From: Saeed Mahameed <saeed@kernel.org>
To: "David S. Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>,
Jason Gunthorpe <jgg@nvidia.com>
Cc: Leon Romanovsky <leonro@nvidia.com>,
linux-rdma@vger.kernel.org, netdev@vger.kernel.org,
Mark Bloch <mbloch@nvidia.com>, Maor Gottlieb <maorg@nvidia.com>,
Saeed Mahameed <saeedm@nvidia.com>
Subject: [for-next v2 08/17] net/mlx5: Lag, record inactive state of bond device
Date: Wed, 23 Feb 2022 15:39:21 -0800 [thread overview]
Message-ID: <20220223233930.319301-9-saeed@kernel.org> (raw)
In-Reply-To: <20220223233930.319301-1-saeed@kernel.org>
From: Mark Bloch <mbloch@nvidia.com>
A bond device will drop duplicate packets (received on inactive ports)
by default. A flag (all_slaves_active) can be set to override such
behaviour. This flag is a global flag per bond device (ALB mode isn't
supported by mlx5 driver so it can be ignored)
When NETDEV_CHANGEUPPER / NETDEV_CHANGEINFODATA event is received check if
there is an interface that is inactive.
Downstream patch will use this information in order to decide if a drop
rule is needed.
Signed-off-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
.../net/ethernet/mellanox/mlx5/core/lag/lag.c | 49 ++++++++++++++++++-
.../net/ethernet/mellanox/mlx5/core/lag/lag.h | 1 +
.../net/ethernet/mellanox/mlx5/core/lag/mp.c | 2 +-
3 files changed, 49 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
index 05e8cbece095..125ac4befd74 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
@@ -31,6 +31,7 @@
*/
#include <linux/netdevice.h>
+#include <net/bonding.h>
#include <linux/mlx5/driver.h>
#include <linux/mlx5/eswitch.h>
#include <linux/mlx5/vport.h>
@@ -619,6 +620,8 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
struct net_device *upper = info->upper_dev, *ndev_tmp;
struct netdev_lag_upper_info *lag_upper_info = NULL;
bool is_bonded, is_in_lag, mode_supported;
+ bool has_inactive = 0;
+ struct slave *slave;
int bond_status = 0;
int num_slaves = 0;
int changed = 0;
@@ -638,8 +641,12 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
rcu_read_lock();
for_each_netdev_in_bond_rcu(upper, ndev_tmp) {
idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev_tmp);
- if (idx >= 0)
+ if (idx >= 0) {
+ slave = bond_slave_get_rcu(ndev_tmp);
+ if (slave)
+ has_inactive |= bond_is_slave_inactive(slave);
bond_status |= (1 << idx);
+ }
num_slaves++;
}
@@ -654,6 +661,7 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
tracker->hash_type = lag_upper_info->hash_type;
}
+ tracker->has_inactive = has_inactive;
/* Determine bonding status:
* A device is considered bonded if both its physical ports are slaves
* of the same lag master, and only them.
@@ -710,6 +718,38 @@ static int mlx5_handle_changelowerstate_event(struct mlx5_lag *ldev,
return 1;
}
+static int mlx5_handle_changeinfodata_event(struct mlx5_lag *ldev,
+ struct lag_tracker *tracker,
+ struct net_device *ndev)
+{
+ struct net_device *ndev_tmp;
+ struct slave *slave;
+ bool has_inactive = 0;
+ int idx;
+
+ if (!netif_is_lag_master(ndev))
+ return 0;
+
+ rcu_read_lock();
+ for_each_netdev_in_bond_rcu(ndev, ndev_tmp) {
+ idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev_tmp);
+ if (idx < 0)
+ continue;
+
+ slave = bond_slave_get_rcu(ndev_tmp);
+ if (slave)
+ has_inactive |= bond_is_slave_inactive(slave);
+ }
+ rcu_read_unlock();
+
+ if (tracker->has_inactive == has_inactive)
+ return 0;
+
+ tracker->has_inactive = has_inactive;
+
+ return 1;
+}
+
static int mlx5_lag_netdev_event(struct notifier_block *this,
unsigned long event, void *ptr)
{
@@ -718,7 +758,9 @@ static int mlx5_lag_netdev_event(struct notifier_block *this,
struct mlx5_lag *ldev;
int changed = 0;
- if ((event != NETDEV_CHANGEUPPER) && (event != NETDEV_CHANGELOWERSTATE))
+ if (event != NETDEV_CHANGEUPPER &&
+ event != NETDEV_CHANGELOWERSTATE &&
+ event != NETDEV_CHANGEINFODATA)
return NOTIFY_DONE;
ldev = container_of(this, struct mlx5_lag, nb);
@@ -734,6 +776,9 @@ static int mlx5_lag_netdev_event(struct notifier_block *this,
changed = mlx5_handle_changelowerstate_event(ldev, &tracker,
ndev, ptr);
break;
+ case NETDEV_CHANGEINFODATA:
+ changed = mlx5_handle_changeinfodata_event(ldev, &tracker, ndev);
+ break;
}
ldev->tracker = tracker;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
index e5d231c31b54..305d9adbe325 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.h
@@ -35,6 +35,7 @@ struct lag_tracker {
enum netdev_lag_tx_type tx_type;
struct netdev_lag_lower_state_info netdev_state[MLX5_MAX_PORTS];
unsigned int is_bonded:1;
+ unsigned int has_inactive:1;
enum netdev_lag_hash hash_type;
};
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c
index 1ca01a5b6cdd..4213208d9ef7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c
@@ -50,7 +50,7 @@ bool mlx5_lag_is_multipath(struct mlx5_core_dev *dev)
static void mlx5_lag_set_port_affinity(struct mlx5_lag *ldev,
enum mlx5_lag_port_affinity port)
{
- struct lag_tracker tracker;
+ struct lag_tracker tracker = {};
if (!__mlx5_lag_is_multipath(ldev))
return;
--
2.35.1
next prev parent reply other threads:[~2022-02-23 23:40 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-23 23:39 [pull request][for-next v2 00/17] mlx5-next 2022-22-02 Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 01/17] mlx5: remove unused static inlines Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 02/17] net/mlx5: Add ability to insert to specific flow group Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 03/17] net/mlx5: E-Switch, reserve and use same uplink metadata across ports Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 04/17] net/mlx5: E-switch, remove special uplink ingress ACL handling Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 05/17] net/mlx5: E-switch, add drop rule support to ingress ACL Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 06/17] net/mlx5: Lag, use local variable already defined to access E-Switch Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 07/17] net/mlx5: Lag, don't use magic numbers for ports Saeed Mahameed
2022-02-23 23:39 ` Saeed Mahameed [this message]
2022-02-23 23:39 ` [for-next v2 09/17] net/mlx5: Lag, offload active-backup drops to hardware Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 10/17] net/mlx5: cmdif, Return value improvements Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 11/17] net/mlx5: cmdif, cmd_check refactoring Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 12/17] net/mlx5: cmdif, Add new api for command execution Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 13/17] net/mlx5: Use mlx5_cmd_do() in core create_{cq,dct} Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 14/17] net/mlx5: cmdif, Refactor error handling and reporting of async commands Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 15/17] RDMA/mlx5: Use new command interface API Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 16/17] net/mlx5: Add reset_state field to MFRL register Saeed Mahameed
2022-02-23 23:39 ` [for-next v2 17/17] net/mlx5: Add clarification on sync reset failure Saeed Mahameed
2022-02-28 22:41 ` [pull request][for-next v2 00/17] mlx5-next 2022-22-02 Saeed Mahameed
2022-03-01 0:31 ` Jakub Kicinski
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=20220223233930.319301-9-saeed@kernel.org \
--to=saeed@kernel.org \
--cc=davem@davemloft.net \
--cc=jgg@nvidia.com \
--cc=kuba@kernel.org \
--cc=leonro@nvidia.com \
--cc=linux-rdma@vger.kernel.org \
--cc=maorg@nvidia.com \
--cc=mbloch@nvidia.com \
--cc=netdev@vger.kernel.org \
--cc=saeedm@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).