On Wed, 1 Apr 2026 05:11:07 GMT, Emanuel Peter <[email protected]> wrote:
>> src/jdk.incubator.vector/share/classes/jdk/incubator/vector/Vector.java line
>> 1039:
>>
>>> 1037: * <li> Bitwise Reinterpretation:
>>> 1038: * {@code f(X)} is the bitwise image of {@code X} of the same vector
>>> 1039: * bit size, only the lane boundaries are redrawn from {@code ETYPE}
>>
>> Suggestion:
>>
>> * {@code f(X)} is the bitwise image of {@code X} of the same shape,
>> * only the lane boundaries are redrawn from {@code ETYPE}
>>
>> ?
>> In that respect I think we may have some incorrect entries for rows of
>> `reinterpretShape` and the logical result, which should be of the same shape.
>
> I think this wording issue actually zipped through my head while trying to
> fall asleep yesterday. Thanks for pointing it out.
> In that respect I think we may have some incorrect entries for rows of
> reinterpretShape and the logical result, which should be of the same shape.
Ah wow, you are right! I think I'll adjust the examples in the table. Because I
think it really does make more sense to order the operation "shape-invariant
reinterpret first, then insert/select". That keeps it in sync with the
conversion operation.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/30113#discussion_r3019808908
