summaryrefslogtreecommitdiffstats
path: root/dbus/dbus-threads.c
Commit message (Collapse)AuthorAgeFilesLines
* 2003-03-16 Havoc Pennington <hp@pobox.com>Havoc Pennington2003-03-161-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Oops - test code was only testing failure of around 30 of the mallocs in the test path, but it turns out there are 500+ mallocs. I believe this was due to misguided linking setup such that there was one copy of dbus_malloc etc. in the daemon and one in the shared lib, and only daemon mallocs were tested. In any case, the test case now tests all 500+ mallocs, and doesn't pass yet, though there are lots of fixes in this patch. * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix this so that it doesn't need to allocate memory, since it has no way of indicating failure due to OOM (and would be annoying if it did). * dbus/dbus-list.c (_dbus_list_pop_first_link): new function * bus/Makefile.am: rearrange to create two self-contained libraries, to avoid having libraries with overlapping symbols. that was resulting in weirdness, e.g. I'm pretty sure there were two copies of global static variables. * dbus/dbus-internals.c: move the malloc debug stuff to dbus-memory.c * dbus/dbus-list.c (free_link): free list mempool if it becomes empty. * dbus/dbus-memory.c (_dbus_disable_mem_pools): new function * dbus/dbus-address.c (dbus_parse_address): free list nodes on failure. * bus/dispatch.c (bus_dispatch_add_connection): free message_handler_slot when no longer using it, so memory leak checkers are happy for the test suite. * dbus/dbus-server-debug-pipe.c (debug_finalize): free server name * bus/bus.c (new_connection_callback): disconnect in here if bus_connections_setup_connection fails. * bus/connection.c (bus_connections_unref): fix to free the connections (bus_connections_setup_connection): if this fails, don't disconnect the connection, just be sure there are no side effects. * dbus/dbus-string.c (undo_alignment): unbreak this * dbus/dbus-auth.c (_dbus_auth_unref): free some stuff we were leaking (_dbus_auth_new): fix the order in which we free strings on OOM failure * bus/connection.c (bus_connection_disconnected): fix to not send ServiceDeleted multiple times in case of memory allocation failure * dbus/dbus-bus.c (dbus_bus_get_base_service): new function to get the base service name (dbus_bus_register_client): don't return base service name, instead store it on the DBusConnection and have an accessor function for it. (dbus_bus_register_client): rename dbus_bus_register() * bus/dispatch.c (check_hello_message): verify that other connections on the bus also got the correct results, not just the one sending hello
* 2003-02-26 Havoc Pennington <hp@pobox.com>Havoc Pennington2003-02-261-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * dbus/dbus-connection.c (dbus_connection_send_message_with_reply_and_block): fix crash where we ref'd the outgoing message instead of the returned reply * dbus/dbus-transport-unix.c (do_authentication): check read watch at the end of this function, so if we didn't need to read for authentication, we reinstall it for receiving messages * dbus/dbus-message.c (dbus_message_new_reply): allow replies to a NULL sender for peer-to-peer case * dbus/dbus-transport-unix.c (check_read_watch): handle !authenticated case correctly * glib/dbus-gmain.c: add support for DBusServer * dbus/dbus-server.c: add data slot support * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): check return values and handle errors * dbus/dbus-dataslot.c: factor out the data slot stuff from DBusConnection * Doxyfile.in (INPUT): add glib subdir * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): rename setup_with_g_main instead of hookup_with_g_main; write docs
* 2003-02-19 Havoc Pennington <hp@pobox.com>Havoc Pennington2003-02-201-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | * Doxyfile.in (PREDEFINED): put DOXYGEN_SHOULD_SKIP_THIS in Doxyfile.in, not Doxyfile * dbus/dbus-keyring.c: do some hacking on this * dbus/dbus-sysdeps.c (_dbus_delete_file): new * dbus/dbus-errors.c (dbus_set_error_const): do not call dbus_error_init (dbus_set_error): remove dbus_error_init, check for message == NULL *before* we sprintf into it, and add @todo about including system headers in this file * dbus/dbus-sysdeps.c (_dbus_create_file_exclusively): new * dbus/dbus-errors.h (DBUS_ERROR_FAILED): add * dbus/dbus-sysdeps.c (get_user_info): break this function out to get various bits of user information based on either username or user ID (_dbus_homedir_from_username): new function
* 2003-02-19 Havoc Pennington <hp@pobox.com>Havoc Pennington2003-02-191-6/+2
| | | | | | | | | | | | | | | | | | | Throughout: mop up all the Doxygen warnings and undocumented stuff. * dbus/dbus-sysdeps.c (do_exec): do not use execvp, we don't want to search any paths. * dbus/dbus-threads.c: move global mutex initializers to dbus-internals.h, multiple prototypes was confusing doxygen besides being kind of ugly * Doxyfile (PREDEFINED): have Doxygen define DOXYGEN_SHOULD_SKIP_THIS so we can exclude things from docs with #ifndef DOXYGEN_SHOULD_SKIP_THIS (do not abuse the feature! it's for stuff like the autogenerated macros in dbus-md5.c, not just for things you don't feel like documenting...)
* 2003-02-16 Anders Carlsson <andersca@codefactory.se>Anders Carlsson2003-02-161-1/+1
| | | | | | | | | | | | * dbus/dbus-auth.c: (client_try_next_mechanism): Plug a leak. * dbus/dbus-threads.c: (dbus_condvar_wait_timeout): Return TRUE if there's no thread implementation around. * glib/dbus-gmain.c: (free_source), (dbus_connection_hookup_with_g_main): Make sure to remove the GSource when the connection is finalized.
* 2003-02-15 Alexander Larsson <alexl@redhat.com>Alexander Larsson2003-02-151-63/+153
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * dbus/dbus-threads.c: * dbus/dbus-threads.h: Add condvars. Remove static mutext from API. Implement static mutexes by initializing them from threads_init. * glib/dbus-gthread.c: * qt/dbus-qthread.cpp: Update with the thread api changes. * dbus/dbus-list.c: * dbus/dbus-list.h: Turn StaticMutex into normal mutex + init function. Export new functions _dbus_list_alloc_link, _dbus_list_free_link, _dbus_list_append_link, _dbus_list_prepend_link * dbus/dbus-sysdeps.c: * dbus/dbus-sysdeps.h: New type dbus_atomic_t, and new functions _dbus_atomic_inc, _dbus_atomic_dec. Only slow fallback implementation at the moment. * dbus/dbus-protocol.h: Add DBUS_MESSAGE_LOCAL_DISCONNECT define * dbus/dbus-message.c: Make ref/unref atomic. Fix some docs. * dbus/dbus-connection-internal.h: * dbus/dbus-connection.c: * dbus/dbus-connection.h: Make threadsafe. Change _peek to _borrow,_return & _steal_borrowed. Change disconnect callback to event. Make dbus_connection_dispatch_messages reentrant. * dbus/dbus-transport.c: Don't ref the connection on calls to the transport implementation. * dbus/dbus-message-handler.c: Make threadsafe. * glib/dbus-gmain.c: Don't use peek_message anymore * test/Makefile.am: * test/debug-thread.c: * test/debug-thread.h: Simple thread implementation that asserts() on deadlocks in single-threaded code. * test/bus-test.c: (main) Call debug_threads_init. * test/watch.c: Use disconnect message instead of disconnect callback. * bus/connection.c: * bus/connection.h: Don't call dbus_connection_set_disconnect_function. Instead export bus_connection_disconnect. * bus/dispatch.c: Call bus_connection_disconnect when we get a disconnected message.
* add a comment about needed fix for dbus_static_mutex_lockHavoc Pennington2003-01-061-1/+7
|
* 2003-01-05 Havoc Pennington <hp@pobox.com>Havoc Pennington2003-01-061-3/+66
| | | | | | | | | | | | | | | | | | * bus/connection.c: implement routines for handling connections, first thing is keeping a list of owned services on each connection and setting up watches etc. * bus/services.c: implement a mapping from service names to lists of connections * dbus/dbus-hash.c: add DBUS_HASH_POINTER * dbus/dbus-threads.c (dbus_static_mutex_lock): add functions to use static mutexes for global data * dbus/dbus-connection.c (dbus_connection_set_data): add new collection of functions to set/get application-specific data on the DBusConnection.
* 2002-12-15 Havoc Pennington <hp@pobox.com>Havoc Pennington2002-12-151-0/+148
* dbus/dbus-threads.c: add thread stubs that a higher library layer can fill in. e.g. the GLib wrapper might fill them in with GThread stuff. We still need to use this thread API to thread-safe-ize the library.