Merge tag 'net-next-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next

Pull networking updates from Jakub Kicinski:
 "Core:

   - support "prefer busy polling" NAPI operation mode, where we defer
     softirq for some time expecting applications to periodically busy
     poll

   - AF_XDP: improve efficiency by more batching and hindering the
     adjacency cache prefetcher

   - af_packet: make packet_fanout.arr size configurable up to 64K

   - tcp: optimize TCP zero copy receive in presence of partial or
     unaligned reads making zero copy a performance win for much smaller
     messages

   - XDP: add bulk APIs for returning / freeing frames

   - sched: support fragmenting IP packets as they come out of conntrack

   - net: allow virtual netdevs to forward UDP L4 and fraglist GSO skbs

  BPF:

   - BPF switch from crude rlimit-based to memcg-based memory accounting

   - BPF type format information for kernel modules and related tracing
     enhancements

   - BPF implement task local storage for BPF LSM

   - allow the FENTRY/FEXIT/RAW_TP tracing programs to use
     bpf_sk_storage

  Protocols:

   - mptcp: improve multiple xmit streams support, memory accounting and
     many smaller improvements

   - TLS: support CHACHA20-POLY1305 cipher

   - seg6: add support for SRv6 End.DT4/DT6 behavior

   - sctp: Implement RFC 6951: UDP Encapsulation of SCTP

   - ppp_generic: add ability to bridge channels directly

   - bridge: Connectivity Fault Management (CFM) support as is defined
     in IEEE 802.1Q section 12.14.

  Drivers:

   - mlx5: make use of the new auxiliary bus to organize the driver
     internals

   - mlx5: more accurate port TX timestamping support

   - mlxsw:
      - improve the efficiency of offloaded next hop updates by using
        the new nexthop object API
      - support blackhole nexthops
      - support IEEE 802.1ad (Q-in-Q) bridging

   - rtw88: major bluetooth co-existance improvements

   - iwlwifi: support new 6 GHz frequency band

   - ath11k: Fast Initial Link Setup (FILS)

   - mt7915: dual band concurrent (DBDC) support

   - net: ipa: add basic support for IPA v4.5

  Refactor:

   - a few pieces of in_interrupt() cleanup work from Sebastian Andrzej
     Siewior

   - phy: add support for shared interrupts; get rid of multiple driver
     APIs and have the drivers write a full IRQ handler, slight growth
     of driver code should be compensated by the simpler API which also
     allows shared IRQs

   - add common code for handling netdev per-cpu counters

   - move TX packet re-allocation from Ethernet switch tag drivers to a
     central place

   - improve efficiency and rename nla_strlcpy

   - number of W=1 warning cleanups as we now catch those in a patchwork
     build bot

  Old code removal:

   - wan: delete the DLCI / SDLA drivers

   - wimax: move to staging

   - wifi: remove old WDS wifi bridging support"

* tag 'net-next-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1922 commits)
  net: hns3: fix expression that is currently always true
  net: fix proc_fs init handling in af_packet and tls
  nfc: pn533: convert comma to semicolon
  af_vsock: Assign the vsock transport considering the vsock address flags
  af_vsock: Set VMADDR_FLAG_TO_HOST flag on the receive path
  vsock_addr: Check for supported flag values
  vm_sockets: Add VMADDR_FLAG_TO_HOST vsock flag
  vm_sockets: Add flags field in the vsock address data structure
  net: Disable NETIF_F_HW_TLS_TX when HW_CSUM is disabled
  tcp: Add logic to check for SYN w/ data in tcp_simple_retransmit
  net: mscc: ocelot: install MAC addresses in .ndo_set_rx_mode from process context
  nfc: s3fwrn5: Release the nfc firmware
  net: vxget: clean up sparse warnings
  mlxsw: spectrum_router: Use eXtended mezzanine to offload IPv4 router
  mlxsw: spectrum: Set KVH XLT cache mode for Spectrum2/3
  mlxsw: spectrum_router_xm: Introduce basic XM cache flushing
  mlxsw: reg: Add Router LPM Cache Enable Register
  mlxsw: reg: Add Router LPM Cache ML Delete Register
  mlxsw: spectrum_router_xm: Implement L-value tracking for M-index
  mlxsw: reg: Add XM Router M Table Register
  ...

1 files changed, 108 insertions, 88 deletions

diff --git a/drivers/net/wireless/mediatek/mt76/sdio.c b/drivers/net/wireless/mediatek/mt76/sdio.c
index 9a4d95a2a707..62b5b912818f 100644
--- a/drivers/net/wireless/mediatek/mt76/sdio.c
+++ b/drivers/net/wireless/mediatek/mt76/sdio.c
@@ -36,46 +36,49 @@ mt76s_alloc_rx_queue(struct mt76_dev *dev, enum mt76_rxq_id qid)
 	return 0;
 }
 
-static int mt76s_alloc_tx(struct mt76_dev *dev)
+static struct mt76_queue *mt76s_alloc_tx_queue(struct mt76_dev *dev)
 {
 	struct mt76_queue *q;
-	int i;
-
-	for (i = 0; i < MT_TXQ_MCU_WA; i++) {
-		q = devm_kzalloc(dev->dev, sizeof(*q), GFP_KERNEL);
-		if (!q)
-			return -ENOMEM;
 
-		spin_lock_init(&q->lock);
-		q->hw_idx = i;
-		dev->q_tx[i] = q;
+	q = devm_kzalloc(dev->dev, sizeof(*q), GFP_KERNEL);
+	if (!q)
+		return ERR_PTR(-ENOMEM);
 
-		q->entry = devm_kcalloc(dev->dev,
-					MT_NUM_TX_ENTRIES, sizeof(*q->entry),
-					GFP_KERNEL);
-		if (!q->entry)
-			return -ENOMEM;
+	spin_lock_init(&q->lock);
+	q->entry = devm_kcalloc(dev->dev,
+				MT_NUM_TX_ENTRIES, sizeof(*q->entry),
+				GFP_KERNEL);
+	if (!q->entry)
+		return ERR_PTR(-ENOMEM);
 
-		q->ndesc = MT_NUM_TX_ENTRIES;
-	}
+	q->ndesc = MT_NUM_TX_ENTRIES;
 
-	return 0;
+	return q;
 }
 
-void mt76s_stop_txrx(struct mt76_dev *dev)
+static int mt76s_alloc_tx(struct mt76_dev *dev)
 {
-	struct mt76_sdio *sdio = &dev->sdio;
+	struct mt76_queue *q;
+	int i;
 
-	cancel_work_sync(&sdio->tx.xmit_work);
-	cancel_work_sync(&sdio->tx.status_work);
-	cancel_work_sync(&sdio->rx.recv_work);
-	cancel_work_sync(&sdio->rx.net_work);
-	cancel_work_sync(&sdio->stat_work);
-	clear_bit(MT76_READING_STATS, &dev->phy.state);
+	for (i = 0; i <= MT_TXQ_PSD; i++) {
+		q = mt76s_alloc_tx_queue(dev);
+		if (IS_ERR(q))
+			return PTR_ERR(q);
 
-	mt76_tx_status_check(dev, NULL, true);
+		q->qid = i;
+		dev->phy.q_tx[i] = q;
+	}
+
+	q = mt76s_alloc_tx_queue(dev);
+	if (IS_ERR(q))
+		return PTR_ERR(q);
+
+	q->qid = MT_MCUQ_WM;
+	dev->q_mcu[MT_MCUQ_WM] = q;
+
+	return 0;
 }
-EXPORT_SYMBOL_GPL(mt76s_stop_txrx);
 
 int mt76s_alloc_queues(struct mt76_dev *dev)
 {
@@ -131,11 +134,32 @@ mt76s_process_rx_queue(struct mt76_dev *dev, struct mt76_queue *q)
 	return nframes;
 }
 
-static void mt76s_process_tx_queue(struct mt76_dev *dev, enum mt76_txq_id qid)
+static void mt76s_net_worker(struct mt76_worker *w)
+{
+	struct mt76_sdio *sdio = container_of(w, struct mt76_sdio,
+					      net_worker);
+	struct mt76_dev *dev = container_of(sdio, struct mt76_dev, sdio);
+	int i, nframes;
+
+	do {
+		nframes = 0;
+
+		local_bh_disable();
+		rcu_read_lock();
+
+		mt76_for_each_q_rx(dev, i)
+			nframes += mt76s_process_rx_queue(dev, &dev->q_rx[i]);
+
+		rcu_read_unlock();
+		local_bh_enable();
+	} while (nframes > 0);
+}
+
+static int mt76s_process_tx_queue(struct mt76_dev *dev, struct mt76_queue *q)
 {
-	struct mt76_queue *q = dev->q_tx[qid];
+	bool wake, mcu = q == dev->q_mcu[MT_MCUQ_WM];
 	struct mt76_queue_entry entry;
-	bool wake;
+	int nframes = 0;
 
 	while (q->queued > 0) {
 		if (!q->entry[q->tail].done)
@@ -144,12 +168,13 @@ static void mt76s_process_tx_queue(struct mt76_dev *dev, enum mt76_txq_id qid)
 		entry = q->entry[q->tail];
 		q->entry[q->tail].done = false;
 
-		if (qid == MT_TXQ_MCU) {
+		if (mcu) {
 			dev_kfree_skb(entry.skb);
 			entry.skb = NULL;
 		}
 
 		mt76_queue_tx_complete(dev, q, &entry);
+		nframes++;
 	}
 
 	wake = q->stopped && q->queued < q->ndesc - 8;
@@ -159,13 +184,35 @@ static void mt76s_process_tx_queue(struct mt76_dev *dev, enum mt76_txq_id qid)
 	if (!q->queued)
 		wake_up(&dev->tx_wait);
 
-	if (qid == MT_TXQ_MCU)
-		return;
+	if (mcu)
+		goto out;
 
-	mt76_txq_schedule(&dev->phy, qid);
+	mt76_txq_schedule(&dev->phy, q->qid);
 
 	if (wake)
-		ieee80211_wake_queue(dev->hw, qid);
+		ieee80211_wake_queue(dev->hw, q->qid);
+out:
+	return nframes;
+}
+
+static void mt76s_status_worker(struct mt76_worker *w)
+{
+	struct mt76_sdio *sdio = container_of(w, struct mt76_sdio,
+					      status_worker);
+	struct mt76_dev *dev = container_of(sdio, struct mt76_dev, sdio);
+	int i, nframes;
+
+	do {
+		nframes = mt76s_process_tx_queue(dev, dev->q_mcu[MT_MCUQ_WM]);
+
+		for (i = 0; i <= MT_TXQ_PSD; i++)
+			nframes += mt76s_process_tx_queue(dev,
+							  dev->phy.q_tx[i]);
+
+		if (dev->drv->tx_status_data &&
+		    !test_and_set_bit(MT76_READING_STATS, &dev->phy.state))
+			queue_work(dev->wq, &dev->sdio.stat_work);
+	} while (nframes > 0);
 }
 
 static void mt76s_tx_status_data(struct work_struct *work)
@@ -194,11 +241,10 @@ static void mt76s_tx_status_data(struct work_struct *work)
 }
 
 static int
-mt76s_tx_queue_skb(struct mt76_dev *dev, enum mt76_txq_id qid,
+mt76s_tx_queue_skb(struct mt76_dev *dev, struct mt76_queue *q,
 		   struct sk_buff *skb, struct mt76_wcid *wcid,
 		   struct ieee80211_sta *sta)
 {
-	struct mt76_queue *q = dev->q_tx[qid];
 	struct mt76_tx_info tx_info = {
 		.skb = skb,
 	};
@@ -209,7 +255,7 @@ mt76s_tx_queue_skb(struct mt76_dev *dev, enum mt76_txq_id qid,
 		return -ENOSPC;
 
 	skb->prev = skb->next = NULL;
-	err = dev->drv->tx_prepare_skb(dev, NULL, qid, wcid, sta, &tx_info);
+	err = dev->drv->tx_prepare_skb(dev, NULL, q->qid, wcid, sta, &tx_info);
 	if (err < 0)
 		return err;
 
@@ -222,10 +268,9 @@ mt76s_tx_queue_skb(struct mt76_dev *dev, enum mt76_txq_id qid,
 }
 
 static int
-mt76s_tx_queue_skb_raw(struct mt76_dev *dev, enum mt76_txq_id qid,
+mt76s_tx_queue_skb_raw(struct mt76_dev *dev, struct mt76_queue *q,
 		       struct sk_buff *skb, u32 tx_info)
 {
-	struct mt76_queue *q = dev->q_tx[qid];
 	int ret = -ENOSPC, len = skb->len, pad;
 
 	if (q->queued == q->ndesc)
@@ -257,7 +302,7 @@ static void mt76s_tx_kick(struct mt76_dev *dev, struct mt76_queue *q)
 {
 	struct mt76_sdio *sdio = &dev->sdio;
 
-	queue_work(sdio->txrx_wq, &sdio->tx.xmit_work);
+	mt76_worker_schedule(&sdio->txrx_worker);
 }
 
 static const struct mt76_queue_ops sdio_queue_ops = {
@@ -266,49 +311,19 @@ static const struct mt76_queue_ops sdio_queue_ops = {
 	.tx_queue_skb_raw = mt76s_tx_queue_skb_raw,
 };
 
-static void mt76s_tx_work(struct work_struct *work)
-{
-	struct mt76_sdio *sdio = container_of(work, struct mt76_sdio,
-					      tx.status_work);
-	struct mt76_dev *dev = container_of(sdio, struct mt76_dev, sdio);
-	int i;
-
-	for (i = 0; i < MT_TXQ_MCU_WA; i++)
-		mt76s_process_tx_queue(dev, i);
-
-	if (dev->drv->tx_status_data &&
-	    !test_and_set_bit(MT76_READING_STATS, &dev->phy.state))
-		queue_work(dev->wq, &dev->sdio.stat_work);
-}
-
-static void mt76s_rx_work(struct work_struct *work)
-{
-	struct mt76_sdio *sdio = container_of(work, struct mt76_sdio,
-					      rx.net_work);
-	struct mt76_dev *dev = container_of(sdio, struct mt76_dev, sdio);
-	int i;
-
-	/* rx processing */
-	local_bh_disable();
-	rcu_read_lock();
-
-	mt76_for_each_q_rx(dev, i)
-		mt76s_process_rx_queue(dev, &dev->q_rx[i]);
-
-	rcu_read_unlock();
-	local_bh_enable();
-}
-
 void mt76s_deinit(struct mt76_dev *dev)
 {
 	struct mt76_sdio *sdio = &dev->sdio;
 	int i;
 
-	mt76s_stop_txrx(dev);
-	if (sdio->txrx_wq) {
-		destroy_workqueue(sdio->txrx_wq);
-		sdio->txrx_wq = NULL;
-	}
+	mt76_worker_teardown(&sdio->txrx_worker);
+	mt76_worker_teardown(&sdio->status_worker);
+	mt76_worker_teardown(&sdio->net_worker);
+
+	cancel_work_sync(&sdio->stat_work);
+	clear_bit(MT76_READING_STATS, &dev->phy.state);
+
+	mt76_tx_status_check(dev, NULL, true);
 
 	sdio_claim_host(sdio->func);
 	sdio_release_irq(sdio->func);
@@ -335,18 +350,23 @@ int mt76s_init(struct mt76_dev *dev, struct sdio_func *func,
 	       const struct mt76_bus_ops *bus_ops)
 {
 	struct mt76_sdio *sdio = &dev->sdio;
+	int err;
 
-	sdio->txrx_wq = alloc_workqueue("mt76s_txrx_wq",
-					WQ_UNBOUND | WQ_HIGHPRI,
-					WQ_UNBOUND_MAX_ACTIVE);
-	if (!sdio->txrx_wq)
-		return -ENOMEM;
+	err = mt76_worker_setup(dev->hw, &sdio->status_worker,
+				mt76s_status_worker, "sdio-status");
+	if (err)
+		return err;
+
+	err = mt76_worker_setup(dev->hw, &sdio->net_worker, mt76s_net_worker,
+				"sdio-net");
+	if (err)
+		return err;
+
+	sched_set_fifo_low(sdio->status_worker.task);
+	sched_set_fifo_low(sdio->net_worker.task);
 
 	INIT_WORK(&sdio->stat_work, mt76s_tx_status_data);
-	INIT_WORK(&sdio->tx.status_work, mt76s_tx_work);
-	INIT_WORK(&sdio->rx.net_work, mt76s_rx_work);
 
-	mutex_init(&sdio->sched.lock);
 	dev->queue_ops = &sdio_queue_ops;
 	dev->bus = bus_ops;
 	dev->sdio.func = func;