Skip to contentSkip to conversation

net: switch drivers to netif_napi_add_weight()

PATCH Applied Jakub Kicinski 7 participants 16 messages May 6, 2022 – May 12, 2022

Subsystems (30)
lore
Share & tools
lore
author maintainer
#14 lore raw

PATCH 6/6 net: wan: switch to netif_napi_add_weight()

In reply to Jakub Kicinski

A handful of WAN drivers use custom napi weights,
switch them to the new API.

---
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 drivers/net/wan/fsl_ucc_hdlc.c | 2 +-
 drivers/net/wan/hd64572.c      | 3 ++-
 drivers/net/wan/ixp4xx_hss.c   | 2 +-
 drivers/net/wan/lapbether.c    | 2 +-
 4 files changed, 5 insertions(+), 4 deletions(-)
drivers/net/wan/fsl_ucc_hdlc.c +1 -1
index 5ae2d27b5da9..22edea6ca4b8 100644
@@ -1231,7 +1231,7 @@ static int ucc_hdlc_probe(struct platform_device *pdev)
 	dev->watchdog_timeo = 2 * HZ;
 	hdlc->attach = ucc_hdlc_attach;
 	hdlc->xmit = ucc_hdlc_tx;
-	netif_napi_add(dev, &uhdlc_priv->napi, ucc_hdlc_poll, 32);
+	netif_napi_add_weight(dev, &uhdlc_priv->napi, ucc_hdlc_poll, 32);
 	if (register_hdlc_device(dev)) {
 		ret = -ENOBUFS;
 		pr_err("ucc_hdlc: unable to register hdlc device\n");
drivers/net/wan/hd64572.c +2 -1
index b89b03a6aba7..534369ffe5de 100644
@@ -173,7 +173,8 @@ static void sca_init_port(port_t *port)
 	sca_out(DIR_EOME, DIR_TX(port->chan), card); /* enable interrupts */
 
 	sca_set_carrier(port);
-	netif_napi_add(port->netdev, &port->napi, sca_poll, NAPI_WEIGHT);
+	netif_napi_add_weight(port->netdev, &port->napi, sca_poll,
+			      NAPI_WEIGHT);
 }
 
 /* MSCI interrupt service */
drivers/net/wan/ixp4xx_hss.c +1 -1
index 863c3e34e136..e46b7f5ee49e 100644
@@ -1504,7 +1504,7 @@ static int ixp4xx_hss_probe(struct platform_device *pdev)
 	port->clock_reg = CLK42X_SPEED_2048KHZ;
 	port->id = pdev->id;
 	port->dev = &pdev->dev;
-	netif_napi_add(ndev, &port->napi, hss_hdlc_poll, NAPI_WEIGHT);
+	netif_napi_add_weight(ndev, &port->napi, hss_hdlc_poll, NAPI_WEIGHT);
 
 	err = register_hdlc_device(ndev);
 	if (err)
drivers/net/wan/lapbether.c +1 -1
index 282192b82404..960f1393595c 100644
@@ -408,7 +408,7 @@ static int lapbeth_new_device(struct net_device *dev)
 	spin_lock_init(&lapbeth->up_lock);
 
 	skb_queue_head_init(&lapbeth->rx_queue);
-	netif_napi_add(ndev, &lapbeth->napi, lapbeth_napi_poll, 16);
+	netif_napi_add_weight(ndev, &lapbeth->napi, lapbeth_napi_poll, 16);
 
 	rc = -EIO;
 	if (register_netdevice(ndev))
Signature · 1 line
2.34.1
author maintainer
#1 lore raw

PATCH 1/6 um: vector: switch to netif_napi_add_weight()

In reply to Jakub Kicinski

UM's netdev driver uses a custom napi weight, switch to the new API for setting custom weight.

---
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 arch/um/drivers/vector_kern.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/um/drivers/vector_kern.c +2 -1
index 1d6f6a66766c..548265312743 100644
@@ -1255,7 +1255,8 @@ static int vector_net_open(struct net_device *dev)
 			goto out_close;
 	}
 
-	netif_napi_add(vp->dev, &vp->napi, vector_poll, get_depth(vp->parsed));
+	netif_napi_add_weight(vp->dev, &vp->napi, vector_poll,
+			      get_depth(vp->parsed));
 	napi_enable(&vp->napi);
 
 	/* READ IRQ */
Signature · 7 lines
2.34.1


_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um

Jakub Kicinski

author maintainer
#0 lore raw

cover net: switch drivers to netif_napi_add_weight()

The minority of drivers pass a custom weight to netif_napi_add(). Switch those away to the new netif_napi_add_weight(). All drivers (which can go thru net-next) calling netif_napi_add() will now be calling it with NAPI_POLL_WEIGHT or 64.

Jakub Kicinski (6):
um: vector: switch to netif_napi_add_weight()
caif_virtio: switch to netif_napi_add_weight()
eth: switch to netif_napi_add_weight()
r8152: switch to netif_napi_add_weight()
net: virtio: switch to netif_napi_add_weight()
net: wan: switch to netif_napi_add_weight()

 arch/um/drivers/vector_kern.c                         | 3 ++-
 drivers/net/caif/caif_virtio.c                        | 3 ++-
 drivers/net/ethernet/3com/typhoon.c                   | 2 +-
 drivers/net/ethernet/adaptec/starfire.c               | 2 +-
 drivers/net/ethernet/amd/amd8111e.c                   | 2 +-
 drivers/net/ethernet/amd/pcnet32.c                    | 3 ++-
 drivers/net/ethernet/arc/emac_main.c                  | 3 ++-
 drivers/net/ethernet/atheros/ag71xx.c                 | 3 ++-
 drivers/net/ethernet/broadcom/bcm63xx_enet.c          | 4 ++--
 drivers/net/ethernet/broadcom/sb1250-mac.c            | 2 +-
 drivers/net/ethernet/dec/tulip/tulip_core.c           | 2 +-
 drivers/net/ethernet/ezchip/nps_enet.c                | 4 ++--
 drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c | 3 ++-
 drivers/net/ethernet/hisilicon/hisi_femac.c           | 3 ++-
 drivers/net/ethernet/huawei/hinic/hinic_rx.c          | 3 ++-
 drivers/net/ethernet/huawei/hinic/hinic_tx.c          | 3 ++-
 drivers/net/ethernet/ibm/emac/mal.c                   | 4 ++--
 drivers/net/ethernet/ibm/ibmveth.c                    | 2 +-
 drivers/net/ethernet/intel/e100.c                     | 2 +-
 drivers/net/ethernet/lantiq_etop.c                    | 8 ++++----
 drivers/net/ethernet/marvell/pxa168_eth.c             | 3 ++-
 drivers/net/ethernet/microchip/sparx5/sparx5_fdma.c   | 3 ++-
 drivers/net/ethernet/microsoft/mana/mana_en.c         | 2 +-
 drivers/net/ethernet/moxa/moxart_ether.c              | 2 +-
 drivers/net/ethernet/mscc/ocelot_fdma.c               | 4 ++--
 drivers/net/ethernet/myricom/myri10ge/myri10ge.c      | 4 ++--
 drivers/net/ethernet/neterion/vxge/vxge-main.c        | 9 +++++----
 drivers/net/ethernet/nxp/lpc_eth.c                    | 2 +-
 drivers/net/ethernet/realtek/8139cp.c                 | 2 +-
 drivers/net/ethernet/sfc/efx_channels.c               | 4 ++--
 drivers/net/ethernet/sfc/falcon/efx.c                 | 4 ++--
 drivers/net/ethernet/smsc/smsc911x.c                  | 3 ++-
 drivers/net/ethernet/toshiba/tc35815.c                | 2 +-
 drivers/net/ethernet/wiznet/w5100.c                   | 2 +-
 drivers/net/ethernet/wiznet/w5300.c                   | 2 +-
 drivers/net/ethernet/xscale/ixp4xx_eth.c              | 2 +-
 drivers/net/usb/r8152.c                               | 6 ++----
 drivers/net/virtio_net.c                              | 4 ++--
 drivers/net/wan/fsl_ucc_hdlc.c                        | 2 +-
 drivers/net/wan/hd64572.c                             | 3 ++-
 drivers/net/wan/ixp4xx_hss.c                          | 2 +-
 drivers/net/wan/lapbether.c                           | 2 +-
 42 files changed, 71 insertions(+), 59 deletions(-)
Signature · 1 line
2.34.1
author maintainer
#9 lore raw

PATCH 4/6 r8152: switch to netif_napi_add_weight()

r8152 uses a custom napi weight, switch to the new API for setting custom weight.

---
CC: [email protected]
CC: [email protected]

---
 drivers/net/usb/r8152.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
drivers/net/usb/r8152.c +2 -4
index d3b53db57c26..c2da3438387c 100644
@@ -9732,10 +9732,8 @@ static int rtl8152_probe(struct usb_interface *intf,
 
 	usb_set_intfdata(intf, tp);
 
-	if (tp->support_2500full)
-		netif_napi_add(netdev, &tp->napi, r8152_poll, 256);
-	else
-		netif_napi_add(netdev, &tp->napi, r8152_poll, 64);
+	netif_napi_add_weight(netdev, &tp->napi, r8152_poll,
+			      tp->support_2500full ? 256 : 64);
 
 	ret = register_netdev(netdev);
 	if (ret != 0) {
Signature · 1 line
2.34.1
author maintainer
#2 lore raw

PATCH 2/6 caif_virtio: switch to netif_napi_add_weight()

In reply to Jakub Kicinski

caif_virtio uses a custom napi weight, switch to the new API for setting custom weights.

---
CC: [email protected]
CC: [email protected]

---
 drivers/net/caif/caif_virtio.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/caif/caif_virtio.c +2 -1
index 444ef6a342f6..5458f57177a0 100644
@@ -714,7 +714,8 @@ static int cfv_probe(struct virtio_device *vdev)
 	/* Initialize NAPI poll context data */
 	vringh_kiov_init(&cfv->ctx.riov, NULL, 0);
 	cfv->ctx.head = USHRT_MAX;
-	netif_napi_add(netdev, &cfv->napi, cfv_rx_poll, CFV_DEFAULT_QUOTA);
+	netif_napi_add_weight(netdev, &cfv->napi, cfv_rx_poll,
+			      CFV_DEFAULT_QUOTA);
 
 	tasklet_setup(&cfv->tx_release_tasklet, cfv_tx_release_tasklet);
 
Signature · 1 line
2.34.1
author maintainer
#4 lore raw

PATCH 3/6 eth: switch to netif_napi_add_weight()

In reply to Jakub Kicinski

Switch all Ethernet drivers which use custom napi weights to the new API.

---
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 drivers/net/ethernet/3com/typhoon.c                   | 2 +-
 drivers/net/ethernet/adaptec/starfire.c               | 2 +-
 drivers/net/ethernet/amd/amd8111e.c                   | 2 +-
 drivers/net/ethernet/amd/pcnet32.c                    | 3 ++-
 drivers/net/ethernet/arc/emac_main.c                  | 3 ++-
 drivers/net/ethernet/atheros/ag71xx.c                 | 3 ++-
 drivers/net/ethernet/broadcom/bcm63xx_enet.c          | 4 ++--
 drivers/net/ethernet/broadcom/sb1250-mac.c            | 2 +-
 drivers/net/ethernet/dec/tulip/tulip_core.c           | 2 +-
 drivers/net/ethernet/ezchip/nps_enet.c                | 4 ++--
 drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c | 3 ++-
 drivers/net/ethernet/hisilicon/hisi_femac.c           | 3 ++-
 drivers/net/ethernet/huawei/hinic/hinic_rx.c          | 3 ++-
 drivers/net/ethernet/huawei/hinic/hinic_tx.c          | 3 ++-
 drivers/net/ethernet/ibm/emac/mal.c                   | 4 ++--
 drivers/net/ethernet/ibm/ibmveth.c                    | 2 +-
 drivers/net/ethernet/intel/e100.c                     | 2 +-
 drivers/net/ethernet/lantiq_etop.c                    | 8 ++++----
 drivers/net/ethernet/marvell/pxa168_eth.c             | 3 ++-
 drivers/net/ethernet/microchip/sparx5/sparx5_fdma.c   | 3 ++-
 drivers/net/ethernet/microsoft/mana/mana_en.c         | 2 +-
 drivers/net/ethernet/moxa/moxart_ether.c              | 2 +-
 drivers/net/ethernet/mscc/ocelot_fdma.c               | 4 ++--
 drivers/net/ethernet/myricom/myri10ge/myri10ge.c      | 4 ++--
 drivers/net/ethernet/neterion/vxge/vxge-main.c        | 9 +++++----
 drivers/net/ethernet/nxp/lpc_eth.c                    | 2 +-
 drivers/net/ethernet/realtek/8139cp.c                 | 2 +-
 drivers/net/ethernet/sfc/efx_channels.c               | 4 ++--
 drivers/net/ethernet/sfc/falcon/efx.c                 | 4 ++--
 drivers/net/ethernet/smsc/smsc911x.c                  | 3 ++-
 drivers/net/ethernet/toshiba/tc35815.c                | 2 +-
 drivers/net/ethernet/wiznet/w5100.c                   | 2 +-
 drivers/net/ethernet/wiznet/w5300.c                   | 2 +-
 drivers/net/ethernet/xscale/ixp4xx_eth.c              | 2 +-
 34 files changed, 58 insertions(+), 47 deletions(-)
drivers/net/ethernet/3com/typhoon.c +1 -1
index ad57209007e1..cad4f354cc76 100644
@@ -2464,7 +2464,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* The chip-specific entries in the device structure. */
 	dev->netdev_ops		= &typhoon_netdev_ops;
-	netif_napi_add(dev, &tp->napi, typhoon_poll, 16);
+	netif_napi_add_weight(dev, &tp->napi, typhoon_poll, 16);
 	dev->watchdog_timeo	= TX_TIMEOUT;
 
 	dev->ethtool_ops = &typhoon_ethtool_ops;
drivers/net/ethernet/adaptec/starfire.c +1 -1
index c6982f7caf9b..8f0a6b9c518e 100644
@@ -772,7 +772,7 @@ static int starfire_init_one(struct pci_dev *pdev,
 	dev->watchdog_timeo = TX_TIMEOUT;
 	dev->ethtool_ops = &ethtool_ops;
 
-	netif_napi_add(dev, &np->napi, netdev_poll, max_interrupt_work);
+	netif_napi_add_weight(dev, &np->napi, netdev_poll, max_interrupt_work);
 
 	if (mtu)
 		dev->mtu = mtu;
drivers/net/ethernet/amd/amd8111e.c +1 -1
index 9421afb950f7..05ac8d9ccb2f 100644
@@ -1828,7 +1828,7 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 	dev->watchdog_timeo = AMD8111E_TX_TIMEOUT;
 	dev->min_mtu = AMD8111E_MIN_MTU;
 	dev->max_mtu = AMD8111E_MAX_MTU;
-	netif_napi_add(dev, &lp->napi, amd8111e_rx_poll, 32);
+	netif_napi_add_weight(dev, &lp->napi, amd8111e_rx_poll, 32);
 
 #if AMD8111E_VLAN_TAG_USED
 	dev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
drivers/net/ethernet/amd/pcnet32.c +2 -1
index c20c369c7eb8..b5ff47283cfe 100644
@@ -1881,7 +1881,8 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev)
 	/* napi.weight is used in both the napi and non-napi cases */
 	lp->napi.weight = lp->rx_ring_size / 2;
 
-	netif_napi_add(dev, &lp->napi, pcnet32_poll, lp->rx_ring_size / 2);
+	netif_napi_add_weight(dev, &lp->napi, pcnet32_poll,
+			      lp->rx_ring_size / 2);
 
 	if (fdx && !(lp->options & PCNET32_PORT_ASEL) &&
 	    ((cards_found >= MAX_UNITS) || full_duplex[cards_found]))
drivers/net/ethernet/arc/emac_main.c +2 -1
index c642c3d3e600..288e2961823e 100644
@@ -981,7 +981,8 @@ int arc_emac_probe(struct net_device *ndev, int interface)
 	dev_info(dev, "connected to %s phy with id 0x%x\n",
 		 phydev->drv->name, phydev->phy_id);
 
-	netif_napi_add(ndev, &priv->napi, arc_emac_poll, ARC_EMAC_NAPI_WEIGHT);
+	netif_napi_add_weight(ndev, &priv->napi, arc_emac_poll,
+			      ARC_EMAC_NAPI_WEIGHT);
 
 	err = register_netdev(ndev);
 	if (err) {
drivers/net/ethernet/atheros/ag71xx.c +2 -1
index ec167af0e3b2..cac509708e9d 100644
@@ -1922,7 +1922,8 @@ static int ag71xx_probe(struct platform_device *pdev)
 		return err;
 	}
 
-	netif_napi_add(ndev, &ag->napi, ag71xx_poll, AG71XX_NAPI_WEIGHT);
+	netif_napi_add_weight(ndev, &ag->napi, ag71xx_poll,
+			      AG71XX_NAPI_WEIGHT);
 
 	err = clk_prepare_enable(ag->clk_eth);
 	if (err) {
drivers/net/ethernet/broadcom/bcm63xx_enet.c +2 -2
index c1b97e8c55ef..698438a2ee0f 100644
@@ -1859,7 +1859,7 @@ static int bcm_enet_probe(struct platform_device *pdev)
 
 	/* register netdevice */
 	dev->netdev_ops = &bcm_enet_ops;
-	netif_napi_add(dev, &priv->napi, bcm_enet_poll, 16);
+	netif_napi_add_weight(dev, &priv->napi, bcm_enet_poll, 16);
 
 	dev->ethtool_ops = &bcm_enet_ethtool_ops;
 	/* MTU range: 46 - 2028 */
@@ -2714,7 +2714,7 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 
 	/* register netdevice */
 	dev->netdev_ops = &bcm_enetsw_ops;
-	netif_napi_add(dev, &priv->napi, bcm_enet_poll, 16);
+	netif_napi_add_weight(dev, &priv->napi, bcm_enet_poll, 16);
 	dev->ethtool_ops = &bcm_enetsw_ethtool_ops;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
drivers/net/ethernet/broadcom/sb1250-mac.c +1 -1
index 5d5f10180158..f02facb60fd1 100644
@@ -2203,7 +2203,7 @@ static int sbmac_init(struct platform_device *pldev, long long base)
 	dev->min_mtu = 0;
 	dev->max_mtu = ENET_PACKET_SIZE;
 
-	netif_napi_add(dev, &sc->napi, sbmac_poll, 16);
+	netif_napi_add_weight(dev, &sc->napi, sbmac_poll, 16);
 
 	dev->irq		= UNIT_INT(idx);
 
drivers/net/ethernet/dec/tulip/tulip_core.c +1 -1
index 79df5a72877b..434d8bf0e8f9 100644
@@ -1689,7 +1689,7 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	dev->netdev_ops = &tulip_netdev_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 #ifdef CONFIG_TULIP_NAPI
-	netif_napi_add(dev, &tp->napi, tulip_poll, 16);
+	netif_napi_add_weight(dev, &tp->napi, tulip_poll, 16);
 #endif
 	dev->ethtool_ops = &ops;
 
drivers/net/ethernet/ezchip/nps_enet.c +2 -2
index 69dbf950d451..f1eb660aaee2 100644
@@ -612,8 +612,8 @@ static s32 nps_enet_probe(struct platform_device *pdev)
 		goto out_netdev;
 	}
 
-	netif_napi_add(ndev, &priv->napi, nps_enet_poll,
-		       NPS_ENET_NAPI_POLL_WEIGHT);
+	netif_napi_add_weight(ndev, &priv->napi, nps_enet_poll,
+			      NPS_ENET_NAPI_POLL_WEIGHT);
 
 	/* Register the driver. Should be the last thing in probe */
 	err = register_netdev(ndev);
drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c +2 -1
index bacf25318f87..b3dae17e067e 100644
@@ -1020,7 +1020,8 @@ static int fs_enet_probe(struct platform_device *ofdev)
 	ndev->netdev_ops = &fs_enet_netdev_ops;
 	ndev->watchdog_timeo = 2 * HZ;
 	INIT_WORK(&fep->timeout_work, fs_timeout_work);
-	netif_napi_add(ndev, &fep->napi, fs_enet_napi, fpi->napi_weight);
+	netif_napi_add_weight(ndev, &fep->napi, fs_enet_napi,
+			      fpi->napi_weight);
 
 	ndev->ethtool_ops = &fs_ethtool_ops;
 
drivers/net/ethernet/hisilicon/hisi_femac.c +2 -1
index a6c18b6527f9..93846bace028 100644
@@ -852,7 +852,8 @@ static int hisi_femac_drv_probe(struct platform_device *pdev)
 	ndev->priv_flags |= IFF_UNICAST_FLT;
 	ndev->netdev_ops = &hisi_femac_netdev_ops;
 	ndev->ethtool_ops = &hisi_femac_ethtools_ops;
-	netif_napi_add(ndev, &priv->napi, hisi_femac_poll, FEMAC_POLL_WEIGHT);
+	netif_napi_add_weight(ndev, &priv->napi, hisi_femac_poll,
+			      FEMAC_POLL_WEIGHT);
 
 	hisi_femac_port_init(priv);
 
drivers/net/ethernet/huawei/hinic/hinic_rx.c +2 -1
index b33ed4d92b71..24b7b819dbfb 100644
@@ -481,7 +481,8 @@ static void rx_add_napi(struct hinic_rxq *rxq)
 {
 	struct hinic_dev *nic_dev = netdev_priv(rxq->netdev);
 
-	netif_napi_add(rxq->netdev, &rxq->napi, rx_poll, nic_dev->rx_weight);
+	netif_napi_add_weight(rxq->netdev, &rxq->napi, rx_poll,
+			      nic_dev->rx_weight);
 	napi_enable(&rxq->napi);
 }
 
drivers/net/ethernet/huawei/hinic/hinic_tx.c +2 -1
index 8d59babbf476..87408e7bb809 100644
@@ -809,7 +809,8 @@ static int tx_request_irq(struct hinic_txq *txq)
 
 	qp = container_of(sq, struct hinic_qp, sq);
 
-	netif_napi_add(txq->netdev, &txq->napi, free_tx_poll, nic_dev->tx_weight);
+	netif_napi_add_weight(txq->netdev, &txq->napi, free_tx_poll,
+			      nic_dev->tx_weight);
 
 	hinic_hwdev_msix_set(nic_dev->hwdev, sq->msix_entry,
 			     TX_IRQ_NO_PENDING, TX_IRQ_NO_COALESC,
drivers/net/ethernet/ibm/emac/mal.c +2 -2
index 075c07303f16..ff5487bbebe3 100644
@@ -605,8 +605,8 @@ static int mal_probe(struct platform_device *ofdev)
 
 	init_dummy_netdev(&mal->dummy_dev);
 
-	netif_napi_add(&mal->dummy_dev, &mal->napi, mal_poll,
-		       CONFIG_IBM_EMAC_POLL_WEIGHT);
+	netif_napi_add_weight(&mal->dummy_dev, &mal->napi, mal_poll,
+			      CONFIG_IBM_EMAC_POLL_WEIGHT);
 
 	/* Load power-on reset defaults */
 	mal_reset(mal);
drivers/net/ethernet/ibm/ibmveth.c +1 -1
index 22fb0d109a68..5c6a04d29f5b 100644
@@ -1674,7 +1674,7 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 	adapter->pool_config = 0;
 	ibmveth_init_link_settings(netdev);
 
-	netif_napi_add(netdev, &adapter->napi, ibmveth_poll, 16);
+	netif_napi_add_weight(netdev, &adapter->napi, ibmveth_poll, 16);
 
 	netdev->irq = dev->irq;
 	netdev->netdev_ops = &ibmveth_netdev_ops;
drivers/net/ethernet/intel/e100.c +1 -1
index 4a8013f20152..36418b510dde 100644
@@ -2848,7 +2848,7 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
 
 	nic = netdev_priv(netdev);
-	netif_napi_add(netdev, &nic->napi, e100_poll, E100_NAPI_WEIGHT);
+	netif_napi_add_weight(netdev, &nic->napi, e100_poll, E100_NAPI_WEIGHT);
 	nic->netdev = netdev;
 	nic->pdev = pdev;
 	nic->msg_enable = (1 << debug) - 1;
drivers/net/ethernet/lantiq_etop.c +4 -4
index 9b6fa27b7daf..7cedbe1fdfd7 100644
@@ -701,11 +701,11 @@ ltq_etop_probe(struct platform_device *pdev)
 
 	for (i = 0; i < MAX_DMA_CHAN; i++) {
 		if (IS_TX(i))
-			netif_napi_add(dev, &priv->ch[i].napi,
-				       ltq_etop_poll_tx, 8);
+			netif_napi_add_weight(dev, &priv->ch[i].napi,
+					      ltq_etop_poll_tx, 8);
 		else if (IS_RX(i))
-			netif_napi_add(dev, &priv->ch[i].napi,
-				       ltq_etop_poll_rx, 32);
+			netif_napi_add_weight(dev, &priv->ch[i].napi,
+					      ltq_etop_poll_rx, 32);
 		priv->ch[i].netdev = dev;
 	}
 
drivers/net/ethernet/marvell/pxa168_eth.c +2 -1
index 52bef50f5a0d..349b8a94e939 100644
@@ -1486,7 +1486,8 @@ static int pxa168_eth_probe(struct platform_device *pdev)
 
 	/* Hardware supports only 3 ports */
 	BUG_ON(pep->port_num > 2);
-	netif_napi_add(dev, &pep->napi, pxa168_rx_poll, pep->rx_ring_size);
+	netif_napi_add_weight(dev, &pep->napi, pxa168_rx_poll,
+			      pep->rx_ring_size);
 
 	memset(&pep->timeout, 0, sizeof(struct timer_list));
 	timer_setup(&pep->timeout, rxq_refill_timer_wrapper, 0);
drivers/net/ethernet/microchip/sparx5/sparx5_fdma.c +2 -1
index 1e9ff365459e..66360c8c5a38 100644
@@ -381,7 +381,8 @@ static int sparx5_fdma_rx_alloc(struct sparx5 *sparx5)
 		}
 		sparx5_fdma_rx_add_dcb(rx, dcb, rx->dma + sizeof(*dcb) * idx);
 	}
-	netif_napi_add(rx->ndev, &rx->napi, sparx5_fdma_napi_callback, FDMA_WEIGHT);
+	netif_napi_add_weight(rx->ndev, &rx->napi, sparx5_fdma_napi_callback,
+			      FDMA_WEIGHT);
 	napi_enable(&rx->napi);
 	sparx5_fdma_rx_activate(sparx5, rx);
 	return 0;
drivers/net/ethernet/microsoft/mana/mana_en.c +1 -1
index 06f853c5c141..b1d773823232 100644
@@ -1602,7 +1602,7 @@ static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 
 	gc->cq_table[cq->gdma_id] = cq->gdma_cq;
 
-	netif_napi_add(ndev, &cq->napi, mana_poll, 1);
+	netif_napi_add_weight(ndev, &cq->napi, mana_poll, 1);
 
 	WARN_ON(xdp_rxq_info_reg(&rxq->xdp_rxq, ndev, rxq_idx,
 				 cq->napi.napi_id));
drivers/net/ethernet/moxa/moxart_ether.c +1 -1
index afb7dcadb8d2..a3214a762e4b 100644
@@ -533,7 +533,7 @@ static int moxart_mac_probe(struct platform_device *pdev)
 	}
 
 	ndev->netdev_ops = &moxart_netdev_ops;
-	netif_napi_add(ndev, &priv->napi, moxart_rx_poll, RX_DESC_NUM);
+	netif_napi_add_weight(ndev, &priv->napi, moxart_rx_poll, RX_DESC_NUM);
 	ndev->priv_flags |= IFF_UNICAST_FLT;
 	ndev->irq = irq;
 
drivers/net/ethernet/mscc/ocelot_fdma.c +2 -2
index dffa597bffe6..083fddd263ec 100644
@@ -799,8 +799,8 @@ void ocelot_fdma_netdev_init(struct ocelot *ocelot, struct net_device *dev)
 		return;
 
 	fdma->ndev = dev;
-	netif_napi_add(dev, &fdma->napi, ocelot_fdma_napi_poll,
-		       OCELOT_FDMA_WEIGHT);
+	netif_napi_add_weight(dev, &fdma->napi, ocelot_fdma_napi_poll,
+			      OCELOT_FDMA_WEIGHT);
 }
 
 void ocelot_fdma_netdev_deinit(struct ocelot *ocelot, struct net_device *dev)
drivers/net/ethernet/myricom/myri10ge/myri10ge.c +2 -2
index fe5e77330f5f..61497c3e4cfb 100644
@@ -3586,8 +3586,8 @@ static int myri10ge_alloc_slices(struct myri10ge_priv *mgp)
 			goto abort;
 		ss->mgp = mgp;
 		ss->dev = mgp->dev;
-		netif_napi_add(ss->dev, &ss->napi, myri10ge_poll,
-			       myri10ge_napi_weight);
+		netif_napi_add_weight(ss->dev, &ss->napi, myri10ge_poll,
+				      myri10ge_napi_weight);
 	}
 	return 0;
 abort:
drivers/net/ethernet/neterion/vxge/vxge-main.c +5 -4
index db4dfae8c01d..d2de8ac44f72 100644
@@ -2720,8 +2720,8 @@ static int vxge_open(struct net_device *dev)
 	}
 
 	if (vdev->config.intr_type != MSI_X) {
-		netif_napi_add(dev, &vdev->napi, vxge_poll_inta,
-			vdev->config.napi_weight);
+		netif_napi_add_weight(dev, &vdev->napi, vxge_poll_inta,
+				      vdev->config.napi_weight);
 		napi_enable(&vdev->napi);
 		for (i = 0; i < vdev->no_of_vpath; i++) {
 			vpath = &vdev->vpaths[i];
@@ -2730,8 +2730,9 @@ static int vxge_open(struct net_device *dev)
 	} else {
 		for (i = 0; i < vdev->no_of_vpath; i++) {
 			vpath = &vdev->vpaths[i];
-			netif_napi_add(dev, &vpath->ring.napi,
-			    vxge_poll_msix, vdev->config.napi_weight);
+			netif_napi_add_weight(dev, &vpath->ring.napi,
+					      vxge_poll_msix,
+					      vdev->config.napi_weight);
 			napi_enable(&vpath->ring.napi);
 			vpath->ring.napi_p = &vpath->ring.napi;
 		}
drivers/net/ethernet/nxp/lpc_eth.c +1 -1
index 756f97dce85b..f606d75b33b4 100644
@@ -1373,7 +1373,7 @@ static int lpc_eth_drv_probe(struct platform_device *pdev)
 	pldat->duplex = DUPLEX_FULL;
 	__lpc_params_setup(pldat);
 
-	netif_napi_add(ndev, &pldat->napi, lpc_eth_poll, NAPI_WEIGHT);
+	netif_napi_add_weight(ndev, &pldat->napi, lpc_eth_poll, NAPI_WEIGHT);
 
 	ret = register_netdev(ndev);
 	if (ret) {
drivers/net/ethernet/realtek/8139cp.c +1 -1
index ad7b9e9d7f95..e0feeec13da6 100644
@@ -1986,7 +1986,7 @@ static int cp_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	eth_hw_addr_set(dev, (u8 *)addr);
 
 	dev->netdev_ops = &cp_netdev_ops;
-	netif_napi_add(dev, &cp->napi, cp_rx_poll, 16);
+	netif_napi_add_weight(dev, &cp->napi, cp_rx_poll, 16);
 	dev->ethtool_ops = &cp_ethtool_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 
drivers/net/ethernet/sfc/efx_channels.c +2 -2
index eec80b024195..3f28f9861dfa 100644
@@ -1316,8 +1316,8 @@ void efx_init_napi_channel(struct efx_channel *channel)
 	struct efx_nic *efx = channel->efx;
 
 	channel->napi_dev = efx->net_dev;
-	netif_napi_add(channel->napi_dev, &channel->napi_str,
-		       efx_poll, napi_weight);
+	netif_napi_add_weight(channel->napi_dev, &channel->napi_str, efx_poll,
+			      napi_weight);
 }
 
 void efx_init_napi(struct efx_nic *efx)
drivers/net/ethernet/sfc/falcon/efx.c +2 -2
index b7282331faec..f619ffb26787 100644
@@ -2017,8 +2017,8 @@ static void ef4_init_napi_channel(struct ef4_channel *channel)
 	struct ef4_nic *efx = channel->efx;
 
 	channel->napi_dev = efx->net_dev;
-	netif_napi_add(channel->napi_dev, &channel->napi_str,
-		       ef4_poll, napi_weight);
+	netif_napi_add_weight(channel->napi_dev, &channel->napi_str, ef4_poll,
+			      napi_weight);
 }
 
 static void ef4_init_napi(struct ef4_nic *efx)
drivers/net/ethernet/smsc/smsc911x.c +2 -1
index c854efdf1f25..3bf20211cceb 100644
@@ -2304,7 +2304,8 @@ static int smsc911x_init(struct net_device *dev)
 		return -ENODEV;
 
 	dev->flags |= IFF_MULTICAST;
-	netif_napi_add(dev, &pdata->napi, smsc911x_poll, SMSC_NAPI_WEIGHT);
+	netif_napi_add_weight(dev, &pdata->napi, smsc911x_poll,
+			      SMSC_NAPI_WEIGHT);
 	dev->netdev_ops = &smsc911x_netdev_ops;
 	dev->ethtool_ops = &smsc911x_ethtool_ops;
 
drivers/net/ethernet/toshiba/tc35815.c +1 -1
index ce38f7515225..47aab9c132c8 100644
@@ -804,7 +804,7 @@ static int tc35815_init_one(struct pci_dev *pdev,
 	dev->netdev_ops = &tc35815_netdev_ops;
 	dev->ethtool_ops = &tc35815_ethtool_ops;
 	dev->watchdog_timeo = TC35815_TX_TIMEOUT;
-	netif_napi_add(dev, &lp->napi, tc35815_poll, NAPI_WEIGHT);
+	netif_napi_add_weight(dev, &lp->napi, tc35815_poll, NAPI_WEIGHT);
 
 	dev->irq = pdev->irq;
 	dev->base_addr = (unsigned long)ioaddr;
drivers/net/ethernet/wiznet/w5100.c +1 -1
index 4fd7c39e1123..acd78120e53c 100644
@@ -1133,7 +1133,7 @@ int w5100_probe(struct device *dev, const struct w5100_ops *ops,
 
 	ndev->netdev_ops = &w5100_netdev_ops;
 	ndev->ethtool_ops = &w5100_ethtool_ops;
-	netif_napi_add(ndev, &priv->napi, w5100_napi_poll, 16);
+	netif_napi_add_weight(ndev, &priv->napi, w5100_napi_poll, 16);
 
 	/* This chip doesn't support VLAN packets with normal MTU,
 	 * so disable VLAN for this device.
drivers/net/ethernet/wiznet/w5300.c +1 -1
index 402d5036f266..773f8c77909a 100644
@@ -603,7 +603,7 @@ static int w5300_probe(struct platform_device *pdev)
 	ndev->netdev_ops = &w5300_netdev_ops;
 	ndev->ethtool_ops = &w5300_ethtool_ops;
 	ndev->watchdog_timeo = HZ;
-	netif_napi_add(ndev, &priv->napi, w5300_napi_poll, 16);
+	netif_napi_add_weight(ndev, &priv->napi, w5300_napi_poll, 16);
 
 	/* This chip doesn't support VLAN packets with normal MTU,
 	 * so disable VLAN for this device.
drivers/net/ethernet/xscale/ixp4xx_eth.c +1 -1
index d947955621ee..89770c2e0ffb 100644
@@ -1480,7 +1480,7 @@ static int ixp4xx_eth_probe(struct platform_device *pdev)
 	ndev->dev.dma_mask = dev->dma_mask;
 	ndev->dev.coherent_dma_mask = dev->coherent_dma_mask;
 
-	netif_napi_add(ndev, &port->napi, eth_poll, NAPI_WEIGHT);
+	netif_napi_add_weight(ndev, &port->napi, eth_poll, NAPI_WEIGHT);
 
 	if (!(port->npe = npe_request(NPE_ID(port->id))))
 		return -EIO;
Signature · 1 line
2.34.1
author maintainer
#10 lore raw

PATCH 5/6 net: virtio: switch to netif_napi_add_weight()

In reply to Jakub Kicinski

virtio netdev driver uses a custom napi weight, switch to the new API for setting custom weight.

---
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 drivers/net/virtio_net.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/virtio_net.c +2 -2
index ebb98b796352..db05b5e930be 100644
@@ -3313,8 +3313,8 @@ static int virtnet_alloc_queues(struct virtnet_info *vi)
 	INIT_DELAYED_WORK(&vi->refill, refill_work);
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		vi->rq[i].pages = NULL;
-		netif_napi_add(vi->dev, &vi->rq[i].napi, virtnet_poll,
-			       napi_weight);
+		netif_napi_add_weight(vi->dev, &vi->rq[i].napi, virtnet_poll,
+				      napi_weight);
 		netif_napi_add_tx_weight(vi->dev, &vi->sq[i].napi,
 					 virtnet_poll_tx,
 					 napi_tx ? napi_weight : 0);
Signature · 1 line
2.34.1
reviewer Reviewed-by on v1 5/6
#11 lore raw

Re: [PATCH net-next 5/6] net: virtio: switch to netif_napi_add_weight()

Quoted 4 lines · Jakub Kicinski

virtio netdev driver uses a custom napi weight, switch to the new API for setting custom weight.

Quoted 26 lines

---
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 drivers/net/virtio_net.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/virtio_net.c +2 -3
index ebb98b796352..db05b5e930be 100644
@@ -3313,8 +3313,8 @@ static int virtnet_alloc_queues(struct virtnet_info *vi)
 	INIT_DELAYED_WORK(&vi->refill, refill_work);
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		vi->rq[i].pages = NULL;
-		netif_napi_add(vi->dev, &vi->rq[i].napi, virtnet_poll,
-			       napi_weight);
+		netif_napi_add_weight(vi->dev, &vi->rq[i].napi, virtnet_poll,
+				      napi_weight);
 		netif_napi_add_tx_weight(vi->dev, &vi->sq[i].napi,
 					 virtnet_poll_tx,
 					 napi_tx ? napi_weight : 0);
--

2.34.1

Acked-by on v1 5/6
#12 lore raw

Re: [PATCH net-next 5/6] net: virtio: switch to netif_napi_add_weight()

In reply to Jakub Kicinski
Quoted 5 lines · Jakub Kicinski

virtio netdev driver uses a custom napi weight, switch to the new API for setting custom weight.

Quoted 26 lines

---
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 drivers/net/virtio_net.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/virtio_net.c +2 -3
index ebb98b796352..db05b5e930be 100644
@@ -3313,8 +3313,8 @@ static int virtnet_alloc_queues(struct virtnet_info *vi)
        INIT_DELAYED_WORK(&vi->refill, refill_work);
        for (i = 0; i < vi->max_queue_pairs; i++) {
                vi->rq[i].pages = NULL;
-               netif_napi_add(vi->dev, &vi->rq[i].napi, virtnet_poll,
-                              napi_weight);
+               netif_napi_add_weight(vi->dev, &vi->rq[i].napi, virtnet_poll,
+                                     napi_weight);
                netif_napi_add_tx_weight(vi->dev, &vi->sq[i].napi,
                                         virtnet_poll_tx,
                                         napi_tx ? napi_weight : 0);
--

2.34.1

maintainer Acked-by on v1 5/6
#13 lore raw

Re: [PATCH net-next 5/6] net: virtio: switch to netif_napi_add_weight()

In reply to Jakub Kicinski
Quoted 4 lines · Jakub Kicinski

virtio netdev driver uses a custom napi weight, switch to the new API for setting custom weight.

Quoted 25 lines

---
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 drivers/net/virtio_net.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/virtio_net.c +2 -2
index ebb98b796352..db05b5e930be 100644
@@ -3313,8 +3313,8 @@ static int virtnet_alloc_queues(struct virtnet_info *vi)
 	INIT_DELAYED_WORK(&vi->refill, refill_work);
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		vi->rq[i].pages = NULL;
-		netif_napi_add(vi->dev, &vi->rq[i].napi, virtnet_poll,
-			       napi_weight);
+		netif_napi_add_weight(vi->dev, &vi->rq[i].napi, virtnet_poll,
+				      napi_weight);
 		netif_napi_add_tx_weight(vi->dev, &vi->sq[i].napi,
 					 virtnet_poll_tx,
 					 napi_tx ? napi_weight : 0);
Signature · 1 line
2.34.1

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

maintainer Acked-by on v1 2/6
#3 lore raw

Re: [PATCH net-next 2/6] caif_virtio: switch to netif_napi_add_weight()

In reply to Jakub Kicinski
Quoted 4 lines · Jakub Kicinski

caif_virtio uses a custom napi weight, switch to the new API for setting custom weights.

Quoted 23 lines

---
CC: [email protected]
CC: [email protected]

---
 drivers/net/caif/caif_virtio.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/caif/caif_virtio.c +2 -1
index 444ef6a342f6..5458f57177a0 100644
@@ -714,7 +714,8 @@ static int cfv_probe(struct virtio_device *vdev)
 	/* Initialize NAPI poll context data */
 	vringh_kiov_init(&cfv->ctx.riov, NULL, 0);
 	cfv->ctx.head = USHRT_MAX;
-	netif_napi_add(netdev, &cfv->napi, cfv_rx_poll, CFV_DEFAULT_QUOTA);
+	netif_napi_add_weight(netdev, &cfv->napi, cfv_rx_poll,
+			      CFV_DEFAULT_QUOTA);
 
 	tasklet_setup(&cfv->tx_release_tasklet, cfv_tx_release_tasklet);
 
Signature · 1 line
2.34.1
#15 lore raw
In reply to Jakub Kicinski
This series was applied to netdev/net-next.git (master) by David S. Miller <[email protected]>:

Hello:

This series was applied to netdev/net-next.git (master) by David S. Miller <[email protected]>:

Quoted 13 lines · On Fri, 6 May 2022 10:07:45 -0700 you wrote:

The minority of drivers pass a custom weight to netif_napi_add(). Switch those away to the new netif_napi_add_weight(). All drivers (which can go thru net-next) calling netif_napi_add() will now be calling it with NAPI_POLL_WEIGHT or 64.

Jakub Kicinski (6):
um: vector: switch to netif_napi_add_weight()
caif_virtio: switch to netif_napi_add_weight()
eth: switch to netif_napi_add_weight()
r8152: switch to netif_napi_add_weight()
net: virtio: switch to netif_napi_add_weight()
net: wan: switch to netif_napi_add_weight()

Quoted 1 line

[...]

Here is the summary with links:
- [net-next,1/6] um: vector: switch to netif_napi_add_weight()

    https://git.kernel.org/netdev/net-next/c/4d92c6275575
  - [net-next,2/6] caif_virtio: switch to netif_napi_add_weight()
    https://git.kernel.org/netdev/net-next/c/be8af67fabcb
  - [net-next,3/6] eth: switch to netif_napi_add_weight()
    https://git.kernel.org/netdev/net-next/c/b707b89f7be3
  - [net-next,4/6] r8152: switch to netif_napi_add_weight()
    https://git.kernel.org/netdev/net-next/c/8ded532cd1cb
  - [net-next,5/6] net: virtio: switch to netif_napi_add_weight()
    https://git.kernel.org/netdev/net-next/c/d484735dcf92
  - [net-next,6/6] net: wan: switch to netif_napi_add_weight()
    https://git.kernel.org/netdev/net-next/c/6f83cb8cbfbf

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
maintainer
#5 lore raw

Re: [PATCH net-next 3/6] eth: switch to netif_napi_add_weight()

In reply to Jakub Kicinski
Quoted 10 lines · Jakub Kicinski

Switch all Ethernet drivers which use custom napi weights to the new API.

---

drivers/net/ethernet/sfc/efx_channels.c +0 -0
index eec80b024195..3f28f9861dfa 100644

drivers/net/ethernet/sfc/efx_channels.c void efx_init_napi_channel(struct efx_channel *channel) · line 1321

Quoted 8 lines
drivers/net/ethernet/sfc/efx_channels.c +2 -2
@@ -1316,8 +1316,8 @@ void efx_init_napi_channel(struct efx_channel *channel)
 	struct efx_nic *efx = channel->efx;
 
 	channel->napi_dev = efx->net_dev;
-	netif_napi_add(channel->napi_dev, &channel->napi_str,
-		       efx_poll, napi_weight);
+	netif_napi_add_weight(channel->napi_dev, &channel->napi_str, efx_poll,
+			      napi_weight);
 }

This isn't really a custom weight; napi_weight is initialised to
64 and never changed, so probably we ought to be just using NAPI_POLL_WEIGHT here and end up on the non-_weight API.
Same goes for Falcon.

-ed

author maintainer
#6 lore raw

Re: [PATCH net-next 3/6] eth: switch to netif_napi_add_weight()

Quoted 7 lines · On Wed, 11 May 2022 18:57:53 +0100 Edward Cree wrote:
Quoted 7 lines
drivers/net/ethernet/sfc/efx_channels.c +0 -0
index eec80b024195..3f28f9861dfa 100644
@@ -1316,8 +1316,8 @@ void efx_init_napi_channel(struct efx_channel *channel)
 	struct efx_nic *efx = channel->efx;
 
Quoted 11 lines

drivers/net/ethernet/sfc/efx_channels.c void efx_init_napi_channel(struct efx_channel *channel) · line 1321

Quoted 6 lines
drivers/net/ethernet/sfc/efx_channels.c +2 -2
 	channel->napi_dev = efx->net_dev;
-	netif_napi_add(channel->napi_dev, &channel->napi_str,
-		       efx_poll, napi_weight);
+	netif_napi_add_weight(channel->napi_dev, &channel->napi_str, efx_poll,
+			      napi_weight);
 }  

This isn't really a custom weight; napi_weight is initialised to
64 and never changed, so probably we ought to be just using NAPI_POLL_WEIGHT here and end up on the non-_weight API.
Same goes for Falcon.

Ack, I wanted to be nice. I figured this must be a stub for a module param in your our of tree driver. Should I send a patch to remove the non-const static napi_weight globals and switch back to non-_weight?

maintainer
#7 lore raw

Re: [PATCH net-next 3/6] eth: switch to netif_napi_add_weight()

Quoted 8 lines · Jakub Kicinski
Quoted 4 lines · On Wed, 11 May 2022 18:57:53 +0100 Edward Cree wrote:

This isn't really a custom weight; napi_weight is initialised to
64 and never changed, so probably we ought to be just using NAPI_POLL_WEIGHT here and end up on the non-_weight API.
Same goes for Falcon.

Ack, I wanted to be nice. I figured this must be a stub for a module param in your our of tree driver.

I mean, it *is*... but there's almost certainly a better way. Configuring
NAPI weight for tuning purposes (as opposed to some kind of correctness limitation in a driver) probably ought to be left to generic infrastructure rather than vendor-specific modparams in driver code.

Quoted 2 lines

Should I send a patch to remove
the non-const static napi_weight globals and switch back to non-_weight?

Yes please, unless Martin has any objections...?

-ed

#8 lore raw

Re: [PATCH net-next 3/6] eth: switch to netif_napi_add_weight()

Quoted 15 lines · Edward Cree
Quoted 8 lines · Jakub Kicinski
Quoted 4 lines · On Wed, 11 May 2022 18:57:53 +0100 Edward Cree wrote:

This isn't really a custom weight; napi_weight is initialised to
64 and never changed, so probably we ought to be just using NAPI_POLL_WEIGHT here and end up on the non-_weight API.
Same goes for Falcon.

Ack, I wanted to be nice. I figured this must be a stub for a module param in your our of tree driver.

I mean, it *is*... but there's almost certainly a better way. Configuring
NAPI weight for tuning purposes (as opposed to some kind of correctness limitation in a driver) probably ought to be left to generic infrastructure rather than vendor-specific modparams in driver code.

Quoted 4 lines
Quoted 2 lines

Should I send a patch to remove
the non-const static napi_weight globals and switch back to non-_weight?

Yes please, unless Martin has any objections...?

Fine for me.

Martin