On Tue, Aug 26, 2008 at 12:21 PM, Øyvind Harboe <[EMAIL PROTECTED]> wrote: > On Tue, Aug 26, 2008 at 12:09 PM, Valentin Longchamp > <[EMAIL PROTECTED]> wrote: >> On Tuesday 26 August 2008 08.34:16 you wrote: >>> On Mon, Aug 25, 2008 at 11:00 PM, Valentin Longchamp >>> >>> I have noticed that the iMX31 is *VERY* sensitive on the precise >>> implementation of the JTAG interface. >> >> I have tried it with the command ordering you proposed above and I have >> experienced the same results as before. >> >>> >>> It didn't work w/the parport when we tried it. >>> >>> On our ZY1000 reset halt works though(which uses OpenOCD): >>> >>> http://www.zylin.com/zy1000.html >>> >>> I'm not convinced that the iMX31 will work without RCLK.... >> >> OK. I have noticed working with RealView ICE that mx31ads board JTAG chain is >> not optimal. Do you work with an ADS board Øyvind ? I will try on our board, >> but I only have one available at work, and I usually work on openOCD at home. >
I can now confirm this. I have tried both our prototype version 1 and 2 with openOCD and JTAGkey, and I experience the same results (reset halt works but not stepping) as Øyvind. So the mx31ADS board JTAG interface has something wrong that was giving me bad results before. Now I can go on on the right tracks. But a little bit of guidance would be nice for me to dive into the ARM11 code. Regards Valentin _______________________________________________ Openocd-development mailing list [email protected] https://lists.berlios.de/mailman/listinfo/openocd-development
