[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 04 Apr 2024 12:40:38 +0200
From: Paolo Abeni <pabeni@...hat.com>
To: Chintan Vankar <c-vankar@...com>, Dan Carpenter
<dan.carpenter@...aro.org>, Siddharth Vadapalli <s-vadapalli@...com>,
Heiner Kallweit <hkallweit1@...il.com>, Andrew Lunn <andrew@...n.ch>, Roger
Quadros <rogerq@...nel.org>, Richard Cochran <richardcochran@...il.com>,
Jakub Kicinski <kuba@...nel.org>, Eric Dumazet <edumazet@...gle.com>,
"David S. Miller" <davem@...emloft.net>
Cc: linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH net-next v5 1/2] net: ethernet: ti: am65-cpts: Enable RX
HW timestamp for PTP packets using CPTS FIFO
On Tue, 2024-04-02 at 17:14 +0530, Chintan Vankar wrote:
> Add a new function "am65_cpts_rx_timestamp()" which checks for PTP
> packets from header and timestamps them.
>
> Add another function "am65_cpts_find_rx_ts()" which finds CPTS FIFO
> Event to get the timestamp of received PTP packet.
>
> Signed-off-by: Chintan Vankar <c-vankar@...com>
> ---
>
> Link to v4:
> https://lore.kernel.org/r/20240327054234.1906957-1-c-vankar@ti.com/
>
> Changes from v4 to v5:
> - Updated commit message.
> - Replaced "list_del_entry()" and "list_add()" functions with equivalent
> "list_move()" function.
>
> drivers/net/ethernet/ti/am65-cpts.c | 64 +++++++++++++++++++++++++++++
> drivers/net/ethernet/ti/am65-cpts.h | 6 +++
> 2 files changed, 70 insertions(+)
>
> diff --git a/drivers/net/ethernet/ti/am65-cpts.c b/drivers/net/ethernet/ti/am65-cpts.c
> index c66618d91c28..bc0bfda1db12 100644
> --- a/drivers/net/ethernet/ti/am65-cpts.c
> +++ b/drivers/net/ethernet/ti/am65-cpts.c
> @@ -906,6 +906,70 @@ static int am65_skb_get_mtype_seqid(struct sk_buff *skb, u32 *mtype_seqid)
> return 1;
> }
>
> +static u64 am65_cpts_find_rx_ts(struct am65_cpts *cpts, u32 skb_mtype_seqid)
> +{
> + struct list_head *this, *next;
> + struct am65_cpts_event *event;
> + unsigned long flags;
> + u32 mtype_seqid;
> + u64 ns = 0;
> +
> + am65_cpts_fifo_read(cpts);
> + spin_lock_irqsave(&cpts->lock, flags);
am65_cpts_fifo_read() acquires and releases this same lock. If moving
to a lockless schema is too complex, you should at least try to acquire
the lock only once. e.g. factor out a lockless __am65_cpts_fifo_read
variant and explicitly acquire the lock before invoke it.
> + list_for_each_safe(this, next, &cpts->events) {
> + event = list_entry(this, struct am65_cpts_event, list);
> + if (time_after(jiffies, event->tmo)) {
> + list_del_init(&event->list);
> + list_add(&event->list, &cpts->pool);
Jakub suggested to use list_move() in v4, you should apply that here,
too.
Cheers,
Paolo
Powered by blists - more mailing lists