>> It is not good idea. Any module (CIPE in this case) outside of
>> core cannot know, what generic fields in skb must be initialized
>> in copying.
> Yes, and it's ugly too :-)
>> Most of fields in skb are strictly private to core. If you
>> believe, that some primitive is missing, it would be better to
>> submit it for inclusion to core.
> Basically I need a version of skb_realloc_headroom that allows me
> to specify needed headroom _and_ tailroom, because my driver adds a
> "footer" in addition to its own header. I can well imagine there
> could be real hardware drivers that need this too.
It sounds like the best option would be to have a generic function
called skb_reallic_head_and_tail_room which allocated either or both
as required, and define a pair of macros called skb_realloc_headroom
and skb_realloc_tailroom which just called the generic function with
an extra zero parameter for the unwanted area...
If set up correctly, that would mean that nothing else had to change
since the macros would implicitly do the necessary tweaks every time
the original functions were used...
Alan: Your thoughts on this?
Best wishes from Riley.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/