Ralf Habacker [Sat, 31 Oct 2015 20:05:03 +0000 (21:05 +0100)]
Fix typo and missing eol on debug message.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92721
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Thu, 29 Oct 2015 05:31:38 +0000 (06:31 +0100)]
refs test: reduce number of repeats under Wine
Under Wine, the API calls we use to do this are implemented via IPC
to wineserver, which makes it unreasonably slow to try to brute-force
bugs by having many threads stress-test refcounting. Do a few
repetitions just to verify that refcounting basically works, but
don't do the full stress-test.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92538
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
[rh: add WINESERVERSOCKET related hint]
Simon McVittie [Mon, 19 Oct 2015 15:02:36 +0000 (16:02 +0100)]
When running TAP tests, filter out trailing \r from Windows .exe
If we're running Windows executables using Wine, then tap-driver.sh
won't accept "1..4\r\n" as TAP syntax.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92538
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Ralf Habacker [Tue, 3 Feb 2015 18:56:32 +0000 (19:56 +0100)]
Fix crash running test-syslog on windows/wine.
Use seperate buffer in _dbus_system_logv() to avoid crash in vsprintf().
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92538
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Ralf Habacker [Sun, 4 Oct 2015 08:33:32 +0000 (10:33 +0200)]
Display autolaunch scope on verbose print of daemon found message on windows.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92080
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Mon, 26 Oct 2015 18:54:04 +0000 (18:54 +0000)]
1.10.3
Simon McVittie [Mon, 26 Oct 2015 12:54:26 +0000 (12:54 +0000)]
1.10.2
Simon McVittie [Mon, 19 Oct 2015 13:50:37 +0000 (14:50 +0100)]
Add a regression test for invalid BecomeMonitor calls
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92298
Simon McVittie [Mon, 26 Oct 2015 12:05:09 +0000 (12:05 +0000)]
NEWS
Jan Alexander Steffens (heftig) [Thu, 22 Oct 2015 17:50:29 +0000 (19:50 +0200)]
user dbus.socket: Set DBUS_SESSION_BUS_ADDRESS
As discussed in <https://github.com/systemd/systemd/issues/1600>.
See also <https://bugs.archlinux.org/task/46721>,
<https://bugzilla.gnome.org/show_bug.cgi?id=756420>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92612
[smcv: use AC_PATH_PROG to find systemctl; ignore systemctl failure]
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
[smcv: add links to earlier bug reports elsewhere]
Simon McVittie [Mon, 19 Oct 2015 14:19:27 +0000 (15:19 +0100)]
When running dbus-daemon --session in tests, override listen address
Otherwise, we can't reliably run tests for Windows, because the default
listening address on Windows is "autolaunch:" which is global to
a machine, resulting in testing an installed dbus-daemon instead of
the one we intended to test.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92538
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Mon, 19 Oct 2015 14:18:02 +0000 (15:18 +0100)]
Disable activation tests on Windows builds
These rely on the --systemd-activation code path, which is not
compiled for Windows.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92538
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Ralf Habacker [Wed, 21 Oct 2015 11:19:48 +0000 (13:19 +0200)]
Keep cmake build system in sync with autotools (add test-monitor).
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92298
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Mon, 19 Oct 2015 15:37:34 +0000 (16:37 +0100)]
NEWS
Simon McVittie [Tue, 6 Oct 2015 11:43:22 +0000 (12:43 +0100)]
BecomeMonitor: do not overwrite error with another error
If the user gave us a syntactically invalid error name, we'd
overwrite the MatchRuleInvalid error with NoMemory, causing an
assertion failure (crash) in the dbus-daemon.
This is not a denial-of-service vulnerability on the system bus,
because monitoring is a privileged action, and root privilege
is checked before this code is reached. However, it's an annoying
bug on the session bus.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92298
Reviewed-by: Philip Withnall <philip.withnall@collabora.co.uk>
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Fri, 2 Oct 2015 10:51:40 +0000 (11:51 +0100)]
NEWS
Simon McVittie [Thu, 1 Oct 2015 18:17:49 +0000 (19:17 +0100)]
On Windows, load local configuration relative to bus setup
This makes an installed tree with
/some-prefix/
etc/
dbus-1/
session-local.conf
share/
dbus-1/
session.conf
relocatable to any location.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92028
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Tested-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Wed, 30 Sep 2015 16:15:53 +0000 (17:15 +0100)]
Cancel pending activation on any activation error
This fixes the error reporting if you make two attempts
to activate a service that cannot be activated due to an
error that is reported synchronously, such as a system
service with no User= line in its .service file.
This is easy to reproduce with the gdbus(1) tool, which
sends an Introspect call in addition to the one you asked
it to. If you try to activate a service using
gdbus call --session -d com.example.FailToActivate \
-o / -m org.freedesktop.DBus.Peer.Ping
then gdbus will actually send two method calls: one
Introspect, and one Ping. The Introspect gets the correct
error reply, but when dbus-daemon enters
bus_activation_activate_service() for the Ping call, it
sees that there is a pending activation and does an
early-return. The pending activation does not finish
until the timeout is reached.
A couple of error cases handled this correctly, but the
majority did not; make them all go into the same code path.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92200
Reviewed-by: Thiago Macieira <thiago@kde.org>
Simon McVittie [Wed, 30 Sep 2015 14:49:06 +0000 (15:49 +0100)]
NEWS
Milan Crha [Fri, 18 Sep 2015 15:22:29 +0000 (16:22 +0100)]
Fix creation of Exec path for files not in prefix
Doing strcat() into a static buffer produces incorrect results for
the second and subsequent services if they are not in the ${prefix};
for example, if the first call should have returned
"C:\bar\bin\service1" and the second should have returned
"C:\bar\bin\service2", the second result would actually be
"C:\bar\bin\service1C:\bar\bin\service2".
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83539
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
[smcv: added commit message; used strncpy/strncat to avoid overflow]
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Ralf Habacker [Sun, 20 Sep 2015 11:47:23 +0000 (13:47 +0200)]
Make Windows dbus-daemon look for the config file we install
The canonical location for bus setup changed from
${sysconfdir}/dbus-1 to ${datadir}/dbus-1 (or their CMake
equivalents) in version 1.9.18.
Also stop trying to use bus/session.conf from the build tree,
which will not work if our ${prefix} contains an older
${sysconfdir}/dbus-1/session.conf.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=92028
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Tue, 25 Aug 2015 18:54:18 +0000 (19:54 +0100)]
mention o.fd.DBus.Verbose in the NEWS too
Simon McVittie [Tue, 25 Aug 2015 18:34:54 +0000 (19:34 +0100)]
start towards 1.10.2
Simon McVittie [Tue, 25 Aug 2015 15:51:59 +0000 (16:51 +0100)]
1.10.0
Simon McVittie [Wed, 19 Aug 2015 22:47:40 +0000 (23:47 +0100)]
audit: make the first few fds close-on-exec
libcap-ng < 0.7.7 leaks one non-close-on-exec fd during initialization.
test-bus asserts that all fds beyond 2 passed to an executed subprocess
have the close-on-exec flag set, which will fail at that leaked fd.
This was unnoticed until commit
517c4685, because libaudit was
previously only initialized if we were configured to switch uid,
which the regression tests do not do; the system bus is normally
the only place that happens, but the system bus is not normally
run with the "embedded tests" enabled (since they are bad
for performance and security).
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=91684
Reviewed-by: Philip Withnall <philip.withnall@collabora.co.uk>
Simon McVittie [Wed, 19 Aug 2015 21:04:30 +0000 (22:04 +0100)]
Distribute tap-test.sh.in unconditionally
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=91684
Reviewed-by: Tyler Hicks <tyhicks@canonical.com>
Simon McVittie [Thu, 6 Aug 2015 23:15:34 +0000 (00:15 +0100)]
1.9.21
Simon McVittie [Thu, 6 Aug 2015 17:55:31 +0000 (18:55 +0100)]
Prepare 1.9.20 (1.10rc1)
Simon McVittie [Thu, 6 Aug 2015 16:44:21 +0000 (17:44 +0100)]
NEWS
Laurent Bigonville [Mon, 15 Jun 2015 13:31:58 +0000 (15:31 +0200)]
Throw a warning if we cannot open the audit socket as a session bus
This will effectively print a warning when failing to open the audit
socket running as a session bus.
The call to audit_open() should succeed even if the dbus-daemon doesn't
have the CAP_AUDIT_WRITE capability.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83856
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Laurent Bigonville [Mon, 15 Jun 2015 13:30:16 +0000 (15:30 +0200)]
Initialize audit subsystem even for the session bus
If SELinux is enabled on the system, dbus will check the permissions but
no audit trails will be generated in case of denial as the audit
subsystem is not initialized. Same should apply for apparmor.
[smcv: without audit, the equivalent of the audit trail goes to stderr
where it can be picked up by systemd-journald]
A unprivileged user should be able to open the audit socket
(audit_open()) but should not have the permissions to log an audit
trail. The CAP_AUDIT_WRITE file capability could be set on the
dbus-daemon executable in order to allow the session bus to log an AVC
denial.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83856
[smcv: s/should/could/ in commit message to reflect lack of consensus that
"setcap cap_audit_write+ep dbus-daemon" is desirable in general]
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Fri, 3 Jul 2015 15:57:28 +0000 (16:57 +0100)]
audit: only check for CAP_AUDIT_WRITE once, during initialization
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89225
Reviewed-by: Colin Walters <walters@verbum.org>
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Thu, 19 Feb 2015 12:08:59 +0000 (12:08 +0000)]
audit: use DBUS_SYSTEM_LOG_WARNING if we cannot open the audit fd
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89225
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Thu, 19 Feb 2015 12:04:26 +0000 (12:04 +0000)]
bus: move shared libaudit code to a new audit.[ch]
This fixes various duplicated libaudit interactions in both
SELinux and AppArmor code paths, including opening two audit sockets
if both SELinux and AppArmor were enabled at compile time.
In particular, audit.c is now the only user of libcap-ng.
This commit is not intended to introduce any functional changes,
except for the de-duplication.
The actual audit_log_user_avc_message() call is still duplicated,
because the SELinux and AppArmor code paths use different mechanisms
to compose the audit message: the SELinux path uses a statically-sized
buffer on the stack which might be subject to truncation, whereas
the AppArmor path uses malloc() (via DBusString) and falls back to
using syslog on a memory allocation failure.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89225
Reviewed-by: Colin Walters <walters@verbum.org>
[smcv: minor issues raised during review are subsequently fixed]
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Rolland Dudemaine [Thu, 6 Aug 2015 13:01:06 +0000 (14:01 +0100)]
Use poll's data types directly, where used at all
[smcv: added commit message; moved fallback implementation
below definition of DBusPollable; more comments; removed unnecessary cast]
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90314
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Tue, 21 Jul 2015 20:13:43 +0000 (21:13 +0100)]
1.9.19
Simon McVittie [Tue, 21 Jul 2015 16:52:21 +0000 (17:52 +0100)]
dbus-1.9.16
Simon McVittie [Tue, 21 Jul 2015 16:32:17 +0000 (17:32 +0100)]
dbus-1.8.20
Simon McVittie [Tue, 21 Jul 2015 16:20:11 +0000 (17:20 +0100)]
Merge branch 'dbus-1.8'
Conflicts:
.gitignore
NEWS
tools/dbus-monitor.c
Simon McVittie [Tue, 21 Jul 2015 16:19:25 +0000 (17:19 +0100)]
NEWS for 1.8.x
Simon McVittie [Fri, 3 Jul 2015 12:50:04 +0000 (13:50 +0100)]
dbus-monitor: disable automatic handling of o.fd.Peer messages
A normal DBusConnection will automatically reply to o.fd.Peer
messages such as Ping. We don't want this: we'll
confuse everyone else by replying to messages that weren't
intended for us.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90952
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Philip Withnall <philip.withnall@collabora.co.uk>
(cherry picked from commit
d9ee040d0bff2b421bca80c2339dcd9347d906db,
commit message adjusted to describe the impact in versions < 1.9)
Conflicts:
tools/dbus-monitor.c
Simon McVittie [Mon, 3 Mar 2014 15:59:10 +0000 (15:59 +0000)]
.gitignore: ignore /test-driver, generated by recent Automake
(cherry picked from commit
1895a8ee5b0e3813b571e9db3528611802ea7f7e)
Simon McVittie [Fri, 3 Jul 2015 12:59:44 +0000 (13:59 +0100)]
Add test-case for the same situation as fd.o #90952
This does not directly test the code in the previous commit, but it does
confirm that calling dbus_connection_set_route_peer_messages() is enough
to fix the observed bug.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90952
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Philip Withnall <philip.withnall@collabora.co.uk>
[smcv: re-worded commit message in response to review]
Simon McVittie [Fri, 3 Jul 2015 12:50:04 +0000 (13:50 +0100)]
dbus-monitor: disable automatic handling of o.fd.Peer messages
A normal DBusConnection will automatically reply to o.fd.Peer
messages such as Ping. We don't want this: if we are using
traditional eavesdropping with an older dbus-daemon, we'll
confuse everyone else by replying to messages that weren't
intended for us. If we are using the new Monitoring
interface (since 1.9.12), the same still applies, but in
addition, the dbus-daemon will disconnect us for not being
a well-behaved monitor.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90952
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Philip Withnall <philip.withnall@collabora.co.uk>
Ralf Habacker [Tue, 21 Apr 2015 11:16:53 +0000 (13:16 +0200)]
bus_registry_new: Assert in case of not valid context parameter to avoid potiental crashes (CID 54764).
Because the context parameter is dereferenced several times in related code without a null check,
we need to make sure to have a valid context.
Reported by Coverity: CID 54764: Dereference after null check (FORWARD_NULL)
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90021
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Wed, 17 Jun 2015 19:41:24 +0000 (20:41 +0100)]
Merge branch 'dbus-1.8'
Conflicts:
NEWS
Simon McVittie [Wed, 17 Jun 2015 19:40:51 +0000 (20:40 +0100)]
NEWS
Jacek Bukarewicz [Wed, 17 Jun 2015 17:53:41 +0000 (18:53 +0100)]
Fix memleak in GetConnectionCredentials handler
Reply message was not unreferenced when GetConnectionCredentials
handler was successful.
Signed-off-by: Jacek Bukarewicz <j.bukarewicz@samsung.com>
[smcv: changed bus_message_unref() to dbus_message_unref()]
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=91008
Simon McVittie [Wed, 27 May 2015 10:42:44 +0000 (11:42 +0100)]
tests: use the new bus setup for `make installcheck`
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89280
Reviewed-by: Dimitri John Ledkov <dimitri.j.ledkov@intel.com>
Simon McVittie [Wed, 27 May 2015 10:42:18 +0000 (11:42 +0100)]
dbus-daemon.1: document the new locations
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89280
Reviewed-by: Dimitri John Ledkov <dimitri.j.ledkov@intel.com>
Simon McVittie [Wed, 27 May 2015 10:02:05 +0000 (11:02 +0100)]
NEWS
Dimitri John Ledkov [Mon, 2 Mar 2015 11:55:54 +0000 (11:55 +0000)]
Adjust cmake build to match autoconf installation locations.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89280
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Dimitri John Ledkov [Wed, 25 Feb 2015 15:39:04 +0000 (15:39 +0000)]
Move session & system bus configuration to datadir, by default.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89280
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Wed, 27 May 2015 09:42:46 +0000 (10:42 +0100)]
Start towards 1.9.18
Simon McVittie [Thu, 14 May 2015 14:04:36 +0000 (15:04 +0100)]
more NEWS, also more release name
Simon McVittie [Thu, 14 May 2015 13:50:24 +0000 (14:50 +0100)]
1.9.16
Simon McVittie [Thu, 14 May 2015 13:45:04 +0000 (14:45 +0100)]
Merge branch 'dbus-1.8'
Conflicts:
NEWS
Simon McVittie [Thu, 14 May 2015 13:43:53 +0000 (14:43 +0100)]
add a missing change, fix some bug numbers
Simon McVittie [Thu, 14 May 2015 13:41:06 +0000 (14:41 +0100)]
Merge branch 'dbus-1.8'
Conflicts:
NEWS
cmake/CMakeLists.txt
configure.ac
Simon McVittie [Thu, 14 May 2015 11:23:09 +0000 (12:23 +0100)]
Fail to generate random bytes instead of falling back to rand()
This is more robust against broken setups where we run out
of memory or cannot read /dev/urandom.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90414
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
[smcv: document @error]
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Tue, 12 May 2015 10:35:04 +0000 (11:35 +0100)]
Make UUID generation failable
Previously, this would always succeed, but might use
weak random numbers in rare failure cases. I don't think
these UUIDs are security-sensitive, but if they're generated
by a PRNG as weak as rand() (<= 32 bits of entropy), we
certainly can't claim that they're universally unique.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90414
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
[smcv: document @error]
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Thu, 14 May 2015 11:17:54 +0000 (12:17 +0100)]
_dbus_server_init_base: raise a DBusError
This can currently only fail from OOM, but I'm about to make
it possible to fail from insufficient entropy.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90414
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
[smcv: document @error]
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Thu, 14 May 2015 11:17:10 +0000 (12:17 +0100)]
_dbus_server_new_for_socket: raise a DBusError
This can currently only fail due to OOM, but I'm about to
make it possible to fail for other reasons.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90414
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
[smcv: correct failure to set error in one case; document @error]
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Tue, 12 May 2015 10:10:58 +0000 (11:10 +0100)]
Security hardening: force EXTERNAL auth in session.conf on Unix
DBUS_COOKIE_SHA1 is dependent on unguessable strings, i.e.
indirectly dependent on high-quality pseudo-random numbers
whereas EXTERNAL authentication (credentials-passing)
is mediated by the kernel and cannot be faked.
On Windows, EXTERNAL authentication is not available,
so we continue to use the hard-coded default (all
authentication mechanisms are tried).
Users of tcp: or nonce-tcp: on Unix will have to comment
this out, but they would have had to use a special
configuration anyway (to set the listening address),
and the tcp: and nonce-tcp: transports are inherently
insecure unless special steps are taken to have them
restricted to a VPN or SSH tunnelling.
Users of obscure Unix platforms (those that trigger
the warning "Socket credentials not supported on this Unix OS"
when compiling dbus-sysdeps-unix.c) might also have to
comment this out, or preferably provide a tested patch
to enable credentials-passing on that OS.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90414
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 14 May 2015 13:30:08 +0000 (14:30 +0100)]
start towards 1.8.20
Simon McVittie [Thu, 14 May 2015 12:28:38 +0000 (13:28 +0100)]
1.8.18
Ralf Habacker [Wed, 13 May 2015 18:17:05 +0000 (20:17 +0200)]
dbus_daemon_publish_session_bus_address: Fix -Wsign-compare issue.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90089
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Wed, 13 May 2015 17:52:23 +0000 (18:52 +0100)]
NEWS
Simon McVittie [Tue, 27 Aug 2013 18:16:42 +0000 (19:16 +0100)]
Avoid reading beyond the length of a variable
Appending &some as DBUS_TYPE_INT64, DBUS_TYPE_UINT64 or DBUS_TYPE_DOUBLE,
where "some" is an int, reads beyond the bounds of that variable.
Use a zero-filled DBusBasicValue instead.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=30350
Simon McVittie [Tue, 27 Aug 2013 17:21:03 +0000 (18:21 +0100)]
Fix whitespace as per Havoc's review (in 2010)
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=30350
Christian Dywan [Thu, 23 Sep 2010 17:22:53 +0000 (19:22 +0200)]
Implement dbus_message_iter_get_element_count
According unit tests are added to _dbus_message_test.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=30350
Reviewed-by: Havoc Pennington <hp@pobox.com>
Simon McVittie [Fri, 17 Apr 2015 13:01:08 +0000 (14:01 +0100)]
DBusSocket: put the #ifdef case before the !defined case
This avoids the confusing #ifndef...#else anti-pattern.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Ralf Habacker [Thu, 16 Apr 2015 21:16:31 +0000 (23:16 +0200)]
Convert mostly DBUS_SOCKET_... and DBUS_POLLABLE_.. macros for more type safety.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Thu, 12 Mar 2015 20:44:42 +0000 (20:44 +0000)]
Turn DBusSocket into a type-safe struct, preventing inappropriate conversion
Fix the remaining platform-specific code to look at the struct's
appropriate platform-specific member.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Simon McVittie [Thu, 12 Mar 2015 18:44:48 +0000 (18:44 +0000)]
Remove _dbus_socket_is_invalid, no longer used
It didn't have many users anyway, and I've replaced them with the
DBUS_SOCKET_IS_VALID macro.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 12 Mar 2015 18:35:48 +0000 (18:35 +0000)]
Convert miscellaneous socket APIs to DBusSocket
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 12 Mar 2015 18:33:46 +0000 (18:33 +0000)]
generic socket transport code: work in terms of DBusSocket
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 12 Mar 2015 20:16:46 +0000 (20:16 +0000)]
Split _dbus_set_fd_nonblocking vs. _dbus_set_socket_nonblocking
The former is Unix-specific, the latter is also portable to Windows.
On Unix, they're really the same thing.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 12 Mar 2015 18:29:40 +0000 (18:29 +0000)]
main: reload_pipe is (despite its name) a socket pair
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 12 Mar 2015 15:21:40 +0000 (15:21 +0000)]
DBusMainLoop, DBusSocketSet: work in terms of DBusPollable
This requires generic support for keying hash tables by DBusPollable:
there are already implementations for int and uintptr_t keys, but not
for "int or uintptr_t depending on platform", which is what
DBusPollable now means.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Simon McVittie [Thu, 12 Mar 2015 14:59:54 +0000 (14:59 +0000)]
Mostly remove the remnants of an older socket abstraction layer
This is only used on Windows, and wasn't even a particularly abstract
abstraction.
I've removed DBUS_SOCKET_IS_INVALID in favour of DBUS_SOCKET_IS_VALID
because I prefer to avoid double-negatives.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 12 Mar 2015 19:54:25 +0000 (19:54 +0000)]
dbus-sysdeps: add more infrastructure around DBusSocket
This is all trivial right now, but will become significant when we
change DBusSocket into a type-safe struct.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Simon McVittie [Thu, 12 Mar 2015 14:29:38 +0000 (14:29 +0000)]
bus_unix_fds_passing_test: the results of _dbus_socketpair are sockets
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
Ralf Habacker [Tue, 31 Mar 2015 19:53:22 +0000 (21:53 +0200)]
Use typedef DBusSocket for sockets fd's to avoid conversion warnings.
[smcv: remove unneeded and invalid dbus-sysdeps.h from public header;
make prototype of _dbus_socketpair() consistent; undo conversion
of getaddrinfo result from int to SOCKET; don't call
_dbus_return_val_if_fail() from internal function]
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89444
Simon McVittie [Tue, 12 May 2015 10:54:50 +0000 (11:54 +0100)]
Security hardening: force EXTERNAL auth in session.conf on Unix
DBUS_COOKIE_SHA1 is dependent on unguessable strings, i.e.
indirectly dependent on high-quality pseudo-random numbers
whereas EXTERNAL authentication (credentials-passing)
is mediated by the kernel and cannot be faked.
On Windows, EXTERNAL authentication is not available,
so we continue to use the hard-coded default (all
authentication mechanisms are tried).
Users of tcp: or nonce-tcp: on Unix will have to comment
this out, but they would have had to use a special
configuration anyway (to set the listening address),
and the tcp: and nonce-tcp: transports are inherently
insecure unless special steps are taken to have them
restricted to a VPN or SSH tunnelling.
Users of obscure Unix platforms (those that trigger
the warning "Socket credentials not supported on this Unix OS"
when compiling dbus-sysdeps-unix.c) might also have to
comment this out, or preferably provide a tested patch
to enable credentials-passing on that OS.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90414
Ralf Habacker [Wed, 6 May 2015 10:09:19 +0000 (12:09 +0200)]
reader_init: Initialize all fields of struct DBusTypeReader (CID 54754, 54772, 54773).
This patch is based on the fix for 'Field reader.array_len_offset is
uninitialized'
Reported by Coverity: CID 54754, 54772, 54773: Uninitialized scalar
variable (UNINIT)
[smcv: also re-order how the class is set when we recurse, so that
the sub-reader's class doesn't end up NULL]
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90021
Simon McVittie [Fri, 8 May 2015 14:37:58 +0000 (15:37 +0100)]
Merge branch 'dbus-1.8'
Simon McVittie [Fri, 8 May 2015 14:36:19 +0000 (15:36 +0100)]
Revert "reader_init: Initialize all fields of struct DBusTypeReader (CID 54754, 54772, 54773)."
This reverts commit
21a7873f20145d561c2de8f084d98200a8de2c83.
This appears to cause a segfault, presumably resulting from something
assuming that reader_init() would not reinitialize all fields:
#0 0x00007ffff7b74777 in _dbus_type_reader_get_current_type (reader=reader@entry=0x7fffffffda50) at .../dbus/dbus-marshal-recursive.c:791
#1 0x00007ffff7b719d0 in _dbus_header_cache_check (header=<optimized out>)
at .../dbus/dbus-marshal-header.c:209
#2 0x00007ffff7b719d0 in _dbus_header_cache_check (header=header@entry=0x624658, field=field@entry=6) at .../dbus/dbus-marshal-header.c:250
#3 0x00007ffff7b72884 in _dbus_header_get_field_basic (header=header@entry=0x624658, field=field@entry=6, type=type@entry=115, value=value@entry=0x7fffffffdbd8) at .../dbus/dbus-marshal-header.c:1365
#4 0x00007ffff7b7d8c2 in dbus_message_get_destination (message=message@entry=0x624650) at .../dbus/dbus-message.c:3457
#5 0x00007ffff7b67be6 in _dbus_connection_send_preallocated_unlocked_no_update (connection=connection@entry=0x6236d0, preallocated=0x0,
preallocated@entry=0x6234c0, message=message@entry=0x624650, client_serial=client_serial@entry=0x7fffffffdcbc)
at .../dbus/dbus-connection.c:2017
Ralf Habacker [Tue, 5 May 2015 15:10:01 +0000 (17:10 +0200)]
auth_set_unix_credentials: Fix calling _dbus_credentials_add_pid without checking return value (CID 54708).
Reported by Coverity: CID 54708: Unchecked return value (CHECKED_RETURN)
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90021
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Ralf Habacker [Tue, 5 May 2015 15:08:18 +0000 (17:08 +0200)]
auth_set_unix_credentials: Fix calling _dbus_credentials_add_unix_uid without checking return value (CID 54722).
Reported by Coverity: CID 54722: Unchecked return value (CHECKED_RETURN)
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90021
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Ralf Habacker [Wed, 6 May 2015 10:14:31 +0000 (12:14 +0200)]
Merge remote-tracking branch 'origin/dbus-1.8'
Ralf Habacker [Wed, 6 May 2015 10:09:19 +0000 (12:09 +0200)]
reader_init: Initialize all fields of struct DBusTypeReader (CID 54754, 54772, 54773).
This patch is based on the fix for 'Field reader.array_len_offset is uninitialized'
Reported by Coverity: CID 54754, 54772, 54773: Uninitialized scalar variable (UNINIT)
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90021
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Ralf Habacker [Wed, 6 May 2015 07:18:20 +0000 (09:18 +0200)]
do_check_nonce: Fix of calling _dbus_string_append_len without checking return value (CID 54720).
Reported by Coverity: CID: Unchecked return value (CHECKED_RETURN)
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90021
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Ralf Habacker [Wed, 6 May 2015 06:25:20 +0000 (08:25 +0200)]
dbus_message_demarshal: Fix calling _dbus_string_append_len without checking return value (CID 54690).
Reported by Coverity: CID 54690: Unchecked return value (CHECKED_RETURN)
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90021
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Simon McVittie [Tue, 5 May 2015 11:50:20 +0000 (12:50 +0100)]
Update NEWS for 1.9 branch
Simon McVittie [Tue, 5 May 2015 11:50:11 +0000 (12:50 +0100)]
Merge branch 'dbus-1.8'
Simon McVittie [Tue, 5 May 2015 11:43:47 +0000 (12:43 +0100)]
NEWS for 1.8 branch
Adrian Szyndela [Tue, 5 May 2015 11:30:30 +0000 (12:30 +0100)]
DBusCounter: add a mutex to protect the refcount and notify function
The overall problem here is that DBusCounter is indirectly linked
to a DBusConnection, but is not actually guaranteed to be protected by
that connection's mutex; and a DBusMessage can carry a reference to the
DBusCounter, resulting in freeing that DBusMessage having an effect on
the DBusCounter.
Making the refcount atomic would not be a sufficient fix, since it would
not protect the notify function: _dbus_counter_notify() could be called
indirectly by dbus_message_unref(), in an arbitrary thread that does not
hold the DBusConnection's lock, at the same time that the holder
of the DBusConnection lock calls _dbus_transport_set_max_message_size().
[smcv: added commit message]
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=89297
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Adrian Szyndela [Tue, 5 May 2015 11:27:15 +0000 (12:27 +0100)]
extend lock's range in live_messages_notify()
The other code paths that ref or unref a transport are protected by
the DBusConnection's lock. This function already used that lock,
but for a narrower scope than the refcount manipulation.
live_messages_notify() could be triggered by unreffing messages
that originated from the same connection in a different thread.
[smcv: added commit message]
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=90312
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Ralf Habacker [Tue, 28 Apr 2015 18:11:27 +0000 (20:11 +0200)]
tools: MSVC compile fixes
unistd.h and sleep() are normally Unix-specific, although mingw also provides them.
The Windows C runtime documents _environ as its equivalent of Unix environ.
https://bugs.freedesktop.org/show_bug.cgi?id=90089
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>