On Thu, 18 Jun 2026 15:07:17 GMT, Ferenc Rakoczi <[email protected]> wrote:
>> An aarch64 implementation of the MontgomeryIntegerPolynomial256.mult() >> method and IntegerPolynomial.conditionalAssign(). Since 64-bit >> multiplication is not supported on Neon and manually performing this >> operation with 32-bit limbs is slower than with GPRs, a hybrid neon/gpr >> approach is used. Neon instructions are used to compute intermediate values >> used in the last two iterations of the main "loop", while the GPRs compute >> the first few iterations. At the method level this improves performance by >> ~9% and at the API level roughly 5%. >> >> >> >> --------- >> - [x] I confirm that I make this contribution in accordance with the >> [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai). > > Ferenc Rakoczi has updated the pull request incrementally with one additional > commit since the last revision: > > Added UseIntPolyIntrinsics to aotCodeCache.hpp > Did you mean the change in commit > [dd7b08b](https://github.com/openjdk/jdk/commit/dd7b08b301a5820776b2b1df24d333301bf9cca4)? > Yes! > What is a GHA run? GitHub Actions. See the red crosses around "Some checks were not successful" at the end of this page. Those are the infrastructure bugs that are already fixed in master. Merge from master to get these built again. Currently, it does not build Windows-AArch64, where interesting bugs might lurk. ------------- PR Comment: https://git.openjdk.org/jdk/pull/30941#issuecomment-4743473874
