RFR: [10] 2018-06-25: Bulk backports to sh/jdk10

Aleksey Shipilev shade at redhat.com
Mon Jun 25 16:51:36 UTC 2018


http://cr.openjdk.java.net/~shade/shenandoah/backports/jdk10-20180625/webrev.01/index.html

This backports recent performance and bugfixing work to sh/jdk10:

 [backport] SH::make_(tlabs)_parsable() should work correctly with/without TLABs
 [backport] Avoid double-enqueueing store values in runtime
 [backport] Removed racy assertion
 [backport] AlwaysPreTouch fails with non-default ConcGCThreads
 [backport] Workaround traversal+C2 bug with individual loads/stores
 [backport] Improve scheduling and interleaving of SATB processing in mark loop
 [backport] Apply ShenandoahEvacOOMScope only for evac-taking paths in ShenandoahBarrierSet
 [backport] Replace risky SBS::need_update_refs_barrier with straightforward check
 [backport] Don't generate arraycopy pre-barrier calls for traversal GC
 [backport] Pre-filter oops before enqueing them in SBS slowpaths
 [backport] Implement arraycopy post-processing in GC threads for Traversal GC
 [backport] SATB buffer filtering/compaction hides unmarked objects until final-mark
 [backport] Process remaining SATB buffers in final mark/traverse loop instead of separate phase
 [backport] Skip RESOLVE in SATBBufferClosure if no forwarded objects are in heap
 [backport] ShenandoahRootProcessor corrupts none Shenandoah VM_Operations
 [backport] VSC++ requires space(s) in between two string literals
 [backport] Missing Shenandoah entry in GCNameHelper::to_string
 [backport] Adaptive thresholds for Traversal

Testing: tier3_gc_shenandoah {fastdebug|release}

Thanks,
-Aleksey



More information about the shenandoah-dev mailing list