是否有人可以在启用CMS GC的Java运行时提供以下verbosegc行的一些细节?明确地:
-
这项活动在哪里举行
29.490
指示用于标记的高水位堆位置?
-
问题是什么
preclean
-
问题是什么
reset
阶段?如果这不仅仅是扫荡的结束,为什么还要花时间来完成呢?
-
Rescan
,
grey
root
remark
活动的区域和指标
29.653
? 我们是否应该假设这些只是后续扫描的报告堆度量?如果是这样,他们为什么要浪费时间?
verbosegc日志摘录:
29.490: [GC [1 CMS-initial-mark: 16225K(17684K)] 16737K(22420K), 0.0030839 secs] [Times: user=0.00 sys=0.00, real=0.00 secs]
29.494: [CMS-concurrent-mark-start]
29.652: [CMS-concurrent-mark: 0.158/0.158 secs] [Times: user=0.16 sys=0.00, real=0.16 secs]
29.652: [CMS-concurrent-preclean-start]
29.652: [CMS-concurrent-preclean: 0.000/0.000 secs] [Times: user=0.00 sys=0.00, real=0.00 secs]
29.653: [GC[YG occupancy: 512 K (4736 K)]29.653: [Rescan (non-parallel) 29.654: [grey object rescan, 0.0001517 secs]29.654: [root rescan, 0.0030887 secs], 0.0033817 secs]29.657: [weak refs processing, 0.0000271 secs] [1 CMS-remark: 16225K(17684K)] 16737K(22420K), 0.0035183 secs] [Times: user=0.00 sys=0.00, real=0.00 secs]
29.658: [CMS-concurrent-sweep-start]
29.679: [CMS-concurrent-sweep: 0.022/0.022 secs] [Times: user=0.03 sys=0.00, real=0.03 secs]
29.680: [CMS-concurrent-reset-start]
29.703: [CMS-concurrent-reset: 0.022/0.022 secs] [Times: user=0.02 sys=0.00, real=0.01 secs]
29.797: [GC 29.797: [DefNew: 4736K->512K(4736K), 0.0207183 secs] 20800K->17172K(31512K), 0.0209119 secs] [Times: user=0.01 sys=0.02, real=0.03 secs]
