From: Leon Romanovsky <leon@kernel.org>
To: Doug Ledford <dledford@redhat.com>, Jason Gunthorpe <jgg@mellanox.com>
Cc: Leon Romanovsky <leonro@mellanox.com>,
RDMA mailing list <linux-rdma@vger.kernel.org>,
Maor Gottlieb <maorg@mellanox.com>,
Mark Bloch <markb@mellanox.com>,
Saeed Mahameed <saeedm@mellanox.com>,
linux-netdev <netdev@vger.kernel.org>
Subject: [PATCH rdma-next 09/12] RDMA/mlx5: Refactor netdev affinity code
Date: Thu, 28 Mar 2019 15:27:39 +0200 [thread overview]
Message-ID: <20190328132742.12070-10-leon@kernel.org> (raw)
In-Reply-To: <20190328132742.12070-1-leon@kernel.org>
From: Mark Bloch <markb@mellanox.com>
The design of representors is such that once an IB representor is created,
the netdev of representor already exists, we can use that fact to simplify
the netdev affinity code.
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
drivers/infiniband/hw/mlx5/ib_rep.c | 2 ++
drivers/infiniband/hw/mlx5/main.c | 47 +++++++++++++++++++++++------
2 files changed, 39 insertions(+), 10 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.c b/drivers/infiniband/hw/mlx5/ib_rep.c
index d3988f6ae2ae..7946cf26421b 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.c
+++ b/drivers/infiniband/hw/mlx5/ib_rep.c
@@ -71,6 +71,8 @@ mlx5_ib_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
ibdev->is_rep = true;
ibdev->port[0].rep = rep;
+ ibdev->port[0].roce.netdev =
+ mlx5_ib_get_rep_netdev(dev->priv.eswitch, rep->vport);
ibdev->mdev = dev;
ibdev->num_ports = num_ports;
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index d9dfc9975164..9ce8ae5565a3 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -156,6 +156,34 @@ static int get_port_state(struct ib_device *ibdev,
return ret;
}
+static struct mlx5_roce *mlx5_get_rep_roce(struct mlx5_ib_dev *dev,
+ struct net_device *ndev,
+ u8 *port_num)
+{
+ struct mlx5_eswitch *esw = dev->mdev->priv.eswitch;
+ struct net_device *rep_ndev;
+ struct mlx5_ib_port *port;
+ int i;
+
+ for (i = 0; i < dev->num_ports; i++) {
+ port = &dev->port[i];
+ if (!port->rep)
+ continue;
+
+ read_lock(&port->roce.netdev_lock);
+ rep_ndev = mlx5_ib_get_rep_netdev(esw,
+ port->rep->vport);
+ if (rep_ndev == ndev) {
+ read_unlock(&port->roce.netdev_lock);
+ *port_num = i + 1;
+ return &port->roce;
+ }
+ read_unlock(&port->roce.netdev_lock);
+ }
+
+ return NULL;
+}
+
static int mlx5_netdev_event(struct notifier_block *this,
unsigned long event, void *ptr)
{
@@ -172,22 +200,17 @@ static int mlx5_netdev_event(struct notifier_block *this,
switch (event) {
case NETDEV_REGISTER:
+ /* Should already be registered during the load */
+ if (ibdev->is_rep)
+ break;
write_lock(&roce->netdev_lock);
- if (ibdev->is_rep) {
- struct mlx5_eswitch *esw = ibdev->mdev->priv.eswitch;
- struct mlx5_eswitch_rep *rep = ibdev->port[0].rep;
- struct net_device *rep_ndev;
-
- rep_ndev = mlx5_ib_get_rep_netdev(esw, rep->vport);
- if (rep_ndev == ndev)
- roce->netdev = ndev;
- } else if (ndev->dev.parent == &mdev->pdev->dev) {
+ if (ndev->dev.parent == &mdev->pdev->dev)
roce->netdev = ndev;
- }
write_unlock(&roce->netdev_lock);
break;
case NETDEV_UNREGISTER:
+ /* In case of reps, ib device goes away before the netdevs */
write_lock(&roce->netdev_lock);
if (roce->netdev == ndev)
roce->netdev = NULL;
@@ -205,6 +228,10 @@ static int mlx5_netdev_event(struct notifier_block *this,
dev_put(lag_ndev);
}
+ if (ibdev->is_rep)
+ roce = mlx5_get_rep_roce(ibdev, ndev, &port_num);
+ if (!roce)
+ return NOTIFY_DONE;
if ((upper == ndev || (!upper && ndev == roce->netdev))
&& ibdev->ib_active) {
struct ib_event ibev = { };
--
2.20.1
next prev parent reply other threads:[~2019-03-28 13:28 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-28 13:27 [PATCH rdma-next 00/12] Move IB representors to single IB device multiple ports Leon Romanovsky
2019-03-28 13:27 ` [PATCH mlx5-next 01/12] net/mlx5: E-Switch, don't use hardcoded values for FDB prios Leon Romanovsky
2019-03-28 13:27 ` [PATCH mlx5-next 02/12] net/mlx5: E-Switch, add a new prio to be used by the RDMA side Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 03/12] RDMA/mlx5: Move netdev info into the port struct Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 04/12] RDMA/mlx5: Free IB device on remove Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 05/12] RDMA/mlx5: Move ports allocation to outside of INIT stage Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 06/12] RDMA/mlx5: Use correct size for device resources Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 07/12] RDMA/mlx5: Move rep into port struct Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 08/12] RDMA/mlx5: Move default representors SQ steering to rule to modify QP Leon Romanovsky
2019-03-28 13:27 ` Leon Romanovsky [this message]
2019-03-28 13:27 ` [PATCH rdma-next 10/12] RDMA/mlx5: Move SMI caps logic Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 11/12] RDMA/mlx5: Move to single device multiport ports in switchdev mode Leon Romanovsky
2019-03-28 13:27 ` [PATCH rdma-next 12/12] RDMA/mlx5: Remove VF representor profile Leon Romanovsky
2019-04-04 13:02 ` [PATCH rdma-next 00/12] Move IB representors to single IB device multiple ports Jason Gunthorpe
2019-04-04 17:42 ` Leon Romanovsky
2019-04-10 6:48 ` Leon Romanovsky
2019-04-10 18:06 ` Jason Gunthorpe
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=20190328132742.12070-10-leon@kernel.org \
--to=leon@kernel.org \
--cc=dledford@redhat.com \
--cc=jgg@mellanox.com \
--cc=leonro@mellanox.com \
--cc=linux-rdma@vger.kernel.org \
--cc=maorg@mellanox.com \
--cc=markb@mellanox.com \
--cc=netdev@vger.kernel.org \
--cc=saeedm@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;
as well as URLs for NNTP newsgroup(s).