[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <B04768ED-4E00-46AD-87A0-8AF89479A87F@flugsvamp.com>
Date: Wed, 15 Feb 2023 11:39:02 -0800
From: Jonathan Lemon <jlemon@...gsvamp.com>
To: Köry Maincent <kory.maincent@...tlin.com>
Cc: Jonathan Lemon <jonathan.lemon@...il.com>,
Richard Cochran <richardcochran@...il.com>,
Florian Fainelli <f.fainelli@...il.com>,
netdev@...r.kernel.org,
Maxime Chevallier <maxime.chevallier@...tlin.com>,
Thomas Petazzoni <thomas.petazzoni@...tlin.com>
Subject: Re: net: phy: broadcom: error in rxtstamp callback?
The code is correct - see Documentation/networking/timestamping.rst
The function returns true/false to indicate whether a deferral is needed
in order to receive the timestamp. For this chip, a deferral is never
required - the timestamp is inline if it is present.
—
Jonathan
On 15 Feb 2023, at 2:07, Köry Maincent wrote:
> Hello,
>
> I am new to PTP API. I am currently adding the support of PTP to a PHY driver.
> I looked at the other PTP supports to do it and I figured out there might be an
> issue with the broadcom driver. As I am only beginner in PTP I may have wrong
> and if it is the case could you explain me why.
> I also don't have such broadcom PHY to test it, but I want to report it if the
> issue is real.
> The issue is on the rxtstamp callback, it never return true nor deliver the
> skb.
>
> Here is the patch that may fix it:
>
> diff --git a/drivers/net/phy/bcm-phy-ptp.c b/drivers/net/phy/bcm-phy-ptp.c
> index ef00d6163061..57bb63bd98c7 100644
> --- a/drivers/net/phy/bcm-phy-ptp.c
> +++ b/drivers/net/phy/bcm-phy-ptp.c
> @@ -412,7 +412,9 @@ static bool bcm_ptp_rxtstamp(struct mii_timestamper *mii_ts,
> __pskb_trim(skb, skb->len - 8);
> }
>
> - return false;
> + netif_rx(skb);
> +
> + return true;
> }
>
> static bool bcm_ptp_get_tstamp(struct bcm_ptp_private *priv,
> --
> 2.25.1
>
>
> Regards,
Powered by blists - more mailing lists