Re: [avr-gcc-list] What Happened to the sbi() and cbi() Macros????

[E. Weddington]

Pertti Kellomäki wrote:

> This is what has been objected to in many posts. I am still
> mystified what metric makes
>
>   SFR &=~mask;
>
> simpler and cleaner than
>
>   cbi(SFR, bit_num);
>
> It is not that I cannot decode the first form. I can still after
> twenty years write RCA 1802 assembly in hex from my head.
> I just prefer "clear this bit in SFR for me" to "negate this bitfield,
> do a bitwise and, and assign the result to SFR".


In practice, when you read that, you don't break it down into individual 
operations. You look at it as a whole and begin to see patterns:

set bit:
v |= m

clear bit:
v &= ~m

toggle bit:
v ^= m

read bit:
v & m

The downside to reading it as a whole, is when you're debugging somebody 
else's code who may not have written it very well, causing strange errors.

However, the upside is that you can then begin to combine them in unique 
ways to achieve a pretty efficient operation:

v = ((v & ~m1) | m2 ^ m3);

YMMV.
Eric