[8] RFR: 2020-05-18, Bulk backports to sh/jdk8

[Aleksey Shipilev]

On 5/19/20 2:32 PM, Zhengyu Gu wrote:
> Nit:
> 
> 1)
> 130 void ShenandoahPhaseTimings::set_cycle_data(Phase phase, double time) {
>   131 #ifdef ASSERT
>   132   double d = _cycle_data[phase];
>   133   assert(d == uninitialized(), err_msg("Should not be set yet: %s, 
> current value: %lf", phase_name(phase), d));
>   134 #endif
>   135   _cycle_data[phase] = time;
>   136 }
>   137
>   138
> 
> Double empty lines

Fixed.

> 2)
> 106 void 
> ShenandoahStringDedup::parallel_oops_do(ShenandoahPhaseTimings::Phase 
> phase, OopClosure* cl) {
>   107   _queues->parallel_oops_do(cl);
>   108   _table->parallel_oops_do(cl);
>   109   _thread->parallel_oops_do(cl);
> 
> 
> Timing for queue and table are added in a separate backport, but seems 
> to miss for thread. It is a 8u specific.

Yeah. I'll go and amend this one:
  https://mail.openjdk.java.net/pipermail/shenandoah-dev/2020-May/012219.html

> Otherwise, looks good.
Thanks, I'll push this one.

-- 
Thanks,
-Aleksey