RFR: 8371637: allocateNativeInternal sometimes return incorrectly aligned memory [v6]

[Harald Eilertsen]

On Thu, 20 Nov 2025 11:02:32 GMT, Almaz <duke at openjdk.org> wrote:

>> I think the proposed simpler patch makes sense. Basically, it artificially expands the allocation size to match the alignment, which is probably not going to cause any side-effect as that is likely what was happening anyway...
>
> @mcimadamore:
> 
> Question regarding `MAX_MALLOC_ALIGN = Unsafe.ADDRESS_SIZE == 4 ? 8 : 16;`
> 
> Is 16 here just a heuristic based on known malloc implementations or is it backed by something? I can't find anything regarding that in the libc standard. 
> 
> Not related to this particular PR: should `MAX_MALLOC_ALIGN` be `MIN_MALLOC_ALIGN` instead? Larger allocations might be page size aligned.

> Should I also add [the test](https://github.com/openjdk/jdk/pull/28235#issuecomment-3532821794) suggested by @Yqwed before we conclude this PR, or is it better to leave it for a later one?

I've tested it a bit, but I'm not sure it provides any extra value. It seems to only trigger on the same condition as the existing `testActualByteAlignment` and `testAlignedAccess` tests in `TestMemoryAlignment.java`..

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

PR Comment: https://git.openjdk.org/jdk/pull/28235#issuecomment-3557849641