Spencer Oliver wrote:
>> We have an STR712 board by Olimex but (actually, an IAR 
>> kickstart board). We can do the "halt reset" with no  errors 
>> but it does not do a clean reset (the PC does not go to 
>> zero). With soft_reset_halt it does go back to zero. There 
>> are no errors logged in either case. 
>> soft_reset_halt just gives a cleaner reset.
>>
>> However, halt reset gives a clean reset with STM32 boards and 
>> I use it for them.
>>
>> Also, when we put in the reset_config ...  shown above the 
>> debug session becomes non-functional but no openocd errors are logged.
>>
>> -gene
> 
> Because the str7 has the srst and trst tied internally a reset halt will
> never halt @ 0.
> A reset halt on this device will always be a 'reset run' then 'halt' command
> - 
> so a few instructions will always be executed.

Exactly! We see it reset then gets to some application code that prints 
to the uart then stops.

> 
> This is why people often put a small reset delay in the startup file for the
> str7 to compensate for this.

Last I checked there was a str710.cfg file which we use. How do you set 
his "small reset delay" you speak of?  Would this be 
jtag_n[st]rst_delay? Should this be in str710.cfg?

> Other cores such as stm32 or str9 will reset correctly @ 0 when a 'reset
> halt' is issued.
> 
> Cheers
> Spen

_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to