| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Write a "test-profile" that does echo client-server with threads;
profile reveals lock contention, memcpy/realloc of buffers, and
UTF-8 validation as hot spots. 20% of lock contention eliminated
with dbus_atomic_inc/dec implementation on x86. Much remaining
contention is global mempool locks for GList and DBusList.
* dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec): add
x86 implementation
* dbus/dbus-connection.c (struct DBusConnection): use
dbus_atomic_t for the reference count
* dbus/dbus-message.c (struct DBusMessage): declare
dbus_atomic_t values as volatile
* configure.in: code to detect ability to use atomic integer
operations in assembly, from GLib patch
* dbus/dbus-internals.c (_dbus_verbose_real): call getpid every
time, tired of it being wrong in threads and forked processes
* glib/test-profile.c: a little program to bounce messages back
and forth between threads and eat CPU
* dbus/dbus-connection.c: add debug spew macros for debugging
thread locks; include config.h at top; fix deadlock in
dbus_connection_flush()
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-auth.c (record_mechanisms): memleak fixes
* dbus/dbus-sysdeps.c (_dbus_string_save_to_file): fix some
memleaks
* dbus/dbus-keyring.c (add_new_key): fix a memleak, and
on realloc be sure to update the pointer in the keyring
* dbus/dbus-string.c (_dbus_string_zero): compensate for align
offset to avoid writing to unallocated memory
* dbus/dbus-auth.c (process_rejected): return FALSE if we fail to
try the next mechanism, so we properly handle OOM
* dbus/dbus-keyring.c (_dbus_keyring_new_homedir): fix double-free
on OOM.
(_dbus_keyring_new): fix OOM bug
(_dbus_keyring_new_homedir): always set error; impose a maximum
number of keys we'll load from the file, mostly to speed up the
test suite and make its OOM checks more useful, but also for
general sanity.
* dbus/dbus-auth.c (process_error_server): reject authentication
if we get an error from the client
(process_cancel): on cancel, send REJECTED, per the spec
(process_error_client): send CANCEL if we get an error from the
server.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-mainloop.c (_dbus_loop_iterate): fix UMR in verbose
debug spew
* dbus/dbus-auth.c (handle_client_data_cookie_sha1_mech): fix OOM
handling problem
* dbus/dbus-keyring.c (_dbus_keyring_new_homedir): only whine
about DBUS_TEST_HOMEDIR once
* bus/Makefile.am (TESTS_ENVIRONMENT): put DBUS_TEST_HOMEDIR in
the environment
* bus/dispatch.c (bus_dispatch_sha1_test): actually load sha1
config file so we test the right thing
Throughout: assorted docs improvements
|
| |
|
|
|
| |
* dbus/dbus-userdb.c, dbus/dbus-sysdeps.c: redo all the passwd
database usage so it all goes via the DBusUserDatabase cache.
|
| |
|
|
|
|
|
|
|
|
|
| |
* bus/config-parser.c (bus_config_parser_new): fix a memleak
* dbus/dbus-sysdeps.c: change DBusCredentials to use longs for
the pid/gid/uid, just for paranoia.
* test/break-loader.c (randomly_do_n_things): find a byte
containing a type code, and randomly change it to a different
type code.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-server.c (dbus_server_set_auth_mechanisms): new
function
* dbus/dbus-auth.c (_dbus_auth_set_mechanisms): new
* dbus/dbus-internals.c (_dbus_dup_string_array): new function
* dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): chmod the
socket 0777, and unlink any existing socket.
* bus/bus.c (bus_context_new): change our UID/GID and fork if
the configuration file so specifies; set up auth mechanism
restrictions
* bus/config-parser.c (bus_config_parser_content): add support
for <fork> option and fill in code for <auth>
* bus/system.conf.in: add <fork/> to default configuration,
and limit auth mechanisms to EXTERNAL
* doc/config-file.txt (Elements): add <fork>
* dbus/dbus-sysdeps.c (_dbus_become_daemon): new function
(_dbus_change_identity): new function
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Fix some annoying DBusString API and fix all affected code.
* dbus/dbus-string.c (_dbus_string_init): get rid of annoying
max_length argument
(_dbus_string_get_data): change to return string instead of using
an out param
(_dbus_string_get_const_data): ditto
(_dbus_string_get_data_len): ditto
(_dbus_string_get_const_data_len): ditto
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* bus/test.c (bus_test_flush_bus): remove the sleep from here,
I think it may have just been superstition. Not sure.
* dbus/dbus-string.c (_dbus_string_base64_decode): catch some OOM
failures that were not being handled.
* dbus/dbus-auth.c (process_auth): fix a memleak in OOM handling
* dbus/dbus-memory.c: add ability to set number of mallocs in a
row that will fail on out-of-memory.
* dbus/dbus-internals.c (_dbus_test_oom_handling): convenience
function for testing out-of-memory handling.
* bus/config-loader-expat.c (memsuite): don't wrap the dbus
allocation functions, they do map exactly to the expat ones.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
All tests pass, no memleaks, no valgrind complaints.
* bus/test.c: refcount handler_slot
* bus/connection.c (bus_connections_new): refcount
connection_data_slot
* dbus/dbus-auth-script.c (_dbus_auth_script_run): delete unused
bytes so that auth scripts pass.
* bus/dispatch.c: init message_handler_slot so it gets allocated
properly
* bus/dispatch.c (message_handler_slot_ref): fix memleak
* dbus/dbus-server-debug-pipe.c (_dbus_server_debug_pipe_new):
dealloc server_pipe_hash when no longer used for benefit of
leak checking
* dbus/dbus-auth.c (process_command): memleak fix
* bus/dispatch.c (check_hello_message): memleak fix
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-watch.c (_dbus_watch_new): handle failure to malloc
the watch
* dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
add some missing dbus_set_result
* bus/dispatch.c (bus_dispatch_add_connection): handle failure to
alloc the DBusMessageHandler
* dbus/dbus-transport.c (_dbus_transport_disconnect): don't ref
the transport here, since we call this from the finalizer; it
resulted in a double-finalize.
* dbus/dbus-transport.c (_dbus_transport_disconnect): fix a bug
where we tried to use transport->connection that was NULL,
happened when transport was disconnected early on due to OOM
* bus/*.c: adapt to handle OOM for watches/timeouts
* dbus/dbus-transport-unix.c: port to handle OOM during
watch handling
* dbus/dbus-auth.c (_dbus_auth_get_unused_bytes): return a
reference to unused bytes instead of a copy
* dbus/dbus-server.c (dbus_server_handle_watch): return FALSE for
out of memory
* dbus/dbus-connection.c (dbus_connection_handle_watch): return
FALSE on OOM
* dbus/dbus-timeout.c (dbus_timeout_handle): return FALSE for out
of memory
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* bus/dispatch.c (bus_dispatch_test): do test using debug-pipe
transport, tests more of the real codepath. Set up clients
with bus_setup_debug_client.
* bus/test.c (bus_setup_debug_client): function to set up debug
"clients" on the main loop
* dbus/dbus-transport.c (_dbus_transport_open): add debug-pipe
support
* dbus/dbus-server.c (dbus_server_listen): add debug-pipe
server type
* dbus/dbus-server-debug.c: support a debug server based on pipes
* dbus/dbus-sysdeps.c (_dbus_full_duplex_pipe): new function
(_dbus_close): new function
* configure.in: check for socketpair
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Mega-patch that gets the message bus daemon initially handling
out-of-memory. Work still needed. Also lots of random
moving stuff to DBusError instead of ResultCode.
* dbus/dbus-list.c (_dbus_list_length_is_one): new function
* dbus/dbus-connection.c
(dbus_connection_send_with_reply_and_block): use DBusError
* dbus/dbus-bus.c: adapt to API changes, make it use DBusError not
DBusResultCode
* dbus/dbus-connection.c (dbus_connection_send): drop the result
code here, as the only failure possible is OOM.
* bus/connection.c (bus_connection_disconnect):
rename bus_connection_disconnected as it's a notification only
* bus/driver.c (bus_driver_handle_acquire_service): don't free
"name" on get_args failure, should be done by get_args;
don't disconnect client for bad args, just return an error.
(bus_driver_handle_service_exists): ditto
* bus/services.c (bus_services_list): NULL-terminate returned array
* bus/driver.c (bus_driver_send_service_lost)
(bus_driver_send_service_acquired): send messages from driver to a
specific client to the client's unique name, not to the broadcast
service.
* dbus/dbus-message.c (decode_header_data): reject messages that
contain no name field
(_dbus_message_get_client_serial): rename to
dbus_message_get_serial and make public
(_dbus_message_set_serial): rename from set_client_serial
(_dbus_message_set_reply_serial): make public
(_dbus_message_get_reply_serial): make public
* bus/connection.c (bus_connection_foreach): allow stopping
iteration by returning FALSE from foreach function.
* dbus/dbus-connection.c (dbus_connection_send_preallocated)
(dbus_connection_free_preallocated_send)
(dbus_connection_preallocate_send): new API for sending a message
without possibility of malloc failure.
(dbus_connection_send_message): rename to just
dbus_connection_send (and same for whole function family)
* dbus/dbus-errors.c (dbus_error_free): make this reinit the error
* dbus/dbus-sysdeps.c (_dbus_exit): new function
* bus/activation.c: handle/return errors
* dbus/dbus-errors.h: add more DBUS_ERROR #define
* dbus/dbus-sysdeps.c (_dbus_directory_open) (_dbus_file_get_contents)
(_dbus_directory_get_next_file): use DBusError instead of DBusResultCode
(_dbus_result_from_errno): move to this file
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* test/data/auth/*: adapt to changes
* dbus/dbus-auth-script.c (_dbus_auth_script_run): add
USERID_BASE64 and change USERNAME_BASE64 to put in username not
userid
* dbus/dbus-keyring.c (_dbus_keyring_validate_context): prevent
more stuff from being in a context name, to make the protocol
simpler to deal with
* dbus/dbus-errors.c (dbus_error_has_name): new function
(dbus_error_is_set): new function
* dbus/dbus-auth.c: replace DBUS_STUPID_TEST_MECH auth
with DBUS_COOKIE_SHA1, implement DBUS_COOKIE_SHA1
* dbus/dbus-connection.c (dbus_connection_flush): also read
messages during a flush operation
* dbus/Makefile.am: remove dbus-md5 since it isn't currently used.
|
| |
|
|
|
|
|
|
|
| |
* dbus/dbus-auth.c: (send_rejected), (process_test_subdir):
* dbus/dbus-keyring.c: (_dbus_keyring_new_homedir),
(_dbus_keyring_test):
* dbus/dbus-md5.c: (_dbus_md5_compute):
* dbus/dbus-sha.c: (_dbus_sha_compute):
Plug memory leaks.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-auth.c (handle_server_data_stupid_test_mech): Just get
credentials from our currently running process.
(get_word): Fix a buglet where we were copying the entire length
instead of relative to our position.
* dbus/dbus-hash.c (_dbus_hash_test): Don't try to allocate the
keys on the stack... it's 640k of data.
* dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): Always
read the credentials byte off the socket, even if we don't have
SO_PEERCRED.
(_dbus_poll): Implement poll() using select() for systems which
don't have it.
* glib/test-dbus-glib.c (main): Print out an error if no
parameters are given.
* test/data/auth/fallback.auth-script: Added. Tests that a client
can fallback to a secondary auth mechanism if the first fails.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
* 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.
|
| |
|
|
|
|
|
| |
* test/data/auth/fail-after-n-attempts.auth-script: new test
* dbus/dbus-auth.c (send_rejected): shutdown_mech() when we
reject the client.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-auth.c (handle_server_data_external_mech): args to
dbus_credentials_match were backward
* dbus/dbus-auth-script.c (_dbus_auth_script_run): support
NO_CREDENTIALS and ROOT_CREDENTIALS
* dbus/dbus-auth.c (_dbus_auth_do_work): move get_state() routine
into here. Never process more commands after we've reached an
end state; store further data as unused bytes.
* test/data/auth/*: add more auth tests
* dbus/dbus-auth-script.c (_dbus_auth_script_run): support EXPECT
command to match exact string and EXPECT_UNUSED to match unused
bytes
* test/Makefile.am (dist-hook): fix to dist all the test stuff
|
| |
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-string.c (_dbus_string_pop_line): fix to also strip
\r off of popped lines
* dbus/dbus-auth.c (_dbus_auth_test): write code to run auth
scripts
* dbus/dbus-auth-script.c (_dbus_auth_script_run): when doing a
SEND, append \r\n
|
| |
|
|
|
|
|
|
|
|
| |
* dbus/dbus-auth-script.c, dbus/dbus-auth-script.h: sync
initial cut at test framework for DBusAuth from laptop.
Doesn't quite work yet but it compiles and I need to get
it off the 266mhz laptop. ;-)
* dbus/dbus-server-debug.c (_dbus_server_debug_accept_transport):
fix a memleak in error case
|
| |
|
|
|
|
|
|
|
|
| |
* dbus/dbus-keyring.c, dbus/dbus-keyring.h: template files
for code to manage cookies in your home directory
* dbus/dbus-sysdeps.c (_dbus_generate_random_bytes): new function
* dbus/dbus-auth.c (get_state): impose a maximum number of tries
to authenticate, then disconnect the client.
|
| |
|
|
|
|
|
|
| |
* dbus/dbus-auth.c: rework to use only REJECTED, no
MECHANISMS
* doc/dbus-sasl-profile.txt: drop MECHANISMS and just
use REJECTED, suggested by Mark McLoughlin
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-auth.c: (process_auth), (_dbus_auth_unref):
* dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
(dbus_connection_unref):
* dbus/dbus-marshal.c: (_dbus_marshal_test):
* dbus/dbus-message.c: (dbus_message_unref),
Plug memory leaks.
(dbus_message_get_fields):
Remove debugging printout.
(_dbus_message_loader_return_buffer):
Don't store the header string.
(_dbus_message_test):
Plug leaks.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* test/watch.c (error_handler): make it safe if the error handler
is called multiple times (if we s/error handler/disconnect
handler/ we should just guarantee it's called only once)
* dbus/dbus-transport.c (_dbus_transport_disconnect): call the
error handler on disconnect (it's quite possible we should
just change the error handler to a "disconnect handler," I'm
not sure we have any other meaningful errors)
* configure.in: check for getpwnam_r
* dbus/dbus-transport.c, dbus/dbus-transport-unix.c,
dbus/dbus-auth.c: add credentials support, add EXTERNAL auth
mechanism as in SASL spec, using socket credentials
* dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): new function
(_dbus_send_credentials_unix_socket): new function
* dbus/dbus-sysdeps.c (_dbus_accept_unix_socket): rename just
dbus_accept()
(_dbus_write): only check errno if <0 returned
(_dbus_write_two): ditto
|
| |
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-auth.c: fixes fixes fixes
* dbus/dbus-transport-unix.c: wire up support for
encoding/decoding data on the wire
* dbus/dbus-auth.c (_dbus_auth_encode_data)
(_dbus_auth_decode_data): append to target string
instead of nuking it.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-marshal.h (DBUS_COMPILER_BYTE_ORDER): #ifdef
WORDS_BIGENDIAN then compiler byte order is DBUS_BIG_ENDIAN,
doh
* dbus/dbus-marshal.c: Add macros to do int swapping in-place and
avoid swap_bytes() overhead (ignoring possible assembly stuff for
now). Main point is because I wanted unpack_uint32 to implement
_dbus_verbose_bytes
(_dbus_verbose_bytes): new function
* dbus/dbus-string.c (_dbus_string_validate_ascii): new function
* dbus/dbus-message.c (_dbus_message_loader_get_is_corrupted): add
mechanism to handle a corrupt message stream
(_dbus_message_loader_new): fix preallocation to only prealloc,
not prelengthen
* dbus/dbus-string.c (_dbus_string_skip_blank): fix this function
(_dbus_string_test): enhance tests for copy/move and fix the
functions
* dbus/dbus-transport-unix.c: Hold references in more places to
avoid reentrancy problems
* dbus/dbus-transport.c: ditto
* dbus/dbus-connection.c (dbus_connection_dispatch_message): don't
leak reference count in no-message case
* test/watch.c (do_mainloop): handle adding/removing watches
during iteration over the watches. Also, ref the connection/server
stored on a watch, so we don't try to mangle a destroyed one.
* dbus/dbus-transport-unix.c (do_authentication): perform
authentication
* dbus/dbus-auth.c (get_state): add a state
AUTHENTICATED_WITH_UNUSED_BYTES and return it if required
(_dbus_auth_get_unused_bytes): append the unused bytes
to the passed in string, rather than prepend
* dbus/dbus-transport.c (_dbus_transport_init_base): create
the auth conversation DBusAuth
* dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd)
(_dbus_transport_new_for_domain_socket): when creating a
transport, pass in whether it's a client-side or server-side
transport so we know which DBusAuth to create
|
|
|
* doc/dbus-sasl-profile.txt: docs on the authentication protocol,
it is a simple protocol that just maps directly to SASL.
* dbus/dbus-auth.h, dbus/dbus-auth.c: authentication protocol
initial implementation, not actually used yet.
* dbus/dbus-string.c (_dbus_string_find): new function
(_dbus_string_equal): new function
(_dbus_string_base64_encode): new function
(_dbus_string_base64_decode): new function
|
