Bluez rfcomm/SPP issue with zephyr hrm
 

Hi, I have a problem reading data from a zephyr heart rate bt monitor using standard rfcomm ports.

Using it on a normal linux laptop with bt dongle runs fine, with:

* rfcomm bind /dev/rfcomm0 hrm_bt_addr
* cat /dev/rfcomm0

I see incoming data correctly (about 60 bytes per seconds) and I'm able to parse the protocol..

Doing the same on the n900 results in only 1-3 bytes per sec.

Using AF_BLUETOOTH sockets I'm able to read data correctly, but I'd like to investigate to understand if the broken rfcomm way is a my problem/mistake or a bug in bluez.

Anyone having a similar problem?

Thanks

Re: Bluez rfcomm/SPP issue with zephyr hrm
 

Oh... it seems that setting tty raw mode fixes the problem.

Re: Bluez rfcomm/SPP issue with zephyr hrm
 

I'm playing with talking to the same device.

Are you developing in QT or scratchbox?

Be curious to swap notes.

Thanks,
David

Re: Bluez rfcomm/SPP issue with zephyr hrm
 

With Nokia Qt SDK driving bluez via dbus, then I open the resulting /dev/rfcomm* with posix api (as QFile did not emit signals when new data arrives) and a socketnotifyer.

I started my experiments for a simple reason, to have a big font display with my heart rate and a disabled screen dimming, to be able to read it during my 1 our cyclette session from 1/2 meters.

Next step is having espeak saying "please come on, you are going too slow" ;)

Re: Bluez rfcomm/SPP issue with zephyr hrm
 

Hmm, I would like to see your dbus code if you don't mind. I have not been able to find a good example of initiating RFcomm over DBUS in C.

After I manually create the /dev/rfcomm device, I am using a QT serial port application as the basis for my code. I'm getting packets, hope to work on my parser this weekend.

David