hi richard,
you may have a clock skew problem, as the bluetooth module uses the uart
to communicate with the mcu on the shimmer, and justfatlogging uses a
high-speed clock configuration.
my casual observation (please don't send me code if this isn't the
problem) sez that you'll need two things to get going: 1) a flag in
your makefile telling the bluetooth module that you're running this
clock configuration (PFLAGS += -DUSE_8MHZ_CRYSTAL) and 2) an updated
bluetooth driver, the public version of which was missing a second set
of uart settings for this case.
good news is that your mail reminded me that i never got around to
releasing this driver update, but i have now!
update ~/tinyos-main/tos/platforms/shimmer/chips/bluetooth/ to get what
you need, then edit the file to swap the comments between the 1mhz and
4mhz smclk calls.
i hope that this helps,
steve
ps: you should not look at FAT_EKG for guidance; it has no relevance to
our mainstream codebase. in fact, thanks for reminding me to remove it
from the repository completely.
On 05/31/2011 10:39 AM, [email protected] wrote:
Hi,
I'm attempting to integrate the bluetooth function into justfatlogging,
and I've run into a couple of problems:
Firstly, after the code compiles and is loaded, it seems that bluetooth
commandModeEnded does not get triggered, as oppose to the case in
HelloWorld. Secondly, bluetooth.writeDone is not triggered either with
no good explanation, thus the ready flag is not changed within and the
write only happens once. Lastly, I'm using hyperterminal to intepret a
test message sent through bluetooth. the same message from HelloWorld,
which displays perfectly as ascii, would become jumbled and not
completely sent when the functionality is brought over to justfatlogging.
All that i'm doing is copying over the code from helloworld, plus adding
in the necessary initialization etc. Does anyone have any idea regarding
this? Also, is there a better way to integrate SD+bluetooth
functionality? I've checked out Reconfigurable and FAT_EKG and both of
them seem to point this way.
Thanks a lot,
Richard
_______________________________________________
Shimmer-users mailing list
[email protected]
https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users
