Arm-USB-TINY + Olimex SAM7-P256 - can not halt processor

Running a Debian + latest openOCD 0.6 (though also tested with 0.5) I have found myself to be unable to halt my processor.

My configuration is as follow :

openocd.cfg :

telnet_port 4444
gdb_port 3333

jtag_khz 3000

script interface/olimex-jtag-tiny.cfg 
script  board/atmel_at91sam7s-ek.cfg

Starting openOCD gives me :

Open On-Chip Debugger 0.6.0-dev-00194-ge875e32 (2011-11-04-21:54)
Licensed under GNU GPL v2
For bug reports, read
        http://openocd.sourceforge.net/doc/doxygen/bugs.html
3000 kHz
Info : only one transport option; autoselect 'jtag'
srst_only srst_pulls_trst srst_gates_jtag srst_open_drain
Warn : use 'at91sam7s256.cpu' as target identifier, not '0'
Info : clock speed 3000 kHz
Info : JTAG tap: at91sam7s256.cpu tap/device found: 0x3f0f0f0f (mfg: 0x787, part: 0xf0f0, ver: 0x3)
Info : Embedded ICE version 1
Info : at91sam7s256.cpu: hardware has 2 breakpoint/watchpoint units

Performing the “halt” command in a telnet 4444 session gives openOCD log information

Info : accepting 'telnet' connection from 4444
Info : Halt timed out, wake up GDB.
Error: timed out while waiting for target halted
in procedure 'halt'

A “poll” command confirms that the system is still considered running from openOCD log information

background polling: on
TAP: at91sam7s256.cpu (enabled)
target state: running

I have verified from Sam-Ba (Sam_I_Am) that the unit is in fact working (I can program and run software).

Any ideas of what im wrong?

Going from jtag_khz 3000 to 1500 gave me back control.

So the problem is solved :slight_smile:

Unfortunately the victory was short lived.

I periodicly still get the “Halt timed out, wake up GDB” problem.

So the problem is not gone.