Running the "embedded tests" through valgrind revealed that before this
commit, we would have been willing to read up to 3 bytes off the end of
a message if the message is truncated part way through a boolean. Any
practical allocator will round up allocations to the next 32-bit (or
larger) boundary, so in practice this will not leave the memory buffer
(and in particular did not crash during unit testing), but it could read
uninitialized contents.
On little-endian CPUs, an attacker might be able to use this to learn
whether up to 3 bytes of uninitialized memory in the dbus-daemon
were all-zero (their crafted message would be relayed) or not (their
connection would be disconnected for sending an invalid message). On
big-endian CPUs, an attacker might be able to use this to learn whether
up to 3 bytes were all-zeroes (relayed to a cooperating peer), 0-2
bytes of all-zeroes followed by 0x01 (relayed to a cooperating peer),
or something else (disconnected). This is not believed to be exploitable
to leak interesting information.
Fixes: 62e46533 "hardcode dbus_bool_t to 32 bits"
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107332
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Thiago Macieira <thiago@kde.org>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
It's just painfully slow, particularly when we fork (as we do in
test-bus to test service activation).
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
Instead of having separate test wrappers for the cases that do and
don't take a DBusConnection, we can just pass a NULL DBusConnection
to the one that doesn't.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
These tests are very reliant on their custom LOG_COMPILER,
which AX_VALGRIND_CHECK replaces.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
We don't need to do this for connections that were never set up
with the main loop.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
If re-initializing the string fails, it will be left in a state
where it has a length of 0 and a NULL buffer. That's valid to
"free", but not valid to pass to rmdir().
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
Not all of these tests will be fully valgrind-clean yet (or perhaps
ever), but it's easier to add this to all of them than to think
about it.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
AX_VALGRIND_CHECK overrides LOG_COMPILER, which means we can't rely
on running under glib-tap-test.sh. Default to TAP mode by modifying
our (effective) argv instead.
If you really want the default behaviour (unstructured output) this
can still be achieved by adding some arguments that are a no-op,
such as `-m quick`.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
test_server_setup() takes a reference to the DBusServer, so we need
to release that ref by calling test_server_shutdown().
test_server_shutdown() also disconnects the server, so we don't need
to do that.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
If an implementation fails to listen, and a subsequent implementation
succeeds, then we would have leaked this. Detected by running
tests/loopback.c under valgrind.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
We currently get away with this because the connection isn't fully
freed before we exit, but the connection is meant to own the result
of _dbus_connection_get_credentials() (it's "(transfer none)" in
GLib terminology).
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
If they didn't, we'd probably leak the server and/or the error.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
The only place this is set in practice is in dbus-server-win.c, which
does not set the error. If it did, dbus_server_listen() would leak it.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107194
Using strncpy (buffer, str, strlen (str)) is a "code smell" that
might indicate a serious bug (it effectively turns strncpy into
strcpy), and gcc 8 now warns about it. In fact we avoided the bug
here, but it wasn't at all obvious.
We already checked that path_len is less than or equal to
_DBUS_MAX_SUN_PATH_LENGTH, which is 99, chosen to be strictly less
than the POSIX minimum sizeof(sun_path) >= 100, so we couldn't
actually be overflowing the available buffer.
The new static assertion in this commit matches a comment above the
definition of _DBUS_MAX_SUN_PATH_LENGTH: we define
_DBUS_MAX_SUN_PATH_LENGTH to 99, because POSIX says struct
sockaddr_un's sun_path member is at least 100 bytes (including space
for a \0 terminator). dbus will now fail to compile on
platforms that are non-POSIX-compliant in this way, except for Windows.
We zeroed the struct sockaddr_un before writing into it, so stopping
one byte short of the end of sun_path ensures that we get \0
termination.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107350
Reviewed-by: Thiago Macieira <thiago@kde.org>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
In gcc 8, -Wall -Wextra includes -Wcast-function-type, which warns
about passing an extra (unwanted) parameter to callbacks. Instead
of using g_list_foreach(), open-code the equivalent.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107349
Reviewed-by: Thiago Macieira <thiago@kde.org>
In gcc 8, -Wall -Wextra includes -Wcast-function-type, which warns
about passing an extra (unwanted) parameter to callbacks. Instead
of using _dbus_list_foreach(), open-code the equivalent here.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107349
Reviewed-by: Thiago Macieira <thiago@kde.org>
In gcc 8, -Wall -Wextra includes -Wcast-function-type, which warns
about passing an extra (unwanted) parameter to callbacks. Instead
of using _dbus_list_foreach(), add a function to do what we actually
wanted here.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107349
Reviewed-by: Thiago Macieira <thiago@kde.org>
This was the only remaining symbol using the long prefix. Renaming it
gives us one consistent rule: symbols starting with dbus are public,
symbols starting with _dbus are not.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107349
Reviewed-by: Thiago Macieira <thiago@kde.org>
As in the previous commit, this prefix is meaningless in translation
units that don't get compiled into libdbus.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107349
Reviewed-by: Thiago Macieira <thiago@kde.org>
The naming convention dbus_internal_do_not_use_foo() was for functions
that had to be exported by libdbus but called by the embedded tests.
This is obsolete (in favour of _dbus_foo()) now that we have
DBUS_PRIVATE_EXPORT, and is doubly useless in this case because these
functions aren't even in libdbus - they're local to dbus-message-util.c.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107349
Reviewed-by: Philip Withnall <withnall@endlessm.com>
In gcc 8, -Wall -Wextra includes -Wcast-function-type, which warns
about casting a function pointer to an incompatible type. In this
case the cast was because we were ignoring the void * argument, which
in this case is NULL. Since this function is only used within
dbus-message-util.c anyway, we might as well just use the correct
signature and remove the cast.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=107349
Reviewed-by: Thiago Macieira <thiago@kde.org>
This is a bit more convenient than fetching it as-needed.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=105656
As this test's coverage expands, this function will have to do more
(clear up name watches, filters, etc.) so it'll be helpful to keep it
all in one place.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=105656
When I introduce per-container message filtering, it'll be useful to
be able to get the instance for a connection without worrying about
whether that connection is NULL (representing the dbus-daemon itself,
or an activatable service that has not yet been activated).
Also make it robust against Containers having not been initialized,
for completeness.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=105656
It isn't completely obvious that connection is allowed to be NULL here.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=105656
This makes it clearer that the only possible error is out-of-memory,
so its use in ListQueuedOwners() is not leaking information to callers
that might not be allowed to know the difference between "doesn't exist"
and "exists but you are not allowed to know that".
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=105656
This connection is the one looking at the name, as opposed to the
one that owns the name (if any).
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=105656
It seems I eventually introduce this into every project where I've
added GLib-based unit tests. Today it's dbus' turn.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=105656
For example, this can be the case in bubblewrap or Debian pbuilder after
unsharing the network namespace:
bwrap \
--bind / / \
--dev-bind /dev /dev \
--bind /dev/shm /dev/shm \
--bind /dev/pts /dev/pts \
--unshare-net \
${builddir}/test/test-loopback --tap
...
ok 1 /connect/tcp # SKIP Name resolution does not work here:
getaddrinfo("127.0.0.1", "0", {flags=ADDRCONFIG, family=INET,
socktype=STREAM, protocol=TCP}): Name or service not known
On some systems this can be circumvented by using nss_wrapper from
<https://cwrap.org/nss_wrapper.html>:
cat > hosts <<EOF
127.0.0.1 localhost
EOF
bwrap \
... \
env \
LD_PRELOAD=libnss_wrapper.so \
NSS_WRAPPER_HOSTS=$(pwd)/hosts \
${builddir}/test/test-loopback --tap
...
# listening at tcp:host=127.0.0.1,port=39219,family=ipv4,guid=...
but for systems where that does't work, we should be prepared to skip
the affected tests.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=106812
Pathological autobuilder environments might not list localhost in
/etc/hosts.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=106812
Minimal autobuilder environments don't always have working TCP,
so we may need to skip TCP tests. Make sure we test the equivalent
code paths via Unix sockets in those environments.
One notable exception is test/fdpass.c, which uses TCP as a transport
that is known not to be able to carry Unix fds; this needs to continue
to use TCP.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=106812
This expands test coverage, and lets us reuse the test for other
address schemes.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=106812
getaddrinfo and getnameinfo have their own error-handling convention
in which the library call returns either 0 or an EAI_* error code
unrelated to errno. If the error code is not EAI_SYSTEM, then
the value of errno is undefined (in particular it might be carried
over from a previous system call or library call). Introduce a
new helper function _dbus_error_from_gai() to handle this.
The equivalent code paths in Windows appear to be OK: the Windows
implementation of getaddrinfo() is documented to return a Winsock
error code, which we seem to be handling correctly.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Reviewed-by: Philip Withnall <withnall@endlessm.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=106395
