Why is svf.c modifying cmd_queue_cur_state and cmd_queue_cur_end_state? This must end in tears as that state information belongs to the private parts of jtag.c layer and modifying only the JTAG layer doesn't take the driver into account.
Magnus explains a bit below a situation that I *believe* is akin to what happens when SRST is asserted and srst_pulls_trst. Somehow the calling code can know that the JTAG controller is in a different state than the JTAG layer expects. The calling code should then be able to tell the JTAG layer and driver what the actual state is. Is that it? https://lists.berlios.de/pipermail/openocd-development/2009-May/007172.html This patch modifies the broken code in svf.c: https://lists.berlios.de/pipermail/openocd-development/2009-May/007133.html -- Øyvind Harboe Embedded software and hardware consulting services http://consulting.zylin.com _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
