Application failure with traversal

[Roman Kennke]

Am 09.02.2018 um 18:36 schrieb Aleksey Shipilev:
> On 02/09/2018 06:27 PM, Lennart Börjeson wrote:
>>>   c) C2: -XX:-TieredCompilation
>>
>> Fails.
>
> OK, both good and bad! And that is without any GC cycle happening, right? If so, that must
> definitely point at something gone wrong in C2, which means we can use passive mode, and gradually
> narrow down which part had failed (in order of most probability):
>
>   0) Run with -XX:-TieredCompilation -XX:ShenandoahGCHeuristics=passive
>   1) Add -XX:+ShenandoahBarriersForConst
>   2) Add -XX:+ShenandoahAsmWB
>   3) Add -XX:+ShenandoahReadBarrier
>   4) Add -XX:+ShenandoahWriteBarrier
>   5) Add -XX:-ShenandoahWriteBarrierRB
>   6) Add -XX:+ShenandoahStoreValEnqueueBarrier
>   7) Add -XX:+ShenandoahCASBarrier
>   8) Add -XX:+ShenandoahAcmpBarrier
>   9) Add -XX:+ShenandoahCloneBarrier

Given it's traversal, I'd suspect -XX:+ShenandoahStoreValEnqueueBarrier
as most likely.

Any chance this could be extracted into a testcase?

Roman