This PR proposes an approach to reduce the thread count of RPL.
The current RPL/Trickle stack needs about 5 threads to handle tasks
like updating the trickle timer, routing entries and the transmission of
DAOs.
This PR modifies RPL to use only one thread with a looped `msg_recv()` call.
The message is then multiplexed to the right task.
Currently, when using vtimer_set_msg the corresponding msg_t is filled
with the MSG_TIMER ("12345") type.
This approach makes it difficult to differentiate between incoming
messages via vtimer_set_msg.
In this PR I introduce another parameter for the vtimer_set_msg
function to specify a custom msg_t type.
This implementation is based on RFC 6550 with addition of RFC 6554 (Source Routing Header for RPL). Both can be found under the following links:
- http://tools.ietf.org/html/rfc6550
- http://tools.ietf.org/html/rfc6554
The PR provides basic functionality for handling and forwarding packages in non-storing mode. In addition the structure of the previous implemented RPL storing mode is now revised, so that readability and modularity is increased. The following features are implemented:
- building function for a SRH and integration in common packets
- source-route build algorithm based on the structure of the DODAG
- an RPL-based interpretation of the SRH and removal at destination
- new structure for RPl-module with extracted beaconing-functionality
- leaf nodes are now supported
There are some missed goals and should be included in future updates:
- building a common routing table structure for different types of routing protocols
- routing tables are statically assigned via source code, future update should have an optional variable at build-time, which sets the size of the routing table depending on the desired functionality of a node in the network (root, node, leaf)
Currently, the tcp and udp implementations are bound to each other in a
module called *destiny*. Thus, when using only one of them then the
other one gets also compiled into the binary and initialized,
which results in unnecessary RAM usage and workload for the CPU.
The approach in this PR defines a common module named *socket_base*,
which contains functions used by the posix layer. Compiled by it's own,
those functions return negative error codes, to symbolize upper layers
that they are not supported. When also including the modules *udp* or
*tcp* respectively, functions from *socket_base* get overwritten with the
correct functionality.
Defining *udp* or *tcp* in a Makefile also includes *socket_base*.
Defining *pnet* in a Makefile also includes *socket_base*.
If there's no IPv6 address configured so far to any interface, 6lowpan
should not try to handle incoming packets. This can easily lead to
looping packets.
Instead of using differing integer types use kernel_pid_t for process
identifier. This type is introduced in a new header file to avoid
circular dependencies.
This PR converts tabs to white spaces.
The statement I used for the conversion:
```find . -name "*.[ch]" -exec zsh -c 'expand -t 4 "$0" > /tmp/e && mv /tmp/e "$0"' {} \;```
Afterwards, I had a quick overview of the converted files to prevent odd indentation.
For many modules the `Makefile` contains a line like
```
MODULE:=$(shell basename $(CURDIR))
```
This conclusively shows that we do not have to set the module name
manually.
This PR removes the need to set the module name manually, if it is the
same as the basename. E.g. for `…/sys/vtimer/Makefile` the variable
make `MODULE` will still be `vtimer`, because it is the basename of the
Makefile.
When sending a packet a workaround is applied when ndp_get_ll_address() returns NULL as is's not implemented properly yet.
The same should happen when a packet is forwarded as it's dropped otherwise.
msg_queue is defined but never used, hook it up so IP packets get queued instead of dropped when there is more than one.
change the name to ip_msg_queue to avoid naming conflicts.
