From: "Nabil S. Alramli" <dev@nalramli.com>
To: netdev@vger.kernel.org, saeedm@nvidia.com, saeed@kernel.org,
kuba@kernel.org, davem@davemloft.net, tariqt@nvidia.com,
linux-kernel@vger.kernel.org, leon@kernel.org
Cc: jdamato@fastly.com, sbhogavilli@fastly.com, nalramli@fastly.com,
"Nabil S. Alramli" <dev@nalramli.com>
Subject: [net-next RFC v2 4/4] mlx5: Add {get,set}_per_queue_coalesce()
Date: Mon, 18 Sep 2023 18:29:55 -0400 [thread overview]
Message-ID: <20230918222955.2066-5-dev@nalramli.com> (raw)
In-Reply-To: <20230918222955.2066-1-dev@nalramli.com>
The mlx-5 driver currently only implements querying or modifying coalesce
configurations globally. It does not allow per-queue operations. This
change is to implement per-queue coalesce operations in the driver.
Signed-off-by: Nabil S. Alramli <dev@nalramli.com>
---
.../ethernet/mellanox/mlx5/core/en_ethtool.c | 46 +++++++++++++++++--
1 file changed, 41 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index daa0aa833a42..492c03c3d5f4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -515,8 +515,8 @@ static int mlx5e_set_channels(struct net_device *dev,
static int mlx5e_ethtool_get_per_queue_coalesce(struct mlx5e_priv *priv,
int queue,
- struct ethtool_coalesce *coal,
- struct kernel_ethtool_coalesce *kernel_coal)
+ struct ethtool_coalesce *coal,
+ struct kernel_ethtool_coalesce *kernel_coal)
{
struct dim_cq_moder *rx_moder, *tx_moder;
struct mlx5e_params *params;
@@ -572,6 +572,23 @@ static int mlx5e_get_coalesce(struct net_device *netdev,
return mlx5e_ethtool_get_coalesce(priv, coal, kernel_coal);
}
+/**
+ * mlx5e_get_per_queue_coalesce - gets coalesce settings for particular queue
+ * @netdev: netdev structure
+ * @coal: ethtool's coalesce settings
+ * @queue: the particular queue to read
+ *
+ * Reads a specific queue's coalesce settings
+ **/
+static int mlx5e_get_per_queue_coalesce(struct net_device *netdev,
+ u32 queue,
+ struct ethtool_coalesce *coal)
+{
+ struct mlx5e_priv *priv = netdev_priv(netdev);
+
+ return mlx5e_ethtool_get_per_queue_coalesce(priv, queue, coal, NULL);
+}
+
#define MLX5E_MAX_COAL_TIME MLX5_MAX_CQ_PERIOD
#define MLX5E_MAX_COAL_FRAMES MLX5_MAX_CQ_COUNT
@@ -643,9 +660,9 @@ static int cqe_mode_to_period_mode(bool val)
static int mlx5e_ethtool_set_per_queue_coalesce(struct mlx5e_priv *priv,
int queue,
- struct ethtool_coalesce *coal,
- struct kernel_ethtool_coalesce *kernel_coal,
- struct netlink_ext_ack *extack)
+ struct ethtool_coalesce *coal,
+ struct kernel_ethtool_coalesce *kernel_coal,
+ struct netlink_ext_ack *extack)
{
struct dim_cq_moder *rx_moder, *tx_moder;
struct mlx5_core_dev *mdev = priv->mdev;
@@ -772,6 +789,23 @@ static int mlx5e_set_coalesce(struct net_device *netdev,
return mlx5e_ethtool_set_coalesce(priv, coal, kernel_coal, extack);
}
+/**
+ * mlx5e_set_per_queue_coalesce - set specific queue's coalesce settings
+ * @netdev: the netdev to change
+ * @coal: ethtool's coalesce settings
+ * @queue: the queue to change
+ *
+ * Sets the specified queue's coalesce settings.
+ **/
+static int mlx5e_set_per_queue_coalesce(struct net_device *netdev,
+ u32 queue,
+ struct ethtool_coalesce *coal)
+{
+ struct mlx5e_priv *priv = netdev_priv(netdev);
+
+ return mlx5e_ethtool_set_per_queue_coalesce(priv, queue, coal, NULL, NULL);
+}
+
static void ptys2ethtool_supported_link(struct mlx5_core_dev *mdev,
unsigned long *supported_modes,
u32 eth_proto_cap)
@@ -2506,6 +2540,8 @@ const struct ethtool_ops mlx5e_ethtool_ops = {
.flash_device = mlx5e_flash_device,
.get_priv_flags = mlx5e_get_priv_flags,
.set_priv_flags = mlx5e_set_priv_flags,
+ .get_per_queue_coalesce = mlx5e_get_per_queue_coalesce,
+ .set_per_queue_coalesce = mlx5e_set_per_queue_coalesce,
.self_test = mlx5e_self_test,
.get_fec_stats = mlx5e_get_fec_stats,
.get_fecparam = mlx5e_get_fecparam,
--
2.35.1
next prev parent reply other threads:[~2023-09-18 22:45 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-23 22:31 [net-next RFC 0/1] mlx5: support per queue coalesce settings Nabil S. Alramli
2023-08-23 22:31 ` [net-next RFC 1/1] mlx5: Add {get,set}_per_queue_coalesce() Nabil S. Alramli
2023-08-24 18:51 ` [net-next RFC 0/1] mlx5: support per queue coalesce settings Saeed Mahameed
2023-08-25 0:26 ` Nabil S. Alramli
2023-09-18 22:29 ` [net-next RFC v2 0/4] " Nabil S. Alramli
2023-09-18 22:29 ` [net-next RFC v2 1/4] mlx5: Add mlx5e_param to individual mlx5e_channel and preserve them through mlx5e_open_channels() Nabil S. Alramli
2023-09-18 22:29 ` [net-next RFC v2 2/4] mlx5: Add queue number parameter to mlx5e_safe_switch_params() Nabil S. Alramli
2023-09-18 22:29 ` [net-next RFC v2 3/4] mlx5: Implement mlx5e_ethtool_{get,set}_per_queue_coalesce() to support per-queue operations Nabil S. Alramli
2023-09-18 22:29 ` Nabil S. Alramli [this message]
2023-09-19 18:55 ` [net-next RFC v2 0/4] mlx5: support per queue coalesce settings Rahul Rameshbabu
2023-09-19 20:42 ` Nabil S. Alramli
2023-10-06 21:46 ` Rahul Rameshbabu
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=20230918222955.2066-5-dev@nalramli.com \
--to=dev@nalramli.com \
--cc=davem@davemloft.net \
--cc=jdamato@fastly.com \
--cc=kuba@kernel.org \
--cc=leon@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=nalramli@fastly.com \
--cc=netdev@vger.kernel.org \
--cc=saeed@kernel.org \
--cc=saeedm@nvidia.com \
--cc=sbhogavilli@fastly.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;
as well as URLs for NNTP newsgroup(s).