On 04/27/2011 09:42 AM, Eric Cooper wrote: > I'm a JTAG newbie and could use some help configuring openocd for use > with a Seagate DockStar, a Marvell Kirkwood (feroceon) board similar > to the Sheevaplug. > > The setup "mostly" works: I can halt, resume, and single-step the > target reliably from the openocd telnet client. I can load images > into RAM and run them. > > Here's what doesn't work: > > 1. In gdb, I can step and continue, but I can't halt. Once I've > continued, typing ^C does nothing. (Typing it a second time just > makes gdb offer to quit.) I can still halt from the openocd telnet > client.
That's odd. I've never experienced this. I'm a little behind the latest in git. Recently introduced bug in gdb server code? In gdb? BTW, be careful using openocd through telnet while gdb is connected. Openocd assumes exclusive use of one or the other, and does things like clear all existing breakpoints when gdb connects. It could be confusing the issue for you. > 2. Breakpoints don't work. I can set one in either the telnet client > or gdb without any error message. But when I continue execution, the > program never halts at the breakpoint. Have you tried hardware breakpoints as well? In the past there have been cache-related problems with software breakpoints on some less mature architectures, though I can't speak for arm7_9. And of course sw breakpoints can't be used in some circumstances; e.g. code running from ROM or not currently mapped into virtual memory (e.g. Linux kernel modules). You should be able to verify that the software breakpoint was succesfully inserted by doing a 'mdw' or 'arm disassemble' at the breakpoint address and seeing if the original instruction was replaced with the bkpt instruction. Hope this helps, Mike _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
