure (default: 16)
parm:           bt_coex_active:enable wifi/bt co-exist (default: enable) (bool)
parm:           led_mode:0=system default, 1=On(RF On)/Off(RF Off), 2=blinking, 3=Off (default: 0) (int)
parm:           power_save:enable WiFi power management (default: disable) (bool)
parm:           power_level:default power save level (range from 1 - 5, default: 1) (int)
parm:           disable_11ac:Disable VHT capabilities (default: false) (bool)
parm:           remove_when_gone:Remove dev from PCIe bus if it is deemed inaccessible (default: false) (bool)
parm:           disable_11ax:Disable HE capabilities (default: false) (bool)

These are the flags I was referencing earlier. So, I stumbled across a new one listed somewhere when trying to confirm some CRDA / DFS info and it's

lar_disable=1

I took off all the other codes / flags and

|rx rate: 1200.9 Mbit/s 80MHz, tx rate: 1200.9 Mbit/s 80MHz

Still waiting on the new cards to ship....

There are some interesting characteristics though wen looking at the reg db output...

global
country US: DFS-FCC
        (902 - 904 @ 2), (N/A, 30), (N/A)
        (904 - 920 @ 16), (N/A, 30), (N/A)
        (920 - 928 @ 8), (N/A, 30), (N/A)
        (2400 - 2472 @ 40), (N/A, 30), (N/A)
        (5150 - 5250 @ 80), (N/A, 23), (N/A), AUTO-BW
        (5250 - 5350 @ 80), (N/A, 24), (0 ms), DFS, AUTO-BW
        (5470 - 5730 @ 160), (N/A, 24), (0 ms), DFS
        (5730 - 5850 @ 80), (N/A, 30), (N/A), AUTO-BW
        (5850 - 5895 @ 40), (N/A, 27), (N/A), NO-OUTDOOR, AUTO-BW, PASSIVE-SCAN
        (5925 - 7125 @ 320), (N/A, 12), (N/A), NO-OUTDOOR, PASSIVE-SCAN
        (57240 - 71000 @ 2160), (N/A, 40), (N/A)

So, the 6ghz looks like it tops out with 320mhz and a huge band of 2160mhz for ~60ghz. But there's nothing out until WIFI7 for 320mhz channels but it's already baked into the card? I wonder if maybe pitting the two cards against each other will get to that 320mhz with some tweaking....

 

[Tech Junky]

Well, the Mt7921K's were unimpressive. I discovered also hostapd doesn't want to enable ieee80211ax even though it's an option in the config templates along with anything that enable HE. I'm fairly certain this is a program issue and not related to the MTK chipset. I was able to get the MTK's working but, the speeds capped at B/G/N under 100mbps. The AX210 at least as a client hits top end 866mbs but can't be used for an AP beyond 2.4ghz bands due to it being a disabled / hobbled feature by Intel.

Now, the debate is whether to push forward with getting Qualcomm based chips and continue down the path of an internal AP w/ AC speeds until hostapd unlocks the AX feature set and enable it at a later time or just forgo the whole idea and wait for WIFI7 to be released.