# Kernel 5.16 6Ghz Disabled AX210

**URL:** <https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675>\
**Category:** Framework Laptop 13\
**Created:** [February 24, 2022, 2:23am UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675 "2022-02-24T02:23:52Z")\
**Posts on this page:** 8\
**Page:** 1

<div class="post-metadata">

**Author:** ![ryanpetris](https://sea1.discourse-cdn.com/flex001/user_avatar/community.frame.work/ryanpetris/32/6963_2.png) [@ryanpetris](https://community.frame.work/u/ryanpetris)\
**Post date:** [February 24, 2022, 2:23am UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/1 "2022-02-24T02:23:52Z")

</div>

It appears that the Linux Kernel version 5.16 has disabled 6Ghz wifi specifically on the Framework Laptop for the AX210 wifi card. While I don’t have a 6Ghz router to test this on, I can see that `iw reg get` and `iw list` reports all 6Ghz channels as disabled. Rolling back to 5.15 results in the channels being enabled again.

I tracked this down to this commit: [[PATCH 03/12] iwlwifi: mvm: read 6E enablement flags from DSM and pass to FW - Luca Coelho](https://lore.kernel.org/all/iwlwifi.20211024181719.2fa34d31383c.I6504005c60882c94e6e58f64cab4e42e6481ce08@changeid/)

This commit is getting a value for oem\_uhb\_allow\_bitmap from somewhere in the BIOS, and therefore I’m guessing this is something that Framework will need to fix on their end.

Enabling debugging for iwlwifi shows the following in the journal:

```auto
kernel: iwlwifi 0000:aa:00.0: iwl_mvm_lari_cfg sending LARI_CONFIG_CHANGE, oem_uhb_allow_bitmap=0x1

```

Other machines show this value as 0x0, and removing this commit fixes the issue.

---

<div class="post-metadata">

**Author:** ![D.H](https://sea1.discourse-cdn.com/flex001/user_avatar/community.frame.work/d.h/32/17958_2.png) [@D.H](https://community.frame.work/u/D.H)\
**Post date:** [February 24, 2022, 1:20pm UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/2 "2022-02-24T13:20:51Z")

</div>

Thorough write-up of the issue.

Just for my own curiosity, what .ucode firmware file for the ax210 is loaded by your distro?

---

<div class="post-metadata">

**Author:** ![ryanpetris](https://sea1.discourse-cdn.com/flex001/user_avatar/community.frame.work/ryanpetris/32/6963_2.png) [@ryanpetris](https://community.frame.work/u/ryanpetris)\
**Post date:** [February 24, 2022, 1:30pm UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/3 "2022-02-24T13:30:40Z")

</div>

For 5.16, it’s version 67:

```auto
loaded firmware version 67.8f59b80b.0 ty-a0-gf-a0-67.ucode op_mode iwlmvm

```

For 5.15, it was version 66. I did try loading version 66 instead of 67 on 5.16 but that didn’t make a difference; that was before I found that commit.

Also I didn’t mention it in my original post but this is on Arch.

---

<div class="post-metadata">

**Author:** ![D.H](https://sea1.discourse-cdn.com/flex001/user_avatar/community.frame.work/d.h/32/17958_2.png) [@D.H](https://community.frame.work/u/D.H)\
**Post date:** [February 24, 2022, 3:11pm UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/4 "2022-02-24T15:11:49Z")

</div>

Also on arch derivative (EndeavourOS), also seeing the 6ghz disabled on 5.16.10, same ucode as you, so it’s not a bug that’s peculiar to your install. Booting into 5.15.24 (current Linux-lts version for me) seems to restore the 6ghz channels according to `iw list` .

Doesn’t really affect me as I only have a WiFi 5 router in the house, but definitely would be good to get this cleared up.

@ryanpetris Were you able to report this bug upstream yet?

---

<div class="post-metadata">

**Author:** ![ryanpetris](https://sea1.discourse-cdn.com/flex001/user_avatar/community.frame.work/ryanpetris/32/6963_2.png) [@ryanpetris](https://community.frame.work/u/ryanpetris)\
**Post date:** [February 24, 2022, 3:27pm UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/5 "2022-02-24T15:27:46Z")

</div>

I haven’t reported it as it doesn’t look like a bug, it looks intentional. As I mentioned I have other machines that don’t exhibit this issue, either because the bios is returning `0` or doesn’t have that value.

To me it appears that this needs to change on Framework’s end.

---

<div class="post-metadata">

**Author:** ![D.H](https://sea1.discourse-cdn.com/flex001/user_avatar/community.frame.work/d.h/32/17958_2.png) [@D.H](https://community.frame.work/u/D.H)\
**Post date:** [February 27, 2022, 3:05pm UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/6 "2022-02-27T15:05:19Z")

</div>

Seems to still be this way on 5.16.11 now too.

I’ll look into reporting this regression upstream. Even if it was intentional, taking something from “just works” to “broken without OEM involvement” is a step in the wrong direction…

---

<div class="post-metadata">

**Author:** ![lbkNhubert](https://avatars.discourse-cdn.com/v4/letter/l/e19b73/32.png) [@lbkNhubert](https://community.frame.work/u/lbkNhubert)\
**Post date:** [August 7, 2022, 10:49pm UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/7 "2022-08-07T22:49:28Z")

</div>

@D.H - were you ever able to report the regression? I just tested on Manjaro with 5.19.0-2 and 5.15.59-1. The former, nothing, the latter, successful connection.

---

<div class="post-metadata">

**Author:** ![D.H](https://sea1.discourse-cdn.com/flex001/user_avatar/community.frame.work/d.h/32/17958_2.png) [@D.H](https://community.frame.work/u/D.H)\
**Post date:** [August 8, 2022, 2:05am UTC](https://community.frame.work/t/kernel-5-16-6ghz-disabled-ax210/15675/8 "2022-08-08T02:05:46Z")

</div>

I totally forgot.

6Ghz is still showing disabled across the board on 5.18.16 kernel with intel firmware version firmware version 71.058653f6.0 ty-a0-gf-a0-71.ucode . Then again, I do not have a 6Ghz base station so it may be disabling those due to not hearing any signals in that band and assuming that it is regulatory banned in this location.

I really need to get a move on on my mainboard as wireless router/AP/NAS project…

Any bright ideas about that oem\_uhb\_allow\_bitmap value, @DHowett ?
