> -----Original Message----- > From: Richard Cochran <richardcochran@xxxxxxxxx> > Sent: 2021年4月18日 23:11 > To: Vladimir Oltean <olteanv@xxxxxxxxx> > Cc: Y.b. Lu <yangbo.lu@xxxxxxx>; 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 Sun, Apr 18, 2021 at 12:18:42PM +0300, Vladimir Oltean wrote: > > > > How about not passing "clone" back to DSA as an argument by reference, > > but instead require the driver to populate DSA_SKB_CB(skb)->clone if > > it needs to do so? > > > > Also, how about changing the return type to void? Returning true or > > false makes no difference. Thank you. That's good idea. And how about letting driver store the skb clone pointer, or not? I copied my comments in patch #3 here, " Can we totally drop dsa_skb_cb in dsa core? The only usage of it is holding a skb clone pointer, for only felix and sja1105. Actually we can move such pointer in <device>_skb_cb, instead of reserving the space of skb for any drivers." > > +1 > > Thanks, > Richard
