[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87wmi6njda.fsf@nvidia.com>
Date: Thu, 17 Oct 2024 21:08:17 -0700
From: Rahul Rameshbabu <rrameshbabu@...dia.com>
To: Vadim Fedorenko <vadfed@...a.com>
Cc: Saeed Mahameed <saeedm@...dia.com>, Leon Romanovsky <leon@...nel.org>,
Tariq Toukan <tariqt@...dia.com>, "David S. Miller"
<davem@...emloft.net>, Vadim Fedorenko <vadim.fedorenko@...ux.dev>,
Jakub Kicinski <kuba@...nel.org>, Eric Dumazet <edumazet@...gle.com>,
<netdev@...r.kernel.org>
Subject: Re: [PATCH net-next] mlx5_en: use read sequence for gettimex64
On Mon, 14 Oct, 2024 10:01:03 -0700 Vadim Fedorenko <vadfed@...a.com> wrote:
> The gettimex64() doesn't modify values in timecounter, that's why there
> is no need to update sequence counter. Reduce the contention on sequence
> lock for multi-thread PHC reading use-case.
>
> Signed-off-by: Vadim Fedorenko <vadfed@...a.com>
> ---
> drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c | 6 +-----
> 1 file changed, 1 insertion(+), 5 deletions(-)
>
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
> index b306ae79bf97..4822d01123b4 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
> @@ -402,9 +402,7 @@ static int mlx5_ptp_gettimex(struct ptp_clock_info *ptp, struct timespec64 *ts,
> struct ptp_system_timestamp *sts)
> {
> struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, ptp_info);
> - struct mlx5_timer *timer = &clock->timer;
> struct mlx5_core_dev *mdev;
> - unsigned long flags;
> u64 cycles, ns;
>
> mdev = container_of(clock, struct mlx5_core_dev, clock);
> @@ -413,10 +411,8 @@ static int mlx5_ptp_gettimex(struct ptp_clock_info *ptp, struct timespec64 *ts,
> goto out;
> }
>
> - write_seqlock_irqsave(&clock->lock, flags);
> cycles = mlx5_read_time(mdev, sts, false);
> - ns = timecounter_cyc2time(&timer->tc, cycles);
> - write_sequnlock_irqrestore(&clock->lock, flags);
> + ns = mlx5_timecounter_cyc2time(clock, cycles);
> *ts = ns_to_timespec64(ns);
> out:
> return 0;
The patch seems like a good cleanup to me. Like Vadim mentioned, we
should not need to update the timecounter since this simply a read
operation.
Reviewed-by: Rahul Rameshbabu <rrameshbabu@...dia.com>
Powered by blists - more mailing lists