Uploaded image for project: 'JDK'
  1. JDK
  2. JDK-8153224

Monitor deflation prolong safepoints

XMLWordPrintable

    • Icon: Enhancement Enhancement
    • Resolution: Fixed
    • Icon: P3 P3
    • 15
    • 8u72, 9
    • hotspot
    • b26
    • generic
    • generic

      In applications with non-trivial amounts of lock contention the time to deflate idle monitors can prolong safepoints by a non-trivial amount.

      Today monitors are deflated by the VM thread in the beginning of every safepoint. In JDK-7021979 it was observed that 250,000 monitors can easily take up to 8ms to process. The 250,000 monitors were created artificially using a small java program.

      At Twitter we have seen programs with more than 600,000 monitors and 6500 threads. The ratio between threads and monitors allocated is less than 10% of the MAXPRIVATE (1024) private monitors per thread, suggesting these program do not suffer from the issue reported in JDK-7021979.

        1.
        		 baseline_cleanups from Async Monitor Deflation project Sub-task Resolved Daniel Daugherty  
        2.
        		 monitor_logging updates from Async Monitor Deflation project Sub-task Resolved Daniel Daugherty  
        3.
        		 more monitor logging updates from Async Monitor Deflation project Sub-task Resolved Daniel Daugherty  
        4.
        		 more baseline cleanups from Async Monitor Deflation project Sub-task Resolved Daniel Daugherty  
        5.
        		 is_busy diagnostics and other baseline cleanups from Async Monitor Deflation project Sub-task Resolved Daniel Daugherty  
        6.
        		 rename, whitespace, indent and comments changes in preparation for lock free Monitor lists Sub-task Resolved Daniel Daugherty  
        7.
        		 baseline cleanups from Async Monitor Deflation v2.0[789] Sub-task Resolved Daniel Daugherty  
        8.
        		 ObjectSynchronizer::FastHashCode() cleanups from Async Monitor Deflation project Sub-task Resolved Daniel Daugherty  
        9.
        		 replace monitor list mux{Acquire,Release}(&gListLock) with spin locks Sub-task Resolved Daniel Daugherty  
        10.
        		 add OM_CACHE_LINE_SIZE and use smaller size on SPARCv9 and X64 Sub-task Resolved Daniel Daugherty  
        11.
        		 refactor ObjectMonitor::set_owner() and _owner field setting Sub-task Resolved Daniel Daugherty  

            dcubed Daniel Daugherty
            cvarming Carsten Varming (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            12 Start watching this issue

              Created:
              Updated:
              Resolved: