< prev index next >
src/hotspot/share/gc/g1/g1_globals.hpp
Print this page
rev 55404 : 8226197: Reducing G1?s CPU cost with simplified write post-barrier and disabling concurrent refinement
Summary: A prototype to add a mode for G1 to use a simplified write post-barrier. Guarded by new flag G1FastWriteBarrier.
@@ -243,10 +243,14 @@
\
product(uintx, G1MixedGCCountTarget, 8, \
"The target number of mixed GCs after a marking cycle.") \
range(0, max_uintx) \
\
+ product(bool, G1FastWriteBarrier, false, \
+ "Use simplified write barrier that does not use per-thread " \
+ "dirty card queue and concurrent remset refinement.") \
+ \
experimental(bool, G1PretouchAuxiliaryMemory, false, \
"Pre-touch large auxiliary data structures used by the GC.") \
\
experimental(bool, G1EagerReclaimHumongousObjects, true, \
"Try to reclaim dead large objects at every young GC.") \
< prev index next >