Because the L2 buffers are only used by 802.15.4 sub-system,
we must not compile and link them to the kernel.
Change-Id: Iab0352582ac1c57e0dd8f54ca8e757e47e4df3be
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
There could be cases where the destroy callback needs more
fine-grained control of step ordering than "1. destroy, 2. put back to
free FIFO". One case could be when the the pool needs to be protected
by a microkernel mutex or semaphore. In such a case the putting back
to the FIFO may need to happen before a custom action in the destroy
callback.
Making the destroy callback responsible for returning to the free FIFO
gives full flexibility regarding the order of the cleanup actions.
Change-Id: Ib9532d1dd70e0a2042af54ebd3e40a853dd42d33
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
We need to have a generic buffer API in order to efficiently transfer
data between different subsystems. The first such case will be the
Networking and Bluetooth subsystems where 6LoWPAN data will be passed
back and forth.
The needed API needs to provide enough flexibility for different
buffer sizes as well as custom protocol-specific context data.
The implementation offered in this patch follows the general design of
the existing Networking and Bluetooth buffer implementations by using
a backing array of buffer which is fed into a "free buffers" FIFO for
management. The main difference is that the API allows specifying
variable sized buffers for each created pool, as well as a minimum
amount of "user data" that's allocated as part of each buffer.
There's also an optional destroy callback that's e.g. useful for HCI
flow control in Bluetooth (for notifying the controller of available
buffers).
Change-Id: I00b7007135a0ff35219f38f48658f31728fbb7ca
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Change-Id: I819d13f0d7a23e3a61dcda6a3ced18810b192158
Signed-off-by: Dan Kalowsky <daniel.kalowsky@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Change-Id: I2f28a4dd2b1b372905638cbccdf987396999d7ec
Signed-off-by: Dan Kalowsky <daniel.kalowsky@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Change-Id: I6da43e41f9c6efee577b70513ec368ae3cce0144
Signed-off-by: Dan Kalowsky <daniel.kalowsky@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Change all the Intel and Wind River code license from BSD-3 to Apache 2.
Change-Id: Id8be2c1c161a06ea8a0b9f38e17660e11dbb384b
Signed-off-by: Javier B Perez Hernandez <javier.b.perez.hernandez@linux.intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
Have a separate network buffer pool for received and transmitted
data. This way incoming packet flood cannot prevent from sending
IP packets out.
Change-Id: I2aae3bcb4ea63f36355f758de44532b167edfdb8
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Print information who allocated or freed mac net buffer.
Change-Id: I106e1703b7f5c3e5363d53d63c599a12b65577cb
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Generalize the stack analyze functions so that they
can be used by other fibers.
Change-Id: If1fae51db45010b75f6ac58cd8d874b31ca336e2
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The macro was incorrectly set and caused in certain case
wrong expansion.
Change-Id: I5198906a5066882d178c0fcff8c6816a5461f37a
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The tinydtls.h include file was missing from earlier tinydtls
commits.
Change-Id: I8b8398313c7f598daf81f4fb8c07fd138e22e22d
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Documenting the function that returns the used UDP connection
pointer. This function is needed for example in tinyDTLS
server implementation.
Change-Id: I08aae798dab0749912694d6338de91dcbd13df84
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Add NET_BUF_IP() and NET_BUF_UDP() macros that return
correct struct containing either IP or UDP data.
Change-Id: I43b828ef78c1f13ccee41e6a60daa36145a35eba
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Fixed the description of what happens if there is a timeout.
Change-Id: Id19a791193b527a8ee086f069e103f42d2d9b0f4
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
We must not touch uip_len(buf) value as Contiki IP stack
will use it to determine what it should do with the
received packet. So after we have received the packet,
we update separate datalen variable in net_buf that will
store the correct length of the packet.
Change-Id: Iab79b741508e95e581d1727645e6b1d1eacded4c
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Allow caller to specify a timeout in ticks while waiting
network data. The timeout value has following meaning:
TICKS_UNLIMITED wait forever
TICKS_NONE do not wait
value > 0 timeout value (only applicaple if
CONFIG_NANO_TIMEOUTS is defined)
Change-Id: Icbb33fa0eeb462659d4006dd9d948a04f39a87a9
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
When sending data to network, make sure to check the return
code from network driver. The driver should return these codes:
0 : If packet could not be sent. In this case buf should
not be released.
1 : If the packet was sent successfully. In this case the buf
should be released by either the send() or some other
lower layer function.
<0: If there is an error, the buf should not be released by
send() function.
Fixed the relevant part in Contiki code so that the tcpip_output()
return correct value
0 = packet was not sent, caller needs to free it
1 = packet was sent, network driver will free the net_buf
after sending it to medium
Change-Id: I4380d7747985fc057f5ef73ca97b76f6e9888a55
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Add infrastructure and checks to validate the net_mbuf during
its lifecycle.
Change-Id: Id4b638fff991325d2247b1f24152552038888915
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Add macros that check that we are not trying to use already
freed net_buf and vice versa.
Change-Id: Ifdb6a7e24cd6d7a0de46fa7b991c358ea8828d6f
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Making sure we do not include anything unless they
are inside include guard.
Change-Id: Ie88cbc0ed6dbbf4033f3c8c310df7d084ab8b126
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The net_reply() can be used if the application wants to reply
something to peer right after receiving the data. The function
will reverse the source and destination addresses and ports,
and set the uIP internal structures in such a way that the
reply is possible. Using this function saves the application
from allocating a separate net_buf for just sending simple
message back.
Change-Id: Ia960c83b77a8f300222159ee085fed99e2b007c2
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Change all occurances of /*! to /** to match javadoc
style.
Change-Id: I3a759d34e0e928216f61252682266e64c5b875f8
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
6lowpan HC06 and HC01 compression/uncompression functions are
adapted to net_mbuf buffers. HC06 compression and uncompression
works properly. HC01 adapted to net_mbuf but functionality is
failing due to inproper IPv6 link local addresses. buf pointers
from net_mbuf and couple of net_mbuf calls are removed (which
are not used at the moment and saves memory from net_mbuf).
Change-Id: I2711b71b1b8df2eaac67d69275baf4b62b03f9f6
Signed-off-by: Ravi kumar Veeramally <ravikumar.veeramally@linux.intel.com>
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
This is needed for enabling RPL and 6lowpan support.
Change-Id: I87f95a08db2cc874bbe841d9b1456803ab16d75d
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Enable 802.15.4 driver that receives the IPv6 packet, does header
compression on it and writes it to the 15.4 stack Tx FIFO.
The 15.4 Tx fiber will pick up the header compressed IPv6
6LoWPAN packet and fragment it into suitable chunks ready
to be sent to the 15.4 hw driver. The 15.4 stack Rx fiber picks
15.4 frames from 15.4 Rx FIFO (which are fed by 15.4 driver)
and reassembles the all fragmented packets and uncompress the
header and writes to the IP Rx FIFO.
The 15.4 loopback radio driver receives the 15.4 frames and
put it back to 15.4 Rx Fifo.
Change-Id: I68d1c7ff6de0e8770616e574206c2091c69a28e6
Signed-off-by: Ravi kumar Veeramally <ravikumar.veeramally@linux.intel.com>
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Move packetbuf from net_buf to net_mbuf, update 802.15.4 stack
with net_mbuf (which is smaller in size, and mostly wrapper for
packetbuf) and split sicslowpan(ipv6/sicslowpan.c|h) functionality
to compression and fragmentataion. This patch contains
null_compression and null_fragmentation which is exactly what
uip_driver.c|h does.
Change-Id: Ifee359e20aa57a50fb00c9ce049d9476b2ac67e3
Signed-off-by: Ravi kumar Veeramally <ravikumar.veeramally@linux.intel.com>
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
After an IP packet has been (reassembled and) received, the driver
submits the IP packet to the IP stack.
Change-Id: Icbd1e7740ce59db16bc1d61002420e86ab101d2d
Signed-off-by: Patrik Flykt <patrik.flykt@linux.intel.com>
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
We do not need large 1280 bytes long buffers in lower IP stack
layers. After long packet has been fragmented to smaller chunks
it depends on the radio technology how long buffers we need
to use. For 802.15.4 we can use 128 bytes long buffers.
Change-Id: I8cd6085944d78fa26a4aa1a93c58fc436856541b
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The uip_len(buf) cannot be used as the Contiki stack zeroes
the buffer len after sending and after processing received packet.
Thus the application can use uip_appdatalen(buf) macro to get
the application data length.
Change-Id: Iaaa508acb96c28919cfc0188ef46bb2b4a693090
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Create IN6ADDR_ANY_INIT and IN6ADDR_LOOPBACK_INIT address
definitions for applications.
Change-Id: Ic864f559cae90a5f85e45ee90d73127b0ed8fde6
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Network driver should call this when it knows the device MAC or
EUI-64 address.
Change-Id: Ie03a80fcdb9a10cd1cb15f12037b3802da13bca9
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The network address union that holds IPv4 and IPv6 address
should be anonymous as there is no need to refer the union
separately.
Change-Id: I70bab6b91dcf49dfb5adb405eb2b9c602f932603
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
This mega patch changes how Contiki uses buffers. In standard
Contiki there is only couple of static and global buffers.
This makes the stack non-reentrant and only usable in one
thread. This patch replaces the global buffer with buffer
supplied by caller (net_buf.h). The Contiki stack is still not
fully re-entrant after this (packet reassembly needs more TLC)
but it is a good start.
Change-Id: I63abc230b36b14f33f687d3ef64cffb0f3a69f5d
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The network context defines a network connection.
Change-Id: I29a186be6c9de9d824f10b6442fa1dfd3711d24d
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Functions to receive data from apps and transmit it to network, and
receive from network and pass data to apps.
Change-Id: I1b1b8c041e6c5e20294081d2cd403636e9909cdc
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Jukka Rissanen