RFR[15] JDK-8243114: Implement montgomery{Multiply, Square}intrinsics on Windows

[Simon Tooke]

On 2020-06-24 4:25 p.m., Andrew Hughes wrote:
> On 22/06/2020 10:00, Andrew Haley wrote:
>> On 18/06/2020 19:33, Martin Balao wrote:
> snip...
>
>>>    * L3724
>>>     * The last argument of 'sub' has type 'int', while in the not-Windows
>>> variant is a long. Can we align this?
>> We should do that, yes. Better it be long everywhere.

Sorry, I should have responded to this sooner.

'long' is 32 bits on Windows x64, and 64 bits on other platforms.

sub() is only ever called with an 'int', so this change would actually 
cause the windows code to move to a jlong (i.e. int64_t), if we're 
talking about function signature.

If we're only talking about aesthetics, the type 'long' works, but it's 
somewhat misleading and doesn't actually change the code

>>
> Patch looks good to me too.
>
> Simon, if you are happy with changing the final argument of sub to long
> on the Windows side, I'll sponsor this patch, make that minor change and
> push:
I'm fine with this change, and thank you.
>
> https://cr.openjdk.java.net/~stooke/webrevs/jdk-8243114-jdk/01/01/jdk.patch
>
> to jdk/jdk
>
> Thanks,