On Tue, 24 Feb 2026 05:06:12 GMT, Jatin Bhateja <[email protected]> wrote:

>> As per https://github.com/openjdk/jdk/pull/28002#issuecomment-3789507594 on 
>> JDK-8370691 pull request, splitting out portion of PR#28002 into a separate 
>> patch.
>> 
>> With Float16 we need to differentiate b/w vector lane element type and 
>> carrier type. while element type corresponds to exact type of lane i.e. 
>> Float16.class, carrier type corresponds to the storage type required to 
>> propagate IEEE 754 float16 values to inline expander.
>> 
>> Patch adds a new carrier type to existing LaneType structure and use it for 
>> following purpose.
>>  - To compute preferred shape of vector.
>>  - To allocate vector backing storage during dummy vector and iota array 
>> creation.
>>  - Sanity checking to ensure vector lane type is a primitive type in inline 
>> expander.
>> 
>> All existing Vector API jtreg test are passing with the patch.
>> 
>> Kindly review and share your feedback.
>> 
>> Best Regards,
>> Jatin
>
> Jatin Bhateja has updated the pull request incrementally with one additional 
> commit since the last revision:
> 
>   Review comments resolution

Changes requested by liach (Reviewer).

Sure, the C++ code changes look trivially fine.

src/hotspot/share/opto/vectorIntrinsics.cpp line 46:

> 44: 
> 45:   ciField* fd1 = ik->get_field_by_name(ciSymbols::CTYPE_name(), 
> ciSymbols::class_signature(), /* is_static */ true);
> 46:   assert(fd1 != nullptr, "element type info is missing");

We probably need separate asserts for CTYPE and ETYPE now, given these asserts 
test all fields. Below assert comment should get an update too.

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

PR Review: https://git.openjdk.org/jdk/pull/29629#pullrequestreview-3854316943
PR Comment: https://git.openjdk.org/jdk/pull/29629#issuecomment-3959290124
PR Review Comment: https://git.openjdk.org/jdk/pull/29629#discussion_r2853092193

Reply via email to