Simon McVittie [Fri, 1 Apr 2022 13:47:27 +0000 (14:47 +0100)]
CI: Don't wait for Windows Docker image before doing Unix builds
The default in Gitlab-CI is that each job waits for all jobs in
previous stages to complete, but we can override this default by
explicitly saying that there are no prerequisites.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Fri, 6 Nov 2020 14:42:19 +0000 (14:42 +0000)]
cmake: Look for dbus-arch-deps.h next to DBus1Config.cmake
<dbus/dbus-arch-deps.h> is architecture-dependent, and compilers have
not traditionally supported an installation path for architecture-specific
headers (Debian-based systems have /usr/include/${multiarch_tuple}, but
that isn't portable beyond Debian). When dbus was built using Autotools,
dependent projects that use CMake need to look for this header in the
right place.
Unfortunately, it seems that at least recent versions of CMake will
ignore the HINTS we get from pkg-config if they are told to search in
a non-standard prefix via ${DBus1_ROOT}.
Look for dbus-arch-deps.h in a directory derived from the filename of
the CMake config file, before trying the normal search algorithm. The
CMake config file is in ${libdir}, and so is the architecture-specific
header, so this should work reasonably reliably.
According to the CMake documentation, if we search for the same thing
multiple times, the first successful result will be used; and searching
with NO_DEFAULT_PATH is the official way to prepend things to the
search order.
Resolves: https://gitlab.freedesktop.org/dbus/dbus/-/issues/314 Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Wed, 16 Mar 2022 13:30:33 +0000 (13:30 +0000)]
dbus-launch: If no X11, look for existing bus before failing
What we are aiming for here is that `dbus-launch --autolaunch` will
exit with an error if it would have been unable to advertise the bus
in a way that will be picked up by other instances of itself, so that
we don't get a proliferation of autolaunched dbus-launch instances,
each with its own dbus-daemon, and none of them talking to the others.
However, we don't need to do that *immediately*: we can try to look for
an existing instance first. If we find one, then we can assume that
it's suitable for use, even if X11 autolaunching was disabled at
compile-time or if connecting to the X server failed at runtime. We
only need to bail out if we get to the point where we are ready to start
forking child processes.
This lets systemd users run `dbus-launch --autolaunch` to find a
systemd-managed dbus-daemon, even if X11 autolaunching is disabled.
When combined with the previous commit, it also lets macOS users run
`dbus-launch --autolaunch` to find a launchd-managed dbus-daemon.
Resolves: dbus/dbus#385 Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Wed, 16 Mar 2022 13:24:53 +0000 (13:24 +0000)]
dbus-launch: Generalize support for discovering an existing bus
Previously we were looking for an existing bus via
_dbus_lookup_user_bus(), meaning a user bus $XDG_RUNTIME_DIR/bus
(in practice this is managed by systemd --user, although in principle
there's nothing to stop other session frameworks from providing the
same thing).
_dbus_lookup_session_address() looks for an externally-managed
dbus-daemon in a more general way: on macOS it uses launchd, and on other
Unix it's a thin wrapper around _dbus_lookup_user_bus(). Let's try that,
so that macOS users can get their existing dbus-daemon from launchd.
This partially resolves dbus/dbus#385, although initially only for macOS
users who have (unusually) enabled X11 autolaunching support.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Alex Richardson [Sat, 19 Mar 2022 14:58:59 +0000 (14:58 +0000)]
cmake: Only add warning flags if the compiler supports them
I am compiling for FreeBSD where the compiler is Clang and doesn't accept
all the GCC warning flags. This breaks the -Werror build:
```
error: unknown warning option '-Wduplicated-branches' [-Werror,-Wunknown-warning-option]
error: unknown warning option '-Wduplicated-cond' [-Werror,-Wunknown-warning-option]
error: unknown warning option '-Wjump-misses-init' [-Werror,-Wunknown-warning-option]
error: unknown warning option '-Wlogical-op'; did you mean '-Wlong-long'? [-Werror,-Wunknown-warning-option]
error: unknown warning option '-Wrestrict' [-Werror,-Wunknown-warning-option]
error: unknown warning option '-Wunused-but-set-variable'; did you mean '-Wunused-const-variable'? [-Werror,-Wunknown-warning-option]
```
With this change we use check_{c,cxx}_compiler_flag to check if the flag
is supported before adding it. In the future this will allow adding
clang-specific warning flags to the list of warnings as well since they
will be ignored for GCC.
Mike Gilbert [Sun, 27 Mar 2022 19:09:48 +0000 (15:09 -0400)]
sysdeps-unix: check fd before calling _dbus_fd_set_close_on_exec()
If /proc/self/oom_score_adj does not exist, fd will invalid (-1).
Attempting to set the CLOEXEC flag will obviously fail, and we lose the
original errno value from open().
Bug: https://bugs.gentoo.org/834725 Signed-off-by: Mike Gilbert <floppym@gentoo.org>
Alex Richardson [Sat, 19 Mar 2022 14:56:12 +0000 (14:56 +0000)]
cmake: Fix DBUS_SESSION_SOCKET_DIR when cross-compiling
Inferring it from the environment is not correct, since the host system
could have a different temporary directory defined. Instead of guessing
based on the host, require the user to pass an explicit directory when
cross-compiling. This is helpful for me since I am cross-compiling for
FreeBSD from macOS and on my host TMPDIR is set to
/var/folders/<random characters>/T/ instead of the expected /tmp.
Ralf Habacker [Tue, 29 Mar 2022 07:36:54 +0000 (09:36 +0200)]
tools/ci-install.sh: on debian|ubuntu reduce the need for mingw package name changes
Instead of having to specify an exact version that needs to be adjusted
with each repository update, it is now possible to specify package names
without version or partially qualified versions, which reduces the
frequency of necessary adjustments.
This is achieved by searching for the package names in a previously
downloaded list of available packages.
Alex Richardson [Sat, 19 Mar 2022 17:56:54 +0000 (17:56 +0000)]
cmake: Only pass -Wl,--export-dynamic when linking
Otherwise we get the following warnings when building .o files with Clang:
clang-13: warning: -Wl,--export-dynamic: 'linker' input unused [-Wunused-command-line-argument]
This is required to allow the -Werror build to pass on FreeBSD.
Alex Richardson [Wed, 23 Mar 2022 22:06:33 +0000 (22:06 +0000)]
cmake: Drop an unnecessary include directory
There should be no need to include the directory above the DBus sources,
if that is actually required users can always pass -I flags to CMake.
I noticed this because CLion started indexing all my cloned projects when
I opened DBus due to this include path.
Alex Richardson [Sun, 20 Mar 2022 23:25:29 +0000 (23:25 +0000)]
cmake: Detect backtrace() support on platforms such as FreeBSD
On FreeBSD use of backtrace requires linking libexecinfo. The current
check_symbol_exists() will fail due to that missing library. Fortunately,
CMake ships with a FindBacktrace module (at least since 3.0) that can
be used to correctly handle platforms such as FreeBSD (and OpenBSD
according to the FindBacktrace source).
Alex Richardson [Sat, 19 Mar 2022 14:48:04 +0000 (14:48 +0000)]
Revert "Fix gcc compile error: redundant redeclaration of ‘environ’ [-Werror=redundant-decls]"
Breaks the build on FreeBSD which doens't have an environ declaration.
The CMake check_symbol_exists call sets `HAVE_DECL_ENVIRON` to an empty
variable (which means if(DEFINED) suceeds). This normalization should not
be necessary as it will be handled correctly by `#cmakedefine01`. If not,
all the other HAVE_* defines would also be wrong.
Simon McVittie [Wed, 16 Mar 2022 14:54:55 +0000 (14:54 +0000)]
sysdeps: Implement _DBUS_DOUBLES_BITWISE_EQUAL with memcmp
memcmp is the Standard C way to compare arbitrary regions of memory
bit-by-bit, so let's use it, instead of reinventing it in a way that
violates Standard C aliasing rules.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Ralf Habacker [Thu, 3 Mar 2022 11:40:17 +0000 (12:40 +0100)]
cmake: Fix race condition on creating docbook based files
With the previous implementation, race conditions could arise because a
generated intermediate file was used by multiple targets.
To fix the mentioned problem, the macro 'generate_docbook_file' has been
integrated into a in a new macro 'add_docbook' to simplify the dependency
chain and make it easier to use.
When using an xml template with the 'TEMPLATE' parameter, a separate
intermediate xml file is used for each generated output file to avoid
overwriting each other, which was the main cause of the described problem.
Due to the adaptation of the calling conventions it was necessary to
introduce the parameter 'MAN_CATEGORY'.
Ralf Habacker [Thu, 3 Mar 2022 08:48:51 +0000 (09:48 +0100)]
cmake: Set DBUS_*_CONFIG_FILE to be in datadir on Unix
The files were installed to the datadir (usually /usr/share) since
commit 70950325, but the macros used to implement dbus-daemon --system
and dbus-daemon --session still looked for them in the sysconfdir
(usually /etc).
Fixes: 70950325 "Adjust cmake build to match autoconf installation locations"
Simon McVittie [Fri, 25 Feb 2022 16:08:32 +0000 (16:08 +0000)]
CI: Use current Debian stable release for mingw-w64 builds
Now that we have resolved the failure to build with newer mingw-w64,
we don't need to hold these back to Debian 10 'buster' and can upgrade
to the current stable release, Debian 11 'bullseye'.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Fri, 25 Feb 2022 16:06:59 +0000 (16:06 +0000)]
build: Opt out from using mingw-w64's replacement printf(), etc.
The Windows code in dbus is careful to use Windows-specific equivalents
of the Standard C features that are not implemented by msvcrt.dll, so
we don't need to substitute a Standard C printf implementation.
This avoids compiler warnings/errors when gcc expects us to be using
Microsoft printf syntax (`ms_printf` attribute), but newer versions of
mingw-w64 expect us to be using GNU or Standard C printf syntax
(`gnu_printf` attribute) as a result of `__USE_MINGW_ANSI_STDIO` being
enabled by default if not otherwise specified.
Resolves: https://gitlab.freedesktop.org/dbus/dbus/-/issues/380 Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Fri, 25 Feb 2022 13:48:05 +0000 (13:48 +0000)]
CI: Build on Debian 11 'bullseye' instead of Debian 10 'buster'
This is the current stable release for Debian, making Debian 10 a much
less interesting target for backports. Add a manually-triggered job
so we can still try buster occasionally.
Continue to use buster for mingw-w64 builds until format string issues
with bullseye toolchains can be sorted out.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Fri, 25 Feb 2022 13:45:10 +0000 (13:45 +0000)]
CI: Remove EOL versions of Debian and Ubuntu
Debian 9 'stretch' is EOL, and Ubuntu 16.04 is EOL unless you enter into
a special subscription with Canonical. This puts them outside our
informal security-support policy, and realistically, anyone sufficiently
change-averse to be following these distributions is not going to be
backporting a current version of dbus.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Fri, 25 Feb 2022 13:53:56 +0000 (13:53 +0000)]
Disable MSVC CI by default for now
The necessary CI runner doesn't seem to be available, and we shouldn't
let that prevent us from merging dbus changes. This can be reverted
when the runner comes back.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Mon, 21 Feb 2022 16:07:26 +0000 (16:07 +0000)]
sysdeps-unix: Diagnose failure to open /proc/self/oom_score_adj
Previously, we silently ignored this, but now that we're more careful
about the contexts in which we try to reset the OOM score and whether
we log failures as a warning, we can let the dbus-daemon-launch-helper
show a message if it can't write there.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Mon, 21 Feb 2022 16:02:13 +0000 (16:02 +0000)]
spawn-unix: Don't log an error if unable to reset Linux OOM score
We cannot safely log between fork() and exec() because it isn't an
async-signal-safe operation (in particular it might allocate memory).
We also don't want to treat a failure here as a real problem, because
it might legitimately not work: in a system dbus-daemon that has dropped
privileges from root, the pseudo-file representing this process parameter
remains owned by root and cannot be altered by the unprivileged user.
For the main use-case for this operation, the system dbus-daemon, we
have another opportunity to do this in the dbus-daemon-launch-helper
(see the previous commit).
Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Mon, 21 Feb 2022 16:00:42 +0000 (16:00 +0000)]
dbus-daemon-launch-helper: Reset Linux OOM score adjustment here
Previously, we were relying on the system bus being able to reset
its OOM score adjustment after it forks, but before it execs the
dbus-daemon-launch-helper. However, it can't actually do that (leading
to dbus#378), because the system bus typically starts as root, uses its
root privileges to adjust resource limits, and then drops privileges
to the `@DBUS_USER@`, typically `dbus` or `messagebus`. This leaves the
pseudo-files in /proc for its process parameters owned by root, and the
`@DBUS_USER@` is not allowed to open them for writing.
The dbus-daemon-launch-helper is setuid root, so it can certainly
alter its OOM score adjustment before exec'ing the actual activated
service. We need to do this before dropping privileges, because after
dropping privileges we would be unable to write to this process
parameter.
This is a non-async-signal-safe context, so we can safely log errors
here, unlike the fork-and-exec code paths.
Resolves: https://gitlab.freedesktop.org/dbus/dbus/-/issues/378 Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Mon, 21 Feb 2022 15:53:38 +0000 (15:53 +0000)]
spawn-unix: On Linux, don't try to increase OOM-killer protection
The oom_score_adj parameter is a signed integer, with increasingly
positive values being more likely to be killed by the OOM-killer,
and increasingly negative values being less likely.
Previously, we assumed that oom_score_adj would be negative or zero,
and reset it to zero, which does not require privileges because it
meant we're voluntarily giving up our OOM-killer protection.
In particular, bus/dbus.service.in has OOMScoreAdjust=-900, which
we don't want system services to inherit.
However, systemd >= 250 has started putting a positive oom_score_adj
on user processes, to make it more likely that the OOM killer will kill
a user process rather than a system process. Changing from a positive
oom_score_adj to zero is increasing protection from the OOM-killer,
which only a privileged process is allowed to do, resulting in warnings
whenever we carry out traditional (non-systemd) service activation
on the session bus.
To avoid this, do the equivalent of:
if (oom_score_adj < 0)
oom_score_adj = 0;
which is always allowed.
Resolves: https://gitlab.freedesktop.org/dbus/dbus/-/issues/374 Signed-off-by: Simon McVittie <smcv@collabora.com>
Simon McVittie [Mon, 21 Feb 2022 10:35:10 +0000 (10:35 +0000)]
tests: Ensure session bus has started before integration test
The session dbus-daemon won't necessarily be run immediately on login
if we are using systemd socket activation for it, and the transient
services directory $XDG_RUNTIME_DIR/dbus-1/services isn't created until
it's actually run. Ping the dbus-daemon to make sure it's available.
Signed-off-by: Simon McVittie <smcv@debian.org>
Bug-Debian: https://bugs.debian.org/1005889
Jan Tojnar [Thu, 11 Feb 2021 05:00:45 +0000 (06:00 +0100)]
doc: Introduce XML catalog
This adds a nice way of loading the DTD files based on the identifier
in the DOCTYPE declaration, no matter where the DTDs are installed.
See also ‘XML catalog’ Wikipedia entry and update-xmlcatalog(8).
projects / thirdparty / dbus.git / log
