netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Saeed Mahameed <saeed@kernel.org>
To: "David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>
Cc: netdev@vger.kernel.org, Tariq Toukan <tariqt@nvidia.com>,
	Saeed Mahameed <saeedm@nvidia.com>
Subject: [net-next 01/10] mlx5: count all link events
Date: Thu,  3 Jun 2021 13:11:46 -0700	[thread overview]
Message-ID: <20210603201155.109184-2-saeed@kernel.org> (raw)
In-Reply-To: <20210603201155.109184-1-saeed@kernel.org>

From: Jakub Kicinski <kuba@kernel.org>

mlx5 devices were observed generating MLX5_PORT_CHANGE_SUBTYPE_ACTIVE
events without an intervening MLX5_PORT_CHANGE_SUBTYPE_DOWN. This
breaks link flap detection based on Linux carrier state transition
count as netif_carrier_on() does nothing if carrier is already on.
Make sure we count such events.

netif_carrier_event() increments the counters and fires the linkwatch
events. The latter is not necessary for the use case but seems like
the right thing to do.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_main.c  |  6 +++++-
 include/linux/netdevice.h                      |  2 +-
 net/sched/sch_generic.c                        | 18 ++++++++++++++++++
 3 files changed, 24 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index ad0f69480b9c..e36d0c6a08db 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -91,12 +91,16 @@ void mlx5e_update_carrier(struct mlx5e_priv *priv)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 	u8 port_state;
+	bool up;
 
 	port_state = mlx5_query_vport_state(mdev,
 					    MLX5_VPORT_STATE_OP_MOD_VNIC_VPORT,
 					    0);
 
-	if (port_state == VPORT_STATE_UP) {
+	up = port_state == VPORT_STATE_UP;
+	if (up == netif_carrier_ok(priv->netdev))
+		netif_carrier_event(priv->netdev);
+	if (up) {
 		netdev_info(priv->netdev, "Link up\n");
 		netif_carrier_on(priv->netdev);
 	} else {
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 5cbc950b34df..be1dcceda5e4 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -4187,8 +4187,8 @@ unsigned long dev_trans_start(struct net_device *dev);
 void __netdev_watchdog_up(struct net_device *dev);
 
 void netif_carrier_on(struct net_device *dev);
-
 void netif_carrier_off(struct net_device *dev);
+void netif_carrier_event(struct net_device *dev);
 
 /**
  *	netif_dormant_on - mark device as dormant.
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index fc8b56bcabf3..e9c0afc8becc 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -540,6 +540,24 @@ void netif_carrier_off(struct net_device *dev)
 }
 EXPORT_SYMBOL(netif_carrier_off);
 
+/**
+ *	netif_carrier_event - report carrier state event
+ *	@dev: network device
+ *
+ * Device has detected a carrier event but the carrier state wasn't changed.
+ * Use in drivers when querying carrier state asynchronously, to avoid missing
+ * events (link flaps) if link recovers before it's queried.
+ */
+void netif_carrier_event(struct net_device *dev)
+{
+	if (dev->reg_state == NETREG_UNINITIALIZED)
+		return;
+	atomic_inc(&dev->carrier_up_count);
+	atomic_inc(&dev->carrier_down_count);
+	linkwatch_fire_event(dev);
+}
+EXPORT_SYMBOL_GPL(netif_carrier_event);
+
 /* "NOOP" scheduler: the best scheduler, recommended for all interfaces
    under all circumstances. It is difficult to invent anything faster or
    cheaper.
-- 
2.31.1


  reply	other threads:[~2021-06-03 20:12 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-03 20:11 [pull request][net-next 00/10] mlx5 updates 2021-06-03 Saeed Mahameed
2021-06-03 20:11 ` Saeed Mahameed [this message]
2021-06-03 22:30   ` [net-next 01/10] mlx5: count all link events patchwork-bot+netdevbpf
2021-06-03 20:11 ` [net-next 02/10] net/mlx5: Fix duplicate included vhca_event.h Saeed Mahameed
2021-06-03 20:11 ` [net-next 03/10] net/mlx5: check for allocation failure in mlx5_ft_pool_init() Saeed Mahameed
2021-06-03 20:11 ` [net-next 04/10] net/mlx5e: Remove the repeated declaration Saeed Mahameed
2021-06-03 20:11 ` [net-next 05/10] net/mlx5e: IPoIB, Add support for NDR speed Saeed Mahameed
2021-06-03 20:11 ` [net-next 06/10] net/mlx5e: Zero-init DIM structures Saeed Mahameed
2021-06-03 20:11 ` [net-next 07/10] net/mlx5e: RX, Re-place page pool numa node change logic Saeed Mahameed
2021-06-03 20:11 ` [net-next 08/10] net/mlx5e: Disable TX MPWQE in kdump mode Saeed Mahameed
2021-06-03 20:11 ` [net-next 09/10] net/mlx5e: Disable TLS device offload " Saeed Mahameed
2021-06-03 20:11 ` [net-next 10/10] net/mlx5e: Remove unreachable code in mlx5e_xmit() Saeed Mahameed

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=20210603201155.109184-2-saeed@kernel.org \
    --to=saeed@kernel.org \
    --cc=davem@davemloft.net \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=saeedm@nvidia.com \
    --cc=tariqt@nvidia.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;
as well as URLs for NNTP newsgroup(s).