080ecc5488
Now tcpip_output() is a function pointer that is supposed to be set via the macro tcpip_set_outputfunc(). Packet drivers do so on process startup. Thus if there are several packet drivers in a Contiki system the one started last is the one actually used. This behaviour is especially useful for the 'IP forwarding' "meta" packet driver.
138 lines
4.3 KiB
C
138 lines
4.3 KiB
C
/*
|
|
* This is an example of how to write a network device driver ("packet
|
|
* driver") for Contiki. A packet driver is a regular Contiki process
|
|
* that does two things:
|
|
* # Checks for incoming packets and delivers those to the TCP/IP stack
|
|
* # Provides an output function that transmits packets
|
|
*
|
|
* The output function is registered with the Contiki TCP/IP stack,
|
|
* whereas incoming packets must be checked inside a Contiki process.
|
|
* We use the same process for checking for incoming packets and for
|
|
* registering the output function.
|
|
*/
|
|
|
|
/*
|
|
* We include the "contiki-net.h" file to get all the network functions.
|
|
*/
|
|
#include "contiki-net.h"
|
|
|
|
/*---------------------------------------------------------------------------*/
|
|
/*
|
|
* We declare the process that we use to register with the TCP/IP stack,
|
|
* and to check for incoming packets.
|
|
*/
|
|
PROCESS(example_packet_driver_process, "Example packet driver process");
|
|
/*---------------------------------------------------------------------------*/
|
|
/*
|
|
* Next, we define the function that transmits packets. This function
|
|
* is called from the TCP/IP stack when a packet is to be transmitted.
|
|
* The packet is located in the uip_buf[] buffer, and the length of the
|
|
* packet is in the uip_len variable.
|
|
*/
|
|
u8_t
|
|
example_packet_driver_output(void)
|
|
{
|
|
let_the_hardware_send_the_packet(uip_buf, uip_len);
|
|
}
|
|
/*---------------------------------------------------------------------------*/
|
|
/*
|
|
* This is the poll handler function in the process below. This poll
|
|
* handler function checks for incoming packets and delivers them to
|
|
* the TCP/IP stack.
|
|
*/
|
|
static void
|
|
pollhandler(void)
|
|
{
|
|
/*
|
|
* We assume that we have some hardware device that notifies us when
|
|
* a new packet has arrived. We also assume that we have a function
|
|
* that pulls out the new packet (here called
|
|
* check_and_copy_packet()) and puts it in the uip_buf[] buffer. The
|
|
* function returns the length of the incoming packet, and we store
|
|
* it in the global uip_len variable. If the packet is longer than
|
|
* zero bytes, we hand it over to the TCP/IP stack.
|
|
*/
|
|
uip_len = check_and_copy_packet();
|
|
|
|
/*
|
|
* The function tcpip_input() delivers the packet in the uip_buf[]
|
|
* buffer to the TCP/IP stack.
|
|
*/
|
|
if(uip_len > 0) {
|
|
tcpip_input();
|
|
}
|
|
|
|
/*
|
|
* Now we'll make sure that the poll handler is executed repeatedly.
|
|
* We do this by calling process_poll() with this process as its
|
|
* argument.
|
|
*
|
|
* In many cases, the hardware will cause an interrupt to be executed
|
|
* when a new packet arrives. For such hardware devices, the interrupt
|
|
* handler calls process_poll() (which is safe to use in an interrupt
|
|
* context) instead.
|
|
*/
|
|
process_poll(&example_packet_driver_process);
|
|
}
|
|
/*---------------------------------------------------------------------------*/
|
|
/*
|
|
* Here we shutdown the hardware in case the process exits.
|
|
*/
|
|
static void
|
|
exithandler(void)
|
|
{
|
|
shutdown_the_hardware();
|
|
}
|
|
/*---------------------------------------------------------------------------*/
|
|
/*
|
|
* Finally, we define the process that does the work.
|
|
*/
|
|
PROCESS_THREAD(example_packet_driver_process, ev, data)
|
|
{
|
|
/*
|
|
* This process has a poll handler, so we declare it here. Note that
|
|
* the PROCESS_POLLHANDLER() macro must come before the PROCESS_BEGIN()
|
|
* macro.
|
|
*/
|
|
PROCESS_POLLHANDLER(pollhandler());
|
|
|
|
/*
|
|
* This process has an exit handler, so we declare it here. Note that
|
|
* the PROCESS_EXITHANDLER() macro must come before the PROCESS_BEGIN()
|
|
* macro.
|
|
*/
|
|
PROCESS_EXITHANDLER(exithandler());
|
|
|
|
/*
|
|
* The process begins here.
|
|
*/
|
|
PROCESS_BEGIN();
|
|
|
|
/*
|
|
* We start with initializing the hardware.
|
|
*/
|
|
initialize_the_hardware();
|
|
|
|
/*
|
|
* Register the driver. This will cause any previously registered driver
|
|
* to be ignored by the TCP/IP stack.
|
|
*/
|
|
tcpip_set_outputfunc(example_packet_driver_output);
|
|
|
|
/*
|
|
* Now we'll make sure that the poll handler is executed initially. We do
|
|
* this by calling process_poll() with this process as its argument.
|
|
*/
|
|
process_poll(&example_packet_driver_process);
|
|
|
|
/*
|
|
* And we wait for the process to exit.
|
|
*/
|
|
PROCESS_WAIT_EVENT_UNTIL(ev == PROCESS_EVENT_EXIT);
|
|
|
|
/*
|
|
* Here ends the process.
|
|
*/
|
|
PROCESS_END();
|
|
}
|
|
/*---------------------------------------------------------------------------*/
|