Currently if you ask gcc -O2 (tried on 3.0) to evaluate
!a
it will give you code of the form:
cmp r4, #0 movne ip, #0 moveq ip, #1
how does one teach the optimiser that this is equivalent and 33% faster:
rsbs ip, r4, #1 movls ip, #0
Hang on a minute, that's not the same algorithm though. For negative values of r4, you'll get bogus output... 1 - -2 is 3 wheras you need zero. Surely? (ie, !-2 is 0, not 3)
Ian.
