Am 21.03.2018 um 22:31 schrieb Erlo Haugen: > Or maybe allow for choosing between the standard compliant way and the > __bit way by means if a flag or compiler directive. > I do realize that linking modules with different implementation of bool > is bound to be problematic or even fail, but I'm not sure if that will > be a problem in real life.
The optimization introduced yesterday that decides for each local bool variable individually seems to work well enough (though naturally it is not as good as the new register allocator could be, but porting that one to mcs51 would have been far more effort). Dhrystone code size is now down to slightly below 3.6.0-levels, which also means it is at at least a 2-year-low (I currently don't have earlier data). Whetstone code size (which had also regressed slightly from 3.6.0 to 3.7.0) looks even better. Philipp ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Sdcc-user mailing list Sdcc-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sdcc-user
