Basically every logging statement is wrapped by a nm_logging_enabled()
to evaluate the function call of the logging lazy.
Make the function a candidate for inlining, it safes some space. On
a default build it goes for me from 2580584 to 2560104 bytes (20k).
$ ./autogen.sh && make && strip ./src/NetworkManager
We really expect this assertion not to be violated.
As we want for nm_logging_enabled() to become smaller and inline,
remove the runtime assertion from regular builds.
Live fast and dangerous.
Instead of calling _ensure_initialized() at various places
to ensure that we setup logging before any logging commands
are executed, initialize the logging fields in the global variable.
This removes code from nm_logging_enabled(), which we want to
become a static inline function.
When one of the configuration snippet is malformed, NM doesn't tell
which file caused the error:
$ NetworkManager --print-config
Failed to read configuration: Key file does not start with a group
Fix this.
$ NetworkManager --print-config
Failed to read configuration: /usr/lib/NetworkManager/conf.d/test.conf: Key file does not start with a group
As explained in the function comment, a NULL value for @val is
acceptable. Otherwise:
$ nmcli connection edit
(process:2276): nmcli-CRITICAL **: check_valid_name: assertion 'val' failed
Fixes: fb8fe1d8cb
Instead of using (only) a hard-coded list of VPN types,
prefer lookup the VPN settings from the .name files.
Still, fallback to a hard-coded list if the plugin cannot
be found, because for connection-add we currently don't
actually need the plugin installed.
GCC can place each symbols into separate sections making it possible for
ld to discard unneeded sections (and thus symbols afterwards). This achieves
almost the same size benefits as LTO without the huge performance
penalty (and the undefined behavior with -g).
The cost of more section headers in .o files seems to be negligible.
Before:
359.73user 71.94system 1:29.67elapsed 481%CPU (0avgtext+0avgdata 137948maxresident)k
text data bss dec hex filename
2772228 74848 9272 2856348 2b959c src/NetworkManager
With LTO:
740.37user 94.89system 4:13.83elapsed 329%CPU (0avgtext+0avgdata 118316maxresident)k
text data bss dec hex filename
2435437 74224 8704 2518365 266d5d src/NetworkManager
With --gc-sections:
365.63user 72.34system 1:30.12elapsed 485%CPU (0avgtext+0avgdata 138016maxresident)k
text data bss dec hex filename
2498498 74040 8408 2580946 2761d2 src/NetworkManager
Currently we don't specify to dnsmasq which interface must be used to
contact a given nameserver and so requests can be sent through the
wrong interface.
Fix this by concatenating a @interface prefix to each server (unless
an IPv6 interface scope-id is already present).
https://bugzilla.gnome.org/show_bug.cgi?id=765153
Under no circumstance should a non-global configuration with NULL or
empty iface be added to the DNS manager. Assert this early and remove
other unnecessary checks.
A large part of "nm-test-utils.h" is only relevant for tests inside "src/"
directory, as they are helpers related to NetworkManager core part.
Split this part out of "nm-test-utils.h" header.
- make NMAccessPoint and NMAccessPointClass internal structs. This means,
they cannot be subclassed anymore, but we also don't want that.
- This way, we can safely embed the private data directly in the now
private access-point instance.
- change type of boolean fields from gboolean to bool.
- some whitespace fixes
- don't include "nm-default.h" from headers. All source files
include this header as first.
- drop G_BEGIN_DECLS/G_END_DECLS. This is not C++ nor public
API.
- drop unnecessary includes from header files. They are either
not required, or already provided via "nm-default.h".
- include in source files the corresponding header file as first
after "nm-default.h". This should ensure that header files are
self-contained (appart from "nm-default.h").
Having a simple accessor print warnings is not nice. At that point there
is no context as to why we are trying to read the value.
Note that the function already handles and expects invalid values, it's
just not clear that printing warnings from a utility function is the right
thing to do.
Just ignore such cases silently (at this point). It's up to the caller
to print a warning or whatever.
(gdb) bt
#0 0x00007ffff4d9e075 in g_type_check_instance_is_fundamentally_a (type_instance=type_instance@entry=0x-1, fundamental_type=fundamental_type@entry=80) at gtype.c:4030
#1 0x00007ffff4d7e447 in g_object_unref (_object=0xffffffffffffffff) at gobject.c:3076
#2 0x00007fffe89cdfa8 in disconnect_context_complete (ctx=ctx@entry=0x555555b4c680) at nm-modem-broadband.c:1062
#3 0x00007fffe89cf6e5 in disconnect (modem=<optimized out>, warn=0, cancellable=0x0, callback=0x0, user_data=0x0) at nm-modem-broadband.c:1126
#4 0x00007fffe89d24cd in nm_modem_deactivate (self=0x555555aba1b0 [NMModemBroadband], device=device@entry=0x555555ab8c20 [NMDeviceModem]) at nm-modem.c:1164
#5 0x00007fffdbd73022 in deactivate (device=0x555555ab8c20 [NMDeviceModem]) at nm-device-modem.c:455
#6 0x00005555555e087b in nm_device_cleanup (self=self@entry=0x555555ab8c20 [NMDeviceModem], reason=reason@entry=NM_DEVICE_STATE_REASON_NOW_MANAGED, cleanup_type=cleanup_type@entry=CLEANUP_TYPE_DECONFIGURE)
at devices/nm-device.c:10392
#7 0x00005555555e0ffd in _set_state_full (self=self@entry=0x555555ab8c20 [NMDeviceModem], state=state@entry=NM_DEVICE_STATE_UNAVAILABLE, reason=reason@entry=NM_DEVICE_STATE_REASON_NOW_MANAGED, quitting=quitting@entry=0) at devices/nm-device.c:10804
#8 0x00005555555e1a16 in nm_device_state_changed (self=self@entry=0x555555ab8c20 [NMDeviceModem], state=state@entry=NM_DEVICE_STATE_UNAVAILABLE, reason=reason@entry=NM_DEVICE_STATE_REASON_NOW_MANAGED)
at devices/nm-device.c:11029
Fixes: 21b50c59ce
