From: Mehmet Fide <mehmet.fide@gmail.com>
To: Johannes Berg <johannes@sipsolutions.net>
Cc: "Simon Wunderlich" <sw@simonwunderlich.de>,
"Toke Høiland-Jørgensen" <toke@toke.dk>,
linux-wireless@vger.kernel.org, linux-kernel@vger.kernel.org,
mehmet.fide@screeningeagle.com
Subject: [PATCH wireless-next] wifi: mac80211: replace the CSA beacon before clearing csa_active
Date: Tue, 17 Nov 2026 16:31:54 +0100 [thread overview]
Message-ID: <20261117153154.1804945-1-mehmet.fide@gmail.com> (raw)
From: Mehmet Fide <mehmet.fide@screeningeagle.com>
__ieee80211_csa_finalize() clears csa_active and only then replaces the
CSA beacon with the one for after the switch. A driver that fetches its
beacons with ieee80211_beacon_get() and skips them once the countdown
is complete, as ath9k does in ath9k_csa_is_finished(), tests csa_active
first. A beacon fetched between the two steps passes that test and gets
the CSA beacon again, which goes out on the new channel, and
ieee80211_beacon_get() reports "beacon TX faster than countdown".
Replace the beacon first, as the finalize work did before the clear was
moved ahead of it. csa_active is still cleared when replacing the
beacon fails.
Fixes: e487eaeb076a ("cfg80211/mac80211/ath6kl: acquire wdev lock outside ch_switch_notify")
Signed-off-by: Mehmet Fide <mehmet.fide@screeningeagle.com>
---
Notes:
Tested on rtw-next on an i.MX8MP board with an AR9271 as AP, ath9k_htc
with "wifi: ath9k_htc: do not fetch the beacon of a finished countdown"
v2 (link below), hostapd switching between channel 1 and 6, one sniffer
on each channel counting CSA beacons heard on the new channel. Before:
3 in 200 switches with a count of 5, and a print in the driver showed 3
fetches between the two steps. After: none in 200 switches with a count
of 5 and 80 with a count of 2. With v1 of that driver patch,
back-to-back switches and AP stops in and out of a countdown gave no
warning from mac80211. The rtw88 AP channel switch under review
(RTL8822BU, RTL8822CE) counts down on the air as before.
https://lore.kernel.org/r/20261117151703.1802556-1-mehmet.fide@gmail.com/
net/mac80211/cfg.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 9e47360ee8f5..9458838bfe63 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -4776,9 +4776,8 @@ static int __ieee80211_csa_finalize(struct ieee80211_link_data *link_data)
&link_data->csa.chanreq.oper))
return -EINVAL;
- link_conf->csa_active = false;
-
err = ieee80211_set_after_csa_beacon(link_data, &changed);
+ link_conf->csa_active = false;
if (err)
return err;
base-commit: 5dbf2415ca9821a611d09308e3b29d2c37e7cd77
--
2.56.0
reply other threads:[~2026-11-17 15:31 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20261117153154.1804945-1-mehmet.fide@gmail.com \
--to=mehmet.fide@gmail.com \
--cc=johannes@sipsolutions.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-wireless@vger.kernel.org \
--cc=mehmet.fide@screeningeagle.com \
--cc=sw@simonwunderlich.de \
--cc=toke@toke.dk \
/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