When a packet is tx_down()ed, should the npkt_t cell, endpoint, and iface
members be set to those of the final target (ie those of the Ethernet
driver) or those of the next hop (ie the converter)?
The tx_up() and rx_up() functions also have a duplicate set of these in
their function call parameters. Am I correct in understanding the set passed
to tx_up() and rx_up() will be those of the previous hop (ie the converter)
whereas the set in the npkt_t will be those of the originator.
The ddk doc says
cell The cell to or from which the npkt_t is headed.
Is it to which' for downward packets and from which’ for upward packets?
The doc should specify.
Thanks,
Shaun
Shaun Jackman <sjackman@nospam.vortek.com> wrote:
When a packet is tx_down()ed, should the npkt_t cell, endpoint, and iface
members be set to those of the final target (ie those of the Ethernet
driver) or those of the next hop (ie the converter)?
The final target. The convertor is really in 2 cells: yours and the
target.
The tx_up() and rx_up() functions also have a duplicate set of these in
their function call parameters. Am I correct in understanding the set passed
to tx_up() and rx_up() will be those of the previous hop (ie the converter)
whereas the set in the npkt_t will be those of the originator.
On up headed packets, the set in the npkt aren’t valid and the function
parameters contain the previous hop (the producer, not the convertor).
-seanb
The ddk doc says
cell The cell to or from which the npkt_t is headed.
Is it to which' for downward packets and from which’ for upward packets?
The doc should specify.
Thanks,
Shaun