* [PATCH mlx5-next] IB/mlx5: Return the administrative GUID if exists
@ 2020-01-16 12:00 Leon Romanovsky
2020-01-25 19:04 ` Jason
0 siblings, 1 reply; 2+ messages in thread
From: Leon Romanovsky @ 2020-01-16 12:00 UTC (permalink / raw)
To: Doug Ledford, Jason Gunthorpe
Cc: Danit Goldberg, RDMA mailing list, Saeed Mahameed, linux-netdev,
Leon Romanovsky
From: Danit Goldberg <danitg@mellanox.com>
A user can achieve the operational GUID (a.k.a affective GUID) through
link/infiniband. Therefore it is preferred to return the administrative
GUID if exists instead of the operational.
This way the PF can query which VF GUID will be set in the next bind.
In order to align with MAC address, zero is returned if
administrative GUID is not set.
For example:
- Before setting administrative GUID:
ip link show
ib0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 4092 qdisc mq state UP mode DEFAULT group default qlen 256
link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
vf 0 link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff,
spoof checking off, NODE_GUID 00:00:00:00:00:00:00:00, PORT_GUID 00:00:00:00:00:00:00:00, link-state auto, trust off, query_rss off
ip link set ib0 vf 0 node_guid 11:00:af:21:cb:05:11:00
ip link set ib0 vf 0 port_guid 22:11:af:21:cb:05:11:00
- After setting administrative GUID:
ip link show
ib0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 4092 qdisc mq state UP mode DEFAULT group default qlen 256
link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
vf 0 link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff,
spoof checking off, NODE_GUID 11:00:af:21:cb:05:11:00, PORT_GUID 22:11:af:21:cb:05:11:00, link-state auto, trust off, query_rss off
Fixes: 9c0015ef0928 ("IB/mlx5: Implement callbacks for getting VFs GUID attributes")
Signed-off-by: Danit Goldberg <danitg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
drivers/infiniband/hw/mlx5/ib_virt.c | 28 ++++++++++++----------------
include/linux/mlx5/driver.h | 5 +++++
2 files changed, 17 insertions(+), 16 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/ib_virt.c b/drivers/infiniband/hw/mlx5/ib_virt.c
index 4f0edd4832bd..b61165359954 100644
--- a/drivers/infiniband/hw/mlx5/ib_virt.c
+++ b/drivers/infiniband/hw/mlx5/ib_virt.c
@@ -164,8 +164,10 @@ static int set_vf_node_guid(struct ib_device *device, int vf, u8 port, u64 guid)
in->field_select = MLX5_HCA_VPORT_SEL_NODE_GUID;
in->node_guid = guid;
err = mlx5_core_modify_hca_vport_context(mdev, 1, 1, vf + 1, in);
- if (!err)
+ if (!err) {
vfs_ctx[vf].node_guid = guid;
+ vfs_ctx[vf].node_guid_valid = 1;
+ }
kfree(in);
return err;
}
@@ -185,8 +187,10 @@ static int set_vf_port_guid(struct ib_device *device, int vf, u8 port, u64 guid)
in->field_select = MLX5_HCA_VPORT_SEL_PORT_GUID;
in->port_guid = guid;
err = mlx5_core_modify_hca_vport_context(mdev, 1, 1, vf + 1, in);
- if (!err)
+ if (!err) {
vfs_ctx[vf].port_guid = guid;
+ vfs_ctx[vf].port_guid_valid = 1;
+ }
kfree(in);
return err;
}
@@ -208,20 +212,12 @@ int mlx5_ib_get_vf_guid(struct ib_device *device, int vf, u8 port,
{
struct mlx5_ib_dev *dev = to_mdev(device);
struct mlx5_core_dev *mdev = dev->mdev;
- struct mlx5_hca_vport_context *rep;
- int err;
-
- rep = kzalloc(sizeof(*rep), GFP_KERNEL);
- if (!rep)
- return -ENOMEM;
+ struct mlx5_vf_context *vfs_ctx = mdev->priv.sriov.vfs_ctx;
- err = mlx5_query_hca_vport_context(mdev, 1, 1, vf+1, rep);
- if (err)
- goto ex;
+ node_guid->guid =
+ vfs_ctx[vf].node_guid_valid ? vfs_ctx[vf].node_guid : 0;
+ port_guid->guid =
+ vfs_ctx[vf].port_guid_valid ? vfs_ctx[vf].port_guid : 0;
- port_guid->guid = rep->port_guid;
- node_guid->guid = rep->node_guid;
-ex:
- kfree(rep);
- return err;
+ return 0;
}
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 27200dea0297..a24937fc56b9 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -461,6 +461,11 @@ struct mlx5_vf_context {
int enabled;
u64 port_guid;
u64 node_guid;
+ /* Valid bits are used to validate administrative guid only.
+ * Enabled after ndo_set_vf_guid
+ */
+ u8 port_guid_valid:1;
+ u8 node_guid_valid:1;
enum port_state_policy policy;
};
--
2.20.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH mlx5-next] IB/mlx5: Return the administrative GUID if exists
2020-01-16 12:00 [PATCH mlx5-next] IB/mlx5: Return the administrative GUID if exists Leon Romanovsky
@ 2020-01-25 19:04 ` Jason
0 siblings, 0 replies; 2+ messages in thread
From: Jason @ 2020-01-25 19:04 UTC (permalink / raw)
To: Leon Romanovsky
Cc: Doug Ledford, Jason Gunthorpe, Danit Goldberg, RDMA mailing list,
Saeed Mahameed, linux-netdev, Leon Romanovsky
On Thu, Jan 16, 2020 at 02:00:48PM +0200, Leon Romanovsky wrote:
> From: Danit Goldberg <danitg@mellanox.com>
>
> A user can achieve the operational GUID (a.k.a affective GUID) through
> link/infiniband. Therefore it is preferred to return the administrative
> GUID if exists instead of the operational.
> This way the PF can query which VF GUID will be set in the next bind.
> In order to align with MAC address, zero is returned if
> administrative GUID is not set.
>
> For example:
> - Before setting administrative GUID:
> ip link show
> ib0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 4092 qdisc mq state UP mode DEFAULT group default qlen 256
> link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
> vf 0 link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff,
> spoof checking off, NODE_GUID 00:00:00:00:00:00:00:00, PORT_GUID 00:00:00:00:00:00:00:00, link-state auto, trust off, query_rss off
>
> ip link set ib0 vf 0 node_guid 11:00:af:21:cb:05:11:00
> ip link set ib0 vf 0 port_guid 22:11:af:21:cb:05:11:00
>
> - After setting administrative GUID:
> ip link show
> ib0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 4092 qdisc mq state UP mode DEFAULT group default qlen 256
> link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
> vf 0 link/infiniband 00:00:00:08:fe:80:00:00:00:00:00:00:52:54:00:c0:fe:12:34:55 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff,
> spoof checking off, NODE_GUID 11:00:af:21:cb:05:11:00, PORT_GUID 22:11:af:21:cb:05:11:00, link-state auto, trust off, query_rss off
>
> Fixes: 9c0015ef0928 ("IB/mlx5: Implement callbacks for getting VFs GUID attributes")
> Signed-off-by: Danit Goldberg <danitg@mellanox.com>
> Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
> ---
> drivers/infiniband/hw/mlx5/ib_virt.c | 28 ++++++++++++----------------
> include/linux/mlx5/driver.h | 5 +++++
> 2 files changed, 17 insertions(+), 16 deletions(-)
Applied to for-next, thanks
Jason
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2020-01-25 19:04 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-01-16 12:00 [PATCH mlx5-next] IB/mlx5: Return the administrative GUID if exists Leon Romanovsky
2020-01-25 19:04 ` Jason
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).