Makes sense. So if the bits are adjusted, I guess the code is good to go?

Regards,
Alex

On Mon, Jan 16, 2012 at 1:47 PM, Don Clugston <dclugston at googlemail.com> wrote:
> On 15 January 2012 23:11, Alex <xtzgzorex at gmail.com> wrote:
>> There seems to be commented out support for detecting AVX already: https://github.com/D-Programming-Language/druntime/blob/master/src/core/cpuid.d#L275
>>
>> Any reason this isn't enabled?
>
> I actually wrote that code before any AVX processors had ever been released. So it was entirely speculative.
>
>>
>> Regards,
>> Alex
>>
>> On Sun, Jan 15, 2012 at 11:08 PM, Alex <xtzgzorex at gmail.com> wrote:
>>> For what it's worth, bit 28 of ECX indicates whether the CPU has AVX
>>> (for the cpuid instruction).
>>>
>>> Regards,
>>> Alex
>>>
>>> On Sun, Jan 15, 2012 at 11:04 PM, Walter Bright <walter at digitalmars.com> wrote:
>>>> http://en.wikipedia.org/wiki/Advanced_Vector_Extensions#CPUs_with_AVX
>>>> _______________________________________________
>>>> D-runtime mailing list
>>>> D-runtime at puremagic.com
>>>> http://lists.puremagic.com/mailman/listinfo/d-runtime
>> _______________________________________________
>> D-runtime mailing list
>> D-runtime at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/d-runtime
> _______________________________________________
> D-runtime mailing list
> D-runtime at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/d-runtime