Hi,
I am attempting to RAM debug an LPC2294 on a custom board with OpenOCD / Eclipse. When I attempt to debug, I get occasional (appear to be totally random) error messages from OpenOCD:
“Error: arm_jtag.c:38 arm_jtag_set_instr_error_handler(): setting the new JTAG instruction failed, debugging is likely to be broken”
If I attmempt to step across a time-kill ‘for’ loop, I get the following error messages from OpenOCD:
"Error: armv4_5.c:186 armv4_5_mode_to_number(): invalid mode value encountered
Error: arm7_9_common.c:1083 arm7_9_debug_entry(): cpsr contains invalid mode value - communication failure
Error: gdb_server.c:608 gdb_get_registers_packet(): gdb requested registers but we’re not halted, dropping connection
Info: server.c:362 server_loop(): dropped ‘gdb’ connection
Warning: arm7_9_common.c:682 arm7_9_poll(): DBGACK set while target was in unknown state. Reset or initialize target before resuming"
after which the Eclipse debugging is halted. I have tried jtag_speed settings of 0, 5, 10 and 100. What is up with this flakey debugging?
Additional information:
-
LPC2294 runs from a 12MHz crystal oscillator
-
Using OLIMEX ARM-JTAG (20-pin header, parallele port connection to PC)
-
PC is 3GHz CPU, Windows XP Pro
-
On PCB: TRST, TDI, TMS, TDO have 10kohm pull-ups, TCK has 10kohm pull-down, RTCK has removable 10kohm pull-down.
-
LPC2294 is unprogrammed (maybe?)
I really appreciate everyone’s thoughts in this matter.
Luke