From: Saeed Mahameed <saeed@kernel.org>
To: "David S. Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>
Cc: netdev@vger.kernel.org, Paul Blakey <paulb@nvidia.com>,
Roi Dayan <roid@nvidia.com>, Oz Shlomo <ozsh@nvidia.com>,
Saeed Mahameed <saeedm@nvidia.com>
Subject: [net-next 04/14] net/mlx5: Allow skipping counter refresh on creation
Date: Fri, 29 Oct 2021 13:56:22 -0700 [thread overview]
Message-ID: <20211029205632.390403-5-saeed@kernel.org> (raw)
In-Reply-To: <20211029205632.390403-1-saeed@kernel.org>
From: Paul Blakey <paulb@nvidia.com>
CT creates a counter for each CT rule, and for each such counter,
fs_counters tries to queue mlx5_fc_stats_work() work again via
mod_delayed_work(0) call to refresh all counters. This call has a
large performance impact when reaching high insertion rate and
accounts for ~8% of the insertion time when using software steering.
Allow skipping the refresh of all counters during counter creation.
Change CT to use this refresh skipping for it's counters.
Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 2 +-
.../net/ethernet/mellanox/mlx5/core/fs_counters.c | 14 +++++++++++---
include/linux/mlx5/fs.h | 4 ++++
3 files changed, 16 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index f44e5de25037..c1c6e74c79c4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -889,7 +889,7 @@ mlx5_tc_ct_counter_create(struct mlx5_tc_ct_priv *ct_priv)
return ERR_PTR(-ENOMEM);
counter->is_shared = false;
- counter->counter = mlx5_fc_create(ct_priv->dev, true);
+ counter->counter = mlx5_fc_create_ex(ct_priv->dev, true);
if (IS_ERR(counter->counter)) {
ct_dbg("Failed to create counter for ct entry");
ret = PTR_ERR(counter->counter);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c
index 60c9df1bc912..31c99d53faf7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c
@@ -301,7 +301,7 @@ static struct mlx5_fc *mlx5_fc_acquire(struct mlx5_core_dev *dev, bool aging)
return mlx5_fc_single_alloc(dev);
}
-struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging)
+struct mlx5_fc *mlx5_fc_create_ex(struct mlx5_core_dev *dev, bool aging)
{
struct mlx5_fc *counter = mlx5_fc_acquire(dev, aging);
struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats;
@@ -332,8 +332,6 @@ struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging)
goto err_out_alloc;
llist_add(&counter->addlist, &fc_stats->addlist);
-
- mod_delayed_work(fc_stats->wq, &fc_stats->work, 0);
}
return counter;
@@ -342,6 +340,16 @@ struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging)
mlx5_fc_release(dev, counter);
return ERR_PTR(err);
}
+
+struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging)
+{
+ struct mlx5_fc *counter = mlx5_fc_create_ex(dev, aging);
+ struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats;
+
+ if (aging)
+ mod_delayed_work(fc_stats->wq, &fc_stats->work, 0);
+ return counter;
+}
EXPORT_SYMBOL(mlx5_fc_create);
u32 mlx5_fc_id(struct mlx5_fc *counter)
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index a7e1155bc4da..cd2d4c572367 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -245,6 +245,10 @@ int mlx5_modify_rule_destination(struct mlx5_flow_handle *handler,
struct mlx5_flow_destination *old_dest);
struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging);
+
+/* As mlx5_fc_create() but doesn't queue stats refresh thread. */
+struct mlx5_fc *mlx5_fc_create_ex(struct mlx5_core_dev *dev, bool aging);
+
void mlx5_fc_destroy(struct mlx5_core_dev *dev, struct mlx5_fc *counter);
u64 mlx5_fc_query_lastuse(struct mlx5_fc *counter);
void mlx5_fc_query_cached(struct mlx5_fc *counter,
--
2.31.1
next prev parent reply other threads:[~2021-10-29 20:56 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-29 20:56 [pull request][net-next 00/14] mlx5 updates 2021-10-29 Saeed Mahameed
2021-10-29 20:56 ` [net-next 01/14] net/mlx5: Add esw assignment back in mlx5e_tc_sample_unoffload() Saeed Mahameed
2021-11-01 13:00 ` patchwork-bot+netdevbpf
2021-10-29 20:56 ` [net-next 02/14] net/mlx5: CT: Remove warning of ignore_flow_level support for VFs Saeed Mahameed
2021-10-29 20:56 ` [net-next 03/14] net/mlx5e: IPsec: Refactor checksum code in tx data path Saeed Mahameed
2021-10-29 20:56 ` Saeed Mahameed [this message]
2021-10-29 20:56 ` [net-next 05/14] net/mlx5: DR, Add check for unsupported fields in match param Saeed Mahameed
2021-10-29 20:56 ` [net-next 06/14] net/mlx5e: Refactor rx handler of represetor device Saeed Mahameed
2021-10-29 20:56 ` [net-next 07/14] net/mlx5e: Use generic name for the forwarding dev pointer Saeed Mahameed
2021-10-29 20:56 ` [net-next 08/14] net/mlx5: E-Switch, Add ovs internal port mapping to metadata support Saeed Mahameed
2021-10-29 20:56 ` [net-next 09/14] net/mlx5e: Accept action skbedit in the tc actions list Saeed Mahameed
2021-10-29 20:56 ` [net-next 10/14] net/mlx5e: Offload tc rules that redirect to ovs internal port Saeed Mahameed
2021-10-29 20:56 ` [net-next 11/14] net/mlx5e: Offload internal port as encap route device Saeed Mahameed
2021-10-29 20:56 ` [net-next 12/14] net/mlx5e: Add indirect tc offload of ovs internal port Saeed Mahameed
2021-10-29 20:56 ` [net-next 13/14] net/mlx5e: Term table handling of internal port rules Saeed Mahameed
2021-10-29 20:56 ` [net-next 14/14] net/mlx5: Support internal port as decap route device 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=20211029205632.390403-5-saeed@kernel.org \
--to=saeed@kernel.org \
--cc=davem@davemloft.net \
--cc=kuba@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=ozsh@nvidia.com \
--cc=paulb@nvidia.com \
--cc=roid@nvidia.com \
--cc=saeedm@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.