[vector] RFR : Removed subAll() and few javadoc changes

[Viswanathan, Sandhya]

Hi Yang,

The helper class methods were for future revision consideration and not targeted for the first revision. We are currently preparing to take the API for community review towards the first revision release and are working to get the API in a clean shape and create a formal Javadoc. At some point in near future we hope to create a separate branch for the first revision.  After which we can bring the methods in the helper class back in for future revisions. Please bear with us in the meantime. 

Best Regards,
Sandhya


-----Original Message-----
From: Yang Zhang (Arm Technology China) [mailto:Yang.Zhang at arm.com] 
Sent: Wednesday, April 03, 2019 7:24 PM
To: Viswanathan, Sandhya <sandhya.viswanathan at intel.com>; Kharbas, Kishor <kishor.kharbas at intel.com>; Ningsheng Jian (Arm Technology China) <Ningsheng.Jian at arm.com>; panama-dev at openjdk.java.net
Cc: nd <nd at arm.com>
Subject: RE: [vector] RFR : Removed subAll() and few javadoc changes

Hi Sandhya

OK, thanks for the clarification. But I think this doesn't answer my question why removing them clearly.
>From my point of view, those helper functions are important to make the API completeness, right?
Or are you going to move them (e.g. div, mod, map) into corresponding Vector classes?

Regards,
Yang

-----Original Message-----
From: Viswanathan, Sandhya <sandhya.viswanathan at intel.com>
Sent: Thursday, April 4, 2019 1:42 AM
To: Yang Zhang (Arm Technology China) <Yang.Zhang at arm.com>; Kharbas, Kishor <kishor.kharbas at intel.com>; Ningsheng Jian (Arm Technology China) <Ningsheng.Jian at arm.com>; panama-dev at openjdk.java.net
Cc: nd <nd at arm.com>
Subject: RE: [vector] RFR : Removed subAll() and few javadoc changes

Hi Yang,

Thanks for pointing out the jmh build failure issue.  The helper class were removed as part of preparing for formal API review. 

Best Regards,
Sandhya


-----Original Message-----
From: panama-dev [mailto:panama-dev-bounces at openjdk.java.net] On Behalf Of Yang Zhang (Arm Technology China)
Sent: Tuesday, April 02, 2019 3:37 AM
To: Kharbas, Kishor <kishor.kharbas at intel.com>; Ningsheng Jian (Arm Technology China) <Ningsheng.Jian at arm.com>; panama-dev at openjdk.java.net
Cc: nd <nd at arm.com>
Subject: RE: [vector] RFR : Removed subAll() and few javadoc changes

Hi Kishor

Sorry I missed your latest patch. I find that there is a jmh build failure which results from the change of removing VectorHelper classes. I have created a JBS https://bugs.openjdk.java.net/browse/JDK-8221808. 

I think these VectorHelper classes are useful when someone wants to use some operations that are not intrinsification candidates for now.  To make the api completeness, where should those helper functions go? Do you have other motivations to remove them?

Regards,
Yang


-----Original Message-----
From: Kharbas, Kishor <kishor.kharbas at intel.com>
Sent: Tuesday, March 19, 2019 9:14 AM
To: Ningsheng Jian (Arm Technology China) <Ningsheng.Jian at arm.com>; Yang Zhang (Arm Technology China) <Yang.Zhang at arm.com>; panama-dev at openjdk.java.net
Cc: nd <nd at arm.com>; Kharbas, Kishor <kishor.kharbas at intel.com>
Subject: RE: [vector] RFR : Removed subAll() and few javadoc changes

Hi,

I have a follow on patch making corrections to and adding missing Javadoc. This patch includes the changes in the earlier patch (below).
Summary of changes -
1. Made Javadoc corrections to fix all doclint errors.
2. Remove VectorHelper classes.
3. Remove suball()

Link: http://cr.openjdk.java.net/~kkharbas/vector-api/webrev.javadoc.00

Requesting a review.

Regards,
Kishor

> -----Original Message-----
> From: Ningsheng Jian (Arm Technology China) 
> [mailto:Ningsheng.Jian at arm.com]
> Sent: Tuesday, February 26, 2019 2:38 AM
> To: Kharbas, Kishor <kishor.kharbas at intel.com>; 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] RFR : Removed subAll() and few javadoc changes
> 
> Hi Kishor,
> 
> On 02/26/2019 03:31 AM, Kharbas, Kishor wrote:
> > Hi Yang,
> >
> >> -----Original Message-----
> >> From: Yang Zhang (Arm Technology China) [mailto:Yang.Zhang at arm.com]
> >> Sent: Monday, February 25, 2019 12:44 AM
> >> To: Kharbas, Kishor <kishor.kharbas at intel.com>; panama- 
> >> dev at openjdk.java.net
> >> Subject: RE: [vector] RFR : Removed subAll() and few javadoc 
> >> changes
> >>
> >> Hi Kishor
> >>
> >> Currently I'm implementing subAll() for AArch64.
> >> Could you please help to explain:
> >> 1. What's the motivation of removing subAll() ?
> >
> > As part of API review, we found that subAll() does not have a clear
> semantic meaning. So it was decided to remove it.
> >
> 
> OK, thanks! I assume the "API review" work is public, and then we will 
> try not to miss any specific API change reviews. We all don't want any 
> API to be biased to any particular architecture. :-)
> 
> >> 2. If subAll() should be removed,  why not remove the middle-end 
> >> and the
> >> x86 backend about subAll() ?
> >
> > Yes, the implementation of it should be removed too. I focused on 
> > api
> changes so did not remove the other parts.
> >
> I think it's OK to do that in a separate patch.
> 
> Thanks,
> Ningsheng