[PATCH] Define CLOCKS_PER_SEC type to the type clock_t

Paul Eggert eggert@cs.ucla.edu
Tue Jan 6 15:58:00 GMT 2015


On 01/06/2015 06:22 AM, Andreas Schwab wrote:
>> This fails only at runtime if the value is wrong, which is not as good as
>> >checking it at compile-time.
> Does it matter?

Not so much here, no, since the constant is always 1 million and the 
assert is optimized away.  Still, we should prefer compile-time checking.



More information about the Libc-alpha mailing list