On Mar 7, 2012, at 9:06 PM, David Relson wrote: > On the other hand, after changing the "int main(void)" statement to "int > test(void)", compile, and disassemble then the expected "ret" > statements appear. It seems that the "main" function gets special > handling.
Hmm. There was a very similar issue that I ran into recently with the AVR gcc compiler, having to do with "naked" functions not having a proper epilogue, resulting in bad code when jump reordering was done. Or something like that. PR42240: http://gcc.gnu.org/ml/gcc-patches/2011-02/msg01441.html http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42240 It looks like the non-hosted main() in msp430-gcc is similar to "naked", so the problem could be similar. I don't much like their patch, since it disables optimizations that could be important in a function of any particular size. avr-gcc also has attribute ((OS_Main)) which is similar and apparently didn't have the bug. BillW ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ _______________________________________________ Mspgcc-users mailing list Mspgcc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mspgcc-users
