Messages in this thread | | | Date | Sat, 18 Mar 2006 22:57:19 +0100 (MET) | From | Jan Engelhardt <> | Subject | Re: [patch 1/1] consolidate TRUE and FALSE |
| |
> >Isn't there a runtime cost converting all "non-false" values to a unique "true" (i.e. converting non-zero values to one) ? >
Somewhat. The answer is "yes, but depends on usage". If you just write
_Bool x = filthy_action(); if(x)
Then the compiler is smart enough to optimize 'x' away if it is not used somewhere else, therefore we do not pay a price for converting the return type of filthy_action (=int) to a _Bool.
The asm output for storing the result of filthy_action() [requires 'volatile int x' in this small example]
call strxcmp mov [ebp-4], eax
While making that a 'volatile _Bool x' makes it:
call strxcmp test eax, eax setnz al mov [ebp-1], al
Makes me prefer typedef int bool over _Bool.
Jan Engelhardt -- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |