On Fri, 27 Mar 2026 20:19:17 GMT, Paul Sandoz <[email protected]> wrote:

>> Emanuel Peter has updated the pull request with a new target base due to a 
>> merge or a rebase. The incremental webrev excludes the unrelated changes 
>> brought in by the merge/rebase. The pull request contains 33 additional 
>> commits since the last revision:
>> 
>>  - Merge branch 'master' into 
>> JDK-8375631-part-number-exception-and-documentation
>>  - more test
>>  - wip part number for unslice
>>  - fix up documentation a little more
>>  - more small updates
>>  - fix small issues
>>  - small fix in test and assert code
>>  - Merge branch 'master' into 
>> JDK-8375631-part-number-exception-and-documentation
>>  - more testing
>>  - cleanup and testReinterpret
>>  - ... and 23 more: https://git.openjdk.org/jdk/compare/3a2b8818...ee08c9ef
>
> src/jdk.incubator.vector/share/classes/jdk/incubator/vector/AbstractVector.java
>  line 573:
> 
>> 571:             return 0 <= part && part < limit;
>> 572:         } else if (limit < 0) {  // insertion (output is logical result 
>> with padding)
>> 573:             return limit < part && part <= 0;
> 
> `return limit <= part ...` ? since you flipped the operands.

No, this is correct. `part` in `[imit+1..0]`. The `limit` is exclusive.

Note: Flipping `a < b` to `b > a` is correct. What may be additionally 
confusing here is that before the `limit` was positive, but now `partLimit` 
returns a negative number, so we drop the `-`.

Maybe you meant something else. Can you clarify?

> test/jdk/jdk/incubator/vector/PartNumberTest.java line 45:
> 
>> 43:     }
>> 44: 
>> 45:     interface TestMethod {
> 
> Annotate with `@FunctionalInterface`.

Done :)

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

PR Review Comment: https://git.openjdk.org/jdk/pull/30113#discussion_r3008275581
PR Review Comment: https://git.openjdk.org/jdk/pull/30113#discussion_r3008284956

Reply via email to