On Tue, 2015-02-17 at 21:02 +0300, Alexey Mednyy wrote:
> > I would think that what calculate_strength would return values that
> > were somewhat variant, and less than 100, but it is not.... am I
> > missing something basic here?
> >
> > Tom
...
> Is there any advance in this question?
The above issue has been fixed with commit
ce82d7c91214dfe9ea4abad036c8d2d8b0d9fd24
> But I've got even more strange situation my wpa_supplicant/driver not
> sure returns positive values eq:
>
> dict entry(
> string "Signal"
> variant int16 46
> )
>
> So all my wifi networks always assumed 100% signal.
>
> I use last revision of connman, wpa_supplicant v2.3, rtl8188eu with
> 2.6.36 kernel on arm device.
> Any advice?
IIRC one cause for this was using ancient wext drivers. The kernel
version you are destined to use is ancient enough for me to believe this
is a plausible reason.
What I don't remember is what the maximum positive value for the
strength reported can be, is it 100 (as in %), or 127 (as in signed
byte). The signal strength is calculated is in plugins/wifi.c,
calculate_strength().
Patches are welcome.
Cheers,
Patrik
_______________________________________________
connman mailing list
[email protected]
https://lists.connman.net/mailman/listinfo/connman
