]> rtime.felk.cvut.cz Git - frescor/forb.git/log
frescor/forb.git
13 years agoforb: Split forb_port_destroy() to stop and destroy phases master
Michal Sojka [Fri, 25 Feb 2011 07:20:28 +0000 (08:20 +0100)]
forb: Split forb_port_destroy() to stop and destroy phases

This allows for proper termination of forb. Now, there are no leaked
file descriptors in the tests.

13 years agoforb: Set cancelation points in receiver thread
Michal Sojka [Fri, 25 Feb 2011 07:17:34 +0000 (08:17 +0100)]
forb: Set cancelation points in receiver thread

This is to ensure that the thread is canceled only in a defined place
and no resources (like references to peer structures) remain in the system.

13 years agoDestroy forb in the correct order
Michal Sojka [Thu, 24 Feb 2011 15:58:36 +0000 (16:58 +0100)]
Destroy forb in the correct order

This is to avoid deadlocks during forb_destroy().

13 years agoforb: Update documentation of forb_executor_run()
Michal Sojka [Thu, 24 Feb 2011 15:40:52 +0000 (16:40 +0100)]
forb: Update documentation of forb_executor_run()

13 years agoforb: Add forb_executor_synchronize()
Michal Sojka [Thu, 24 Feb 2011 15:24:16 +0000 (16:24 +0100)]
forb: Add forb_executor_synchronize()

This is used to wait for executors to finish processing ongoing requests
during FORB termination. This ensures that all peer sockets are closed
and wvtest does not complain about open file descriptors at the end of
tests.

13 years agoforb: Update documentation of forb_request_send()
Michal Sojka [Thu, 24 Feb 2011 13:47:28 +0000 (14:47 +0100)]
forb: Update documentation of forb_request_send()

13 years agoforb: Update forb_server_id_to_string() documentation
Michal Sojka [Tue, 22 Feb 2011 11:41:42 +0000 (12:41 +0100)]
forb: Update forb_server_id_to_string() documentation

13 years agoforbrun: Support --serverid command line switch
Michal Sojka [Tue, 22 Feb 2011 11:41:22 +0000 (12:41 +0100)]
forbrun: Support --serverid command line switch

13 years agoforb: forbrun passes the parameters to shlibs properly
Petr Benes [Tue, 22 Feb 2011 11:24:27 +0000 (12:24 +0100)]
forb: forbrun passes the parameters to shlibs properly

13 years agoforbrun: Correctly handle arguments to -i switch
Michal Sojka [Tue, 22 Feb 2011 10:38:05 +0000 (11:38 +0100)]
forbrun: Correctly handle arguments to -i switch

13 years agoforb: Fix synchronization error in forbrun
Michal Sojka [Sun, 20 Feb 2011 09:42:58 +0000 (10:42 +0100)]
forb: Fix synchronization error in forbrun

forbrun starts servers synchronously, i.e. the next server is started
only after the previous one is fully initialized (see
forb_signal_server_ready()). This synchronization was broken because
forb_signal_server_ready() was incorrectly called in
forb_execute_object() called from forb_execution_thread() called from
forb_init(). This caused that the semaphore for the synchronization had
the value of one (instead of zero) when forb_init() finished.
Therefore, the first waiting for the semaphore returned immediately
without actually waiting for the server.

This is now fixed by introduction of default_executor which is declared
in struct forb. This executor is initialized "manually", without
forb_execute_object(), so that the forb_signal_server_ready() is not
called.

In future, we might want use the default executor for executing requests
to other objects, which do not require a separate thread.

13 years agoforb: Regref test converted to wvtest style
Michal Sojka [Sun, 20 Feb 2011 08:56:06 +0000 (09:56 +0100)]
forb: Regref test converted to wvtest style

Propagation of references from one orb to another is currently broken
and the test for that is commented out.

13 years agoforb: Convert syncobj test to wvtest style
Michal Sojka [Sun, 20 Feb 2011 08:37:39 +0000 (09:37 +0100)]
forb: Convert syncobj test to wvtest style

13 years agoforb: Increase log level for executor test
Michal Sojka [Sat, 19 Feb 2011 22:59:33 +0000 (23:59 +0100)]
forb: Increase log level for executor test

13 years agoforb: Fix deadlock in executor
Michal Sojka [Sat, 19 Feb 2011 22:57:48 +0000 (23:57 +0100)]
forb: Fix deadlock in executor

The first iteration of executor loop waited for the request without
actually checking that there is no request in the queue. Therefore,
when the client sent the request before the executor called
fosa_cond_wait()  for the first time, the executor blocked indefinitely.

13 years agoforb: Convert some older tests to wvtest framework
Michal Sojka [Sat, 19 Feb 2011 15:22:37 +0000 (16:22 +0100)]
forb: Convert some older tests to wvtest framework

13 years agoforb: Fix incorrect parameter of forb_peer_put()
Michal Sojka [Sat, 19 Feb 2011 15:00:11 +0000 (16:00 +0100)]
forb: Fix incorrect parameter of forb_peer_put()

13 years agoforb: Free "new_peers" on destroy
Michal Sojka [Sat, 19 Feb 2011 14:59:15 +0000 (15:59 +0100)]
forb: Free "new_peers" on destroy

The so called new_peers were not freed and their sockets were not closed.
We are fixing it now.

13 years agoforb: Do not compile inet_broadcast if it is not used
Michal Sojka [Sat, 19 Feb 2011 13:42:16 +0000 (14:42 +0100)]
forb: Do not compile inet_broadcast if it is not used

When inet is the default protocol, broadcasts are not used (because they
are not reliable). Previously, if inet was the default, we didn't bind the
socket to multicast address and thus we were not able to receive the
broadcasts. This situation was confusing and therefore it is better omit
the implementation of inet_broadcast entirely to make this clear that it
is not used.

13 years agoforbrun: Enable uLUt-based logging
Michal Sojka [Fri, 18 Feb 2011 23:42:20 +0000 (00:42 +0100)]
forbrun: Enable uLUt-based logging

It had no sense to disable it -- it works fine for FORB logging. If it
is needed to change log levels of server, this could be implemented in
forb_main() of the particular server.

13 years agoforbrun: Do not set ID if it is not specified on command line
Michal Sojka [Fri, 18 Feb 2011 23:35:20 +0000 (00:35 +0100)]
forbrun: Do not set ID if it is not specified on command line

13 years agoforb: Destroy forb at the end of executor_test
Michal Sojka [Fri, 18 Feb 2011 23:27:42 +0000 (00:27 +0100)]
forb: Destroy forb at the end of executor_test

This make wvtest happy because it closes all the file descriptors
that were open by FORB so the final check in wvmain passes.

Now, all wvtests passes!

13 years agoforbrun: Comment out not passing test
Michal Sojka [Fri, 18 Feb 2011 23:19:46 +0000 (00:19 +0100)]
forbrun: Comment out not passing test

This test does not pass because the registered references are not
distributed to other servers. This will be hopefully implemented later
through some kind of name server.

13 years agoforb: Rephrase documentation
Michal Sojka [Fri, 18 Feb 2011 23:15:44 +0000 (00:15 +0100)]
forb: Rephrase documentation

13 years agoforbrun: Handle errors in test_client
Michal Sojka [Fri, 18 Feb 2011 22:32:40 +0000 (23:32 +0100)]
forbrun: Handle errors in test_client

13 years agoforbrun: Simplify argument processing
Michal Sojka [Fri, 18 Feb 2011 22:25:28 +0000 (23:25 +0100)]
forbrun: Simplify argument processing

13 years agoforb: Fix races between instances of forb_main_thread
Michal Sojka [Fri, 18 Feb 2011 22:03:58 +0000 (23:03 +0100)]
forb: Fix races between instances of forb_main_thread

The data passed to forb_main_thread must be allocated for each thread
separately. Otherwise, threads may access data intended for other threads.

13 years agoforb: Add tests for forbrun command line parameters
Michal Sojka [Fri, 18 Feb 2011 21:40:50 +0000 (22:40 +0100)]
forb: Add tests for forbrun command line parameters

13 years agoforb: Update forbrun usage help
Michal Sojka [Fri, 18 Feb 2011 21:19:23 +0000 (22:19 +0100)]
forb: Update forbrun usage help

13 years agoforb: fixed passing parameters to shared libraries in forbrun
Petr Benes [Thu, 17 Feb 2011 15:59:16 +0000 (16:59 +0100)]
forb: fixed passing parameters to shared libraries in forbrun

13 years agoforbrun: Exit whenever the first forb_main() returns
Michal Sojka [Thu, 17 Feb 2011 14:32:30 +0000 (15:32 +0100)]
forbrun: Exit whenever the first forb_main() returns

The exit code of forbrun is the returned value of forb_main().

13 years agoforb: Use new syntax for forbrun command line
Michal Sojka [Thu, 17 Feb 2011 14:29:37 +0000 (15:29 +0100)]
forb: Use new syntax for forbrun command line

We have added '--' before the name of the first library as it simplifies
command line parsing.

13 years agoforb: Add tests for forbrun
Michal Sojka [Thu, 17 Feb 2011 13:55:50 +0000 (14:55 +0100)]
forb: Add tests for forbrun

13 years agoforb: Move shared library tests under forbrun/ directory
Michal Sojka [Thu, 17 Feb 2011 11:18:57 +0000 (12:18 +0100)]
forb: Move shared library tests under forbrun/ directory

13 years agoforb: First implementation of forbrun
Michal Sojka [Tue, 15 Feb 2011 16:12:15 +0000 (17:12 +0100)]
forb: First implementation of forbrun

Not tested! forbrun should executed FORB servers implemented as shared
libraries. Its aim is to allow execution of multiple independent servers
in a single process. This is done in preparation of porting FRSH/FORB to
RTEMS (and other single-address space executives).

13 years agoforb: Add support for synchronized initialization of multiple servers
Michal Sojka [Tue, 15 Feb 2011 15:24:32 +0000 (16:24 +0100)]
forb: Add support for synchronized initialization of multiple servers

This is needed in forbrun (future commit) to start multiple servers in a
single address space.

13 years agoforb: Add comments to forb_daemon_prepare()
Michal Sojka [Tue, 15 Feb 2011 15:23:06 +0000 (16:23 +0100)]
forb: Add comments to forb_daemon_prepare()

13 years agoforb: Added shared library for testing of forbrun-based server execution
Michal Sojka [Tue, 15 Feb 2011 13:40:05 +0000 (14:40 +0100)]
forb: Added shared library for testing of forbrun-based server execution

13 years agoforb: object serialization working even for inter-executor invocation
Petr Benes [Tue, 15 Feb 2011 12:24:14 +0000 (13:24 +0100)]
forb: object serialization working even for inter-executor invocation

in case the object being deserialized indicates being local,
it is not being created from scratch, however just copied

13 years agoforb: End of header index removed from stub.
Petr Benes [Fri, 11 Feb 2011 15:22:43 +0000 (16:22 +0100)]
forb: End of header index removed from stub.

It is now embedded into the request structure. Does not make
so much mess in stubs.

13 years agoforb: Objects must be properly reference-counted
Michal Sojka [Fri, 11 Feb 2011 15:07:40 +0000 (16:07 +0100)]
forb: Objects must be properly reference-counted

This fixes some nasty segfaults.

13 years agoInter-executor invocation bugs fixed.
Petr Benes [Fri, 11 Feb 2011 14:35:55 +0000 (15:35 +0100)]
Inter-executor invocation bugs fixed.

Double free bug on request codec buffer fixed.
Stub index indicating a codec read position for inter-executor invocation fixed.

Test application->executor->executor does not fire properly. Needs a bug fix.

13 years agoStruct request adjusted for inter-executor invocation.
Petr Benes [Thu, 10 Feb 2011 14:27:26 +0000 (15:27 +0100)]
Struct request adjusted for inter-executor invocation.

Bug fix required.

13 years agoMerge branch 'master' of rtime.felk.cvut.cz:/frescor/frsh-forb
Petr Benes [Tue, 8 Feb 2011 17:05:38 +0000 (18:05 +0100)]
Merge branch 'master' of rtime.felk.cvut.cz:/frescor/frsh-forb

13 years agoforb: Add test for various invocation methods
Michal Sojka [Tue, 8 Feb 2011 17:01:37 +0000 (18:01 +0100)]
forb: Add test for various invocation methods

This tests the following:
1) remote invocation (currently broken)
2) local invocation from application to the executor
3) local invocation from an executor to another one
4) direct invocation from an executor to the same one

13 years agoIDL compiler: Invoke methods directly only within the same executor
Michal Sojka [Tue, 8 Feb 2011 16:45:13 +0000 (17:45 +0100)]
IDL compiler: Invoke methods directly only within the same executor

13 years agoSimplify frob_get_current_executor()
Michal Sojka [Tue, 8 Feb 2011 16:40:54 +0000 (17:40 +0100)]
Simplify frob_get_current_executor()

Return the executor as return value and not via a parameter.

13 years agoFixed synchronization for codec buffer.
Petr Benes [Tue, 8 Feb 2011 15:30:23 +0000 (16:30 +0100)]
Fixed synchronization for codec buffer.

13 years agoFixed typography.
Petr Benes [Tue, 8 Feb 2011 14:38:21 +0000 (15:38 +0100)]
Fixed typography.

13 years agoMerge branch 'master' of rtime.felk.cvut.cz:/frescor/frsh-forb
Petr Benes [Tue, 8 Feb 2011 13:53:32 +0000 (14:53 +0100)]
Merge branch 'master' of rtime.felk.cvut.cz:/frescor/frsh-forb

13 years agoPrepared for multiple executors in a single address space
Petr Benes [Tue, 8 Feb 2011 13:52:56 +0000 (14:52 +0100)]
Prepared for multiple executors in a single address space

Object implementation invocation prepared for a use of multiple
executors running in a single address space.

One can call the iop methods in the same way as for the case of
a remote invocation.

13 years agoDocument better forb_init()
Michal Sojka [Mon, 7 Feb 2011 20:49:25 +0000 (21:49 +0100)]
Document better forb_init()

13 years agoUsing plain POSIX instead of FOSA for thread specific data.
Petr Benes [Fri, 28 Jan 2011 17:47:01 +0000 (18:47 +0100)]
Using plain POSIX instead of FOSA for thread specific data.

13 years agoInitialize thread specific data in forb_init()
Michal Sojka [Fri, 28 Jan 2011 16:04:37 +0000 (17:04 +0100)]
Initialize thread specific data in forb_init()

This was done to avoid race conditions which causes errors in code.
Unfortunately this change does not help since FOSA is broken and we will
better use plain POSIX here.

13 years agoPrepared test cases for executors (thread specific data) benesp16
Petr Beneš [Tue, 25 Jan 2011 16:13:36 +0000 (17:13 +0100)]
Prepared test cases for executors (thread specific data)

tests/executor_id.c as a simple test
tests/executor_calls.c as a manually created stub

13 years agoAdded a function forb_object_get_executor() in object.h
Petr Beneš [Tue, 25 Jan 2011 16:10:14 +0000 (17:10 +0100)]
Added a function forb_object_get_executor() in object.h

13 years agoAdded a function forb_get_current_executor()
Petr Beneš [Tue, 25 Jan 2011 16:08:54 +0000 (17:08 +0100)]
Added a function forb_get_current_executor()

13 years agoAdded thread specific data to executor
Petr Beneš [Sun, 17 Oct 2010 23:48:07 +0000 (01:48 +0200)]
Added thread specific data to executor

Edited executor.h and executor.c
Requires a testcase

13 years agoLog time
Michal Sojka [Fri, 25 Jun 2010 22:54:47 +0000 (00:54 +0200)]
Log time

13 years agoAllow specifying loglevels in environment
Michal Sojka [Fri, 25 Jun 2010 22:54:17 +0000 (00:54 +0200)]
Allow specifying loglevels in environment

13 years agoAdd ul_log to executor
Michal Sojka [Fri, 25 Jun 2010 22:51:35 +0000 (00:51 +0200)]
Add ul_log to executor

13 years agoGet compilable output for sequences of a typedeffed type
Michal Sojka [Thu, 24 Jun 2010 22:38:26 +0000 (00:38 +0200)]
Get compilable output for sequences of a typedeffed type

This change is a workaround rather than a real fix. This way be generate
unnecessarily more [de]serialization functions even if they do the same
as the function of the original type.

13 years agotest: Sequence is initilized to zero
Michal Sojka [Wed, 23 Jun 2010 11:16:28 +0000 (13:16 +0200)]
test: Sequence is initilized to zero

13 years agoAdd forb_sequence_ensure_allocated()
Michal Sojka [Tue, 22 Jun 2010 15:30:19 +0000 (17:30 +0200)]
Add forb_sequence_ensure_allocated()

13 years agoUnify parameters of forb_sequence_*()
Michal Sojka [Tue, 22 Jun 2010 11:04:32 +0000 (13:04 +0200)]
Unify parameters of forb_sequence_*()

Now, all macros take the pointer to the sequence.

13 years agoAdd sequence tests
Michal Sojka [Mon, 21 Jun 2010 16:45:04 +0000 (18:45 +0200)]
Add sequence tests

13 years agoAdd more sequence helper macros
Michal Sojka [Mon, 21 Jun 2010 16:44:46 +0000 (18:44 +0200)]
Add more sequence helper macros

13 years agoDocument forb_daemon_ready()
Michal Sojka [Sun, 20 Jun 2010 08:45:14 +0000 (10:45 +0200)]
Document forb_daemon_ready()

13 years agoAdd helper functions for daemonizing servers
Michal Sojka [Fri, 18 Jun 2010 23:37:51 +0000 (01:37 +0200)]
Add helper functions for daemonizing servers

13 years agoRemove no longer valid comment (forb_destroy())
Michal Sojka [Fri, 18 Jun 2010 15:10:18 +0000 (17:10 +0200)]
Remove no longer valid comment (forb_destroy())

13 years agoDo not destroy forb automatically on exit
Michal Sojka [Fri, 18 Jun 2010 14:08:38 +0000 (16:08 +0200)]
Do not destroy forb automatically on exit

I do not know how to deregister the exit handler which is needed if
forb_destroy() is called manually. If the handler is not deregister
we get double free error.

14 years agoDo not export config.h
Michal Sojka [Mon, 19 Apr 2010 07:53:38 +0000 (09:53 +0200)]
Do not export config.h

It collides with other packages and it is not necessary to export it at
all.

14 years agoDebug messages display also the orb-id of disconnected peer
Michal Sojka [Tue, 8 Dec 2009 15:56:41 +0000 (16:56 +0100)]
Debug messages display also the orb-id of disconnected peer

14 years agoDon't print the annoying loglevel in log messages
Michal Sojka [Fri, 27 Nov 2009 10:24:46 +0000 (11:24 +0100)]
Don't print the annoying loglevel in log messages

14 years agoIntroduce new ul_log function which prints also process name
Michal Sojka [Fri, 27 Nov 2009 10:15:07 +0000 (11:15 +0100)]
Introduce new ul_log function which prints also process name

This is useful when debugging FRSH with multiple programs/resources
running.

14 years agoSuppressed unused function warning
Michal Sojka [Sun, 12 Jul 2009 15:01:22 +0000 (17:01 +0200)]
Suppressed unused function warning

14 years agoFixed warnings on printf() on 64 bit systems
Michal Sojka [Sun, 12 Jul 2009 14:44:07 +0000 (16:44 +0200)]
Fixed warnings on printf() on 64 bit systems

14 years agoAdded optional TCP_NODELAY to INET proto
Michal Sojka [Wed, 17 Jun 2009 17:22:17 +0000 (19:22 +0200)]
Added optional TCP_NODELAY to INET proto

This is to have graphs in benchmarks not influenced by Nagle's algorithm
in TCP.

14 years agoLoglevel of proto_inet changed to trash
Michal Sojka [Thu, 28 May 2009 11:34:33 +0000 (13:34 +0200)]
Loglevel of proto_inet changed to trash

14 years agoAdded redistribution of HELLO packets
Michal Sojka [Thu, 28 May 2009 03:29:59 +0000 (05:29 +0200)]
Added redistribution of HELLO packets

In FRSH_FORB, contract broker acts as central node and redistributes
HELLO packets to all connected applications. This way every application
can connect to every other.

14 years agoLength of all messages is set to be multiple of 8
Michal Sojka [Thu, 28 May 2009 02:13:15 +0000 (04:13 +0200)]
Length of all messages is set to be multiple of 8

This is to achieve proper alignment of headers on stream oriented connections.

14 years agoModified forb_init_attr to be C++ compatible
Michal Sojka [Thu, 28 May 2009 01:27:11 +0000 (03:27 +0200)]
Modified forb_init_attr to be C++ compatible

14 years agoAdded handling of FORB_EXTERNAL_IP environment variable
Michal Sojka [Wed, 27 May 2009 22:56:03 +0000 (00:56 +0200)]
Added handling of FORB_EXTERNAL_IP environment variable

This variable specifies the IP address of some local interface, which FORB
announces for contacting by others.

14 years agoInet protocol sends HELLO as the first message after connect()
Michal Sojka [Wed, 27 May 2009 21:34:29 +0000 (23:34 +0200)]
Inet protocol sends HELLO as the first message after connect()

This allows the receiver to know sender's listening address, which is different
from the address of the connect()ed socket. Besides the address, also orb_id is
sent, which makes debugging easier.

It this approach is found successful, this functionality should be moved to
iop.c and protocols should have new connect callback which will be called by iop
if the peer was never contacted.

14 years agoHandle EINTR in epoll
Michal Sojka [Wed, 27 May 2009 20:27:25 +0000 (22:27 +0200)]
Handle EINTR in epoll

14 years agoCorrect handling of closed TCP connections in proto_inet
Michal Sojka [Wed, 27 May 2009 18:33:14 +0000 (20:33 +0200)]
Correct handling of closed TCP connections in proto_inet

In FRSH_FORB this allows restarting resource managers without restart of fcb.
The problem of previous implementation was, that if the socket was closed but
peer structure was not unregistered the disconnected peer structure stored
fd of closed socket and when a new connection was made, the fd was the same
as stored in disconnected peer structure. Therefore the disconnected peer
appeared as alive.

14 years agoDiscovery handling moved to discovery.c
Michal Sojka [Wed, 27 May 2009 18:32:15 +0000 (20:32 +0200)]
Discovery handling moved to discovery.c

This was done to defined better (more higher level) interface to peer
manipulation. The interface is represented by forb_new_peer_discovered(),
forb_peer_disconnected() and forb_peer_find[_timer]().

14 years agoAdded more debugging messages to protocols
Michal Sojka [Wed, 27 May 2009 17:03:12 +0000 (19:03 +0200)]
Added more debugging messages to protocols

14 years agoFix error reporting in protocols
Michal Sojka [Wed, 27 May 2009 17:01:21 +0000 (19:01 +0200)]
Fix error reporting in protocols

Error is reported by returning -1, but size_t is unsigned type. All the
needed prototypes were changed to return signed ssize_t.

14 years agoforb_request_send() moved to iop.c
Michal Sojka [Wed, 27 May 2009 15:32:18 +0000 (17:32 +0200)]
forb_request_send() moved to iop.c

The reason is that we want all inter-orb protocol related messages in the same
log domain.

14 years agoFixed errors in INET proto test
Michal Sojka [Tue, 26 May 2009 17:08:00 +0000 (19:08 +0200)]
Fixed errors in INET proto test

14 years agoYet another debug message
Michal Sojka [Tue, 26 May 2009 16:50:13 +0000 (18:50 +0200)]
Yet another debug message

14 years agoAdded REUSEADDR to TCP socket
Michal Sojka [Tue, 26 May 2009 16:49:37 +0000 (18:49 +0200)]
Added REUSEADDR to TCP socket

14 years agoAdded mutex to protect request send from multiple threads
Michal Sojka [Tue, 26 May 2009 16:26:27 +0000 (18:26 +0200)]
Added mutex to protect request send from multiple threads

On TCP, requests sent from multiple threads could be theoretically
intermixed.

14 years agoAutomatically register FCB peer when INET is default
Michal Sojka [Tue, 26 May 2009 15:03:27 +0000 (17:03 +0200)]
Automatically register FCB peer when INET is default

FCB is always found on a fixed port. If FCB_ADDR environment variable
it not set, it is registered on 127.0.0.1, otherwise it is registered
on address according to the FCB_ADDR.

14 years agoDisable discovery protocol when INET is default
Michal Sojka [Tue, 26 May 2009 15:01:50 +0000 (17:01 +0200)]
Disable discovery protocol when INET is default

14 years agoAllow bidirectional communication over TCP connection
Michal Sojka [Tue, 26 May 2009 15:00:28 +0000 (17:00 +0200)]
Allow bidirectional communication over TCP connection

14 years agoAdded register callback for protocols
Michal Sojka [Tue, 26 May 2009 14:57:14 +0000 (16:57 +0200)]
Added register callback for protocols

14 years agoAdd more debug messages
Michal Sojka [Tue, 26 May 2009 14:56:28 +0000 (16:56 +0200)]
Add more debug messages

14 years agoAutomatic registration of FCB's reference
Michal Sojka [Tue, 26 May 2009 14:54:36 +0000 (16:54 +0200)]
Automatic registration of FCB's reference

This is enabled only if FORB is compiled with FRSH_FORB and
INET protocol is default. We can do this because FCB has now fixed ID.