I am using avr-gcc 2.4.0 20061007 When I compile for the atmega324 I get
avr-gcc -g -Wall -O2 -mmcu=atmega324 -I /usr/include -I/opt/avr/avr/include -W1,-Map,main.map -o main.elf main.c -lm unknown MCU 'atmega324' specified *then a list of all the known MCU names with one name per line including atmega324p(not atmega324)* main.c:1: error: MCU 'atmega324' supported for assembler only When I comple for the atmega324p gcc just coughs up that it doesnt recognize the 324p at all. avr-gcc -g -Wall -O2 -mmcu=atmega324 -I /usr/include -I/opt/avr/avr/include -W1,-Map,main.map -o main.elf main.c -lm Known MCU names: *here a bunch of names with 7 to 10 names on a line(including the atmega324(not 324p)* Assembler messages: Fatal error: unknown MCU: atmega324p It looks like binutils was updated to have just the 324 where gcc is still looking for the 324p I just checked and I have regular 324 here. Free samples are a wonderful thing. Thanks, Ray Moore On 1/25/07, Joerg Wunsch <[EMAIL PROTECTED]> wrote:
As Raymond Moore wrote: > Probably related to this P/nonP thing. I have recently acquired > some atmega324 chips and when attempting to compile for them, GCC > and binutils differ as to whether it is atmega324 or atmega324p... You are not supposed to be able to get your hands on non-P ATmega324 silicon, so always pick -mmcu=atmega324p for them. In fact, avr-libc cannot handle anything but the `P' version anyway, as the non-P version has been discontinued by Atmel in its infant days. -- cheers, J"org .-.-. --... ...-- -.. . DL8DTL http://www.sax.de/~joerg/ NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)
_______________________________________________ AVR-libc-dev mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/avr-libc-dev
