[PATCH 2/7] GPS is configurable by DGPS_ENABLE variable

[Tomas Jura]

On 4.2.2016 17:04, Dan Williams wrote:
> On Tue, 2016-02-02 at 15:30 +0100,tomas.jura1 at gmail.com  wrote:
>> From: Tomas Jura<tomas_jura1 at gmail.com>
>>
>> ---
>>   plugins/mbm/mm-broadband-modem-mbm.c | 6 ++++--
>>   1 file changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/plugins/mbm/mm-broadband-modem-mbm.c b/plugins/mbm/mm
>> -broadband-modem-mbm.c
>> index bfd26da..efd4f63 100644
>> --- a/plugins/mbm/mm-broadband-modem-mbm.c
>> +++ b/plugins/mbm/mm-broadband-modem-mbm.c
>> @@ -1433,8 +1433,10 @@ parent_enable_location_gathering_ready
>> (MMIfaceModemLocation *self,
>>   
>>       if (start_gps) {
>>           gchar *s_gps_interval = getenv("GPS_INTERVAL");
>> -        gchar *buf = g_strdup_printf("AT*E2GPSCTL=1,%d,0",
>> -                                     s_gps_interval ?
>> CLAMP(atoi(s_gps_interval),1,60) : MBM_GPS_NMEA_INTERVAL );
>> +        gchar *s_dgps = getenv("DGPS_ENABLE");
>> +        gchar *buf = g_strdup_printf("AT*E2GPSCTL=1,%d,%d",
>> +                                     s_gps_interval ?
>> CLAMP(atoi(s_gps_interval),1,60) : MBM_GPS_NMEA_INTERVAL,
>> +                                     s_dgps ?
>> CLAMP(atoi(s_dgps),0,1) : 0 );
>>           mm_base_modem_at_command_full (MM_BASE_MODEM (self),
>>                                         
>>   mm_base_modem_peek_port_primary (MM_BASE_MODEM (self)),
>>                                          buf,
> Is there ever a reason to *not* use DGPS if it's available?  Does it
> increase startup time when AGPS is not used?
>
> Dan
Hi

I made a practical measurement of fix time for DGPS is on and off. 12 
measurements for each case.
DGPS off average 40.4s ( stdev: 2.9 )
DGPS on average 39.8s ( stdev: 1.6 )
The precision measurement is limited to minimum GPS reporting interval 
to 1s.

Without modem reset between measurements the time is also same - around 
3 seconds. The fix times are same for both cases (DGPS on/off).   Seems 
that enabling DGPS on has no effect to fix time. I could not try 
DGPS+AGPS, because my modem does not supports it.

BTW: During the measurement I had serious problems with modem 
disconnections. The modem is MiniPCI card. The accompanying kernel 
message was:

Mar 27 14:34:02 het kernel: cdc_wdm 4-4:1.5: wdm_int_callback - 0 bytes
Mar 27 14:34:02 het kernel: cdc_wdm 4-4:1.6: wdm_int_callback - 0 bytes
Mar 27 14:36:08 het kernel: usb 4-4: USB disconnect, device number 29
Mar 27 14:36:08 het kernel: cdc_ether 4-4:1.7 wwan0: unregister 
'cdc_ether' usb-0000:00:1d.7-4, Mobile Broadband Network Device
Mar 27 14:36:08 het kernel: cdc_acm 4-4:1.9: failed to set 
dtr/rts                 <========== HERE
Mar 27 14:36:13 het kernel: usb 4-4: new high-speed USB device number 30 
using ehci-pci
Mar 27 14:36:14 het kernel: usb 4-4: New USB device found, 
idVendor=0bdb, idProduct=1900
Mar 27 14:36:14 het kernel: usb 4-4: New USB device strings: Mfr=1, 
Product=2, SerialNumber=3
Mar 27 14:36:14 het kernel: usb 4-4: Product: Ericsson F3507g Mobile 
Broadband Minicard Composite Device
Mar 27 14:36:14 het kernel: usb 4-4: Manufacturer: Ericsson
Mar 27 14:36:14 het kernel: usb 4-4: SerialNumber: 3541430203035820

Have you seen something like this? What could be the reason? Power 
instability on battery?

Tomas