[jdk17u-dev] RFR: 8298472: AArch64: Detect Ampere-1 and Ampere-1A CPUs and set default options

Goetz Lindenmaier goetz at openjdk.org
Mon Jan 16 08:34:25 UTC 2023


On Sun, 15 Jan 2023 14:31:32 GMT, Patrick Zhang <qpzhang at openjdk.org> wrote:

> Clean backport of CPU detection function for Ampere-1 and Ampere-1A, and performance enhancement with turning on UseSIMDForMemoryOps option for these two CPUs by default. Verified OpenJDK bundled JMH test cases, `make run-test TEST="micro:java.lang.ArrayCopy*"` , SIMD vs NoSIMD ratios are mostly positive as expected, aligned with that on jdk mainline. Low risk to other functions/systems due to the CPU detection logic.
> 
> Additional tests done:
> 1. Linux aarch64 release tier1 on Ampere-1 system, without explicitly specifying UseSIMDForMemoryOps 
> 2. OpenJDK GHA Sanity Checks on forked repo: https://github.com/cnqpzhang/jdk17u-dev/actions/runs/3919027362
> 
> Backport-of: 706845743699efb01994e2d12c65023a3e972b77
> Signed-off-by: Patrick Zhang <patrick at os.amperecomputing.com>

Hi @cnqpzhang 
This is obviously not a clean backport. I get 3!!!! conflicts when I patch the change to 17.
Please correct your description above and describe what you resolved. Then get a review.
After all this, you can tag the change again.
Thansk.

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

PR: https://git.openjdk.org/jdk17u-dev/pull/1067


More information about the jdk-updates-dev mailing list