On Thu, May 26, 2011 at 2:06 PM, Mahr, Stefan <stefan.m...@sphairon.com> wrote: >>> Btw.: There is still an endianness issue with mips target. Drasko adds >>> endianness swapping (that I removed two years ago) >> Why did you remove it ? > > I removed swapping in mips_m4k, because for commands like mdw the swapping > was already done in target.c. If a target was selected as big endian, > swapping was done twice. So setting endianness had no effect. I did not > realized that dump_image works different.
What do you concern under "dump_image" ? A command that dumps image ? To be honest, I did not try it for MIPS, but there are some people on the list who did (after my patch was applied, I think). Point is that nothing worked, not only dump_image. Simply told, memory write/read was byte-inversed - so you can say goodbye to the SDRAM configuration, to start with. We were not even capable to load image, far away from dumping it... > Since you revert this change, same issue again :) It's not an issue, it's a feature ;). Now that we come again to this point, could it be that MIPS JTAGs sometimes take microcode in big endian ? I have noticed that one implementation works for some people, and represents "bytes flipped" issue for other... Stefan, if you wold like to go further investigating the issue, maybe you can take look at small doc I wrote about MIPS EJTAG : http://www.mail-archive.com/openocd-development@lists.berlios.de/msg16208.html. It might be helpful. If you have some comments, they are very welcome. I hope that I'll have some more time to craft it well and commit it upstream one day :). BR, Drasko _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
