After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 755610 - Export S390Subchannels property on Device.Wired and use that in libnm
Export S390Subchannels property on Device.Wired and use that in libnm
Status: RESOLVED FIXED
Product: NetworkManager
Classification: Platform
Component: general
git master
Other Linux
: Normal normal
: ---
Assigned To: NetworkManager maintainer(s)
NetworkManager maintainer(s)
Depends on:
Blocks:
 
 
Reported: 2015-09-25 08:02 UTC by Jiri Klimes
Modified: 2015-09-29 08:00 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Jiri Klimes 2015-09-25 08:02:43 UTC 
S390Subchannels is not exported on D-Bus.
Let's export the property and add it to libnm:NMDeviceEthernet too.
Comment 1 Jiri Klimes 2015-09-25 08:04:11 UTC 
The code is in branch jk/device-s390-subchannels.
Comment 2 Thomas Haller 2015-09-25 08:48:59 UTC 
LGTM
Comment 3 Thomas Haller 2015-09-25 09:04:53 UTC 
hold on, it breaks:

make check -C libnm-glib/
Comment 4 Jiri Klimes 2015-09-25 11:39:56 UTC 
(In reply to Thomas Haller from comment #3)
> hold on, it breaks:
> 
> make check -C libnm-glib/

Fixed, branch re-pushed.
Comment 5 Dan Williams 2015-09-25 21:28:59 UTC 
> libnm: match s390 subchannels and MAC blacklist in connection_compatible()

A device can have either 2 or 3 subchannels, so I think we need to compare the # of channels in match_subchans() too.  Right now if the connection has 2 but the device has 3, it could match.  Should probably have a testcase for that too.

IIRC ctc uses only 2 but qeth uses 3.
Comment 6 Jiri Klimes 2015-09-29 08:00:51 UTC 
(In reply to Dan Williams from comment #5)
> > libnm: match s390 subchannels and MAC blacklist in connection_compatible()
> 
> A device can have either 2 or 3 subchannels, so I think we need to compare
> the # of channels in match_subchans() too.  Right now if the connection has
> 2 but the device has 3, it could match.  Should probably have a testcase for
> that too.
> 
Done both in libnm and core, testcase added.

Committed to master:
edec77f merge:  S390 subchannels improvements (bgo #755610)
82e4364 tests: add a test for connection_compatible() for wired devices
e7ed259 cli: print s390 subchannels for devices that have that
9abe74d libnm: match s390 subchannels and MAC blacklist in connection_compatible()
bbaca1b libnm: add 390-subchannels property to NMDeviceEthernet
4219aa9 device: export S390Subchannels property on Ethernet device