> On Nov 15, 2013, at 4:08 PM, Ben Pfaff <b...@nicira.com> wrote:
>
>> On Fri, Nov 15, 2013 at 04:03:32PM -0800, Jarno Rajahalme wrote:
>>
>>>> On Nov 15, 2013, at 3:48 PM, Ben Pfaff <b...@nicira.com> wrote:
>>>>
>>>> On Wed, Nov 13, 2013 at 01:32:42PM -0800, Jarno Rajahalme wrote:
>>>> Add raw_ctz64(), ctz64(), and popcount64() using builtins when
>>>> available.
>>>>
>>>> I'm not sure if the "UINT64_MAX == ~0UL" and "UINT64_MAX == ~0ULL"
>>>> work in all cases as I imagine they would.
>>>
>>> I think you could use ULONG_MAX and ULLONG_MAX, any reason not to?
>>
>> It doesn't matter if they are the same in some cases, right?
>
> Actually, I guess we can just skip the test entirely. GCC always
> supports 64-bit integers as unsigned long long, so this by itself
> ought to do the trick:
>
> #elif __GNUC__ >= 4
> static inline int
> raw_ctz64(uint64_t n)
> {
> return __builtin_ctzll(n);
> }
> static inline int
> popcount64(uint64_t n)
> {
> return __builtin_popcountll(n);
> }
Thanks Ben, I'll fold this in.
Jarno
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
