2012/11/25 Franck Jullien <[email protected]>:
> 2012/11/25 Andreas Fritiofson <[email protected]>:
>>
>> On Sat, Nov 24, 2012 at 11:11 PM, Franck Jullien <[email protected]>
>> wrote:
>>>
>>> 2012/11/24 Franck Jullien <[email protected]>:
>>> > 2012/11/24 Franck Jullien <[email protected]>:
>>>
>>> >> I'm doing some tests with an opencores TAP and debug system using
>>> >> openOCD and an ftdi based JTAG interface.
>>> >> I'm using the new ftdi driver but I have a problem.
>>> >>
>>> >> When I want to do a DR SCAN of one bit and then stay in DR SCAN, I
>>> >> have TMS signal rising....
>>> >> If someone familiar with the ftdi driver could take a look at the
>>> >> waveform and test code:
>>> >>
>>> >> http://pastie.org/5429202
>>> >> http://picpaste.com/pics/jtag-22Kzxqqn.1353794728.png
>>> >>
>>> >> (I also tested the ft2232_libftdi driver with the same result).
>>> >>
>>> >> Franck.
>>> >
>>> > I forgot to paste the debug logs:
>>> >
>>> > ------------------------------
>>> > tap state is RUN/IDLE
>>> > DRSCAN type:1
>>> > start=RUN/IDLE goal=DRSHIFT
>>> > in field 0/1 1 bits
>>> > DR scan, 1 bits, end in DRSHIFT
>>> > ------------------------------
>>> > tap state is DRSHIFT
>>> > DRSCAN type:1
>>> > in field 0/1 1 bits
>>> > DR scan, 1 bits, end in DRSHIFT
>>> >
>>> >
>>> > Franck.
>>>
>>> And I also forgot to say that after this 1 bit scan, on the next 32
>>> bits DR SCAN, there is not TCK generated, JTAG signals are not moving
>>> anymore....
>>>
>>
>> Hi!
>>
>> I'll see if if I can find some time to look into this. It would be good if
>> you could extend your code snippet above to generate the complete sequence
>> that shows the issue, i.e. including the following scan that doesn't work.
>> Looking only at your example it seems to work as expected (apart from the
>> peculiarity that tms is toggling but that in itself is not a problem since
>> there are no clocks).
>>
>> What are you trying to do? Repeatedly clocking data out of DR without going
>> through DRCAPTURE? Is it supposed to work like that?
>>
>> /Andreas
>>
>
> Hi,
>
> Thanks for your answer.
>
> I need to stay in DRSHIFT until I get a '1' out of TDO, then read some
> bytes without moving the TAP.
>
> So I'm doing something like this:
>
> int adbg_wb_burst_read(struct or1k_jtag *jtag_info, int word_size_bytes,
>                         int word_count, unsigned long start_address, void 
> *data)
> {
>         /*....*/
>
>         /* Get 1 status bit, then word_size_bytes*8 bits */
>         status = 0;
>
> #if REMOVE_THIS_AND_IT_WORKS
>
>         field[0].num_bits = 1;
>         field[0].out_value = NULL;
>         field[0].in_value = (uint8_t *)&status;
>
>         while (!(status & BURST_READ_READY) && retry < MAX_READ_STATUS_WAIT) {
>                 jtag_add_dr_scan(tap, 1, &field[0], TAP_DRSHIFT);
>                 jtag_execute_queue();
>                 retry++;
>         }
>
> #endif
>
>         /*......*/
>         /* When we are ready, read some bytes */
>
>         jtag_add_dr_scan(tap, nb_fields, field, TAP_IDLE);
>         jtag_execute_queue();
>
>         /*......*/
>
>         return ERROR_OK;
> }
>
> Here you can find the waveform with and without the polling for status bit:
>
> http://picpaste.com/pics/jtag_1-NtI4rr5S.1353850723.png
>
> I'll try to run openocd with gdb to see what's going on.....
>
> This code works with the usb-blaster driver.
>
> Franck.

My waveform wasn't complete, the TAP doesn't hang while "polling DR
SHIFT", it's just longer...
However, my problem is coming from this TMS set to '1' after the
jtag_add_dr_scan of
1 bit with an end state set to DRSHIFT (plus this clock pulse in the
middle of the waveform) :

http://picpaste.com/pics/jtag_2-sZk1MbUP.1353879598.png

Still searching....

Franck.

------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
_______________________________________________
OpenOCD-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openocd-devel

Reply via email to