AcoSail/kernel_optimize_test
8
0
Fork 0
forked from luck/tmp_suning_uos_patched
Code Pull Requests Activity

net/mlx4: Check if Granular QoS per VF has been enabled before updating QP qos_vport

Browse Source 
The Granular QoS per VF feature must be enabled in FW before it can be
used.

Thus, the driver cannot modify a QP's qos_vport value (via the UPDATE_QP FW
command) if the feature has not been enabled -- the FW returns an error if
this is attempted.

Fixes: 08068cd568 ("net/mlx4: Added qos_vport QP configuration in VST mode")
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Ido Shamay 2017-06-05 10:44:56 +03:00 committed by David S. Miller
parent 6c6ab3e73b
commit 269f9883fe
2 changed files with 17 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
drivers/net/ethernet/mellanox/mlx4/qp.c
View File

@ -484,6 +484,12 @@ int mlx4_update_qp(struct mlx4_dev *dev, u32 qpn,
}
if (attr & MLX4_UPDATE_QP_QOS_VPORT) {
if (!(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_QOS_VPP)) {
mlx4_warn(dev, "Granular QoS per VF is not enabled\n");
err = -EOPNOTSUPP;
goto out;
}
qp_mask |= 1ULL << MLX4_UPD_QP_MASK_QOS_VPP;
cmd->qp_context.qos_vport = params->qos_vport;
}

16
drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
View File

@ -5255,6 +5255,13 @@ void mlx4_delete_all_resources_for_slave(struct mlx4_dev *dev, int slave)
mutex_unlock(&priv->mfunc.master.res_tracker.slave_list[slave].mutex);
}
static void update_qos_vpp(struct mlx4_update_qp_context *ctx,
struct mlx4_vf_immed_vlan_work *work)
{
ctx->qp_mask |= cpu_to_be64(1ULL << MLX4_UPD_QP_MASK_QOS_VPP);
ctx->qp_context.qos_vport = work->qos_vport;
}
void mlx4_vf_immed_vlan_work_handler(struct work_struct *_work)
{
struct mlx4_vf_immed_vlan_work *work =
@ -5369,11 +5376,10 @@ void mlx4_vf_immed_vlan_work_handler(struct work_struct *_work)
qp->sched_queue & 0xC7;
upd_context->qp_context.pri_path.sched_queue |=
((work->qos & 0x7) << 3);
upd_context->qp_mask |=
cpu_to_be64(1ULL <<
MLX4_UPD_QP_MASK_QOS_VPP);
upd_context->qp_context.qos_vport =
work->qos_vport;
if (dev->caps.flags2 &
MLX4_DEV_CAP_FLAG2_QOS_VPP)
update_qos_vpp(upd_context, work);
}
err = mlx4_cmd(dev, mailbox->dma,