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 23/27] wifi: mt76: mt7996: set up block ack sessions on station links added later
Date: Tue,  6 Oct 2026 22:59:32 +0200	[thread overview]
Message-ID: <20261006205936.669125-23-nbd@nbd.name> (raw)
In-Reply-To: <20261006205936.669125-1-nbd@nbd.name>

mt7996_ampdu_action() sends STA_REC_BA to the links that the station
has when mac80211 starts a session. A link that is brought up later
gets no STA_REC_BA, so it transmits without aggregation and, as a
recipient, is not set up for hardware reordering. mac80211 does not
start the sessions again when a link is added.

Send STA_REC_BA for each running TX and RX session on the links that
mt7996_mac_sta_change_links() brings up. For an RX session, use the
current head of the reorder window, which mt76_rx_aggr_start() keeps
on the default link, as the starting sequence number.

That head moves only when the host reorders: with hardware RRO, mt7996
marks the frames as already aggregated and mt7996_mac_fill_rx() skips
the host reorder state; with WED or the NPU active,
mt76_rx_poll_complete() does not call mt76_rx_aggr_reorder(). The
driver has no MCU query for the current window. In those modes, stop
the RX sessions with ieee80211_stop_rx_ba_session() instead, so that
the originator sets up new sessions, which then reach all links through
the regular IEEE80211_AMPDU_RX_START path.

Fixes: ef3f5941e560 ("wifi: mt76: mt7996: rework mt7996_ampdu_action to support MLO")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt7996/main.c  | 81 ++++++++++++++++++-
 1 file changed, 80 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index 324d3b195a3f..89e9ef855a78 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -1614,6 +1614,85 @@ mt7996_mac_sta_links_set_keys(struct mt7996_dev *dev,
 	}
 }
 
+static bool
+mt7996_rx_reorder_host(struct mt7996_dev *dev)
+{
+	return !mt7996_has_hwrro(dev) &&
+	       !mtk_wed_device_active(&dev->mt76.mmio.wed) &&
+	       !mt76_npu_device_active(&dev->mt76);
+}
+
+static u16
+mt7996_mac_sta_ba_rx_ssn(struct mt7996_dev *dev, struct mt7996_sta *msta,
+			 unsigned int tid)
+{
+	struct mt76_rx_tid *tid_rx;
+	u16 ssn;
+
+	tid_rx = mt76_dereference(msta->deflink.wcid.aggr[tid], &dev->mt76);
+	if (!tid_rx)
+		return msta->ba_rx[tid].ssn;
+
+	spin_lock_bh(&tid_rx->lock);
+	ssn = tid_rx->head;
+	spin_unlock_bh(&tid_rx->lock);
+
+	return ssn;
+}
+
+static int
+mt7996_mac_sta_links_ba_add(struct mt7996_dev *dev, struct ieee80211_vif *vif,
+			    struct ieee80211_sta *sta, unsigned long links)
+{
+	struct mt7996_sta *msta = (struct mt7996_sta *)sta->drv_priv;
+	u16 rx_restart = 0;
+	unsigned int tid;
+	int err;
+
+	if (!links)
+		return 0;
+
+	for (tid = 0; tid < IEEE80211_NUM_TIDS; tid++) {
+		struct ieee80211_ampdu_params params = {
+			.sta = sta,
+			.tid = tid,
+		};
+
+		if (msta->ba_tx_mask & BIT(tid)) {
+			params.ssn = msta->ba_tx[tid].ssn;
+			params.buf_size = msta->ba_tx[tid].buf_size;
+			params.amsdu = msta->ba_tx[tid].amsdu;
+			err = mt7996_mcu_add_tx_ba(dev, &params, vif, links,
+						   true);
+			if (err)
+				return err;
+		}
+
+		if (!(msta->ba_rx_mask & BIT(tid)))
+			continue;
+
+		/* without host reordering, the reorder head stays at the
+		 * session start, so have the originator set up a new session
+		 */
+		if (!mt7996_rx_reorder_host(dev)) {
+			rx_restart |= BIT(tid);
+			continue;
+		}
+
+		params.ssn = mt7996_mac_sta_ba_rx_ssn(dev, msta, tid);
+		params.buf_size = msta->ba_rx[tid].buf_size;
+		params.amsdu = msta->ba_rx[tid].amsdu;
+		err = mt7996_mcu_add_rx_ba(dev, &params, vif, links, true);
+		if (err)
+			return err;
+	}
+
+	if (rx_restart)
+		ieee80211_stop_rx_ba_session(vif, rx_restart, sta->addr);
+
+	return 0;
+}
+
 static int
 mt7996_mac_sta_links_sync(struct mt7996_dev *dev, struct ieee80211_vif *vif,
 			  struct ieee80211_sta *sta, unsigned long links,
@@ -1672,7 +1751,7 @@ mt7996_mac_sta_links_sync(struct mt7996_dev *dev, struct ieee80211_vif *vif,
 			return err;
 	}
 
-	return 0;
+	return mt7996_mac_sta_links_ba_add(dev, vif, sta, new_links);
 }
 
 static void
-- 
2.47.3


  parent reply	other threads:[~2026-10-06 20:59 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-06 20:59 [PATCH mt76-next 01/27] wifi: mt76: abort a scan only for a link change on the scanning radio Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 02/27] wifi: mt76: mt7996: enable the SKU power limit table Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 03/27] wifi: mt76: mt7996: set sniffer mode on a radio that is already running Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 04/27] wifi: mt76: mt7996: update the address of an AP link that is added again Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 05/27] wifi: mt76: mt7996: size the MLD setup TLV by the links it describes Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 06/27] wifi: mt76: mt7996: find the last MLD link among the links torn down Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 07/27] wifi: mt76: mt7996: upload only the keys of the link being set up Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 08/27] wifi: mt76: mt7996: do not walk the vif keys when a link is destroyed Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 09/27] wifi: mt76: mt7996: report station statistics from the primary link Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 10/27] wifi: mt76: mt7996: clear all link wcid entries in sta_pre_rcu_remove Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 11/27] wifi: mt76: mt7996: free the WTBL index if a link allocation fails Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 12/27] wifi: mt76: mt7996: pick the secondary link from links with driver state Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 13/27] wifi: mt76: mt7996: make a returning link primary if a station has none Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 14/27] wifi: mt76: mt7996: do not reuse the deflink of a removed station link Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 15/27] wifi: mt76: mt7996: keep a returning link when adding station links fails Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 16/27] wifi: mt76: mt7996: add the station BMC entry as new once per BSSID Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 17/27] wifi: mt76: mt7996: keep 4-address and decap state per station Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 18/27] wifi: mt76: mt7996: do not add a station link in disconnected state Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 19/27] wifi: mt76: mt7996: record the connection state of a station Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 20/27] wifi: mt76: mt7996: keep the block ack session parameters per station Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 21/27] wifi: mt76: mt7996: set up the BSS of a late station link before its peer Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 22/27] wifi: mt76: mt7996: bring up station links added after association Felix Fietkau
2026-10-06 20:59 ` Felix Fietkau [this message]
2026-10-06 20:59 ` [PATCH mt76-next 24/27] wifi: mt76: mt7996: restore a station link that returns after a removal Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 25/27] wifi: mt76: mt7996: fix PLE free page and AC queue register offsets Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 26/27] wifi: mt76: mt7996: fix the hw-queues debugfs queue requests Felix Fietkau
2026-10-06 20:59 ` [PATCH mt76-next 27/27] wifi: mt76: mt7996: disable the RX NAPIs before the DMA cleanup Felix Fietkau

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=20261006205936.669125-23-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