hg: jdk/submit: 8217561: X86: Add floating-point Math.min/max intrinsics (reduction pattern)

[bsrbnd at gmail.com]

Changeset: 9aa98249f99c
Author:    bsrbnd
Date:      2019-03-04 21:22 +0100
URL:       http://hg.openjdk.java.net/jdk/submit/rev/9aa98249f99c

8217561: X86: Add floating-point Math.min/max intrinsics (reduction pattern)
Summary: Implementation taking care of +/-0.0 and NaN which uses a specific pattern for reductions
Reviewed-by: aph, kvn, neliasso, sviswanathan, adinn
Contributed-by: Jatin Bhateja <jatin.bhateja at intel.com>

! src/hotspot/cpu/x86/macroAssembler_x86.hpp
! src/hotspot/cpu/x86/x86_64.ad
! src/hotspot/share/opto/library_call.cpp
! src/hotspot/share/opto/loopTransform.cpp
! test/hotspot/jtreg/compiler/intrinsics/math/TestFpMinMaxIntrinsics.java