linux-rdma.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Leon Romanovsky <leon@kernel.org>
To: Doug Ledford <dledford@redhat.com>, Jason Gunthorpe <jgg@nvidia.com>
Cc: Aharon Landau <aharonl@nvidia.com>,
	"David S. Miller" <davem@davemloft.net>,
	Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>,
	Gal Pressman <galpress@amazon.com>,
	Jakub Kicinski <kuba@kernel.org>,
	linux-kernel@vger.kernel.org, linux-rdma@vger.kernel.org,
	Maor Gottlieb <maorg@nvidia.com>,
	Mark Zhang <markzhang@nvidia.com>,
	Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>,
	Mustafa Ismail <mustafa.ismail@intel.com>,
	Naresh Kumar PBS <nareshkumar.pbs@broadcom.com>,
	Neta Ostrovsky <netao@nvidia.com>,
	netdev@vger.kernel.org, Potnuri Bharat Teja <bharat@chelsio.com>,
	Saeed Mahameed <saeedm@nvidia.com>,
	Selvin Xavier <selvin.xavier@broadcom.com>,
	Shiraz Saleem <shiraz.saleem@intel.com>,
	Yishai Hadas <yishaih@nvidia.com>,
	Zhu Yanjun <zyjzyj2000@gmail.com>
Subject: [PATCH rdma-next v1 10/11] RDMA/mlx5: Add modify_op_stat() support
Date: Wed, 15 Sep 2021 02:07:29 +0300	[thread overview]
Message-ID: <d3a6b47b67add7d155748f0485d4f3511c15c124.1631660727.git.leonro@nvidia.com> (raw)
In-Reply-To: <cover.1631660727.git.leonro@nvidia.com>

From: Aharon Landau <aharonl@nvidia.com>

Add support for ib callback modify_op_stat() to add or remove an
optional counter. When adding, a steering flow table is created
with a rule that catches and counts all the matching packets;
When removing, the table and flow counter are destroyed.

Signed-off-by: Aharon Landau <aharonl@nvidia.com>
Reviewed-by: Mark Zhang <markzhang@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/infiniband/hw/mlx5/counters.c | 79 +++++++++++++++++++++++++--
 drivers/infiniband/hw/mlx5/mlx5_ib.h  |  1 +
 include/rdma/ib_verbs.h               |  2 +
 3 files changed, 76 insertions(+), 6 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/counters.c b/drivers/infiniband/hw/mlx5/counters.c
index 6aa54ee441db..627077514e14 100644
--- a/drivers/infiniband/hw/mlx5/counters.c
+++ b/drivers/infiniband/hw/mlx5/counters.c
@@ -12,6 +12,7 @@
 struct mlx5_ib_counter {
 	const char *name;
 	size_t offset;
+	u32 type;
 };
 
 #define INIT_Q_COUNTER(_name)		\
@@ -75,19 +76,19 @@ static const struct mlx5_ib_counter ext_ppcnt_cnts[] = {
 	INIT_EXT_PPCNT_COUNTER(rx_icrc_encapsulated),
 };
 
-#define INIT_OP_COUNTER(_name)                                          \
-	{ .name = #_name }
+#define INIT_OP_COUNTER(_name, _type)		\
+	{ .name = #_name, .type = MLX5_IB_OPCOUNTER_##_type}
 
 static const struct mlx5_ib_counter basic_op_cnts[] = {
-	INIT_OP_COUNTER(cc_rx_ce_pkts),
+	INIT_OP_COUNTER(cc_rx_ce_pkts, CC_RX_CE_PKTS),
 };
 
 static const struct mlx5_ib_counter rdmarx_cnp_op_cnts[] = {
-	INIT_OP_COUNTER(cc_rx_cnp_pkts),
+	INIT_OP_COUNTER(cc_rx_cnp_pkts, CC_RX_CNP_PKTS),
 };
 
 static const struct mlx5_ib_counter rdmatx_cnp_op_cnts[] = {
-	INIT_OP_COUNTER(cc_tx_cnp_pkts),
+	INIT_OP_COUNTER(cc_tx_cnp_pkts, CC_TX_CNP_PKTS),
 };
 
 static int mlx5_ib_read_counters(struct ib_counters *counters,
@@ -453,6 +454,7 @@ static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev,
 	for (i = 0; i < ARRAY_SIZE(basic_op_cnts); i++, j++) {
 		descs[j].name = basic_op_cnts[i].name;
 		descs[j].flags |= IB_STAT_FLAG_OPTIONAL;
+		descs[j].priv = &basic_op_cnts[i].type;
 	}
 
 	if (MLX5_CAP_FLOWTABLE(dev->mdev,
@@ -460,6 +462,7 @@ static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev,
 		for (i = 0; i < ARRAY_SIZE(rdmarx_cnp_op_cnts); i++, j++) {
 			descs[j].name = rdmarx_cnp_op_cnts[i].name;
 			descs[j].flags |= IB_STAT_FLAG_OPTIONAL;
+			descs[j].priv = &rdmarx_cnp_op_cnts[i].type;
 		}
 	}
 
@@ -468,6 +471,7 @@ static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev,
 		for (i = 0; i < ARRAY_SIZE(rdmatx_cnp_op_cnts); i++, j++) {
 			descs[j].name = rdmatx_cnp_op_cnts[i].name;
 			descs[j].flags |= IB_STAT_FLAG_OPTIONAL;
+			descs[j].priv = &rdmatx_cnp_op_cnts[i].type;
 		}
 	}
 }
@@ -537,7 +541,7 @@ static void mlx5_ib_dealloc_counters(struct mlx5_ib_dev *dev)
 {
 	u32 in[MLX5_ST_SZ_DW(dealloc_q_counter_in)] = {};
 	int num_cnt_ports;
-	int i;
+	int i, j;
 
 	num_cnt_ports = is_mdev_switchdev_mode(dev->mdev) ? 1 : dev->num_ports;
 
@@ -552,6 +556,18 @@ static void mlx5_ib_dealloc_counters(struct mlx5_ib_dev *dev)
 		}
 		kfree(dev->port[i].cnts.descs);
 		kfree(dev->port[i].cnts.offsets);
+
+		for (j = 0; j < MLX5_IB_OPCOUNTER_MAX; j++) {
+			if (!dev->port[i].cnts.opfcs[j].fc)
+				continue;
+
+			mlx5_ib_fs_remove_op_fc(dev,
+						&dev->port[i].cnts.opfcs[j],
+						j);
+			mlx5_fc_destroy(dev->mdev,
+					dev->port[i].cnts.opfcs[j].fc);
+			dev->port[i].cnts.opfcs[j].fc = NULL;
+		}
 	}
 }
 
@@ -731,6 +747,56 @@ void mlx5_ib_counters_clear_description(struct ib_counters *counters)
 	mutex_unlock(&mcounters->mcntrs_mutex);
 }
 
+static int mlx5_ib_modify_stat(struct ib_device *device, u32 port,
+			       int index, bool enable)
+{
+	struct mlx5_ib_dev *dev = to_mdev(device);
+	struct mlx5_ib_counters *cnts;
+	struct mlx5_ib_op_fc *opfc;
+	u32 num_hw_counters, type;
+	int ret = 0;
+
+	cnts = &dev->port[port - 1].cnts;
+	num_hw_counters = cnts->num_q_counters + cnts->num_cong_counters +
+		cnts->num_ext_ppcnt_counters;
+	if ((index < num_hw_counters) ||
+	    (index >= num_hw_counters + cnts->num_op_counters))
+		return -EINVAL;
+
+	if (!(cnts->descs[index].flags & IB_STAT_FLAG_OPTIONAL))
+		return -EINVAL;
+
+	type = *(u32 *)cnts->descs[index].priv;
+	if (type >= MLX5_IB_OPCOUNTER_MAX)
+		return -EINVAL;
+
+	opfc = &cnts->opfcs[type];
+
+	if (enable) {
+		if (opfc->fc)
+			return -EEXIST;
+
+		opfc->fc = mlx5_fc_create(dev->mdev, false);
+		if (IS_ERR(opfc->fc))
+			return PTR_ERR(opfc->fc);
+
+		ret = mlx5_ib_fs_add_op_fc(dev, port, opfc, type);
+		if (ret) {
+			mlx5_fc_destroy(dev->mdev, opfc->fc);
+			opfc->fc = NULL;
+		}
+	} else {
+		if (!opfc->fc)
+			return -EINVAL;
+
+		mlx5_ib_fs_remove_op_fc(dev, opfc, type);
+		mlx5_fc_destroy(dev->mdev, opfc->fc);
+		opfc->fc = NULL;
+	}
+
+	return ret;
+}
+
 static const struct ib_device_ops hw_stats_ops = {
 	.alloc_hw_port_stats = mlx5_ib_alloc_hw_port_stats,
 	.get_hw_stats = mlx5_ib_get_hw_stats,
@@ -739,6 +805,7 @@ static const struct ib_device_ops hw_stats_ops = {
 	.counter_dealloc = mlx5_ib_counter_dealloc,
 	.counter_alloc_stats = mlx5_ib_counter_alloc_stats,
 	.counter_update_stats = mlx5_ib_counter_update_stats,
+	.modify_hw_stat = mlx5_ib_modify_stat,
 };
 
 static const struct ib_device_ops hw_switchdev_stats_ops = {
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index d81ff5078e5e..cf8b0653f0ce 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -821,6 +821,7 @@ struct mlx5_ib_counters {
 	u32 num_ext_ppcnt_counters;
 	u32 num_op_counters;
 	u16 set_id;
+	struct mlx5_ib_op_fc opfcs[MLX5_IB_OPCOUNTER_MAX];
 };
 
 int mlx5_ib_fs_add_op_fc(struct mlx5_ib_dev *dev, u32 port_num,
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index e825e8e7accf..3e8d570b5852 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -553,10 +553,12 @@ enum ib_stat_flag {
  * struct rdma_stat_desc
  * @name - The name of the counter
  * @flags - Flags of the counter; For example, IB_STAT_FLAG_OPTIONAL
+ * @priv - Driver private information; Core code should not use
  */
 struct rdma_stat_desc {
 	const char *name;
 	unsigned int flags;
+	const void *priv;
 };
 
 /**
-- 
2.31.1


  parent reply	other threads:[~2021-09-14 23:08 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-14 23:07 [PATCH rdma-next v1 00/11] Optional counter statistics support Leon Romanovsky
2021-09-14 23:07 ` [PATCH mlx5-next v1 01/11] net/mlx5: Add ifc bits to support optional counters Leon Romanovsky
2021-09-14 23:07 ` [PATCH mlx5-next v1 02/11] net/mlx5: Add priorities for counters in RDMA namespaces Leon Romanovsky
2021-09-14 23:07 ` [PATCH rdma-next v1 03/11] RDMA/counter: Add a descriptor in struct rdma_hw_stats Leon Romanovsky
2021-09-14 23:07 ` [PATCH rdma-next v1 04/11] RDMA/counter: Add an is_disabled field " Leon Romanovsky
2021-09-27 16:53   ` Jason Gunthorpe
2021-09-28  3:28     ` Mark Zhang
2021-09-28 11:51       ` Jason Gunthorpe
2021-09-14 23:07 ` [PATCH rdma-next v1 05/11] RDMA/counter: Add optional counter support Leon Romanovsky
2021-09-27 17:03   ` Jason Gunthorpe
2021-09-28  9:03     ` Mark Zhang
2021-09-28 11:51       ` Jason Gunthorpe
2021-09-29 12:14         ` Leon Romanovsky
2021-09-14 23:07 ` [PATCH rdma-next v1 06/11] RDMA/nldev: Add support to get status of all counters Leon Romanovsky
2021-09-27 17:30   ` Jason Gunthorpe
2021-09-28  9:12     ` Mark Zhang
2021-09-28 11:52       ` Jason Gunthorpe
2021-09-28 12:51         ` Mark Zhang
2021-09-29 12:26         ` Leon Romanovsky
2021-09-29 23:56           ` Jason Gunthorpe
2021-09-14 23:07 ` [PATCH rdma-next v1 07/11] RDMA/nldev: Allow optional-counter status configuration through RDMA netlink Leon Romanovsky
2021-09-27 17:20   ` Jason Gunthorpe
2021-09-29 12:27     ` Leon Romanovsky
2021-09-14 23:07 ` [PATCH rdma-next v1 08/11] RDMA/mlx5: Support optional counters in hw_stats initialization Leon Romanovsky
2021-09-14 23:07 ` [PATCH rdma-next v1 09/11] RDMA/mlx5: Add steering support in optional flow counters Leon Romanovsky
2021-09-14 23:07 ` Leon Romanovsky [this message]
2021-09-14 23:07 ` [PATCH rdma-next v1 11/11] RDMA/mlx5: Add optional counter support in get_hw_stats callback Leon Romanovsky

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=d3a6b47b67add7d155748f0485d4f3511c15c124.1631660727.git.leonro@nvidia.com \
    --to=leon@kernel.org \
    --cc=aharonl@nvidia.com \
    --cc=bharat@chelsio.com \
    --cc=davem@davemloft.net \
    --cc=dennis.dalessandro@cornelisnetworks.com \
    --cc=dledford@redhat.com \
    --cc=galpress@amazon.com \
    --cc=jgg@nvidia.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=maorg@nvidia.com \
    --cc=markzhang@nvidia.com \
    --cc=mike.marciniszyn@cornelisnetworks.com \
    --cc=mustafa.ismail@intel.com \
    --cc=nareshkumar.pbs@broadcom.com \
    --cc=netao@nvidia.com \
    --cc=netdev@vger.kernel.org \
    --cc=saeedm@nvidia.com \
    --cc=selvin.xavier@broadcom.com \
    --cc=shiraz.saleem@intel.com \
    --cc=yishaih@nvidia.com \
    --cc=zyjzyj2000@gmail.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).