support for CC2451 ?

[ericcastro]

I spent a good amount of time doing all the wiring only to find out this is not compatible with the CC2451 :-(

would it be too difficult to support it ?

[wavesoft]

Hello ericcastro,

the library itself implements the CC Debugger protocol from TI, therefore it should be able to flash the whole CC253x/4x family. However there need to be some changes to the examples, since they were tailored to the BLE112/BLE113 commercial modules (that need a licence key in some particular memory address etc.)

• The arduino proxy can stay as-is, since it's very low-level
• The cclib python library provides all the functions you will need
• Only the cc_write_flash.py needs to be changed in order to match your needs

In detail you will only need to comment-out the BLE functionality (getBLEInfo + the block of code for checking the license, and getBLEPStoreSize + the block of code to save the PStore).

If you want to test, modify the cc_info.py and comment-out the lines 46-50. If it displays the correct chip information you are ready to go!

If it works I can write an alternative write_flash.py script.

[wavesoft]

You will also need to update the chipID guard (ccdebugger.py:95-96)

        if ((self.chipID & 0xff00) != 0x8d00) && (self.chipID & 0xff00) != 0x4100)):
            raise IOError("This class works ONLY with CC2540/2541 TI chips (This is a 0x%04x)!" % self.chipID)

[ericcastro]

trying this now. will let you know how it goes!

[ericcastro]

ok, output looks much better now

Using CCDebugger on port /dev/tty.usbmodem1421

Chip information:
      Chip ID : 0x4113
   Flash size : 16 Kb
    SRAM size : 1 Kb
          USB : No

Debug status:
 [ ] CHIP_ERASE_BUSY
 [ ] PCON_IDLE
 [X] CPU_HALTED
 [ ] PM_ACTIVE
 [ ] HALT_STATUS
 [X] DEBUG_LOCKED
 [X] OSCILLATOR_STABLE
 [ ] STACK_OVERFLOW

Debug config:
 [ ] SOFT_POWER_MODE
 [ ] TIMERS_OFF
 [ ] DMA_PAUSE
 [ ] TIMER_SUSPEND

let me know if this is correct / expected, and what do we need to flash it. thanks for the help!

[wavesoft]

Hmm.. the device seems to respond properly, however the memory values seems a bit off. This either means that it couldn't read the CHIPINFO0 (0x6276) register (ex. different instructions between the two chips), or the register is actually zero.

Try to read the CHIPID Register (that should be 0x4113) or any other register that you know it's value in order to check if this is register issue:

chipid = dbg.readXDATA(0x624A, 1)
print "0x%04x" % chipid[0]

[ericcastro]

thanks, I already left home for work but will try this tonight and will let you know.

[ericcastro]

ok I tried this and not getting anything very interesting so far
the lines:

chipid = dbg.readXDATA(0x624A, 1)
print "0x%04x" % chipid[0]

will output:

0x0000

also, I uncommented some of the output lines you mentioned to remove just to see if there was anything there, I got:

Device information:
 IEEE Address : 000000000000
           PC : 0000

Firmware information:
      License : 0000000000000000000000000000000000000000000000000000000000000000
   BT Address : 00:00:00:00:00:00
 Hardware Ver : 00

I suspect there could be something wrong with the wiring I made ? I may have misunderstood something with the voltage dividers maybe ? Here is a picture of my breadboard so you can take a look as well if you want:
http://i.imgur.com/8yU1WNQ.jpg

and this is my pin config in CCLib proxy on my Leonardo:

int LED      = 13;
int CC_RST   = 10;
int CC_DD_I  = 8 ;
int CC_DD_O  = 7;
int CC_DC    = 9;

[wavesoft]

Hello @ericcastro, sorry for my late reply, but I was a bit buried with work.
I see some problems with your wiring (top right voltage divider).

You can check this diagram for reference:

Let me know if you manage to get this working in order to update the my compatibility table ;)

[wavesoft]

Hey @ericcastro do you have any update on this? Otherwise I will close this issue.

[ericcastro]

oh, I'm sorry I never did reply. Actually I was able to use a different utility (CCLoader by RedBearLabs) in order to successfully flash my CC2451 - so from my end this is no longer relevant.

Thanks - Eric

[wavesoft]

Thanks for the reply @ericcastro