CR for RFR 8142980

[Vladimir Ivanov]

No problem, I'm fine with that.

Best regards,
Vladimir Ivanov

On 11/20/15 12:58 AM, Berg, Michael C wrote:
> Vladimir, since I have folks requesting more additions anyways, I will clean them up on the next round.
>
> Regards,
> Michael
>
> -----Original Message-----
> From: Vladimir Ivanov [mailto:vladimir.x.ivanov at oracle.com]
> Sent: Thursday, November 19, 2015 12:38 PM
> To: Berg, Michael C; 'hotspot-compiler-dev at openjdk.java.net'
> Subject: Re: CR for RFR 8142980
>
> I'm curious why some of CPU capability checks are 32-bit only, e.g.:
> +  NOT_LP64(assert(VM_Version::supports_evex(), ""));
>
> +  NOT_LP64(assert(VM_Version::supports_avx512dq(), ""));
>
> Best regards,
> Vladimir Ivanov
>
> On 11/13/15 11:50 PM, Berg, Michael C wrote:
>> Hi Folks,
>>
>> I would like to contribute bug fixes for AVX 512 which address 64-bit
>> derby on EVEX enabled targets as well as KNL support and instrinsic
>> support for AVX512.
>>
>> These changes are also needed for a follow on changeset which
>> introduces the use of these additions in the AVX512 instrinsic code.
>>
>> Bug-id: https://bugs.openjdk.java.net/browse/JDK-8142980
>>
>>
>> webrev:
>>
>> http://cr.openjdk.java.net/~mcberg/8142980/webrev.01/
>>
>> The derby metric fails to run on 64-bit SKX and KNL EVEX enabled
>> platforms. Also code support is needed for AVX512 intrinsics and
>> include KNL support.
>>
>> Thanks,
>>
>> Michael
>>