RFR: 8333890: Fatal error in auto-vectorizer with float16 kernel. [v2]

[Emanuel Peter]

On Sat, 6 Jul 2024 23:55:04 GMT, Jatin Bhateja <jbhateja at openjdk.org> wrote:

>> Hi,
>> 
>> This bug fix patch fixes an assertion failure seen while auto-vectorizing conversion chain involving float16 type.
>> Constraining the container type for ConvF2HF IR to short type upfront , as relying on backpropagating narrow
>> type from memory operation  may not be sufficient if IR is succeeded by ConvHF2F as depicted by the include test case.
>> 
>> Best Regards,
>> Jatin
>
> Jatin Bhateja has updated the pull request incrementally with one additional commit since the last revision:
> 
>   IR rule extension for ANY vector size

A further question:
Why is the `basic_type` of `F2HF` a `T_INT` and not `T_SHORT`? Because I see: 

`src/java.base/share/classes/java/lang/Float.java
`

    @IntrinsicCandidate
    public static short floatToFloat16(float f) {


It just seems to me that special-casing here is ad-hoc, and not the best over-all design.

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

PR Comment: https://git.openjdk.org/jdk/pull/20062#issuecomment-2213651503