This is the mail archive of the newlib@sourceware.org mailing list for the newlib project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Compiling newlib for the msp430x architecture


Hello,

I failed to compile newlib for with the msp430x architecture with 92kB Flash.

Trying to compile newlib (in my case for the MSP430F2617) with the large memory model (20bit pointers) failed for me because the Make-System automatically injected the -mcpu=msp430 into the gcc flags. This gives me the following warning:

> ../../../../../../newlib/libc/argz/dummy.c:1:0: warning: MCU 'msp430f2617' supports 430X ISA but -mcpu option is set to 430

because in order to get the large memory addressing instructions you need to set -mlarge which requires the -mmcu=msp430f2617 option.

I tried to debug this as i thought it was my fault but it turns out that some Makefile extracts the -mcpu=msp430 option from the

$ msp430-elf-gcc /* couple of -L and -B's */ --print-multi-lib

command. According to the GCC manpage this option is deprecated and -mmcu= should be used instead.

I'm not sure if I am missing something here or if this is really a bug in the make system which is why I am asking here.

Thanks, Stephan


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]