Peter A. Bigot - 2013-11-25

This might be a justifiable change. On the other hand, the mspgcc manual is mostly of historical interest, and much of its advice is no longer relevant. AFAIK normal integer promotion and optimization will naturally eliminate any performance advantages to explicitly using word-sized types in the current 4.7-based release. I'd also want to be very careful, since unsigned char (or uint8_t) will become a (signed) int under arithmetic conversion rules[*]; using a larger-typed alias might introduce behavioral differences.

[*] at least in C++11 which is the only standard I've recently reviewed in this area.