Re: [PATCH] hashed device lookup (Does NOT meet Linus' sumission

Matti Aarnio (matti.aarnio@zmailer.org)
Sun, 7 Jan 2001 21:24:14 +0200


On Sun, Jan 07, 2001 at 02:10:52PM -0500, jamal wrote:
> On Sun, 7 Jan 2001, Matti Aarnio wrote:
> > Read what I wrote about the issue to Alan.
> > Ben's code has no problems with receiving VLANs with network
> > cards which have "hardware support" for VLANs.
>
> OK. I suppose an skb->vlan_tag is passed to the driver and it will know
> what to do with it (pass it on a descriptor etc).

Sure, nice. WHY SHOULD THERE BE MORE LAYER-2 STUFF ADDED TO
SKB OBJECTS ?

One of important abstraction issues is to isolate device specific
new things (like what VLAN/PVC/SVC is used at your favourtite
802.1Q/ATM/X.25/FrameRelay connection).

The less we leak that kind of things to SKB, the better, IMO.
They are net_device issues, after all.

Tell me (if you can), why packet sender calls hardware-header
generation for packet, if the card can insert it for you ?
Consider the structure of Ethernet MAC header, where is source
address ? Where is the destination address ? If you write the
destination, why should you not write the source there too ?

No doubt some cards can fill in the source address while doing
frame transmit, but is it worth the hazzle ?

> cheers,
> jamal

/Matti Aarnio
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/