RFR: Aarch64 cmpxchg_oop() fix

Aleksey Shipilev shade at redhat.com
Wed Apr 3 07:54:22 UTC 2019


On 4/2/19 11:11 PM, Roman Kennke wrote:
> Aleksey spotted a bug that we apparently introduced with LRB. Let's
> revert this little part:
> 
> diff -r c5a159a6b36d
> src/hotspot/cpu/aarch64/gc/shenandoah/shenandoahBarrierSetAssembler_aarch64.cpp
> ---
> a/src/hotspot/cpu/aarch64/gc/shenandoah/shenandoahBarrierSetAssembler_aarch64.cpp
> Mon Apr 01 13:33:58 2019 +0200
> +++
> b/src/hotspot/cpu/aarch64/gc/shenandoah/shenandoahBarrierSetAssembler_aarch64.cpp
> Tue Apr 02 23:09:55 2019 +0200
> @@ -414,7 +414,9 @@
>    }
>    __ bind(done);
> 
> -  if (!is_cae) {
> +  if (is_cae) {
> +    __ mov(result, tmp1);
> +  } else {
>      __ cset(result, Assembler::EQ);
>    }
>  }
> 
> Testing: hotspot_gc_shenandoah on aarch64

OK, looks good. Was any test failing because of this?

-Aleksey



More information about the shenandoah-dev mailing list