[vector api] RFR: Implement vector max and min for AArch64 NEON

[Vladimir Ivanov]

Thanks, pushed [1]

Best regards,
Vladimir Ivanov

[1] http://hg.openjdk.java.net/panama/dev/rev/dd9e5f49b77a

On 07/11/2018 23:05, Yang Zhang (Arm Technology China) wrote:
> Hi Vladimir
> 
> Please check this updated patch:
> http://cr.openjdk.java.net/~yzhang/vectorapi.max_min/webrev.01/
> 
> Regards
> Yang
> 
> -----Original Message-----
> From: Vladimir Ivanov <vladimir.x.ivanov at oracle.com>
> Sent: Thursday, November 8, 2018 10:44 AM
> To: Yang Zhang (Arm Technology China) <Yang.Zhang at arm.com>; panama-dev at openjdk.java.net
> Cc: nd <nd at arm.com>
> Subject: Re: [vector api] RFR: Implement vector max and min for AArch64 NEON
> 
> There are some conflicts with latest vectorIntrinsics branch (rebased today). Please, send me updated patch.
> 
> Best regards,
> Vladimir Ivanov
> 
> On 07/11/2018 18:33, Yang Zhang (Arm Technology China) wrote:
>> Hi Vladimir
>>
>>> I assume NEON is unconditionally supported on AArch64.
>>> No need to adjust Matcher::match_rule_supported[_vector] to make use of those new instructions?
>>
>> Yes, NEON is mandatory feature on AArch64. Usually we don't need to adjust Matcher::match_rule_supported[_vector].
>>
>>> Otherwise, looks fine to me (not an expert in AArch64 port).
>>
>> Thanks for your review.
>>
>> Regards
>> Yang
>>
>>
>> -----Original Message-----
>> From: Vladimir Ivanov <vladimir.x.ivanov at oracle.com>
>> Sent: Thursday, November 8, 2018 3:13 AM
>> To: Yang Zhang (Arm Technology China) <Yang.Zhang at arm.com>;
>> panama-dev at openjdk.java.net
>> Cc: nd <nd at arm.com>
>> Subject: Re: [vector api] RFR: Implement vector max and min for
>> AArch64 NEON
>>
>>
>>
>> On 06/11/2018 23:15, Yang Zhang (Arm Technology China) wrote:
>>> Hi,
>>>
>>> I have a patch which implements vector api max() and min() for AArch64 NEON.
>>> Could you please help to review this?
>>>
>>> Webrev:
>>> http://cr.openjdk.java.net/~yzhang/vectorapi.max_min/webrev.00/
>>
>> I assume NEON is unconditionally supported on AArch64.
>>
>> No need to adjust Matcher::match_rule_supported[_vector] to make use of those new instructions?
>>
>> Otherwise, looks fine to me (not an expert in AArch64 port).
>>
>> Best regards,
>> Vladimir Ivanov
>>