[PATCH rdma-next 0/3] rdma: move to the new NAPI helpers

[PATCH rdma-next 0/3] rdma: move to the new NAPI helpers

[Jakub Kicinski]

I'm dropping the wight argument from netif_napi_add()
because most callers just want the default. This is
prep taking care of the few callers in RDMA.

Jakub Kicinski (3):
  IB/hfi1: switch to netif_napi_add_tx()
  IB/hfi1: switch to netif_napi_add_weight()
  ipoib: switch to netif_napi_add_weight()

 drivers/infiniband/hw/hfi1/ipoib_tx.c     | 4 +---
 drivers/infiniband/hw/hfi1/netdev_rx.c    | 2 +-
 drivers/infiniband/ulp/ipoib/ipoib_main.c | 6 ++++--
 3 files changed, 6 insertions(+), 6 deletions(-)

-- 
2.36.1

[PATCH rdma-next 1/3] IB/hfi1: switch to netif_napi_add_tx()

[Jakub Kicinski]

Switch to the new API not requiring the NAPI_POLL_WEIGHT argument.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
CC: dennis.dalessandro@cornelisnetworks.com
CC: jgg@ziepe.ca
CC: leon@kernel.org
CC: linux-rdma@vger.kernel.org
---
 drivers/infiniband/hw/hfi1/ipoib_tx.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/infiniband/hw/hfi1/ipoib_tx.c b/drivers/infiniband/hw/hfi1/ipoib_tx.c
index d6bbdb8fcb50..5d9a7b09ca37 100644
--- a/drivers/infiniband/hw/hfi1/ipoib_tx.c
+++ b/drivers/infiniband/hw/hfi1/ipoib_tx.c
@@ -742,9 +742,7 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
 				kzalloc_node(sizeof(*tx->sdma_hdr),
 					     GFP_KERNEL, priv->dd->node);
 
-		netif_tx_napi_add(dev, &txq->napi,
-				  hfi1_ipoib_poll_tx_ring,
-				  NAPI_POLL_WEIGHT);
+		netif_napi_add_tx(dev, &txq->napi, hfi1_ipoib_poll_tx_ring);
 	}
 
 	return 0;
-- 
2.36.1

[PATCH rdma-next 2/3] IB/hfi1: switch to netif_napi_add_weight()

[Jakub Kicinski]

Since we'll remove the last argument from netif_napi_add()
soon switch this RDMA driver to netif_napi_add_weight()
for now to avoid cross-tree patches.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
CC: dennis.dalessandro@cornelisnetworks.com
CC: jgg@ziepe.ca
CC: leon@kernel.org
CC: linux-rdma@vger.kernel.org
---
 drivers/infiniband/hw/hfi1/netdev_rx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/hfi1/netdev_rx.c b/drivers/infiniband/hw/hfi1/netdev_rx.c
index 03b098a494b5..3dfa5aff2512 100644
--- a/drivers/infiniband/hw/hfi1/netdev_rx.c
+++ b/drivers/infiniband/hw/hfi1/netdev_rx.c
@@ -216,7 +216,7 @@ static int hfi1_netdev_rxq_init(struct hfi1_netdev_rx *rx)
 		 * right now.
 		 */
 		set_bit(NAPI_STATE_NO_BUSY_POLL, &rxq->napi.state);
-		netif_napi_add(dev, &rxq->napi, hfi1_netdev_rx_napi, 64);
+		netif_napi_add_weight(dev, &rxq->napi, hfi1_netdev_rx_napi, 64);
 		rc = msix_netdev_request_rcd_irq(rxq->rcd);
 		if (rc)
 			goto bail_context_irq_failure;
-- 
2.36.1

[PATCH rdma-next 3/3] ipoib: switch to netif_napi_add_weight()

[Jakub Kicinski]

We want to remove the weight argument from the basic
netif_napi_add() API and just default to 64.
Switch ipoib to the new API for explicitly specifing
the weight.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
CC: jgg@ziepe.ca
CC: leon@kernel.org
CC: linux-rdma@vger.kernel.org
---
 drivers/infiniband/ulp/ipoib/ipoib_main.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 2a8961b685c2..a4904371e2db 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -1664,8 +1664,10 @@ static void ipoib_napi_add(struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = ipoib_priv(dev);
 
-	netif_napi_add(dev, &priv->recv_napi, ipoib_rx_poll, IPOIB_NUM_WC);
-	netif_napi_add(dev, &priv->send_napi, ipoib_tx_poll, MAX_SEND_CQE);
+	netif_napi_add_weight(dev, &priv->recv_napi, ipoib_rx_poll,
+			      IPOIB_NUM_WC);
+	netif_napi_add_weight(dev, &priv->send_napi, ipoib_tx_poll,
+			      MAX_SEND_CQE);
 }
 
 static void ipoib_napi_del(struct net_device *dev)
-- 
2.36.1

Re: [PATCH rdma-next 0/3] rdma: move to the new NAPI helpers

[Leon Romanovsky]

On Tue, Jul 05, 2022 at 04:02:05PM -0700, Jakub Kicinski wrote:
> I'm dropping the wight argument from netif_napi_add()
> because most callers just want the default. This is
> prep taking care of the few callers in RDMA.
> 
> Jakub Kicinski (3):
>   IB/hfi1: switch to netif_napi_add_tx()
>   IB/hfi1: switch to netif_napi_add_weight()
>   ipoib: switch to netif_napi_add_weight()
> 
>  drivers/infiniband/hw/hfi1/ipoib_tx.c     | 4 +---
>  drivers/infiniband/hw/hfi1/netdev_rx.c    | 2 +-
>  drivers/infiniband/ulp/ipoib/ipoib_main.c | 6 ++++--
>  3 files changed, 6 insertions(+), 6 deletions(-)

Thanks, applied.

> 
> -- 
> 2.36.1
>