NetworkManager merge requestshttps://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests2021-06-04T08:07:56Zhttps://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/874vapi: Update NM-1.0.metadata to include WireGuard declarations2021-06-04T08:07:56ZPeter van der Veldevapi: Update NM-1.0.metadata to include WireGuard declarationsCurrently, the WireGuard VAPI constants are incorrectly generated and not added to the correct namespaces.
This tiny patch will move, for example, NM.SETTING_WIREGUARD_SETTING_NAME to NM.SettingWireGuard.SETTING_NAMECurrently, the WireGuard VAPI constants are incorrectly generated and not added to the correct namespaces.
This tiny patch will move, for example, NM.SETTING_WIREGUARD_SETTING_NAME to NM.SettingWireGuard.SETTING_NAMEhttps://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/875[th/libnm-systemd-core-move] systemd: move "src/core/systemd" to "src/libnm-s...2021-05-30T07:52:10ZThomas Haller[th/libnm-systemd-core-move] systemd: move "src/core/systemd" to "src/libnm-systemd-core"This follows the recently introduced naming scheme and directory layout.
"libnm-systemd-core" is an independent component, and as such should no
be inside "src/core/".
Move it.This follows the recently introduced naming scheme and directory layout.
"libnm-systemd-core" is an independent component, and as such should no
be inside "src/core/".
Move it.https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/876[th/systemd] systemd: update code from upstream (2021-05-29)2021-06-01T12:27:37ZThomas Haller[th/systemd] systemd: update code from upstream (2021-05-29)https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/878libnm: introduce the new get_ports() in bridge/bond API2021-06-02T07:20:48ZFernando Fernández Manceralibnm: introduce the new get_ports() in bridge/bond APIThis patch is introducing get_ports() in bridge and bond API. In
addition, the existing get_slaves() functions are being marked as
deprecated.
Signed-off-by: Fernando Fernandez Mancera <ffmancera@riseup.net>This patch is introducing get_ports() in bridge and bond API. In
addition, the existing get_slaves() functions are being marked as
deprecated.
Signed-off-by: Fernando Fernandez Mancera <ffmancera@riseup.net>https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/879[th/wifi-ms-cost-info-ie] wifi: fix parsing Microsoft Network Cost IE2021-06-10T14:23:09ZThomas Haller[th/wifi-ms-cost-info-ie] wifi: fix parsing Microsoft Network Cost IEhttps://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/734
Fixes: 5307b1ed733f ('wifi: guess metered flag based on Network Cost information element')https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/734
Fixes: 5307b1ed733f ('wifi: guess metered flag based on Network Cost information element')https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/880[th/uuid-normalize]2021-06-04T07:47:57ZThomas Haller[th/uuid-normalize]with commit 207cf3d5d4c9, now `connection.uuid` gets normalized.
- update some places that still used `nm_utils_is_uuid()` when then can perform a stricter check (for the normalized UUID). This is mostly under `src/core/settings`.
- `c...with commit 207cf3d5d4c9, now `connection.uuid` gets normalized.
- update some places that still used `nm_utils_is_uuid()` when then can perform a stricter check (for the normalized UUID). This is mostly under `src/core/settings`.
- `connection.secondaries` contains a list of UUIDs. Rework the implementation of this property, but also add a normalization so that it only contains normalized, unique UUIDs.
---
the `parent` properties can be either an interface name or a UUID. Here we also should perform some normalization, that is not yet done....https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/881[th/va-args-join] change implementation of `NM_UTILS_{LOOKUP,ENUM2STR}_DEFINE()`2021-06-08T10:52:21ZThomas Haller[th/va-args-join] change implementation of `NM_UTILS_{LOOKUP,ENUM2STR}_DEFINE()`this is to avoid a lgtm.com warning:
https://lgtm.com/projects/g/NetworkManager/NetworkManager/?mode=tree&ruleFocus=1506698658528this is to avoid a lgtm.com warning:
https://lgtm.com/projects/g/NetworkManager/NetworkManager/?mode=tree&ruleFocus=1506698658528https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/882bash-completion: localize the prev variable2021-06-08T09:38:07ZSibo Dongbash-completion: localize the prev variableThe prev variable is not localzed in the nmcli Bash completion script,
which calls _init_completion.
Even though prev does not appear in the completion script, it should
still be localized. This variable may otherwise appear in the user...The prev variable is not localzed in the nmcli Bash completion script,
which calls _init_completion.
Even though prev does not appear in the completion script, it should
still be localized. This variable may otherwise appear in the user's
environment and clobber a user-defined variable of the same name, which
is bad.
Localize the prev variable.
Closes https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/741https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/883[th/coverity-fixes]2021-06-11T20:45:54ZThomas Haller[th/coverity-fixes]some more fixes.
See https://bugzilla.redhat.com/show_bug.cgi?id=1938672#c25some more fixes.
See https://bugzilla.redhat.com/show_bug.cgi?id=1938672#c25https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/885Configure secondary ip in Aliyun cloud2021-07-19T15:55:01ZWen LiangConfigure secondary ip in Aliyun cloudSigned-off-by: Wen Liang <liangwen12year@gmail.com>Signed-off-by: Wen Liang <liangwen12year@gmail.com>https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/886iwd: Don't add connection.interface-name by default2021-06-10T21:32:51ZAndrew Zaborowskiiwd: Don't add connection.interface-name by defaultThe `[connection].interface-name` setting can't be handled IWD (e.g. during autoactivations performed by IWD) and although for manual activations NM will probably check it, there's no reason for the IWD backend to restrict new profiles t...The `[connection].interface-name` setting can't be handled IWD (e.g. during autoactivations performed by IWD) and although for manual activations NM will probably check it, there's no reason for the IWD backend to restrict new profiles to one interface plus when running IWD the udev permanent interface naming is likely to be broken.
I wouldn't mind if this got merged into the new nm-1-32 branch too.
See also #743https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/887libnm: Don't crash if service tells us a new key management mode2021-07-16T16:41:35ZSimon McVittielibnm: Don't crash if service tells us a new key management modeThe NetworkManager service sometimes adds new key management modes.
If it does, an older client library (perhaps in a container) shouldn't
crash when talking to a newer NetworkManager service over D-Bus.
Resolves: NetworkManager#744
--...The NetworkManager service sometimes adds new key management modes.
If it does, an older client library (perhaps in a container) shouldn't
crash when talking to a newer NetworkManager service over D-Bus.
Resolves: NetworkManager#744
---
This is impossible to test as-is, because there's no newer version of NetworkManager; but it's a forward-port of a tested change made in a package based on libnm 1.6.2, which stopped Steam crashing with this assertion failure when talking to a NetworkManager that supports WPA3-Personal (SAE).https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/888[th/libnm-core-hide-structs]2021-06-11T20:42:33ZThomas Haller[th/libnm-core-hide-structs]I think our headers should not expose the GObject structs. Hide them.
This is of course an API change, but I think really nobody should be relying on this.I think our headers should not expose the GObject structs. Hide them.
This is of course an API change, but I think really nobody should be relying on this.https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/890[th/connection-track-setting-rework]2021-06-21T15:30:58ZThomas Haller[th/connection-track-setting-rework]rework how `NMConnection` tracks `NMSetting` instances.
Instead of a `GHashTable`, use a fixed size array (of 52 entries, for each setting type).
This also makes accessing the arrays predictable and stable.rework how `NMConnection` tracks `NMSetting` instances.
Instead of a `GHashTable`, use a fixed size array (of 52 entries, for each setting type).
This also makes accessing the arrays predictable and stable.https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/891tools: suppress duplicate state change in deactivation step of active connection2021-06-23T11:58:03ZFrédéric Martinsonstools: suppress duplicate state change in deactivation step of active connectionSigned-off-by: Frederic Martinsons <frederic.martinsons@sigfox.com>Signed-off-by: Frederic Martinsons <frederic.martinsons@sigfox.com>https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/892libnm: introduce nm_device_get_ports() to NMDevice2021-08-02T08:00:46ZFernando Fernández Manceralibnm: introduce nm_device_get_ports() to NMDevicehttps://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/893[th/cloud-setup-preserve-ip-from-profile] cloud-setup: preserve IPv4 addresse...2021-10-05T07:35:55ZThomas Haller[th/cloud-setup-preserve-ip-from-profile] cloud-setup: preserve IPv4 addresses/routes/rules from profilenm-cloud-setup automatically detects routes, addresses and rules and configures them
on the device using the emphermal Reapply() API. That is, it does not modify the
existing profile (on disk), but changes the runtime configuration only....nm-cloud-setup automatically detects routes, addresses and rules and configures them
on the device using the emphermal Reapply() API. That is, it does not modify the
existing profile (on disk), but changes the runtime configuration only.
As such, it used to wipe otherwise statically configured IP addresses, routes and
rules. That seems unnecessary. Let's keep the configuration from the (persistent)
configuration.
There is of course the problem that nm-cloud-setup doesn't really
understand the existing IP configuration, and it can only hope that
it can be meaningfully combined with what nm-cloud-setup wants to
configure. This should cover most simple cases, for more complex setups,
the user probably should disable nm-cloud-setup and configure the
network explicitly to their liking.
https://bugzilla.redhat.com/show_bug.cgi?id=1971527https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/894[th/config-match-lookup]2021-06-21T15:31:13ZThomas Haller[th/config-match-lookup]`nm_config_data_get_connection_default()` gets called *a lot*.
Mostly by `nm_device_get_route_metric()`.
In an example profiling run activating a 100 devices it thus shows up to consume 13.3% of the CPU time. That's mainly due to the o...`nm_config_data_get_connection_default()` gets called *a lot*.
Mostly by `nm_device_get_route_metric()`.
In an example profiling run activating a 100 devices it thus shows up to consume 13.3% of the CPU time. That's mainly due to the overhead of `g_key_file_get_string()` (interestingly, it is due to `glib_gettext()` to construct the `GError` message). Anyway, it sucks.
Add our own lookup index for these values and use binary search.https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/896[th/dhcp-factory-cleanup] minor cleanup of NMDhcpClientFactory2021-06-23T11:15:14ZThomas Haller[th/dhcp-factory-cleanup] minor cleanup of NMDhcpClientFactoryhttps://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/897[th/systemd] systemd: update code from upstream (2021-06-17)2021-06-23T12:37:44ZThomas Haller[th/systemd] systemd: update code from upstream (2021-06-17)