On Monday 06 July 2009, Øyvind Harboe wrote:
> on wi-9c.cfg which I'm testing against "reset run" + "halt" does not work.
> "step" sits in an infinite loop(or times out now).

Hmm, on a dm365 I saw some odd behavior but I attributed that
to it being beta hardware, since last time I tried anything
with a dm355 those things behaved.  "reset halt" doesn't work
as expected (i.e. stop with PC at zero); and for that matter,
"reset" didn't always seem to work ... after resetting it a few
times, the CPU said it was still running with MMU active.

I did see misbehavior with an arm920t where "step" and
"vector catch" didn't seem to work, a month or so back.
This is without GDB.  The "vector catch" didn't seem to
be effective at writing the value ... readback didn't
show it was written, it didn't act as if it had been,
but a breakpoint at address zero showed kicked in.  And
the "step" mostly seemed like a "resume".

Personally, I'm still having a bit of a hard time telling
the difference between operator error, buggy chips, and
buggy software.  Those dm365 things seem more on the side
of hardware bugs... though maybe not entirely.  The arm920t
issues seemed more like software.

It's just *some* things which seem broken.

- Dave


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

Reply via email to