RFR: 8289552: Make intrinsic conversions between bit representations of half precision values and floats

[Joe Darcy]

On Fri, 5 Aug 2022 22:52:10 GMT, Smita Kamath <svkamath at openjdk.org> wrote:

>> src/hotspot/share/runtime/sharedRuntime.cpp line 450:
>> 
>>> 448: JRT_END
>>> 449: 
>>> 450: JRT_LEAF(jshort, SharedRuntime::f2hf(jfloat  x))
>> 
>> Is it necessary/desirable to have a transliteration of the Java implementation here rather than just using the Java library implementation?
>
> @jddarcy Thanks for your comment. I am not sure if there is a way of using Java library implementation here.

I was under the impression that if a platform didn't have special support for the functionality in question it could not have the intrinsic and just use the Java library implementation?

-------------

PR: https://git.openjdk.org/jdk/pull/9781