public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Caleb Sander Mateos <csander@purestorage.com>
To: Saeed Mahameed <saeedm@nvidia.com>,
	Leon Romanovsky <leon@kernel.org>,
	Tariq Toukan <tariqt@nvidia.com>,
	Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>
Cc: Caleb Sander Mateos <csander@purestorage.com>,
	netdev@vger.kernel.org, linux-rdma@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH net-next 1/2] mlx5/core: avoid memory barrier in eq_update_ci()
Date: Thu, 31 Oct 2024 21:46:39 -0600	[thread overview]
Message-ID: <20241101034647.51590-1-csander@purestorage.com> (raw)

The memory barrier in eq_update_ci() after the doorbell write is a
significant hot spot in mlx5_eq_comp_int(). Under heavy TCP load, we see
3% of CPU time spent on the mfence. As explained in [1], this barrier is
only needed to preserve the ordering of writes to the doorbell register.

Use writel() instead of __raw_writel() for the doorbell write to provide
this ordering without the need for a full memory barrier.
memory-barriers.txt guarantees MMIO writes using writel() appear to the
device in the same order they were made. On strongly-ordered
architectures like x86, writel() adds no overhead to __raw_writel();
both translate into a single store instruction. Removing the mb() avoids
the costly mfence instruction.

[1]: https://lore.kernel.org/netdev/CALzJLG8af0SMfA1C8U8r_Fddb_ZQhvEZd6=2a97dOoBcgLA0xg@mail.gmail.com/

Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
index 4b7f7131c560..f03736711343 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
@@ -68,13 +68,12 @@ static inline struct mlx5_eqe *next_eqe_sw(struct mlx5_eq *eq)
 static inline void eq_update_ci(struct mlx5_eq *eq, int arm)
 {
 	__be32 __iomem *addr = eq->doorbell + (arm ? 0 : 2);
 	u32 val = (eq->cons_index & 0xffffff) | (eq->eqn << 24);
 
-	__raw_writel((__force u32)cpu_to_be32(val), addr);
-	/* We still want ordering, just not swabbing, so add a barrier */
-	mb();
+	/* Ensure ordering of consecutive doorbell writes */
+	writel((__force u32)cpu_to_be32(val), addr);
 }
 
 int mlx5_eq_table_init(struct mlx5_core_dev *dev);
 void mlx5_eq_table_cleanup(struct mlx5_core_dev *dev);
 int mlx5_eq_table_create(struct mlx5_core_dev *dev);
-- 
2.45.2


             reply	other threads:[~2024-11-01  3:48 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-11-01  3:46 Caleb Sander Mateos [this message]
2024-11-01  3:46 ` [PATCH net-next 2/2] mlx5/core: deduplicate {mlx5_,}eq_update_ci() Caleb Sander Mateos
2024-11-03  3:55   ` Parav Pandit
2024-11-03 22:18     ` Caleb Sander
2024-11-05  5:22       ` Parav Pandit
2024-11-05 16:06         ` Caleb Sander
2024-11-06  5:44           ` Parav Pandit
2024-11-06 23:44             ` Caleb Sander
2024-11-07  2:36               ` Saeed Mahameed
2024-11-07  2:43                 ` Parav Pandit
2024-11-07  4:45                 ` Caleb Sander
2024-11-07  5:14                   ` Saeed Mahameed
2024-11-07 18:30                     ` [PATCH net-next v2 1/2] mlx5/core: relax memory barrier in eq_update_ci() Caleb Sander Mateos
2024-11-07 18:30                       ` [PATCH net-next v2 2/2] mlx5/core: deduplicate {mlx5_,}eq_update_ci() Caleb Sander Mateos
2024-11-08 10:49                         ` Parav Pandit
2024-11-11 11:59                           ` Tariq Toukan
2024-11-08 10:46                       ` [PATCH net-next v2 1/2] mlx5/core: relax memory barrier in eq_update_ci() Parav Pandit
2024-11-11 11:59                         ` Tariq Toukan
2024-11-12  2:38                       ` Jakub Kicinski

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=20241101034647.51590-1-csander@purestorage.com \
    --to=csander@purestorage.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=kuba@kernel.org \
    --cc=leon@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=saeedm@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox