On Mon, Sep 24, 2012 at 6:46 PM, Spencer Oliver <[email protected]> wrote:
> On 24 September 2012 17:20, Andreas Fritiofson
> <[email protected]> wrote:
>> On Mon, Sep 24, 2012 at 4:50 PM, Spencer Oliver <[email protected]> wrote:
>>> They do not include the ftdi debug part in the schematics - i had to
>>> buzz the boards when i added the configs.
>>
>> Added the configs as in you didn't use the ones shipped with OpenOCD
>> (stm32-stick.cfg and hitex-str9-something.cfg)? In that case please
>> compare your own configs against those.
>>
>
>  i mean the ones shipped with OpenOCD are the ones i added and
> originally buzzed the connections.
> These configs have been in OpenOCD for a long time and do work with
> the older driver, i will revalidate the connections just to be sure.
>
> Spen

I'm sure the connections are fine if it works with the old driver.
However I'm not so certain that the layout is correct. For the STM32
board, it specifies

ftdi_layout_signal nSRST -data 0x0080

which means that the reset signal is connected from the FTDI chip via
a push-pull buffer to the CPUs. But OpenOCD is told that srst is open
drain, which gives rise to the following:

Error: 280 23 ftdi.c:165 ftdi_set_signal(): interface can't tri-state 'nSRST'

So that would mean that you could set srst_push_pull in your
reset_config. However, if the layout config is correct in srst being
push-pull, that would be really bad HW design, since it could prevent
the STM32 from resetting itself properly. So I'm more inclined to
believe that   the board actually has an open drain buffer or even the
FTDI pin connected straight to the reset line.

So assuming the FTDI pin is correct since you buzzed it (0x0080) one
of the following layout specs could be better.

If there's a tristate buffer with input grounded and #OE controlled by the FTDI:
ftdi_layout_signal nSRST -noe 0x0080

If there's a simple open-collector npn transistor stage:
ftdi_layout_signal nSRST -oe 0x0080

If the FTDI pin is unbuffered:
ftdi_layout_signal nSRST -data 0x0080 -oe 0x0080

Can you reverse-engineer that part of the schematics?

/Andreas

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
OpenOCD-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openocd-devel

Reply via email to