On Mon, 17 Nov 2025 15:03:15 GMT, Jorn Vernee <[email protected]> wrote:
>> Harald Eilertsen has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Fix calculation of allocationSize when byteAlignment < MAX_MALLOC_ALIGN
>>
>> This work was sponsored by: The FreeBSD Foundation
>
> src/java.base/share/classes/jdk/internal/foreign/SegmentFactories.java line
> 207:
>
>> 205: long result;
>> 206: if (byteAlignment > MAX_MALLOC_ALIGN || alignedSize %
>> byteAlignment != 0) {
>> 207: allocationSize = alignedSize + byteAlignment -
>> Math.min(MAX_MALLOC_ALIGN, alignedSize);
>
> This doesn't look correct to me. Let's say we want to do a 7 byte allocation
> (with `init == false`), aligned to 4. We get `7 % 4 -> 3` so we enter this
> branch, and then `allocationSize = 7 + 4 - 7`, and we end up allocating only
> 4 bytes. The previous computation only works if `MAX_MALLOC_ALIGN` is greater
> than `byteAlignment`.
>
> I think it's simpler to just replace `MAX_MALLOC_ALIGN` in this `if`
> statement with an 'expected alignment' derived from the size.
Maybe something like `expectedAlignment = Math.min(MAX_MALLOC_ALIGN,
alignedSize % byteAlignment == 0 ?alignedSize : 0)` (i.e. don't assume any
alignment for e.g. 7 byte allocations.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/28235#discussion_r2534734059
