On 04/16/2018 10:29 PM, Vasyl Vavrychuk wrote:
>> Documentation on BackgroundScanning says
>> When BackgroundScanning is false, ConnMan will not perform any scan
>> regardless of wifi is connected or not, unless it is requested by
>> the user through a D-Bus call.
>> but code in connman_device_set_powered thinks other way. It starts
>> scan after device power enable.
> The BackgroundScanning variable is supposed to by honored in plugins/wifi.c.
> Unfortunately, the scanning logic is not simple. I wouldn't be surprised
> that there are still bugs in there. Do see scans when BackgroundScanning is
Actually I am talking about different thing. I do not see any scan
scheduled by wifi.c plugin by itself if BackgroundScanning is false.
What I see is that connman_device_set_powered issue scan upon every
power on. Plugin can not (or should not) distinguish this scan from
user requested scans.
So this should be solved on the device.c level if we want.
I digged into the history of this scan call. It seems to exist since the
early days. The code base as evolved quite a lot since then.
I am not sure that it is a problem form user stand point. That is
It might be that we don't need it anymore. At least the initial commit
leads to this conclusion 9b05cebe93ad ("Add support for automatic
connection policy"). Though I am a bit hesitant to remove it without