[aarch64-port-dev ] [PATCH] 8217561 : X86: Add floating-point Math.min/max intrinsics, approval request

Andrew Haley aph at redhat.com
Thu Feb 28 11:32:54 UTC 2019

On 2/28/19 6:45 AM, Pengfei Li (Arm Technology China) wrote:
> But for an already (almost) sorted array, fmin/fmax instructions do make the performance worse, but not too much.

I have seen a 3x performance degradation on a min reduction over an array.

Andrew Haley
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671

More information about the hotspot-compiler-dev mailing list