Hi!
2008/6/5 Paul Stickney <[EMAIL PROTECTED]>:
> What verifications are used to verify sendDone is not being called?
> That the initial send is SUCCESS?
I use printf() to check the error_t state (did not post this to make
the code more readable).
For send it looks like this:
switch( call CollectionSend.send(&paket, sizeof(measuringData_t)) ) {
case SUCCESS: printf("SensorMoteC.nc:CollectionSend.send(...) Error:
SUCCESS\n");
break;
case FAIL: printf("SensorMoteC.nc:CollectionSend.send(...) Error:
FAIL\n");
break;
case EBUSY: printf("SensorMoteC.nc:CollectionSend.send(...) Error:
EBUSY\n");
break;
case EOFF: printf("SensorMoteC.nc:CollectionSend.send(...) Error:
EOFF\n");
break;
case ESIZE: printf("SensorMoteC.nc:CollectionSend.send(...) Error:
ESIZE\n");
break;
default: printf("SensorMoteC.nc:CollectionSend.send(...) Error:
UNKNOWN\n");
break;
}
The output is:
$ java PrintfClient -comm $MOTECOM2
Thread[Thread-1,5,[EMAIL PROTECTED]:57600: resynchronising
SensorMoteC.nc:CollectionSend.send(...) Error: SUCCESS
SensorMoteC.nc:CollectionSend.send(...) Error: EBUSY
SensorMoteC.nc:CollectionSend.send(...) Error: EBUSY
SensorMoteC.nc:CollectionSend.send(...) Error: EBUSY
...
Similar procedure for sendDone(..)...
> The logic looks inverted.
> Shouldn't collectionSendBusy only be set to true IFF it was successful?
>
>> if (error != SUCCESS) collectionSendBusy = TRUE;
>> else call DbgMessenger.beep(10);
Yes, you are right. It is inverted :-O
But I have bypassed the collectionSendBusy-check in my latest tests.
So this should not be the real problem.
Any ideas what I could try or look at? I am working with micaz motes,
MIB510 programmingboard. and MTS 310 sensorboards.
Thanks
Nicola
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
