> -----Original Message----- > From: Richard Cochran <richardcochran@xxxxxxxxx> > Sent: 2021年4月18日 23:06 > To: Y.b. Lu <yangbo.lu@xxxxxxx> > Cc: netdev@xxxxxxxxxxxxxxx; Vladimir Oltean <vladimir.oltean@xxxxxxx>; > David S . Miller <davem@xxxxxxxxxxxxx>; Jakub Kicinski <kuba@xxxxxxxxxx>; > Jonathan Corbet <corbet@xxxxxxx>; Kurt Kanzenbach <kurt@xxxxxxxxxxxxx>; > Andrew Lunn <andrew@xxxxxxx>; Vivien Didelot <vivien.didelot@xxxxxxxxx>; > Florian Fainelli <f.fainelli@xxxxxxxxx>; Claudiu Manoil > <claudiu.manoil@xxxxxxx>; Alexandre Belloni > <alexandre.belloni@xxxxxxxxxxx>; UNGLinuxDriver@xxxxxxxxxxxxx; > linux-doc@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx > Subject: Re: [net-next 1/3] net: dsa: optimize tx timestamp request handling > > On Fri, Apr 16, 2021 at 08:36:53PM +0800, Yangbo Lu wrote: > > Optimization could be done on dsa_skb_tx_timestamp(), and dsa device > > drivers should adapt to it. > > > > - Check SKBTX_HW_TSTAMP request flag at the very beginning, instead of in > > port_txtstamp, so that most skbs not requiring tx timestamp just return. > > > > - No longer to identify PTP packets, and limit tx timestamping only for PTP > > packets. If device driver likes, let device driver do. > > > > - It is a waste to clone skb directly in dsa_skb_tx_timestamp(). > > For one-step timestamping, a clone is not needed. For any failure of > > port_txtstamp (this may usually happen), the skb clone has to be freed. > > So put skb cloning into port_txtstamp where it really needs. > > This patch changes three things ^^^ at once. These are AFAICT independent > changes, and so please break this one patch into three for easier review. Will split it in next version. Thank you. > > Thanks, > Richard
