public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed
From: Saeed Mahameed <saeed@kernel•org>
To: "David S. Miller" <davem@davemloft•net>,
	Jakub Kicinski <kuba@kernel•org>, Paolo Abeni <pabeni@redhat•com>,
	Eric Dumazet <edumazet@google•com>
Cc: Saeed Mahameed <saeedm@nvidia•com>,
	netdev@vger•kernel.org, Tariq Toukan <tariqt@nvidia•com>,
	Shay Drory <shayd@nvidia•com>, Mark Bloch <mbloch@nvidia•com>
Subject: [net 10/15] net/mlx5: E-switch, Devcom, sync devcom events and devcom comp register
Date: Mon, 22 May 2023 22:42:37 -0700	[thread overview]
Message-ID: <20230523054242.21596-11-saeed@kernel.org> (raw)
In-Reply-To: <20230523054242.21596-1-saeed@kernel.org>

From: Shay Drory <shayd@nvidia•com>

devcom events are sent to all registered component. Following the
cited patch, it is possible for two components, e.g.: two eswitches,
to send devcom events, while both components are registered. This
means eswitch layer will do double un/pairing, which is double
allocation and free of resources, even though only one un/pairing is
needed. flow example:

	cpu0					cpu1
	----					----

 mlx5_devlink_eswitch_mode_set(dev0)
  esw_offloads_devcom_init()
   mlx5_devcom_register_component(esw0)
                                         mlx5_devlink_eswitch_mode_set(dev1)
                                          esw_offloads_devcom_init()
                                           mlx5_devcom_register_component(esw1)
                                           mlx5_devcom_send_event()
   mlx5_devcom_send_event()

Hence, check whether the eswitches are already un/paired before
free/allocation of resources.

Fixes: 09b278462f16 ("net: devlink: enable parallel ops on netlink interface")
Signed-off-by: Shay Drory <shayd@nvidia•com>
Reviewed-by: Mark Bloch <mbloch@nvidia•com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia•com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h        | 1 +
 .../net/ethernet/mellanox/mlx5/core/eswitch_offloads.c   | 9 ++++++++-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 9f007c5438ee..add6cfa432a5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -342,6 +342,7 @@ struct mlx5_eswitch {
 		u32             large_group_num;
 	}  params;
 	struct blocking_notifier_head n_head;
+	bool paired[MLX5_MAX_PORTS];
 };
 
 void esw_offloads_disable(struct mlx5_eswitch *esw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 7c34c7cf506f..8d19c20d3447 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -2742,6 +2742,9 @@ static int mlx5_esw_offloads_devcom_event(int event,
 		    mlx5_eswitch_vport_match_metadata_enabled(peer_esw))
 			break;
 
+		if (esw->paired[mlx5_get_dev_index(peer_esw->dev)])
+			break;
+
 		err = mlx5_esw_offloads_set_ns_peer(esw, peer_esw, true);
 		if (err)
 			goto err_out;
@@ -2753,14 +2756,18 @@ static int mlx5_esw_offloads_devcom_event(int event,
 		if (err)
 			goto err_pair;
 
+		esw->paired[mlx5_get_dev_index(peer_esw->dev)] = true;
+		peer_esw->paired[mlx5_get_dev_index(esw->dev)] = true;
 		mlx5_devcom_set_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS, true);
 		break;
 
 	case ESW_OFFLOADS_DEVCOM_UNPAIR:
-		if (!mlx5_devcom_is_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS))
+		if (!esw->paired[mlx5_get_dev_index(peer_esw->dev)])
 			break;
 
 		mlx5_devcom_set_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS, false);
+		esw->paired[mlx5_get_dev_index(peer_esw->dev)] = false;
+		peer_esw->paired[mlx5_get_dev_index(esw->dev)] = false;
 		mlx5_esw_offloads_unpair(peer_esw);
 		mlx5_esw_offloads_unpair(esw);
 		mlx5_esw_offloads_set_ns_peer(esw, peer_esw, false);
-- 
2.40.1


  parent reply	other threads:[~2023-05-23  5:43 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-23  5:42 [pull request][net 00/15] mlx5 fixes 2023-05-22 Saeed Mahameed
2023-05-23  5:42 ` [net 01/15] net/mlx5: Collect command failures data only for known commands Saeed Mahameed
2023-05-24  7:50   ` patchwork-bot+netdevbpf
2023-05-23  5:42 ` [net 02/15] net/mlx5: Handle pairing of E-switch via uplink un/load APIs Saeed Mahameed
2023-05-23  5:42 ` [net 03/15] net/mlx5: DR, Fix crc32 calculation to work on big-endian (BE) CPUs Saeed Mahameed
2023-05-23  5:42 ` [net 04/15] net/mlx5: DR, Check force-loopback RC QP capability independently from RoCE Saeed Mahameed
2023-05-23  5:42 ` [net 05/15] net/mlx5e: Use correct encap attribute during invalidation Saeed Mahameed
2023-05-23  5:42 ` [net 06/15] net/mlx5: Fix error message when failing to allocate device memory Saeed Mahameed
2023-05-23  5:42 ` [net 07/15] net/mlx5e: Fix deadlock in tc route query code Saeed Mahameed
2023-05-23  5:42 ` [net 08/15] net/mlx5e: Fix SQ wake logic in ptp napi_poll context Saeed Mahameed
2023-05-23  5:42 ` [net 09/15] net/mlx5e: TC, Fix using eswitch mapping in nic mode Saeed Mahameed
2023-05-23  5:42 ` Saeed Mahameed [this message]
2023-05-23  5:42 ` [net 11/15] net/mlx5: Devcom, fix error flow in mlx5_devcom_register_device Saeed Mahameed
2023-05-23  5:42 ` [net 12/15] net/mlx5: Devcom, serialize devcom registration Saeed Mahameed
2023-05-23  5:42 ` [net 13/15] net/mlx5: Free irqs only on shutdown callback Saeed Mahameed
2023-05-23  5:42 ` [net 14/15] net/mlx5: Fix irq affinity management Saeed Mahameed
2023-05-23  5:42 ` [net 15/15] net/mlx5: Fix indexing of mlx5_irq Saeed Mahameed

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=20230523054242.21596-11-saeed@kernel.org \
    --to=saeed@kernel$(echo .)org \
    --cc=davem@davemloft$(echo .)net \
    --cc=edumazet@google$(echo .)com \
    --cc=kuba@kernel$(echo .)org \
    --cc=mbloch@nvidia$(echo .)com \
    --cc=netdev@vger$(echo .)kernel.org \
    --cc=pabeni@redhat$(echo .)com \
    --cc=saeedm@nvidia$(echo .)com \
    --cc=shayd@nvidia$(echo .)com \
    --cc=tariqt@nvidia$(echo .)com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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