Hi Sergey, On Wed, Jan 21, 2009 at 4:30 PM, Sergey Lapin <slapi...@gmail.com> wrote: > Hi, all! > > I write support for mc13224 - arm7tdmi-based ieee8021.15.4 -compliant > communication chip. Chip is quite new, documentation is in weird state > and is not full, but core and basic peripherals like SPI are > documented normally. I already managed to make core work, > but I can't manage it to reset properly. >
Can you please send your patches to OpenOCD? Even if it is not working correctly your patch base help other people who wants to test it. > 1. It does not have TRST, only SRST, and it works (resets), but then > OpenOCD is unable to talk to it anymore, full restart of openocd and > board reset button press works to some extent. > > 2. halt works only first time, then the same problem is as with reset. > Startup process and limitations is not described in manual. > > Now I have managed to make GPIO and SPI work via mww, > can write stuff to internal RAM and run it, so basically it works. > > Could anybody advice how to debug/gather information to fix this issue? > > I could provide -d3 log of openocd, but it just tells that chip is in > invalid mode. > (will submit in another email as soon as I get home). > I am also testing Freescale ARM9/11 processors with OpenOCD. Sometimes I got this "invalid mode" error message from OpenOCD and sometimes I got also this warning message: Warn : captured: 0x00 check_value: 0x01 check_mask: 0x0f Warn : in_handler: w/o "in_value", mismatch in SIR > Freescale chose strategy of making very hard to work with this chip, > demanding closed binaries every now and then, and depending on > IAR Workbench for closed libraries, etc, but I'd like to > make it work fully (at least as just controller) hoping to reverse engineer > the rest for proper gcc support, because it is only integrated chip > on market at the moment, and I'd like to experiment with this kind of stuff. > By any chance if anybody had any experience with this particular chip, > have any information, I'd like to know about it. > Please go ahead! It is nice to see open-source alternatives to closed solution. Be sure your work will help more people. > Thanks a lot. > S. Best Regards, Alan _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
