Jason Thorpe wrote:
> 
> On Mar 15, 2007, at 11:26 AM, matthew green wrote:
> 
>>   Anyway, I've decided to stick with "cas" as "compare-and-store".
>>
>> i really wish you wouldn't.  as an example, google gives:
>>
>>    Results 1 - 10 of about 19,400 for "compare and swap" cas
>>    Results 1 - 10 of about 50 for "compare and store" cas
> 
> And how many "compare-and-swap" functions return the old value?  Many  
> of them just return true or false.

I was just wondering, is the returned value even meaningful? I mean, 
either it returns the same value as what is compared against, in which 
case a swap has been performed, or it returns something else. That 
something else, is that ever possible to use in some meaningful way? I 
mean, it's a copy of a value that very well could have changed before 
you even have a chance to act upon it.

However, for code already using a cas then, I would guess that they 
check for a return value of ==cmp or !=cmp. So why not have your cas 
function return that, Jason? That should then not break anything, and 
that can be implemented internally as a compare_and_store with a simple 
wrapper to return cmp or ~cmp.

Or is some code actually using the return value in some more clever way 
in case it's not ==cmp.

	Johnny