linux-rdma.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Eli Cohen <eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
To: dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	liranl-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org,
	Eli Cohen <eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Subject: [PATCH for-next 08/10] IB/mlx5: Implement callbacks for manipulating VFs
Date: Tue,  1 Mar 2016 18:52:21 +0200	[thread overview]
Message-ID: <1456851143-138332-9-git-send-email-eli@mellanox.com> (raw)
In-Reply-To: <1456851143-138332-1-git-send-email-eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Implement the IB defined callbacks used to manipulate the policy for the
link state, set GUIDs or get statistics information. This functionality
is added into a new file that will be used to add any SRIOV related
functionality to the mlx5 IB layer.

The following callbacks have been added:

mlx5_ib_get_vf_config
mlx5_ib_set_vf_link_state
mlx5_ib_get_vf_stats
mlx5_ib_set_vf_guid

In addition, publish whether this device is based on a virtual function.

In mlx5 supported devices, virtual functions are implemented as vHCAs.
vHCAs have their own QP number space so it is possible that two vHCAs
will use a QP with the same number at the same time.

Signed-off-by: Eli Cohen <eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
 drivers/infiniband/hw/mlx5/Makefile  |   2 +-
 drivers/infiniband/hw/mlx5/ib_virt.c | 194 +++++++++++++++++++++++++++++++++++
 drivers/infiniband/hw/mlx5/main.c    |  10 ++
 drivers/infiniband/hw/mlx5/mlx5_ib.h |   8 ++
 include/linux/mlx5/driver.h          |   5 +-
 include/linux/mlx5/mlx5_ifc.h        |   6 ++
 6 files changed, 223 insertions(+), 2 deletions(-)
 create mode 100644 drivers/infiniband/hw/mlx5/ib_virt.c

diff --git a/drivers/infiniband/hw/mlx5/Makefile b/drivers/infiniband/hw/mlx5/Makefile
index 27a70159e2ea..82e074f47cf2 100644
--- a/drivers/infiniband/hw/mlx5/Makefile
+++ b/drivers/infiniband/hw/mlx5/Makefile
@@ -1,4 +1,4 @@
 obj-$(CONFIG_MLX5_INFINIBAND)	+= mlx5_ib.o
 
-mlx5_ib-y :=	main.o cq.o doorbell.o qp.o mem.o srq.o mr.o ah.o mad.o
+mlx5_ib-y :=	main.o cq.o doorbell.o qp.o mem.o srq.o mr.o ah.o mad.o ib_virt.o
 mlx5_ib-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += odp.o
diff --git a/drivers/infiniband/hw/mlx5/ib_virt.c b/drivers/infiniband/hw/mlx5/ib_virt.c
new file mode 100644
index 000000000000..c1b9de800fe5
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/ib_virt.c
@@ -0,0 +1,194 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/module.h>
+#include <linux/mlx5/vport.h>
+#include "mlx5_ib.h"
+
+static inline u32 mlx_to_net_policy(enum port_state_policy mlx_policy)
+{
+	switch (mlx_policy) {
+	case MLX5_POLICY_DOWN:
+		return IFLA_VF_LINK_STATE_DISABLE;
+	case MLX5_POLICY_UP:
+		return IFLA_VF_LINK_STATE_ENABLE;
+	case MLX5_POLICY_FOLLOW:
+		return IFLA_VF_LINK_STATE_AUTO;
+	default:
+		return __IFLA_VF_LINK_STATE_MAX;
+	}
+}
+
+int mlx5_ib_get_vf_config(struct ib_device *device, int vf, u8 port,
+			  struct ifla_vf_info *info)
+{
+	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;
+
+	err = mlx5_query_hca_vport_context(mdev, 1, 1,  vf + 1, rep);
+	if (err) {
+		mlx5_ib_warn(dev, "failed to query port policy for vf %d (%d)\n",
+			     vf, err);
+		goto free;
+	}
+	memset(info, 0, sizeof(*info));
+	info->linkstate = mlx_to_net_policy(rep->policy);
+	if (info->linkstate == __IFLA_VF_LINK_STATE_MAX)
+		err = -EINVAL;
+
+free:
+	kfree(rep);
+	return err;
+}
+
+static inline enum port_state_policy net_to_mlx_policy(int policy)
+{
+	switch (policy) {
+	case IFLA_VF_LINK_STATE_DISABLE:
+		return MLX5_POLICY_DOWN;
+	case IFLA_VF_LINK_STATE_ENABLE:
+		return MLX5_POLICY_UP;
+	case IFLA_VF_LINK_STATE_AUTO:
+		return MLX5_POLICY_FOLLOW;
+	default:
+		return MLX5_POLICY_INVALID;
+	}
+}
+
+int mlx5_ib_set_vf_link_state(struct ib_device *device, int vf,
+			      u8 port, int state)
+{
+	struct mlx5_ib_dev *dev = to_mdev(device);
+	struct mlx5_core_dev *mdev = dev->mdev;
+	struct mlx5_hca_vport_context *in;
+	int err;
+
+	in = kzalloc(sizeof(*in), GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	in->policy = net_to_mlx_policy(state);
+	if (in->policy == MLX5_POLICY_INVALID) {
+		err = -EINVAL;
+		goto out;
+	}
+	in->field_select = MLX5_HCA_VPORT_SEL_STATE_POLICY;
+	err = mlx5_core_modify_hca_vport_context(mdev, 1, 1, vf + 1, in);
+
+out:
+	kfree(in);
+	return err;
+}
+
+int mlx5_ib_get_vf_stats(struct ib_device *device, int vf,
+			 u8 port, struct ifla_vf_stats *stats)
+{
+	int out_sz = MLX5_ST_SZ_BYTES(query_vport_counter_out);
+	struct mlx5_core_dev *mdev;
+	struct mlx5_ib_dev *dev;
+	void *out;
+	int err;
+
+	dev = to_mdev(device);
+	mdev = dev->mdev;
+
+	out = kzalloc(out_sz, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
+
+	err = mlx5_core_query_vport_counter(mdev, true, vf, port, out, out_sz);
+	if (err)
+		goto ex;
+
+	stats->rx_packets = MLX5_GET64_PR(query_vport_counter_out, out, received_ib_unicast.packets);
+	stats->tx_packets = MLX5_GET64_PR(query_vport_counter_out, out, transmitted_ib_unicast.packets);
+	stats->rx_bytes = MLX5_GET64_PR(query_vport_counter_out, out, received_ib_unicast.octets);
+	stats->tx_bytes = MLX5_GET64_PR(query_vport_counter_out, out, transmitted_ib_unicast.octets);
+	stats->multicast = MLX5_GET64_PR(query_vport_counter_out, out, received_ib_multicast.packets);
+
+ex:
+	kfree(out);
+	return err;
+}
+
+static int set_vf_node_guid(struct ib_device *device, int vf, u8 port, u64 guid)
+{
+	struct mlx5_ib_dev *dev = to_mdev(device);
+	struct mlx5_core_dev *mdev = dev->mdev;
+	struct mlx5_hca_vport_context *in;
+	int err;
+
+	in = kzalloc(sizeof(*in), GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	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);
+	kfree(in);
+	return err;
+}
+
+static int set_vf_port_guid(struct ib_device *device, int vf, u8 port, u64 guid)
+{
+	struct mlx5_ib_dev *dev = to_mdev(device);
+	struct mlx5_core_dev *mdev = dev->mdev;
+	struct mlx5_hca_vport_context *in;
+	int err;
+
+	in = kzalloc(sizeof(*in), GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	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);
+	kfree(in);
+	return err;
+}
+
+int mlx5_ib_set_vf_guid(struct ib_device *device, int vf, u8 port,
+			u64 guid, int type)
+{
+	if (type == IFLA_VF_IB_NODE_GUID)
+		return set_vf_node_guid(device, vf, port, guid);
+	else if (type == IFLA_VF_IB_PORT_GUID)
+		return set_vf_port_guid(device, vf, port, guid);
+
+	return -EINVAL;
+}
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 399049b70a72..00ca1e998254 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -549,6 +549,9 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 	if (MLX5_CAP_GEN(mdev, cd))
 		props->device_cap_flags |= IB_DEVICE_CROSS_CHANNEL;
 
+	if (!mlx5_core_is_pf(mdev))
+		props->device_cap_flags |= IB_DEVICE_VIRTUAL_FUNCTION;
+
 	return 0;
 }
 
@@ -686,6 +689,7 @@ static int mlx5_query_hca_port(struct ib_device *ibdev, u8 port,
 	props->qkey_viol_cntr	= rep->qkey_violation_counter;
 	props->subnet_timeout	= rep->subnet_timeout;
 	props->init_type_reply	= rep->init_type_reply;
+	props->grh_required	= rep->grh_required;
 
 	err = mlx5_query_port_link_width_oper(mdev, &ib_link_width_oper, port);
 	if (err)
@@ -2266,6 +2270,12 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
 	dev->ib_dev.map_mr_sg		= mlx5_ib_map_mr_sg;
 	dev->ib_dev.check_mr_status	= mlx5_ib_check_mr_status;
 	dev->ib_dev.get_port_immutable  = mlx5_port_immutable;
+	if (mlx5_core_is_pf(mdev)) {
+		dev->ib_dev.get_vf_config	= mlx5_ib_get_vf_config;
+		dev->ib_dev.set_vf_link_state	= mlx5_ib_set_vf_link_state;
+		dev->ib_dev.get_vf_stats	= mlx5_ib_get_vf_stats;
+		dev->ib_dev.set_vf_guid		= mlx5_ib_set_vf_guid;
+	}
 
 	mlx5_ib_internal_fill_odp_caps(dev);
 
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index d2b9737baa36..752e3e33bcbc 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -719,6 +719,14 @@ void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp);
 void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp);
 void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start,
 			      unsigned long end);
+int mlx5_ib_get_vf_config(struct ib_device *device, int vf,
+			  u8 port, struct ifla_vf_info *info);
+int mlx5_ib_set_vf_link_state(struct ib_device *device, int vf,
+			      u8 port, int state);
+int mlx5_ib_get_vf_stats(struct ib_device *device, int vf,
+			 u8 port, struct ifla_vf_stats *stats);
+int mlx5_ib_set_vf_guid(struct ib_device *device, int vf, u8 port,
+			u64 guid, int type);
 
 #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
 static inline void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev)
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 8edcd08853dd..ea00549e45aa 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -613,7 +613,10 @@ struct mlx5_pas {
 };
 
 enum port_state_policy {
-	MLX5_AAA_000
+	MLX5_POLICY_DOWN	= 0,
+	MLX5_POLICY_UP		= 1,
+	MLX5_POLICY_FOLLOW	= 2,
+	MLX5_POLICY_INVALID	= 0xffffffff
 };
 
 enum phy_port_state {
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 9f404de5f99b..7a146074d428 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -3661,6 +3661,12 @@ struct mlx5_ifc_query_hca_vport_pkey_in_bits {
 	u8         pkey_index[0x10];
 };
 
+enum {
+	MLX5_HCA_VPORT_SEL_PORT_GUID	= 1 << 0,
+	MLX5_HCA_VPORT_SEL_NODE_GUID	= 1 << 1,
+	MLX5_HCA_VPORT_SEL_STATE_POLICY	= 1 << 2,
+};
+
 struct mlx5_ifc_query_hca_vport_gid_out_bits {
 	u8         status[0x8];
 	u8         reserved_at_8[0x18];
-- 
1.8.3.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2016-03-01 16:52 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-01 16:52 [PATCH for-next 00/10] IB SR-IOV support Eli Cohen
     [not found] ` <1456851143-138332-1-git-send-email-eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-03-01 16:52   ` [PATCH for-next 01/10] net/core: Add support for configuring VF GUIDs Eli Cohen
     [not found]     ` <1456851143-138332-2-git-send-email-eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-03-01 17:37       ` Jason Gunthorpe
     [not found]         ` <20160301173751.GA25176-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2016-03-01 17:49           ` Eli Cohen
     [not found]             ` <20160301174951.GA19366-lgQlq6cFzJSjLWYaRI30zHI+JuX82XLG@public.gmane.org>
2016-03-01 18:25               ` Jason Gunthorpe
     [not found]                 ` <20160301182516.GA12495-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2016-03-01 21:08                   ` Or Gerlitz
     [not found]                     ` <CAJ3xEMgrAUCj7PS6fegmuSUsjMruH3gzSHZmuzAX+ZbHZOpL9w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-03-02 16:50                       ` Doug Ledford
     [not found]                         ` <56D719E3.2000206-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-03-02 18:40                           ` Or Gerlitz
     [not found]                             ` <CAJ3xEMh5vJAZVO03=rRVCvqqXzXvah3idrMtMQfFP-wBxR7R_Q-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-03-04 14:35                               ` Doug Ledford
     [not found]                                 ` <56D99D3D.4000606-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-03-07  7:23                                   ` Or Gerlitz
2021-10-26 15:16     ` Eugene Syromiatnikov
2016-03-01 16:52   ` [PATCH for-next 02/10] IB/mlx5: Fix decision on using MAD_IFC Eli Cohen
2016-03-01 16:52   ` [PATCH for-next 03/10] IB/core: Support accessing SA in virtualized environment Eli Cohen
     [not found]     ` <1456851143-138332-4-git-send-email-eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-03-01 17:44       ` Jason Gunthorpe
     [not found]         ` <20160301174401.GC25176-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2016-03-01 18:17           ` Eli Cohen
     [not found]             ` <20160301181742.GB19366-lgQlq6cFzJSjLWYaRI30zHI+JuX82XLG@public.gmane.org>
2016-03-01 18:32               ` Jason Gunthorpe
     [not found]                 ` <20160301183256.GB12495-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2016-03-01 19:07                   ` Eli Cohen
     [not found]                     ` <20160301190742.GC19366-lgQlq6cFzJSjLWYaRI30zHI+JuX82XLG@public.gmane.org>
2016-03-01 19:31                       ` Jason Gunthorpe
     [not found]                         ` <20160301193153.GA25755-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2016-03-01 19:46                           ` Eli Cohen
     [not found]                             ` <20160301194608.GF19366-lgQlq6cFzJSjLWYaRI30zHI+JuX82XLG@public.gmane.org>
2016-03-01 21:15                               ` Or Gerlitz
2016-03-04 14:37                               ` Doug Ledford
     [not found]                                 ` <56D99D8E.5020900-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-03-07  7:52                                   ` Or Gerlitz
2016-03-01 16:52   ` [PATCH for-next 04/10] IB/core: Add interfaces to control VF attributes Eli Cohen
2016-03-01 16:52   ` [PATCH for-next 05/10] IB/ipoib: Add ndo operations for configuring VFs Eli Cohen
2016-03-01 16:52   ` [PATCH for-next 06/10] net/mlx5_core: Add VF param when querying vport counter Eli Cohen
2016-03-01 16:52   ` [PATCH for-next 07/10] net/mlx5_core: Implement modify HCA vport command Eli Cohen
2016-03-01 16:52   ` Eli Cohen [this message]
     [not found]     ` <1456851143-138332-9-git-send-email-eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-03-06 12:33       ` [PATCH for-next 08/10] IB/mlx5: Implement callbacks for manipulating VFs Yuval Shaia
2016-03-01 16:52   ` [PATCH for-next 09/10] IB/ipoib: Allow mcast packets from other VFs Eli Cohen
     [not found]     ` <1456851143-138332-10-git-send-email-eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-03-06 11:50       ` Yuval Shaia
     [not found]         ` <20160306115006.GA23975-Hxa29pjIrETlQW142y8m19+IiqhCXseY@public.gmane.org>
2016-03-06 12:13           ` Or Gerlitz
2016-03-01 16:52   ` [PATCH for-next 10/10] IB/core: Use GRH when the path hop-limit > 0 Eli Cohen
     [not found]     ` <1456851143-138332-11-git-send-email-eli-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-03-01 17:38       ` Jason Gunthorpe
     [not found]         ` <20160301173846.GB25176-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2016-03-03 15:55           ` Doug Ledford

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=1456851143-138332-9-git-send-email-eli@mellanox.com \
    --to=eli-vpraknaxozvwk0htik3j/w@public.gmane.org \
    --cc=dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=liranl-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
    /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).