Netdev List
 help / color / mirror / Atom feed
From: Saeed Mahameed <saeedm@mellanox.com>
To: "David S. Miller" <davem@davemloft.net>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	Aya Levin <ayal@mellanox.com>, Tariq Toukan <tariqt@mellanox.com>,
	Jiri Pirko <jiri@mellanox.com>,
	Saeed Mahameed <saeedm@mellanox.com>
Subject: [net-next v2 09/16] net/mlx5e: Split open/close ICOSQ into stages
Date: Tue, 20 Aug 2019 20:24:28 +0000	[thread overview]
Message-ID: <20190820202352.2995-10-saeedm@mellanox.com> (raw)
In-Reply-To: <20190820202352.2995-1-saeedm@mellanox.com>

From: Aya Levin <ayal@mellanox.com>

Align ICOSQ open/close behaviour with RQ and SQ. Split open flow into
open and activate where open handles creation and activate enables the
queue. Do a symmetric thing in close flow: split into close and
deactivate.

Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../mellanox/mlx5/core/en/xsk/setup.c         |  2 ++
 .../ethernet/mellanox/mlx5/core/en/xsk/tx.c   |  7 +++++++
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 19 +++++++++++++++----
 3 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c
index 2c4d1f415968..d360750b25b7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c
@@ -150,6 +150,7 @@ void mlx5e_close_xsk(struct mlx5e_channel *c)
 
 void mlx5e_activate_xsk(struct mlx5e_channel *c)
 {
+	mlx5e_activate_icosq(&c->xskicosq);
 	set_bit(MLX5E_RQ_STATE_ENABLED, &c->xskrq.state);
 	/* TX queue is created active. */
 
@@ -162,6 +163,7 @@ void mlx5e_deactivate_xsk(struct mlx5e_channel *c)
 {
 	mlx5e_deactivate_rq(&c->xskrq);
 	/* TX queue is disabled on close. */
+	mlx5e_deactivate_icosq(&c->xskicosq);
 }
 
 static int mlx5e_redirect_xsk_rqt(struct mlx5e_priv *priv, u16 ix, u32 rqn)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
index 35e188cf4ea4..fd2c75b4b519 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c
@@ -26,6 +26,13 @@ int mlx5e_xsk_async_xmit(struct net_device *dev, u32 qid)
 		return -ENXIO;
 
 	if (!napi_if_scheduled_mark_missed(&c->napi)) {
+		/* To avoid WQE overrun, don't post a NOP if XSKICOSQ is not
+		 * active and not polled by NAPI. Return 0, because the upcoming
+		 * activate will trigger the IRQ for us.
+		 */
+		if (unlikely(!test_bit(MLX5E_SQ_STATE_ENABLED, &c->xskicosq.state)))
+			return 0;
+
 		spin_lock(&c->xskicosq_lock);
 		mlx5e_trigger_irq(&c->xskicosq);
 		spin_unlock(&c->xskicosq_lock);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 1cec30627bb0..2c4e1cb54402 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1375,7 +1375,6 @@ int mlx5e_open_icosq(struct mlx5e_channel *c, struct mlx5e_params *params,
 	csp.cqn             = sq->cq.mcq.cqn;
 	csp.wq_ctrl         = &sq->wq_ctrl;
 	csp.min_inline_mode = params->tx_min_inline_mode;
-	set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state);
 	err = mlx5e_create_sq_rdy(c->mdev, param, &csp, &sq->sqn);
 	if (err)
 		goto err_free_icosq;
@@ -1389,12 +1388,22 @@ int mlx5e_open_icosq(struct mlx5e_channel *c, struct mlx5e_params *params,
 	return err;
 }
 
-void mlx5e_close_icosq(struct mlx5e_icosq *sq)
+static void mlx5e_activate_icosq(struct mlx5e_icosq *icosq)
 {
-	struct mlx5e_channel *c = sq->channel;
+	set_bit(MLX5E_SQ_STATE_ENABLED, &icosq->state);
+}
 
-	clear_bit(MLX5E_SQ_STATE_ENABLED, &sq->state);
+static void mlx5e_deactivate_icosq(struct mlx5e_icosq *icosq)
+{
+	struct mlx5e_channel *c = icosq->channel;
+
+	clear_bit(MLX5E_SQ_STATE_ENABLED, &icosq->state);
 	napi_synchronize(&c->napi);
+}
+
+void mlx5e_close_icosq(struct mlx5e_icosq *sq)
+{
+	struct mlx5e_channel *c = sq->channel;
 
 	mlx5e_destroy_sq(c->mdev, sq->sqn);
 	mlx5e_free_icosq(sq);
@@ -1971,6 +1980,7 @@ static void mlx5e_activate_channel(struct mlx5e_channel *c)
 
 	for (tc = 0; tc < c->num_tc; tc++)
 		mlx5e_activate_txqsq(&c->sq[tc]);
+	mlx5e_activate_icosq(&c->icosq);
 	mlx5e_activate_rq(&c->rq);
 	netif_set_xps_queue(c->netdev, c->xps_cpumask, c->ix);
 
@@ -1986,6 +1996,7 @@ static void mlx5e_deactivate_channel(struct mlx5e_channel *c)
 		mlx5e_deactivate_xsk(c);
 
 	mlx5e_deactivate_rq(&c->rq);
+	mlx5e_deactivate_icosq(&c->icosq);
 	for (tc = 0; tc < c->num_tc; tc++)
 		mlx5e_deactivate_txqsq(&c->sq[tc]);
 }
-- 
2.21.0


  parent reply	other threads:[~2019-08-20 20:24 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-20 20:24 [pull request][net-next v2 00/16] Mellanox, mlx5 devlink RX health reporters Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 01/16] net/mlx5e: Rename reporter header file Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 02/16] net/mlx5e: Change naming convention for reporter's functions Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 03/16] net/mlx5e: Generalize tx reporter's functionality Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 04/16] net/mlx5e: Extend tx diagnose function Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 05/16] net/mlx5e: Extend tx reporter diagnostics output Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 06/16] net/mlx5e: Add cq info to tx reporter diagnose Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 07/16] net/mlx5e: Add helper functions for reporter's basics Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 08/16] net/mlx5e: Add support to rx reporter diagnose Saeed Mahameed
2019-08-20 20:24 ` Saeed Mahameed [this message]
2019-08-20 20:24 ` [net-next v2 10/16] net/mlx5e: Report and recover from CQE error on ICOSQ Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 11/16] net/mlx5e: Report and recover from rx timeout Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 12/16] net/mlx5e: RX, Handle CQE with error at the earliest stage Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 13/16] net/mlx5e: Report and recover from CQE with error on RQ Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 14/16] Documentation: net: mlx5: Devlink health documentation updates Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 15/16] net/mlx5e: Fix deallocation of non-fully init encap entries Saeed Mahameed
2019-08-20 20:24 ` [net-next v2 16/16] net/mlx5: Fix the order of fc_stats cleanup Saeed Mahameed
2019-08-21  3:36 ` [pull request][net-next v2 00/16] Mellanox, mlx5 devlink RX health reporters Jakub Kicinski
2019-08-21  6:36 ` David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190820202352.2995-10-saeedm@mellanox.com \
    --to=saeedm@mellanox.com \
    --cc=ayal@mellanox.com \
    --cc=davem@davemloft.net \
    --cc=jiri@mellanox.com \
    --cc=netdev@vger.kernel.org \
    --cc=tariqt@mellanox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox