Date: Thu, 5 Mar 2026 20:03:04 +0100 From: =?UTF-8?Q?Tilman_Keskin=C3=B6z?= <arved@FreeBSD.org> To: wireless@freebsd.org Subject: Re: LinuxKPI (iwlwifi, rtw8x) Please test! Important! (also in stable/15 now) Message-ID: <a25027df-b981-49eb-982c-ea86cfaeedbd@FreeBSD.org> In-Reply-To: <p836qqn7-6pq4-nrq-p346-8qp5ors7p170@mnoonqbm.arg> References: <r158r59-93o7-9767-1r57-805sq1s07q4p@mnoonqbm.arg> <p836qqn7-6pq4-nrq-p346-8qp5ors7p170@mnoonqbm.arg>
index | next in thread | previous in thread | raw e-mail
[-- Attachment #1 --]
Hi,
I am a bit behind, but rtw89 works with these changes in stable.
Only issue I have (not cause by these changes, already started a few
weeks ago, but I forgot to report it):
On bootup i get:
wlan0: link state changed to UP
wlan0: link state changed to DOWN
rtw890: lkpi_sta_a_to_a:3325: lvif 0xfffffe00d3dd1000 vap
0xfffffe00d3dd1010 iv_bss 0xfffffe00cc1c0000 lvif_bss 0xfffff8001f5ea00
0 lvif_bss->ni 0xfffffe0121674000 synched 0
rtw890: lkpi_iv_newstate: error 95 during state transition 2 (AUTH) -> 2
(AUTH)
wlan0: link state changed to UP
wlan0: link state changed to DOWN
rtw890: lkpi_sta_a_to_a:3325: lvif 0xfffffe00d3dd1000 vap
0xfffffe00d3dd1010 iv_bss 0xfffffe0120887000 lvif_bss 0xfffff8002c21f00
0 lvif_bss->ni 0xfffffe00cc1c0000 synched 0
rtw890: lkpi_iv_newstate: error 95 during state transition 2 (AUTH) -> 2
(AUTH)
wlan0: link state changed to UP
Which takes a few seconds, but afterwards it works just fine.
On 2/27/26 03:45, Bjoern A. Zeeb wrote:
> On Wed, 25 Feb 2026, Bjoern A. Zeeb wrote:
>
> I've done the MFCs and all changes are also all in stab;e/15 now.
> So please test!
>
> Lots of health,
> /bz
>
>
>> Hi,
>>
>> about 3 years ago, some of you may remember, we were hunting down the
>> constant (firmware) crashes with iwlwifi due to net80211 not playing
>> nice
>> and the LinuxKPI 802.11 sta state machine not able to deal with it to
>> make
>> the firmware happy.
>>
>> In that course I unfolded some state machine bits, which resultet in
>> code
>> duplication and some other complications. Fast forward, I just
>> undone some
>> of that in main. This is probably one of the more risky changes as
>> of late.
>> I tried to test it but as we also have learnt in the last years, the
>> testing
>> of many will reveal problems more likely than me trying to test in my
>> limited
>> capacity.
>>
>>
>>
>> So please all test main now!
>>
>> Prepare yourself in case of trouble (boot envorments, or otherwise,
>> in case
>> you need to roll back know how to).
>> Let it run for a few hours or a day or two and then please report
>> back if
>> it is still working without extra troubles compared to before.
>> In case you hit any hiccup please report right away!
>> I will try to reply promptly (I do sleep occassionally) but real life
>> currently also takes some share so it may be a day before I get to the
>> emails.
>> Please include driver and chipset/firmware you are using if possible
>> along with any possible problem report.
>>
>>
>>
>> Why is this important (now)?
>>
>> There'll be follow-up changes needed to get the iwlwifi update into main
>> as well so that all drivers are on the same level (v6.19) for a change
>> again. The reason this is important now is that this will be our base
>> for 15.1-RELEASE so we will have 6ish weeks still to sort out any
>> problems.
>>
>> In addition (bonus points) if things go smoothly there is the next
>> branch
>> waiting for me to merge to bring further improvements still during March
>> (at least to main and possibly stable/15).
>>
>>
>> What if you are on stable/15?
>>
>> I believe you could just grab the last changes and merge them yourself.
>> I will go and have a look and try to merge anything I can Thu or Fri
>> (before the weekend) to keep the diff minimal in case that helps!
>>
>>
>> Lots of health!
>> Bjoern
>>
>>
>
[-- Attachment #2 --]
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Hi,</p>
<p>I am a bit behind, but rtw89 works with these changes in stable.</p>
<p>Only issue I have (not cause by these changes, already started a
few weeks ago, but I forgot to report it):</p>
<p>On bootup i get:</p>
<p>
<span style="font-family:monospace"><span
style="color:#000000;background-color:#ffffff;">wlan0: link
state changed to UP</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">wlan0:
link state changed to DOWN</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">rtw890:
lkpi_sta_a_to_a:3325: lvif 0xfffffe00d3dd1000 vap
0xfffffe00d3dd1010 iv_bss 0xfffffe00cc1c0000 lvif_bss
0xfffff8001f5ea00</span><br>
<span style="color:#000000;background-color:#ffffff;">0
lvif_bss->ni 0xfffffe0121674000 synched 0</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">rtw890:
lkpi_iv_newstate: error 95 during state transition 2 (AUTH)
-> 2 (AUTH)</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">wlan0:
link state changed to UP</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">wlan0:
link state changed to DOWN</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">rtw890:
lkpi_sta_a_to_a:3325: lvif 0xfffffe00d3dd1000 vap
0xfffffe00d3dd1010 iv_bss 0xfffffe0120887000 lvif_bss
0xfffff8002c21f00</span><br>
<span style="color:#000000;background-color:#ffffff;">0
lvif_bss->ni 0xfffffe00cc1c0000 synched 0</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">rtw890:
lkpi_iv_newstate: error 95 during state transition 2 (AUTH)
-> 2 (AUTH)</span><span
style="color:#000000;background-color:#ffffff;">
</span><br>
<span style="color:#000000;background-color:#ffffff;">wlan0:
link state changed to UP</span><br>
<span style="color:#000000;background-color:#ffffff;">
</span><br>
</span>Which takes a few seconds, but afterwards it works just
fine.</p>
<p><br>
</p>
<p><br>
</p>
<p><br>
</p>
<div class="moz-cite-prefix">On 2/27/26 03:45, Bjoern A. Zeeb wrote:<br>
</div>
<blockquote type="cite"
cite="mid:p836qqn7-6pq4-nrq-p346-8qp5ors7p170@mnoonqbm.arg">On
Wed, 25 Feb 2026, Bjoern A. Zeeb wrote:
<br>
<br>
I've done the MFCs and all changes are also all in stab;e/15 now.
<br>
So please test!
<br>
<br>
Lots of health,
<br>
/bz
<br>
<br>
<br>
<blockquote type="cite">Hi,
<br>
<br>
about 3 years ago, some of you may remember, we were hunting
down the
<br>
constant (firmware) crashes with iwlwifi due to net80211 not
playing nice
<br>
and the LinuxKPI 802.11 sta state machine not able to deal with
it to make
<br>
the firmware happy.
<br>
<br>
In that course I unfolded some state machine bits, which
resultet in code
<br>
duplication and some other complications. Fast forward, I just
undone some
<br>
of that in main. This is probably one of the more risky changes
as of late.
<br>
I tried to test it but as we also have learnt in the last years,
the testing
<br>
of many will reveal problems more likely than me trying to test
in my limited
<br>
capacity.
<br>
<br>
<br>
<br>
So please all test main now!
<br>
<br>
Prepare yourself in case of trouble (boot envorments, or
otherwise, in case
<br>
you need to roll back know how to).
<br>
Let it run for a few hours or a day or two and then please
report back if
<br>
it is still working without extra troubles compared to before.
<br>
In case you hit any hiccup please report right away!
<br>
I will try to reply promptly (I do sleep occassionally) but real
life
<br>
currently also takes some share so it may be a day before I get
to the
<br>
emails.
<br>
Please include driver and chipset/firmware you are using if
possible
<br>
along with any possible problem report.
<br>
<br>
<br>
<br>
Why is this important (now)?
<br>
<br>
There'll be follow-up changes needed to get the iwlwifi update
into main
<br>
as well so that all drivers are on the same level (v6.19) for a
change
<br>
again. The reason this is important now is that this will be
our base
<br>
for 15.1-RELEASE so we will have 6ish weeks still to sort out
any problems.
<br>
<br>
In addition (bonus points) if things go smoothly there is the
next branch
<br>
waiting for me to merge to bring further improvements still
during March
<br>
(at least to main and possibly stable/15).
<br>
<br>
<br>
What if you are on stable/15?
<br>
<br>
I believe you could just grab the last changes and merge them
yourself.
<br>
I will go and have a look and try to merge anything I can Thu or
Fri
<br>
(before the weekend) to keep the diff minimal in case that
helps!
<br>
<br>
<br>
Lots of health!
<br>
Bjoern
<br>
<br>
<br>
</blockquote>
<br>
</blockquote>
</body>
</html>
home |
help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?a25027df-b981-49eb-982c-ea86cfaeedbd>