RFR 8231293: Shenandoah: Traversal fails JVMTI tests
zgu at redhat.com
Mon Sep 23 14:27:55 UTC 2019
You may have a point, I wonder if it is an upstream bug.
JvmtiTagMap::do_weak_oops() -> entry->object_peek() triggers native load
barrier, while here should probably be a RawAccess.
On 9/23/19 10:00 AM, Roman Kennke wrote:
> Why is that a problem?
> As far as I know, we don't concurrently clean native roots with
> traversal GC. Or do we?
>> The problem is caused by LRB trying to evacuate dead oops.
>> At this point, traversal is completed, therefore, no more evacuation
>> should happen. We should complete the marking bitmap and arm native
>> barrier to prevent from-space oops from leaking to system.
>> There are duplicated works among fixup roots and parallel cleaning,
>> let's address them in DKJDK-8231324.
>> Bug: https://bugs.openjdk.java.net/browse/JDK-8231293
>> Webrev: http://cr.openjdk.java.net/~zgu/JDK-8231293/webrev.00/index.html
>> hotspot_gc_shenandoah (fastdebug and release)
>> vmTestbase/nsk/jvmti (fastdebug and release) normal and traversal
More information about the hotspot-gc-dev