Yes, I know I'm replying to myself, but I have found something.

On Fri, Dec 19, 2008 at 08:59:25AM -0600, [email protected] 
wrote:
> On Fri, Dec 19, 2008 at 02:04:06PM -0000, Spencer Oliver wrote:
> > It correctly finds the rlink now and the jtag id's but fails to stop a stm32
> > - look into it a bit more when i have some time.
> 
> Hmmm.  What's involved in that operation?  Anything at this level?
> I had to get this working before I could even think about trying to
> learn to actually use OpenOCD.

I have found that "reset run" seems to bring the target back up in some
sort of odd state.  The left half (a little less than half, actually) of
the screen on the Primer (which is running a CircleOS image) is white,
and the cursor isn't moving.  That particular condition probably means
little in the context of OpenOCD, except that it means that something
isn't in the right condition when the target starts running again.

Either "reset halt" or "reset init" followed by "resume" appear to work.
And by work, I mean that the reset blanks the screen, and the resume
gets the target running again.  That definition of work may differ from
the real one.  The eariler (wedged after reset run) condition can be
cleared by doing this.  Presumably just because the target is being
reset.

Maybe this is related to the problem you're seeing, though you seem
to be saying that an attempt to stop the target simply has no effect.
Or maybe this is not a problem, and I'm just interpreting it wrong.
This is just where I started looking when you said you couldn't stop the
target.

In any case, any idea what I should be looking at?  I understand the
hardware interface somewhat intimately now, but get lost fast outside of
rlink.c.
_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to