RE: [EXTERNAL] [PATCH] netpoll: support sending over raw IP interfaces

[Ratheesh Kannoth]

> From: Mark <mark@xxxxxxxxxx>
> Sent: Thursday, March 21, 2024 6:03 PM
> To: Ratheesh Kannoth <rkannoth@xxxxxxxxxxx>
> Cc: netdev@xxxxxxxxxxxxxxx; Hans de Goede <hdegoede@xxxxxxxxxx>; Eric
> Dumazet <edumazet@xxxxxxxxxx>; Jakub Kicinski <kuba@xxxxxxxxxx>; Breno
> Leitao <leitao@xxxxxxxxxx>; Ingo Molnar <mingo@xxxxxxxxxx>; David S.
> Miller <davem@xxxxxxxxxxxxx>; Paolo Abeni <pabeni@xxxxxxxxxx>; linux-
> kernel@xxxxxxxxxxxxxxx
> Subject: Re: [EXTERNAL] [PATCH] netpoll: support sending over raw IP
> interfaces
> 
> Hi Ratheesh,
> 
> > Op 18 mrt 6 Reiwa, om 15:06 heeft Ratheesh Kannoth
> <rkannoth@xxxxxxxxxxx> het volgende geschreven:
> >
> >> […]
> > Is below network topology possible ?
> > Netpoll()- ------> netdev A ----> raw interface Where netdev A's
> > netdev->header_ops != NULL
> 
> I believe so, this is not uncommon in tunnel devices like gretap.
> However, those fully encapsulate the link layer header in the packet to the
> lower interface. I am not aware of a interface driver that removes a header
> upon xmit, so to speak. However, I have just posted a v2 that instead uses the
> documented `dev_has_header()` API, which seems to fit the check exactly,
> here:
ACK.  I understand the complexity to implement it in driver than a easy fix in netpoll.