| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| |
| |
| |
| |
| | |
This prevents assertion failures in the IPCP in some cases. IPCPs can
now safely assert the type.
|
|/
|
|
|
|
|
| |
When starting two IRMds, the second one exits, but applications can't
contact the first because the second instance took the UNIX socket.
Now the lockfile is checked for running IRMd instances before opening
the UNIX socket.
|
| |
|
|
|
|
|
|
|
|
| |
When there is a burst of successive flow allocations for a certain
name, each such request will block a thread in the IRMD for
IRMD_REQ_ARR_TIMEOUT ms to allow the application some time to respond.
This refactors some parts of the IRMd.
|
|
|
|
|
| |
The acceptor will not log disconnects with IRMd. Unexpected
disconnects will be reported and handled by management components.
|
|
|
|
|
|
|
|
| |
The IPCP will now respond with an ipcp_create_r message when it fails,
informing the IRMd.
Also adds some const qualifiers in the public headers and fixes
some formatting in dev.c.
|
|
|
|
|
|
|
|
|
|
| |
This removes the logfile and outputs log messages to the logging
system. The creation of the logfiles (as well as the ap_init() call)
were moved into ipcp_init() to simplify the IPCP creation and
shutdown.
Fixes #25
Fixes #27
|
|
|
|
|
|
|
| |
If an IPCP SEGVs abruptly, the sanitizer in the IRMd would also cause
a SEGV because of looping over a registry linked list inside another
safe loop over the same list. A function is added to the registry to
avoid this.
|
| |
|
|
|
|
|
|
| |
When trying to double bootstrap, the entry name was strdup'd before
the bootstrap leaking the previous name. Now it bootstraps first and
sets the name after a successful bootstrap only.
|
|\ |
|
| |
| |
| |
| |
| | |
Copyright is set to 2016 - 2017. License text on includes and sources
in the library are changed to indicate the LGPLv2.1 license.
|
| |
| |
| |
| | |
Adds LGPL license to the ouroboros lists.
|
|/
|
|
|
|
|
| |
This allows IPCPs to bind a name, so that they can announce their name
to neighbors which can then allocate a flow to them. Registering of
the name happens by an administrator. It also moves the irmd_api to
common ground, since it is used by all IPCPs.
|
|
|
|
|
|
|
| |
This changes the return value of the register and unregister
functions. If they were able to reg/unreg some entries a positive
value is returned, otherwise an error. Some spelling errors in the
IRMd have also been fixed.
|
|
|
|
|
|
|
|
|
|
| |
Now correctly relays the qoscube end-to-end in the stack.
A simple function specifying the cube in the spec is used for initial
testing.
The translation is now done in dev.c, but it could be moved elsewhere
when qos cabability matures and the need arises.
|
| |
|
|
|
|
|
| |
This corrects the license statements on all files. Installed headers
are LGPLv2.1, the rest of the code is GPLv2.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Receiver timeouts can now be set on a flow using the flow_set_timeout
function. Specifying NULL disables the timeout. The flow_get_timeout
function gets the value for the timeout.
This commit also deprecates fcntl in favor of flow_get_flags and
flow_set_flags functions.
struct qos_spec is typedef'd as a qosspec_t.
The tools and cdap.c are updated to use the new API.
Fixes a bug in operf client where the client's writer thread wouldn't
cancel on SIGINT.
|
|
|
|
|
|
| |
*BSD accept() doesn't timeout on the timeval set by setsockopt
SO_RCVTIMEO. The irmd and ipcp will now call select on the irmd fd
with the same timeout.
|
|
|
|
| |
Also cleans up the flow_set if an AP-I dies abruptly.
|
|
|
|
|
|
| |
Now that we have an rbuff per flow per AP, there is no more need to
keep the dst_api in the rdrbuff. This also simplifies the sanitizer in
the irmd.
|
|
|
|
|
| |
Creation and destruction of the rbuffs is now completely handled in
the irm_flow.
|
|
|
|
|
| |
The IRMd should not send messages to the IPCPs after its mainloop
terminated. IPCPs should cleanup al state when they get the SIGTERM.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The steps for flow deallocation have been further refined. An
operation ipcp_flow_fini() which wait for all SDUs to be read from a
flow has been added. The shim IPCPs and the local IPCP have been
adapted to this new API.
Deallocation messages have been removed from the shim IPCPs, since
there is insufficient state synchronisation between them to make this
work reliably.
|
|
|
|
| |
This fixes 2 wrong unlocks in the bind API method.
|
|\
| |
| |
| | |
lib, irmd, ipcpd: Add name querying to IPCPs
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This adds the ability to query IPCPs if a name can be reached through
them, e.g. if a name is available in a DIF. This means that in the
shim-udp a DNS query is performed, in the shim-eth-llc an ARP-like
query has been added, in the local a check is done to see if the name
is registered, and in the normal currently no application is reachable
through it.
|
|/
|
|
|
|
| |
In the rewrite of the flow allocator for per-port ring buffers, the
messages were removed because they were not accurate anymore. This
adds messages when a port is allocated.
|
|
|
|
|
|
|
| |
Deallocation was reverted to a synchronoous operation between the AP,
IRMd and IPCP in order to avoid inconsistent states of the port_id.
Fixes some memory leaks, particularly the shm_flow_set is now closed
upon deallocation.
|
|
|
|
|
|
|
|
|
|
|
| |
The BSD compiler requires explicit initializers for all fields with
the -Wmissing-field-initializers flag enabled
On branch be-fix-build # Changes to be committed: # modified:
src/irmd/main.c # modified: src/tools/irm/irm.c # modified:
src/tools/irm/irm_bind.c # modified: src/tools/irm/irm_ipcp.c #
modified: src/tools/irm/irm_unbind.c # # Untracked files: #
killouroboros.sh # start.sh #
|
|
|
|
|
| |
This has the code checked with -Wcast-qual and -Wconversion flags.
These flags were removed because SWIG generated code fails.
|
|
|
|
|
| |
This reduces the risk for some bugs, for instance due to
signed/unsigned mismatches and unused variables.
|
|
|
|
|
| |
This stabilises flow allocation now that the rbuffs are created upon
flow allocation. Only the IRMd can sync this process sufficiently.
|
|
|
|
|
|
|
|
| |
The fast path will now use an incoming ring buffer per flow per
process. This necessitated the development of a new method for the
asynchronous io call, which is now based on an event queue system for
scalability (fqueue). The ipcpd's and tools have been updated to this
API.
|
|
|
|
|
| |
Both the N and N-1 entity must now call flow_dealloc for the port_id
to be released.
|
|
|
|
|
| |
Flow deallocation from the application will immediately return (void
call). The IRMd will not send a reply message.
|
|
|
|
|
|
|
|
|
|
|
|
| |
IPCPs can now use ap_init() to initialize the memory. All flows are
accessed using flow descriptors, this greatly simplifies IPCP
development. Reverts the fast path to a single ap_rbuff per process.
Splits lib/ipcp into irmd/ipcp and lib/ipcp-dev. Adds a lib/shim-dev
holding tailored functions for shims. Moves the buffer_t to utils.h.
Fixes the shim-eth-llc length field. Removes the flow from shared.h.
Fixes #4
Fixes #5
|
|
|
|
|
| |
Permissions are now set correctly upon creation, removing the need to
call fchmod.
|
|
|
|
|
| |
If there are unregistered IPCPs (no DIF name set) the lookup should
skip over them without segfaulting.
|
|
|
|
| |
Fast path is split in north and southbound paths.
|
| |
|
|
|
|
|
| |
Flow will now allocate correctly when the name is registered in a DIF
before it is bound to an AP.
|
|
|
|
|
|
|
|
|
|
| |
The shm_du_map is renamed to shm_rdrbuff to reflect the Random
Deletion Ringbuffer used in the implementation. The close_on_exit call
is removed and SDUs are cleaned up by the application in the ap_fini()
call. This required a non-blocking peek() operation in the shm_ap_rbuff.
Some initial implementation for future support of qos cubes has been
added to the shm_rdrbuff.
|
|
|
|
| |
Prevents double destruction.
|
|
|
|
|
| |
Flows are deallocated first, then the socket is closed to prevent
IPCPs hanging on the connect.
|