Error while running simple bluetooth demo on AM335 + WL18xx COM8 Module

Maybe this would have been a better question - What is the minimum udev version requirement needed to to get the bluetooth function to work properly?  We have udev-168 in our system. Should that work with the bluetooth ST and tty HCI driver that are part of the kernel?

Hi,

udev version on my set up is 182.

I did not build it, It was available by default in the file system, any further questions related to udev, please check it on am335x forum.

Hello,

My original question was: "What is the minimum version of udev required in order to make the bluetooth function work?  My first attempt to build system a system from scratch had udev-168.  It does not appear to work with the software included in your system - i.e. UIM and the kernel drivers.  Yes you have udev-182 in your system, but if I ask someone from TI where to get the tar file for that (and any support libraries) , no one can point me to that information.

Thanks,

Michael

Okay I seemed to have stumped the panel.  If you can not answer that question can you at least look over the following observances that I have made thus far:

1) To reiterate the problem: I cannot do the simple SetDevicePower 1 command after boot up and then launching the Bluetopia SS1BTPM & and LinuxSPPM on my rootfs built system. 

I have two systems: 1) the TI SDK 7.0 with binaries and 2) my system that I built using our rootfs structure and a kernel whose source I took from your am335x-evm-sdk-src-07.00.00.00.tar.gz and whose uim utility I copied from the TI SDK

2) I can successfully carry out the Bluetopia commands using TI's SDK rootfs, binaries and linux kernel there.  So the H/W is not an issue.

3) On TI rootfs system I can interchange my linux kernel that I built and STILL run the Bluetopia command successfully.

4) So that means that the st_drv and tty_hci in my kernel is good/works

5) I copied the uim utility from the TI rootfs system and it works and runs without  any problems.

6) So the problem Has to be with the environment on my rootfs system.  Something is still missing or incorrect.

What else is required to get a communication path between the host and the WL1835MOD working????

I can even delete the bluetooth.ko module from the TI's system and just issue the Bluetopia commands and it still works on the TI system.  So not even that is required to get the communication path working. 

I am using the same firmware that you have in /lib/firmware  i.e -> TIInit_11.8.32.bts and that is not the problem.

Our company wants to buy your WL1835MOD daughter boards but it won't happen if we can't find a way to make in work in linux. 

Thanks for any insight that you might have.  This has been tedious endeavor just to get a proof of concept!

Michael

I have observed a particular difference in my two systems - the TI SDK system in which bluetooth works and my home grown system that does not work.  The difference seems to be that /sys/devices/kim.12/install does not change.  I am not seeing the "uim: signal received , opening /dev/ttyO1" message in my system. 

I have two sets of logs, the first is the working TI SDK system. The 2nd is my system.

***** TI SDK SYSTEM *****

SPPM>Initialize 1

BTPM_Initialize() Success: 0.
DEVM_RegisterEventCallback() Success: 5.
SPPM>SetDevicePower 1

[  100.901752] (hci_tty): inside hci_tty_open (dd2cb380, dd1072c0)
[  100.908049] (stc):  chnl_id list empty :4
[  100.912300] (stk) : st_kim_start(stk) :ldisc_install = 1uim:poll broke due to event 10(PRI:2/ERR:8)

uim:read 1 from install

uim:@ st_uart_config
uim: signal received, opening /dev/ttyO1
uim:@ set_baud_rate
uim:set_baud_rate() done
uim:Setting speed to 3000000
uim:@ read_command_complete
uim: Command complete started
uim:@ read_hci_event
uim: read_hci_event
uim:Command complete done
uim:Speed changi
[  101.075981] (stc): st_tty_open ng to 3000000, 1
uim:@ set_custom_baud_rate
(stk) :line discipline installed
[  101.090253] (stk) :TIInit_11.8.32.btsuim:Installed N_TI_WL Line displine
uim:begin polling...
(stk) :change remote baud rate command in firmware
[  101.108775] (stk) :skipping the wait event of change remote baud(stk) :sleep command in scr
[  101.805195] (stk) :sleep command in scr[  102.755455] (stc): add_channel_to_table: id 4
[  102.760077] (stc): add_channel_to_table: id 2
[  102.764718] (stc): add_channel_to_table: id 3
DEVM_PowerOnDevice() Success: 0.

Device Powered On.

******* MY System ***********

SPPM>Initialize 1

BTPM_Initialize() Success: 0.
DEVM_RegisterEventCallback() Success: 5.
SPPM>SetDevicePower 1

uim:poll broke due to event 10(PRI:2/ERR:8)

uim:read 1 from install

uim: Inside st_uart_configuim: Inside set_baud_rateuim:set_baud_rate() done

 => Notice NO signal received message ! <=
uim: Inside read_command_completeuim: Inside read_hci_eventuim:Command complete done
uim: Inside set_custom_baud_rateuim: Inside read_firmware_version
uim:Installed N_TI_WL Line displine
uim:begin polling...
uim:poll broke due to event 10(PRI:2/ERR:8)

uim:read 0 from install

uim: Inside st_uart_configuim:Un-Installed N_TI_WL Line displine
uim:begin polling...
DEVM_PowerOnDevice() Failure: -3, Unknown Error.
Function Error.
SPPM>

I checked the code in uim.c and if install is not equal to 1 when the signal is supposed to be received then it exits and the line descipline is never changed. 

So I guess the problem is that /sys/devices/kim.12/install is not getting changed to a 1.  What would prevent that from happening? 

Thanks  for any insight that you might have.

Michael

Hi,

This happens when the GPIO is not toggled.

Do you see "line disc installation timed out" in the kernel logs.

Hello - No, I checked (grep-ed) the file log as well as the boot console start-up messages and there was no evidence of "line disc installation timed out" in there.

Hello-

Just noticed that by settting the clock (date) to the current time solved my problem. No kidding!!!

And just for a test I went back to the TI SDK 7.00 system and voided the time and guess what?  The same problem manifests itself and the system fails!

Hope everyone out there is aware of this phenomenon.

thanks for you patience and all the help.

Michael

HI Michael.

I guess i'm facing the same problem as you are in our custom boards. We are getting the same DEVM_PowerOnDevice() Failure: -3, Unknown Error.

Can you please clarify what you meant by setting the clock to the current time? Have you set the time in the board?

Regards,
Balaji

Hi Balaji,

It's not somuch of setting the date as it was running the hwclock.sh script.  If you look  in the SDK7.0 version of the TI /etc/rcS.d/ directory and see their S55bootmisc.sh script, you will see what I am referring to.  I think it had to do with making sure the clock was running with a reasonable initialization factor. 

I know if I fail to run the hwclock.sh, the Bluetooth functions stalled/failed every time.  If I did execute the hwclock.sh the system behaved successfully. 

Hope this helps

Michael

Hi Michael,

Thanks for the pointers. The BT stack is working now.

Hi,

I am testing file transfer from Bluetooth.

BT is pairing with my handset ok.but when file transfer it's gives me connection error on handset.

and also i was run LinuxFTPM_CLT application for file transfer.

FTPM>Connect 103B5910218A 1 0
Remote Device Properties Changed.
BD_ADDR:       103B5910218A
Device Flags:  0x800005CD
Connect State: TRUE
FTPM>
etFTPMConnectConfirmation, FTPConnectionID: 0x0001, Status 0x0002
BD_ADDR: 103B5910218A

FTPM>dir
You must open an FTP Client before issuing this command.
Function Error.


How to open FTP Client ? Any idea ?

Thanks
Tejas