[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <469dfbf5-4b5f-4457-8f88-f180b2c3a8ae@gmail.com>
Date: Thu, 24 Oct 2024 19:59:34 +0300
From: Tariq Toukan <ttoukan.linux@...il.com>
To: Caleb Sander Mateos <csander@...estorage.com>,
Saeed Mahameed <saeedm@...dia.com>, Leon Romanovsky <leon@...nel.org>,
Tariq Toukan <tariqt@...dia.com>, Andrew Lunn <andrew+netdev@...n.ch>,
"David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>
Cc: netdev@...r.kernel.org, linux-rdma@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mlx5: simplify EQ interrupt polling logic
On 23/10/2024 23:51, Caleb Sander Mateos wrote:
> Use a while loop in mlx5_eq_comp_int() and mlx5_eq_async_int() to
> clarify the EQE polling logic. This consolidates the next_eqe_sw() calls
> for the first and subequent iterations. It also avoids a goto. Turn the
> num_eqes < MLX5_EQ_POLLING_BUDGET check into a break condition.
>
> Signed-off-by: Caleb Sander Mateos <csander@...estorage.com>
> ---
> drivers/net/ethernet/mellanox/mlx5/core/eq.c | 22 +++++++-------------
> 1 file changed, 8 insertions(+), 14 deletions(-)
>
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
> index 68cb86b37e56..859dcf09b770 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
> @@ -114,15 +114,11 @@ static int mlx5_eq_comp_int(struct notifier_block *nb,
> struct mlx5_eq *eq = &eq_comp->core;
> struct mlx5_eqe *eqe;
> int num_eqes = 0;
> u32 cqn = -1;
>
> - eqe = next_eqe_sw(eq);
> - if (!eqe)
> - goto out;
> -
> - do {
> + while ((eqe = next_eqe_sw(eq))) {
> struct mlx5_core_cq *cq;
>
> /* Make sure we read EQ entry contents after we've
> * checked the ownership bit.
> */
> @@ -140,13 +136,14 @@ static int mlx5_eq_comp_int(struct notifier_block *nb,
> "Completion event for bogus CQ 0x%x\n", cqn);
> }
>
> ++eq->cons_index;
>
> - } while ((++num_eqes < MLX5_EQ_POLLING_BUDGET) && (eqe = next_eqe_sw(eq)));
> + if (++num_eqes >= MLX5_EQ_POLLING_BUDGET)
> + break;
> + }
>
> -out:
> eq_update_ci(eq, 1);
>
> if (cqn != -1)
> tasklet_schedule(&eq_comp->tasklet_ctx.task);
>
> @@ -213,15 +210,11 @@ static int mlx5_eq_async_int(struct notifier_block *nb,
> eqt = dev->priv.eq_table;
>
> recovery = action == ASYNC_EQ_RECOVER;
> mlx5_eq_async_int_lock(eq_async, recovery, &flags);
>
> - eqe = next_eqe_sw(eq);
> - if (!eqe)
> - goto out;
> -
> - do {
> + while ((eqe = next_eqe_sw(eq))) {
> /*
> * Make sure we read EQ entry contents after we've
> * checked the ownership bit.
> */
> dma_rmb();
> @@ -229,13 +222,14 @@ static int mlx5_eq_async_int(struct notifier_block *nb,
> atomic_notifier_call_chain(&eqt->nh[eqe->type], eqe->type, eqe);
> atomic_notifier_call_chain(&eqt->nh[MLX5_EVENT_TYPE_NOTIFY_ANY], eqe->type, eqe);
>
> ++eq->cons_index;
>
> - } while ((++num_eqes < MLX5_EQ_POLLING_BUDGET) && (eqe = next_eqe_sw(eq)));
> + if (++num_eqes >= MLX5_EQ_POLLING_BUDGET)
> + break;
> + }
>
> -out:
> eq_update_ci(eq, 1);
> mlx5_eq_async_int_unlock(eq_async, recovery, &flags);
>
> return unlikely(recovery) ? num_eqes : 0;
> }
LGTM.
Reviewed-by: Tariq Toukan <tariqt@...dia.com>
Thanks.
Powered by blists - more mailing lists