RFR: 8365017: The SegmentBulkOperations::copy method can be improved using overlaps [v6]

[Emanuel Peter]

On Mon, 11 Aug 2025 16:31:17 GMT, Per Minborg <pminborg at openjdk.org> wrote:

>> This PR proposes to use overlapping memory areas in `SegmentBulkOperations::copy`, similar to what is proposed for `SegmentBulkOperations::fill` in https://github.com/openjdk/jdk/pull/25383.
>> 
>> This PR passes `tier1`, `tier2`, and `tier3`testing on multiple platforms.
>
> Per Minborg has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Fix comment

src/java.base/share/classes/jdk/internal/foreign/AbstractMemorySegmentImpl.java line 274:

> 272:             // A long value that is less than zero has it's 63:th bit set and so,
> 273:             // we can just AND the expressions (and the sign bit 63)
> 274:             return (int) (((thisStart - thatEnd) & (thatStart - thisEnd)));  // overlap occurs -> negative value

Why not move the branch to here? Is there really a good reason why you move it to the call-site?
That way, you could save the comment:
`// Returns a negative value if the regions overlap, otherwise a non-negative value.`
Also: I see that your conditions at the call-site are not consistent with the current semantics. You have:
`if (overlaps(that) >= 0) {`
and
`src.overlaps(dst) == 0`
The second seems wrong to me - do you agree?

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

PR Review Comment: https://git.openjdk.org/jdk/pull/26672#discussion_r2270199905