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
next prev 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.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=kuba@kernel.org \
--cc=mbloch@nvidia.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=saeedm@nvidia.com \
--cc=shayd@nvidia.com \
--cc=tariqt@nvidia.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.