Instead of returning only TRUE/FALSE, return the number of signals
that were received while waiting. This make the API cleared, because
previously I always had to check anew whether wait-for-signal returns
TRUE or FALSE on timeout.
Also, add nmtstp_assert_wait_for_signal() and nmtstp_assert_wait_for_signal_until()
macros.
(cherry picked from commit af55476bf1)
This allows tests to use these functions on a different platform instance
then on the singleton. The change makes the argument list longer, which is
unfortunate. On the other hand, it makes those functions more useful
in general.
You can't have it all.
Also, they now follow the pattern of most functions in NM where the type
is a singleton: you always pass the singleton to the function, although
in the usual case there is only one singleton instance. This allows to
use the function also on the non-singleton instance.
(cherry picked from commit c4151ebb5b)
On netlink layer, this field is uint8_t/uchar.
A larger (signed) plen makes no sense. Adjust the signatures
to have only guint8.
(cherry picked from commit 44768f0311)
On netlink layer, this field is uint8_t/uchar.
A larger (signed) plen makes no sense. Adjust the signatures
to have only guint8.
(cherry picked from commit 14ee5dd2f8)
NMPCacheId is a union with fields for all known NMPCacheIdTypes.
Up to now, we always cloned the entire union, computed the hash
over all (possibly unset) fields and used memcmp() unanimously.
That was ok, because NMPCacheId was 16 bytes in total and cache-id
types that consumed less bytes didn't have a large overhead.
Next, we will add a new cache id type which increases the size of
NMPCacheId to 24 bytes. So, while possibly only a fraction of the
instances is that large, they would all have to pay that price.
Change that to consider and clone only those parts of the id
that are actually used.
(cherry picked from commit b1e3deaf2f)
As we get more NMPCacheIdType values, it's better to have for
each type a pre-declared list of supported types, instead of
iterating over all types and letting _nmp_object_init_cache_id()
figure out that the cache-id-type is unsupported on that object.
(cherry picked from commit fe78ae0b6a)
NM_UTILS_LOOKUP_DEFAULT_NM_ASSERT() is useful because unless
compiled with NM_MORE_ASSERTS, there is no assertion.
An assertion includes the function name, and can make the
function ineligible for inlining.
(cherry picked from commit fbfe2ef216)
Since long, dnsmasq supports scoping the IPv6 address
with '@<interface-name>'. Since 2.58, it also supports
'%' as delimiter, which is the standard way to specify
the zone-id (rfc6874).
Since 2.73, specifying the scope with '@' as "server"
address is no longer working properly, thus breaking
NetworkManager with dnsmasq >= 2.73.
To work around that, use '%' delimiter. That breaks pre-2.58
users that have a DNS server on a link local address, but that
seems acceptable as that version was released in January 2012.
https://bugzilla.gnome.org/show_bug.cgi?id=764839
(cherry picked from commit c328cf52f2)
The applied connection must not be modified during the activation. If
the PPP setting needs to be changed when activating a PPPoE
connection, make a copy to prevent the following error:
could not get secrets:
GDBus.Error:org.freedesktop.NetworkManager.Settings.Failed:
The connection was modified since activation
https://bugzilla.redhat.com/show_bug.cgi?id=1324895
(cherry picked from commit 76309ebe79)
For internal compilation we want to be able to use deprecated
API without warnings.
Define the version min/max macros to effectively disable deprecation
warnings.
However, don't do it via CFLAGS option in the makefiles, instead hack it
to "nm-default.h". After all, *every* source file that is for internal
compilation needs to include this header as first.
Otherwise we fail since they don't exist. We have to carefully
implement an "or" condition for the cases of having prebuilt manpages
in a tarball, vs actually building them from source.
This way it's consistently used across all manual page without a need
for XSL templating.
Also, the entities file could in future possibly be used to template the
build-time configurables such as filesystem paths or bug tracker URL.
It's injected from the makefile, but not even used consistently or included in
the resulting render of manual page. Which is good, otherwise we'd have a
non-reproducible build with possible multilib conflicts if rendered around
midnight.
