| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit has a first implementation of flow allocation (the "slow
path") and read/write (the "fast path") for ouroboros. It provides
basic but unstable communications over the shared memory.
It required a lot of changes all over the stack, and fixes a number of
previously undetected issues.
This PR still need heavy revision regarding data model, locking and
cleanup.
lib/dev: modifications to the API. It now uses an ap_init() call to
set the AP name and sets the Instance ID to the pid of the process. It
also binds the AP to the shared memory and creates tables for mappings
in the fast path. A call to ap_fini() releases the resources.
lib/shm_ap_rbuff: added ring buffer for data exchange between
processes in the fast path. It passes an index in the shm_du_map.
lib/shm_du_map: rewrote API to work with calls from dev.c. Garbage
collector added. Tests updated to new API.
ipcpd/ipcp-data: removed everything related to flows, as these are
universal for all ap's and kept in ap_data (dev.c), or similar structs
for shim ipcps.
shim-udp: added flow allocator and read/write functions and shm
elements.
irmd: revised data model and structures necessary for flow allocation.
tools: echo updated to new dev.h API.
messaging system was updated to comply with new flow allocation
messages. All exchanges use pid and port_id to bootstrap the fast
path.
|
|
|
|
|
| |
flow allocation now propagates on the client side up to the IPCP.
added UNKNOWN_AP and UNKNOWN_AE definitions to dev.h
|
|
|
|
|
| |
ipcp_create now returns the pid of the created process to allow for
more efficient scripting.
|
|
|
|
|
|
|
|
| |
All instance-id's in ouroboros will be set by the system to the pid of
the process associated with this application process instance. This
means that the user has no way to choose the instance id's. Function
calls that assumed manually defined instance id's have been replaced
throughout the system.
|
|
|
|
|
|
|
|
|
|
|
|
| |
Unregistering ap's now works. An AP now registers/unregisters its AP-I
by sending its AP name and its pid to the IRMd. The IPCPs register
whatevercast names. An AP name is currently mapped on a whatevercast
name represented by the same string literal. The IRMd allows
registration of only one AP-I per AP. A Name Space Management system
is needed in the processing system so we can resolve this completely.
Changing the stack to register whatevercast names required some changes
all over the ipcpd implemented and in the library.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This implements the API for flow allocation. The shims currently does
the following.
The shim IPCP binds to an interface (IP address) and listens for flow
allocation requests on UDP port 0x0D1F (3359), referenced as the
listen port (lp). It will treat any datagram received on lp as a flow
allocation request.
Upon receiving an allocation request IRM_MSG_CODE__IPCP_FLOW_ALLOC
from the IRMd, the shim IPCP will bind a UDP socket to a port (cp)
allocated by the host OS. From that port it will send a UDP packet
containing the destination ap_name to server_host:lp and wait for a
response.
Upon reception of a packet on server_host:lp, the shim_IPCP creates a
UDP socket for the flow with a port set by the host os (sp), binds to
it and echoes the received datagram back from server_host:sp to
client_host:cp. It will also notify the IRMd of an incoming flow
allocation request IRM_MSG_CODE__IPCP_FLOW_REQ_ARR, with as
src_ap_name ("John Day"). It will get the port_id as a return value
of that message and create a flow with status FLOW_PENDING with that
port_id. If the server responds negatively to the flow allocation
request (i.e. the shim IPCP on the server side receives a
IRM_MSG_CODE__IPCP_FLOW_ALLOC_RESPONSE with a response != 0, it will
delete the pending flow. If response == 0, it will set the status to
FLOW_ALLOCATED.
On the client machine the IPCP will learn sp upon reception of the
echoed datagram. It will then create a flow with the port_id it
received with the message from the IRMd and set it to ALLOCATED.
Pending implementation:
DNS support, this PR only supports local flows on the loopback adapter
127.0.0.1.
A thread to listen for the echoed message, to avoid the
entire IPCP to block when the echoed message is lost.
This PR compiles but is untested pending necessary implementations
elsewhere in the stack.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Basic functions for implementation of IPC processes, and
implementation of core functions of the shim IPCP over UDP. Updates
to the build system to compile these IPC processes, as well as some
fixes in the irmd (rudimentary capturing exit signals) and some fixes
in the library, mainly relating to the messaging.
Basic implementation of creation / bootstrapping / deletion of the
shim UDP. Placeholders for other functions.
|
| |
|
|\
| |
| |
| | |
Dif config now correctly passed to the IPCP
|
| |
| |
| |
| |
| | |
This changes the name of the IPCP types after discussions with
Dimitri.
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This adds dif_config to the prototype, in which one is able to specify
the parameters a DIF should have. The bootstrap operation of an IPCP
takes this as parameter and is oblivious to whether it is a shim or a
normal IPCP. The dif_config struct is also correctly serialized and
deserialized and passed opaquely to the correct IPCP. This IPCP is in
charge of deserializing it correctly.
|
|/ |
|
|
|
|
|
| |
Removes name_is_equal. Just using instance_name_cmp is encouraged. It
also removes a wrong initialization in the irm tool.
|
|
|
|
| |
This moves strdup to utils to make it globally available.
|
|\ |
|
| |
| |
| |
| |
| |
| | |
This adds the messages that are sent to the IPCPs related to
flows. Some messages are also sent to the IRMd (e.g. when a new flow
arrives).
|
| |\
| | |
| | |
| | | |
lib, irmd: Update communication with IRMd
|
| | |
| | |
| | |
| | |
| | | |
All messages sent to the IRMd now also get a reply back with the
result of the operation.
|
| |\| |
|
| | |
| | |
| | |
| | |
| | | |
fixes wrong check, checks now use lazy evaluation
changed the order of instance_name_cpy to (dst, src)
|
| | |
| | |
| | |
| | |
| | | |
all functions taking a char * ap_name and uint id now take either a
instance_name_t or instance_name_t *
|
| | |
| | |
| | |
| | |
| | | |
FLOW_O_INVALID now defined in terms of conflicting options
bugfix in setopts
|
| | |
| | |
| | |
| | |
| | | |
forgotten return statement
forgotten NULL check
|
| |\| |
|
| | |\ |
|
| | | |
| | | |
| | | |
| | | |
| | | | |
This removes the custom ser/des methods for communicating with the
IPCP daemon and also uses GPB instead.
|
| | | |
| | | |
| | | |
| | | |
| | | | |
IRMd and dev.c now also use GPB instead of our own ser/des. irm_msg
struct has been dropped as well as the methods associated with it.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Removes rina_name_t from that API. Passing ap_name and api_id as
params instead. The IRM tool has been updated accordingly. Some errors
in the build related to protobuf-c have also been resolved.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This adds a cmake file so that the build can ask to generate
protobuf-c files from .proto files. The messages between the IRM and
the library are compiled into the library.
|
| | |/
| |/|
| | |
| | |
| | | |
moved the flow definition to the library and made it public.
thread-safety implemented without compiler checks.
|
|/ / |
|
|/ |
|
|
|
|
| |
logs should be included in source, not in header
|
|
|
|
|
|
|
|
| |
Updated tests. Added code to drop a corner case (packet where the tail
PCI would cross the edge of the ring buffer) because solving this very
rare case is not worth the performance hit on the ringbuffer the extra
code would incur. This means the ringbuffer might drop a very small
percentage of packets.
|
|\ |
|
| |
| |
| |
| |
| |
| | |
This adds several messages for the dev.h API calls to communicate with
the IRM daemon. The deserializing of these messages is still missing
and the irmd hasn't been updated with them either.
|
|\| |
|
| |\
| | |
| | |
| | | |
tools: Simple echo application
|
| | |
| | |
| | |
| | |
| | |
| | | |
This adds a simple echo application written to show application how to
use the API. It also updates the dev.h header file with the insights
gained from performing this excercise.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Got rid of divisions and modulo operations.
Change in the meaning of a #define.
SHM_DU_MAP_SIZE: is now an exponent of 2. Default is 10, allowing
for 1024 blocks in the map.
|
| | |
| | |
| | |
| | |
| | |
| | | |
Elements must be created/destroyed in order, but IPCPs can access PCI
while the PDU is in the ring buffer.
Test updated, no more waits are needed.
|
|/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The shared memory map will allow passing SDU's between IPCPs
efficiently. It is a single block of managed memory. SDU's are stored
in a shm_du_buff datastructure, which is a list of blocks inside the
shared memory region. The mechanism for passing SDU's is not yet
implemented and could be either within the region or using another
local IPC mechanism.
The following things are configured at compile time:
SHM_FILE_NAME : the name. Default is "ouroboros_du_map".
SHM_DU_BUFF_BLOCK_SIZE: size of a block inside the shared memory
region. Default is the pagesize of the system.
SHM_DU_MAP_SIZE : the approximate size in bytes of the shared
memory map. The actual filesize may differ by
a small margin. Default is 4MB.
|
|/
|
|
|
| |
fixes memleaks in du_buff and du_buff_test
fixes bug in copy operation
|
|
|
|
|
| |
Some allocations were not freed in case the creation of a new IPCP
failed.
|
|
|
|
|
| |
This adds the messages required for the IRMd to communicate with an
IPC Process through the library.
|
|
|
|
|
| |
This removes some memleaks present in the sockets layer. It also fixes
the bad initialization of the difs_size param in the irm_msg.
|
|
|
|
| |
du_buff.c: LOG_WARN changed to LOG_ERR
|
| |
|
|\ |
|