For those of you who were at the meeting Sunday and heard me ramble on about my furnace issues...
I think I figured it out. The gumstix could start a new i2c transaction so soon after the previous i2c transaction ended, that the r8c's didn't have enough time to properly close the old transaction and be ready for the new one, even when they were for different chips. I added a 50uS delay in the gumstix after each i2c transaction and it seems to have solved all the i2c problems. I also moved the serial transmit task from an interrupt handler to the main loop. The old handler occasionally got confused and stopped transmitting, but I don't know if that was due to other problems. It also tried to be fancy about detecting the end of physical transmission vs just an empty queue (the lcd's have some timing constraints) which I don't worry about any more, I just added an extra mS to the appropriate delays. Anyway, it's working fine now so I'm not going to change it back ;-) _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
