diff mbox series

[net-next] mlx4: handle non-napi callers to napi_poll

Message ID 20201008184526.3196768-1-jonathan.lemon@gmail.com (mailing list archive)
State Not Applicable
Headers show
Series [net-next] mlx4: handle non-napi callers to napi_poll | expand

Commit Message

Jonathan Lemon Oct. 8, 2020, 6:45 p.m. UTC 
From: Jonathan Lemon <bsd@fb.com>

netcons calls napi_poll with a budget of 0 to transmit packets.
Handle this by:
 - skipping RX processing
 - do not try to recycle TX packets to the RX cache

Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_rx.c | 3 +++
 drivers/net/ethernet/mellanox/mlx4/en_tx.c | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)

Comments

Jakub Kicinski Oct. 11, 2020, 6:35 p.m. UTC | #1 
On Thu, 8 Oct 2020 11:45:26 -0700 Jonathan Lemon wrote:
> From: Jonathan Lemon <bsd@fb.com>
> 
> netcons calls napi_poll with a budget of 0 to transmit packets.
> Handle this by:
>  - skipping RX processing
>  - do not try to recycle TX packets to the RX cache
> 
> Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>

Tariq, Saeed - how does this look to you?
Tariq Toukan Oct. 12, 2020, 6:40 a.m. UTC | #2 
On Mon, 12 Oct 2020 at 05:47, Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Thu, 8 Oct 2020 11:45:26 -0700 Jonathan Lemon wrote:
> > From: Jonathan Lemon <bsd@fb.com>
> >
> > netcons calls napi_poll with a budget of 0 to transmit packets.
> > Handle this by:
> >  - skipping RX processing
> >  - do not try to recycle TX packets to the RX cache
> >
> > Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
>
> Tariq, Saeed - how does this look to you?

Reviewed-by: Tariq Toukan <tariqt@nvidia.com>

Thanks,
Tariq
Jakub Kicinski Oct. 12, 2020, 9:07 p.m. UTC | #3 
On Mon, 12 Oct 2020 09:40:51 +0300 Tariq Toukan wrote:
> On Mon, 12 Oct 2020 at 05:47, Jakub Kicinski <kuba@kernel.org> wrote:
> >
> > On Thu, 8 Oct 2020 11:45:26 -0700 Jonathan Lemon wrote:  
> > > From: Jonathan Lemon <bsd@fb.com>
> > >
> > > netcons calls napi_poll with a budget of 0 to transmit packets.
> > > Handle this by:
> > >  - skipping RX processing
> > >  - do not try to recycle TX packets to the RX cache
> > >
> > > Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>  
> >
> > Tariq, Saeed - how does this look to you?  
> 
> Reviewed-by: Tariq Toukan <tariqt@nvidia.com>

Applied, thanks!
diff mbox series

Patch

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 99d7737e8ad6..502d1b97855c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -943,6 +943,9 @@  int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget)
 	bool clean_complete = true;
 	int done;
 
+	if (!budget)
+		return 0;
+
 	if (priv->tx_ring_num[TX_XDP]) {
 		xdp_tx_cq = priv->tx_cq[TX_XDP][cq->ring];
 		if (xdp_tx_cq->xdp_busy) {
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 45869b29368f..3ddb7268e415 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -350,7 +350,7 @@  u32 mlx4_en_recycle_tx_desc(struct mlx4_en_priv *priv,
 		.dma = tx_info->map0_dma,
 	};
 
-	if (!mlx4_en_rx_recycle(ring->recycle_ring, &frame)) {
+	if (!napi_mode || !mlx4_en_rx_recycle(ring->recycle_ring, &frame)) {
 		dma_unmap_page(priv->ddev, tx_info->map0_dma,
 			       PAGE_SIZE, priv->dma_dir);
 		put_page(tx_info->page);