[PATCH AUTOSEL 5.12 013/116] net/mlx5e: Use net_prefetchw instead of prefetchw in MPWQE TX datapath

[PATCH AUTOSEL 5.12 013/116] net/mlx5e: Use net_prefetchw instead of prefetchw in MPWQE TX datapath

[Sasha Levin]

From: Maxim Mikityanskiy <maximmi@mellanox.com>

[ Upstream commit 991b2654605b455a94dac73e14b23480e7e20991 ]

Commit e20f0dbf204f ("net/mlx5e: RX, Add a prefetch command for small
L1_CACHE_BYTES") switched to using net_prefetchw at all places in mlx5e.
In the same time frame, commit 5af75c747e2a ("net/mlx5e: Enhanced TX
MPWQE for SKBs") added one more usage of prefetchw. When these two
changes were merged, this new occurrence of prefetchw wasn't replaced
with net_prefetchw.

This commit fixes this last occurrence of prefetchw in
mlx5e_tx_mpwqe_session_start, making the same change that was done in
mlx5e_xdp_mpwqe_session_start.

Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index bdbffe484fce..d2efe2455955 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -576,7 +576,7 @@ static void mlx5e_tx_mpwqe_session_start(struct mlx5e_txqsq *sq,
 
 	pi = mlx5e_txqsq_get_next_pi(sq, MLX5E_TX_MPW_MAX_WQEBBS);
 	wqe = MLX5E_TX_FETCH_WQE(sq, pi);
-	prefetchw(wqe->data);
+	net_prefetchw(wqe->data);
 
 	*session = (struct mlx5e_tx_mpwqe) {
 		.wqe = wqe,
-- 
2.30.2

[PATCH AUTOSEL 5.12 048/116] IB/hfi1: Correct oversized ring allocation

[Sasha Levin]

From: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>

[ Upstream commit b536d4b2a279733f440c911dc831764690b90050 ]

The completion ring for tx is using the wrong size to size the ring,
oversizing the ring by two orders of magniture.

Correct the allocation size and use kcalloc_node() to allocate the ring.
Fix mistaken GFP defines in similar allocations.

Link: https://lore.kernel.org/r/1617026056-50483-4-git-send-email-dennis.dalessandro@cornelisnetworks.com
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/hw/hfi1/ipoib.h    |  3 ++-
 drivers/infiniband/hw/hfi1/ipoib_tx.c | 14 +++++++-------
 2 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/infiniband/hw/hfi1/ipoib.h b/drivers/infiniband/hw/hfi1/ipoib.h
index f650cac9d424..d30c23b6527a 100644
--- a/drivers/infiniband/hw/hfi1/ipoib.h
+++ b/drivers/infiniband/hw/hfi1/ipoib.h
@@ -52,8 +52,9 @@ union hfi1_ipoib_flow {
  * @producer_lock: producer sync lock
  * @consumer_lock: consumer sync lock
  */
+struct ipoib_txreq;
 struct hfi1_ipoib_circ_buf {
-	void **items;
+	struct ipoib_txreq **items;
 	unsigned long head;
 	unsigned long tail;
 	unsigned long max_items;
diff --git a/drivers/infiniband/hw/hfi1/ipoib_tx.c b/drivers/infiniband/hw/hfi1/ipoib_tx.c
index edd4eeac8dd1..cdc26ee3cf52 100644
--- a/drivers/infiniband/hw/hfi1/ipoib_tx.c
+++ b/drivers/infiniband/hw/hfi1/ipoib_tx.c
@@ -702,14 +702,14 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
 
 	priv->tx_napis = kcalloc_node(dev->num_tx_queues,
 				      sizeof(struct napi_struct),
-				      GFP_ATOMIC,
+				      GFP_KERNEL,
 				      priv->dd->node);
 	if (!priv->tx_napis)
 		goto free_txreq_cache;
 
 	priv->txqs = kcalloc_node(dev->num_tx_queues,
 				  sizeof(struct hfi1_ipoib_txq),
-				  GFP_ATOMIC,
+				  GFP_KERNEL,
 				  priv->dd->node);
 	if (!priv->txqs)
 		goto free_tx_napis;
@@ -741,9 +741,9 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
 					     priv->dd->node);
 
 		txq->tx_ring.items =
-			vzalloc_node(array_size(tx_ring_size,
-						sizeof(struct ipoib_txreq)),
-				     priv->dd->node);
+			kcalloc_node(tx_ring_size,
+				     sizeof(struct ipoib_txreq *),
+				     GFP_KERNEL, priv->dd->node);
 		if (!txq->tx_ring.items)
 			goto free_txqs;
 
@@ -764,7 +764,7 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
 		struct hfi1_ipoib_txq *txq = &priv->txqs[i];
 
 		netif_napi_del(txq->napi);
-		vfree(txq->tx_ring.items);
+		kfree(txq->tx_ring.items);
 	}
 
 	kfree(priv->txqs);
@@ -817,7 +817,7 @@ void hfi1_ipoib_txreq_deinit(struct hfi1_ipoib_dev_priv *priv)
 		hfi1_ipoib_drain_tx_list(txq);
 		netif_napi_del(txq->napi);
 		(void)hfi1_ipoib_drain_tx_ring(txq, txq->tx_ring.max_items);
-		vfree(txq->tx_ring.items);
+		kfree(txq->tx_ring.items);
 	}
 
 	kfree(priv->txqs);
-- 
2.30.2