[8u] RFR: Revert parts of x86_64.ad to 8u upstream state

[Roman Kennke]

Sure, go!

Roman


> Continuing to scrub the differences against 8u upstream. These are the funny bits in x86_64.ad:
>   https://builds.shipilev.net/patch-openjdk-jdk8-aarch64/hotspot/src/cpu/x86/vm/x86_64.ad.sdiff.html
> 
> The first block got there via backport that reverts x86_64.ad to "upstream" state. But that state is
> about jdk/jdk, not about jdk8u, where those white-space differences are actually present:
>   http://hg.openjdk.java.net/shenandoah/jdk8/hotspot/rev/3f325d1f67d3
> 
> The second block is the leftover from two successive backports. The last one removed the only
> matcher that used immU8, and so we don't need it. See:
>   http://hg.openjdk.java.net/shenandoah/jdk8/hotspot/rev/ec5c2ccc5dbf
>   http://hg.openjdk.java.net/shenandoah/jdk8/hotspot/rev/3b015c5e7a5e
> 
> sh/jdk8 patch:
> 
> diff -r 16898268d728 src/cpu/x86/vm/x86_64.ad
> --- a/src/cpu/x86/vm/x86_64.ad  Thu Jul 18 18:22:24 2019 +0200
> +++ b/src/cpu/x86/vm/x86_64.ad  Fri Jul 19 10:40:25 2019 +0200
> @@ -213,5 +213,5 @@
>  reg_class ptr_reg_with_rbp(RAX, RAX_H,
>                             RDX, RDX_H,
> -                           RBP, RBP_H,
> +                           RBP, RBP_H,
>                             RDI, RDI_H,
>                             RSI, RSI_H,
> @@ -810,5 +810,5 @@
>    __ bind(done);
>  }
> -
> +
> 
>  //=============================================================================
> @@ -2940,14 +2940,4 @@
>  %}
> 
> -operand immU8()
> -%{
> -  predicate((0 <= n->get_int()) && (n->get_int() <= 255));
> -  match(ConI);
> -
> -  op_cost(5);
> -  format %{ %}
> -  interface(CONST_INTER);
> -%}
> -
>  operand immI16()
>  %{
> 
> Testing: hotspot_gc_shenandoah
>