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>,
Maxim Mikityanskiy <maximmi@nvidia.com>
Subject: [net-next 14/16] net/mlx5e: Improve the MTU change shortcut
Date: Tue, 27 Sep 2022 13:36:09 -0700 [thread overview]
Message-ID: <20220927203611.244301-15-saeed@kernel.org> (raw)
In-Reply-To: <20220927203611.244301-1-saeed@kernel.org>
From: Maxim Mikityanskiy <maximmi@nvidia.com>
Normally, the MTU change requires reopening the channels, but it can be
skipped if the new MTU doesn't change any of the queue parameters and if
MTU is not used in the data path.
The shortcut is applicable to the non-linear mode of striding RQ,
because the only thing affected by MTU is the queue length. As ethtool
sets the queue length in packets, but striding RQ length is defined in
strides or bytes, we estimate the RQ length to be at least as big as the
requested number of MTU-sized packets, that's why it depends on MTU.
Improve the shortcut by actually checking whether the RQ length stayed
the same, instead of an intermediate step in the calculation.
As MTU also affects the SHAMPO parameters, skip the shortcut if SHAMPO
is in use.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en/params.c | 4 ++--
drivers/net/ethernet/mellanox/mlx5/core/en/params.h | 2 --
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 10 ++++++----
3 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index 5dd3567d02d8..9a58f8f978b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -57,8 +57,8 @@ static u32 mlx5e_rx_get_linear_stride_sz(struct mlx5e_params *params,
return roundup_pow_of_two(mlx5e_rx_get_linear_sz_skb(params, false));
}
-u8 mlx5e_mpwqe_log_pkts_per_wqe(struct mlx5e_params *params,
- struct mlx5e_xsk_param *xsk)
+static u8 mlx5e_mpwqe_log_pkts_per_wqe(struct mlx5e_params *params,
+ struct mlx5e_xsk_param *xsk)
{
u32 linear_stride_sz = mlx5e_rx_get_linear_stride_sz(params, xsk);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
index 3e148a00fa73..f2c1a23dca61 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h
@@ -101,8 +101,6 @@ void mlx5e_init_rq_type_params(struct mlx5_core_dev *mdev, struct mlx5e_params *
u16 mlx5e_get_linear_rq_headroom(struct mlx5e_params *params,
struct mlx5e_xsk_param *xsk);
-u8 mlx5e_mpwqe_log_pkts_per_wqe(struct mlx5e_params *params,
- struct mlx5e_xsk_param *xsk);
bool mlx5e_rx_is_linear_skb(struct mlx5e_params *params,
struct mlx5e_xsk_param *xsk);
bool mlx5e_rx_mpwqe_is_linear_skb(struct mlx5_core_dev *mdev,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index f8d45360a643..a38f0c6f06d9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4080,19 +4080,21 @@ int mlx5e_change_mtu(struct net_device *netdev, int new_mtu,
if (params->packet_merge.type == MLX5E_PACKET_MERGE_LRO)
reset = false;
- if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) {
+ if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ &&
+ params->packet_merge.type != MLX5E_PACKET_MERGE_SHAMPO) {
bool is_linear_old = mlx5e_rx_mpwqe_is_linear_skb(priv->mdev, params, NULL);
bool is_linear_new = mlx5e_rx_mpwqe_is_linear_skb(priv->mdev,
&new_params, NULL);
- u8 ppw_old = mlx5e_mpwqe_log_pkts_per_wqe(params, NULL);
- u8 ppw_new = mlx5e_mpwqe_log_pkts_per_wqe(&new_params, NULL);
+ u8 sz_old = mlx5e_mpwqe_get_log_rq_size(params, NULL);
+ u8 sz_new = mlx5e_mpwqe_get_log_rq_size(&new_params, NULL);
/* Always reset in linear mode - hw_mtu is used in data path.
* Check that the mode was non-linear and didn't change.
* If XSK is active, XSK RQs are linear.
+ * Reset if the RQ size changed, even if it's non-linear.
*/
if (!is_linear_old && !is_linear_new && !priv->xsk.refcnt &&
- ppw_old == ppw_new)
+ sz_old == sz_new)
reset = false;
}
--
2.37.3
next prev parent reply other threads:[~2022-09-27 20:37 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-27 20:35 [pull request][net-next 00/16] mlx5 updates 2022-09-27 Saeed Mahameed
2022-09-27 20:35 ` [net-next 01/16] net/mlx5: Add the log_min_mkey_entity_size capability Saeed Mahameed
2022-09-29 2:50 ` patchwork-bot+netdevbpf
2022-09-27 20:35 ` [net-next 02/16] net/mlx5e: Convert mlx5e_get_max_sq_wqebbs to u8 Saeed Mahameed
2022-09-27 20:35 ` [net-next 03/16] net/mlx5e: Remove unused fields from datapath structs Saeed Mahameed
2022-09-27 20:35 ` [net-next 04/16] net/mlx5e: Make mlx5e_verify_rx_mpwqe_strides static Saeed Mahameed
2022-09-27 20:36 ` [net-next 05/16] net/mlx5e: Validate striding RQ before enabling XDP Saeed Mahameed
2022-09-27 20:36 ` [net-next 06/16] net/mlx5e: Let mlx5e_get_sw_max_sq_mpw_wqebbs accept mdev Saeed Mahameed
2022-09-27 20:36 ` [net-next 07/16] net/mlx5e: Use mlx5e_stop_room_for_max_wqe where appropriate Saeed Mahameed
2022-09-27 20:36 ` [net-next 08/16] net/mlx5e: Fix a typo in mlx5e_xdp_mpwqe_is_full Saeed Mahameed
2022-09-27 20:36 ` [net-next 09/16] net/mlx5e: Use the aligned max TX MPWQE size Saeed Mahameed
2022-09-27 20:36 ` [net-next 10/16] net/mlx5e: kTLS, Check ICOSQ WQE size in advance Saeed Mahameed
2022-09-27 20:36 ` [net-next 11/16] net/mlx5e: Simplify stride size calculation for linear RQ Saeed Mahameed
2022-09-27 20:36 ` [net-next 12/16] net/mlx5e: xsk: Remove dead code in validation Saeed Mahameed
2022-09-27 20:36 ` [net-next 13/16] net/mlx5e: xsk: Fix SKB headroom calculation " Saeed Mahameed
2022-09-27 20:36 ` Saeed Mahameed [this message]
2022-09-27 20:36 ` [net-next 15/16] net/mlx5e: Make dma_info array dynamic in struct mlx5e_mpw_info Saeed Mahameed
2022-09-27 20:36 ` [net-next 16/16] net/mlx5e: Use runtime values of striding RQ parameters in datapath Saeed Mahameed
2022-09-29 2:35 ` [pull request][net-next 00/16] mlx5 updates 2022-09-27 Jakub Kicinski
2022-09-29 7:20 ` Saeed Mahameed
2022-09-29 15:33 ` 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=20220927203611.244301-15-saeed@kernel.org \
--to=saeed@kernel.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=kuba@kernel.org \
--cc=maximmi@nvidia.com \
--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;
as well as URLs for NNTP newsgroup(s).