Wed, 18 Nov 2015 10:46:21 -1000 8143151: [JVMCI] assertion for allocation of "too many" CompileTasks must take JVMCI into account
twisti [Wed, 18 Nov 2015 10:46:21 -1000] rev 34197
8143151: [JVMCI] assertion for allocation of "too many" CompileTasks must take JVMCI into account Reviewed-by: iignatyev, twisti Contributed-by: Doug Simon <doug.simon@oracle.com>
Wed, 18 Nov 2015 09:43:31 -1000 8141330: [JVMCI] avoid deadlock between application thread and JVMCI compiler thread under -Xbatch
twisti [Wed, 18 Nov 2015 09:43:31 -1000] rev 34196
8141330: [JVMCI] avoid deadlock between application thread and JVMCI compiler thread under -Xbatch Reviewed-by: twisti Contributed-by: Doug Simon <doug.simon@oracle.com>
Wed, 18 Nov 2015 03:03:43 +0300 8139595: MethodHandles::remove_dependent_nmethod is not MT safe
vlivanov [Wed, 18 Nov 2015 03:03:43 +0300] rev 34195
8139595: MethodHandles::remove_dependent_nmethod is not MT safe Reviewed-by: jrose, coleenp
Mon, 02 Nov 2015 15:52:37 +0100 8141137: C2 fails rematerializing nodes using flag registers.
goetz [Mon, 02 Nov 2015 15:52:37 +0100] rev 34194
8141137: C2 fails rematerializing nodes using flag registers. Summary: Don't rem. if input stretches several live ranges. If rem., don't add SpillCopy on RegFlags edge. Reviewed-by: kvn, vlivanov
(0) -30000 -10000 -3000 -1000 -300 -100 -30 -10 -4 +4 +10 +30 +100 +300 +1000 +3000 +10000 tip