Linux wireless drivers development
 help / color / mirror / Atom feed
From: Ping-Ke Shih <pkshih@realtek.com>
To: Kadir Yavuz Kurt <k.yavuzkurt1@gmail.com>
Cc: "linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>
Subject: RE: [BUG] wifi: rtw89: 8922ae: no association until module reload
Date: Wed, 18 Nov 2026 00:41:55 +0000	[thread overview]
Message-ID: <95f8f42c4d844a778cbeac6aae10ec5a@realtek.com> (raw)
In-Reply-To: <20261117190249.8814-1-k.yavuzkurt1@gmail.com>

Kadir Yavuz Kurt <k.yavuzkurt1@gmail.com> wrote:
> > This is a good point. I think you can add rtw89 into blacklist, and
> > manually load the modules after booting.
> 
> That works. Blacklisted rtw89_8922ae, booted, loaded it by hand once
> the desktop was up:
> 
>     [ 7.427775] Bluetooth: hci0: RTL: fw version 0x41c0c905
>     [50.304704] rtw89_8922ae 0000:07:00.0: loaded firmware
>                                            rtw89/rtw8922a_fw-4.bin
>     [53.787826] wlp7s0: authenticate with xx:xx:xx:xx:xx:xx
>     [53.788897] wlp7s0: authenticated
>     [53.789371] wlp7s0: associate with xx:xx:xx:xx:xx:xx (try 1/3)
>     [53.907495] wlp7s0: associated
> 
> First attempt, single load. So the chip doesn't need a second init,
> which I'd been wondering about.
> 
> > If this works, add a udev rule to delay loading rtw89.
> 
> This is where it fell apart. I automated it as a systemd unit
> (After=bluetooth.service, wait for 'bluetoothctl show' to report
> Powered: yes, then modprobe):
> 
>     [ 7.464834] Bluetooth: hci0: RTL: fw version 0x41c0c905
>     [ 9.994965] rtw89_8922ae 0000:07:00.0: loaded firmware
>                                            rtw89/rtw8922a_fw-4.bin
> 
> No authenticate line, no association. BT had finished 2.5 s earlier, so
> waiting for BT isn't enough.

Did you use NetworkManager to connect? If so, maybe turn GUI off/on
to see if it can try to connect. 

> 
> Which pushed me to test what I should have tested first: no BT at all.
> blacklist btusb + btrtl, rtw89 left to load normally at boot. lsmod
> shows no btusb, there's no hci0, and no BT firmware download anywhere
> in the log.
> 
>     [ 3.037204] rtw89_8922ae 0000:07:00.0: loaded firmware
>                                            rtw89/rtw8922a_fw-4.bin
>     [54.735452] rtw89_8922ae 0000:07:00.0: loaded firmware  (my reload)
>     [58.218773] wlp7s0: authenticate with xx:xx:xx:xx:xx:xx
>     [58.219916] wlp7s0: authenticated
> 
> Still broken with Bluetooth completely gone.
> 
> So I was wrong about Bluetooth and I'd like to withdraw that part of my
> first mail. The 1 ms overlap was real but it wasn't causing anything.
> Load at 3 s: fails. At 10 s: fails. At 3 s with no BT: fails. At 50 s:
> works. The only thing that tracks the outcome is how late the probe
> runs, and the reload only helps because it happens late.

I don't know the exact time BT needs to initialize, but 10s should be
enough. No idea....

> 
> What I can't tell you is what the driver is sensitive to that early in
> boot. This box starts a lot at once (amdgpu, nvidia, docker, libvirt,
> tailscale, ollama), so contention around probe time is my guess, but
> it's only a guess.
> 
> All of the above is on 7.0.0-43-generic.
> 
> I can bisect the delay to find where it flips. Each point needs its own
> boot, since once a probe has failed every load after it succeeds, but
> reboots are cheap here. Anything you'd like instrumented at probe time,
> or a debug_mask that would show what differs? I'll test patches on this
> hardware whenever.

Have you tried manually scanning AP by iw command? Just want to know
if WiFi works while it doesn't trigger connection. 




      reply	other threads:[~2026-11-18  0:41 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-03 19:44 [BUG] wifi: rtw89: 8922ae: no association until module reload Kadir Yavuz Kurt
2026-11-11  2:46 ` Ping-Ke Shih
2026-11-17 19:02   ` Kadir Yavuz Kurt
2026-11-18  0:41     ` Ping-Ke Shih [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=95f8f42c4d844a778cbeac6aae10ec5a@realtek.com \
    --to=pkshih@realtek.com \
    --cc=k.yavuzkurt1@gmail.com \
    --cc=linux-wireless@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox