Why do we rfkill in Sugar?

[Paul Fox]

[ moving to devel ]

martin wrote:
 > Hi Paul, techteam,
 > 
 > you patched the codepath that controls the wireless disable to run
 > rfkill to *really* kill the network when wwe mean to kill the network.
 > 
 > Why did we do this? Regulatory reasons? Is there a tracking bug where
 > I can get more background? The commit msg was rather terse :-)
 > 
 > Why do we care? It is confounding GNOME/NM/nm-applet, and depending on
 > the rationale, my workarounds may be valid or not.

we needed a way to control power to the wireless card for power
management.  on XO-1, we used a private mechanism
(/sys/power/wlan-enabled), but on 1.5, we wanted to something
"standard".  so we implemented rfkill.  we later, also
implemented rfkill on xo-1, which makes common code at user level
possible.

NM will react to changes in a devices rfkill status.  it will not,
however, actually control rfkill.  and yes, i was aware that the
NM applet "disable" function was incompatible.

what are your workarounds?

paul

 > 
 > cheers,
 > 
 > 
 > 
 > m
 > -- 
 >  martin at laptop.org -- School Server Architect
 >  - ask interesting questions
 >  - don't get distracted with shiny stuff  - working code first
 >  - http://wiki.laptop.org/go/User:Martinlanghoff

=---------------------
 paul fox, pgf at laptop.org