We call GetConnectionUnixProcessID and GetConnectionUnixUser *a lot*.
And we do so synchronously. Both is a problem.
To avoid the first problem, cache the last few requests with each cached
value being valid for one second.
On a quick test, this saves 98% of the requests:
59 GetConnectionUnixProcessID(*)
3201 GetConnectionUnixProcessID(*) (served from cache)
59 GetConnectionUnixUser(*)
3201 GetConnectionUnixUser(*) (served from cache)
Note that now as we serve requests from the cache, it might be the case
that the D-Bus endpoint already disconnected. Previously, the request would
have failed but now we return the cached user-id and process-id. This
problem is mitigated by only caching the values for up to one second.
Also, it's not really a problem because we cache sender names. Those
are supposed to be unique and not repeat. So, even if the peer already
disconnected, it is still true that the corresponding PID/UID was as
we have cached it. We don't use this API for checking whether the peer
is still connected, but what UID/PID it has/had. That answer is still
correct for the cached value after the peer disconnected.
The NMDBusManager owns a reference to the system bus. Expose it, so we
can use it. Note that g_bus_get() -- as alternative to get the systembus
singleton -- is asynchronous, and g_bus_get_sync() has an API that makes
one wonder what it does. Since we already have a reference to the connection
object we want to use, expose it.
NMConnection is an interface, which is implemented by the types
NMSimpleConnection (libnm-core), NMSettingsConnection (src) and
NMRemoteConnection (libnm).
NMSettingsConnection does a lot of things already:
1) it "is-a" NMDBusObject and exports the API of a connection profile
on D-Bus
2) it interacts with NMSettings and contains functionality
for tracking the profiles.
3) it is the base-class of types like NMSKeyfileConnection and
NMIfcfgConnection. These handle how the profile is persisted
on disk.
4) it implements NMConnection interface, to itself track the
settings of the profile.
3) and 4) would be better implemented via delegation than inheritance.
Address 4) and don't let NMSettingsConnection implemente the NMConnection
interface. Instead, a settings-connection references now a NMSimpleConnection
instance, to which it delegates for keeping the actual profiles.
Advantages:
- by delegating, there is a clearer separation of what
NMSettingsConnection does. For example, in C we often required
casts from NMSettingsConnection to NMConnection. NMConnection
is a very trivial object with very little logic. When we have
a NMConnection instance at hand, it's good to know that it is
*only* that simple instead of also being an entire
NMSettingsConnection instance.
The main purpose of this patch is to simplify the code by separating
the NMConnection from the NMSettingsConnection. We should generally
be aware whether we handle a NMSettingsConnection or a trivial
NMConnection instance. Now, because NMSettingsConnection no longer
"is-a" NMConnection, this distinction is apparent.
- NMConnection is implemented as an interface and we create
NMSimpleConnection instances whenever we need a real instance.
In GLib, interfaces have a performance overhead, that we needlessly
pay all the time. With this change, we no longer require
NMConnection to be an interface. Thus, in the future we could compile
a version of libnm-core for the daemon, where NMConnection is not an
interface but a GObject implementation akin to NMSimpleConnection.
- In the previous implementation, we cannot treat NMConnection immutable
and copy-on-write.
For example, when NMDevice needs a snapshot of the activated
profile as applied-connection, all it can do is clone the entire
NMSettingsConnection as a NMSimpleConnection.
Likewise, when we get a NMConnection instance and want to keep
a reference to it, we cannot do that, because we never know
who also references and modifies the instance.
By separating NMSettingsConnection we could in the future have
NMConnection immutable and copy-on-write, to avoid all unnecessary
clones.
During shutdown, we will need to still iterate the main loop
to do a coordinated shutdown. Currently we do not, and we just
exit, leaving a lot of objects hanging.
If we are going to fix that, we need during shutdown tell
NMDBusManager to reject all future operations.
Note that property getters and "GetManagerObjects" call is not
blocked. It continues to work.
Certainly for some operations, we want to allow them to be called even
during shutdown. However, these have to opt-in.
This also fixes an uglyness, where nm_dbus_manager_start() would
get the set-property-handler and the @manager as user-data. However,
NMDBusManager will always outlife NMManager, hence, after NMManager
is destroyed, the user-data would be a dangling pointer. Currently
that is not an issue, because
- we always leak NMManager
- we don't run the mainloop during shutdown
We currently start the bus manager only after the creation of a
NMManager because the NMManager is needed to handle set-property bus
calls. However, objects created by NMManager
(e.g. NMDnsSystemdResolved) need a bus connection and so their
initialization currently fail.
To fix this, split nm_dbus_manager_start() in two parts: first only
create the connection and acquire the bus. After this step the
NMManager can be set up. In the second step, set NMManager as the
set-property handler and start exporting objects on the bus.
Fixes: 297d4985ab
Previously, we used the generated GDBusInterfaceSkeleton types and glued
them via the NMExportedObject base class to our NM types. We also used
GDBusObjectManagerServer.
Don't do that anymore. The resulting code was more complicated despite (or
because?) using generated classes. It was hard to understand, complex, had
ordering-issues, and had a runtime and memory overhead.
This patch refactors this entirely and uses the lower layer API GDBusConnection
directly. It replaces the generated code, GDBusInterfaceSkeleton, and
GDBusObjectManagerServer. All this is now done by NMDbusObject and NMDBusManager
and static descriptor instances of type GDBusInterfaceInfo.
This adds a net plus of more then 1300 lines of hand written code. I claim
that this implementation is easier to understand. Note that previously we
also required extensive and complex glue code to bind our objects to the
generated skeleton objects. Instead, now glue our objects directly to
GDBusConnection. The result is more immediate and gets rid of layers of
code in between.
Now that the D-Bus glue us more under our control, we can address issus and
bottlenecks better, instead of adding code to bend the generated skeletons
to our needs.
Note that the current implementation now only supports one D-Bus connection.
That was effectively the case already, although there were places (and still are)
where the code pretends it could also support connections from a private socket.
We dropped private socket support mainly because it was unused, untested and
buggy, but also because GDBusObjectManagerServer could not export the same
objects on multiple connections. Now, it would be rather straight forward to
fix that and re-introduce ObjectManager on each private connection. But this
commit doesn't do that yet, and the new code intentionally supports only one
D-Bus connection.
Also, the D-Bus startup was simplified. There is no retry, either nm_dbus_manager_start()
succeeds, or it detects the initrd case. In the initrd case, bus manager never tries to
connect to D-Bus. Since the initrd scenario is not yet used/tested, this is good enough
for the moment. It could be easily extended later, for example with polling whether the
system bus appears (like was done previously). Also, restart of D-Bus daemon isn't
supported either -- just like before.
Note how NMDBusManager now implements the ObjectManager D-Bus interface
directly.
Also, this fixes race issues in the server, by no longer delaying
PropertiesChanged signals. NMExportedObject would collect changed
properties and send the signal out in idle_emit_properties_changed()
on idle. This messes up the ordering of change events w.r.t. other
signals and events on the bus. Note that not only NMExportedObject
messed up the ordering. Also the generated code would hook into
notify() and process change events in and idle handle, exhibiting the
same ordering issue too.
No longer do that. PropertiesChanged signals will be sent right away
by hooking into dispatch_properties_changed(). This means, changing
a property in quick succession will no longer be combined and is
guaranteed to emit signals for each individual state. Quite possibly
we emit now more PropertiesChanged signals then before.
However, we are now able to group a set of changes by using standard
g_object_freeze_notify()/g_object_thaw_notify(). We probably should
make more use of that.
Also, now that our signals are all handled in the right order, we
might find places where we still emit them in the wrong order. But that
is then due to the order in which our GObjects emit signals, not due
to an ill behavior of the D-Bus glue. Possibly we need to identify
such ordering issues and fix them.
Numbers (for contrib/rpm --without debug on x86_64):
- the patch changes the code size of NetworkManager by
- 2809360 bytes
+ 2537528 bytes (-9.7%)
- Runtime measurements are harder because there is a large variance
during testing. In other words, the numbers are not reproducible.
Currently, the implementation performs no caching of GVariants at all,
but it would be rather simple to add it, if that turns out to be
useful.
Anyway, without strong claim, it seems that the new form tends to
perform slightly better. That would be no surprise.
$ time (for i in {1..1000}; do nmcli >/dev/null || break; echo -n .; done)
- real 1m39.355s
+ real 1m37.432s
$ time (for i in {1..2000}; do busctl call org.freedesktop.NetworkManager /org/freedesktop org.freedesktop.DBus.ObjectManager GetManagedObjects > /dev/null || break; echo -n .; done)
- real 0m26.843s
+ real 0m25.281s
- Regarding RSS size, just looking at the processes in similar
conditions, doesn't give a large difference. On my system they
consume about 19MB RSS. It seems that the new version has a
slightly smaller RSS size.
- 19356 RSS
+ 18660 RSS
The next commit will completely rework NMBusManager and replace
NMExportedObject by a new type NMDBusObject.
Originally, NMDBusObject was added along NMExportedObject to ease
the rework and have compilable, intermediate stages of refactoring. Now,
I think the new name is better, because NMDBusObject is very strongly related
to the bus manager and the old name NMExportedObject didn't make that
clear.
I also slighly prefer the name NMDBusObject over NMBusObject, hence
for consistancy, also rename NMBusManager to NMDBusManager.
This commit only renames the file for a nicer diff in the next commit.
It does not actually update the type name in sources. That will be done
later.
Our gdbus generated types use the same names as their corresponding
"real" types, but with "NM" changed to "NMDBus".
Unfortunately, that means that introspection/nmdbus-manager.c (the
generated type for src/nm-manager.c) uses the same type name as the
entirely unrelated src/nm-dbus-manager.c.
Fix this by removing the "d" from src/nm-dbus-manager.c. (We could
rename the generated type instead, but then it becomes inconsistent
with all the other generated types, and we're already using it as
"NMDBusManager" in libnm/nm-manager.c.)
Add NMExportedObject, make it the base class of all D-Bus-exported
types, and move the nm-properties-changed-signal logic into it. (Also,
make NMSettings use the same properties-changed code as everything
else, which it was not previously doing, presumably for historical
reasons).
(This is mostly just shuffling code around at this point, but
NMExportedObject will be more important in the gdbus port, since
gdbus-codegen doesn't do a very good job of supporting objects that
export multiple interfaces [as each NMDevice subclass does, for
example], so we will need more glue/helper code in NMExportedObject
then.)
Rather than randomly including one or more of <glib.h>,
<glib-object.h>, and <gio/gio.h> everywhere (and forgetting to include
"nm-glib-compat.h" most of the time), rename nm-glib-compat.h to
nm-glib.h, include <gio/gio.h> from there, and then change all .c
files in NM to include "nm-glib.h" rather than including the glib
headers directly.
(Public headers files still have to include the real glib headers,
since nm-glib.h isn't installed...)
Also, remove glib includes from header files that are already
including a base object header file (which must itself already include
the glib headers).
Clean up some of the cross-includes between headers (which made it so
that, eg, if you included NetworkManagerUtils.h in a test program, you
would need to build the test with -I$(top_srcdir)/src/platform, and if
you included nm-device.h you'd need $(POLKIT_CFLAGS)) by moving all
GObject struct definitions for src/ and src/settings/ into nm-types.h
(which already existed to solve the NMDevice/NMActRequest circular
references).
Update various .c files to explicitly include the headers they used to
get implicitly, and remove some now-unnecessary -I options from
Makefiles.
Change the way that nm-properties-changed-signal works, and parse the
dbus-binding-tool-generated info to get the exact list of properties
that it's expected to export.
This makes NM_PROPERTY_PARAM_NO_EXPORT unnecessary, and also fixes the
problem of properties like NMDevice:hw-address being exported on
classes where it shouldn't be.
Since dbus senders are faked for private connections, we can't just
call dbus_bus_get_unix_user() on fake senders. They need to be
checked against the NMDBusManager's list of private connections
first.
For cases where NM may run without a bus daemon in root-only
environments, like an initramfs. For disconnection, since private
connection just get a disconnect message instead of NameOwnerChanged
signals broadcast by a bus daemon, just synthesize the NameOwnerChanged
signals using our fake owner name. It's just easier to do this rather
than modify any code that cares about disconnects.
Note that the new private socket is only enabled if built with
dbus-glib >= 0.100 as there are bugs in previous versions in the
implementation of dbus_g_proxy_new_for_peer() which clients must
use to talk to the private socket.
When providing a service on the bus daemon and a private connection,
we'll need to track objects so we can register them with the
private connection too. Thus all registration/unregistration
calls have to go through the NMDBusManager, not straight to
dbus-glib.
* src/nm-dbus-manager.c
src/nm-dbus-manager.h
- (nm_dbus_manager_get_name_owner): return error
* src/nm-manager.c
- (impl_manager_activate_connection): perform additional validation on
ActivateConnection calls of user connections
- (is_user_request_authorized): ensure that the requestor is the same
UID as the UID that owns the user settings service; users shouldn't
be able to control another user's connections
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@4325 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/nm-dbus-manager.h
src/nm-hal-manager.c
- Include the correct headers now that NetworkManagerDbusUtils.h doesn't
do it for them
* src/Makefile.am
src/NetworkManagerDbusUtils.c
src/NetworkManagerDbusUtils.h
- Remove these two source files; they are unused
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@3111 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Move ppp-manager over to dbus-glib. The big deal is that it was
the last piece of
code that used NM's own version of dbus signal handling and
custom dictionary
marshalling/unmarshalling. With this change, all that obsolete
code can disappear
and we get to maintain over 2000 lines less code.
* libnm-util/dbus-dict-helpers.c:
* libnm-util/dbus-dict-helpers.h: Remove.
* src/ppp-manager/nm-pppd-plugin.c: Convert it to use dbus-glib.
* src/ppp-manager/nm-pppd-plugin.xml: Implement.
* src/ppp-manager/nm-ppp-manager.c: Use dbus-glib instead of
* home-brewed dbus signal
handlers.
* src/nm-dbus-manager.c: Remove all the manual dbus signal
* handling.
* configure.in: Remove test/libnm-util/Makefile creation.
* test/Makefile.am: Remove libnm-util from SUBDIRS.
* test/libnm-util/: Remove the whole directory.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2965 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/vpn-manager/nm-vpn-connection.[ch]:
* src/vpn-manager/nm-vpn-manager.[ch]:
* src/vpn-manager/nm-vpn-service.[ch]: Rewrite the vpn handling
* code. Using
dbus-glib, GObjects, signals etc.
* libnm-glib/nm-vpn-manager.[ch]:
* libnm-glib/nm-vpn-connection.[ch]: Now that the NM
* implementation changed
so much, rewrite these too.
* libnm-glib/Makefile.am: Add new files to build, build new
* binding files for
the new introspection files.
* libnm-glib/nm-client.[ch]: Remove all VPN related stuff from
* here.
* libnm-glib/nm-dbus-utils.[ch]: Renamed from nm-utils.[ch] that
* was shadowing
the header with the same name from libnm-utils.
* libnm-glib/nm-vpn-plugin.[ch]: Implement.
* libnm-util/Makefile.am: Add nm-utils.[ch] to build.
* introspection/nm-vpn-plugin.xml: Implement.
* introspection/nm-vpn-connection.xml: Implement.
* introspection/nm-vpn-manager.xml: Implement.
* src/NetworkManagerSystem.c
* (nm_system_vpn_device_set_from_ip4_config): Remove
the named manager argument, it can just as easily get it as the
caller.
(nm_system_vpn_device_unset_from_ip4_config): Ditto.
* src/vpn-manager/nm-dbus-vpn.[ch]: Remove.
* src/nm-dbus-manager.h: Fix up the name_owner signal signature.
* src/dhcp-manager/nm-dhcp-manager.c (garray_to_string): Remove,
* use one from
libnm-utils.
* libnm-util/nm-connection.c: Ditto.
* src/NetworkManagerMain.h: Remove, it's finally empty.
* configure.in: Remove utils/ from build.
* include/NetworkManagerVPN.h: Add some more defines to reduce
* the amount
of hard-coded strings.
* utils/: Move it over to libnm-util.
* test/Makefile.am: Link against libnm-util now that util/ is
* gone.
* dispatcher-daemon/Makefile.am: Ditto.
* src/Makefile.am: Ditto.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2798 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Totally break NetworkManager. Please use 0.6 branch until futher notice.
* src/:
- Remove old low-level dbus interface implementations and replace them
with dbus-glib one.
* configure.in:
- Require dbus-glib >= 0.72.
- Plug in new sources to build.
* libnm-glib/:
- Implement GObject wrappers on top of DBUS glib auto-generated bindings
to make it more convenient to use from GObject based programs.
* introspection/:
- Implement DBUS XML introspection files, used by both NM and libnm-glib.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2309 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/nm-dbus-manager.c:
* src/nm-dbus-manager.h:
- Convert all internal DBUS code to use dbus-glib bindings.
- Remove GObject properties, we don't need them here.
- Don't explicitly set things to NULL after freeing, glib is
happy to do it if asked nicely (G_DEBUG=gc-friendly).
- Make public API argument checks type safe.
- Remove unnecessary (and wrong) cast to GObject for the first
argument to g_signal_* calls - The first argument is a gpointer.
- Export DBusGConnection to other cool classes that (are going to)
use dbus-glib.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2260 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Threading removal related cleanups:
- Use the glib default main context. Remove the device main context
member from NMDevice, and the main_context member from NMData. Change
all the idle and timeout scheduler functions to use plain
g_idle_add() and g_timeout_add().
- As a side-effect of the first change, nm_dbus_manager_get() no longer
takes an argument; fix that up too.
- Remove all locking, which is useless since we no longer use threads. For
example, nm_get_device_by_iface_locked() has been removed. The global
device list lock, the AP List lock, and all static locks in
NetworkManagerPolicy.c have been removed. The locking utility functions
in NetworkManagerUtils.c have also been removed.
- Other cleanups in spacing and code style
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2205 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Rework DBus manager signal handling to be more flexible. Previously,
only one signal handler could be registered for a particular interface.
The DBus manager now reference counts DBus bus matches and allows multiple
clients to register signal handlers for the same interface and sender.
* src/NetworkManager.c
- (main): track NMI signal handler ID and remove it when we quit
* src/NetworkManagerMain.h
- Keep track of NMI signal handler ID
* src/nm-dbus-manager.c
src/nm-dbus-manager.h
- rework signal handling; each signal handler references one signal
match, but a signal match may be referenced by one or more
signal handlers. Matches are refcounted and are destroyed when the
last signal handler that references the match is removed. This is
necessary because two signal handlers may end up requiring the same
dbus bus match, so the match must live until the last signal handler
is destroyed (for example, with the wpa_supplicant network interface
dbus interface).
* src/dhcp-manager/nm-dhcp-manager.c
- (nm_dhcp_manager_new): track DHCP signal handler id
- (nm_dhcp_manager_dispose): remove DHCP signal handler
* src/vpn-manager/nm-vpn-service.c
- (nm_vpn_service_add_watch): track VPN service signal handler id
- (nm_vpn_service_remove_watch): remove VPN service signal handler
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2124 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* Huge DBus refactor:
- Create a "DBus Manager" object which manages the connection and
sends signals on NameOwnerChanged and connection/disconnection events,
handles reconnection to the bus if NM gets kicked off, and abstracts
signal handling
- Remove DBusConnection members from places where they are no
longer needed due to the refactor, like the dbus-connection
property of the named manager, and from NMData
- Reformats a bunch of the code to gnome style
(8-space tabs, braces on same line as statement, 80-col width).
Consider it open season to reformat any bits to gnome style.
style that aren't already.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2061 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
