That's odd. In my experience debugging with the -rtos riscv hack in the OpenOCD debug script allows OpenOCD (and gdb) to deal with all harts (and gdb treats then as if they were threads) ok. Maybe you can describe your configuration a bit more? And what version of OpenOCD are you using/building?
On Sebastian Huber <[email protected]>, 19 Sep 2018 1:13 p.m. wrote: Hello Tommy, thanks for your hints. We already use the OpenOCD from the RISC-V Github site. We have done some instrumentation for JTAG and we see that OpenCD ignores second hart (it can enumerate it correctly but it does not initialize debug interface for it). On the first hart, single stepping doesn't work. Once we hit a break point we are stuck on this break point, e.g. continue stops immediately on the break point we hit without executing other code in between. -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : [email protected]<mailto:[email protected]> PGP : Public key available on request. Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
_______________________________________________ OpenOCD-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/openocd-devel
