Linux wireless drivers development
 help / color / mirror / Atom feed
From: Felix Fietkau <nbd@nbd.name>
To: linux-wireless@vger.kernel.org
Subject: [PATCH mt76-next 25/25] wifi: mt76: mt7996: keep a unicast management frame on its MLO link
Date: Wed, 28 Oct 2026 13:24:03 +0100	[thread overview]
Message-ID: <20261028122403.796394-25-nbd@nbd.name> (raw)
In-Reply-To: <20261028122403.796394-1-nbd@nbd.name>

mt7996_mac_write_txwi_80211() sets MT_TXD5_FL for an MLD interface only
on multicast, EAPOL and 4-address null frames. A unicast management
frame to an MLD peer goes out on a link that the hardware picks among
the links of the peer, not on the link of its TXD. ML reconfiguration
depends on that link: mac80211 removes a link before it sends the Link
Reconfiguration Request, so that the request leaves on a remaining
link, and the AP MLD sends the Response on the link where the request
was received. With hardware link selection, the request leaves on the
removed link, or the AP sends the Response there. The station then
drops the Response, and the link removal ends with a disconnection.

Set MT_TXD5_FL for unicast management frames to a station of an MLD
interface, so that they keep the link that mac80211 or the driver
selected.

Fixes: f0b0b239b8f3 ("wifi: mt76: mt7996: rework mt7996_mac_write_txwi() for MLO support")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
index a92536180dbe..03670293a78f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c
@@ -851,7 +851,8 @@ mt7996_mac_write_txwi_80211(struct mt7996_dev *dev, __le32 *txwi,
 	}
 
 	if (vif && ieee80211_vif_is_mld(vif) &&
-	    (multicast || unlikely(skb->protocol == cpu_to_be16(ETH_P_PAE))))
+	    (multicast || unlikely(skb->protocol == cpu_to_be16(ETH_P_PAE)) ||
+	     (ieee80211_is_mgmt(fc) && wcid->sta)))
 		txwi[5] |= cpu_to_le32(MT_TXD5_FL);
 
 	if (ieee80211_is_nullfunc(fc) && ieee80211_has_a4(fc) &&
-- 
2.47.3


      parent reply	other threads:[~2026-10-28 12:24 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-28 12:23 [PATCH mt76-next 01/25] wifi: mt76: mt7996: take the link of an 802.3 frame from its station Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 02/25] wifi: mt76: mt7996: do not update beacons while CSA is active Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 03/25] wifi: mt76: mt7996: only remove a BMC STA_REC that was added Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 04/25] wifi: mt76: mt7996: pass the puncturing bitmap to the firmware Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 05/25] wifi: mt76: abort a scan when the chanctx changes Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 06/25] wifi: mt76: set the channel again when the chanctx puncturing changes Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 07/25] wifi: mt76: mt7996: advertise preamble puncturing support Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 08/25] wifi: mt76: mt7996: keep scan and ROC links out of the MLD group Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 09/25] wifi: mt76: mt7996: factor the vif link teardown out of remove_interface Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 10/25] wifi: mt76: mt7996: destroy station links before the MLD group is freed Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 11/25] wifi: mt76: mt7996: send MLD link operations for an AP MLD only Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 12/25] wifi: mt76: mt7996: set up a vif link again on another radio Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 13/25] wifi: mt76: mt7996: describe an MLD peer with the PEER_MLD command Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 14/25] wifi: mt76: mt7915: honour the valid flag in efuse block reads Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 15/25] wifi: mt76: mt7996: skip MLD address translation for encap offload frames Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 16/25] wifi: mt76: mt7921: honour the valid flag in efuse block reads Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 17/25] wifi: mt76: mt7915: upload the CSA beacon only once Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 18/25] wifi: mt76: read the chanctx of a link once Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 19/25] wifi: mt76: restart the beacon monitor of a link that gets a chanctx again Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 20/25] wifi: mt76: publish the beacon monitor interval after its timestamp Felix Fietkau
2026-10-28 12:23 ` [PATCH mt76-next 21/25] wifi: mt76: mt7996: keep rx header translation off while a monitor runs Felix Fietkau
2026-10-28 12:24 ` [PATCH mt76-next 22/25] wifi: mt76: mt7996: apply monitor mode again after a full reset Felix Fietkau
2026-10-28 12:24 ` [PATCH mt76-next 23/25] wifi: mt76: check the RX PN of an MLD peer once per peer Felix Fietkau
2026-10-28 12:24 ` [PATCH mt76-next 24/25] wifi: mt76: mt7996: send no MLD_SETUP when station links only go away Felix Fietkau
2026-10-28 12:24 ` Felix Fietkau [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=20261028122403.796394-25-nbd@nbd.name \
    --to=nbd@nbd.name \
    --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