RFR: 8371450: AES performance improvements for key schedule generation [v4]

[Shawn M Emery]

On Sat, 8 Nov 2025 17:07:55 GMT, Jamil Nimeh <jnimeh at openjdk.org> wrote:

>> I believe my original changes here utilize a "MergeStore" technique that the compiler optimizes.  I've asked @minborg to see if I got this right.  To verify the optimization here, I used the separate local int variable technique and saw a 0.7% decrease in benchmark performance.
>
> Well, your experiment answers my question.  Unless @minborg says otherwise I'm fine with what you have.  You're already seeing improvements so that's great.

At first look, @minborg believes that "MergeStore" is not being activated here because the xors are being performed at the "element level".  So the speedup is more likely to come from eliminating the temp object and the extra instructions for the assignments and array copy.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/28188#discussion_r2512826553