I just committed a fix for the following bugs, probably the most annoying bug that remained since the introduction of the STK500v2 code:
bug #17199: EEPROM fails verification on ATmega645 with pony-stk200 hardware bug #16849: EEPROM write fails for AT90USB1287 with mode 0x41 bug #15146: stk500v2_paged_write: loadpage instruction not defined for part The fix was to add loadpage and writepage instructions to the EEPROM sections of all AVRs that have an EEPROM organized in pages (well, a *really* boring task), and to prevent avr_write_byte_default() from trying to use paged memory access unless the memory is marked "paged". Please everybody who is able to build from CVS, verify this fixes all outstanding issues. EEPROM write should now be possible for all AVRs, regardless of whether using the -U eeprom:w:... option or from within terminal mode. The only restriction is that both, high-voltage programming as well as JTAG page programming (i. e. the -U option when applied to a JTAG ICE) are not performing a memory erase on EEPROM before programming, so they can only program bits from logic `1' to logic `0'. Thanks to everybody reporting these bugs, and in particular to Bob Paddock who analyzed the problem, so it's been easy to implement the workaround. -- cheers, J"org .-.-. --... ...-- -.. . DL8DTL http://www.sax.de/~joerg/ NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) _______________________________________________ avrdude-dev mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/avrdude-dev
